ANTHRACENYL-TETRALACTAM MACROCYCLES AND THEIR USE IN DETECTING A TARGET SACCHARIDE

Abstract

A water-soluble compound of the formula (I):

##STR00001##

wherein R.sup.9 and R.sup.10 are suitably hydrophilic substituents, which may be used to selectively bind to a target saccharide such as glucose and which exhibits a detectable spectroscopic response to such binding, thus enabling its use in the detection and correction of blood glucose concentrations in vivo.

Claims

1. A water-soluble compound of the formula (I): ##STR00010## wherein R.sup.1 to R.sup.8 are each independently selected from hydrogen; optionally substituted alkyl groups; optionally substituted cycloalkyl groups; optionally substituted heterocyclyl groups; optionally substituted alkenyl groups; optionally substituted alkynyl groups; optionally substituted aryl groups; optionally substituted heteroaryl groups; alkoxyl groups; ketone and aldehyde groups; carboxylic acids and carboxylate ions; carboxylate esters; SO.sub.3H; SO.sub.3.sup.; OSO.sub.3H; OSO.sub.3.sup.; PO.sub.3XY where X and Y are independently hydrogen, alkyl or a negative charge; OPO.sub.3XY where X and Y are independently hydrogen, alkyl or a negative charge; amines; amides; halo groups; CN; NO.sub.2; OH; and imino and imido groups, provided that in any one or more of the pairs R.sup.1R.sup.2, R.sup.3R.sup.4, R.sup.5R.sup.6 and R.sup.7R.sup.8, the two substituents may be joined together to form part of an optionally substituted cyclic group; and R.sup.9 and R.sup.10 are each independently selected from hydrogen; optionally substituted alkyl groups; optionally substituted cycloalkyl groups; optionally substituted heterocyclyl groups; optionally substituted alkenyl groups; optionally substituted alkynyl groups; optionally substituted aryl groups; optionally substituted heteroaryl groups; alkoxyl groups; ketone and aldehyde groups; carboxylic acids and carboxylate ions; carboxylate esters; SO.sub.3H; SO.sub.3.sup.; OSO.sub.3H; OSO.sub.3.sup.; PO.sub.3XY where X and Y are independently hydrogen, alkyl or a negative charge; OPO.sub.3XY where X and Y are independently hydrogen, alkyl or a negative charge; amines; amides; halo groups; CN; NO.sub.2; OH; and imino and imido groups; or R.sup.9 and R.sup.10 are each independently selected from hydrogen or hydrophilic substituents containing one or more hydrophilic functional groups selected from polar groups; and wherein one or more of the substituents R.sub.1 to R.sub.10 is a hydrophilic substituent containing one or more hydrophilic functional groups selected from polar groups.

2. A compound according to claim 1, wherein R.sup.9 and R.sup.10 are each independently selected from hydrogen and hydrophilic substituents containing one or more hydrophilic functional groups selected from polar groups, provided that at least one of R.sup.9 and R.sup.10 is a hydrophilic substituent containing one or more hydrophilic functional groups selected from polar groups.

3. A compound according to claim 1 or claim 2, wherein said one or more hydrophilic functional groups are selected from carboxylic acids, carboxylate ions, carboxylate esters, hydroxyl, amines, amides, ethers, ketone and aldehyde groups, NO.sub.2, sulphates, sulphonates, phosphates, phosphonates, and combinations thereof.

4. A compound according to any preceding claim, wherein said one or more hydrophilic functional groups are selected from carboxylic acids, carboxylate ions, amides, ethers and combinations thereof.

5. A compound according to any preceding claim, wherein R.sup.1 to R.sup.8 are selected from hydrogen, carboxylate esters, alkoxyl groups, optionally substituted cyclic imido groups, hydroxyl and sulphonates.

6. A compound according to any preceding claim, wherein R.sup.1 to R.sup.8 are all hydrogen.

7. A compound according to any preceding claim, wherein the at least one hydrophilic substituent is selected from groups of the formula C(O)R.sup.14, where R.sup.14 is selected from: a. groups NR.sup.15C(R.sup.16CO.sub.2H).sub.3 in which R.sup.15 is selected from hydrogen and C1 to C4 alkyl; and R.sup.16 is a group (CH.sub.2).sub.n, where n is an integer from 1 to 6, optionally containing an ether group O; b. groups NR.sup.15C(R.sup.17).sub.3 in which R.sup.15 is as defined above; R.sup.17 is a group R.sup.18C(O)NR.sup.15C(R.sup.18CO.sub.2H).sub.3; and each R.sup.18 is independently selected from groups R.sup.16 as defined above; and c. groups NR.sup.15C(R.sup.25).sub.3 in which R.sup.15 is as defined above; R.sup.25 is a group R.sup.18C(O)NR.sup.15C(R.sup.26).sub.3; R.sup.26 is a group R.sup.18C(O)NR.sup.15C(R.sup.18CO.sub.2H).sub.3; and each R.sup.18 is independently selected from groups R.sup.16 as defined above.

8. A compound according to any one of claims 1 to 7, which has one of the formulae shown below: ##STR00011##

9. A compound according to any one of claims 1 to 7, which has the formula: ##STR00012## or a salt or protected form thereof.

10. A compound according to any one of the preceding claims, which exhibits a spectroscopic response on complexing with a target saccharide, in particular glucose, which spectroscopic response is preferably detectable in the visible and/or near-infrared region of the electromagnetic spectrum.

11. A compound according to any one of the preceding claims, which is immobilised on or in a solid or semi-solid support.

12. A compound according to claim 11, wherein the solid or semi-solid support is a polymeric matrix and/or a gel, such as a hydrogel.

13. A compound according to claim 12, wherein the polymeric matrix and/or gel is a polymer selected from cross-linked polyethylene glycol and/or polyacrylamide.

14. A compound according to claim 12 or 13, wherein the compounds are chemically linked to the polymeric matrix and/or gel.

15. A compound according to claim 12 or 13, wherein the compounds are physically incorporated within the polymeric matrix and/or gel via non-covalent interactions.

Description

[0193] The present invention will now be further described with reference to the following non-limiting examples, and the accompanying illustrative drawings, of which:

[0194] FIG. 1A shows a synthetic lectin 2, as described above, which has been previously reported for use in the detection of glucose; FIG. 1B shows a compound 3, also for use in the detection of glucose, in accordance with the present invention; FIG. 1C shows the compound 3 complexed with glucose;

[0195] FIGS. 2A, 2B, and 2C illustrate, schematically, aspects of the design and synthesis of the compound 3 and of its interactions with glucose molecules;

[0196] FIG. 3 illustrates the ground state conformation of a molecule of 3 as predicted by Monte Carlo molecular mechanics calculations;

[0197] FIG. 4 shows a reaction scheme for preparing compound 3, as discussed in Example 1 below;

[0198] FIGS. 5A, 5B and 5C show reaction schemes suitable for preparing asymmetric versions of compounds according to the invention;

[0199] FIGS. 6A and 6B show in more detail a reaction scheme for preparing compound 9, as discussed in Example 2 below;

[0200] FIG. 7 shows the structures of test substrates used in binding studies with the compounds 3 and 2;

[0201] FIGS. 8A, 8B, 8C, and 8D show data from binding studies on compound 3 and glucose, in the form of partial .sup.1H NMR spectra, binding curves, fluorescence titration data and ITC (isothermal titration microcalorimetry) data, as referred to in Examples 3 to 6 below;

[0202] FIG. 9 shows the labelling system used for NMR binding and structural studies on the compound 3 and a methyl--D-glucose molecule 10;

[0203] FIGS. 10A and 10B show NMR-based structures for a complex formed between the compound 3 and methyl--D-glucoside;

[0204] FIG. 11 illustrates schematically a detection device, and a detection and supply system, according to the invention;

[0205] FIGS. 12A and 12B show further compounds 13 and 14 according to the invention;

[0206] FIGS. 13A, 13B and 13C show retrosynthetic schemes for the synthesis of compounds 13 and 14, as described in Examples 9 and 11 respectively below;

[0207] FIGS. 14A-141 show the structural formulae for compounds prepared in Examples 9 and 11;

[0208] FIGS. 15A-15D show data from studies on compound 13, in the form of fluorescence emissions spectra, partial .sup.1H NMR spectra and binding selectivity data, as referred to in Example 10 below;

[0209] FIG. 16 shows the structural formula for the compound 3 bound to a poly[acryloyl-bis(aminopropyl)polyethylene glycol], as described in Example 12 below;

[0210] FIGS. 17A and 17B show schematically the methods by which substituted anthracene precursor compounds were synthesised in Example 13 below, in order to test the effects of their substituents on their emissions spectra; and

[0211] FIG. 18 shows fluorescence spectra for four anthracene precursor molecules, as tested in Examples 13A, 13B and 13C.

EXAMPLE 1DESIGN & SYNTHESIS OF COMPOUND 3

[0212] The compound 3, shown in FIG. 1B, is a compound according to the first aspect of the invention, designed as a synthetic lectin analogue for the purpose of detecting glucose in human blood. The compound 3 has been shown to be capable of associating with a molecule of glucose (ie D-glucose), which is able to occupy the cavity defined by the two aromatic anthracene moieties and the two bridging isophthaloyl groupssee FIG. 1C. Compound 3 can be seen to have a much simpler structure than that of the previously reported synthetic lectin 2 which is shown in FIG. 1A and described above.

[0213] Underlying the present invention is the unexpected discovery that condensed aromatic units can play a useful role in the design of improved synthetic lectins. Contact between aromatic surfaces and carbohydrate CH groups is often observed in lectin-saccharide complexes, and it is widely thought that CH- interactions, allied to hydrophobic effects, can make important contributions to binding, as indeed in the compound 2 of FIG. 1A. In previously prepared synthetic lectins, for example compound 2, the aromatic surfaces have been provided by oligophenyl units. However, though helpful synthetically, the biphenyl bond tends to twist due to steric interference between ortho hydrogens, and this can disturb the interactions between rigidly positioned axial CH groups and the aromatic surfaces. In contrast, a condensed aromatic unit can make ideal contact with an array of axial CH groups. Moreover, a carbohydrate molecule can slide across the surface of the aromatic unit without significant loss of binding energy, so that (a) other interactions can be maximised and (b) some freedom of movement can be retained within the complex (hence less entropy loss on binding). Such effects are illustrated schematically in FIGS. 2A and 2B, which show the interactions of both biphenyl (FIG. 2A) and condensed aromatic (FIG. 2B) units with a -D-glucose molecule.

[0214] The use of condensed aromatic moieties in the compounds of the present invention can provide additional advantages in the context of the detection of saccharides. These moieties tend to be strongly absorbent of electromagnetic radiation and also fluorescent, with their emissions being modulated on association with a target saccharide such as glucose.

[0215] It has been found that a compound such as 3 can be prepared in just two steps: cyclisation of suitably protected forms of the constituent bis-anthracenyl and isophthaloyl moieties, followed by deprotection of the pendant solubilising groups (in this case, NHC(CH.sub.2OCH.sub.2CH.sub.2CO.sub.2.sup.).sub.3 groups, which can be protected during the cyclisation step with, for example, t-butyl groups). Such a reaction is shown schematically in FIG. 2C, according to which the monocycle 3 is prepared by reacting the diamine 4 (bis-(aminomethyl)anthracene) with the isophthaloyl spacer component 5.

[0216] The PFP groups function as leaving groups in 5, whilst the C(O)Y group is a protected form of a hydrophilic, water-solubilising substituent.

[0217] The FIG. 2C reaction is a method according to the eleventh aspect of the invention.

[0218] Monte Carlo molecular mechanics calculations suggested that the molecule 3 could adopt a range of conformations with different angles between its aromatic surfaces, but that all low-energy structures would feature a cleft or cavity (as seen in FIG. 3; solubilising side chains removed for clarity; anthracene units shown in space-filling mode). It was not clear, however, that this simple and rather flexible architecture would favour any particular saccharide, or indeed that it would show any notable carbohydrate-binding properties. However the enclosed, amphiphilic nature of the cavity did seem generally suitable for carbohydrate recognition, although the tilted arrangement of the anthracene moieties might not be ideal for certain saccharide molecules.

[0219] Compound 3 was prepared in 23% yield using the route shown in FIG. 2C. The diamine component 4 is available commercially, but can also conveniently be synthesised by bis-bromomethylation of anthracene followed by treatment with hexamethylenetetramine [Gunnlaugsson et al, Org Lett 4: 2449-2452 (2002)]. Diester 5 was prepared via a 3-step procedure which involved treatment of tris(hydroxymethyl)aminomethane with t-butyl acrylate; reaction of the resulting amine with 1,3,5-benzenetricarbonyl trichloride (followed by hydrolysis of unreacted acid chloride groups); and conversion of carboxylic acid groups to PFP esters using N,N-dicyclohexylcarbodiimide (DCC).

[0220] The detailed method for synthesising 3 was as follows (reaction scheme shown in FIG. 4).

[0221] t-Butyl Protected Macrocycle S1:

[0222] Firstly, the bis-pentafluorophenyl ester 5 was prepared in three steps from tris(hydroxymethyl)aminomethane, t-butyl acrylate and benzene-1,3,5-tricarbonyl chloride [see Klein, E et al, Angew Chem, Int Ed, 44: 298-302 (2005)]. A solution of 5 (1.6 g, 1.55 mmol) was then prepared in anhydrous THF (45 mL), and the solution was added dropwise over 30 hours (syringe pump) to a solution of 9,10-bis(aminomethyl)anthracene 4 (367 mg, 1.55 mmol) and DIPEA (5 mL) in anhydrous THF (1 L) under nitrogen. After stirring for a further 24 hours, the solvent was removed under reduced pressure. The residue was dissolved in DCM (100 mL) and washed with saturated aqueous NH.sub.4Cl (100 mL), water (100 mL) and brine (100 mL). The organic solution was dried over Na.sub.2SO.sub.4 and evaporated in vacuo. The residue was taken up in DMSO (12 mL) and insoluble material removed by a syringe filter (0.45 m). The DMSO solution was injected into a preparative HPLC apparatus fitted with a reverse phase column (Hichrom Kromasil, 15021.2 mm, 5 m) and eluted with methanol/water (90:10 to 100:0 over 5 minutes, then 100:0 for a further 15 minutes; flow rate=20 mL/min). The component eluting at 8.5 minutes was collected and freeze-dried to yield macrocycle S1 (370 mg, 0.21 mmol, 27%) as a pale yellow powder. .sup.1H NMR (500 MHz, CDCl.sub.3, TMS standard) =8.49 (d, 4H, J=1.7 Hz, ArH), 8.30 (dd, 8H, .sup.2J=7.1 Hz, .sup.3J=3.8 Hz, AnH), 7.46 (dd, 8H, .sup.2J=7.1 Hz, .sup.3J=3.8 Hz, AnH), 7.43 (s, 2H, ArH), 6.70 (s, 2H, NHC(CH.sub.2O).sub.3), 6.45 (t, 4H, .sup.3J=4.5 Hz, AnCH.sub.2NH), 5.52 (d, 8H, .sup.2J=5.1 Hz, AnCH.sub.2NH), 3.87 (s, 12H, C(CHO).sub.3), 3.73 (t, 12H, .sup.3J=6.4 Hz, CH.sub.2CH.sub.2O), 2.51 (t, 12H, .sup.3J=6.4 Hz, CH.sub.2CH.sub.2O), 1.39 (s, 54H, C(CH.sub.3).sub.3). .sup.13C NMR (125 MHz, CDCl.sub.3) =171. 13 (CH.sub.2CO.sub.2), 165.82 (AnCH.sub.2NHCOAr), 161.27 (ArCONHC), 134.75 (Ar), 130.20 (An), 129.69 (An), 126.52 (An), 124.70 (An), 80.68 (C(CH.sub.3).sub.3), 69.13 (C(CH.sub.2O).sub.3), 67.26 (OCH.sub.2CH.sub.2CO.sub.2t-Bu), 60.57 (C(CH.sub.2O).sub.3), 37.28 (AnCH.sub.2NH), 36.46 (OCH.sub.2CH.sub.2CO.sub.2t-Bu), 28.15 (C(CH.sub.3).sub.3). Ananthracenyl, Arisophthalamide aryl. HRMS (ESI): m/z calculated for C.sub.100H.sub.126O.sub.24N.sub.6Na.sub.2.sup.2+[M+2Na.sup.2+]=920.4304, found: 920.4272.

[0223] Also isolated was the corresponding [3+3] macrocycle (100 mg, 0.037 mmol, 5% yield, retention time=10 min).

[0224] Receptor 3 (Sodium Salt):

[0225] Macrocycle S1 (200 mg, 0.11 mmol) was dissolved in DCM (20 mL) and cooled in ice. Trifluoroacetic acid (TFA) (5 mL) was added dropwise to the solution. The reaction was allowed to warm to room temperature and stirred for 3 hours.

[0226] The solvent was removed in vacuo, and the residue was suspended in water (5 mL).

[0227] NaOH aq (0.5 M) was added dropwise until the suspended material dissolved, forming a clear solution. The clear solution was freeze-dried and further purified by preparative HPLC (apparatus as above), eluting with methanol/water (5:95 to 30:70 over 15 minutes, then to 100:0 over a further 15 minutes; flow rate=20 mL/min). The component with retention time=15 minutes was collected and freeze-dried to yield macrocycle 3 (150 mg, 85%). .sup.1H NMR (500 MHz, D.sub.2O) =8.39 (s, 4H, ArH), 8.27 (s, 8H, AnH.sub.A (for labelling see FIG. 9; anthracene protons A and B were distinguished through the intensities of the NOESY cross-peaks to AnCH.sub.2NH)), 7.93 (bs, 2H, ArH), 7.51 (bs, 8H, AnH.sub.B, 5.46 (bs, 8H, AnCH.sub.2NH), 3.90 (s, 12H, C(CHO).sub.3), 3.78 (bs, 12H, OCH.sub.2CH.sub.2CO.sub.2Na), 2.48 (bs, 12H, OCH.sub.2CH.sub.2CO.sub.2Na). .sup.13C NMR (125 MHz, D.sub.2O) =179.98 (CH.sub.2CO.sub.2Na), 168.32 (AnCH.sub.2NHCOAr), 162.87 (ArCONHC), 136.56 (Ar), 134.44 (Ar), 130.26 (An), 129.45 (An), 126.87 (An), 124.88 (An), 69.48 (C(CH.sub.2O).sub.3), 69.37 (OCH.sub.2CH.sub.2CO.sub.2Na), 61.51 (C(CH.sub.2O).sub.3), 38.13 (AnCH.sub.2NH), 37.53 (OCH.sub.2CH.sub.2CO.sub.2Na). Ananthracenyl, Arisophthalamide aryl. HRMS (ESI): m/z calculated for C.sub.76H.sub.79N.sub.6O.sub.24.sup.+[hexacarboxylic acid form+H.sup.+]=1459.5169, found: 1459.5140.

EXAMPLE 2SYNTHESIS OF ASYMMETRIC COMPOUNDS

[0228] For comparison purposes, an asymmetric alternative, compound 9, was also synthesised using a method according to the invention, as depicted in FIG. 5A. In compound 9, a single anthracene unit is paired with a smaller p-xylyl unit. Although a longer sequence was required to prepare 9, the process was straightforward.

[0229] Firstly, the isophthaloyl moiety 6 was substituted with a t-butyl-protected form of the hydrophilic, water-solubilising group NHC(CH.sub.2OCH.sub.2CH.sub.2CO.sub.2.sup.).sub.3, by reacting 6 with the amine 7. This yielded the compound 8, in which one of the potentially reactive C(O) groups was protected with a methoxyl group MeO. Compound 8 was then reacted with the diamine-substituted bis-anthracene 4, to yield an intermediate (referred to below as compound 11) in which a single anthracene moiety was bound to two isophthaloyl moieties.

[0230] The methoxyl protecting group on compound 11 was then replaced by the leaving group O-PFP, using LiOH followed by PFP-OH and DCC, to yield a further reactive intermediate (referred to below as compound 12). Subsequently, 12 was reacted with p-xylylenediamine, in the presence firstly of TFA and then NaOH, to yield the final compound 9. In compound 9, the solubilising groups R.sup.9 and R.sup.10 are also now in their deprotected (ie carboxylate) forms.

[0231] Alternatively, the intermediate 12 could be prepared by directly combining the reactants 4 and 5 (as in the preparation of the symmetric compound 3), so long as the compound 5 is present in moderate excess, for example at a molar ratio of 4:5 of around 1:3. This method is shown in FIG. 5B.

[0232] Aside from delivering a useful control compound, the routes shown in FIGS. 5A and 5C can be adapted to prepare a variety of asymmetric analogues of compound 3. It should thus be possible to tune the binding and/or optical properties of a compound according to the invention, for example by varying the substituents on the two anthracene units.

[0233] The detailed method for synthesising 9 was as follows (reaction scheme shown in FIGS. 6A and 6B).

[0234] Methyl 1,3,5-Benzenetricarboxylate S2.

[0235] Trimethyl 1,3,5-benzenetricarboxylate (22.0 g, 87.2 mmol) was dissolved in MeOH (700 mL). NaOH aq (6.97 g, 174.3 mmol NaOH in 100 mL water) was added dropwise with stirring. Stirring was continued at room temperature overnight, then the solvent was removed and the crude white solid was dissolved in saturated NaHCO.sub.3 aq (600 mL). The pH of the solution was adjusted to 5.5 by adding HCl aq (1 M), and the aqueous solution was extracted with EtOAc (250 mL3) to remove dimethyl 1,3,5-benzenetricarboxylate. The pH of the aqueous solution was then further adjusted to 4.4 and extracted with EtOAc (250 mL3). The organic phases were combined, washed with brine, dried over MgSO.sub.4 and concentrated in vacuo to afford S2 as a white solid (R.sub.f=0.5, EtOAc:MeOH:H.sub.2O=80:20:1). Yield 62% (12.2 g, 54.5 mmol). .sup.1H NMR (400 MHz, (CD.sub.3).sub.2CO) =8.86 (t, J=1.7 Hz, 1H), 8.82 (d, J=1.7, 2H), 3.98 (s, 3H). .sup.13C NMR (100 MHz, (CD.sub.3).sub.2CO) =166.1, 165.9, 135.3, 135.0, 132.8, 132.8, 132.4, 53.1. This material was used without further purification.

[0236] Pentafluorophenyl Ester 8.

[0237] Dicarboxylic acid S2 (6.00 g, 26.8 mmol) was dissolved in anhydrous THF (500 mL). Pentafluorophenol (11.04 g, 60.0 mmol) and N,N-dicyclohexylcarbodiimide (DCC) (12.8 g, 62 mmol) were added under nitrogen atmosphere at room temperature and the mixture was stirred overnight. Amine S3 [Klein, E et al, Carbohydrate recognition in water by a tricyclic polyamide receptor, Angew Chem, Int Ed 44: 298-302 (2005)] (13.5 g, 26.8 mmol) was dissolved in anhydrous THF (150 mL) with N,N-diisopropylethylamine (6.93 g, 53.6 mmol) and a catalytic amount of 4-dimethylaminopyridine (330 mg, 2.7 mmol, 5 mol %). This solution was added to the reaction mixture dropwise over 1 hour, after which the mixture was stirred for a further 24 hours under nitrogen. The solvent was evaporated, the crude product was suspended in diethyl ether (75 mL) and insoluble residues were removed by filtration. Concentration of the filtrate and purification by column chromatography on silica gel (EtOAc/hexane, 15:85 to 30:70), gave the product 8 as a clear oil (15.5 g, 17.4 mmol, 65%). .sup.1H NMR (400 MHz, CDCl.sub.3) =8.91 (t, J=1.6 Hz, 1H), 8.75 (t, J=1.6 Hz, 1H), 8.70 (t, J=1.6 Hz, 1H), 7.09 (s, 1H), 3.98 (s, 3H), 3.85 (s, 6H), 3.69 (t, J=6.2 Hz, 6H), 2.47 (t, J=6.2 Hz, 6H), 1.36 (s, 27H). .sup.13C NMR (100 MHz, CDCl.sub.3) =28.0, 36.2, 52.9, 61.0, 67.2, 69.1, 81.1, 125.4 (t, J.sub.CF=13.2 Hz), 127.9, 131.6, 133.8, 134.0, 134.2, 136.9, 137.3, 139.2, 140.6, 141.1, 142.6, 161.4, 165.3, 166.5, 171.6. HRMS (ESI): m/z calculated for C.sub.41H.sub.51F.sub.5NNaO.sub.14.sup.+[M+Na.sup.+]=900.3200, found: 900.3178.

[0238] Bis-Methyl Ester Intermediate S4.

[0239] Diamine 4 (200 mg, 0.85 mmol) and pentafluorophenyl ester 8 (880 mg, 1.00 mmol) were dissolved in anhydrous THF (30 mL) under nitrogen. N,N-diisopropylethylamine (2 mL, 1.51 g, 12 mmol) was added. The mixture was stirred overnight at room temperature, after which analysis by TLC indicated that the reaction was complete. The solvent was removed and the residue was purified by column chromatography on silica gel (hexane/EtOAc, 1:1 then 3:4) to obtain intermediate S4 as a yellow solid (630 mg, 78%). R.sub.f=0.5 (hexane/EtOAc, 2:3). .sup.1H NMR (400 MHz, CDCl.sub.3) =8.73 (t, J=1.6 Hz, 2H), 8.57 (t, J=1.6 Hz, 2H), 8.46 (dd, J=6.9, 3.3 Hz, 4H), 8.15 (t, J=1.7 Hz, 2H), 7.60 (dd, J=6.9, 3.2 Hz, 4H), 7.50 (t, J=4.4 Hz, 2H), 6.50 (s, 2H), 5.69 (d, J=4.6 Hz, 4H), 3.93 (s, 6H), 3.71 (s, 12H), 3.52 (t, J=6.2 Hz, 12H), 2.17 (t, J=6.2 Hz, 12H), 1.20 (s, 54H). .sup.13C NMR (100 MHz, CDCl.sub.3) =171.1, 165.9, 165.7, 165.6, 135.6, 134.8, 131.9, 131.7, 131.2, 130.4, 129.6, 129.1, 126.4, 124.9, 80.6, 68.9, 67.0, 60.3, 52.4, 36.0, 27.8. HRMS (ESI): m/z calculated for C.sub.90H.sub.127N.sub.4O.sub.26.sup.+[M+H.sup.+]=1623.8584, found: 1623.8610.

[0240] Dicarboxylic Acid Intermediate S5.

[0241] Intermediate S4 (630 mg, 0.39 mmol) was dissolved in THF (30 mL) at room temperature. LiOH.H.sub.2O (170 mg, 3.90 mmol) was added, followed by H.sub.2O (3 mL). The mixture was stirred overnight at room temperature then the solvent was removed by evaporation, keeping the temperature below 40 C. The residue was dissolved in H.sub.2O (30 mL), and the pH was adjusted to ca 4-5 by addition of HCl aq. The mixture was extracted with EtOAc (260 mL) and the combined organic phases were dried over Na.sub.2SO.sub.4. Evaporation of the solvent gave diacid S5 as a clear oil, which was used without further purification (593 mg, 95%). .sup.1H NMR (400 MHz, DMSO-d.sub.6) =9.25 (s, 2H), 8.55 (dd, J=7.0, 3.3 Hz, 4H), 8.47 (d, J=1.5 Hz, 2H), 8.39 (s, 2H), 8.37 (s, 2H), 7.70-7.54 (m, 6H), 5.56 (d, J=4.4 Hz, 4H), 3.64 (s, 12H), 3.54 (t, J=6.1 Hz, 12H), 2.35 (t, J=6.1 Hz, 12H), 1.31 (s, 54H). .sup.13C NMR (125 MHz, DMSO-do) =170.53, 170.44, 165.89, 138.84, 135.24, 134.14, 130.27, 129.77, 129.35, 126.83, 125.23, 124.28, 120.45, 79.40, 67.66, 66.20, 66.13, 59.97, 59.89, 38.15, 37.98, 37.81, 37.64, 37.47, 37.31, 37.14, 35.08, 35.01, 26.38, 26.33. HRMS (ESI): m/z calculated for C.sub.84H.sub.114N.sub.4O.sub.26Na.sup.+[M+Na.sup.+]=1617.7637, found: 1617.7614.

[0242] Bis-Pentafluorophenyl Ester S6.

[0243] Pentafluorophenol (170 mg, 0.93 mmol), DCC (191 mg, 0.93 mmol) and diacid S5 (593 mg, 0.37 mmol) were dissolved in anhydrous THF (100 mL) under nitrogen. 4-Dimethylaminopyridine (DMAP) (5 mg, 0.04 mmol) was added, and the mixture was stirred at room temperature overnight. The solvent was then removed and the residue was purified by column chromatography on silica gel (hexane/EtOAc, 2:1) to obtain the activated ester S6 as a yellow solid (420 mg, 60%). R.sub.f=0.8 (hexane: EtOAc, 1:1). .sup.1H NMR (500 MHz, CDCl.sub.3) =8.94 (t, J=1.7 Hz, 2H), 8.76 (t, J=1.6 Hz, 2H), 8.48 (dd, J=7.0, 3.2 Hz, 4H), 8.25 (s, 2H), 7.72 (d, J=4.7 Hz, 2H), 7.62 (dd, J=6.9, 3.2 Hz, 4H), 6.58 (s, 2H), 5.72 (d, J=4.7 Hz, 4H), 3.70 (s, 12H), 3.51 (t, J=6.1 Hz, 12H), 2.14 (s, 11H), 1.19 (s, 54H). .sup.13C NMR (100 MHz, CDCl.sub.3) =171.16, 165.42, 164.97, 161.47, 136.02, 135.24, 133.15, 130.42, 129.57, 128.09, 126.48, 124.90, 80.66, 68.85, 67.07, 66.94, 60.44, 36.85, 36.02, 27.92, 27.79, 27.71. HRMS (ESI): m/z calculated for C.sub.96H.sub.112N.sub.4O.sub.26F.sub.10N.sup.+[M+Na.sup.+]=1949.7275, found: 1949.7297.

[0244] t-Butyl Protected Macrocycle S7.

[0245] Bis-pentafluorophenyl ester S6 (400 mg, 0.21 mmol) was dissolved in anhydrous THF (40 mL) to make solution A. p-Xylenediamine (29 mg, 0.21 mmol) was dissolved in anhydrous THF (40 mL) to make solution B. Using a syringe pump, solutions A and B were then added simultaneously over 30 hours to a solution of DIPEA (5 mL, 53.8 mmol) in anhydrous THF (700 mL) under nitrogen. The reaction was stirred for a further 24 hours, then the solvent was evaporated and the residue was re-dissolved in CH.sub.2Cl.sub.2 (150 mL). The solution was washed with saturated aqueous NH.sub.4Cl (100 mL), brine (100 mL) and NaHCO.sub.3 (100 mL). The organic layer was collected and dried over MgSO.sub.4. The solvent was evaporated and the residue was purified by preparative HPLC using the previously-described apparatus, eluting with methanol/water (90:10 to 100:0 over 20 min; flow rate=20 mL/min). The component with retention time=11.3 min was collected and freeze-dried to yield macrocycle S7 (120 mg, 34%) as a light yellow solid. .sup.1H NMR (500 MHz, CDCl.sub.3) =8.64 (s, 2H), 8.39 (dd, J=6.9, 3.2 Hz, 4H), 8.13 (s, 2H), 7.81 (s, 2H), 7.64 (s, 2H), 7.59 (dd, J=6.9, 3.0 Hz, 4H), 7.10 (s, 4H), 6.68 (s, 2H), 6.28 (s, 2H), 5.69 (d, J=4.2 Hz, 4H), 4.43 (d, J=6.0 Hz, 4H), 3.86 (s, 12H), 3.72 (t, J=6.2 Hz, 12H), 2.48 (t, J=6.1 Hz, 12H), 1.39 (s, 54H). .sup.13C NMR (125 MHz, CDCl.sub.3) =171.2, 166.3, 166.2, 165.5, 137.2, 136.7, 135.0, 134.9, 130.6, 130.3, 129.5, 127.7, 127.6, 127.2, 126.8, 124.6, 80.7, 69.1, 67.2, 60.4, 43.8, 36.9, 36.5, 28.1.

[0246] Macrocyclic Hexacarboxylate 9.

[0247] Macrocycle S7 (120 mg, 0.07 mmol) was dissolved in DCM (20 mL) and cooled in ice. TFA (3 mL) was added dropwise and the solution was stirred for 3 hours at room temperature. The solvent was removed in vacuo, and the residue was suspended in water (5 mL). NaOH aq (0.5 M) was added drop-wise until the suspended material dissolved, forming a clear solution. The amount of NaOH was calculated as ca 8 equivalents with respect to S7. The clear solution was freeze-dried to obtain a light yellow powder (99% yield). This product was further purified by preparative HPLC (apparatus as above), eluting with methanol/water (5:95 to 30:70 over 15 minutes, then to 100:0 over a further 15 minutes; flow rate=20 mL/min). The component with retention time=12.1 min was collected and freeze-dried to yield macrocycle 9 (75 mg, 0.05 mmol, 71%). .sup.1H NMR (500 MHz, D.sub.2O) =8.48 (dd, J=6.9, 3.4 Hz, 4H), 8.38 (q, J=1.2, 0.8 Hz, 2H), 8.22 (q, J=1.2, 0.8 Hz, 2H), 7.84 (s, 2H), 7.67 (dd, J=6.9, 3.2 Hz, 4H), 7.14 (d, J=1.0 Hz, 4H), 5.65 (s, 4H), 4.44 (s, 4H), 3.89 (m, 12H), 3.79 (m, 12H), 2.48 (m, 12H). .sup.13C NMR (125 MHz, D.sub.2O) =180.2, 135.3, 134.4, 129.4, 127.2, 126.1, 124.5, 117.5, 115.1, 112.8, 68.8, 68.6, 60.9, 48.8, 37.7. HRMS (ESI): m/z calculated for C.sub.68H.sub.75N.sub.6O.sub.24.sup.+[hexacarboxylic acid form+H.sup.+]=1359.4817, found: 1359.4827.

EXAMPLE 3STRUCTURAL ANALYSIS OF COMPOUND 3

[0248] The structure and properties of the compound 3 prepared in Example 1 were investigated as follows.

[0249] The compound was dissolved in D.sub.2O at concentrations up to 4 mM giving clean, if slightly broadened, .sup.1H NMR spectra. Minor signal movements were observed on dilution to 1 mM but no further effect was observed below this concentration, implying that the system is monomeric at 1 mM or less.

[0250] The partial .sup.1H NMR spectrum for compound 3 alone can be seen in FIG. 8A, closest to the baseline.

EXAMPLE 4BINDING STUDIES.SUP.1.H NMR

[0251] The binding of receptor compound 3 to carbohydrate substrates, in aqueous solution, was studied initially by .sup.1H NMR titrations at 298 K. The saccharides used as test substrates are shown in FIG. 7.

[0252] Titrations were performed on a Varian 500 MHz spectrometer. Solutions of reducing carbohydrates were prepared in D.sub.2O and kept overnight at room temperature before the titration experiments, in order to ensure equilibration of anomers. In a typical titration, aliquots of carbohydrate solution were added to receptor solution (DSS as internal standard) and the .sup.1H NMR spectra were recorded. Variations in chemical shifts were entered into a specifically written non-linear least squares curve-fitting program implemented within Excel. Assuming 1:1 stoichiometry, the program calculates K.sub.a and the limiting change in chemical shift . The assumption is supported by the generally good fits between observed and calculated data.

[0253] The partial NMR spectra for binding studies with glucose (ie D-glucose), using 1.1 mM 3 with glucose at 0 to 200 mM, are shown in FIG. 8A. FIG. 9 provides a key to the peak assignments in the spectra.

[0254] The NMR data showed that the addition of some carbohydrates to 3 caused substantial changes to its NMR spectrum, especially to the chemical shifts of the isophthaloyl protons E and F. For example, the addition of glucose caused a downfield movement of the signal due to internally-directed protons E, with tending towards 0.8 ppm (FIG. 8A). The signal due to externally-directed protons F also shifted downfield, by 0.15 ppm. Small movements of the signals due to anthracene protons A and B were observed, while the spectrum sharpened considerably during the titration. The movements of protons E and F gave excellent fits to a 1:1 binding model: FIG. 8B shows the overlapping observed and calculated binding curves for the NMR proton E, yielding values for the association constant K.sub.a of 58 and 54 M.sup.1 respectively (average=56 M.sup.1).

[0255] Data for the other test substrates were analysed similarly, to give the binding constants listed in Table 1 below. Values for the tricyclic system 2 are recorded for comparison purposes.

TABLE-US-00001 TABLE 1 Table 1: Association constants measured by .sup.1H NMR titration in D.sub.2O at 298 K. For structures of test substrates see FIG. 7. Values denoted ~0 were too small for analysis. Errors were estimated at 10% for most cases where K.sub.a 10 M.sup.1. K.sub.a (M.sup.1) Substrate 3 2.sup. All-equatorial D-glucose 56 (58*, 60 monosaccharides 55.sup.) methyl -D-glucoside 10 96 (121*, 130 101.sup.) 2-deoxy-D-glucose 36 29 N-acetyl-D-glucosamine 10 7 D-xylose 9 17 D-glucuronic acid, sodium salt 24 methyl -D-glucuronide, 87 sodium salt Other methyl -D-glucoside 6 15 monosaccharides D-galactose 1 3 D-mannose 1 ~1 D-fructose ~0 L-rhamnose ~0 ~0 L-fucose ~0 3 D-arabinose 1 4 D-lyxose ~0 ~1 D-ribose ~0 6 Disaccharides and D-cellobiose 28 71 miscellaneous D-maltose 35 ~0 substrates D-lactose 16 (9*) 8 D-sucrose ~0 D-trehalose ~0 Mannitol ~0 sodium lactate ~0 *Measured by fluorescence titration in phosphate buffer solution (pH 7.1, 0.1M) at 298 K. .sup.Measured by ITC titration in water at 298 K. .sup.Data from Harwell et al, supra.

[0256] Given the relative simplicity of 3, one might expect reduced performance in comparison to 2. Remarkably, however, the two systems behave quite similarly, the main difference being that 3 is the more selective for glucose vs other monosaccharides. Thus, both 2 and 3 prefer the all-equatorial carbohydrate moieties, binding well to molecules containing the all-equatorial -glucosyl unit and compounds containing it, for example glucose (for which the K.sub.a values are almost identical), methyl -D-glucoside and, to a lesser extent, 2-deoxyglucose, N-acetylglucosamine and xylose (all three of which are all-equatorial molecules). Compound 3 also binds fairly strongly to anionic glucuronic acid derivatives. Selectivity against other monosaccharides is generally good for both systems, but again 3 is appreciably superior. Aside from methyl -D-glucoside, all non-target monosaccharides were bound by 3 with K.sub.a1 m.sup.1. With disaccharides, 3 seems to bind significantly to any system containing -glucosyl (cellobiose, maltose, lactose). Here there is a qualitative difference from 2, which binds cellobiose well but shows no affinity for maltose. Such binding affinities are not generally problematic, however, in the context of blood glucose monitoring, since molecules such as cellobiose, maltose and lactose are unlikely to be present in the bloodstream at significant concentrations compared to the likely glucose concentration.

[0257] Thus, it can be seen that the compound 3, according to the invention, demonstrates a surprisingly good affinity, and selectivity, for glucose, whilst also being simpler in structure and thus more readily accessible than the prior art compound 2. This illustrates the likely utility of compound 3, and related compounds, in the detection of blood glucose levels.

EXAMPLE 5BINDING STUDIESFLUORESCENCE SPECTROSCOPY

[0258] Receptor-carbohydrate complex formation can also be studied by fluorescence spectroscopy.

[0259] Fluorescence titrations were carried out at 298 K on a PerkinElmer LS45 spectrometer in PBS (phosphate buffered saline) buffer solution (pH 7.1, 100 mM). The carbohydrate stock solutions were prepared by dissolving the carbohydrates in buffer containing the receptor at the concentration to be used in the titration (thus avoiding dilution of the receptor during the experiment). The solutions were kept overnight at room temperature before the titration experiments, in order to ensure equilibration of anomers. The wavelength to be used for fluorescence excitation was determined by measurement of the UV-visible spectrum of receptor 3 in the presence of carbohydrates. 394 nm was chosen, because at this wavelength the absorption of receptor was almost independent of carbohydrate concentration. In a typical titration, aliquots of carbohydrate-receptor solution were added to receptor solution (2.5 mL) in a quartz cuvette (3 mL, 10 mm pathway). The solution was stirred for 2 minutes and left to stand for another 1 minute before the emission spectrum was recorded. Binding constants were calculated using non-linear curve fitting assuming 1:1 binding stoichiometry, employing both Kaleidagraph and a customised Excel spreadsheet. Errors were estimated at <5%.

[0260] The spectrum for compound 3 (18.8 M) with glucose is shown in FIG. 8C. It can be seen that on excitation with UV light at 394 nm, 3 emitted in the blue-violet region, peaking at 423 nm, with a quantum yield of 2.4% (20 m aqueous solution). Addition of glucose caused the emission intensity to increase by factors of up to 2.5. Analysis of the changes gave, again, an excellent fit to a 1:1 binding curve (shown as an inset in FIG. 8C; binding data at 423 nm; K.sub.a=58 M.sup.1). Binding constants obtained by this method were in good agreement with those measured by NMR titrations (see Table 1). Moreover these fluorescence characteristics are promising for practical glucose sensing, especially when compared to biphenyl-based synthetic lectins. The excitation wavelength is only just outside the visible region, thus relatively safe and obtainable with inexpensive UV LEDs. In contrast, the biphenyl-based systems require light at 280 nm for excitation, and produce far weaker emissions which do not always change on binding. The excitation wavelength for a compound according to the invention can moreover be tuned, for instance to bring it within the visible spectrum, by modification of the substituents R.sup.1 to R.sup.8 on the anthracene units. The observed mode of binding of compound 3 to glucose suggests that changes to the anthracene unit, in particular at its two ends, should have only minor effects on binding.

EXAMPLE 6BINDING STUDIESISOTHERMAL TITRATION CALORIMETRY

[0261] The binding of 3 to glucose and methyl -D-glucoside was studied by a third technique, isothermal titration calorimetry (ITC). Experiments were performed on a VP-ITC (Microcal, Inc., Northampton, Mass.) at 298 K. Stock solutions of carbohydrates were made up in pure HPLC grade water and allowed to equilibrate overnight. Receptor solutions were made up in pure water. All the solutions were degassed and thermostated using the ThermoVac accessory before titration. The sample cell volume was 1.4226 mL. Each titration experiment included 25-35 successive injections. The output trace for 3 and glucose is shown in FIG. 8D; analysis of these data yielded a value for K.sub.a of 55 M.sup.1).

[0262] Measured affinities were again consistent with those determined by NMR titrations (Table 1), and revealed that complexation was enthalpy-driven with significant negative entropies (eg for glucose, H=3.8 kcal mol.sup.1, TS=1.4 kcal mol.sup.1). This contrasts with the oligophenyl-based synthetic lectins, where binding entropies are positive (eg for 2+10, H=0.6, TS=2.3 kcal mol.sup.1). However, negative binding entropies are common for natural lectins. The observation of negative S does not preclude a role for hydrophobic interactions. Indeed, with fewer polar spacers, it seems likely that 3 is less dependent on polar interactions than tricyclic cages such as 2, and thus more reliant on the displacement of high-energy water. This is supported by experiments in less polar media, where H-bonding must dominate. Thus the organic soluble (t-butyl protected) precursor of 3 bound octyl -D-glucoside in chloroform with K.sub.a=3200 M.sup.1 . The corresponding value for a biphenyl-based system was 100 times higher.

[0263] The role of non-polar interactions was highlighted by studies on the control macrocycle 9. This compound possesses the same polar groups as 3, but provides less apolar surface for hydrophobic CH- interactions. Addition of some carbohydrates (eg glucose and xylose) to 9 yielded minor changes in the .sup.1H NMR spectrum of the macrocycle. However signal movements were almost linear with concentration, implying K.sub.a was too small to measure.

EXAMPLE 7FURTHER STRUCTURAL STUDIES

[0264] The 3D structure of 3 and its complex with methyl -D-glucoside 10 was studied by 2D nOe spectroscopy (NOESY). The resultant structures are illustrated in FIGS. 10A and 10B. FIG. 10A shows a view of the complex roughly parallel to the tetralactam ring.

[0265] The anthracene units and the substrate are shown in space-filling mode, while the solubilising side-chains are removed for clarity. FIG. 10B shows a view roughly perpendicular to the tetralactam ring. The figure shows the shortest intermolecular distances (according to NOESY) D-4 and D-5, the longer D-6.sub.R distance, and the intramolecular D-E contacts. The four NH O hydrogen bonds appear as dotted lines.

[0266] In the case of 3 itself, a key issue to be investigated was the orientation of the annular amide groups, which in theory can be positioned such that either NH or CO point inward. Strong NOESY cross-peaks between NH protons D and spacer CH E (see FIG. 10B), and the absence of connections D-F, indicated that inward-directed NH groups are preferred. The data thus support the calculated structure shown in FIGS. 10A and 10B.

[0267] To study the complex 3.10, an excess of 10 was added such that 90% of 3 was in the bound state. Again, the intramolecular NOESY signals showed a strong preference for the NH-in arrangement. A large number of intermolecular cross-peaks were observed at long nOe mixing times, but at short mixing times the connections D-4 and D-5 stood out strongly, followed by D-6.sub.R. These data are best accommodated by structures in which the substrate CH.sub.2OH passes through the tetralactam ring so that H4 and H5 can come into contact with two diametrically opposite protons D. One such structure is shown in FIGS. 10A and 10B. This substrate positioning allows the formation of four intermolecular NH O bonds to four substrate oxygens, as well as 6 CH- contacts. Interestingly, the distance between the aromatic surfaces in this structure is smaller than previously determined for a biphenyl-based synthetic lectin [see Ferrand et al, Angew Chem Int Ed, 48: 1775-1779 (2009)]. This suggests a tight fit, which may contribute to the negative entropy of binding.

[0268] The conformation in FIGS. 10A and 10B can help to explain the selectivity of 3 for specific saccharides. An axial OH group, as in galactose or mannose, would clearly disrupt this structure, while the loss of CH.sub.2OH from the substrate (to give xylose) would remove both polar and apolar binding interactions. On the other hand, several of the better test substrates (glucuronides, cellobiose, maltose, lactose) do not seem compatible with this binding geometry, so other modes of interaction could be possible.

EXAMPLE 8FURTHER PROPERTIES OF COMPOUND 3

[0269] A number of further experiments were performed to test the potential of compound 3 for glucose monitoring in vivo.

[0270] Lactate and mannitol are carbohydrate-like molecules which can be present in blood, and which often bind to the boron-based receptors of the prior art. Neither produced any response when added to 3, thus confirming its selectivity for the desired target analyte. Binding to glucose was also studied at physiological temperature (310 K). The affinity measured by NMR titration was 33 M.sup.1, lower than at room temperature (as expected). However this is still potentially useful, implying receptor occupancy of 6-25% across the physiological range of 2-10 mM glucose.

[0271] Photobleaching of 3 was found to be relatively slow. Under continuous UV irradiation in a fluorescence spectrometer, emission decayed by <10% in 5 hours. In a practical glucose monitoring device this would translate to a long lifetime as the receptor compound would be subjected only to short pulses of light every few seconds.

EXAMPLE 9SYNTHESIS OF COMPOUND 13

[0272] The compound 13, shown in FIG. 12A, is another compound according to the first aspect of the invention, designed for the purpose of detecting glucose in blood. Due to the modified R.sup.9 and R.sup.10 groups attached to its isophthaloyl moieties, this compound benefits from enhanced aqueous solubility, rendering it particularly suitable for use in vivo for blood glucose monitoring.

[0273] A potential route to the synthesis of compound 13 is illustrated by the retrosynthetic scheme shown on FIGS. 13A-13C, and described in more detail below. This method, which accords with the eleventh aspect of the invention, can be seen in its latter stages to be analogous to the method proposed for the preparation of compound 3. However, it begins with the preparation and attachment, to an isophthaloyl precursor of formula (III), of the solubilising moiety which will represent the substituents R.sup.9 and R.sup.10 in the final product.

[0274] In FIGS. 13A-13C and the following description, the compound 13 appears as its sodium salt AnR-G2-Na; the isophthaloyl precursor of formula (III) as G2-Linker; the amine used to link the hydrophilic moiety to the isophthaloyl precursor as G2-NH.sub.2; and the corresponding nitro-substituted form of the amine as G2-NO.sub.2. FIGS. 14A to 14D show the structures for G2-NH.sub.2, G2-Linker and AnR-G2-Na, and also (FIG. 14C) for a t-butyl-protected form (AnR-G2-tBu) of the eventual sodium salt 13, in each case with the protons assigned.

[0275] Compound 13 was prepared using the route shown in FIGS. 13A-13C, as described below.

[0276] G2-NH.sub.2. To an autoclave (250 mL) the commercially available compound G2-NO.sub.2 (3.89 g, 2.65 mmol), Raney Ni (7.5 mL, water slurry) and ethanol (100 mL) were added.

[0277] The autoclave was then sealed, pressured with H.sub.2 (50 bar) and left stirring for 24 hours at 60 C. After cooling to room temperature the mixture was filtered through Celite and washed with DCM (50 mL), then the solvent was removed under reduced pressure to yield the product (3.46 g, 91%). .sup.1H-NMR (500 MHz, CDCl.sub.3): 1.42 (s, 81H, H1), 1.97 (t, 18H, H5), 2.15 (t, 6H, H10), 2.23 (m, 24H, H9/4), 2.38 (s, 2H, H12), 6.22 (s, 3H, H7). MS(ESI) calculated for C.sub.76H.sub.134O.sub.21N.sub.4H.sup.+=1439.50; found 1439.96.

[0278] G2-Linker.

[0279] To a stirred suspension of tri-PFP (1.64 g, 2.31 mmol) and G2-NH.sub.2 (1.70 g, 1.16 mmol) in a mixture of THF (10 ml, anhydrous, degassed) and CH.sub.2Cl.sub.2 (2 ml, anhydrous, degassed), DIPEA (1.81 ml, 10.4 mmol) was added. The reaction mixture was heated for two hours at 40 C., after which the clear solution was concentrated to dryness with a rotary evaporator. The resulting oil was purified via column chromatography (10:90 to 60:40 EA:HEX) to yield the product as a white solid (1.68 g, 74.0%). R.sub.f=0.34 (40:60 EA:HEX). .sup.1H-NMR (500 MHz, CDCl.sub.3): 1.31 (s, 81H, H1), 1.84 (t, 18H, H5), 2.08 (m, 24H, H4/9), 2.23 (t, 6H, H10), 8.96 (t, 1H, H17), 9.15 (d, 2H, H15), 9.47 (s, 1H, H12). .sup.19F-NMR (500 MHz, CDCl.sub.3): 152.42 (d, 4F, F20), 157.75 (t, 2F, F22), 162.30 (t, 4F, F21). MS(ESI) calculated for C.sub.97H.sub.136F.sub.10N.sub.4O.sub.26Na.sup.+=1987.21; found 1987.10.

[0280] AnR-G2-t-Bu.

[0281] A solution of G2-Linker (439 mg, 0.22 mmol) in THF (100 mL, anhydrous) was added dropwise over 36 hours (syringe pump) to a solution of 9,10-bis(aminomethyl)anthracene (52.8 mg, 0.22 mmol) and DIPEA (2 mL, 12 mmol) in THF (900 mL, anhydrous) under nitrogen. After stirring for a further 36 hours the solvent was removed under vacuum and the residue dissolved into chloroform (200 mL) and washed with NH.sub.4Cl (sat aq, 200 mL), water (200 mL) and brine (200 mL), then dried over MgSO.sub.4, filtered and evaporated in vacuo. The crude was dissolved in acetone/water (5:2, 6 mL) and filtered through a syringe filter (0.45 m). The solution was injected into a preparative reverse phase HPLC apparatus fitted with a reverse phase column (WatersXselect, 25019 mm, 5 m) and eluted with acetone/water (80:20 to 90:10 over 20 min; flow rate 19 mL/min). The component eluting at 19 min was collected, concentrated under vacuum and freeze-dried to yield AnR-G2-tBu (58 mg, 14%) as a pale yellow powder. R.sub.f=0.7 (70:30 EtoAc:Hexane). .sup.1H NMR (500 MHz, CDCl.sub.3): =1.42 (s, 162H, H24), 2.01 (t, J.sub.HH=7.3 Hz, 36H, H20), 2.22 (t, J.sub.HH=7.0 Hz, 12H, H15), 2.24 (t, J.sub.HH=7.3 Hz, 36H, H21), 2.32 (t, J.sub.HH=7.0 Hz, 12H, H16), 5.53 (d, J.sub.HH=4.9 Hz 8H, H5), 6.18 (s, 6H, H18), 6.60 (t, J.sub.HH=4.9 Hz, 4H, H6), 7.38 (t, J.sub.HH=1.3 Hz, 2H, H9), 7.45 (dd, J.sub.HH=6.9, 3.3 Hz 8H, H1), 8.32 (dd, J.sub.HH=6.9, 3.3 Hz 8H, H2), 8.73 (s, 4H, H10), 8.81 (s, 2H, H13). .sup.13C NMR (125 MHz, CDCl.sub.3): =28.22 C24, 29.99 C20, 30.03 C21, 31.92/31.98 C15/16, 37.49 C5, 57.88 C19, 58.56 C14, 80.78 C23, 124.90 C2/9, 126.44 C1, 129.85/130.27 C3/4, 130.35 C10, 165.55 C12, 166.21 C7, 172.87 C23, 173.22 C17. MS(ESI) calculated for C.sub.202H.sub.300O.sub.48N.sub.12Na.sub.2.sup.2+=1855.28; found 1855.14.

[0282] AnR-G2-Na.

[0283] AnR-G2-tBu (54.4 mg, 14.8 mol) was dissolved in DCM (6 mL) and cooled to 0 C. over ice. TFA (2 mL) was added dropwise over 5 minutes and the solution stirred under N.sub.2 for 16 hours at room temperature. The solvent was then removed under vacuum and the residue dissolved in H.sub.2O/MeOH (6:4, 10 mL), and NaOH (0.1 M) was added dropwise until pH 7. The solution was then filtered through a syringe filter (0.45 m) and the remaining solution freeze-dried to obtain AnR-G2-Na as a pale yellow powder (43.7 mg, 97%). .sup.1H NMR (500 MHz, CDCl.sub.3): =1.97 (t, J.sub.HH=7.4 Hz, 36H, H20), 2.20 (m, 48H, H15/21), 2.39 (t, J.sub.HH=7.5 Hz, 12H, H21), 5.48 (s, 8H, H5), 7.56 (dd, J.sub.HH=7.0, 2.6 Hz 8H, H1), 7.99 (s, 4H, H9), 8.29 (dd, J.sub.HH=7.0, 2.6 Hz 8H, H2), 8.53 (s, 4H, H10). .sup.13C NMR (125 MHz, D.sub.2O)): =30.32 C20, 30.41 C15, 30.83 C16, 31.11 C21, 37.21 C5, 58.17 C19, 58.91 C14, 124.49 C2, 127.25 C1, 127.25 C9, 128.58 C4, 129.74 C3, 130.14 C10, 133.75 C8, 135.95 C11, 168.03/168.22 C12/7, 175.07 C17, 182.12 C22.

EXAMPLE 10TESTING OF COMPOUND 13

[0284] Relevant properties of compound 13 were tested, in order to assess its suitability for use as a blood glucose sensor in vivo. The results are summarised in FIGS. 15A-15D.

[0285] Firstly with regard to its optical properties, FIG. 15A shows fluorescence titration curves for the compound at 18.8 M with glucose in phosphate buffer solution (pH 7.1, 0.1 M) at 298 K. The cell path length was 10 mm and the excitation wavelength 395 nm, and glucose concentrations from 0 to 171 mM were investigated. The inset shows binding data (423 nm) and fitting curve (Kaleidagraph), which gives K.sub.a=87 M. It can be seen from these data that compound 13 maintains its optical output from the anthracene core. Upon binding with D-glucose it exhibits an approximately three-fold increase in fluorescence emission (em 423 nm) when the system is excited (ex 395 nm).

[0286] FIG. 15B shows partial .sup.1H NMR spectra from the titration of compound 13 (0.125 mM) with D-glucose (/=36:64) in D.sub.2O at 298 K. The inset illustrates both experimental and calculated values for the NMR binding of proton E of compound 13 (see the structure at the top right of FIG. 15C) with D-glucose in D.sub.2O: these can be seen to be in good agreement, yielding a K.sub.a value of 89 M.sup.1.

[0287] FIG. 15C is a table of NMR- and fluorescence-derived binding constants for compound 13 with three different saccharides. The data demonstrate the compound's selectivity towards all-equatorial saccharides such as D-glucose (K.sub.a=89 and 87 M.sup.1), as compared to D-mannose and methyl--glucoside.

[0288] FIG. 15D shows partial .sup.1H NMR spectra for compound 13 at concentrations from 0.125 mM to 2 mM in D.sub.2O at 298 K, with assignments based on the structure at the top right of FIG. 15C. The compound exhibited good solubility in water, with no indication of aggregation during these NMR dilution studies.

[0289] Overall, these data indicate that compound 13 would be suitable for use as a glucose sensor in human blood serum. Its enhanced binding and selectivity for glucose over other sugars will make it sensitive to glucose levels even within the hypoglycemic range, and its fluorescence output can provide a detectable indication of saccharide binding in such contexts.

EXAMPLE 11SYNTHESIS OF COMPOUND 14

[0290] The compound 14, shown in FIG. 12B, is a yet further compound according to the first aspect of the invention, designed for detecting glucose in blood. Its modified solubilising groups R.sup.9 and R.sup.10 give it even greater aqueous solubility than compound 13. These groups not only make compound 14 highly suitable for use in the bloodstream, but also allow for greater flexibility in the design of modified versions of compound 14 carrying alternative substituents on the anthracene units and/or additional functional groups.

[0291] A potential route to the synthesis of compound 14 is illustrated by the retrosynthetic scheme of FIGS. 13A-13C, and described in more detail below. This method also accords with the eleventh aspect of the invention, and in its latter stages is analogous to the method proposed for the preparation of compound 3. It begins with the preparation and attachment, to an isophthaloyl precursor of formula (III), of the solubilising moiety which will represent the substituents R.sup.9 and R.sup.10 in the final product.

[0292] In FIGS. 13A-13C and the following description, the compound 14 appears as its sodium salt AnR-G3-Na; the isophthaloyl precursor of formula (III) as G3-Linker; the amine used to link the hydrophilic moiety to the isophthaloyl precursor as G3-NH.sub.2; and the corresponding nitro-substituted form of the amine as G3-NO.sub.2. FIGS. 14E to 141 show the structures for these compounds, and also for a t-butyl-protected form (AnR-G3-tBu) of the eventual salt 14, in each case with the protons assigned.

[0293] Compound 14 was prepared using the route shown in FIGS. 13A-13C, as described below.

[0294] G3-NO.sub.2.

[0295] G2-NH.sub.2 (6.31 mg, 4.4 mmol, prepared as in Example 9), NO.sub.2-triPFP (1.00 g, 1.30 mmol) and DIPEA (1 mL) were dissolved in THF (20 mL, anhydrous) under N.sub.2. The reaction was heated to 50 C. and left stirring over molecular sieves (4 A) for 48 hours. The solvent was removed and toluene added and evaporated three times to remove the DIPEA. The crude was purified via column chromatography (30:70 to 50:50 EA:HEX to 100:5 EA: MeOH) to yield G3-NO.sub.2 as a white solid (3.04 g, 52%). R.sub.f=0.65 (50:50 EA:HEX). .sup.1H NMR (500 MHz, CDCl.sub.3): =1.42 (, 243H, H1), 1.94 (m, 78H, H5/10/15), 2.16 (m, 78H, H4/9/14), 6.28 (s, 9H, H7), 7.00 (s, 3H, H12). MS(HiRes-ESI) calculated for C.sub.238H.sub.411O.sub.68N.sub.13Na.sub.3.sup.3+=1536.2938; found 1536.2926.

[0296] G3-NH.sub.2.

[0297] To an autoclave (250 mL) G3-NO.sub.2 (2.93 g, 0.64 mmol), Raney Ni (10 mL, water slurry) and ethanol (40 mL) were added. The autoclave was then sealed, pressured with H.sub.2 (50 bar) and left stirring for 24 hours at 60 C. The mixture was then filtered through celite, washed with DCM (50 mL) and the solvent removed under reduced pressure to yield G2-NH.sub.2 (2.90 g, 99%). R.sub.f=0.5 (EA). .sup.1H NMR (500 MHz, CDCl.sub.3): =1.40 (, 243H, H1), 1.92 (m, 78H, H5/10/15), 2.17 (m, 78H, H4/9/14), 6.41 (s, 9H, H7), 7.66 (s, 3H, H12). MS(HiRes-ESI) calculated for C.sub.238H.sub.414O.sub.66N.sub.13Na.sub.2.sup.3+=1518.9739; found 1518.9682.

[0298] G3-Linker.

[0299] G3-NH.sub.2 (0.50 g, 111 mol) and tri-PFP (0.54 g, 333 mol) were dissolved in THF (1 mL, anhydrous) under N.sub.2 over molecular sieves (4 ). DIPEA (1 mL, 10.4 mmol) was injected and the reaction heated to 40 C. and left stirring for 4 hours at room temperature under N.sub.2. The solvent was removed under vacuum and toluene (60 mL) added and removed three times on the rotary evaporator to remove the DIPEA. The crude was purified via column chromatography (40:60 to 60:40 to 100:0 EA:HEX) to yield G3-Linker as a white foam (280 mg, 50%). R.sub.f=0.5 (50:50, EA:HEX). .sup.19F NMR (470 MHz, CDCl.sub.3): =151.92 (d, J.sub.FF=19.1 Hz, 4F, F25), 157.32 (t, J.sub.FF=22.0 Hz, 2F, F27), 161.89 (t, J.sub.FF=20.2 Hz, 4F, F26). .sup.1H NMR (500 MHz, CDCl.sub.3): =1.41 (s, 243H, H1), 1.92 (m, 78H, H5/10/15), 2.16 (m, 78H, H4/9/14), 6.27 (s, 9H, H7), 6.88 (s, 3H, H12), 9.06 (s, 1H, H22), 9.14 (s, 2H, H20), 9.49 (s, 1H, H17). MS(HiRes-ESI) calculated for C.sub.259H.sub.415O.sub.71N.sub.13F.sub.10Na.sub.3.sup.3+=1700.9593; found 1700.9530.

[0300] AnR-G3-tBu.

[0301] 9,10-Bis-amino(methyl)anthracene (13.1 mg, 55.6 mol) was dissolved in THF (250 mL, anhydrous) and DIPEA (2 mL, 21.9 mmol). Next a solution of G3-Linker (280 mg, 55.6 mol) in THF (50 mL, anhydrous) was injected into the solution of amine over 36 hours with an automated syringe pump under N.sub.2 with stirring. After the addition the reaction was left for a further 36 hours. The solvent was removed under vacuum and the crude dissolved in DCM (50 mL) and washed with NH.sub.4Cl (sat aq, 50 mL), water (50 mL) and brine (50 mL), then dried over MgSO.sub.4, filtered and evaporated in vacuo. The crude was dissolved in acetone/water (85:15, 4 mL) and filtered through a syringe filter (0.45 m). The solution was injected into a preparative reverse phase HPLC apparatus fitted with a reverse phase column (WatersXselect, 25019 mm, 5 m) and eluted with acetone/water (85:15 to 100:0 over 30 min; flow rate 19 mL/min). The component eluting at 22 min was collected, concentrated under vacuum and freeze-dried to yield AnR-G3-tBu (130 mg, 48%) as a white powder. .sup.1H NMR (500 MHz, CDCl.sub.3): =1.42 (s, 243H, H30), 1.96 (t, J.sub.HH=8.2 Hz, 108H, H25), 2.05 (s, 54H, H20/15), 2.20 (t, J.sub.HH=8.2 Hz, 162H, H26/21/16), 5.58 (s, 8H, H5), 6.40 (s, 6H, H18), 6.51 (s, 18H, H23), 6.94 (s, 4H, H6), 7.44 (m, 10H, H1/8), 8.41 (m, 8H, H2), 8.67 (s, 4H, H10), 8.81 (s, 2H, H13). MS(ESI) calculated for C.sub.526H.sub.858N.sub.30O.sub.138Na.sub.3.sup.3+=3291.02; found 3292.90.

[0302] AnR-G3-Na.

[0303] AnR-G3-tBu (80 mg, 8.2 mol) was dissolved in DCM (6 mL) and cooled to 0 C. over ice. Next TFA (2 mL) was added dropwise over 5 minutes and the reaction left for 16 hours at room temperature under N.sub.2. The TFA was then removed under vacuum and the product dissolved in H.sub.2O:MeOH (6:4, 10 mL). Next NaOH (0.1 M) was added until pH 7 and the solution freeze-dried to yield the product (65 mg, 98%). .sup.1H NMR (500 MHz, CDCl.sub.3): =1.42 (s, 243H, H30), 1.96 (t, J.sub.HH=8.2 Hz, 108H, H25), 2.05 (s, 54H, H20/15), 2.20 (t, J.sub.HH=8.2 Hz, 162H, H26/21/16), 5.58 (s, 8H, H5), 6.40 (s, 6H, H18), 6.51 (s, 18H, H23), 6.94 (s, 4H, H6), 7.44 (m, 10H, H1/8), 8.41 (m, 8H, H2), 8.67 (s, 4H, H10), 8.81 (s, 2H, H13).

[0304] Compound 14 is expected to bind selectively to glucose in a similar manner to compounds 3 and 13, demonstrating selectivity over other saccharides and indeed over other species which are likely to be present in the bloodstream. It is also expected to generate a similar spectroscopic response, dependent upon glucose binding. It will be highly soluble in an aqueous environment such as human blood serum.

EXAMPLE 12IMMOBILISATION OF COMPOUND 3

[0305] The compound 3 was immobilised within a hydrogel support by the following method. The polymer used was a poly[acryloyl-bis(aminopropyl)polyethylene glycol] (PEGA), purchased from Sigma Aldrich in the form of beads with an average diameter of 300-500 m. The PEGA beads were stored in 90% MeOH with an amine functionality of 0.2 mmol/g.

[0306] Firstly, the sodium salt of compound 3 (AnR-G1-Na, obtained as in Example 1) was converted to the free acid form AnR-G1-H. AnR-G1-Na (12.1 mg, 7.59 mol) was dissolved in water (0.8 mL) and HCl (1 M) was added dropwise until pH 2. The precipitate was collected, washed with water (32 mL) and freeze-dried to yield AnR-G1-H as a yellow powder (10.9 mg, 99%).

[0307] Next, PEGA beads (2.9 mg, 0.58 mol (NH.sub.2)) were weighed into an eppendorf tube (1 mL) and centrifuged at 6000 rpm for 2 minutes. DMSO (250 L, anhydrous) was added and the mixture was centrifuged at 6000 rpm for 5 minutes and decanted three times. AnR-G1-H (2.96 mg, 2.03 mol) was dissolved in DMSO (400 L) and added to the beads under N.sub.2. NHS (N-hydroxysuccinimide) (1.40 mg, 12.2 mol) and EDCI (1-ethyl-3-(3-dimethylaminopropyl)carbodiimide) (2.33 mg, 12.18 mol) were dissolved in DMSO (200 L) and added, followed by DIPEA (3.03 L, 17.4 mol). The reaction was gently rocked for 16 hours, after which the tube was centrifuged at 6000 rpm for 2 minutes and the DMSO removed. DCM (0.5 mL) and water (0.5 mL) were added and shaken and the DCM layer removed three times. The beads were then washed with DMSO (21.5 mL) and MeOH (21.5 mL). NaOH (1.5 mL, 1 M) was added to the beads and they were shaken for 2 hours. The beads were then washed with water (31.5 mL) and PBS (phosphate buffered saline) (0.1 M, pH 7.4, 21.5 mL). This method was analogous to that disclosed by Shapiro et al in Measuring Binding of Protein to Gel-Bound Ligands Using Magnetic Levitation, J Am Chem Soc (2012), 134(12): 5637-5646.

[0308] Thus immobilised, the compound 3 is in a form suitable for introduction into a patient's body, to allow the in vivo monitoring of blood glucose levels. The hydrogel may for example form part of an implant for introduction into the bloodstream, or may be provided on a fibre optic probe.

EXAMPLE 13SHIFTING OF FLUORESCENCE

[0309] By altering the substituents on the anthracene moieties of compounds such as 3, 13 and 14, it is possible to alter their spectroscopic responses. This example demonstrates the tailoring of fluorescence emissions spectra in anthracene-containing precursor compounds which are usable to prepare compounds of formula (II) and in turn compounds of formula (I).

[0310] Four such precursor compounds were prepared and tested, in which the anthracene substituents R.sup.1 to R.sup.4 were all either (a) hydrogen, (b) OCH.sub.3, (c) CO.sub.2CH.sub.3 or (d) N-substituted cyclic imido, with the nitrogen atom being substituted with CH.sub.2CO.sub.2-t-butyl. In these compounds, the CH.sub.2NH.sub.2 groups of formula (II) were instead methyl groups.

[0311] FIGS. 17A and 17B shows schematically the methods by which the substituted precursor compounds were prepared. The tested products are labelled CMR 1 (R.sup.1-R.sup.4=OCH.sub.3); CMR 4 (R.sup.1-R.sup.4=CO.sub.2CH.sub.3); and CMR 6 (R.sup.1-R.sup.4=N-substituted cyclic imido).

[0312] In FIGS. 17A and 17B, BINAP is 2,2-bis(diphenylphosphino)-1,1-binaphthyl and HMDS is hexamethyldisilazane, also known as bis(trimethylsilyl)amine. The compound CMR 1 was prepared as described in the literature (Modjewski et al, Tetrahedron Lett (2009) 50: 6687-6690). The unsubstituted analogue bis-methylanthracene is commercially available.

[0313] The emissions spectra of the precursor compounds were recorded using a PerkinElmer LS45 fluorescence spectrometer, at wavelengths between 400 and 550 nm. The results are shown in FIG. 18. It can be seen that the addition of ester groups to the anthracene unit shifts its emissions peak towards the red (longer wavelength) end of the spectrum. Substitution with the cyclic imido groups shifts the peak yet further. The methoxyl substituents, in contrast, shift the emissions peak in the opposite direction, to a shorter wavelength. Similar trends can be expected in the emissions spectra of compounds of formula (I) derived from these precursors. Thus, the compounds can be tailored to provide a spectroscopic response in a desired region of the spectrum. An emissions peak at a longer wavelengthfor example about 550 nm or greateris expected to be of particular value for in vivo glucose detection systems.

CONCLUSIONS

[0314] It can be seen from the above that compounds 3, 13 and 14, and other analogous compounds according to the invention, are likely to provide an excellent starting point for a new approach to blood glucose monitoring. Their simplicity, accessibility and tunability can make them suitable for use in continuous glucose monitoring systems, in a practical and cost-effective manner.

[0315] In particular, an analogue of compound 3, or of compound 13 or 14, in which the substituents on the two anthracene units are chosen so as to increase the peak emissions wavelength of the compound (for example, by extending conjugation), could be of particular value for the in vivo monitoring of blood glucose levels. In the compounds 13 and 14 especially, which carry highly hydrophilic solubilising groups R.sup.9 and R.sup.10, it should be possible to modify the anthracene units in order to tune their spectroscopic responses, without undue detriment to the aqueous solubility of the overall compounds.

[0316] Explanation of FIG. 11

[0317] FIG. 11 shows schematically a detection and supply system according to the seventh aspect of the invention. This incorporates a device according to the fifth aspect and a detection system according to the sixth aspect, and may be used in a method according to the eighth, ninth or tenth aspect of the invention. In this case the system is for use in detecting the concentration of glucose in the bloodstream of a living patient, and for supplying insulin to the bloodstream when necessary: it can thus be used as or in an artificial pancreas.

[0318] The system comprises a detection device 50, which is introduced into the patient's bloodstream (denoted generally as 51). The device 50 may for example take the form of an implantable chip or capsule, or a probe such as a fibre optic probe. Carried in or on the device 50, in an appropriate physical form such as a hydrogel, is a compound according to the first or second aspect of the invention, or a polymer according to the third aspect. This compound or polymer exhibits a spectroscopic response in the presence of glucose, the nature and/or magnitude of the response being dependent on the glucose concentration in the bloodstream 51.

[0319] The system also comprises a detector 52, in the form of a small device which can be strapped to the patient's body at or close to the location of the implanted device 50. The detector 52 is capable of detecting the spectroscopic response of the compound or polymer to its environment. The detector 52 comprises interrogation means 53, by which it can apply electromagnetic radiation 54 at a wavelength suitable to excite the compound or polymer and to cause it to emit electromagnetic radiation 55 in response. The emitted wave 55 can be detected by the detector 52, which then sends a signal 56 to the control means 57. The signal 56 thus carries information regarding the concentration of glucose in the patient's bloodstream. Such information may be relayed from the detector 52 and/or the control means 57 to an output device 58, such as a screen, from which a user may obtain information about glucose concentrations and/or associated warnings. Information may also be output from the detector and/or the control means to another device or system such as is shown at 59.

[0320] The control means 57 incorporates a comparator means 60. This compares the signal 56 with pre-programmed information regarding safe blood glucose concentrations. If the control means detects a predetermined minimum difference between the signal 56 and the programmed safe concentrations, it sends a signal 61 to the delivery means 62, which is connected to a supply 64 of insulin. The signal 61 controls the delivery of insulin from the supply 64 back into the patient's bloodstream, via a pump 63 and appropriate intravenous conduits 65. In this way, the control means 57 can maintain the patient's blood glucose levels within safe ranges, supplying insulin when necessary in response to detected changes. Monitoring of the patient's blood glucose levels, and their maintenance within safe ranges, can be done continuously due to the presence of the device 50 in the bloodstream and the simplicity, and ready detectability, of the response of the detector compound (I) or (Ia) to changing glucose concentrations.

[0321] The control means 57 may comprise one or more of: a microprocessor or other data processing and/or operational control means; a data storage means such as a flash memory; and a connector or connection port for connecting to another device or system 59 (for example a computer) in order to transfer data between the two. Instead or in addition, conventional wireless communication and data transfer systems may be used to control operation of, and communicate with, the detection system remotely.