The present invention relates to 4,6-di-(0-thiophosphate)-inositol-1 ,2,3,5-tetra-0-sulfate and its use as an enteric activator of Clostridium difficile toxin in prevention and therapy of the toxic symptoms associated with C. difficile infection. Background of the invention

Infections with Clostridium difficile can lead to severe, even life-threatening diarrhoea. The symptoms are caused by two toxins synthesized by C. difficile, TcdA and TcdB. The toxins can enter the cells lining the colon, where they are activated by cytosolic inositol hexakisphosphate (IP6) and exert their toxic function. One suggested therapeutic intervention is to activate the toxins in the extracellular space of the colon lumen. The activated toxins are no longer able to enter the colon cells, and within the colon lumen they cannot exert their toxic function. Thus, activation of the toxins within the colon lumen renders the toxins harmless for the affected patient.

IP6 cannot be used for therapeutic intervention, because it precipitates due to the high calcium concentration in the colon lumen.

WO2013045107A1 shows PEG-modified inositol phosphate compounds and mixed inositol phosphate-sulfate compounds and their use in activation of C. difficile toxin activation. The activity of the compounds shown therein is promising, however improvements upon the results related therein would be of advantage. Thus, the problem underlying the present invention is to provide activators of C. difficile toxin that exhibit stronger activity at high calcium concentration.

This problem is solved by the subject matter of the independent claims. Description of the invention

During an investigation of the compound class first shown in WO2013045107A1 , it was surprisingly found that a small subset of mixed inositol tetrakissulfate bisthiophosphate compounds is far superior in effect to previously investigated mixed sulfate-phosphate compounds (see Fig. 1 ). According to a first aspect of the invention, a compound characterized by the general formula (I)

is provided, wherein two out of six X are OPS0 ₂ ^2" and the remaining X are OS0 ₃ ^" .

The straight lines in formula I are meant to indicate that the stereochemistry of the individual ring carbon atoms is undefined. The formula is meant to encompass any diastereomer.

In certain embodiments, the compound is characterized by the general formula (lla) or (lib)

OSO3- (lla) OSO3- (llb).

According to a second aspect of the invention, the compound as specified by formulae (I), (II), (lla) or (Mb) is provided for use as a medicament in the therapy or prevention of a disease.

According to a third aspect of the invention, the compound as specified by formulae (I), (lla) or (Mb) is provided for use in the therapy or prevention of C. difficile infection, or in the therapy or prevention of symptoms associated with C. difficile infection.

The compounds according to the invention do not need to penetrate mammalian or bacterial membranes to be active. In addition, the compounds according to the invention are unlikely to put selective pressure on the C. difficile bacteria, thus avoiding problems related to resistance.

According to a fourth aspect of the invention, a dosage form comprising the compound as specified by formulae (I), (lla) or (Mb) is provided, particularly for use in the therapy or prevention of C. difficile infection, or in the therapy or prevention of symptoms associated with C. difficile infection. In certain embodiments, the dosage form is a peroral formulation, particularly a tablet, capsule, lozenge, powder, solution or syrup.

According to an alternative aspect of the invention, the compound as specified in the above aspects of the invention is provided as a medicament, particularly a medicament formulated for use in the prevention or therapy of symptoms associated with C. difficile infection.

In certain embodiments, the medicament comprises the compound as specified in the above aspects of the invention alone or together with one or more pharmaceutically acceptable excipients or carriers.

The medicament may be administered alone or in combination with one or more therapeutic agents, particularly in combination with an antibacterial drug, more particularly in combination with an antibacterial drug selected from the group comprising (by way of non-limiting examples) metronidazole, vancomycin or fidaxomicin.

According to yet another aspect of the invention, a method of treatment or prevention of symptoms associated with C. difficile infection is provided, comprising the administration of the compound as specified by formulae (I), (Ma) or (Mb) to a subject in need thereof. Administration may be effected by any of the aforementioned means.

The compound may be given to a patient already diagnosed with C. difficile infection, or to a patient being suspected of suffering from C. difficile infection. Alternatively, the compound may be used as a prophylactic for patients that are at risk of contracting the infection, such as patients under treatment with antibacterial drugs in hospital settings.

According to yet another aspect of the invention, a method of treatment of C. difficile infection is provided, comprising the administration of an antibacterial drug, particularly metronidazole, vancomycin or fidaxomicin in combination with the compound as specified by formulae (I), (Ma) or (Mb) to a subject in need thereof. Short description of the figures

Fig. 1 shows the extent of cleavage of TcdB in the presence and absence of Ca ²⁺ (10 mM) for IP6, IP2S4 and activator compound IT2S4 (Ma).

Fig. 2 shows ¹ H-NMR and ³¹ P-NMR of compound (Ma).

Fig. 3 shows the extent of TcdB cleavage in presence of 10 mM CaCI ₂ for inositol hexaphosphate (IP6), inositol hexasulphate (IS6), and two mixed phosphate-sulfate compounds. Examples

Example 1 : Synthesis of compound (Ma)

The synthesis followed the sequence depicted in the scheme below:

PTSA: p-toluenesulfonic acid; DMF: dimethylformamide; TBDMSCI: tert-butyldimethylsilyl chloride; DCM: dichlormethane; S8: elemental sulphur; pyr.: pyridine; TMSBr: trimethylsilyl bromide, TFA: trifluoroacetic acid

Phosphorylation

The known 2-tertbutyldimethylsilyl inositol orthoformate was co-evaporated 3x with toluene and dissolved in dichlormethane (DCM). 1 H-tetrazole (4 eq.) followed by phosphoramidite (8 eq.) were added to the reaction and stirred overnight. Pyridine, followed by crushed sulphur flakes (20 eq.) were added to the reaction and stirred overnight. The resulting crude mixture was diluted with DCM and washed with saturated NaHC0 _3, dried with Na ₂ S0 ₄ , filtered and concentrated. The product was purified by flash chromatography with DCM in toluene.

¹ H-NMR (400 MHz; CDCIs): δ 7.35-7.29 (m, 4H), 7.15 (dd, J= 6.6, 2.1 Hz, 2H), 7.07-7.04 (m, 2H), 5.54 (d, J= 1 .1 Hz, 1 H), 5.45-5.41 (m, 2H), 5.30-4.97 (m, 8H), 4.51-4.49 (m, 1 H), 4.33- 4.32 (m, 2H), 4.27 (d, J= 1 .3 Hz, 1 H), 0.93 (s, 9H), 0.13 (s, 6H);

3 ¹ P-NMR (162 MHz; CDCIs): δ 70.1 Deprotection

The following deprotection conditions are in analogy to the synthesis published in the Journal of the American Chemical Society [JACS 2005, 127, 5288].

Starting material (50mg) was treated with thiophenol (300 μΙ), 77-cresol (300 μΙ), trifluoroacetic acid (1 .8 ml). Then added TMSBrOH slowly (360 μΙ). Stirred 2 h at room temperature. Evaporated twice from toluene. Diluted with DCM, and ca. 5 ml water. Neutralized with 1 N NaOH. Poured aqueous layer (slightly cloudy) directly on SolEx C18 cartridge (Thermofisher, 1 g, 6 ml). Eluted with water. In some cases some aromatic impurities were found but would precipitate over time in water and could be filtered-off.

1H-NMR (500 MHz; D2O): δ 4.36 (q, J = 9.6 Hz, 2H), 4.02 (t, J = 2.7 Hz, 1 H), 3.64 (dd, J = 9.7, 2.8 Hz, 2H), 3.50 (t, J= 9.3 Hz, 1 H).

3 ¹ P-NMR (203 MHz; D2O): δ 45.7

Sulfation

The sulfation reaction of the thiophosphate has to be performed carefully because the thiophosphate is eventually converted to the phosphate under the reaction conditions. We thus monitored the sulfation carefully and saw that the reaction was complete after ca. 30 min. and that no decomposition could be observed in this time. Thus, sulphurtrioxide dimethylformamide (SO ₃ -DMF) complex (12 eq.) was added to a suspension of inositol phosphate in DMF and the reaction was stirred 35 min. The reaction was quenched by adding 1 N NaOH, until ca. pH 8 followed by ca. 3 ml methanol (MeOH) to precipitate salts. The solid was purified by Sephadex LH-20 column, eluting with water.

¹ H-NMR (500 MHz; D ₂ 0): δ 5.06 (s, 1 H), 5.04-4.98 (m, 4H), 4.79-4.76 (m, 1 H).

³¹ P-NMR (203 MHz; D ₂ 0): δ 44.5

¹ H-NMR and ³¹ P-NMR results are shown in Fig. 2.

Example 2: Comparison of cleavage efficiency

IP6, activator compound (Ma) and IP2S4 were compared with regard to the extent of cleavage of TcdB (Fig. 1 ). The compound to be tested was added at 1 mM to 150 ng toxin B in presence or absence of 10 mM Ca ²⁺ in 100 mM Tris pH7.4 and incubated for 3 h at 37°C. Cleaved protein fragments were separated by SDS-PAGE and visualized by silver staining. The extent of cleavage was quantified from protein band intensities using the ImageJ software package. Signals were normalized to cleavage of positive and negative controls.

The results show that the di-thiophosphate-tetra-sulfate inositols are surprisingly superior even in comparison to di-phosphate-tetra-sulfate inositols, which in turn are significantly superior to the inositolhexasulfate and inositol hexaphosphate previously published (Fig. 3 and comparative example 3). Example 3 (comparative): P2S4 inositol, IP6 and IS6 cleavage efficiency

Samples were prepared and processed as described in example 2. Error bars show s.d.; Asterisk indicates statistical difference compared to IP2S4 (P < 0.05); n = 3. Methods:

Analogue solubility measurements by ICP-MS. 100 mM solutions of inositol

hexakisphosphate (IP6) analogues with or without 10 mM CaCI ₂ were prepared in 10 mM tris pH 7.4 and incubated with agitation for 2 h at 37 °C. The solutions were immediately filtered through 0.2 mm nylon filters equilibrated to 37 °C. The phosphorous content in each filtrate was determined by inductively coupled plasma-mass spectrometry (ICP-MS). The values obtained were divided by the number of phosphates in each IP6 analogue to determine the concentration of the compound in the solutions.

Free calcium ion quantification. A fresh 1 mM solution of murexide (Merck, Germany) was prepared in 10 mM tris pH 7.4. For each IP6 analogue, samples containing 0.5 mM analogue, murexide and CaCI ₂ in 50 mL 10 mM tris were prepared in triplicate. After 5 min incubation at room temperature, the samples were centrifuged at 20Ό00 g for 2 min and the upper 40 mL of the supernatant transferred to a 384-well plate. Samples without IP6 analogue containing CaCI ₂ ranging from 1 mM to 20 mM, and 20 mM were also prepared and used for calibrating each experiment. The absorbance was measured at 474 nm and 544 nm and the data analyzed as reported by Ohnishi.[/4na/. Biochem. 85, 165 (1978)] The experiment was repeated in triplicate.

Cleavage assays with holotoxin. 1 mM IP6 analogues were equilibrated with 10 mM CaCI ₂ in 100 mM tris pH 7.4 for 15 min at 37 °C before addition of 150 ng TcdB (TgcBiomics, Germany) in a total volume of 20 mL. A negative control (no IP6, 10 mM CaCI ₂ ) and a positive control (1 mM IP6) were also included on every gel. The reaction mixtures were incubated for 3 h at 37 °C and then placed on ice. Laemmli sample buffer (5X) was added to stop the reactions and 10 mM EDTA was added to the samples containing CaCI ₂ before heating at 95 °C for 3 minutes. The cleavage products were visualized by SDS-PAGE (using 15-well 8 % acrylamide Precise™ Tris-Glycine gels, ThermoScientific, USA) followed by silver staining according to a modified Vorum protocol [ Proteomics 1 , 1359 (2001 )] with the thiosulphate sensitization step extended to 10 min. The linearity of the staining protocol was verified with serial dilutions of TcdB starting at 160 ng/lane down to 20 ng/lane. The band intensities were quantified as described for the cleavage assays with recombinant toxin. The experiment was done in triplicate.