Inhibition of vascular calcification by inositol phosphates derivatized with ethylene glycol oligomers

Myo-inositol hexakisphosphate (IP6) is a natural product known to inhibit vascular calcification (VC), but with limited potency and low plasma exposure following bolus administration. Here we report the design of a series of inositol phosphate analogs as crystallization inhibitors, among which 4,6-di-O-(methoxy-diethyleneglycol)-myo-inositol-1,2,3,5-tetrakis(phosphate), (OEG2)2-IP4, displays increased in vitro activity, as well as more favorable pharmacokinetic and safety profiles than IP6 after subcutaneous injection. (OEG2)2-IP4 potently stabilizes calciprotein particle (CPP) growth, consistently demonstrates low micromolar activity in different in vitro models of VC (i.e., human serum, primary cell cultures, and tissue explants), and largely abolishes the development of VC in rodent models, while not causing toxicity related to serum calcium chelation. The data suggest a mechanism of action independent of the etiology of VC, whereby (OEG2)2-IP4 disrupts the nucleation and growth of pathological calcification.


Supplementary Figures Supplementary
: a, Schematic illustration of the assay principle. Human serum was spiked with 6 mM phosphate and 10 mM calcium. CPP structural rearrangement from CPP1 to CPP2, indicated by shift in absorbance (550 nm), was monitored over time at 37 °C. Addition of calcification inhibitors delays CPP transition. b, Delay in serum calcification propensity by lower phosphorylated inositol phosphate control compounds (n = 4 for myo-inositol, n = 3 for inositol phosphates). All data points represent the T 50 (min) as mean ± s.d. from n independent experiments and the dotted horizontal line represents the c350, (i.e., the concentration necessary to delay T 50 to 350 min), which was used to rank the inhibitor activity of all tested compounds.
Supplementary Fig. 2. Delay in serum calcification propensity by functionalized inositol phosphates (n = 5 for OEG 12 -IP5 and OEG 11 -IP2S3, n = 6 for (OEG 2 ) 2 -IP4, n = 26 for IP6, and n = 3 for all others). All data points represent the T 50 (min) as mean ± s.d. from n independent experiments and the dotted horizontal line represents the c350, (i.e., the concentration necessary to delay T 50 to 350 min), which was used to rank the inhibitor activity of all tested compounds.

Supplementary Fig. 3
Delay in serum calcification propensity by control compounds (n = 3 for magnesium citrate, PEG-alendronate, IT6 and PP i , n = 4 for etidronate and ITPP, and n = 6 for alendronate). All data points represent the T 50 (min) as mean ± s.d. from n independent experiments and the dotted horizontal line represents the c350, (i.e., the concentration necessary to delay T 50 to 350 min), which was used to rank the inhibitor activity of all tested compounds.

Supplementary Fig. 4: CPP hydrodynamic diameter (D h ) maturation and polydispersity index (PDI)
analyzed by dynamic light scattering. a, Serum background particle size did not change over time and were polydisperse (D h = 66 ± 5, 68 ± 2 and 64 ± 1 nm after 1, 3 and 5 h, respectively; PDI range 0.356 -0.405). b, Compounds at 100 µM did not alter D h of serum background particles. c, CPP formation was initiated by addition of 6 mM phosphate and 10 mM calcium. CPP1 present at 1 h (D h = 137 ± 14 nm; PDI 0.217 -0.295) and CPP2 present at 5 h (D h = 979 ± 170 nm; PDI 0.355 -0.409). d,-f, Human serum was mixed with increasing concentrations of compounds and subsequently spiked with calcium and phosphate to initiate CPP maturation (pre-setup). g,-i, Calcium and phosphate was added to human serum to initiate CPP maturation and subsequently, without further incubation time, increasing concentrations of test compounds were added (post-setup). Interestingly, CPP crystal growth retardation was stronger when (OEG 2 ) 2 -IP4 was added after formation of CPP1, compared to when the compound was added to serum before CPPs were formed. Data are expressed as mean ± s.d. (n = 3).

Supplementary Fig. 5:
Measurement of cell-free particle growth and formation of precipitate in nGM (normal growth medium), CaPM (normal growth medium spiked with calcium and phosphate to final concentrations of 2.7 mM calcium and 2.5 mM phosphate) and CaPM + (OEG 2 ) 2 -IP4.  Statistical difference was derived from Kruskal-Wallis test followed by non-parametric Mann-Whitney test with * p < 0.05 and *** p < 0.001 vs. vehicle.
The crude product was purified by silica gel column chromatography using a gradient of 0-7% Compound 11. Compound 10 (40 mg, 0.0346 mmol) was dissolved in DCM (0.2 mL) and TFA (2 mL) was dissolved in water (0.2 mL). The TFA solution was slowly added to the reaction mixture and allowed to react for 1 h at RT. The mixture was co-evaporated with toluene under vacuum and purified by silica gel chromatography using a gradient of 0-10% MeOH/DCM, yielding 32 mg (81%). DMR microscope, 200x magnification). In order to perform 3D analysis of mineralized bone structure, ex vivo µCT analyses of the methylmethacrylate embedded left tibia was performed using a highresolution µCT scanner (Skyscan 1076, SkyScan, Belgium) also at the proximal metaphysis. The Xray source was operated at 80 kV and 110 µA and a 25µm titanium filter was used. Images were captured using 10 megapixel CCD-camera.
Calibration of BMD was performed by scanning appropriate phantoms using identical X-ray settings.
In order to be able to compare the bone analysis results with that of control (non-CKD) rats, bone of control rats (n = 5) of the same age and genetic background as used for the current study was also analyzed.