Modulation of metal-azolate frameworks for the tunable release of encapsulated glycosaminoglycans

Glycosaminoglycans (GAGs) are biomacromolecules necessary for the regulation of different biological functions. In medicine, GAGs are important commercial therapeutics widely used for the treatment of thrombosis, inflammation, osteoarthritis and wound healing. However, protocols for the encapsulation of GAGs in MOFs carriers are not yet available. Here, we successfully encapsulated GAG-based clinical drugs (heparin, hyaluronic acid, chondroitin sulfate, dermatan sulfate) and two new biotherapeutics in preclinical stage (GM-1111 and HepSYL proteoglycan) in three different pH-responsive metal-azolate frameworks (ZIF-8, ZIF-90, and MAF-7). The resultant GAG@MOF biocomposites present significant differences in terms of crystallinity, particle size, and spatial distribution of the cargo, which influences the drug-release kinetics upon applying an acidic stimulus. For a selected system, heparin@MOF, the released therapeutic retained its antithrombotic activity while the MOF shell effectively protects the drug from heparin lyase. By using different MOF shells, the present approach enables the preparation of GAG-based biocomposites with tunable properties such as encapsulation efficiency, protection and release.

process. This improves the determination of the amount of GAG in the solution by reducing the interference of the degraded MOF components and the buffer media with the carbazole assay.

Carbazole assay
Ammonium sulfamate (20 µL, 4 M) was added to an aliquot of sample (200 µL) or water (blank control) and the resultant mixture was vortexed for 1 min. Then, sodium tetraborate in sulfuric acid (1 mL, 25 mM) was added and carefully mixed. The mixture was heated at 100 ºC for 5 min and cooled to room temperature. Afterwards, the carbazole solution (0.1%, 40 µL) was added and the resultant mixture was heated again at 100 ºC for 15 min and then cooled down to room temperature (color develops during this step). Finally, the resultant solution was analyzed by UV-vis spectroscopy; the absorbance at 520 nm was used to quantify the amount of the analyte by comparison with the corresponding calibration curve (Fig. S23). All the experiments were performed in triplicate.

Release test of GAG@MOFs
1 mL of citrate buffer (80 mM, pH = 6) was added to a pellet of GAGs@MOF, the sample was kept under bidimensional stirring.
Aliquots of 200 µL of the supernatant were collected by centrifugation (13400 rpm, 1 min) and replaced with the same volume of fresh medium. The amount of GAG released in the incubation media was the determined by UV-vis spectroscopy using carbazole assay.
Preliminary studies for determining the synthetic conditions to optimize the encapsulation efficiency (EE%)

Determination of the encapsulation efficiency of FITC-CMD@ZIF-8, FITC-CMD@ZIF-90, and FITC-CMD@MAF-7 biocomposites
The quantitative assessment of the cargo loaded within the biocomposites was carried out by re-dissolving the MOF matrix under acidic conditions, soaking the samples in 2 mL of citrate buffer (100 mM, pH = 6). The resultant solution was analyzed by UV-vis spectroscopy, where the absorbance at 490 nm was used to quantify the amount of the analyte by comparison with the corresponding calibration curve (Fig. S1-S3). To avoid any interference of the ligand, metal, and/or the citrate buffer during the determination process, the calibration curves were performed adding a known amount of FITC-CM-dextran to a solution of Zn 2+ and the corresponding ligand (HICA or Hmtz) in citrate buffer media. The amount of Zn 2+ and ligand added to this mixture depends of the amount of material formed for each metal to ligand ratio. These experiments were performed in triplicate for each sample described in the Tables S1 and S2.   Results reveal that the FITC-CMD@ZIF-8 biocomposites obtained from 0.36 (3) and 0.72 (4) mg mL -1 of FITC-CMD present higher EE% than those obtained from 0.18 (2) and 1.44 (5) mg mL -1 of FITC-CMD (Fig. S1). In addition, the EE% is influenced by the metal to ligand ratio: the optimal encapsulation efficiency is reached when using 0.36 (3) and 0.72 (4) mg mL -1 of FITC-CMD and 1:4 and 1:3.47 metal to ligand ratio (8DXA3, 8DXA4, 8DXB3, and 8DXB4). for FITC-CMD@ZIF-90 biocomposites (90DXAn; where n = 2 -5), the EE% decreases drastically with the increase in the initial concentration of FITC-CMD from ca. 84% for 90DXA2 to ca. 34% for 90DXA5 (Fig. S2). Such results suggest that the EE% strongly depends on the initial concentration of the model drug.
FITC-CMD@MAF-7 biocomposites obtained from 1:4 metal to ligand ratio (7DXAn; where n = 2 -5) present exceptional polysaccharide payloads regardless of the initial concentration of FITC-CMD. For instance, the EE% is almost quantitative for the samples prepared with low initial concentrations of FITC-CMD (EE > 90% for 7DXA2, 7DXA3, and 7DXA4). The EE% decreases slightly as the concentration of FITC-CMD increases (ca. 85% for 7DXA5) (Fig. S3). These findings are consistent with the previous reports about the biomineralization of carbohydrates within ZIF-8. [1] However, such reports also declare that Zn 2+ :L ratio affects the polysaccharide payloads. Thus, in concordance with these studies, another two different metal to ligand ratios were tested (Zn 2+ :L = 1:3.47 (B) and 1:2.52 (C)) to corroborate the role of this parameter in the EE of the resultant FITC-CMD@MAF-7 and FITC-CM-dextran@ZIF-90 biocomposites (7DXBn, 7DXCn, 90DXBn, and 90DXCn; where n = 2 -5) (Table S1 and

Release test of FITC-CMD@ZIF-8 FITC-CMD@MAF-7 and FITC-CMD@ZIF-90
The drug release performance of the resultant MAF-7 and ZIF-90-based biocomposites (8DXAn, 8DXBn, 8DXCn, 7DXAn, 7DXBn, 7DXCn and 90DXAn, 90DXBn, and 90DXCn; where n = 2 -5) was assessed by monitoring the amount of FITC-CM-dextran released over time upon applying an external acidic stimulus (Fig. S4-S7). Thus, the powder material was soaked in 1 mL of citrate buffer (100 mM, pH = 6) under bidimensional continuous stirring (500 rpm). At different incubation times the sample was centrifuged, and 1 mL of the supernatant was taken to be analyzed by UV-vis spectroscopy (max 490 nm). It is worth to mention that the samples prepared with MAF-7 keeping Zn 2+ :L = 1:2.52 ratio degrades almost immediately upon the addition of citrate buffer.    In general, the release profiles of ZIF-8, ZIF-90 and MAF-7 biocomposites present an initial burst release, followed by a sustained delivery process (Fig. S4-S7).
After a close inspection of the release profiles obtained from different metal to ligand ratio, it is evident that the delivery rate increases as the Zn 2+ :L decreases. For instance, a comparative analysis of the release kinetics obtained from 90DXA3, 90DXB3, and 90DXC3 (Zn 2+ :L = 1:4, 1:3.47, and 1:2.52; respectively) reveals that 90DXA3 required around 50 min to achieve the full release of the model drug; whereas its analogous 90DXB3 and 90DXC3 achieved the complete release of the cargo within 25 and 15 min, respectively (Fig. S5). Similarly, for MAF-7-based biocomposites, it was observed that the samples obtained from Zn 2+ :L= 1:2.52 (7DXCn) degraded immediately upon soaking them into the acidic media. By contrast, for those prepared from 1:4 (A) and 1:3.47 (B) Zn 2+ :L ratio took from 15 min (7DXA5 and 7DXB5) to 30 min (7DXA2, 7DXA3, 7DXB2 and 7DXB3) to achieve the full release of the cargo (Fig. S6).
In light of such findings, we conclude that the optimal synthetic conditions to ensure acceptable encapsulation efficiencies and drug release kinetics, for all the three different metal-azolate systems, requires the usage of 0.36 mg mL -1 of the biomacromolecule keeping the ratio Zn 2+ :L = 1:3.47.

Determination of the anticoagulant activity of HP@MOFs
The anticoagulant activity of heparin was assessed by a chromogenic method for anti-IIa assay using a commercial kit (Iduron ANTI-IIA HEPARIN KIT). This assay is a two-step chromogenic method based on the inhibition of an excess of factor IIa in presence of antithrombin (AT).

Biopreservation experiments
The heparinase I (0.1 IU) was purchased from Iduron, and was supplied as frozen solution. A preliminary experiment was performed in order to determine the time required to complete degradation of the HP used in this work. This kinetic experiment was performed by UV spectroscopy following the formation of uronic acid (Amax = 232 nm) produced during the enzymatic degradation of the HP (Fig. S26c)

Gas Sorption
Gas adsorption isotherm measurements were performed on an ASAP 2020 Surface Area and Pore Size Analyzer. Samples were activated by heating in vacuum at 100 ºC for 3 h under N2 and 100 ºC for 18 h under vacuum (2X10 -6 mm of Hg). UHP grade (99.999%) N 2 and He were used for all measurements. The temperatures were maintained at 77 K (liquid nitrogen bath).

Thermogravimetric analysis (TGA)
TGA data confirmed the estimation made via carbazole assay (Fig. S28). However, the TGA data suggest that GAGs@MOF biocomposites possess a higher amount of zinc cations compared to the pure MOF materials (Fig. S28).
This observation may be attributed to the electrostatic interactions between the Zn 2+ and the negatively charged sulphate and carboxylate groups of the GAGs as reported by Parrish et al. 2 To confirm this hypothesis, we used energy dispersive X-ray spectroscopy (EDS) to determine the elemental composition of the HP@MOF biocomposites and related control samples (i.e. neat MOFs, and free HP) (Fig. S29, ESI †). In all the samples, as shown in plot S29 we observed an excess of Zn associated to the S assigned to sulphate groups. (Fig. S29 and Table S4-S7, ESI †). This excess of Zn validates the hypothesis and explain the TGA results.

Estimation of the USP units of heparin released form the maximum safe dosage of HP@MOF
Herein, we provide an estimation of the USP units of heparin released form the maximum safe dosage of HP@MOF biocomposites. Based on the cytotoxicity of ZIF-8 (30 mg L -1 ), 3 we can estimate the threshold concentration for ZIF-90 (32 mg L -1 ), and MAF-7 (30 mg L -1 ) to ensure the biocompatibility of those materials (IC20) (see Table S8 and Fig S30). Considering the amount of HP encapsulated in 100 mg of biocomposite, we calculated the USP units of HP released from the maximum dose of HP@MOF (Table S8, Fig. S30). According to the current dose regulations for heparin, 4 MAF-7 seems to be a suitable carrier for intravenous injection of HP. The particle size of this biocomposite is compatible with this administration route. 5 Sample %HP Loading Capacity    Table S9. Comparative overview of the properties reported for other drug delivery systems designed for heparin release. MS= mesoporous silica, CL=cargo loading, PEDOT = poly(3,4-ethylenedioxythiophene), PLGA = poly(lactic-co-glycolic acid), PMMA-b-PMAETMA = poly(methyl methacrylate-b-trimethyl aminoethyl methacrylate), PHB = polyhydroxybutyrate-co-hydroxyvalerate, PEG = polyethylene glycol, PCL = poly(ecaprolactone), TCP = tricalcium phosphate, PLLA = poly(L-lactic acid). When needed, we used the efficacy of our heparin (180 USP/mg) for the calculation of the lading capacity. *from Aldrich, our Heparin can be treated at 120 C