Enhanced Antibacterial Activity of Vancomycin Loaded on Functionalized Polyketones

1. Introduction

Today, the fight against infectious diseases still poses a serious challenge worldwide for healthcare. According to the literature, bacteria are becoming quickly resistant to commercially available antibiotics and antiseptics [1,2,3,4,5,6,7]. Antibiotics are often used in massive doses to maintain their therapeutic effects because of their low bioavailability and unfavorable interactions with biological barriers promoting bacterial resistance. Moreover, resistant bacteria develop different mechanisms such as efflux pumps [8,9,10], inactivation of enzymes expression [10,11,12], modification of drug penetration through barriers [10,13], further reducing antibiotic efficiency. Additionally, bacteria may promote biofilm formation, which is a protective self-produced extracellular polymeric matrix, contributing to their difficult eradication [14,15,16]. Therefore, a deep comprehension of resistance mechanisms and the development of alternative antimicrobials are necessary to address the problem of bacteria resistance [1,17,18,19,20,21,22].

Drug Delivery Systems (DDS) are widely used to prevent antibiotics and antiseptics resistance, improving drugs efficiency, selectivity, biodistribution and pharmacokinetics, and preventing biofilm formation [6,23,24,25,26,27,28,29,30,31,32]. Biocompatible materials, such as polymers, polysaccharides, proteins, lipids, and peptides are widely used to produce DDS, avoiding drug degradation and adverse effects, controlling drug release [6,33,34,35,36,37,38].

Scheme 1. Synthetic scheme for the preparation of post-functionalized PK with a) taurinate (PKT) and b) sulfanilate (PKSK).

Scheme 1. Synthetic scheme for the preparation of post-functionalized PK with a) taurinate (PKT) and b) sulfanilate (PKSK).

Specifically devised polymers are also a promising alternative for biomedical and biotechnological applications, synthesized with different functional groups, promoting interactions with drugs [34,35,39,40,41,42,43,44,45,46]. DDS based materials can be used as carriers of therapeutic molecules (nanoparticles, microparticles, implants) or as conjugates, bonding drug molecules through linkers designed to release them specifically on targeted sites [10,34,39,47,48,49,50,51]. To date, DDS prepared with biocompatible polymers have been widely described in the literature and many are also commercially available [52,53,54].

Within this panorama, aliphatic PK appear as an interesting yet scantly explored solution as polymeric carriers for DDS production [55,56]. In fact, aliphatic PK, produced by catalytic copolymerization of CO and ethylene, may be easily post-functionalized by means of the so-called Paal-Knorr reaction [57], allowing to obtain customized PK with specific desired functional groups [58,59,60,61,62,63,64,65,66,67,68,69,70,71].

Even though few works have been published in the literature on the preparation of post-functionalized PK, data reported in the literature demonstrate their high potential as supports for different types of enzymes [72], germicide polymeric coatings[73] or as scaffolds for urothelial cells [56].

Vancomycin (VCM), a branched tricyclic glycosylated peptide antibiotic (Figure 1), is a very interesting last line defense antibiotic used against serious infections caused by Staphylococci, Enterococci, and other Gram-positive pathogenic bacteria [74,75,76,77,78,79,80,81,82,83], inhibiting cell wall synthesis in sensitive bacteria. Because of its wide use, to limit the rise of bacterial resistance to VCM, development of DDS is fundamental.

Physical chemical studies have demonstrated that VCM polymeric formulations are able to penetrate the bacterial phospholipidic double membrane by releasing the drug locally, inhibiting biofilm formation and eradicating mature biofilms [14,38,84,85,86,87,88,89]. In this context, the aim of this work is to prepare polymeric DDS employing specifically customized PK able to interact with VCM through ionic and hydrophobic bonds [38,90,91]. PK was prepared according to literature procedure [68], post functionalized with sulfanilate (PKSK) or taurinate (PKT) and loaded with Vancomycin (VCM) at different pH. All polymers were characterized by FT-IR, DSC, TGA, SEM, elemental analysis, while NMR analysis was not performed due to the insolubility of PKs, even after functionalization. PKSK and PKT before and after VCM loading were further tested against Staphylococcus aureus ATCC 25923 strain and their antimicrobial activity compared to pure VCM.

2. Materials and Methods

Carbon monoxide and ethene were supplied by SIAD Company Italy (“research grade”, purity > 99.9%). Bis(diphenylphosphino)propane (Dppp), TsOH (p-toluene sulfonic acid), Methanol (99.8%), Diethyl ether (>99.5 % ), Sulfanilic acid (≥ 98%), Taurine (≥99 %), Sodium hydroxide (≥98 % ), Potassium hydroxide (≥85 %), m-Cresol (99 %), Dimethyl sulfoxide (99%), Agar, Phosphate-buffered saline (PBS), Nutrient broth (NB), Mueller Hinton Broth (MHB) and Vancomycin hydrochloride were purchased from Sigma Aldrich Co. (St. Louis, MO). MilliQ water was used for all experiments. The complexes [Pd(OAc)₂(dppp)] were prepared as reported in literature [92]. Staphylococcus aureus ATCC 25923 strain was obtained from the American Type Culture Collection. 96-Well Microtiter Microplates was purchased from Thermo Fisher.

The average viscosity molecular weight of PK has been evaluated as Limit Viscosity Number (LVN), used m-cresol as solvent and the viscosity was measured by a Cannon-Fenske type capillary viscosimeter, thermostated at 25°C, as reported in literature [93].

FT-IR spectroscopy was measures on a Nicolet AVATAR Nexus FT-IR Thermo Fisher Scientific spectrometer in KBr, with 32 scans and a resolution of 4 cm^-1 in the wave number range between 4000 cm^-1 and 400 cm^-1. The melting and decomposition temperature and thermogravimetric analysis of polymers was determined using a DSC/TGA Linseis PTA ST1000 under nitrogen flow of 100 mL/ min with a temperature ramp of 20 °C min^-1 from 30 °C to 500 °C. Elemental analysis for carbon, hydrogen, nitrogen, oxygen, and sulphur (CHNSO) were performed using UNICUBE organic elemental analyzer (Elemental). It offers highly sensitive analysis at a detection limit of 10 µg/g or 10 ppm. UV-Vis spectra were recorded on UV Carey 100 (Agilent) spectrophotometer in a grass cuvette in the range 400 – 200 cm^-1. Scanning electron microscopy (SEM) was carried out with a Zeiss Sigma|VP (Jena, Germany), variable pressure instrument (VP-SEM), at 20 kV and variable magnifications. For SEM imaging, the samples were prepared by Au deposition (layer about 40 nm) using AC sputtering.

Optical density measurement (OD600) was conducted by BioTek Synergy H1 a modular multimode microplate reader, with monochromator-based optics and filter-based optics. The optical density at 600 nm is referred to as OD600. The light at λ=600 nm is easy to produce and does little to damage or hinder microbial growth.

2.2. Synthesis of Aliphatic Polyketones (PK)

The reaction was conducted in a 250 mL stainless steel autoclave equipped using a 150 mL Pyrex glass beaker. In typical experiment, 2.10 mg of [Pd(OAc)₂(dppp)] (0.003 mmol), 246.23 mg of p-TsOH (1.29 mmol) and 60 mL of methanol were mixed in a 150 mL Pyrex glass beaker, and then placed in the autoclave. The autoclave was then sealed and pressurized with a 1/1 mixture of CO/C₂H₄ at 5.0 MPa and heated to 90 °C for 4 hours over stirring. After cooling to room temperature, a solid product was collected, filtered, and washed three times with methanol, acetone and diethyl ether. The resulting white powder polyketone (PK) was dried in vacuum for approximately 8 hours [92,93,94,95,96]. M.p. 250 °C; FT-IR (KBr, cm^-1): ν˜= 1690 (vs) (C=O), 2900 (s) (C-H), 1406 (s) (C-H), 1333 (s) (C-H), 1057 (s) (C-H); Elemental analysis (%) for PK found: C 63.45, H 7.06, 0 29.49.

2.3. Synthesis of PKSK and PKT by Paal-Knorr Modification of PK with Potassium Sulfanilate Or Sodium Taurinate

In a round-bottom flask equipped with a mechanical stirrer and a reflux condenser, 1.0 g of polyketone (PK) was suspended in 30 mL of methanol with 1.88g potassium sulfanilate (0.09 mol) or 1.3g sodium taurinate (0.09 mol). The reaction mixture was heated at 70 °C for 48h, and after cooling was collected by filtration, washed with methanol, acetone and diethyl ether. The resulting brown powder polyketone with potassium sulfanilate (PKSK) and orange powder polyketone with sodium taurinate (PKT) were dried in a vacuum for 8 hours and characterized by FT-IR, DSC, TGA, Elemental analysis, SEM and OD600[96,97,98,99,100].

PKSK: M.p. 234⁰C; FT-IR (KBr, cm^-1): ν˜= 685 (s), 813 (w), 832 (w), 1010 (s), 1035 (s), 1057 (w), 1124 (s), 1164 (s), 1246 (s), 1259 (w), 1335 (w), 1408 (s), 1426 (w), 1499 (s), 1693 (vs), 2852 (w, br), 2912 (w, br), 2963 (w), 3437 (vs, br), (see Table 1S in supporting information); Elemental analysis (%) for PKSK found: C 51.18, H 5.38, N 4.02, S 9.31.

PKT: M.p. 214⁰C; FT-IR (KBr, cm^-1): ν˜=812 (w),1054 (s), 1198 (m, br), 1259 (w), 1336 (s), 1409 (s), 1427 (w), 1470 (w), 1693 (vs), 2853 (w), 2913 (w), 2953 (w), 3439 (vs, br), (see Table 2S in supporting information); Elemental analysis (%) for PKT found: C 54.76, H 6.5, N 1.66, S 4.23.

2.4. Loading of Vancomycin on PKSK or PKT

Loading of VCM on functionalized polyketones (PKSK or PKT) following the same protocol and the sensitiveness to pH was evaluated in DMSO solutions, in the 2.3-5.0-8.8 pH range (obtained by HCl or NaOH solutions respectively) (pH 8+ DSA potentiometer, favs Scientific Equipment) at room temperature. Polymer and Vancomycin 2/1 wt/wt ratio were suspended in 2mL DMSO (pH 2.3,5.0,8.8) under stirring. After 20 min, the solution was dialysed against MilliQ water with Spectra/Por 1 Dialysis Membrane Standard RC Tubing MWCO: 6 – 8 kD for 24 h to remove the remained unabsorbed antibiotic. The concentration of unabsorbed antibiotic (Free-VCM) resulted in the MilliQ water was measured using UV at a wavelength of 281 nm [101,102,103]. VCM encapsulation efficiency % (EE%) and loading capacity % (LC%) were determined as follow:

$$EE\%=\frac{TotalVCM-FreeVCM}{TotalVCM}x100$$

$$LC\%=\frac{TotalVCM-FreeVCM}{Polymer+(TotalVCM-FreeVCM)}x100$$

Total VCM = the starting VCM amount used for the loading.

Polymer = the starting PKSK or PKT amount used for the loading considering no losses.

Samples were lyophilized and characterized by FT-IR, DSC, TGA, SEM, OD600.

PKSK-VCM pH 2.3: M.p. 234 °C; FT-IR (KBr, cm^-1): ν˜= 806 (w), 1015 (w), 1035 (w), 1057 (m), 1124 (w), 1195 (w, br), 1260 (w), 1335 (m), 1408 (s), 1426 (w), 1499 (w), 1693 (vs), 2852 (w), 2912 (w), 2963 (w), 3437 (s, br) (see Table 1S in supporting information); Elemental analysis (%) for PKSK-VCM pH 2.3 found: C 50.87, H 4.96, N 4.36, S 7.67.

PKSK-VCM pH 5.0: M.p. 234 °C; FT-IR (KBr, cm^-1): ν˜= 806 (w), 1015 (w), 1035 (w), 1057 (m), 1124 (m), 1195 (m, br), 1260 (w), 1335 (m), 1384 (w), 1408 (s), 1426 (w), 1499 (w), 1693 (vs), 2852 (w), 2912 (w), 2963 (w), 3437 (s, br), (see Table 1S in supporting information); Elemental analysis (%) for PKSK-VCM pH 5.0 found: C 47.38, H 3.87, N 7.13, S 2.37.

PKSK-VCM pH 8.8: M.p. 234 °C; FT-IR (KBr, cm^-1): ν˜= 806 (m), 1015 (w), 1035 (w), 1057 (m), 1124 (w), 1195 (w, br), 1260 (m), 1335 (m), 1384 (s), 1408 (s), 1426 (w), 1499 (w), 1693 (vs), 2852 (w), 2912 (w), 2963 (w), 3437 (vs, br), (see Table 1S in supporting information); Elemental analysis (%) for PKSK-VCM pH 8.8 found: C 48.76, H 4.28, N 5.86, S 5.10.

PKT-VCM pH 2.3: M.p. 214 °C; FT-IR (KBr, cm^-1): ν˜=812 (m), 1023 (w), 1054 (s), 1095 (w, br), 1198 (w, br), 1259 (w), 1336 (m), 1384 (w), 1409 (s), 1427 (w), 1470 (w), 1693 (vs), 2852 (w), 2918 (w), 2962 (w), 3439 (vs, br) (see Table 2S in supporting information); Elemental analysis (%) for PKT-VCM pH 2.3 found: C 53.14, H 5.63, N 2.10, S 3.75.

PKT-VCM pH 5.0: M.p. 214 °C; FT-IR (KBr, cm^-1): ν˜=812 (m), 1023 (w), 1054 (s), 1095 (w, br), 1198 (w, br), 1259 (w), 1336 (m), 1384 (w), 1409 (s), 1427 (w), 1470 (w), 1693 (vs), 2852 (w), 2918 (w), 2962 (w), 3439 (vs, br), (see Table 2S in supporting information); Elemental analysis (%) for PKT-VCM pH 5.0 found: C 49.35, H 3.46, N 7.73, S 1.12.

PKT-VCM pH 8.8: M.p. 214 °C; FT-IR (KBr, cm^-1): ν˜=812 (m), 1023 (w), 1054 (s), 1095 (w, br), 1198 (w, br), 1259 (w), 1336 (m ), 1384 (w), 1409 (s), 1427 (w), 1470 (w), 1693 (vs), 2852 (w), 2918 (w), 2962 (w), 3439 (vs, br), (see Table 2S in supporting information); Elemental analysis (%) for PKT-VCM pH 8.8 found: C 52.78, H 4.71, N 3.66, S 3.41.

2.5. Vancomycin Release

Vancomycin release from PKT-VCM and PKSK-VCM was assessed using test tubes containing 5.0 mL of PBS phosphate buffer pH 7.4 with 5.0 mg of samples and thermostated at 37 °C. The experiments were carried out in triplicate (without stirring) under sink conditions. Released antibiotics were quantified by UV at a wavelength of 281 nm [99,103,104].

2.6. Antibacterial Activity

Antimicrobial activity was determined by standard liquid dilution method in Nutrient broth (NB) medium. S. aureus ATCC 25923 cells were grown overnight at 37 °C in NB broth and diluted in the same medium for assays. In 96-well sterile microtiter tray, 50 µL of bacteria from overnight culture (adjusted to 1x106 cells/mL) was added to the serial dilution of PKT-VCM and PKSK-VCM from 25 µg/ml and final concentration of 0 mg/ml as blank, in a total volume of 150 µl of NB. The 96-well microtiter tray was incubated at 37°C, overnight, in a shaking incubator and the cells growth was assessed by measuring OD at 600 nm. MICs (minimum inhibition concentrations) were determined as the lowest antibiotic.

3. Results and Discussion

3.1. Synthesis of Post Functionalized PKs

PK prepared according to literature method [68] (Mw = 15.272 g mol⁻¹ determined as reported by Vavasori and coworkers [93]), was post-functionalized by Paal-Knorr reaction in the presence of taurinate or sulfanilate (1/1 molar ratio) at 70 °C, for 48 h in methanol as solvent (Scheme 1a, 1b). After work up (see experimental section), the solid polymers, recovered in high yields (2.3g, 80% and 1.96g, 85% respectively for PKSK and PKT), were characterized by FT-IR, DSC, TGA, SEM and elemental analysis. Taurinate and sulfanilate were chosen to prepare a specific modified PK (PKSK and PKT) containing anionic functional groups, which should promote the interaction and immobilization of VCM

Moreover, it is interesting to note that taurinate and sulfanilate were expected to improve the solubility of the post-functionalized polymers, yet both PKSK and PKT were not soluble in any conventional organic solvent nor in water (at any pH).

3.2. Influence of pH on Vancomycin Loading

To find best loading conditions, it is important to consider that VCM is a large molecule bearing six different acidic and basic functional groups with specific pKa (Figure 1). Thus, VCM is differently charged, going from -4 at pH > 13, to +2 at acidic pH (~0) [105], and consequently its reactivity towards PKSK and PKT will be influenced by the pH.

PKSK and PKT were loaded with VCM at room temperature using a common protocol and operating at three different pH values (2.3, 5.0, 8.8) [104]. Polymers loaded with VCM, named PKSK-VCM and PKT-VCM, were purified by dialysis and unabsorbed VCM measured by UV at 281 nm (Figure 1S) [101,102,103], to determine encapsulation efficiency (EE%) and loading capacity (LC%) (reported in Table 1; all experiments were performed in triplicate and all data are expressed as the mean ± standard deviation).

Data reported in Table 1 show that, with both polymers, maximum LC% and highest EE% were achieved at pH 5.0. These data where further confirmed by elemental analysis, since highest nitrogen content was determined at pH 5.0, confirming the presence of higher amounts of Vancomycin. At pH 2.3 or 8.8, a decrease of EE% of up to 34% for PKSK-VCM and 57% for PKT-VCM and LC% between 8% and 21%, respectively, was registered. According to the literature, at pH 2.3 VCM has the higher positive charge (+2) and therefore this pH was supposed to be the optimal one for the loading on anionic polyketones tested[105]. Nonetheless data reported in Table 1 clearly show that highest VCM loading was obtained at pH 5.0, confirming that not only ionic but also hydrophobic interactions are important to promote VCM loading. Moreover, higher encapsulation efficiency by PKSK-VCM as compared to PKT-VCM, may be attributed to higher hydrophobic interactions between VCM and the aromatic functional groups present in PKSK [38,90,91,106].

3.3. PKT-VCM and PKSK-VCM Characterisation

PKT, PKSK, PKT-VCM and PKSK-VCM were lyophilized and characterized by FT-IR, DSC, TGA, SEM and elemental analysis. FT-IR of PKSK, PKT and PKT-VCM, PKSK-VCM prepared at different pH show very modest differences, and no clear evidence of the presence of VCM in the FT-IR of PKSK-VCM and PKT-VCM could be highlighted, probably due to the very low amount of antibiotic present with respect to the polymer. Nevertheless, comparison of FT-IR of polymers before and after VCM loading show a slight shift of the peaks in the range 1010-1095 cm^-1, corresponding to the SO₃^- groups which are the functional groups mostly affected by the loading of VCM (see Figure 2S and Tables 1S, 2S). Additionally, picks corresponding to the aromatic rings present in PKSK also show a modest shift (813-832 cm^-1 to 806 cm^-1), (see Figure 2S and Tables 1S, 2S), suggesting that not only ionic but also hydrophobic interactions may contribute to drug retention [38,90].

To gain further information on the characteristics of polymers prepared, thermal analyses (DSC, TGA) were performed (Figure 3S, Figure 4S). In agreement with the literature [107], due to the very low loading of VCM on the polymers, no significant difference was observed between PKT and PKT-VCM or PKSK and PKSK-VCM.

Interestingly, SEM analysis of PKT-VCM and PKSK-VCM with higher VCM loading (Table 1), showed evident differences between the polymers before and after VCM loading (Figure 2). The SEM images of PK (Figure 2a) show pores with lamellate edges of average size 1.13 ± 0.05 μm, while on PKT surface (Figure 2b), pores are smoother, less laminated with poorly defined edges (0.48 ± 0.03 μm), probably due to the functionalization of PK with taurinate, leading to morphological changes. Similarly, PKSK images (Figure 2c) showed a different morphology compared to PK, due to the partial functionalization, although also in this case pores were visible (0.25 ± 0.02 μm). When VCM was loaded on PKT or PKSK (Figure 2d and Figure 2e respectively), the morphology of the polymers further changed. New particles with a smoother and rounder surface, with spherical geometry for PKT-VCM, were observed on the polymeric surface (PKT-VCM 0.64 ± 0.04 μm; PKSK-VCM 0.43 ± 0.04 μm). These particles are probably due to the presence of VCM in the samples.

3.4. Release of Vancomycin from PK Polymers

Different polymeric samples of PKSK and PKT loaded with VCM at different pH (PKT-VCM 2.3, 5.0, 8.8 and PKSK-VCM 2.3, 5.0, 8.8, see Figure 3) were employed for the study of VCM release at physiological pH 7.4, analogously to the procedure reported by Ruiz et al. for a formulation of VCM on surface-modified polypropylene[104]. According to data reported in Figure 3, it emerges that, PKSK-VCM release trends are rather similar, while a wider gap exists for release trends by PKT-VCM. Further, release of VCM from PKSK-VCM 5.0 is slower compared to PKSK-VCM 2.3 and 8.8, further suggesting the higher strength of the interactions formed between PKSK and VCM at this specific pH. On the contrary, release trends of VCM from PKT-VCM prepared at different pH, show higher differences compared to those of PKSK-VCM, and the slowest release was obtained with PKT-VCM prepared at pH 2.3.

The difference in VCM release profiles between PKSK and PKT could be ascribed to the different chemical composition of the post-functionalized polyketones. In fact, PKSK and PKT were deliberately modified with different functional groups (all bearing an SO₃-) but with different chemical backbones to highlight possible differences between aromatic (PKSK) and aliphatic (PKT) functionalities present. Some minor differences between the two polymers have been evidenced by FT-IR, but release tests clearly highlight the importance of the functional group present on the polymer. It may be speculated that not only ionic but also hydrophobic interactions between the aromatic functionality of sulfanilate and VCM, contribute to strengthen the interaction between the drug and the polymer.

3.5. Antimicrobial Tests

To test PKSK-VCM and PKT-VCM antibacterial activity, minimal inhibitory concentration (MIC) analyses were performed against Gram positive bacterial strains Staphylococcus aureus ATCC 25923 (MRSA), one of the main pathogens that can form biofilm infections [108,109]. Preliminarily PK, PKSK and PKT were tested against Staphylococcus aureus and their inefficacy verified (Figure 5S).

Data obtained from the antimicrobial tests carried out in the presence of PKSK-VCM and PKT-VCM are reported in Table 2 (corresponding growth curves Figure 6S). These data once more clearly show a significant dependence of the efficiency of VCM loading from the pH. In particular, loading at pH 5.0 allowed to achieve MIC values ~6 times lower than VCM alone, with both polymers, while PKSK-VCM and PKT-VCM loaded at pH 2.3 and 8.8 showed MIC values comparable to or higher than VCM alone.

4. Conclusions

In this work VCM has been successfully loaded on two polymers (PKT, PKSK) obtained from PK modified with the Paal-Knorr reaction using taurinate (T) and sulfanilate (SK). The loading of VCM was studied at three different pH, showing a substantial influence in the characteristics of the final polymeric formulation. Indeed, for both polymers the best EE% and LC% were obtained at pH 5.0 (for PKSK EE% 75.00, LC% 31.03; PKT EE% 80.00, LC% 31.58), together with prolonged release. Interestingly, MIC values obtained with PKSK-VCM and PKT-VCM were 6 times lower than those of VCM alone. These optimal achievements might be the result of different interactions between VCM and polymers that were supposed to involve both ionic and hydrophobic interactions. Moreover, pH used for loading has proved to be a very important variable, probably affecting the interactions occurring between the drug and the polymer. These results confirm the promising role of functionalized polyketones as DDS for VCM delivery and therefore further studies are ongoing to understand the influence of the polymer structure and the mechanisms of VCM loading, and the use of these polymers to produce transdermal patches for controlled release of Vancomycin. Moreover, the loading of different antibiotics on these polymers and their antibacterial activities will also be investigated.