(193t) Sustained Delivery of Phosphates from Crosslinked Peg Hydrogel Nanoparticles Suppress Collagenase Activity of Intestinal Pathogens

Iatrogenic injury to the intestinal tract, such as that which occurs when surgeons remove tumors via endoscopy or major surgical resection, can disrupt the normal microbiota and leave a large wound to heal in the presence of highly pathogenic bacteria. Previous studies indicate that certain bacteria (i.e. E. faecalis, S. marcescens & P. aeruginosa) are capable of secreting high levels of collagenase, a key phenotype involved in healing impairment. Our previous studies have shown that normal inhabitants of the intestinal microbiota (E. faecalis) can be provoked to express enhanced collagenase in the gut during surgical injury leading to post-operative complications such as obstruction and anastomotic leak (a surgical problem that occurs when the resected and reconnected intestine fails to heal)¹. A common approach to address this issue is the administration of oral and intravenous antibiotics to eliminate pathogens in the gut. However, clinical studies indicate that patients remain colonized by strains of E. faecalis and P. aeruginosa for as long as one week after surgery². We have previously shown that depletion of extracellular phosphate occurs in the intestinal tract following surgical injury and that this depletion is a major cue that triggers bacterial virulence. As a result of extracellular phosphate depletion, pathogens must scavenge phosphate from host tissues, leading to the disruption of the epithelium and impaired healing. Earlier in-vivo and in-vitro studies have shown that the maintenance of monophosphate (Pi) in the epithelium via oral supplementation prevents virulence expression while maintaining bacterial survival³. Our recent studies indicate that a polyphosphate hexamer, sodium hexametaphosphate (PPi), suppresses virulence⁴ as well as collagenase production in-vitro and prevents anastomotic leak when administered orally in mice subjected to intestinal injury and intestinally inoculated with collagenolytic bacteria⁵. Therapeutic drug delivery approaches focused on localized and sustained release of phosphates from nanoparticles to suppress collagenase production, while allowing commensal bacteria to proliferate normally offer great promise for the promotion of healing following surgical interventions in the intestinal tract.

We have developed Pi- as well as PPi- loaded poly(ethylene) glycol (PEG) hydrogel nanoparticles, NP-Pi or NP-PPi, respectively, using inverse phase miniemulsion polymerization, which result in sustained release of phosphates. Our previous studies have shown that NP-PPi suppress P. aeruginosa virulence in a dose-dependent manner in-vitro while maintaining bacterial survival. The goal of the present study is to determine the effects of NP-Pi and NP-PPi on the suppression of clinically relevant gram-negative and gram-positive collagenolytic bacterial strains.

Methods:

Nanoparticles were created using inverse miniemulsion polymerization⁴. This process involved free-radical initiated co-polymerization of PEG diacrylate (M_n= 575 Da) and N-vinylpyrrolidone (NVP). These reactants, as well as the water-soluble surfactant TWEEN 20 constituted the aqueous phase. The organic phase consisted of cyclohexane and Span 80 (an organic-soluble surfactant). The total concentration of reactive double bonds in the aqueous precursor was fixed at 0.685 M. A thermal initiator, potassium persulfate, was used to begin polymerization and was added to the precursor at a concentration based on a 1.5% molar ratio of initiator to total reactive double bonds. The reaction was allowed to proceed for 4.5 hours at 56°C under an ultra-high purity nitrogen blanket. Monophosphate (Pi) was physically encapsulated within the hydrogel nanoparticles during the inverse miniemulsion polymerization. After the reaction was complete, nanoparticles were precipitated in acetone and purified via three cycles of acetone washes with sonication and centrifugation. Particle size distribution was quantified using nanoparticle tracking analysis on a Nanosight LM10 system. Nanoparticle surface charge characteristics and swelling ratio were quantified using dynamic light scattering on a Malvern ZetaSizer NanoZS and gravimetric swelling measurements, respectively. Phosphate release kinetics from nanoparticles into bacterial tryptone yeast (TY) medium was determined using a molybdenum blue absorbance method. Blank nanoparticles were created by rinsing NP-Pi with excess deionized water until phosphate could no longer be detected by absorbance. NP-PPi were created by post-loading blank NPs with PPi. Blank NPs were exposed to a high concentration PPi solution for 14 days at 40°C ± 5°C. Post-loaded PPi-loaded NPs were then rinsed briefly with deionized water and characterized as described above. Clinically relevant collagenolytic pathogens were obtained from the Center for Surgical Infection Research at the University of Chicago Medical Center. Pathogens were prepared from frozen stock and were plated on tryptic soy broth agarized medium and incubated overnight at 37°C. A few colonies were transferred to 2mL of TY medium and grown in suspension for 24 hours at 37°C under shaking, after which experiments were performed to quantify bacterial virulence and survival. In order to quantify bacterial growth in solutions supplemented with NP treatment, bio-luminescent strains of S. marcescens and P. aeruginosa were obtained. Bio-luminescent P. aeruginosa with collagenase activity analogous to the isolated strain was available for purchase through Perkin Elmer in-vivo imaging reagents. Bio-luminescent S. marcescens was constructed by transformation with pAKlux2 plasmid containing luxCDABE luciferase cassette. To determine the ability of NP-Pi or NP-PPi to suppress collagenase, S. marcescens and P. aeruginosa were cultured in phosphate depleted media with varying concentrations of the following groups, free phosphate (Pi or PPi), NP-Pi or NP-PPi and blank (unloaded) nanoparticles. Bacterial growth was measured by bio-luminescence measurements. Collagenase activity was measured using a DQ^TM gelatin and fluorescein conjugate assay kit with fluorescence excitation at 495nm and emission at 515nm.

Results:

NP-Pi and NP-PPi were found to have an average particle diameter of 181.7 nm ± 52.5 nm and 179.6 nm ± 71.8 nm respectively, a zeta potential of -17.92 mV ± 1.05 mV and -18.34 mV ± 1.44 mV respectively, a swelling ratio of 2.8 ± 0.08 and 4.0 ± 0.24 respectively, and allowed for sustained release of phosphate for over three days.

Our in-vitro studies indicate that for the gram-negative pathogen S. marcescens, NP-PPi are effective at suppressing collagenase activity and have no significant effect on bacterial survival when compared to control. Secondly, the blank NP control elicits no significant response in bacterial growth or collagenase levels for S marcescens. For the gram-negative pathogen P. aeruginosa, NP-PPi demonstrate a significant suppression of collagenase activity and a slight promotion of bacterial growth. The blank NP control unexpectedly shows mild attenuation of collagenase activity as well as significant promotion of bacterial growth. In the case of both gram-negative pathogens, our data indicate that PPi is more effective than Pi at suppressing collagenase activity. PPi (either delivered free or from NP-PPi) suppresses collagenase activity of the gram-positive pathogen E. faecalis, however, this effect was not found to be as significant as that observed in the case of the gram-negative strains.

Conclusions & Future Directions:

PEG hydrogel nanoparticles allow for sustained release of phosphates (Pi and PPi). NP-PPi allow for significant attenuation in collagenase production of gram-negative pathogens as a result of PPi release from these nanoparticles.

Current studies are being performed to optimize the effectiveness of NP formulations in suppressing collagenase production across gram-positive and gram-negative pathogens. Future work includes (1) determining the resistance of NP formulations to washout in a parallel plate flow chamber, (2) in-vitro and in-vivo studies to determine the mucoadhesiveness of NP polymer formulations on cultured epithelial cells and on intestinal mucosa, and (3) determining the functional durability and efficacy of NP formulations in animal models of intestinal injury.

References:

1. (Shogan BD, et al. Sci Transl Med. 2015:7(286):286ra68.)

2. (Ohigashi, S, et al. J Gastrointest Surg. 2013:17(9):1657-64.)

3. (Zaborin A, et al. Proc Natl Acad Sci. 2009:106(5):6327-32.)

4. (Yin Y, et al. Ann Biomed Eng. 2016:45(4):1058-1068.)

5. (Hyoju, SK, et al. Ann Surg. 2017 [Accepted].)