Synthesis of Amphiphilic Amino Poly-amido-saccharide and Poly Lactic Acid Block Copolymers and Fabrication of Paclitaxel-Loaded Mucoadhesive Nanoparticles

Abstract

Drug delivery to the esophagus through systemic administration remains challenging as minimal drug reaches the desired target. Local delivery offers the potential for improved efficacy while minimizing off-target toxicities, but necessitates bioadhesive properties for mucosal delivery. Herein, we describe the synthesis of two new mucoadhesive amphiphilic copolymers prepared by sequential ring-opening copolymerization or post-polymerization click conjugation. Both strategies yield block copolymers containing a hydrophilic amine-functionalized poly-amido-saccharide and either a hydrophobic alkyl derivatized poly-amido-saccharide or poly lactic acid, respectively. The latter resulting copolymers readily self-assemble into spherical, ≈180 nm diameter, positively charged mucoadhesive nanoparticles. The NPs entrap ultra-high levels of paclitaxel via encapsulation of free paclitaxel and paclitaxel conjugated to a biodegradable, biocompatible poly(1,2-glycerol carbonate). Paclitaxel-loaded NPs rapidly enter cells, release paclitaxel, are cytotoxic to esophageal OE33 and OE19 tumor cells in vitro, and, importantly, demonstrate improved mucoadhesion compared to conventional polyethylene glycol-poly lactic acid nanoparticles to ex vivo esophageal tissue.

Graphical Abstract

Introduction

Diseases of the esophagus present a unique challenge for targeted treatment delivery due to the repetitive swallowing of saliva and ingestion of solid and liquid foods that both wash drugs away from the esophageal mucosal surface and trigger peristalsis into the stomach, where their efficacy is further affected by exposure to gastric acid, alkaline bile, and metabolism through the portal venous system. The esophagus is an attractive target for development and testing of novel local treatment strategies, as drugs may be delivered orally or intraluminal directly to the esophageal mucosal surface. The esophageal mucosa comprises a stratified squamous epithelium with underlying stromal layer (lamina propria) and thin muscle layer (muscularis mucosa). It is the site of many inflammatory, infectious, and neoplastic disorders, including eosinophilic esophagitis, candidiasis, viral infections, Barrett’s esophagus, squamous dysplasia, squamous cell carcinoma, and adenocarcinoma¹. Additionally, persistent disparities exist in the treatment of esophageal cancer among the different racial, ethnic, and socially economic populations, calling for treatments that are more universal². Current treatments for most of these diseases are via systemic drug administration, whether oral or intravenous, resulting in off-target toxicities and low local drug concentrations within the esophageal mucosa. Liquid oral formulations can improve local delivery to the esophageal mucosal surface, however these are often not widely available and suffer a short residence time on the tissue³. Thus, there is a critical need to develop targeted treatment strategies that specifically localize to the esophageal mucosal surface to enhance drug concentration, duration, and efficacy.

Local administration reduces off-target side effects and increases retention time in the diseased tissue. One example of this strategy is to use a polymeric mesh. The mesh is surgically implanted with sutures or staples, and allows for the slow release of its drug payload overtime. However, these delivery platforms require surgery, exhibit variable drug release due to implantation site location, and lose their mechanical properties over time^4-8. Local administration of nanoparticles (NPs) eliminates the need for surgery; however, this approach is disease-specific and depends on the surrounding tissue compartment. Tumors in confined spaces like the breast or intraperitoneal (IP) space are well-suited, as injected nanoparticles will remain near the tumor^9-13. On the other hand, for areas such as in the gastrointestinal tract, local delivery is challenging, as administered nanoparticles may be readily ushered away via fluid movement and tissue peristalsis^14-17.

One solution to increase local tissue retention time is to use NPs composed of a ‘bioadhesive’ or ‘mucoadhesive’ material. Mucoadhesive materials are well known, with naturally derived materials such as chitosan^18,19 and alginate^20,21 being prime examples. Both polysaccharides are adhesive due to their high density of charge, capacity to partake in hydrogen bonding, and high molecular weight^22,23. Despite these advances, naturally occurring polysaccharides possess limitations. First, due to their natural origin, polysaccharide samples are poorly defined with varying polydispersity, length, and branching. Additionally, harvest of these polysaccharides from their natural sources results in endotoxins and other potential contaminates that reduce their biocompatibility^24-27. For these reasons, there is a need for synthetic polymers which mimic the mucoadhesive properties with greater tunability and improved batch-to-batch control over composition and structure.

Several synthetic polymers are being investigated for their mucoadhesive properties^28-31. For example, polyacrylic acid (PAA) is a current candidate due to its high density of carboxylic acids for interaction with the mucin layer^32-34. Similarly, cationic polymers such as poly(allylamine) hydrochloride (PAH)³⁵ and poly((2-dimethylamino)ethyl methacrylate) (PDMAEMA)³⁶ electrostatically bind to mucin’s negative charge. Despite being mucoadhesive, these positively-charged delivery platforms suffer from increased toxicties^37-39, establishing a need for less toxic synthetic polysaccharide mimetics. Poly-amido-saccharides (PASs) address some of these limitations^40-44. These carbohydrate-like polymers are synthesized via an anionic ring-opening polymerization of a bicyclic β-lactam sugar monomer and possess an amide linkage between the repeat units. By introducing an amine functionality at the 6’ position, amine-functionalized poly-amido-saccharide (AmPAS) are mucoadhesive like chitosan^45,46.

Of the methods to incorporate mucoadhesive polymers into a nanoparticle, amphiphilic copolymers (i.e., containing a charged mucoadhesive polymer and a hydrophobic polymer) are attractive due to their innate ability to self-assemble into polymeric nanoparticles in aqueous conditions. Herein, we describe the synthesis of several new cationic amphiphilic diblock copolymers based on poly-amido-saccharides via sequential ring-opening copolymerization of hydrophilic and hydrophobic bicyclic β-lactam monomers or post-polymerization click conjugation strategy to couple a hydrophobic poly lactic acid to a hydrophilic poly-amido-saccharide. Of the two block polymers, the latter readily forms ~200 nm diameter nanoparticles (mucoNPs) using a miniemulsion method and encapsulates paclitaxel (PTX), a common clinical chemotherapeutic for esophageal cancers, with a 50% encapsulation efficiency. The resulting nanoparticles release PTX over two weeks, are internalized and cytotoxic against esophageal cancer cells, and adhere to ex vivo esophageal tissue with greater retention than common PEG-based nanoparticles.

EXPERIMENTAL SECTION

Materials.

All materials were purchased from Sigma Aldrich or TCI chemicals unless otherwise specified. Cell lines were donated from collaborators at Massachusetts General Hospital. PEG-PLA was ordered from PolySciences, and PEG-tetrazine was ordered from Nanocs. The polylactic acid amine polymer (PDLLA-amine/PDLLA-NH2) was purchased from Nanosoft Polymers. We synthesized poly(1,2-glycerol carbonate)-graft-succinate-paclitaxel (PGC-PTX) following a published procedure¹³.

Synthesis of amphiphilic copolymers.

AmPAS and EthylPAS monomers were synthesized following previously published procedures⁴⁵. The new block copolymers were synthesized via a one-pot, block copolymerization. Specifically, EthylPAS monomer was solubilized in dry THF at a temperature of 0 degrees, followed by 0.02 equivalents of initiator and 0.1 LiHMDS base, following similar protocols^44,46. The EthylPAS polymerization is very rapid⁴⁴, so after 10 minutes, the AmPAS monomer was added, achieving a diblock copolymer.

PLA was purchased from Nanocs and functionalized with a transcyclooctane moiety. The thiol on the terminal group of the AmPAS was modified to contain a tetrazine functional group. These two polymers were then coupled together using established click chemistry^47-51. The polymers were characterized by NMR, FT-IR, and GPC. For determination of the copolymer molecular weight, we used the hydrophobic (protected) version as it is compatible with GPC solvent system.

Critical Micelle Concentration Assays.

1-1.5 mg of amphiphilic copolymers were solubilized into ethyl acetate, then diluted 2x into subsequent centrifuge tubes. After evaporation overnight, we solubilized the polymer dilutions into DI water saturated with pyrene. After 30 minutes of sonication, fluorescence was measure using a Horiba Jobin Yvon FluoroMax 3 Fluorimeter at 373nm and 384nm, then the ratio was taken to determine pyrene encapsulation.

Nanoparticle Formulation.

Nanoparticles were formulated via a miniemulsion process. Specifically, the amphiphilic copolymer 1:2 AmPAS-PLA was solubilized with paclitaxel and/or PGC-PTX in acetone. The aqueous solution was added to the organic phase, and then sonicated in a bath for 10 minutes, with subsequent sonication via a probe tip for 1 minutes. Nanoparticles were stirred overnight to evaporate the acetone, then put onto dialysis to remove unencapsulated material. Evaluation of the size, dispersity, and charge was performed via dynamic light scattering.

In Vitro Cytotoxicity Assays.

OE33 and OE19 cells were plated onto a 96-well plate at 5,000 cells/well and allowed to adhere overnight. MucoNPs and PTX were diluted with media, then applied to the cells. After three days of incubation, cell viability was evaluated via MTS assay.

Paclitaxel Release Assays.

NPs were placed into dialysis tubing and left in release buffer for the entirety of the study. The release medium was composed of 10mM pH 7.4 phosphate buffer, and was consistently replaced to ensure sink conditions. Quantification of PTX released from the NPs was determined via UPLC and absorbance at 240 nm using a standard curve.

Ex Vivo Mucoadhesion Assays.

Mucin was coated onto microscope slides and allowed to adhere overnight. Rhodamine-labeled mucoNPs and non-mucoNPs were pipetted onto the mucin-coated surface and allowed to dry. The slides were analyzed for fluorescent signal at the ‘time 0’ point. The slides were then washed with DI water, then imaged again. This process was repeated 10 times. Porcine esophagus was purchased from Animal Biotech Industries. Tissue samples were cut into 1cm strips and coated with rhodamine-labeled NPs. Samples were imaged under UV excitation, and a small sample was taken for later quantification. Tissues were then washed, imaged, and collected, up to 10 times. Fluorescence was quantified using a Horiba Jobin Yvon FluoroMax 3 Fluorimeter.

RESULTS AND DISCUSSION

Nanoparticles (NPs) are widely investigated to enhance the delivery of small molecule agents to specific tissue sites^52,53. Unfortunately, systemic administration of NPs typically leads to a small fraction of the administered dose (<10%) reaching the target⁵⁴, unless the target tissue is the liver, with off-target toxicity remaining. To improve efficacy and achieve higher doses of the agent at the tissue site, NPs are functionalized with targeting moieties or locally administered. Targeted therapies employ antibodies (or antibody fragments)^55-58, aptamers^59-62, peptides^63-66, or other moieties to increase the percentage of payload to the delivery site and to mitigate the off-target effects. Each active targeting strategy possesses advantages and disadvantages, but none overcome the issues of systemic delivery^67,68. Local administration ensures the delivery of NPs to the target tissue site, but this route of administration is only applicable to a handful of anatomical locations without the aid of invasive procedures. Given our interest in drug delivery to the esophagus, local administration is a viable option. Furthermore, the esophagus is coated in a negatively charged mucosal layer providing a means for retaining the NP at the esophagus via incorporation of a positively charged targeting moiety on the NP.

Chitosan is a positively charged polymer used in NP formulations to increase mucosal targeting and resident time⁶⁹. For example, Bigucci et. al., describe chitosan-based nanoparticles that penetrate the mucin allowing for delivery of vancomycin to the underlying epithelium. The vancomycin-loaded NPs adhere to mucin and adhesion increases with increasing positive zeta potential, while maintaining antimicrobial activity⁷⁰. Similarly, Chamsai et al. report chemotherapy-loaded chitosan-based nanoparticles for colorectal targeting⁷¹. Chitosan-based nanoparticles are also finding increasing uses in intra-nasal vaccines, and this active area is summarized by Yoo et al.⁷² and Mehrabi et al.⁷³ Although chitosan is widely used, alternatives to chitosan are of interest as discussed earlier, and amine-functionalized poly-amido-saccharides (AmPASs) are one potential candidate polymer system.

We established the following design requirements to guide our polymer and NP studies: 1) amphiphilic diblock copolymers comprised of a cationic block and hydrophobic block; 2) amphiphilic diblock copolymers of > 15,000 g/mol; 3) self-assembled NP formation; 4) NP diameters between 150-400 nm; 5) NP zeta potential greater than +15 mV; 6) amenability to drug encapsulation with an efficiency > 40%; 7) drug release over 2-weeks; and 8) mucoadhesive capabilities. Additionally, we are mindful of the composition of the NP with the goal to minimize toxicity. We selected PLA as one of the polymer components as it is a commercially used biocompatible polymer in FDA regulatory approve products. We chose the poly-amido-saccharides as this glycopolymer linked by amide groups will break down into sugars. Finally, we selected the PGC -succinate polymer as it is biocompatible and degrades into glycerol, succinic acid, and CO₂.

Synthesis of Amphiphilic Block Copolymers.

Amphiphilic copolymers have gained popularity as nanocarriers due to their ability to self-assemble and easily functionalize^74-76. We investigated two potential synthetic routes to prepare these new block copolymers, either one-pot sequential copolymerization or post-polymerization conjugation via click chemistry. For sequential copolymerization, we will polymerize the two bicyclic β-lactam monomers using anionic ring opening polymerization as the same reaction conditions will be used throughout the process. In contrast, the post-polymerization conjugation route to block copolymers is not restrictive to the polymerization methodology, employing simple click chemistry to link the two polymers together.

For the anionic ring opening polymerization, we prepared the AmPAS and EthylPAS monomers from D-glucal, following previously published protocols^44,45. In short, we converted the 6’-OH to an amine with a N-nosyl-N-boc protecting group, while the remaining glycal alcohols were protected with a benzyl group. To synthesize the lactam AmPAS monomer, we performed a cycloaddition with trichloroacetyl isocyanate (TCAI) (Figure S1A). We prepared the EthylPAS monomer using a similar cycloaddition reaction with tri-O-ethyl glucal (Figure S1B). We performed the copolymerization in dry conditions at 0°C in the presence of the initiator, an activated pentafluorophenyl ester and lithium bis(trimethylsilyl)-amide (LiHMDS) (0.1 equivalence with respect to the first monomer) (Scheme 1). For this sequential copolymerization, we added the EthylPAS monomer first, as the polymerization kinetics are rapid with full polymerization occurring in minutes, followed by the AmPAS monomer. Subsequent deprotection of the Nos/Boc and benzyl protection groups afforded the final diblock copolymer. By adjusting the ratio of the AmPAS monomer to the EthylPAS monomer, we prepared four diblock copolymers with monomer feed ratios (AmPAS:EthylPAS) of 5:1, 2:1, 1:1, and 1:2. Preparation of the corresponding 1:5 diblock polymer resulted in a copolymer primarily of EthylPAS with a lower molecular weight of 9K, consistent with just polymerization of the ethyl monomer. Otherwise, the copolymerization results in molecular weights with low dispersity (Đ) (Table 1). Polymer molecular weights, as determined via gel permeation chromatography (GPC), ranged from ~ 15-33kDa, and spectra obtained using nuclear magnetic resonance (NMR) were used to determine the copolymer ratios. For example, the ¹H NMR spectrum of the 5:1 AmPAS:EthylPAS is shown in Figure S2. The determined ratio is 4:1, as opposed to the initial 5:1 feed monomer ratio, reflecting that the AmPAS monomer did not polymerize to completion.

Scheme 1.

Synthetic route to the AmPAS-EthylPAS deblock copolymer.

Table 1.

Table of protected, amphiphilic copolymer Mw, dispersity, and critical micelle concentration

Copolymer Name	Experimentally determined Ratio	Mw - Theory (Da)	Mw – Actual (Da)	Đ	CMC (μM)
5:1 AmPAS-EthylPAS	4:1	28,084	32,210	1.21	130
2:1 AmPAS-EthylPAS	1.5:1	29,863	31,402	1.19	175
1:1 AmPAS-EthylPAS	1:1	18,148	19,738	1.05	75
1:2 AmPAS-EthylPAS	1:2	16,457	15,271	1.06	20
1:5 AmPAS-EthylPAS	0:5	14,978	9,782	1.09	n/a
1:2 AmPAS-PLA	1:2	43,430	43,902	1.43	0.5
1:5 AmPAS-PLA	1:5	31,715	34,573	1.41	1.5

Copolymer name refers to initial feed ratio of monomers for the polymerization or ratios of polymers used for the post-polymerization click conjugation.

For the post-polymerization conjugation strategy, we will again use the AmPAS but will replace the EthylPAS with poly(lactic acid) (PLA) as the hydrophobic polymer and perform a click reaction to couple the two together. PLA is widely used in FDA approved products and preclinical devices and is a commodity polymer^77-80. We purchased PLA with a terminal amine group (MW 20kDa) from Nanosoft Polymers. Next, we installed a trans-cyclooctene unit on the end of the PLA via NHS chemistry (Scheme 2). We polymerized the AmPAS monomer as before except that we used the initiator, pentafluorophenyl S-benzylthioglycolate. Subsequently, we attached a tetrazine with a short PEG spacer to the AmPAS terminus via a thiol-malemide reaction as this reaction is facile and specific. The PLA (20 kDa) was coupled to two lengths of the polymerized AmPAS, resulting in ratios of 5:1 and 2:1 via a Diels-Alder cycloaddition affording the final diblock copolymers (Scheme 2). A Diels-Alder reaction was chosen for its high yield, quick reaction kinetics, and specificity^47-49,81.

Scheme 2.

Synthetic route to AmPAS-PLA block copolymer.

The molecular weights of the copolymers are 44 and 35kg/mol respectively, and the PLA-AmPAS polymers exhibit greater dispersity than the AmPAS-EthylPAS due to the PLA block. FTIR data are consistent with the synthesized structure and indicate the presence of both polymers, as NMR was not feasible due to overlapping signals (Figure S3). The PEG-PLA (MW 15kDa) copolymer was purchased from PolySciences.

Next, we measured the critical micelle concentration (CMC) to gauge the ability of the copolymers to form nanoparticles. Of the copolymers, the 2:1 PLA-AmPAS copolymers exhibit significantly lower CMCs, prompting further investigation with these copolymers for this mucoadhesive application (Table 1, Figure S4). The CMC values for the AmPAS:EthylPAS polymers decrease with increasing hydrophobic content or EthylPAS. At the 1:5 ratio, the copolymer is hydrophobic, is minimally soluble in aqueous solution, and no does not form micelles. The AmPAS-PLA CMC values are considerably smaller, by more than 10X, and likely reflects PLA being more hydrophobic than EthylPAS, which possess three polar ether linkages within the repeat unit. Given the lowest CMC value is for the 1:2 AmPAS-PLA copolymer, we selected this block copolymer for the subsequent studies.

Preparation of Mucoadhesive Nanoparticles.

These amphiphilic copolymers naturally form NPs when in aqueous solution, however the size distribution is large. Thus, we employed emulsion precipitation, a common and scalable technique, to prepare well defined NPs of ~200 nm diameters with low polydispersity (PDI). This approach is also amenable to load hydrophobic payloads (Figure 1). Previously published work indicated that size, along with charge, are the two most important factors for mucoadhesive nanoparticles⁸². Specifically, positively charged nanoparticles with diameters between 150-400nm are mucoadhesive, with the 215 nm diameter particles achieving the greatest adhesion⁸³. We prepared nanoparticles (mucoNPs) from the 1:2 AmPAS:PLA using the mini-emulsion process and obtained particles with ~200 nm diameters with a low PDI (1.2) and a positive zeta potential of ~25 mV (Figure 2A-D), achieving the design requirements of size and charge. Additionally, we prepared neutral non-mucoadhesive nanoparticles (non-mucoNPs; 200 nm diameters; PDI 1.2) from PEG-PLA as a control group.

Figure 1.

Chemical structures of the selected mucoNP composed of the block copolymer of PLA and AmPAS and illustrations of the various PTX-loaded mucoNPs under investigation.

Figure 2.

Diameter (A) and (B) PDI of the library of mucoNPs; (C) Zeta potential of the mucoNPs compared to non-mucoadhesive control; (D) SEM of the mucoNPs; (E) Encapsulation efficiency of PTX loading into the mucoNPs; (F) Total PTX encapsulation; (G) Stability of mucoNPs versus non-mucoadhesive PEG NPs determined via diameter change over time (* p<0.05; ** p<0.01; *** p < 0.001)

We selected paclitaxel as the drug for NP encapsulation given its widespread use in the clinic for treatment of multiple cancer types, including esophageal cancers. Paclitaxel (PTX), which is isolated from the bark of the Pacific yew (Taxus brevifolia), binds to the β subunit of tubulin, halting cell division. PTX possess limited aqueous solubility requiring formulation with either cremphor EL (Taxol^®) or entrapped within an albumin nanoparticle (Abraxane^®) to be intravenously delivered. It is a potent anti-cancer agent with a typical in vitro IC₅₀ values between 1-10 ng/mL. By itself, Taxol has limited activity in esophageal cancer and is recommended for patients who are unable to tolerate combination chemotherapy⁸⁴. Chemotherapy regimens for esophageal cancer can be given in the preoperative, perioperative, postoperative, or definitive settings. Management for locally advanced disease is typically multimodal with various combinations of chemoradiotherapy, surgical resection, and immune checkpoint inhibition. Combination chemotherapy typically consist of Taxol with Carboplatin, 5-FU, or capecitabine^85-87. Abraxane, the NP formulated version of paclitaxel, demonstrates elevated in vitro efficacy in experimental advanced esophageal cancer⁸⁸. Abraxane and cisplatin are highly effective and well-tolerated first-line treatments for metastatic or inoperable locally advanced esophageal squamous cell carcinoma⁸⁹, suggesting that the nanoparticle format may provide paclitaxel delivery advantages⁹⁰.

We learned that loading free paclitaxel into the NPs (mucoNPs_freePTX) during the miniemulsion process results in significantly larger particles (380.5 ± 60.3 nm) with greater PDI (0.3) and inefficient encapsulation (<25%), indicating poor stability and formulation. To overcome these limitations, we added a second hydrophobic polymer to the formulation to drive and stabilize the NPs. Additionally, we sought to increase the drug loading capacity of the NPs, and, thus, selected a biodegradable and biocompatible polymer amenable to drug attachment (Figure 1). Specifically, we synthesized poly(1,2-glycerol carbonate)-graft-succinate-paclitaxel (PGC-PTX; Mw = 12kDa; Đ = ~1.2) following a published procedure (see Figure S5)^13,91,92. This polymer contains 34 mol% (~70 wt%) of PTX and degrades into glycerol, carbon dioxide, and succinic acid. Loading of this polymer into the core of the AmPAS-PLA nanoparticles (mucoNPsPGC-PTX) affords similar diameter particles to the unloaded nanoparticles (181.6 ± 12.1 nm and 176.3 ± 8.9 nm, respectively) with low PDIs (Figure 2). Additionally, nearly 100% of the polymer is encapsulated, resulting in ultraefficient drug loading. Loading both free PTX and the PGC-PTX results in NPs (mucoNPs_combo) with a diameter of 187.1 ± 14.2 nm and a PDI below 0.1. SEM imaging reveals a slightly smaller average nanoparticle diameter, 170.5 nm, but with a larger standard deviation, 35 nm (Figure 2D). The encapsulation efficiency (60%) is significantly greater than free PTX alone, resulting in nearly 1mg/mL concentration of PTX in the final NP formulation (Figure 2E,F). The weight percent loading of PTX in the mucoNPs_combo is 20%, significantly more than most PTX encapsulated NPs such as Abraxane^™ (10%) or PLGA NPs (5%)⁹³. Next, we evaluated nanoparticle stability over several weeks, measuring both diameter and PDI via dynamic light scattering (DLS). The AmPAS mucoNPs are stable maintaining their diameter over the span of 1 month, unlike the PEG-PLA, which aggregate over the same time span (Figure 2G).

Paclitaxel Release Profile.

A key aspect of a nanoparticle formulation is to maintain a prolonged drug release at the desired site. The esophageal epithelium turnover is 3 weeks in a healthy adult⁹⁴. However this timeline can shorten in esophageal cancer patients, especially those diagnosed with adenocarcinoma, which typically stems from GERD and acid reflux that damages the epithelium⁹⁵. Therefore, the ideal release profile for the mucoNPs is likely one in which a majority of the drug is released within a 2-3 week window, before the NPs in the mucosa are washed away into the stomach.

To identify the NP with the appropriate release profile, we evaluated the in vitro release of three different nanoparticle formulations: mucoNPs_freePTX, mucoNPs_PGC-PTX, and mucoNPs_combo (Figure 1). We placed nanoparticles in a dialysis membrane against a release buffer under sink conditions, and we removed aliquots at 1, 4, 8, 12 hours, and 1, 2, 3, 4, 7, 14, 21, and 28 days to measure PTX by UPLC along with a standard curve. The mucoNPs_freePTX exhibit the fastest release profile, with over 25% released in the first 24 hours (Figure 3A). The mucoNPs_PGC-PTX release PTX the slowest, with roughly 50% released over the 28-day period (Figure 3B). The rate constants are 0.5661, 0.3511, and 0.3784 for the mucoNPs_freePTX, mucoNPs_PGC-PTX, and mucoNPs_combo, respectively. The mucoNPs_combo release ~50% of their payload within the first 4-days, and then slowly release PTX over the following 4-week period, fulfilling the pharmacokinetic design requirement for delayed and prolonged drug release.

Figure 3.

Paclitaxel release over 24 hours (A) and 28 days (B)

In Vitro Evaluation.

We first examined the cellular uptake of the mucoNPs via flow cytometry. Specifically, we loaded rhodamine-B covalently labeled PGC polymer into the mucoNPs_combo or the non-muco NPs affording fluorescently tagged NPs. We selected two human cancer cell lines OE33 and OE19, established from adenocarcinoma of the lower esophagus and the gastric cardia/esophageal gastric junction, respectively. The former was classified as stage IIA while the later was stage III. It is well known that NPs are readily taken up in cancerous cells and to a lesser extent in healthy cells due to healthy cells being slower to divide and less metabolically active. This is another reason we envisioned a local approach, as opposed for systemic, for delivery of the nanoparticles. We treated OE33 and OE19 cells with mucoNPsrho or non-muco NPsrho, and measured NPs uptake via flow cytometry at 1, 2, 4, 8, 12, or 24 hours post treatment (Figure 4A). In the OE19s, the rhodamine signal saturates after 12-24 hours for both the mucoNPs and the control non-mucoNPs. Interestingly, in the OE33s, the mucoNPs exhibit significantly greater signal after 8 hours, indicating greater cellular uptake than the non-mucoNPs.

Figure 4.

(A) Timeline of the cellular uptake of rhodamine-labeled mucoNPs (black bars) versus the non-mucoadhesive PEG-NPs (grey bars) in OE19 and OE33 cancer cells; (B) Cytotoxicity of the mucoNP library against OE19 and OE33 esophageal cancer cells.

We then evaluated cytotoxicity against these two different esophageal cancer cell lines: OE33 and OE19. Both the mucoNPs_combo and the mucoNPs_freePTX are cytotoxic, with an IC₅₀ of 12.8 ng/mL and 12.9 ng/mL for OE33 cells, and 1.68 ng/mL and 3.34 ng/mL for OE19 cells, respectively (Figure 4B). The mucoNPs_PGC-PTX are the least cytotoxic against OE33 and OE19 with IC50s of 20.2 ng/mL and 14.9 ng/mL, respectively, likely due to their slower release over the 3-day period of this assay. The nanocarrier itself is not cytotoxic except at the highest concentrations (>1000 ng/mL). The difference between Taxol’s impact on OE19 and OE33 is well documented^96,97, although the reasoning behind how differences in paclitaxel efficacy are altered within each of the cell lines is currently unknown. However, the overall heightened cytotoxicity of the mucoNPs_combo is apparent and is derived from a combination of ultra-high PTX loading, rapid internalization, and PTX release.

Ex Vivo Mucoadhesion

Next, we evaluated the mucoadhesivity of the NPs using a published wash assay on both porcine gastric mucin-coated slides and ex vivo porcine esophageal tissue samples. We loaded the mucoNPs_rho or non-mucoNPs_rho onto the mucin samples, washed the samples with DI water, and recorded the fluorescence. The washing and fluorescent measurement steps were repeated 10 times. Treatment with the mucoNPs affords greater signal after subsequent washes, indicating greater mucoadhesion compared to the non-mucoNPs (Figure 5A). This trend is similar with the porcine tissue, as the fluorescent signal remains after more washes compared to the non-mucoadhesive NPs (Figure 5B). Exponential decay analysis of the tissue data reveals that the half-life of the mucoNPs was 5.19 washes compared to 2.06 washes for the non-mucoNPs (Figure 5C,D). Figure 5E qualitatively shows that the tissue treated with the non-mucoNPs (left) displays less rhodamine signal after 5 and 10 washes, compared to the mucoNPs (right) treated tissue. These data highlight the importance of the AmPAS and mucoadhesion over traditional nanocarriers.

Figure 5.

Fluorescent signal of the rhodamine-loaded mucoNPs vs non-mucoNPs after several washes on mucin coated slides (A) and porcine esophagus (B); Exponential decay analysis of fluorescent signal on the procine tissue for both mucoNPs (C) and the PEG NPs (D) with representative fluorescent samples of the mucoNP treated (left) and PEG NP treated (right) tissue samples (E).

CONCLUSIONS

Polymers as the building blocks for nanoparticles offer several advantages over other nanoparticles systems, including exquisite control of size, ease of functionality, and increased drug loading. We describe the synthesis of two new block copolymers composed of the cationic AmPAS and either EthylPAS or PLA. These amphiphilic copolymers self-assemble in aqueous environments and the AmPAS-PLA possesses lower CMC values. Using a miniemulsion process, we formulated mucoNPs with a diameter of ≈180 nm and a zeta potential of +25 mV from the AmPAS-PLA block copolymers. We encapsulated PTX within the mucoNPs as free PTX or PTX conjugated to poly(1,2-glycerol carbonate). The latter affords mucoNPs capable of loading high concentrations of PTX. These nanoparticles rapidly enter tumor cells in vitro and are cytotoxic. Finally, the mucoNPs are mucoadhesive to ex vivo esophageal tissue and show prolonged retention. These results support the further development of mucoNPs and future in vivo evaluation. Changing the paradigm in the clinic from systemic delivery to local administration, even for difficult disease such as esophageal disorders, opens the door to new therapeutic modalities. Specifically, the mucoadhesive nature of these drug carriers would enable precise application to the disease site, when delivered via an endoscope, potentially limiting off-target morbidities and in turn increasing the treatment efficacy. In summary, the opportunities for drug delivery at the mucosa are significant since mucosal surfaces line the endoluminal cavities of the digestive, respiratory, and reproductive organs, and, as such, continued research into new mucoadhesive polymers and mucoadhesive micro- and nano-carriers is warranted.

Data Availability.

The raw data required to reproduce these findings are available from the authors upon request.