Chitosan Hydrogels for Synergistic Delivery of Chemotherapeutics to Triple Negative Breast Cancer Cells and Spheroids

Abstract

Purpose

This study aimed to develop a hydrogel system for treating aggressive triple negative breast cancer (TNBC) via kinetically-controlled delivery of the synergistic drug pair doxorubicin (DOX) and gemcitabine (GEM). A 2D assay was adopted to evaluate therapeutic efficacy by determining combination index (CI), and a 3D assay using cancer spheroids was implemented to assess the potential for translation in vivo.

Methods

The release of DOX and GEM from an acetylatedchitosan (ACS, degree of acetylation χ_Ac = 40 ± 5%) was characterized to identify a combined drug loading that affords release kinetics and dose that are therapeutically synergistic. The selected DOX/GEM-ACS formulation was evaluated in vitro with 2-D and 3-D models of TNBC to determine the combination index (CI) and the tumor volume reduction, respectively.

Results

Therapeutically desired release dosages and kinetics of GEM and DOX were achieved. When evaluated with a 2-D model of TNBC, the hydrogel afforded a CI of 0.14, indicating a stronger synergism than concurrent administration of DOX and GEM (CI = 0.23). Finally, the therapeutic hydrogel accomplished a notable volume reduction of the cancer spheroids (up to 30%), whereas the corresponding dosages of free drugs only reduced growth rate.

Conclusions

The ACS hydrogel delivery system accomplishes drug release kinetics and molar ratio that affords strong therapeutically synergism. These results, in combination with the choice of ACS as affordable and highly abundant source material, provide a strong pre-clinical demonstration of the potential of the proposed system for complementing surgical resection of aggressive solid tumors.

INTRODUCTION

Combination chemotherapy has become ubiquitous in the field of drug delivery owing to its distinct advantages over single-drug therapies [1, 2]. Single-drug therapies cause acquired drug resistance following prolonged administration, leading to non-responsiveness to therapy, the need of increased dosages, and ultimately poor prognosis and quality of life [3–5]. Combination chemotherapy overcomes these limitations by disrupting multiple metabolic pathways simultaneously, resulting in three progressively desirable therapeutic outcomes: additive effect, potentiation, and synergism [2]. Among these, synergism is achieved when the combined efficacy of multiple drugs is remarkably higher than the sum of the individual therapies administered independently and represents the most auspicious therapeutic outcome [2, 6]. In the past, synergy of chemotherapy regimens has been mostly attributed to the properties and molar ratio of the administered drugs; however, recent studies demonstrate that the order by which the drugs are administered, or schedule, is critical towards maximizing the synergistic outcome [7–9] by comparing scheduled regimens with their single-drug therapy counterparts in vitro, in animal models, and in the clinical setting [7–17].

Current studies on drug delivery systems (DDSs) that deliver synergistic combination chemotherapy regimens utilize a variety of drug delivery carriers, namely nano/micro-particles and liposomes [2, 18, 19], polymer-drug conjugates [20], and hydrogels [21]. Among these, hydrogels present unique advantages. Owing to their flexible microstructure, hydrogels can be tuned by controlling the gel fraction and crosslinking density to achieve the desired release kinetics, dose, and molar ratio [22, 23]. Notably, hydrogels can serve as “depots” from which sustained release - and thus a local high concentration - is maintained towards the surrounding tissue, thus bypassing the pharmacokinetic limitations of other DDSs and free chemotherapeutics [24–26]. In hydrogel-based DDSs drug loading and release depends strongly on molecular interactions between the drug molecules and the polymer chains that form the hydrogel scaffold, as well as the gel morphology, both of which can be tuned through judicious chemical modification of the polymer [24, 26].

One polymer that has captivated researchers for development of DDSs is chitosan [27, 28]. Chitosan, a biopolymer derived from naturally abundant chitin, is biocompatible and biodegradable [29]. Chitosan chains display primary amine groups that enable facile chemical modification and have been utilized to tailor the release kinetics of the therapeutic cargo [23, 30]. Further, because chitosan has an endogenous short physiological half-life (i.e., days – week(s)), physical chitosan hydrogels present an optimal short-time DDS and release no toxic degradation byproducts [31]. Consequently, a variety of chitosan-based products have been developed for biomedical applications, such as oral and ocular delivery systems as well as wound dressings, some of which have received FDA approval, demonstrating clinical feasibility [32, 33].

In this work, we utilized the drug pair of gemcitabine (GEM) and doxorubicin (DOX), which were selected for their extensively documented synergism [7, 8, 34, 35]. Previous work has defined the optimum molar ratio (GEM: DOX > 1) and delivery kinetics (GEM prior to DOX) for this drug pair to treat breast cancer [7, 8, 34]. In previous work, we studied the physicochemical factors governing release of GEM and DOX from chitosan hydrogels modified with acetyl, butanoyl, and heptanoyl moieties, using an integrated in silico – experimental approach [23]. We demonstrated that acetyl-modified chitosan (ACS) affords the optimum delivery ratio and kinetics for this drug pair, in both single-drug loaded and co-loaded systems. In this work, we selected one particular ACS with degree of modification (χ_Ac) of 40 ± 5%, which demonstrated favorable kinetics and molar ratio of GEM/DOX during release. To assess the viability of this system as an injectable depot, initial release studies were performed with DOX and GEM at different single-drug loadings. Specifically, we optimized this system by examining different loading conditions for the individual drugs to achieve the release of individual DOX and GEM at a therapeutically viable concentration, namely 0.26 μM – 0.64 μM for DOX within 72 h, and 2.8–19.4 μM for GEM within 72 h. The optimized hydrogel co-loaded with GEM/DOX afforded release doses of 0.07–0.13 μM for DOX and 4.5–20.7 μM for GEM, after 72 h, which fall within the desired (synergistic) therapeutic window [7, 8, 34]. The ACS hydrogel was evaluated in vitro against the TNBC cell line MDA-MB-231 in the form of 2-D (cell monolayer) and 3-D (spheroids) systems. Notably, the select DOX/GEM-loaded ACS afforded a combination index (CI) of 0.142 ± 0.010 in the 2-D assay. This was considerably lower than the CI obtained with the free drug DOX-GEM combination (0.223 ± 0.002) at the same global drug concentration and molar ratio, indicating that the ACS delivery system provides a significant contribution towards achieving therapeutic synergism; for comparison, Vogus et al. obtained a comparable CI (0.12) using a complex microfluidic system that provides precise control of the delivery scheduled and dose of DOX and GEM against TNBC cells [8]. In the 3-D assay, the DOX/GEM-loaded ACS gel afforded a significant decrease in tumor volume, reducing the spheroid growth percentage from a 30 ± 2.9% increase in volume (no treatment) to −32.7 ± 2.2%; in comparison, the concurrent administration of a GEM:DOX ratio of 10:1 only reduced volume growth to 2.4 ±5.2%. Collectively, these results demonstrate the efficacy of the proposed ACS delivery system in achieving scheduling-based therapeutic synergism and its translational potential towards future clinical studies.

MATERIALS AND METHODS

Materials

Low-molecular-weight chitosan (85% maximum degree of deacetylation, 15 kDa) was obtained from Polysciences Inc. (Warrington, PA). Acetic anhydride, phosphate-buffered saline (PBS), potassium hydroxide, and the Kaiser Test kit were from Sigma Aldrich (St. Louis, MO). Doxorubicin hydrochloride (DOX) and Gemcitabine hydrochloride (GEM) were obtained from LC Laboratories (Woburn, MA). Triple negative breast cancer cells MDA-MB-231 were purchased from ATCC (Manassas, VA). Dulbecco’s Modified Eagle Medium (DMEM) and fetal bovine serum (FBS) were obtained from Genesee Scientific (San Diego, CA). Penicillin Streptomycin (Pen Strep) was obtained from Gibco (Gaithersburg, MD). 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT), was purchased from Invitrogen (Carlsbad, CA). Corning Matrigel Growth Factor Reduced (GFR) Basement Membrane Matrix was purchased from Fisher Scientific (Hampton, NH). All other reagents were of reagent grade or higher.

Modification of Chitosan

Chitosan modification was performed following an adapted protocol from previous work [23]. Briefly, chitosan was dissolved at 6 mg mL⁻¹ in 1% wt. acetic acid (aq.) and filtered to remove insoluble components. A stoichiometric amount of acetic anhydride was added dropwise, and the mixture was stirred at room temperature for 1 h. The reaction was stopped by precipitating the chitosan solution dropwise into a 3x volume of 5 M methanolic potassium hydroxide. The precipitate was collected by centrifugation and washed with methanol until neutral pH. The precipitate was then washed copiously with DI water, followed by 10 mM PBS, pH 7.4, to yield a physical hydrogel; for the experiments performed with cells, all washing steps were carried out aseptically.

The level of acetyl modification (χ_Ac) was determined by a modified version of the Ninhydrin assay developed by Kaiser et al. [36]. Lyophilized ACS was dissolved in 1% acetic acid (aq) at 3 mg mL ⁻¹, and unmodified chitosan was dissolved at varying concentrations (0.5–6 mg mL ⁻¹). Aliquots of 100 μL of chitosan solution were combined with 30 μL of potassium cyanide in water/pyridine and 30 μL of ninhydrin (6% v/v in ethanol). The solutions were then incubated at 100°C for 5 min, diluted 200x and analyzed via UV – vis spectroscopy at 570 nm. The level of modification was determined using Eq. 1.

$$\chi =\frac{\left(1-\frac{D\cdot M\cdot C^{*}}{C}\right)}{\left(1-\frac{D\cdot M_{mod}\cdot C^{*}}{C}\right)}$$ (1)

where χ is the degree of modification, C* is the molar concentration of amine groups on a chitosan chain, D is the dilution factor, C is the mass concentration used for the Kaiser test, M is the molecular weight of the glucosamine monomer, and M_mod is the molecular weight of the acetyl modification.

Single-Drug Loading and Release

Single-drug loading measurements were performed by incubating 150 mg of ACS hydrogel with 1.2 mL of drug solutions in PBS at 0.5–2 μg mL ⁻¹ for DOX and 25–100 μg mL ⁻¹ for GEM, for 48 h at RT. After loading, the drug-depleted supernatant was collected to collected to quantify the amount of drug absorbed in the hydrogel. An aliquot of 150 mg of loaded hydrogel was then combined with 1 mL of 10 mM PBS, pH 7.4, and placed in an orbital shaker at 50 rpm at 37°C. At selected time points, 200 μL of supernatant was withdrawn and replenished with PBS. After the release studies, samples containing DOX (loading supernatant and time point collections) were analyzed by fluorescent spectroscopy on a Biotek Synergy H1 Microplate Reader (Agilent, Winooski, VT) at λ_ex = 480 nm and λ_em = 580 nm. Samples containing GEM were analyzed using liquid chromatography on a Waters 2690 HPLC system (Waters, Milford, MA) equipped with an Aeris 3.6 μm C18 column (50 × 4.6 mm). The chromatographic method utilized an isocratic 5% acetonitrile in water (0.1% v/v formic acid) for 5 min, while monitoring the effluent via UV spectrophotometry at 290 nm. GEM concentration was determined by integrating the peak area. All experiments were performed in triplicate.

Dual-Drug Loading and Release

Dual-drug loading experiments were performed by incubating 150 mg of ACS hydrogel with 1.2 mL of drug solution at either low (25 μg mL⁻¹ GEM and 0.5 μg mL⁻¹ DOX in PBS) or high concentration (50 μg mL ⁻¹ GEM and 1 μg mL ⁻¹ DOX in PBS) for 48 h at RT. After loading, the drug-depleted supernatant was collected to quantify drug loading. Drug loading, supernatant sampling, and quantification of DOX release were performed as described in Section 2.3. The collected samples were analyzed using a Waters 2690 HPLC system (Waters, Milford, MA) equipped with an Aeris 3.6 μm C18 column (50 × 4.6 mm). The chromatographic method utilized a 5–100% gradient of acetonitrile (0.1% v/v formic acid) in water (0.1% v/v formic acid) over 10 min, while monitoring the effluent via UV spectrophotometry at 290 nm. GEM concentration was determined by integrating the peak area. All experiments were performed in triplicate.

2-D Cell Culture

Triple negative breast cancer cells, MDA-MB-231, were cultured in DMEM supplemented with 10% (v/v) FBS and 1% (v/v) Pen Strep in a humidified incubator at 37°C and 5% CO₂.

2-D Cell Viability Assay

Cells were seeded at a density of 3 × 10⁴ cells/well in a 24-well plate, corresponding to 5 × 10³ cells/well in a 96-well format, and allowed to adhere overnight. Trans-well inserts were added to the plate, and the treatment condition was applied. Treatments with free drug employed 100 μL of drug-infused media, whereas treatments with drug-loaded ACS were performed by adding the gel (25, 50, 75, 100, or 125 mg) followed by 100 μL of media. The 24-well plate was then placed in a humidified incubator at 37°C and 5% CO₂ for 72 h. The drug-infused media were obtained by diluting stock solutions of DOX and GEM in sterile DMSO with cell culture media (DMEM supplemented with 10% v/v FBS and 1% v/v Pen-Strep) such that the maximum concentration of DMSO did not exceed 0.5% (v/v). The mass of hydrogel used in the treatments was chosen to achieve a concentration of DOX released after 72 h equal to that utilized in free drug cell viability experiments. After 72 h of treatment the media was aspirated, and the cells were incubated with 500 μL of MTT (0.5 mg mL⁻¹ in cell culture media) for 4 h at 37°C. The MTT solution was aspirated and 500 μL of DMSO was added to the to dissolve formazan crystals. The concentration of formazan was quantified by UV-Vis spectroscopy at 540 nm on a Biotek Synergy H1 Microplate Reader (Agilent, Santa Clara, CA).

3-D Spheroid Culture

Spheroids constituted by MDA-MB-231 cells were cultured following the protocol established by Froehlich et al. [37] using the liquid overlay technique. Briefly, 100 μL of cell suspension, 1 × 10⁴ cells, in cell culture media supplemented in 3.5% Matrigel were seeded in a round bottom Cellstar® cell-repellent surface 96-well plate. Cells were cultured in a 37°C incubator with 5% CO2 and 100% humidity for 3 days forming spheroids.

3-D Spheroid Assay

After 72 h of culture, the 3-D spheroids were imaged to determine initial size prior to treatment using a Nikon Eclipse TE-2000E inverted fluorescence optical microscope with a mechanical stage (Nikon, Tokyo, Japan). Following imaging, the spheroids were treated via incubation for 72 h with either (i) free drug solution of DOX, (ii) free drug solution of GEM, (iii) free drug solution containing both GEM and DOX at a molar ratio of 10:1, or (iv) a dual-drug-loaded hydrogel releasing a molar ratio of GEM:DOX ≥ 10:1. After treatment, the spheroids were imaged via phase contrast microscopy. For treatments using gels, the spheroids were transferred to a 24-well plate and the volume of media was brought to 500 μL. Trans-well inserts were added to the plate, and the drug-loaded gel was added to the transwell insert and covered with 100 μL of media. The mass of gel added (25, 50, 75, 100, or 125 mg) was chosen to expose the cells to a concentration of DOX/GEM comparable to that utilized in case (iii).

Growth inhibition was adopted as a metric to quantify treatment efficacy against spheroids, as done in the current literature [38, 39]. Following treatment, the spheroids were collected and imaged via optical microscopy. All images of cells were processed using ImageJ (NIH) by removing background and measuring all spheroids as ellipses. The major axis (d_max) and minor axis (d_min) were measured and the volume of the spheroid was calculated as V = (Π x d_max x d_min)/6. The change in spheroid volume was defined by (V-V*)/V*, where V is the spheroid volume post-treatment and V* was the spheroid volume immediately prior to treatment.

In addition to growth inhibition, live-dead staining was performed to visualize cell death within the spheroids. Following treatment, spheroids were incubated with 100 μL of staining solution for a final concentration of 2 μM calcein AM (λ_excitation/λ_emission: 494/517 nm) and 4 μM EthD-1 (λ_excitation/λ_emission: 528/617 nm), for 15 min at 37°C. Following incubation, the spheroids were fluorescently imaged, using FITC and TRITC fluorescence filter cubes.

Statistical Analysis

Statistical significance was determined by evaluation of the Student’s t test using Microsoft Excel. Statistical significance for samples was evaluated against the control test condition, and p < 0.05 was considered significant. For analysis *, and ** represent p < 0.05, 0.01, respectively.

RESULTS

Single-Drug Loading and Release Studies

In previous work, we conducted an in-depth study of the transport mechanism of DOX and GEM on ACS hydrogels featuring different levels of acetylation (χ_Ac) [23], and selected a value of χ_Ac (40 ± 5%) that grants favorable release properties, namely, faster release of GEM relative to DOX, and a GEM:DOX molar ratio ≥ 10:1. In this work, we sought to optimize the loading of GEM and DOX on the select ACS hydrogel to obtain the release of a therapeutically-viable and synergistic dose of the drug combination.

Preliminary single-drug loading and release tests with both DOX and GEM were conducted. As anticipated, we observed an increase in loading for both DOX and GEM as the drug concentration in the feedstock increased (Fig. 1A and B, respectively). Specifically, loading of DOX was not significantly affected by loading concentration and ranged from 1.8 ± 0.2 to 3.1 ± 0.04 ng per mg of ACS hydrogel, while the loading of GEM ranged from 55.7 ±0.7 μg to 214 ± 6.7 ng per mg of ACS hydrogel. Release was performed in PBS at pH 7.4 and 37°C to simulate physiological conditions. We observed that the release of DOX was relatively independent of the loading drug concentration (i.e., no statistically significant correlation was observed, Fig. 1C), varying between 51.3 ± 13.6% and 65.1 ±0.7% at 72 h. Similarly, GEM release varied between 10.9 ± 1.0% and 16.3 ± 2.2% over 72 h (Fig. 1D).

Fig. 1.

Loading of (A) DOX and (B) GEM as a function of loading concentration utilized. Release profiles for (C) DOX at various loading concentrations: 0.5 μg mL⁻¹ (squares), 1 μg mL⁻¹ (circles), and 2 μg mL⁻¹ (triangles), and (D) GEM at various loading concentrations: 25 μg mL⁻¹ (squares), 50 μg mL⁻¹ (circles), and 100 μg mL⁻¹ (triangles). Data are represented as mean ± S.D. (n ≥ 3); ** represents p < 0.01, as obtained from the Student’s t-test.

We finally calculated the corresponding dose of DOX and GEM released from the ACS hydrogels based on the last time point (72 h) on the temporal release profiles (Fig. S1): DOX released dose ranged between 0.263 ± 0.04 μM and 0.632 ± 0.12 μM, while GEM dose ranged between 2.87 ± 0.3 μM to 19.4 ±4.3 μM, both of which are expected to be therapeutically efficacious.

Optimization of Dual-Drug Loading and Release

We then sought to optimize the dual-drug loading of ACS hydrogels to achieve the release of DOX and GEM with global kinetics and dosage ratio (1:10) that are conducive to therapeutic synergism [7, 8, 34]. To this end, we utilized the results of single-drug release studies to guide to design of dual-drug release studies. Specifically, two drug solutions were adopted: (i) a diluted solution featuring a GEM concentration of 25 μg mL ⁻¹ and DOX concentration of 0.5 μg mL ⁻¹, which granted final (72 h) doses of 2.87 ± 0.3 μM of GEM and 0.263 ± 0.04 μM of DOX in single-drug release studies; and (ii) a concentrated solution featuring a GEM concentration of 50 μg mL⁻¹ GEM and DOX concentration of 1 μg mL⁻¹, which granted final (72 h) doses of 7.43 ±0.6 μM for GEM and 0.330 ± 0.13 μM for DOX in single-drug release studies. The resulting loading in ACS hydrogel (χ_Ac of 40 ± 5%) was 65.1 ±8 ng of GEM per mg of ACS and 1.71 ±0.07 ng of DOX per mg of ACS when using the diluted loading solution, and 90.7 ± 15.3 ng of GEM per mg of ACS and 3.32 ± 0.3 ng of DOX per mg of ACS when using the concentrated loading solution (Fig. 2A). Notably, these loading values fall within experimental error relative to the single-drug loading values.

Fig. 2.

(A) Amounts of DOX and GEM loaded on ACS hydrogels using either diluted drug solutions (25 μg mL⁻¹ GEM and 0.5 μg mL⁻¹ DOX, black column) or concentrated drug solutions (50 μg mL⁻¹ GEM and 1 μg mL⁻¹ DOX, grey column). (B) Release profiles of DOX (red) and GEM (black) from the ACS hydrogels loaded with either concentrated (circles) or diluted (squares) drug solutions. Data are represented as mean ± S.D. (n ≥ 3); ns and * represent no statistical significance and p < 0.05, respectively, as obtained from the Student’s t-test.

The resulting temporal release profiles indicate that 22.0 ± 5.0% of GEM and 14.3 ± 1.6% of DOX are released over 72 h from the ACS hydrogel prepared using diluted loading solution (squares in Fig. 2B), whereas 41.1 ± 6.8% of GEM and 13.7 ± 1.1% of DOX are released from the ACS hydrogel prepared using concentrated loading solution (circles in Fig. 2B). These correspond to GEM and DOX doses of 4.56 ± 1.4 μM and 0.067 ± 0.01 μM released over 72 h from the ACS hydrogel loaded with diluted drug solutions, and 20.7 ±2.4 μM and 0.126 ±0.02 μM released from the ACS hydrogel loaded with concentrated drug solutions (Fig. S2).

2-D in Vitro Viability Assay

We then resolved to evaluate the therapeutically relevance of our DDS against the TNBC cell line MDA-MB-231 in a 2D (cell monolayer) cytotoxicity assay. The dose response curve and the corresponding fit with the median effect equation [40] are presented in Fig. 3A. From these results we derived the IC50 values for the various treatments, namely 23.1 ± 2.5 μM for GEM, 0.641 ±0.06 μM for DOX, 0.112±0.01 μM for the GEM+DOX free drug combination (10:1 M ratio), and 0.071 ±0.002 μM for the GEM+DOX drug combination released from the ACS hydrogel (we note that the IC50 of DOX is higher than the value that we reported in prior work on ACS hydrogels and found in the literature, namely 0.08–0.2 μM; in this study, however, we utilized a high seeding cell density to build our 2-D assay, and the resulting high cell density is known to increase the effective IC50 [41]). To evaluate drug synergism, we utilized the IC50 values provided by the combination therapies to calculate the combination index (CI) [42]. For the free drug combination, the CI was found to be 0.223 ± 0.002, indicating a strong synergism, consistent with previous findings [7, 8, 34]. Notably, the DDS afforded a CI of 0.142 ± 0.010, indicating even higher synergism. For comparison, Vogus et al. obtained only a slightly lower CI (0.12) using a model microfluidic system capable of precisely controlling the delivery scheduled and dose of the same drug pair (DOX and GEM) against TNBC cells [8]. This indicates that the DDS successfully delivers the desired kinetics and drug dose, demonstrating the effectiveness of the proposed design. In addition to the IC50 and CI values, we recognize that other metrics besides IC50 or CI, such as the Hill coefficient, have been shown to be valid predictors of the in vivo efficacy [43]. Accordingly, the Hill coefficients resulting from the treatments tested in this work are tabulated in Table S1, along with the corresponding values for IC50 and CI.

Fig. 3.

Loading of (A) DOX and GEM for the two concentrations tested; a low concentration (25 μg mL⁻¹ GEM and 0.5 μg mL⁻¹ DOX) and a high concentration (50μg mL⁻¹ GEM and 1 μg mL⁻¹ DOX). (B) Viability of untreated cells and cells exposed to 1 50 mg of ACS hydrogel (unloaded). Data are represented as mean ± S.D. (n ≥3).

To ensure that the improved treatment resulted solely from the DDS implementing favorable delivery kinetics, we evaluated the biocompatibility of the hydrogel with no drug cargo. The data in Fig. 3B, reporting the comparison in the viability of gel-treated and non-treated cells, demonstrate excellent biocompatibility of ACS hydrogel.

3-D Spheroid Assay

To further assess the therapeutic efficacy of the proposed DDS, we evaluated the GEM/DOX-loaded ACS hydrogel using a 3-D tumor spheroid assay, which has been shown to reflect the likelihood of successful translation of DDS in vivo [38, 39]. With this assay, we sought to achieve two goals: demonstrate that the DDS offers a better treatment outcome than the prescribed dual free drug administration and determine the necessary dosing needed to maximize tumor volume reduction. Fig. 4 reports the achieved change in tumor volume following a mid-dose drug treatment (~ 2 μM).

Fig. 4.

Change in volume of spheroids corresponding to a selected dose of treatment. The control (black) and ACS hydrogel (grey) represent spheroids receiving no treatment and exposure to 1 50 mg of unloaded hydrogel, respectively. Treatments were performed using drug concentrations of 1.67 μM GEM (green), 2.5 μM DOX (red), 20 μM GEM + 2μM DOX as free drugs concurrently administered (blue), and 20.8 μM GEM + 2.08μM released from the ACS hydrogel (orange). Data are represented as mean ± S.D. (n ≥ 3); ns represents no statistical significance, * and ** represent p < 0.05, and p < 0.01, respectively, as obtained from the Student’s t-test.

After 72 h, the untreated spheroid showed a 30 ±2.9% increase in volume, nearly identical to that incubated with the unloaded ACS hydrogel (29.6 ± 7.3%), thus corroborating the claim of biocompatibility of the DDS. The volume increase of spheroids treated with 1.67 μM GEM dropped to 20.5±2.1%, yet without showing a statistically significant difference compared to the control. When treated with 2.5 μM DOX, the increase in spheroid volume was further reduced to 8.7 ± 4.5%, with statistically significant difference compared to the control (p < 0.05). Similarly, the concurrently free drug combination treatment (2.0 μM DOX and 20 μM GEM) limited the spheroid growth to 2.4 ± 5.2%, indicating a statistically significant volume reduction (p < 0.05). Remarkably, the DDS delivering a dose of 2.08 μM DOX and 20.8 μM GEM afforded a reduction in spheroid volume of −32.7 ±2.2%, which corresponds to a statistically significant volume drop compared to both no treatment and 2.5 μM DOX treatment (p < 0.01), as well as concurrent free drug combination at a similar dose (p < 0.05).

We also utilized fluorescence-based live-dead staining to identify a relationship between volume change in the spheroids upon treatment and the viability of the cells constituting those spheroids (Fig. 5, where green fluorescence and red fluorescence represents viable and non-viable cells, respectively). Notably, the increase in red fluorescence trends in the same direction as volume change, namely substantial decreases in volume change results in increased red fluorescence relative to the control. Although both free and ACS-based combination treatments yield a similar red fluorescence, the spheroids treated with the dual-drug-loaded ACS are visibly smaller (Fig. 5 K and L) compared to those treated via concurrent free drug administration at a similar dosage (Fig. 5I and J). Collectively, these data demonstrate the efficacy of treatment stemming from the proposed DDS.

Fig. 5.

Images of spheroids corresponding to the conditions utilized in Fig. 4; all grey-scale images represent spheroids prior to treatment and all fluorescent images represent spheroids following treatment; green and red, represent viable and non-viable cells, respectively. (A) and (B) are the control, (C) and (D) are the unloaded hydrogel, (E) and (F) represent treatment with 1.67 μM GEM, (G) and (H) represent treatment with 2.5 μM DOX, (I) and (J) represent treatment with 2.0 μM DOX and 20 μM GEM concurrently administered, and (K) and (L) represent treatment with 2.08 μM DOX and 20.8 μM GEM released from the ACS gel. The scale bar in all images is 400 μm

To fully elucidate the impact of drug dosage on spheroid volume reduction, we finally explored the effect of concentration of both single and dual-drugs on the spheroid volume change (Fig. 6). Treatment with free GEM lowered the spheroid volume increase from 30.1 ±9.2% (0.21 μM GEM) to −7.29 ±3.0% (53.3 μM); considering the viability in spheroid volume upon treatment (Fig. 5F), it is evident that GEM fails to reduce the viability of the spheroids (Fig. S3A 1–5), although it substantially reduces the ability of the spheroid to proliferate and grow as dosing increases. Treatment with DOX, on the other hand, afforded a drastic decrease in spheroid volume, from 19.4 ± 3.6% (0.02 μM) to −10.2 ± 2.4 (1.25 μM). At higher DOX doses, an inverse trend in volume increase is observed (Fig. 5H). This, however, can be attributed to a strong reduction in viability (Fig. S3B 1–5) that results in loss of cohesiveness, and ultimately in a high observed volume increase (51.9 ± 6.3% at 5 μM). Spheroids exposed to the combination of free DOX and GEM showed a trend comparable to that of DOX-only treatment, namely a reduction in volume from 25.6 ±0.5% volume change at 0.016 μM DOX (0.160 μM GEM) to 2.4 ±5.2% at 2 μM DOX (20 μM GEM), followed by a drastic increase in volume change, up to 66.9 ± 19.3% at 4 μM DOX (40 μM GEM), and loss of viability at higher doses (Fig. S3C 1–5). Finally, the ACS DDS produced an initial volume decrease from −21.3 ± 3.8% at 0.42 μM DOX (4.2 μM GEM) to −32.7 ±2.2% at 2.08 μM DOX (20.8 GEM), which represent the highest reduction among all treatments, and the highest reduction in viability of the spheroid (Fig. 5 L). Upon increasing dosage to 4.02 μM DOX (40.2 μM GEM), the volume change increased to a − 12.7 ±3.3%, concomitant with a high loss in viability of the spheroid (Fig. S3D 1–5). We have tabulated the values of normalized red fluorescence per unit of area to account for any loss in tumor integrity due to treatment that could otherwise be misinterpreted as tumor growth. The values corresponding to the images of the spheroids contained in Fig. 5 and Fig. S3 are collated in Table S2 and S3, respectively.

Fig. 6.

Tumor organoid volume change observed as a function of drug concentration for free GEM (green), free DOX (red), free GEM:DOX concurrently administered at a molar ratio of 10:1 (blue), and GEM:DOX released from ACS hydrogel at molar ratio ≥ 10:1. The desired therapeutic window is indicated by the black box. Data are represented as mean ± S.D. (n ≥ 3).

DISCUSSION

Single-Drug Loading and Release Studies

Single-drug loading and release studies were initially undertaken to evaluate the transport of DOX and GEM through the ACS hydrogel and to ensure that the loading of both drugs enabled therapeutically efficacious release doses (Fig. S1). Our results demonstrate that (i) GEM release represents the first front of drug delivery, with the GEM flux exhausted within a 24 h time frame (Fig. 1D), whereas (ii) DOX is release at ‘steady state’ across the entire 72 h window (Fig. 1C). This release scheme, where the GEM total dose is released prior to a steady flux of DOX, represents the desired delivery schedule for this drug pair. Notably, the release profiles of both GEM and DOX (Fig. 1C and D) did not show dependence upon the initial drug loading on the ACS hydrogel, indicating that drug transport and release kinetics are controlled by the drug-polymer interactions and that the proposed DDS is efficient and robust. This study of single-drug loading and release are functional to the design and interpretation of the subsequent dual-drug release study and represent a necessary step in the pre-clinical characterization pathway [23].

Dual-Drug Loading and Release Studies

We subsequently utilized the results of single-drug studies to guide dual-drug loading and release studies. Notably, the amounts of GEM and DOX co-loaded on the ACS hydrogel (Fig. 2A) were within experimental error relative to single-drug loading. Whilst seemingly minor, this feature greatly simplifies the screening of the loading conditions, as it enables performing the loading studies in a facile single-payload format and subsequently translating them accurately in a multipayload format. On the other hand, a deviation from ideality can be observed in the release profile from a co-loaded hydrogel, especially for multimodal drugs like DOX (Fig. 2B) whose complex amphiphilic character drives complex drug-drug and drug-polymer interactions [28], making their transport mechanism difficult to predict; nonetheless, the release profile is independent of the loading concentration, as observed in the single-drug study. GEM release from the co-loaded hydrogel (Fig. 2B), on the other hand, was consistent with single-drug release at the lower concentration utilized and only deviated at higher loading concentrations (Fig. 1D). Notably, the dual-drug release profiles (Fig. 2B) deviate from those obtained with single drug loaded hydrogels (Fig. 1C and D). This phenomenon has been observed in prior work [23] and can be attributed to drug-drug interactions resulting in the formation of aggregates that comprise two GEM and one DOX molecule and migrate through the chitosan network.

In Vitro 2-D and 3-D Viability Assays

We finally evaluated the therapeutic efficacy of the proposed DDS against 2-D (cell monolayers) and 3-D (tumor spheroids) models of the TNBC cell line MDA-MB-231. In view of combination chemotherapy, we adopted (i) the “combination index” (CI) and (ii) the tumor volume reduction as critical performance parameters. For references, when DOX and GEM were concurrently released as free drugs in a 1:10 M ratio, we observed a synergistic treatment outcome, indicated by a CI of 0.23, consistent with published studies focusing on this drug pair and target cell line [7, 8, 34]. The treatment with dual-drug-loaded ACS hydrogel further reduced the CI to 0.14, providing evidence that the ability of the hydrogel to control the delivery kinetics and ratio improves the therapeutic efficacy. Notably, this result compares well with the CI value obtained by Vogus et al. (CI ~ 0.12) [8], who utilized a microfluidic setup to achieve a completely segregated delivery schedule across the same 72-h window. We recognize that other metrics, such as the Hill coefficient, may be valuable predictors of the in vivo efficacy compared to the IC50 or CI values [43]. Accordingly, we reported the Hill coefficients for the treatments tested in this work are tabulated in Table S1, along with the corresponding values for IC50 and CI. We note, however, that the Hill equation assumes simultaneous interaction of two ligands to the same target [44]. In this work, however, DOX and GEM are released with different kinetics and therefore impact the target metabolic pathways at different characteristic times. This motivates the discrepancy between the CI values, which indicate strong synergism of DOX and GEM released from the chitosan hydrogel, and the Hill coefficients. Another metric, namely the tumor spheroid volume reduction, has earned wide popularity in evaluating drug regimens as it is also an accurate predictor of successful in vivo translation [38, 39]. Accordingly, we also utilized a 3-D tumor spheroid model to evaluate the ACS DDS.

To our knowledge, the present study represents the first report evaluating the GEM+DOX pair against a TNBC spheroid model. When treated with GEM as a free drug in solution, the viability of the spheroids was hardly impacted (Fig. S3A 1–5), although higher doses clearly limit their growth, as shown in Fig. 6. The treatment with free DOX, on the other hand, achieved both reduction of spheroid growth and viability, resulting in the disintegration of the spheroids (Fig. 6 and Fig. S3B 1–5). The combined treatment with free GEM and DOX in solution offered a trade-off between volume reduction and viability, as reflected in the volume change curve (blue squares in Fig. 6), which appears to be an average of the results of the single free drug treatments (red and green squares in Fig. 6). Furthermore, the combined free drug treatment afforded a decrease in viability at increasing at higher doses comparable to that achieved by DOX alone (Fig. S3C 1–5). Spheroids treated with GEM/DOX-loaded ACS hydrogel, on the other hand, showed a more drastic volume decrease (orange squares in Fig. 6), combined with a major loss in viability (Fig. S3D 1–5). To evaluate nanoparticle delivery and penetration, other researchers have utilized models comprising heterogenous spheroids containing cancer cells and fibroblasts [45]. The cellular makeup endows these spheroids with a characteristic resistance to nanoparticle access, which makes them less vulnerable compared to their homogeneous (single cell type) counterparts. In this work, we resolved to focus on homogenous spheroids solely consisting of MDA-MB-231 cells. Our ACS DDS is a macroscopic hydrogel from which diffusion of the drug payload, rather than site-dependent DDS accumulation, is the mechanism for inducing effective spheroid volume reduction. While a heterogenous spheroid may represent a more realistic model of a tumor, for our DDS it is unlikely that the diffusion of the small drug molecules into and throughout a heterogenous spheroid would be significantly different from a homogenous model. Collectively, these results demonstrate that hydrogel-mediated scheduling is critical to improve therapeutic efficacy, advocating for the use of the proposed ACS hydrogel for the treatment of TNBC.

CONCLUSIONS

The use of hydrogels constructed from modified polysaccharides and loaded with chemotherapeutics has been extensively studied over the last decade and remains an active area of research. Next-generation therapeutic biomaterials are becoming increasingly focused on integrating the paradigm of scheduled and synergistic combination within the design of drug-delivery systems. These systems, by achieving a precise control of dosages and delivery schedules, provide enhanced drug synergism and enable a drastic reduction of the required therapeutic doses, while maintaining highly efficacious outcomes. This study focuses on the development of an ACS hydrogel that (i) is constructed with scalable chemistry on a naturally abundant material, (ii) is capable of delivering a recognized synergistic drug pair (DOX and GEM) with precise dosages and release kinetics, and (iii) can be delivered locally at the malignancy site. When evaluated against the MDA-MB-231 (TNBC) cell line, a cancer with notoriously poor prognosis, the proposed dual-drug-loaded ACS hydrogel afforded remarkable results in terms of both combination index a reduction of spheroid growth. This work provides a roadmap for developing naturally-derived hydrogels and evaluating their efficacy, thereby laying the groundwork for the future development of materials targeting aggressive metastatic solid tumors that - to this day - greatly impact the health and quality of life of millions of people worldwide.

ABBREVIATIONS

ACS

acetylated chitosan

CI

Combination index

DDS

drug delivery system

DMEM

Dulbecco’s Modified Eagle Media

DMSO

dimethylsulfoxide

DOX

Doxorubicin

FBS

Fetal bovine serum

GEM

Gemcitabine

GFR

Matrigel Growth Factor Reduced (GFR) Basement Membrane Matrix

HPLC

high pressure liquid chromatography

MTT

3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide

PBS

phosphate-buffered saline

TNBC

Triple Negative Breast Cancer