Co-administration of a tumor-penetrating peptide improves the therapeutic efficacy of paclitaxel in a novel air-grown lung cancer 3D spheroid model

Abstract

Three-dimensional (3D) cell culture platforms are increasingly being used in cancer research and drug development since they mimic avascular tumors in vitro. In the present study, we focused on the development of a novel air-grown multicellular spheroid (MCS) model to mimic in vivo tumors for understanding lung cancer biology and improvement in the evaluation of aerosol anticancer therapeutics. 3D MCS were formed using A549 lung adenocarcinoma cells, comprising cellular heterogeneity with respect to different proliferative and metabolic gradients. The growth kinetics, morphology, and 3D structure of air-grown MCS were characterized by brightfield, fluorescent, and scanning electron microscopy. MCS demonstrated a significant decrease in growth when the tumor-penetrating peptide iRGD and paclitaxel (PTX) were co-administered as compared to PTX alone. It was also found that when treated with both iRGD and PTX, A549 MCS exhibited an increase in apoptosis and decrease in clonogenic survival capacity in contrast to PTX treatment alone. This study demonstrated that co-administration of iRGD resulted in the improvement of the tumor penetration ability of PTX in an in vitro A549 3D MCS model. In addition, this is the first time a high-throughput air-grown lung cancer tumor spheroid model has been developed and evaluated.

1. Introduction

Despite continuous efforts in the development of anticancer therapeutics, cancer is still the second most frequent cause of death in the United States, with lung cancer resulting in the highest cancer-related mortality rate.¹ The 5-year survival rate for lung cancer is only 18% as most cases are detected at an advanced stage and are associated with poor prognosis.¹ Therefore, there is a vital need to improve research methods and develop better treatment alternatives for more effective ways of combating lung cancer. Conventional anticancer drug screening relies on two-dimensional (2D) monolayer cell culture, which lacks the ability to create a physiologically representative tumor and results in unpredictable and conflicting results when compared to animal studies or clinical trials.² It has previously been reported that not only the cancer cells but the overall three-dimensional (3D) microenvironment of tumors are responsible for influencing drug sensitivity and development of resistance against anticancer therapeutics.³ Therefore, 3D multicellular spheroids (MCS), consisting of cellular aggregates in different proliferative and metabolic states that closely mimic microenvironmental features of in vivo tumors within intratumoral space, have been investigated for the evaluation of anticancer drugs.

MCS can range in size from 20 µm to 1 mm in diameter, depending on the cell type and growth conditions.⁴ MCS have been reported to simulate the conditions of more than 40 different types of malignant cancers in several aspects including protein expression, pH and oxygen gradients, poor vascularization, hypoxia, diffusion rates of growth factors within the spheroids, and interactions with their extracellular matrix.^{5, 6} When present in in vivo tumors, cells are subjected to different environments and the above-mentioned factors instigate 3D spheroids to undergo morphological and phenotypic changes in vitro that allow them to better mimic the cellular heterogeneity seen in solid in vivo tumors.

Cells located in the periphery of spheroids reflect actively growing tumor cells adjacent to capillaries in vivo. Cells located in the center of the spheroid become quiescent and eventually undergo apoptosis or necrosis and form a necrotic core.⁷ At a critical size of 300–600 µm, a necrotic core occurs in most, but not all, spheroid types. The necrotic core is typically surrounded by a viable rim of cells and the formation of the core is mainly due to limited diffusion of oxygen and nutrients as well as the accumulation of waste products, often leading to low pH and oxygen levels.⁸ The hypoxic cells present in the center of MCS can mimic drug-resistant populations of cells similar to in vivo tumors.

MCS have been formed using a variety of methods such as liquid overlay⁹, stirred culture¹⁰, encapsulation into natural or polymeric matrices¹¹, hanging drop culture¹², micromolding¹³, and centrifugation pellet culture¹⁴, among others. Spheroid models comprised of a wide variety of cancer types have been developed and characterized. Of these, breast cancer MCS have been the most prevalent¹⁵ and spheroids comprised of prostate¹⁶, brain (glioma)¹⁷, osteosarcoma¹⁸, and pancreatic¹⁹ cells have also been developed. Of the current MCS development efforts, less than 5% have focused on lung cancer models. Lung cancer MCS have been formed by seeding the cells into low attachment 96-well plates followed by centrifugation¹⁴, a bench top roller method²⁰, liquid overlay in agarose-coated 96-well plates²¹, spinner flask method²², embedment in collagen²³, microfluidic device²⁴, and liquid overlay on agar.⁹

Anticancer drugs are only able to penetrate tumor tissues 3–5 cell diameters from blood vessels, either due to low perfusion of blood in the tumor vessels or high interstitial pressure, which limits the flow of fluid in the tumor and prevents the drug from entering the tumor.²⁵ Tumors and the corresponding blood vessel surfaces express various different kinds of molecular signatures (nucleolin²⁶, annexin 1²⁷, plectin-1²⁸, p32 protein²⁹, αv-integrins³⁰), which can be used as targets for delivering anti-cancer drugs. In this study, we utilized αv-integrins as a potential target using the tumor homing and penetrating iRGD peptide. iRGD is a cyclic peptide (CRGDK/RGPDC) which initially binds to αv-integrins using its vascular recognition (CendR) motif, where it gets proteolytically cleaved and the truncated peptide (CRGDK/R) gains affinity for neuropilin-1 (NRP1).³¹ These changes have been reported to enhance the penetration of drugs into tumors, either when coupled as cargo with iRGD or when co-administered with iRGD (bystander effect).³² The bystander activity of iRGD when co-administered with paclitaxel (PTX), an anti-cancer drug that interferes with microtubule disassembly in actively proliferating cells³³, has not been investigated with a 3D air-grown lung MCS model. In this study, we evaluate the therapeutic efficacy of PTX when co-administered with iRGD to improve the penetrability and diffusion of PTX in air-grown lung MCS.

Previously studied lung MCS models involved the growth of spheroids in liquid media, which is not a completely accurate representation of tumors present in the airways of the lung. In the present study, we hypothesize that air-grown 3D MCS will better simulate the in vivo lung tumor microenvironment, and can potentially provide better connectivity between in vitro screening and in vivo studies in the evaluation of the effect of anticancer therapeutics iin vitro air-grown 3D lung MCS model was developed³⁴ using A549 lung adenocarcinoma cells, characterized their morphological changes and growth pattern with respect to time, and assessed their response to anticancer therapeutics.

2. Materials and methods

2.1 Materials

Dulbecco’s Modified Eagle Medium (DMEM), Pen-Strep, and Fungizone^® were obtained from Life Technologies (Norwalk, CT). Trypsin-EDTA, sodium pyruvate, and Dulbecco’s phosphate buffered saline (PBS) were obtained from Fisher Scientific (Waltham, MA). Sodium alginate (Protonal^® LF20/40) was purchased from FMC BioPolymer (Philadelphia, PA). Fetal bovine serum (FBS) was obtained from Atlanta Biologics (Flowery Branch, GA). Paclitaxel (PTX) was obtained from LC Laboratories (Woburn, MA). Annexin V-FITC apoptosis detection kit and Quant-iT PicoGreen dsDNA reagent and kit (Invitrogen) were purchased from ThermoFisher Scientific. iRGD peptide was purchased from Biomatik LLC (Wilmington, DE). A549 cells were obtained from ATCC (Manassas, VA).

2.2 Methods

2.2.1 Fabrication of 3D alginate scaffolds

The alginate was purified (Supplementary S1) and a hydrogel scaffold was fabricated using a simple three-step procedure: (i) 3 ml of alginate (2.5% w/v) was mixed with 3 ml of PBS, (ii) 1.35 ml of PBS was mixed with 150 µl of calcium sulfate (1.5 mM), and (iii) 6 ml of solution (i) and 1.5 ml of solution (ii) were mixed in two syringes connected to a syringe connector. The crosslinked alginate solution was cast in micromolds from 3D Petri Dishes^® (MicroTissues, Providence, RI) and allowed to gel at room temperature. The resulting alginate scaffold gels contained uniformly sized highly ordered microwells (7 × 5 array). These gels were equilibrated with DMEM at 37 °C in an incubator for overnight before cell seeding. Figure 1 illustrates the schematic showing the fabrication of the alginate scaffold matrix using 3D Petri Dishes^®.

Figure 1.

(a) Schematic of the development of an air-grown tumor spheroid platform. (i) Alginate scaffolds were formed in a 3D Petri Dish® micromold system, (ii) the scaffold was removed from the mold, (iii) transferred into a Transwell, and the A549 cells were seeded in the microwells over the alginate scaffold to allow the formation of multicellular spheroid (MCS). Eventually, (iv) alginate gels were degraded to allow for the culturing of MCS at air-media interface. (b) Effect of starting the cell number on A549 multicellular spheroids (MCS) size. (Left) Spheroids were developed from 575, 1425, 2285, 2850, 3700 cells/spheroid seeding densities and cultured for 7 days, and (Right) growth curve of air-grown A549 MCS over 30 days at a cell density 2850 cells/spheroid. Data is represented as mean ± SD, n = 5.

2.2.2 Mechanical characterization of alginate scaffolds

The mechanical properties of the alginate gels were assessed and compared with that of lung tissue before using them as scaffolds for MCS formation. Cylindrical shaped gels were prepared (10 mm in diameter and 7 mm high) and equilibrated with DMEM at 1 and 12 hours. The scaffolds were then subjected to compression testing using a mechanical testing apparatus (Instron model 3342) with a 50 N load cell, compressive strain rate of 1 mm/min, and no preload. The Young's modulus was calculated from the slope of the linear region of the resulting stress/strain curve.

2.2.3 Cell culture and development of high-throughput air-grown 3D multicellular spheroids

Lung adenocarcinoma cells (A549) were cultured in DMEM supplemented with 10% (v/v) FBS, Pen-Strep (100 U/ml penicillin, 100 µg/ml streptomycin), and Fungizone (0.5 µg/ml amphotericin B, 0.41 µg/ml sodium deoxycholate) at standard conditions (37 °C, 5% CO₂ and saturated humidity). Equilibrated alginate scaffolds were placed in 12-well Transwells and 700 µl of DMEM was added to the basolateral side of the Transwell. Cells were seeded (100 µl/micromold) in alginate scaffolds to induce spheroid formation. To optimize the number of cells for proper spheroid formation and growth, different cell concentrations (575, 1425, 2285, 2850, 3700 cells/spheroid) were used. The gels with cells were incubated at 37°C in 5% CO₂ incubator. Air-interface conditions were initiated 24 hours after cell seeding by carefully removing the media from the apical side of the Transwell.

2.2.4 Spheroid formation, growth kinetics, and morphology of air-grown 3D spheroids

Analysis of cellular association and the growth kinetics of spheroids were accomplished by capturing brightfield images at regular time intervals and cell concentrations (575, 1425, 2285, 2850, 3700 cells/spheroid; 35 spheroids/well). Brightfield images were captured using a Cytation 3 Imaging Reader (BioTek, Winooski, VT). The 3D structures of the spheroids were observed using a Zeiss SIGMA VP Scanning Electron Microscope (Germany).

2.2.5 Spheroid fixation, cryosectioning, and H&E staining

Three and seven days after formation, MCS were fixed with 10 % (v/v) formalin for 1 hour followed by washing in PBS. The spheroids were then transferred onto a vinyl mold and Tissue-Tek O.C.T. compound (Sakura Finetek, Torrance, CA) was added to the mold and kept in dry ice for 1 hour. Several thin sections (10 µm) were collected on glass slides (Tissue path Superfrost™ plus gold slides). The sections were stained with Mayer’s hematoxylin solution and Eosin Y staining and images were taken using a Nikon Eclipse E600 upright microscope.

2.2.6 Cellular viability of air-grown lung spheroids

Cellular viability of MCS was assessed at days 3, 5 and 7 following cell seeding in alginate molds. 1 µM of Calcein AM (live cells, green) and 4 µM of ethidium homodimer-1 (EthD-1, dead cells, red) were used to prepare the cell staining solution in PBS. The MCS were incubated in the staining solution for 30 minutes at 37°C. After washing the spheroids with PBS, fluorescent images of the spheroids were captured using the Cytation 3 Imaging Reader.

2.2.7 Cellular proliferation in air-grown lung spheroids

To quantify the proliferation of the cells present in the air-grown MCS, the DNA content of spheroids was measured as a function of time using a Quant-iT PicoGreen dsDNA assay kit. MCS (2850 cells/spheroid) were collected 3 and 7 days after spheroid formation. The spheroids were lysed using a spheroid lysis buffer (SCIVAX Life Sciences) and incubated for 20 minutes at 37°C followed by sonication for 5 minutes. 20 µl of the cell lysate was diluted with 80 µl of Tris-EDTA buffer and was then incubated with 100 µl of the PicoGreen working solution in 96-well plates for 5 minutes at room temperature. The DNA content of the spheroids was determined via fluorescence spectroscopy using a Cytation 3 plate reader with excitation and emission wavelengths of 480 nm and 520 nm, respectively. The rate of spheroid growth was evaluated by calculating the average spheroid area using the cellular analysis program of the Cytation 3 image reader, where a significant increase in projected area represented an increase in cellular proliferation.

2.2.8 Effect of paclitaxel and iRGD on spheroid growth

PTX stock (5000 µM) solution was made in DMSO, and it was serially diluted in DMEM to achieve the desired working concentration. Air-grown MCS (2850 cells/spheroid) were formed over 4 days. On day 5, the spheroids were incubated with PTX (0, 5, 10, 50 and 100 µM) in DMEM. Following treatment, brightfield images were taken at 4, 24, and 48 hours to observe the effect of PTX on the growth of spheroids by measuring their average size. In another set of experiments, iRGD (1 mg/ml) was co-administered with PTX to enhance the effect of PTX penetration into the spheroids. 1 mg of iRGD was initially mixed with 20 µl of DMSO and then 980 µl of media was used to completely dissolve the iRGD solution. Spheroids were prepared as above and on day 5, they were exposed to the following conditions: (i) 5 µM PTX, (ii) 5 µM PTX plus iRGD, (iii) 10 µM PTX, and (iv) 10 µM PTX plus iRGD. Brightfield imaging was performed to determine the effect of treatments on the growth of spheroids.

2.2.9 Quantitative evaluation of apoptosis via flow cytometry

Apoptosis was evaluated in A549 MCS treated with 10 µM PTX and 10 µM PTX plus iRGD at day 7 using an Annexin V-FITC and propidium iodide (PI) apoptosis detection kit (BD Biosciences, San Jose, CA) according to the manufacture’s instructions and then analyzed using FACS Verse flow cytometer (BD Biosciences). Following treatment, the alginate scaffold was degraded using EDTA (100 mM) and the spheroids were dissociated into single cell suspensions using trypsin-EDTA and washed twice with PBS. The resuspended single cells were stained with Annexin V-FITC and PI for 15 minutes in the dark. Stained cells were immediately subjected to flow cytometry to determine the percentage of live and dead cells. MCS treated with PTX-free DMEM was used as a negative control.

2.2.10 Spheroid clonogenic survival assay

A clonogenic assay was performed on MCS treated with 10 µM PTX and 10 µM PTX plus iRGD 7 days after treatment. A549 MCS treated with PTX-free DMEM was used as a negative control. The potential of A549 cells in the treated spheroid to form colonies (>50 cells) was evaluated by a clonogenic method as described previously.^{35, 36} In brief, following treatment, the alginate scaffold was degraded, MCS were collected and rinsed with PBS, and were dissociated into single cell suspensions using trypsin-EDTA followed by gentle agitation. The cells were suspended in fresh DMEM and were observed microscopically to confirm the absence of any cellular clumps. The single cell suspensions were then plated in 6-well plates (3 × 10³ cells/well) in triplicate and were incubated at 37 °C in 5% CO₂. After 10 days, cells were fixed with 2% formaldehyde for 10 minutes and stained with 0.5% crystal violet solution (methanol: H₂O: acetic acid 1:2:2). After staining, the plates were washed twice with PBS and left at room temperature for 30 minutes to dry. Clusters of at least 50 cells were counted as colonies using automated counting software (OpenCFU) and the total number of colonies were reported for control and treated spheroids. The surviving fraction and plating efficiency were determined using the following equations:

$$\mathit{\text{Survival Fraction}}=\frac{\mathit{\text{Number of colonies in treated well}}}{\mathit{\text{Number of colonies in control}}}$$

$$\mathit{\text{Plating Efficiency }}(\%)=\frac{\mathit{\text{Number of colonies}}}{\mathit{\text{Number of seeded cells}}}\times 100$$

2.2.11 Statistical analysis

For all of the studies, experiments were performed in at least triplicate. Statistical analyses for all the experiments were performed with a two-tailed unpaired student’s t-test to determine any significant differences in observed data using PRISM version 5.0 software (GraphPad Software, Inc., CA). Data were expressed as mean ± standard deviation.

3. Results

3.1 Mechanical characterization of alginate scaffolds

Alginate scaffolds were fabricated and evaluated for their similarity to native lung tissue with respect to mechanical strength. The Young’s modulus of three scaffolds (7.34 ± 0.52 mm thick and 9.69 ± 0.27 mm diameter) was tested at 1 and 12 hours after equilibration in DMEM. After 1 hour, the Young’s modulus was found to be 2.24 ± 0.41 kPa, whereas after 12 hours the modulus increased to 3.73 ± 0.21 kPa (Figure S1).

3.2 Air-grown MCS formation and growth kinetics

3D Petri Dishes® were used as template to imprint 5×7 microwells in alginate hydrogels, which served as the location for cellular self-assembly (Figure 1a) with each well of the 12-well plate containing 35 spheroids. The average diameters of the spheroids were examined quantitatively. Figure 1b (left) shows the growth kinetics of spheroids at different concentrations (575, 1425, 2285, 2850, 3700 cells/spheroid) for up to 7 days of growth. Figure 1b (right) illustrates the growth kinetics of spheroids at 2850 cells/spheroid for 30 days. At all cell concentrations, the spheroid size increased with time and growth kinetics remained steady up to 30 days of growth.

3.3 Image-based evaluation of MCS growth and morphology

The change in morphology and growth of the spheroids with respect to time was observed via brightfield imaging. Figure S2 shows the formation of spheroids at different cell concentrations (575, 1425, 2285, 2850, 3700 cells/spheroid; 35 spheroids/well) through 7 days of growth. Figure 2a shows representative brightfield images obtained up to day 30 of MCS with 2850 cells/spheroid and these spheroids increased in size as a function of time. In each spheroid, cells were loose aggregates up to day 3 and became more tightly organized spheroids at day 6. At days 13 through 20, the centers of the spheroids were lighter than the remainder of the spheroid. At days 22 and 30, the spheroids exhibited a well-formed center and looser aggregates of cells were seen at the surface of the spheroid. Figure 2b shows scanning electron micrographs (SEM) of the MCS. On day 3, the cells were not completely fused and the boundary of A549 cells was distinguishable whereas on day 7 the cells were fused together, forming tight intercellular contacts within the spheroid. Figure 2c demonstrates the high throughput formation of 35 spheroids (5×7 array) in a single well of 12-well Transwell.

Figure 2.

(a) Representative brightfield images of A549 air-grown spheroids from days 1 to 30, showing the variation in the morphological features, (b) scanning electron micrographs (200× magnification) of A549 multicellular spheroids (MCS) at days 3 and 7, and (c) brightfield image of alginate scaffold containing 35 MCS (5×7 array) in a single Transwell of a 12-well plate.

3.4 Cellular morphology, viability, and proliferation of A549 spheroids

To assess the cellular morphology and viability of A549 cells within the spheroids, the spheroids were stained with H&E and Calcein AM/EthD-1 at days 3, 5, and 7 of spheroid formation. In Figure 3a, the viability of A549 cells is seen at the outer rim of the spheroid (green) and dead cells in the center (red). As seen in Figure 3b, there was a significant increase in the total DNA content and the average area of spheroids at day 7 in comparison to day 3, indicating the proliferation of A549 cells in the MCS. Figure 3c shows H&E staining of a cross section of A549 MCS at days 3 and 7, demonstrating the arrangement of cells within the MCS and the progressive change in the spheroid structure (transitioning from loosely packed cells at day 3 to compact spheroid formation by day 7).

Figure 3.

(a) A549 cell viability within multicellular spheroids (MCS) was demonstrated using Calcein AM (green) and Ethidium homodimer-1 (red) staining at days 3, 5 and 7. (b) Spheroid internal structure and morphology was evaluated using H&E staining at days 3, 5 and 7. (c) Proliferation of A549 cells within MCS was evaluated by quantifying the total DNA content and spheroid area at days 3 and 7. Data is represented as mean ± SD, n = 3. ** p < 0.005. Scale bar = 200 µm.

3.5 Drug study using A549 multicellular spheroids

To evaluate the effect of the chemotherapeutic paclitaxel (PTX) and tumor-penetrating peptide iRGD on lung MCS microenvironment, the chemosensitivity of MCS to PTX was examined. MCS were incubated with PTX (5, 10, 50 and 100 µM) and evaluated at 4, 24, and 48 hours. As seen in Figure S3, there was no significant change in the average diameters of the spheroids exposed up to 100 µM of PTX. In another set of experiments, iRGD was co-administered with PTX to sensitize air-grown MCS to PTX. Figure 4a and Figure 4b contain brightfield images of MCS at 1, 3, 5, and 7 days after treatment with PTX and PTX plus iRGD, respectively. Figure 4c shows the quantification of MCS growth kinetics following treatment with PTX and PTX plus iRGD over a period of 7 days. Figure 4d shows the rate of change of A549 MCS growth after 7 days of treatment. A549 MCS showed a significant decrease in average diameter when PTX was co-administered with iRGD as compared to PTX treatment alone. There was also a progressive decrease in the growth kinetics and growth rate of MCS when treated with 10 µM PTX plus iRGD as compared to both PTX alone and the untreated control.

Figure 4.

Study of A549 multicellular spheroid (MCS) sensitization using the chemotherapeutic drug paclitaxel (PTX) and the tumor-penetrating peptide iRGD. Sequential brightfield images of spheroids (a) after exposure with PTX, (b) co-administration of PTX and iRGD, (c) spheroid growth inhibition curve following treatments, and (d) change in the growth rate of A549 MCS after 7 days of treatment. Data is represented as mean ± SD, n = 3. ** p < 0.005, and *** p < 0.0005. Scale bar = 300 µm.

3.6 Quantitative evaluation of apoptosis in air-grown spheroids

The viability of A549 cells in spheroids after treatment with PTX and PTX plus iRGD was evaluated using the Annexin V-FITC apoptosis assay. After 7 days of exposure, the spheroids were dissociated into single cells and were treated with Annexin V-FITC and PI to differentiate between live and necrotic/dead cells.³⁷ Quantitative evaluation of apoptosis (Figure 5) indicated a significant difference in the percentage of dead cells after incubation with 10 µM PTX plus iRGD (23.4%) in comparison to 10 µM PTX (9.3%) and untreated control (0.12%).

Figure 5.

Apoptosis assay using Annexin/PI staining (a) A549 spheroids were treated with 10 µM PTX and iRGD plus 10 µM PTX and (b) quantification of viability (live cells) and apoptosis (dead cells) after treatment with 10 µM PTX and iRGD plus 10 µM PTX. Data is represented as mean ± SD, n = 3. * p<0.05, ** p < 0.005 and *** p<0.0005.

3.7 Clonogenic survival assay

A clonogenic assay was used to evaluate the survival potential of A549 cells and their potential to regrow and form colonies (>50 cells) following treatment with PTX and PTX plus iRGD. The total number of colonies formed, plating efficiency, and survival fraction was calculated (Figure 6). Figure 6a shows that the spheroids treated with PTX plus iRGD exhibited greater inhibition in reproductive potential and clonogenic ability. In addition, there was a significant reduction in number of colonies (Figure 6b), plating efficiency (Figure 6c), and surviving fractions (Figure 6d) in comparison to PTX alone and untreated control. These results indicated an increase in efficacy and sensitivity of MCS to PTX in presence of iRGD.

Figure 6.

Clonogenic capacity of A549 cells after no treatment (control), 10 µM PTX treatment and 10 µM PTX plus iRGD treatment of MCS. (a) Crystal violet staining of the reproduced colonies, (b) total number of surviving colonies formed on the culture plate 10 days after treatments, (c) plating efficiency and (d) data representing the number of colonies, plating efficiency (%), and survival fraction (%) for A549 lung spheroids exposed to PTX and PTX plus iRGD. Data is represented as percentage of colonies compared with untreated control (means ± SD, n = 3). ** p<0.005 compared to control.

4. Discussion

Adenocarcinoma, squamous cell carcinoma, and large cell carcinoma represent different types of non-small cell lung cancer (NSCLC), of which adenocarcinomas (epithelial tumors) are the most commonly diagnosed types.³⁸ In the present study, 3D air-grown MCS comprised of A549 lung adenocarcinoma cells were developed in alginate hydrogels in a high throughput model (Figure 1a). The alginate hydrogels served as a non-adherent surface as they are relatively inert with respect to cell-substrate interactions. As a result, the cell-to-cell interactions in the MCS were greater than the cell-to-substrate interactions due to the lack of mammalian receptors (integrins) in alginate, which results in low protein adsorption on the substrate.³⁹ For the mechanical compliance of the alginate substrate, hydrogels after 12 hours equilibration in media demonstrated a significant increase in Young’s modulus in comparison to gels equilibrated for 1 hour (Figure S1). This significant improvement in Young’s modulus might be due to the presence of calcium in the media, prompting stronger physical interactions between Ca⁺² and alginate chains with the progress of time and the stability of Ca⁺² ions in the gels. Previous studies have reported that if the alginate concentration is varied then the modulus also changes accordingly, where an increase in alginate results in a higher modulus.^{40, 41} Thus, A549 cells were grown over alginate hydrogels that have mechanical properties comparable to healthy lung tissue (Young’s modulus ~ 5kPa), biocompatibility, and inertness for development of air-grown multicellular spheroids.

To mimic the lung in vivo microenvironment, 3D spheroids were developed and grown at air interface. Air interface condition was induced by culturing spheroids on a non-adherent surface (alginate gel) in a Transwell, degrading the gel, and then allowing the MCS to grow in air on the apical side of the Transwell while still allowing for nutrient access through the basolateral side, which had access to cell media. The single cell suspensions seeded over alginate scaffold microwells spontaneously self-assembled to form 3D multicellular aggregates, and when a majority of the cells aggregated into a tight structure, a spheroid was successfully formed. A549 cells were successfully able to form spheroids in air-interface condition in 100% of the wells at various cell concentrations with tight and well-defined boundaries. The diameter of A549 spheroids increased with respect to time and the size of the spheroids primarily varied as a function of the initial cell-seeding density (Figure S2). However, of the cell concentrations evaluated, the diameters of the 2850 cell/spheroid concentration were highly uniform and consistent in size, which indicated favorable reproducibility of the spheroids (Figure 1b), and this was therefore selected as an optimal cell concentration for A549 MCS formation for further culture.

Air-grown A549 lung spheroids were maintained in culture for 30 days and the complete spheroid formation resulted in significant morphological changes (Figure 2). Initially (day 1–3) individual cells spontaneously self-assembled to form multicellular aggregates, in which the cells were spherical and the boundary of single cells could be easily identified. At days 4–6, the cells began to adhere to one another due to the interplay of complex intercellular interactions, which resulted in the formation of more compact 3D multicellular aggregates. At this stage the surface of the spheroids became smooth and continuous and the single cells were no longer distinguishable. It has been reported that the cell surface receptors E-cadherin and β1-integrin play a major role in spheroid formation at this step.^{42, 43} Subsequently, at days 7–20, the multicellular aggregates developed into compact solid spheroids as a result of cell-mediated contractility^{44, 45}, but the periphery of the spheroids looked disorganized, suggesting cellular proliferation at the outer rim of MCS. Eventually, by days 21–30, a concentric layer structure was present within the spheroids comprising an outer layer of proliferating cells, middle layer of viable quiescent non-proliferating cells, and necrotic core at the center of spheroid due to the hypoxic environment. The cells in the center of the MCS have less access to oxygen and nutrients so that they firstly become hypoxic and then necrotic.⁴⁶ The brightfield images of the spheroids (Figure 2) also exhibited formation of lumen in the center, which represents dead cells due to lack of direct access of the cells with media. These lumens are also seen in the H&E stained cross-sections (Figure S4) of A549 spheroids (14 days old), which corroborate with the findings of previous studies suggesting that when the epithelial cells gets detached from a matrix, they undergo programmed cell death (anoikis).^{47, 48}

Apart from studying the morphological characteristics and proliferation of A549 cells within MCS, the viability and DNA content were also determined (Figure 3). Calcein AM and EthD-1 staining allowed for the visual evaluation of viable A549 cells within the MCS. Calcein AM was used as the vital stain marker for metabolically active viable cells (green color) and EthD-1 penetrates compromised cell membranes and binds to nucleic acids indicating dead cells (red color). The live/dead staining of 3 and 5-day-old spheroids showed live and metabolically active proliferating cells within the MCS, but for 7-day-old spheroids there was formation of a necrotic core in the center of the spheroids due to restricted nutrient diffusion and waste removal, with the presence of live proliferating cells in the peripheral rim. A549 MCS were cross-sectioned and stained with H&E staining at days 3, 5 and 7. Hematoxylin stains the basophilic structure blue (cell nucleus) and eosin stains acidophilic structures (cell cytoplasm) pink.⁴⁹ On day 3, A549 spheroids were comprised of a loose collection of cells, indicating initial multicellular aggregation, and on day 5 the size of the spheroids had increased. On day 7, the spheroids clearly demonstrated an outer peripheral rim comprised of cells tightly bound to one another while the middle layer was comprised of loosely aggregated cells with significant spaces in between them. A progressive increase in the DNA content and total area of the spheroids suggests that cell proliferation occurred and also corroborated the findings of staining results.

The objective of developing the described A549 air-grown MCS model was to evaluate the therapeutic index of PTX with and without the presence of a tumor homing and penetrating peptide (iRGD). A549 spheroids were grown for 4 days, when a diameter of approximately 400 µm was reached prior to drug exposure, as the phenomenon of low drug penetration has been reported in spheroids > 400µm⁵⁰. The spheroids were exposed to PTX on the fifth day. As PTX is known to exert its therapeutic effect only on the proliferating/dividing cells, cells located near the spheroid surface responded to PTX treatment. However, the quiescent cells that were located in the intermediate and center regions of the A549 spheroids were shown to be less sensitive to PTX treatment, which might be due to both the restricted transport/limited permeability and mechanisms of drug resistance associated with the tumor microenvironment. It was observed that when PTX was co-administered with iRGD, there was a decrease in the spheroid size and growth kinetics (Figure 4), increase in apoptosis (Figure 5), and decrease in the survival potential (Figure 6) of the A549 cells. This indicates that in the presence of iRGD, PTX was able to penetrate further into the intermediate and central regions of the spheroid, which could be due to iRGD-mediated internalization and tumor penetration.^{31, 32} iRGD binds to αv-integrins molecules, which are overexpressed in cancerous tissues, using its vascular recognition (CendR) motif, where it is proteolytically cleaved and the truncated peptide (CRGDK/R) gains affinity for neuropilin-1.³¹ These changes have been reported to increase the permeability of drugs into tumors. Sugahara and co-workers demonstrated that iRGD enhances the penetration of drugs, either when coupled as cargo or when co-administered with chemotherapeutic drugs into the tumors.³²

Thus, 3D cultures such as MCS could serve as an important tool in understanding how chemotherapeutics and other therapies impact in vivo tumors before conducting expensive clinical studies, which would make the drug screening process much more efficient and successful. Notably, the use of iRGD along with the drug could be a promising approach to increase the amount of accessible drug inside cancerous tissue and prevent recurrence. In conclusion, the development of in vitro A549 lung MCS provided a new representation of in vivo epithelial tumors and the bystander activity of iRGD enhanced the penetration of PTX into A549 MCS, thereby increasing the therapeutic efficacy of PTX.

5. Conclusions

In the present work, we demonstrated the high-throughput formation of A549 spheroids grown at the air-interface using micromolds from 3D Petri Dishes^®. Air-grown 3D lung spheroids were successfully developed and these tissues demonstrated a progressive increase in size over 30 days of continuous culture. The cytotoxic effect of PTX in A549 MCS was found to be restricted only to the outer layer of proliferating cells, which indicated limited drug diffusion and penetration into 3D spheroids. However, the depth of PTX diffusion was enhanced via co-administration with the tumor-penetrating peptide iRGD. The gradual decrease in the growth of A549 MCS, increase in the apoptosis, and decrease in the clonogenic reproducibility of A549 cells indicated enhanced sensitization of spheroids to PTX in the presence of iRGD. Thus, air-grown 3D MCS could be an important tool to understand the mechanism of resistance in lung cancer and co-administration of iRGD along with chemotherapeutic drugs could be a potential approach with substantial improvement for preclinical drug development for lung cancer treatment.

Supplementary Material

Supp FigS1-4

Figure S1. Mechanical strength evaluation of 1% alginate scaffolds after equilibration in DMEM for 1 and 12 hours. Data is represented as mean ± SD, n = 3.

Figure S2. Effect of the starting cell number on A549 multicellular spheroid (MCS) formation. Spheroids were developed from 575, 1425, 2285, 2850, 3700 cells/spheroid seeding densities and were cultured for 7 days. Brightfield images were taken each day through day 7. Scale bar = 500 µm.

Figure S3. Drug study on A549 multicellular spheroid (MCS). Spheroids were cultured for 4 days and on the fifth day were treated with 5, 10, 50 and 100 µM of PTX. (i) Sequential brightfield images of A549 MCS at 4, 24 and 48 h for all the treatments. (ii) Growth analysis following treatment and plotted as the diameter (µm) versus time. Scale bar = 300 µm.

Figure S4. H&E stained cross-sections of A549 spheroids (14 days old) demonstrating formation of lumen in the center of spheroids representing dead cells due to lack of direct contact between A549 cells and the alginate matrix.

Novelty and Impact.

The simulation of in vivo microenvironmental features present in lung tumors is important to provide improved translation from in vitro drug screening to in vivo animal models. Here, the authors have developed a novel high-throughput lung adenocarcinoma multicellular 3D spheroid assay, with cells grown at air-interface, which were utilized to evaluate the effect of anticancer therapeutics on lung tumors. These findings emphasize the potential for the co-administration of the tumor penetrating peptide (iRGD) and anticancer drugs as a new therapeutic approach for lung cancer treatment.