Pancreatic cancer cell line in responsive hydrogel microcapsules for drug evaluation

Drug therapies are the cornerstone of systemic treatment for pancreatic cancer patients. However, the relative outcome of drug evaluation is often hampered by the complex microenvironment of pancreatic cancer due to the lack of reasonable tumor models. Here, we proposed a novel platform that integrated pancreatic adenocarcinoma cells encapsulated into hydrogel microcapsules for three‐dimensional (3D) tumor cultivation and antitumor agent evaluation. These hydrogel microcapsules contain alginate/poly (N‐isopropyl acrylamide) (alginate/PNIPAM) shells and carboxymethyl cellulose cores, which are generated through the microfluidic electrospray technique. The microcapsules have the feature of rapid response to temperature, by which they can regulate the internal pressure environment. Besides, benefiting from good monodispersity, precise size control, and biocompatibility of these microcapsules, these wrapped tumor cells have the capacity for proliferating spontaneously and forming 3D tumor spheroids with good cell viability. We have demonstrated that pancreatic adenocarcinoma cells encapsulated in the composite microcapsules with different PNIPAM concentrations showed different drug sensitivity, which could be ascribed to the influence of external pressures environment. These results indicate that the tumor spheroids coated in these responsive microcapsules have great potential in the analysis of antitumor drug sensitivity.


INTRODUCTION
Pancreatic cancer with high malignancy and poor prognosis is one of the most lethal tumors in the digestive system. 1 In clinics, drug therapy plays an irreplaceable role in controlling pancreatic cancer progression and prolonging patients' life. 2 Unfortunately, their medication efficacy is hampered by the complex microenvironments of tumors among individuals. 2,3Thus, the in vitro drug screening platform for clinical medication guidance is pretty significant.Currently, multifarious two-dimensional (2D) and 3D cell culture systems have emerged vigorously for drug evaluation. 4Despite some positive effects, the 2D cell culture system shows poor similarity to in vivo tissue, preventing them from revealing real effects in bodies. 5Besides, the 3D culture of pancreatic cancer cells can mimic the characteristics of pancreatic cancer in vivo to some extent. 6However, these 3D tumor models can only simulate the appearance of pancreatic cancer while failing to reproduce the complex tumor microenvironment. 7The local microenvironment has been demonstrated to strongly influence the biological behavior of pancreatic cancer, such as mechanical factors based on extensive dense fibrous stroma among pancreatic cancer cells.5a,8 Therefore, a novel drug evaluation platform with adjustable mechanical properties for pancreatic tumors is still anticipated.Herein, we design a type of tumor cell-encapsulated microfluidic hydrogel microcapsule with mechanical regulatory function for antitumor drug evaluation, as schemed in Figure 1.Microfluidics techniques have the ability to modulate fluid on a submillimeter scale precisely in micro-or nano-scale space, which miniaturizes basic laboratory functions such as sample preparation, reaction, and detection to a few square centimeters on a chip. 9With this feature, the microfluidics technique has been widely employed to generate uniform microparticles and microcapsules. 10Especially, microfluidic electrospray microcapsules with adjustable properties have been developed for tumor cell encapsulation and 3D cultivation, which contributed to in vitro drug evaluation. 11However, these microcapsule-based screening systems exhibited non-responsiveness to mechanical extrusion similar to the actual tumor microenvironment.By contrast, poly(Nisopropyl acrylamide) (PNIPAM) is a kind of hydrogel with a remarkable thermal response, which implements a size reduction at the critical temperature (32 • C). 12 This feature allows the PNIPAM hydrogel to increase pressure on the substance within the hydrogel when it is heated.Nevertheless, the combination of drug-screening microcapsule and hydrogel with mechanical regulation capability has seldom been studied.
In this work, we constructed a composite hydrogel microcapsule with temperature responsiveness for pancreatic cancer cell encapsulation and biomimetic drug evaluation.The generated microcapsules comprised carboxymethyl cellulose (CMC) cores containing pancreatic tumor spheroids and alginate (ALG)/PNIPAM shells.Both ALG and CMC exhibit favorable biocompatibility and accessibility.Furthermore, the crosslinking mechanism involving ions for sodium alginate is characterized by its simplicity and rapidity, while also causing minimal harm to the encapsulated cells.Therefore, the composite hydrogels exhibited low toxicity, high stability, and precise mechanical regulation.In view of its excellent properties, the wrapped pancreatic tumor cells could proliferate rapidly and form 3D tumor spheroids with good cell viability and homogeneous size distribution.It was demonstrated that these hydrogel microcapsules could respond to temperature variation rapidly, thereby regulating the internal pressure environment for the encapsulated cells.Notably, the pancreatic tumor spheroids within the microcapsules exhibited significant heterogeneity in drug sensitivities to relative chemotherapeutic agents when they were exposed to different external pressure environments.These experiments revealed that integrating the 3D cancer models and hydrogel microcapsules can provide a reliable platform for simulating pancreatic cancer microenvironment and guiding clinical drug treatment.

RESULTS AND DISCUSSION
In the present experiment, the thermo-responsive behaviors of the shell hydrogels (ALG/PNIPAM) were investigated and optimized first.The heart-shaped hydrogel film doped with dark substances was fabricated to observe the shrinkage effect intuitively.As elaborated in Figure 2 and Figure S1, once the film was put onto the heating table with the temperature set as 37 • C, they shrank strikingly, accompanied by the water displacement.The rapid size change of the mixed hydrogels indicated that they had good temperature responsiveness.Besides, the shrinkage efficiency of these composite hydrogels is impaired with the increasing concentration of either PNIPAM or ALG.These results could be attributed to the reason that the hydrophilic ionically crosslinked ALG would not be affected as temperature continuously increased, and therefore, It weakens the hydrophobicity of the hydrogel mixture at high temperatures.
The one-step preparation of the microcapsules was realized by photopolymerizing the composite droplets, which were generated from a simple microfluidic capillary device with a concentric structure using the microfluidic electrospray technique.Briefly, the coaxial fluids of CMC encasing pancreatic cancer cells (core flow) and ALG/PNIPAM (1 wt%/5 wt%) with photoinitiator (shell flow) were pumped into the corresponding fluid microchannels.With the help of a high-voltage power supply, the continuous fluids formed a "Taylor cone" at the end of the chip and were then cut into core-shell droplets.These droplets reacted with calcium chloride in the collection pool, followed by the irradiation of ultraviolet (UV) light to form the stable ALG/PNIPAM shells.The fast-cross-linking property of the hydrogel ensured the encapsulation of the CMC inside the microcapsules to form uniform core-shell microcapsules, as indicated in Figure 3A-C.It was worth noting that there were many key parameters related to the volume and morphology of the microcapsules.When the fluid rate of the inner phase was kept constant and the fluid rate of the outer phase was elevated from 30 to 55 μL/min, the overall diameter of these microcapsules increased (Figure S2A).Similarly, when the outer flow rate remained unchanged, the CMC solution flow rate rose from 1 to 6 μL/min leading to a marked increase in the diameters of microcapsules, as recorded in Figure S2B.Moreover, the size of the microcapsules could be regulated by the electric field (voltage) and the distance between this microfluidic device outlet and its below CaCl 2 solution.It was found that the larger voltage and smaller distance would result in smaller microcapsules, as described in Figure S2C,D.Additionally, the monodispersity of the microcapsules was also studied.As shown in Figure 3D, benefiting from the precise manipulation of the microfluidic system, the prepared microgels exhibited good monodispersity.Interestingly, the diameter of shell thickness and core of the microcapsules were uniform distribution, indicating that the microfluidic electrospray technology was a reliable strategy for generating microcapsules (Figure 3E,F).The scanning electron microscope (SEM) was also employed to characterize these core/shell microcapsules.As the SEM images are shown in Figure S3, the microcapsule exhibited numerous large pores on the shell and a huge hollow within the shell, confirming the successful core-shell structure of the microcapsules and the possibility of mass transfer for cell culture.
To evaluate the contractile capacity of these composite microcapsules under temperature change, we employed a hot stage with constant temperature and recorded the temperature-responsive behavior of the microcapsules through an optical microscope.As shown in Figure 4Ai,ii, the overall size of the ALG/PNIPAM shell contracted in the range of 1-4 min, and the size reduced by 35% after 5 min heating (Figure 4B,C and Figure S4).When the heating procedure was turned off, these microcapsules completely returned to their initial morphology and color (Figure 4Aiii).Additionally, these microcapsules could recover to their initial volume after 10 cycles, demonstrating the good thermal stability and repeatability of microcapsules (Figure S5).The influence of concentration gradation of PNIPAM on the respon-siveness of microcapsules was also defined.As shown in Figure 4D, all the microcapsules containing PNIPAM exhibited thermo-triggered deformability, and their volume shrinkage ratio decreased along with the increasing concentration of PNIPAM, which was consistent with the above results.A typical compressive stress-strain curve for different ALG/PNIPAM hydrogels was presented.The 5wt% PNIPAM/ALG hydrogel had a compression strength of 0.12 MPa at the maximum compressive strain of 50%.We also obtained the tensile stress-strain curve of PNIPAM/ALG hydrogels with different proportions.The fracture toughness of 5wt% PNIPAM/ALG hydrogel was lower than that of the others, as shown in Figure S6.Meanwhile, no significant thermo-responsiveness was found in the hydrogel microcapsules made without PNIPAM, as shown in Figure S7.
To ascertain the potential cytotoxicity of the PNI-PAM/ALG composite hydrogel as the shell structure of microcapsules on nuclear layer cells, we first conducted an experiment wherein hydrogels of varying concentrations of PNIPAM were introduced to 2D cultured tumor cells (Patu8902 cells, the pancreatic cancer cell line).Our findings revealed a positive correlation between increasing PNIPAM concentrations and enhanced cytotoxicity, with 5wt% PNIPAM demonstrating the most optimal outcome (Figure S8).To investigate the biological activity of using the composite microcapsules as 3D cell culture scaffolds, Patu8902 cells were encapsulated into the microcapsules and observed for 9 days.On the first day, it was found that the tumor cells began to aggregate within 24 h.In the subsequent few days, the cells exhibited good viability and rapid growth and gradually grew into a well-defined spheroid.Both Calcein AM/propidium iodide (PI) fluorescence imaging and quantitative analysis revealed a rapid cell proliferation process and excellent cell viability during the whole cultivation, confirming the feasibility of the culture system in pancreatic cancer cell 3D culture, as shown in Figure 5A-D.Besides, to investigate the influence of UV illumination on the live/dead ratio of encapsulated tumor cells, Calcein AM/PI kit was employed, and the fluorescence intensity was quantified.As demonstrated in Figure 5E, the cell viability was negatively correlated to UV exposure time, indicating that prolonged UV irradiation could lead to cell death.On the basis of this result, we chose 20 s as the UV exposure time, at which time 80% of cells maintained viability.In order to ascertain the efficacy of shell contraction in microcapsules for facilitating pressure conduction within the internal tumorsphere, a digital simulation based on the Thermal Strain and elastic mechanics formula was conducted.The simulation results revealed that the microcapsule shell, through a volume phase change, consistently conducted pressure from the external to the internal environment, thereby establishing a stable pressure environment for the polymer layer of the tumor spheroid, as shown in Figure S9.The tumor-relevant biological molecules were analyzed by immunofluorescence to evaluate the neoplasia.CD44 is often used as the marker of cancer stem cells in pancreatic cancer.Meanwhile, as a member of the mucin family, MUC1 is often defined as a special marker for the diagnosis of pancreatic adenocarcinoma.As indicated in Figure S10A, there were a significant number of CD44/MUC1 positive pancreatic cancer cells within the spheroids.Furthermore, the pancreatic ductal epithelial marker expression of CYFRA21-1 (CK19) and ki67 (proliferation marker) indicated the effective proliferation of pancreatic tumor cells.These results confirmed that the Patu8902 cells encapsulated in the composite microcapsules had characteristics similar to that of the orthotopic tumor, revealing their potential as an in vitro model for drug evaluation (Figure S10B).Prior research has demonstrated that the robust extracellular matrix of pancreatic cancer can influence the expression of genes associated with internal tumor cells, thereby augmenting the stem cell characteristics of tumor cells and facilitating the epithelial-mesenchymal transition of tumor cells. 13hese processes collectively contribute to the development of drug resistance in tumor cells to some degree.In our study, we conducted gene expression correlation analysis on tumor spheroids and 2D culture within PNIPAM microcapsules.Our findings reveal that, in comparison to two-dimensional culture, the expression of genes linked to stem cell properties and mesenchymal properties was significantly up-regulated, as shown in Figure S11A,B.
To determine the drug sensitivity of Patu8902 cells encapsulated in these microfluidic composite microcapsules, we prepared pancreatic tumor spheroidsencapsulated microcapsules without PNIPAM hydrogel (Figure S12).The tumor spheroids were separately treated with 5-fluorouracil (5-FU), gemcitabine (Gem), and oxaliplatin (Oxa), which was consistent with the clinical drug regimens.It was found that the cell viability of the tumor cells coated in the microcapsules decreased with the increased drug concentrations and prolonged time, as shown in Figure 6A-C.This result proved that all of these drugs were able to induce tumor cell apoptosis in a dose and time-dependent pattern effectively.After that, the influence of the external pressure environment on drug sensitivity was evaluated.As a result, the tumor spheroids encapsulated in the microcapsules with ALG/PNIPAM shells exhibited different sensitivities to the drugs when treated with the same procedures, as demonstrated in Figure 6D-F.Such a consequence might be ascribed to the tighter cell junction triggered by external squeezing, which limits the rapid penetration of chemotherapy drugs.These results revealed a great potential of composite hydrogel microcapsules with thermo-responsiveness for actual tumor environment mimicking and more precise drug screening.

CONCLUSION
In total, we have constructed composite hydrogel scaffolds with ALG/PINPAM shells and CMC cores through the microfluidic encapsulation technique for tumor spheroids formation and drug screening.The resultant microcapsules had well-defined 3D morphologies and good uniformity.The size could be precisely manipulated by adjusting the core and shell flow rates, the voltage, and the distance.
It has been demonstrated that the cells within the microcapsules could spontaneously form 3D tumor masses with a homogeneous distribution of morphology and excellent viabilities.Besides, the composite microcapsules with PNIPAM shells showed excellent volume-shrinkage ability to produce mechanical squeezing, which could mimic actual tumor environments.Moreover, the Patu8902 cells spheroids encapsulated in composite hydrogel scaffolds with PNIPAM shells possessed heterogeneous sensitivities to the same chemotherapy drug compared with ordinary hydrogel microcapsules.On the basis of these results, we believe that constructing microcapsules with mechanical extrusion responsiveness will provide new insights into the design of precise drug screening platforms.
The 96-well polystyrene plates were purchased from COR-ING.The CellTiter-Glo kits were obtained from Promega.Both the primary antibodies and second antibodies were bought by Servicebio (Wuhan).

Mechanical property test
The mechanical properties of the PNIPAM/ALG hydrogel were assessed using a Universal Testing Machine.
The hydrogel samples had a diameter of 6 mm and a height of 4 mm.The compression ratio employed was 2 mm/min.Additionally, the tensile properties were determined through a uniaxial stretching test.The crosshead speed was set at 10 mm/min, and the specimen was securely held between clamps with a length of H = 40 mm.The specimen had a thickness of 1 mm and a width of 18 mm.To evaluate the cyclic tensile behavior, various predetermined maximum strains were applied.

Fabrication of composite microcapsules
Unless otherwise specified, the outer (shell) flow was composed of 1 wt% ALG, PNIPAM (0, 5wt%), and 1 vol% HMPP (2-hydroxy-2-methylpropiophenone), 0.2 wt% BIS, and the inner (core) flow was composed of 0.5 wt% CMC, respectively, which were dissolved in sterile water.The outer/inner phase flows were introduced into the microfluidic device constructed with three parts (including circular structure fluid inlets, syringe pumps, and a microcapsule fabrication unit).Briefly, the transverse diameter of the inner/outer capillary was approximately 100 μm and 200 μm at the rate of 4, and 40 μL/min.The microfluidic capillary and the collecting bath (2% CaCl 2 ) were connected to positive and negative voltage power supply.Then, microcapsules were produced through sheared and cross-linked with CaCl2 in the collecting bath.Finally, these microcapsules were gelled after irradiating with a UV lamp.

Patu8902 cell spheroids culture
The Patu8902 cells harvested from a culture plate were prepared into single-cell suspensions and passaged with Trypsin-EDTA (0.25%) once they reached a confluency of 75%-80%.The Patu8902 cells suspension mixed with CMC were encapsulated in the composite microcapsules and cultured under the complete culture medium composed of Dulbecco's Modified Eagle Medium (high-glucose), 10 vol% fetal bovine serum and 1 X penicillin/streptomycin were rested under a thermostat-controlled incubator (37 • C, 5% CO 2 ).

Patu8902 cells spheroids viability
The viability of Patu8902 cells spheroids was evaluated through the Calcein AM/PI kit in the hydrogel scaffolds after 1, 3, 5, and 9 days.After washing the cell-laden microcapsules three times with sterile saline, the samples were stained with the Calcein AM/PI solution (1:1000 DMEM) and incubated in darkness at 37 • C for 45 mins.Then, the viable and dead cells were stained green and red, respectively.For quantifying the proportion of live cells, encapsulated tumor spheroids were transferred to 96-well plates, and a 50-vol% CellTiter-Glo kit was injected into the complete medium and incubated at 37 • C for 30 min.Then, the absorbance of relative plates was performed at a multifunctional microplate reader (luminescence; Intergration: 500).

Quantitative polymerase chain reaction
It was conducted to analyze the expression levels of CD44, CD133, ALDH2, SOX2, OCT4, E-cadherin, Vimentin, Snail, Cxcr4, and GAPDH in tumor spheroids encapsulated under composite capsules and 2D cells.Total RNAs were extracted from the samples using the miRVana miRNA isolation kit (Thermo Fisher Scientific), and cDNAs were synthesized using the high-capacity cDNA reverse transcription Kit (Thermo Fisher Scientific).The TaqMan gene expression assay (Thermo Fisher Scientific) with the THUNDERBIRD Probe qPCR Mix (TOYOBO) was utilized for the analysis, with normalization to 18S rRNA.The CFX96 Real-Time System was employed for quantitative analyses.All experiments were carried out in triplicate.

Drug evaluation
The working solution of gemcitabine, 5-FU, and oxaliplatin were freshly prepared, and these chemotherapeutic agents were supplemented to the complete medium at a stable concentration (Gem: 1 μM, 5-FU: 100 μM, and Oxa: 100 μM).Tumor spheroids were incubated in these cell media for 2, 8, 16, and 24 h.After 2, 8, 16, and 24 h of treatment with the agents, the cell viability of tumor spheroids was evaluated by a CellTiter-Glo kit (the same as above).0.1% dimethyl sulfoxide (DMSO) was used as a control to treat tumor spheroids.

Characteristics
These composite microstructures of microcapsules were characterized by SEM (JSM-IT200).These composite microstructures of microcapsules were lyophilized and gold-plated, then observed by SEM.

Statistical analyses
The experimental data are presented as mean ± standard deviation (M±SD).Relative statistical differences were analyzed by the student's t-test, and the p-value was described as :* < 0.

C O N F L I C T O F I N T E R E S T S TAT E M E N T
The authors declare no conflict of interest.

R E F E R E N C E S
The schematic of the high-pressure environment that pancreatic cancer cells faced in vivo.B Schematic of generating hydrogel microcapsules mimicking tumor microenvironment and drug reaction by microfluidic electrospray method.F I G U R E 2 A The thermo-responsive property of these heart-shaped biological gels on a heating stage with the temperature set to 37 • C. Scale bar: 500 μm.B The volume reduction of the biological gels at varying poly(N-isopropyl acrylamide) (PNIPAM) concentrations (n = 3 for each group).

F I G U R E 3 A
Schematic diagram of fabricating composite microcapsules through microfluidic electrospray technology and ultraviolet (UV) irradiation.B The real-time process of generating droplets under proper flow rates of the core and shell fluid.C Optical images of composite hydrogel microcapsules.D-F Size distributions of the microcapsules' overall diameter D, shell thickness E, and core diameter F. 100 microcapsules were measured for each graph.The scale bar is 500 μm in B and C. The droplets were first ion-crosslinked and then exposed to ultraviolet light for 20 s.F I G U R E 4 A The composite hydrogel microcapsules under room temperature (i), after heating at 37 • C for 5 min (ii), and recovered to retention time (RT) (iii).B The volume change of the poly(N-isopropyl acrylamide) (PNIPAM) microcapsules at 37 • C from 0 to 5 min.C The size distribution of composite microcapsules after heating.D The volume shrinkage ratio of the composite microcapsules with different concentrations of PNIPAM.HT represents heating time.The scale bar is 400 μm.

F
I G U R E 5 A Light microscopy images of Patu8902 cells after being encapsulated in poly(N-isopropyl acrylamide) (PNIPAM) capsules for 1, 3, 5, and 9 days.The scale bar is 100 μm.B Live/dead fluorescence imaging of tumor spheroids in microcapsules taken by laser scanning confocal microscope at days 1, 3, 5, and 9.The scale bar is 100 μm.C Cell viability of tumor spheroids embedded in PNIPAM capsules was monitored for 9 days.D Amplification efficiency of Patu8902 cells taken by CellTiter-Glo kit at days 1, 3, 5, 7, and 9. E Patu8902 cell viability under different UV irradiation times.

F
I G U R E 6 A-C Cell viability of encapsulated Patu8902 cells spheroids without poly(N-isopropyl acrylamide) (PNIPAM) and treated with 5-fluorouracil (5-FU) A, gemcitabine (Gem) B, and oxaliplatin (Oxa) C. D-F Cell viability of pancreatic tumor spheroids encapsulated in the microcapsules with PNIPAM and treated with 5-FU A, Gem B, and Oxa C.
05, **< 0.01, ***< 0.001, and not significant (NS, p > 0.05).Sample sizes (n) were shown in relative figure legends.A U T H O R C O N T R I B U T I O N S T.Y.S. conceived the idea, designed and carried out the experiments.T.Y.S., H.Z., and H.W. analyzed the data and wrote the paper.G.L.L. and Y.D.Q.contributed to the scientific discussion of the article.A C K N O W L E D G M E N T S This work was supported by the National Natural Science Foundation of China (22302231 and 82303960), Guangdong Basic and Applied Basic Research Foundation (2023A1515011986), Shenzhen Fundamental Research Program (JCYJ20190808120405672), Futian Healthcare Research Project (FTWS2022013 and FTWS2023080), and the Fundamental Research Funds for the Central Universities, Sun Yat-sen University (23qnpy153).