Nano-fibre Integrated Microcapsules: A Nano-in-Micro Platform for 3D Cell Culture

Nano-in-micro (NIM) system is a promising approach to enhance the performance of devices for a wide range of applications in disease treatment and tissue regeneration. In this study, polymeric nanofibre-integrated alginate (PNA) hydrogel microcapsules were designed using NIM technology. Various ratios of cryo-ground poly (lactide-co-glycolide) (PLGA) nanofibres (CPN) were incorporated into PNA hydrogel microcapsule. Electrostatic encapsulation method was used to incorporate living cells into the PNA microcapsules (~500 µm diameter). Human liver carcinoma cells, HepG2, were encapsulated into the microcapsules and their physio-chemical properties were studied. Morphology, stability, and chemical composition of the PNA microcapsules were analysed by light microscopy, fluorescent microscopy, scanning electron microscopy (SEM), Fourier-Transform Infrared spectroscopy (FTIR), and thermogravimetric analysis (TGA). The incorporation of CPN caused no significant changes in the morphology, size, and chemical structure of PNA microcapsules in cell culture media. Among four PNA microcapsule products (PNA-0, PNA-10, PNA-30, and PNA-50 with size 489 ± 31 µm, 480 ± 40 µm, 473 ± 51 µm and 464 ± 35 µm, respectively), PNA-10 showed overall suitability for HepG2 growth with high cellular metabolic activity, indicating that the 3D PNA-10 microcapsule could be suitable to maintain better vitality and liver-specific metabolic functions. Overall, this novel design of PNA microcapsule and the one-step method of cell encapsulation can be a versatile 3D NIM system for spontaneous generation of organoids with in vivo like tissue architectures, and the system can be useful for numerous biomedical applications, especially for liver tissue engineering, cell preservation, and drug toxicity study.

Preparation and characterization of PNA hydrogel microcapsules. Several electrospray parameters (e.g. voltage, distance to the collector, solution viscosity, and flow rate) were optimized to obtain reproducible PNA microcapsules with 10, 30, 50, 75 and 100 wt.% CPN particles. At 75 and 100% CPN particle concentrations, the spray solution was clogged into a needle which resulted in the random shape and poor quality of PNA microcapsules, hence were excluded for further experimental study. Figure 3 (top row) shows optical microscopic images of PNA microcapsules containing 0, 10, 30 and 50% (w/w) of CPN particles symbolled as PNA-0, PNA-10, PNA-30, and PNA-50, respectively. Microscopy images clearly indicated that PNA-0 microcapsules had opaque background and was changed into light dark background with increasing percentage of CPN particles. The external morphology of all PNA microcapsules was smooth and spherical. PNA-0 microcapsules were used as control microcapsules for comparative study. Figure 3 (bottom row) represents the size distribution of PNA microcapsules in the range of 400 to 600 µm. The average diameter was slightly decreased with increasing percentage of CPN particles and measured as 489 ± 31 µm, 480 ± 40 µm, 473 ± 51 µm and 464 ± 35 µm for PNA-0, PNA-10, PNA-30, and PNA-50, respectively.
The chemical structure and degradation behaviour of PNA microcapsules were analysed by FT-IR and TGA study, respectively (Fig. S2). The FT-IR spectra show that there were no changes in the position of absorption peaks for alginate hydrogel and CPN. CPN loaded microcapsules showed the peaks resulting from the superimposition of their separated components in the infrared spectra. All peaks found in CPN particles and alginate hydrogel alone were also present in the microcapsules containing CPN. PNA microcapsules stability was studied in cell culture media (DMEM with 10% FBS and 1% Pen-strip) by placing the PNA microcapsules on a rocker at 37 °C. There was no substantial change in morphology of all PNA microcapsules products up to 28 days (results not shown).
HepG2 cells encapsulation in 3D PNA microcapsules. The optical images of 3D PNA/HepG2 microcapsules produced in this study are shown in Fig. 4 (top row). 3D PNA-0/HepG2 microcapsules were used as control microcapsules for comparative study. The size of 3D PNA/HepG2 microcapsules was found in the range of 400-600 µm with spherical morphology and porous nature. The respective size distribution of 3D PNA/HepG2 microcapsules is shown in Fig. 4 (bottom row). The size of 3D PNA/HepG2 microcapsules was slightly bigger than control microcapsules. Among 3D PNA/HepG2 microcapsules, the average diameter was decreased with an increasing percentage of CPN particles and measured as 500 ± 52 µm, 490 ± 50 µm, 482 ± 58 µm and 475 ± 49 µm for PNA-0, PNA-10, PNA-30, and PNA-50, respectively. All cells were well distributed throughout the microcapsules with almost no cells on the surface of 3D PNA microcapsules.
Fluorescence imaging of 3D PNA/Hepg2 microcapsules. The 3D PNA/HepG2 microcapsules were stained with acridine orange propidium iodide (AOPI) dye, where the live and dead cells are stained as green and red, respectively (Fig. 5). The fluorescent images support the uniform cell distribution and morphological stability of 3D PNA/HepG2 microcapsules up to 14 days of culture. In all microcapsules the HepG2 grew with the formation of some clonal aggregates, and growth did not appear to be affected by the fibres. The external morphology  of all the microcapsules was smooth and spherical with no visible difference in morphology over 14 days of culture observation. The spherical morphology and porous nature of the microcapsules were observed by the SEM (Fig. S4) and distribution of cells within the microcapsules was confirmed. However, the PNA-10/HepG2 microcapsules showed better viability (p < 0.05) compared to other compositions as they had a fewer number of dead cells (i.e. stained red) (Fig. S3).
Cell viability of 3D PNA/HepG2 microcapsules. Viability of HepG2 cells in all four types of 3D PNA/ HepG2 microcapsules i.e PNA-0, PNA-10, PNA-30, and PNA-50 was measured by trypan blue and LDH assay, and AOPI dye staining. The effect of CPN on the growth of encapsulated cells within the microcapsules was also studied by counting the number of cells per microcapsule at different time points. HepG2 cell growth was increased with increasing culture time (Fig. 6A). The cell growth in all CPN containing microcapsules (PNA-10, PNA-30, and PNA-50) was higher compared to control microcapsule (PNA-0). Significant effects of CPN on cell growth was noticed at day 14, and 10% CPN microcapsule showed the highest rate of cell growth (p < 0.01) compared to control microcapsule (Fig. 6A). Hence, subsequent studies mainly compared between PNA-0 and  www.nature.com/scientificreports www.nature.com/scientificreports/ PNA-10 microcapsules. The viability of HepG2 cells at day 14 was 80 ± 9% and 85 ± 8% for PNA-0 and PNA-10 microcapsules, respectively. The PNA-10 microcapsules showed substantial higher viability at days 7 and 14 compared to PNA-0, (Fig. 6B). A similar pattern was observed in LDH levels ( Fig. 6C) supporting the viability data (Fig. 6B).
Metabolic activity of 3D PNA/HepG2 microcapsules. The level of urea production of 3D PNA/HepG2 microcapsules at day 14 is shown in Fig. 7A and found significantly (p < 0.05) higher in PNA-10 compared to PNA-0 microcapsules. In PNA-0/HepG2 microcapsules, the urea production level was decreased over time and found lower at day 14 compared to day 1 (data not shown), suggesting that ammonia detoxification may not be possible in 3D PNA-0/HepG2 encapsulates for long term culture. CYP1A1 enzyme activity was determined using luminescent CYP450-specific substrate after induction with omeprazole. The induced enzyme activity was increased, approximately 9-folds in PNA-10 compared to that of untreated microcapsules, Fig. 7B. These results indicate healthy maintenance of HepG2 cells within the PNA microcapsules. The fibrous structures in PNA microcapsules can allow encapsulated cells to grow well in terms of increasing cell population as well as enhancing cellular functions 60 . 3D PNA-10/primary hepatocytes microcapsules. Figure 8 represents 3D cultivation of primary hepatocytes (culture conditions: dynamic vs. static) in PNA-10 microcapsules. Fluorescent images show qualitative cell viability where cells were stained with AOPI dye, Fig. 8A. Green and red indicates live and dead cells, respectively. MTT viability was normalized to that of control microcapsule on day 1. Both static and dynamic cell culture condition, cell viability was decreased significantly at day 5 compared to day 1. However, the train of dramatic dcreasing of cell viability over the culture days was found in static condition. Our results indicate that the cell viability of primary rat hepatocytes in PNA-10 microcapsule favor dynamic culture conditions.

Discussion
In this study, we report a novel strategy to design PNA hydrogel microcapsules using NIM technology and one step method of HepG2 cell encapsulation. Alginate hydrogel microcapsules (AHM) meet several requirements for cell encapsulation. However, one major limitation of AHM is the lack of enough physical and chemical cues at the core of microsphere that impacts the long-term cell growth and development 39,61 . With the aim to enhance physio-chemical properties of AHM for long term cell culture and to build a strategy towards future clinical translations of AHM, we synthesized PNA microcapsule platforms. Prior to PNA microcapsule preparation, PLGA nanofibre mesh was prepared using electrospinning technique. Polymer concentration and types of solvents used to dissolve PLGA are known major game-changing parameters to control the morphology of electrospun fibre 62 . We optimized 20 wt.% PLGA solution (prepared in Chloroform: DMF; 80:20 as solvent) to get a uniform and defect-free nanofibre mesh ( Fig. 2A). An air-dried PLGA mesh was used to obtain CPN particles by using grinding mechanism of chattering, chipping, and erosion of material with applied shear forces under liquid nitrogen. Material hardness and temperature of the grinding condition determine the quality of cryo-grinding particles 63,64 . SEM analysis of the CPN particles indicates that the cryogrinding process was successful to convert electrospun nanofibre mesh into intact short fibrous particles at the size of 10-60 µm (Fig. 2B). www.nature.com/scientificreports www.nature.com/scientificreports/ Next, the CPN particles were loaded into alginate solution to make PNA microcapsules (400 to 600 µm) with spherical morphology. In this stage, our main aim was to investigate suitability of electrospray technique as a method for preparing ionotropically crosslinked alginate microcapsules containing CPN particles under the conditions that are compatible for cell encapsulation [65][66][67][68] . Our material fabrication method indicates that the CPN particles were precisely encapsulated within the PNA microcapsules without any loss of spherical surface morphology up to 28 days of incubation in the cell culture media. This result confirms that PNA microcapsules have good mechanical stability in the cell culture media. The mechanical stability of microcapsules is a major concern in the design of cell immobilization for a therapeutic delivery device, where microcapsule stability is needed to prolong in vivo functions 69,70 . Efforts have been given into increasing the mechanical strength of AHM either by adding multilayers of oppositely charged polymer coating 71,72 or by covalently crosslinking with chemical agents such as glutaraldehyde 73,74 . Both coating or covalent crosslinking techniques require either multi-step process or introduce toxic crosslinking agents into the microcapsules (e.g. glutaraldehyde) which can complicate the encapsulation process. Therefore, the present method of PNA microcapsules preparation provides a simple and straight forward approach to enhance the mechanical strength of microcapsules compared to previously reported approaches.
FTIR results suggest that the addition of CPN in PNA microcapsules does not significantly alter the chemical structure of AHM. The PNA microcapsules containing CPN (10, 30 and 50%) showed greater mass loss and degradation behavior compared to the PNA microcapsules with no CPN i.e. PNA-0 microcapsules indicating that the presence of CPN in AHM play a significant role in altering the thermal degradation pattern of the PNA microcapsules according to loaded concentration 75 . Similarly, the plot from TGA showed the difference in mass loss and thermal stability of the PNA microcapsules (Fig. S2).
The next objective of this study was to encapsulate the HepG2 cells into PNA microcapsules by optimizing electrostatic encapsulation conditions. The working hypothesis was to test whether the PNA provides additional ECM mimicking microstructural and mechanical supports inside the microcapsules. We expected that CPN particles being short nanofibres and spheres being in 3D environments will have the ability to provide cellular contact guidance to facilitate cell growth and development. Although the optimum size of AHM for cell microencapsulation is still debatable, microcapsule of 500 µm diameter has been found to be a good compromise in many studies 66,76,77 . Therefore, in the present study, we optimized our electrostatic encapsulation conditions,  mainly flow rate, voltage, distance from tip to the gelling solution, and concentration of CaCl 2 , to produce smooth microcapsules with average diameter 500 µm (Fig. 4). Production of highly controlled monodisperse 3D PNA/ HepG2 microcapsules is needed, as uniform cell-encapsulates offer more consistent properties, such as mechanical strength, diffusion and transport of oxygen and nutrients to the core of encapsulated cells.
The viability of HepG2 cells at 24 hours post-encapsulation was more than 97% (Fig. 6A). This indicates that CPN particles did not have short-term detrimental effects on HepG2 cells. However, the viability was dropped down to 80% for PNA microcapsules by day 14 but with considerable differences among all PNA microcapsules. The higher cell viability in PNA-10 microcapsules in an even longer period of culture (Fig. 6B,C) indicates that CPN can contribute to cell adhesion and proliferation under 3D environment. A decrease cell proliferation rate in PNA-30 and PNA-50 microcapsules could be due to the decrease in porosity and poor mass transfer which ultimately affected the permeability of the hydrogel 60 . To confirm the liver-specific metabolic functionality of the 3D PNA/HepG2 microcapsules, urea production, and cytochrome P450 (CYP450) activity assay was performed (Fig. 7). Urea production is an important physiological function of liver cells through active detoxification of ammonia through the urea cycle 78 . CYP450 enzyme activity is a complex phenomenon mediated by activation of nuclear receptors and gene transcription. The enzyme activity is necessary for the detoxification of foreign chemicals and metabolism of drugs. CYP1A1 is one of the major CYP450s involved in drug metabolism [79][80][81] . CYP1A1, inducible in liver as well as extrahepatic tissue, shows high catalytic activity toward environmental chemicals 82 . In the present study, we also studied the activity of CYP1A1 enzyme and examined significantly higher activity in 3D PNA-10/HepG2 microcapsule compared to control (3D PNA-0/HepG2) at day 14, suggesting that the 3D PNA-10/HepG2 encapsulates may provide a superior conducting environment where HepG2 cells undergo structural and metabolic differentiation. Since our experiments are targeted towards understanding the response of the encapsulated cells, understanding the induction CYP1A1 has been shown to provide important information to predict drug interaction 83 . Further investigation is required to determine whether the 3D PNA-10/HepG2 microcapsule improve liver-specific metabolic functionality. So far, many promising systems for the 3D cultivation of primary hepatocytes have been developed 84,85 . NIM system that we proposed in our current system is conceptually attractive, as NIM mimics many key properties of natural ECMs and provide adequate microenvironments for embedding cells under real 3D conditions.
In summary, we designed and fabricated alginate-based 3D microcapsules by incorporating cryo-fractured electrospun nanofibres of PLGA in different weight percentages. The incorporation of nanofibres caused no change in the chemical structure of the alginate hydrogel microcapsule and its spherical morphology for several weeks in a cell culture media. We optimized our electrostatic encapsulation conditions to produce 3D microcapsules of HepG2 with an average diameter of 500 µm. Our finding suggests that the 3D PNA-10/HepG2 microcapsule improved overall suitability for cell proliferation, toxicity, and metabolic activity compared to PNA-0/HepG2 microcapsule as a control. These results provide a baseline information to improve the material's design and configurations for optimal in vitro cell growth and development. Factors including culture conditions (dynamic vs. static) also have an important influence on the 3D cultivation of primary hepatocytes (Fig. 8). Development of this 3D cell microcapsules by incorporating co-cultures of other hepatic cells, such as primary hepatocytes, kupffer cells, stellate cells, etc., may further improve characteristics, and prediction of human liver-related functionalities in in vitro 3D system. However, several additional assessments are still needed to validate and develop 3D PNA-10 microcapsule into a fully functional in vivo mimicking architecture of liver which has potential use for preclinical study, toxicology, and pharmacological drug screening.  www.nature.com/scientificreports www.nature.com/scientificreports/ needle. The needle-to collector distance was maintained at 75 mm, with an applied voltage of 20 kV. The feeding rate of the solution was precisely controlled by a syringe pump system, which was adjusted to a flow rate of 4 mL/h. The fibres deposited onto an aluminum foil-wrapped rotating collector were left overnight in dust-free conditions at room temperature, to allow complete solvent evaporation. The nanofibre meshes were removed from the collector and detached from the aluminum foil for further characterization.

Methods
Production of cryoground PLGA nanofibre (CPN). Electrospun nanofibre mesh of PLGA was subjected through the cryogrinding process as previously described 64 with some modification to the obtained ground powder of nanofibres. A cryogenic impact grinder (6770 Freezer Mill, Spex, USA) with a self-contained liquid nitrogen bath (4-5 L) was used. About 0.2 g of PLGA nanofibre mesh was cut into the small pieces of unit cm 2 size and put into a 25 ml polycarbonate grinding vials for pulverization. After a pre-cooling period (5 min), six working cycles were used for each grinding. Each cycle consisted of grinding and re-cooling periods (1 min). The applied impact bar frequency was 14 Hz. The ground fibre was dispersed in ethanol and filtered through sterile 70 µm sieve for the smaller and uniform size of CPN. Dry powder of CPN was collected after complete evaporation of ethanol under the hood.
Surface morphology analysis. The surface morphology of electrospun PLGA nanofibre mesh and CPN was analyzed using a scanning electron microscope (SEM, Hitachi SU8000, Tokyo, Japan). Small cut pieces of mesh and CPN samples were deposited on copper tape and sputter-coated with gold using a Polaron SEM coating system (Quorum Technologies, East Sussex, UK) for 90 seconds at 15 mA. SEM images were taken at an accelerating voltage of 10 kV and 10 μA current. Image-Pro Plus 6.0 software (Media Cybernetics, Inc., Rockville, MD, USA) was used to measure nanofibre diameters in SEM images.
Fabrication of PNA microcapsules. Ultra-pure alginate (PRONOVA UP LVG) was purchased from Novamatrix (Industriveien 33 N-1337 Sandvika Norway). 2% (w/v) alginate solution was prepared in 1X HBSS buffer (Life Technology, Gaithersburg, MD, USA). CPN was mixed with an alginate solution in different weight percent of the dry weight of alginate. CPN was first dispersed in DI water under sonication and then combined with alginate solution. For better dispersion of CPN, the mixture solution was pipetted and vortexed for 5 min. PNA microcapsules were obtained by electro-spraying the dispersed solution of CPN-alginate using a previously published method 65,66 . In brief, the CPN-alginate solution was drawn into a syringe fitted with a 24-gauge angiocatheter, pierced at the hub with a 23-gauge needle to serve as the positive electrode in the electrostatic encapsulation process. The syringe was loaded onto a syringe pump and arranged in such a way that the droplets ejected from the angiocatheter would fall orthogonally onto the surface of 150-mM calcium chloride (CaCl 2 ) solution. The distance from the angiocatheter tip to the surface of the CaCl 2 was fixed at 25 mm with an applied voltage of 5.5 kV. Pump flow rates were set at 18 mL/min and the ground electrode was immersed in the CaCl 2 receiving bath. After fabrication, PNA microcapsules were washed twice with DI water before subjected for further characterization. The control microcapsules without CPN were also fabricated using the alginate solution only.
Morphology and stability study. Morphology of the fabricated PNA microcapsules was studied by using an optical microscope (EVOS ® XL Core Imaging System) and SEM. Optical images were taken while microcapsules were in the cell culture condition. SEM images were taken after PNA microcapsules were lyophilized utilizing the same procedure stated previously. PNA microcapsules were cryo-fractured in liquid nitrogen and lyophilized to observe their internal morphology using SEM. Size distributions of the microcapsule were measured from optical images using Image J software. Stability of the PNA microcapsules was studied by incubating them into 24 well plates with complete DMEM culture media in the rotator at 50 rpm speed for four weeks. www.nature.com/scientificreports www.nature.com/scientificreports/ Primary hepatocyte encapsulation in PNA microcapsules (3D PNA/PH). Method for encapsulation of primary hepatocytes (PH) in PNA microcapsules was exactly the same as HepG2 cells. Briefly, freshly isolated primary rat (Wistar) hepatocytes were purchased from the Triangle Research Laboratory (Research Triangle Park, NC). Cell counting and viability measurement of fresh hepatocytes obtained from the company was performed by trypan blue (TB) assay before these cells were assigned to a designated experimental purpose and processed accordingly. Cells were used for encapsulation within 6-8 hours of their isolation. After encapsulation as previously described, microcapsules were cultured in static and dynamic (3D rotator at 40 rpm speed) condition. The viability of encapsulated hepatocytes was evaluated by measuring the reduction of tetrazolium salt [3-(4, 5-dimethylthiazol-2-yl)-2-5 diphenyltetrazolium bromide; MTT as previously reported 88  HepG2 cell viability and attachment. Cell viability was monitored with Trypan Blue (TB) assay after retrieving the encapsulated cells. The cultured PNA/HepG2 microcapsules were quickly rinsed with 1X DBPS (Life Technology, Gaithersburg, MD, USA) twice and de-gelled by using sodium citrate (100 mM) solution. The resulted solution was centrifuged to retrieve the cell pellet which was resuspended in fresh media after decanting the supernatant. The retrieved suspension was stained with TB reagent and cell viability was calculated after counting the live or dead cells using haemocytometer.
Cell attachment and distribution within the microcapsule were studied by SEM analysis. The samples were prepared by rinsing the 3D PNA/HepG2 with DPBS (2 brief rinses) followed by fixation with 4% glutaraldehyde (Sigma Aldrich, St. Louis, MO, USA) for 30 min at 4 °C. After fixation, samples were briefly rinsed with deionized (DI) water (2 times) and dehydrated (sequential incubations in 30, 50, 75 and 100% ethanol, 10 mins each) at room temperature. The samples were left to dry in a sterile fume hood for 24 h before SEM imaging.
Fluorescence imaging and analysis. Fluorescence imaging of 3D PNA/HepG2 was performed by staining with acridine orange and propidium iodide (AOPI) dye (Nexcelom Bioscience, Lawrence, MA). At different time points, cultured media was aspirated from the wells, and microcapsules were washed with DPBS twice to remove FBS. Then, stained with 15 µl dye and incubated at 37 °C for 10 min. Z-stack fluorescence images were photographed under an Olympus IX83 microscope using Olympus cell Sens Dimension software (Olympus Corporation, Shinjuku, Tokyo, Japan).
Lactase dehydrogenase (LDH) assay. LDH was quantified in collected media at different time periods with a Pierce LDH cytotoxicity assay kit (Thermo Scientific, CatLog No: 88953, Waltham, MA, USA). Briefly, 50 µL of each collected sample medium was transferred to a 96-well flat-bottom plate in triplicate wells along with LDH positive control (as mentioned in the kit) and blank media as a negative control. Then, 50 µL of the reaction mixture was added in each well, and the plate was incubated at room temperature for 30 minutes at dark condition. The reaction was stopped by adding 50 µL of Stop Solution to each sample wells and mixed by gentle tapping. The absorbance of the assay solution was measured on a microplate reader (BioTek Inc., Winooski, VT, USA) at 490 nm and 680 nm wavelength to calculate the cytotoxicity.
Urea and CYP450 assay. Urea assay was performed after stimulating the microcapsules with 5 mM NH 4 Cl for 24 h as described previously 88 with some modification. And, commercially available urea assay kit (BioChain, Newark, CA, USA, catalog No: Z5030016) was used to calculate the urea production in collected cultured media according to the manufacturer's instruction. Briefly, 50 µL of water (blank), standard (as provided in the kit) and samples were taken in triplicate into separate wells of 96 well plates. Then, 200 µL working reagent was added and incubated 50 min at room temperature. Optical density was read at 430 nm using microplate reader and urea concentration (µg/ml) in the collected sample was calculated.
CYP1A1 enzyme activities were measured by P450-Glo ™ CYP1A1 assay kit (Promega Co., Madison, WI, USA, catalog No: V8751) as described previously 89 with some modification. Briefly, CYP1A1 activity was induced by incubating microcapsules in media supplemented with 300 μM omeprazole. Cultured media alone was used as a control. All microcapsules were incubated with complete media supplemented with 100 μM Luciferin-CEE for 5 h. An aliquot (25 μl) of the medium was transferred to 96-well opaque white luminometer, and luciferin detection reagent (25 µl) was added to each well. After sitting the samples at room temperature for 20 min, luminescence was measured using microplate reader. Statistical analysis. All results were expressed as mean ± S.D. Data were analysed for significance with OriginPro software (Origin Lab, Northampton, MA, USA) using a one-way analysis of variance (ANOVA). Post hoc Tukey's test was performed with ANOVA for multiple comparisons. The α-value was set to 0.05 and p-values less than 0.05 were considered statistically significant.

Data Availability
The datasets generated during and/or analysed during the current study are available from the corresponding author upon request.