A thermally responsive cationic nanogel-based platform for three-dimensional cell culture and recovery

Abstract

Thermo-sensitive nanogels have attracted great interest due to their versatile biological applications, but little research has been carried out to study their abilities in three-dimensional (3D) cell culture and recovery. A new insight is reported into exploring a low-charged thermosensitive cationic nanogel, poly(N-isopropylacrylamide-co-2-hydroxyethyl methacrylate-co-2-dimethylaminoethyl methacrylate) (P(NIPAM-HEMA-DMAEMA) or PNHD), for 3D cell culture. The positively charged DMAEMA moieties of the PNHD nanogels facilitate cellular attachment, and the low charge density renders those nanogels minimally cytotoxic. The volume phase transition temperatures (VPTTs) of the synthesized PNHD nanogels can be tailored by adjusting monomer feeds during the course of polymerization. Dynamic rheological measurements confirm the fully reversible sol–gel transition for 50 mg mL⁻¹ PNHD nanogels. At room temperature, PNHD nanogels are in their dispersion states, allowing them to be thoroughly mixed with inoculated cells. Above their gelation temperatures, hydrogels containing a mixture of cells and nanogels are formed in situ. The increase in viable cells with culture time suggests that the in situ formed hydrogels support cell proliferation. Recovery of these cultured cells can be simply achieved by cooling the mixed hydrogels below their gelation temperature. The spherical shape of the released cells indicates their 3D growth in the absence of a physical scaffold. The released cells are not only alive, but also retain the capability of migration.

---

1. Introduction

Traditional cell culture is performed in two dimension (2D). However, in such a 2D platform, cells are typically exposed to a rigid flat solid surface on the basal side and to a liquid at the apical surface. Inhabiting such a 2D rigid substrate requires dramatic adaption by the surviving cells because they lack exposure to the extracellular matrix (ECM) that is unique to each cell type, which may alter cell metabolism and functionality. Furthermore, 2D systems cannot provide a complex and dynamic microenvironment for cells, and thus lead to spurious findings to some extent by forcing cells to adjust to an artificial flat and rigid surface. To overcome these problems, three-dimensional (3D) matrices have been developed to replace the conventional 2D cell culture platform. Most strategies for developing 3D matrices aim to mimic natural ECM as the templates onto which cells can attach, proliferate, migrate and function, and subsequently direct the cells to form final products.¹ There are certain properties that the 3D matrices should exhibit, including a suitable surface chemistry to promote cellular attachment, migration, proliferation and matrix secretion, and an appropriate pore size and interconnectivity for cell infiltration.² The matrices fabricated from naturally occurring materials have the advantage over synthetic ones in that they can serve as a potential source for biological cues and signalling molecules, but their mechanical properties are difficult to control. On the other hand, the major benefit of synthetic matrices is the ability to design and manipulate them in order to achieve the desired mechanical properties.³ However, synthetic matrices inherently lack biological cues for cellular attachment, proliferation and migration, which partially hinder their biological applications. As a result, considerable efforts have been put into designing biomimetic synthetic matrices for 3D cell expansion, regenerative medicine and tissue-engineering applications.^4–6

The biomimetic synthetic matrices with in situ pore formation can be of great interest for cell culture. Cells and bioactive molecules are homogeneously distributed in a liquid suspension before solid matrices are formed. After matrices are constructed in situ, cells are randomly entrapped inside the 3D environment. The matrices generally necessitate solidification within a short period of time. The solidification mechanisms directly affect the kinetics of the process and stability of the resultant supports.⁷ Typical solidification approaches to matrix formation include thermally or photochemically activated radical polymerization or crosslinking,^8–11 ionic crosslinking,^12,13 self-assembly,¹⁴ thermal gelation,^7,15 and others. Among them, the matrices solidified via thermal gelation employing thermoresponsive polymers are more attractive since such physical transition is fully reversible. The thermosensitive hydrogels formed from N-isopropylacrylamide (NIPAM)-based copolymers,¹⁶ pluronic series,¹⁷ and polypeptides¹⁸ have been utilized for two-dimensional or three-dimensional cell culture and for a gentle recovery of cells by reducing the temperature to 25 °C. However, these hydrogels cannot offer well controlled pore size, and loose polymer chains may attach to the surface of recovered cells, which may induce detrimental side effects inside the human body. NIPAM-based nanogels/microgels may provide an alternative for cell culture and recovery. To date, although thermosensitive nanogels/microgels have attracted great interest over the last decades due to their versatile applications^19,20 in drug delivery,^21,22 biosensing,²³ chemical separation²⁴ and catalysis,^25–27 little research has been conducted to examine the capability of thermosensitive nanogels for 3D cell culture. Recently, we reported the application of anionic thermosensitive poly(N-isopropylacrylamide-co-acrylic acid) (P(NIPAM-AA)) nanogels for 3D cell culture.^28,29 However, their negative surface charge may not be suitable for cellular attachment due to the negatively charged proteoglycans on the cell membrane.

In this study, we developed new thermoreversible low-positively-charged cationic nanogels for 3D cell culture and recovery, in which positive charges on the nanogel surface facilitate cell attachment while low charge density minimizes cytotoxicity. The details of the preparation of the cationic thermoresponsive poly (N-isopropylacrylamide-co-2-hydroxyethyl methacrylate-co-2-dimethylaminoethyl methacrylate) [P(NIPAM-HEMA-DMAEMA) or PNHD] nanogels and the possible mechanism for the in situ formation of thermal-reversible 3D cell culture microenvironments are depicted in Fig. 1. The amount of positive charges and the volume phase transition temperatures (VPTT) of the PNHD nanogels can be tuned by adjusting reaction recipes.³⁰ Below VPTT, cells are homogenously mixed with nanogels in solution. Above VPTT, cells can be easily encapsulated inside the in situ formed hydrogels and proliferated in a 3D manner, mimicking the natural environment. Because the gelation is thermoreversible, the in situ formed hydrogels turn into solutions upon cooling to room temperature. Therefore, the 3D cultured cells can be easily released and recovered without using trypsin, a chemical reagent typically required to remove adherent cells from a 2D surface.

Fig. 1 (a) Two-stage emulsion polymerization protocol for the synthesis of cationic thermosensitive PNHD nanogels; (b) schematic description on the 3D cell culture microenvironment using temperature-triggered PNHD nanogels. The inoculated cells can homogeneously mix with PNHD nanogels at room temperature. Upon heating the culture system above 37 °C, the cell-laden hydrogel composed of cells and PNHD nanogels are formed in situ for 3D cell culture. The cultured cells can be recovered and separated by cooling the hydrogel system to room temperature.

2. Materials and methods

2.1 Materials

N-Isopropylacrylamide (NIPAM, 99%+), 2-hydroxyethyl methacrylate (HEMA, 98%+), 2-dimethylaminoethyl methacrylate (DMAEMA, 98%+), N,N-methylenebisacrylamide (MBA, 98%+), potassium persulfate (KPS, 99%+) and sodium dodecyl sulfate (SDS, 98.5%+) were purchased from Sigma-Aldrich. NIPAM was purified by recrystallization in n-hexane and dried in vacuum at room temperature. Dulbecco's Modified Eagle's Medium (DMEM), trypsin-EDTA, penicillin–streptomycin and fetal bovine serum (FBS) were purchased from Gibco-BRL (Grand Island, NY). The Live/Dead viability/cytotoxicity kit (L3224) was purchased from Molecular Probes (Oregon). 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) was received from Merck (Germany). All other chemicals were of analytical grades and directly used without further purification.

2.2 Synthesis of thermo-responsive cationic nanogels

The P(NIPAM-HEMA-DMAEMA) (PNHD) nanogels were synthesized by a two-stage free radical emulsion copolymerization according to the recipes listed in Table 1. To achieve low positive charges on nanogels, the molar ratios of the cationic DMAEMA monomer to total monomer feed were kept low from 0.05 to 0.20. The molar ratio of the MBA crosslinker to the total monomer feed was set to 0.02 for all nanogels. The molar ratios of the emulsifier SDS and the initiator KPS to total monomer feed were fixed at 0.05 and 0.01 for the PNHD1–3 nanogels. To avoid the possible electrostatic reaction between the DMAEMA and the KPS initiator or SDS emulsifier, DMAEMA was charged after 1 h polymerization, and the polymerization was carried out at pH 9 (Fig. 1). In a typical experiment, known amounts of NIPAM, HEMA, MBA and SDS were dissolved in 97 mL water. pH of the mixing solution was adjusted to 9, and then the solution was transferred to a 250 mL three-necked flask fitted with a condenser, a stirrer, and gas inlet/outlet. After degassing for 30 min, the system was soaked in a preheated oil-bath at 70 °C under a nitrogen atmosphere. In addition, 3.0 mL of KPS aqueous solution was injected to initiate the polymerization. After 1 h reaction, DMAEMA (ranging from 0.5 to 2.0 mmol) was charged to the reactor. The polymerization was further carried out for another 4 h. After cooling, the nanogels were purified by dialysis (cut-off M_W 12–14 kDa) against MilliQ water for 1 week with daily change of water. The purified nanogels were adjusted to pH 7 using NaOH and concentrated using a rotary evaporator. An aliquot of concentrated nanogel dispersion was dried in an oven to determine its final concentration.

Table 1 Synthesis and characterization of different PNHD nanogels

Nanogels	Monomer content in feed (mmol)			MBA (mmol)	SDS (mmol)	KPS (mmol)	DMAEMA content (mmol g^−1 polymer)		Yield (%)	dh^c (nm)	Zeta potential^c (mV)	VPTT (°C)	Tgel (°C)
	NIPAM	HEMA	DMAEMA				Observed^a	Theoretical^b
a Calculated from pH & conductometric titration.b Calculated from monomer content in feed.c Determined from the DLS at 25 °C; zeta potential: mean ± SE, n = 5.
PNHD1	9.34	0.16	0.50	0.20	0.50	0.10	0.34	0.43	88	78	−0.77 ± 0.12	36	34.0
PNHD2	8.68	0.32	1.00	0.20	0.50	0.10	0.64	0.85	87	59	0.80 ± 0.14	34	33.1
PNHD3	7.36	0.64	2.00	0.20	0.50	0.10	1.14	1.63	84	93	1.35 ± 0.41	33	—
PNHD4	8.68	0.32	1.00	0.20	0.10	0.10	—	0.85	—	—	—	—	—
PNHD5	8.68	0.32	1.00	0.20	0.50	0.02	0.73	0.85	83	164	1.03 ± 0.64	35	33.7

---

2.3 Conductometric and potentiometric titrations

The amount of cationic DMAEMA in the synthesized nanogels was quantified by conductometric and pH titration. The details of the titration have been described elsewhere.³¹ Typically, the pH of 100 mL PNHD nanogel dispersion (0.70 mg mL⁻¹ in MilliQ water) was first adjusted to around 10 using NaOH, and the nanogel was then back titrated by a 0.1 M HCl solution under N₂ protection. After each addition of HCl, the solution conductivity was monitored by a H18733 conductivity meter, and the pH was recorded using a pre-calibrated pH/mV meter (smartCHEM-pH, TPS Australia).

2.4 Dynamic light scattering and ζ-potential

The purified PNHD nanogel dispersions (pH 7.0) were diluted to 1.0 mg mL⁻¹ using MilliQ water. The hydrodynamic diameters (d_h) and zeta potentials of the nanogels were measured by dynamic light scattering (DLS) and electrophoresis at different measurement temperatures on a Malvern Zetasizer (Nano-ZS). The DLS data were collected by an autocorrelator, and the CONTIN software was used to analyze the intensity–intensity autocorrelation functions. For the electrokinetics study, the Smoluchowski model was used to convert electrophoresis mobilities to ζ-potentials.

2.5 2D cell culture

A murine embryonic mesenchymal progenitor cell line (C3H/10T1/2) from Riken Cell Bank (Japan) was cultured in a complete growth medium (DMEM supplemented with 10% FBS, 100 units per mL penicillin, 100 μg mL⁻¹ streptomycin and 2 mM L⁻¹ L-glutamine) and at 37 °C in a humidified atmosphere containing 5% CO₂.

2.6 Cell viability assays

The cytotoxicity of the synthesized nanogels was evaluated by the MTT test.^32,33 Typically, 150 μL of C3H/10T1/2 cells were seeded in individual wells of a 96-well plate at a density of 5.0 × 10⁴ cells mL⁻¹ and allowed to adhere overnight. The growth medium was replaced with 150 μL of fresh medium containing the synthesized PNHD nanogels with concentrations varying from 50 to 500 μg mL⁻¹ (pH, ∼7.0). Cells were then incubated at 37 °C for 48 h. Subsequently, 10 μL of MTT (5.0 mg mL⁻¹ in PBS) was added to each well. After an additional 4.0 h incubation, the growth medium was removed, and 150 μL DMSO was added to each well to ensure complete solubilization of the formed formazan crystals. Finally, to determine the number of living cells, the absorbance of the solubilized formazan crystals was recorded using an ELX808 Absorbance Microplate Reader (Bio-Tek Instruments, Inc., USA) at a wavelength of 595 nm.

2.7 Rheological characterization

Dynamic oscillation experiments of 50 mg mL⁻¹ of PNHD nanogel dispersions (pH, ∼7.0) were performed using a Bohlin CVO rheometer (Bohlin Instruments), equipped with 40 mm parallel plate geometry. The gap was set to 0.07 mm, and the temperature was controlled by a Peltier system connected to a water bath. Stress sweeps were first performed to determine the linear viscoelastic region at 37 °C. Within the linear viscoelastic region and under a controlled stress of 0.1 Pa, oscillation frequency dependences of the elastic (storage) modulus G′ and viscous (loss) modulus G′′ were measured at different temperatures.

2.8 Hydrogel morphology

The in situ-formed hydrogels, composing of thermosensitive PNHD nanogels, prepared at 37 °C were immediately quenched by liquid nitrogen and then dried in a freeze dryer. The cross-sectional morphologies of these samples were observed using a Philips XL 30 FE-SEM (scanning electron microscope) after surface platinum coating. The accelerating voltage was 10 kV.

2.9 Three-dimensional cell culture

C3H/10T1/2 cell dispersion (in complete growth medium, 1.1 × 10⁶ cells mL⁻¹) was mixed with 55 mg mL⁻¹ of nanogel dispersions (pH, ∼7.0) at a volumetric ratio of 1 : 10 at room temperature, resulting in an initial cell concentration of 1.0 × 10⁵ cells per mL and a nanogel concentration of 50 mg mL⁻¹. Furthermore, 1.0 mL of the above mixture suspension was seeded to each well of a 12-well plate and placed in an incubator at 37 °C. After a 2.0 h thermal gelation of the dispersion, 4.0 mL of complete growth medium at 37 °C was added to each well of the 12-well plate. Cells were then incubated at 37 °C for up to 14 days. During this culture period, 3 mL of the growth medium in each well was exchanged with fresh medium every 3 days. After a predetermined period of cell culture, 1.0 mL of the growth medium was replaced with 1.0 mL of MTT solution (5.0 mg mL⁻¹ in PBS). After subsequent 4.0 h incubation at 37 °C, the culture medium was removed, the in situ-formed hydrogels were cooled down to room temperature, and 4.0 mL of DMSO was added to ensure the solubilization of the formed formazan crystals. Finally, to determine the number of living cells, the absorbance of the solubilized formazan crystals was recorded using an ELX808 Absorbance Microplate Reader (Bio-Tek Instruments, Inc., USA) at a wavelength of 595 nm. To eliminate the absorbance contribution from the nanogels, the same experiments without MTT addition were used as a reference. The absorbance difference between the system with MTT and the reference was used in our data presentation to represent relative cell density.

2.10 Confocal laser scanning microscopy

The Live/Dead cytotoxicity/viability kit was used to stain live and dead cells. Briefly, fresh 1 μM calcein AM (the acetomethoxy derivate of calcein) and 2.5 μM ethidium homodimer-1 (EthD-1) working solutions were prepared according to the manufacturer's protocol. At day 1 or 7, the growth medium was removed from the culture system and the cell/PNHD5 mixed hydrogels were washed two times using 1.0 mL PBS at 37 °C. PBS used for washing was then replaced with 2 mL of working dye solutions and further incubated at 37 °C for another 45 min. The top layer of the hydrogel was gently scraped to remove cells growing on the top surface, and the middle part of the hydrogel was carefully transferred to a glass slide for imaging analysis. Hydrogel samples were observed under a Leica SP5 spectral scanning confocal microscope (Leica Microsystems, Germany). Excitation wavelengths of 494 and 528 nm and emission wavelengths of 517 and 617 nm were employed for live (green) and dead (red) cells, respectively.

2.11 Recovery of 3D cultured C3H/10T1/2 cells

The morphology of C3H/10T1/2 cells cultured in the in situ-formed hydrogels for 6 days was observed using an Olympus IX50 inverted microscope after cooling the hydrogels to room temperature so as to release the 3D cultured cells from the hydrogels. The released cells were further cultured on the flat bottom of a 12-well plate (2D) for one day to allow cells to attach on the substrate, and the morphology of the attached cells was examined using the Olympus microscope.

2.12 Statistical analysis

Statistical significance was determined by the application of a Student's t-test or by one-way ANOVA followed by Student–Newman–Keuls test using Sigma Stat version 3.5. A difference was considered significant if P < 0.05.

3. Results and discussion

3.1 Nanogel synthesis and characterization

In this study, we propose a new concept of developing a 3D cell culturing environment by exploring the in situ-formed hydrogel scaffolds made up of thermosensitive cationic nanogels. The presence of a weak positive charge on the nanogels facilitates cell attachment but with low cytotoxicity.³⁴ The thermosensitivity renders the in situ formation of 3D mixed hydrogels with cells by varying experimental temperatures, and the sol–gel transition is fully reversible (Fig. 1).

Nanogels are the nanometric counterparts of gels, which exhibit network structures in solvents. The most commonly used approach to produce nanogels is the free radical crosslinking copolymerization of monovinylic monomer(s) and di- or multi-functional comonomers (crosslinkers). In our system, a series of thermosensitive cationic nanogels, P(NIPAM-HEMA-DMAEMA) (PNHD) nanogels, were synthesized by free radical emulsion copolymerization, where MBA was used as the crosslinker to copolymerize with NIPAM, HEMA and DMAEMA comonomers. NIPAM contributes to the thermosensitivity, DMAEMA provides cationic characteristic and HEMA is used to adjust the volume phase transition temperature (VPTT). In order to avoid the formation of macroscopic gels, our experiments were carried out in dilute solutions. In addition, we also adopted low concentrations of crosslinkers and the emulsion polymerization technique to minimize inter-particle aggregation. The detailed synthesis recipes and key nanogel characterization information are summarized in Table 1. The synthesized PNHD1, 2, 3 and 5 nanogels were in the latex form, while the PNHD4 nanogel agglomerated during synthesis due to insufficient amount of emulsifier. After polymerization, membrane dialysis was used to remove residual unreacted monomers and SDS emulsifier. From weight analysis, the yields were found to be above 83% for all the nanogels.

The DMAEMA moiety in the nanogels is the cationic functional group for facilitating cellular attachment and proliferation.³⁵ On the other hand, low charge density is crucial to maintain the low cytotoxicity of the materials. The average amount of functional DMAEMA in the nanogels was quantified by conductometric & pH titrations. The representative back titration curves of PNHD1–3 nanogels are shown in the ESI (Fig. S1†). After a gradual addition of HCl, the conductivity initially decreases (descending leg) as excess alkali is neutralized. After a transition point, it slowly increases with the addition of HCl. After passing a second transition point at high HCl concentrations, the conductivity rapidly increases (ascending leg) in association with the introduction of excessive HCl. In the buffering zone between the descending and ascending legs, the conductivity increases slightly with HCl addition due to the protonation of functional DMAEMA groups. Therefore, the buffering zone can be used to quantify the DMAEMA content in the synthesized nanogels. The details on the titration curve explanation and DMAEMA content calculation are given in our previous publication.^32,33 The trends of all nanogel titration curves are similar, but a broader buffering zone is evident for those with higher feed DMAEMA contents. The averaged DMAEMA contents in the nanogels from titration are summarized in Table 1. The measured DMAEMA contents are slightly lower than the theoretical values calculated for the feed, which indicates a relatively high conversion of DMAEMA in the nanogel copolymerization process.

3.2 Dynamic light scattering of dilute nanogels

The hydrodynamic diameters (d_h) of the synthesized nanogels were measured by dynamic light scattering (DLS) at different experimental temperatures (1.0 mg mL⁻¹; pH, ∼7.0), and the CONTIN software was used to analyze the intensity autocorrelation function. For all synthesized nanogels except for the PNHD4, only one peak or decay mode is observed at different temperatures, which is related to the translational diffusion of nanogels. The representative size distributions of the nanogels at 25 °C are shown in the ESI (Fig. S2†). Cumulant study shows the mean d_h of 78, 59, 93 and 164 nm for PNHD1–3 and PNHD5 nanogels at 25 °C (Table 1). They are identical to the apparent particle sizes calculated from CONTIN analysis. The size of PNHD5 nanogels is larger than that of other nanogels due to the smaller feed of KPS initiator.

To further understand the phase behavior as a function of temperature, plots of PNHD nanogels' d_h versus temperature are shown in Fig. 2. The d_h of PNHD nanogels almost remains constant at temperatures below 30 °C. With an increase in temperature to a critical value, the d_h sharply increases. The critical transition temperature is known as the volume phase transition temperature (VPTT) of nanogels [corresponding to the lower critical solution temperature (LCST) for linear thermosensitive polymers]. In this study, the VPTT of the synthesized PNHD nanogels is defined as the temperature at which the steepest slope is determined in Fig. 2 with their detailed values shown in Table 1. At low temperatures, PNIPAM segments are soluble due to the formation of H-bonds with water, and the hydrated nanogels are in their swollen states. Beyond VPTT, the dehydration of the PNIPAM segments gives rise to chain shrinkage. When the electrostatic or steric interaction from the hydrophilic segments of the nanogel are strong enough to balance the hydrophobic interaction associated with the PNIPAM segments, no nanogel aggregation can be observed, and the hydrodynamic radius of the nanogel decreases after VPTT.³⁶ However, when the hydrophilic segments of the nanogel are not able to compensate for the hydrophobic contribution, nanogel aggregates will be formed, and the hydrodynamic radius sharply increases beyond VPTT as shown in Fig. 2.

Fig. 2 Temperature dependence of the hydrodynamic diameters of various PNHD nanogels (C, ∼1.0 mg mL⁻¹, pH, ∼7.0). (a) PNHD1; (b) PNHD2; (c) PNHD3; (d) PNHD5.

PNIPAM nanogels are temperature responsive with a VPTT of 32 °C.³⁷ Introducing hydrophilic DMAEMA as a comonomer to the PNIPAM nanogels will increase the VPTT of the resulting nanogels. Therefore, a relatively hydrophobic HEMA moiety has to be introduced during the copolymerization to ensure that the VPTT of the nanogels remains within the most suitable temperature range for cell growth. From our experiments, the VPTTs of the synthesized PNHD nanogels with different DMAEMA contents can be controlled between 30 and 36 °C by adjusting the feed of HEMA. The amounts of DMAEMA and HEMA co-monomers are crucial to control hydrogel formation beyond the VPTT.

The surface charges of synthesized nanogels were examined from zeta potential analysis. As shown in Table 1, the zeta potential values are −0.77 ± 0.12, 0.80 ± 0.14, 1.35 ± 0.41 and 1.03 ± 0.64 mV for PNHD1–3 and PNHD5 nanogels. The negative zeta-potentials for the nanogel with the smallest amount of DEMEMA comonomer are attributed to the negative change from the KPS initiator. Zeta potentials increase with the increasing molar fraction of DMAEMA in the nanogels because of the presence of more positively charged amino groups in the nanogels. The weak positive charges of the nanogels result in their low cytotoxicity.

3.3 Cytotoxicity of the nanogels

The biocompatibility of the synthesized thermosensitive nanogels was examined with C3H/10T1/2 stem cells by the MTT assay. The MTT assay is based on the principle that the mitochondrial dehydrogenase of viable cells can metabolize yellow-colored water-soluble tetrazolium salt, MTT, into water-insoluble, blue-colored formazan crystals. The extent of MTT conversion, measured by the optical density values at 595 nm, is dependent on the number of viable cells. Damaged or dead cells are devoid of any mitochondrial dehydrogenase activity.^38,39 Thus, the optical intensities in the MTT assay represent the number of viable cells in a cell culture system.⁶ Fig. 3 shows the cytotoxicity of the synthesized thermosensitive cationic nanogels to C3H/10T1/2 cells at various concentrations. From the figure, the cytotoxicity of the nanogels depends on their DMAEMA content. As the nanogel concentrations range from 50 to 500 μg mL⁻¹, only the PNHD3 nanogel shows cytotoxicity with a cell number ratio (% of controls) lower than 100%, whereas PNHD1, 2, and 5 nanogels slightly stimulate cell growth with a cell number ratio higher than 100%. The mechanism by which the PNHD1, 2, and 5 nanogels (with 0.34, 0.64 and 0.73 mmol DMAEMA per g polymer) stimulate cell growth is still unclear, but it may be caused by the presence of weak positive charges. The cytotoxicity of PNHD3 (with 1.14 mmol DMAEMA per g polymer) is ascribed to its high DMAEMA content, and similar toxicity to chondrosarcoma cells has been reported for a copolymer of DMAEMA and HEMA at a molar ratio of 1 : 1.⁴⁰ By carefully adjusting the molar ratio of DMAEMA in the copolymer, it is possible to maintain the cellular mitochondrial dehydrogenase activity at a low positive charge. The low cytotoxicity of PNHD1, 2, and 5 nanogels supports their potential biological applications.

Fig. 3 Cytotoxicity of PNHD nanogels at various concentrations to C3H/10T1/2 cells. Incubation time: 48 h.

3.4 Thermal gelation of concentrated nanogels

Fig. 4A compares photographs of 50 mg mL⁻¹ of PNHD nanogel dispersions at 25 and 37 °C. All three PNHD nanogel dispersions are liquids at 25 °C, but form hydrogels at 37 °C. After cooling down to 25 °C, the hydrogels turn into solutions again. The physical gelation behavior can be triggered by temperature and is fully reversible. The temperature-responsive sol–gel transition properties of these nanogel dispersions were further investigated using dynamic rheometric measurement at different temperatures. Fig. 4B shows the temperature dependence of the dynamic modulus of 50 mg mL⁻¹ of PNHD nanogel dispersions (pH, ∼7.0). At low temperatures (25 to 32 °C), the PNHD nanogel dispersions are in their liquid forms and the solution behavior is predominantly viscous flow, where the slope of ln G′′ vs. ln(frequency) is 1 (Fig. S3†). Since viscous modulus (G′′) dominates the flow property, the elastic modulus (G′) cannot be accurately measured under a low oscillation frequency. Within this temperature range (dashed lines shown in the Fig. 4B), low shear viscosity η′ should be more appropriate to quantify the solution behaviour rather than dynamic modulus. From the relationship of η′ = G′′/ω (ω is the angular frequency 2πf and f is the oscillation frequency), we can calculate the η′ to be around 0.01 to 0.05 mPa s. With increasing temperature, solution viscosity drops slightly due to the increase in thermal movement. Beyond 30 °C, the temperature increase results in a dramatic increase in both G′ and G′′ and the dispersions display viscoelastic properties. Above 34 °C, the G′ is higher than the G′′ at low oscillation frequencies, which indicates the formation of hydrogels. Such temperature-driven sol–gel transition is fully reversible. The temperature where G′ and G′′ cross over during the heating process is defined as the gelation temperature (T_gel).⁴¹ T_gel values obtained from Fig. 4B are included in Table 1, and the T_gel of PNHD1, 2, and 5 nanogels are close to their corresponding VPTTs. This result indicates that hydrophobic interaction among these shrunk nanogels is the main driving force for the nanogel thermal gelation, which is associated with the transformation of NIPAM moieties from hydrophilic to hydrophobic beyond VPTTs. Moreover, at 37 °C, the G′ of PNHD5 hydrogel is larger than those of PNHD1 and PNHD2 hydrogels, which implies that the in situ-formed PNHD5 hydrogel has stronger mechanical properties compared with PNHD1 and PNHD2 hydrogels. Since the particle size of PNHD5 nanogel is larger than that of other PNHD nanogels, the dangling chains of the nanogels are relatively longer. The longer dangling chains facilitate a stronger inter-nanogel interaction in both the number of active junctions and the strength of each junction, which give rise to a higher elastic modulus as evident from the dynamic oscillation measurement.

Fig. 4 (A) Comparison of 50 mg mL⁻¹ of PNHD nanogels (pH, ∼7.0) in the dispersed state (25 °C) and the hydrogel state (37 °C). (B) Temperature dependence of the dynamic rheological moduli of 50 mg mL⁻¹ of PNHD nanogels (pH = 7.0). (Blue-colored circles): PNHD1; (red-colored triangles): PNHD2; (green-colored squares): PNHD5. The open circles, triangles and squares (○, △, □) correspond to the elastic (or storage) modulus (G′), and the closed circles, triangles and squares (●, ▲, ■) correspond to the viscous (or lose) modulus (G′′).

3.5 Microstructure of the hydrogels

SEM was utilized to investigate the microstructure of the resultant temperature-triggered in situ hydrogels. Since the hydrogel formation is thermosensitive and fully reversible, the immediate quenching of the hydrogels using liquid nitrogen minimizes the induced changes to the hydrogel microstructure and results in the formation of an amorphous structure.⁴² After lyophilization, the interior structure information of the hydrogel can be revealed from the cross-section images. Fig. 5 compares the SEM images of the cross-sectional morphologies for the in situ-formed hydrogels of PNHD1, 2, and 5. All PNHD hydrogels present interconnected porous structures with a pore size of 5–10 μm. The interconnected porous microstructure will be beneficial for cell culture because it allows for transport of nutrients and waste products in and out of the hydrogels.

Fig. 5 Cross-sectional SEM morphologies of the in situ-formed PNHD hydrogels. (a) PNHD1; (b) PNHD2; (c) PNHD5.

3.6 Three-dimensional culture of stem cells

C3H/10T1/2 stem cells, murine mesenchymal progenitor cells, were used to examine the potential of 3D cell culture in the presence of thermoreversible nanogels. The cells are well mixed with the nanogel dispersions at room temperature. When the temperature of the mixture is raised to 37 °C, the cell/nanogel mixed hydrogel constructs in situ. After being cultured for a certain period, the number of viable cells is evaluated by the MTT assay. Fig. 6 compares the absorbance of the MTT assays for C3H/10T1/2 cells cultured in the in situ formed thermoreversible PNHD1, 2, or 5 hydrogels. The extent of MTT conversion, measured by the optical intensity values at 595 nm, is dependent on the number of viable cells. The optical density values in the MTT assay represent viable cell number in the cell culture system.⁶ Cells continuously proliferate in the in situ-formed PNHD hydrogels in the first 9 days, and the cell number increases 4-fold. After that, the number of viable cells ceases to increase. The cell density inside the hydrogels may be so high after 9 days that the oxygen and nutrient supply achieved by passive diffusion may not meet the need of the cells. The increase in viable cells in culture suggests that the in situ-formed PNHD hydrogels support cell proliferation.

Fig. 6 Proliferation of C3H/10T1/2 cells cultured in the in situ-formed thermoreversible PNHD hydrogels. (Mean ± SD, n = 3). ‘*’ indicates P < 0.05 for cell number of other PNHD hydrogels in comparison with that in the PNHD5 hydrogels.

On day 1 and 3, the number of viable cells cultured in PNHD5 nanogel are much higher than that in PNHD1 and PNHD2 nanogels (*P < 0.05). After 6 days, no significant difference is found among the PNHD1, 2, and 5 nanogels. This result indicates that the larger hydrodynamic size of the nanogels may assist initial cell growth by providing a relatively stronger mechanical support for cells, which enhances nutrient and waste transportation.

The cell proliferation is further confirmed via the Live/Dead cytotoxicity/viability assay. The assay is based on a special two-color fluorescence dye, where green and red fluorescence represent live and dead cells, respectively. Live cells are able to produce ubiquitous intracellular esterase, and this enzyme can convert the virtually non-fluorescent cell-permeant calcein AM to the intensely green fluorescent calcein, which is well-retained within live cells. The red color is generated as ethidium homodimer-1 binds to the DNA of dying or dead cells, because the cell-impermeant dye can only enter the dying and dead cells with damaged membranes. Fig. 7 shows the confocal laser scanning images of MSCs cultured inside the hydrogels. An increase in green fluorescence between day 1 (Fig. 7a and b) corresponds to an increase in the cell number, which verifies cell viability and proliferation during this seven-day period. These results are consistent with those of the MTT analysis. Moreover, as the samples are taken from the middle of the mixed hydrogels, the results strongly support that live cells are well distributed within the 3D structured nanogels.

Fig. 7 Confocal laser scanning microscopy images of the MSC cells cultured in PNHD5 nanogels at: (a) day 1 and (b) day 7. Scale bars are 250 μm. 10× oil immersion objective.

To further understand cell behaviors inside the in situ-formed PNHD hydrogels, the cellular morphologies were examined using an inverted microscope. After reducing the temperature from 37 to 25 °C, the cell-laden nanogel-based hydrogels turn into solution containing cells and nanogel dispersions. The images of C3H/10T1/2 cells released from the PNHD1, 2, or 5 nanogels after 6 days are presented in Fig. S4(b),† 8a and b. Fig. S4(a)† shows the image of trypsinized adherent C3H/10T1/2 cells in 2D as a reference of cell size. From Fig. S4(b),† 8a and b, it is found that the cells are singularly dispersed and circular, the cell size is similar to the reference, and there is no cell attachment to the bottom surface of the 12-well plates. These results demonstrate that cells expand inside the in situ-formed hydrogels (3D culture) rather than on the surface of the 12-well plate (2D), although no physical scaffolds are introduced.

Fig. 8 Characterization of C3H/10T1/2 stem cells recovered from PNHD hydrogels. (a and b) Microscopic image analysis of cell morphology recovered from in situ-formed PNHD2 (a) or PNHD5 (b) hydrogels after culturing for 6 days by cooling the culture system to room temperature. (c and d) Microscopic image analysis of the C3H/10T1/2 cells reattached to a 2D substrate after release from the PNHD2 (c) or PNHD5 (d) hydrogels. Scale bar: 50 μm.

The cell viability is further confirmed by reculturing the cells released from the PNHD hydrogels on the flat bottom of a 12-well plate. After culturing these cells for one day, the morphology of the cells released from the PNHD1, 2, and 5 hydrogels is shown in Fig. S4(c),† 8c and d. It can be seen that cells can adhere to the 2D substrate, spread flat and show significant cellular structural changes compared to those observed in 3D culture. This result clearly indicates that the cells residing inside the in situ-formed hydrogels are not only alive, but also retain the capability of migration.⁴³ Consequently, our results reinforce that the temperature-triggered in situ cationic hydrogels can be employed for 3D cell culture and cell recovery.

To date, very few synthetic nanogels have been applied for 3D cell culture, and this is the first report of the use of a cationic weakly positively charged nanogel system for stem cell culture. Recently, negatively charged NIPAM-AA nanogels have been employed for 3D cell culture,^28,29 and the cell proliferation within the hydrogel network formed by nanogels is greater than that in the 2D culture, considering the same initial cell density. However, anionic nanogels may not promote cell adhesion due to the presence of negatively charged proteoglycans in extracellular matrix. The cationic nanogels of NIPAM-HEMA-DMAEMA are thus explored for improving the initial cell attachment, which often significantly influences further cellular functions. It is well known that the presence of DMAEMA in hydrogels strongly supports the attachment of macrophages due to its attraction of fibroblasts,⁴⁴ but the amount of DMAEMA must be carefully monitored because cells will not adhere or proliferate with a higher content of DMAEMA.⁴⁰ A small amount of DMAEMA (<5 mol%) in NIPAM-based nanogels promotes the proliferation of L929 cells.⁴⁵ Our results also reinforce that the PNHD3 nanogel with a high percentage of DEAMEA shows cytotoxicity, while PNHD1, 2, and 5 nanogels are suitable for cell culture. In this study, the cationic nanogel packed hydrogels have been shown to support stem cell proliferation, whereas the thermoresponsive NIPAM moieties allow us to culture and recover cells by physically adjusting operation temperatures. In practice, hydrogels can be simply made from linear polymers. However, these polymer chains easily attach to the membrane of 3D cultured cells via a variety of intermolecular interactions, which may induce side effects or increase the operation costs associated with the downstream processing of stem cells after harvesting. A way to circumvent this problem is to fabricate thermoresponsive nanogels, in which linear polymers are tangled. After the temperature is cooled down to room temperature, the swelling of nanogels renders their detachment from cell membranes.

4. Conclusions

A series of thermosensitive cationic nanogels, poly(N-isopropylacrylamide-co-2-hydroxyethyl methacrylate-co-2-dimethylaminoethyl methacrylate) (PNHD) nanogels, are developed and used for 3D cell culture. The ζ-potentials of the nanogels increase with the increasing molar fraction of DMAEMA. The low-charge density renders the PNHD1, 2, and 5 nanogels minimally toxic. The hydrodynamic sizes of the nanogels are sensitive to temperature due to the presence of thermosensitive NIPAM segments. The VPTTs of PNHD nanogels can be controlled at 30 to 36 °C by adjusting the HEMA and DMAEMA comonomer feeds. At high nanogel concentrations, the hydrogels composed of PNHD nanogels are formed as evident from dynamic rheological measurement. The increase in viable cells cultured in the in situ-formed PNHD hydrogels demonstrates that these nanogels support cell proliferation. Confocal fluorescence images confirm cell viability and proliferation inside the hydrogels. The cells recovered from the hydrogels upon cooling to room temperature can adhere to a 2D substrate, which indicates that the cells residing inside the in situ-formed hydrogels are not only alive, but also retain the capability of migration. Therefore, the in situ-formed thermoreversible PNHD hydrogels can be a good platform for cell 3D culture and recovery in the absence of a physical scaffold.

Acknowledgements

This project is financially supported by the Faculty of Engineering, Computer and Mathematic Sciences, the University of Adelaide. SD is grateful for the financial support from the Australian Research Council (ARC) (Discovery Grant no. DP110102877). HZ would like to acknowledge the support from the 111 Project (B12034).

References

1. H. Zhang, S. Dai, J. Bi and K.-K. Liu, Biomimetic three-dimensional microenvironment for controlling stem cell fate, Interface Focus, 2011, 1, 792–803.
2. E. S. Place, N. D. Evans and M. M. Stevens, Complexity in biomaterials for tissue engineering, Nat. Mater., 2009, 8, 457–470.
3. E. S. Place, J. H. George, C. K. Williams and M. M. Stevens, Synthetic polymer scaffolds for tissue engineering, Chem. Soc. Rev., 2009, 38, 1139–1151.
4. H. Shin, S. Jo and A. G. Mikos, Biomimetic materials for tissue engineering, Biomaterials, 2003, 24, 4353–4364.
5. G. Dubey and K. Mequanint, Conjugation of fibronectin onto three-dimensional porous scaffolds for vascular tissue engineering applications, Acta Biomater., 2011, 7, 1114–1125.
6. T. Gan, Y. Zhang and Y. Gu, In situ gelation of P(NIPAM-HEMA) microgel dispersion and its applications as injectable 3D cell scaffold, Biomacromolecules, 2009, 10, 1410–1415.
7. Q. Hou, P. A. De Bank and K. M. Shakesheff, Injectable scaffolds for tissue regeneration, J. Mater. Chem., 2004, 14, 1915–1923.
8. J. Elisseeff, W. McIntosh, K. Anseth, S. Riley, P. Ragan and R. Langer, Photoencapsulation of chondrocytes in poly(ethylene oxide)-based semi-interpenetrating networks, J. Biomed. Mater. Res., 2000, 51, 164–171.
9. E. Behravesh, S. Jo, K. Zygourakis and A. G. Mikos, Synthesis of in situ cross-linkable macroporous biodegradable poly(propylene fumarate-co-ethylene glycol) hydrogels, Biomacromolecules, 2002, 3, 374–381.
10. J. Elisseeff, K. Anseth, D. Sims, W. McIntosh, M. Randolph and R. Langer, Transdermal photopolymerization for minimally invasive implantation, Proc. Natl. Acad. Sci. U. S. A., 1999, 96, 3104–3107.
11. K. T. Nguyen and J. L. West, Photopolymerizable hydrogels for tissue engineering applications, Biomaterials, 2002, 23, 4307.
12. E. Alsberg, K. W. Anderson, A. Albeiruti, R. T. Franceschi and D. J. Mooney, Cell-interactive alginate hydrogels for bone tissue engineering, J. Dent. Res., 2001, 80, 2025–2029.
13. J. J. Marler, A. Guha, J. Rowley, R. Koka, D. Mooney, J. Upton and J. P. Vacanti, Soft-tissue augmentation with injectable alginate and syngeneic fibroblasts, Plast. Reconstr. Surg., 2000, 105, 2049–2058.
14. A. K. Salem, F. Rose, R. O. C. Oreffo, X. B. Yang, M. C. Davies, J. R. Mitchell, C. J. Roberts, S. Stolnik-Trenkic, S. J. B. Tendler, P. M. Williams and K. M. Shakesheff, Porous polymer and cell composites that self-assemble in situ, Adv. Mater., 2003, 15, 210–213.
15. A. Gutowska, B. Jeong and M. Jasionowski, Injectable gels for tissue engineering, Anat. Rec., 2001, 263, 342–349.
16. M. T. Moran, W. M. Carroll, I. Selezneva, A. Gorelov and Y. Rochev, Cell growth and detachment from protein-coated PNIPAAm-based copolymers, J. Biomed. Mater. Res., Part A, 2007, 81, 870–876.
17. A. Higuchi, N. Aoki, T. Yamamoto, T. Miyazaki, H. Fukushima, T. M. Tak, S. Jyujyoji, S. Egashira, Y. Matsuoka and S. H. Natori, Temperature-induced cell detachment on immobilized pluronic surface, J. Biomed. Mater. Res., Part A, 2006, 79, 380–392.
18. B. Li, D. O. V. Alonso and V. Daggett, The molecular basis for the inverse temperature transition of elastin, J. Mol. Biol., 2001, 305, 581–592.
19. S. Nayak and L. A. Lyon, Soft nanotechnology with soft nanoparticles, Angew. Chem., Int. Ed., 2005, 44, 7686–7708.
20. S. Dai, P. Ravi and K. C. Tam, Thermo- and photo-responsive polymeric systems, Soft Matter, 2009, 5, 2513–2533.
21. T. Zhou, W. Wu and S. Zhou, Engineering oligo(ethylene glycol)-based thermosensitive microgels for drug delivery applications, Polymer, 2010, 51, 3926–3933.
22. M. Das, S. Mardyani, W. C. W. Chan and E. Kumacheva, Biofunctionalized pH-responsive microgels for cancer cell targeting: rational design, Adv. Mater., 2006, 18, 80–83.
23. Z. Hu, Y. Chen, C. Wang, Y. Zheng and Y. Li, Polymer gels with engineered environmentally responsive surface patterns, Nature, 1998, 393, 149–152.
24. H. Kawaguchi and K. Fujimoto, Smart latexes for bioseparation, Bioseparation, 1999, 7, 253–258.
25. A. Biffis, N. Orlandi and B. Corain, Microgel-stabilized metal nanoclusters: size control by microgel nanomorphology, Adv. Mater., 2003, 15, 1551–1555.
26. Y. Lu, Y. Mei, M. Drechsler and M. Ballauff, Thermosensitive core-shell particles as carriers for Ag nanoparticles: modulating the catalytic activity by a phase transition in networks, Angew. Chem., Int. Ed., 2006, 45, 813–816.
27. A. Biffis and L. Minati, Efficient aerobic oxidation of alcohols in water catalysed by microgel-stabilised metal nanoclusters, J. Catal., 2005, 236, 405–409.
28. D. Wang, D. Cheng, Y. Guan and Y. Zhang, Thermoreversible hydrogel for in situ generation and release of HepG2 spheroids, Biomacromolecules, 2011, 12, 578–584.
29. Z. Shen, J. Bi, B. Shi, D. Nguyen, C. J. Xian, H. Zhang and S. Dai, Exploring thermal reversible hydrogels for stem cell expansion in three-dimensions, Soft Matter, 2012, 8, 3250.
30. Z. Shen, K. Terao, Y. Maki, T. Dobashi, G. Ma and T. Yamamoto, Synthesis and phase behavior of aqueous poly(N-isopropylacrylamide-co-acrylamide), poly(N-isopropylacrylamide-co-N,N- dimethylacrylamide) and poly(N-isopropylacrylamide-co-2-hydroxylethyl methacrylate), Colloid Polym. Sci., 2006, 284, 1001–1007.
31. S. Dai, P. Ravi, C. H. Tan and K. C. Tam, Self-assembly behavior of a stimuli-responsive water-soluble fullerene-containing polymer, Langmuir, 2004, 20, 8569–8575.
32. Z. Shen, Y. Li, K. Kohama, B. Oneill and J. Bi, Improved drug targeting of cancer cells by utilizing actively targetable folic acid-conjugated albumin nanospheres, Pharmacol. Res., 2011, 63, 51–58.
33. Z. Shen, W. Wei, H. Tanaka, K. Kohama, G. Ma, T. Dobashi, Y. Maki, H. Wang, J. Bi and S. Dai, A galactosamine-mediated drug delivery carrier for targeted liver cancer therapy, Pharmacol. Res., 2011, 64, 410–419.
34. S. Hattori, J. D. Andrade, J. B. Hibbs, D. E. Gregonis and R. N. King, Fibroblast cell-proliferation on charged hydroxyethyl methacrylate copolymers, J. Colloid Interface Sci., 1985, 104, 72–78.
35. Z. Shen, B. Shi, H. Zhang, J. Bi and S. Dai, Exploring low-positively-charged thermosensitive copolymers as gene delivery vectors, Soft Matter, 2012, 8, 1385–1394.
36. T. Hoare and R. Pelton, Highly pH and temperature responsive microgels functionalized with vinylacetic acid, Macromolecules, 2004, 37, 2544–2550.
37. T. Gan, Y. Guan and Y. Zhang, Thermogelable PNIPAM microgel dispersion as 3D cell scaffold: effect of syneresis, J. Mater. Chem., 2010, 20, 5937–5944.
38. S. Crisp, B. G. Lewis and A. D. Wilson, Conductometric titration of aqueous solutions of polyacrylic acid and its copolymers, J. Dent. Res., 1975, 54, 1238.
39. P. Liu, P. Davis, H. Liu and T. R. Krishnan, Evaluation of cytotoxicity and absorption enhancing effects of melittin-a novel absorption enhancer, Eur. J. Pharm. Biopharm., 1999, 48, 85–87.
40. M. J. Bruining, H. G. T. Blaauwgeers, R. Kuijer, E. Pels, R. M. M. A. Nuijts and L. H. Koole, Biodegradable three-dimensional networks of poly(dimethylamino ethyl methacrylate). Synthesis, characterization and in vitro studies of structural degradation and cytotoxicity, Biomaterials, 2000, 21, 595–604.
41. A. Au, J. Ha, A. Polotsky, K. Krzyminski, A. Gutowska, D. S. Hungerford and C. G. Frondoza, Thermally reversible polymer gel for chondrocyte culture, J. Biomed. Mater. Res., Part A, 2003, 67, 1310–1319.
42. M. Negre, P. Leone, J. Trichet, C. Defarge, V. Boero and M. Gennari, Characterization of model soil colloids by cryo-scanning electron microscopy, Geoderma, 2004, 121, 1–16.
43. J. Lee, M. J. Cuddihy, G. M. Cater and N. A. Kotov, Engineering liver tissue spheroids with inverted colloidal crystal scaffolds, Biomaterials, 2009, 30, 4687–4694.
44. K. Smetana Jr, Cell biology of hydrogels, Biomaterials, 1993, 14, 1046–1050.
45. D. Liu, T. Wang, X. X. Liu and Z. Tong, Cell Proliferation and Cell Sheet Detachment from the Positively and Negatively Charged Nanocomposite Hydrogels, Biopolymers, 2014, 101, 58–65.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c4ra02852j

---