Polymer Thin Films with Tunable Acetylcholine-like Functionality Enable LongTerm Culture of Primary Hippocampal Neurons Seungyoon B. Yu,†,⊥ Jieung Baek,‡,⊥ Minsuk Choi,†,⊥ Youjin Oh,§ Hak Rae Lee,‡ Seung Jung Yu,‡ Eunjung Lee,‡ Jong-Woo Sohn,§ Sung Gap Im,*,‡ and Sangyong Jon*,† †
KAIST Institute for the BioCentury, Department of Biological Sciences, ‡KAIST Institute for the NanoCentury, Department of Chemical and Biomolecular Engineering, and §Department of Biological Sciences, Korea Advanced Institute of Science and Technology, 291 Daehak-ro, Daejeon 34141, Republic of Korea S Supporting Information *
ABSTRACT: In vitro culture systems for primary neurons have served as useful tools for neuroscience research. However, conventional in vitro culture methods are still plagued by challenging problems with respect to applications to neurodegenerative disease models or neuron-based biosensors and neural chips, which commonly require long-term culture of neural cells. These impediments highlight the necessity of developing a platform capable of sustaining neural activity over months. Here, we designed a series of polymeric bilayers composed of poly(glycidyl methacrylate) (pGMA) and poly(2-(dimethylamino)ethyl methacrylate) (pDMAEMA), designated pGMA:pDMAEMA, using initiated chemical vapor deposition (iCVD). Harnessing the surface-growing characteristics of iCVD polymer films, we were able to precisely engraft acetylcholine-like functionalities (tertiary amine and quaternary ammonium) onto cell culture plates. Notably, pGD3, a pGMA:pDMAEMA preparation with the highest surface composition of quaternary ammonium, fostered the most rapid outgrowth of neural cells. Clear contrasts in neural growth and survival between pGD3 and poly-L-lysine (PLL)-coated surfaces became apparent after 30 days in vitro (DIV). Moreover, brain-derived neurotrophic factor level continuously accumulated in pGD3-cultured neurons, reaching a 3-fold increase at 50 DIV. Electrophysiological measurements at 30 DIV revealed that the pGD3 surface not only promoted healthy maturation of hippocampal neurons but also enhanced the function of hippocampal ionotropic glutamate receptors in response to synaptic glutamate release. Neurons cultured long-term on pGD3 also maintained their characteristic depolarization-induced Ca2+ influx functions. Furthermore, primary hippocampal neurons cultured on pGD3 showed longterm survival in a stable state up to 90 daysfar longer than neurons on conventional PLL-coated surfaces. Taken together, our findings indicate that a polymer thin film with optimal acetylcholine-like functionality enables a long-term culture and survival of primary neurons. KEYWORDS: neuron culture, polymer thin films, initiated chemical vapor deposition, iCVD, acetylcholine, brain-derived neurotrophic factor, BDNF approaches has remained an elusive goal.3 Most primary neurons cultured on TCPS coated with PLL or other materials become degraded and die within 1 month of culture at most. However, more demanding applications, such as potential uses in models of neurodegenerative disease (e.g., Alzheimer’s and Parkinson’s disease) or utilization in neuron-based biosensors and neural chips, commonly require substantially longer-term
In vitro culture systems for primary neural cells have served as useful tools for neuroscience research as they provide easy access for examination and manipulations of living neurons. In conventional use, poly-L-lysine (PLL) and extracellular matrix (ECM) proteins have been widely employed as surface coatings for tissue culture polystyrene (TCPS) plates.1−3 Various neurotrophic factors added during culture have also been shown to further facilitate active neurogenesis and neural plasticity.4,5 However, these conventional methods still pose a challenge for the reliable, long-term culture of neural cells, and achieving survival for more than a month using these © 2016 American Chemical Society
Received: May 27, 2016 Accepted: October 28, 2016 Published: October 28, 2016 9909
DOI: 10.1021/acsnano.6b03527 ACS Nano 2016, 10, 9909−9918
Article
www.acsnano.org
Article
ACS Nano
Figure 1. Polymerization concept of a surface using initiated chemical vapor deposition (iCVD). (a) Chemical structure of acetylcholine, DMAEMA, and GMA and the conjugation reaction between the tertiary amine group in DMAEMA and epoxy group in GMA. (b) Schematic illustration of the procedure for fabricating the polymer-coated surface through sequential introduction of GMA and DMAEMA in the iCVD chamber and its application to the culture of hippocampal neurons.
culture of neural cells.6−13 For example, for the construction of in vitro neurodegenerative disease models, neural cells are transfected with viruses carrying disease-initiating mutant genes, followed by long-term culture in a stable, healthy state until the signature of the disease onset is detected, which usually requires far more than a month.6,9 In addition, there has been considerable interest in interfacing neural cells with devices for application in neuron-based biosensors and neural prostheses, such as neural chips. A key to the success of neural−interface devices is establishing a method for maintaining the activity and survival of neurons in vitro for as long as possible.10−12 In this regard, developing a culture system that enables long-term culture and survival of primary neurons over months is imperative to satisfy the unmet need for these potential applications. Specific surfaces with neuron-friendly functionalities have been constructed as substitutes for PLL or ECM components to enhance the viability of neural cells in vitro.14−19 Specifically, 2-(dimethylamino)ethyl methacrylate (DMAEMA)16−19 has garnered research interest due to its similarity in the chemical structure of the neurotransmitter acetylcholine (Figure 1a). Moreover, the methacrylate moiety in DMAEMA enables the molecules to form a polymer via a well-known free-radical polymerization reaction.20 Thus, DMAEMA-incorporated biomimetic polymeric surfaces have been developed for use in primary neural cell culture and shown to accelerate neurite outgrowth compared to the conventional PLL-based culture system.16 However, none of the previous DMAEMA-incorporated polymers has been shown to support long-term survival of primary neural cells. Hydrogels with a tethered 2-acetoxyN,N,N-trimethylethanaminium segmentan acetylcholine-like functionalitywere recently reported.18 This quaternary ammonium functionality more closely mimics acetylcholine than tertiary amine-containing DMAEMA for use as a potential biomaterial in neural engineering. However, most hydrogels harboring the quaternary ammonium were found to be highly toxic toward neural cells;16,21,22 only hydrogels with a low fraction of the acetylcholine-like moiety (∼14%) showed good performance in neural cell culture.18 This observation strongly suggests that the chemical nature of the acetylcholine-like
functionality must be appropriately tuned for successful primary neuron culture. Based on these previous findings, we hypothesized that a composite surface of tertiary amine (pseudo-acetylcholine-like) and quaternary ammonium (acetylcholine-like) with the proper composition might provide an optimal environment for neurons. Here, we report rationally designed polymer thin films with a specific neuron-friendly functionality that enable accelerated neurite outgrowth during neural development. Surprisingly, primary hippocampal neurons cultured on the developed functional polymer surface showed long-term survival, maintaining a stable and healthy state for up to 90 daysfar longer than neurons cultured on conventional PLL surfaces.
RESULTS AND DISCUSSION The polymer thin films harboring a neuron-friendly functionality were constructed from the key functional monomers, glycidyl methacrylate (GMA)23 and DMAEMA (Figure 1a), using a vapor-phase polymerization method, initiated chemical vapor deposition (iCVD) process.23−26 First, an epoxycontaining pGMA thin film was deposited on the TCPS surface. Poly(DMAEMA) was successively deposited onto the pGMA layers, generating the grafted polymeric bilayers of pGMA:pDMAEMA, designated pGD (Figure 1b). In general, in the iCVD process, a polymeric layer grows on arbitrary substrates based on a surface-growth mechanism. Therefore, the method enables grafting of two successively deposited polymeric layers through the linkage of vinyl groups, forming covalent bonds between the two layers that greatly stabilize the bilayer.24,27,28 Thus, a bilayer of pGMA and pDMAEMA could be obtained through the vinyl-group-mediated grafting. In addition, we also expected that part of the tertiary amine moiety in the pDMAEMA layer would react with an epoxide in the pre-existing pGMA layer upon iCVD-mediated polymerization, thereby generating an acetylcholine-like quaternary ammonium functionality,29 as shown in Figure 1a. Furthermore, the polymeric adlayer comprising pDMAEMA could be firmly grafted through the stable covalent bond between the two polymeric layers. Because the amount of pDMAEMA film 9910
DOI: 10.1021/acsnano.6b03527 ACS Nano 2016, 10, 9909−9918
Article
ACS Nano
Figure 2. Composition of poly(GMA-DMAEMA) surfaces and their effects on neural growth. (a) XPS spectra of survey scan (left) and N 1s high-resolution scan (right) of pGMA, pGD1, pGD2, pGD3, pGD4, and pDMAEMA. (b) Surface atomic composition of tertiary N (black) and quaternary N+ (red) of each polymer film and the dendritic length (gray bar) of primary hippocampal neurons cultured on the surfacemodified plates, measured at 8 DIV. (c) Images of neural cells on the surfaces immunostained for MAP2 (green) and costained with the nuclear dye DAPI (blue) at 8 DIV. Scale bars, 50 μm (applies to all images).
deposited onto the pre-existing pGMA film can be tuned precisely by simply controlling the reaction time, we were able to construct a series of pGD thin films with various ratios of the two acetylcholine-like functionalities: a tertiary amine of DMAEMA (pseudo-acetylcholine-like) and a quaternary ammonium (acetylcholine-like). Thus, delicate control on pDMAEMA thickness less than ∼100 nm enabled us to obtain an optimized surface composition of two functionalities on the surface for neural cell culture. A series of pGD films with four different pDMAEMA compositions were prepared by controlling the pDMAEMA deposition time as follows: pGD1, 60 s; pGD2, 180 s; pGD3, 360 s; and pGD4, 600 s. The synthesized pGD films were characterized using various spectroscopic methods, including Fourier transform infrared (FT-IR) spectroscopy, X-ray photoelectron spectroscopy (XPS), and atomic force microscopy (AFM) (Supplementary Figure S1). The peaks for the CC group30 at 1640 cm−1 in the spectra of GMA and DMAEMA monomers disappeared in the spectrum of a pGD prepared with a DMAEMA deposition time of 360 s, indicating successful radical polymerization in the iCVD chamber. Figure 2a shows an XPS survey scan of each surface. The fraction of pDMAEMA coverage in each pGD film was quantitatively estimated by measuring the atomic composition of C, O, and N (Supplementary Figure S2), which revealed that the composition of the DMAEMA functionality increased from 0.11 for pGD1 to 0.85 for pGD4. As expected, the static water contact angle also decreased with the increase in pDMAEMA coverage, indicating the increased hydrophilicity of the pGD film with increasing incorporation of the hydrophilic DMAEMA moiety into the pGD copolymer (Supplementary Figure S3a). AFM surface images showed that the roughness of the pGD3 film lay between that of the two homopolymer films, pGMA and
pDMAEMA, without any apparent grain-like morphology or phase segregation (Supplementary Figure S3b). Notably, the roughness of the pGD3 film was extremely low, with a rootmean-square (rms) roughness less than 1 nm. Because the surface composition ratio of neutral tertiary amine (N) and charged quaternary ammonium (N+) in pGD films could be a key determinant of successful neural cell culture, we compared N and N+ values for all four pGD films with those of the two control homopolymers, pGMA and pDMAEMA (Figure 2b). As expected, no N+ peak was detected in the spectrum of pGMA or pDMAEMA homopolymers, indicating that the peak corresponding to the quaternary ammonium resulted from the amine−epoxy reaction at the interface between pGMA and pDMAEMA layers. Whereas the surface composition of neutral tertiary amine (N) increased with increasing coverage of pDMAEMA in pGD films, that of the positively charged quaternary ammonium (N+) increased in pGD3 but decreased in pGD4. It would be reasonable to hypothesize that the actual N+ composition exposed in the outermost surface of pGD4 was masked by the excess pDMAEMA adlayer, thereby resulting in less exposure of the quaternary ammonium in pGD4 compared to pGD3. The four different polymeric films with acetylcholine-like functionality (pGD1−4) were tested for their ability to support the adhesion and growth of primary hippocampal neurons. Figure 2c shows immunohistochemical images of neural cells cultured on the polymeric surfaces at 8 days in vitro (DIV). The average length of dendrites in cells on each pGD surface was measured and quantified, as shown in Figure 2b (gray bar graph). Whereas neural cells on pGD1 did not adhere or grow well, those on pGD2−4 showed normal growth behavior, including adhesion and neurite outgrowth. Among pGD films, pGD3, with the highest N+ functionality, showed the best 9911
DOI: 10.1021/acsnano.6b03527 ACS Nano 2016, 10, 9909−9918
Article
ACS Nano
Figure 3. Short-term neural growth accelerations in vitro. (a) Confocal microscopy images of neurons immunostained for MAP2 (green) and Tau (red) and costained with the nuclear dye DRAQ5 (blue). Neurons were seeded on PLL and PGD3 at low density (1 × 104 cells/well) for single-cell observations. Cells were grown in vitro for 2, 4, 6, and 8 days. Scale bars, 50 μm. (b,c) Quantitative analysis of average neurite number per neuron (b) and mean axonal length (c). *p < 0.05, ***p < 0.0001, student’s t-test. (d) Neurons classified into five developmental stages at 2 and 4 DIV. The numbers indicate data points for the statistics.
and stained for confocal microscopy imaging analysis every 2 days. It has been shown that neural culture in vitro for 8 days is long enough to observe neuronal growth stage transitions, neurite outgrowth, and polarization into axon and dendrites.13,16,31 It is generally known that MAP2 (microtubuleassociated protein 2) and Tuj1 (neuron-specific class III βtubulin) are usually expressed in neural soma and dendrites, whereas Tau is more prevalent in axons in the early development of neurons.32,33 Confocal microscopy images of neurons cultured on either pGD3 or PLL with MAP2 and Tuj1 staining showed that neurons on pGD3 showed more extensive protrusion of their dendrites and axonal projection than those on PLL at all time points examined (Figure 3a and Supplementary Figure S5a). Already at 2 DIV, neurons on pGD3 displayed clear distinctions in axonal length and expression of MAP2, indicating a neuronal growth spurt. Furthermore, MAP2 and Tau expression at 8 DIV were clearly detected and distributed in the neuronal parts of pGD3-
performance in terms of neurite outgrowth ability (Figure 2b,c). Unlike the case for pGD films, primary neurons did not adhere or grow at all on pGMA and pDMAEMA homopolymers (Supplementary Figure S4a,b). Interestingly, it turned out that pDMAEMA with 100% coverage of the tertiary amine (pseudo-acetylcholine-like functionality) was not an appropriate surface for neural cell culture. These results clearly suggest that neural adhesion and growth are highly dependent on the composition of both tertiary amine and quaternary ammonium functionalities, and an optimal point in the composition does exist for the best neural adhesion and outgrowth. Next, we examined various aspects of the best-performing pGD3 film to test its suitability as a culture surface for primary neural cells. For comparison, conventional PLL-coated glass was used as a positive control. Hippocampal neurons were seeded at a low cell density (1 × 104 cells/mL) on glass substrates coated with PLL or pGD3 film and were then fixed 9912
DOI: 10.1021/acsnano.6b03527 ACS Nano 2016, 10, 9909−9918
Article
ACS Nano cultured neural cells. In contrast, axonal and dendritic microtubules in neurons grown on a PLL-coated surface were not yet fully polarized, with less broadly extended neurites.13,34 Similarly, expression of NeuN (neuronal nuclear antigen) became apparent in the neural cells on pGD3 at 4 DIV, whereas its expression on PLL-cultured neurons began to be detected at 6 DIV (Supplementary Figure S5b). This suggests that neurons grown on pGD3 entered the postmitotic stage earlier than those on PLL. Further culture until 10 DIV revealed that both surfaces were capable of serving as low-density neural culture platforms on which primary neurons could develop into fully mature neural cells, although neurons grown on pGD3 surfaces reached maturity sooner (Supplementary Figure S6). For more precise comparison, we carried out a quantitative analysis of the numbers and lengths of neurites (Figure 3b). In accord with confocal microscopy images, pGD3-cultured neurons showed appreciable acceleration of neurite and axonal outgrowth and greater neurite branching than neurons on PLL-coated control surface, even at early DIV. To better illustrate the difference in general growth tendency between the two surfaces, we categorized neurons at 2 DIV and 4 DIV on the basis of their developmental stages. Neuronal developmental can be classified into five stages: stage 1, lamellipodia appear around the neuronal soma to aid attachment to the surface; stage 2, a few 20−30 μm long neurites spread out; stage 3, one neurite overextends and forms an axon; stages 4 and 5, dendrite branches stretch and further produce spines, respectively.31 The general population of neurons cultured on pGD3 entered advanced developmental stages much faster than those grown on PLL. This was especially evident at 4 DIV, when ∼50% of pGD3-grown neurons reached stage 4, but only ∼30% of PLLgrown neurons had reached the same stage, and a majority was at earlier developmental stages. These data clearly indicate that the pGD3 surface is superior to conventional PLL in supporting primary neurons and accelerating neuronal growth, even at low cell densities. Surfaces that enable long-term culture of primary neural cells over months could be utilized for various applications, ranging from neurodegenerative disease models to neuron-interfaced devices. Thus, we further explored the suitability of the pGD3 film for culturing primary neural cells for long time periods while maintaining their activity and stability. For this experiment, we seeded cells at a somewhat higher density (5 × 104 to 1 × 105 cells/mL) to allow neurons to form synapses and send signals to other neurons. Confocal microscopy images revealed clear differences in neural growth and survival between pGD3 and PLL-coated surfaces that became apparent at 30 DIV and grew more distinct with increasing culture time from 50 to 80 DIV (Figure 4a). Whereas PLL-cultured neurons exhibited many clusters of nuclear-stained cells (blue) with only a few neural-marker-positive cells (green), the pGD3 surface was well populated with evenly distributed neural-markerpositive cells. By 80 DIV, most PLL-cultured neurons seemed to have lost their characteristic neuronal morphologies and MAP2 expression and showed cleavage of dendrites and axons, suggesting that microtubule degeneration had taken place. In contrast, most pGD3-cultured neurons remained healthy with high levels of MAP2 expression and maintained their neuronlike morphologies; dendrites of single neurons spread into all directions, forming an extensive network with other neurons. Moreover, it should be noted that pGD3-cultured neurons at 90 DIV still retained their morphologies and relative positions, interacting with each other while strongly expressing MAP2
Figure 4. Neural network morphologies distinguished after 30 DIV. (a) Immunocytochemical images of neurons grown on pGD3 (left) and PLL (right) surfaces for 30, 50, and 80 days. (b) Neural network maintained for 90 DIV. Cells were immunostained for MAP2 (green) and costained with the nuclear dye DAPI (blue). Scale bars, 200 μm. (c) Relative brain-derived neurotrophic factor mRNA expression levels in neurons cultured for 10, 30, and 50 DIV were analyzed by qRT-PCR.
(Figure 4b). To our knowledge, conventional PLL-based in vitro cultures of neurons are sustainable up to 8 weeks at maximum,35 which is not sufficient to observe neural aging and degeneration. In this sense, our platform pGD3, extending the culture period to more than 12 weeks, yielded results that could provide insight into mature neurons. The axonal microtubule-associated protein Tau, one of the most important factors in the etiology of Alzheimer disease, becomes hyperphosphorylated and entangled with other Tau proteins to form plaques as the disease progresses, leading to neurodegeneration.6,34,36 Accordingly, we assessed Tau expression in both culture surfaces at 90 DIV (Supplementary Figure S7). Neurons grown on PLL barely expressed Tau and exhibited a deteriorated neuronal morphology, whereas Tau expression was readily detected in neurons grown on pGD3. Interestingly, Tau expression in pGD3-cultured neurons was not appropriately associated with axons, suggesting that longterm culture on pGD3 is accompanied by microtubule degeneration. Thus, the present culture system may have potential for use in preparing in vitro neurodegenerative disease models. The morphologies of neuronal networks cultured on pGD3, with their wide-ranging dendritic branches and axonal stretching, resembled the arrangement of neurons treated with neurotrophic factors.37,38 The best-known and most extensively expressed factor involved in neural circuit development is brain-derived neurotrophic factor (BDNF). Accordingly, we evaluated expression levels of BDNF in pGD3cultured neurons at 10, 30, and 50 DIV. As expected, BDNF mRNA levels did not increase in neurons cultured on PLL. In stark contrast, BDNF mRNA in pGD3-cultured neurons continuously accumulated over time, reaching a 3-fold increase at 50 DIV (Figure 4c), a finding previously reported only for culture surfaces treated with neurotropic factors.39 BDNF expression is known to decrease with neural age and thus can cause impairment of connections that could lead to neural disorders.40 Although the mechanism by which the acetylcholine-like functionality on pGD3 induces upregulation of BDNF 9913
DOI: 10.1021/acsnano.6b03527 ACS Nano 2016, 10, 9909−9918
Article
ACS Nano
Figure 5. Electrophysiological properties of hippocampal neurons cultured on PLL and pGD3. (a) Whole cell patch clamp recordings of membrane potential in the current clamp mode demonstrate the RMP and AP frequency of hippocampal neurons cultured on PLL (left) and pGD3 (right) at 30 DIV. (b) Summary of RMP from neurons cultured on PLL and pGD3 at 15 and 30 DIV. (c) Summary of AP frequency from neurons cultured on PLL and pGD3 at 15 and 30 DIV. (d) Whole cell patch clamp recordings in the voltage clamp mode at a holding potential of −60 mV demonstrate sEPSCs recorded from hippocampal neurons cultured on PLL (left) and pGD3 (right) at 30 DIV. (e) Summary of sEPSC frequency onto neurons cultured on PLL and pGD3 at 15 and 30 DIV. (f) Summary of sEPSC mean amplitude onto neurons cultured on PLL and pGD3 at 15 and 30 DIV. *p < 0.05, unpaired t-test; n.s., not significant.
Figure 6. Neural depolarization characteristics. (a) Ca2+ influx was detected using Fluo-4 before and after treatment with a high KCl (50 mM) concentration. Scale bars, 50 μm. (b) Relative changes in Ca2+ influx revealed by the change in Fluo-4 fluorescence intensity (ΔF/F0) recorded every 10 s for 5 min before and after KCl treatment.
between two culture surfaces (Supplementary Figure S8), suggesting that the neural−glial interactions may not be directly linked to the long-term neuronal culture performance of pGD3 surface. Next we compared electrophysiological properties of neurons cultured on PLL and pGD3 (Figure 5). At 30 DIV, hippocampal neurons on PLL had baseline or resting membrane potential (RMP) of −52.8 ± 0.9 mV (n = 4), but
remains unclear and needs further study, the enhanced production of BDNF apparently enables long-term culture and increases the survival rate of primary neurons on the pGD3 surface. We also examined the effect of hippocampal glial cells coexisting with neuron cells on the performance of PLL and pGD3 surfaces. Immunostaining for glial cells using a glial fibrillary acidic protein (GFAP)-specific antibody showed no appreciable difference in overall population of glial cells 9914
DOI: 10.1021/acsnano.6b03527 ACS Nano 2016, 10, 9909−9918
Article
ACS Nano
like), resulting from the tertiary amine−epoxy reaction. The polymer thin film with the highest concentration of acetylcholine-like functionality among the polymer films, designated pGD3, showed the best performance, fostering acceleration of neuronal growth and retention of neuronal activity over 90 days, a result far superior to that achieved using a conventional PLL surface. The pGD3 film may ultimately have applications in various fields ranging from basic neuroscience to neuron− device interface applications, such as neuroprosthetics and biosensors. Because the pGD3-based culture platform maintains long-term neuronal functionality, we anticipate that it would enable natural neural aging to be observed in vitro and support research on degenerative diseases. Furthermore, beyond a focus on building in vitro models, the pGD3 surface could be utilized as an interface in devices such as neuron-based biosensors and neural chips. Further investigations into the key determinants responsible for the performance of the present polymer surface are currently underway to better understand the underlying mechanism.
neurons cultured on pGD3 had significantly hyperpolarized RMP of −66.5 ± 3.6 mV (n = 4, p = 0.010) (Figure 5a,b). At 15 DIV, by contrast, there was no significant difference between RMP of neurons cultured on PLL and pGD3 (Figure 5b). It was also noted that RMP of neurons cultured on PLL remained unchanged between 15 DIV and 30 DIV (Figure 5b). Given that mature hippocampal neurons have more hyperpolarized RMP compared to immature neurons,41−44 it is proposed that pGD3, compared to PLL, has beneficial effects of promoting healthy maturation of hippocampal neurons in primary culture. As expected from hyperpolarized RMP, hippocampal neurons on pGD3 fired significantly (p = 0.046) smaller number of action potentials (APs) (0.2 ± 0.1 Hz, n = 4) compared to neurons cultured on PLL (0.7 ± 0.2 Hz, n = 4) (Figure 5a,c). AP frequency was not different between pGD3 and PLL groups at 15 DIV (Figure 5c). We also evaluated and compared synaptic function of hippocampal neurons cultured on PLL and pGD3 by measuring spontaneous excitatory postsynaptic currents (sEPSCs) in the voltage clamp mode at a holding potential of −60 mV. We found that sEPSC frequency was not significantly different between neurons cultured on PLL and pGD3 at either 15 or 30 DIV (Figure 5d,e). However, it was noted that at 30 DIV, the mean amplitude of sEPSC was significantly larger in neurons cultured on pGD3 (200.9 ± 38.6 pA, n = 4) compared to those on PLL (96.3 ± 21.8 pA, n = 5, p = 0.041) (Figure 5d,f). We also noted that the mean amplitude of sEPSC was significantly increased between 15 DIV and 30 DIV when cultured on pGD3, while it remained unchanged on PLL (Figure 5f). These results suggest that pGD3 helps to enhance the function of hippocampal ionotropic glutamate receptors (e.g., AMPA receptors) in response to synaptic glutamate release. We also examined whether long-term-cultured neurons retained their characteristic functions. Exploiting the fact that a sudden change in extracellular K+ concentration depolarizes the cell membrane of neurons and induces rapid Ca2+ signals that propagate into neuronal networks, we measured the Ca2+ influx response to application of KCl as an indicator of neuronal activity. Application of a high concentration of KCl (50 mM) caused an immediate surge of Ca2+ influx in both pGD3- and PLL-cultured neurons (Figure 6a). However, the magnitude of the local Ca2+ increase, measured as the mean change in fluorescence of the Ca2+ indicator dye Fluo-4 relative to baseline fluorescence (ΔF/F0), was much greater and more sustained in neurons grown on pGD3 than those grown on PLL (Figure 6b), where the intensity of signals rapidly diminished within 4 min of KCl application. At the moderately elevated KCl concentration of 5 mM (∼2× the normal concentration of extracellular K+), an appreciable increase in Ca2+ influx was detected in the case of pGD3, whereas no appreciable signal change was observed in neurons grown on PLL (Supplementary Figure S9). These results indicate that neurons cultured on a pGD3 surface, unlike those on PLL, are able to maintain normal neuronal functions and remain healthy, even after long-term in vitro culture.
MATERIALS AND METHODS Deposition of the Polymers via iCVD. Cover glasses or cell culture dishes were placed in the iCVD chamber. Then, preheated (40 °C) GMA (Aldrich, 97%), as a monomer, and tert-butyl peroxide (TBPO) (Aldrich, 98%), as an initiator, were introduced into the iCVD chamber at a volumetric flow rate ratio of GMA/TBPO = 2:1 (in sccm), to yield an underlying pGMA polymeric layer with a thickness of 100 nm. The pressure inside the chamber was set to 160 mTorr. The pGMA film was formed successfully with a deposition rate of 10 nm/60 s. After pGMA fabrication, the process pressure was set to 120 mTorr. Then, without breaking the vacuum, DMAEMA (Tokyo Chemical Industry Co., 98.5%) and TBPO were introduced into the chamber, also at a volumetric flow rate ratio of 2:1 (in sccm), to obtain partial coverage of the ultrathin pDMAEMA on the pGMA surface. The pDMAEMA surface was deposited at a rate of 2 nm/60 s. For optimization of partial coverage, the process time of pDMAEMA deposition was set to 0, 60, 180, 360, and 600 s, yielding pGD1, pGD2, pGD3, and pGD4, respectively. The temperatures of filament and substrate were kept at 180 and 38 °C, respectively, throughout the deposition process. Characterization of iCVD-Deposited Polymer. Polymerization of pGMA, pDMAEMA, and pGD3 was confirmed by FT-IR spectroscopy using an ALPHA FT-IR spectrometer (Bruker Optics, USA) in normal transmission mode. The surface chemical composition of iCVD polymer films was analyzed by XPS (Sigma Probe, Thermo VG Scientific Inc.), and surface morphologies of the polymer films were assessed using AFM (psia XE-100). The static contact angle was measured with a contact angle analyzer (Phoenix 150, SEO, Inc.) using 5 μL of deionized water droplets. Hippocampal Dissection from Embryonic Rat Brain. Primary hippocampal neurons were prepared from E18 Sprague−Dawley rats (Charles River Technology), as reported previously.2 Briefly, a pregnant rat bearing 18-day-old embryos was euthanized, and the uterus was removed, washed, and transferred to a dish containing HBSS (Hank’s balanced salt solution, Sigma) for separation of embryos. Embryonic tissues were kept submerged at all times during the process. Brains were dissected out of the embryos, and hippocampi were removed. A trypsin/EDTA solution (0.05% trypsin, 0.53 mM EDTA in HBSS; Welgene) was then added, and the tube was incubated in a 37 °C water bath for 30 min. In some cases, DNase I was added to prevent clumping of cells. Next, hippocampal tissues were dissociated by triturating first with a Pasteur pipet and then with a narrow-bore pipet for further dissociation into single cells. Cell density was determined by cell counting after no tissue pieces were visible in the tube. Primary Neuron Culture on Polymers. Before seeding hippocampal neural cells, 18 mm coverslips were placed in 12-well plates,
CONCLUSION In summary, we have reported a type of polymer thin film that enables long-term culture and survival of primary neurons. The iCVD process enables the synthesis of a series of polymer thin films with a precisely tunable composition of two acetylcholinelike functionalities: a tertiary amine of DMAEMA (pseudoacetylcholine-like) and a quaternary ammonium (acetylcholine9915
DOI: 10.1021/acsnano.6b03527 ACS Nano 2016, 10, 9909−9918
Article
ACS Nano
Calcium Imaging and Analysis. Neuron samples on 18 mm round coverslips were incubated at 37 °C for 30 min with 2 mL of Tyrode’s solution diluted with a 2 mM mixture of the Ca2+ indicator Fluo-4 AM and Pluronic F-127. After incubation and washing with normal Tyrode’s solution, Ca2+ in the neuronal cytoplasm was automatically imaged using a confocal microscope equipped with a 488 nm excitation filter; serial time-lapse images were taken every 10 s. Cells were treated with Tyrode’s solution containing high (50 mM) or low (5 mM) KCl after 5 min of time-lapse recording to instantly induce neuronal depolarization. Ca2+ influx via voltage-dependent Ca2+ channels was then imaged for an additional 5 min. Fluo-4 AM fluorescence was imaged using a Nikon A1R confocal microscope, and acquired images were analyzed using NIS Elements software.
and a 1 mg/mL solution of PLL (Sigma-Aldrich), dissolved in 0.1 M borate buffer (pH 8.5), was dropped onto the glass. PLL-covered coverslips were allowed to stand for 12−24 h at room temperature. After the PLL solution was removed, the PLL-coated coverslip was sterilized. Neural cells were then seeded on both PLL- and iCVD polymer-coated surfaces in neuronal plating medium (MEM with Earle’s salts, L-glutamine [Invitrogen], and 0.6% D-glucose) containing fetal bovine serum and amino acids. Cell seeding density ranged from 1 × 104 to 1 × 105 cells/well. Neurons were maintained at 37 °C in a humidified 5% CO2 incubator for more than 6 h before changing the medium to neurobasal maintenance medium (Gibco) containing GlutaMAX-I (Gibco) and B27 supplement (Gibco). The maintenance medium is preferable for long-term culture of neurons since it does not include serum glutamate. Maintenance medium was refreshed by replacing ∼30% of the medium volume every 6 days with fresh medium. Immunocytochemical Staining of Primary Neurons. Neural cells grown on coverslips were fixed with 4% paraformaldehyde (Sigma-Aldrich), washed with Dulbecco’s phosphate-buffered saline (DPBS), and permeabilized by treating with 0.25% Triton X-100 for 5 min, followed by additional washing. Blocking proved to be unnecessary. Primary antibodies for MAP2, Tau, Tuj1, GFAP (1:500; Santa Cruz Biotechnology), and NeuN (1:100; Millipore) were mixed in PBS containing 3% bovine serum albumin (BSA) and incubated for 1 h at room temperature or overnight at 4 °C. After appropriate washing, cells were incubated for 5 min with secondary antibodies (1:2000 anti-Ms Rhodamine, 1:1000 anti-Rb FITC; Santa Cruz Technology), dissolved in PBS/3% BSA, and then incubated with DRAQ5 to stain nuclei. Cells were rinsed three times with DPBS, then soaked in distilled water, dried, mounted onto slide glass with mounting medium (Sigma-Aldrich), and imaged by confocal or fluorescence microscopy. Statistical Analysis. Acquired confocal and fluorescence microscopy images were analyzed using the ImageJ plugin NeuronJ. Neurite numbers and lengths were analyzed in randomly selected fields. Neurotrophic Factor mRNA Levels. About 6 × 105 cells were detached by treating with lysis buffer (GeneAll) for 5 min. Total neuronal RNA (30 ng/μL for each sample) was isolated using an RNA Prep procedure. cDNA was prepared from total RNA by reverse transcription (RT) with QIAGEN Rotor-Gene using AccuPower RT PreMix (Bioneer) and the following primer pairs: BDNF, 5′-AGC TGA GCG TGT GTG ACA GT-3′ (forward) and 5′-ACC CAT GGG ATT ACA CTT GG-3′ (reverse); and β-actin, 5′-GGG AAA TCG TGC GTG ACA TT-3′ (forward) and 5′-CGG ATG TCA ACG TCA CAC TT-3′ (reverse). Quantitative real-time RT-PCR (qRT-PCR) was carried out with 50 ng of neuronal RNA per reaction. Fold changes in BDNF mRNA levels relative to those of the endogenous control β-actin were determined by analysis of SYBR green fluorescence using the ΔΔCT method. Electrophysiology. Coverslips that contain cultured hippocampal neurons were carefully placed on the recording chamber of a Nikon Eclipse FN1 microscope equipped with a fixed stage and an optiMOS scientific CMOS camera. The neurons were bathed and perfused with normal Tyrode’s solution (32−34 °C) that contains 150 mM NaCl, 2.5 mM KCl, 2 mM CaCl2, 1 mM MgCl2, 10 mM glucose, and 10 mM HEPES, pH 7.4. Neurons were visualized using infrared differential interference contrast imaging to obtain the whole cell recording. Electrophysiological signals were recorded using an Axopatch 700B amplifier (Molecular Devices), low-pass filtered at 2−5 kHz, and analyzed offline on a PC with pCLAMP (Molecular Devices) and Mini Analysis Programs (Synaptosoft). Recording electrodes were pulled from glass microcapillaries and had resistances of 2.5−5 MΩ when filled with potassium-based internal solutions that contain 120 mM Kgluconate, 10 mM KCl, 10 mM HEPES, 5 mM EGTA, 1 mM CaCl2, 1 mM MgCl2, and 2 mM MgATP, pH 7.3. Upon patch break in, recordings of sEPSCs were performed for ∼2 min at a holding potential of −60 mV in voltage clamp mode. Subsequently, baseline or RMP and action potential was recorded in current clamp mode for ∼5 min.
ASSOCIATED CONTENT S Supporting Information *
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsnano.6b03527. FT-IR spectra of GMA, DMAEMA, pGMA, pDMAEMA, and pGD (Figure S1); atomic composition and partial coverage of pDMAEMA on pGMA caculated from XPS (Figure S2); water contact angles and AFM images of the polymeric surfaces (Figure S3); optical microscopy images of hippocampal neural cells on the polymeric surfaces (Figure S4); confocal microscopy images of neurons immunostained for Tuj1 and NeuN (Figure S5); confocal microscopy images of single neurons grown for 10 DIV (Figure S6); neural network morphology and tau expression on 90 DIV (Figure S7); Tuj1 and GFAP expression in neurons and glial cells (Figure S8); and neuronal Ca2+ influx with subtle KCl stimuli (Figure S9) (PDF)
AUTHOR INFORMATION Corresponding Authors
*E-mail:
[email protected]. *E-mail:
[email protected]. Author Contributions ⊥
S.B.Y., J.B., and M.C. contributed equally to this work.
Author Contributions
The manuscript was written through contributions of all authors. S.J. and S.G.I. conceived the project; S.B.Y., J.B., and M.C. performed all of the experiments; H.L., S.Y., and E.L. provided polymers; S.B.Y., J.B., M.C., Y.O., J.-W.S., S.G.I., and S.J. analyzed the results; and S.B.Y., J.B., M.C., S.G.I., and S.J. wrote the manuscript. Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS This work was supported by the Korea Health Technology R&D Project (Grant No. HI13C21810301) through the Korea Health Industry Development Institute (HHIDI) funded by the Ministry of Health & Welfare, Republic of Korea, and the National Research Foundation of Korea (NRF) grants funded by the Ministry of Science, ICT & Future Planning (MSIP) of Korea under Contract Number NRF-2015M1A2A2056542 (climate change program). REFERENCES (1) Brewer, G. J.; Torricelli, J. R. Isolation and Culture of Adult Neurons and Neurospheres. Nat. Protoc. 2007, 2, 1490−1498. 9916
DOI: 10.1021/acsnano.6b03527 ACS Nano 2016, 10, 9909−9918
Article
ACS Nano (2) Kaech, S.; Banker, G. Culturing Hippocampal Neurons. Nat. Protoc. 2006, 1, 2406−2415. (3) Bertrand, S. J.; Aksenova, M. V.; Aksenov, M. Y.; Mactutus, C. F.; Booze, R. M. Endogenous Amyloidogenesis in Long-Term Rat Hippocampal Cell Cultures. BMC Neurosci. 2011, 12, 38. (4) Lu, B.; Nagappan, G.; Guan, X.; Nathan, P. J.; Wren, P. BDNFBased Synaptic Repair as a Disease-Modifying Strategy for Neurodegenerative Diseases. Nat. Rev. Neurosci. 2013, 14, 401−416. (5) Cardenas-Aguayo, M. C.; Kazim, S. F.; Grundke-Iqbal, I.; Iqbal, K. Neurogenic and Neurotrophic Effects of BDNF Peptides in Mouse Hippocampal Primary Neuronal Cell Cultures. PLoS One 2013, 8, e53596. (6) Choi, S. H.; Kim, Y. H.; Hebisch, M.; Sliwinski, C.; Lee, S.; D’Avanzo, C.; Chen, H. C.; Hooli, B.; Asselin, C.; Muffat, J.; Klee, J. B.; Zhang, C.; Wainger, B. J.; Peitz, M.; Kovacs, D. M.; Woolf, C. J.; Wagner, S. L.; Tanzi, R. E.; Kim, D. Y. A Three-Dimensional Human Neural Cell Culture Model of Alzheimer’s Disease. Nature 2014, 515, 274−293. (7) Zhou, W.; Dai, X.; Fu, T.-M.; Xie, C.; Liu, J.; Lieber, C. M. Long Term Stability of Nanowire Nanoelectronics in Physiological Environments. Nano Lett. 2014, 14, 1614−1619. (8) Haughey, N. J.; Nath, A.; Chan, S. L.; Borchard, A. C.; Rao, M. S.; Mattson, M. P. Disruption of Neurogenesis by Amyloid Beta-Peptide, and Perturbed Neural Progenitor Cell Homeostasis, in Models of Alzheimer’s Disease. J. Neurochem. 2002, 83, 1509−1524. (9) Park, I.-H.; Arora, N.; Huo, H.; Maherali, N.; Ahfeldt, T.; Shimamura, A.; Lensch, M. W.; Cowan, C.; Hochedlinger, K.; Daley, G. Q. Disease-Specific Induced Pluripotent Stem Cells. Cell 2008, 134, 877−886. (10) Edwards, D.; Stancescu, M.; Molnar, P.; Hickman, J. J. Two Cell Circuits of Oriented Adult Hippocampal Neurons on Self-Assembled Monolayers for Use in the Study of Neuronal Communication in a Defined System. ACS Chem. Neurosci. 2013, 4, 1174−1182. (11) Froeter, P.; Huang, Y.; Cangellaris, O. V.; Huang, W.; Dent, E. W.; Gillette, M. U.; Williams, J. C.; Li, X. Toward Intelligent Synthetic Neural Circuits: Directing and Accelerating Neuron Cell Growth by Self-Rolled-up Silicon Nitride Microtube Array. ACS Nano 2014, 8, 11108−11117. (12) Zhu, B.; Luo, S. C.; Zhao, H.; Lin, H. A.; Sekine, J.; Nakao, A.; Chen, C.; Yamashita, Y.; Yu, H. H. Large Enhancement in Neurite Outgrowth on a Cell Membrane-Mimicking Conducting Polymer. Nat. Commun. 2014, 5, 4523. (13) Kunze, A.; Tseng, P.; Godzich, C.; Murray, C.; Caputo, A.; Schweizer, F. E.; Di Carlo, D. Engineering Cortical Neuron Polarity with Nanomagnets on a Chip. ACS Nano 2015, 9, 3664−3676. (14) Gumera, C. B.; Wang, Y. Modulating Neuronal Responses by Controlled Integration of Acetylcholine-Like Functionalities in Biomimetic Polymers. Adv. Mater. 2007, 19, 4404−4409. (15) Gao, J.; Kim, Y. M.; Coe, H.; Zern, B.; Sheppard, B.; Wang, Y. A Neuroinductive Biomaterial Based on Dopamine. Proc. Natl. Acad. Sci. U. S. A. 2006, 103, 16681−16686. (16) Tu, Q.; Li, L.; Zhang, Y.; Wang, J.; Liu, R.; Li, M.; Liu, W.; Wang, X.; Ren, L.; Wang, J. The Effect of Acetylcholine-Like Biomimetic Polymers on Neuronal Growth. Biomaterials 2011, 32, 3253−3264. (17) Wang, S.; Jeffries, E.; Gao, J.; Sun, L.; You, Z.; Wang, Y. Polyester with Pendent Acetylcholine-Mimicking Functionalities Promotes Neurite Growth. ACS Appl. Mater. Interfaces 2016, 8, 9590−9599. (18) Zhou, Z.; Yu, P.; Geller, H. M.; Ober, C. K. The Role of Hydrogels with Tethered Acetylcholine Functionality on the Adhesion and Viability of Hippocampal Neurons and Glial Cells. Biomaterials 2012, 33, 2473−2481. (19) Zhou, Z.; Yu, P.; Geller, H. M.; Ober, C. K. Biomimetic Polymer Brushes Containing Tethered Acetylcholine Analogs for Protein and Hippocampal Neuronal Cell Patterning. Biomacromolecules 2013, 14, 529−537.
(20) Yang, R.; Gleason, K. K. Ultrathin Antifouling Coatings with Stable Surface Zwitterionic Functionality by Initiated Chemical Vapor Deposition (iCVD). Langmuir 2012, 28, 12266−12274. (21) Coronas, V.; Durand, M.; Chabot, J.; Jourdan, F.; Quirion, R. Acetylcholine Induces Neuritic Outgrowth in Rat Primary Olfactory Bulb Cultures. Neuroscience 2000, 98, 213−219. (22) Elsas, S. M.; Kwak, E. M.; Stent, G. S. Acetylcholine-Induced Retraction of an Identified Axon in the Developing Leech Embryo. J. Neurosci. 1995, 15, 1419−1436. (23) Bakker, R.; Verlaan, V.; Van der Werf, C.; Rath, J.; Gleason, K.; Schropp, R. Initiated Chemical Vapour Deposition (iCVD) of Thermally Stable Poly-Glycidyl Methacrylate. Surf. Coat. Technol. 2007, 201, 9422−9425. (24) Tenhaeff, W. E.; Gleason, K. K. Initiated and Oxidative Chemical Vapor Deposition of Polymeric Thin Films: iCVD and oCVD. Adv. Funct. Mater. 2008, 18, 979−992. (25) Paxson, A. T.; Yagüe, J. L.; Gleason, K. K.; Varanasi, K. K. Stable Dropwise Condensation for Enhancing Heat Transfer via the Initiated Chemical Vapor Deposition (iCVD) of Grafted Polymer Films. Adv. Mater. 2014, 26, 418−423. (26) Ryu, S.; Yoo, J.; Jang, Y.; Han, J.; Yu, S. J.; Park, J.; Jung, S. Y.; Ahn, K. H.; Im, S. G.; Char, K.; et al. Nanothin Coculture Membranes with Tunable Pore Architecture and Thermoresponsive Functionality for Transfer-Printable Stem Cell-Derived Cardiac Sheets. ACS Nano 2015, 9, 10186−10202. (27) Sojoudi, H.; Walsh, M. R.; Gleason, K. K.; McKinley, G. H. Designing Durable Vapor-Deposited Surfaces for Reduced Hydrate Adhesion. Adv. Mater. Interfaces 2015, 2, 1500003. (28) Sojoudi, H.; McKinley, G. H.; Gleason, K. K. Linker-Free Grafting of Fluorinated Polymeric Cross-Linked Network Bilayers for Durable Reduction of Ice Adhesion. Mater. Horiz. 2015, 2, 91−99. (29) You, J. B.; Min, K.-I.; Lee, B.; Kim, D.-P.; Im, S. G. A Doubly Cross-Linked Nano-Adhesive for the Reliable Sealing of Flexible Microfluidic Devices. Lab Chip 2013, 13, 1266−1272. (30) Kim, M.-J.; Lee, B.; Yang, K.; Park, J.; Jeon, S.; Um, S. H.; Kim, D.-I.; Im, S. G.; Cho, S.-W. BMP-2 Peptide-Functionalized Nanopatterned Substrates for Enhanced Osteogenic Differentiation of Human Mesenchymal Stem Cells. Biomaterials 2013, 34, 7236−7246. (31) Banker, G. Culturing Nerve Cells; MIT Press, 1998. (32) Dehmelt, L.; Halpain, S. The Map2/Tau Family of MicrotubuleAssociated Proteins. Genome Biol. 2005, 6, 204. (33) Cobos, I.; Borello, U.; Rubenstein, J. L. Dlx Transcription Factors Promote Migration through Repression of Axon and Dendrite Growth. Neuron 2007, 54, 873−888. (34) Jha, S.; Rasband, M. N. Di-Rectifying Tau. EMBO J. 2011, 30, 4699−4700. (35) Lesuisse, C.; Martin, L. J. Long-Term Culture of Mouse Cortical Neurons as a Model for Neuronal Development, Aging, and Death. J. Neurobiol. 2002, 51, 9−23. (36) Hardy, J.; Selkoe, D. J. The Amyloid Hypothesis of Alzheimer’s Disease: Progress and Problems on the Road to Therapeutics. Science 2002, 297, 353−356. (37) Kellner, Y.; Gödecke, N.; Dierkes, T.; Thieme, N.; Zagrebelsky, M.; Korte, M. The BDNF Effects on Dendritic Spines of Mature Hippocampal Neurons Depend on Neuronal Activity. Front. Synaptic Neurosci. 2014, 6, 5. (38) Ji, Y.; Lu, Y.; Yang, F.; Shen, W.; Tang, T. T.; Feng, L.; Duan, S.; Lu, B. Acute and Gradual Increases in BDNF Concentration Elicit Distinct Signaling and Functions in Neurons. Nat. Neurosci. 2010, 13, 302−309. (39) Musazzi, L.; Rimland, J.; Ieraci, A.; Racagni, G.; Domenici, E.; Popoli, M. Pharmacological Characterization of BDNF Promoters I, II and IV Reveals That Serotonin and Norepinephrine Input Is Sufficient for Transcription Activation. Int. J. Neuropsychopharmacol. 2014, 17, 779−791. (40) Mattson, M. P.; Magnus, T. Ageing and Neuronal Vulnerability. Nat. Rev. Neurosci. 2006, 7, 278−294. 9917
DOI: 10.1021/acsnano.6b03527 ACS Nano 2016, 10, 9909−9918
Article
ACS Nano (41) Ben-Ari, Y.; Cherubini, E.; Corradetti, R.; Gaiarsa, J. L. Giant Synaptic Potentials in Immature Rat CA3 Hippocampal Neurones. J. Physiol. 1989, 416, 303−325. (42) Spruston, N.; Johnston, D. Perforated Patch-Clamp Analysis of the Passive Membrane Properties of Three Classes of Hippocampal Neurons. J. Neurophysiol. 1992, 67, 508−529. (43) Staley, K. J.; Otis, T. S.; Mody, I. Membrane Properties of Dentate Gyrus Granule Cells: Comparison of Sharp Microelectrode and Whole-Cell Recordings. J. Neurophysiol. 1992, 67, 1346−1358. (44) Zhang, L.; Spigelman, I.; Carlen, P. Development of GabaMediated, Chloride-Dependent Inhibition in CA1 Pyramidal Neurones of Immature Rat Hippocampal Slices. J. Physiol. 1991, 444, 25.
9918
DOI: 10.1021/acsnano.6b03527 ACS Nano 2016, 10, 9909−9918
