If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
While exposure of C17.2 neural progenitor cells (NPCs) to nanomolar concentrations of carbon nanotubes (NTs) yields evidence of cellular substructure reorganization and alteration of cell division and differentiation, the mechanisms of NT entry are not understood. This study examines the entry modes of (GT)20 DNA-wrapped single-walled carbon nanotubes (SWCNTs) into NPCs. Several endocytic mechanisms were examined for responsibility in nanomaterial uptake and connections to alterations in cell development via cell-cycle regulation. Chemical cell-cycle arrest agents were used to synchronize NPCs in early G1, late G1/S, and G2/M phases at rates (>80%) aligned with previously documented levels of synchrony for stem cells. Synchronization led to the highest reduction in SWCNT internalization during the G1/S transition of the cell cycle. Concurrently, known inhibitors of endocytosis were used to gain control over established endocytic machineries (receptor-mediated endocytosis (RME), macropinocytosis (MP), and clathrin-independent endocytosis (CIE)), which resulted in a decrease in uptake of SWCNTs across the board in comparison with the control. The outcome implicated RME as the primary mechanism of uptake while suggesting that other endocytic mechanisms, though still fractionally responsible, are not central to SWCNT uptake and can be supplemented by RME when compromised. Thereby, endocytosis of nanomaterials was shown to have a dependency on cell-cycle progression in NPCs.
Why it matters
Every cell relies on the uptake or endocytosis of materials, such as proteins, cytokines, and even synthetic carbon nanomaterials, to perform its required cellular fate functions. Therefore, endocytosis is of interest for bringing therapeutic targets into cells. Recently, endocytosis has resurfaced as a topic of heavy debate, discussion, and discovery. Studying how materials get into the cell can aid in detangling trafficking to design higher-efficiency targeted drug- and gene-delivery therapies. Here we report the involvement of multiple endocytic pathways for bringing nanomaterials into neural stem cells and find a strong dependency of nanotube internalization on the cell cycle. This information can be harnessed to augment delivery of therapeutic materials based on the developmental stage of the cell.
Carbon nanomaterial-based therapeutics have moved to the forefront of medicine in the last two decades, dominating applications such as controlled-release drug delivery, cell labeling, nanosensors, and scaffolds for tissue regeneration (
). The nanoscale size of these materials and applicability of surface functionalization to assure biocompatibility of the nanotubes (NTs) make their transport across the blood-brain barrier (BBB) especially valuable, opening delivery access to the central nervous system. The unique optical and electronic properties of single-walled carbon nanotubes (SWCNTs) as well as their size range (1 nm diameter, ≥100 nm length) and mechanical strength, enable them to serve in a variety of applications involving crossing the BBB, from tumor targeting to gene therapy (
). While NTs have been shown to be instrumental in accessing areas previously thought of as difficult to reach, other works have debated cytotoxicity and cellular damage related to size, purity, concentration, and functionalization (
As this was an early concern when developing experimental conditions, the material used in this work has been thoroughly characterized. The SWCNTs used in our studies are CoMoCat NTs wrapped with (GT)20 single-stranded DNA oligomers (DNA-SWCNTs) (
). The DNA conjugation serves as a biomolecule mask to the cells, also substantially increasing nanomaterial solubility in aqueous buffer and inhibiting NT coalescence. Such a functionalization facilitates NT entry via endocytosis upon introduction in vitro (
). Our previous studies in C17.2 neural progenitor cells (NPCs) allowed us to define the acceptable range of DNA-SWCNT concentrations (10 pM or lower) to be used for biological applications. Even in this range, there will be noticeable differences in cell morphology post DNA-SWCNT introduction (
). Therefore, it is postulated that the presence of the nanomaterials, in optimized concentrations and external conditions can upregulate aspects of natural cellular fate processes that can be harnessed to further understand the system, which can be used in the future to develop improved nanosensors and delivery therapeutics. Thus, a cellular- and molecular-scale analysis of the NT-cell system is required to comprehend the modification of cell developmental behavior in response to potential nanomaterial-based therapies as well as the downstream implications of cell-material interactions on these therapies.
A crucial step in determining the influence of NTs on NSCs was to prioritize the trafficking of materials through the cell, beginning with the method of entry. Nanomaterial entry mechanisms are not fully understood, and new studies have appeared recently detailing aspects of endocytic mechanisms and pathways that have eluded the field for decades (
). These entry mechanisms are accompanied by respective stimuli that trigger various downstream cellular responses and, accordingly, signal transduction changes in the cell. Unsurprisingly, this results in biochemical changes to the cell, vital to its growth and development, in addition to the cell’s innate dynamic internal component reorganization.
While there are many aspects of neural stem processes that are yet to be understood, it is well known that the cell cycle lies at the crux of division, differentiation, and development of stem cells. Changes in cellular fate processes are generally linked to the cell cycle, the regulator of growth and development (
), and previous reports identify a link between endocytosis and cell cycle, defending the idea that cellular growth and development is interwoven with the trafficking of materials into and through the cell (
). Gaining control of the cell cycle via synchronization and maximizing single-phase yield are vital to resultant signal transduction pathways that lead to changes in downstream cellular fate processes. The overall mechanism of entry and the effect of specific phases of the cell on the endocytic mechanism will be key factors in deconvoluting the localization and correlation of SWCNTs among the subcellular organelles of NPCs.
Persisting speculation surrounding the entry, internalization, trafficking, and correlation of NTs within stem cells is a consequence, to some extent, of the lack of understanding of cell-cycle restructuring. As much as entry of NTs into the cell is a product of the cell cycle, the altered regulation of the cell cycle is also a by-product of the entrance of the NTs into the cell (
). Previously witnessed modifications in cell morphology, cytoskeletal organization, and differentiation could be tied to altered cell development. In the situation where growth, development, and differentiation are all in question, the cell cycle remains a critical underlying factor for the majority of the endocytic and developmental alterations observed. Therefore, studying how uptake translates further into downstream biochemical reactions is critical. Our study is focused primarily on the modes of entry of SWCNTs and how internalization changes based on the phase state of the cell.
Materials and methods
C17.2 mouse v-myc immortalized NSCs, a gift of Dr. Evan Snyder (Burnham Institute, La Jolla, CA), were cultured according to accepted protocol (
). NSCs were grown in Dulbecco’s modified Eagle’s medium (DMEM; Corning, Corning, NY) supplemented with 10% fetal bovine serum and 5% horse serum (Stasis Stem Cell Serum) at standard culture conditions of 37°C and 5% CO2. 1× low potassium Locke’s buffer (10 mM HEPES, 5.6 mM KCl, 154 mM NaCl, 5.6 mM glucose, 1.2 mM MgCl2, 2.3 mM CaCl2, pH 7.4) was used for washing steps and all points where buffers are typically used.
Arrest and synchronization of cell cycle
Early G1, late G1/S, and G2/M arrest was achieved through lovastatin (10 μM; AdipoGen, San Diego, CA), double thymidine (0.25 mM; Alfa Aesar, Ward Hill, MA) and nocodazole (400 nM; Sigma, St. Louis, MO) block, respectively. All agents were made up in DMEM (15% growth serum), and optimal concentrations and exposure times were determined to maximize percentage of cells in synchrony. Removal of chemicals and washing and replacement with medium or buffer was used to test release from arrest. NSCs were treated with arrest agents before being trypsinized and processed for flow cytometry. Cells were washed and fixed dropwise using 70% cold ethanol while vortexing, placed at 4°C for 30 min for complete fixation, and subsequently stained for DNA content using propidium iodide/RNase (BD Biosciences, Franklin Lakes, NJ). Flow cytometry was conducted using a BD FACSCanto hardware system and the BD FACSDiva software package. Gating of aggregates and doublets in the software resulted in high reliability of singlet cell data. FCS file analysis was conducted in both FCSalyzer and FlowJo. Data were fitted to a Watson-Pragmatic model where applicable.
“Leave-one-out” inhibition of endocytic mechanisms
Inhibition of endocytic mechanism receptor-mediated endocytosis (RME), macropinocytosis (MP), and clathrin-independent endocytosis (CIE) was effected by solutions of 0.45 M hypertonic sucrose (Fisher Chemical, Waltham, MA), 1 mM amiloride hydrochloride (Sigma-Aldrich, St. Louis, MO), and 1 μg/mL filipin complex (Sigma-Aldrich), respectively in 1× low K+ Locke’s buffer for 30 min of incubation at 37°C, knocking out one form of endocytosis and leaving the others undisturbed. Uninhibited control populations remained in 1× low K+ Locke’s buffer alone for 30 min. After inhibition the cultures were rinsed, leading into uptake testing with either SWCNTs or Tf-AF488.
Uptake assays and visualization
DNA-SWCNT uptake assay
NSCs seeded at low concentrations (3–5 × 103 cells/cm2) on tissue-culture-treated borosilicate glass coverslips were allowed to grow for 48 h. Growth medium was removed from the cells, followed by a few gentle washes in Locke’s buffer to remove any remnants of serum proteins. At this point, any synchronization/arrest (nocodazole, lovastatin, or thymidine) or endocytic inhibition (hypertonic sucrose, amiloride, or filipin) treatments were introduced. Cells were washed again with Locke’s buffer post treatments. CoMoCAT SWCNTs wrapped with oligomeric (GT)20 repeat DNA sequences were then suspended in sterile 1× low K+ Locke’s buffer at a low concentration of 5 ng/mL and incubated at 37°C for 8 h. After the incubation period the solution was aspirated, and the cells were washed in Locke’s buffer to remove any uninternalized nanomaterials off the cell surface before being fixed in 4% paraformaldehyde. Coverslips were finally sealed onto slides with diamond antifade mountant prior to Raman imaging.
Hyperspectral imaging of SWCNTs in vitro
Raman hyperspectral imaging (532 nm excitation laser) was done on an Alpha300 R WITec confocal micro-Raman microscope (WITec, Ulm, Germany). Scans were taken at 50× and 100× magnification with a 0.1s integration time. Cell autofluorescence (2700–3015 cm−1) and NT signature lines (310 cm−1, 1590 cm−1, 2630 cm−1) were used to create spectral images, which were collected for a single set of 10 cells per sample. All 10 samples showed the same statistical distribution for SWCNTs in cells. Images were juxtaposed to show localization of internalized nanomaterials. 100× depth scans were taken starting from below the cell to above it, at intervals of z = 0.1 μm, to confirm NT residence. Visual (in a spectral map) and spectral (in single-point spectra) confirmation was used to collect internalization data.
Transferrin conjugate uptake assay
Cells grown to confluency were incubated with chemical synchronization agents for 24 h or endocytic inhibitory treatments for 30 min at 37°C. Cell monolayers were subsequently rinsed with warmed 1× low K+ Locke’s buffer and incubated with 20 μg/mL Transferrin-AlexaFluor 488 (Tf-AF488; Molecular Probes, Eugene, OR) in 1× Locke’s buffer for 30 min at 37°C. Excess transferrin was rinsed off and cells were harvested, fixed with cold ethanol, and processed for flow cytometry. Uptake was discerned through relative fluorescence intensities read in the FITC channel of BD FACSCanto. Debris and aggregates were gated out to select for singlets, from which median fluorescence intensity was calculated from the resultant histogram peaks using FCSalyzer.
The LIVE/DEAD Viability/Cytotoxicity Kit (Invitrogen, Waltham, MA) was used to determine the average percentage of live cells in a population after each treatment. Live and dead cells were counted across three fields of view for three separate monolayers for each synchrony and endocytic inhibition treatment.
Where applicable, statistical analysis was determined by standard error measurements and post hoc analysis using Student’s t-test. Significance levels were set at p < 0.05 (∗) and p < 0.01 (∗∗). Raman image scans were collected across two to three independent trials. All flow cytometry and cell assay data detailed throughout the study are presented as means of three or more independent experiments (across cell passages n = 14 to n = 19).
Results and discussion
Cell-cycle synchronization of embryonic neural stem cells
Many drugs and chemicals have been multipurposed for use in cell-cycle arrest and synchronization (
). While the range of reagents harnessed for this purpose is extensive, three methods of arrest were chosen for their ubiquity as well as their simplicity in mechanism of action. Agents with complex molecular mechanisms of action tend to affect multiple natural processes within the cell, resulting in higher rates of apoptotic response and permanent alterations. Therefore, molecular agents with evidence of direct effects on the actin cytoskeleton were avoided because of the vital role actin plays in conjunction with clathrin during RME. In an effort to maintain the natural state of the cell as best as possible, low concentrations of the pharmacological arrest agents lovastatin (early G1), thymidine (late G1/S), and nocodazole (G2/M) were chosen.
Non-pharmacological methods of arrest, such as serum starvation and contact inhibition, have been used for decades to synchronize mammalian cells. However, it has been noted that serum deprivation while achieving synchrony simultaneously poises cells for a much higher rate of differentiation in embryonic stem cells (
). Single mechanism arrest followed by release into subsequent phases yielded low rates of synchrony that were transient; hence, three different mechanisms were chosen to achieve sustainable synchrony.
Existing literature on the synchronization of primary cell lines (
). These values from literature determined the upper and lower bounds of concentrations tested for each drug to find the highest synchrony possible for C17.2 NPCs. Nocodazole had the widest range of concentrations from nM to μM. While nM ranges led to lower rates of synchrony, μM concentrations were claimed to achieve higher synchrony but at the cost of permanently damaged microtubules. Thymidine was used fairly consistently in the lower mM range, while lovastatin was shown to work over a range of μM concentrations. Fig. 1, A–C details the optimization range for concentration-dependent synchrony by each treatment. Cells were treated with one of the three synchrony agents for 24 h for nocodazole and lovastatin, or for 12 h in thymidine, 10 h in serum medium, and again 12 h in thymidine, to achieve a double thymidine block. The cells were then harvested and stained for cellular DNA content with propidium iodide to analyze via flow cytometry.
Fig. 1A shows the results of the concentrations of lovastatin tested from 0 to 40 μM. While synchrony up to even 90% in early G1 is achievable with lovastatin, a lower concentration accomplishing 80% synchrony was chosen to minimize damage and apoptotic cell density. With a double thymidine block, a consistent even split between late G1 cells and S phase cells is evident, while G2/M cells are at a normally functioning population size, showing that the cells are still proliferating and dividing. The nocodazole treatment (Fig. 1C) shows that at 200 nM the crossover of G2/M cells into the majority begins, but significant consistency is witnessed only above 400 nM. The flow cytometric histograms of DNA content shown in Fig. 1, E–G correspond to the final treatment concentrations chosen, as compared with the asynchronous population (Fig. 1D). Ultimately, the optimization process revealed that 15 μM lovastatin, 2 mM thymidine, and 400 nM nocodazole led to the highest optimal synchronies of 80% early G1, 85% G1/S transition, and 89% G2/M, respectively.
Additionally, it was important to rule out any adverse effects and chemical alterations caused by the arrest agents used. Table 1 shows the percentage viability of the cell populations after they were placed in each treatment for the times detailed in the methods below. This shows that all synchrony treatments had exceptionally high rates of viability, were comparable with the control, and remained unaffected by chemical alteration.
Table 1Reagents used for cell-cycle arrest and chemical endocytic inhibition, their target processes, and resultant cell viabilities after 30 min (endocytic inhibition) and 24 h (cell-cycle arrest) treatments on neural progenitor cells
Cell-cycle dependency of DNA-SWCNT endocytic delivery
The synchronized cell populations were subsequently assessed for SWCNT uptake, the aim being to ascertain the influence of NT introduction on the cell-cycle arrested populations as compared with a control population. The results would also indicate whether NTs are accepted into the cell at varying levels depending on their phase, and ultimately whether this has any repercussions on how the cell processes the endocytosed material.
Previous studies support that the endocytic methods introducing certain materials into the cell vary based on the state of the cell (
). This implies that as the cell cycles through its natural phases, the corresponding primary method(s) of uptake oscillates accordingly. Inhibitory testing of all endocytic methods available for SWCNT entry shows how uptake varies according to the isolated individual mechanisms. A side-by-side comparison of SWCNT uptake based on cell cycle and methods of endocytosis will explain the process dependency and give a clearer understanding of how the mode of entry leads to localization and to downstream molecular changes as a result of their presence.
Fig. 2 shows the data detailing the SWCNT uptake and a summary of its dependence on phases of the cell cycle. Representative 50× hyperspectral scans of cells treated with each arrest agent are shown. The scans were collected by Raman spectroscopy, an established method of identifying carbon NTs in cells, as detailed previously (
). The SWCNTs in residence (blue boxes) are clearly distinguishable from the ones outside of the cell (white boxes). Several spectral signatures of NTs are recognized. We use mapping of the wavelengths for the radial breathing mode (near 300 cm−1), G peak (1590 cm−1), and G′ peak (2630 cm−1). Cell autofluorescence should be differentiated from NT signal (C-H vibrations between 2700 and 3015 cm−1). In some cases, spectral overlap with C-H vibrations will result in anomalies in the NT signature spectral maps (red boxes) to be eliminated. SWCNT residence was confirmed accordingly and, as needed, was investigated with depth profile scans under 100× magnification as shown with an example of an edge case in Fig. S1.
The experimental strategy developed for identifying NTs with Raman imaging is to begin with focusing on the bottom glass substrate and moving into the cell stepwise to focus on the bottom plane of the cell (where the spectral map is at its clearest or highest resolution). The height of the cell was then taken into account to move stepwise in the z direction up through the cell in order to characterize only internalized carbon NTs. Any NTs that were suspected to be surface-bound (marked with white boxes in Figs. 2, A–H and 3, A–H), were confirmed to be external and were not included in the uptake numbers (Figs. 2I and 3I). Only residence-confirmed cells (blue boxes) were included and plotted in Fig. 2I.
This strategy also includes several procedures to mitigate the concern of aggregated nanomaterials. First, as previously explained in previous work (
), RBM modes can be used to discriminate the aggregates from individual NTs. The G and 2D modes also show characteristic broadening upon aggregation that can be resolved from data statistics. In the event that aggregation of NTs is detected, the instance would be included in the statistical count. However, the use of an extremely low concentration of NTs is an important distinction of this work from a number of previous ones, where the SWCNT concentration used was many orders of magnitude larger and aggregation happened permanently.
The data in Fig. 2I, obtained by counting the average number of NTs per cell from hyperspectral images (Figs. 2, A–H), summarizes the findings and indicates a large increase in uptake for G2/M arrested cells, as well as an insignificant increase in uptake in early G1 cells, as compared with an asynchronous cell population. A reduction in uptake is noted in late G1/S arrested cells.
Double thymidine block treatment resulting in majority G1/S transition cells led to a 40% decrease in endocytic uptake, suggesting that the late G1 and early S phases are selective to uptake of materials. Simultaneously, samples of primarily early G1 and G2/M cells led to 40% and 390% increases in uptake of SWCNTs, respectively. Existing studies contradictorily propose that clathrin-mediated endocytosis (CME) is either inhibited or unaffected during the G2/M phase (
). Our data show that the G2/M phase is an avid acceptor of materials, as it has a fourfold higher yield of internalized NTs per cell on average when compared with control cells. This proves the dependency of uptake on the mitotic phase of the cell cycle.
The disproportionately high uptake of SWCNTs during nocodazole arrested G2/M cells suggests either one method of uptake is augmented during mitosis or that multiple pathways are contributing simultaneously to internalization. To unravel this further, CME and the importance of the vital protein clathrin during the mitotic phase is an important and well-studied place to start. Clathrin is involved in the generation of new vesicles as well as mitotic spindle stabilization (
). Reduced dynamics of clathrin will result in lower production of CCVs to carry forward typically observed rates of RME. Some reports suggest that clathrin is impaired during mitosis, reducing the activity of CME during mitosis (
). Overall, the sharp increase in uptake of SWCNTs in cells treated with nocodazole for G2/M arrest indicates that NT uptake is heavily a premitotic phase event.
Simultaneously, the increase in endocytosis in early G1 and reduction during late G1/S suggest that a percentage of uptake happens in each phase and there is heterogeneity in mechanism of SWCNT uptake that is heavily cell cycle dependent. This aligns with the recent work done by the Gopal et al. on another material, which showed that the uptake of silicon nanoneedles was regulated by caveolae-mediated endocytosis and CME as well as MP simultaneously (
). They noted nanodiamond presence in all phases of the cell cycle, including all phases of mitosis, causing no alteration of the spindle fibers nor chromosome segregation.
This, however, does not point to CME as solely responsible for the uptake of SWCNTs. Heterogeneity of clathrin-dependent uptake is based on variation in plasma membrane tension, which can depend highly on the state of the cell. Membrane tension is inversely related to the success of endocytosis. The stiffer the membrane, the less likely that endocytic machinery can generate forces to overcome that tension to bring materials into the cell. Membrane tension is also related to cell adhesion. When cell adhesion decreases, as in the case of cells preparing for mitosis, membrane tension is at its highest, indicating that CME should not be able to operate to traffic materials into the cell at the G2/M phase in the cell cycle. The G1, S, and G2 phases themselves have shown nonhomogeneous adhesion of the cell membrane (
), and will continue to see below, that RME plays a role in the uptake of SWCNTs, there is strong heterogeneity of the contribution of diverse mechanisms of uptake that is heavily dependent on the cell membrane and cell-cycle dynamics.
It is important to note the role of the size of the cell on uptake. The surface area of the cell is known to vary highly with the progression of the cell cycle, often one of the determinant factors in advancement to the next phase of the cycle. Therefore, it is a realistic concern that a change in surface area of the cell could proportionately affect uptake during one phase as compared with another. To address this concern, the distribution of SWCNT uptake per cell per treatment and the corresponding cross-sectional surface area of each of those cells are shown in Fig. S2. Some treatments show a normal distribution of SWCNT uptake, where the mode of the data set corresponds to the mean/average presented in Fig. 2. There are a few outliers in the case of lovastatin and thymidine, for example, where we see the mode of the uptake data set is lower compared with the mean, yet this reduction is roughly consistent across all synchrony treatments. Along with low-value correlation coefficients, the data show a lack of correlation between the surface area of the cell and SWCNT uptake.
Finally, the percentage chance of a cell being asynchronous is roughly 10% for G2/M, 27% for early G1, and 15% for G1/S. Therefore, of the cell scans collected, one, or at most two cell scans have a statistical chance of being an unsynchronized cell. This was counteracted by analyzing cells that showed a similar statistical distribution and assessing and excluding, if needed, the minimal number of outliers with significantly higher or lower NTs internalized. Ultimately, the synchronization was performed to maximally increase the yield of single-phase cells and increase the confidence in uptake numbers for each phase.
Inhibition of endocytic mechanisms for the study of DNA-SWCNT uptake pathways
To deepen the understanding of cell-cycle-based alteration of endocytic mechanisms and to test the contributions of specific methods of uptake for SWCNTs, “leave-one-out” endocytic inhibitory testing was conducted using: 1) hypertonic sucrose against RME; 2) amiloride against MP; and 3) filipin against CIE.
Each treatment knocked out one mechanism of endocytosis while leaving the remaining methods intact. Hypertonic sucrose causes clathrin to create microcages, rendering them inactive and bringing RME to a halt (
). Cells incubated with each of the endocytic inhibition treatments were assessed for the rates of viability to confirm the health of the cell populations. Table 1 shows high viabilities for each treatment, comparable with untreated, control NPCs. It is important to note that there is a lot of variability in endocytic inhibitors, and certain chemicals will often affect more than one method of endocytosis. Amiloride, for example, at certain concentrations, can arrest both MP and fast endophilin-mediated endocytosis (FEME) (
). This is an area of ongoing research, and many groups are simultaneously detangling new pathways, validating new cargoes, and deepening our understanding of conventional endocytic mechanisms and their inhibitors (
Fig. 3 shows the detailed data for post-endocytic inhibitory uptake of SWCNTs. Immediately evident is the reduction in uptake across the board from all treatments. The 55% reduction in endocytosis post inhibition of RME aligns with previous results and hypotheses, but the 63% drop in SWCNT uptake in filipin-treated CIE-inhibited cells and 52% in MP-inhibited cells implies that RME is aided in NT uptake by CIE and MP, as corroborated by studies previously cited on uptake of other nanomaterials in a range of cell types. These results reveal the equal participation of CIE and MP alongside RME in the internalization of SWCNTs. For complete inhibition, endocytosis was ceased by incubation at 4°C during NT uptake, knocking out all ATP-based endocytosis, including RME, MP, and CIE and leading to a 78% drop in uptake. This supports that the majority of DNA-SWCNT uptake in NPCs is energy dependent. While there is some evidence of energy-independent uptake via the lipid bilayer, studies by colleagues support the primary role of energy-dependent uptake of SWCNTs (
).The control remains the highest in uptake given that all methods of endocytosis, no matter whether primary or secondary, are accepting nanomaterials intracellularly.
Transferrin was used as a control and comparison for the above results, where its uptake is mediated by TfR only via RME. Transferrin is an iron-binding glycoprotein that facilitates the internalization of iron into the cell. It binds to iron tightly but reversibly and associates with the TfR membrane receptor to deliver iron intracellularly and be trafficked back out via recycling endosomes (
). If SWCNTs were internalized only by RME, a similar uptake trend would be seen for transferrin to match that of the SWCNTs, as both cargoes, though recognized by different receptors, would be internalized by the same CCVs. The comparison between transferrin and NTs is shown in Fig. 4. The largest drop in transferrin uptake is after inhibition of RME, as expected, and during early G1. This reduction specifies that RME does not serve as a primary method of uptake during the earlier portion of the G1 growth phase. Contrary to what was seen for NTs, transferrin internalization looks unchanged during late G1/S. Transferrin uptake is shown in Fig. 4A to actually increase when MP and CIE are inhibited, with an internalization higher than what is seen in the control cells, following the hypothesis that when one endocytic method is stunted, in this case RME, other methods will substitute. This phenomenon has been observed in other cargoes (
The juxtaposition of the transferrin onto NT uptake proposes that, while RME is not the primary or sole mechanism responsible for SWCNT uptake during early G1, the consistently higher internalization of transferrin during late G1/S and G2/M arrested cells shows that RME does have a role to play. Most interestingly, the data continue to indicate that there is likely multiple-mechanism co-uptake via each of the three endocytic pathways, leading to the drastic uptake of NTs during the G2/M phase. The overall results strongly correlate with the hypothesis that nanomaterials gain entry into the cell by more than just one method, and that more than one method of endocytosis can be responsible for the uptake of a single cargo. The uptake of cargo (
) by multiple endocytic pathways have been recorded, and this field requires further investigation to understand the dynamics of nanomaterials, endocytic co-transport, and substitution during each phase of the cell cycle.
A potential contributing factor to multiple pathway uptake is the diversity in length of the SWCNTs introduced to the NSCs. This diversity comes from prolonged sonication utilized for single-NT dispersion—it is known to cause breakage at defective sites along the nanomaterial, leading to a diverse range of the lengths from 100 to 500 nm. Shortening of NTs caused by additional sonication of solution is a ubiquitously accepted side effect first mentioned by Shelimov et al. (
). Size dependency of endocytic uptake is a well-studied but case-based phenomenon that is also affected by shape and functionalization of the material. For DNA-SWCNTs specifically, evidence shows that uptake of cylindrical NTs was higher in rate than spherical gold nanoparticles, and that SWCNTs less than 100 nm in length have been shown to be taken up via RME (
) are approximately 100 nm in diameter for CME and clathrin-independent carrier/glycosylphosphatidylinositol-anchored protein enriched early endocytic compartment endocytosis. Sizes are thought to be 60 nm in diameter for caveolin-dependent endocytosis and roughly 60–80 nm in cylindrical diameter for FEME, but up to hundreds of nanometers in length. MP and phagocytosis lie at a particle acceptance size of >200 nm. Therefore, the range of lengths in the NTs used in this study is likely to instigate a range of endocytic mechanisms that would require further experimentation to unravel them.
Moving forward, there are also several important questions surrounding post-entry movement and trafficking of nanomaterials through the cell via RME, MP, and CIE individually to be studied. An interaction mapping from entry to exit of the SWCNTs in NPCs with focus on intracellular component reorganization and an examination of expression levels of the key proteins involved will build a better picture of endocytosis as it relates to cell-cycle regulation and cellular fate processes. Evidence that endocytosis is cell cycle dependent garners interest to revisit the phenomenon that the intracellular presence of the NTs has been shown to trigger an increase in the rate of differentiation of C17.2 NPCs (
). Cellular fate processes are well intertwined such that changes in one process triggers a cascade of downstream changes in associated processes. The endocytosis of nanomaterials triggers the typical endocytic pathway, but the size, shape, and functionalization alters the normal cascade in such a way that this path is thought to lead downstream to neural differentiation.
Conversely, at the disproportionate high dose of NTs administered for targeted gene- or drug-delivery applications without appropriate stable biofunctionalization, in some works evidence of cytotoxicity and reorganization of, or NT interaction with, intracellular components in various cell types has been reported (
). Many, in turn, argue against claims of cytotoxicity, stating that the specific type of nanomaterial, type of functionalization, and final working concentrations of the material will dramatically influence the degree of biocompatibility and are critical to take into account (
). Therefore, a deeper investigation of the influence of NTs on various cell types is required to debunk the controversial issue of the biocompatibility of nanomaterials and to better understand the mechanisms by which intracellular interactions with well-defined nanomaterials unfold.
Examining cell mechanics and biochemistry of a dynamic and adapting cell will strengthen the understanding of nanomaterial cell downstream interactions, including the positive and/or negative effects that NTs and nanomaterial-based therapeutics fundamentally have on neuronal cell populations and patients treated using these methods.
The progress in SWCNT and NPC interactions over this study uncovered the roles of RME, MP, and CIE as mechanisms of SWCNT uptake into NPCs. It unraveled the dependency of endocytosis mechanism on cell-cycle regulation in SWCNT uptake, where all three methods of endocytosis (RME, MP, and CIE) were shown to play differing phase-dependent roles that are thought to change based on variations in lipid bilayer membrane tension. Ultimately, uptake of SWCNTs was shown to drastically increase in premitotic G2/M phase arrested cells and to decrease in G1/S transition arrested cells.
A clearer picture of the relationship between endocytosis, cell cycle, and differentiation will be evident through the propagation of this field of work. For ailments such as Alzheimer’s disease and Parkinson’s disease, natural augmentation of neuronal differentiation as a supplement for poor neurogenesis (
) can ease symptoms and be the basis for a more permanent and safe treatment option. Although therapeutics are an extended goal, the mechanistic understanding of neural development can inspire the development of biosensors. Finally, a deeper understanding of the mechanisms by which SWCNTs alter NSCs both in positive and negative respects will give fuel to determining therapeutics capable of ensuring cell health and renewal against neurodegeneration.
S.C. and S.S.J. designed the research. S.C., S.K., and A.I.V. performed the research. S.C. analyzed the data and wrote the paper. T.I. and S.V.R. provided guidance and valuable insights.
The authors would like to acknowledge the use of shared Raman facilities supported by the LU CREF grant, and BioRender for facilitating design of the graphical abstract. SC was funded by a Lehigh University fellowship.