Understanding cellular internalization pathways of silicon nanowires
- 898 Downloads
Understanding how cells interact with nanomaterials is important for rational design of nanomaterials for nanomedicine and transforming them for clinical applications. Particularly, the mechanism for one-dimensional (1D) nanomaterials with high aspect ratios still remains unclear.
In this work, we present amine-functionalized silicon nanowires (SiNW-NH2) entering CHO-β cells via a physical membrane wrapping mechanism. By utilizing optical microscopy, transmission electron microscopy, and confocal fluorescence microscopy, we successfully visualized the key steps of internalization of SiNW-NH2 into cells.
Our results provide insight into the interaction between 1D nanomaterials and confirm that these materials can be used for understanding membrane mechanics through physical stress exerted on the membrane.
KeywordsCellular interaction Silicon nanowires Membrane wrapping
amine-functionalized silicon nanowires
transmission electron microscopy
single-walled carbon nanotubes
chemical vapor deposition
Chinese hamster ovary cells transfected with folate receptor beta
Roswell Park Memorial Institute
phosphate buffered saline
Fourier transform infrared spectroscopy
X-ray photoelectron spectroscopy
As nanotechnology advances as an innovative option in clinical settings, researchers continue to explore a wide array of nanomaterials for applications as imaging and anti-cancer agents, for drug delivery purposes, and for therapeutics. While this progress has been exciting for the future of medicine, these materials have not overcome the barrier of translating from benchtop to clinic. In order for nanomaterials to advance as viable options for biological applications, further understanding of the basic interactions between mammalian cells and nanomaterials must be achieved.
In the past few decades, to understand how the cellular membrane can respond to the entry of external nanomaterials research has been mainly focused on finding the endocytosis pathways of zero dimensional (0D) nanomaterials [1, 2, 3, 4, 5]. Limited efforts have also been made to understand the uptake of various 1D nanomaterials into cells. For example, gold nanorods and magnetic nanowires have been heavily studied for imaging and tracking purposes [3, 6, 7, 8], but their specific uptake pathways were not well-studied. Single-walled carbon nanotubes (SWCNTs) have also been of interest due to their high aspect ratio and uniqueness as a material. Yaron and coworkers showed that the uptake of SWCNTs was energy-dependent, suggesting that the pathway is endocytosis and not membrane penetration, but the surface and size is so dissimilar to that of other longer 1D materials that these uptake pathways cannot be translated . Additionally, Kostarelos and co-authors investigated previously investigated functionalized carbon nanotubes (f-CNTs) with a variety of functional groups, including an ammonium functionalization. The authors observed that the ammonium functionalized wires enter the cells at both 37 and 4 °C, ruling out a receptor-mediated pathway . Notably, the dimensions of the f-CNTs that were studied were on the scale of 1 nm in diameter and 1000 nm in length, making them unique from other 1D nanomaterials.
Theoretical studies on both 0D and 1D nanostructures suggest that understanding the membrane mechanics during endocytosis is a critical aspect to explain internalization pathways of these nanostructures. Huang and co-authors designed a nanorod model using coarse-grained molecular dynamics to demonstrate that for endocytosis such model system needs to initially bind in an upright docking position, on the membrane plane, and then be wrapped by the membrane in order to proceed through a laying-down-then-standing-up sequence to enter cells . Based on this work, it is reasonable to believe that membrane wrapping occurs as one of the first steps in endocytosis. Shi and co-authors used multi-walled carbon nanotubes to experimentally and theoretically illustrate that the cell entry of 1D nanomaterials can occur by tip recognition and rotation, but the authors do not delve into the details of internalization . Yi and co-workers theoretically proved that cell membrane internalizes 1D nanomaterials following a near-perpendicular entry mode at small membrane tension but switches to a near-parallel interaction mode at large membrane tension . These theoretical models illustrate the necessity to experimentally understand the membrane interactions of high aspect ratio 1D nanomaterials.
Due to the advantages of anisotropy and higher surface area to volume ratios than 0D nanomaterials, 1D nanomaterials can produce a stronger interaction with cells during the entering process. These features indicate that 1D nanomaterials can be considered as a better system to explore the possible membrane wrapping mechanism in the uptake pathways of nanomaterials [14, 15]. To this end, in this work we use multiple microscopy methods to visualize key steps during the cellular internalization process of silicon nanowires. Since silicon nanowires are fabricated with a complete control of dimensions [16, 17], have flexible chemistry for surface modification , have unique optical properties for in vitro and in vivo imaging [19, 20, 21], they are excellent candidates for the studies of cellular uptake pathways. Silicon nanowires modified with amine groups are the focus of the study, as compared to as-grown SiNWs with hydroxyl groups and SiNWs with specific targeting groups conjugated, SiNW-NH2 are able to be internalized in cells without targeting receptor mediated processes . We demonstrate the uptake pathway of 5 µm SiNW-NH2 to be a physical membrane wrapping mechanism using CHO-β and HeLa cell lines. Studies at two different incubation temperatures, 37 and 4 °C, were carried out in order to evaluate temperature dependence of the membrane mechanics as well as to elucidate that the process is physically driven rather than receptor-mediated. We chose 4 °C because it is well understood that many endocytic pathways are temperature dependent and that these pathways are limited to high temperature due to the large activation barrier, so uptake at a lower temperature would indicate that the mechanism is physically driven .
Synthesis and functionalization of SiNWs
Silicon nanowires were synthesized by chemical vapor deposition (CVD) with 40 nm gold nanoparticles as growth catalysts and silane as a precursor. SiNWs in length of 5 µm were chosen in this works because of the appropriate aspect ratio, low cytotoxicity and considerable high endocytosis rate to study the cellular internalization . The CVD growth was carried out in with a growth pressure of 100 Torr, silane flow rate of 5 sccm, and growth time of 5 min . A representative TEM micrograph (Additional file 1: Figure S1) shows that the dimensions of the wires are consistent with the growth conditions.
The as-grown SiNWs were first treated with thermal oxidation in atmosphere at 900 °C for 2 min in order to clean and oxidize the surface for further modification. The thermally oxidized SiNWs were then submerged in a solution of 1% (3-aminopropyl) trimethoxysilane (APTMS) (Sigma-Aldrich, St. Louis, MO, USA) in pure ethanol overnight at room temperature. The substrates were then rinsed and submerged in pure ethanol at 80 °C in order to stabilize the functional groups. After 2 h, the substrates were removed and dried under a stream of nitrogen. SiNWs-NH2 were then removed from the substrate into pure ethanol via sonication for several seconds. The SiNW-NH2 suspensions were centrifuged at 14,000 rpm three times to wash away impurities. The wires were then resuspended in a complete Roswell Park Memorial Institute (RPMI) 1640 cell culture medium with 10% FBS, 1% l-glutamine, 1% penicillin/streptococcus for future use.
In order to confirm that the amine modification was successful, unmodified SiNWs and SiNWs-NH2 were sonicated in ultrapure water and placed in disposable folded capillary cells for zeta potential measurements. The concentration of SiNWs and SiNW-NH2 in each capillary cell was estimated to be 10 µg/mL. Measurements were carried out on a Malvern Zetasizer Nano Series.
To further confirm and quantify amine functionalization on the modified silicon nanowires, the authors performed Fourier transform infrared spectroscopy (FTIR) (Thermo Nicolet Nexus FTIR) as well as X-ray photoelectron spectroscopy (XPS) analysis. As grown nanowires on growth substrates were modified with amine and used as the modified samples for measurements here. Comparison between the FTIR spectrum for the unmodified silicon nanowires (blue curve in Additional file 1: Figure S2) and the FTIR for the amine-modified wires (red curve in Additional file 1: Figure S2) shows that a peak near 1600 cm−1 in the spectrum for the amine-modified wires is visible and corresponds to the NH2 bending mode, confirming the presence of amine groups. The XPS spectrum measured from the modified SiNW-NH2 sample (Additional file 1: Figure S3) shows atomic percentages of N and Si of 1.2 and 20.3%, respectively. Based on this result, the amine coverage was estimated to be 0.61 mol/nm2, which is close to the coverage previously modeled based on the total covalently bonded APTES coverage on silica .
We cultured immortalized Chinese hamster ovary cells transfected with folate receptor beta (CHO-β) for our cellular interaction studies. Our previous study investigated the uptake of SiNW-NH2 using both CHO and CHO-β cells. The findings showed that SiNW-NH2 were successfully internalized by both cell types. Therefore, CHO-β was chosen as the cell line of choice due to their success in internalizing SiNW-NH2 . Typically, CHO-β cells were cultured in the complete RPMI 1640 mentioned above at 37 °C in a humidified atmosphere with 5% CO2. Cell viability was maintained and confirmed by cell morphology under optical microscopy during the testing periods discussed in the work [25, 26]. For optical studies, the CHO-β cells were cultured on sterile glass cover slides in 35 × 10 mm tissue culture dishes with one million cells per milliliter. Once they came to confluency, the cells were ready for microscopy studies.
To prepare samples for the optical images, the medium was removed from the culture dish and the dish was rinsed with 1 mL of phosphate buffered saline (PBS). 1 mL of the prepared nanowire solution was added to the dish and the cells were incubated with the wires for various time points.
Before imaging, the SiNW-NH2 solution was removed and the dishes were washed with 1 mL PBS. To achieve better focus of live cells, we prepare samples as follows. A microscope slide was prepared by adhering double sided tape along the long edges of the slide. The glass cover slide taken from the cell dish was then placed on top of the prepared microscope slide. One side of the cover slide was sealed with nail varnish. Fresh medium was added between the cover glass and the slide followed by sealing the other side. Bright field, fluorescent, and dark field images were collected using an Olympus BX-51 optical microscope.
Transmission electron microscopy
Once the cells were incubated with the wires for a given amount of time, the nanowire solution was removed via pipette and rinsed with 1 mL of PBS. The PBS was replaced with 2 mL of 2.5% glutaraldehyde and 0.1 M sodium cacodylate buffer solution (fix). This solution was allowed to sit for several minutes before being poured off. Another 2 mL of the fix was added and the cells were scraped from the bottom of the dish and transferred to a small conical centrifuge tube. The cells were spun down, the old fix was removed and 1.5 mL of fresh fix was added.
The remainder of the processing was done at the Purdue Life Sciences Microscopy Facility. Cells were embedded in 2% agarose, and post-fixed in buffered 1% osmium tetroxide containing 0.8% potassium ferricyanide. Cells were then en bloc stained in 4% uranyl acetate, dehydrated with a graded series of ethanol, and embedded in LX-112 resin. Sections with a 90 nm thickness for the 37 °C and 180 nm for the 4 °C samples were cut using a Reichert-Jung Ultracut E ultramicrotome and stained with 2% uranyl acetate and lead citrate. Images were acquired on a FEI Tecnai G220 electron microscope equipped with a LaB6 source and operated at 100 kV.
After incubation, the SiNW-NH2 solution was removed and the dishes were washed twice with 1 mL PBS. PBS was removed and 1 mL of 5 µM 1,1′-dioctadecyl-3,3,3′,3′-tetramethylindocarbocyanine perchlorate (DiI) was added and allowed to incubate at 37 °C for 20 min. The DiI was then removed and the dish was rinsed twice with 1 mL of PBS. PBS was replaced with 2 mL of fresh culture medium and the cells were incubated at 37 °C for another 10 min to rinse off excess DiI. The old medium was removed before cells were imaged.
The glass cover slide was placed on a microscope slide using the aforementioned preparation. Images were taken using an Olympus FluoView 300 Confocal Laser Scanning Microscope with a 543 nm excitation. For each area, approximately 100 images were taken at 10 µm steps through the Z-axis and the images obtained towards the center of the top and bottom of the cells are presented here.
Results and discussion
Zeta potential measurement confirms successful surface functionalization
The as-grown SiNWs were surface modified with amine groups to introduce a positive surface charge. The zeta potential measurements were carried out in triplicate to confirm the success of functionalization. The zeta potential of the unmodified SiNWs was measured to be −29.7 ± 7.85 mV, due to the oxide layer on the SiNW surface. The zeta potential for the SiNW-NH2 was measured to be +28.1 ± 5.11 mV, consistent with that the amine groups are protonated under neutral conditions. Such change in zeta potential after the modification confirms a successful functionalization. Since the cell membrane has slightly negative charge , the positive charge of the SiNW-NH2 will ensure charge–charge interaction between the wires and cell membrane.
The interaction between SiNW-NH2 and cells is insensitive to temperature
As we previously reported , no obvious binding or internalization was observed in the CHO-β cells treated with the same concentration of unmodified SiNWs even after 11 h incubation (see Additional file 1: Figure S4). We attributed the observation of no interaction to the lack of charge–charge interaction between the unmodified SiNWs and the cell membrane . This is the also the reason why we chose to focus on amine-functionalized SiNWs in this paper.
Figure 2 plotted the quantitative analysis of wires interacting with cells as a function of incubation time at 37 and 4 °C. Wires were counted from the overlay of the bright and dark field images. Five areas measuring 70 µm × 50 µm were examined for each sample. Numbers of wires interacting per cell measured and standard deviation of the data presented as error bars are plotted in Fig. 2. Figure 2 showed similar trends for both temperatures as functions of the incubation period with more wires interacting with cells at 37 °C. The average rate of interaction over the first 5 h of the incubation period was estimated to be 0.5 wires per cell per hour at 37 °C (green curve), while 0.4 wires per cell per hour at 4 °C, likely indicating a slightly slower internalization for 4 °C. Such small difference can be attributed to the fact that the cells incubated at 4 °C were not as active as they were at 37 °C. It is well-known that receptor mediated endocytosis processes do not happen at low temperatures [28, 29, 30]. A small decrease in the number of wires interacting with cells after 5 h at 37 °C was observed. We attributed the decrease after 5 h in the number of wires interacting with cells to exocytosis, as it is consistent with the reported values between 6 and 8 h at 37 °C for the typical time scale for exocytosis of nanoparticles and nanorods . Since it is a competing process between endocytosis and exocytosis once the SiNWs enter the cells, we attributed the decrease in the number of wires interacting with cells after 5 h at 37 °C to exocytosis. The interaction between SiNW-NH2 and cells was found to be insensitive to temperature in our studies, indicating that the charge–charge attraction induced interaction between SiNW-NH2 and cells is not through the receptor mediated pathway but a physically driven process.
Cross-section TEM images indicate membrane wrapping
Confocal fluorescence confirms the physical membrane wrapping
Figure 5b, c show bright field and fluorescence images obtained from the control group without wires (Fig. 5b) and the CHO-β cells co-cultured with SiNW-NH2 for 10 h at 37 °C (Fig. 5c). Bright field optical images on the left column of Fig. 5 show the presence of wires and the fluorescence images on the right column were used to locate the cell membrane, therefore indicating the interaction between wires and cellular membrane. Specifically, since the images plotted here were obtained towards the center of the top and bottom of the cells along the z scanning under the confocal microscope, the wires shown in focus in the bright field are in the focal plane therefore considered to be inside the cell. Clearly different from Fig. 5b, c demonstrated internalization of SiNW-NH2 by CHO-β cells. Importantly, Fig. 5c also shows that wire-shaped fluorescent signals present in the cells, indicating the internalized SiNW-NH2 became fluorescent. Together with observation that SiNWs were not directly labeled by Dil from Fig. 5a, these results also suggest that internalized SiNW-NH2 is through membrane wrapping. More fluorescence images obtained from 2- and 5-h incubation periods along with a plot showing average number of wires per cell internalized as a function of incubation time period are included in Additional file 1 (Figures S5, S6, respectively).
KM and YH carried out experiments, analyzed data and wrote the manuscript. CY supervised the entire project, was involved in the design of all experiments and revised the manuscript. All authors read and approved the final manuscript.
Authors gratefully acknowledge the Purdue University Life Sciences Microscopy Facility, Purdue University Amy Facility, Dr. Dmitry Zemlyanov at the Surface Analysis Facility of the Birck Nanotechnology Center, Purdue University for XPS analysis, and Dr. J. X. Cheng for access to measurement facilities. The authors would also like to acknowledge Justine Arrington for helpful discussions.
The authors declare that they have no competing interests.
Availability of data
The authors declare that the data supporting the findings of this study are available within the article and its Additional file 1.
This work was funded by Purdue University.
- 10.Bianco A, Kostarelos K, Partidos CD, Prato M. Biomedical applications of functionalised carbon nanotubes. Chem Commun (Camb). 2005;(5):571-577. doi: 10.1039/b410943k
- 17.Zhong Z, Yang C, Lieber CM, Vijay K. Nanosilicon, vol. 5. Oxford: Elsevier; 2008. p. 176–216Google Scholar
- 19.Park H, Crozier KB. Multispectral imaging with vertical silicon nanowires. Sci Rep. 2013;3:2460.Google Scholar
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.