Decoding Live Cell Interactions with Multi-Nanoparticle Systems: Differential Implications for Uptake, Trafficking and Gene Regulation

Abstract

Surface modification with oligonucleotides renders gold nanoparticles to endocytose through very different pathways as compared to unmodified ones. Such oligonucleotide-modified gold nanoparticles (OGNs) have been exploited as effective nanocarriers for gene regulation therapies. Notably, in an effort to reduce overall dosage and provide safer transition to the clinic, cooperative systems composed of two or more discrete nanomaterials have been recently proposed as an alternative to intrinsically multifunctional nanoparticles. Yet, our understanding of such systems designed to synergistically cooperate in their diagnostic or therapeutic functions remains acutely limited. Specifically, cellular interactions and uptake of OGNs are poorly understood when the cell simultaneously interacts with other types of nanoparticles. Here, we investigated the impact of simultaneous uptake of similar-sized iron oxide nanoparticles (IOPs) on the endocytosis and gene regulation function of OGNs, whose analogues have been proposed for sensitization, targeting and treatment of tumors. We discovered that both the OGN uptake amount and, remarkably, the gene regulation function remained stable when exposed to a very wide range of extracellular concentrations of IOPs. Additionally, the co-localization analysis showed that a proportion of OGNs was co-localized with IOPs inside cells, which hints at the presence of similar trafficking pathways for OGNs and IOPs following endocytosis. Taken together, our observations indicate that while the OGN endocytosis is highly independent of the IOP endocytosis, it shares transport pathways inside cells but does so without affecting the gene regulation behavior. These results provide key insights into concomitant interactions of cells with diverse nanoparticles, and offer a basis for the future design and optimization of cooperative nanomaterials for diverse theranostic applications.

1. Introduction

Personalized medicine relies on the incorporation of appropriate molecular species to target specific cells — more specifically, molecularly targeted therapy. The need to visualize the performance of molecularly targeted therapies, in turn, has motivated the development of the field of nanoparticle-based imaging. Nanoparticles provide an attractive agent to address the molecular imaging requirement, as they possess the ability to amplify a contrast signal by incorporating large numbers of reporting elements, unique physicochemical properties,¹ the ability to modulate pharmacokinetics through surface chemistry ,^2–4 and the ability to combine multiple functions in a single scaffold.⁵ Building on this idea, the integration of imaging and therapeutic capabilities into single-nanoparticle systems has also been realized, permitting confirmation of drug delivery to tumor sites^6–8 as well as image-guided selective tumor ablation.⁹

While the integration of multiple functions into a single nanostructure appears attractive, it can diminish the efficacy of the individual functions because of space and surface-chemistry limitations in these tiny platforms. For instance, magnetic nanoparticles and drug molecules have been co-encapsulated in liposomes to simultaneously achieve multiple functions,¹⁰ however, this often results in a reduced loading capacity and stability with respect to a single-component liposome. Alternatively, recent reports have suggested separation of functions into two or more nanoparticle formulations to simplify this drawback.^11–14 Such research is aimed at engineering combinations of diverse nanoparticles that can cooperate in their diagnostic or therapeutic functions. Designing such combinations, however, remains challenging owing to a rudimentary knowledge of simultaneous cellular interactions with multiple, distinct types of nanoparticles. An urgent need in developing imaging and therapeutic strategies, therefore, is the cataloging of differences in nanoparticle endocytosis and intracellular transport as well as their effect on gene regulation behavior when two or more nanoparticle formulations concomitantly interact with the cell.

Prior work focusing on the endocytosis of single-nanoparticle formulations has uncovered the different factors, notably the physico-chemical properties,^15–18 that govern the process. The endocytosis process is usually energy-dependent and receptor-mediated, as has been carefully elucidated by recent studies ^{16, 19, 20} The surface properties of the nanoparticles can be adjusted by a variety of modifications in order to target specific endocytic receptors and boost uptake rate, and has been widely exploited for smart nanocarrier design.^{21, 22} Following endocytosis, the nanoparticles undergo intracellular transport and localization that can be broadly divided into the following steps: formation of early endosome, fusion of endosome and transport to lysosome, which is the general destination except for cases with special targeting ligand modification or lysosome escaping design.^{23, 24}

To catalog and probe these processes when the cells interact with a combination of nanoparticles, we selected oligonucleotide-modified gold nanoparticles (OGNs) and streptavidin-coated iron oxide nanoparticles (IOPs) as our model system. Our choice of the model system was governed by the versatility and demonstrated utility of these individual nanomaterials in the literature,^{8, 25} and, crucially, by use of their close analogues (gold nanorods and iron-oxide nanoworms) in designing a cooperative nanomaterial system.¹¹ Gold nanoparticles (AuNPs) are bioinert, can be easily synthesized and functionalized, and have proven to be effective nanocarriers in gene delivery systems,^{7, 26} Becuse antisense oligonucleotide therapies are powerful gene-therapy candidates for clinical treatments of cancer and HIV/AIDS among other disorders,^27–31 significant attention has been focused on developing antisense OGNs. OGNs display prominent advantages over traditional liposome carriers, including greater knockdown of gene expression, higher endocytic efficiency, stronger binding affinity for targets, nuclease resistance, and lower cytotoxicity.^{25, 32} On the other hand, IOPs have been harnessed in the clinical setting as magnetic resonance contrast probes, for instance, in the imaging of lesions in the reticuloendothelial system organs, such as the liver and lymph nodes.^{33, 34} In particular, IOPs decorated with streptavidin have shown the ability to home to diseased tissues through interactions with multiple tissue-specific receptors.⁸

In this article, we specifically focus on understanding the (differences in) endocytosis, trafficking and downstream impact of the OGNs while the cells are simultaneously exposed to the streptavidin-coated IOPs (Figure 1A). Here, the antisense oligonucleotides for the surface modification of AuNPs were designed to suppress the expression of green fluorescent protein (GFP). For visualization and quantification purposes, OGNs were labeled with cyanine-3 (Cy3) dye through oligonucleotide hybridization, and IOPs were labeled with Alexa 488 through streptavidin-biotin conjugation (Figure 1B). This design allowed us to shed light on the relatively unexplored and poorly understood phenomena of simultaneous cellular interactions with diverse nanoparticles, which resides at the heart of emergent multi-nanoparticle systems that are being developed for theranostic purposes.

Figure 1.

(A) Schematic depicting the process used to synthesize the Cy3-labeled OGNs and Alexa-488-labeled IOPs as well as the subsequent uptake and trafficking of the nanoparticles co-incubated with the mammalian cell. Here, AuNPs were first conjugated with thiol-modified oligonucleotides and then hybridized with complementary Cy3-labeled oligonucleotides. Streptavidin-modified IOPs bind with biotin-labeled Alexa 488. (B) Representative duo-color confocal image showing the distribution of OGNs (red) and IOPs (green) inside a HeLa cell following 2 h of incubation. Yellow signal depicts the colocalization of OGNs and IOPs. Scale bar 10μm.

2. Results and Discussion

2.1. Decoding Cellular Interactions with Streptavidin-Coated IOPs

Commonly used 15 nm diameter IOPs, which were coated with 5 nm streptavidin, were employed as one of the nanomaterials in the model system. IOP concentrations ranging from 0.5 to 2μg/mL have been regularly employed for in vitro cellular studies without significant cytotoxicity.³⁵ As can be evidenced by comparing Figure 2A (control, i.e., where no nanoparticles are used) with Figure 2B–D, fluorescent imaging of NIH 3T3 fibroblast cells displays substantial changes in cell morphology and size (Figure S1) when the cells were exposed to a high concentration of IOPs. We also separately tested the cytotoxicity of such streptavidin-modified IOPs in the 3T3 cells as well as HeLa cells at IOP concentrations of 1, 5 and 25μg/mL after an incubation period of 2 h. Overall, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assays revealed an increasing adverse impact of the nanoparticles with increasing concentration. Both kinds of cells suffer from significant cytotoxicity at an IOP concentration of 25μg/mL, although HeLa cells display higher tolerance to the IOPs (Figure 2F). Additionally, we quantified the endocytosis of Alexa-488-labeled IOPs (Figure S2) in the HeLa cells using flow cytometry (Figure 2E). The fluorescence-activated cell sorting (FACS) results revealed that the amount of nanoparticle uptake in the HeLa cells is nearly linear at the lower end of the IOP concentration range – but this behavior changes markedly when the cells are exposed to higher IOP concentrations. Such saturation curves are commonly observed for receptor-mediated endocytosis, where the particles (at high concentrations) compete for receptor sites on the cell surface,³⁶ while the initial uptake rate is proportional to the active surface area.³⁷ Crucially, the plateauing behavior of IOP endocytosis at the higher concentrations indicates that if other nanoparticles were to share the same set of endocytosis pathways (as the IOPs), it would create a competitive (rather than a cooperative) landscape likely resulting in lower-than-expected uptake rates for both sets of nanoparticles.

Figure 2.

(A-D) Fluorescent microscopy images of 3T3 cells with GFP expression following a 2 h incubation period with IOPs, at different concentrations of 0, 1, 5 and 25μg/mL. Scale bar: 10μm. (E) Amount of Alexa-488-labeled IOPs internalized by the Hela cells was quantified using FACS, with measurements being performed in triplicate. Average and standard deviation values of the IOP intensity recorded over ca. 10 000 cells are provided along with the nonlinear curve-fit (Hill). (F) MTT analysis [(average and standard deviation (N =5)] of HeLa and 3T3 cell viability when incubated with different concentrations of IOPs. Significant differences in viability were noted when the cell is incubated with 25μg/mL IOPs and when no nanoparticles are used (control) (p-value < 0.05).

2.2. Probing the Simultaneous Endocytosis of OGNs and IOPs

Because the endocytosis rate is strongly related to the nanoparticle size,^{36, 38} we sought to minimize the size difference between IOPs and OGNs to reduce the impact of such dimensional variations on our observations. Here, we used AuNPs with 15nm diameter, that is the same as that of the aforementioned IOPs. It is worth noting that the antisense oligonucleotide has 27 bases, corresponding to a 9nm-long double-helix strand. Considering that the single strand is usually shorter than the full length of the double strand, the size of the modified nanoparticle was estimated to be ~20nm.³⁹ Cy3-labeled oligonucleotides were hybridized with the antisense oligonucleotide for visualization and quantification purposes with confocal imaging and FACS, respectively (Figure S3). We mixed the OGNs (at a concentration of 1 μg/mL) with IOPs of different concentrations in the cell culture medium, and performed confocal imaging and FACS after 2 h of incubation with the HeLa cells.

The fluorescent images (Figure 3A–D) show the endocytosis of OGNs while cells were also incubated with IOPs at the previous set of studied concentrations, that is 0, 1, 5 and 25 μg/mL, respectively. These images do not exhibit any obvious differences in terms of the amount of OGNs taken up by the HeLa cells. The flow cytometry measurements (Figure 3E) further reinforce the confocal imaging observations. For the FACS measurements, ca. 10 000 cells were measured for each of the samples. While there is an evident shift in the fluorescence levels from the control sample, no differences between any of the other samples could be discerned thereby indicating the similarity in the level of uptake of the OGNs. Together, the confocal imaging and FACS measurements confirm that the endocytosis of IOPs do not have an impact on the number of OGNs internalized by the cell.

Figure 3.

(A-D) Representative confocal images showing the uptake of OGNs in HeLa cells when both OGNs and IOPs (at 0, 1, 5, and 25μg/mL) are co-incubated with the cells. Red signals emanate from the Cy3 labeling of the OGNs. Scale bar: 10μm. (E) Flow cytometry results showing the endocytosis of Cy3-labeled OGNs as the IOP concentration was increased from 0 to 25 μg/mL. 10 000 cells were counted for each sample.

In combination with the previous result of the IOPs uptake as a function of its concentration, one can reasonably infer that there is little competition between these two types of nanoparticles owing to the utilization of distinct endocytic pathways by them. We attribute this primarily to the DNA surface modification of the AuNPs as opposed to the streptavidin coating of the IOPs. Previous research suggests that the pathway of endocytosis of spherical nucleic acid nanoparticle conjugates is mediated by lipid rafts, in particular caveolae.^{17, 40} Mirkin and co-workers have noted the importance of scavenger reporters (specifically, SR-A) in specific recognition of and strong binding to the dense oligonucleotide shell of such nanoparticle conjugates.¹⁷ Following such binding, OGNs were postulated to enter cells via the caveolae-mediated pathway, owing to the close proximity of the nanoparticle conjugate, SR-A, and the lipid raft microdomains. In contrast, unmodified gold nanoparticles do not exhibit the same predilection for caveolae-mediated pathway.⁴¹ Indeed, cells that were pretreated with chlorpromazine, which inhibits the formation of clathrin-coated vesicles, were observed to have significantly reduced uptake of 15 nm AuNPs.

2.3. Putative Interference of IOPs in the Gene Regulation Function of OGNs

While spherical nucleic acids have been proposed as therapeutic payloads via gene regulation, less is known at this time about how other nanoparticles may interfere in this key function during simultaneous entry and uptake. To address this unmet need, we examined if and how the gene regulation activity of OGNs is affected by the concomitant uptake of the IOPs. Previous work has demonstrated the efficacy of DNA-AuNP for antisense gene regulation, where the unique ensemble properties of the conjugate confer several important advantages in the context of intracellular target recognition and binding.²⁵ Consistent with the demonstrated results, the use of the antisense oligonucleotide for the OGNs in our experiments induces downregulation of GFP level in the 3T3 fibroblasts.

Clearly, the confocal images (Figure 4A–B) obtained following a 24-hour incubation period show that the OGNs induce significantly lower GFP intensity in the fibroblasts. In sharp contrast, the control case where nonsense oligonucleotides were employed exhibit similar fluorescence levels to the nontreated cells. FACS measurements verified that the GFP expression was reduced by ca. 10% when the 3T3 cells were incubated with 1 μg/mL of OGNs (Figure 4F). Subsequently, we performed an independent set of measurements where the IOPs and OGNs were simultaneously added to the cell culture medium. Remarkably, the level of GFP downregulation, as seen from the confocal images, do not show any perceptible difference – irrespective of whether the IOP concentrations was 1, 5, or 25 μg/mL (Figure 4C–E). This observation was also confirmed by FACS, as shown in Figure 4F. Although GFP downregulation of OGNs alone is well known and has been demonstrated by previous research studies (as OGNs efficiently scavenge intracellular DNA or RNA²⁵), our results provide the first indication that such gene regulation activity may remain unperturbed even when the cell is exposed to other nanoparticles. We note, though, that this observation may not be generalizable to all other nanoparticles, and further investigation is required to understand the specific parameters of the types of nanoparticles that preclude interference in gene regulation.

Figure 4.

(A-E) Representative confocal images displaying the lack of interference of IOPs in the gene regulation activity of the OGNs. Scale bar: 10μm. (F) Flow cytometry results based on triplicate measurements confirmed that the action of OGNs remains consistent, despite the different concentrations of IOPs used during the incubation process (p-value < 0.05).

2.4. Analysis of Intracellular Localization of OGNs and IOPs

Intracellular trafficking of NPs plays a critical role in the cellular fate of nanoparticle-based therapeutics and imaging agents. Hence, a better understanding of the simultaneous intracellular transport of multiple types of nanoparticles is imperative in designing efficient cooperative systems. Standard intracellular transport of nanoparticles is typically initiated by endocytic vesicles pinching off the cellular membrane, and the early endosomes act as a hub to guide further trafficking of the cargo to various cellular destinations (such as lysosomes) based on the sorting signals.^{23, 42, 43} Here, we are interested in studying the localization (or the lack thereof) of the diverse nanoparticles inside the mammalian cells to elucidate whether these share the trafficking pathways. To visualize this phenomenon, we initially used equal concentrations of IOPs and OGNs and performed time-lapse co-localization analysis. Because the size of the nanoparticles (here, 20nm) and the relevant subcellular structures are smaller than the diffraction-limited resolution of an optical imaging system, the different nanoparticle(s) or cellular component(s) are imaged as a combined color because of spatial co-localization of the two linked objects.⁴⁴ This makes co-localization analysis a powerful tool to reveal the intricate interactions of cells and nanoparticles.^{17, 23, 45}

Figure 5A shows representative false-colored confocal images obtained from the HeLa cells following incubation of the two sets of nanoparticles for 1, 2, and 4 h, respectively. The red color indicates the presence of OGNs, and the green color represents the IOPs. The yellow shade is indicative of the presence of both types of nanoparticles within one voxel. Evidently, most of IOPs and OGNs are located near the perinuclear area after 1 h incubation but spread more widely in the cytoplasm with time. The time-lapse images also reveal the partial co-localization of the two types of nanoparticles. To better quantify the degree of co-localization between the OGNs (manifested through Cy3 fluorescence) and IOPs (indicated by Alexa 488 expression), we used the threshold Mander’s co-localization coefficients (tM).⁴⁵ The quantitative analysis (Figure 5B) showed that, on average, ca. 61% (std. dev. of 10%) of OGNs are co-located with IOPs following 1 h of incubation. The co-localization percentage of OGNs decreased slightly when the incubation time increased to 2 hours, while no substantial difference was observed between 2 h and 4 h of incubation. A similar trend, with a relatively smaller change as a function of incubation time, is also observed by analyzing the tM for IOPs. On the basis of these observations, we reason that both types of nanoparticles are initially partially co-localized in endosomes and lysosomes in the perinuclear region but are subsequently dispersed more widely in the cytoplasm. The latter step is critical, as the presence of the OGNs in the lysosomes alone would inhibit the gene regulation activity.⁴⁰ Our hypothesis at the 1 h mark is strengthened by prior reports of high degree of co-localization of diverse nanostructures (including photoluminescent nanodiamonds,⁴⁶ quantum dots,⁴⁷ and AuNPs⁴⁸) with early endosomes and lysosomes. We also performed co-localization studies using different concentrations of IOPs while keeping the concentration of OGN fixed at 1 μg/mL, whose results are provided in the Supporting Information (Figure S4).

Figure 5.

(A) Time-lapse duo-color confocal images recorded after the HeLa cells were co-incubated with 1 μg/mL of OGNs (red) and IOPs (green) for 1, 2, and 4 h. Yellow color is indicative of the presence of both types of nanoparticles within one voxel. Scale bar: 5μm. (B) Co-localization analysis shows that OGNs were partially colocalized with IOPs initially, but the degree of co-localization decreases with incubation time. On the basis of independent measurements of a set of 10 cells, the threshold Mander’s co-localization coefficients (tM) for the OGNs were computed as 61±10, 46±13 and 45±13% at 1, 2, and 4 h respectively, whereas those for the IOPs were 41±9, 35±12 and 34±10% for 1, 2, and 4 h, respectively.

3. Conclusions

Understanding the interaction of nanoparticles with cells is of high interest because of their widespread prevalence in medicine, nanotechnology, and environment. While potential applications for transporting drugs and genes in disease therapeutics have received considerable attention,^{1, 13, 22} some of the most fundamental aspects such as the processes of cellular uptake, trafficking and gene regulation activity are less known, particularly when the cell interacts with a multiplicity of nanoparticles. In an effort to improve the thematic understanding of multiple nanoparticle cell interactions, we report a series of systematic experiments featuring clinically relevant nanomaterials, that is OGNs and IOPs as the model system. Our choice of this system was motivated by their individual therapeutic/diagnostic attributes as well as recent endeavors to use this combination of materials to design cooperative, synergistic therapies that could potentially reduce the required dose of anticancer drugs and mitigate toxic side effects.^{11, 12}

Here, in competitive uptake studies, the OGN and IOP nanoparticles were co-incubated with HeLa cells. Our observations indicate that there is minimal, if any, influence of each particle on the other’s uptake behavior. Specifically, no reduction in the internalization of the OGNs was observed despite significant increase in the IOP concentration indicating the presence of different (and, hence, non-competitive) endocytosis pathways for IOPs and OGNs. In general, such behavior may be attributed to a combination of factors, notably nanoparticle size, composition and surface properties.^49–51 Although IOPs and OGNs have a similar size of ca. 20 nm, there are major differences in core composition and surface modification. To definitively determine the major driver behind the non-competitive uptake process, we repeated the co-incubation study using a combination of unmodified AuNPs with the same core composition as the OGNs (but naturally distinct surface properties). Quantitative flow cytometry measurements revealed that the content of OGNs internalized was not modulated by the presence of the unmodified AuNPs– even when the concentration of the unmodified nanoparticles was increased to 25 times that of their oligonucleotide-modified counterparts (Figure S5). Our data underscore that varying the surface properties of these nanoparticles results in significant changes in their uptake mechanism that could be leveraged for subcellular targeting into specific endocytic pathways and cellular organelles of choice.

Our fluorescence microscopy and flow cytometry analyses also show that the presence of IOPs in the co-incubation studies does not impede the gene regulation activity of the OGNs. Notably, though, the time-dependent co-localization studies reveal a complex pattern of intracellular trafficking and sorting of these two types of nanoparticles. Our observations indicate that, initially, the OGNs are partially co-localized in endosomes with the IOPs in the perinuclear region. At later times, a fraction of them appear to escape the endosomes and are found dispersed in the cytosol, where they are likely involved in the antisense gene regulation pathways. Previous research combining immunofluorescence and transmission electron microscopy ultrastructural analysis supports this inference.⁴⁰

In addition to focusing on the above features for designing superior cooperative nanoparticle systems, future investigations will also need to probe the possible intracellular degradation of these types of nanoparticles. Such degradation of IOPs in particular and the associated release of free iron ions can have substantial effects on cell homeostasis⁵² with high intracellular concentrations of IOPs reported to affect the actin cytoskeleton resulting in diminished cell proliferation. However, driven in part by the wide variability in types of IOPs or cells used and differences in incubation conditions, much of the reported evidence are not concordant thereby necessitating further in-depth investigations into this aspect.⁵³

Overall, our work proposes a research model for studying the interactions of diverse nanomaterials and cells with the goal of guiding the future design of cooperative, synergistic therapies that combine distinct tumor treatment (or imaging) functionalities while reducing overall dosage levels. Our measurements provide important collective insight into the uptake, transport, and gene regulation when the cell simultaneously interacts with multiple types of nanoparticles. Our ongoing investigations will focus on advancing precise, site-specific delivery of these cooperative nanoparticles as well as on addressing non-specific, differential uptake of the two types of nanoparticles by the mononuclear phagocyte system.

4. Experimental Section

4.1. Preparation of AuNPs and Oligo DNA Conjugation

A detailed protocol for the preparation of AuNPs and their surface modification is provided in a previous article.⁵⁴ AuNPs (15nm BBI) were stabilized with the adsorption of bis (p-sulfonatophenyl) phenylphosphine dihydrate dipotassium (BSPP, Sigma Aldrich). The choice of nanoparticle size was based on previous studies featuring spherical nucleic acid nanoparticle conjugates^{17, 25, 40} and superparamagnetic IOPs⁵⁵ that sought to optimize nanoparticle circulation in vivo as well as their interactions with biosystems in manners mimicking biomolecules. BSPP was added to the colloidal nanoparticle solution and the mixture was shaken overnight at room temperature. The resulting mixture was centrifuged at 3000 rpm for 30 min and the supernatant was removed. AuNPs were then resuspended in 1mL of BSPP solution (2.5mM). Upon mixing with 1mL of methanol, the mixture was centrifuged, the supernatant was removed and the AuNPs were resuspended in 1 mL of BSPP solution (2.5mM).

The disulfide bond in the thiol-modified oligonucleotides was reduced to monothiol using tris(2- carboxyethyl)phosphine (20mM, 1h). The oligonucleotides were purified using size exclusion columns (G-25, GE Healthcare) to get rid of the small molecules. Monothiol-modified oligonucleotides and phosphinated AuNPs were then incubated with DNA-to-Au molar ratio more than 200:1 in 0.5 × TBE buffer containing 50 mM NaCl for 40 hours at room temperature. AuNP-DNA conjugates were washed with 0.5 × TBE buffer in an Amicon Ultra 0.5 mL centrifugal filter (Millipore, Billerica, MA) to get rid of the extra oligonucleotides. The concentration of these AuNP-DNA conjugates was estimated from the optical absorbance at ~ 520 nm.

The following DNA sequences purchased from Integrated DNA Technologies were used in our work:

1. 5′ ATGATATAGACGTTGTGGCAAAAAAAA-thiol 3`

2. 2. 5′ GCCACAACGTCTATATCAT-Cy3 3`

3. 3. 5′ GCCACAACGTCTATATCATAAAAAAAA-thiol 3`

4.2. Preparation of IOPs and Alexa 488 Conjugation

Streptavidin-functionalized IOPs (15 nm, Millipore Sigma) were first incubated with Alexa 488-Biotin (1ug/mL, Millipore Sigma) for 30min. Subsequently, the IOP Alexa-488 conjugates were washed with phosphate-buffered saline (PBS) in an Amicon Ultra 0.5 mL centrifugal filter (Millipore, Billerica, MA) to get rid of the extra Alexa 488-biotin.

4.3. Agarose Gel Electrophoresis

Agarose (1%, Sigma) dissolved with 0.5 × TBE buffer was used for gel electrophoresis. Nanoparticles (20μL) containing 50% sucrose were added to each lane, and the gel was run at 100 V for 40 min. The running buffer was 0.5 × TBE buffer. The images of gels were recorded using a gel imager (Bio-Rad).

4.4. Cell Culture

HeLa and NIH 3T3 cells were incubated at 37℃, 5%CO2 and Dulbecco’s modified Eagle’s medium (DMEM) (Millipore Sigma) culture medium with 10% fetal bovine serum (Millipore Sigma). HeLa cells were chosen because of their extensive use in understanding nanoparticle endocytosis,⁵⁶ while the 3T3 fibroblasts provide a suitable test bed for transfection investigations.⁵⁷

4.5. Confocal Imaging

The DNA oligo attached on the AuNPs hybridizes with complimentary strands with Cy3 fluorophore. Leica SP8 confocal was used for Cy3 imaging with 561nm laser excitation, and the emission channel was set to 570 – 620 nm. The IOPs were labeled with Alexa 488 while the chosen 3T3 cell line expresses GFP stably. For Alexa 488 and GFP imaging, 488nm laser was used for excitation while the emission channel was set to 500 – 550 nm.

4.6. Cell Viability Assay

Cells were cultured in a 96-well microplate overnight and then incubated with nanoparticles. Next, the MTT labeling reagent (Sigma) was added and the cells were incubated in a humidified atmosphere for 4 h. Solubilization solution is then added followed by further incubation in a humidified atmosphere overnight. Finally, the microplate was evaluated with the use of an ELISA reader at 550– 600 nm with a reference wavelength of >650 nm.

4.7. Flow Cytometry

Cells were incubated with nanoparticles for the time, as detailed in the Results and Discussion section. Before measurements, the cell growth medium was removed, and the cells were washed three times with 1×PBS. Then, trypsin was added to each sample and the samples were incubated for 2 min at 37 ºC. Then, DMEM was added and the resulting cell suspensions were transferred to tubes. Cells (10 000) were analyzed using an FACS canto flow cytometer (BD Biosciences, USA). Consistent gating based on cell size and granularity (forward and side scatter) was applied to select the fluorescence signals of counted cells.