Subscriber access provided by READING UNIV
Article
Impact of crosslinker valency on gold nanoparticle aggregate formation and cellular uptake Alice Liu, and Jacob M. Berlin Langmuir, Just Accepted Manuscript • DOI: 10.1021/acs.langmuir.7b03524 • Publication Date (Web): 22 Nov 2017 Downloaded from http://pubs.acs.org on November 27, 2017
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Langmuir is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Langmuir
Impact of crosslinker valency on gold nanoparticle aggregate formation and cellular uptake Alice T. Liu, Jacob M. Berlin* Department of Molecular Medicine, Irell and Manella Graduate School of Biological Sciences, Beckman Research Institute, City of Hope National Medical Center, Duarte, California 91010, United States
ABSTRACT: Synthesis of spherical, biocompatible nanoparticle aggregates using a small molecular crosslinker is a simple and flexible approach for the controlled assembly of gold nanoparticles. This strategy can be extended to a variety of crosslinkers, making it possible to the test the effect of crosslinker properties on aggregate formation and physicochemical properties. Here we synthesized aggregates using a series of structurally homologous crosslinkers with differing valencies. These aggregates have the same size, morphology, surface charge, surface coating, and stability in salt, media, and low pH conditions, but they differ in their stability to cyanide etching and uptake by cells. This highlights the fine tuning of nanoparticle aggregate properties that can be achieved by using small molecule crosslinkers.
INTRODUCTION Inorganic nanoparticle aggregates are an emerging class of nanomaterials that have attracted interest in a variety of biological applications, including imaging1, ex vivo diagnostics2, and tumor targeting3, 1 ACS Paragon Plus Environment
Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 2 of 22
and as model systems for understanding the clearance and elimination of nanoparticles4. The most common strategy for preparing aggregates is through the hybridization of DNA-functionalized nanoparticles with complementary DNA linker strands4-5, which are highly programmable and allow for the rational design of nanoparticle superstructures. However, one disadvantage with this strategy is that the DNA linkers themselves could potentially be bioactive. Other strategies include using block copolymers6 or micelles1c for non-covalent assembly, which has promising applications but limited control over aggregate assembly. We have previously reported on a synthesis method for the controlled assembly of biocompatible inorganic nanoparticle aggregates using a commercially available small molecule crosslinker, pentaerythritol tetrakis(3-mercaptopropionate) (TetLink)7. Using a small molecule crosslinker offers the flexibility for modifying linker properties (e.g., hydrophobicity, length, and valency), and consequently, modifying aggregate properties. In this study we investigate the effect of crosslinker valency on aggregate synthesis and physicochemical and biological properties. Two commercially available crosslinkers were chosen based on their structural similarity to TetLink: trimethylol tris(3-mercaptopropionate) (TriLink) and 1,4-butanediol bis(3-mercaptopropionate) (BiLink). Here we report that our synthesis method can be extended to crosslinkers with different valencies, allowing us to probe the effect of the crosslinker on the physicochemical properties of the resulting nanoparticle aggregates.
EXPERIMENTAL SECTION Materials Citrate-stabilized gold colloid suspensions (5, 10, 15, 20, and 30 nm). Pentaerythritol tetrakis(3mercaptopropionate) (TetLink) and trimethylol tris(3-mercaptoproprionate) (TriLink) were purchased from Sigma-Aldrich, and 1,4-butanediol bis(3-mercaptopropionate) (BiLink) was purchased from Wako 2 ACS Paragon Plus Environment
Page 3 of 22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Langmuir
Pure Chemical Industries, Japan. PEG2000-maleimide (M.W. 2000) was purchased from Nanocs. Phenol free-RPMI 1640 media was purchased from Life Technologies. DMEM (high glucose, L-glutamine, no sodium pyruvate) was purchased from LifeTechnologies (ThermoFisher Scientific). MDA-MB-231 cells were purchased from ATCC (ATCC HTB-26). 10x phosphate-buffered solution (PBS) (Potassium chloride 2 g L-1, Monopotassium phosphate 2.4 g L-1, Sodium chloride 80 g L-1, Disodium phosphate 14.4 g L-1, Tris Ultrapure 24.2 g L-1) was purchased from Corning. CellTiter 96® AQueous One Solution Cell Proliferation Assay (MTS) was purchased from Promega. ICP-MS grade nitric acid (70%) and hydrochloric acid (37%) were purchased from BD Aristar. ICP-MS gold calibration solution (100 µg mL-1) was purchased from Spex CertiPrep. UltraTrace Elemental Analysis Grade water was purchased from ThermoFisher. All other chemical compounds were purchased from Sigma-Aldrich. Aggregate Synthesis The ability of the crosslinkers to form aggregates was tested according to the synthesis protocol previously reported. Briefly, for TetLink and TriLink, a solution of 5 nm AuNPs in water (5 x 1013 particles mL-1) was added dropwise to a prepared solution of crosslinker in ethanol and water, and the reaction solution was shaken on a tabletop shaker at top speed for 2 h (Table S1). For BiLink, the order-ofaddition of the previous protocol had to be changed to form aggregates: a prepared solution of crosslinker in ethanol was added dropwise to a solution of 5nm AuNPs in water (5 x 1013 particles mL-1) (Table S2). Water was then added dropwise to the mixture, and the reaction solution was shaken on a tabletop shaker at top speed for 4 h. After shaking, the solutions were left on the benchtop for 24 h, and then terminated by the addition of an excess of PEG2000-maleimide and an additional 2 h of shaking (Table S3). Aggregates were then filtered using a 25 mm 0.2 µm polycarbonate track etch membrane in a Swinnex housing. The filtrate was washed three times by centrifugation (10 000 x g for 10 min) and resuspended in diH2O. As an example, reagent calculations for synthesizing 100nm aggregates out of 5
3 ACS Paragon Plus Environment
Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 4 of 22
nm AuNPs are given in Table 1. Reaction parameters for varying crosslinker concentration and AuNP size are included in the supplemental information (Table S4). Crosslinker
Final cross- Volume linker con- AuNPs (µL) centration
Crosslinker concentration in ethanol (mg mL-1)
Volume crosslinker solution (µL)
Additional ethanol (µL)
Volume H2O (µL)
Volume PEGmaleimide (100 mg mL-1 in H2O) (µL)
TetLink 2.4 mM 500 8 116 0 173.2 167.1 TriLink 0.48 mM 500 4 37.7 78.3 173.2 86.1 BiLink 3.2 mM 500 20 42.7 N/A 457.3 141.0 Table 1. Reagent calculations for synthesizing 100 nm aggregates using the three crosslinkers and 5 nm AuNPs.
Aggregate characterization Hydrodynamic diameter and surface charge of prepared aggregates were characterized by dynamic light scattering (DLS) and zeta potential measurements on the ZetaPALS instrument (Brookhaven Instruments Corporation). UV absorption spectra were taken on the Ultrospec 3000pro (GE Lifesciences). Aggregates were visualized by transmission electron microscopy (TEM). 3.5 µL of aggregates solution was dried onto a formvar-stabilized 200 mesh copper–carbon grid purchased from Ted Pella. TEM images were taken using FEI Tecnai 12 Twin. Particle concentration and the number-weighted average size of aggregates were also measured by Nanosight (Malvern Instruments). Concentrated aggregate samples were diluted 1000x to 2000x with diH2O and injected onto the Nanosight. Based on the measured particle concentrations, the TriLink, TetLink, and BiLink aggregates were then normalized to have the same particle concentration (8.3 x 1011 particles mL-1). Organic composition of aggregates was determined by analyzing the same batch of aggregates using thermogravimetric analysis (TGA) and inductively coupled plasma mass spectrometry (ICP-MS). TGA was used to measure total organic content and ICP-MS was used to measure sulphur to gold ratio. Thus, Ultra Trace Elemental Analysis Grade (EA) water was used for all steps of aggregate synthesis and 4 ACS Paragon Plus Environment
Page 5 of 22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Langmuir
washing that require H2O in order to minimize any background sulphur. Calculations and equations used to determine weight% of crosslinker and PEG for each aggregate using paired ICP-MS and TGA analysis are shown in Table S5. Aggregate stability measurements Aggregates with a HD of approximately 100 nm were synthesized from 5nm gold particles for each crosslinker and normalized by optical density (OD) as measured by UV-vis absorption spectrometry (Table S6-7). For media stability, aggregates were incubated in complete cell media (phenol-free RPMI 1640 supplemented with 10% fetal bovine serum). For salt stability, aggregates were incubated in a final concentration of 1x PBS. For cyanide stability, aggregates were incubated in solution containing complete cell media and final cyanide concentration of 0.1 M. For pH stability, aggregates were incubated in either pH5 or pH 7 solution buffered with 0.02M sodium acetate. Aggregate stability was measured by UV-vis spectroscopy over a period of 24 h. Peak shift or broadening was determined by overlay of the UV spectra with that of control aggregates in diH2O. Normalized OD (absorbance) was calculated as the ratio of the optical density at λmax of the experimental aggregates to control aggregates. Cell viability and uptake Aggregates with a HD of approximately 100 nm were synthesized from 5nm gold particles for each crosslinker and normalized to the same particle concentration by Nanosight (Table S8). For cell viability experiments, MDA-MB-231 cells were incubated for 24 h with concentrations of TetLink, TriLink, or BiLink aggregates between 3.24 x 108 to 4.15 x 1010 particles mL-1. Viability was measured using the MTS cell proliferation assay in triplicate, with three experimental replicates. For cell uptake experiments, MDA-MB-231 cells were incubated for 24 h with TetLink, TriLink, or BiLink aggregates at a concentration of 1.04 x 1010 particles mL-1. The amount of gold within cells was quantified using ICPMS and normalized to the amount of gold initially added. Cell uptake was also visualized using TEM.
5 ACS Paragon Plus Environment
Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 6 of 22
RESULTS AND DISCUSSION Impact of Crosslinker Valency on Aggregate Formation In our previous report, TetLink at a concentration range of 0.12 to 2.4 mM resulted in the controlled assembly of spherical nanoparticle aggregates7. For TriLink, 0.16 mM TriLink seemed to be the lower threshold for aggregate formation. At this concentration misshapen aggregates are formed, whereas below this concentration the particles rapidly precipitated out of solution, and above this concentration (0.24 mM TriLink) spherical aggregates are formed. 0.16mM TriLink corresponds to an equivalent number of thiols per nm2 as TetLink, suggesting that a minimum excess of thiols to particle surface area (calculated to be 93 thiols nm-2) is required for controlled assembly. Whereas testing concentrations of TetLink above 2.4 mM was hindered due to limited linker solubility in ethanol, TriLink did not have such solubility issues. Therefore, concentrations of TriLink up to 16.0 mM were tested and found to be able to form spherical aggregates (Scheme 1A). Interestingly, when BiLink was tested for aggregate formation using the same protocol over the same range of concentrations, the nanoparticles rapidly agglomerated and precipitated out of solution. Optimization of aggregate synthesis using BiLink revealed some key parameters that had to be changed in order for controlled assembly to occur (Scheme 1B). Firstly, changing the order of addition made the biggest impact on aggregate formation. Instead of adding the gold nanoparticles dropwise to the linker solution, using BiLink required dropwise addition of the linker to the gold nanoparticles. Secondly, the reaction solution was shaken for 4 h instead of 2 h. TEM of the aggregates at various time points during the aggregation process revealed that aggregate formation via this protocol required 4 h to complete (Figure 1A). Extending the reaction to 24 h showed no difference in aggregate morphology compared to 4 h. Finally, with the above modifications, the minimum concentration of BiLink needed for aggregate formation was 2.4 mM, which is ten times the number of thiols per nm2 compared to TetLink and TriLink. Like TriLink, BiLink concentrations up to 16.0 mM were able to form spherical 6 ACS Paragon Plus Environment
Page 7 of 22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Langmuir
aggregates. However, aggregation using BiLink was much more sensitive to synthesis conditions (e.g., reaction scale, reaction vessel material, different batches of purchased gold colloid) compared to TriLink or TetLink, resulting in higher batch-to-batch variability in final aggregate size. Despite the need for reaction optimization and different synthesis conditions for BiLink, we hypothesized that the size of aggregates synthesized using these crosslinkers could still be tuned by changing crosslinker concentration, as outlined in our previous report7.
Scheme 1. Reaction optimization for aggregate synthesis using TetLink, TriLink, and BiLink crosslinkers. A) Synthesis schema for TetLink and TriLink and the range of crosslinker concentrations which are able to form stable aggregates. B) Synthesis schema for BiLink and the range of crosslinker concentrations which are able to form aggregates. 7 ACS Paragon Plus Environment
Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 8 of 22
As previously reported for TetLink, the final aggregate size can be tuned by changing either the AuNP concentration or the crosslinker concentration. Indeed, both BiLink and TriLink follow this same trend, where increasing the crosslinker concentration within its working concentration range resulted in a logarithmic increase in aggregate hydrodynamic diameter (Figure 1B). TEM images confirm this increase in size, and show that the aggregates maintain their spherical morphology over the entire range of sizes (Figure S1A). For 5 nm gold particles held at a constant particle concentration, increasing TetLink concentration from 0.12 to 2.4 mM resulted in a logarithmic increase in aggregate hydrodynamic diameter (HD) from 60 to 100 nm. Under the same conditions, increasing TriLink concentration from 0.24 to 16.0 mM resulted in a logarithmic increase in aggregate HD from 80 to 180 nm. Likewise, increasing BiLink concentration from 2.4 to 16.0 mM resulted in a logarithmic increase in aggregate HD from 80 to 150 nm. While the upper size limit of the aggregates does plateau due to filtration through a 0.2 µm membrane, the overall trend that increasing crosslinker concentration increases aggregate size is consistent with and without filtration (Figure S1B). Given the ability for all three linkers to tune the size of aggregates assembled out of 5 nm AuNPs, we hypothesized that the same rule would apply to aggregates assembled out of larger AuNP sizes. We first tested the ability of the three crosslinkers to assemble aggregates out of a variety of AuNP sizes. Surprisingly, decreasing linker valency resulted in an increase in the size range of nanoparticles that could be assembled into aggregates. TetLink was previously reported to also assemble aggregates out of 10 and 15 nm AuNPs, but 20 nm particles could not be assembled7. With TriLink, aggregates could be assembled out of 5, 10, 15, and 20 nm AuNPs; with 30nm AuNPs, there were only a few small clusters, and the solution was dominated by free particles. However, with BiLink, aggregates could be assembled out of 5, 10, 15, 20, and 30 nm AuNPs; 50 nm particles could not be assembled beyond small clusters of 2 to 3 particles. We hypothesize that TriLink and TetLink fail to assemble larger nanoparticles due to the decreased degrees of freedom for their thiol arms in comparison to BiLink, with TetLink 8 ACS Paragon Plus Environment
Page 9 of 22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Langmuir
being the most constrained crosslinker and having the smallest range of NP sizes it can assemble. As the ratio of AuNP diameter to crosslinker length increases, we hypothesize that it becomes more and more difficult for the linker to stably bridge the individual nanoparticles. We expect that linkers with more degrees of freedom can offset some of this instability by allowing the nanoparticles to rearrange into more stable conformations (Figure 1A), and thus expand the range of nanoparticle sizes that can be assembled. However, our hypothesis will need to be further tested using a larger set of linkers with varying degrees of rigidity.
Figure 1. A) Progression of aggregation using BiLink crosslinker and 10 nm AuNPs. One batch of aggregates was shaken at room temperature on a tabletop shaker at top speed for 24 h, and 3.5 μL aliquots were removed at each time point for imaging. At early timepoints the AuNPs were clumped together in disorganized clusters, which then rearranged over time into discrete, dense spherical assemblies. Aggregate formation was complete by 4 h, and there was no difference in aggregates at 4 h versus at 24 h. Scale bar 100 nm. B) Size range of aggregates that can be assembled out of 5 nm AuNPs for the different crosslinkers. Varying crosslinker concentration allows for the tuning of final aggregate size. Inset is zoomed in for 0.12 to 3.2 mM concentrations. 9 ACS Paragon Plus Environment
Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 10 of 22
To demonstrate that the size of aggregates assembled out of the larger AuNPs can also be tuned similar to the 5 nm AuNPs, we synthesized a set of aggregates with the same HD out of varying particle sizes for each crosslinker. Figure 2 shows the UV-vis absorption spectra and representative TEM image of aggregates with a HD of approximately 100 nm made with a range of nanoparticle sizes for the TetLink, TriLink, and BiLink crosslinkers. Interestingly, the plasmon resonance of the aggregates red-shift only slightly at smaller AuNP sizes, but aggregates assembled out of 20 and 30 nm AuNPs have a more dramatic shift in plasmon resonance. Modeling suggested that this phenomenon is due to magnetic dipole resonances, which have applications for plasmonic metamaterials8. For TetLink and TriLink, successful aggregation with the larger particles required not only changing crosslinker concentrations, but particle concentration as well, as previously reported for TetLink7 (Table S4). For BiLink, it should be noted that the sensitivity of the reaction made it difficult to control the final aggregate size of larger AuNPs. We found that washing the AuNPs once before aggregation allowed better control of final aggregate size, similar to a previous report that found washing gold particles prior to antibody conjugation improved particle stability9. Overall, these results show that by tuning linker valency, linker concentration, and particle concentration, a variety of permutations of final aggregate size and starting AuNP size can be synthesized and tailored for different applications. The flexibility of this platform allows for the synthesis of a set of matched aggregates that differ in only one parameter and probing the effects of that parameter on aggregate properties.
10 ACS Paragon Plus Environment
Page 11 of 22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Langmuir
Figure 2.Synthesizing 100 nm aggregates out of 5, 10, 15, 20, and 30 nm AuNPs using TetLink, TriLink, and BiLink crosslinkers. Changing the crosslinker changes the range of AuNP subunit sizes that can assemble into aggregates. Crosslinker concentration used and resulting aggregate hydrodynamic diameter are given for each crosslinker and AuNP size. Representative TEM images (scale bar 50 nm) show the structure of the aggregates and UV–vis spectroscopy indicates the characteristic red shift of maximal absorbance upon aggregation.
11 ACS Paragon Plus Environment
Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 12 of 22
Impact of Crosslinker Valency on Aggregate Physicochemical Properties Accordingly, we decided to further investigate the effect of crosslinker valency on the physicochemical properties of 100 nm aggregates composed of 5 nm AuNPs, which will be designated throughout the rest of the report as TetLink aggregates, TriLink aggregates, and BiLink aggregates. The surface charge of the aggregates was measured by zeta potential and found to be similarly negative for all three aggregates: -33.6 mV for TetLink aggregates, -32.8 mV for TriLink aggregates, and -23.0 mV for BiLink aggregates. The stock citrate-stabilized 5 nm gold particles also had a negative surface charge of -33.8 mV. Aggregate composition was determined by pairing thermogravimetric analysis (TGA) and ICP-MS. TGA was used to measure the weight percent of organic material (crosslinker + PEG2000-maleimide) incorporated into the different aggregates. ICP-MS was then used to measure the gold to sulphur ratio of the same batch of aggregates, from which the weight percent of crosslinker could be calculated. The weight percent of PEG2000-maleimide could then be calculated by subtracting the amount of crosslinker from total organic content (Table S7). Based on this analysis, all three aggregates had very similar PEGylation (4-5%) and crosslinker content (17-25%) (Figure 3). Thus, the physicochemical properties of all three aggregates are similar.
Figure 3. Aggregate composition showing combined weight% of PEG-maleimide and crosslinker as determined by paired TGA and ICP-MS compositional analysis. 12 ACS Paragon Plus Environment
Page 13 of 22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Langmuir
Crosslinker Valency Affects Cell Uptake Based on many studies in the literature showing that the size, shape, surface chemistry of nanoparticles affect their uptake by cells10, and the fact that these properties were all very similar between the TetLink, TriLink, and BiLink aggregates, we hypothesized that the aggregates would also would exhibit very similar cell uptake behavior. However, when these aggregates were incubated with MDA-MB-231 breast cancer cells, there was a surprising difference in cell uptake. BiLink aggregates were taken up the most, followed by the TetLink and TriLink aggregates, which both had extremely low cell uptake (Figure 4A). Cell uptake was also visualized using TEM, revealing internalization of all three aggregates within cellular vesicles (Figure 4B-D). Concurrent with the ICP-MS results, the BiLink aggregates were more abundant in the cells than either TetLink or TriLink aggregates. In fact, TEM imaging of multiple cell pellet sections showed that the majority of cells treated with TetLink or TriLink aggregates were completely devoid of aggregates. In order to understand why the aggregates showed such different cell uptake behavior, we investigated whether the aggregates were cytotoxic or were unstable and agglomerating11 under cell culture conditions. Cell viability experiments showed that none of the aggregates were cytotoxic, even at concentrations above that which was used for the cell uptake studies (Figure 4E). The stability of the aggregates under different conditions was measured by monitoring changes in the UV-vis absorption spectra. A broadening of the extinction peak or a red-shift in λmax would indicate further aggregation or agglomeration, while a decrease in absorbance at λmax would indicate a decrease in aggregate concentration. In complete cell media, all three aggregates were completely stable for 24 h with no change in peak shape or drop in absorbance (Figure 4F). In 1x PBS, all three aggregates showed the same changes in UV-vis spectra — while there was no change in peak shape, there was a drop in absorbance at their λmax after 60 min, indicating a decrease in aggregate concentration (Figure S2). Indeed, we observed that in the presence of salt, the aggregates became “sticky” and irreversibly coated plastic and glass surfaces. pH stabil13 ACS Paragon Plus Environment
Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 14 of 22
ity was also tested because the ester linkages in the crosslinkers could theoretically be hydrolyzed at low pH, leading to aggregate instability. However, the aggregates were similarly stable at pH 5 and pH 7, where the salt buffer caused a decrease in absorbance but otherwise no differences in UV-vis spectra (Figure S3). In summary, the three aggregates were completely non-toxic and showed no differences in stability under cell culture conditions. Therefore, these factors do not account for the differences in cell uptake.
14 ACS Paragon Plus Environment
Page 15 of 22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Langmuir
Figure 4. Crosslinker valency affects cell uptake and cyanide stability. A) ICP-MS analysis of cell uptake by MDA-MB-231 cells treated with aggregates (1.04 x 1010 particles mL-1) for 24 h. Cell uptake was done in triplicate, with 3 experimental replicates. **** p
