Communication pubs.acs.org/bc
Affinity-Based Assembly of Peptides on Plasmonic Nanoparticles Delivered Intracellularly with Light Activated Control Demosthenes P. Morales, William R. Wonderly, Xiao Huang, Meghan McAdams, Amanda B. Chron, and Norbert O. Reich* Department of Chemistry and Biochemistry, University of California, Santa Barbara, California 93106, United States S Supporting Information *
ABSTRACT: We report a universal strategy for functionalizing near-infrared light-responsive nanocarriers with both a peptide “cargo” and an orthogonal cell-penetrating peptide. Modularity of both the cargo and the internalization peptide attachment is an important feature of these materials relying on the robust affinity of polyhistidine tags to nitrilotriacetic acid in the presence of nickel as well as the affinity of biotin labeled peptides to streptavidin. Attachment to the gold surface uses thiol-labeled scaffolds terminated with the affinity partner. These materials allow for unprecedented spatiotemporal control over the release of the toxic α-helical amphipathic peptide (KLAKLAK)2 which disrupts mitochondrial membranes and initiates apoptotic cell death. Laser treatment at benign near-infrared wavelengths releases peptide from the gold surface as well as breaches the endosome barrier for cytosolic activity (with 105-fold improved response to peptide activity over the free peptide) and can be monitored in real time.
T
that intracellular localization could be achieved by fusion of cell-penetrating peptides to the protein of interest. To make the platform completely modular, here we employed an additional affinity handle to separate the cargo and the internalizing moiety, leading to a new class of customizable biomolecule delivery vehicles. We sought to translate this technology to biomedical applications. A promising peptide therapeutic is the 14 amino acid amphiphilic α-helical proapoptotic peptide domain (PAD), (KLAKLAK)2, that disrupts the extracellular membrane of prokaryotes and the mitochondrial membrane in eukaryotes.7,8 Unassisted, (KLAKLAK)2 inefficiently penetrates through the eukaryotic cytoplasmic membrane, requiring the fusion to an internalization moiety or introduction by a carrier.9 Penetration peptide fusions of (KLAKLAK)2 have shown significant improvements in eliciting cellular apoptosis having IC50 values in the 10 −7 −10 −6 M range compared to unmodified (KLAKLAK)2 in similar cell lines (10−4−10−3 M), but require at least 24 h to assay peptide efficacy.10,11 Furthermore, (KLAKLAK)2 delivered into mouse models of glioblastoma and primary Lewis lung carcinoma tumors using iron oxide nanoparticles and HPMA copolymers, respectively, shows dramatic improvements.3,12 However, further improvements in controlling when and where (KLAKLAK)2 is released are likely to minimize off-target effects. Our approach synergizes
he ability to intracellularly deliver proteins and peptides with spatiotemporal control has extensive basic science and biomedical applications. Unfortunately, despite many innovative approaches, severe limitations for the robust delivery of proteins and peptides into cells remain problematic. Significant challenges involve the stabilization of proteins, increasing their cellular uptake, and enhancing their release from endosomes. For example, recent efforts to decorate large protein complexes with nucleic acids to enhance cellular uptake shows some promise.1 Also, supercharged fusion proteins have been shown to promote cellular uptake when combined with commercial transfection reagents.2 Neither, however, provides spatial and temporal control of biomolecule release. Furthermore, light responsive delivery systems based on photosensitive polymers are often restricted to the UV and visible light regions which are limited by low penetration depths and possible UV damage.3 We recently demonstrated the use of hollow gold nanoshells (HGN) appended with the green fluorescent protein (GFP) as a means to deliver and control the release of histidine tagged proteins within individual cells.4 This technology provides a means to control the release of proteins with subcellular resolution, opening up new areas of basic research. Upon irradiation with a near-infrared (NIR) laser the HGN releases the thiol-scaffold-nitrilotriacetic acid moiety, thereby causing the his-tagged protein to be released as well. Coincidentally, this irradiation also results in the release of the protein from the endosome through membrane disruption caused by vapor bubble formation, thereby overcoming a critical obstacle to intracellular protein delivery.4−6 Previously, we demonstrated © XXXX American Chemical Society
Received: May 16, 2017 Revised: May 17, 2017 Published: May 19, 2017 A
DOI: 10.1021/acs.bioconjchem.7b00276 Bioconjugate Chem. XXXX, XXX, XXX−XXX
Communication
Bioconjugate Chemistry
STV to HGN was used to maximize biotin labeling and prevent self-assembly of particles. The STV activated particles were subsequently labeled with biotinylated cell-penetrating peptides (CPP), TAT (Biotin-YGRKKRRQRRRPQ, Abbr. STV-TAT), which facilitates efficient nanoparticle internalization through endocytosis.14 Furthermore, a polyhistidine variant with a GG spacer of the (KLAKLAK)2 (H6PAD) peptide was bound to the particle in the presence of Ni2+ providing high specificity attachment. Irradiation of these modular peptide carriers (HGN-MPC) with NIR light allows for the cleavage of the thiol-gold bond as well as the endosomal escape of the cytotoxic peptide for mitochondrial membrane disruption (Scheme 1). This construct provides a high level of modularity of various peptide targets relying only on the facile fusion of a polyhistidine tag and biotin to the termini of the peptides of interest. Release of bound cargo from HGN upon laser treatment is highly efficient and a result of the oscillation of surface electrons of the nanoparticle when the incident wavelength is resonant at its plasmon causing the cleavage of the thiol-gold dative bond.4,15,16 We sought to promote a modular and uniform delivery system and utilized a nucleic acid scaffold to promote high-efficiency release and stability. Nucleic acids are a robust coating on gold nanoparticle surfaces and eliminate possible incompatibilities associated with directly adsorbing peptides to gold surfaces. Furthermore, a double-stranded nucleic acid scaffold was selected due to the increased efficiency of thiol gold cleavage relative to single-stranded nucleic acids and flexible PEG based scaffolds (Figure 1A). This discrepancy may be attributed to the possible nonspecific base adsorption on the gold surface and random bends in the scaffold due to similar low persistence lengths of single-stranded nucleic acids and PEG compared to double-stranded oligos.17−19 Extension of nucleic acid strands may minimize possible contact of the scaffolds to the particle surface permitting increased proportion of cleaved strands. Upon assembly of the HGN-MPC, the average hydrodynamic diameter increases to ∼140 nm as observed by DLS. However, further TEM analysis indicates that that the overall shift in average size can be attributed to bridging of particles during streptavidin assembly but does not affect the plasmon at 800 nm (SI Figure 1). Furthermore, it was observed that the amount of STV-TAT on the particle surface influenced how many H6PAD were associated with the NTA sites on the HGN. Varying the number of STV-TAT on the HGN surface by varying the mole ratio of biotinylated DNA complement to unmodified DNA complement (nul) yields an altered loading capacity of the H6PAD. A maximal loading of ∼20 000 H6PAD per particle was determined after etching fluorescein labeled peptide from HGN with KCN (Figure 1B) then compared against a standard curve (SI Figure 2). It is important to note that the NTA is functionalized on the anchoring strand of the DNA scaffold; although only ∼6000 strands are found adsorbed to the nanoparticle surface (determined by Qubit fluorimeter assay after KCN etch), which may be due to the self-association of peptides in solution.20 Incorporating a 1:1 ratio of biotinylated complement with unmodified DNA reduced loading to ∼6500 peptides per particle (Figure 1B). Because the STV share some mole fraction of the nucleic acids that contain the NTA site (Scheme 1) we conclude that the decrease in PAD loading with increased biotin complement concentration can be attributed to possible steric hindrance of STV blocking available NTA sites for binding.
internalization and irradiation of the nanoparticle in cells to allow for real-time assaying of peptide activity. Plasmonic hollow gold nanoshells (40 nm average diameter) offer spatiotemporal delivery of the peptide cargo due to their strong absorption of biologically benign NIR light at 800 nm (SI Figure 1) to elicit the cleavage of the thiol scaffold tethering the peptides to the surface.13 The orthogonal assembly of two peptides to the hollow gold nanoshell relies on the use of robust affinity reagents tethered to the surface of the nanoparticle (Scheme 1). A thiol-DNA-amine anchoring strand Scheme 1. Hollow Gold Nanoshell (HGN) Mediated Delivery of H6PAD Peptide in Cellsa
a
HGN decorated with a proapoptotic peptide via nickel-NTA (NTA) affinity are internalized by an orthogonal peptide presented on the surface via streptavidin-biotin (STV, biotin-TAT) linkage using a double-stranded DNA scaffold. During the double-stranded nucleic acid assembly, varying ratios of DNA complement with either a biotin functional group or free base (nul) were employed to tailor loading capacity When activated with 800 nm (NIR) light the peptides are released to disrupt the membrane of mitochondria eliciting a cascade of signaling initiating apoptosis.
was adsorbed to the gold surface by self-assembly and further functionalized with a thiol derivative of nitrilotriacetic acid (NTA) using a heterobifunctional linker, NHS-PEG4-maleimide (Scheme 1).4 The thiol-NTA served two purposes: presentation of the protein binding handle at the terminus of the scaffold and to passivate free gold surfaces unoccupied by DNA tethers to improve stability. A biotinylated complement was then hybridized to the anchoring strand to assemble streptavidin (STV) to the particle. A large excess of ∼500 000:1 B
DOI: 10.1021/acs.bioconjchem.7b00276 Bioconjugate Chem. XXXX, XXX, XXX−XXX
Communication
Bioconjugate Chemistry
neuropilin-1 receptor (SI Figure 5).22 HeLa cells that lack nueropilin-1 overexpression, however, showed no indication of particle internalization. H6PAD was released by the cleavage of the associated thioldsDNA-NTA on the HGN surface after laser treatment using pulsed near-infrared laser light with a 1 kHz repetition rate. After exposing the HGN constructs to a pulsed 800 nm laser operating at 2.5 mJ cm−2 for 10 s, particles were then concentrated by centrifugation to extract the released DNA and peptide. The retained peptides on the HGN were eluted from NTA handles with an acidic buffer containing a 1% solution of TFA, and the peptide depleted particles were removed by centrifugation. Comparison of supernatants from laser treated and untreated HGN by HPLC showed that >80% of the H 6 PAD were released from the HGN (Figure 1C). Furthermore, this correlates well with the release efficiency described in Figure 1A where etching the nanoparticle cores with a buffer containing KCN to chemically release bound dsDNA showed ∼87% of the dsDNA scaffold was released following laser irradiation via DNA quantitation assay. The effect of peptide release on PPC-1 cellular viability upon laser excitation was then explored using varying concentrations of HGN constructs internalized in cells after 2 h and irradiation with 800 nm light. We utilized a femtosecond pulse NIR laser operating at 2.5 mJ cm−2 for 10 s which was shown to have no effect on PPC-1 cell viability in previous studies.4,15 PPC-1 cells treated with the free H6PAD peptide alone is largely ineffective and has an IC50 of 3.8 × 10−3 M (SI Figure 6). We reasoned that with facilitated delivery by HGN and escape from endosomes we could elicit a cellular response at much lower peptide concentrations relative to free H6PAD. When treated with nanoparticles at particle concentrations of 0.8 to 6.4 pM the HGN-MPC are benign and show no effect on cellular viability when left unexposed to 800 nm light, highlighting the temporal control of the particle system (Figure 2B, blue). The
Figure 1. (a) Percent release of scaffolds including single- and doublestranded nucleic acids and a 3k PEG based scaffold after 800 nm laser exposure with pulse fluence of 2.5 mJ cm−2 and 10 s exposure relative to non-irradiated HGN. Efficient release was observed in doublestranded nucleic acid scaffolds. (b) Loading capacity of H6PAD on dsDNA coated nanoparticles hybridized with varying mole ratios of nonbiotinylated (Nul) and biotinylated complement (Biotin). 1:1 ratio allows more streptavidin to bind to HGN, but limits available sites for polyhistidine tails of H6PAD to ∼7000 per particle. Likewise, the 1:9 ratio reduces the number of streptavidin on the surface, allowing for increased H6PAD association to ∼20 000. (c) HPLC chromatogram of H6PAD retained on HGN not exposed to laser (blue) and after 800 nm laser exposure with pulse fluence of 2.5 mJ cm−2 and 10 s exposure (magenta). ∼15% of H6PAD is observed to be retained on HGN after laser treatment.
The STV-TAT promotes efficient internalization of the nanoparticle construct within 2 h at 37 °C and 5% CO2 into a variety of different cell types, influenced by its high positive charge.21 Darkfield microscopy was used to detect the characteristic scattering of HGN. Visualization of constructs with varying STV-TAT concentrations demonstrated that while a 1:1 ratio showed significant internalization of HGN, a 1:9 ratio also showed adequate internalization in PPC-1 at a 0.8 pM particle concentration (SI Figure 3). To maximize PAD concentration for delivery subsequent experiments were thus conducted with particles presenting a 1:9 complement ratio. To show that H6PAD remained associated with the HGN once internalized by endocytosis, we employed fluorescence microscopy of HGN-MPC after 2 h treatment with cells. Colocalization analysis shows localization in cells of both fluorescein labeled H6PAD and a Cy3 labeled complement of the DNA scaffold of the HGN (SI Figure 4) indicating that the HGN retained PAD after endocytosis. Furthermore, we also observed that delivery into cells was reliant on TAT facilitated endocytosis as omission of the TAT peptide results in the lack of particle internalization. Leveraging the need for internalization peptides on the HGN to achieve cellular localization we demonstrated the modularity of the HGN-MPC and explored the use of targeting peptides as well. Substitution of the TAT peptide for a biotinylated C-end rule peptide, RPARPAR, conferred specific cell targeting where the HGN-MPC preferentially internalize in PPC-1 cells that overexpress the
Figure 2. Temporal controlled cytotoxicity of HGN-MPC. Viability of PPC-1 cells 2 days post-laser treatment with varying concentrations of HGN with H6PAD without laser (blue), HGN-TAT with laser (green), and HGN with H6PAD and exposed to laser (red).
HGN lacking the H6PAD peptide exposed to NIR light shows increased cell death with increasing concentrations of HGN likely due to localized particle heating (green). However, exposure of the HGN-MPC to laser light irradiation demonstrates an augmented effect facilitated by the H6PAD to the particle heating increasing cell death relative to the HGN without the H6PAD when exposed to laser light (Figure 2B, green and red). These types of synergistic effects with other C
DOI: 10.1021/acs.bioconjchem.7b00276 Bioconjugate Chem. XXXX, XXX, XXX−XXX
Communication
Bioconjugate Chemistry
HGN, on the other hand, shows little change in the fluorescence profile of the JC-1 dye 60 min post-irradiation (bottom panels) indicating that release of the mitochondrial targeting peptide is required for depolarization. In conclusion, we have developed a novel approach to deliver toxic peptides into cells with control over peptide release. This technique enables the real-time visualization of peptide activity post-treatment within a reasonable time period. This work extends our initial proof-of-concept to the delivery of biomedically relevant proteins. Further, by showing that small and large proteins can be delivered with the same platform, we highlight the modularity of this approach. Applications involving the assembly of orthogonal cell-targeting moieties as well as different protein cargoes to the same particle are ongoing.
particles have been described previously using pulsed laser irradiation.23,24 Overall, in combination with the nanoparticle delivery platform, a cellular response was observed at an effective concentration of 10−8 M of peptide. Furthermore, to reinforce the effect of peptide release on apoptosis it was imperative to demonstrate alterations in mitochondrial phenotype as a consequence to laser exposure. Using a commercial two-photon confocal microscope equipped with a tunable NIR laser and an x−y scanner allowing for raster scanning of laser light across the field of view we demonstrate that the HGN-MPC are capable of spatially and temporally controlled release of the cargo and allow for real-time visualization of phenotypic changes.4,25 To explore the direct effect of the peptide release in PPC-1 cells, we sought to visualize the membrane destabilization after laser irradiation using the mitochondrial JC-1 stain that aggregates and fluoresces red in polarized mitochondria and fluoresces green in its monomeric form when mitochondria are depolarized. Using the two-photon confocal microscope’s NIR-laser we irradiated select PPC-1 cells treated with the lower concentration of 0.8 pM HGN-MPC and monitored the red and green fluorescence 60 min after laser treatment. Initially the fluorescence profile of the PPC-1 shows the characteristic red fluorescence of the JC-1 aggregates in the mitochondria (Figure 3, top left). After 60 min, however, we see a significant decrease in red fluorescence and subsequent evolution of the green fluorescence in the irradiated cells (top right), indicating that a membrane depolarization event has occurred, preluding apoptosis. Omission of the H6PAD and irradiation of the
■
ASSOCIATED CONTENT
S Supporting Information *
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.bioconjchem.7b00276. Materials and methods; UV−vis and mass spectra; calibration and viability curves; particle imaging (PDF)
■
AUTHOR INFORMATION
Corresponding Author
*E-mail:
[email protected]. ORCID
Norbert O. Reich: 0000-0001-6032-2704 Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS This work was supported by National Institutes of Health (NIH) grant R01 EB012637. Author W. Wonderly would like to acknowledge NSF EAGER grant 1506539. The authors thank H. Waite, A. Pallaoro, and G. Braun, for support and helpful advice. Also, the authors thank A. Mikhailovsky for laser use and B. Lopez for aid with microscopy and acknowledge the support for the Olympus Fluoview 1000 MPE microscope from the NIH (1S10RR022585-01A1).
■
REFERENCES
(1) Brodin, J. D., Sprangers, A. J., McMillan, J. R., and Mirkin, C. A. (2015) DNA-Mediated Cellular Delivery of Functional Enzymes. J. Am. Chem. Soc. 137, 14838−14841. (2) Zuris, J. A., Thompson, D. B., Shu, Y., Guilinger, J. P., Bessen, J. L., Hu, J. H., Maeder, M. L., Joung, J. K., Chen, Z.-Y., and Liu, D. R. (2014) Cationic lipid-mediated delivery of proteins enables efficient protein-based genome editing in vitro and in vivo. Nat. Biotechnol. 33, 73−80. (3) Shamay, Y., Adar, L., Ashkenasy, G., and David, A. (2011) Light induced drug delivery into cancer cells. Biomaterials 32, 1377−1386. (4) Morales, D. P., Braun, G. B., Pallaoro, A., Chen, R., Huang, X., Zasadzinski, J. A., and Reich, N. O. (2015) Targeted intracellular delivery of proteins with spatial and temporal control. Mol. Pharmaceutics 12, 600−9. (5) Lukianova-Hleb, E., Hu, Y., Latterini, L., Tarpani, L., Lee, S., Drezek, R. A., Hafner, J. H., and Lapotko, D. O. (2010) Plasmonic Nanobubbles as Transient Vapor Nanobubbles Generated around Plasmonic Nanoparticles. ACS Nano 4, 2109−2123. (6) Lukianova-Hleb, E. Y., Belyanin, A., Kashinath, S., Wu, X., and Lapotko, D. O. (2012) Plasmonic nanobubble-enhanced endosomal
Figure 3. Real-time monitoring of mitochondrial depolarization postlaser irradiation of HGN-MPC. PPC-1 cells treated with 0.8 pM HGN-MPC and treated with pulsed NIR laser. H6PAD activity visualized after release from HGN-MPC by fluorescence shift of the JC-1 dye. The JC-1 dye aggregates in polarized environments (in mitochondria) and exhibits red fluorescence. In depolarized environments the JC-1 dye is a monomer fluorescing green. HGN-MPC are treated with pulsed laser irradiation and a decrease in JC-1 aggregate is observed after 60 min characterized by decrease in red fluorescence and evolution of the green fluorescence from the JC-1 monomer (top panels). Cells treated with HGN without H6PAD show minimal change in fluorescence post-irradiation after 60 min. Scale bar 50 μm. D
DOI: 10.1021/acs.bioconjchem.7b00276 Bioconjugate Chem. XXXX, XXX, XXX−XXX
Communication
Bioconjugate Chemistry escape processes for selective and guided intracellular delivery of chemotherapy to drug-resistant cancer cells. Biomaterials 33, 1821−6. (7) Mai, J. C., Mi, Z., Kim, S.-H., Ng, B., and Robbins, P. D. (2001) A Proapoptotic Peptide for the Treatment of Solid Tumors. Cancer Res. 61, 7709−7712. (8) Ellerby, H. M., Arap, W., Ellerby, L. M., Kain, R., Andrusiak, R., Rio, G. D., Krajewski, S., Lombardo, C. R., Rao, R., Ruoslahti, E., et al. (1999) Anti-cancer activity of targeted pro-apoptotic peptides. Nat. Med. 5, 1032−1038. (9) Javadpour, M. M., Juban, M. M., Lo, W.-C. J., Bishop, S. M., Alberty, J. B., Cowell, S. M., Becker, C. L., and McLaughlin, M. L. (1996) De Novo Antimicrobial Peptides with Low Mammalian Cell Toxicity. J. Med. Chem. 39, 3107−3113. (10) Barnhart, K. F., Christianson, D. R., Hanley, P. W., Driessen, W. H., Bernacky, B. J., Baze, W. B., Wen, S., Tian, M., Ma, J., Kolonin, M. G., et al. (2011) A peptidomimetic targeting white fat causes weight loss and improved insulin resistance in obese monkeys. Sci. Transl. Med. 3, 108ra112. (11) Futaki, S., Niwa, M., Nakase, I., Tadokoro, A., Zhang, Y., Nagaoka, M., Wakako, N., and Sugiura, Y. (2004) Arginine carrier peptide bearing Ni(II) chelator to promote cellular uptake of histidinetagged proteins. Bioconjugate Chem. 15, 475−81. (12) Agemy, L., Friedmann-Morvinski, D., Kotamraju, V. R., Roth, L., Sugahara, K. N., Girard, O. M., Mattrey, R. F., Verma, I. M., and Ruoslahti, E. (2011) Targeted nanoparticle enhanced proapoptotic peptide as potential therapy for glioblastoma. Proc. Natl. Acad. Sci. U. S. A. 108, 17450−17455. (13) Jain, P. K., Qian, W., and El-Sayed, M. A. (2006) Ultrafast cooling of photoexcited electrons in gold nanoparticle-thiolated DNA conjugates involves the dissociation of the gold-thiol bond. J. Am. Chem. Soc. 128, 2426−33. (14) Huang, X., Hu, Q., Braun, G. B., Pallaoro, A., Morales, D. P., Zasadzinski, J., Clegg, D. O., and Reich, N. O. (2015) Light-activated RNA interference in human embryonic stem cells. Biomaterials 63, 70−79. (15) Huang, X., Pallaoro, A., Braun, G. B., Morales, D. P., Ogunyankin, M. O., Zasadzinski, J., and Reich, N. O. (2014) Modular Plasmonic Nanocarriers for Efficient and Targeted Delivery of CancerTherapeutic siRNA. Nano Lett. 14, 2046−51. (16) Goodman, A. M., Hogan, N. J., Gottheim, S., Li, C., Clare, S. E., and Halas, N. J. (2017) Understanding Resonant Light-Triggered DNA Release from Plasmonic Nanoparticles. ACS Nano 11, 171. (17) Park, S., Brown, K. A., and Hamad-Schifferli, K. (2004) Changes in Oligonucleotide Conformation on Nanoparticle Surfaces by Modification with Mercaptohexanol. Nano Lett. 4, 1925−1929. (18) Tinland, B., Pluen, A., Sturm, J., and Weill, G. (1997) Persistence Length of Single-Stranded DNA. Macromolecules 30, 5763−5765. (19) Lee, H., Venable, R. M., MacKerell, A. D., Jr, and Pastor, R. W. (2008) Molecular Dynamics Studies of Polyethylene Oxide and Polyethylene Glycol: Hydrodynamic Radius and Shape Anisotropy. Biophys. J. 95, 1590−1599. (20) Javadpour, M. M., and Barkley, M. D. (1997) Self-Assembly of Designed Antimicrobial Peptides in Solution and Micelles. Biochemistry 36, 9540−9549. (21) Torchilin, V. P. (2008) Tat peptide-mediated intracellular delivery of pharmaceutical nanocarriers. Adv. Drug Delivery Rev. 60, 548−58. (22) Teesalu, T., Sugahara, K. N., Kotamraju, V. R., and Ruoslahti, E. (2009) C-end rule peptides mediate neuropilin-1-dependent cell, vascular, and tissue penetration. Proc. Natl. Acad. Sci. U. S. A. 106, 16157−16162. (23) Abadeer, N. S., and Murphy, C. J. (2016) Recent Progress in Cancer Thermal Therapy Using Gold Nanoparticles. J. Phys. Chem. C 120, 4691−4716. (24) Shao, J., Griffin, R. J., Galanzha, E. I., Kim, J.-W., Koonce, N., Webber, J., Mustafa, T., Biris, A. S., Nedosekin, D. A., and Zharov, V. P. (2013) Photothermal nanodrugs: potential of TNF-gold nanospheres for cancer theranostics. Sci. Rep. 3, 1293.
(25) Huang, X., Hu, Q., Lai, Y., Morales, D. P., Clegg, D. O., and Reich, N. O. (2016) Light-Patterned RNA Interference of 3DCultured Human Embryonic Stem Cells. Adv. Mater. 28, 10732− 10737.
E
DOI: 10.1021/acs.bioconjchem.7b00276 Bioconjugate Chem. XXXX, XXX, XXX−XXX
