422
Bioconjugate Chem. 2009, 20, 422–426
COMMUNICATIONS Knocking (Anti)-Sense into Cells: The Microsphere Approach to Gene Silencing Lois M. Alexander, Rosario M. Sa´nchez-Martı´n, and Mark Bradley* School of Chemistry, University of Edinburgh, West Mains Road, Edinburgh, EH9 3JJ United Kingdom. Received December 9, 2008; Revised Manuscript Received February 2, 2009
200 and 500 nm polymeric microspheres have been conjugated to siRNA targeted against EGFP expressed in human cervical cancer (HeLa) cells and shown to efficiently silence protein expression over 72 h, without detrimental cytotoxicity. Furthermore, with use of an independent Cy5 tracking label on the siRNA-laden microsphere, silencing of EGFP could be assessed by selecting only those cells that contained the delivery vehicle (and thus the siRNA) generating a more accurate picture of microsphere-induced gene silencing.
Since it was first described in the 1990s (1, 2), interest in the area of RNA interference or the so-called RNAi pathway has grown dramatically. This interest has been heightened by the increased understanding of the role of short interfering RNA (siRNA) (3) and the possible therapeutic applications it may have (4, 5). However, naked, synthetic siRNA is cell-impermeable and requires a delivery system, which includes, for example, viral methods (6, 7) which generate siRNA in vivo, microinjection (8), lipofection (9), and carbon nanotubes (10). Microspheres have been shown to be efficient delivery agents of a range of biological cargos, including sensors, proteins, and reporters (11-14). As well as having high uptake efficiencies in a wide range of cells, they have additionally been shown to be nontoxic (14) and may be readily and easily functionalized. Here, we report the attachment of double-stranded siRNA to 200 and 500 nm polystyrene microspheres via cleavable (disulfide) and noncleavable (amide) linkages and demonstrate the efficient silencing of enhanced green fluorescent protein (EGFP) stably expressed in human cervical cancer (HeLa) cells. Polystyrene microspheres (200 and 500 nm) with divinylbenzene cross-linking were prepared by dispersion or emulsion polymerization and PEGylated as previously reported (14). Initially, in order to quantify the uptake of these microspheres in HeLa cells, fluoresceinamine was coupled to PEGylated microspheres via a noncleavable (amide) linker (Scheme 1, 2a-b) using standard solid-phase techniques (14). Analysis of cellular uptake was made by flow cytometry following incubation with the microspheres for 6-72 h and was observed to be over 70% after 24 h and over 80% after 72 h (see Supporting Information, Graph S-1). All flow cytometric analysis was performed in 0.2% trypan blue, which quenches any extracellular fluorescence associated with microspheres on the outside of the cell, ensuring that only fluorescence originating from microspheres inside the cell is measured (15, 16). Uptake was additionally confirmed by confocal microscopy, showing fluo* Corresponding author. Professor Mark Bradley, School of Chemistry, University of Edinburgh, West Mains Road, Edinburgh, EH9 3JJ (United Kingdom), E-mail:
[email protected], Tel: +44 (0) 131 651 3307, Fax: +44 (0) 131 650 6453.
Figure 1. Pseudoconfocal microscopy of HeLa-EGFP cells (green) incubated with (a) 200 nm TAMRA siRNA microspheres (red); (b) 500 nm TAMRA siRNA microspheres (red). Images taken after 24 h (scale bar is 200 µm). Inset images are a 4 × magnification, where nuclei are stained with Hoechst 33342.
rescein microspheres localized within the cytoplasmic regions of the cell (see Supporting Information, Figure S-3). In order to achieve an intracellularly cleavable functionality, PEGylated microspheres (1a-b) were functionalized with a disulfide linker, yielding 3a-b. To evaluate the cleavability of the disulfide linkage, fluoresceinamine was coupled, yielding disulfide-FAM microspheres (Scheme 1, 4a-b) and the constructs were treated with glutathione (10 mM) to facilitate release of thiolated fluoresceinamine, which was assessed by flow cytometry. Reduction of the disulfide linkage was evident after just 1 h and maximal after 24 h (Supporting Information, Graph S-2). 4a and 4b were subsequently incubated with HeLa cells and release of the fluorophore into the intracellular environment examined and was found to be prevalent after 12 h when microspheres as fluorescently localized points within cells were no longer extensively visible (see Supporting Information, Figure S-2a,b). In contrast, microspheres exhibiting fluoresceinamine via a noncleavable (amide) linkage retained their fluorescent label regardless of glutathione treatment or the length of incubation time within HeLa cells (see Supporting Information, Graph S-2 and Figure S-3). Subsequently, the efficiency of siRNA loading onto microspheres was studied using siRNA with a 5′-amino functionality on the sense strand for coupling to the microspheres (Scheme 1) and a 5′ fluorescent label (TAMRA) on the antisense strand allowing quantification of coupling and hybridization by spec-
10.1021/bc800529r CCC: $40.75 2009 American Chemical Society Published on Web 02/26/2009
Communications
Bioconjugate Chem., Vol. 20, No. 3, 2009 423
Scheme 1. Preparation of siRNA and Fluoresceinamine Loaded 200 and 500 nm Microspheres with Noncleavable (Amide) and Cleavable (Disulfide) Linkagesa
a
For siRNA sequences, see Supporting Information.
Figure 2. EGFP fluorescence expression in HeLa cells (a) by flow cytometry after 24, 48 and 72 h incubations with siRNA microspheres (200 and 500 nm, amide and disulphide, 5a–d) and siRNA transfected via Lipofectamine 2000. siRNA concentration ) 28 nM in all incubations. Negative GFP ) HeLa cells untreated and not transfected with EGFP; Positive GFP ) HeLa cells untreated and stably transfected with EGFP. Right: by microscopy after 72 h (b) Untreated cells; (c) with 200 nm amide-siRNA microspheres (5a); (d) with 200 nm disulphide-siRNA microspheres (5c); (e) with 500 nm amide-siRNA microspheres (5b); (f) with 500 nm disulphide-siRNA microspheres (5d). Scale bar is 350 µm. Inset are phase contrast images, showing approximately 70-80% cell confluency.
trofluorimetric analysis (see Supporting Information). The quantity of siRNA present on the microspheres was found to
be ca. 13 000 molecules (2.2 × 10-20 mol) of siRNA per 500 nm microsphere and ca. 780 chains (1.3 × 10-21 mol) of siRNA
424 Bioconjugate Chem., Vol. 20, No. 3, 2009
Communications
Scheme 2. Preparation of Dual-Labeled 500 nm Cy5-co-siRNA Microspheres with Disulfide and Amide Linkages (Fmoc-PEG-OH is Fmoc-NH-CH2-(CH2-O-CH2)3-CH2-NH-C(O)-CH2-CH2-COOH)
per 200 nm microsphere. As such, with a microsphere concentration of 86 µg mL-1 (which has been demonstrated to afford excellent and uniform cellular uptake), this equates to a siRNA
concentration of 28 nM in cell cultures (assuming all siRNA on the microspheres is cleaved in solution; see Supporting Information, Graphs S4-6). Examining this on a single bead
Figure 3. (a) EGFP Intensity in HeLa-EGFP cells after 24, 48, and 72 h incubation with Cy5-co-siRNA microspheres (9a-b). Negative GFP ) HeLa cells untreated with beads and not transfected with EGFP; positive GFP ) HeLa cells untreated with beads and stably transfected with EGFP. Inset: Dot plot of EGFP (FITC-A) vs Cy5 (APC-A) showing beadfected cells (Q1 and Q2) have a substantial reduction in EGFP intensity. Right: Confocal microscopy of HeLa-EGFP cells (green) treated with (b) amide-siRNA 500 nm Cy5 microspheres (9b) (scale bar is 70 µm); (c) DisulfidesiRNA 500 nm Cy5 microspheres (9a) (scale bar is 100 µm). Microspheres (blue) are indicated with white arrows. Images were collected using a Leica DM IRE2 inverted confocal microscope.
Communications
per cell basis, we equate this to an intracellular siRNA concentration of 0.3 nM (200 nm microspheres) to 4 nM (500 nm microspheres), assuming that all siRNA is disengaged from the delivery device and cell volume average of 5000 µ3 (17). However, it must be noted that, in general, cells take up more 200 nm microspheres than 500 nm microspheres. Accordingly, in order to qualify siRNA-laden microsphere uptake, HeLaEGFP cells were incubated with microspheres loaded with TAMRA labeled siRNA (5a-b) and assessed for uptake via pseudoconfocal microscopy (Figure 1) (a control was established using TAMRA-siRNA not conjugated to microspheres to measure background staining). Uptake was evident after 24 h, and the siRNA label could be seen localized on the polymer microspheres within cells. Previous studies have indicated that these microspheres enter cells by a mechanism that does not result in their endosomal capture (11, 13), avoiding the need to apply endosomal disrupting agents to achieve cytoplasmic localization. While siRNA functionalized with a fluorophore can achieve efficient gene knockdown (18), labeling of the antisense strand can result in poor silencing (see Supporting Information, Graph S-7) (19). As a consequence, 2a-b and 3a-b were conjugated to unlabeled siRNA, and HeLa-EGFP cells were subsequently treated with these microspheres (86 µg mL-1, 28 nM siRNA) for 24, 48, and 72 h, before analysis of EGFP expression by microscopy and flow cytometry. After 48 h, excellent reductions in EGFP intensities were observed in cells incubated with all siRNA bead types, which was even more prominent after 72 h (Figure 2), with up to a 90% reduction in EGFP intensity. Positive controls were established using a commercially available transfection agent, Lipofectamine 2000, which achieved a 70% reduction in EGFP after 72 h using the same concentration of siRNA as the microspheres (28 nM, Figure 2). Negative controls included lipofection and beadfection of scrambled siRNA, siRNA without a carrier system, and microspheres without siRNA (Supporting Information, Graph S-8). In addition, the concentration of siRNA-microspheres incubated with HeLa-EGFP cells was varied and the knockdown efficiency was found to be dose- as well as time-dependent (see Supporting Information, Graph S-9). Furthermore, siRNA microspheres were shown to be nontoxic by MTT assay, even after 72 h, at all concentrations analyzed (see Supporting Information, Graph S-3). Although EGFP silencing was clear, it was considered that a more accurate assessment of the knockdown would be to obtain a system whereby the beadfected cells could be assessed independently from non-beadfected cells. To this affect, a dual-functionalized system was developed. Thus, orthogonally protected lysine (20) was employed to allow functionalization of the microspheres with both a fluorophore (Cy5) and siRNA (Scheme 2). As such, HeLa-EGFP cells were incubated with Cy5-co-siRNA 500 nm microspheres (9a-b, 86 µg mL-1, 28 nM siRNA), and EGFP silencing was analyzed after 24, 48, and 72 h by flow cytometry selecting only those cells that were positive for Cy5 (beadfected cells, thus containing siRNA) (Figure 3a). Silencing of EGFP was evident after 24 h and extensive at 48 and 72 h using both cleavable (disulfide) and noncleavable (amide) linkages to the microspheres. Reduction in EGFP expression was confirmed by confocal microscopy (Figure 3b,c), whereby cells containing microspheres (indicated by white arrows) were negative for EGFP expression. Using this method, the gene silencing capabilities of microspheres could be more accurately depicted, since only those cells known to contain the delivery device were considered in analysis of the EGFP expression.
Bioconjugate Chem., Vol. 20, No. 3, 2009 425
In conclusion, enhanced green fluorescent protein stably expressed in human cervical cancer (HeLa) cells was successfully silenced (by approximately 90%) using siRNA linked to 500 and 200 nm microspheres via cleavable (disulfide) and noncleavable (amide) linkers, exceeding the silencing capabilities of commercially available lipofection products. In addition, a system employing dual-functionalized (Cy5-co-siRNA) microspheres allowed independent evaluation of only those cells that had been beadfected with siRNA, yielding gene silencing data based only on those cellular populations that had received the delivery vehicle, generating a more accurate analysis of the gene silencing capabilities of microspheres. This approach additionally allowed HeLaEGFP cells beadfected with siRNA to be tracked over time, both visually by microscopy and by fluorescence-based flow cytometric methods. Importantly, the nontoxic nature of the microspheres along with their facile functionalization and controllability over cellular loading makes these constructs particularly suitable to this application in a day and age when efficient, nontoxic, and versatile delivery devices are strongly desired.
ACKNOWLEDGMENT The authors would like to thank the EPSRC and the BBSRC for funding. R.M. Sa´nchez-Martı´n would like to thank the Royal Society for a Dorothy Hogdkin Fellowship. The authors are grateful to M. Lopalco for the Cy5 dye. Supporting Information Available: Microsphere preparation, functionalization, and cellular uptake (including siRNA sequences); disulfide cleavage; MTT toxicity assays; controls in gene silencing; and dose-response of siRNA microspheres. This material is available free of charge via the Internet at http:// pubs.acs.org.
LITERATURE CITED (1) Fire, A., Xu, S., Montgomery, M. K., Kostas, S. A., Driver, S. E., and Mello, C. C. (1998) Potent and specific genetic interference by double-stranded RNA in Caenorhabditis elegans. Nature 391, 806–810. (2) Hamilton, A. J., and Baulcombe, D. C. (1999) A species of small antisense RNA in post-transcriptional gene silencing in plants. Science 286, 950–952. (3) Elbashir, S. M., Harborth, J., Lendeckel, W., Yalcin, A., Weber, K., and Tuschl, T. (2001) Duplexes of 21-nucleotide RNAs mediate RNA interference in cultured mammalian cells. Nature 411, 494–497. (4) Dykxhoorn, D. M., Novina, C. D., and Sharp, P. A. (2003) Killing the messenger: Short RNAs that silence gene expression. Nat. ReV. Mol. Cell Biol. 4, 457–467. (5) Song, E. W., Zhu, P. C., Lee, S. K., Chowdhury, D., Kussman, S., Dykxhoorn, D. M., Feng, Y., Palliser, D., Weiner, D. B., Shankar, P., Marasco, W. A., and Lieberman, J. (2005) Antibody mediated in ViVo delivery of small interfering RNAs via cellsurface receptors. Nat. Biotechnol. 23, 709–717. (6) Brummelkamp, T. R., Bernards, R., and Agami, R. (2002) Stable suppression of tumorigenicity by virus mediated RNA interference. Cancer Cell 2, 243–248. (7) Xia, H., Mao, Q., Paulson, H. L., and Davidson, B. L. (2002) siRNA-mediated gene silencing in vitro and in vivo. Nat. Biotechnol. 20, 1006–1010. (8) Kennerdell, J. R., and Carthew, R. W. (1998) Use of dsRNAmediated genetic interference to demonstrate that frizzled and frizzled 2 act in the wingless pathway. Cell 95, 1017–1026. (9) Arnold, A. S., Tang, Y. L., Qian, K., Shen, L., Valencia, V., Phillips, M. I., and Zhang, Y. C. (2007) Specific β1-adrenergic receptor silencing with small interfering RNA lowers high blood pressure and improves cardiac function in myocardial ischemia.
426 Bioconjugate Chem., Vol. 20, No. 3, 2009 J. Hypertens. 25, 197–205. (10) Liu, Z., Winters, M., Holodniy, M., and Dai, H. (2007) siRNA Delivery into Human T cells and primary cells with carbon-nanotube transporters. Angew. Chem., Int. Ed. 46, 2023– 2027. (11) Sa´nchez-Martı´n, R. M., Cuttle, M., Mittoo, S., and Bradley, M. (2006) Microsphere-based real-time calcium sensing. Angew. Chem., Int. Ed. 45, 5472–5474. (12) Bradley, M., Alexander, L., and Sa´nchez-Martı´n, R. M. (2008) Cellular uptake of fluorescent labelled biotin-streptavidin microspheres. J. Fluoresc. 18, 733–739. (13) Bradley, M., Alexander, L., Duncan, K., Chennaoui, M., Jones, A. C., and Sa´nchez-Martı´n, R. M. (2008) pH sensing in living cells using fluorescent microspheres. Bioorg. Med. Chem. Lett. 18, 313–317. (14) Sa´nchez-Martı´n, R. M., Muzerelle, M., Chitkul, N., How, S. E., Mittoo, S., and Bradley, M. (2005) Bead-based cellular analysis, sorting and multiplexing. ChemBioChem 6, 1341– 1345. (15) Mosiman, V. L., Patterson, B. K., Canterero, L., and Goolsby, C. L. (1997) Reducing cellular autofluorescence in flow cytometry: An in situ method. Cytometry Part B: Clinical Cytometry
Communications 30, 151–156. (16) Rejman, J., Oberle, V., Zuhorn, I. S., and Hoesktra, D. (2004) Size-dependent internalization of particles via the pathways of clathrin- and caveolae-mediated endocytosis. Biochem. J. 377, 159–170. (17) Korchev, Y. E., Gorelik, J., Lab, M. J., Sviderskaya, E. V., Johnston, C. L., Coombes, C. R., Vodyanoy, I., and Edwards, C. R. W. (2000) Cell volume measurement using scanning ion conductance microscopy. Biophys. J. 78, 451–457. (18) Berezhna, S. Y., Supekova, L., Supek, F., Schultz, P. G., and Deniz, A. A. (2006) siRNA in human cells selectively localizes to target RNA sites. Proc. Natl. Acad. Sci. U.S.A. 103, 7682–7687. (19) Ja¨rve, A., Mu¨ller, J., Kim, I., Rohr, K., MacLean, C., Fricker, G., Massing, U., Eberle, F., Dalpke, A., Fischer, R., Trendelenburg, M. F., and Helm, M. (2007) Surveillance of siRNA integrity by FRET imaging. Nucleic Acids Res. , e124. (20) Dı´az-Mocho´n, J. J., Bialy, L., and Bradley, M. (2004) Full orthogonality between Dde and Fmoc: The direct synthesis of PNA-peptide conjugates. Org. Lett. 6, 1127–1129. BC800529R
