Evaluation of carrier-mediated siRNA delivery

0 downloads 0 Views 579KB Size Report
Evaluation of carrier-mediated siRNA delivery: Lessons for the design of a stem-loop. qPCR-based approach for quantification of intracellular full-length siRNA.
GENE DELIVERY

Journal of Controlled Release 166 (2013) 220–226

Contents lists available at SciVerse ScienceDirect

Journal of Controlled Release journal homepage: www.elsevier.com/locate/jconrel

Evaluation of carrier-mediated siRNA delivery: Lessons for the design of a stem-loop qPCR-based approach for quantification of intracellular full-length siRNA Stefano Colombo, Hanne Mørck Nielsen, Camilla Foged ⁎ Department of Pharmacy, Faculty of Health and Medical Sciences, University of Copenhagen, Universitetsparken 2, DK-2100 Copenhagen O, Denmark

a r t i c l e

i n f o

Article history: Received 27 September 2012 Accepted 4 January 2013 Available online 11 January 2013 Keywords: siRNA qPCR Stem-loop qPCR Nanomedicine Drug delivery

a b s t r a c t Harnessing the RNA interference (RNAi) process with chemically synthesized small interfering RNA (siRNA) is dependent on the development of efficient delivery vehicles that can help overcome the numerous barriers existing for siRNA delivery. However, quantifying the intracellular amount of siRNA delivered by use of carriers remains an analytical challenge. The purpose of the present study was to optimize and validate an analytical protocol based on stem-loop reverse transcription quantitative polymerase chain reaction (RT qPCR) to quantitatively monitor the carrier-mediated intracellular siRNA delivery. An in vitro cell culture model system expressing enhanced green fluorescent protein (EGFP) was used to develop the assay, which was based on the intracellular quantification of a full-length double-stranded Dicer substrate siRNA by stem-loop RT qPCR. The result is a well-documented protocol for accurate and sensitive determination of the effective intracellular siRNA concentration upon transfection with different reagents. Specific guidelines for the customization of the protocol are provided and reported together with an example of its application for studying a specific siRNA delivery case. The outcome of the present study is a thoroughly discussed analytical protocol generally applicable to characterize carrier-mediated siRNA delivery processes. © 2013 Elsevier B.V. All rights reserved.

1. Introduction Recent results in the small interfering RNA (siRNA) delivery research field towards developing RNA interference (RNAi) therapeutics highlight the importance of adequate siRNA dosing and improving the understanding of the delivery dynamics mediated by different types of carrier systems [1,2]. However, despite the vast number of nanocarrier systems tested for siRNA delivery to date, little attention has been drawn to quantifying the ability of the nanocarriers to deliver full length siRNA to the intracellular compartment and make it available for functional RNAi. Such quantification is of ultimate importance for both in vitro and in vivo purposes, where an accurate siRNA dosing is crucial for achieving an optimal therapeutic effect and for avoiding non-specific side effects [3–5]. An analytical method enabling the quantitative intracellular siRNA determination is therefore highly in demand to reach a better understanding of the carrier's mechanism of action. Several analytical techniques based on different approaches are available to measure the cellular content of small RNA [6–11]. However, a validated protocol to quantify the intracellular concentration of full length siRNA delivered via nanocarriers remains to be developed, but is nevertheless mandatory for the future rational design of innovative and efficient siRNA nanocarriers. Among the most recent applications of the polymerase chain reaction (PCR), the stem-loop reverse transcription ⁎ Corresponding author. Tel.: +45 35 33 64 02; fax: +45 35 33 60 01. E-mail address: [email protected] (C. Foged). 0168-3659/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.jconrel.2013.01.006

(RT) quantitative (q)PCR has drawn much attention because of its simplicity and convenience [12,13]. The application of the stem-loop primer technology to the siRNA delivery research field has been discussed previously in the literature [14], and, more recently, an experimental protocol has been proposed for micro RNAs [15]. Despite the information that can be found in the literature regarding quantification techniques for the small RNAs, the application of these assays for determining the concentration of double-stranded Dicer substrate siRNA [16] is poorly discussed and documented. This study reports an analytical method, which relies on the intracellular full-length siRNA quantification of double-stranded Dicer substrate siRNA by stem-loop RT qPCR. It is fully based on commercially available components, which are well-documented in the scientific literature to assure the reproducibility of the data and allow for independent customization and development. The result is a detailed description of a scientifically highly relevant analytical assay, which is critically discussed and assessed in a stepwise manner, corroborated by relevant experimental observations and useful guidelines for specific customization. 2. Materials and methods 2.1. Materials 2′-O-Methyl modified Dicer substrate asymmetric siRNA duplexes directed against enhanced green fluorescent protein (EGFP) and firefly

luciferase (FLuc, used as negative control sequence) were provided by Integrated DNA Technologies (IDT, Coralville, IA, USA) as dried, purified and desalted duplexes, and re-annealed as recommended by the supplier in the IDT duplex buffer consisting of 30 mM 4-(2-hydroxyethyl)piperazine-1-ethanesulfonic acid (HEPES) and 100 mM potassium acetate, pH 7.5. The siRNAs had the following sequences and modification patterns: EGFP sense 5′-pACCCUGAAGUUCAUCUGCACCACcg-3′, antisense 5′-CGGUGGUGCAGAUGAACUUCAGGGUCA-3′ (the siRNA duplex is referred to as EGFPS1 R25D/27 below), and FLuc sense 5′-pGGUUC CUGGAACAAUUGCUUUUAca-3′ and antisense 5′-UGUAAAAGCAAUUGU UCCA-GGAACCAG-3′, where lowercase letters represent deoxyribonucleotides, underlined capital letters represent 2′-O-methylribonucleotides and p represents phosphate residues [16]. RNase-free diethyl pyro carbonate (DEPC)-treated Milli-Q water was used for all buffers and dilutions. The concentration of the re-annealed siRNA was determined using the Quant-iT™ RiboGreen® RNA Reagent (Molecular Probes, Invitrogen, Paisley, UK). Additional chemicals were obtained commercially at analytical grade. 2.2. Cell culture The human non-small cell lung carcinoma cell line H1299 stably expressing EGFP (EGFP-H1299) was used [17]. The cells were maintained in RPMI 1640 medium (Fisher Scientific, Waltham, MA, USA) supplemented with 100 U/ml penicillin, 100 μg/ml streptomycin, 2 mM L-glutamine (all from Sigma-Aldrich, St. Louis, MO, USA), 0.2 mg/ml Geneticin (Invitrogen, Carlsbad, CA, USA) and 10% (v/v) fetal bovine serum (FBS, PAA Laboratories, Pasching, Austria). The cells were grown in an atmosphere of 5% CO2/95% O2 at 37 °C changing the growth medium three times a week and sub-cultured approximately 1:5 twice a week using TrypLE™ Express (Invitrogen) for 5 min at 37 °C to detach the cells. 2.3. Transfections EGFP-H1299 cells were seeded in standard tissue culture 24-well plates (Corning, Corning, NY, USA) at a density of 8 × 10 4 cells/well. After 24 h (80% confluency), the cells were transfected and incubated with 300 μl of pre-mixed samples in RMPI 1640 medium containing 10% FBS (v/v) for 48 h. For the complex formation between transfection reagent and siRNA, 1 μl of Lipofectamine 2000 (LF, Invitrogen) or TransIT-TKO (TKO, Mirus Bio, Madison, WI, USA) was added to 49 μl siRNA solution (0.012–1.5 pg siRNA, corresponding to concentrations of 0.24–30 nM) in RMPI 1640, and 250 μl RMPI 1640 was added after complex formation. The final siRNA concentrations in the transfection medium were 0.04–5 nM. 2.4. Flow cytometry The cells were harvested using TrypLE™ Express. The EGFP silencing was evaluated by flow cytometry using a Gallios flow cytometer (Beckman Coulter, Fullerton, CA, USA). The cell viability was estimated by propidium iodide (PI, Invitrogen) staining applying 1 μg/ml PI. Flow cytometry data was analyzed using the Cyflogic program (CyFlo, Turku, Finland).

221

RNA Stat-60 reagent was precipitated overnight at 4 °C from the water phase in one volume of isopropanol (Sigma) and collected by centrifugation for 20 min at 12,000 ×g at 4 °C. The RNA quality and concentration were verified by absorbance measurements using the Nanodrop 2000c Spectrophotometer (Thermo Scientific, Wilmington, USA). 2.6. Reverse transcription An RNA amount of 700 ng was reverse transcribed in a total reaction volume of 20 μl (35 ng/μl) that also included 0.5 mM deoxynucleotide mix, 1× Transcriptor Reverse Transcriptase buffer, 10 U Transcriptor Reverse Transcriptase and 20 U Protector RNase Inhibitor (all from Roche, Basel, Switzerland). The RNA template was denatured for 10 min at 70 °C and immediately cooled on ice for 2 min. The reaction mixture and the stem-loop primers were then added during the cooling phase (10 nM of each, final concentration). The pulsed RT program consisted of 15 min at 14 °C, 10 min at 42 °C, 25 cycles (15 s at 14 °C, 10 s at 42 °C and 15 s at 65 °C), 5 min transcriptase inactivation at 85 °C, followed by cooling at 4 °C. The stem-loop reverse transcription primers used were (the specific 3′ sequences are in bold): EGFPS1 R25/D27 antisense strand (AS): 5′-CCATCATGCTCTCGACCTGTCCGAT CACGACGAGAGCATGATGGTGACCC-3′; snoRNA U109: 5′-CCATCATGC TCTCGACCTGTCCGATCACGACGAGAGCATGATGGTTATGT-3′. 2.7. qPCR The qPCR was performed with a LightCycler® 480 (Roche) using Sybr Green® Master mix (Roche). The primer sets (TAGC, Copenhagen, Denmark) were designed and analyzed using the OligoAnalyzer 3.1 (IDT) and the UNAFold program [19]. The housekeeping gene snoRNA U109 (Genbank ID: AM055742.1) was used for normalization and was selected from a set of nine small RNAs (Supplementary Material, Table 1) using the NormFinder algorithm (Aarhus University Hospital, Aarhus, Denmark). The applied PCR primers had the following sequences; Forward EGFPS1 R25/D27 AS: 5′-CGGTGGTGCAGATGAACTTCAGG-3′; Forward snoRNA U109 5′-CCAACCTTCTAGTAAAGGTTGAGTGGT-3′; Reverse universal: 5′-GACCTGTCCGATCACGACGAG-3′. The PCR program was 95 °C for 5 min, 37 cycles (95 °C for 15 s, 62 °C for 15 s, 72 °C for 1 s), followed by cooling at 4 °C. The PCR data analyses were performed using the qpcR package in R (R Foundation for Statistical Computing, Vienna, Austria), and the Cp values were determined by fitting and calculation of the maximum of the second derivative [20,21]. The ΔΔCP method was used to quantify the siRNA [22]. 2.8. Statistics The experiments were performed in triplicate, unless otherwise stated. Values are given as means ± standard deviations (SD). For the statistics and plotting, PRISM (GraphPad, La Jolla, CA, USA) was used. Statistically significant differences were assessed by an analysis of variance (ANOVA) at a 0.05 significance level and followed by Tukey's post test. 3. Results and discussion

2.5. RNA isolation and purification The cells were washed three times with 1 ml phosphate-buffered saline (PBS, Sigma-Aldrich) and trypsinized with TrypLE™ Express according to the manufacturer's instructions. Subsequently, the detached cells were pelleted by centrifugation for 5 min at 1100 ×g and washed with 1 ml PBS, and the operation was repeated two times [18]. Total RNA was isolated and purified using the RNA Stat-60 (AMS Biotechnology, Abingdon, UK) or the Norgen total RNA Purification Kit (Norgen, Thorold, ON, Canada). The RNA isolated with the

Evaluating the carrier-mediated siRNA delivery process through quantification of intracellular full length siRNA remains a challenge, especially due to the present poor understanding of the factors that influence the analytical result. The stem-loop RT qPCR technique is one of the most commonly used techniques to quantify small RNAs due to its design simplicity and high sensitivity. In the current study, we designed and tested an assay based on stem-loop RT qPCR adapted to double-stranded siRNA quantification and identified several pitfalls in the use of this approach for evaluating carrier-mediated siRNA delivery

GENE DELIVERY

S. Colombo et al. / Journal of Controlled Release 166 (2013) 220–226

GENE DELIVERY

222

S. Colombo et al. / Journal of Controlled Release 166 (2013) 220–226

(Table 1). Based on thorough investigations, an optimized analytical protocol was developed and is critically discussed. 3.1. Rationale for choice of model system The model system was designed based on the following criteria: i) The opportunity to cross-refer the obtained results via the use of different techniques and ii) the application of a technique and materials well-documented in the literature. The human non-small cell lung carcinoma cell line H1299 stably expressing EGFP was chosen as the model cell line to enable verification and corroboration of the results obtained by qPCR at the mRNA level with data generated by flow cytometry from analysis of EGFP expression at the protein level. A Dicer substrate asymmetric duplex siRNA directed against EGFP (EGFPS R25D/27) was chosen for the analyses (Fig. 1A), because it possesses several functional features that make it particularly useful as a model sequence for the stem-loop RT qPCR-based siRNA delivery analysis. It is highly potent [23] and has a single Dicer cleavage site that has been determined experimentally by electrospray ionization mass spectrometry (ESI-MS) [16]. The ESI-MS analysis furthermore showed that no enzymatic cleavage by Dicer occurs at the 3′ end of the AS (Fig. 1A). This allows for the design of a specific stem-loop primer for the AS of the siRNA duplex, which can be used for the RT of both the diced and the non-diced sequences with no distinctions (Fig. 1B and C). We suggest in general avoiding the use of siRNAs with multiple Dicer cleavage sites, since a differential quantification among the Dicer variants could compromise the overall assay validity. 3.2. Primer set design The design of stem-loop folding primers was initially described by Chen et al. [18], and only a few modifications of the original technique have subsequently been reported [24,25]. These are included in the studies by Varkonyi-Gasic et al. [26,27], who investigated the stem-loop primer RT process and identified the main drawbacks for the use of stem-loop folded primers. According to their results, a recurrent problem using stem-loop qPCR is the appearance of a background signal produced as a result of dimer formation due to self-interactions during the PCR. The stem-loop RT of the EGFPS1 R25D/27 AS applied in the present study is particularly challenging since the 3′ end is C/G rich and contains a G triplet. For this reason, a stem-loop primer characterized by a C triplet at the specific 3′ end was required. This particular feature makes the sequence highly prone to self-interactions, during both the RT and the PCR, and an unspecific signal was detected by qPCR due to accumulation of Table 1 Critical steps and challenges. Step

Challenges

Actions

Design of the model system Design of the primer set

Reproducibility of the results Improved specificity

Use of components well-documented in the literature Sequence modification to decrease the extendable dimer formation and misfolding In silico study of self-dimers, hetero-dimers, and secondary structure formation In silico target prediction Experimental validation by melting curve analysis and negative controls PCR and RT optimization and comparative efficiency study

Enhanced sensitivity

Reduction of background signal PCR and RT optimization RNA sample preparation

Data analysis

Enhanced assay sensitivity Assay reproducibility Optimal yield and RNA quality Avoid false positives/ negatives Generation of high-quality data

Optimization of cell washing procedure Study RNA isolation procedure Assessment of RNA quality Selection of normalizers and negative controls

concatenated primers (results not shown). This phenomenon was independent of the presence of the RNA template and depended only on the interactions of such a featured stem-loop primer during the RT reaction. The stem-loop primer sequence designed by Chen et al. [18] was modified at specific sites to obtain a PCR optimized for the detection with Sybr Green® qPCR (Fig. 1D). The RNA sequence was heavily modified to reduce self-interactions, but the overall structural properties of the stem-loop primer remained unchanged. The folding of the primer and its thermodynamic stability were analyzed by using UNAFold [19] (Fig. 2). The features conserved during the structure optimization were: i) the Δ free energy (ΔG), the change in enthalpy (ΔH) and the melting temperature of the prevalent folding, and ii) the folding element features (G/C pairs in the beginning and the end of the stem structure, a similar content of G/C pairs in the stem region and the loop size) (Fig. 2). The sequence modifications allowed for i) avoiding inactive misfolded structures and ii) the formation of extendable primer dimers with ΔG b − 5 kcal × mol − 1 during the PCR (Supplementary Material, Fig. 1). The primer set specificity was predicted by verifying the absence of secondary amplification targets in humans for the qPCR primer set in silico by BLAST analysis (http://blast.ncbi.nlm.nih.gov/). In addition, the miRBase (http://www.mirbase.org/) database was used to search for small RNA sequences prone to be reverse transcribed by the stem-loop primers [12] and successively amplified by qPCR. The PCR reaction was studied and optimized according to Pfaffl et al. [28]. The efficiency of the designed PCR scheme was verified using serial dilutions of standard amounts of EGFPS1 R25D/27 spiked to a pool of RNAs isolated from untreated EGFP-H1299 cells (0.028–17.5 ng total RNA). There was no amplification of unwanted products, which was confirmed by melting curve analysis (results not shown). Even though a background signal was detected, the siRNA quantification was not impaired by its presence, as reported previously [14]. In this specific case, the background signal was observed at cycle 33. However it is generally recommended to identify such a background signal by including in each experiment “no RT template”, “no primer” and “no transcriptase” negative controls [27]. 3.3. The efficiency of the RT reaction is the analytical bottleneck The efficiency of the RT reaction appeared to be the major analytical bottleneck for the development of the assay, because even small variations in the efficiency of the RT reaction appeared to considerably impair the assay sensitivity. The RT efficiency was evaluated by spiking EGFPS1 R25D/27 in the RT mixture to a final concentration of 1.88 −11–7.52 −16 M, and the reaction was stabilized by the addition of RNA isolated from EGFP-H1299 cells to a final concentration of 35 ng/μl (Supplementary Material, Table 2). The RT efficiency was significantly influenced by parameters such as the specific RT protocol, the primer sequence and concentration, as well as the amount of RNA template; this effect was most pronounced at lower target RNA concentrations (Supplementary Material, Fig. 2). An optimized pulsed RT protocol was designed by modifying the protocol developed by Tang et al. [29] to keep the RT efficiency constant at a wide range of template concentrations. Thus, the siRNA duplex was denatured at every cycle to increase the melting rate between the primers and the template siRNA strand (Fig. 1E). The use of the pulsed RT protocol resulted in a constant RT efficiency (Fig. 3) throughout the entire experimental range (slope = 3.25, R = 0.97) and an acceptable batch-to-batch and inter-experimental variation (each independent experiment showed a slope of 3.3± 0.3, R > 0.94, Supplementary Material, Table 2). To explain these results, it was hypothesized that the AS of the double-stranded siRNA preferentially melts with the sense strand rather than the other sequences, such as the stem-and-loop primers. This suggests that the melting of the primer–target duplex is influenced by the stem-loop primer concentration in the reaction, and that the reaction efficiency is determined by the relative concentration of primer/

223

template. The function of the pulsed protocol is therefore to minimize the impact of this bias through gradual depletion of the template. The optimal stem-loop primer concentration was 10–15 nM (result not shown). In most cases, it was possible to mix stem-loop primers specific for different targets in the same RT reaction (maintaining the total primer concentration below 30 nM) without a significant increase in the background signal (results not shown). However, a condition for using a primer mix is to avoid applying primers with 3′ ends prone to extensive dimer formation. 3.4. Quantifying intracellular siRNA: RNA sample preparation

Fig. 1. EGFP siRNA sequence, Dicer cleavage site, primers and RT protocol used for quantification. The following aspects were considered in the design of the protocol: The Dicer cleavage site of the target sequence and the optimization of the RT program to achieve constant amplification efficiency. A: The siRNA sequence and its Dicer cleavage site (red line). B: The stem-loop primer, which is specific for the 3′ end of the AS (red) as it is not cleaved by Dicer. C: The difference between the diced and the non-diced sequence of the cDNA product was 6 base pairs in the 3′ end (blue). D: The forward PCR primer (light blue, black arrow) was designed to melt specifically the conserved part of the siRNA AS sequence, and the universal reverse primer was designed on the loop region to avoid formation of primer dimers (red arrow). E: The applied RT program.

Fig. 2. Stem-loop RT primer design. (A) The sequence by Chen et al. [15] was modified to minimize the primer–primer interactions during the EGFPS1 R25/D27 AS quantification (B). The * indicates the mutation sites, while a double * indicates a transversion. The specific sequence is in bold. The folding into the stem-loop was modeled by using UNAFold, red circles indicate the G/C pairs, while blue circles indicate the A/T pairs.

The sample preparation technique should be carefully considered when planning an in vitro siRNA delivery experiment. Different technical challenges must be addressed to avoid false positives as well as underestimation of the results, depending on the physicochemical properties of the delivery system. Therefore, some guidelines are provided in this section to ensure the protocol validity under different experimental conditions. The choice of cell washing procedure is critical for the outcome of an siRNA quantification experiment. The purpose of the cell washing is to remove the siRNA molecules that are not present in the intracellular compartment, but associated with other components of the system. We observed that both the physicochemical properties of the transfection system and the experimental conditions influenced the impact of the cell washing on the final results. To obtain more reproducible results, it has been proposed to remove non-internalized carrier by washing the cells with PBS before and after trypsinization [14]. Comparing the results obtained by washing LF or TKO transfected cells with PBS or trypsinization plus PBS wash, it was observed that the results varied significantly (results not shown). This suggests that the physicochemical properties of the siRNA-loaded delivery system complex, such as the surface charge and the hydrophobicity, are decisive for the interactions with the components of the in vitro system, generating phenomena such as cell membrane association and non-specific absorption to plastic ware surfaces [30,31]. Another important issue during the design of the analytical protocol is the choice of RNA isolation technique and the subsequent quality control. Despite the fact that several small RNA purification kits are commercially available, the main technologies to purify the small RNAs are the TRI reagent RNA purification (RNA Stat-60) and spin column chromatography purification (Total RNA Purification Kit). The results obtained in our

Fig. 3. RT efficiency and reproducibility. The RT protocol allowed for a constant RT efficiency in the entire concentration range tested. Standard concentrations of siRNA spiked in RNA samples from untreated EGFP-H1299 showed reproducible results and a good amplification efficiency. The RT efficiency was tested in four different experiments using two different batches of siRNA: batch 1 experiment 1 (circles), batch 1 experiment 2 (crosses), batch 1 experiment 3 (asterisks) and batch 3 experiment 4 (squares). Data represents means of three technical replicates and standard deviation (n=3).

GENE DELIVERY

S. Colombo et al. / Journal of Controlled Release 166 (2013) 220–226

GENE DELIVERY

224

S. Colombo et al. / Journal of Controlled Release 166 (2013) 220–226

laboratory show that the TRI reagent-based RNA purification guaranteed an effective dissociation of the siRNA from the delivery system (results not shown). For instance, studying poly(DL-lactide-co-glycolide acid) (PLGA)-based, siRNA-loaded nanoparticles, the phenol–chloroform treatment eased the release of the entrapped siRNA by effectively dissolving the polymeric scaffold (results not shown). In contrast, only a partial elution of the siRNA loaded in the polymeric carrier system was obtained using a column-based purification, which may bias the experimental result. On the other hand, the use of the column-based system to analyze transfection reagents such as LF and TKO assures superior purity and reproducibility. Therefore, the RNA purification system should be selected depending on the specific delivery system studied, even if the phenol–chloroform-based siRNA isolation procedure is suggested [32]. An additional option when using the TRI reagent isolation procedure is to optimize siRNA elution by adding proteinase K or heparin when analyzing carriers based on electrostatic complexation of siRNA with lipids or proteins [33–35]. It is fundamental to check the RNA purity and assure the absence of chemical contaminants that could have any inhibitory effect on the reverse transcriptase activity because the efficiency of enzymatic reactions is the bottleneck of the assay. Samples with an A260/A230 ratio above 1.8 were considered of an acceptable chemical purity. Samples with lower A260/A230 ratios were re-purified by precipitation overnight at 4 °C in double volumes of 95% ethanol and 0.3 M sodium acetate. Samples with an A260/A280 ratio above 1.8 were considered of an acceptable quality free from genomic DNA contamination. In addition, it is important that the RNA integrity is preserved during the sample preparation. For more specific recommendations to the design and quality assessment of specific experiments we refer to the Minimum Information for Publication of Quantitative Real-Time PCR Experiments (MIQE) guidelines [36]. 3.5. Result analysis: Normalization and reference genes The quantification of qPCR results relative to the expression level of housekeeping small RNAs allows for obtaining more accurate and reproducible results, as well as providing a further quality control [22,28]. The choice of stable and reliable normalizers and normalization techniques is an important issue during establishment of the analytical assay, especially because it could substantially affect the assay results [37], depending on the platform and experimental setup used. The NormFinder was chosen among the available qPCR normalizer selection programs due to its flexibility and informative output [38]. The main criterion for selecting the normalizer gene was to avoid any influence of the cell treatment on the expression level of the normalizer. Therefore, a selection of normalizers was ranked according to their ability to not undergo changes in their expression profile due to the siRNA delivery procedure and/or components. Thus, the expression level of a pool of nine small RNAs (Supplementary Material, Table 1) was studied under the following conditions: Untreated cells (control group) and cells transfected with i) naked EGFP siRNA, ii) LF or TKO complexed with negative control FLuc siRNA (25 nM), iii) TKO or LF with no siRNA and iv) four different concentrations of EGFP siRNA complexed with LF and TKO. The results were analyzed using different groupings and criteria (Supplementary Material, Table 3). For each control/sample group, the small RNA that showed the lowest intragroup variation was selected. When analyzing the negative controls independently, the intergroup variation and the stability score were also considered as selection criteria. The intergroup variation and stability score calculated for the transfected sample was not considered relevant information, because it cannot be assumed that there is any correlation between the normalizer regulation and the siRNA transfection process [39]. The small nucleolar RNA U109 (snoRNAU109) appeared to be the most suitable normalizer gene because it showed the least degree of variation among all the samples. In addition, the Let7-a (Genbank ID: NR_029476) was promising, but

this reference showed a high variability among different cell passage numbers and was therefore not included in the analyses. However, we generally recommend, whenever possible, to use a combination of two reference small RNAs [40]. 3.6. Protocol test: Results and data analysis The developed protocol was used to analyze the siRNA delivery using the two commercial transfection reagents LF and TKO. Both reagents possess similar characteristics being lipoplexes upon complexation with siRNA. From the preliminary experiments, we verified that both the trypsinization and PBS washing were required when preparing the RNA sample (results not shown). The TRI-reagent RNA isolation was preferred to enable easy comparison with the results obtained with the polymeric nanoparticles as transfection reagent. The transfection reagents were studied at constant experimental conditions and varying the siRNA concentration during the lipoplex formation process. The LF reagent mediated a more efficient EGFP silencing compared to TKO (Fig. 4). However, the simultaneous intracellular siRNA quantification allowed for unambiguous demonstration that the enhanced RNAi effect mediated by LF was not a result of a more efficient siRNA delivery through the cell membrane, but rather due to an effective intracellular behavior that made the siRNA more active (Fig. 5). In fact, both transfection reagents were able to transfect high amounts of siRNA, but, especially at low concentrations, LF produced a more effective EGFP silencing than TKO. It was confirmed by correlating the quantified intracellular siRNA and the RNAi effect (Fig. 5) that increasing the number of siRNA molecules in the cytoplasm reduced the RNAi triggering ability. This observation has previously been reported as a result of saturation of the RNAi machinery [11]. As a final verification, the assay results were also quantified by absolute quantification to cross-refer the results obtained with previous independent findings [11,14], which showed similar transfection efficiencies for LF (0.5–2%, results not shown). The data discussed shows that the intracellular siRNA quantification by stem-loop RT qPCR is an informative analysis when studying the siRNA delivery. It can be used to define the delivery dynamics of siRNA transfection reagents. However, this analysis requires thorough optimization to avoid false or inconsistent results. It must be noticed that the standard experimental setup chosen for the reported delivery study was not optimized for the transfection systems. The presented results should therefore be considered as an example of siRNA delivery

Fig. 4. Comparison of gene silencing and siRNA delivery by EGFPS1 R25/D27 AS quantification. LF or TKO was complexed with different amounts of EGFPS1 R25D/27 and used for transfection at a final concentration of 0.04–5 nM. The % reduction in EGFP expression was recorded by flow cytometry, and was higher using LF (circles) than TKO (squares), even though both reagents mediated an effective transfection at low concentrations. The intracellular siRNA concentration resulting from LF transfection (rhombi) was similar and, at low siRNA concentrations, lower compared to the one resulting from TKO transfection (crosses). The LF produced a higher siRNA concentration only at 5 nM. Data represents mean values±SD (n=3).

225

Research Academy is kindly acknowledged for funding the material. The funding sources had no involvement in the study design, and in the collection, analysis and interpretation of data, just as they had no involvement in the writing of the report and the decision to submit the paper for publication. We wish to thank Mark A. Behlke (IDT), Kim A. Lennox (IDT) and Andrej N. Spiess (University Hospital Hamburg-Eppendorf) for the valuable scientific discussions. Appendix A. Supplementary data Supplementary data to this article can be found online at http:// dx.doi.org/10.1016/j.jconrel.2013.01.006. References Fig. 5. Relative siRNA efficacy. The siRNA concentration was divided by the EGFP silencing produced to obtain the relative number of siRNA corresponding to a depletion of 1% of EGFP (relative siRNA efficacy) and plotted as a function of the total silencing observed. Lower values correspond to higher siRNA delivery efficacy. The results suggest a significant relation between the silencing effect and the logarithmic value of the relative siRNA efficacy using LF (R2 =0.97, Pb 0.001). LF (circles) compared to TKO (squares) results in a higher relative siRNA efficacy (P=0.0005). The data suggests that LF mediated a higher EGFP silencing primarily by mediating a more favorable siRNA release in the cytoplasm. Data represents mean values±SD (n=3).

characterization by RT-qPCR intracellular siRNA quantification and not a quality dissertation on the commercial transfection reagents used. 4. Conclusion An accurate and customizable analytical protocol is required to be able to analyze the siRNA delivery dynamics of different carriers for nucleic acid-based drugs. In the present study, we thoroughly analyzed the main steps required for setting up an assay to quantify the intracellular siRNA delivery by carrier-mediated transfection. The challenges and the pitfalls regarding i) model system, ii) primer design, iii) RT and qPCR optimization, iv) sample preparation and v) normalizer selection were critically discussed, and clear guidelines allowing for the assay customization were provided. The result of this work is a convenient and documented analytical protocol to study the siRNA delivery in vitro by using EGFP expressing cell models. The system tested produced reliable results and provided useful information regarding the siRNA delivery process. Through the data obtained by stem-loop RT qPCR siRNA quantification, it was possible to characterize the delivery dynamics of two different siRNA carrier systems applied under different experimental conditions. The future perspectives are to use the protocol to study siRNA carriers in vitro as well as in vivo to improve our understanding of the intracellular delivery mechanism, dynamics and kinetics. Conflict of interests The authors declare that they have no competing interests. Acknowledgement We gratefully acknowledge the Danish Agency for Science, Technology and Innovation for funding this project. This study has been carried out with financial support from the Commission of the European Communities, Priority 3 “Nanotechnologies and Nanosciences, Knowledge Based Multifunctional Materials, New Production Processes and Devices” of the Sixth Framework Programme for Research and Technological Development (Targeted Delivery of Nanomedicine: NMP4-CT-2006-026668). We are grateful to the Danish Agency for Science, Technology and Innovation, the Drug Research Academy (University of Copenhagen) and the Carlsberg Foundation for the financial support for the Zetasizer Nano ZS, the Nanodrop 2000C spectrophotometer, the LightCycler® 480 system and the Gallios Flow Cytometer, respectively. In addition, the Drug

[1] C. Alabi, A. Vegas, D. Anderson, Attacking the genome: emerging siRNA nanocarriers from concept to clinic, Curr. Opin. Pharmacol. 12 (2012) 427–433. [2] K.A. Whitehead, R. Langer, D.G. Anderson, Knocking down barriers: advances in siRNA delivery, Nat. Rev. Drug Discov. 8 (2009) 129–138. [3] S. Benter, I. Akhtar, Toxicogenomics of non-viral drug delivery systems for RNAi: potential impact on siRNA-mediated gene silencing activity and specificity, Adv. Drug Deliv. Rev. 59 (2007) 164–182. [4] P. Lingor, U. Michel, U. Scholl, M. Bahr, S. Kugler, Transfection of “naked” siRNA results in endosomal uptake and metabolic impairment in cultured neurons, Biochem. Biophys. Res. Commun. 315 (2004) 1126–1133. [5] C. Foged, siRNA delivery with lipid-based systems: promises and pitfalls, Curr. Top. Med. Chem. 12 (2012) 97–107. [6] C.K. Raymond, B.S. Roberts, P. Garrett-Engele, L.P. Lim, J.M. Johnson, Simple, quantitative primer-extension PCR assay for direct monitoring of microRNAs and short-interfering RNAs, RNA 11 (2005) 1737–1744. [7] P. Mestdagh, T. Feys, N. Bernard, S. Guenther, C. Chen, F. Speleman, J. Vandesompele, High-throughput stem-loop RT-qPCR miRNA expression profiling using minute amounts of input RNA, Nucleic Acids Res. 36 (2008) e143. [8] S. Sharbati-Tehrani, B. Kutz-Lohroff, R. Bergbauer, J. Scholven, R. Einspanier, miR-Q: A novel quantitative RT-PCR approach for the expression profiling of small RNA molecules such as miRNAs in a complex sample, BMC Mol. Biol. 9 (2008) 34. [9] Y. Landesman, N. Svrzikapa, A. Cognetta, X. Zhang, B.R. Bettencourt, S. Kuchimanchi, K. Dufault, S. Shaikh, M. Gioia, A. Akinc, R. Hutabara, R. Meyers, In vivo quantification of formulated and chemically modified small interfering RNA by heating-in-Triton quantitative reverse transcription polymerase chain reaction (HIT qRT-PCR), Silence 1 (2010) 16. [10] H. Akita, R. Ito, I.A. Khalil, S. Futaki, H. Harashima, Quantitative three-dimensional analysis of the intracellular trafficking of plasmid DNA transfected by a nonviral gene delivery system using confocal laser scanning microscopy, Mol. Ther. 9 (2004) 443–451. [11] M. Overhoff, W. Wünsche, G. Sczakiel, Quantitative detection of siRNA and single-stranded oligonucleotides: relationship between uptake and biological activity of siRNA, Nucleic Acids Res. 32 (2004) e170. [12] S. Stratford, S. Stec, V. Jadhav, J. Seitzer, M. Abrams, M. Beverly, Examination of real-time polymerase chain reaction methods for the detection and quantification of modified siRNA, Anal. Biochem. 379 (2008) 96–104. [13] V. Benes, M. Castoldi, Expression profiling of microRNA using real-time quantitative PCR, how to use it and what is available, Methods 50 (2010) 244–249. [14] A. Cheng, M. Li, Y. Liang, Y. Wang, L. Wong, C. Chen, A.V. Vlassov, S. Magdaleno, Stem-loop RT-PCR quantification of siRNAs in vitro and in vivo, Oligonucleotides 19 (2009) 203–208. [15] C. Chen, R. Tan, L. Wong, R.H.J. Fekete, J. Halsey, Quantitation of microRNAs by real-time RT-qPCR, Methods Mol. Biol. 687 (2011) 113–134. [16] S.D. Rose, D.H. Kim, M. Amarzguioui, J.D. Heidel, M.A. Collingwood, M.E. Davis, J.J. Rossi, M.A. Behlke, Functional polarity is introduced by Dicer processing of short substrate RNAs, Nucleic Acids Res. 33 (2005) 4140–4156. [17] D.M.K. Jensen, D. Cun, M.J. Maltesen, S. Frokjaer, H.M. Nielsen, C. Foged, Spray drying of siRNA-containing PLGA nanoparticles intended for inhalation, J. Control. Release 142 (2010) 138–145. [18] C. Chen, D.A. Ridzon, A.J. Broomer, Z. Zhou, D.H. Lee, J.T. Nguyen, M. Barbisin, N.L. Xu, V.R. Mahuvakar, M.R. Andersen, K.Q. Lao, K.J. Livak, K.J. Guegler, Real-time quantification of microRNAs by stem-loop RT-PCR, Nucleic Acids Res. 33 (2005) e179. [19] N.R. Markham, M. Zuker, UNAFold: software for nucleic acid folding and hybridization, Methods Mol. Biol. 453 (2008) 3–31. [20] C. Ritz, A.N. Spiess, qpcR: an R package for sigmoidal model selection in quantitative real-time polymerase chain reaction analysis, Bioinformatics 24 (2008) 1549–1551. [21] A.N. Spiess, C. Feig, C. Ritz, Highly accurate sigmoidal fitting of real-time PCR data by introducing a parameter for asymmetry, BMC Bioinforma. 9 (2008) 221. [22] K.J. Livak, T.D. Schmittgen, Analysis of relative gene expression data using real-time quantitative PCR and the 2(−Delta Delta C(T)) Method, Methods 25 (2001) 402–408. [23] D.H. Kim, M.A. Behlke, S.D. Rose, M.S. Chang, S. Choi, J.J. Rossi, Synthetic dsRNA Dicer substrates enhance RNAi potency and efficacy, Nat. Biotechnol. 23 (2005) 222–226. [24] G. Wan, Q.E. Lim, H.P. Too, High-performance quantification of mature microRNAs by real-time RT-PCR using deoxyuridine-incorporated oligonucleotides and hemi-nested primers, RNA 16 (2010) 1436–1445.

GENE DELIVERY

S. Colombo et al. / Journal of Controlled Release 166 (2013) 220–226

GENE DELIVERY

226

S. Colombo et al. / Journal of Controlled Release 166 (2013) 220–226

[25] H. Yang, J.J. Schmuke, L.M. Flagg, J.K. Roberts, E.M. Allen, S. Ivashuta, L.A. Gilbertson, T.A. Armstrong, A.T. Christian, A novel real-time polymerase chain reaction method for high throughput quantification of small regulatory RNAs, Plant Biotechnol. J. 7 (2009) 621–630. [26] E. Varkonyi-Gasic, R. Wu, M. Wood, E.F. Walton, R.P. Hellens, Protocol: a highly sensitive RT-PCR method for detection and quantification of microRNAs, Plant Methods 3 (2007) 12. [27] E. Varkonyi-Gasic, R.P. Hellens, qRT-PCR of small RNAs, Methods Mol. Biol. 631 (2010) 109–122. [28] M.W. Pfaffl, A new mathematical model for relative quantification in real-time RT-PCR, Nucleic Acids Res. 29 (2001) e45. [29] F. Tang, P. Hajkova, S.C. Barton, D.A. O'Carroll, C. Lee, K. Lao, M.A. Surani, 220-plex microRNA expression profile of a single cell, Nat. Protoc. 1 (2006) 1154–1159. [30] D. Vercauteren, R.E. Vandenbroucke, A.T. Jones, J. Rejman, J. Demeester, S.C. De Smedt, N.N. Sanders, K. Braeckmans, The use of inhibitors to study endocytic pathways of gene carriers: optimization and pitfalls, Mol. Ther. 18 (2010) 561–569. [31] J.K. Vasir, V. Labhasetwar, Quantification of the force of nanoparticle–cell membrane interactions and its influence on intracellular trafficking of nanoparticles, Biomaterials 29 (2008) 4244–4252. [32] P. Chomczynski, N. Sacchi, Single-step method of RNA isolation by acid guanidinium thiocyanate–phenol–chloroform extraction, Anal. Biochem. 162 (1987) 156–159. [33] S.T. Glenn, K.L. Head, B.T. Teh, K.W. Gross, H.L. Kim, Maximizing RNA yield from archival renal tumors and optimizing gene expression analysis, J. Biomol. Screen. 15 (2010) 80–85.

[34] M.L. Johnson, C. Navanukraw, A.T. Grazul-Bilska, L.P. Reynolds, D.A. Redmer, Heparinase treatment of RNA before quantitative real-time RT-PCR, Biotechniques 35 (2003) 1140–1144. [35] A. Kotorashvili, A. Ramnauth, C. Liu, J. Lin, K. Ye, R. Kim, R. Hazan, T. Rohan, S. Fineberg, O. Loudig, Effective DNA/RNA co-extraction for analysis of microRNAs, mRNAs, and genomic DNA from formalin-fixed paraffin-embedded specimens, PLoS One 7 (2012) e34683. [36] S.A. Bustin, V. Benes, J.A. Garson, J. Hellemans, J. Huggett, M. Kubista, R. Mueller, T. Nolan, M.W. Pfaffl, G.L. Shipley, J. Vandesompele, C.T. Wittwer, The MIQE guidelines: minimum information for publication of quantitative real-time PCR experiments, Clin. Chem. 55 (2009) 611–622. [37] S.U. Meyer, S. Kaiser, C. Wagner, C. Thirion, M.W. Pfaffl, Profound effect of profiling platform and normalization strategy on detection of differentially expressed microRNAs—a comparative study, PLoS One 7 (2012) e38946. [38] C.L. Andersen, J.L. Jensen, T.F. Ørntoft, Normalization of real-time quantitative reverse transcription-PCR data: a model-based variance estimation approach to identify genes suited for normalization, applied to bladder and colon cancer data sets, Cancer Res. 64 (2004) 5245–5250. [39] H.J. Peltier, G.J. Latham, Normalization of microRNA expression levels in quantitative RT-PCR assays: identification of suitable reference RNA targets in normal and cancerous human solid tissues, RNA 14 (2008) 844–852. [40] J. Vandesompele, K.D. Preter, F. Pattyn, B. Poppe, N.V. Roy, A.D. Paepe, F. Speleman, Accurate normalization of real-time quantitative RT-PCR data by geometric averaging of multiple internal control genes, Genome Biol. 3 (2002) 0034.1–0034.11.