Analytical Biochemistry 484 (2015) 136–142
Contents lists available at ScienceDirect
Analytical Biochemistry journal homepage: www.elsevier.com/locate/yabio
Optimized luciferase assay for cell-penetrating peptide-mediated delivery of short oligonucleotides Henrik Helmfors ⇑, Jonas Eriksson, Ülo Langel Department of Neurochemistry, Stockholm University, Stockholm 10691, Sweden
a r t i c l e
i n f o
Article history: Received 22 April 2015 Received in revised form 20 May 2015 Accepted 31 May 2015 Available online 4 June 2015 Keywords: Cell-penetrating peptide (CPP) Luciferase Short interfering RNA (siRNA) Splice-correcting oligonucleotide (SCO) High-throughput screening (HTS)
a b s t r a c t An improved assay for screening for the intracellular delivery efficacy of short oligonucleotides using cell-penetrating peptides is suggested. This assay is an improvement over previous assays that use luciferase reporters for cell-penetrating peptides because it has been scaled up from a 24-well format to a 96-well format and no longer relies on a luciferin reagent that has been commercially sourced. In addition, the homemade luciferin reagent is useful in multiple cell lines and in different assays that rely on altering the expression of luciferase. To establish a new protocol, the composition of the luciferin reagent was optimized for both signal strength and longevity by multiple two-factorial experiments varying the concentrations of adenosine triphosphate, luciferin, coenzyme A, and dithiothreitol. In addition, the optimal conditions with respect to cell number and time of transfection for both short interfering RNA (siRNA) and splice-correcting oligonucleotides (SCOs) are established. Optimal transfection of siRNA and SCOs was achieved using the reverse transfection method where the oligonucleotide complexes are already present in the wells before the cells are plated. Z0 scores were 0.73 for the siRNA assay and 0.71 for the SCO assay, indicating that both assays are suitable for high-throughput screening. Ó 2015 Elsevier Inc. All rights reserved.
An increasing portion of the discovery phase in life science research is moving toward an automated high-throughput screening (HTS)1 approach to assays. This development leads to a quicker gain of data, allowing a researcher to swiftly weed out the uninteresting and quickly discover what may work and what may be a waste of time. However, automation up to the HTS level in academia is available only to those academic labs with the resources to use one of the HTS facilities available in some countries and universities. Most academic labs seldom use any automated approaches. During the past decades, a new class of peptide-based carriers that have the ability to cross the cellular membrane has emerged. These peptides are called cell-penetrating peptides (CPPs). These peptides are (mostly) cationic (and/or amphipathic) and, in addition to self-translocation, also have the ability to carry cargo with them into the intracellular environment [1,2]. The cargo that these peptides are able to carry is diverse, ranging from small molecules [3] to plasmids [4] and other oligo- and polynucleotides such as ⇑ Corresponding author. E-mail address: [email protected] (H. Helmfors). Abbreviations used: HTS, high-throughput screening; CPP, cell-penetrating peptide; siRNA, short interfering RNA; SCO, splice-correcting oligonucleotide; DTT, dithiothreitol; CoA, coenzyme A; EDTA, ethylenediaminetetraacetic acid; ATP, adenosine triphosphate; HPLC, high-performance liquid chromatography; mRNA, messenger RNA; RNAi, RNA interference; LH2, D-luciferin; CV, coefficient of variation. 1
http://dx.doi.org/10.1016/j.ab.2015.05.023 0003-2697/Ó 2015 Elsevier Inc. All rights reserved.
short interfering RNA (siRNA) and splice-correcting oligonucleotides (SCOs) [5,6] all the way to proteins [7]. The therapeutic potential seems unlimited, but the CPP technology has so far not led to any clinical trials. Studies investigating toxicity and biodistribution are ongoing, and CPP oligonucleotide therapies are anticipated to reach clinical trials in the near future [8]. So far, the standardized way to describe the internalization properties of CPPs has been to measure the uptake of fluorescently labeled peptides either through a fluorimetric measurement or by confocal microscopy [9]. This is a good way to measure the uptake of individual CPPs with small hydrophobic (the fluorophore) covalently coupled cargoes; however, this is less interesting if the main objective is to use the CPPs as delivery vehicles for macromolecules. Successful delivery of a macromolecule should preferably result in some form of biological activity that is easy to measure. Many assays rely on either one of the two most popular reporter genes: green fluorescent protein [10,11] or luciferase [12]. The main benefits of luciferase-based assays are that the readout is simple and the signal-to-background ratio is very high. One such assay that we previously found to be practical [13] is the splice-correction assay developed by Kole and coworkers [14]. We previously investigated the oligonucleotide cotranslocation potency of various peptide vectors using this assay [5,6,15,16]. It is a cellular assay
CPP-mediated short oligonucleotide delivery / H. Helmfors et al. / Anal. Biochem. 484 (2015) 136–142
in which the activity of an antisense oligonucleotide results in upregulation of luciferase gene expression. In the past, we have relied on small-scale use of 24-well cell culture plates for all of our experiments. This has been adequate when assessing one or at most a few CPPs at the same time. Many publications about CPPs report on the discovery of, and sometimes applications for, a single new peptide [5,6,15–17], whereas there are fewer publications that present a series of peptides [18,19]. Being able to compare multiple peptides and transfection conditions simultaneously over a shorter time would increase the rate of discovery. For this purpose, here we present a method that aims to reduce the size of our previously very successful 24-well assay for splice correction [13] into a 96-well format. At this format, the reagent usage for CPPs and oligonucleotides is the same as for a 24-well plate with the added benefit of being able to test up to four times as many experimental conditions simultaneously; however, the luciferase reagent is consumed at a four times higher amount. That rate of reagent consumption makes a commercial reagent almost prohibitively expensive. Taking four times as many measurements simultaneously reveals a bottleneck for efficiency. The simplest luminometers, which lack reagent injectors, read one well at a time and take more than 3 min to measure one 96-well plate. During this time, the signal may change by up to 20% between the first and last measurements of a 96-well plate. Here we present a method for evaluating transfection via CPPs that has the potential to scale up to true HTS. It is based on a cost-effective homemade reagent for the firefly luminescence assay. The work presented is optimized for transfection of short oligonucleotides such as SCOs and siRNAs. This journal recently published a miniaturized gene transfection assay using luciferase plasmid in 384- and 1536-well plates [20]. The transfection was done using CaPO4–DNA or PEI–DNA with automated robotic liquid handling and was read using high-end equipment and a commercial luciferase reagent. Our assay was done manually using handheld pipettes, making it also useful for those without specialized equipment. In addition, our assay reagent significantly reduces the cost of one assay; for example, according to current list price (April 2015), the Promega ONE-Glo reagent costs approximately 90 euro per 96-well plate, whereas our assay reagent costs approximately 3 euro per 96-well plate, which is a 30-fold cost savings. Materials and methods Reagents For delivery experiments and luciferin assay buffer, PS-20 -OMe splice-correcting oligonucleotides (50 -CCU CUU ACC UCA GUU ACA-30 ) were purchased from RiboTask (Denmark), D-luciferin was purchased from PerkinElmer (Sweden), and siRNA against firefly luciferase [ACGCCAAAAACAUAAAGAAAG(dTdT)] was purchased from Eurofins Genomics (Germany). All other reagents, MgCO3, MgSO4, Tricine, dithiothreitol (DTT), coenzyme A (CoA), ethylenediaminetetraacetic acid (EDTA), and adenosine triphosphate (ATP) were obtained from Sigma–Aldrich (Sweden). Peptide synthesis PepFect14 [4,6] was synthesized on a Rink Amide ChemMatrix resin (PCAS Biomatrix, Canada), on a Biotage Initiator + Alstra peptide synthesis machine (Biotage, Sweden), using N,N0 -diisopropylcarbodiimide (DIC) and OxymaPure as coupling reagents and standard Fmoc protected amino acid monomers (Iris Biotech, Germany). Peptide was cleaved using 95% trifluoroacetic acid (TFA, Iris Biotech), 2.5% H2O, and 2.5%
137
triisopropylsilane (Sigma–Aldrich) and was precipitated in cold diethyl ether. The obtained crude peptide was dissolved in 5% acetic acid and lyophilized. The peptide was purified by high-performance liquid chromatography (HPLC) using a preparative BioBasic C8 column (ThermoFisher, Sweden) with a gradient elution made up of acetonitrile and water. The identity of the purified product was verified using matrix-assisted laser desorption ionization mass spectrometry (MALDI–MS) (Voyager-DE STR, Applied Biosystems). After HPLC purification, the peptide was lyophilized and peptide solutions used later were based on dilutions of accurately weighed substance. Cell cultures HeLa pLuc 705 and U-87 MG-luc2 cells were both maintained in Dulbecco’s modified Eagle’s medium (DMEM, Life Technologies) supplemented with glutamax, 1 mM sodium pyruvate, 10% fetal bovine serum (FBS, Life Technologies), 100 U/ml penicillin, and 100 lg/ml streptomycin in a water-jacketed incubator at 37 °C and 5% CO2. For experiments that investigated the composition of the luciferin reagent, 1 104 HeLa pLuc 705 cells were seeded into white, tissue culture-treated, clear flat-bottom 96-well plates (Corning) 24 h prior to transfection. On the day of transfection, the cell culture medium was replaced with fresh medium prior to treatment with peptide/oligonucleotide complexes. Optimal results in the splice-correcting assay were achieved when 7 103 cells were seeded into the same kind of 96-well plates as above that already contained the peptide/oligonucleotide complexes—so-called reverse transfection. After 24 h, most of the cell medium was carefully aspirated, and the plates were then centrifuged upside down at low (200–300) rpm for 15 s in a plate centrifuge to completely empty them of cell medium. Cells were subsequently frozen at 80 °C and lysed with a single freeze–thaw cycle, the frozen plates were allowed to reach room temperature before being assayed for luciferase activity. For siRNA experiments, 7 103 U-87 MG-luc2 cells stably expressing an enhanced version of the firefly luciferase enzyme were seeded into the same type of plates as above, with the plates already containing the peptide/siRNA complexes. Then, 24 h post-transfection, the plates were treated the same way as for the SCO experiments, aspirated, centrifuged upside-down, freeze–thawed, and assayed as above. In vitro transfection SCOs and PF14 were mixed differently than as described previously. The peptide was dissolved in ultrapure water (MilliQ) at 1 mM concentration, and aliquots of this solution were kept frozen at 20 °C. Aliquots were thawed on the day of transfection and diluted in ultrapure water to 100 lM. SCOs were delivered lyophilized, and on arrival they were dissolved in RNase-free ultrapure water at 100 lM concentration and aliquoted. On the day of transfection, aliquots were thawed and diluted to 10 lM concentration. Peptide and SCO were mixed at a 5:1 M ratio in a phosphate/sodium buffer at pH 7.4 to a final concentration of 7.5/1.5 lM, and 10 ll of the complex solution was added to the 96-well plate prior to plating 90 ll of HeLa pLuc 705 cell suspension, resulting in a final concentration of 150 nM SCO and 750 nM PF14. The principle behind the SCO delivery assay is as follows. A plasmid, HeLa pLuc 705, carrying the luciferase gene is interrupted with an insertion of intron 2 from b-globin pre-mRNA (messenger RNA) carrying a cryptic splice site. If the aberrant splice site is not masked by antisense oligonucleotides (SCOs), the pre-mRNA of luciferase will be improperly processed. Successful delivery leads to an increase in luminescence.
138
CPP-mediated short oligonucleotide delivery / H. Helmfors et al. / Anal. Biochem. 484 (2015) 136–142
Table 1 Reagent concentrations used for the multiple two-factorial experiments. LH2 (mM)
ATP (mM)
CoA (lM)
DTT (mM)
0.1 0.5 1 10
0.1 0.5 1 10
0.1 1 25 100
0.1 1 25 100
Note. LH2, D-luciferin.
For PF14/siRNA complexes, the peptide was treated as above, and the siRNA was dissolved using the buffer provided by the manufacturer according to the manufacturer’s recommendations. This was then further diluted to 1 lM solution in phosphate buffer (pH 7.4). Complexes between siRNA and PF14 were made in a molar ratio of 30:1 with the siRNA at a concentration of 400 nM and the PF14 concentration at 12 lM, and 10 ll of the complex solution was added to white 96-well plates prior to plating 90 ll of U-87 cell suspension, resulting in a final concentration of 40 nM siRNA and 1.2 lM PF14. Successful delivery of siRNA leads to degradation of the luciferase mRNA by recruitment of the RNAi (RNA interference) pathway [21], which in turn leads to a decrease in luminescence. Assay reagent The luciferin assay reagent is based on the reagent reported by Siebring-van Olst and coworkers [22]. The components of the assay buffer are MgCO3, MgSO4, ATP, CoA, D-luciferin (LH2), DTT, and EDTA in Tricine buffer. Stock concentrations of the components were as follows: DTT, 500 mM; LH2, 50 mM; ATP, 50 mM; CoA, 1 mM; EDTA, 10 mM; Tricine, 200 mM; MgCO3, 10 mM; and MgSO4, 50 mM. The pH was set to 7.8 for Tricine and magnesium carbonate. The sensitive components ATP, CoA, LH2, and DTT were kept frozen, whereas the others were kept at room temperature. The first four components were then mixed at different concentrations as indicated in Table 1 to find the optimal composition. The final concentrations for Tricine, MgCO3, MgSO4, and EDTA were 20, 1, 5, and 1 mM, respectively. Results and discussion When the assay format is changed from 24 to 96 wells, more data points need to be read after each experiment. Working with a luminometer that lacks injectors has revealed a problem with the commercial reagent; it has the inherent property that the luminescence signal changes by up to 20% between the first and last measurements in one 96-well plate (the time that passes between two measurements in all figures corresponds to the time it takes to measure 96 wells, which is 3 min) and then rapidly loses approximately 70% of the signal within the first 30 min (see Fig. S1 in online Supplementary Material). This is disadvantageous for two reasons. First, the large difference between subsequent measures is a major source of uncertainty; small differences between samples would routinely be missed because the signal decreases faster than the instrument can measure. Second, in terms of multiplexing analysis, plates will need to be prepared and measured sequentially; luciferin reagent needs to be added to a plate, and that plate then immediately needs to be measured before the next plate can be treated. To counter the changing conditions, it is suggested to omit CoA from the luciferase reagent [22]. However, we find that in our experiments, although not having any CoA does indeed minimize the rate of change in the signal, the signal strength is also decreased by more than 10-fold (data not shown) compared with the assay reagent composition that we finally settled on. That
presents a different problem; there is a loss of sensitivity. As illustrated in Fig. 1, having the lowest amount of CoA also gives the lowest and most stable signal. CoA is a major effector of maximum initial signal, and its concentration needs to be controlled in order to get high persistent luminescence. Concentrations of CoA below 1 lM do not seem to affect signal strength, whereas higher concentrations of CoA scale with higher signal strength but also lead to a more rapid decrease in signal. DTT seems to be the main effector of the longevity of the signal; as can be seen in Fig. 1, the signals that decrease the fastest both contain the least amount of DTT, and the strongest signal in accordance with the above also has the most CoA. The reagent composition providing the most stable signal (80%). Reducing the size from 24-well plates to 96-well plates has changed how cells should be transfected with short oligonucleotides such as SCO and siRNA using CPPs; the 24-well assay requires transfer from 24-well plates to 96 wells for luminescence reading. This assay reduces the time needed from seeding of cells to attaining results by 33% while simultaneously reducing the number of pipetting steps from seven to three and reducing the cost by more than 95% and the waste by one microplate. In addition, this report establishes PF14 as a de facto positive control both for CPP-mediated delivery of splice-correcting oligonucleotides in the HeLa pLuc 705 cell line and for siRNA delivery in the U-87 MG-luc2 cell line. Acknowledgments This work was supported by the Swedish Research Foundation (VR-NT, VR-Med) and the Swedish Cancer Society.
Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.cej.2014.04.115. References [1] R. Brasseur, G. Divita, Happy birthday cell penetrating peptides: already 20 years, Biochim. Biophys. Acta 1798 (2010) 2177–2181. [2] M. Lindgren, M. Hällbrink, A. Prochiantz, Ü. Langel, Cell-penetrating peptides, Trends Pharmacol. Sci. 21 (2000) 99–103. [3] C. Rousselle, P. Clair, J. Temsamani, J.-M. Scherrmann, Improved brain delivery of benzylpenicillin with a peptide-vector-mediated strategy, J. Drug Target. 10 (2002) 309–315. [4] K.-L. Veiman, I. Mäger, K. Ezzat, H. Margus, T. Lehto, K. Langel, et al., PepFect14 peptide vector for efficient gene delivery in cell cultures, Mol. Pharm. 10 (2013) 199–210. [5] S. El-Andaloussi, T. Lehto, I. Mäger, K. Rosenthal-Aizman, I.I. Oprea, O.E. Simonson, et al., Design of a peptide-based vector, PepFect6, for efficient delivery of siRNA in cell culture and systemically in vivo, Nucleic Acids Res. 39 (2011) 3972–3987.
[6] K. Ezzat, S. El-Andaloussi, E.M. Zaghloul, T. Lehto, S. Lindberg, P.M.D. Moreno, et al., PepFect 14, a novel cell-penetrating peptide for oligonucleotide delivery in solution and as solid formulation, Nucleic Acids Res. 39 (2011) 5284–5298. [7] S. Futaki, T. Suzuki, W. Ohashi, T. Yagami, S. Tanaka, K. Ueda, et al., Argininerich peptides: an abundant source of membrane-permeable peptides having potential as carriers for intracellular protein delivery, J. Biol. Chem. 276 (2001) 5836–5840. [8] P. Boisguerin, S. Deshayes, M.J. Gait, L. O’Donovan, C. Godfrey, C.A. Betts, et al., Delivery of therapeutic oligonucleotides with cell penetrating peptides, Adv. Drug Deliv. Rev. (2015), http://dx.doi.org/10.1016/j. addr.2015.02.008. [9] T. Holm, H. Johansson, P. Lundberg, M. Pooga, M. Lindgren, Ü. Langel, Studying the uptake of cell-penetrating peptides, Nat. Protoc. 1 (2006) 1001–1005. [10] O. Shimomura, F.H. Johnson, Y. Saiga, Extraction, purification, and properties of aequorin, a bioluminescent protein from the luminous hydromedusan, Aequorea, J. Cell. Comp. Physiol. 59 (1962) 223–239. [11] R.Y. Tsien, The green fluorescent protein, Annu. Rev. Biochem. 67 (1998) 509– 544. [12] H. Fraga, Firefly luminescence: a historical perspective and recent developments, Photochem. Photobiol. Sci. 7 (2008) 146–158. [13] S. El-Andaloussi, P. Guterstam, Ü. Langel, Assessing the delivery efficacy and internalization route of cell-penetrating peptides, Nat. Protoc. 2 (2007) 2043– 2047. [14] S.H. Kang, M.J. Cho, R. Kole, Up-regulation of luciferase gene expression with antisense oligonucleotides: implications and applications in functional assay development, Biochemistry 37 (1998) 6235–6239. [15] S. Lindberg, A. Muñoz-Alarcón, H. Helmfors, D. Mosqueira, D. Gyllborg, O. Tudoran, et al., PepFect15, a novel endosomolytic cell-penetrating peptide for oligonucleotide delivery via scavenger receptors, Int. J. Pharm. 441 (2013) 242–247. [16] S. El-Andaloussi, H. Johansson, T. Holm, Ü. Langel, A novel cell-penetrating peptide, M918, for efficient delivery of proteins and peptide nucleic acids, Mol. Ther. 15 (2007) 1820–1826. [17] L. Crombez, G. Aldrian-Herrada, K. Konate, Q.N. Nguyen, G.K. McMaster, R. Brasseur, et al., A new potent secondary amphipathic cell-penetrating peptide for siRNA delivery into mammalian cells, Mol. Ther. 17 (2009) 95–103. [18] J. Regberg, A. Srimanee, M. Erlandsson, R. Sillard, D.A. Dobchev, M. Karelson, et al., Rational design of a series of novel amphipathic cell-penetrating peptides, Int. J. Pharm. 464 (2014) 111–116. [19] K. Langel, S. Lindberg, D. Copolovici, P. Arukuusk, R. Sillard, Ü. Langel, Novel fatty acid modifications of transportan 10, Int. J. Pept. Res. Ther. 16 (2010) 247–255. [20] J. Li, S.T. Crowley, J. Duskey, S. Khargharia, M. Wu, K.G. Rice, Miniaturization of gene transfection assays in 384- and 1536-well microplates, Anal. Biochem. 470 (2015) 14–21. [21] A. Fire, S. Xu, M.K. Montgomery, S.A. Kostas, S.E. Driver, C.C. Mello, Potent and specific genetic interference by double-stranded RNA in Caenorhabditis elegans, Nature 391 (1998) 806–811. [22] E. Siebring-van Olst, C. Vermeulen, R.X. de Menezes, M. Howell, E.F. Smit, V.W. Van Beusechem, Affordable luciferase reporter assay for cell-based highthroughput screening, J. Biomol. Screen. 18 (2013) 453–461. [23] K.A. Barriscale, S.A. O’Sullivan, T.V. McCarthy, A single secreted luciferasebased gene reporter assay, Anal. Biochem. 453 (2014) 44–49. [24] K. Ezzat, H. Helmfors, O. Tudoran, C. Juks, S. Lindberg, K. Padari, et al., Scavenger receptor-mediated uptake of cell-penetrating peptide nanocomplexes with oligonucleotides, FASEB J. 26 (2012) 1172–1180. [25] S. Lindberg, J. Regberg, J. Eriksson, H. Helmfors, A. Muñoz-Alarcón, A. Srimanee, et al., A convergent uptake route for peptide- and polymer-based nucleotide delivery systems, J. Control. Release 206 (2015) 58–66.
Contents lists available at ScienceDirect
Analytical Biochemistry journal homepage: www.elsevier.com/locate/yabio
Optimized luciferase assay for cell-penetrating peptide-mediated delivery of short oligonucleotides Henrik Helmfors ⇑, Jonas Eriksson, Ülo Langel Department of Neurochemistry, Stockholm University, Stockholm 10691, Sweden
a r t i c l e
i n f o
Article history: Received 22 April 2015 Received in revised form 20 May 2015 Accepted 31 May 2015 Available online 4 June 2015 Keywords: Cell-penetrating peptide (CPP) Luciferase Short interfering RNA (siRNA) Splice-correcting oligonucleotide (SCO) High-throughput screening (HTS)
a b s t r a c t An improved assay for screening for the intracellular delivery efficacy of short oligonucleotides using cell-penetrating peptides is suggested. This assay is an improvement over previous assays that use luciferase reporters for cell-penetrating peptides because it has been scaled up from a 24-well format to a 96-well format and no longer relies on a luciferin reagent that has been commercially sourced. In addition, the homemade luciferin reagent is useful in multiple cell lines and in different assays that rely on altering the expression of luciferase. To establish a new protocol, the composition of the luciferin reagent was optimized for both signal strength and longevity by multiple two-factorial experiments varying the concentrations of adenosine triphosphate, luciferin, coenzyme A, and dithiothreitol. In addition, the optimal conditions with respect to cell number and time of transfection for both short interfering RNA (siRNA) and splice-correcting oligonucleotides (SCOs) are established. Optimal transfection of siRNA and SCOs was achieved using the reverse transfection method where the oligonucleotide complexes are already present in the wells before the cells are plated. Z0 scores were 0.73 for the siRNA assay and 0.71 for the SCO assay, indicating that both assays are suitable for high-throughput screening. Ó 2015 Elsevier Inc. All rights reserved.
An increasing portion of the discovery phase in life science research is moving toward an automated high-throughput screening (HTS)1 approach to assays. This development leads to a quicker gain of data, allowing a researcher to swiftly weed out the uninteresting and quickly discover what may work and what may be a waste of time. However, automation up to the HTS level in academia is available only to those academic labs with the resources to use one of the HTS facilities available in some countries and universities. Most academic labs seldom use any automated approaches. During the past decades, a new class of peptide-based carriers that have the ability to cross the cellular membrane has emerged. These peptides are called cell-penetrating peptides (CPPs). These peptides are (mostly) cationic (and/or amphipathic) and, in addition to self-translocation, also have the ability to carry cargo with them into the intracellular environment [1,2]. The cargo that these peptides are able to carry is diverse, ranging from small molecules [3] to plasmids [4] and other oligo- and polynucleotides such as ⇑ Corresponding author. E-mail address: [email protected] (H. Helmfors). Abbreviations used: HTS, high-throughput screening; CPP, cell-penetrating peptide; siRNA, short interfering RNA; SCO, splice-correcting oligonucleotide; DTT, dithiothreitol; CoA, coenzyme A; EDTA, ethylenediaminetetraacetic acid; ATP, adenosine triphosphate; HPLC, high-performance liquid chromatography; mRNA, messenger RNA; RNAi, RNA interference; LH2, D-luciferin; CV, coefficient of variation. 1
http://dx.doi.org/10.1016/j.ab.2015.05.023 0003-2697/Ó 2015 Elsevier Inc. All rights reserved.
short interfering RNA (siRNA) and splice-correcting oligonucleotides (SCOs) [5,6] all the way to proteins [7]. The therapeutic potential seems unlimited, but the CPP technology has so far not led to any clinical trials. Studies investigating toxicity and biodistribution are ongoing, and CPP oligonucleotide therapies are anticipated to reach clinical trials in the near future [8]. So far, the standardized way to describe the internalization properties of CPPs has been to measure the uptake of fluorescently labeled peptides either through a fluorimetric measurement or by confocal microscopy [9]. This is a good way to measure the uptake of individual CPPs with small hydrophobic (the fluorophore) covalently coupled cargoes; however, this is less interesting if the main objective is to use the CPPs as delivery vehicles for macromolecules. Successful delivery of a macromolecule should preferably result in some form of biological activity that is easy to measure. Many assays rely on either one of the two most popular reporter genes: green fluorescent protein [10,11] or luciferase [12]. The main benefits of luciferase-based assays are that the readout is simple and the signal-to-background ratio is very high. One such assay that we previously found to be practical [13] is the splice-correction assay developed by Kole and coworkers [14]. We previously investigated the oligonucleotide cotranslocation potency of various peptide vectors using this assay [5,6,15,16]. It is a cellular assay
CPP-mediated short oligonucleotide delivery / H. Helmfors et al. / Anal. Biochem. 484 (2015) 136–142
in which the activity of an antisense oligonucleotide results in upregulation of luciferase gene expression. In the past, we have relied on small-scale use of 24-well cell culture plates for all of our experiments. This has been adequate when assessing one or at most a few CPPs at the same time. Many publications about CPPs report on the discovery of, and sometimes applications for, a single new peptide [5,6,15–17], whereas there are fewer publications that present a series of peptides [18,19]. Being able to compare multiple peptides and transfection conditions simultaneously over a shorter time would increase the rate of discovery. For this purpose, here we present a method that aims to reduce the size of our previously very successful 24-well assay for splice correction [13] into a 96-well format. At this format, the reagent usage for CPPs and oligonucleotides is the same as for a 24-well plate with the added benefit of being able to test up to four times as many experimental conditions simultaneously; however, the luciferase reagent is consumed at a four times higher amount. That rate of reagent consumption makes a commercial reagent almost prohibitively expensive. Taking four times as many measurements simultaneously reveals a bottleneck for efficiency. The simplest luminometers, which lack reagent injectors, read one well at a time and take more than 3 min to measure one 96-well plate. During this time, the signal may change by up to 20% between the first and last measurements of a 96-well plate. Here we present a method for evaluating transfection via CPPs that has the potential to scale up to true HTS. It is based on a cost-effective homemade reagent for the firefly luminescence assay. The work presented is optimized for transfection of short oligonucleotides such as SCOs and siRNAs. This journal recently published a miniaturized gene transfection assay using luciferase plasmid in 384- and 1536-well plates [20]. The transfection was done using CaPO4–DNA or PEI–DNA with automated robotic liquid handling and was read using high-end equipment and a commercial luciferase reagent. Our assay was done manually using handheld pipettes, making it also useful for those without specialized equipment. In addition, our assay reagent significantly reduces the cost of one assay; for example, according to current list price (April 2015), the Promega ONE-Glo reagent costs approximately 90 euro per 96-well plate, whereas our assay reagent costs approximately 3 euro per 96-well plate, which is a 30-fold cost savings. Materials and methods Reagents For delivery experiments and luciferin assay buffer, PS-20 -OMe splice-correcting oligonucleotides (50 -CCU CUU ACC UCA GUU ACA-30 ) were purchased from RiboTask (Denmark), D-luciferin was purchased from PerkinElmer (Sweden), and siRNA against firefly luciferase [ACGCCAAAAACAUAAAGAAAG(dTdT)] was purchased from Eurofins Genomics (Germany). All other reagents, MgCO3, MgSO4, Tricine, dithiothreitol (DTT), coenzyme A (CoA), ethylenediaminetetraacetic acid (EDTA), and adenosine triphosphate (ATP) were obtained from Sigma–Aldrich (Sweden). Peptide synthesis PepFect14 [4,6] was synthesized on a Rink Amide ChemMatrix resin (PCAS Biomatrix, Canada), on a Biotage Initiator + Alstra peptide synthesis machine (Biotage, Sweden), using N,N0 -diisopropylcarbodiimide (DIC) and OxymaPure as coupling reagents and standard Fmoc protected amino acid monomers (Iris Biotech, Germany). Peptide was cleaved using 95% trifluoroacetic acid (TFA, Iris Biotech), 2.5% H2O, and 2.5%
137
triisopropylsilane (Sigma–Aldrich) and was precipitated in cold diethyl ether. The obtained crude peptide was dissolved in 5% acetic acid and lyophilized. The peptide was purified by high-performance liquid chromatography (HPLC) using a preparative BioBasic C8 column (ThermoFisher, Sweden) with a gradient elution made up of acetonitrile and water. The identity of the purified product was verified using matrix-assisted laser desorption ionization mass spectrometry (MALDI–MS) (Voyager-DE STR, Applied Biosystems). After HPLC purification, the peptide was lyophilized and peptide solutions used later were based on dilutions of accurately weighed substance. Cell cultures HeLa pLuc 705 and U-87 MG-luc2 cells were both maintained in Dulbecco’s modified Eagle’s medium (DMEM, Life Technologies) supplemented with glutamax, 1 mM sodium pyruvate, 10% fetal bovine serum (FBS, Life Technologies), 100 U/ml penicillin, and 100 lg/ml streptomycin in a water-jacketed incubator at 37 °C and 5% CO2. For experiments that investigated the composition of the luciferin reagent, 1 104 HeLa pLuc 705 cells were seeded into white, tissue culture-treated, clear flat-bottom 96-well plates (Corning) 24 h prior to transfection. On the day of transfection, the cell culture medium was replaced with fresh medium prior to treatment with peptide/oligonucleotide complexes. Optimal results in the splice-correcting assay were achieved when 7 103 cells were seeded into the same kind of 96-well plates as above that already contained the peptide/oligonucleotide complexes—so-called reverse transfection. After 24 h, most of the cell medium was carefully aspirated, and the plates were then centrifuged upside down at low (200–300) rpm for 15 s in a plate centrifuge to completely empty them of cell medium. Cells were subsequently frozen at 80 °C and lysed with a single freeze–thaw cycle, the frozen plates were allowed to reach room temperature before being assayed for luciferase activity. For siRNA experiments, 7 103 U-87 MG-luc2 cells stably expressing an enhanced version of the firefly luciferase enzyme were seeded into the same type of plates as above, with the plates already containing the peptide/siRNA complexes. Then, 24 h post-transfection, the plates were treated the same way as for the SCO experiments, aspirated, centrifuged upside-down, freeze–thawed, and assayed as above. In vitro transfection SCOs and PF14 were mixed differently than as described previously. The peptide was dissolved in ultrapure water (MilliQ) at 1 mM concentration, and aliquots of this solution were kept frozen at 20 °C. Aliquots were thawed on the day of transfection and diluted in ultrapure water to 100 lM. SCOs were delivered lyophilized, and on arrival they were dissolved in RNase-free ultrapure water at 100 lM concentration and aliquoted. On the day of transfection, aliquots were thawed and diluted to 10 lM concentration. Peptide and SCO were mixed at a 5:1 M ratio in a phosphate/sodium buffer at pH 7.4 to a final concentration of 7.5/1.5 lM, and 10 ll of the complex solution was added to the 96-well plate prior to plating 90 ll of HeLa pLuc 705 cell suspension, resulting in a final concentration of 150 nM SCO and 750 nM PF14. The principle behind the SCO delivery assay is as follows. A plasmid, HeLa pLuc 705, carrying the luciferase gene is interrupted with an insertion of intron 2 from b-globin pre-mRNA (messenger RNA) carrying a cryptic splice site. If the aberrant splice site is not masked by antisense oligonucleotides (SCOs), the pre-mRNA of luciferase will be improperly processed. Successful delivery leads to an increase in luminescence.
138
CPP-mediated short oligonucleotide delivery / H. Helmfors et al. / Anal. Biochem. 484 (2015) 136–142
Table 1 Reagent concentrations used for the multiple two-factorial experiments. LH2 (mM)
ATP (mM)
CoA (lM)
DTT (mM)
0.1 0.5 1 10
0.1 0.5 1 10
0.1 1 25 100
0.1 1 25 100
Note. LH2, D-luciferin.
For PF14/siRNA complexes, the peptide was treated as above, and the siRNA was dissolved using the buffer provided by the manufacturer according to the manufacturer’s recommendations. This was then further diluted to 1 lM solution in phosphate buffer (pH 7.4). Complexes between siRNA and PF14 were made in a molar ratio of 30:1 with the siRNA at a concentration of 400 nM and the PF14 concentration at 12 lM, and 10 ll of the complex solution was added to white 96-well plates prior to plating 90 ll of U-87 cell suspension, resulting in a final concentration of 40 nM siRNA and 1.2 lM PF14. Successful delivery of siRNA leads to degradation of the luciferase mRNA by recruitment of the RNAi (RNA interference) pathway [21], which in turn leads to a decrease in luminescence. Assay reagent The luciferin assay reagent is based on the reagent reported by Siebring-van Olst and coworkers [22]. The components of the assay buffer are MgCO3, MgSO4, ATP, CoA, D-luciferin (LH2), DTT, and EDTA in Tricine buffer. Stock concentrations of the components were as follows: DTT, 500 mM; LH2, 50 mM; ATP, 50 mM; CoA, 1 mM; EDTA, 10 mM; Tricine, 200 mM; MgCO3, 10 mM; and MgSO4, 50 mM. The pH was set to 7.8 for Tricine and magnesium carbonate. The sensitive components ATP, CoA, LH2, and DTT were kept frozen, whereas the others were kept at room temperature. The first four components were then mixed at different concentrations as indicated in Table 1 to find the optimal composition. The final concentrations for Tricine, MgCO3, MgSO4, and EDTA were 20, 1, 5, and 1 mM, respectively. Results and discussion When the assay format is changed from 24 to 96 wells, more data points need to be read after each experiment. Working with a luminometer that lacks injectors has revealed a problem with the commercial reagent; it has the inherent property that the luminescence signal changes by up to 20% between the first and last measurements in one 96-well plate (the time that passes between two measurements in all figures corresponds to the time it takes to measure 96 wells, which is 3 min) and then rapidly loses approximately 70% of the signal within the first 30 min (see Fig. S1 in online Supplementary Material). This is disadvantageous for two reasons. First, the large difference between subsequent measures is a major source of uncertainty; small differences between samples would routinely be missed because the signal decreases faster than the instrument can measure. Second, in terms of multiplexing analysis, plates will need to be prepared and measured sequentially; luciferin reagent needs to be added to a plate, and that plate then immediately needs to be measured before the next plate can be treated. To counter the changing conditions, it is suggested to omit CoA from the luciferase reagent [22]. However, we find that in our experiments, although not having any CoA does indeed minimize the rate of change in the signal, the signal strength is also decreased by more than 10-fold (data not shown) compared with the assay reagent composition that we finally settled on. That
presents a different problem; there is a loss of sensitivity. As illustrated in Fig. 1, having the lowest amount of CoA also gives the lowest and most stable signal. CoA is a major effector of maximum initial signal, and its concentration needs to be controlled in order to get high persistent luminescence. Concentrations of CoA below 1 lM do not seem to affect signal strength, whereas higher concentrations of CoA scale with higher signal strength but also lead to a more rapid decrease in signal. DTT seems to be the main effector of the longevity of the signal; as can be seen in Fig. 1, the signals that decrease the fastest both contain the least amount of DTT, and the strongest signal in accordance with the above also has the most CoA. The reagent composition providing the most stable signal (80%). Reducing the size from 24-well plates to 96-well plates has changed how cells should be transfected with short oligonucleotides such as SCO and siRNA using CPPs; the 24-well assay requires transfer from 24-well plates to 96 wells for luminescence reading. This assay reduces the time needed from seeding of cells to attaining results by 33% while simultaneously reducing the number of pipetting steps from seven to three and reducing the cost by more than 95% and the waste by one microplate. In addition, this report establishes PF14 as a de facto positive control both for CPP-mediated delivery of splice-correcting oligonucleotides in the HeLa pLuc 705 cell line and for siRNA delivery in the U-87 MG-luc2 cell line. Acknowledgments This work was supported by the Swedish Research Foundation (VR-NT, VR-Med) and the Swedish Cancer Society.
Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.cej.2014.04.115. References [1] R. Brasseur, G. Divita, Happy birthday cell penetrating peptides: already 20 years, Biochim. Biophys. Acta 1798 (2010) 2177–2181. [2] M. Lindgren, M. Hällbrink, A. Prochiantz, Ü. Langel, Cell-penetrating peptides, Trends Pharmacol. Sci. 21 (2000) 99–103. [3] C. Rousselle, P. Clair, J. Temsamani, J.-M. Scherrmann, Improved brain delivery of benzylpenicillin with a peptide-vector-mediated strategy, J. Drug Target. 10 (2002) 309–315. [4] K.-L. Veiman, I. Mäger, K. Ezzat, H. Margus, T. Lehto, K. Langel, et al., PepFect14 peptide vector for efficient gene delivery in cell cultures, Mol. Pharm. 10 (2013) 199–210. [5] S. El-Andaloussi, T. Lehto, I. Mäger, K. Rosenthal-Aizman, I.I. Oprea, O.E. Simonson, et al., Design of a peptide-based vector, PepFect6, for efficient delivery of siRNA in cell culture and systemically in vivo, Nucleic Acids Res. 39 (2011) 3972–3987.
[6] K. Ezzat, S. El-Andaloussi, E.M. Zaghloul, T. Lehto, S. Lindberg, P.M.D. Moreno, et al., PepFect 14, a novel cell-penetrating peptide for oligonucleotide delivery in solution and as solid formulation, Nucleic Acids Res. 39 (2011) 5284–5298. [7] S. Futaki, T. Suzuki, W. Ohashi, T. Yagami, S. Tanaka, K. Ueda, et al., Argininerich peptides: an abundant source of membrane-permeable peptides having potential as carriers for intracellular protein delivery, J. Biol. Chem. 276 (2001) 5836–5840. [8] P. Boisguerin, S. Deshayes, M.J. Gait, L. O’Donovan, C. Godfrey, C.A. Betts, et al., Delivery of therapeutic oligonucleotides with cell penetrating peptides, Adv. Drug Deliv. Rev. (2015), http://dx.doi.org/10.1016/j. addr.2015.02.008. [9] T. Holm, H. Johansson, P. Lundberg, M. Pooga, M. Lindgren, Ü. Langel, Studying the uptake of cell-penetrating peptides, Nat. Protoc. 1 (2006) 1001–1005. [10] O. Shimomura, F.H. Johnson, Y. Saiga, Extraction, purification, and properties of aequorin, a bioluminescent protein from the luminous hydromedusan, Aequorea, J. Cell. Comp. Physiol. 59 (1962) 223–239. [11] R.Y. Tsien, The green fluorescent protein, Annu. Rev. Biochem. 67 (1998) 509– 544. [12] H. Fraga, Firefly luminescence: a historical perspective and recent developments, Photochem. Photobiol. Sci. 7 (2008) 146–158. [13] S. El-Andaloussi, P. Guterstam, Ü. Langel, Assessing the delivery efficacy and internalization route of cell-penetrating peptides, Nat. Protoc. 2 (2007) 2043– 2047. [14] S.H. Kang, M.J. Cho, R. Kole, Up-regulation of luciferase gene expression with antisense oligonucleotides: implications and applications in functional assay development, Biochemistry 37 (1998) 6235–6239. [15] S. Lindberg, A. Muñoz-Alarcón, H. Helmfors, D. Mosqueira, D. Gyllborg, O. Tudoran, et al., PepFect15, a novel endosomolytic cell-penetrating peptide for oligonucleotide delivery via scavenger receptors, Int. J. Pharm. 441 (2013) 242–247. [16] S. El-Andaloussi, H. Johansson, T. Holm, Ü. Langel, A novel cell-penetrating peptide, M918, for efficient delivery of proteins and peptide nucleic acids, Mol. Ther. 15 (2007) 1820–1826. [17] L. Crombez, G. Aldrian-Herrada, K. Konate, Q.N. Nguyen, G.K. McMaster, R. Brasseur, et al., A new potent secondary amphipathic cell-penetrating peptide for siRNA delivery into mammalian cells, Mol. Ther. 17 (2009) 95–103. [18] J. Regberg, A. Srimanee, M. Erlandsson, R. Sillard, D.A. Dobchev, M. Karelson, et al., Rational design of a series of novel amphipathic cell-penetrating peptides, Int. J. Pharm. 464 (2014) 111–116. [19] K. Langel, S. Lindberg, D. Copolovici, P. Arukuusk, R. Sillard, Ü. Langel, Novel fatty acid modifications of transportan 10, Int. J. Pept. Res. Ther. 16 (2010) 247–255. [20] J. Li, S.T. Crowley, J. Duskey, S. Khargharia, M. Wu, K.G. Rice, Miniaturization of gene transfection assays in 384- and 1536-well microplates, Anal. Biochem. 470 (2015) 14–21. [21] A. Fire, S. Xu, M.K. Montgomery, S.A. Kostas, S.E. Driver, C.C. Mello, Potent and specific genetic interference by double-stranded RNA in Caenorhabditis elegans, Nature 391 (1998) 806–811. [22] E. Siebring-van Olst, C. Vermeulen, R.X. de Menezes, M. Howell, E.F. Smit, V.W. Van Beusechem, Affordable luciferase reporter assay for cell-based highthroughput screening, J. Biomol. Screen. 18 (2013) 453–461. [23] K.A. Barriscale, S.A. O’Sullivan, T.V. McCarthy, A single secreted luciferasebased gene reporter assay, Anal. Biochem. 453 (2014) 44–49. [24] K. Ezzat, H. Helmfors, O. Tudoran, C. Juks, S. Lindberg, K. Padari, et al., Scavenger receptor-mediated uptake of cell-penetrating peptide nanocomplexes with oligonucleotides, FASEB J. 26 (2012) 1172–1180. [25] S. Lindberg, J. Regberg, J. Eriksson, H. Helmfors, A. Muñoz-Alarcón, A. Srimanee, et al., A convergent uptake route for peptide- and polymer-based nucleotide delivery systems, J. Control. Release 206 (2015) 58–66.
