Assessing the modulation of cells by modified mRNAs in vitro and in vivo

Stefan Wild1, Marion Jurk1, Yvonne Diener1, Christophe Beclin2, Rohollah Babaeikelishomi3, Alexandros Vegiopoulos3, Ute Bissels1, Olaf Hardt1, Sebastian Knoebel1, Harold Cremer2, and Andreas Bosio1 1Miltenyi Biotec GmbH, Bergisch Gladbach, Germany; 2IBDML, UMR7288 (CNRS/Aix-Marseille Université), Marseille, France; 3DKFZ Molecular Metabolic Control (A170), German Cancer Research Center, Heidelberg, Germany

Introduction 4 Efficient reprogramming of fibroblasts into iPS cells The transfer of genetic information to cells allows deciphering limited by the defense mechanisms of cells and potentially (A) Workflow for efficient reprogramming of human day 14 (12 consecutive ). Starting numbers the function of genes and as well as programming of genotoxic. In our study we aimed at exploring the versatility fibroblasts into iPS cells. After adaption of the cells to the were 7000, 4000, and 2000 in 6-well plates. At day 14, cells cells. Current methods for transient gain- or loss-of-function of RNA to i) extend DNA-free gain-of-function experiments to reprogramming medium, cells are transfected daily for 12 days were fixed and stained with an anti-human Oct-4 antibody. experiments and modification of genes by recombinases are cells that have been difficult to address in vitro and in vivo and with modified mRNAs coding for reprogramming (C) Alkaline phosphatase staining of iPS colonies of mainly based on transfer of DNA. However, delivery of DNA is ii) to increase the efficacy of recombination events. factors. Medium is replaced 4 h after and immune reprogrammed human newborn fibroblasts at day 14 in suppressor B18R is added to inhibit innate immune responses. 24-well plates (12 consecutive transfections). (D) Example of iPS cell colonies can be detected usually at day 14. (B) Oct-4 iPS colonies. staining of reprogrammed human newborn fibroblasts at Results A B 7000 Cells 4000 Cells C Workflow Coat plates with Start Stop iPSC colonies CTS™ CELLStart™ transfec- transfec- are ready and plate cells tions tions

Day -3 0 11 14 Generation of animal component–free, sterile, modified mRNAs Pre-culture cells in 2000 Cells 1 Pluriton™ Reprogram- Allow cells Daily transfection for Allow iPSC Pick, analyze, D Transmitted light ming Medium in tissue to adjust 12 consecutive days colonies to grow or sort iPSCs culture flasks (optional) We have set up a standardized workflow to generate mRNAs. Gene of interest Reprogramming Reprogramming Reprogramming StemMACS™ The cDNA of a gene of interest, optimized with regard to T7 RNA polymerase Capping medium medium + B18R medium iPS-Brew XF human codon usage, is cloned into a basic vector system G AAAA Figure 4 containing the T7 promotor and a 5’ and 3’ UTR. RNA is 3‘ UTR RNA Poladenylation generated in an animal component–free production process by T7-based transcription and subsequent 5’ capping and 3’ T7 promotor polyadenylation. Optionally, Ψ-UTP- and 5-mCTP-modified 5‘ UTR Transfection of iCre and FlpO mRNA yields very high recombination rates are introduced to reduce innate immune Basic transcription vector 5 responses against RNA in downstream applications. RNA is Figure 1 DNase treated, sterile filtered, and lyophilized. (A) We explored the recombination efficiency of iCre and FlpO TK-Neo STOP ΔhLNGFR- A IRES-EGFP mRNA when transfecting (by lipofection) or electroporating iCre recombination

mES cells carrying a GFP ON-OFF cassette (GAGG-Stop StemMACS™ mRNAs show a very fast, efficient, and non-toxic ΔhLNGFR-IRES-EGFP expression (flox)-eGFP(FRT)). (B) We observed a very high recombination FlpO recombination expression in vitro + genomic region IoxP site FRT site 2 efficiency in electroporated mES cells. The rate of GFP cells ROSA26 homology polyA site correlated with the amount of iCre mRNA, indicating the B 100 We have validated the performance of mRNA generated with (C) RNA transfection of cultured myoblasts. Satellite cells were possibility to control the degree of recombination. (C) The 80 60 the new process on different cell lines, primary cells, and isolated from mouse skeletal muscle using MACS Technology, recombination efficiency of up to 90% after 40 pluripotent stem cells. We first aimed at optimizing the based on the depletion of cells positive for lineage markers or lipofection with mRNA was about 10 times higher than after 20 0 transfection of primary human fibroblasts. To this end, we (CD31, CD45, CD11b, Sca-1). As a proof of concept, nuclear GFP electroporation with DNA. (D) Electroporation with FlpO 15 µg 7.5 µg 3 µg 1 µg 0.5 µg screened for reagents showing the highest transfection mRNA (100 ng per well) was transfected into myoblasts cultured for mRNA also resulted in an about 3-fold higher recombination C 100 proportion of eGFP+ cells efficiency and lowest toxicity. One particular reagent (now five days in 24-well plates, resulting in a transfection efficiency efficiency than electroporation with FlpO DNA. 80 (transfection control) 60 total recombination efficiency 40 called StemMACS™ mRNA Transfection Reagent) showed a higher than 75%. We tested up to four consecutive transfections (E) Efficient transfection and iCre recombination in primary recombination efficiency 20 within transfected cells transfection efficiency for fibroblasts of up to 90% (n=12) with and did not observe cell death or other negative effects. mesenchymal stem/stromal cells (MSC) from murine adipose 0 7.5 µg mRNA 3.75 µg DNA 1 µg mRNA (eGFP control) an average survival of 90–99%. tissue using nuclear GFP and iCre mRNA. Primary MSCs were electroporation lipofection

+ A GFP (P1/P2/UR3) * + 0.6 µL Reagent B isolated from collagenase-digested murine subcutaneous fat, 100 + (A) RNA transfection of fibroblasts. 2.5×104 human fibroblasts FITC-A Mean (P1/P2) D proportion of mCherry cells 100 100 80 (transfection control) 60 (FITC Mean of live cells) using a two-step procedure involving MACS Technology, of different origins were plated (24-well plate, 500 µL/well) 90 60 recombination efficiency

50 GFP expression 80 – – – + overnight and transfected the next day with 0.6 µL of StemMACS 80 resulting in TER119 CD31 CD45 Sca-1 cells. The cultured cells 40 recombination efficiency 70 HSCs 40 + 60 within transfected cells cells 60 20 + were transfected with 250 ng of nuclear GFP or iCre mRNA. mRNA Transfection Reagent and with indicated amounts of 50 30 40 0 (mCherry control) 40 7.5 µg mRNA 7.5 µg DNA 20 20 GFP % of StemMACS eGFP mRNA. GFP expression was determined 24 h % GFP 30 Three days following transfection, iCre-mediated recombination 20 10 0 later on the MACSQuant® Analyzer using the MFI of viable cells 10 1 3 5 7 was assessed by measuring expression of the floxed gene. 0 0 80 1.2 and the percentage of GFP-positive cells among viable cells. Day after electropration Relative mRNA expression was determined by qRT-PCR. E nGFP mRNA nGFP 60 Control iCre Transfection efficiency and GFP expression depended on the 0.8 Medium Medium Medium Medium Medium 40 90.12% M1 levels 0.4 * ratio of StemMACS Transfection Reagent and RNA. Overall cell number 20 100 ng mRNA* ng 100 mRNA* ng 100 mRNA* ng 100 mRNA* ng 100 mRNA* ng 100 200 ng mRNA* ng 200 mRNA* ng 200 mRNA* ng 200 mRNA* ng 200 mRNA* ng 200 400 mRNA* ng 400 mRNA* ng 400 mRNA* ng 400 mRNA* ng 400 mRNA* ng Relative cell viability was above 95% for all cells and ratios tested. newborn newborn adult dermal adult dermal adult dermal 0 0 1 2 3 4 mRNA Relative 0 foreskin foreskin fibroblasts 1 fibroblasts 2 fibroblasts 3 0 10 10 10 10 Gene1 (floxed) (B) RNA electroporation of hematopoietic stem cells. CD133+ fibroblasts 1 fibroblasts 2 GFP cells were isolated from cord blood by magnetic cell separation Figure 5 (MACS® Technology), cultured overnight, and subsequently electroporated with GFP mRNA (2.4–8 µg per sample). 24 h after C nGFP Transmitted light nGFP/Transmitted light electroporation 97+/–2.9% of the cells were GFP+ (n=10), and mRNA-based expression of transcription factors the overall viability was 45+/–17.8%. Viability of cells 6 electroporated with mRNA relative to mock-electroporated cells was about 100%, showing that the mRNA is not toxic (data We have set up and validated by immunocytochemistry a Primary fibroblasts were seeded in 24-well or 96-well plates not shown). After electroporation of cells with 8 µg GFP mRNA, number of mRNAs coding for transcription factors known to and transfected with 50 to 83 ng mRNA per 10,000 cells. These GFP expression persisted for at least seven days, indicating a modulate cell fate towards neural (Ascl1 (Mash1), Lmx1a, Nurr1 will be used to explore the RNA technology for directed Figure 2 high stability of GFP expression (mean+/–SD; n=3). (Nr4a2), Myt1l, NeuroG2, Pou3F2, NeuroD1), cardiac (Gata4, conversion of fibroblasts. Mef2c, Mesp1) and brown adipocyte (Cebpb, Pparg) lineages. 3 In vivo electroporation with mRNA is more efficient than with DNA Ascl1 (Mash1) Lmx1a Nurr1 (NR4a2) Myt1l NeuroG2 Pou3f2

(A) Efficiency of in vivo electroporation of the mouse postnatal A forebrain with mRNA or DNA was analyzed by in vivo analysis NeuroD1 Cebpb Mesp1 of GFP expression. DNA- or mRNA-containing solutions were Pparg Gata4 Mef2c directly injected into the lateral ventricles. After application of electrical pulses, neural stem cells (NSC) in the subventricular – + zone (SVZ) were transfected. (B) Seven days after Anesthetized pup (P0) Injection: Pcx-EGFP-N1 Electrical pulses or EGFP mRNA 100 V electroporation with a DNA construct, GFP was expressed in (5 µg each) Figure 6 all derivatives of NSC – from the SVZ to the olfactory bulb (OB). (C) Compared to electroporation with 5 µg cDNA B C D (Pcx-EGFP-N1), electroporation with the same amount of GFP- Plasmid mRNA OB RNA led to a considerably stronger GFP expression in more SVZ LV LV NSC lining the ventricular wall. (D) Left: In vivo electroporation RMS striatum Conclusion with RNA of the ventricular wall at p5 allowed efficient striatum expression of GFP in ependymocytes, a cell population that is refractory to DNA transfection. Right: transfection of GFP in • We developed a workflow for the animal component–free • Transfection of a variety of transgenic cells with iCre or FlpO ependymocytes validated by S100-beta staining. Figure 3 generation of sterile, modified mRNAs. showed a much higher recombination rate than observed • Modified mRNAs showed a very fast, efficient, and harmless with iCre or FlpO DNA. expression in vitro and in vivo in a variety of primary cell • A number of mRNAs coding for transcription factors known types, exceeding expression rates observed with DNA to modulate cell fate towards different lineages could be transfection. generated and expressed in primary fibroblasts. • Consecutive transfections with reprogramming mRNAs efficiently converted newborn and adult patient fibroblasts into iPS cells.

Unless otherwise specifically indicated, Miltenyi Biotec products and services are for research use only and not for therapeutic or diagnostic use. MACS, StemMACS, and MACSQuant are registered trademarks or trademarks of Miltenyi Biotec GmbH. All other trademarks mentioned in this document are the property of their respective owners and are used for identification purposes only. Copyright © 2014 Miltenyi Biotec. All rights reserved.