[0001] The present invention concerns a nucleotide-modified messenger RNA for the permanent correction of a genetic alteration on a DNA. The invention further concerns a nucleotide-modified messenger RNA in combination with a repair template. It finally concerns a pharmaceutical composition.

[0002] The gene therapy refers to the insertion of nucleic acids such as DNA or RNA into somatic cells of an individual, e.g. in order to treat a disease. By doing so, usually an intact gene should be inserted into the genome of the target cell in order to replace a defective gene which is causally related to the development of the disease. Basically a gene therapy has chances of success only for such diseases which are based on the alteration of only one or a few number of genes.

[0003] Surfactant protein B deficiency and cystic fibrosis (CF) are severe, congenital, fatal diseases for which currently no satisfying therapies do exist. Surfactant protein B deficiency is rare and occurs in about one out of one million newborns. Surfactant protein B (SP-B) is a pulmonary surfactant associated protein that plays an essential role in the alveolar stability by lowering the surface tension at the air-liquid interface in the lung. Mutations of the SP-B encoding gene (SFTPB) results in a rapidly fatal respiratory failure associated with alveolar proteinosis, within the first year of life. The cystic fibrosis is the most prevalent life-limiting autosomal-recessive disease in Caucasian populations. It can be found in one out of 2,500 newborns and affects more than 70,000 people worldwide. Mutations in the gene coding for the "cystic fibrosis transmembrane conductance regulator" (CFTR), a chloride channel, result in an impaired anion secretion and hyper- absorption of sodium across epithelia. The chronic lung disease is the major factor contributing to mortality and morbidity in CF patients. Even with the current therapy, the mean survival is only between 30 and 40 years.

[0004] Current gene therapeutic efforts with DNA-based or viral vectors remain largely unsuccessful in treating these fatal illnesses. As the airways evolved in direct contact to the environment, their inherent defense mechanisms present a significant barrier for the delivery of foreign vectors into the lung. In addition, the use of most of the viral vectors presents a health risk since they can act as oncogenes.

[0005] Kormann et al. (201 1 ), Expression of therapeutic proteins after delivery of chemically modified mRNA in mice, Letters to Nature Biotechnology, pages 1 -6, describe a therapeutic approach for the treatment of SP-B deficiency where a functional nucleotide modified messenger RNA (mRNA) encoding SP-B is introduced into alveolar cells of the mouse. This is affected by intratracheal high pressure application of the SP-B mRNA. Through the nucleotide modification 25 % of the uridine and cytidine were replaced by 2-thiouridine (s2U) and 5-methylcytidine (m5C), respectively. As a result the modified mRNA has less immunogenicity and is more stable than its unmodified counterpart. However, this approach has the disadvantage that the nucleotide modified mRNA can compensate the SP-B deficiency only for a limited time period, i.e. until it will be degraded by RNases so that a detectable effect will soon disappear. A permanent gene supplementation or gene correction cannot be realized.

[0006] McCaffrey et al. (2013), Targeted genome engineering with zinc-finger nucleases, TALENs and CRISPR. The buzz on the cut: from dream to reality, internet article posted on July 1 , 2013, in Therapeutics and tagged gene engeneering, genome editing, mRNA [http://zon.trilinkbiotech.com/2013/07/01/the-buzz-on-thecut -from -dream- to-reality/], suggests the use of the ZFN and TALEN nucleases encoded by synthetic mRNAs for transient expression in genome engineering. He mentions that mRNAs could be made less immunogenic and non-toxic by substitution of cytosine and uridine with 5- methylcytosine and pseudouridine.

[0007] US 2013/01 17870A1 discloses the use of mRNA encoding the TALEN nuclease for producing genetically modified or transgenic animals, respectively. It also discloses the transfection of swine fibroblasts with TALEN encloding nucleotide-modified mRNA. [0008] Against this background it is an object of the present invention to provide a new substance which can be used as a tool within the framework of the gene therapy. With this substance the preconditions should be established for a permanent correction of a genetic alteration on a DNA.

[0009] This measure is realized by the provision of a nuclease encoding nu- cleotide-modified messenger RNA (nec-mRNA).

[0010] The inventors have surprisingly recognized that by the gene therapeutically use of a nec-mRNA the preconditions are established to correct a genetic alteration on the DNA in a permanent manner. For that purpose the nec-mRNA is transfected into the cytoplasm of a target cell and will there be translated into a nuclease. The nuclease is then transported into the nucleus. In the nucleus it can bind to the DNA which comprises the genetic alteration and can initiate a double-strand break (DSB). The DSB as a repair mechanism stimulates a homologous recombination, thus establishing the precondition for an exchange of the genetic alteration against e.g. the wild type sequence of the corresponding DNA section. Here the temporarily existing nuclease activity is advantageous, especially in contrast to any DNA or virus encoded nuclease activities where there is the risk of an integration into the genome of the host, and gives additional therapeutic safety for the system according to the invention.

[0011] This finding was surprising. In the art so far nucleotide-modified mRNA is mostly used for the direct substitution of the deficient gene or protein, respectively. For example Kormann et al. (cit. loc.) describe a nucleotide-modified mRNA which encodes the red fluorescent protein (RFP), the mouse erythropoietin (mEpo) or the surfactant protein B. The WO 201 1/012316 also describes nucleotide-modified mRNA which encodes the surfactant protein B.

[0012] Kariko et.al. (2012), Increased Erythropoiesis in Mice Injected With Submicrogram Quantities of Pseudouridine-containing mRNA Encoding Erythropoietin, Molecular Therapy, Vol. 16, No. 1 1 , pages 1833 - 1844, describe the use of nucleotide- modified mRNA for the synthesis of erythropoietin in a mouse model and propose the therapeutic use of nucleotide-modified mRNA.

[0013] The use of nec-mRNA as a molecular tool for establishing a permanent gene correction is not described in the prior art.

[0014] According to the invention "nucleotide-modified messenger RNA" refers to such an mRNA, where a part of the nucleotides, or nucleosides or nucleobases is modified, i.e. changed. In this respect the terms "nucleotides" and "nucleosides" are used interchangeably. Preferably it is referred to a chemical modification. This modification has the result that the mRNA is more stable and has less immunogenicity. Nucleotide- modified messenger RNA is generally known in the prior art, cf. for example WO

201 1/012316. The content of the before-mentioned publication is incorporated herein by reference. Examples for chemically-modified nucleotides or nucleosides are pseudouri- dine (Ψ), 5-methylcytidine (m5C), N6-methyladenosine (m6A), 5-methyluridine (m5U) or 2-thiouridine (s2U).

[0015] According to the invention the use of a nec-mRNA also encompasses the use of different nec-mRNAs, such as a pair of nec-mRNAs, where each nec-mRNA could encode different nucleases. For example, one nuclease might bind and cleave upstream and the other nuclease downstream of the genetic alteration and, in doing so, create the optimum preconditions for a homologous recombination.

[0016] A "genetic alteration" refers to any change of the sequence on the DNA in comparison to the wild type, e.g. caused by a mutation or a polymorphism. Preferably the genetic alteration can be found in a gene resulting in a loss of function of the encoded protein or even in a complete knockout.

[0017] A "correction" of a genetic alteration on the DNA or a "gene correction" refers to a permanent exchange of the genetic alteration or the genetically altered gene for a nucleic acid section without such alteration, for example of the wild type. [0018] The problem underlying the invention is herewith completely solved.

[0019] According to a preferred embodiment up to including approx. 100 % of the uridine nucleotides and/or up to including approx. 100 % of the cytidine nucleotides, preferably up to including approx. 70 % of the uridine nucleotides and/or up to including approx. 70 % of the cytidine nucleotides, further preferably up to including approx. 50 % of the uridine nucleotides and/or up to including approx. 50 % of the cytidine nucleotides, further preferably up to including approx. 25 % of the uridine nucleotides and/or up to including approx. 25 % of the cytidine nucleotides, and highly preferably approx. 10 % of the uridine nucleotides and/or approx. 10 % of the cytidine nucleotides of the nec-mRNA are modified, further preferably by exchanging uridine for 2-thiouridine (s2U) and/or pseudouridine (Ψ) and/or by exchanging cytidine for 5-methylcytidine (m5C).

[0020] This measure has the advantage that through the prescribed content of nucleotide modifications an mRNA is provided which is significantly stable and little immunogenic. Even more, the inventors could surprisingly realize that it is sufficient if only up to including about 10 % of the cytidines are replaced by 5-methylcytidine (m5C) and/or up to including approx. 10 % of the uridines are replaced by 2-thiouridine (s2U). The inventors could provide evidence that also such slightly modified nec-mRNA is stable and little immunogenic. Since the nucleotide modification is complex, this has the advantage that the nec-mRNA according to the invention, because of the low concentration of nucleotide modifications, can be produced in a cost-saving manner. Besides of reducing costs, the lowering of the portion of modified nucleotides has also the advantage that the efficiency of the translation is increased. This is because very high portions of specifically modified nucleotides, such as 2-thiouridine (s2U), significantly interfere with the translation of the modified mRNA. However, with lower portions an optimum translation can be observed.

[0021] According to a preferred embodiment the genetic alteration is located in a lung protein, preferably in a surfactant protein, further preferably in the surfactant protein B (SP-B), further preferably a receptor protein including cystic fibrosis transmembrane and conductance regulator (CFTR), further preferably a transcription factor including Foxp3. [0022] This measure has the advantage that the invention can be used as a therapeutic tool in the therapy of lung diseases for which currently no satisfying therapies do exist.

[0023] According to a preferred embodiment the nec-mRNA encodes a nuclease which is configured to bind to the DNA upstream and/or downstream of the genetic alteration.

[0024] This measure has the advantage that the encoded nuclease binds next to the genetic alteration, there catalyzes a double-strand break (DSB), thereby initiate cellular repair mechanisms including a homologous recombination (HR). In this way only the genetic alteration can be replaced in a targeted manner, e.g. by the wild type.

[0025] The nuclease is preferably selected from the group consisting of: zinc- finger nucleases (ZFNs), transcription activator-like effector nucleases (TALEN),

CRISPR/Cas9, and dimeric CRISPR RNA guided Fokl nucleases.

[0026] Zinc-finger nucleases (ZFNs) and transcription activator-like effector nucleases (TALEN) are artificial endonucleases which bind to specific sequences upstream and/or downstream of the genetic alteration on the DNA via DNA-binding polypeptides. The targeted configuration, the structure and the functionality of these nucleases are known to the person skilled in the art. Reference is made in this connection to the document of Carlson et al. (2012), Targeting DNA with fingers and TALENs, Molecular Therapy-Nucleic Acids 1 , e3. "Cas" stands for "CRISPR associated". CRISPR stands for "Clustered Regularly Interspaced Short Palindromic Repeats". Cas9 is a nuclease which has originally discovered in bacteria, which binds in a targeted manner to distinct sections of the DNA via the CRISP/Cas system by means of a short complementary single- stranded RNA; cf. Mail et al. (2013), RNA-guided human genome engineering via Cas9, Science 339(6121 ), pages 823-826 and Cong et al. (2013), Multiplex genome engineering using CRISPR/Cas systems, Science 339(6121 ), pages 819-823; and Tsai et al. (2014), Dimeric CRISPR RNA-guided Fokl nucleases for highly specific genome editing, Nature Biotechnology 32, p. 569-576. These publications are incorporated herein by reference. [0027] According to a preferred embodiment of the invention the nec-mRNA is coupled to an aptamer.

[0028] This measure has the advantage that the binding site of the nec-mRNA can be adjusted to defined target cells by a sequence-specific design of the aptamer. In this way it can be ensured that the nec-mRNA corrects the genetic alteration in a targeted manner only in specific cells where it is necessary. If, for example, the correction of a lung protein is intended, such aptamers are coupled to the nec-mRNA which selectively bind to cells of the lung tissue. Cells which are not affected by the genetic alteration remain untouched. This measure results in additional therapeutic safety. By the so-called SELEX method such aptamer sequences can be enriched which bind to the desired cell or the desired cell membrane structure, respectively.

[0029] According to a preferred embodiment the nec-mRNA is packed into nanoparticles.

[0030] This measure has the advantage that the absorption of the nec-mRNA into the cell is significantly improved, in particular into cells of the lung tissue. In particular, nanoparticle associated or packed nec-mRNA can be administered intravenously (i. v.) and still can reach the lung cells. Such a lung-cell targeted i.v. administration probably won't be possible without using nanoparticles. Examples for appropriate nanoparticles are the lipid GL67/DOPE and biocompatible chitosan-coated nanoparticles. In this connection "nanoparticle" refers to a particle between approx. 1 and approx. 300 nanometers in size (hydrodynamic diameter), preferably between approx. 50 nm and approx. 250 nm, further preferably between approx. 75 nm and approx. 200 nm, further preferably between approx 100 nm and approx. 175 nm, and highly preferably between approx. 150 nm and approx. 160 nm.

[0031 ] According to a preferred embodiment the nanoparticle is coated with chi- tosan. [0032] This measure has the advantage that the respirability is further increased. In addition, chitosan has been proven as being particularly biocompatible resulting in an increase of the tolerance of the nec-mRNA by a living being.

[0033] Another subject matter of the present invention is the nec-mRNA according to the invention in combination with a repair template.

[0034] A "repair template" refers to a nucleic acid molecule, such as a DNA fragment, which comprises a nucleotide section which should be exchanged by homologous recombination (HR) for the section on the DNA comprising the genetic alteration. For example, this nucleotide section corresponds to the wild type or the "healthy gene", respectively, which does not comprise the genetic alteration. Upstream and downstream of the genetic alteration the repair template comprises sections which are significantly homologous to the DNA that a hybridization and homologous recombination can take place after the nuclease has induced a double-strand break.

[0035] The combination according to the invention of the nec-mRNA and the repair template allows a permanent correction of the genetic alteration by a lifelong expression of the corrected protein. As a consequence, nec-mRNA and repair template are in this way a "gene correction set" according to the invention.

[0036] It goes without saying that the repair template can comprise an inducible promoter by means of which the expression of the repaired gene can be controlled in a targeted manner.

[0037] According to a preferred embodiment of the invention the repair template is packed into an adeno-associated viral vector (AAV), and/or is encoded by a plasmid DNA, and/or is packed into a lentiviral vector, and/or is packed into a protein-capped adenoviral vector (AdV).

[0038] This measure has the advantage that it is made use of an established principle of the introduction of genetic information into the cell. AAV, lentiviral and AdV vectors have been proven successful in the practice of gene transfer because of the absence of gene toxic side effects.

[0039] Another subject matter of the present invention is a pharmaceutical composition comprising a nec-mRNA, preferably the before-mentioned nec-mRNA according to the invention, and further preferably in addition a repair template. Furthermore, the pharmaceutical composition is provided for the treatment of a lung disease which might be surfactant protein B deficiency and/or cystic fibrosis (CF) and/or Asthma and/or chronic obstructive pulmonary disease (COPD).

[0040] The features, characteristics and advantages of the nec-mRNA according to the invention also apply to the composition according to the invention.

[0041] Also disclosed is a method for the correction of a genetic alteration on a DNA comprising the following steps: (1 ) introducing a repair template into a DNA- containing cell, which comprises the genetic alteration to be corrected, (2) introducing a nec-mRNA into the cell. The cell can preferably be a lung cell and the introduction is preferably realized by means of high pressure application of the repair template and the nec-mRNA into the lung.

[0042] Both, the repair template, preferably packed into a AAV vector, as well as the nec-mRNA, preferably packed into nanoparticles, can be administered systemically or intravenously, respectively. This is of particular advantage if an application into the respiratory tract is not possible because of a mucus obstruction of the lung.

[0043] Another subject-matter of the present invention relates to a method for the correction of a genetic alteration on a DNA comprising the following steps: (1 ) introducing a repair template into a living being having a genetically altered DNA to be corrected, (2) introducing a nec-mRNA into the living being. The living being can be preferably a human being, and the introduction is preferably realized by means of high pressure application of the repair template and the nec-mRNA into the lung of the living being. [0044] The features, advantages, further developments, embodiments disclosed in relation to the nec-mRNA, combination and pharmaceutical composition according to the invention apply to the methods according to the invention in equal measure.

[0045] It goes without saying that the before-mentioned features and those to be explained in the following embodiments cannot only be used in the combination specifically indicated but also in other combinations or in isolated manner without departing from the scope of the invention.

[0046] The invention will now be explained on the basis of embodiments resulting in further advantages, characteristics and features.

[0047] Reference is made to the enclosed figures which show the following:

Fig. 1 shows the optimizing of the nec-mRNA by varying the portion of modified nucleotides;

Fig. 2 shows the principle of the permanent gene correction by the use of nec- mRNA;

Fig. 3 shows the in vivo gene correction at the SP-B locus of a SP-B knockout mouse by means of a nec-mRNA and the resulting increase of the life span of the mouse; and

Fig. 4 shows the homology directed repair of the SP-B locus in mouse fibroblasts in vitro by TALEN, encoded by modified mRNA.

Fig. 5 SP-B expression in lung BALF. The difference in SP-B expression levels per μg total protein in BALF between transgenic SP-B mice on Doxycycline and BALB/c mice was not significant (n.s.). Boxes represent medians ± IQRs (interquartile ranges), n = 4 mice per group. Fig. 6 nec-mRNA cleaves the SP-B cassette, induces HDR in vitro, and is expressed in lung cells in vivo, (a) TALEN and ZFN candidates relative to the transgenic SP-B cassette. Transgenic SP-B mice-derived fibroblasts were used for (b-d); n.d., not detectable, (b) T7 assays to determine the frequency of TALEN- and ZFN-induced indels in genomic DNA harvested 3d post-transfection (5 g/ce\\). (c) T1 - and Z3-induced indels following delivery as either mRNA or pDNA (0.5 or 5μg). (d) Percent HDR 3d following co-transfection of 5μg T1 or Z3 mRNA (or pDNA) with O^g donor plasmid. Arrows denote /V/?el-sensitive cleavage products resulting from HDR. (e) Time-course showing kinetics and stability of 3xFLAG-tagged Z3 mRNA versus Z3 AAV in A549 cells (n=3). (f) An- ti-3xFLAG flow cytometry shows protein expression in total lung cells and ATM cells. Boxes, medians ± IQRs; whiskers, minimum and maximum; ^* , P < 0.05 versus unmodified; ^** and ^*** , P < 0.01 and P < 0.001 versus no NPs. (g) Immunostaining for 3xFLAG in lung sections from mice described in f. Scale bar, δθμηη. Arrows indicate 3xFLAG expression.

Fig. 7 Expression of FLAG-tagged TALENs and ZFNs in MLE12 cells, a,

MLE12 cells (murine ATM cells) were transfected with 1 μg of each TALEN and ZFN plasmid (left bar = left TALEN/ZFN, right bar = right TALEN/ZFN) or untransfected (control). After 24 h the expression was determined by flow cytometry. Transfection efficiency (% 3xFLAG expression, left y axis) and median fluorescence intensity (right axis, blue lines) from three pooled samples each are shown, b, Representative FACS dot plots of MLE12 cells expressing 3xFLAG. MLE12 cells were transfected with plasmids encoding for TALEN (T) 1 , 2, 3, and ZFN (Z) 2, 3, 4 and 5. L, left arm. R, right arm. All assays were performed in biological triplicates.

Fig. 8 Schematic of the T7 assay to prove cleavage of the transgenic SP- B promoter region. Fibroblasts from transgenic SP-B mice were transfected with TALEN or ZFN plasmid pairs (5 μg) (1 .), leading to cellular repair, with NHEJ leading to approximated 1 -5% indels (insertions and / or deletions) (2.). Genomic DNA was harvested 4 d after transfection and a locus-specific PCR was performed (3.)- PCR products were melted at 95°C, re-annealed at gradually decreasing temperatures (4.) and treated with T7 endonuclease (5.)- T7 endonuclease only cuts heterodi- mers at sites of mismatch, resulting in smaller fragments which then can be visualized on agarose gels (6.) to determine the frequency of nucle- ase-induced indels in the samples.

Fig. 9 Z3 pair amino acid sequence. Amino acid sequence of FLAG-tagged

a, left Z3 arm (SEQ ID no. 35) and b, right Z3 arm (SEQ ID no. 36).

Fig. 10 Z3 nec-mRNA is efficiently deposited in the lung in vivo. 10ΟμΙ PBS,

2C^g unmodified Z3 mRNA, or 2C^g Z3 nec-mRNA containing a 5' 3xFLAG tag (naked or complexed with nanoparticles, 10ΟμΙ total volume) were i.t. administered to BALB/c mice (n = 5 mice per group). After 24h, total RNA was extracted from lungs, reverse transcribed, and Z3 mRNA was quantified by qPCR. Mean+SD ist shown. ^*** , P < 0.001 versus "unmod. Z3 mRNA naked"; ^§ , P < 0.05 versus "unmod. Z3 mRNA + NP".

Fig. 1 1 3xFLAG+ Clara cells 24h after i.t. administration of PBS or unmodified or differently modified Z3 nec-mRNA, with or without NPs. Z3 protein expression was quantified viai flow cytometry against 3xFLAG (n = 5 mice per group). Percentages of 3xFLAG ⁺ Clara cells ^* , P < 0.05 versus unmodified mRNA; ^** , P < 0.01 versus "without NP". Boxes represent medians ± IQRs. Whiskers represent the minimum and maximum observations.

Fig. 12 In vivo immune reaction to nec-mRNA. 1 μg of Z3 mRNA panel i.v. or i.p. injected into mice (n = 3 per group). 6h and 24h post-injection, IFN-a was measured by ELISA in duplicates. Relative mRNA deposition amounts were determined by RT-qPCR of isolated lung tissue. ^* , P < 0.05 versus "IMP only".

Rescue of SP-B deficient mice by in vivo gene manipulation, (a)

Treatment scheme and Kaplan-Meier survival curves of transgenic SP-B mice treated i.t. with donor (2.5x10 ¹¹ v.g. AAV6-donor, AAV6-mock, or none) and nuclease (20 g Z3 nec-mRNA-NP, mock-mRNA-NP, 5x10 ¹⁰ v.g. Z3 AAV, or none), then withdrawn from doxycycline. Groups C-F, n=6; groups A and B, n=13, reduced to n=4 20d post-doxycycline removal. Log-rank tests were performed. (b,c) Representative SP-B expression (brown) in lung tissue (c) and anti-SP-B blots on cell-free BALF supernatant (10μg total protein/lane) (c) from mice described in a. Scale bar, δθμηη. Lavages and tissue were harvested 20 days after doxycycline removal, (d) Lung compliance normalized to respective body weight (n=4), 20d after doxycycline removal. Baseline measurement performed for 20min; values calculated prior to each hyperinflation. ^*** , P < 0.001 versus groups C-F; §, §§, and §§§, P < 0.05, P < 0.01 , and P < 0.001 versus group D. (e,f) PCR on lung-isolated DNA from groups A and B or untargeted lungs; each lane represents an individual mouse. Samples were taken 20d after doxycycline removal, (e) PCR of the targeted locus followed by T7 assays. Arrows show expected bands, (f) PCR using P1/P3 or P1/P2, followed by gel electrophoresis. #, untargeted control; §, DNA pool of groups A and B. Arrow indicates band resulting from HDR. (g) Schematic of the transgenic SP-B cassette, CAG integration and primer (P1 , P2 and P3) locations for in-out PCRs. (h) Representative immunohistochemistry for groups A, B, and a doxycycline- control group (+Doxy) using two different anti-3xFLAG antibodies. Scale bar, δθμηη. Tissue was collected 20d after doxycycline removal.

Fig. 14 Structures of TUB07-pFB-ZFN3-repair-template (A), TUB09-pFB-CMV- 3Flag-NLS-38561 -Fok-KKR-bGHpA (B), and TUB08-pFB-CMV-3Flag- NLS-38558-Fok-ELD-bGHpA (C). Fig. 15 Ex vivo transgene integration. Fibroblasts, derived from transgenic SP-B mice, were transduced with 2.5 x 10 ⁵ v.g. of AAV6-donor and either a mock control ("untargeted"), 2 μg Z3 nec-mRNA ("mRNA targeted) or 1x10 ⁵ v.g. AAV6-Z3 ("AAV targeted"). L, ladder. Lanes marked with "-" are the respective no-template negative controls. Given are the expected amplicon sizes.

Fig. 16 SP-B expression in mouse experimental groups A-F as measured by Western blot. SP-B expression in ng^g total protein was determined by Quantity One software (www.bio-rad.de). Boxes represent medians ± IQRs (interquartile ranges). Whiskers represent the minimum and maximum observations, n = 6 mice per group were used.

Fig. 17 Semiquantitative analysis of the immunohistochemistry shown in

Fig. 2b). Boxes represent medians ± IQRs (interquartile range), n = 6 mice per group were used.

Fig. 18 % Resistance of mouse lungs after gene manipulation. Calculated by dividing the mean resistance values at the end of challenge by mean values at end of each washout period of the +Doxy group and main groups A and B, challenged with methacholine in rising concentrations over time to determine airway hyper responsiveness, n = 3 mice per group were used.

Fig. 19 Representative photographs of the lungs from groups +Doxy, A and B, before and after perfusion and lung function measurements. n = 3 mice per group were examined.

Fig. 20 Representative photographs of the lungs from groups C to F, before and after perfusion and lung function measurements, n = 3 mice per group were examined. Fig. 21 Comparison of hemorrhagic counts (semiquantitive analysis of data from Supplementary Fig. 14 and 15). If left lung showed partial hemorrhage it was counted as 1 , when more than half of the left lung area was hemorrhagic it was counted as 2. For all four right lung lobes, signs of hemorrhage were counted as 1 (resulting in an maximum count of 6). The straight lines represent the means, a, hemorrhage count before perfusion; b, hemorrhage count after perfusion. ^* , P < 0.05 versus Doxy-control, groups A and B; ^§ , P < 0.05 versus Doxy-control and group B. (Mann-Whitney test, two-sided, asymptotic significance).

Fig. 22 Differential cell counts. Cells from lung lavages were stained with May- Grijnwald/Giemsa, counted and related to 1 ml of BALF, 20d after doxycycline removal.

Fig. 23 IL-12 ELISA in BALF. Cytokine levels were quantified in mice BALF by

ELISA at sacrificing date (mean ± s.e.m); n.s., not significant. Serum was tested 20 days after doxycycline removal.

Fig. 24 3xFLAG expression score (combined semiquantitative analysis of the immunohistochemistry shown in Fig. 2h). Boxes represent medians ± IQRs (interquartile range). ^* , P < 0.05 versus Doxy-control and group A. The lavage was harvested 20d after doxycycline removal.

Fig. 25 Expression of 3xFLAG+ in a) total lung cells and b) ATM cells. ^* , P <

0.05 versus Doxy-control; ^** , P < 0.01 versus Doxy-control. The tissue was harvested 20d after doxycycline removal.

Fig. 26 Target site sequencing. We pooled sorted AT II cell samples within different experimental groups (A-F), one pool per group, performed single- cell separation, cloned PCR amplicons of the DNA (P1/P2) from those single cells in TOPO vectors, and sequenced the amplicons with primers P1 and P2. Subsequently, we performed an alignment of the sequences with the donor reference sequence, thereby identifying corrected cells (lower lanes depicting part of intron 1 (named "rev 3. NBT1 P1..." and "fwd 4. NBT1 P1...").

Fig. 27 ln-out PCRs and T7 assays on DNA samples from donor only treated mice (group C). The non-appearance of any secondary band(s) demonstrates that no Tl accidentally took place in the donor only group C. L, ladder. Lanes marked with "-" are the respective non-template negative controls.

Embodiments

1. Optimizing the nec-mRNA

[0048] In Kormann et al. (201 1 ; cit. loc.) it is described that the replacement of 25 % of each uridine and cytidine in the mRNA by 2-thiouridine and 5-methylcytidine in the SP-B deficient mouse results in a significantly stable and low immunogenic SP-B mRNA. The inventors have tested in an experiment whether the portion of modified nucleotides can be further reduced. The result of such an experiment is shown in Fig. 1.

[0049] In a first approach the inventors have manufactured an mRNA encoding the red fluorescent protein (RFP), where different levels of uridine and cytidine were replaced by 2-thiouridine (s2U) and 5-methylcytidine (m5C), namely 25 % of each, 10 % of each, and 100 % of m5C and 10 % of s2U. With these RNA molecules A-549 cells were transfected and after 24 hours the median of the fluorescence intensity (MIT) as the size of the transfection efficiency and a positive expression were measured. The result is shown in Fig. 1A. It can be seen that with a substitution of each of 25 % or 10 % an optimum expression is detectable; cf. 4th and 5th column.

[0050] In a further approach the immunogenicity of the modified mRNA molecules was examined. For this purpose, in addition to the s2U and m5C modified mRNAs also pseudouridine (Psi) modified mRNA molecules were manufactured which all encode zinc-finger nucleases (ZFN-5; both directions). With the chemically-modified mRNAs PBMCs were transfected via liposome fusion (Lipofectamin-2000). In the following, in an ELISA the expression of IFN-alpha was measured as a rate for the immunogenicity. The result is shown in Fig. 1 B. Here it becomes evident that the use of an unmodified mRNA results in a very strong immunoreaction (column far right). When using 10 % of each m5C/s2U (6th approach) and 25 % of each m5C/s2U (5th approach), the immunogenicity is significantly reduced.

The low replacement of modified nucleotides has the advantage that the nec-mRNA is producible in a significantly cheaper manner than the nucleotide-modified mRNAs of the prior art where considerably higher portions of the nucleotides are replaced. It has further the advantage that the efficiency of the translation is optimized.

2. Principle of the permanent gene correction by the use of nec-mRNA

[0051] The inventors have developed a system to achieve a permanent correction of the gene loci which are present in mutated form in several diseases such as severe congenital lung diseases, in order to allow a stable lifelong expression of the corrected protein. This system is shown in Fig. 2. A lung efficient AAV vector (1 .) shuttles the repair template into the cell and the nucleus (2.). Subsequently, a modified, aptamer-coupled mRNA encoding a specific nuclease pair (3.) is transfected into the cytoplasm of the target cell (4.), where it is translated into the nuclease pair proteins. The duration and strength of the expression can be influenced or controlled by means of different chemical modifications (5.). The nuclease pair is transported into the nucleus, where it binds to the target region and creates a double-strand break (DSB) (6.). The DSB stimulates a HR as a repair mechanism, exchanging the genetic defect for the corrected repair template (7.).

3. In vivo gene correction at the SP-B locus of the mouse by nec-mRNA

[0052] Next, it was examined whether a modified mRNA encoding a nuclease can catalyze an effective gene correction in the lung cell of the mouse in vivo. For this purpose, experiments were performed with a mouse having SP-B deficiency. The employed mouse model is described in detail in Kormann et al. (201 1 ; cit. loc).

[0053] To this end, the inventors designed an mRNA encoding a TALEN pair which is specific for the SP-B locus. Furthermore, a repair template was designed which is encoded by an adeno-associated viral vector (AAV). It comprises a constitutive promoter upstream of the SP-B cDNA to make the gene expression independent from doxycycline. The repair template is schematically shown in Fig. 3a. The repair template which carries a fully-functional CAG promoter with Tet0 ₇ -CMV and a truncated SP-B cDNA as homology arms is integrated into the genome via homologous recombination (HR) to overcome doxycycline dependency.

[0054] The nucleotide-modified mRNA (25 % s2U/m5C) and the vector- encoded repair template were administered into the lung of the mice via a singular high pressure application. In the following, the delivery of doxycycline was stopped which causes acute respiratory failure in the mice and, as a consequence, determines the life span of the mice. The result is shown in Fig. 3b. It turns out, mice which were treated with a combination of 25 % of s2U/m5C modified SP-B locus specific TALEN mRNA and the repair template (n = 6) survive significantly longer than the controls; cf. continuous right curve in comparison to the left dotted curve. The controls were treated with a corresponding nucleotide modified mRNA which encodes the read fluorescent protein (RFP) (n = 5, Kaplan Meier survival curves, Wilcoxon-Gehan test).

4. TALEN-encoding nucleotide-modified mRNA induces homology-directed repair in vitro

[0055] In a further experiment it should be examined whether and to which extent a replacement or a correction of the genetic alteration on the DNA can be obtained. For this purpose, the DNA of SP-B fibroblasts was cleaved by TALEN, encoded by modified mRNA, and a repair template with a Nhel restriction site was introduced. In the following, the extent of the homologous recombination was measured. The result is shown in Fig. 4. It becomes evident that a homologous recombination of 31 % was reached which demonstrates the suitability of the combination of nec-mRNA and repair template according to the invention.

5. Material and Methods

[0056] TALEN and ZFN reagents. TALENs and ZFNs targeting the transgenic SP-B cassette were screened by Dual Luciferase Single Strand Annealing Assay

(DLSSA) and assembled using an archive of zinc-finger proteins, as previously described; Urnov, F.D. et al. Nature 435, 646-651 (2005). The full amino acid sequences of the Z3 pair are shown in Fig. 7. The ZFN expression vector was assembled as previously described; Doyon, Y. et al. Nat Biotechnol 26, 702-708 (2008).

[0057] Dual Luciferase Single Strand Assay (DLSSA). ZFNs were screened using a luciferase-based reporter system. This reporter-based assay system is composed of four mammalian expression vectors each under the control of the cytomegalovirus (CMV) immediate early promoter. The vectors are (1 ) ZFN1 , (2) ZFN2, (3) pDLSSA- Firefly, and (4) pDLSSA-Renilla. pDLSSA-Firefly vector contains a Firefly luciferase gene derived from the pGL3-Promoter vector (www.promega.com) with an internal -600 bp duplication of the middle part of the Firefly luciferase gene. DNA fragments that contain individual ZFN pair binding sites are inserted between these duplicated regions. pDLSSA- Renilla is derived from pRL-TK (www.promega.com) and expresses the Renilla luciferase gene. One day before transfection, 20,000 mouse Neruo2A cells (www.atcc.org) are seeded in a 96-well plate with Dulbecco's Modified Eagle Medium (DMEM;

www.cellgro.com) plus 5 mM L-glutamine and 10% FBS. The four expression vectors described above (6.25 ng each) are co-transfected using Lipofectamine 2000 (Life Technologies). ZFN cleavage of the target plasmid followed by 5' to 3' end resection generates single-stranded DNA from the duplicated portion of the Firefly luciferase gene. Annealing of this complementary DNA and subsequent DNA repair creates an intact Firefly luciferase gene that reports the activity of the test ZFNs. Detection of Renilla luciferase serves as an internal control and allows for normalization of intra-transfection variability. Cells are harvested 24-hours post-transfection and the activities of both Firefly and Renilla luciferase are measured using the Dual-Glo Luciferase System (www.promega.com). ZFN activity is scored as the ratio of Firefly luciferase activity to Renilla luciferase activity. [0058] Targeting vectors. The targeting vector carrying the CAG promoter was assembled from synthetic oligonucleotides (www.lifetechnologies.com) and PCR products, and was verified by sequencing. The Nhel RFLP donor plasmid was constructed by removing the CAG promoter from the targeting vector by Nhel digestion, leaving a single Nhel restriction site, which was used in the RFLP assays.

[0059] Cell culture and transfection. For the T7 and HDR assays 1 x10 ⁶ fibroblasts in 6-well plates were transfected as indicated in the respective figure legends using the Neon electroporation system (www.lifetechnologies.com) with 100 μΙ tips. The electro- poration settings were 1 ,650 Volts, 20 ms, 1 pulse. A549 cells (human ATM cells, the cell type responsible for SP-B expression in the lungs) were maintained at 37°C under 5% C0 ₂ and grown in minimal essential medium (www.lifetechnologies.com), supplemented with 10% FCS, 1 % penicillin-streptomycin. One day before transfection, 50,000 or 80,000 cells/well/500 μΙ were plated in 24-well plates. The cells (70-90% confluent) were transfected with 5 μg (T7 assays, fragment analyses and RFLP) or 1 μg Z3 pair nec-mRNA (time-course experiment) using Neon electroporation (www.lifetechnologies.com) with a transfection mix volume of 100 μΙ according to manufacturer's instructions or transduced with MOI of 1 x10 ⁵ v.g. of each Z3 AAV6. For transfection experiments demonstrated in Fig. 6d, we equilibrated the DNA amounts by adding inert (empty vector) DNA to a total of 9 μg each. For transduction, the cells were washed once with PBS and cultured in Opti- MEM; 6 h after transduction 10% FCS was supplied. After 24 h the medium was removed, the cells washed once with PBS and fresh culture medium was added. Primary fibroblasts from transgenic SP-B mice were obtained by removing the dorsal skin, followed by separation of epidermis from the dermis using dispase. After further digestion of the dermis using collagenase, the suspension was passaged through a 70 μηη strainer. After wash and centrifugation steps, the cell pellet was resuspended in fibroblast culture medium (DMEM/Ham's F-12 medium with L-glutamine, 10% MSC grade Fetal Calf Serum, 1x MEM non-essential amino acids, 1 x sodium pyruvate, 1 % penicillin/streptomycin, 0.1 mM 2-mercaptoethanol). For the time course experiments: after 1 , 2, 3, 4, 5, and 14 days after transfection the A549 cells were harvested, permeabilized using BD Cytofix/Cytoperm plus (www.bd.com), stained with APC anti-DYKDDDK clone L5 (www.biolegend.com) antibody, and analysed on an LSR-I flow cytometer (www.bd.com) and data were analysed with BD FACSDiva software (www.bd.com). [0060] Generation of (nec-)mRNA. To generate templates for in vitro transcription the 3xFLAG-tagged T1 and Z3 were cut out of their original vectors and sub- cloned into a PolyA-120 containing pVAX1 (www.lifetechnologies.com). The plasmids were linearized with Xbal and transcribed in vitro using the MEGAscript T7 Transcription kit (www.lifetechnologies.com), incorporating 25% 2-thio-UTP and 25% 5-methyl-CTP or 100% PseudoUTP and 100% 5-methyl-CTP (all from www.trilink.com). The anti reverse CAP analog (ARCA) capped synthesized nec-mRNAs were purified using the MEGAclear kit (www.lifetechnologies.com) and analyzed for size on agarose gels and for purity and concentration on a NanoPhotometer (www.implen.com).

[0061] T7 nuclease assay. Genomic DNA was extracted from fibroblasts using the DNeasy Blood & Tissue Kit (www.qiagen.com). A 50 μΙ PCR reaction was set up using 100 ng of gDNA derived from fibroblasts previously transfected with 5 μg T1 or Z3 pair, 0.5 μΜ primers (for T1 : fwd, GTAGGCGTGTACGGTGGGAG [SEQ ID No. 1 ]; rev, CAGCAGAGGGTAGGAAGCAGC [SEQ ID No. 2]; for Z3: fwd,

TGTACGGTGGGAGGCCTAT [SEQ ID No. 3]; rev, CCTGGCAGGTGATGTGG [SEQ ID No. 4]), and AmpliTaq Gold 360 Mastermix (www.lifetechnologies.com). Another PCR reaction was performed using the same primer sets, but with gDNA from untransfected cells. The PCR products were run on agarose gels to verify size and sufficient amplification, pooled, purified by ethanol precipitation, dissolved in 20 μΙ water and the DNA concentration was measured on a NanoPhotometer. 2 μΙ NEBuffer 2 (www.neb.com), 2 μg purified PCR product and water were brought to a total volume of 19 μΙ. The DNA was hybridized in a thermocycler according to the following protocol: 95°C for 5 min, 95-85°C at -2°C/sec, 85-25°C at -O.TC/sec, hold at 4°C. 1 μΙ (10 U) of T7E1 (www.neb.com, M0302L) was added and incubated at 37°C for 15 min. The reaction was stopped by adding 2 μΙ of 0.25 M EDTA. The reaction was again purified by ethanol precipitation and dissolved in 15 μΙ water. The nuclease specific cleavage products were determined on agarose gels. The band intensities were quantified using ImageJ

(http://rsb.info.nih.gov/ij/).

[0062] For measuring off-target effects, A549 cells were transfected 5 μg mRNA or transduced with 1x10 ⁵ v.g. AAV6-Z3. PCR and T7 was performed as described above (primers: off-target 1 : fwd, GCAAGTTTGGCGTCGCTCCA [SEQ ID No. 5]; rev, AGAGGAAGGCGCGGCAGG [SEQ ID No. 6]; off-target 2: fwd,

TTCTTGCTCCAGTGACTCTCTTA [SEQ ID No. 7]; rev,

AGCCTAGTAAAGACAACACTAGTG [SEQ ID No. 8]; off-target 3: fwd,

CAACGTGACCTGCGAGCG [SEQ ID No. 9]; rev, GTGCACGCTCCACTTCTCG [SEQ ID No. 10]; off-target 4: fwd, CTGGAGATGCATCCTTGTCTGT [SEQ ID No. 1 1 ]; rev, GAG G GTGAAGACTTTTG G AG CT [SEQ ID No. 12]; off-target 5: fwd,

CAGCACCAGATGTTCCCTGTTA [SEQ ID No. 13]; rev,

TGGAAAGCAATAGTTCTAGGATGA [SEQ ID No. 14]).

[0063] HDR / RFLP assay. Genomic DNA was extracted from fibroblasts or lung tissue using the DNeasy Blood & Tissue Kit (www.qiagen.com). T1 or Z3 target loci were amplified by PCR (40 cycles, 58°C annealing and 30s elongation at 72°C; 5 min at 72°C to assure completion of amplicons) using 0.5 μΜ of primers P1

(CCTGGCAGGTGATGTGG [SEQ ID No. 15]) and P3 (TGTACGGTGGGAGGCCTAT [SEQ ID No. 16]) with AmpliTaq Gold 360 Mastermix. In addition, in-out PCR reactions were performed using primers P1 and P2 (AGGCACTGGGCAGGTAAGTA [SEQ ID No. 17]).

[0064] Flow Cytometry. Harvested lungs were digested at 37°C for 1 hour on a rotating shaker in 1 mg/ml collagenase type I (www.lifetechnologies.com), 1 % (500 U) DNase (www.epibio.com) solution. Digested lung was passed through a 40-μηι nylon cell strainer and erythrocytes were lysed using ACK Lysing Buffer (www.lifetechnologies). PE anti-CD45 clone 30-F1 1 , PE anti-CD31 clone C13.3, APC anti-mouse Ly-6A (Sca-1 ) clone D7 (www.biolegend.com), FITC anti-FLAG M2 and anti-clara cell secretory protein (www.sigmaaldrich.com) were used to stain lung cells. After staining for extracellular markers, cells were fixed and permeabilized using BD Cytofix/Cytoperm plus

(www.bd.com), then stained with intracellular antibodies. Flow cytometer analyses were performed on a LSR-I flow cytometer (www.bd.com) and data were analysed with BD FACSDiva software (www.bd.com). ATM and Clara cells sorting were performed with a FACSAria (www.bd.com).

[0065] Nanoparticles. Chitosan (83% deacetylated (Protasan UP CL 1 13, www.novamatrix.biz)) coated PLGA (Poly-d,l-lactide-co-glycolide 75:25 (Resomer RG 752H, www.evonik.de) nanoparticles (short: NPs) were prepared by using emulsion- diffusion-evaporation15 with minor changes. In brief, 100 mg PLGA was dissolve in ethyl acetate and added dropwise to an aqueous 2.5% PVA solution (Polyvinyl alcohol, Mowiol 4-88, www.kuraray.eu) containing 15 mg Chitosan. This emulsion was stirred (1 .5 h at RT) and followed by homogenization at 17,000 rpm for 10 min using a Polytron PT 2500E (www.kinematica.ch). These positive charged NPs were sterile filtered and characterized by Malvern ZetasizerNano ZSP (hydrodynamic diameter: 157.3±0.87 nm, PDI 0.1 1 , zeta potential +30.8 ± 0.1 15 mV). After particle formation they were loaded with mRNA by mixing (weight ratio: 25:1 ).

[0066] Transgenic SP-B cassette, mRNA templates and AAVs. AAV serotype 6 vectors from the Z3 pair and the donor sequence were produced and purchased from Virovek (www.virovek.com). The sequence information can be retrieved from the Sequence Listing at SEQ ID nos. 24-34 and from Fig. 14.

[0067] Transgenic SP-B cassette (before gene manipulation): the sequence at nucleotide positions 427-450 of SEQ ID no. 24 is deleted when transgene integration occurs.

[0068] AAV6_CAG_SP-B_donor: 5' AAV ITR: 3933 - 4051 (1 19bp); ZFN3- repair-template: 4087 - 6074 (1988bp); 3' AAV ITR: 61 12 - 6241 (130bp)

[0069] AAV6-ZFN 3-LEFT: 5' AAV ITR: 3933 - 4051 (1 19bp); CMV Promoter: 4060 - 4638 (579bp); 3Flag-NLS-38561 -Fok-KKR: 4844 - 5992 (1 149bp); bGHpA: 5999 - 6223 (225bp); 3' AAV ITR: 6240 - 6369 (130bp).

[0070] AAV6-ZFN 3-RIGHT: 5' AAV ITR: 3933 - 4051 (1 19bp); CMV Promoter: 4060 - 4638 (579bp); 3Flag-NLS-38558-Fok-ELD: 4766 - 6031 (1266bp); bGHpA: 6038 - 6262 (225bp); 3' AAV ITR: 6279 - 6408 (130bp).

[0071] Animal experiments. 6-8 week old BALB/c mice (www.criver.com) and transgenic SP-B mice6 [SP-C rtTA (teto) ₇ SP-B/SP-B ^" ' ^" ] were maintained under specific pathogen-free conditions and were kept with a 12 h /12 h light/dark cycle. All animals were provided with food and water ad libitum, and were acclimatized for at least 7 d before the start of the respective experiment. Transgenic SP-B mice were fed with doxycycline containing food until cessation (day 0 of the control and main groups). All animal procedures were approved and controlled by the local ethics committee and carried out according to the German law of protection of animal life.

[0072] Intratracheal injection. BALB/c or transgenic SP-B mice were anesthetized intraperitoneally with a mixture of medetomidine (0.5 mg/kg), midazolam (5 mg/kg) and fentanyl (50 Mg/kg), and suspended on a mouse intubation platform

(www.penncentury.com, Model MIP) at a 45° angle by the upper teeth. A small animal laryngoscope (www.penncentury.com) was used to provide optimal illumination of the trachea. A Microsprayer Aerosolizer - Model IA-1 C connected to a FMJ-250 High Pressure Syringe (both from www.penncentury.com) was endotracheally inserted and PBS, 20 g Z3 (nec-)mRNA naked or complexed with Nanoparticles or AAV6 (www.virovek.com) (was applied in a volume of 100 μΙ. The Microsprayer tip was withdrawn after 10 s, antidot was injected subcutaneously (atipamezol (50 Mg/kg), flumazenil (10 Mg/kg) and naloxon (24 Mg/kg)), and the mouse was taken off the support after 2 min.

[0073] Airway compliance. Compliance was determined by using the ex vivo model of the isolated perfused lung as described previously (IPL, Harvard Apparatus). In short, in situ mouse lungs were placed in a thorax chamber and mice were ventilated via a tracheal cannula. Ventilation rate was set to 90 breaths per minute with negative pressure ventilation between -2.8 cm H ₂ 0 and 8.5 cm H ₂ 0. To prevent atelectasis a hyperinflation was triggered every 5 minutes (-25 cm H ₂ 0). Perfusion of lungs was done with a 4% hydroxyethyl starch containing perfusion buffer via the pulmonary artery (flow 1 ml/min). Lung function parameters were recorded automatically and compliance calculated by HSE-HA Pulmodyn W Software (Harvard Apparatus). For graphical and statistical analysis, the mean compliance values were calculated from the last 10 timestamps (40 sec) of each 5-minute period (between two hyperinflations).

[0074] Airway resistance. Airway resistance in response to methacholine (MCh, acetyl-3-methylcholine chloride; Sigma-Aldrich) was again determined using the ex vivo model of the isolated perfused lung (IPL, Harvard Apparatus). In brief, after a 20- minutes baseline measurement, lungs were perfused with increasing concentrations of MCh (0.1 μΜ, 1 μΜ, 10 μΜ, and 100 μΜ) for 10 minutes each, separated by a 10-minute washout period with perfusion buffer. Lung function parameters were recorded automatically and airway resistance was recorded by HSE-HA Pulmodyn W Software (Harvard Apparatus). For graphical and statistical analysis, the mean resistance values were calculated from the last 10 timestamps (40 sec) of each 10-minute MCh exposure.

[0075] Histopathology. Mouse lungs were fixed in 4.5% Histofix

(www.carlroth.com) at 4°C overnight. Fixed lungs were embedded in paraffin, and slices were stained with either H&E or Surfactant Protein-B DAB (mouse monoclonal anti-SP-B antibody (www.abcam.com, ab3282), Zytochem Plus HRP One-Step Polymer anti- mouse/rabbit/rat (www.zytomed.com, ZUC53-006) and DAB substrate kit for peroxidase (www.vectorlabs.com, SK-4100). 3xFLAG FITC fluorescence staining (monoclonal anti- FLAG M2-FITC antidbody (www.sigma-aldrich.com, F4049) and DAPI counterstaining (www.applichem.com, A1001 ) was examined using a Zeiss Axio Imager. For 3xFLAG Cy3 fluorescence staining, rabbit polyclonal to DDDDK tag antibody (www.abcam.com, ab21536) was used as primary antibody and goat anti rabbit Cy3 antibody

(www.jacksonimmuno.com, 1 1 1 -165-144) was used as secondary antibody together with DAPI (www.applichem.com, A1001 ).

[0076] Western Blot. Protein from BALF was separated on NuPAGE 10% Bis- Tris Plus gels and a NuPAGE Mini Gel Tank (all from www.lifetechnologies.com), and immunoblotting was performed by standard procedures according to manufacturer's instructions using the XCell II Mini-Cell and blot modules (www.lifetechnologies.com). After blocking for 2 hours at room temperature, primary antibody against SP-B (kindly provided by Prof. Griese, Munich) or ANTI-FLAG M2 (www.sigmaaldrich.com) was incubated overnight, HRP-conjugated secondary antibodies (anti rabbit from

www.dianova.com) were incubated for 1 hour. Blots were processed by using ECL Prime Western Blot Detection Reagents (www.gelifesciences.com). Semiquantitative analysis was performed with the Quantity One software (www.bio-rad.de). [0077] Target-site sequencing. Genomic DNA from primary fibroblasts (in vitro transfected/transduced) or sorted ATM cells (after in vivo transfection/transduction) was isolated using the NucleoSpin Tissue Kit (www.mn-net.com) according to the manufacturer's protocol. Amplicons were derived from PCR with Primers P1 and P2 (sequences see above) using the following conditions: AmpliTaq Gold 360 master mix

(www.lifetechnologies.com) at 95 °C for 10 min, 95 °C for 30 sec, 60 °C for 30 sec, 72 °C for 60 sec, with in total 35 cycles and a final extension step at 72 °C for 7 min. The amplicons were cloned into the pCR-TOPO vector (www.lifetechnologies.com) and sequenced using the primers M13forward (GTAAAACGACGGCCAGTG [SEQ ID No. 18]) and M13reverse (CAGGAAACAGCTATGACCATG [SEQ ID No. 19]). The alignments have been performed with Geneious R6 (www.biomatters.com) using the "multiple align" function, choosing a cost matrix of 65% similarity (5.0/-4.0), a gap open penalty of 12 and a gap extension penalty of 3.

[0078] RealTime RT PCR. The lung cell separations were washed vigorously three times with PBS to avoid carrying over RNA not taken up by lung cells (the third supernatant was later tested for RNA contamination using the qPCR procedure described below). RNA was then isolated with the RNeasy purification kit (www.qiagen.com).

Reverse transcription of 50 ng RNA was carried out using iScript cDNA synthesis kit (www.bio-rad.com). Detection of Z3 cDNA was performed by SYBR-Green based quantitative Real-Time PCR in 20 μΙ reactions on a VNA7 (www.lifetechnologies.com). Reactions were incubated for 10 min at 95°C, followed by 40 cycles of 15 sec at 95°C and 2 min at 50°C (annealing and extension), followed by standard melting curve analysis. The following primer pairs were used: Z3 left fwd: TGTACGGCTACAGGGGAA [SEQ ID No. 20], Z3 left rev GCCGATAGGCAGATTGTA [SEQ ID No. 21]; optimal determined house-keeping gene beta-actin: fwd TAGGCACCAGGGTGATG [SEQ ID No. 22], rev

GCCATGTTCAATGGGGTACT [SEQ ID No. 23].

[0079] Statistics. Differences in mRNA expression between groups were analyzed by pair-wise fixed reallocation randomization tests with REST 2009 software17. All other analyses were performed using the Wilcoxon-Mann-Whitney test with SPSS 21 (www.ibm.com). Data are presented as mean ± s.e.m. or as the median ± IQR (interquartile ranges) and P < 0.05 (two-tailed) was considered statistically significant. For survival studies Log-rank tests were performed. Statistics for lung compliance was performed using 2way ANOVA with Bonferroni-post tests with GraphPad Prism 5.0 software. Lung function data are presented as mean ± s.d. and P < 0.05 (two-tailed) was considered statistically significant. No randomization was used for animal experiments. In all cases but at administration of AAV6/mRNA i.t, the investigators were blinded when assessing outcomes.

6. Results

[0080] Nuclease-mediated genome editing holds enormous potential to knockout unwanted genes or repair disease-causing mutations. An ideal nuclease delivery vehicle is (i) short-lived, (ii) non-integrating, and (iii) able to enter target cells efficiently. A variety of vectors have been utilized to deliver nuclease pairs, however, to date, none have achieved direct in vivo gene correction while simultaneously being transient and non-integrating.

[0081] The inventors have used modified mRNA as an alternative to traditional viral vectors, one which naturally avoids genomic integration and provides a transient pulse of protein expression. By using nucleotide-modified mRNA, the inventors reached therapeutic protein expression levels in vivo in mouse models of surfactant protein B (SP- B) deficiency and experimental asthma. Here, the inventors utilize modified mRNA to deliver site-specific nucleases to the lung to demonstrate the value of "nec-mRNA" as a tool for in vivo genome editing.

[0082] To illustrate the effectiveness of nec-mRNA as a nuclease-delivery vehicle, the inventors chose a well-established transgenic mouse model of SP-B deficiency, where SP-B cDNA is under the control of a Tetracycline-inducible promoter. Administration of doxycycline drives SP-B expression levels comparable to those observed in wild- type mice (Fig. 5). Following cessation of doxycycline, this model closely mimics the phenotypic changes seen in the human version of the disease: thickened alveolar walls, heavy cellular infiltration, increased macrophages and neutrophils, interstitial edema, augmented cytokines in the lavage, a significant drop in lung function, and fatal respiratory distress leading to death within days. Here, the inventors insert a constitutive CAG promoter immediately upstream of the SP-B cDNA to allow doxycycline-independent expression and prolonged life in treated mice.

[0083] First, a panel of ZFNs and TALENs was customized to target the transgenic SP-B cassette (Fig. 6a and Fig. 7). Due to their high activity and proximity to the desired site of promoter integration, TALEN #1 (T1 ) and ZFN #3 (Z3) were selected (Fig. 6a, b). In comparison to plasmid DNA, T1 and Z3 delivered as mRNA showed a significant increase in both DSB-induction (Fig. 6c Fig. 8; P < 0.05) and homology directed repair (HDR) (Fig. 6d, P < 0.05). As Z3 mRNA was more efficient than T1 mRNA in both cases, Z3 was chosen for further experimentation (amino acid sequences in Fig. 9; SEQ ID nos. 35 and 36). Comparison with a Z3-encoding AAV vector ("Z3 AAV") highlights the shortlived expression pattern of Z3 mRNA (Fig. 6e), limiting the time during which off-target cleavage activity could occur.

[0084] To optimize Z3 expression in the lung, the inventors administered a panel of 3xFLAG-tagged Z3 mRNAs with various modification schemes, with or without mRNA-complexation to nanoparticles (NPs); cf. Nafee, N., Taetz, S., Schneider, M., Schaefer, U.F. & Lehr, CM. Nanomedicine 3, 173-183 (2007). Following intratracheal (i.t.) delivery, NP-complexing significantly increased mRNA expression levels (Fig. 10).

3xFLAG protein expression was most robust for the s2Uo.25/m5C ₀ .25-modified, NP- complexed group (Fig. 6f,g and Fig. 1 1 ), and no immune activation was observed following i.t. delivery (Fig. 12). Hence, subsequent in vivo studies utilized i.t. delivery of this candidate, referred to as "Z3 nec-mRNA-NP".

[0085] Next, a complementary donor template was designed to insert a constitutive CAG promoter at the Z3 nec-mRNA-NP cut site, upstream of the transgenic SP-B cDNA (Fig. 13g and Fig. 14). Successful site-specific HDR would allow mice to survive and produce SP-B in the absence of doxycycline. As it is critical to deliver the donor template in excess to ensure it is favored over the homologous chromosome during HDR, for this proof-of-principle, the inventors utilized a vector known to transduce lung cells with high efficiency, AAV-serotype 6 (integration-deficient lentiviruses will be tested in future studies). Ex vivo delivery of the AAV6-donor with Z3 nec-mRNA-NP resulted in successful HDR in primary fibroblasts (Fig. 15).

[0086] Moving in vivo, AAV6-donor and Z3 nec-mRNA-NP (or a Z3 AAV positive control) were then delivered to the lung of transgenic SP-B mice, followed by cessation of doxycycline (Fig. 13a). Notably, mice in these groups lived significantly longer in comparison to matched controls groups (Fig. 13a, P < 0.001 ), while maintaining SP-B expression levels comparable to mice receiving doxycycline, as far as 20d post- doxycycline-removal (Fig. 13b,c and Figs. 16 and 17).

[0087] Phenotypically, combining gene correction with AAV6-donor and Z3 nec- mRNA-NP (or Z3 AAV) prevented the significant drop in lung function (Fig. 13d and Fig. 18), severe hemorrhagic infiltrations and large-scale edema (Figs. 19-21 ), and neutrophilia (Fig. 22) observed in the lungs of negative controls. A non-significant increase of IL-12 was observed in nec-mRNA-NP- versus PBS-treated mice (Fig. 13), however, no IFN-a elevation was detected (data not shown). DSBs and HDR rates (the latter determined by in-out PCR, see Fig. 13g) were concomitant with successful gene manipulation (Fig.

13e,f), which was also determined by target site sequencing (Fig. 26). If achieved in humans, HDR rates of -9% would likely be sufficient to avoid severe disease progression (see Discussion). Results also confirmed that nuclease expression was longer-lived if administered via AAV, rendering nec-mRNA-NP a more favorable delivery vehicle (Fig. 13h and Figs. 24 and 25).

[0088] Inherent key limitations of our approach are (i) the co-transfection of an AAV-DNA donor template in conjunction with nec-mRNA, (ii) the temporal delimitation of our curative in vivo treatment, probably owing to the natural turnover of the transfected lung cell populations, and (iii) the engineering of an artificial, transgenic cassette compared to humanized models. However, the use of nec-mRNA will have immediate implications for all nuclease platforms, including CRISPR/Cas9 systems, targeted gene knockout, as well as therapeutic gene correction strategies for the treatment of SP-B deficiency and other diseases, such as cancer. The inventors will test this technology in humanized models when available and are confident to move nec-mRNA and nec-mRNA-NP (for efficient lung transfection) finally to the clinic. [0089] Overall, the inventors conclude that co-delivery of Z3 nec-mRNA-NP and AAV6-donor results in successful site-specific genome editing in vivo, documenting the first report of life-prolonging gene correction in the lung.

7. Discussion

[0090] This proof-of-principle in a transgenic model of SP-B deficiency demonstrates that nec-mRNA can achieve therapeutic levels of gene correction in vivo, while possessing the three main criteria of an optimal genome editing reagent: (i) transience, (ii) an inability to integrate, and (iii) sufficient transfection of target cells. As lung cell turnover likely prevented survival beyond 30-35 days in this model, animals in future studies will undergo repeated nec-mRNA administration to target additional differentiated cells of the lung.

[0091] The inventors made sure that the truncated SP-B gene fragment in the right homology arm does not express functional SP-B protein by testing the administration of AAV6 donor w/o functional nucleases (Fig. 13a, group C): all mice died within three days. Although there was still some residual SP-B detectable in the Western Blot (about 10%, which the inventors usually see if lavages are tested only three days after Doxycy- cline cessation (data not shown)), it is highly unlikely that this signal is derived from the donor construct as the molecular weight of the band is normal (and therefore inconsistent with any truncated form).

[0092] The ability to target and correct lung progenitor cells will also be the subject of ongoing investigation. The inventors also want to emphasize that the main safety gain by our nec-mRNA technology concerns the reduction of nuclease-derived off-target effects; it does not eliminate the integration of donor template. Since SP-B acts extra- cellularly in the alveolar space, modification of a small number of cells could functionally correct a larger area of lung tissue. The inventors found in vivo HDRs of about 9%: in humans 5-10% of SP-B levels in the lung are sufficient and show only a mild disease (in humans there are SNPs in the SP-B gene causing about 10% of normal SP-B levels, and many of those people were completely healthy. Also, there is no linear correlation be- tween achieved HDR in lung cells (see Fig. 13f) and SP-B expression in the lungs (see Figs. 13b,c and Fig. 17); together supporting the notion that an in vivo HDR of -9% - if achievable - should have therapeutic effects in human.

[0093] Though AAV vectors have not been associated with genotoxicity, further development of nec-mRNA-mediated gene correction approaches may also benefit from pairing with a non-viral or integration-deficient lentiviral donor template. The inventors chose to not look for AAV donor integration for several reasons: a) the inventors wanted the vector to be as coherent as possible and AAV donor integration measurements are at best mere estimations; b) any experiment that uses a transgene donor will require the use of a DNA-based donor and therefore has some risk of insertional mutagenesis. Consequently, it is not possible to achieve HDR in vivo, while avoiding background vector integration. A multitude of papers describing use of AAV and lentivirus have been published in NBT, all of which have necessarily had some background level of donor integration; however no pathological effects of AAV utilization could be demonstrated in extensive murine studies; and c) the big advantage of the inventors' work is that mRNA delivery prevents the persistent expression of the nucleases. This is a far larger worry than background AAV integration.

[0094] Off-target cleavage in vitro on the top five in silico predicted sites was a minor issue with an average of only 0.78% indels at day 14 of transfected or transduced A549 cells (data not shown).

[0095] We found it also important to perform in/out PCRs (and T7 endonucle- ase reactions) on lung samples of mice that received only AAV6-donor (group C). This is important because, given the large overlap between the donor and the chromosome in our case, recombination can occur in the PCR itself. Therefore, we did control PCRs (P1/P3, P1/P2) and T7 reactions on all mice (P1/P3, group C) or pooled samples (P1/P2, groups A, B and C) and could strenghten the positive results found in groups A and B, as no HDR event could have been detected in DNA samples from group C (Fig. 27). [0096] With respect to human SP-B deficiency, it is important to note that the site-specific nucleases designed to target the transgenic locus in this mouse model will not be directly applicable to the human condition. Also, the CAG promoter is very strong, so manipulated cells likely produce significantly higher amounts of SP-B than normal, endogenous cells. Further, the PCR assay used for quantification likely underestimates the true amount of CAG promoter integration. Despite this, the inventors feel that transgenic SP-B deficient mice serve as an excellent proof-of-principle model for several reasons: first, cessation of doxycycline results in phenotypic changes closely modeling those observed in the human condition. Second, administration of doxycycline drives SP- B expression levels comparable to wild-type mice; and finally, the outcome of survival in this model is a definitive measure of efficacy. Together with Chitosan-coated NP's, nec- mRNA presents a strong tool to approach lung diseases still currently uncorrectable in the human system. Combining nec-mRNA with other structured NPs (cf. Young C. et al. Nat Prot 9, 1900-1915 (2014); incorporated herein by reference) may expand the capabilities of gene manipulation to other large disease fields such a cancer therapeutics.

7. Conclusion

[0097] The inventors were able to demonstrate in an impressive manner by means of a mouse model that by using a nuclease-encoding nucleotide-modified messenger RNA (nec-mRNA) a genetic alteration on a DNA can be permanently corrected. The nec-mRNA is administered together with a repair template which comprises the genetic information to be inserted or to be replaced, respectively.