WO2018141371A1 - Purification and/or formulation of rna - Google Patents

Abstract

The present invention is concerned with a method of purifying and/or formulating RNA from samples comprising the RNAand TEAA, wherein the method can be carried out in multi-well plates. The present invention is further directed to RNA purified and/or formulated with such a method.

Description

Purification and/or formulation of RNA

FIELD OF THE INVENTION

The present invention is concerned with a method of purifying and/or formulating nucleic acids. In particular, the present invention provides a method of purifying and/or formulating RNA from a sample comprising the RNA and triethylammonium acetate, wherein the method can be carried out in multi-well plates. Said method is particularly suitable for high- throughput settings, e.g. for the small-scale production of multiple nucleic acid therapeutics in parallel (e.g., for a personalized medicament) or for screening approaches. BACKGROUND OF THE INVENTION

Therapeutic nucleic acids represent an emerging class of drugs. Such therapeutic nucleic acids comprise e.g. mRNA molecules encoding antigens for use as vaccines (Fotin-Mleczek et al. 2012. J. Gene Med. 14(6):428-439). In addition, RNA molecules are suitable for replacement therapies, e.g. providing lacking proteins to patients (such as growth factors or enzymes, see e.g. Thess, et al. Molecular Therapy (2015)). Moreover, RNA molecules for the provision of RNA-coded antibodies for passive immunization or cancer immunotherapies have been developed (e.g., WO2008083949). Furthermore, the therapeutic use of noncoding

immunostimulatory RNA molecules (e.g., WO2009095226) and other noncoding RNAs such as microRNAs or RNAs suitable for genome editing (e.g., CRISPR/Cas9 guide RNAs) are under development.

Given the huge therapeutic potential of nucleic acids, there is an urgent need to develop improved methods for the purification and formulation of nucleic acids to lower the costs for the nucleic acid production. However, efficient purification continues to be a challenge. This is partly due to the different types and combinations of undesired contaminants in a sample that need to be separated and removed from a desired RNA species to obtain a pure RNA sample. Such contaminants are typically components and by-products of upstream processes, in particular the RNA manufacture. Short RNA molecules can be synthesized by chemical methods, whereas long RNAs are typically produced by in vitro transcription using suitable DNA templates with a promoter and RNA polymerases, for example bacteriophage SP6, T3 or T7 RNA polymerases which are able to bind to said promoter. Where in vitro transcription is used to manufacture large RNAs, the non-purified sample typically contains the desired RNA species alongside various contaminants, such as undesired RNA species (abortive sequences), proteins, spermidine, DNA or fragments thereof, pyrophosphates, free nucleotides, endotoxins, detergents, and organic solvents.

HPLC purification (PureMessenger® system; WO2008077592) has been developed as a particularly suitable method for the removal of small RNA contaminants (e.g. abortive RNA sequences), proteins, residual nucleotides and the residuals of the DNA template after its hydrolysis. However, the RNA-containing HPLC fractions contain an organic solvent, mostly acetonitrile, and triethylammonium acetate (TEAA), which have to be removed before the produced RNA can be applied, or before the RNA is subjected to further formulation steps.

The removal of organic solvents from an HPLC sample comprising nucleic acids is typically performed by manual salt/alcohol precipitation, also in pharmaceutical grade nucleic acid production processes (WO2016180430). However, such a precipitation procedure introduces additional chemical components into the production process (isopropanol, alcohol, lithium chloride), requires additional washing steps (e.g., ethanol washing steps), drying steps (the precipitated RNA has to be dried to remove ethanol), and a dissolving step (typically in water at 4°C, which can take several days). All these procedural steps also lead to loss of RNA product. Furthermore, the re-dissolved RNA has to be stored at -20°C. Once such a precipitation procedure to remove TEAA and the organic solvent from a pharmaceutical RNA product has been performed, it makes an analysis for isopropanol contamination mandatory to guarantee for a safe product (WO2016180430). To obtain long-term stability, the dissolved RNA has to be lyophilized, which is an additional, time-consuming step (WO2016165831). Thus, precipitation and re-solubilization of the precipitated nucleic acids are time consuming and often associated with problems (e.g., precipitates are not well-dissolvable due to residual acetonitrile and/or TEAA, contamination with isopropanol, loss of RNA). Moreover, the use of alcohols and other organic solvents should be avoided in cGMP compatible manufacturing processes since additional quality controls are then required. Finally, precipitation steps are merely applicable in large-scale formats.

Given the above outlined problems associated with nucleic acid purification, there is the need for an improved method that results in the removal of in particular TEAA from a sample comprising RNA and TEAA. It would furthermore be of great value for the field if such an improved method can be carried out in a high-throughput setting.

SUMMARY OF THE INVENTION

The present invention provides a method that allows for the removal of TEAA without the need for a precipitation step. An organic solvent (such as e.g. acetonitrile), if additionally present, may be removed by a rotary vacuum concentration and/or evaporation, if necessary. Furthermore, the method can be carried out in multi-well plates, i.e. it is suitable for high throughput and/or robot-assisted settings, and yields a pure, stable and readily soluble RNA product. In a first aspect, the present invention is directed to a method of purifying and/or formulating RNA from a sample comprising the RNA and triethylammonium acetate, the method comprising the step of lyophilizing said sample.

In a particularly preferred embodiment, the method is performed in a multi-well plate. A multi-well plate typically used in laboratory setting is particularly preferred, such as e.g. a 12- well plate, a 24-well plate, a 48-well plate or a 96-well plate. It can furthermore be particularly preferred that the multi-well plate is a glass plate or a glass-coated plate.

In another particularly preferred embodiment, the method is performed in a high throughput or robot-assisted setting.

In yet another embodiment, the RNA subject to the inventive method is in vitro transcribed RNA. Said in vitro transcribed RNA has furthermore been subject to an HPLC purification step in order to remove contaminants from the RNA production, where the RNA may preferably be eluted in the HPLC purification with a mixture of triethylammonium acetate and an organic solvent, such as e.g. acetonitrile. Most preferably, said in vitro transcribed R A has been purified via HPLC using the PureMessenger® system. The method may comprise a step of adjusting the RNA concentration in the sample to about 0.2 to about 2.0 g/1 prior to carrying out the lyophilizing step. Depending on the starting concentration of the RNA, this adjustment may be a concentration or a dilution, wherein a dilution is preferably carried out with ddH₂0. In a preferred embodiment, the lyophilizing step comprises steps of freezing, primary drying, and secondary drying.

It can be preferred that the temperature of the freezing step is in the range of from about -60 to about -40°C, preferably about -50°C, more preferably -48°C. It is further preferred that the sample is cooled to said temperature at a rate of about -0.5°C / minute. Furthermore, the freezing step takes preferably between about 3 to about 6 hours, preferably between about 4 to about 5 hours, and more preferably about 4.5 hours.

In another preferred embodiment, the target temperature of the primary drying step is in the range of from about -30 to about 15°C, preferably from about -20 to about 5°C, and more preferably about -15°C. It can also be preferred that the primary drying step takes between about 3 to about 12 hours, preferably between about 5 to about 10 hours, and more preferably about 9 hours. In yet another preferred embodiment, the target temperature of the second drying step is in the range of from about 10 to about 20°C, preferably about 15°C. It can be preferred that the secondary drying step takes between about 2 to about 5 hours, preferably about 3 to about 4 hours, more preferably about 3.5 hours. In still another embodiment, step b) is performed using an aluminum thermo block and/or at about 80% resistivity.

If no co-solvent is used (see below), it is preferred that the lyophilization step is repeated twice. In this case, the RNA is dissolved in between the steps, preferably in ddH₂0. Thus, if no co-solvent is used, it is preferred that three lyophilization cycles are carried out in total, wherein the R A is dissolved in between the steps in ddH₂0.

In another embodiment, the sample is substantially free from triethylammonium acetate after the lyophilization step.

Embodiments if a co-solvent is used

For the present method, it is preferred that a co-solvent is added to the sample prior to the (first) lyophilization step. Said co-solvent may be selected from the group consisting of t- Butanol, 2-Butanol, Ethanol, n-Propanol, n-Butanol, Isopropanol, Ethyl acetate, Dimethyl carbonate, Dichloromethane, Methyl ethyl ketone, Methyl isobutyl ketone, Acetone, 1- pentanol, Methyl acetate, Methanol, Carbon tetrachloride, Dimethyl sulfoxide,

Hexafluoroacetone, Chlorobutanol, Dimethyl sulfone, Acetic acid, Acetate, Cyclohexane, Tetrahydrofuran, Tetrahydropyran, Dioxane, Trioxane and other cyclic mono-, di- and tri- ethers, PEG 600-6000, and Tocopherol Succinate. It is particularly preferred that said co- solvent is t-Butanol.

Preferably, the t-Butanol is added such that the resulting t-Butanol concentration in the sample is between about 5 and about 50% (f.c). Even more preferably, the t-Butanol is added such that the resulting t-Butanol concentration in the sample is about 20% (f.c).

In another preferred embodiment, the freeze point of the sample after t-Butanol addition is in the range of from about -50 to about -10°C. Preferably, the freeze point of the sample after the addition of t-Butanol is in the range of from about -30 to about -20°C.

In a particularly preferred embodiment, the lyophilization step is repeated once if a co-solvent is used. In this situation, the co-solvent is only added once, namely prior to the first lyophilization step (and not prior to the second lyophilization step; for the second

lyophilization step, ddH₂0 is preferably used). Thus, two lyophilization steps are preferably carried out if a co-solvent is present.

In yet another embodiment, the sample is substantially free from the co-solvent after the lyophilization step. Embodiments relating to the additional presence of an organic solvent

In yet another embodiment, the sample additionally comprises an organic solvent, and the method additionally comprises the step of rotary vacuum concentration and/or evaporation using nitrogen, argon or carbon dioxide prior to the lyophilization step. Said concentration and/or evaporation is carried out in order to remove the organic solvent from the sample and to purify said sample from the organic solvent, respectively.

Preferably, the rotary vacuum concentration is performed at a temperature ranging from about 10 to about 40°C, more preferably from about 10 to about 30°C and most preferably at a temperature of about 30°C, with 28°C being most preferred. It can be preferred that the rotary vacuum concentration takes between about 2 to about 10 hours, preferably about 2 to about 6 hours, more preferably about 4 hours. Prior to the rotary vacuum concentration, a salt may be added to the sample, wherein said salt is preferably NaCl. NaCl may be added to a final concentration of about 10 to 2 mmol/g R A, preferably about 8 to 4 mmol/g RNA and most preferably about 6 mmol/g RNA.

It can be preferred that said organic solvent is selected from the group consisting of acetonitrile, methanol, ethanol, 1-propanol, 2-propanol, ethyl acetate, tetrahydrofuran, acetone and a mixture of any of the foregoing. Most preferably, said organic solvent is acetonitrile.

In an embodiment, the sample is substantially free from the organic solvent after the rotary vacuum concentration and/or evaporation.

General formulation embodiments

In yet another embodiment, the method comprises the step of adding a lyoprotectant, wherein said lyoprotectant is added prior to the final lyophilization cycle. Said lyoprotectant may be selected from the group consisting of sucrose, fructose, glucose, mannose, trehalose, mannitol, polyvinylpyrrolidone, and Ficoll 70.

In still another embodiment, the method comprises the step of adding a complexing agent, wherein said complexing agent is added prior to the final lyophilization cycle. Said complexing agent is preferably a cationic or polycationic compound, wherein said cationic or polycationic compound is preferably a cationic or polycationic peptide or protein.

In a second aspect, the present invention is directed to R A purified and/or formulated according to the method described and claimed herein.

In a third aspect, the present invention is directed to the RNA of the second aspect for use in therapy.

In a fourth aspect, the present invention is directed to the use of a method according to the first aspect for the small-scale production of multiple RNAs in parallel.

In a fifth aspect, the present invention is directed to the use of a method according to the first aspect for the screening of RNAs in parallel.

BRIEF DESCRIPTION OF THE DRAWINGS

The figures shown in the following are merely illustrative and shall describe the present invention in a further way. These figures shall not be construed to limit the present invention thereto.

Figure 1: shows the effect of the number of lyophilization cycles on TEAA removal by single-solvent lyophilization from RNA samples in deep well plates. For experimental details, see Example 2.

Figure 2: shows the integrity of RNA in 6 different samples before (black bars) and after

(white bars) heat dissolution of RNA purified by single-solvent lyophilization.

For experimental details, see Example 2.

Figure 3: shows the RNA yield in 6 different vials, for each vial before (left column), after one (second column), after two (third column) and after three (right column) lyophilization cycles. For experimental details, see Example 3.

Figure 4: shows the impact of TEAA concentration and content of t-BuOH on the freeze point of the sample. For experimental details, see Example 5.2.

Figure 5: shows the duration of single-solvent and co-solvent lyophilization. For

experimental details, see Example 5.2. shows the RNA recovery after single-solvent lyophilization compared to four different co-solvent lyophilization procedures (1 = Program 1 of Table 5; 2 = Program 2 of Table 5; 3 = Program 3 of Table 5; 4 = Program 4 of Table 5). For experimental details, see Examples 2, 5.2, and 6.3.

shows the RNA integrity after four different co-solvent lyophilization procedures (1 = Program 1 of Table 5; 2 = Program 2 of Table 5; 3 = Program 3 of Table 5; 4 = Program 4 of Table 5). For experimental details, see Example 6.1.

shows the residual TEAA content after single-solvent lyophilization compared to four different co-solvent lyophilization procedures (1 = Program 1 of Table 5; 2 = Program 2 of Table 5; 3 = Program 3 of Table 5; 4 = Program 4 of Table 5). For experimental details, see Example 6.2.

shows the residual t-Butanol content after four different co-solvent lyophilization procedures (1 = Program 1 of Table 5; 2 = Program 2 of Table 5; 3 = Program 3 of Table 5; 4 = Program 4 of Table 5). For experimental details, see Example 6.4.

shows the RNA solubility after single solvent (right side) and co-solvent (left side) lyophilization combined with heat dissolution at 45-50 °C. For experimental details, see Example 6.5.

DETAILED DESCRIPTION OF THE INVENTION

DEFINITIONS

For the sake of clarity and readability the following definitions are provided. Any technical feature mentioned for these definitions may be read on each and every embodiment of the invention. Additional definitions and explanations may be specifically provided in the context of these embodiments.

As used in the specification and the claims, the singular forms of "a" and "an" also include the corresponding plurals unless the context clearly dictates otherwise.

The terms "about" and "approximately" in the context of the present invention denotes an interval of accuracy that a person skilled in the art will understand to still ensure the technical effect of the feature in question. The term typically indicates a deviation from the indicated numerical value of ±10% and preferably ±5%.

It needs to be understood that the term "comprising" is not limiting. For the purposes of the present invention, the term "consisting of is considered to be a preferred embodiment of the term "comprising of. If hereinafter a group is defined to comprise at least a certain number of embodiments, this is also meant to encompass a group which preferably consists of these embodiments only. RNA: RNA is the usual abbreviation for ribonucleic acid. It is a nucleic acid molecule, i.e. a polymer consisting of nucleotide monomers. These nucleotides are usually adenosine monophosphate (AMP), uridine monophosphate (UMP), guanosine monophosphate (GMP) and cytidine monophosphate (CMP) monomers or analogues thereof (particularly as defined herein below), which are connected to each other along a so-called backbone. The backbone is formed by phosphodiester bonds between the sugar, i.e. ribose, of a first and a phosphate moiety of a second, adjacent monomer. The specific order of the monomers, i.e. the order of the bases linked to the sugar/phosphate-backbone, is called the RNA sequence. Usually RNA may be obtainable by transcription of a DNA sequence, e.g., inside a cell. In eukaryotic cells, transcription is typically performed inside the nucleus or the mitochondria. In vivo, transcription of DNA usually results in the so-called premature RNA (also called pre-mRNA, precursor mRNA or heterogeneous nuclear RNA), which has to be processed into so-called messenger RNA, usually abbreviated as mRNA. Processing of the premature RNA, e.g. in eukaryotic organisms, comprises a variety of different posttranscriptional modifications such as splicing, 5 '-capping, polyadenylation, export from the nucleus or the mitochondria and the like. The sum of these processes is also called maturation of RNA. The mature messenger RNA usually provides the nucleotide sequence that may be translated into an amino acid sequence of a particular peptide or protein. Typically, a mature mRNA comprises a 5 '-cap, optionally a 5 'UTR, an open reading frame, optionally a 3 'UTR and a poly(A) tail. In addition to messenger RNA, several non-coding types of RNA exist which may be involved in regulation of transcription and/or translation, and immunostimulation.

Within the present invention the term "RNA" further encompasses any type of single stranded (ssRNA) or double stranded RNA (dsRNA) molecule known in the art, such as viral RNA, retroviral RNA and replicon RNA, small interfering RNA (siRNA), antisense RNA (asRNA), circular RNA (circRNA), CRISPR/Cas9 guide RNA, ribozymes, aptamers, riboswitches, immunostimulating RNA, transfer RNA (tRNA), ribosomal RNA (rRNA), small nuclear RNA (snRNA), small nucleolar RNA (snoRNA), microRNA (miRNA), and Piwi-interacting RNA (piRNA).

The term furthermore encompasses any type of RNA comprising nucleotide analogs as defined herein below. HPLC: HPLC (abbreviation for "High Performance (High Pressure) Liquid

Chromatography") is an established method of separating mixtures of substances, which is widely used in biochemistry, analytical chemistry and clinical chemistry. An HPLC apparatus consists in the simplest case of a pump with eluent reservoir containing the mobile phase, a sample application system, a separation column containing the stationary phase, and the detector. In addition, a fraction collector may also be provided, with which the individual fractions may be separately collected after separation and are thus available for further applications.

RNA can be purified from various contaminations from previous manufacturing steps. These include buffer contaminations, protein impurities (Escherichia coli proteins, Restriction enzymes, T7-RNA-Polymerase, RNase-Inhibitor, DNase I, and BSA), impurities from RNA- RNA hybrids, from DNA-RNA hybrids or their fragments, from pDNA contaminations and bacterial genomic DNA contaminations, and solvent contaminations (Acetonitrile,

Chloroform, TEAA, 2-Propanol and Phenol) and free nucleotides. Moreover, size exclusion occurs during that procedure (smaller and larger RNAs can be excluded).

Reversed phase chromatography is preferably used when purifying RNA. Reversed phase HPLC consists of a non-polar stationary phase and a moderately polar mobile phase. The retention time is therefore longer for molecules, which are more non-polar in nature, allowing polar molecules to elute more readily. Retention time is increased by the addition of polar solvent to the mobile phase and decreased by the addition of more hydrophobic solvent. Reversed phase HPLC may comprise the use of a porous reserved phase as stationary phase, which may be provided with a particle size of 8.0 μιη to 50 μιη, in particular 8.0 to 30 μιη, still more preferably about 30 μιη. The reversed phase material may be present in the form of small spheres. Alternatively, the reversed phase may be porous and may have specific particle sizes. Preferably, the reversed phase has a pore size of 1000 A to 5000 A, in particular a pore size of 1000 A to 4000 A, more preferably 1500 A to 4000 A, 2000 A to 4000 A or 2500 A to 4000 A. Particularly preferred pore sizes for the reversed phases are 1000 A to 2000 A, more preferably 1000 A to 1500 A and most preferably 1000 A to 1200 A or 3500-4500 A. Most preferred is a pore size of 4000 A. With a reversed phase having these pore sizes, particularly good results are achieved with regard to purification of the RNA, in particular the elevated pressures built up in the method according to A. Azarani and K.H. Hecker are thus avoided, whereby preparative separation is made possible in a particularly favourable manner. At pore sizes of below 1000 A separation of RNA molecules becomes poorer.

A pore size of 1000 A to 5000 A, in particular a pore size of 1000 A to 4000 A, more preferably 1500 A to 4000 A, 2000 A to 4000 A or 2500 A to 4000 A may be suitable to separate a RNA from other components of a mixture, the RNA having a size as mentioned above of up to about 15000 nucleotides (as single stranded RNA molecule) or base pairs (as double stranded RNA molecule), in particular 100 to 10000, more preferably 500 to 10000 nucleotides or base pairs, even more preferably 800 to 5000 nucleotides or base pairs and even more preferably 800 to 2000 nucleotides or base pairs. However, the pore size of the reversed phase may also be selected in dependence of the size of the RNA to be separated, i.e. a larger pore size may be selected, if larger RNA molecules are to be separated and smaller pore sizes may be selected, if smaller RNA molecules may be selected. This is due to the effect that the retention of the RNA molecules and the separation not only depends on the interaction of the (reversed) phase but also on the possibility of molecules to get inside the pores of the matrix and thus provide a further retention effect. Without being limited thereto, e.g. a pore size for the reversed phase of about 2000 A to about 5000 A, more preferably of about 2500 to about 4000, most preferably of about 3500 to about 4500 A, may thus be used to separate larger RNA molecules, e.g. RNA molecules of 100 to 10000, more preferably 500 to 10000 nucleotides or base pairs, even more preferably 800 to 5000 nucleotides or base pairs and even more preferably 800 to 2000 nucleotides or base pairs. Alternatively, without being limited thereto, a pore size for the reversed phases of about 1000 A to about 2500 A, more preferably of about 1000 A to about 2000 A, and most preferably of about 1000 A to 1200 A may be used to separate smaller RNA molecules, e.g. RNA molecules of about 30- 1000, 50-1000 or 100-1000 or 20-200, 20-100, 20-50 or 20-30 nucleotides may also be separated in this way. In general, any material known to be used as reverse phase stationary phase, in particular any polymeric material may be used, if that material can be provided in porous form. The stationary phase may be composed of organic and/or inorganic material. Examples for polymers are (non-alkylated) polystyrenes, (non-alkylated) polystyrenedivinylbenzenes, silica gel, silica gel modified with non-polar residues, particularly silica gel modified with alkyl containing residues, more preferably with butyl-, octyl and/or octadecyl containing residues, silica gel modified with phenylic residues, polymethacrylates, etc. or other materials suitable e.g. for gel chromatography or other chromatographic methods as mentioned above, such as dextran, including e.g. Sephadex® and Sephacryl® materials, agarose, dextran/agarose mixtures, polyacrylamide, etc.

Most preferably, the material for the reversed phase is a porous polystyrene polymer, a (non- alkylated) (porous) polystyrenedivinylbenzene polymer, porous silica gel, porous silica gel modified with non-polar residues, particularly porous silica gel modified with alkyl containing residues, more preferably with butyl-, octyl and/or octadecyl containing residues, porous silica gel modified with phenylic residues, porous polymethacrylates, wherein in particular a porous polystyrene polymer or a non-alkylated (porous)

polystyrenedivinylbenzene may be used. Stationary phases with polystyrenedivinylbenzene are known per se and may be used.

A non-alkylated porous polystyrenedivinylbenzene is one which, without being limited thereto, may have in particular a particle size of 8.0 ± 1.5 μιη, in particular 8.0 ± 0.5 μιη, and a pore size of 1000- 1500 A, in particular 1000-1200 A or 3500-4500 A and most preferably a particle size of 4000 A.

This stationary phase described in greater detail above is conventionally located in a column. V2A steel is conventionally used as the material for the column, but other materials may also be used for the column provided they are suitable for the conditions prevailing during HPLC. Conventionally the column is straight. It is favorable for the HPLC column to have a length of 5 cm to 100 cm and a diameter of 4 mm to 50 cm. Columns may in particular have the following dimensions: 25 cm long and 20 mm in diameter or 25 cm long and 50 mm in diameter, or 25 cm long and 10 cm in diameter or any other dimension with regard to length and diameter, which is suitable for preparative recovery of RNA, even lengths of several metres and also larger diameters being feasible in the case of upscaling. The dimensions are here geared towards what is technically possible with liquid chromatography.

Selection of the mobile phase depends on the type of separation desired. This means that the mobile phase established for a specific separation, as may be known for example from the prior art, cannot be straightforwardly applied to a different separation problem with a sufficient prospect of success. For each separation problem, the ideal elution conditions, in particular the mobile phase used, have to be determined by empirical testing. In a preferred embodiment of the HPLC method, a mixture of an aqueous solvent and an organic solvent is used as the mobile phase for eluting the RNA. It is favorable for a buffer to be used as the aqueous solvent which has in particular a pH of 6.0-8.0, for example of about 7, for example. 7.0; preferably the buffer is triethylammonium acetate (TEAA), particularly preferably a 0.02 M to 0.5 M, in particular 0.08 M to 0.12 M, very particularly an about 0.1 M TEAA buffer, which also acts as a counterion to the RNA in the ion pair method.

In a preferred embodiment, the organic solvent which is used in the mobile phase comprises acetonitrile, methanol, ethanol, 1-propanol, 2-propanol, ethyl acetate, tetrahydrofuran and acetone or a mixture thereof, very particularly preferably acetonitrile. With these organic solvents, in particular acetonitrile, purification of the RNA proceeds in a particularly favourable manner.

In a particularly preferred embodiment of the method according to the invention, the mobile phase is a mixture of 0.1 M triethylammonium acetate, pH 7, and acetonitrile.

It has proven favorable for the mobile phase to contain 5.0 vol.% to 25.0 vol.% organic solvent, relative to the mobile phase, and for this to be made up to 100 vol.% with the aqueous solvent. Typically, in the event of gradient separation, the proportion of organic solvent is increased, in particular by at least 10%, more preferably by at least 50% and most preferably by at least 100%, optionally by at least 200%, relative to the initial vol.% in the mobile phase. In a preferred embodiment, the proportion of organic solvent in the mobile phase amounts in the course of HPLC separation to 3 to 9, preferably 4 to 7.5, in particular 5.0 vol.%, in each case relative to the mobile phase. More preferably, the proportion of organic solvent in the mobile phase is increased in the course of HPLC separation from 3 to 9, in particular 5.0 vol.% to up to 20.0 vol.%, in each case relative to the mobile phase. Still more preferably, the method is performed in such a way that the proportion of organic solvent in the mobile phase is increased in the course of HPLC separation from 6.5 to 8.5, in particular 7.5 vol.%, to up to 17.5 vol.%, in each case relative to the mobile phase.

It has proven even more particularly favorable for the mobile phase to contain 7.5 vol.% to 17.5 vol.%) organic solvent, relative to the mobile phase, and for this to be made up to 100 vol.%) with the aqueous buffered solvent. The elution may proceed isocratically or by means of gradient separation. In isocratic separation, elution of the RNA proceeds with a single eluent or a constant mixture of a plurality of eluents, wherein the solvents described above in detail may be used as eluent. For gradient separation, the composition of the eluent is varied by means of a gradient program. The equipment necessary for gradient separation is known to a person skilled in the art. Gradient elution may here proceed either on the low pressure side by mixing chambers or on the high pressure side by further pumps.

Preferably, the proportion of organic solvent, as described above, is increased relative to the aqueous solvent during gradient separation. The above-described agents may here be used as the aqueous solvent and the likewise above-described agents may be used as the organic solvent. For example, the proportion of organic solvent in the mobile phase may be increased in the course of HPLC separation from 5.0 vol.% to 20.0 vol.%, in each case relative to the mobile phase. In particular, the proportion of organic solvent in the mobile phase may be increased in the course of HPLC separation from 7.5 vol.% to 17.5 vol.%, in particular 9.5 to 14.5 vol.%, in each case relative to the mobile phase.

The flow rate of the eluent is so selected that good separation of the RNA from the other constituents contained in the sample to be investigated takes place. The eluent flow rate may amount to from 1 ml/min to several litres per minute (in the case of upscaling), in particular about 1 to 1000 ml/min, more preferably 5 ml to 500 ml/min, even more preferably more than 100 ml/min, depending on the type and scope of the upscaling. This flow rate may be established and regulated by the pump. Detection proceeds favorably with a UV detector at 254 nm, wherein a reference measurement may be made at 600 nm. However, any other detection method may

alternatively be used, with which the R A described above in greater detail may be detected in satisfactory and reliable manner.

In preferred embodiments, the HPLC is performed as described in WO2008077592 in order to obtain the HPLC sample comprising the RNA.

Sample comprising RNA and triethylammonium acetate (and optionally acetonitrile): The term "sample" as used herein describes a sample or fraction obtained after purifying RNA via HPLC as set out above. An HPLC sample will therefore be essentially free of the

contaminants removed by HPLC, e.g. small RNA contaminants (e.g. abortive RNA sequences), proteins, residual nucleotides and the residuals of the DNA template after its hydrolysis. The HPLC sample will comprise RNA molecules, as well triethylammonium acetate (and optionally also acetonitrile) as remnants / contaminants from the mobile phase described above. The focus of the present invention is on the removal of the contaminant triethylammonium acetate since this is generally more difficult to achieve compared to the removal of the acetonitrile. Purification/purifying : The terms "purification", "purified" or "purifying" as used herein mean that RNA is separated and/or isolated from unwanted contaminants.

For example, HPLC is used to purify RNA from by-products and components of the RNA production (e.g. chemical synthesis, RNA in vitro transcription). In the context of RNA in vitro transcription, the RNA is separated and/or isolated by HPLC from the by-products and the components of the RNA in vitro transcription reaction present in the sample after the RNA in vitro transcription reaction is complete. Thus, after HPLC purification, the purified RNA sample has a higher purity than the RNA-containing sample after the production and prior to purification.

The HPLC sample can then further be purified by a method as claimed herein by removing triethylammonium acetate (and optionally also the acetonitrile), which was introduced during the HPLC purification. Thus, after purification according to the present invention, the purified HPLC sample has a higher RNA purity than the HPLC sample prior to purification, i.e. the amount of TEAA (and optionally also of the acetonitrile) is substantially lower, if not completely absent.

Preferably, the content of acetonitrile will be lower than 25%, more preferably lower than 20%, lower than 15%, lower than 10%, lower than 5%, lower than 2.5%, lower than 1%. Most preferably, the purified HPLC sample will be substantially free of acetonitrile.

Preferably, the content of TEAA will be lower than 5 g per g RNA, more preferably lower than 4 g per g RNA, lower than 3 g per g RNA, lower than 2 g per g RNA, and even more preferably lower than 1 g per g RNA. Most preferably, the purified HPLC sample will be substantially free of acetonitrile.

Substantially free of: The term "substantially free of is used to describe a sample from which virtually all of a specific substance has been removed. Specifically, a sample is designated to be substantially free of acetonitrile at a concentration of < about 1000 ppm (preferably at a concentration of < about 400 ppm), substantially free of TEAA at a concentration of < about 1.0 g per g RNA (preferably at a concentration of < about 0.5 g per g RNA), and substantially free of t-Butanol at a concentration of < about 0.10 g per g RNA (preferably at a

concentration of < about 0.05 g per g RNA). If reference is made herein to a substance being removed, this means that the sample after removal of the substance is substantially free of this substance.

Formulation/ formulating : The terms„formulation" and„formulating" describes the process in which different components, including an active component, are combined to produce a final medicinal or therapeutic product. The term formulation is often used synonymous with a dosage form in the meaning of the dosage form ready for use. However, the term formulation can also refer to the state of each component when these components are combined to result in a medicinal or therapeutic product. For the present case, this means that the RNA as the active component is formulated in a specific way prior to combining the RNA with the further components of the dosage form to be applied. Lyophilization is often a preferred formulation for therapeutic components, such as nucleic acids and in particular RNAs, because the long-term stability of many materials increases in the lyophilized state. In the present application, "formulating" RNA means in particular providing the RNA in a lyophilized form prepared according to the inventive method as one component of a dosage form. When preparing the final therapeutic dosage form comprising the RNA, the lyophilized RNA formulated according to the present invention may then be dissolved in ddH₂0 or buffer or the like during this preparation process.

Lyoprotectants and/or complexing agents (such as e.g. cationic or polycationic compounds) may be added prior to the final lyophilization when generating formulated RNA, which may accordingly be provided as complexed RNA and/or as RNA additionally comprising a lyoprotectant (e.g. trehalose).

Rotary vacuum concentration: Rotary vacuum concentration (RVC) is based on lowering the pressure above a bulk liquid, which in turn lowers the boiling points of the component liquids in it. Generally, the component liquids of interest in applications of rotary evaporation are solvents that one desires to remove from a sample after an extraction, such as following a natural product isolation or a step in an organic synthesis. Liquid solvents can be removed without excessive heating of what are often complex and sensitive solvent-solute

combinations.

Rotary evaporation is most often and conveniently applied to separate "low boiling" solvents such a n-hexane or ethyl acetate from compounds which are solid at room temperature and pressure. However, careful application also allows removal of a solvent from a sample containing a liquid compound if there is minimal co-evaporation (azeotropic behavior), and a sufficient difference in boiling points at the chosen temperature and reduced pressure.

Rotary vacuum evaporation using centrifuges in particular enables high throughput settings, i.e. the processing of many samples in parallel, for example in multi well plates.

A liquid sample subjected to RVC may be referred to as "concentrated" sample since the sample is more concentrated after RVC then prior to this step.

Lyophilization/lyophilizing : The terms "lyophilization" and "lyophilizing" describe a dehydration process typically used to preserve a perishable material or make the material more convenient for transport. It is also referred to as freeze-drying or cryodessication.

Lyophilization works by freezing the material and then reducing the surrounding pressure to allow the frozen water in the material to sublimate directly from the solid phase to the gas phase. It usually comprises three stages: freezing, primary drying, and secondary drying.

During freezing, the material is cooled to below its triple point, the lowest temperature at which the solid and liquid phases of the material can coexist. This ensures that sublimation rather than melting will occur in the following steps. Larger crystals are easier to freeze-dry. To produce larger crystals, the product should be frozen slowly or can be cycled up and down in temperature. This cycling process is called annealing. Usually, the freezing temperatures are between -50 °C and -80 °C (-58 °F and -112 °F). The freezing phase is the most critical in the whole freeze-drying process. During the primary drying phase, the pressure is lowered to the range of a few millibars, and enough heat is supplied to the material for the ice to sublime. The amount of heat necessary can be calculated using the sublimating molecules' latent heat of sublimation. In this initial drying phase, about 95% of the water in the material is sublimated. This phase may be slow (in some instances even up to several days), because, if too much heat is added, the material's structure could be altered. In this phase, pressure is controlled through the application of partial vacuum. The vacuum speeds up the sublimation, making it useful as a deliberate drying process.

The secondary drying phase aims to remove unfrozen water molecules, since the ice was removed in the primary drying phase. This part of the freeze-drying process is governed by the material's adsorption isotherms. In this phase, the temperature is raised higher than in the primary drying phase, and can even be above 0 °C, to break any physico-chemical interactions that have formed between the water molecules and the frozen material. Usually the pressure is also lowered in this stage to encourage desorption.

At the end of this procedure, the final residual water content in the product is extremely low, around 1% to 4%, which leads to prolonged shelf life at room temperature, so long as the product is protected from reabsorption of moisture.

Resistivity: The term„resistivity" describes a setting during lyophilization, specifically during primary drying, to avoid melting of samples. If a sensor drops under a designated value, heating is switched off and the freezing of samples starts again. Co-solvent: A solvent is a substance that dissolves a solute (a chemically distinct liquid, solid or gas, e.g. R A) resulting in a solution. A solvent is usually a liquid but can also be a solid or a gas. The quantity of solute that can dissolve in a specific volume of solvent varies with temperature. The term„co-solvent" describes a second solvent that is added (usually concomitantly) with a first solvent, for example pre-mixed. For the present invention, the mixture of the first solvent ddH₂0 and the second solvent (i.e. the "co-solvent") t-Butanol is particularly preferred.

Freeze point: The "freeze point" or "freezing point" or "crystallization point" is the temperature at which a liquid changes state from liquid to solid.

Multi well plate: The term "multi well plate" describes a plate with 6, 12, 24, 48, 96, 192, 384, or 1536 wells for sample containment. Some multi well plates have 3456 or 9600 wells. Multiwell plates may be deep well plates, which typically can hold larger volumes than standard well plates, e.g.1 to 2 ml volumes in a 96 well plate format. Multi well plates most commonly are made from plastic materials, e.g. polystyrene, polypropylene, polycarbonate, cyclo-olefms, but can also be made from glass or quartz, or from plastic coated with a thin layer of glass or quartz. Multi well plates allow for processing of several samples in parallel, keeping conditions and handling times consistent for comparison between samples. They make possible high throughput settings, e.g. high throughput screening, sequencing, purification, formulation, etc.

High throughput setting, robot-assisted setting: The term "high throughput setting" describes any setting in which several, often up to thousands, of samples are quickly processed in parallel using multi well plates, with consistent experimental or production conditions. This leads to the advantages of consistency of results or products across large numbers of samples and shortening of experimental and production processes.

High throughput processes are often handled in "robot-assisted settings", in which some or all of the sample handling is performed by robots, data processing and control software, and other liquid handling devices, affording even greater sample processing speed and precision compared to entirely manually operated high throughput processes. Dilution/ diluting : The terms "dilution" and "diluting" describe the reduction of the concentration of a compound, e.g. RNA, in a liquid sample, typically by increasing the volume of the sample by addition of a diluent, e.g. water. Readily soluble: The solubility of lyophilized RNA samples can be measured by determining the RNA content of samples that had been re-dissolved in water under identical conditions, and the coefficient of variation is determined. The ratio of samples that show a higher coefficient of variation than 5 % is used as a measure for the "insolubility" of the RNA samples ("CV"), i.e. a sample with a CV of < 5% is considered "readily soluble".

Therapy/therapeutic: The term "therapy" is used to designate the administration of nucleic acids, e.g. RNA, to a subject in need thereof, for example for the purpose of immunization or prophylaxis or treatment of a medical condition. The term "therapeutic" is used to describe nucleic acids, e.g. RNA, that are suitable for such a therapy. Such therapeutic nucleic acids comprise e.g. mRNA molecules encoding antigens for use as vaccines (Fotin-Mleczek et al. 2012. J. Gene Med. 14(6):428-439), RNA molecules for replacement therapies (Thess, et al. Molecular Therapy (2015)), RNA molecules for the provision of RNA-coded antibodies for passive immunization or cancer immunotherapies (e.g., WO2008083949), noncoding immunostimulatory RNA molecules (e.g., WO2009095226) and other noncoding RNAs such as microRNAs or RNAs suitable for genome editing (e.g., CRISPR/Cas9 guide RNAs).

Chemical synthesis of RNA: Chemical synthesis of relatively short fragments of

oligonucleotides with defined chemical structure provides a rapid and inexpensive access to custom-made oligonucleotides of any desired sequence. Whereas enzymes synthesize DNA and RNA only in the 5' to 3' direction, chemical oligonucleotide synthesis does not have this limitation, although it is most often carried out in the opposite, i.e. the 3' to 5' direction.

Currently, the process is implemented as solid-phase synthesis using the phosphoramidite method and phosphoramidite building blocks derived from protected nucleosides (A, C, G, and U), or chemically modified nucleosides.

To obtain the desired oligonucleotide, the building blocks are sequentially coupled to the growing oligonucleotide chain on a solid phase in the order required by the sequence of the product in a fully automated process. Upon the completion of the chain assembly, the product is released from the solid phase to the solution, deprotected, and collected. The occurrence of side reactions sets practical limits for the length of synthetic oligonucleotides (up to about 200 nucleotide residues), because the number of errors increases with the length of the

oligonucleotide being synthesized. Products are often isolated by HPLC to obtain the desired oligonucleotides in high purity.

Chemically synthesized oligonucleotides find a variety of applications in molecular biology and medicine. They are most commonly used as antisense oligonucleotides, small interfering RNA, primers for DNA sequencing and amplification, probes for detecting complementary DNA or RNA via molecular hybridization, tools for the targeted introduction of mutations and restriction sites, and for the synthesis of artificial genes.

DNA: DNA is the usual abbreviation for deoxyribonucleic acid. It is a nucleic acid molecule, i.e. a polymer consisting of nucleotide monomers. These nucleotides are usually deoxy- adenosine-monophosphate, deoxy-thymidine-monophosphate, deoxy-guanosine- monophosphate and deoxy-cytidine-monophosphate monomers or analogs thereof which are - by themselves - composed of a sugar moiety (deoxyribose), a base moiety and a phosphate moiety, and polymerize by a characteristic backbone structure. The backbone structure is, typically, formed by phosphodiester bonds between the sugar moiety of the nucleotide, i.e. deoxyribose, of a first and a phosphate moiety of a second, adjacent monomer. The specific order of the monomers, i.e. the order of the bases linked to the sugar/phosphate-backbone, is called the DNA-sequence. DNA may be single-stranded or double-stranded. In the double stranded form, the nucleotides of the first strand typically hybridize with the nucleotides of the second strand, e.g. by A/T -base-pairing and G/C-base-pairing. In vitro transcription: The term "in vitro transcription" relates to a process wherein RNA is synthesized in a cell-free system (in vitro). DNA is used as template for the generation of RNA transcripts. The promoter (also referred to herein as "promoter sequence") for controlling in vitro transcription can be any promoter for any DNA dependent RNA polymerase (referred to herein also as "RNA polymerase"). Particular examples of DNA dependent RNA polymerases are the T7, T3, and SP6 RNA polymerases. A DNA template for in vitro RNA transcription may be obtained by cloning of a nucleic acid, in particular cDNA corresponding to the target RNA to be in vitro transcribed, and introducing it into an appropriate DNA for in vitro transcription, for example into plasmid DNA. The cDNA may be obtained by reverse transcription of mRNA, chemical synthesis, or oligonucleotide cloning. Moreover, the DNA template for in vitro RNA synthesis may also be obtained by gene synthesis.

Methods for in vitro transcription are known in the art (Geall et al. (2013) Semin. Immunol. 25(2): 152-159; Brunelle et al. (2013) Methods Enzymol. 530: 101-14). Reagents used in said method typically include:

1) a DNA template with a promoter sequence that has a high binding affinity for its respective RNA polymerase such as bacteriophage-encoded RNA polymerases;

2) ribonucleoside triphosphates (NTPs) for the four bases (adenine, cytosine, guanine and uracil);

3) optionally a cap analog as defined below (e.g. m7G(5')ppp(5')G (m7G));

4) a DNA-dependent RNA polymerase capable of binding to the promoter sequence within the DNA template (e.g. T7, T3 or SP6 RNA polymerase);

5) optionally a ribonuclease (RNase) inhibitor to inactivate any contaminating RNase;

6) optionally a pyrophosphatase to degrade pyrophosphate, which may inhibit transcription;

7) MgCl₂, which supplies Mg²⁺ ions as a co-factor for the polymerase;

8) a buffer to maintain a suitable pH value, which can also contain antioxidants (e.g. DTT) and polyamines such as spermidine at optimal concentrations. The (transcription) buffer may be selected from the group consisting of 4-(2-hydroxyethyl)-l- piperazineethanesulfonic acid (HEPES) and tris(hydroxymethyl)aminomethane (Tris).

Preferably the buffer is used at a concentration from 10 to 100 mM, 10 to 75 mM, 10 to 50 mM, 10 to 40 mM, 10 to 30 mM or 10 to 20 mM. The pH value of the buffer can be adjusted with, for example, NaOH, KOH or HC1. Preferably the buffer has a pH value from 6 to 8.5, from 6.5 to 8.0, from 7.0 to 7.5, even more preferred 7.5. Most preferred is a buffer selected from the group consisting of 80 mM HEPES/KOH, pH 7.5 and 40 mM Tris/HCl, pH 7.5.

The RNA polymerase is preferably selected from the group consisting of T3, T7 and SP6 RNA polymerase. Preferably, the concentration of the RNA polymerase is from about 1 to 100 nM, 1 to 90 nM, 1 to 80 nM, 1 to 70 nM, 1 to 60 nM, 1 to 50 nM, 1 to 40 nM, 1 to 30 nM, 1 to 20 nM, or about 1 to 10 nM. Even more preferred, the concentration of the RNA polymerase is from about 10 to 50 nM, 20 to 50 nM, or 30 to 50 nM. Most preferred is a RNA polymerase concentration of about 40 nM. The person skilled in the art will understand that the choice of the RNA polymerase concentration is influenced by the concentration of the DNA template. Therefore, in specific embodiments the concentration of the RNA polymerase is between 1 and 1000 U/μg template DNA, preferably between 10 and 100 U/μg DNA, particularly if plasmid DNA is used as template DNA. The in vitro transcription is preferably performed in the presence of pyrophosphatase.

Preferably, the concentration of the pyrophosphatase is from about 1 to 20 units/ml, 1 to 15 units/ml, 1 to 10 units/ml, 1 to 5 units/ml, or 1 to 2.5 units/ml. Even more preferred the concentration of the pyrophosphatase is about 5 units/ml. The in vitro transcription reaction mixture preferably comprises Mg²⁺ ions. Preferably, the Mg²⁺ ions are provided in the form of MgCl₂ or Mg(OAc)₂. Preferably, the initial free Mg²⁺ concentration is from about 1 to 100 mM, 1 to 75 mM, 1 to 50 mM, 1 to 25 mM, or 1 to 10 mM. Even more preferred the initial free Mg²⁺ concentration is from about 10 to 30 mM or about 15 to 25 mM. Most preferred is an initial free Mg²⁺ concentration of about 24 mM. The person skilled in the art will understand that the choice of the Mg²⁺ concentration is influenced by the initial total NTP concentration.

The in vitro transcription reaction mixture preferably comprises a reducing agent

(antioxidant) to keep the RNA polymerase in its active state. Preferably, the reducing agent is selected from the group consisting of dithiothreitol (DTT), dithioerythritol (DTE), Tris(2- carboxyethyl)phosphine (TCEP) and β-mercaptoethanol. Preferably the concentration of the reducing reagent is from about 1 to 50 mM, 1 to 40 mM, 1 to 30 mM, or 1 to 20 mM, or 1 to 10 mM. Even more preferred the concentration of the reducing reagent is from 10 to 50 mM or 20 to 40 mM. Most preferred is a concentration of 40 mM of DTT.

Further, the in vitro transcription reaction mixture preferably comprises a polyamine.

Preferably, the polyamine is selected from the group consisting of spermine and spermidine. Preferably the concentration of the polyamine is from about 1 to 25 mM, 1 to 20 mM, 1 to 15 mM, 1 to 10 mM, 1 to 5 mM, or about 1 to 2.5 mM. Even more preferred the concentration of the polyamine is about 2 mM. Most preferred is a concentration of 2 mM of spermidine.

The in vitro transcription reaction mixture preferably comprises a ribonuclease inhibitor. Preferably, the concentration of the ribonuclease inhibitor is from about 1 to 500 units/ml, 1 to 400 units/ml, 1 to 300 units/ml, 1 to 200 units/ml, or 1 to 100 units/ml. Even more preferred the concentration of the ribonuclease inhibitor is about 200 units/ml.

The total NTP concentration in the in vitro transcription reaction mixture may be between 1 and 100 mM, preferably between 10 and 50 mM, and most preferably between 10 and 20 mM. The term total nucleotide concentration means the total concentration of NTPs, e.g. the sum of the concentrations of ATP, GTP, CTP, UTP, and/or cap analog present initially in the in vitro transcription when the various components of the reaction have been assembled in the final volume for carrying out the in vitro transcription reaction. Naturally, as the reaction proceeds, the nucleotides will be incorporated into the RNA molecule and consequently the total nucleotide concentration will be progressively reduced from its initial value. In this context it is particularly preferred that the single nucleotides are provided in a concentration between 0.1 and 10 mM, preferably between 1 and 5 mM and most preferably in a

concentration of 4 mM.

In case a 5' cap as defined below has to be generated at the 5 '-end of the RNA, the in vitro transcription reaction mixture preferably further comprises a cap analog. In this context the concentration of GTP is preferably reduced compared to the other nucleotides (ATP, CTP and UTP). Preferably the cap analog is added with an initial concentration in the range of about 1 to 20 mM, 1 to 17.5 mM, 1 to 15 mM, 1 to 12.5 mM, 1 to 10 mM, 1 to 7.5 mM. Most preferably the cap analog is added in a concentration of 5.8 mM and the GTP concentration is reduced to a concentration of 1.45 mM whereas ATP, CTP and UTP are comprised in the reaction in a concentration of 4 mM each. The ribonucleoside triphosphates (NTPs) GTP, ATP, CTP and UTP or analogs thereof may be provided with a monovalent or divalent cation as counterion. Preferably the monovalent cation is selected from the group consisting of Li⁺ , Na⁺ , K⁺ , NH4⁺ or tris(hydroxymethyl)- amino methane (Tris). Preferably, the divalent cation is selected from the group consisting of Mg²⁺, Ba²⁺ and Mn²⁺.

As outlined in detail below, a part or all of at least one ribonucleoside triphosphate in the in vitro transcription reaction mixture may be replaced with a modified nucleoside triphosphate (as defined herein). Preferably, said modified nucleoside triphosphate is selected from the group consisting of pseudouridine-5 '-triphosphate, l-methylpseudouridine-5 '-triphosphate, 2- thiouridine-5 '-triphosphate, 4-thiouridine-5 '-triphosphate and 5-methylcytidine-5 '- triphosphate. Other modified nucleotides which can be used in this context are listed below.

Following transcription, the DNA template can optionally be removed using methods known in the art comprising DNase I digestion. In this context, it is particularly preferred to add 6 μΐ DNAse I (1 mg/ml) and 0.2 μΐ CaCb solution (0.1 M) / μg DNA template to the transcription reaction, and to incubate it for at least 3 h at 37°C.

Nucleoside/Nucleotide : Nucleosides are glycosylamines that correspond to nucleotides without a phosphate group. A nucleoside consists simply of a nucleobase (also termed a nitrogenous base) and a 5-carbon sugar (either ribose or deoxyribose), whereas a nucleotide is composed of a nucleobase, a five-carbon sugar, and one or more phosphate groups. In a nucleoside, the base is bound to either ribose or deoxyribose via a beta-glycosidic linkage. Examples of nucleosides include cytidine, uridine, adenosine, guanosine, thymidine and inosine.

Modified nucleoside triphosphate: The term "modified nucleoside triphosphate" as used herein refers to chemical modifications comprising backbone modifications as well as sugar modifications or base modifications. These modified nucleoside triphosphates are also termed herein as (nucleotide) analogs, modified nucleosides/nucleotides or nucleotide/nucleoside modifications.

In this context, the modified nucleoside triphosphates as defined herein are nucleotide analogs/modifications, e.g. backbone modifications, sugar modifications or base

modifications. A backbone modification in connection with the present invention is a modification, in which phosphates of the backbone of the nucleotides are chemically modified. A sugar modification in connection with the present invention is a chemical modification of the sugar of the nucleotides. Furthermore, a base modification in connection with the present invention is a chemical modification of the base moiety of the nucleotides. In this context nucleotide analogs or modifications are preferably selected from nucleotide analogs which are applicable for transcription and/or translation.

Sugar Modifications: The modified nucleosides and nucleotides, which may be used in the context of the present invention, can be modified in the sugar moiety. For example, the 2' hydroxyl group (OH) can be modified or replaced with a number of different "oxy" or "deoxy" substituents. Examples of "oxy" -2' hydroxyl group modifications include, but are not limited to, alkoxy or aryloxy (-OR, e.g., R = H, alkyl, cycloalkyl, aryl, aralkyl, heteroaryl or sugar); polyethyleneglycols (PEG), -0(CH2CH2o)nCH2CH2OR; "locked" nucleic acids (LNA) in which the 2' hydroxyl is connected, e.g., by a methylene bridge, to the 4' carbon of the same ribose sugar; and amino groups (-O-amino, wherein the amino group, e.g., NRR, can be alkylamino, dialkylamino, heterocyclyl, arylamino, diarylamino, heteroarylamino, or diheteroaryl amino, ethylene diamine, polyamino) or aminoalkoxy. "Deoxy" modifications include hydrogen, amino (e.g. NH2; alkylamino, dialkylamino, heterocyclyl, arylamino, diaryl amino, heteroaryl amino, diheteroaryl amino, or amino acid); or the amino group can be attached to the sugar through a linker, wherein the linker comprises one or more of the atoms C, N, and O. The sugar group can also contain one or more carbons that possess the opposite

stereochemical configuration than that of the corresponding carbon in ribose. Thus, a modified nucleotide can include nucleotides containing, for instance, arabinose as the sugar.

Backbone Modifications: The phosphate backbone may further be modified in the modified nucleosides and nucleotides. The phosphate groups of the backbone can be modified by replacing one or more of the oxygen atoms with a different substituent. Further, the modified nucleosides and nucleotides can include the full replacement of an unmodified phosphate moiety with a modified phosphate as described herein. Examples of modified phosphate groups include, but are not limited to, phosphorothioate, phosphoroselenates, borano phosphates, borano phosphate esters, hydrogen phosphonates, phosphoroamidates, alkyl or aryl phosphonates and phosphotriesters. Phosphorodithioates have both non-linking oxygens replaced by sulfur. The phosphate linker can also be modified by the replacement of a linking oxygen with nitrogen (bridged phosphoroamidates), sulfur (bridged phosphorothioates) and carbon (bridged methylene-phosphonates).

Base Modifications: The modified nucleosides and nucleotides can further be modified in the nucleobase moiety. Examples of nucleobases found in RNA include, but are not limited to, adenine, guanine, cytosine and uracil. For example, the nucleosides and nucleotides described herein can be chemically modified on the major groove face. In some embodiments, the major groove chemical modifications can include an amino group, a thiol group, an alkyl group, or a halo group.

The nucleotide analogs/modifications may be selected from base modifications, which are preferably selected from 2-amino-6-chloropurineriboside-5'-triphosphate, 2-Aminopurine- riboside-5 '-triphosphate; 2-aminoadenosine-5'-triphosphate, 2'-Amino-2'-deoxycytidine- triphosphate, 2-thiocytidine-5'-triphosphate, 2-thiouridine-5'-triphosphate, 2'- Fluorothymidine-5 '-triphosphate, 2'-0-Methyl inosine-5 '-triphosphate 4-thiouridine-5'- triphosphate, 5-aminoallylcytidine-5'-triphosphate, 5-aminoallyluridine-5'-triphosphate, 5- bromocytidine-5 '-triphosphate, 5-bromouridine-5'-triphosphate, 5-Bromo-2'-deoxycytidine-5'- triphosphate, 5-Bromo-2'-deoxyuridine-5'-triphosphate, 5-iodocytidine-5'-triphosphate, 5- Iodo-2'-deoxycytidine-5'-triphosphate, 5-iodouridine-5'-triphosphate, 5-Iodo-2'-deoxyuridine- 5 '-triphosphate, 5-methylcytidine-5'-triphosphate, 5-methyluridine-5'-triphosphate, 5- Propynyl-2'-deoxycytidine-5'-triphosphate, 5-Propynyl-2'-deoxyuridine-5'-triphosphate, 6- azacytidine-5'-triphosphate, 6-azauridine-5'-triphosphate, 6-chloropurineriboside-5'- triphosphate, 7-deazaadenosine-5 '-triphosphate, 7-deazaguanosine-5 '-triphosphate, 8- azaadenosine-5 '-triphosphate, 8-azidoadenosine-5'-triphosphate, benzimidazole-riboside-5'- triphosphate, Nl-methyladenosine-5'-triphosphate, Nl-methylguanosine-5'-triphosphate, N6- methyladenosine-5'-triphosphate, 06-methylguanosine-5'-triphosphate, pseudouridine-5'- triphosphate, or puromycin-5 '-triphosphate, xanthosine-5'-triphosphate. Particular preference is given to nucleotides for base modifications selected from the group of base-modified nucleotides consisting of 5-methylcytidine-5'-triphosphate, 7-deazaguanosine-5'-triphosphate, 5-bromocytidine-5'-triphosphate, and pseudouridine-5'-triphosphate. Modified nucleosides include pyridin-4-one ribonucleoside, 5-aza-uridine, 2-thio-5-aza- uridine, 2-thiouridine, 4-thio-pseudouridine, 2-thio-pseudouridine, 5-hydroxyuridine, 3- methyluridine, 5-carboxymethyl-uridine, 1-carboxymethyl-pseudouridine, 5-propynyl- uridine, 1-propynyl-pseudouridine, 5-taurinomethyluridine, 1-taurinomethyl-pseudouridine, 5-taurinomethyl-2-thio-uridine, l-taurinomethyl-4-thio-uridine, 5-methyl-uridine, 1-methyl- pseudouridine, 4-thio- 1-methyl-pseudouridine, 2-thio-l-methyl-pseudouridine, 1 -methyl- 1- deaza-pseudouridine, 2-thio- 1 -methyl- 1 -deaza-pseudouridine, dihydrouridine,

dihydropseudouridine, 2-thio-dihydrouridine, 2-thio-dihydropseudouridine, 2- methoxyuridine, 2-methoxy-4-thio-uridine, 4-methoxy-pseudouridine, and 4-methoxy-2-thio- pseudouridine. Modified nucleosides further include 5-aza-cytidine, pseudoisocytidine, 3-methyl-cytidine, N4-acetylcytidine, 5-formylcytidine, N4-methylcytidine, 5-hydroxymethylcytidine, 1-methyl- pseudoisocytidine, pyrrolo-cytidine, pyrrolo-pseudoisocytidine, 2-thio-cytidine, 2-thio-5- methyl- cytidine, 4-thio-pseudoisocytidine, 4-thio-l-methyl-pseudoisocytidine, 4-thio-l- methyl-l-deaza-pseudoisocytidine, 1 -methyl- 1-deaza-pseudoisocytidine, zebularine, 5-aza- zebularine, 5-methyl-zebularine, 5-aza-2-thio-zebularine, 2-thio-zebularine, 2-methoxy- cytidine, 2-methoxy-5-methyl-cytidine, 4-methoxy-pseudoisocytidine, and 4-methoxy-l- methyl-pseudoisocytidine .

Modified nucleosides also include 2-aminopurine, 2, 6-diaminopurine, 7-deaza-adenine, 7- deaza-8-aza-adenine, 7-deaza-2-aminopurine, 7-deaza-8-aza-2-aminopurine, 7-deaza-2,6- diaminopurine, 7-deaza-8-aza-2, 6-diaminopurine, 1-methyladenosine, N6-methyladenosine, N6-isopentenyladenosine, N6-(cis-hydroxyisopentenyl)adenosine, 2-methylthio-N6-(cis- hydroxyisopentenyl) adenosine, N6-glycinylcarbamoyladenosine, N6- threonylcarbamoyladenosine, 2-methylthio-N6-threonyl carbamoyladenosine, N6,N6- dimethyladenosine, 7-methyladenine, 2-methylthio -adenine, and 2-methoxy-adenine.

In other embodiments, modified nucleosides include inosine, 1 -methyl- ino sine, wyosine, wybutosine, 7-deaza-guanosine, 7-deaza-8-aza-guanosine, 6-thio-guanosine, 6-thio-7-deaza- guanosine, 6-thio-7-deaza-8-aza-guanosine, 7-methyl-guanosine, 6-thio-7-methyl-guanosine, 7-methylinosine, 6-methoxy-guanosine, 1-methylguanosine, N2-methylguanosine, N2,N2- dimethylguanosine, 8-oxo-guanosine, 7-methyl-8-oxo-guanosine, l-methyl-6-thio-guanosine, N2-methyl-6-thio-guanosine, and N2,N2-dimethyl-6-thio-guanosine.

The nucleotide can be modified on the major groove face and can include replacing hydrogen on C-5 of uracil with a methyl group or a halo group. In specific embodiments, a modified nucleoside is 5 '-O-(l-Thiophosphate)- Adenosine, 5'-0-(l-Thiophosphate)-Cytidine, 5'-0-(l- Thiophosphate)-Guanosine, 5'-0-(l-Thiophosphate)-Uridine or 5'-0-(l-Thiophosphate)- Pseudouridine.

The modified nucleotides may include nucleoside modifications selected from 6-aza-cytidine, 2-thio-cytidine, a-thio-cytidine, Pseudo-iso-cytidine, 5-aminoallyl-uridine, 5-iodo-uridine, Nl-methyl-pseudouridine, 5,6-dihydrouridine, a-thio -uridine, 4-thio-uridine, 6-aza-uridine, 5- hydroxy-uridine, deoxy-thymidine, 5-methyl-uridine, Pyrrolo-cytidine, inosine, a-thio- guanosine, 6-methyl-guanosine, 5-methyl-cytdine, 8-oxo-guanosine, 7-deaza-guanosine, Nl- methyl-adenosine, 2-amino-6-Chloro-purine, N6-methyl-2-amino-purine, Pseudo-iso- cytidine, 6-Chloro-purine, N6-methyl-adenosine, a-thio-adenosine, 8-azido-adenosine, 7- deaza-adenosine. Further modified nucleotides have been described previously (see, e.g., WO 2013/052523).

RNA integrity: The term "integrity" describes whether the complete RNA molecule is obtained and maintained. Low integrity could be due to, amongst others, degradation, cleavage, incorrect basepairing, incorporation of modified nucleotides or the modification of already incorporated nucleotides, lack of or incomplete capping, lack of or incomplete polyadenylation, or incomplete transcription. It can be measured, e.g., by capillary

electrophoresis. RNA recovery (rate): The term "RNA recovery" describes how much of the original RNA content of a sample remains after subjecting the sample to a procedure. Specifically, in the case of lyophilization, the RNA recovery rate, expressed as "YRNA", is calculated by comparison of the total RNA content before and after a lyophilization-cycle according to the following formula, where "CRNAI/2" is taken as RNA concentration before/after lyophilisation in mg/ml and Vl/2 as sample volume before after lyophilisation in ml: YRNA = (CRNA2 * V2)/ (CRNAI * VI) * 100 %.

Lyoprotectant: As used herein, the term "lyoprotectant" (sometimes also referred to as "cryoprotectant") typically refers to an excipient, which partially or totally replaces the hydration sphere around the RNA molecule and thus prevents catalytic and/or hydrolytic processes leading to damage of the RNA or a less stable RNA lyophilisate.

Suitable lyoprotectants in the context of the invention comprise monosaccharides, such as e.g. glucose, fructose, galactose, sorbose, mannose, etc., and mixtures thereof; disaccharides, such as e.g. lactose, maltose, sucrose, trehalose, cellobiose, etc., and mixtures thereof;

polysaccharides, such as raffmose, melezitose, maltodextrins, dextrans, dextrins, cellulose, starches, etc., and mixtures thereof; and alditols, such as glycerol, mannitol, xylitol, maltitol, lactitol, xylitol sorbitol, pyranosyl sorbitol, myoinositol, etc., and mixtures thereof. Generally, a sugar that is preferred in this context has a high water displacement activity and a high glass transition temperature. Furthermore, a suitable sugar is preferably hydrophilic but not hygroscopic. In addition, the sugar preferably has a low tendency to crystallize, such as trehalose. A lyoprotectant of the inventive method is preferably selected from the group consisting of mannitol, sucrose, glucose, mannose and trehalose. Trehalose is particularly preferred as a lyoprotectant.

Furthermore any of the below defined further components may be used as lyoprotectant. Particularly alcohols such as PEG, mannitol, sorbitol, cyclodextran, DMSO, amino acids and proteins such as prolin, glycine, phenylanaline, arginine, serine, albumin and gelatine may be used as lyoprotectant. Additionally metal ions, surfactans and salts as defined below may be used as lyoprotectant. Furthermore polymers may be used as lyoprotectant, particularly po lyviny lpyrro lidone .

The weight ratio of R A to the lyoprotectant is preferably in a range from about 1 :2000 to about 1 : 10, more preferably from about 1 : 1000 to about 1 : 100, more preferably a in a range from about 1 :250 to about 1 : 10 and more preferably in a range from about 1 : 100 to about 1 : 10 and most preferably in a range from about 1 : 100 to about 1 :50.

Complexing agent: a preferred complexing agent in the present invention is a "cationic or polycationic compound". This compound in the context of the present invention comprises cationic or polycationic polymers, cationic or polycationic peptides or proteins, e.g.

protamine, cationic or polycationic polysaccharides and/or cationic or polycationic lipids.

Cationic or polycationic compounds being particularly preferred agents in this context include protamine, nucleoline, spermine or spermidine, or other cationic peptides or proteins, such as poly-L-lysine (PLL), poly-arginine, basic polypeptides, cell penetrating peptides (CPPs), including HIV-binding peptides, HIV-1 Tat (HIV), Tat-derived peptides, Penetratin, VP22 derived or analog peptides, HSV VP22 (Herpes simplex), MAP, KALA or protein

transduction domains (PTDs), PpT620, prolin-rich peptides, arginine-rich peptides, lysine- rich peptides, MPG-peptide(s), Pep-1, L-oligomers, Calcitonin peptide(s), Antennapedia- derived peptides (particularly from Drosophila antennapedia), pAntp, plsl, FGF, Lactoferrin, Transportan, Buforin-2, Bac715-24, SynB, SynB(l), pVEC, hCT-derived peptides, SAP, or histones. More preferably, the mRNA/RNA according to the invention is complexed with one or more polycations, preferably with protamine or oligofectamine, most preferably with protamine. In this context, protamine is particularly preferred.

Further preferred cationic or polycationic compounds, which can be used in the context of the present invention, may include cationic polysaccharides, for example chitosan, polybrene, cationic polymers, e.g. polyethyleneimine (PEI), cationic lipids, e.g. DOTMA: [l-(2,3- sioleyloxy)propyl)]-N,N,N-trimethylammonium chloride, DMRIE, di-C14-amidine, DOTIM, SAINT, DC-Choi, BGTC, CTAP, DOPC, DODAP, DOPE: Dioleyl phosphatidylethanol- amine, DOSPA, DODAB, DOIC, DMEPC, DOGS: Dioctadecylamidoglicylspermin, DIMRI: Dimyristo-oxypropyl dimethyl hydroxyethyl ammonium bromide, DOTAP: dioleoyloxy-3- (trimethylammonio)propane, DC-6-14: 0,0-ditetradecanoyl-N-(- trimethylammonioacetyl)diethanolamine chloride, CLIPl : rac-[(2,3-dioctadecyloxypropyl)(2- hy droxy ethyl)] -dimethylammonium chloride, CLIP6: rac-[2(2,3-dihexadecyloxypropyl- oxymethyloxy)ethyl]trimethylammonium, CLIP9 : rac-2(2,3-dihexadecyloxypropyl- oxysuccinyloxy)ethyl-trimethylammonium, oligofectamine, or cationic or polycationic polymers, e.g. modified polyaminoacids, such as alpha-aminoacid-polymers or reversed polyamides, etc., modified poly ethylenes, such as PVP (poly(N-ethyl-4-vinylpyridinium bromide)), etc., modified acrylates, such as pDMAEMA (poly(dimethylamino ethyl methylacrylate)), etc., modified amidoamines such as pAMAM (poly(amidoamine)), etc., modified polybetaaminoester (PBAE), such as diamine end modified 1,4 butanediol diacrylate-co-5-amino-l-pentanol polymers, etc., dendrimers, such as polypropylamine dendrimers or pAMAM based dendrimers, etc., polyimine(s), such as PEI:

poly(ethyleneimine), poly(propyleneimine), etc., polyallylamine, sugar backbone based polymers, such as cyclodextrin based polymers, dextran based polymers, chitosan, etc., silan backbone based polymers, such as PMOXA-PDMS copolymers, etc., blockpolymers consisting of a combination of one or more cationic blocks (e.g. selected from a cationic polymer as mentioned above) and of one or more hydrophilic or hydrophobic blocks (e.g. polyethyleneglycole); etc.

Preferably the ratio of the RNA molecules as defined herein to the cationic or polycationic compound and/or the polymeric carrier, preferably as defined herein, is selected from a range of about 6: 1 (w/w) to about 0,25: 1 (w/w), more preferably from about 5: 1 (w/w) to about 0,5 : 1 (w/w), even more preferably of about 4: 1 (w/w) to about 1 : 1 (w/w) or of about 3 : 1 (w/w) to about 1 : 1 (w/w), and most preferably a ratio of about 3 : 1 (w/w) to about 2: 1 (w/w). Alternatively, the ratio of the RNA molecules as defined herein to the cationic or polycationic compound and/or the polymeric carrier, preferably as defined herein, in the component of the complexed RNA molecules, may also be calculated on the basis of the nitrogen/phosphate ratio (N/P-ratio) of the entire complex. In the context of the present invention, an N/P -ratio is preferably in the range of about 0.1-10, preferably in a range of about 0.3-4 and most preferably in a range of about 0.5-2 or 0.7-2 regarding the ratio of RNA molecules : cationic or polycationic compound and/or polymeric carrier, preferably as defined herein, in the complex, and most preferably in a range of about 0.7-1,5, 0.5-1 or 0.7-1, and even most preferably in a range of about 0.3-0.9 or 0.5-0.9, preferably provided that the cationic or polycationic compound in the complex is a cationic or polycationic cationic or polycationic protein or peptide and/or the polymeric carrier as defined above.

DESCRIPTION OF THE FINDINGS UNDERLYING THE PRESENT INVENTION

The removal of the contaminant TEAA is a major problem after HPLC purification of RNA if the RNA is to be used in sensitive applications, such as e.g. in therapy.

The inventors replaced the state-of-the art precipitation step, which is commonly performed after HPLC purification of RNA, by a lyophilization step to remove the TEAA contaminant from the RNA. The lyophilization step was further optimized by adding a co-solvent, namely tert-Butanol (also referred to as "t-BuOH" herein), which inter alia dramatically decreases the duration of the lyophilization process and increases the recovery of the lyophilized RNA.

An organic solvent, which may be present in addition to the TEAA (acetonitrile is a prominent organic solvent often used together with TEAA) may be removed using rotary vacuum concentration (RVC) and/or evaporation.

Advantageously, the inventive method yields lyophilized RNA, which displays a high recovery rate and a high integrity, and is readily soluble. An RNA lyophilized according to the present invention may also be referred to as "formulated RNA" (since it is ready to be formulated into a therapeutic dosage form, see above) and, accordingly, the present method may either be referred to as method of purifying (since TEAA is removed) or as method of formulating RNA (since lyophilized RNA "ready to be used" is provided) or as a combination of both. EXAMPLES

The Examples shown in the following are merely illustrative and shall describe the present invention in a further way. The Examples shall not be construed to limit the present invention thereto.

Example 1: RNA production, RNA purification by HPLC, and removal of acetonitrile from HPLC-purified RNA by rotary vacuum concentration in multi well plates: 1. Transcription and HPLC purification of RNA:

DNA templates were linearized and transcribed in vitro using T7 RNA polymerase (Thermo Fisher Scientific Inc.) according to standard procedures in the presence of a suitable buffer, a nucleotide mixture and a cap analog (m7GpppG). Subsequently, the obtained RNA products (ranging from about 1700 to 2000 nucleotides in size) were purified using HPLC following the manufacturer's instructions (PureMessenger®; CureVac AG, Tubingen, Germany; see WO2008077592). Due to the elution with a buffer comprising triethylammonium acetate (TEAA) and acetonitril, the samples comprising the RNA comprise also TEAA and acetonitril. 2. Removal of acetonitrile by rotary vacuum concentration

After HPLC purification, the fractions comprising RNA (referred to in the following as "HPLC samples") containing acetonitrile (maximum 25% acetonitrile) and 0.1 M TEAA were collected in commercially available deep well plates (multi well, polypropylene). RNA concentrations ranged from 0.2 to 2 g/1. Before rotary vacuum concentration, NaCl (6 mmol/g RNA) was added to provide a counter-ion, which facilitates the removal of (potentially) bound TEAA from the RNA. Acetonitrile was then removed by subjecting the HPLC samples to rotary vacuum concentration (RVC) using a RVC-2-33IR (Christ, Germany) and the program provided in Table 1. Fractions of the resulting RNA solutions were analysed for residual acetonitrile via headspace gas chromatography-flame ionization detection (GC FID), performed by an analytical laboratory according to standard methods (SAS Hagmann GmbH, Horb, Germany). All rotary vacuum concentrated HPLC samples were substantially free from acetonitrile (< 40 ppm) and subject to a further purification step as outlined in Example 2. Table 1 : Program for rotary vacuum concentration of RNA to remove acetonitrile

Example 1 shows that the RVC step resulted in acetonitrile-free RNA. Example 2: Lyophilization of naked RNA in multi well plates to remove TEAA:

After RVC, the samples were diluted with double distilled H₂0 (ddH₂0) to obtain a TEAA concentration of 0.1 M and an RNA concentration of 0.2 - 2 g/1 in the resulting diluted samples. All lyophilization cycles were performed in an Epsilon 2-6D LSCplus (Christ, Germany) using again the deep well plates (multi well, polypropylene). An aluminium thermo block was used to improve heat transfer between shelf and plate. For lyophilization, 1 ml of a sample was subjected to lyophilization (freezing, primary drying, secondary drying) using the program provided in Table 2. The LyoControl RX sensor was set to 80% resistivity to avoid melting of samples. After one lyophilization cycle, the samples were dissolved in 600 μΐ water for injection and incubated in the rotatory vacuum concentrator (RVC-2-33IR, Christ) at 80°C and 100 rpm for 20 min to dissolve the RNA for the next lyophilization cycle. The procedure was repeated twice to arrive at three cycles in total (final protocol: LYO; RCV (80°C); LYO; RCV (80°C) LYO; RCV (80°C)). For analysis purposes, the final lyophilized samples were dissolved in ddH₂0 (in the last RCV (80°C) step) and tested for residual TEAA (see Figure 1), RNA integrity (see Figure 2) and the total RNA recovery rate (see Figure 6, left column). A detailed description of the analytic methods is provided in Example 6.

As can be derived from Figure 1, the residual TEAA concentration (g/g RNA) was substantially reduced after each lyophilization cycle. Figure 2 shows that the integrity of the RNA samples before and after RVC at 80°C was comparable, with values of about 80% integrity. Figure 6 shows that the RNA recovery rate was at about 80% (Figure 6, first column of the diagram, "single solvent lyophilization"). Table 2: Program for single solvent lyophilization of RNA to remove TEAA

Examples 1 and 2 show that the procedure yielded TEAA- and acetonitrile-free RNA with recovery rates of about 80% and RNA integrity of about 80%. Furthermore, using the present method, it was possible to achieve a homogeneous lyophilization of all samples in the multi well plate.

Example 3: Lyophilization of naked RNA in single vials to remove TEAA: Individual samples in six separate glass vials, rather than multi well plates as in Example 2, were processed as described above in Example 2, and the RNA yield before and after each of the three LYO; RCV (80 °C) cycles was determined (see Figure 3). No loss of RNA was detected over all three cycles, which is difficult to achieve by the alternative method of precipitation.

Example 3 shows that the lyophilization method is generally preferred over a precipitation method since no loss of RNA was detected. Example 4: Lyophilization of naked RNA in glass or glass-coated plates to remove TEAA:

TEAA removal by lyophilization resulted in an RNA recovery of about 80% in polystyrene multi well plates (see Example 2). Multi well plates made from plastic materials like polypropylene show low heat conductivity compared to glass multi well plates. Glass or glass-coated plates show better heat transfer rates, which are likely to reduce the duration of the lyophilization cycle and are likely to lead to higher RNA recovery rates. The method described in Example 2 can therefore be performed with glass or glass-coated plates using a shortened lyophilization program (i.e. less incubation times and/or less lyophilization cycles). Example 4 shows that the lyophilization program might even be further shortened, e.g. if different materials and in particular glass or glass-coated plates are used.

Example 5: Co-solvent lyophilization of naked RNA in multi well plates to remove TEAA:

1. Determination of freeze point and optimal t-Butanol concentration:

To further improve the lyophilization procedure outlined in Examples 2, 3, and 4, t-Butanol (t-BuOH) was used as a co-solvent. t-BuOH has a high vapor pressure, freezes completely and quickly, and sublimes during primary drying of the lyophilization cycle.

In a first step, the freeze point of aqueous solutions of TEAA and t-BuOH was determined using a LyoRX resistivity sensor (Epsilon 2-6D LSCplus, Christ) according to the manual of the LyoRX resistivity sensor. These experiments were performed in the absence of RNA. The results were used to determine the freeze-point of aqueous solutions of TEAA and t-BuOH (see Table 3). Table 3: Sample composition for freeze-point determination of aqueous solutions of TEAA and t-BuOH.

The freeze point should not be exceeded during the freezing step of the lyophilization cycle. Based on these results, the t-BuOH concentration for the following experiments was set to 20% since the freeze points at this concentration were the lowest (i.e. about -30°C) under almost all conditions tested.

2. Co-solvent lyophilization of RNA using 20% t-BuOH:

All lyophilization cycles were performed using the Epsilon 2-6D LSCplus (Christ, Germany). An aluminum thermoblock was used for better heat transfer between shelf and plate. RNA samples from RVC (see Example 1) were diluted with water for injection (WFI) and t-BuOH to obtain samples containing RNA amounts from 0.2-1.5 g/1 (f.c), liquid volumes from 500- 1000 μΐ and 20 % (m/V, f.c.) t-BuOH prepared according to Table 3. All samples were transferred into a 96 deep well plate (see Table 4). Table 4: Sample composition for RNA co-solvent lyophilization to remove TEAA.

Empty wells of the plate were filled with 1 ml control solution (1 g/1 Heparin, 0.1 M TEAA and 20% t-BuOH in WFI). The plate was then sealed with a pierced aluminum foil. Five different lyophilization programs were developed that ensured that the temperature of the sample in the freezing step stays below the identified freeze point of about -30°C (see Table 5).

The vacuum for the primary drying phase was adjusted according to the vapour pressure over ice, to ensure that samples would not thaw in the primary drying phase (the sample temperature should be kept below the freeze point). The optimal vacuum was calculated according to the vacuum pressure over ice after first determining the freeze point (see Table 3 and Figure 4). Additionally, the LyoControl RX sensor was set to 80% resistivity to avoid melting of samples, i.e. in case of the sensor dropping under 80%>, the shelf heating is switched off and the freezing of samples starts again. The shelf heating continued when all samples were frozen. To determine the end of the primary drying phase, a periodic pressure increase test with a progression condition of 20% was set. After the secondary drying phase, the samples were re-suspended in 500 μΐ WFI. For better reconstitution, the samples were shaken at 750 rpm under heating at 45°C for 30 minutes (compared to 80°C without co- solvent; see Example 2). After dissolving the RNA in WFI, the lyophilization cycle was repeated with the programs used for the first cycle (LYO - re-suspension step - LYO). For analysis purposes, the lyophilized samples were dissolved in ddH20 and analyzed for residual TEAA, RNA integrity, total RNA recovery rate and residual t-BuOH (a detailed description of analytic methods is provided in Example 6). The co-solvent lyophilization method shortened the TEAA removal from about 8 workdays (using a single-solvent lyophilization method) to only three workdays (see Figure 5) with RNA recovery rates of nearly 100 % (see Figure 6). Furthermore, various other product parameters such as RNA integrity, dissolvability and RNA recovery rates could be improved using co-solvent lyophilization (as outlined in Example 6). Program 4 (of Table 5 shown below) yielded the most preferable results, and represents a particularly preferred embodiment of the present invention.

Table 5: Lyophilization programs to remove TEAA from RNA samples by co-solvent lyophilization (Programs 1 to 4 tested, Program 5 presently undergoing testing)

Freezing Time 00:00h 00: 15h l :36h 03:00h

Target 15 0 -48 -48

Temperature

[°C]

Temperature — -1 -0.5 0 change [°C/min]

Primary Time 00:40h 04:30h 04:00h ... drying Target -48 -20 -20 ...

Temperature

[°C]

a Temperature 0 0.1 0 — α

00 change [°C/min]

Ο Vacuum 0.040mbar 0.040mbar 0.040mbar ...

Safety pressure 0.200mbar 0.200mbar 0.200mbar ...

Lyo PvX control 80% 80% 80% ...

Secondary Time 00:20h 03:00h hold ... drying Target -20 15 15

Temperature

[°C]

Temperature 0 0.19 0 — change [°C/min]

Vacuum O.OlOmbar O.OlOmbar O.O l Ombar ...

Safety pressure 0.080mbar 0.080mbar 0.080mbar ...

Example 5 shows that the lyophilization step can be shortened and improved if a co-solvent, preferably t-BuOH, is added.

Example 6: RNA product characterization after lyophilization:

1. Analysis of R A integrity via fragment analyzer:

The RNA integrity after normalization was analyzed by capillary electrophoresis (Fragment Analyzer Advanced Analytical) with a DNF-471 Standard Sensitivity RNA Kit according to the manufacturer's instructions. The "% (Cone.)" given by the software PROsize™ with the minimal peak height set to 15 RFU and peak width set to 3 sec was used as a value to measure the RNA integrity. As seen in Figure 2, the integrity values of samples after single solvent lyophilization and heat treatment were stable (about 80% RNA integrity), showing that the herein established protocol is a suitable method for purification of HPLC samples. As seen in Figure 7, the integrity values of samples that underwent Programs 1-4 for lyophilization with the co-solvent tert-Butanol were higher compared to the values measured after single solvent lyophilization and heat treatment. Co-solvent lyophilization is therefore preferred.

2. Detection of residual TEAA:

The residual TEAA content after lyophilization was detected via headspace GC FID by a commercial analytic supplier (SAS Hagmann GmbH according to Eu.Pharm. 7.0) with the RNA concentration of the samples after lyophilization adjusted to 1 g/1.

After single solvent lyophilization, the TEAA content decreased to about 0.5 g TEAA/ g RNA (see Figure 1) after about 50 h of lyophilization. For single solvent lyophilization in multi- well plates, three cycles were needed to yield levels of about 0.5 g TEAA per g RNA (see Figure 8, left column). The TEAA content after the much faster co-solvent lyophilization process was comparable to or less than the TEAA content after single solvent lyophilization (Figure 8). Again, co-solvent lyophilization is preferred.

3. Spectrophotometric detection of RNA recovery:

The RNA content after solvation of the lyophilization cake in typically 600 μΐ WFI was measured via Nanodrop (Thermo scientific). The RNA recovery expressed as "YRNA" of every run was calculated by comparison of the total RNA yield before and after the lyophilization cycle according to the following formula, where "CRNAI/2" is taken as RNA concentration before/after lyophilization in mg/ml and V 1/2 as sample volume before after lyophilization in ml:

YRNA = (CRNA₂ * V2)/ (CRNAI * VI) * 100 % With single-solvent lyophilization in plastic-based multi well plates, the RNA recovery rate was about 76% (see Figure 6, left column). RNA recovery rates of up to 100 % were achieved with co-solvent lyophilization in plastic-based multi well plates, with programs 2-4 (see Figure 6, right columns as indicated).

No position-dependent effect in the multi well plate was observed regarding RNA recovery rate, RNA concentrations, or sample volume (i.e. no "edge effect") was observed showing that the process is robust and suitable for a multi- well format. 4. Determination of residual t-BuOH in co-solvent lyophilized RNA:

The residual t-BuOH content was determined by Headspace GC FID by SAS Hagmann GmbH with a RNA concentration of lg/1. Residual t-BuOH was reduced to levels below 0.05 g t-BuOH per g RNA for all programs tested (see Figure 9).

5. Determination of the RNA solubility after lyophilization:

Depending on the application, the lyophilized RNA needs to be dissolved in ddH20 for further processing (e.g., LNP encapsulation, protamine complexation etc.). The solubility was measured by determining the RNA content of at least 16 samples on at least two different timepoints. For each sample, the coefficient of variation was determined. The ratio of samples that showed a higher coefficient of variation than 5 % was used as a measure for the

"insolubility" of the RNA samples ("CV"). RNA samples were lyophilized either by single solvent lyophilization or by co-solvent lyophilization and incubated at 50 or 45°C, respectively, for 30 min. Thereafter, the "insolubility" was measured (see Figure 10).

Samples lyophilized by co-solvent lyophilization were readily soluble and no sample showed fluctuation upon RNA content measurement, whereas about 1/4 of the samples lyophilized by single solvent lyophilization showed fluctuations upon RNA content measurement.

These results show that co-solvent lyophilization leads to formulated RNA that is readily soluble, which is advantageous for further processing and/or application of the RNA.

Claims

1. A method of purifying and/or formulating RNA from a sample comprising the RNA and triethylammonium acetate, the method comprising the step of lyophilizing said sample.

2. The method according claim 1, wherein the method is performed in a multi-well plate.

3. The method according to claim 2, wherein the method is performed in a high throughput or robot-assisted setting.

4. The method according to any one of the preceding claims, wherein the RNA is in vitro transcribed RNA that has been subjected to HPLC purification.

5. The method according to any one of the preceding claims, wherein the RNA concentration in the sample is adjusted to about 0.2 to about 2.0 g/1 prior to carrying out the lyophilizing step.

6. The method according to any one of the preceding claims, wherein the step of lyophilizing comprises steps of freezing, primary drying, and secondary drying.

7. The method according to claim 6, wherein the temperature of the freezing step is in the range of from about -60 to about -40°C, preferably about -50°C, more preferably -48°C.

8. The method according to claim 7, wherein the sample is cooled to said temperature at a rate of about -0.5°C / minute.

9. The method according to any one of claims 6 to 8, wherein the freezing step takes

between about 3 to about 6 hours, preferably between about 4 to about 5 hours, and more preferably about 4.5 hours.

10. The method according to any one of claims 6 to 9, wherein the target temperature of the primary drying step is in the range of from about -30 to about 15°C, preferably from about -20 to about 5°C, and more preferably about -15°C.

11. The method according to any one of claims 6 to 10, wherein the primary drying step takes between about 3 to about 12 hours, preferably between about 5 to about 10 hours, and more preferably about 9 hours.

12. The method according to any one of claims 6 to 11, wherein the target temperature of the second drying step is in the range of from about 10 to about 20°C, preferably about 15°C.

13. The method according to any one of claims 6 to 12, wherein the secondary drying step takes between about 2 to about 5 hours, preferably about 3 to about 4 hours, more preferably about 3.5 hours.

14. The method according any one of the preceding claims, wherein the lyophilization step is performed using an aluminum thermo block.

15. The method according to any one of the preceding claims, wherein the lyophilization step is performed at about 80% resistivity.

16. The method according to any of the preceding claims, wherein the lyophilization step is repeated twice and the R A is dissolved in between the steps, preferably in ddH₂0.

17. The method according to any one of claims 1 to 15, wherein a co-solvent is added to the sample prior to the lyophilization step.

18. The method according to claim 17, wherein the co-solvent is selected from the group consisting of t-Butanol, 2-Butanol, Ethanol, n-Propanol, n-Butanol, Isopropanol, Ethyl acetate, Dimethyl carbonate, Dichloromethane, Methyl ethyl ketone, Methyl isobutyl ketone, Acetone, 1-pentanol, Methyl acetate, Methanol, Carbon tetrachloride, Dimethyl sulfoxide, Hexafluoroacetone, Chlorobutanol, Dimethyl sulfone, Acetic acid, Acetate, Cyclohexane, Tetrahydrofuran, Tetrahydropyran, Dioxane, Trioxane and other cyclic mono-, di- and tri-ethers, PEG 600-6000, and Tocopherol Succinate.

19. The method according to claim 18, wherein the co-solvent is t-Butanol.

20. The method according to claim 19, wherein the t-Butanol is added such that the resulting t-Butanol concentration in the sample is between about 5 and about 50% (fc).

21. The method according to claim 20, wherein the t-Butanol is added such that the resulting t-Butanol concentration in the sample is about 20% (f.c).

22. The method according to any one of claims 19 to 21, wherein the freeze point of the sample after t-Butanol addition is in the range of from about -50 to about -10°C.

23. The method according to claim 22, wherein the freeze point of the sample after the

addition of t-Butanol is in the range of from about -30 to about -20°C.

24. The method according to any one of claims 17 to 23, wherein the lyophilization step is repeated once and the co-solvent is only added prior to the first lyophilization step.

25. The method according to any one of the preceding claims, wherein the sample is

substantially free from triethylammonium acetate after the lyophilization step.

26. The method according to any one of claims 19 to 25, wherein the sample is substantially free from t-Butanol after the lyophilization step.

27. The method according to any one of the preceding claims, wherein the sample

additionally comprises an organic solvent, and the method additionally comprises the step of rotary vacuum concentration and/or evaporation using nitrogen, argon or carbon dioxide prior to the lyophilization step.

28. The method according to claim 27, wherein the rotary vacuum concentration is performed at a temperature ranging from about 10 to about 40°C.

29. The method according to claim 27 or 28, wherein the rotary vacuum concentration is performed at a temperature ranging from about 10 to about 30°C.

30. The method according to any one of claims 27 to 29, wherein the rotary vacuum

concentration is performed at a temperature of about 30°C, preferably 28°C.

31. The method according to any one of claims 27 to 30, wherein the sample is substantially free from the organic solvent after the rotary vacuum concentration and/or evaporation.

32. The method according to any one of claims 27 to 31, wherein said organic solvent is selected from the group consisting of acetonitrile, methanol, ethanol, 1-propanol, 2- propanol, ethyl acetate, tetrahydrofuran, acetone, and a mixture of any of the foregoing.

33. The method according to any one of claims 27 to 32, wherein said organic solvent is acetonitrile.

34. The method according to any one of claims 2 to 33, wherein the multi-well plate is a glass plate or glass-coated plate.

35. The method according to any one of the preceding claims, wherein the method comprises the step of adding a lyoprotectant, wherein said lyoprotectant is added prior to the final lyophilization step.

36. The method according to any one of the preceding claims, wherein the method comprises the step of adding a complexing agent, wherein said complexing agent is added prior to the final lyophilization step.

37. R A purified and/or formulated according to any one of the preceding claims.

38. The RNA according to claim 37 for use in therapy.

39. Use of a method according to any one of claims 2 to 36 for the small-scale production of multiple RNAs in parallel.

40. Use of a method according to any one of claims 2 to 36 for the screening of RNAs in parallel.