WO2025257280A1

WO2025257280A1 - Method for purification of rna

Info

Publication number: WO2025257280A1
Application number: PCT/EP2025/066312
Authority: WO
Inventors: Akai YUSUKE
Original assignee: Merck Patent GmbH
Current assignee: Merck Patent GmbH
Priority date: 2024-06-14
Filing date: 2025-06-11
Publication date: 2025-12-18
Anticipated expiration: 2026-12-14

Abstract

The present invention relates to a method for the chromatographic purification of mRNA. For this anion exchange chromatography is performed with dual pH and salt concentration change as well as a chaotrope in the elution buffer.

Description

Method for purification of RNA

The present invention relates to a method for the chromatographic purification of mRNA. For this anion exchange chromatography is performed with dual pH and salt concentration change as well as a chaotrope in the elution buffer. mRNA is a new candidate for a plurality of pharmaceutical applications like mRNA vaccines. But effective purification of mRNA for introduction into the body remains a problem. One challenge are the different types and combinations of undesired contaminants in a sample that need to be separated from a desired RNA species to obtain a pure RNA sample. Examples of such contaminants are components and by-products of any upstream processes, like undesired RNA species, proteins, DNA or fragments thereof, free nucleotides, endotoxins, detergents, and organic solvents.

Methods for purification of RNA are known in the art. Anion exchange chromatography has been described as one suitable method. Typically, short mRNA including less than 500 nucleotides can be purified by the traditional anion exchangers using salt linear gradient like e.g. described in L. E. Easton et al., Rapid, nondenaturing RNA purification using weak anion exchange fast performance liquid chromatography, RNA, 2010 Mar; 16(3): 647-653. Larger mRNA molecules with over 500 nucleotides are difficult to elute from strong anion exchangers due to the strong binding of mRNA. In addition, larger mRNA molecules can form highly complex formations and are also able to interact with other RNA transcripts, creating quaternary structures. The strong interaction of these molecules with the stationary phase can be diminished by operating under denaturing or partially denaturing conditions. This can be achieved by elevating temperature or by adding of chaotropic reagents like suggested in WO2014144767. However, elevated temperature is not suitable for GMP manufacturing. Additionally, there is a need for scalable methods with high binding capacity and high yield for GMP manufacturing.

It has been found that a dual elution gradient by increasing pH and salt concentration in the presence of a chaotropic salt achieve an effective mRNA elution with high yield and effective removal of process related impurities. The combination of pH, salt and chaotropic salt work cooperatively to improve the mRNA elution especially from strong anion exchangers. With these findings, there is no need of column heater and similar facility to elevate temperature for mRNA elution. This novel scalable method enables a wide range of especially larger ss(single stranded) RNA purification and is applicable for the GMP manufacturing of mRNA-based therapeutics and vaccines.

The present invention is therefore directed to a method for purifying RNA (ribonucleic acid) comprising a) Providing a sample comprising said RNA b) Loading the sample onto a chromatography matrix comprising anion exchange groups c) Optionally washing the chromatography matrix with a wash buffer d) eluting RNA bound to the chromatography matrix by increasing pH and salt concentration of the elution buffer during elution whereby the elution buffer comprises a chaotrope and the method is performed at a temperature below 35 °C.

In a preferred embodiment the method is performed at a temperature between 20 and 28 °C. In another preferred embodiment in step d) during elution the pH of the elution buffer is changed, preferably increased, in the range between 7.5 and 14, preferably in the range between 8 and 11 .

In another preferred embodiment, in step d) during elution the salt concentration of the elution buffer is changed, preferably increased, in the range between 0 and 4 M salt, preferably between 0 and 2 M salt.

In a preferred embodiment the salt is NaCI.

In a preferred embodiment the chaotrope is selected from the group of arginine, guanidine carbonate, urea and/or sodium perchlorate.

In another preferred embodiment the pH of the elution buffer is induced by a buffering system using TRIS, TAPS, CAPS, CHES, Bis Tris propane, Glycine or combinations thereof.

In a preferred embodiment the chromatography matrix is a resin or membrane, most preferred a hydrogel membrane.

In one embodiment the RNA is 500 to 10,000 nucleotides or base pairs in length, preferably 800 to 5,000.

In another embodiment, the removal of impurities in the purified RNA obtained in step d) compared to the sample provided in step a) is at least 80%, preferably at least 90%, most preferred at least 95%. That means at least 80%, preferably at least 90%, most preferred at least 95% (w/w) of the impurities which are present in the sample provided in step a) are removed by the method of the present invention.

In another embodiment the yield of the purified RNA in step d) is at least 65%, preferably at least 85%, most preferred at least 90%. In a preferred embodiment the anion exchange groups comprise trimethylammonium (-CH2N(CH3)3⁺) or triethylammonium (-CH₂CH₂N(CH₂CH3)3⁺).

Figures:

Figure 1A shows chromatogram of mRNA EPO by salt gradient elution at pH 8.0 in the absence of chaotropic salt on a Natrix® Q membrane. Figure 1 B shows chromatogram of mRNA EPO by dual gradient with increasing pH and salt concentration in the presence of chaotropic salt on Natrix® Q. Details can be found in Example 1 .

Figure 2A shows chromatogram of mRNA FLuc by salt gradient elution at pH 8.0 in the absence of chaotropic salt on a Natrix® Q membrane. Figure 2B shows chromatogram of mRNA FLuc by dual gradient with increasing pH and salt concentration in the presence of chaotropic salt on Natrix® Q. Details can be found in Example 1 .

Figure 3A shows chromatogram of step elution in the presence of chaotropic salt using a Natrix® Q membrane for mRNA EPO.

Figure 3B shows chromatogram of step elution in the presence of chaotropic salt using Eshmuno® Q for mRNA EPO. Details can be found in Example 2.

Figure 4A shows chromatogram of step elution in the presence of chaotropic salt using a Natrix® Q membrane for mRNA FLuc.

Figure 4B shows chromatogram of step elution in the presence of chaotropic salt using a Eshmuno® Q resin for mRNA FLuc. Details can be found in Example 2. Figure 5 shows chromatogram of salt gradient elution in the presence of chaotropic salt at fixed pH 10.5 using a Natrix® Q membrane for mRNA FLuc. Details can be found in Example 3.

Definitions

Before describing the present invention in detail, it is to be understood that this invention is not limited to specific compositions or process steps, as such may vary. It must be noted that, as used in this specification and the appended claims, the singular form "a", "an" and "the" include plural referents unless the context clearly dictates otherwise. Thus, for example, reference to "a ligand" includes a plurality of ligands and reference to "an antibody" includes a plurality of antibodies and the like.

Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention is related. The following terms are defined for purposes of the invention as described herein.

As used herein, and unless stated otherwise, the term “sample” refers to any composition or mixture that contains RNA. Samples may be derived from biological or other sources. Biological sources include eukaryotic and prokaryotic sources, such as plant and animal cells, tissues and organs. The sample may also include diluents, buffers, detergents, and contaminating species, debris and the like that are found mixed with the target molecule, in this case RNA. The sample may be "partially purified" (i.e. , having been subjected to one or more purification steps, such as filtration steps) or may be obtained directly from a host cell or organism producing the nucleic acid (e.g., the sample may comprise harvested cell culture fluid). The term "impurity" or “contaminant” as used herein, refers to any foreign or objectionable molecule, including one or more host cell proteins, endotoxins, lipids, nucleic acids and one or more additives which may be present in a sample containing the RNA that is being separated from one or more of the foreign or objectionable molecules using the process of the present invention.

The terms "purifying," "separating," or "isolating," as used interchangeably herein, refer to increasing the degree of purity of the target RNA from a composition or sample comprising the target RNA and one or more impurities. Typically, the degree of purity of the target RNA is increased by removing (completely or partially) at least one impurity from the composition.

The term "chromatography" refers to any kind of technique which separates an analyte of interest (e.g. a target RNA) from other molecules present in a sample like impurities. Usually, the target RNA is separated from other molecules as a result of differences in rates at which the individual molecules of the mixture bind to and/or migrate through a chromatography matrix under the influence of a moving phase.

The term "matrix" or "chromatography matrix" are used interchangeably herein and refers to a solid phase though which the sample migrates in the course of a chromatographic separation. The matrix typically comprises a base material and ligands covalently bound to the base material. The matrix of the present invention can comprise or consist of particles, a membrane or a monolith, preferably the base material is a membrane or monolith, most preferred a resin or a membrane.

A “ligand” is a functional group that is part of the chromatography matrix, typically it is attached to the base material of the matrix, and that determines the binding properties and interaction properties of the matrix. Examples of "ligands" include, but are not limited to, ion exchange groups, hydrophobic interaction groups, hydrophilic interaction groups, thiophilic interactions groups, metal affinity groups, affinity groups, bioaffinity groups, and mixed mode groups (combinations of the aforementioned). It is also possible that one ligand has more than one binding I interaction property. The matrix of the present invention comprises at least anion exchange groups. The matrix of the present invention may comprise between 1 and 6 mmol/g anion exchange groups. Anion exchange groups can be subdivided into strong anion exchange groups, such as quaternary ammonium groups like trimethylammonium (-CH2N(CH3)3⁺) or triethylammonium (- CH₂CH₂N(CH₂CH3)3⁺) as well as weak anion exchange groups, such as ammonium (NHs⁺), N,N diethylamino or diethyl-aminoethyl DEAE. The matrix may additionally comprise further other types of ligands so that the matrix is a mixed mode matrix. Such ligands may e.g., have hydrophobic interaction groups, such as phenyl, butyl, propyl, hexyl. The groups may be part of the base material, they may also be part of a ligand. One ligand may comprise one or several different anion exchange groups. According to the present invention, the term anion exchange chromatography matrix covers matrices which only comprise one or more types of anion exchange groups as well as matrices comprising one or more types of anion exchange groups in combination with other types of functional groups like hydrophobic interaction groups, i.e. , mixed mode matrices.

The ligands can be attached to the base material of the matrix by any type of covalent attachment. Covalent attachment can for example be performed by directly bonding the functional groups to suitable residues on the base material like OH, NH₂, carboxyl, phenol, anhydride, aldehyde, epoxide or thiol etc. It is also possible to attach the ligands via suitable linkers. It is also possible to generate the matrix by polymerizing monomers comprising the ligands and a polymerizable moiety. Examples of matrices generated by polymerization of suitable monomers are polystyrene, polymethacrylamide or polyacrylamide-based matrices generated by polymerizing suitable styrole or acryloyl monomers.

In another embodiment the stationary phase can be generated by grafting the ligands onto the base material or from the base material. For grafting from processes with controlled free-radical polymerisation, such as, for example, the method of atom-transfer free-radical polymerisation (ATRP), are suitable. A very preferred one-step grafting from polymerisation reaction of acrylamides, methacrylates, acrylates, methacrylates etc. which are functionalized e.g. with ionic, hydrophilic or hydrophobic groups can be initiated by cerium(IV) on a hydroxylcontaining support, without the support having to be activated.

When the chromatography matrix is used in a chromatographic separation it is typically used in a separation device, also called housing, as a means for holding the matrix.

In one embodiment, the device comprises a housing with an inlet and an outlet and a fluid path between the inlet and the outlet. In a preferred embodiment the device is a chromatography column. Chromatography columns are known to a person skilled in the art. They typically comprise cylindrical tubes or cartridges filled with the stationary phase as well as filters and/or means for fixing the stationary phase in the tube or cartridge and optionally connections for solvent delivery to and from the tube or cartridge. The size of the chromatography column varies depending on the application, e.g. analytical or preparative. In one embodiment the column or generally the separation device is a single use device.

The term "anion exchange matrix" is thus used herein to refer to a chromatography matrix which carries at least anion exchange groups. That means it typically has one or more types of ligands that are positively charged under the chromatographic conditions used, such as quaternary amino groups.

A "buffer" is a solution that resists changes in pH by the action of its acid-base conjugate components. Various buffers which can be employed depending, for example, on the desired pH of the buffer are described in Buffers. A Guide for the Preparation and Use of Buffers in Biological Systems, Gueffroy, D., ed. Calbiochem Corporation (1975). Non- limiting examples of buffers include TAPS, MES, MOPS, MOPSO, Tris, HEPES, phosphate, acetate, citrate, succinate, and ammonium buffers, as well as combinations of these.

According to the present invention the term “buffer” or “solvent” is used for any liquid composition that is used to load, wash, elute, re-equilibrate, strip and/or sanitize the chromatography matrix.

Detergents are defined as “surface active molecules”. The term "detergent" as used herein refers to compounds that lower the surface tension between two liquids or between a liquid and a solid. Detergents may act as wetting agents, emulsifiers, foaming agents, and dispersants. Examples of detergents are non-ionic, ionic or zwitterionic detergents.

When “loading” a chromatography column in bind and elute mode, the sample or composition comprising the target molecule and one or more impurities is loaded onto a chromatography column. In preparative chromatography, the sample can be loaded directly without the addition of a loading buffer. If a loading buffer is used, the buffer has a composition, a conductivity and/or pH such that the target nucleic acid is bound to the stationary phase while ideally all the impurities are not bound and flow through the column. Typically, the loading buffer, if used, has the same or similar composition as the equilibration buffer used to prepare the column for loading. The final composition of the sample loaded on the column is called feed.

The feed may comprise the sample obtained from in vitro transcription and the loading buffer.

By “wash” or "washing" a chromatography matrix is meant passing an appropriate liquid, e.g. a buffer through or over the matrix. Typically washing is used after loading to remove weakly bound contaminants from the matrix in bind/elute mode prior to eluting the target molecule. Additionally, wash steps can be used to reduce levels of residual detergents, enhance viral clearance and/or alter the conductivity carryover during elution.

To "elute" a molecule (e.g. the target RNA) from a matrix means that the molecule is removed therefrom. Elution may take place by altering the solution conditions such that a buffer different from the loading and/or washing buffer competes with the molecule of interest for the ligand sites on the matrix or alters the equilibrium of the target molecule between stationary and mobile phase such that it favors that the target molecule is preferentially present in elution buffer.

A non-limiting example is to elute a molecule from an ion exchange resin by altering the ionic strength of the buffer surrounding the ion exchange material such that the buffer competes with the molecule for the charged sites on the ion exchange material.

Elution can be performed by using a buffer gradient. A gradient typically starts with an elution buffer similar to the equilibration buffer or wash buffer and is then, in the course of the gradient, changed so that some of its properties, like pH, conductivity or composition, differ more and more from the equilibration buffer. This causes elution conditions which at the beginning provide adequate compound retention so that not all compounds will immediately elute from a chromatography column. Compounds that are retained less elute first. By further changing the elution buffer in the course of the gradient compounds that are retained stronger on the matrix will then also elute from the matrix. An elution gradient thus ideally separates the target molecule from its impurities by causing elution at different elution buffer composition.

There are two main gradient designs:

-Linear gradient - a linear change of a least one property like pH or conductivity in the elution buffer

-Step gradient - a stepwise change in least one property like pH or conductivity in the elution buffer, whereby the property is changed at least 2 times in the course of the gradient, typically 3 to 10 times.

Linear pH gradients are difficult to generate because changing ratios between acidic and basic components leads to a constant change of the ionic strength of the mobile phase, which itself has a significant impact on pH. As a consequence, a linear pH gradient according to the present invention is also a pH gradient which is generated by linear, constant change in the ratio of the more acidic and the more basic buffer component though the resulting mixture might not show linear change in pH. Simple mixing of single buffers with two different pH values in linear range volumes is not applicable in all instances due to limitations in buffering capacity depending on the pH to be mixed. With the use of a single buffer system it is often possible to achieve relatively linear gradients over approximately 1.5 pH units. If wider pH ranges are needed, it is often favorable to mix different buffer salts. The preferred buffer system for the present invention is Tris.

The term “conductivity" as used herein, refers to an inherent property of most materials, that quantifies how strongly it resists or conducts electric current. In aqueous solutions, such as buffers, the electrical current is carried by charged ions. The conductivity is determined by the number of charged ions, the amount of charge they carry and how fast they move. Hence, for most aqueous solutions, the higher the concentration of dissolved salts, the higher the ionic strength and the higher the conductivity. Raising the temperature enables the ions to move faster, hence increasing the conductivity. Typically, the conductivity is defined at room temperature, if not otherwise indicated. The basic unit of conductance is Siemens (S). It is defined as the reciprocal of the resistance in Ohms, measured between the opposing faces of a 1 cm cube of liquid. Therefore, the values are estimated in mS/cm.

A membrane as chromatographic matrix can be distinguished from particle-based chromatography by the fact that the interaction between a solute, e.g. the target nucleic acids or contaminants, and the matrix does not take place in the dead-ended pores of a particle, but mainly in the throughpores of the membrane. Exemplary types of membranes are flat sheet systems, stacks of membranes, microporous polymer sheets with incorporated cellulose, polystyrene or silica-based membranes as well as radial flow cartridges, hollow fiber modules and hydrogel membranes. Preferred are hydrogel membranes. Such membranes comprise a membrane support and a hydrogel formed within the pores of said support. The membrane support provides mechanical strength to the hydrogel. The hydrogel determines the properties of the final product, like pore size and binding chemistry.

The membrane support can consist of any porous membrane like polymeric membranes, ceramic based membranes and woven or nonwoven fibrous material. Suitable polymeric materials for membrane supports are cellulose or cellulose derivatives as well as other preferably inert polymers like polyethylene, polypropylene, polybutylenterephthalate or polyvinylidene-difluoride.

The hydrogels can be formed through in-situ reaction of one or more polymerizable monomers with one or more crosslinkers and/or one or more cross-linkable polymers to form a cross-linked gel that has preferably macropores. Suitable polymerizable monomers include monomers containing vinyl or acryl groups. Preferred are monomers comprising an additional functional group that either directly forms the ligand of the matrix or is suitable for attaching the ligands. Suitable crosslinkers are compounds containing at least two vinyl or acryl groups. Further details about suitable membrane supports, monomers, crosslinkers etc. as well as suitable production conditions can be found in WO04073843 and WO2010/027955. Especially preferred are membranes made of an inert, flexible fiber web support comprising assembly within and around the fiber web support a porous polyacrylamide hydrogel with quaternary ammonium groups (strong anion exchange groups), like Natrix® Q Chromatography membrane, Merck KGaA, Germany.

Depending on the membrane device used, the respective processes are conducted by different operating principles like dead-end operation, cross-flow operation and radial flow operation systems. Dead-end operation is preferred.

Examples of suitable membranes to be used in the method of the present invention are

- Membranes with a polyethersulfone (PES)-based support and a cross-linked polymeric coating, functionalized with quaternary ammonium groups (strong anion exchange groups), like Mustang® Q, Pall.

- Membranes made of stabilized reinforced cellulose, functionalized with quaternary ammonium groups (strong anion exchange groups) or with DEAE groups (diethylaminoethyl, weak ion exchange groups), like Sartobind® membranes, Sartorius.

- Membranes made of stabilized reinforced cellulose, comprising a hydrogel with quaternary ammonium groups (strong anion exchange groups), like Sartobind® Jumbo Membranes made of stabilized reinforced cellulose, functionalized with quaternary ammonium groups (strong anion exchange groups), like Sartobind® Jumbo membranes, Sartorius.

- Membranes made of a fine fiber non-woven scaffold comprising a hydrogel with quaternary ammonium groups (strong anion exchange groups), like 3M^TM Emphaze™ AEX Hybrid Purifier, 3M.

- Membranes made of an inert, flexible fiber web support comprising within and around the fiber web support a porous polyacrylamide hydrogel with quaternary ammonium groups (strong anion exchange groups), like Natrix® Q Chromatography membrane, Merck KGaA, Germany.

A monolith or a monolithic sorbent, similar to a membrane, has throughpores, like interconnected channels, so that liquid can flow from one side of the monolith, through the monolith, to the other side of the monolith.

Since the mobile phase is flowing through these throughpores, molecules to be separated are transported by convection rather than by diffusion. Due to their structure monolithic sorbents show flow rate independent separation efficiency and dynamic capacity.

The monolith is typically formed in situ from reactant solutions and can have any shape or confined geometry, typically with frit-free construction, which guarantees convenience of operation. Preferably, monolithic materials have a binary porous structure, mesopores and macropores. The micron-sized macropores are the throughpores and ensure fast dynamic transport and low backpressure in applications; mesopores contribute to sufficient surface area and thus high loading capacity.

The monoliths can be made of organic, inorganic or organic/inorganic hybrid materials. Preferred are organic polymer-based monoliths. The synthesis of organic polymer monoliths is typically done by a one- step polymerization providing a tunable porous structure with tailored functional groups. Generally, a pre-polymerization mixture consisting of the monomers, crosslinkers, porogenic solvents, and initiators in an appropriate ratio is polymerized in a suitable container, also called mould, determining the format of the monolith. Polymerization is typically initiated by heating, use of UV radiation, microwave or y-ray radiation in the presence of initiators. After reaction for the prescribed time at an appropriate temperature, the resulting material is typically washed with solvents to remove unreacted components and porogenic solvents. Suitable organic polymers are polymethacrylates, polyacrylamides, polystyrenes, polyurethanes, etc., like Poly(methacrylic acid-ethylene dimethacrylate), Poly(glycidyl methacrylate-ethylene dimethacrylate) or Poly(acrylamide-vinylpyridine-N,N'-methylene bisacrylamide).

Inorganic monoliths can be made of silica or other inorganic oxides. Preferably they are made of silica. Silica monoliths are normally prepared via a sol-gel method with phase separation. This mainly includes hydrolysis, condensation, and polycondensation of silica precursors. Typically, tetraethoxysilane (TEOS) or tetramethylorthosilicate (TMOS) is distributed in a suitable solvent in the presence of a porogen (e.g. poly(ethylene glycol) (PEG)), followed by the addition of a catalyst, acid or base, or a binary catalyst, acid and base in sequence. After reaction for a prescribed time, the resulting gellike product is washed with solvents to remove unreacted precursor, porogen, and catalyst, followed by the proper post treatment, typically a heat treatment.

The monoliths can be modified with suitable functional groups, preferably at least ion exchange groups, to generate the targeted interaction with the sample comprising the target molecule and thus the targeted separation. Membranes and monoliths can also be produced by 3D printing processes.

Typically, the monoliths are contained in a housing like a column.

Particle-based resins intended for liquid chromatography are normally comprised of particles that are packed together in a tubular cylinder called column to form a bed. The packed bed shows a distinct space between the particles, so called void volume, which mainly defines the liquid fluid permeability and hydrodynamic properties of the packed bed.

The particles typically consist of a cross-linked polymer matrix in spherical, bead-like or granular shape with relatively uniform size for improved chromatographic and hydrodynamic characteristics of the packed bed. They can have a dense structure with discrete or very small pores but usually exhibit a porous multichannel or reticular structure forming an inner pore volume and additional surface area inside the particle. The particle surface area can be modified with a variety of functional groups suitable for chromatography applications either by using functional monomers for the backbone-polymer structure, coupling of functional groups to the particle surface directly of via ligands or short polymers structures (grafts).

Particulate base materials can be prepared, for example, from organic polymers. Organic polymers of this type can be polysaccharides, such as agarose, dextranes, starch, cellulose, etc., or synthetic polymers, such as poly(acrylamides), poly(methacrylamides), poly(acrylates), poly(methacrylates), hydrophilic substituted poly(alkyl allyl ethers), hydrophilic substituted poly(alkyl vinyl ethers), poly(vinyl alcohols), poly(styrenes) and copolymers of the corresponding monomers. These organic polymers can preferably also be employed in the form of a crosslinked hydrophilic network. This also includes polymers made from styrene and divinylbenzene, which can preferably be employed, like other hydrophobic polymers, in a hydrophilized form.

Alternatively, inorganic materials, such as silica, zirconium oxide, titanium dioxide, aluminium oxide, etc., can be employed as particulate base materials. It is equally possible to employ composite materials, i.e. , for example, particles which can themselves be magnetised by copolymerisation of magnetisable particles or of a magnetisable core. It is also possible to use core shell materials whereby the shell, i.e.at least the surface or a coating, has OH groups.

However, preference is given to the use of hydrophilic base materials which are stable to hydrolysis or can only be hydrolysed with difficulty since the materials according to the invention should preferably withstand alkaline cleaning or regeneration at e.g., basic pH over an extended use duration.

The base matrix may consist of irregularly shaped or spherical particles, whose particle size can be between 2 and 1000 pm. Preference is given to average particle sizes between 3 and 300 pm, in a most preferred embodiment the average particle size is between 20 - 63 pm.

The particulate base material may be in the form of non-porous or preferably porous particles. The average pore sizes can be between 2 and 300 nm. Preference is given to pore sizes between 5 and 200 nm, most preferred average pore size is between 40 - 110 nm.

In a very preferred embodiment, the particulate base material is formed by copolymerisation of a hydrophilic substituted alkyl vinyl ether selected from the group of 1 ,4-butanediol monovinyl ether, 1 ,5-pentanediol monovinyl ether, diethylene glycol monovinyl ether or cyclo-'hexane-'dimethanol monovinyl ether and divinylethyleneurea (1 ,3- divinylimidazolin-2-one) as crosslinking agent. An example of a suitable commercially available vinylether based base material is Eshmuno®, Merck KGaA, Germany.

In a preferred embodiment the polymer to be used as a particulate matrix in the method of the present invention is derivatised by graft polymerisation with tentacle-like structures, which can in turn carry the corresponding ligands or be functionalised by means of the latter. The grafting is preferably carried out in accordance with EP 0 337 144 page 12 example 8 or US 5453186 page 9 example 8 using N-(2- Trimethylammoniumethyl)-acrylamide and/or another monomer carrying suitable functional groups. The polymerisation catalyst employed is cerium(IV) ions, since this catalyst forms free-radical sites on the surface of the base material, from which the graft polymerisation of the monomers is initiated.

The polymerisation is terminated by termination reactions involving the cerium salts. For this reason, the (average) chain length can be influenced by the concentration ratios of the base material, the initiator and the monomers. Furthermore, uniform monomers or also mixtures of different monomers can be employed; in the latter case, grafted copolymers are formed.

Suitable monomers for the preparation of the graft polymers and further details about the grafting procedure are e.g., disclosed in WO 2007/014591 , EP 0337 144, especially page 12, example 8 and US 5453186 page 9, example 8.

Preferably the matrix is derivatised with cationic groups by graft polymerisation whereby the resulting chains that are grafted onto the base material have a length of between 2 and 100, preferably 5 and 60, in particular between 10 and 30 monomer units, each unit typically carrying one cationic group. The matrix might carry additional other functional groups like hydrophobic or hydrophilic groups in addition to the anion exchange groups but in any case, it has anion exchange groups.

Preferred particulate matrices are matrices with weak anion exchange and/or strong anion exchange groups, e.g., with trimethylammoniumethyl (TMAE) groups, like Eshmuno® Q or Fractogel® TMAE, Merck KGaA, Germany, or with dimethylethanolamine (DMAE) groups, like Fractogel® DMAE, Merck KGaA, Germany, or with diethylaminoethyl (DEAE) groups like Fractogel® DEAE, Merck KGaA, Germany

The base material may equally also be in the form of fibres, hollow fibres or coatings.

The term "chaotrope" as used herein, refers to a substance that causes disorder in a protein or nucleic acid by, for example, but not limited to, altering the secondary, tertiary or quaternary structure of a protein or a nucleic acid while leaving the primary structure intact. Exemplary chaotropes include, but are not limited to, guanidine hydrochloride (GuHCI), guanidinium thiocyanate (GuSCN), sodium thiocyanate (KSCN), sodium iodide, sodium perchlorate, urea, arginine, and the like. A typical anionic chaotropic series, shown in order of decreasing chaotropic strength, includes: CCl3COO’»CNS >CF3COO >CIO4>l’ >CH3COO²’>Br, Cl or CHO2’. Descriptions of chaotropes and chaotropic salts can be found in, for instance, in K. Hamaguchi et al. (Proc. Natl. Acad. Sci. (1962) 62:1129-1136).

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, uridinemonophosphate, guanosine-monophosphate and cytidinemonophosphate monomers, 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, 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 O'UTR, an open reading frame, optionally a 3'UTR and a poly(A) sequence. Aside from messenger RNA, several non-coding types of RNA exist which may be involved in regulation of transcription and/or translation, and immunostimulation. The term "RNA" further encompass other coding RNA molecules, such as viral RNA, retroviral RNA and replicon RNA, small interfering RNA (siRNA), antisense RNA, CRISPR 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).

RNA can also result from in vitro transcription. In this case the RNA molecule that has been synthesized from a template DNA, commonly a linearized and purified plasmid template DNA, a PCR product, or an oligonucleotide. RNA synthesis occurs in a cell free ("in vitro") system catalyzed by DNA dependent RNA polymerases. In a process called RNA in vitro transcription, nucleotides or nucleotide analogues are transcribed into RNA. Particular examples of DNA dependent RNA polymerases are the T7, T3, and SP6 RNA polymerases. An in vitro transcribed RNA may comprise elements such as 5'- cap, optionally a O'UTR, an open reading frame, optionally a 3'UTR and a poly(A) sequence. Aside from proteinogenic messenger RNA, several noncoding types of RNA exist which may be involved in regulation of transcription and/or translation. All RNA molecules as defined herein may also be synthesized by in vitro transcription.

Detailed Description

The RNA to be purified according to the method of the present invention may originate from any natural, genetic-engineering or biotechnological source.

The RNA is particularly preferably selected from mRNA, viral RNA, retroviral RNA and replicon RNA, small interfering RNA (siRNA), antisense RNA, clustered regularly interspaced short palindromic repeats (CRISPR) 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) or whole-cell RNA (total RNA extract). The RNA to be isolated may be single-stranded or doublestranded. Single-stranded RNA may optionally form secondary structures by refolding. The RNA to be separated is preferably singlestranded. The RNA may be unlabelled or also labelled (with a fluorescent label or a radiolabel or an antibody epitope). One example of a labelled RNA is digoxigenin-labelled RNA. The RNA may also contain modifications like modified nucleotides or a backbone modification. In a preferred embodiment of the method according to the invention, the RNA to be separated has a size of up to about 15000 nucleotides (as single stranded RNA 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. For this size of RNA, it has proved possible to achieve very good results with regard to purification of the RNA, since the method according to the invention is particularly well suited to RNA of this size. Optionally, however, smaller RNA fragments, for example with a length of 30-500 nucleotides may also be separated in this way.

In one embodiment the RNA to be purified results from in vitro transcription. The process of in vitro transcription is known to the skilled person.

Typically, the sample resulting from in vitro transcription is first subjected to a chromatography step, e.g. affinity chromatography, and/or filtered so that impurities like RNA fragments, DNA fragments, proteins, organic solvents, nucleoside triphosphates, spermidine and buffer components such as salts and detergents are removed.

Suitable filters include depth filtration, charged depth filtration and similar microfiltration techniques. A preferred filtration step is TFF (tangential flow filtration).

Optionally, the sample resulting from in vitro transcription can also be subjected to other chromatographic purification steps prior to subjecting it to the anion exchange chromatography step according to the present invention.

The sample comprising the target RNA and potentially other impurities from which the RNA shall be purified is then subjected to a chromatographic separation on a chromatography matrix comprising anion exchange groups. For this the sample is loaded onto the chromatography matrix. The final composition of the sample loaded onto the matrix is called the feed.

Preferably, the feed has a pH between 6.5 and 8.5.

Typically, the feed preparation and also the chromatographic separation are performed at or around room temperature. But it is also possible to work at other temperatures, e.g. between 10 and 35 °C, most preferred between 20 and 28 °C.

The feed preferably is adjusted to an electrolytic conductivity between 10 to 50 mS/cm, most preferably to 15 and 35 mS/cm.

Conductivity adjustment is done by addition of salt, salt concentrate solutions, or, respectively, dilution with a low conductivity buffer or neat water. For feed conductivity adjustment by salt supplementation preferably sodium or potassium chloride are used, but any other salt commonly used in purification applications such as e.g. salts from sulfate, acetate, carbonate/bicarbonate, phosphate or citrate might be considered as well depending on the adjustment of the concentration of chloride, the adjustment of the conductivity, and effect on the binding capacity of the matrix.

Column equilibration buffer are typically buffers matching the pH and conductivity of the feed loaded onto the chromatography material. Typically buffers with pH between 6.5 and 8.5 and conductivity between 10 to 50 mS/cm are selected but buffers out of that range are applicable as well.

Suitable, exemplary equilibration buffers comprise a chloride concentration between 100 and 200 mM. The equilibration buffer pH is for example around 8.0, adjusted with TRIS buffer, and conductivity between 10 and 50 mS/cm. After loading the matrix is preferably washed with at least one wash buffer. The wash buffer might be identical to the equilibration buffer or different from the equilibration buffer. The matrix might also be washed with 2, 3 or 4 different wash buffers. Preferably, one of the wash buffers comprises a chaotrope and is more basic than the equilibration buffer, typically around 0.5 to 1 pH unit higher.

Preferably the pH and the conductivity of the first wash buffer is identical or similar to the pH and the conductivity of the equilibration buffer and the load feed.

Preferably, more than one wash buffer is used whereby the pH of the second or third wash buffer is between pH 8 and 9 and the conductivity and conductivity of the second or third wash buffer is preferably different from that of the equilibration buffer and the load feed.

In one embodiment the wash buffer comprises a chaotrope and has a conductivity similar to the conductivity of the equilibration buffer, i.e. e.g. a chloride concentration between 100 and 200 mM.

In one embodiment the wash buffer has a neutral pH, e.g. between 6.5 and 7.5, does not comprise a chaotrope and has a conductivity and thus a salt concentration higher than the equilibration buffer.

In one embodiment the wash buffer has a pH below 7.5, e.g. between 6.0 and 7.5, and comprises a chaotrope.

The use and the combination of the wash buffers depends on the impurities present in the sample. The skilled person is able to adjust the composition and combination of wash buffers to the respective sample properties. In a preferred embodiment, at least two wash buffers are used whereby the pH and the conductivity of the first wash buffer is identical or similar to the pH and the conductivity of the equilibration buffer and the load feed and one of the following wash buffers, preferably the second wash buffer comprises a chaotrope and has a pH above the pH of the equilibration buffer. Typically, the concentration of the one or more chaotropes is in the range between 0.1 to 1.5M.

Elution of the target RNA is then done by using an elution buffer.

The elution buffer has a pH higher than the pH of the equilibration buffer and the elution buffer has a higher conductivity than the equilibration buffer. In addition, the elution buffer comprises a chaotrope.

In one embodiment the pH of the elution buffer is above pH 8, preferably between pH 9.0 and 11 . In one embodiment the elution buffer comprises between 500 and 2500 mM chloride, typically sodium chloride. The elution buffer also comprises a chaotrope.

Elution can be performed by directly changing from the last wash buffer to 100% of the elution buffer. But elution is preferably performed by gradient elution with linear or stepwise change of the elution buffer. Preferably, elution starts with 100% of the last wash buffer and then the composition of the elution buffer is changed with linear or stepwise gradient to 100% of a buffer that has a pH between 9 and 11 and comprises between 1000 and 2500 mM chloride.

The term “column volume” (CV) refers to the volume inside of a packed column or generally packed housing. This volume includes the chromatography matrix, the interstitial volume (volume outside of the matrix), and the own internal porosity (pore volume) of the matrix. If the matrix is a membrane the column volume is typically called the membrane volume (MV).

In a preferred embodiment, for elution, the applied linear or step gradient lasts around 50 to 500 column/membrane volumes (CV/MV) plus an optional additional hold step at target elution buffer for at least 10 CV/MV, typically around 20 to 100 CV/MV.

In a very preferred embodiment, elution is done with a buffer comprising chloride. Typically, in the course of the elution, the chloride content is increased, e.g. from 0 to 100 mM to 500 to 2500 mM and the pH is increased from a pH between 7.5 and 8.5 to a pH between 9 and 11 . The concentration of the chaotrope is preferably kept constant. Typically, the chaotrope is present in a concentration between 0.1 and 1 .5 M, preferably between 0.25 and 1 M.

In a preferred embodiment the chaotrope is selected from one or more of the group of arginine, sodium perchlorate and guanidine carbonate. Most preferred the chaotrope is arginine.

Preferably, in the method of the present invention a sample comprising a RNA and 50 to 400 mM chloride and having a pH between 6.5 and 8.5 is loaded on a chromatography matrix comprising anion exchange groups. Preferably, the matrix is washed with at least one wash buffer. Elution is performed with a linear or stepwise gradient up to a final elution buffer comprising between 500 to 2000 mM chloride and having a pH between 9.5 and 11.

In a preferred embodiment, the chromatography matrix is a membrane. It has been found that the method of the present invention even works when using membranes at flow rates between 5 and 25 MV/min. By performing the method of the present invention, the target RNA can be obtained with high yields and high purity. The inventors have found that, unexpectedly, performing elution with a buffer comprising a chaotrope in combination with increasing pH and conductivity of the elution buffer has an influence on the elution of the target RNA and impurities. Even longer mRNA with 1 ,000 to 10,000 nucleotides can be effectively purified.

The purification can be performed at ambient temperature, e.g. at a temperature between 20 and 28 °C. There is no need to apply higher temperatures.

This finding provides a scalable process with higher yield at ambient temperature so that disadvantages relating to the use of a higher temperature are avoided.

The present invention is further illustrated by the following figures and examples, however, without being restricted thereto.

The entire disclosure of all applications, patents, and publications cited above and below as well as corresponding European patent application EP24182183.4, filed on June 14, 2024, are hereby incorporated by reference.

Examples

The following examples represent practical applications of the invention.

Materials: Following feed material used for this experiment: In vitro transcription mRNA EPO: 858nt contained chemically modified 5moU and purified by oligo(dT) affinity chromatography (TriLink Bio Technologies) In vitro transcription mRNA FLuc: 1929nt purified by oligo(dT) affinity chromatography (TriLink Bio Technologies) Membrane: membrane absorber Natrix® Q Micro 0.2mL (Merck KGaA) Resin: minichrom Eshmuno® Q 1 mL column (Merck KGaA), Chrom System: AKTA Avant™ 25 (Cytiva)

Example 1

Preparative mRNA purification - Salt gradient vs dual gradient

Two elution techniques of salt gradient (according to prior art) and dual gradient with chaotrope (according to the present invention) were evaluated as successive process steps using strong anion exchange membrane absorber Natrix® Q Micro 0.2mL with quaternary amine as ligand. An 858nt mRNA encoding EPO protein containing chemically modified 5moU and chemically unmodified 1929nt mRNA encoding FLuc protein were utilized as the feed material for this experiment. Experimental procedures are summarized in Table 1. Figures 1 and 2 illustrate the results of the purification runs.

Table 1 . Chromatographic procedure for Natrix® Q operation used in Example 1 first with salt gradient according to prior art and then with dual gradient with chaotrope according to the present invention.

In terms of behavior of mRNA EPO, the partial mRNA EPO eluted broadly during NaCI gradient elution at pH 8.0 while the rest of the mRNA still bound to Natrix® Q (Fig. 1A). Subsequently, bound mRNA EPO was completely eluted by increasing pH and NaCI concentration in the presence of 0.5M !_(+)- Arginine as a chaotropic salt (Fig. 1 B). Regarding to longer mRNA FLuc, most of mRNA FLuc showed low elution efficiency by NaCI gradient due to the strong hydrogen bonds as expected (Fig. 2A). Subsequently, bound mRNA FLuc was completely eluted by the dual gradient as well as mRNA EPO (Fig. 2B). mRNA with different nucleotides in length showed similar behavior during chromatographic operation. In addition to major peak of ssRNA, other three peaks were observed on later phase of dual gradient elution. They seem to be mRNA-related impurities generated in the process of in vitro transcription. As a result of this evaluation, dual gradient method at ambient temperature enables the effective elution, high selectivity and impurities removal for RNA transcripts of greater than 500nt in length.

Example 2

In chromatography processes, a step elution method is typically utilized for pilot and commercial manufacturing scale. In addition, to estimate the mRNA yield using strong anion exchange membrane and resin, Natrix® Q and Eshmuno® Q were evaluated for these experiments and eluate with step elution were fractionated for analytics. Experimental procedures are summarized in Tables 2 and 3. Figures 3 and 4 illustrate the results of the step elution.

Table 2. Chromatographic procedure for step elution operation for mRNA EPO used in Example 2.

* MV means the required membrane volume for Natrix® Q operation while CV means the required column volume for Eshmuno® Q operation.

Table 3. Chromatographic procedure for step elution operation for mRNA FLuc used in Example 2.

For EPO, total yield was 92.6% and 95.0% for Natrix® Q and Eshmuno®

Q, respectively (Table 4). For FLuc, total yield was 77.2% and 75.3% for Natrix® Q and Eshmuno® Q, respectively (Table 5). In this case, slight variations in yield were observed depending on the length of mRNA. In particularly, longer mRNA FLuc had a high proportion of impurities, resulting in lower yield compared to ssRNA alone of EPO. Even with this assumption, this dual step elution method was still able to maintain high yield. With this result, strong anion exchanger membrane and resin shows the comparable performance in terms of yield and impurities separation, and then demonstrated to switch easily to step elution with high yield for preparative mRNA purification (Fig. 3A-B, 4A-B).

Table 4 Mass balance of step elution for mRNA EPO.

Table 5 Mass balance of step elution for mRNA FLuc.

* Total yield is likely to be improved further - approximately 10% higher by using appropriate buffer blank.

In order to determine impurities removal, the levels of dsRNA of load and eluate were also quantified by double-strand RNA ELISA kit (J2 based, Exalpha Biologicals, 10613002). As shown in Table 6, anion exchange chromatography shows superior dsRNA removal with more than 97% compared to the load.

Table 6. Mass balance of dsRNA.

ND: not detected

Example 3: Salt gradient elution at fixed high pH

To further compare the method of the present invention with prior art, the potential of salt gradient elution method at fixed pH (as known in the art) was evaluated. In this trial, pH 10.5 was selected to avoid the alkaline hydrolysis of mRNA, which is degraded rapidly at the alkaline pH with > 11 .0. So, salt gradient elution at fixed pH 10.5 using Natrix® Q are conducted. Experimental procedures are summarized in Table 7. Figure 5 illustrates the results of purification run.

During salt gradient, small three peaks were observed on the chromatogram. Surprisingly, most of mRNA FLuc was not eluted by salt gradient at pH 10.5 in the presence of 0.5M Arginine (Fig. 5). Natrix® Q maintains the peak separation performance while most of mRNA FLuc eluted during CIP step, result in lower yield compared to dual gradient elution. This data indicates that optimal elution condition combined pH, salt concentration and chaotropic salt work cooperatively to improve the mRNA elution especially from strong anion exchangers. Table 7. Chromatographic procedure for salt gradient elution at pH 10.5 used in Example 3.

Claims

1 . A method for purifying RNA (ribonucleic acid) comprising a) Providing a sample comprising said RNA b) Loading the sample onto a chromatography matrix comprising anion exchange groups c) Optionally washing the chromatography matrix with a wash buffer d) eluting RNA bound to the chromatography with a buffer, having a pH and salt concentration higher than that of the loading and wash buffer and comprising a chaotrope.

2. Method according to claim 1 characterized in that the method is performed at a temperature below 35°C.

3. Method according to claim 1 or claim 2 characterized in that the method is performed at a temperature between 20 and 28 °C.

4. Method according to one or more of claims 1 to 3, characterized in that in step d) during elution the pH of the elution buffer is changed to a higher pH in the range between 8.5 and 11 .

5. Method according to one or more of claims 1 to 4, characterized in that in step d) during elution the salt concentration of the elution buffer is changed to a higher salt concentration in the range between 0 and 2 M salt.

6. Method according to one or more of claims 1 to 5, characterized in that the salt is NaCI.

7. Method according to one or more of claims 1 to 6, characterized in that the chaotrope is selected from the group of arginine, guanidine carbonate and/or sodium perchlorate.

8. Method according to one or more of claims 1 to 7, characterized in that the pH of the elution buffer is induced by a buffering system using TRIS.

9. Method according to one or more of claims 1 to 8, characterized in that the chromatography matrix is a resin or a membrane.

10. Method according to one or more of claims 1 to 9, characterized in that the RNA is 800 to 5,000 nucleotides or base pairs in length.

11 . Method according to one or more of claims 1 to 10, characterized in that the removal of impurities in the purified RNA obtained in step d) compared to the sample provided in step a) is at least 90%.

12. Method according to one or more of claims 1 to 11 , characterized in that the yield of the purified RNA in step d) is at least 65%.

13. Method according to one or more of claims 1 to 12, characterized in that the anion exchange groups comprise trimethylammonium (- CH₂N(CH₃)3⁺) or triethylammonium ( CH₂CH₂N(CH₂CH3)3⁺).

14. Method according to one or more of claims 1 to 13, characterized in that elution in step d) is done by linear or stepwise gradient elution.