WO2025073922A1

WO2025073922A1 - Enhanced chromatographic separations of components in a mixture by employing chaotropic elution

Info

Publication number: WO2025073922A1
Application number: PCT/EP2024/077990
Authority: WO
Inventors: Andreja Gramc LIVK; Timotej ŽVANUT; Aleš ŠTRANCAR; Tjaša PLESNIČAR
Original assignee: Sartorius Bia Separations doo
Current assignee: Sartorius Bia Separations doo
Priority date: 2023-10-04
Filing date: 2024-10-04
Publication date: 2025-04-10
Anticipated expiration: 2026-04-04

Abstract

A method of chromatographic separation of components of an aqueous mixture by means of a chromatographic material comprising the steps (i) to (iv) of: (i) providing the aqueous mixture of the components, (ii) contacting the aqueous mixture with the chromatographic material, (iii) followed by starting the separation by contacting the chromatographic material with an aqueous environment having a low chaotropicity level (S) which is adjusted by means of a first chaotropic salt (1), (iv) followed by increasing the low chaotropicity level (S) to a terminal aqueous environment having a higher chaotropicity level (T) which is adjusted by means of a second chaotropic salt (2), wherein the second chaotropic salt (2) has a higher chaotropicity than the first chaotropic salt (1), and, optionally, (v) collecting the components becoming separated due to the increasing chaotropicity level of the aqueous environment.

Description

Enhanced Chromatographic Separation of Components in a Mixture by Employing Chaotropic Elution

The invention is related to a method of separation of components in an aqueous mixture by means of chromatography.

Background of The Invention

There is a growing need for achieving better characterization of components in complex mixtures such as those obtained from biological sources. A challenging task is to ensure a higher purity of a target entity in a final product, in particular when biomolecules such as proteins, nucleic acids, or even more complex biological materials, e. g. virus particles, whole cells, or cell components are concerned. The removal of impurities is one of the most challenging steps in the production of biomolecules or isolation of the vehicles, such as AAV capsids loaded with genetic material. The same is true to produce virus derived products, as some impurities in virus-based vaccines can cause serious side effects. These impurities need to be detected and removed to achieve high potency and safety throughout the production process.

This may be very difficult when impurities resemble the final product (e.g., as partially filled capsids resemble full AAV capsids, similar length of host cell DNA inside AAV capsids, different isoforms of gene of interest). In these cases, fine elution tuning is crucial to gain selectivity.

There is also a lack of understanding and methods to specifically detect, identify, and separate impurities from a complex mixture of impurities and contaminants, which often behave in a similar way as the target entity. An example of such behaviour are rAAVs where there is a small charge difference between empty, full, and partially filled capsids. Full capsids have almost the same size as the empty capsids but have a slightly lower isoelectric point (pl) than empty rAAV capsids (a difference in the range of 0.4 pH units). Therefore, removing the unwanted contaminants and impurities by conventional methods often leads to a loss of yield of the target entity.

Chromatography has been proven to be a powerful tool in separation, purification, and isolation of target entities in the production process. In particular, small molecules can be purified by different types of chromatographic methods such as ion exchange chromatography, hydrophobic interaction chromatography, affinity chromatography, and others. The gold standard for sample elution in chromatography is using an increasing salt gradient, typically in a linear mode.

Anion exchange chromatography (AEX) is used when the target entity - for example a protein - is negatively charged. The protein becomes negatively charged when chromatography is performed at a pH higher than the one corresponding to the isoelectric point (pl) of a protein. In this type of chromatography, the stationary phase is positively charged and negatively charged molecules are loaded to be attracted to the protein. Elution of the target entity, for example a protein, is mostly performed by means of high salt concentration in an elution buffer. The target entity is collected in fractions of high salt concentration. Such conditions are sometimes disadvantageous since the target entity might be damaged by partial or complete denaturation. Another drawback is the need for downstream removal of the high salt concentration from the target entity by e.g., dialyses, tangential flow filtration, or another separation approach, which may cause loss of yield of the target entity due to high shear forces applied.

Although AEX chromatography is a powerful tool in many manufacturing processes it sometimes reaches its limits in purifying and separating complex samples as described hereinabove.

Object of The Invention

It is an object of the present invention to provide a chromatographic method that avoids the drawbacks of the method of the state-of-the-art AEX chromatography. Another object is to provide a chromatographic method that enables separation, isolation, and purification of multiple subpopulations or isoforms of various biomolecules such as viruses, virus and lipid nanoparticles, exosomes, proteins, nucleic acids, and the like.

A further object of the present invention is to provide a chromatographic method utilizing binary, ternary, or multiple anion and/or cation exchange gradient conditions that enable a separation of multiple subpopulations, or isoforms, of different biomolecules such as viruses, bacteriophages, virus and lipid nanoparticles, exosomes, proteins, nucleic acids, and the like.

Still another object is to provide a method, which can be used analytically as well as in a preparative scale. Summary of The Invention

The objects underlying the present invention are accomplished by a method of chromatographic separation of components, especially biomolecules, in an aqueous mixture by means of a chromatographic material comprising the steps

(i) to (iv):

(i) providing the aqueous mixture of the components,

(ii) contacting the aqueous mixture with the chromatographic material,

(iii) followed by starting the separation by contacting the chromatographic material with an aqueous environment having a low chaotropicity level (S) which is adjusted by means of a first chaotropic salt (1),

(iv) followed by increasing the low chaotropicity level (S) to a terminal aqueous environment having a higher chaotropicity level (T) which is adjusted by means of a second chaotropic salt (2), wherein the second chaotropic salt (2) has a higher chaotropicity than the first chaotropic salt (1), and, optionally,

(v) collecting the components becoming separated due to the increasing chaotropicity level of the aqueous environment.

Alternatively worded, the invention is a method of chromatographic separation of components in an aqueous mixture by means of a chromatographic material comprising the steps (i) to (iv) of:

(i) providing the aqueous mixture of the components,

(ii) contacting the aqueous mixture with the chromatographic material,

(iii) followed by starting the separation by contacting the chromatographic material with an aqueous environment having a first level of chaotropicity(S) which is adjusted by means of a first chaotropic salt (1),

(iv) followed by linearly increasing the first level of chaotropicity(S) to a terminal aqueous environment having a second level of chaotropicity(T) which is adjusted by means of a second chaotropic salt (2), wherein the second chaotropic salt (2) has a higher chaotropicity than the first chaotropic salt (1), thereby eluting the components and, optionally, (v) collecting the components that are separated due to the increasing level of chaotropicity of the aqueous environment, wherein the second level of chaotropicity is higher than the first level of chaotropicity, wherein the components are biomolecules.

Conventionally, after contacting a mixture of components with a stationary phase, e. g. loading the mixture on an ion exchange chromatography material, the separation of components is affected by treating the chromatographic material with a gradient buffer. The gradient can be formed by increasing the ion strength of the buffer by increasing the concentration of buffer salts. The gradient can also be formed by varying the pH of the buffer contacting the chromatographic material.

However, according to the present invention the separation of the components is effected by replacing salt (e.g., anion, cation, or combination of both) of low chaotropicity with a salt (e.g., anion, cation, or combination of both) of higher chaotropicity. Advantageously, the concentration of the salt of higher chaotropicity can even be lower than the concentration of the low chaotropicity salt.

Preferably, the changes are gradual changes obtained by providing the aqueous environment having a first chaotropicity level (S) and providing the aqueous environment having a higher chaotropicity level (T) and changing the ratio between the aqueous environments, linear or step-wise, with linear being preferred.

The process of the invention provides the solution of the objects underlying the present invention.

Whereas chaotropicity was originally used to describe the impact of ions on nucleic-acid structure without any implied mechanism [1], a recent convention has arisen to define chaotropes as water-structure breakers that may modify the behaviour of liquid water even though this assumption remains controversial [2]. The use of entropic measurements of binary liquid mixtures [3] to predict chao- /kosmotropicity is based on this assumption.

Well-known is the so called Hofmeister series of chaotropic substances, ordered in increasing chaotropicity. The effects of these changes were first worked out by Franz Hofmeister, who studied the effects of cations and anions on the solubility of proteins [4].

Hofmeister discovered a series of salts that have consistent effects on the solubility of proteins and (it was discovered later) on the stability of their secondary and tertiary structure. Anions appear to have a larger effect than cations, [5] and are usually ordered.

Chaotropicity, as used in this application, is a property of a compound (especially a salt), independent from a concentration.

When employing the method of the invention for example for an anion exchange chromatography the following ions are preferred (starting from the less to more chaotropic) :

Anions: CHsCHzCOO’ < CH₃COO’ < HCOO’ < F < Ch < Br < I < CIO < SCN’

Cations: N(CH₃)₄ ⁺ < NH₄ ⁺ < Cs⁺ < Rb⁺< K⁺< Na⁺< Li⁺< Mg²⁺< Ca²⁺< Zn²⁺< Ba²⁺

This order can differ based on the nature of the chromatographic material selected for the particular object of separation.

For example, for cation exchangers the following ions are preferred (starting from the less to more chaotropic) :

Anions: CH₃CH2COO’ < Cl’ < Br

Cations: Li⁺ < Na⁺ < K⁺

These orders are preferably used for protein related analytes.

The skilled person is readily able to establish a proper series of chaotropicity valid for chromatographic materials other than anion exchange chromatography. In view of the teaching of the invention it is just a simple trial and error, far below the threshold of any undue experimentation. When adjusting the proper series of chaotropicity the skilled person takes - based on his or her common knowledge - into account that each salt differs in its ability to promote hydrophobic interactions with the chromatographic material (e.g. arranged in a column), ligand and sample. Additionally, when comparing or choosing appropriate salts, ion valence should also be considered.

Chaotropicity level or level of chaotropicity, as used in the application, is a property of an aqueous mixture comprising of one or more compounds (especially salts) and is a result of the type(s) and concentration(s) of compounds. In one embodiment of the invention the chaotropicity level (S) can be linearly increased in step (iv) to at least one intermediate chaotropicity level (I) having a lower chaotropicity level than the chaotropicity level (T), and thereafter the chaotropic level (I) can be increased to the higher chaotropic level (T).

In an alternative of the method of the invention the chaotropicity level (S) can be linearly increased in step (iv) to at least one intermediate chaotropicity level (I) having a lower chaotropicity level than the chaotropicity level (T), whereby at least one intermediate chaotropicity level (I) is adjusted by increasing the concentration of a first chaotropic salt or mixture of salts (1), and thereafter increasing the chaotropic level (I) to the higher chaotropic level (T).

Typically, the low chaotropicity level (S) can be adjusted by a first chaotropic salt or mixture of salts (1).

The terminal chaotropicity level (T) can be adjusted by a second chaotropic salt or mixture of salts (2) having a higher chaotropicity compared to the first chaotropic salt or mixture of salts (1).

In one of the simplest embodiments of the method of the invention a binary salt gradient is used which is established by the combination of the starting chaotropicity level (S) and the terminal chaotropicity level (T) or the intermediate chaotropicity level (I).

At least one intermediate chaotropic level (I) can be adjusted by an at least third chaotropic salt or mixture of salts (3) having a higher chaotropicity compared to the second chaotropic salt or mixture of salts (2).

If the chaotropicity levels (S), (I), and (T) are employed, a ternary gradient is used.

In another embodiment of the invention the component can be a biomolecule such as LNP, nucleic acids, DNA, RIMA, a protein, an enzyme, a glycoprotein, a polysaccharide, a lipid, and/or a fatty acid.

In still another embodiment of the invention the component can be a virus, a virus-like particle, a bacteriophage, an extracellular vesicle, such as an exosome or combinations thereof-.

Particularly, the virus can be:

(i) a parvovirus, (ii) an adeno associated virus (AAV), in particular AAV selected from the group consisting of serotype AAV1, AAV2, AAV3, AAV4, AAV5, AAV6, AAV7, AAV8, AAV9, AAVrh.10, AAV11, AAV 12,

(iii) a hybrid serotype, in particular a recombinant hybrid serotype like AAV2/8,

(iv) a chimera of AAV,

(v) a surface modified AAV, or

(vi) a synthetically derived AAV like particle.

In still another embodiment of the invention the chromatographic material can be:

(i) an ion exchange chromatography material,

(ii) a hydrophobic interaction chromatography material,

(iii) a mixed mode chromatography material,

(iv) a monolith anion exchanger

(v) a monolith cation exchanger

(vi) a monolith multimodal material,

(vii) a particulate anion exchanger and/or

(viii) a multimodal material.

In yet another embodiment of the invention the chromatographic material can be arranged in a membrane, which can be present in the form of particles arranged in a column, and/or a fibre column.

In still another embodiment of the invention the aqueous environment can be an aqueous solution comprising:

(i) chaotropic cations selected from the group consisting of tetramethyl ammonium, ammonium, caesium, rubidium, potassium, sodium, lithium, magnesium, calcium, zinc, and barium, protonated arginine, protonated guanidine, ordered in increasing chaotropicity and/or

(ii) chaotropic anions selected from the group consisting of propionate, acetate, formate, fluoride, chloride, bromide, iodide, nitrate, chlorate, perchlorate, and thiocyanate, ordered in increasing chaotropicity.

In a further embodiment of the invention the aqueous environment may be an aqueous solution comprising chaotropic salts, buffering agents, isotonic substances, auxiliary agents, and/or organic modifiers. In still another embodiment of the invention the isotonic substances can be selected from the group consisting of sucrose, sorbitol, mannitol, xylitol, and mixtures thereof.

In yet another embodiment of the invention the auxiliary agents can be non-ionic surfactants, such as poloxamer 188 or urea. The organic modifiers can be selected from the group consisting of acetonitrile, 1-butanol, t-butanol, propylene carbonate, isopropanol, ethanol, methanol, ethylene glycol, 1-propanol, and mixtures thereof.

Brief Description of The Figures

Figure 1 depicts a chromatogram showing the tryptophan fluorescence resolution of the separation of full/empty AAV capsids in a binary salt (more specifically cation) exchange gradient.

Figure 2A and 2B depicts a chromatogram showing the resolution of the separation of AAV capsids in a ternary salt cation exchange gradient.

Figure 3A and 3B depicts a chromatogram showing an enhanced effect on the resolution of empty/full AAV capsids by exchanging anions in the elution buffer in a ternary anion salt exchange gradient. Figure 3C depicts orthogonal analytics using mass photometry for collected fractions.

Figure 4A depicts a chromatogram showing the UV260/280 elution profile of an AAV sample and collection of fractions from semi-preparative run.

Figure 4B depicts a zoom-in chromatogram of full AAV capsids (fractions E2 and E3).

Figure 4C depicts a chromatogram showing on the left UV absorbance overlay and on the right tryptophan fluorescence overlay.

Figure 5 depicts a chromatogram showing the separation of AAV8 capsids using a two-dimensional chromatographic system.

Figure 6A depicts a chromatogram showing the tryptophan fluorescence resolution of the separation of ovalbumin isoforms.

Figure 6B depicts a silver stained SDS-PAGE gel of the ovalbumin sample.

Figure 7A depicts a chromatogram of the resolution of the separation of main pDNA isoforms. Figure 7B depicts AGE gel of three individual pDNA isoforms as well as a mixture of pFIX15 and lin pFIX15.

Figure 8 depicts a chromatogram showing the resolution of the separation of different exosome subpopulations.

Figure 9 depicts a chromatogram showing the separation of AAV8 capsids on monolithic QA column compared to particle QA column.

Figure 10 depicts a chromatogram showing the separation of AAV8 capsids on cation exchange column.

Figure 11A depicts a chromatogram of the resolution of the separation of main LNP species on strong AEX column (QA).

Figure 11B depicts a chromatogram of the resolution of the separation of main LNP species on weak AEX column (DEAE).

Detailed Description of The Invention

The method of the invention enables enhanced or even baseline separation of multiple subpopulations or isoforms of components of complex mixtures for example biomolecules of similar biophysical characteristics regarding their charge, hydrophobicity, solubility, ion binding capacity, etc., of proteins, nucleic acids, and viruses. The method employs moderate buffer conditions during chromatography, comprising of the following steps, contacting the mixture with a chromatographic material, for example strong or weak anion exchanger material using mild chaotropic salt conditions, eluting of different subpopulations from moderate to stronger chaotropic salt conditions.

In more detail, it is a method of chromatographic separation of components of an aqueous mixture by means of a chromatographic material comprising the steps (i) to (iv) of:

(i) providing the aqueous mixture of the components,

(ii) contacting the aqueous mixture with the chromatographic material,

(iii) followed by starting the separation by contacting the chromatographic material with an aqueous environment having a low chaotropicity level (S) which is adjusted by means of a first chaotropic salt (1), (iv) followed by increasing the low chaotropicity level (S) to a terminal aqueous environment having a higher chaotropicity level (T) which is adjusted by means of a second chaotropic salt (2), wherein the second chaotropic salt (2) has a higher chaotropicity than the first chaotropic salt (1), and, optionally,

According to the method of the invention, components of an aqueous mixture are separated and purified by chromatography using a solid material as stationary phase and a buffer as mobile phase.

In a preferred embodiment, the change of chaotropicity from level (S) to level (T) is made with a lower or identical concentration of salt, but with a higher chaotropicity.

The components which can be separated from contaminants are viruses, virus-like particles, bacteriophages, extracellular vesicles, and/or biomolecules such as LNPs, nucleic acids, DNA, RIMA, proteins, glycoproteins, polysaccharides, lipids, and/or fatty acids.

In case of viruses, the method of the invention enables purification of viruses like parvovirus, adeno associated virus (AAV), in particular AAV selected from the group consisting of serotype AAV1, AAV2, AAV3, AAV4, AAV5, AAV6, AAV7, AAV8, AAV9, AAVrh.10, AAV11, AAV12. Furthermore, AAV hybrid serotypes can be separated and purified, in particular recombinant hybrid serotype like AAV2/8; chimera of AAV, surface modified AAV, and synthetically derived AAV like particles.

In case of proteins, the method of the invention enables separation of isoforms of proteins e. g. ovalbumin isoforms (Fig. 6A and B). Figure 6A depicts the tryptophan fluorescence resolution of the separation of ovalbumin isoforms in binary salt anion exchange gradient. The loading of the ovalbumin sample is performed with a mild chaotropic salt, followed by elution of the isoforms with a strong chaotropic salt. Figure 6B shows a silver stained SDS-PAGE gel of the initial ovalbumin sample.

In case of nucleic acid separation, the method of the invention enables separation of the main pDNA isoforms (oc, lin and sc) and multimers. Figure 7A depicts the resolution of the separation of the main pDNA isoforms (oc, lin and sc) in a ternary salt (more specifically cation) exchange gradient. Loading is performed in mild chaotropic salt conditions, elution of oc and lin isoforms with moderate chaotropic salt and elution of sc isomer with strong chaotropic salt. Figure 7B depicts AGE gel of all three individual pDNA isoforms and a mixture of pFIX15 and lin pFIX15.

In case of exosomes, the method of the invention enables separation of some exosome isoforms. Figure 8 depicts the resolution of the separation of exosome species in a binary mixed salt exchange gradient. Loading is performed in mild chaotropic salt conditions, elution of exosomes with strong chaotropic salt is enabled.

In case of LNPs, the method of the invention enables separation some of exosome isoforms. Loading is performed with no salt, due to the poor retention of LNPs on QA or DEAE monolith. Elution of first LNP species is enabled by commonly used increasing salt gradient to a mild chaotropic conditions. However, elution of additional separated species is achieved by implementation of binary chaotropic gradient to strong chaotropic conditions (Figure 11A and B).

The stationary phase is a chromatographic material such as an ion exchange chromatography material, a hydrophobic interaction chromatography material, a mixed mode chromatography material, a monolith anion exchanger, a monolithic cation exchanger, a monolith multimodal material, a particulate anion exchanger and/or a multimodal material.

For example the ion exchange material is a strong or weak anion exchanger material with hydrogen bond properties and compounded with positively charged metal affinity ligand as a multimodal material, a monolith anion exchanger or multimodal material, a particulate anion exchanger or multimodal material, and/or an anion exchanger or multimodal material arranged in membranes, and/or particle packed anion exchanger or multimodal columns and/or fibre chromatography anion exchanger or multimodal fibre columns. The strong anion exchanger material comprises a quaternary amine ligand with commercial names such as Q, QA, QAE, QAM, TEAE, TMAM or TMAE maintaining consistent charge over the range of about pH 2 to pH 13. Quaternary amine anion exchanger materials are described for separation of empty and full capsid in [6].

The weak anion exchanger DEAE (diethylaminoethyl) material comprises a tertiary amine ligand with a pKa of about 11.5 and allows elution of empty and full capsid at moderate pH values [7]. The chromatographic material may be a cation exchanger. It may be a strong cation exchanger having for example sulfonic acid groups or a weak cation exchanger having for example carboxylic acid groups. Also, phosphonic acid groups are commonly used. It may be a monolith cation exchanger or multimodal material, a particulate cation exchanger or multimodal material, and/or a cation exchanger or multimodal material arranged in membranes, and/or particle packed cation exchanger or multimodal columns and/or fibre chromatography cation exchanger or multimodal fibre columns.

Hydrophobic interaction Chromatography (HIC) utilizes hydrophobic ligands, such as alkyl and aryl ligands. Typical groups are butyl, octyl and phenyl groups.

Multimodal chromatography also known as mixed-mode chromatography (MMC), refers to chromatographic methods that utilize more than one form of interaction between the stationary phase and analytes in order to achieve their separation. In one aspect, the MMC can be regarded as a subtype of AEX. MMC can be classified into physical MMC and chemical MMC. In the former method, the stationary phase is constructed of two or more types of packing materials. In the chemical method, one type of packing material containing two or more functionalities is used. One approach is to connect two commercial columns in series, which is termed a "tandem column". Another approach is a "biphasic column", produced by packing two stationary phases separately in two ends of the same column. A further approach is to homogenize two or more different types of stationary phases in a single column, which is termed a "hybrid column" or "mixed- bed column".

The monolith anion exchanger material is particularly capable of performing the method of the invention. Experiments described in the "Examples" section have been performed on such materials.

Also, other materials are suitable which are provided in the form of particulate anion exchanger material, and/or an anion exchanger material arranged in membranes, and/or particle packed anion exchanger columns and/or fibre chromatography anion exchanger or multimodal fibre columns.

The material may be a multimodal metal affinity exchanger material. That material compromises of properties of positively charged metal affinity ligand and weak anion exchanger with hydrogen bond properties [8]. It enables separation of a sub-population of empty capsids, first followed by full capsids in a linear magnesium chloride gradient and later in a high salt step where mostly empty capsids elute.

The separation, purification, or isolation of the target is performed in essentially aqueous environment. This environment contains the sample to be separated and contaminants. The contaminants can be impurities or contaminants originating from the production process of the target.

In the case of AAV which have been loaded with genetic material for gene therapeutic purposes, several unwanted substances are present which have to be removed before administering to a patient. For example, besides the desired loaded AAV (full AAV) empty AAV are present which have to be removed. Furthermore, contaminants like AAV aggregates or partially loaded AAV are present in the production batch. It goes without saying that the contaminants or impurities have to be removed.

In the following, for the sake of simplicity, the invention is described in detail by the separation process of full from empty AAV capsids using an anion exchange chromatographic material.

In a generic mode of conduction of the method of the invention it is performed in a binary gradient elution.

The process of removal of unwanted substances and separation of full AAV capsids from empty AAV capsids is started by providing a mixture from the manufacturing process of loaded AAV capsids in an aqueous environment. Contacting the mixture with the chromatographic material happens by loading the mixture, optionally in a loading buffer of low chaotropicity, e. g. comprising potassium chloride, on a column containing the chromatographic material. Particularly useful are monolithic anion exchangers. The loading buffer can have a chaotropicity lower than chaotropicity level (S).

When employing the method of the invention on anion exchange column the following ions are preferred for adjusting the chaotropicity level (starting from the less to more chaotropic) :

Anions: CHsCHzCOO’ < CHsCOO’ < HCOCF < F < Ch < Br < I < CIO < SCN’

A binary chaotropicity gradient is established by replacing the starting chaotropic salt (1) by the salt of the higher chaotropicity (2). The gradient can be generated by keeping the cation of the chaotropic salts (1) and (2) the same but changing the anion from an anion of lower chaotropicity to an anion of higher chaotropicity. For example, starting with sodium acetate as chaotropic salt (1) and changing the buffer conditions to sodium perchlorate as chaotropic salt (2).

Alternatively, the gradient can be generated by keeping the anion of the chaotropic salts (1) and (2) the same but changing the cation from a cation of lower chaotropicity to a cation of higher chaotropicity. For example, starting with sodium acetate as chaotropic salt (1) and changing the buffer conditions to barium acetate as chaotropic salt (2).

Moreover, the gradient can be generated by changing both the cations and the anions of lower chaotropicity (1) to cations and anions of higher chaotropicity (2). For example, starting with sodium acetate as chaotropic salt (1) and changing the buffer conditions to calcium bromide as chaotropic salt (2). The gradient can be also generated by combining and exchanging different multiple mixtures of salts. For example, starting with a mixture of sodium acetate and guanidinium hydrochloride as chaotropic mix (1) and changing the buffer conditions to calcium bromide and magnesium chloride as chaotropic mix (2).

After loading, the separation starts by rinsing the chromatographic material with a buffer comprising a low chaotropicity level (S) which is adjusted by a first chaotropic salt (1), for example magnesium chloride. The low chaotropicity level can also be adjusted by a mixture of chaotropic salts, for example a mixture of magnesium chloride and magnesium bromide. The second chaotropic salt (2) is selected dependent on the chaotropic salt or mixture of salts (1), by considering the chaotropicity level (S) and choosing a salt or mixture of salts resulting in a higher chaotropicity level (T). For example, if magnesium chloride is chosen as first chaotropic salt (1), the second chaotropic salt (2) can be magnesium perchlorate. The higher chaotropicity level (T) is governed by the anion shift from chloride to perchlorate because the cation magnesium is the same.

If the chaotropicity level (S) is adjusted by a mixture of magnesium chloride and magnesium bromide, the salt or mixture of salts (2) providing the higher chaotropicity level (T) can be magnesium perchlorate or composed of magnesium chlorate and magnesium iodide. Also, in this example the higher chaotropicity level (T) is governed by the anion shift from chloride/sulfate to chlorate/perchlorate because the cation magnesium is the same. The selection rules are explained in more detail as follows: The cation is e. g. magnesium. In case the anion for adjusting chaotropicity level (S) is CHsCHzCOO' then the higher chaotropicity level (T) can be adjusted by using the anion(s) CHsCOO', HCOO', F', Cl', Br, T, CIOT, and/or SCN'. In case the anion for adjusting chaotropicity level (S) is CHsCOO', then the higher chaotropicity level (T) can be adjusted by using the anion(s) HCOO', F', Cl', Br, T, CIOT, and/or SCN'. In case the anion for adjusting chaotropicity level (S) is HCOO', then the higher chaotropicity level (T) can be adjusted by using the anion(s), F', Cl', Br, T, CIOT, and/or SCN'. In case the anion for adjusting chaotropicity level (S) is F', then the higher chaotropicity level (T) can be adjusted by using the anion(s), Cl', Br, T, CIO₄', and/or SCN'. In case the anion for adjusting chaotropicity level (S) is Cl', then the higher chaotropicity level (T) can be adjusted by using the anion(s), Br, T, CIO₄', and/or SCN'. In case the anion for adjusting chaotropicity level (S) is Br , then the higher chaotropicity level (T) can be adjusted by using the anion(s), T, CIO₄', and/or SCN'. In case the anion for adjusting chaotropicity level (S) is T, then the higher chaotropicity level (T) can be adjusted by using the anion(s), CIC ' , and/or SCN'. In case the anion for adjusting chaotropicity level (S) is CIC ', then the higher chaotropicity level (T) can be adjusted by using the anion SCN'.

Regarding the selection rules considering a constant anion are explained in more detail as follows: The anion is e. g. acetate. In case the cation for adjusting chaotropicity level (S) is N(CH3)4⁺, then the higher chaotropicity level (T) can be adjusted by using the cation(s) NH4⁺, Cs⁺, Rb⁺, K⁺, Na⁺, Li⁺, Mg²⁺, Ca²⁺, Zn²⁺, Ba²⁺. In case the cation for adjusting chaotropicity level (S) is NH4⁺, then the higher chaotropicity level (T) can be adjusted by using the cation(s), Cs⁺, Rb⁺, K⁺, Na⁺, Li⁺, Mg²⁺, Ca²⁺, Zn²⁺, Ba²⁺. In case the cation for adjusting chaotropicity level (S) is Cs⁺, then the higher chaotropicity level (T) can be adjusted by using the cation(s), Rb⁺, K⁺, Na⁺, Li⁺, Mg²⁺, Ca²⁺, Zn²⁺, and/or Ba²⁺. In case the cation for adjusting chaotropicity level (S) is Cs⁺, then the higher chaotropicity level (T) can be adjusted by using the cation(s), Rb⁺, K⁺, Na⁺, Li⁺, Mg²⁺, Ca²⁺, Zn²⁺, and/or Ba²⁺. In case the cation for adjusting chaotropicity level (S) is Rb⁺, then the higher chaotropicity level (T) can be adjusted by using the cation(s), K⁺, Na⁺, Li⁺, Mg²⁺, Ca²⁺, Zn²⁺, and/or Ba²⁺. In case the cation for adjusting chaotropicity level (S) is K⁺, then the higher chaotropicity level (T) can be adjusted by using the cation(s), Na⁺, Li⁺, Mg²⁺, Ca²⁺, Zn²⁺, and/or Ba²⁺. In case the cation for adjusting chaotropicity level (S) is Na⁺, then the higher chaotropicity level (T) can be adjusted by using the cation(s), Li⁺, Mg²⁺, Ca²⁺, Zn²⁺, and/or Ba²⁺. In case the cation for adjusting chaotropicity level (S) is Li⁺, then the higher chaotropicity level (T) can be adjusted by using the cation(s), Mg²⁺, Ca²⁺, Zn²⁺, and/or Ba²⁺. In case the cation for adjusting chaotropicity level (S) is Mg²⁺, then the higher chaotropicity level (T) can be adjusted by using the cation(s), Ca²⁺, Zn²⁺, and/or Ba²⁺. In case the cation for adjusting chaotropicity level (S) is Ca²⁺, then the higher chaotropicity level (T) can be adjusted by using the cation(s), Zn²⁺ and/or Ba²⁺. In case the cation for adjusting chaotropicity level (S) is Zn²⁺, then the higher chaotropicity level (T) can be adjusted by using the cation Ba²⁺.

The skilled person readily recognises further combinations from these comprehensive explanations. The skilled person also recognises that it is possible to choose a particular anion/cation pair for adjusting chaotropicity level (S), e.g., ammonium acetate (providing rather low chaotropicity), and to select a different anion/cation pair for adjusting chaotropicity level (T), e. g. magnesium chloride (providing intermediate chaotropicity by higher than chaotropicity level (S)) or magnesium perchlorate (providing rather high chaotropicity).

In contrast to conventional anion exchange chromatography which is based on a salt or pH gradient and not by a chaotropicity gradient provoked by exchanging chaotropic cations or anions, the concentration of the chaotropic salts (1), (2) or (3) in the applied buffers do not vary significantly.

Rinsing of the column is continued by gradually increasing the chaotropicity level (S) to a terminal higher chaotropicity level (T) adjusted by a second chaotropic salt (2) having a higher chaotropicity e. g. magnesium perchlorate.

According to the series of chaotropicity according to the invention (magnesium) chloride is a weaker chaotrope than (magnesium) perchlorate. These relations are exemplified in Example 2 and the chromatograph of Fig. 1. The concentration of buffer A representing chaotropicity level (S) gradually decreases and to a complementary extent the concentration of buffer B, representing the terminal higher chaotropicity level (T), increases. Under such conditions first empty AAV capsids (peak E) elute followed by full AAV capsids (peak F).

Alternatively, the gradient can be generated by keeping the anion of the chaotropic salts (1) and (3) the same but changing the cation from a cation of lower chaotropicity to a cation of higher intermediate chaotropicity (I). For example, starting with ammonium acetate as chaotropic salt (1) the buffer conditions are changed to sodium acetate as intermediate chaotropic salt (3). Subsequently, the intermediate chaotropicity level (I) is increased by barium acetate as chaotropic salt (2) establishing the terminal level of chaotropicity (T).

It is also possible to change in the series from chaotropic salt (1) to the intermediate chaotropic salt (3) both the anions and cations as long as the resulting chaotropicity increases from chaotropic salt (1) to chaotropic salt (3), e. g. from ammonium acetate [chaotropic salt (1)] to sodium chloride [chaotropic salt (I)]. Likewise, the change from chaotropic salt (3) to chaotropic salt (2) may be performed, e. g. from sodium chloride [chaotropic salt (3)] to potassium perchlorate [chaotropic salt (2)].

Figure 2A and B depicts the resolution of the separation of AAV capsids in a ternary salt cation exchange gradient. The sample is loaded at mild chaotropic salt conditions and elution of empty AAV capsids with moderate chaotropic salt and elution of full AAV capsids with strong chaotropic salt is achieved. Figure 2B is a detailed view of Figure 2A. Figure 2B shows multiple signal detection (tryptophan fluorescence, UV absorbance 260 and 280 nm and light scattering). From the comparison of the results achieved with the binary gradient of Fig. 1 and the ternary gradient of Fig. 2, the improved resolution of the peaks representing empty and full AAV capsids becomes evident: OSEF = 1.26 in Fig. 1 versus ResEF=1.88.

In an alternative of the method of the invention the chaotropicity level (S) is increased in step (iv) to at least one intermediate chaotropicity level (I) having a lower chaotropicity than the chaotropicity level (T), whereby at least one intermediate chaotropic level (I) is adjusted by increasing the concentration of the first chaotropic salt (1), and thereafter increasing the chaotropic level (I) to the higher chaotropic level (T).

Figure 3A and 3B depict the resolution of the separation of AAV capsids in a ternary anion salt exchange gradient.

In this embodiment of the invention - addressed as ternary gradient elution - the chaotropicity level (S), e. g. determined by potassium chloride, is increased to an at least one intermediate chaotropicity level (I) having a lower chaotropicity than the chaotropicity level (T), e. g. determined by magnesium perchlorate, whereby the at least one intermediate chaotropic level (I) can be adjusted by an at least third chaotropic salt (3), e. g. lithium chloride, and thereafter the chaotropic level (I) is be increased to the higher chaotropic level (T). A ternary chaotropicity gradient is established by replacing the starting chaotropic salt (1) by a salt of intermediate chaotropicity (I). The gradient can be generated by keeping the cation of the chaotropic salts (1) and (3) the same but changing the anion from an anion of lower chaotropicity to an anion of higher intermediate chaotropicity (I). For example, starting with ammonium acetate as chaotropic salt (1) the buffer conditions are changed to ammonium chloride as intermediate chaotropic salt (3). Subsequently, the intermediate chaotropicity level (I) is increased by ammonium perchlorate as chaotropic salt (2) establishing the terminal level of chaotropicity (T).

In this case, a baseline separation of empty and full AAV capsids has been achieved where each individual peak, empty and full can be accurately measured. Loading of the sample is performed in a buffer having a low chaotropicity level provided by magnesium acetate. The subsequent elution of empty AAV capsids happens by employing a buffer comprising magnesium chloride, providing a moderate chaotropicity level (intermediate level of chaotropicity). The elution of full AAV capsids is performed by employing a third buffer comprising the strong chaotropic salt (2) magnesium perchlorate, providing a high level of chaotropicity (terminal level of chaotropicity). Please note besides the baseline separation of the two AAV capsid species the excellent resolution of ResEF=6.10. For isolation of the enriched fraction containing full AAV capsids the elute between 8 min to 9 min retention time can be collected as shown in Fig 3A.

Figure 3B is a detailed view of Figure 3A. Figure 3B shows multiple signal detection (tryptophan fluorescence, absorbance at 260 nm and 280 nm and light scattering). Figure 3C represents mass photometry results of anion exchanged collected fractions.

One advantage provided by the method of the invention is that the set-up can be used for preparative purposes. A typical chromatograph of a preparative separation according to the invention is depicted in Fig. 4A.

The preparative separation of empty from full AAV capsids was performed by using a monolithic a strong anion exchange column. The preparative separation employed a ternary salt gradient. The chaotropic salt (1) was magnesium acetate The chaotropicity was increased to the intermediate level of chaotropicity by the salt magnesium chloride (chaotropic salt (2)). The terminal chaotropicity level was adjusted by the salt magnesium perchlorate (chaotropic salt (3)). The fractions El to E3 from preparative run were collected as shown in Figure 4A. Fig. 4B shows a zoom-in chromatogram of possible subpopulations of full AAV capsids.

The collected fractions El, E2, E3, and E4 from the preparative run were individually analysed by an orthogonal analytical method using a strong anion exchanger, CIMac™ AAV full/empty column. The elution was achieved in a linear salt gradient of magnesium acetate. A multidetector setup was used, UV 260 and 280 nm, intrinsic protein fluorescence that mostly induced by tryptophan. Tryptophan fluorescence was monitored at extinction 280 nm and emission at 348 nm.

Figure 4C shows UV 260/280 nm results where, the El fraction represents empty capsids with a 260/280 wavelength ratio of 0.64. E2 and E3 fractions are populated with full capsids with 260/280 wavelength ratio of 1.33 and 1.34 respectively. Fraction E4 (high salt wash) highly likely represents damaged AAV, heavy AAV capsids and/or aggregates. Partially filled AAV capsids were not observed (possibly present in low concentrations - under the limit of detection).

The method of invention was used for analysis of AAV8 lysate samples. Two- dimensional chromatographic system, PATfix® AAV Switcher (Sartorius BIA Separations, Ajdovscina, Slovenia) enabled analysis of complex lysate samples in the upstream of process development. The first column is served for purification of sample and was a strong cation exchange and the second column was anion exchange column where empty, partially filled, and full capsid separation was achieved.

The novel method enabled resolution between empty, partially filled, and full capsids for purified AAV8 standard and lysate sample as shown in Figure 5. The figure depicts light scattering overlay of AAV capsids separation in a ternary anion salt exchange gradient. Overlay shows separation in purified AAV8 standard (contains empty and full fraction) and AAV8 lysate sample.

Figure 6A depicts the tryptophan fluorescence resolution of the separation of ovalbumin isoforms which was performed by means of a strong anion exchanger (QA) using a binary chaotropic gradient. The figure depicts results showing the fluorescence detection of four distinct peaks corresponding to ovalbumin isomers.

To confirm possible ovalbumin isoforms a silver stained SDS-PAGE gel of the ovalbumin sample (different sample dilutions loaded) was run. The respective result is shown in Figure 6B. At least two bands were observed at approximately 45 kDa. Additional bands were also observed from 85 to 150 kDa representing possible ovalbumin dimers or trimers. Possible impurities or individual ovalbumin subunits were observed at approximately 30 kDa, 25 kDa, 17 kDa and 10 kDa. For size assessment a PageRuler prestained protein ladder (Thermo Fisher Scientific Inc., USA) was used.

Figure 7A depicts the resolution of the separation of main pDNA isoforms using a weak anion exchanger (DEAE). A Ternary salt (more specifically cation) exchange gradient was used.

The separation of the main plasmid DNA isoforms was performed using a CIMac™ pDNA column (Sartorius BIA Separations, Ajdovscina, Slovenia).

The sample was a mixture of pFIX 15 (15 kbp plasmid DNA containing open circular and super coiled plasmid DNA) and of linear pFIX 15 (15 kbp linear plasmid DNA), (all Sartorius BIA Separations, Ajdovscina, Slovenia). The gradient was composed of low chaotropicity (cesium chloride), intermediate chaotropicity (lithium chloride) and terminal chaotropicity (calcium chloride). Figure 7A depicts results showing UV 260 nm absorbance detection of three peaks corresponding to each pDNA isoforms.

Figure 7B depicts an AGE gel of all three individual pDNA isoforms as well as a mixture of pFIX15 and lin pFIX15. For size assessment Supercoiled DNA ladder (Figure 6B left edge) and GeneRuler™ DNA ladder (Thermo Fisher Scientific Inc., USA) was used (Figure 6B right edge). The first two bands represent oc and sc respectively, the last band represents the linear pDNA isoform.

Figure 8 depicts the resolution of the separation of exosome isoforms using a strong anion exchanger (QA). A binary salt (more specifically mixed ion) exchange gradient was used.

The separation of exosome isoforms was performed using a CIMac™ AAV e/f column (Sartorius BIA Separations, Ajdovscina, Slovenia).

The gradient was composed of low chaotropicity (potassium acetate) and terminal chaotropicity (sodium perchlorate). Figure 8 depicts results showing UV 260 nm and UV 280 nm absorbance as well as light scattering at 90°angle and tryptophan fluorescence detection of three distinct peaks corresponding exosome subpopulations. The method of invention was also used for analysis of AAV8 sample on QA particle column compared to QA monolithic column. Figure 9 depicts the resolution of the separation of predominantly empty and full AAV8 capsids separated by strong anion exchanger (QA) columns: CIMac™ QA HR. monolithic column (Sartorius BIA Separations, Ajdovscina, Slovenia) and DNAPac™ PA200 particle column (Thermo Fischer Scientific, USA). The gradient was composed of low chaotropicity (magnesium acetate) and intermediate chaotropicity (magnesium chloride) where elution of predominantly empty AAV8 occurred followed by the elution of predominantly full AAV8 capsids by increasing salt concentration of low chaotropicity salt (magnesium acetate). Figure 9 depicts results showing UV 260 nm and UV 280 nm absorbance of two distinct peaks corresponding empty and full AAV8 capsids.

Figure 10 depicts the resolution of the separation of AAV8 capsids using a strong cation exchanger (SO3). A binary salt (more specifically mixed ion) exchange gradient was used. This separation was performed using a CIMac™ SO3 column (Sartorius BIA Separations, Ajdovscina, Slovenia).

The gradient was composed of low chaotropicity (sodium acetate) and terminal chaotropicity (potassium bromide) where protein enriched AAV8 species are eluted. Figure 10 depicts results showing UV 260 nm and UV 280 nm absorbance as well as tryptophan fluorescence detection of two distinct peaks corresponding protein enriched species and (UV ratio below 1) and DNA enriched AAV8 species (UV ratio above 1).

The separation of LNP species was performed using strong and weak anion exchanger CIMac™ QA HR column and CIMac™ DEAE respectively (Sartorius BIA Separations, Ajdovscina, Slovenia). Sample was loaded in low conductivity buffer without salt. The gradient was composed of increasing salt concentration of low chaotropicity (sodium chloride) followed by a chaotropic gradient to terminal chaotropicity (lithium perchlorate). Figures 11A and 11B depict results showing separation of several LNP subpopulations at UV 260 nm and UV 280 nm absorbance as well as light scattering at 90°angle.

Examples

Example 1A

Preparation of rAAV2/8 capsids for method development and orthogonal analysis The rAAV2/8 was generated through triple transfection of suspension HEK293 cell line. Rep2-Cap8 and Helper plasmids were used together with cis construct containing GFP expression cassette flanked by inverted terminal repeats (ITRs) regions from AAV2. The plasmids were combined in a molar ratio 1 : 1 : 1 and transfected to cells using PEI MAX transfection reagent (Polysciences, Inc., USA). The transfection was performed in 5L stirred-tank Biostat B-DCU bioreactor (Sartorius, Germany) in fed-batch mode. Cell lysis was performed 72h posttransfection by adding Tween20 (Sigma-Aldrich, USA) detergent directly into bioreactor. The material was harvested and frozen at -80°C until further use. The lysed harvest of AAV 2/8 serotype was clarified and then processed by a TFF precapture step coupled with a DNase treatment. The sample was captured and additionally purified using a cation exchange chromatography column- CIMmultus™ SO3 (Sartorius BIA Separations, Ajdovscina, Slovenia). The capture step eluate was concentrated and enriched for full AAV capsids using an anion exchange CIMmultus™ QA column (Sartorius BIA Separations, Ajdovscina, Slovenia). Finally, the separately collected empty and full AAV capsids containing fractions were buffer exchanged into a 20 mM Tris, 150 mM NaCI, 2 mM MgCI2, 0.01 % poloxamer 188, pH 7.5 buffer using the Vivaspin Turbo 100 kDa PES concentrator. A sample prepared with a pool of empty and full fractions in ratio 2: 1 was used for chromatographic method development and orthogonal analysis.

Example IB

Preparation of rAAV2/8 capsids for preparative run

The rAAV2/8 was generated through triple transfection of a suspension of a HEK293 cell line in chemically defined media. Rep2-Cap8 and Helper plasmids were used together with a cis construct containing GFP expression cassette flanked by inverted terminal repeats (ITRs) regions from AAV2. Plasmids were combined in molar ratio 1 : 1 : 1 and transfected to cells using PEI MAX transfection reagent (Polysciences, Inc., USA). The transfection was performed in a 5L stirred-tank Biostat B-DCU bioreactor (Sartorius, Germany) in a fed-batch mode. The cell lysis was performed by a 72h post-transfection by adding Tween20 (Sigma-Aldrich, USA) detergent directly into the bioreactor. The material was harvested and frozen at -80°C until further use. The lysed harvest of AAV 2/8 serotype was clarified and then processed by a TFF pre-capture step coupled with DNase treatment. The sample was captured and additionally purified using a cation exchange chromatography column CIMmultus™ SO3 (Sartorius BIA Separations, Ajdovscina, Slovenia). Finally, the virus main elution fraction consisting of both empty and full AAV particles, was buffer exchanged into a formulation buffer using the Sartocon Slice 200 100 kDa PES TFF (Sartorius, Germany) membrane. The sample prepared was used for all preparative runs.

Example 1C

Preparation of rAAV2/8 capsids for separation of AAV8 capsids by strong anion exchanger QA monolithic column, QA particle column and cation exchanger SO3 column. rAAV2/8 samples were generated through triple transfection of suspension HEK293 cell line in chemically defined media. Rep2-Cap8 and Helper plasmids were used together with cis construct containing GFP expression cassette flanked by inverted terminal repeats (ITRs) regions from AAV2. Plasmids were combined in molar ratio 1 : 1 : 1 and transfected to cells using PEI MAX transfection reagent (Polysciences). Transfection was performed in 5L stirred-tank Biostat B-DCU bioreactor (Sartorius) in fed-batch mode. Cell lysis was performed 72 hours posttransfection by adding Tween20 (Sigma-Aldrich) detergent directly into bioreactor. Material was harvested and frozen at -80 °C until further use. Lysed harvest of AAV 2/8 serotype was clarified (this is a crude AAV8 sample used in the analysis of crude AAV8 samples by the improved AEX method). Clarified lysed harvest was then processed by a TFF pre-capture step coupled with DNase treatment. The sample was captured and additionally purified using a CEX chromatography column-CIMmultus™ SO3 1 mL column (Sartorius BIA Separations). The capture step SO3 eluate (consisting of total AAV8 capsids) was buffer exchanged using Vivaspin 20; 100,000 MWCO PES and stored into 50 mM sodium formate, 800 mM NaCI, 0.1% Poloxamer 188, 1% sucrose, pH 3.5 + 10% v/v 1 M Tris pH 9.0.

Example 2

Effect of exchanging only the cations in elution salt on the resolution of empty/full AAV capsids

The following buffers were used:

Buffer A: 20 mM TRIS + 75 mM KCI + 2 mM MgC + 1% sorbitol + 2.5% EtOH; pH 8.50

Buffer B: 37.5 mM MgC + 2 mM MgC + 1% sorbitol + 2.5% EtOH; pH 8.50 Analytical separations of empty and full AAV capsid samples were performed on a 100 pL strong anion exchanger, CIMac™ AAV full/empty column (Sartorius BIA Separations, Ajdovscina, Slovenia). The column was equilibrated with loading conditions (loading and equilibrium buffer, buffer A) using mild chaotropic salt 75 mM potassium chloride in buffer also containing 2 mM magnesium chloride, 2.5% ethanol, 20 mM TRIS and 1% sorbitol at pH 8.5. Sample was eluted with 60 CVs linear salt gradient to strong chaotropic salt (elution buffer, buffer B): 37.5 mM magnesium perchlorate in buffer also containing 2 mM magnesium chloride, 2.5% ethanol, 20 mM TRIS and 1% sorbitol at pH 8.5. The volumetric flow rate was 1 mL/min.

The corresponding buffer combination provides the high resolution of 1.39 of empty and full AAV capsids as shown in Figure 1. No indication of other AAV capsids subpopulations was observed.

Example 3

Enhanced effect of exchanging the cations in elution salt on the resolution of empty/full AAV capsids by means of a ternary cation salt exchange

The following buffers were used:

Buffer A: 20 mM TRIS + 75 mM KCI + 2 mM MgC + 1% sorbitol + 2.5% EtOH; pH 8.50

Buffer B: 20 mM TRIS + 75 mM LiCI + 2 mM MgC + 1% sorbitol + 2.5% EtOH; pH 8.50

Buffer C: 39.5 mM MgC + 1% sorbitol + 2.5% EtOH; pH 8.50

Analytical separations of empty and full AAV capsid samples were performed on a 100 pL strong anion exchanger, CIMac™ AAV full/empty column. The column was equilibrated with loading conditions using mild chaotropic salt 75 mM potassium chloride, 2 mM magnesium chloride, 2.5% ethanol, 20 mM TRIS and 1% sorbitol at pH 8.5. Empty and partially filled AAV capsids were eluted with 30 CVs linear salt gradient to medium chaotropic salt 75 mM lithium cloride, 2 mM magnesium chloride, 2.5% ethanol, 20 mM TRIS and 1% sorbitol at pH 8.5. Full AAV capsids then were eluted with 30 CVs linear salt gradient to strong chaotropic salt 39.5 mM magnesium chloride, 2.5% ethanol, 20 mM TRIS and 1% sorbitol at pH 8.5. The volumetric flow rate was 1 miymin. Corresponding buffer combination provides a high resolution of 1.88 of empty and full AAV capsids as shown in Figure 2A and 2B. By this approach (chaotropic salt gradient of three cations) other AAV capsids subpopulations were also observed, especially partially filled AAV capsids.

Example 4

Enhanced effect of exchanging the anions in elution salt on the resolution of empty/full AAV capsids by means of a ternary anion salt exchange

The following buffers were used:

Buffer A: 20 mM TRIS + 20 mM Mg(CH₃COO)₂ + 1% sorbitol + 2.5% EtOH; pH 8.50

Buffer B: 20 mM TRIS + 20 mM MgC + 1% sorbitol + 2.5% EtOH; pH 8.50

Buffer C: 20 mM TRIS + 20 mM Mg(CIO₄)₂ + 1% sorbitol + 2.5% EtOH; pH 8.50

Analytical separations of empty and full AAV capsid samples were performed on a 100 pL strong anion exchanger, CIMac™ AAV full/empty column. The column was equilibrated with loading conditions (loading and equilibration buffer A) using mild chaotropic salt 20 mM magnesium acetate, 2.5% ethanol, 20 mM TRIS and 1% sorbitol at pH 8.5. Empty and partially filled AAV capsids were eluted with 30 CVs linear salt gradient to medium chaotropic salt (elution buffer 1, buffer B) 20 mM magnesium chloride, 2.5% ethanol, 20 mM TRIS and 1% sorbitol at pH 8.5. Full AAV capsids were eluted with 30 CVs linear salt gradient to strong chaotropic salt (elution buffer 2, buffer C) 20 mM magnesium perchlorate, 2.5% ethanol, 20 mM TRIS and 1% sorbitol at pH 8.5. The volumetric flow rate was 1 miymin.

Corresponding buffer combination provides peak resolution of 6.10 of empty and full AAV capsids as shown in Figure 3A and 3B. By this approach, using ternary anion salt gradient other AAV capsids subpopulations were observed, especially partially filled AAV capsids.

Figure 3C illustrates mass photometry results for collected fractions. Peak from 3.3 - 5.0 min represents empty capsids, followed by the peak from 5.2 - 8.0 min which is a mixture of empty, partially filled and full AAV8 capsids. Similar profile is detected in peak eluting from 9.5 - 10.5 min (high salt wash). Peak from 8.3 - 9.0 min is predominantly full AAV8.

Example 5

Preparative separation with QA using ternary anion salt exchange gradient The following buffers were used:

Buffer A: 20 mM TRIS + 20 mM Mg(CH₃COO)₂ + 1% sorbitol + 2.5% EtOH; pH 8.50

Buffer B: 20 mM TRIS + 20 mM MgCI₂ + 1% sorbitol + 2.5% EtOH; pH 8.50

Buffer C: 20 mM TRIS + 20 mM Mg(CIO₄)₂ + 1% sorbitol + 2.5% EtOH; pH 8.50

Buffer D: 20 mM TRIS + 2 M KAc + 2.5% EtOH; pH 8.50

A preparative separation of empty from full AAV capsids was performed by using a CIMmultus™ QA-1 mL (2pm) column, loaded at 8.16E+12vg/per mL column. The sample was first diluted with buffer A, followed by an additional dilution with Mili-Q® water to meet the binding conditions with conductivity of 2 mS/cm and pH 8.5. CIMmultus™ The QA-1 mL column was conditioned in the manner of buffers: A/B/C/A/D/A: 5/5/10/5/10 CVs each. The sample was detected using a sample pump on an a AKTA Pure 25 chromatographic system. Loading of the sample was followed by a wash with buffer A (10 CVs) and then several elution strategies were implemented. First a linear gradient elution to 100% of buffer B over 30 CVs was used and hold for another 10 CVs of 100% buffer B. A second linear gradient elution to 100% of buffer C over 30 CVs was used and hold for 10 CVs of 100% buffer C. Finally, a wash with buffer A (10 CVs, preferably 20 CVs) was performed with 10 CVs of linear gradient to 100% D. Before buffer D is introduced to perform a high salt wash, it is the recommended to perform a 10 CVs wash with buffer A to avoid precipitation of Mg(CIO₄)₂ in the high salt. The fractions El to E4 from preparative run were collected as shown in Figure 4A. 4B shows a zoom-in chromatogram of possible subpopulations of full AAV capsids, fraction E2 and E3.

The collected fractions were individually analysed by an orthogonal analytical method using 100 pL strong anion exchanger, CIMac™ AAV full/empty column. The column was equilibrated with 2 mM magnesium acetate, 2.5% acetonitrile, 20 mM TRIS and 1% sorbitol at pH 8.5 and eluted with a linear salt gradient to 50 mM magnesium acetate, 2.5% acetonitrile, 20 mM TRIS and 1% sorbitol at pH 8.5. The volumetric flow rate was 1 mb min. The elution was achieved in a linear salt gradient from 2 mM to 80 mM magnesium acetate with 160 column volume (CV). A multidetector setup was used, UV 260 and 280 nm, light scattering (not shown in figure), intrinsic protein fluorescence that mostly induced by tryptophan. Tryptophan fluorescence was monitored at extinction 280 nm and emission at 348 nm. The Figure 4C shows UV 260/280 nm results where, the El fraction represents empty capsids with a 260/280 wavelength ratio of 0.64. E2 and E3 fractions are populated with full capsids with 260/280 wavelength ratio of 1.33 and 1.34 respectively. Fraction E4 (high salt wash) highly likely represents damaged AAV, heavy AAV capsids and/or aggregates. Partially filled AAV capsids were not observed (possibly present in low concentrations - under limit of detection).

Example 6

PATfix AAV Switcher (Sartorius BIA Separations, Ajdovscina, Slovenia) of purified AAV8 standard and AAV8 lysate

The following buffers for AEX separation were used:

Buffer A: 20 mM TRIS + 20 mM Mg(CH₃COO)₂ + 1% sorbitol + 2.5% EtOH; pH 8.5

Buffer B: 20 mM TRIS + 20 mM MgC + 1% sorbitol + 2.5% EtOH; pH 8.5

Buffer C: 20 mM TRIS + 40 mM Mg(CIO₄)₂ + 1% sorbitol + 2.5% EtOH; pH 8.5

Buffer D: 20 mM TRIS + 2 M KAc + 2.5% EtOH; pH 8.5

A separation of empty, partially filled, and full capsids in complex lysate samples was achieved on a CIMac™ AAV full/empty column by using a ternary anion salt exchange gradient composed of magnesium acetate, chloride and perchlorate. The lysate sample was first analysed on a cation exchange column using a pH gradient, from pH 4.00 to pH 8.50. The proportion of the pH gradient elution was then redirected on the second AEX column where a multiple subpopulations separation was achieved. In addition to the empty and full capsids also partially filled and other impurities were observed (Figure 5).

Example 7

Separation of ovalbumin (45 kDa) isoforms using a strong anion exchanger (QA) using a binary anion exchange gradient

The following buffers were used:

Buffer A: 20 mM BTP + 100 mM NaCH₃COO + 2.5% ACN; pH 9.00

Buffer B: 20 mM BTP + 100 mM NaCIO₄ + 2.5% ACN; pH 9.00 A separation of ovalbumin isoforms was performed using a 100 pL CIMac™ AAV full/empty (1.3pm) column. A lyophilized ovalbumin powder (Albumin from chicken egg white, Sigma-Aldrich, cat. No. : 9006-59-1, PN : A 5503) was diluted in double distilled water to a concentration of 0.4 mg/mL. The sample was further diluted with a loading buffer (explained below) to a final loading amount 0.5 pg of ovalbumin.

The column was equilibrated with loading buffer 100 mM sodium acetate, (buffer A) 20 mM BTP and 2.5% acetonitrile at pH 9.0 eluted with a linear 100 CVs gradient to elution buffer B that consists of 100 mM sodium perchlorate, 20 mM BTP and 2.5% acetonitrile at pH 9.0. The volumetric flow rate was 1 mL/min. Figure 6A depicts results showing a fluorescence detection of four distinct peaks corresponding to ovalbumin isomers.

To confirm presence of possible ovalbumin isoforms a silver stained SDS-PAGE gel of the ovalbumin sample (different sample dilutions loaded) was run and depicted in Figure 6B. At least two bands were observed at approximate 45 kDa. Additional bands were also observed from 85 to 150 kDa representing possible ovalbumin dimers or trimers. Possible impurities or individual ovalbumin subunits were observed at approximate 30 kDa, 25 kDa, 17 kDa and 10 kDa. For size assessment a PageRuler prestained protein ladder (Thermo Fisher Scientific Inc., USA) was used.

Example 8

Separation of plasmid DNA isoforms using a weak anion exchanger (pDNA) by means of a ternary salt (more specifically cation) exchange gradient

The following buffers were used:

Buffer A: 100 mM TRIS + 300 mM GuHCI + 400 mM CsCI; pH 8.00

Buffer B: 100 mM TRIS + 300 mM GuHCI + 350 mM LiCI; pH 8.00

Buffer C: 100 mM TRIS + 300 mM GuHCI + 175 mM CaCI₂; pH 8.00

The separation of main plasmid DNA isoforms was performed using a 300 pL CIMac™ pDNA (6pm) column. The sample was a mixture of 0.5 pg of pFIX 15 (15 kbp plasmid DNA containing open circular and super coiled plasmid DNA) and 0.5 pg of linear pFIX 15 (15 kbp linear plasmid DNA).

The column was equilibrated with loading and equilibration buffer (Buffer A) 400 mM caesium chloride, 300 mM guanidinium hydrochloride, 100 mM TRIS at pH 8.0. To speed up the method 10 CVs linear gradient were introduced from 100% Buffer A to 85% Buffer B (which is elution buffer 1). The elution of open circular and linear pDNA isoforms was enabled with a linear 20 CVs gradient from 85% Buffer B to 100% Buffer B 350 mM lithium chloride, 300 mM guanidinium hydrochloride, 100 mM TRIS at pH 8.0. Super coiled pDNA isoform was eluted with 40 CVs linear gradient from Buffer B to Buffer C (elution buffer 2), which is 175 mM calcium chloride, 300 mM guanidinium hydrochloride, 100 mM TRIS at pH 8.0. The volumetric flow rate was 1 miymin. Figure 7A depicts results showing UV 260 nm absorbance detection of three peaks corresponding to each pDNA isoforms.

Orthogonal analytics - AGE gel was also performed to confirm pDNA isoforms. Figure 7B depicts AGE gel of all three individual pDNA isoforms as well as a mixture of pFIX15 and lin pFIX15. For size assessment, Supercoiled DNA ladder (Figure 7B left edge) and GeneRuler™ DNA ladder (Figure 7B right edge) was used. The first two bands represent oc and sc respectively, the last band represents the linear pDNA isoform.

Example 9

The separation of exosome of different subpopulations was performed using a 100 pL CIMac™ AAV e/f (1.3pm) column (Figure 8). The sample was a pool mixture from QA preparative run of lysed harvest of HEK CD63-eGFP exosomes sample.

Sample was prepared on a preparative scale using the following procedure:

The original sample, lysed harvest of HEK CD63-eGFP exosomes was clarified and buffer exchanged (processed by TFF pre-capture step without a DNase treatment). After that the sample was captured and additionally purified using a strong anion exchange chromatography column-CIMmultus™ QA-1 (Sartorius BIA Separations, Ajdovscina, Slovenia). QA preparative fractions El, E2 and E3 were collected (storage buffer: 50 mM HEPES, ~2 M NaCI at pH 7.0). The sample for analytical method development depicted in Example 9 was prepared with a pool of El, E2 and E3 fractions in ratio 1 :2: 1 and diluted with Buffer A.

Separation of exosome subpopulations using a strong anion exchanger (QA) by means of a binary salt exchange gradient was conducted using the following buffers:

Buffer A: 20 mM TRIS + 1% sorbitol + 1 % EtOH + 200 mM KAc; pH 8.50

Buffer B: 20 mM TRIS + 1% sorbitol + 1 % EtOH + 200 mM NaCIO₄; pH 8.50 Buffer C: 20 mM TRIS + 1% sorbitol + 2 M KAc + 1 % EtOH; pH 8.50

The QA column was equilibrated with loading and equilibration buffer (Buffer A) 200 mM potassium acetate, 20 mM TRIS, 1% sorbitol and 1% ethanol at pH 8.50. The elution of exosome subpopulations was enabled with a linear 50 CVs gradient from 100% Buffer A to 100% Buffer B 200 mM sodium perchlorate, 20 mM TRIS, 1% sorbitol and 1% ethanol at pH 8.50. After that high salt wash was applied 2000 mM potassium acetate, 20 mM TRIS, 1% sorbitol and 1% ethanol at pH 8.50. The volumetric flow rate was 1 miymin.

Figure 8 depicts results showing UV 260 nm and 280 nm, light scattering at 90° angle and tryptophan fluorescence. Three distinct peaks corresponding exosome subpopulations with UV 260/280 wavelength ratio of 0.7, 0.67 and 0.61 were observed.

Example 10

Separation of empty and full AAV8 capsids performed on QA particle column.

The following buffers were used:

Buffer A: 15 mM TRIS + 20 mM Mg(CH₃COO)₂ + 1% sorbitol + 2.5% EtOH; pH 8.50

Buffer B: 15 mM TRIS + 20 mM MgC + 1% sorbitol + 2.5% EtOH; pH 8.50

Buffer C: 15 mM TRIS + 40 mM Mg(CH₃COO)₂ + 1% sorbitol + 2.5% EtOH; pH 8.50

Buffer D: 15 mM TRIS + 2 M NaCI + 2.5% EtOH; pH 8.50

Analytical separation of empty and full AAV capsid samples was performed on a 1000 pL (8 pm pores) strong anion exchanger DNAPac™ PA200 particle column.

The QA particle column was equilibrated with loading conditions (loading and equilibration buffer A) using mild chaotropic salt 20 mM magnesium acetate, 2.5% ethanol, 15 mM TRIS and 1% sorbitol at pH 8.5. Mostly empty AAV capsids were eluted with 2 CVs linear salt gradient from 100% buffer A to 100% of medium chaotropic salt (elution buffer 1, buffer B) 20 mM magnesium chloride, 2.5% ethanol, 15 mM TRIS and 1% sorbitol at pH 8.5. Mostly full AAV capsids were eluted with 2 CVs linear salt gradient from 20% buffer A and 80% buffer B to 100% of higher magnesium acetate salt concentration (elution buffer 2, buffer C) 40 mM magnesium acetate, 2.5% ethanol, 15 mM TRIS and 1% sorbitol at pH 8.5. After that high salt wash was applied - 1 CV linear gradient from 100% buffer C to 100% buffer D followed by 1 CV hold in 100% buffer D 2000 mM sodium chloride, 2.5% ethanol, and 15 mM TRIS at pH 8.5. The volumetric flow rate was 2 mL/min.

Corresponding buffer combination provides peak resolution of 2.98 (UV260 nm) of empty and full AAV capsids as shown in Figure 9.

Example 11

The indication of separation of protein enriched AAV8 capsids (predominantly empty AAV8 capsids) and DNA-enriched AAV8 capsids (predominantly full AAV8 capsids) was performed using a 100 pL CIMac™ SO3 (1.3pm) column as depicted in Figure 10.

Separation of LNP species using a strong anion exchanger (QA) by means of a binary salt exchange gradient was conducted using the following buffers:

Buffer A: 10 mM acetic acid + 5 mM Mg(CH3COO)2 + 150 mM Na(CH3COO)2 + 1% sorbitol + 2.5 % EtOH; pH 5.00

Buffer B: 10 mM acetic acid + 5 mM Mg(CH3COO)2 + 150 mM KBr + 1% sorbitol + 2.5 % EtOH; pH 5.00

Buffer C: 10 mM acetic acid + 2000 mM NaCI + 2.5 % EtOH; pH 5.00

Buffer D: 1 M NaOH + 2 M NaCI

The SO3 column was equilibrated with loading and equilibration buffer (Buffer A) 10 mM acetic acid, 1% sorbitol, 2.5% ethanol, 5 mM magnesium acetate and 150 mM sodium acetate at pH 5.00. The elution of protein enriched AAV8 capsids was enabled with an increasing linear 20 CVs chaotropic gradient from 100% Buffer A to 100% Buffer B 10 mM acetic acid, 1% sorbitol, 2.5% ethanol, 5 mM magnesium acetate and 150 mM potassium bromide at pH 5.00. After that high salt wash in 20 CVs gradient from Buffer B to Buffer C was applied. Buffer C was 10 mM acetic acid, 2.5% ethanol and 2000 mM sodium chloride at pH 5.00. To elute the most retained species (DNA-enriched AAV8 capsids) cleaning in place (CIP) was introduced with buffer D I M sodium hydroxide and 2 M sodium chloride. The volumetric flow rate was 2 mb/min.

Figure 10 depicts results showing UV 260 nm, 280 nm, and tryptophan fluorescence. First 2 implied peaks (between 1 and 2 minutes) indicate protein enriched AAV8 capsids with UV ratio below 1. The rest of predominantly DNA- enriched AAV8 capsids elute in CIP (at ~6 minutes) with UV ratio above 1.

Example 12

The separation of LNPs with encapsulated Cy5-eGFP mRNA and with stained phospholipids TMR.-PC was performed using a 100 pL CIMac™ QA (6pm) column (Figure 11A) and CIMac™ Low DEAE (6pm) column (Figure 11B).

Buffer A: 20 mM BTP + 1% sorbitol + 2.5 % EtOH; pH 10.00

Buffer B: 20 mM BTP + 1% sorbitol + 2.5 % EtOH + 600 mM NaCI; pH 10.00

Buffer C: 20 mM BTP + 1% sorbitol + 2.5 % EtOH + 600 mM LiCIO₄; pH 10.00

Buffer D: 20 mM BTP + 1% sorbitol + 2.5 % EtOH; 2000 mM NaCI; pH 10.000

The QA column was equilibrated with loading and equilibration buffer (Buffer A) 20 mM BTP, 1% sorbitol and 2.5% ethanol at pH 10.00. The elution of initial LNP subpopulations was enabled with an increasing linear 25 CVs gradient from 100% Buffer A to 100% Buffer B 600 mM sodium chloride, 20 mM BTP, 1% sorbitol and 2.5% ethanol at pH 10.00. After that the first chaotropic gradient was applied to elute additional LNP species (25 CVs gradient from 100% Buffer B to 100% Buffer C 600 mM lithium perchlorate, 20 mM BTP, 1% sorbitol and 2.5% ethanol at pH 10.00) followed by high salt wash 25 CVs gradient from Buffer C to Buffer D 2000 mM sodium chloride, 20 mM BTP, 1% sorbitol and 2.5% ethanol at pH 10.00. The volumetric flow rate was 1 mb/min.

Figure 11A depicts results showing UV 260 nm, 280 nm, and light scattering at 90° angle. Flow-through, 2 implied peaks (between 2 and 4 minutes) and 2 additional distinct peaks (between 4.5 and 6 minutes) corresponding different LNP species with UV 260/280 wavelength ratios of 1.26, 1.26, 1.66 and 1.43 were observed.

Separation of LNP species using a weak anion exchanger (DEAE) by means of a binary salt exchange gradient was conducted using the following buffers:

Buffer A: 20 mM Tris + 1% sorbitol + 2.5 % EtOH; pH 8.50

Buffer B: 20 mM Tris + 1% sorbitol + 2.5 % EtOH + 600 mM NaCI; pH 8.50 Buffer C: 20 mM Tris + 1% sorbitol + 2.5 % EtOH + 600 mM LiCIC ; pH 8.50

Buffer D: 20 mM Tris + 1% sorbitol + 2.5 % EtOH; 2000 mM NaCI; pH 8.50

The DEAE column was equilibrated with loading and equilibration buffer (Buffer A) 20 mM Tris, 1% sorbitol and 2.5% ethanol at pH 8.50. The elution of initial LNP subpopulations was enabled with a linear 25 CVs increasing salt gradient from 100% Buffer A to 100% Buffer B 600 mM sodium chloride, 20 mM Tris, 1% sorbitol and 2.5% ethanol at pH 8.50. After that the first chaotropic gradient was applied to elute additional LNP species (25 CVs gradient from 100% Buffer B to 100% Buffer C 600 mM lithium perchlorate, 20 mM Tris, 1% sorbitol and 2.5% ethanol at pH 8.50), followed by high salt wash 25 CVs gradient from Buffer C to Buffer D 2000 mM sodium chloride, 20 mM Tris, 1% sorbitol and 2.5% ethanol at pH 8.50. The volumetric flow rate was 1 mb/min.

Figure 11B depicts results showing UV 260 nm, 280 nm, and light scattering at 90° angle. Flow-through, 2 separated peaks (between 1.5 and 5 minutes) and third complex additional peak (between 7.5 and 8.5 minutes) corresponding different LNP species with UV 260/280 wavelength ratios of 1.19, 1.75, and 1.45 were observed.

List of abbreviations:

AAV2 Adeno associated virus serotype 2

ACN Acetonitrile

AGE Agarose gel electrophoresis

BTP Bis-tris propane

CV Column volume

Cy5 Cyanine 5 eGFP Enhanced green fluorescent protein

EtOH Ethanol

GFP Green fluorescent protein

GuHCI Guanidine hydrochloride kbp Kilobase pair

KAc Potassium acetate lin Linear

LNP Lipid nanoparticle mRNA Messanger ribonucleic acid oc Open circular pDNA Plasmid DNA

PES Polyethersulfone rAAV2/8 Recombinant AAV2 vector genom with the type 8 capsid

Rep2-Cap8 Plasmid expressing Rep/Cap genes sc Supercoil

TFF Tangential flow filtration

TMR-PC Tetramethylrhodamine - Phosphatidylcholine

References

[1] Hamaguchi & Geiduschek (1962). "The Effect of Electrolytes on the Stability of the Deoxyribonucleate Helix". J. Am. Chem. Soc. 84 (8): 1329- 1338. doi: 10.1021/ja00867a001.

[2] Dixit, S., Crain, J., Poon, W.C., Finney, J.L., and Soper, A.K. (2002) Molecular segregation observed in a concentrated alcohol-water solution. Nature 416: 829-832.

[3] Frank, H.S., and Evans, M.W. (1945) Free volume and entropy in condensed systems III. Entropy in binary liquid mixtures; partial molal entropy in dilute solutions; structure and thermodynamics in aqueous electrolytes. J Chem Phys 13: 507-532.

[4] Hofmeister, F (1888). "Zur Lehre von der Wirkung der Salze". Arch. Exp. Pathol. Pharmakoi. 24 (4-5): 247-260. doi : 10.1007/bf01918191. S2CID 27935821.

[5] Yang Z (2009). "Hofmeister effects: an explanation for the impact of ionic liquids on biocatalysis". Journal of Biotechnology. 144 (1): 12-22. doi : 10.1016/j.jbiotec.2009.04.011. PMID 19409939.

[6] US 2016/0040137 Al 2016.

[7] P. Gagnon, B. Goricar, S.D. Prebil, H. Jug, M. Leskovec, A. Strancar, Separation of Empty and Full Adeno-Associated Virus Capsids from a Weak Anion Exchanger by Elution with an Ascending pH Gradient at Low Ionic Strength, 2021.

[8] WO 2022/038164 Al.

Claims

1. A method of chromatographic separation of components of an aqueous mixture by means of a chromatographic material comprising the steps (i) to (iv) of:

(i) providing the aqueous mixture of the components,

(ii) contacting the aqueous mixture with the chromatographic material,

2. The method of claim 1, wherein in step (iv) of claim 1 the chaotropicity level (S) is increased to at least one intermediate chaotropicity level (I) having a lower chaotropicity than the chaotropicity level (T), and thereafter increasing the chaotropic level (I) to the higher chaotropic level (T).

3. The method of claim 2, wherein the at least one intermediate chaotropic level (I) is adjusted by increasing a concentration of the first chaotropic salt.

4. The method of claim 2, wherein the at least one intermediate chaotropic level (I) is adjusted by means of a third chaotropic salt (3) having a higher chaotropicity than the first chaotropic salt (1) and a lower chaotropicity than the second chaotropic salt (2).

5. The method of any one of the claims 1 to 4 wherein the chaotropicity levels (S), (I) and/or (T) are individually adjusted by a mixture of chaotropic salts.

6. The method of any one of the claims 1 to 5, wherein the component is a biomolecule such as LNP, nucleic acids, DNA, RIMA, a protein, an enzyme, a glycoprotein, a polysaccharide, a lipid, and/or a fatty acid.

7. The method of any one of the claims 1 to 5, wherein the change from the low chaotropicity level (S) to the higher chaotropicity level (T), the change from the low chaotropicity level (S) to at least one intermediate chaotropicity level (I) and/or the change from at least one intermediate chaotropicity level (I) to the higher chaotropicity level (T) are gradient changes.

8. The method of any one of the claims 1 to 6, wherein the component is a virus, a virus-like particle, a bacteriophage, an extracellular vesicle, such as an exosome, or combinations thereof.

9. The method of claim 8, wherein the virus is selected from a group consisting of:

(i) a parvovirus,

(ii) an adeno associated virus (AAV), in particular AAV selected from the group consisting of serotype AAV1, AAV2, AAV3, AAV4, AAV5, AAV6, AAV7, AAV8, AAV9, AAVrh.10, AAV11, AAV 12,

(iv) a chimera of AAV,

(v) a surface modified AAV, and

(vi) a synthetically derived AAV like particle.

10. The method of any one of the claims 1 to 9, wherein the chromatographic material is:

(i) an ion exchange chromatography material,

(ii) a hydrophobic interaction chromatography material,

(iii) a mixed mode chromatography material,

(iv) a monolith anion or cation exchanger

(v) a monolith multimodal material,

(vi) a particulate anion or cation exchanger and/or

(vii) a multimodal material.

11. The method of claim 10 wherein the chromatographic material

(i) is arranged in a membrane or monolith, (ii) is in form of particles arranged in a column, and/or

(iii) is a fibre column.

12. The method of any one of the claims 1 to 11, wherein the aqueous environment is an aqueous solution comprising:

(i) chaotropic cations selected from the group consisting of tetramethyl ammonium, ammonium, caesium, rubidium, potassium, sodium, lithium, magnesium, calcium, zinc and barium, protonated arginine, protonated guanidine and/or

(ii) chaotropic anions selected from the group consisting of sulfate, propionate, acetate, formate, fluoride, chloride, bromide, iodide, nitrate, chlorate, perchlorate, and thiocyanate.

13. The method of any one of the claims 1 to 12, wherein the aqueous environment comprises chaotropic salts, buffering agents, isotonic substances, auxiliary agents, and/or organic modifiers.

14. The method of claim 13, wherein the isotonic substances are selected from the group consisting of sucrose, sorbitol, mannitol, xylitol, and mixtures thereof.

15. The method of claims 13 to 14, wherein the auxiliary agents are non-ionic surfactants, such as poloxamer 188 or urea.

16. The method of any one of the claims 13 to 15, wherein the organic modifiers are selected from the group consisting of acetonitrile, 1-butanol, t-butanol, propylene carbonate, isopropanol, ethanol, methanol, ethylene glycol, 1-propanol, and mixtures thereof.

17. A method of chromatographic separation of components in an aqueous mixture by means of a chromatographic material comprising the steps (i) to

(iv) of:

(i) providing the aqueous mixture of the components,

(ii) contacting the aqueous mixture with the chromatographic material,

(iii) followed by starting the separation by contacting the chromatographic material with an aqueous environment having a first level of chaotropicity(S) which is adjusted by means of a first chaotropic salt (1), (iv) followed by linearly increasing the first level of chaotropicity (S) to a terminal aqueous environment having a second level of chaotropicity(T) which is adjusted by means of a second chaotropic salt (2), the second chaotropic salt (2) having a higher chaotropicity than the first chaotropic salt (1), thereby eluting the components and, optionally,

(v) collecting the components that are separated due to the increasing level of chaotropicity of the aqueous environment, wherein the second level of chaotropicity is higher than the first level of chaotropicity, wherein the components are biomolecules.

18. The method of claim 17, wherein in step (iv) of claim 17 the first level of chaotropicity(S) is increased to at least one intermediate level of chaotropicity (I) having a lower chaotropicity than the second level of chaotropicity (T), but higher than the first level of chaotropicity (S) and thereafter increasing the chaotropic level (I) to the higher chaotropic level (T).