[0001] The present application claims priority to United States Provisional patent application numbers 62/644,237 and 62/644,241, each filed on March 16, 2018, and the entire disclosures of each of which is incorporated herein by reference.

BACKGROUND

[0002] A number of gene therapy products are currently being developed to treat human diseases. Many of these gene therapy products use recombinant adeno-associated viruses (AAVs) that have been engineered to deliver a heterologous nucleic acid of interest (e.g., a gene encoding a therapeutic protein, an antisense nucleic acid molecule, a ribozyme, a miRNA, an siRNA, a nucleic acid encoding a CRISPR/Cas system, or the like) to the diseased target cells of a patient.

[0003] As is well known in the art, these recombinant AAVs are engineered by deleting, in whole or in part, the internal portion of the AAV genome and inserting the heterologous nucleic acid of interest or“payload” between the inverted terminal repeats (ITRs). The ITRs remain functional in such vectors allowing replication and packaging of the AAV particle containing the nucleic acid payload enclosed within the AAV capsid. Typically, the nucleic acid payload is also operably linked to regulatory sequences (e.g., promoter, enhancer, etc.) capable of driving expression of the payload in the patient’s target cells.

[0004] Various methods have been developed to manufacture large quantities of these recombinant AAVs. These methods typically involve an“upstream” phase where the recombinant AAVs are produced within a host cell (e.g., a mammalian or insect cell) and a “downstream” phase where the recombinant AAVs are collected and purified. Inefficiencies in the packaging of the nucleic acid payload during the upstream phase generally lead to recombinant AAV preparations that include a mixture of“full” AAV particles (i.e., AAV particles that include the nucleic acid payload) and“empty” AAV particles (i.e., AAV particles that lack in whole or in part the nucleic acid payload).

[0005] The presence of empty AAV particles in a gene therapy product may increase the overall dose needed to achieve therapeutic efficacy and may also cause an immune response upon administration to a patient (e.g., neutralization of the AAV particles by host development of neutralizing antibodies, or clearance of AAV-transduced cells due to T-cell activation).

[0006] There is therefore a need in the art for downstream purification methods that can be used to separate full AAV particles from empty AAV particles in the recombinant AAV preparations that are generated by current upstream processes.

SUMMARY

[0007] The present disclosure is based in part on the insight that the separation of full

AAV particles and empty AAV particles in recombinant AAV preparations using anion- exchange chromatography can be improved when magnesium chloride (MgCl ₂ ) is included in one or more of the solutions used for the separation.

[0008] In some embodiments, the methods involve a gradient elution with solutions that include particular concentrations of MgCl ₂ and optionally sodium chloride (NaCl). In some embodiments, the methods involve isocratic elutions with solutions that include particular concentrations of MgCl ₂ and optionally NaCl. In some embodiments, a combination of MgCl ₂ and NaCl is used in one or more of the solutions.

[0009] In some embodiments, the present disclosure provides a gradient elution method that comprises (a) contacting an anion-exchange medium with a recombinant adeno-associated virus (AAV) preparation comprising full AAV particles and empty AAV particles; (b) applying to the anion-exchange medium a gradient solution comprising magnesium chloride (MgCl ₂ ) and optionally sodium chloride (NaCl); and (c) collecting a series of elution fractions from the anion- exchange medium, wherein the series includes a first elution fraction and a second elution fraction collected after the first elution fraction and wherein, (i) the concentration of MgCl ₂ in the gradient solution is increased during the step of applying, (ii) the concentration of NaCl in the gradient solution is increased during the step of applying, or (iii) the concentration of MgCl ₂ and the concentration of NaCl in the gradient solution are both increased during the step of applying. In some embodiments, the concentration of MgCl ₂ is increased during the step of applying. In some embodiments, the concentration of NaCl is increased during the step of applying. In some embodiments, the concentration of MgCl ₂ is held constant during the step of applying. In some embodiments, the concentration of NaCl is increased during the step of applying and the concentration of MgCl ₂ is held constant during the step of applying. In some embodiments, the concentration of MgCl ₂ and the concentration of NaCl are both increased during the step of applying.

[0010] In some embodiments, the present disclosure provides an isocratic elution method that comprises (a) contacting an anion-exchange medium with a recombinant adeno-associated virus (AAV) preparation comprising full AAV particles and empty AAV particles; (b) applying to the anion-exchange medium a wash solution comprising magnesium chloride (MgCl ₂ ) and optionally sodium chloride (NaCl), and collecting a wash elution fraction; and (c) then applying to the anion-exchange medium an elution solution comprising MgCl ₂ and optionally NaCl, and collecting a elution fraction, wherein, (i) NaCl is not present in the wash and elution solutions and the concentration of MgCl ₂ is greater in the elution solution than in the wash solution; (ii) NaCl is present in the elution solution and the concentration of NaCl is greater in the elution solution than in the wash solution; or (iii) NaCl is present in the wash solution, the concentration of NaCl is less in the elution solution than in the wash solution, and the concentration of MgCl ₂ is greater in the elution solution than in the wash solution. In embodiments where NaCl is present in the elution solution and optionally in the wash solution, the concentration of MgCl ₂ can be (i) greater in the elution solution than in the wash solution, (ii) the same in the wash and elution solutions, or (iii) less in the elution solution than in the wash solution. In embodiments where NaCl is not present in the wash and elution solutions, the concentration of MgCl2 may be greater in the elution solution than the wash solution. In embodiments where NaCl is present in the elution solution, the concentration of NaCl may be greater in the elution solution than the wash solution. In embodiments where NaCl is present in the wash solution, the concentration of NaCl may be less in the elution solution than in the wash solution and the concentration of MgCl ₂ may be (i) greater in the elution solution than the wash solution; (ii) the same in the wash solution and elution solution; or (iii) less in the elution solution than the wash solution.

[0011] AAV capsid proteins have a greater UV absorbance at 280 nm than at 260 nm and the nucleic acid payload has greater UV absorbance at 260 nm than at 280 nm. Accordingly, as empty AAV particles lack in whole or in part the nucleic acid payload, they have a lower UV absorbance at 260 nm (A260) than full AAV particles (which include the nucleic acid payload). As a result, the A260/A280 ratio for an elution fraction that is enriched in empty AAV particles will be less than the A260/A280 ratio for an elution fraction that is enriched in full AAV particles. In some embodiments, the methods of the present disclosure preferentially release empty AAV particles from the anion-exchange medium before releasing full AAV particles. In these embodiments, the A260/A280 ratio of the first elution fraction will be less than the second elution fraction (e.g., as shown in Figure 1). In some embodiments, the number and charge of the AAV capsid proteins and nucleic acid payload of a particular recombinant AAV may cause the full AAV particles to be preferentially released from the anion-exchange medium before the empty AAV particles are released. In these embodiments, the A260/A280 ratio of the first elution fraction will be greater than the second elution fraction.

[0012] In some embodiments, the methods of the present disclosure improve the resolution between empty AAV particles and full AAV particles as compared to anion-exchange separation methods where the solutions comprise a single salt (e.g., sodium chloride) and separation therefore relies on differences in the single salt (e.g., sodium chloride) concentration.

[0013] The present disclosure is also based in part on the unexpected discovery that full and empty AAV particles can be separated by anion-exchange chromatography using a single elution that includes a fixed concentration of MgCl ₂ and optionally NaCl (i.e., instead of traditional isocratic separations which use an initial“wash” solution to remove one species from the separation medium, e.g., empty AAV particles, followed by a distinct“elution” solution with higher salt concentration than the“wash” solution to remove the other species, e.g., full AAV particles).

[0014] Without being bound to any theory, in some embodiments, we propose that the

MgCl ₂ in the single elution solution interacts with AAV capsid proteins bound to the separation medium and does not exit during initial applications of the elution solution to the separation medium. As interaction sites get gradually saturated, the remaining MgCl ₂ being added to the separation medium starts to increase the concentration of“free” MgCl ₂ and hence the conductivity of the elution solution that exits the separation medium increases along a gradient (even though the elution solution being added to the column is not changing). [0015] In some of the experiments described herein, the empty AAV particles were less tightly bound to the anion-exchange medium and eluted earlier. The more strongly bound full AAV particles eluted later as the concentration of“free” MgCl ₂ increased. Thus, in some embodiments, the methods of the present disclosure involve separating full particles and empty AAV particles by anion-exchange chromatography using a single elution solution that includes a fixed concentration of MgCl ₂ and optionally NaCl. In some embodiments, a combination of MgCl ₂ and NaCl is used in the single elution solution.

[0016] In some embodiments, the present disclosure provides an isocratic elution method that comprises (a) contacting an anion-exchange medium with a recombinant adeno-associated virus (AAV) preparation comprising full AAV particles and empty AAV particles; (b) applying to the anion-exchange medium an elution solution comprising a fixed concentration of magnesium chloride (MgCl ₂ ) and optionally sodium chloride (NaCl); and (c) collecting a series of elution fractions from the anion-exchange medium during the step of applying the elution solution, wherein the series includes a first elution fraction and a second elution fraction collected after the first elution fraction, and wherein (i) an A260/A280 ratio of the first elution fraction is less than an A260/A280 ratio of the second elution fraction, for example wherein an A260/A280 ratio of the first elution fraction is less than one and an A260/A280 ratio of the second elution fraction is greater than one, or (ii) an A260/A280 ratio of the first elution fraction is greater than an A260/A280 ratio of the second elution fraction, for example wherein an A260/A280 ratio of the first eluate is greater than one and an A260/A280 ratio of the second elution fraction is less than one.

[0017] AAV capsid proteins have a greater UV absorbance at 280 nm than at 260 nm and the nucleic acid payload has greater UV absorbance at 260 nm than at 280 nm. Accordingly, as empty AAV particles lack in whole or part the nucleic acid payload, they have a lower UV absorbance at 260 nm (A260) than full AAV particles (which include the nucleic acid payload). As a result, the A260/A280 ratio for an elution fraction that is enriched in empty AAV particles will be less than the A260/A280 ratio for an elution fraction that is enriched in full AAV particles. In some embodiments, the methods of the present disclosure preferentially release empty AAV particles from the anion-exchange medium before releasing full AAV particles. In these embodiments, the A260/A280 ratio of a first elution fraction will be less than that of a second elution fraction (e.g., as shown in Figure 5). In some embodiments, the number and charge of the AAV capsid proteins and nucleic acid payload of a particular recombinant AAV may cause the full AAV particles to be preferentially released from the anion-exchange medium before the empty AAV particles are released. In these embodiments, the A260/A280 ratio of the first elution fraction will be greater than that of the second elution fraction.

BRIEF DESCRIPTION OF THE DRAWING

[0018] Figure 1 is an exemplary chromatogram depicting A260 and A280 measurements of eluate during an isocratic elution. As shown, a first fraction (which is released from the anion-exchange medium with a first“wash” solution) is enriched in“empty” AAV particles (A260/A280 ratio of first elution peak is less than one) while a second fraction (which is released from the anion-exchange medium with a second“elution” solution) is enriched in“full” AAV particles (A260/A280 ratio of second elution peak is greater than one).

[0019] Figure 2 is an exemplary chromatogram depicting separation of“empty” AAV particles and“full” AAV particles (as evidenced by peak measurements of A260 and A280 in the eluate) with a 0 to 300 mM NaCl gradient on a CIMMULTUS QA column over 80 column volumes at a constant MgCl ₂ concentration of 1 mM at pH 9. The rising straight line shows the gradual increase in solution conductivity caused by the gradual increase in NaCl concentration during the gradient.

[0020] Figure 3 is an exemplary chromatogram depicting separation of“empty” AAV particles and“full” AAV particles using a gradient elution where the concentration of NaCl (0 to 200 mM) and MgCl ₂ (0 to 40 mM) were both increased on a CIMMULTUS QA column over 80 column volumes at pH 9.

[0021] Figure 4 is an exemplary chromatogram depicting separation of“empty” AAV particles and“full” AAV particles during an isocratic elution with a first“wash” solution and a second“elution” solution. The first“wash” solution with 9 mM MgCl ₂ , 50 mM NaCl, and pH 9 was applied to a CIMMULTUS QA column over 4 column volumes, followed by the second “elution” solution with 1 mM MgCl ₂ , 110 mM NaCl and pH 9 over 5 column volumes. Solution conductivity measurements are shown below the chromatograms. [0022] Figure 5 is an exemplary chromatogram depicting A260 and A280 measurements of the eluate during an isocratic elution with a single elution solution (applied over 14 column volumes). As shown, the first elution peak is enriched in“empty” AAV particles (A260/A280 ratio is less than one) while a second elution peak is enriched in“full” AAV particles

(A260/A280 ratio is greater than one). Solution conductivity measurements are shown below the chromatograms and indicate that there was an upward gradient in the eluate (despite the fact that the same elution solution was being used during this time period). An additional (separate) elution solution (with higher salt concentrations than the initially applied elution solution) was subsequently applied to the column (as evidenced by the significant rise in the conductivity trace) but did not lead to release of any further AAV particles from the column (i.e., all the AAV particles had been released from the column by the initial elution solution).

[0023] Figure 6 is an exemplary chromatogram depicting A260 and A280 measurements of the eluate during a control experiment that mirrored the experiment of Figure 5 expect that no recombinant AAV preparation had been loaded onto the column. The lack of peaks in the A260 and A280 measurements at any stage of the elution confirm that the column did not include any AAV particles. Solution conductivity measurements are shown below the chromatograms and do not include the gradual upward gradient that was observed in Figure 5 after the column was contacted with the initial elution solution.

DEFINITIONS

[0024] As used herein, the term“recombinant adeno-associated virus preparation” or

“recombinant AAV preparation”, refers to a product that results from a method of manufacturing recombinant AAV in a host cell (e.g., in a mammalian cell or an insect cell). In some embodiments a recombinant AAV preparation includes a mixture of full AAV particles and empty AAV particles. In some embodiments, a recombinant AAV preparation has been subjected to one or more downstream steps after an initial upstream phase, e.g., host cell lysis, filtration to remove host cell impurities, affinity purification using antibodies that bind AAV capsids, etc., as will be known to those of skill in the art. [0025] As used herein, the term“full AAV particle”, refers to an AAV particle that includes AAV capsid proteins encapsidating a heterologous nucleic acid of interest which is flanked on both sides by AAV ITRs. In some embodiments, the heterologous nucleic acid of interest is operably linked to additional regulatory sequences (e.g., promoter, enhancer, etc.).

[0026] As used herein, the term“empty AAV particle”, refers to an AAV particle that includes AAV capsid proteins but lacks in whole or in part the heterologous nucleic acid of interest flanked on either side by AAV ITRs.

[0027] As used herein, the term“column volume”, refers to the volume inside of a column (e.g., a monolithic or particles in a packed column) not occupied by the anion-exchange medium that makes up the column. This volume includes both the interstitial volume (e.g., volume outside of particles in a packed column) and the media’s own internal porosity (or“pore volume”).

[0028] As used herein, the term“membrane volume”, refers to the volume inside of a membrane not occupied by the anion-exchange medium that makes up the membrane. This volume includes the membrane’s internal porosity (or“pore volume”).

[0029] As used herein, the term“separation medium” refers to a physical structure, such as a column (e.g., a monolith) or a membrane, to which a recombinant AAV preparation is applied in order to achieve separation of certain fractions of the preparation. For example, a recombinant AAV preparation may be applied to a column, which column is then washed with one or more solutions in order to separate (and collect separated fractions) empty and full AAV particles from one another. In some embodiments, a separation medium is an anion-exchange medium. In some embodiments, a separation medium is a column (e.g., a monolithic column or particles in a packed column). In some embodiments, a separation medium is a membrane.

[0030] As used herein, the terms“gradient elution” or“gradient separation” refer to a mode of chromatographic separation wherein the concentration of one or more salts in the elution solution that is applied to the separation medium is gradually changed during the separation.

[0031] As used herein, the terms“isocratic elution” or“isocratic separation” refer to a mode of chromatographic separation where the concentration of all salts in the solution is kept constant during a defined period of the separation (e.g., the“wash” solution during a“wash” period and the“elution” solution during an“elution” period). In some embodiments, an isocratic elution uses a series of two or more separate solutions during the separation, each of which may have different fixed concentrations of one or more salts relative to another solution in the series.

DETAILED DESCRIPTION

[0032] The present disclosure describes methods for purifying recombinant adeno- associated viruses (AAVs). Specifically, the methods involve separating AAV particles that include a nucleic acid payload (full AAV particles) and AAV particles that lack in whole or in part the nucleic acid payload (empty AAV particles) using a simplified anion-exchange purification.

Anion-Exchange Separation Media

[0033] The methods of the present disclosure are not limited to any particular column structure or type of separation media. In some embodiments, a separation medium of the present disclosure includes an anion-exchange medium. In some embodiments, the anion exchange medium may be in the form of a column. In some such embodiments, the column may be a monolith of anion-exchange media. In some embodiments, the column may include packed particles of anion-exchange media. In some embodiments, a separation medium may be an anion-exchange membrane.

[0034] Exemplary anion-exchange media include, without limitation, CIMMULTUS QA

(available from BIA Separations, Ajdovscina, Slovenia), CIMMULTUS DEAE (available from BIA Separations, Ajdovscina, Slovenia), MACRO PREP Q (available from BioRad, Hercules, CA), MACRO PREP DEAE (available from BioRad, Hercules, CA), UNOSPHERE Q (available from BioRad, Hercules, CA), NUVIA Q (available from BioRad, Hercules, CA), POROS 50HQ (available from Thermo Fisher Scientific, Waltham, MA), POROS 50XQ (available from Thermo Fisher Scientific, Waltham, MA), POROS 50D (available from Thermo Fisher

Scientific, Waltham, MA), POROS 50PI (available from Thermo Fisher Scientific, Waltham, MA), SOURCE 30Q (available from GE Healthcare, Uppsala, Sweden), MACROCAP Q (available from GE Healthcare, Uppsala, Sweden), DEAE SEPHAROSE FAST FLOW (available from GE Healthcare, Uppsala, Sweden), Q SEPHAROSE FAST FLOW (available from GE Healthcare, Uppsala, Sweden), Q SEPHAROSE HIGH PERFORMANCE (available from GE Healthcare, Uppsala, Sweden), CAPTO Q (available from GE Healthcare, Uppsala, Sweden), CAPTO Q IMPRES (available from GE Healthcare, Uppsala, Sweden), CAPTO DEAE (available from GE Healthcare, Uppsala, Sweden), Q SEPHAROSE (available from GE Healthcare, Uppsala, Sweden), Q SEPHAROSE XL (available from GE Healthcare, Uppsala, Sweden), STREAMLINE Q XL (available from GE Healthcare, Uppsala, Sweden),

STREAMLINE DEAE (available from GE Healthcare, Uppsala, Sweden), ANX SEPHAROSE 4 FAST FLOW (available from GE Healthcare, Uppsala, Sweden), ESHMUNO Q (available from MilliporeSigma, Billerica, MA), FRACTOGEL TMAE (available from MilliporeSigma, Billerica, MA), FRACTOGEL DEAE (available from MilliporeSigma, Billerica, MA),

CELLUFINE MAX Q (available from JNC corporation, Tokyo, Japan), CELLUFINE MAX DEAE (available from JNC corporation, Tokyo, Japan), Q CERAMIC HYPERD F (available from Pall Corporation, Westborough, MA), DEAE CERAMIC HYPERD F (available from Pall Corporation, Westborough, MA), SARTOBIND Q (available from Sartorius Stedim Biotech, Germany), SARTOBIND STIC (available from Sartorius Stedim Biotech, Germany),

MUSTANG Q (available from Pall Corporation, Westborough, MA), and NATRIFLO HD-Q (available from Burlington, Ontario, Canada).

[0035] Anion-exchange media (e.g., columns, membranes, etc.) are typically equilibrated using standard buffers (e.g., a sodium phosphate buffer) and according to manufacturer’s specifications prior to loading with a recombinant AAV preparation. In some embodiments, an anion-exchange medium may be equilibrated to approximately the same conductivity and pH conditions as the recombinant AAV preparation. In some embodiments, the anion-exchange medium may be equilibrated to below 4 mS/cm and to a final pH of 8 to 10. In some

embodiments, the anion-exchange medium may be equilibrated to below 4.5 mS/cm and to a final pH of 8.5 to 9.5. A recombinant AAV preparation is then loaded onto the anion-exchange medium and allowed to interact with the anion-exchange medium.

[0036] As noted earlier, in some embodiments, the present disclosure is based in part on the insight that the separation of full AAV particles and empty AAV particles in recombinant AAV preparations using anion-exchange chromatography can be improved when magnesium chloride (MgCl ₂ ) is included in one or more of the elution solutions. In some embodiments, the methods involve a gradient elution with elution solutions that include particular concentrations of MgCl ₂ and optionally sodium chloride (NaCl). In some embodiments, the methods involve isocratic elutions with wash and elution solutions that include particular concentrations of MgCl ₂ and optionally NaCl. In some embodiments, a combination of MgCl ₂ and NaCl is used in one or more of the elution solutions. In some embodiments, the methods of the present disclosure improve the resolution between empty AAV particles and full AAV particles as compared to anion-exchange chromatography methods where the elution solutions comprise a single salt (e.g., NaCl) and separation therefore relies on differences in the single salt (e.g., NaCl) concentration.

[0037] In some embodiments, the present disclosure is based in part on the insight that the separation of full AAV particles and empty AAV particles in recombinant AAV preparations using anion-exchange chromatography can be performed using a single elution solution which includes a fixed concentration of magnesium chloride (MgCl ₂ ) and optionally sodium chloride (NaCl). In some embodiments, the methods involve an isocratic elution with a solution that includes particular concentrations of magnesium chloride and optionally sodium chloride. In some embodiments, a combination of magnesium chloride and sodium chloride is used in the elution solution.

Gradient Separation

[0038] In some embodiments, the present disclosure provides a gradient elution method that comprises (a) contacting an anion-exchange medium with a recombinant adeno-associated virus (AAV) preparation comprising full AAV particles and empty AAV particles; (b) applying to the anion-exchange medium a gradient solution comprising magnesium chloride (MgCl ₂ ) and optionally sodium chloride (NaCl); and (c) collecting a series of elution fractions from the anion- exchange medium, wherein the series includes a first elution fraction and a second elution fraction collected after the first elution fraction and wherein, (i) the concentration of MgCl ₂ in the gradient solution is increased during the step of applying, (ii) the concentration of NaCl in the gradient solution is increased during the step of applying, or (iii) the concentration of MgCl ₂ and the concentration of NaCl in the gradient solution are both increased during the step of applying.

[0039] As described herein, gradient separation methods of the present disclosure use gradient solutions that comprise magnesium chloride (MgCl ₂ ) and optionally sodium chloride (NaCl) where the MgCl ₂ and/or NaCl concentration is gradually increased along an elution gradient. In some embodiments, the concentration of one or both of these salts is increased during the gradient elution (e.g., linearly). In some embodiments, the concentration of one of the salts (e.g., MgCl ₂ ) may be decreased (e.g., linearly) while the concentration of the other salt (e.g., NaCl) is increased. In some embodiments, the concentration of one of the salts (e.g., MgCl ₂ ) is held at a constant concentration while the concentration of the other salt (e.g., NaCl) is increased. In some embodiments, a gradient is a linear gradient.

[0040] The art includes a variety of methods and devices that can be used to generate a gradient elution. For example, as is well known in the art, devices are available that can automatically mix different ratios of two (or more) stock or“buffer” solutions with low (e.g., 0 mM MgCl ₂ ) and high (e.g., 50 mM MgCl ₂ ) concentrations of a salt upstream of the separation medium to deliver any desired salt concentration gradient to the separation medium (e.g., a linear gradient from 0 to 50 mM MgCl ₂ ).

[0041] In some embodiments, the gradient solution does not include NaCl and the concentration of MgCl ₂ is increased during the elution gradient (e.g., linearly). In some of these embodiments, the concentration of MgCl ₂ is increased between two values within the range of 0 to 300 mM. In some embodiments, the concentration of MgCl ₂ is increased from a value in the range of 0 to 40 mM to a value in the range of 40 to 200 mM. In some embodiments, the concentration of MgCl ₂ is increased from a value in the range of 0 to 10 mM to a value in the range of 40 to 200 mM. In some embodiments, the concentration of MgCl ₂ is increased from a value in the range of 0 to 5 mM to a value in the range of 40 to 200 mM. In some embodiments, the concentration of MgCl2 is increased from 2 mM to 200 mM.

[0042] In some embodiments, the elution gradient includes NaCl and the concentration of

MgCl ₂ in the elution gradient is held constant while the concentration of NaCl is increased during the gradient (e.g., linearly). In some embodiments, the concentration of MgCl ₂ is held constant in the range of 0 to 50 mM and the concentration of NaCl is increased within a range of 10 to 300 mM. In some such embodiments, the concentration of NaCl is increased from 10 mM to 300 mM. In some embodiments, the concentration of MgCl ₂ is held constant in the range of 1 to 50 mM and the concentration of NaCl are both increased during the gradient between two values within the range of 0 to 500 mM. In some embodiments, the concentration of MgCl ₂ is held constant in the range of 1 to 20 mM and the concentration of NaCl is increased during the gradient between two values within the range of 0 to 400 mM. In some embodiments, the concentration of MgCl ₂ is held constant in the range of 2 to 10 mM and the concentration of NaCl is increased between two values within the range of 0 to 300 mM. In some embodiments, the concentration of MgCl ₂ is held constant in the range of 5 to 20 mM and the concentration of NaCl is increased between two values within the range of 0 to 300 mM. In some embodiments, the concentration of MgCl ₂ is held constant in the range of 5 to 20 mM and the concentration of NaCl is increased between two values within the range of 0 to 150 mM. In some embodiments, the concentration of MgCl ₂ is held constant and the concentration of NaCl is increased from 0 mM to within a range of 200 to 300 mM.

[0043] In some embodiments, the elution gradient includes NaCl and the concentration of

MgCl ₂ and NaCl in the gradient solution are both increased during the gradient (e.g., linearly).

In some embodiments, the concentration of MgCl ₂ is increased during the gradient between two values within the range of 0 to 100 mM and the concentration of NaCl is increased during the gradient between two values within the range of 0 to 400 mM. In some embodiments, the concentration of MgCl ₂ is increased during the gradient between two values within the range of 0 to 50 mM and the concentration of NaCl is increased during the gradient between two values within the range of 0 to 250 mM. In some embodiments, the concentration of MgCl ₂ is increased from a value in the range of 0 to 20 mM to a value in the range of 20 to 50 mM and the concentration of NaCl is increased from a value in the range of 0 to 100 mM to a value in the range of 50 to 300 mM. In some embodiments, the concentration of MgCl ₂ is increased from a value in the range of 0 to 10 mM to a value in the range of 30 to 50 mM and the concentration of NaCl is increased from a value in the range of 0 to 20 mM to a value in the range of 50 to 300 mM. In some embodiments, the concentration of MgCl ₂ is increased from a value in the range of 0 to 5 mM to a value in the range of 30 to 50 mM and the concentration of NaCl is increased from a value in the range of 0 to 10 mM to a value in the range of 100 to 300 mM.

[0044] In some embodiments, a gradient solution has a pH in the range of 6 to 10, for example in the range of 7 to 9, for example 6, 6.5, 7, 7.5, 8, 8.5, 9, 9.5, or 10. In some embodiments the pH of a gradient solution is constant during the gradient. In some

embodiments, the pH of a gradient solution changes during the gradient. It will be appreciated that a variety of well-known buffers (e.g., Bis-Tris Propane) can be used to control the pH of the gradient solutions within these ranges or at any of these values.

[0045] During the elution gradient, a series of eluates (or“fractions”) are collected (e.g., in separate containers that are connected in turn to the column or membrane outlet). In some embodiments, each eluate has approximately the same volume and the eluates are collected in a consecutive and continuous fashion from the column or membrane. In some embodiments, each eluate corresponds to a column volume or membrane volume. In some embodiments, each eluate corresponds to a fraction of a column volume or membrane volume. It will be appreciated that the number of column/membrane volumes required to separate empty AAV particles from full AAV particles will vary and depends on the nature of the recombinant AAV preparation, the type of chromatography media and the particular gradient elution solutions used. For example, a “steeper” gradient elution solution would likely require fewer column/membrane volumes than a “shallow” gradient elution solution but may also provide poorer resolution between elution peaks that are enriched for empty AAV particles and full AAV particles. In some embodiments, at least 2 column/membrane volumes, at least 5 column/membrane volumes, at least 10

column/membrane volumes, at least 15 column/membrane volumes, at least 25

column/membrane volumes, at least at least 30 column/membrane volumes, at least 40 column/membrane volumes, at least 50 column/membrane volumes, at least 60

column/membrane volumes, at least 70 column/membrane volumes, at least 80

column/membrane volumes, at least 90 column/membrane volumes, or at least 100

column/membrane volumes of a gradient elution solution are applied to a column or membrane after it has been loaded with a recombinant AAV preparation.

[0046] As noted earlier, AAV capsid proteins have a greater UV absorbance at 280 nm than at 260 nm and the nucleic acid payload has greater UV absorbance at 260 nm than at 280 nm. Accordingly, as empty AAV particles lack in whole or in part the nucleic acid payload, they have a lower UV absorbance at 260 nm (A260) than full AAV particles (which include the nucleic acid payload). As a result, the A260/A280 ratio for an elution fraction that is enriched in empty AAV particles will be less than the A260/A280 ratio for an elution fraction that is enriched in full AAV particles. In some embodiments, the methods of the present disclosure preferentially release empty AAV particles from the anion-exchange medium before releasing full AAV particles. In these embodiments, the A260/A280 ratio of the first elution fraction will be less than the second elution fraction (e.g., as shown in Figure 1). In some embodiments, the number and charge of the AAV capsid proteins and nucleic acid payload of a particular recombinant AAV may cause the full AAV particles to be preferentially released from the anion- exchange medium before the empty AAV particles. In these embodiments, the A260/A280 ratio of the first elution fraction will be greater than the second elution fraction.

[0047] In some embodiments, the methods of the present disclosure comprise a step of measuring UV absorbance (at 260 nm and 280 nm) of the eluate as it exits the anion-exchange medium during the collection step. In some embodiments, the measured A260 and A280 values (or A260/A280 ratio) are then used to determine which elution fractions to discard (e.g., those that are enriched for empty AAV particles) and which elution fractions to keep (e.g., those that are enriched in full AAV particles). In some embodiments, two or more elution fractions that are enriched in full AAV particles are pooled after they have been collected. In some embodiments, the methods of the present disclosure comprise a step of measuring the solution conductivity of the elutions fractions as they exit the anion-exchange medium during the collection step.

Isocratic Separations

[0048] In some embodiments, isocratic separations are particularly suited to large scale manufacturing because they are simpler to implement than gradient separations. In some embodiments, isocratic separations may also be advantageous over gradient separations because they require lower solution volumes. An isocratic separation that relies on a single elution solution, versus the traditional“wash” and“elution” solutions would be even simpler to implement on a large scale. Isocratic elution with a first“wash” solution and a second“ elution” solution

[0049] As described in the examples, in some embodiments, the elution profile obtained from experiments performed using a gradient elution may be used to design isocratic solutions (i.e., with suitable concentrations of MgCl ₂ and optionally NaCl) that cause preferential release of empty AAV particles or full AAV particles from an anion-exchange medium. It will be appreciated that isocratic separations are particularly suited to large scale manufacturing because they are much simpler to implement than gradient separations. In addition, less buffer volume is used during isocratic separations than gradient separations.

[0050] Thus, in some embodiments, the present disclosure provides an isocratic elution method that comprises (a) contacting an anion-exchange medium with a recombinant adeno- associated virus (AAV) preparation comprising full AAV particles and empty AAV particles; (b) applying to the anion-exchange medium a wash solution comprising magnesium chloride (MgCl ₂ ) and optionally sodium chloride (NaCl), and collecting a wash fraction; and (c) then applying to the anion-exchange medium an elution solution comprising MgCl ₂ and optionally NaCl, and collecting an elution fraction, wherein, (i) NaCl is not present in the wash and elution solutions and the concentration of MgCl ₂ is greater in the elution solution than in the wash solution; (ii) NaCl is present in the elution solution and the concentration of NaCl is greater in the elution solution than in the wash solution; or (iii) NaCl is present in the wash solution, the concentration of NaCl is less in the elution solution than in the wash solution, and the concentration of MgCl ₂ is greater in the elution solution than in the wash solution. In

embodiments where NaCl is present in the elution solution, the concentration of NaCl may be greater in the elution solution than the wash solution. In embodiments where NaCl is present in the wash solution, the concentration of NaCl may be less in the elution solution than in the wash solution and the concentration of MgCl ₂ may be greater in the elution solution than the wash solution. In embodiments where NaCl is not present in the wash and elution solutions, the concentration of MgCl2 may be greater in the elution solution than the wash solution. In embodiments where NaCl is present in the elution solution, the concentration of NaCl may be greater in the elution solution than the wash solution. In embodiments where NaCl is present in the wash solution, the concentration of NaCl may be less in the elution solution than in the wash solution and the concentration of MgCl ₂ may be (i) greater in the elution solution than the wash solution; (ii) the same in the wash solution and elution solution; or (iii) less in the elution solution than the wash solution.

[0051] As described herein, after an anion-exchange medium has been loaded with a recombinant AAV preparation, a“wash” solution is applied to the anion-exchange medium that comprises magnesium chloride (MgCl ₂ ) and optionally sodium chloride (NaCl) and then an “elution” solution is applied to the anion-exchange medium that comprises MgCl ₂ and optionally NaCl. One or more“wash” fractions are then collected while the“wash” solution is applied to the anion-exchange medium and one or more“elution” fractions are collected while the“elution” solution is applied to the anion-exchange medium. In some embodiments a single“wash” fraction is collected while the anion-exchange medium is washed with the“wash” solution and a single“elution” fraction is collected while the anion-exchange medium is eluted with the “elution” solution. In some embodiments, wash and elution fractions have the same volume and they are collected in a consecutive and continuous fashion from the anion-exchange medium. In some embodiments, each of the wash and elution fractions corresponds to multiple anion- exchange medium (e.g., column/membrane) volumes. In some embodiments, each wash or elution fraction corresponds to a column/membrane volume. In some embodiments, each wash or elution fraction corresponds to a fraction of a column/membrane volume. It will be appreciated that the number of column/membrane volumes of the wash and elution solutions that are required to separate empty AAV particles from full AAV particles will vary and depend on the nature of the recombinant AAV preparation, the type of chromatography media and the particular“wash” and“elution” solutions used. In some embodiments, at least 1

column/membrane volume, at least 5 column/membrane volumes, at least 10 column/membrane volumes, or at least 15 column/membrane volumes of each of the wash and elution solutions are applied to the anion-exchange medium after it has been loaded with a recombinant AAV preparation.

[0052] In some embodiments, NaCl is not present in the wash and elution solutions and the concentration of MgCl ₂ is greater in the elution solution than in the wash solution. In some of these embodiments, the concentration of MgCl ₂ is increased from the wash solution to the elution solution between two values within the range of 0 to 200 mM. In some embodiments, the concentration of MgCl ₂ in the wash solution is in the range of 0 to 30 mM and the concentration of MgCl ₂ in the elution solution is in the range of 10 to 50 mM. In some embodiments, the concentration of MgCl ₂ in the wash solution is in the range of 5 to 10 mM and the concentration of MgCl ₂ in the elution solution is in the range of 10 to 40 mM.

[0053] In some embodiments, the wash solution does not include NaCl and comprises 1 to 20 mM MgCl ₂ , for example 1 to 5 mM, 1 to 10 mM, 5 to 20 mM, 5 to 15 mM, or 5 to 10 mM MgCl _2. In some embodiments, the elution solution does not include NaCl and comprises a concentration of MgCl ₂ that is greater than in the wash solution, for example in the range of 10 to 50 mM, 10 to 40 mM, 10 to 30 mM, 30 to 50 mM, or 40 to 50 mM MgCl _2.

[0054] In some embodiments, NaCl is present in the elution solution and optionally in the wash solution. In some of these embodiments, the concentration of NaCl is greater in the elution solution than in the wash solution. In some of these embodiments, the concentration of MgCl ₂ is (i) greater in the elution solution than in the wash solution, (ii) the same in the wash and elution solutions, or (iii) less in the elution solution than in the wash solution.

[0055] In some embodiments, the wash solution includes 1 to 30 mM MgCl ₂ and 1 to 100 mM NaCl. In some such embodiments, a concentration of MgCl ₂ in the wash solution is in the range of 1 to 5 mM, 1 to 10 mM, 5 to 10 mM, 8 to 10 mM, 5 to 20 mM, or 5 to 15 mM, and a concentration of NaCl in the wash solution is in the range of 1 to 90 mM, 5 to 90 mM, 10 to 80 mM, 20 to 70 mM, 30 to 60 mM, 40 to 60 mM, or 45 to 55 mM.

[0056] In some embodiments, the elution solution includes 1 to 50 mM MgCl ₂ and 1 to

500 mM NaCl. In some such embodiments, a concentration of MgCl ₂ in the elution solution is in the range of 1 to 5 mM, 1 to 10 mM, 5 to 50 mM, 10 to 40 mM, 20 to 30 mM, 30 to 50 mM, 10 to 30 mM, 15 to 40 mM, or 15 to 30 mM, and a concentration of NaCl in the elution solution is in the range of 25 to 450 mM, 40 to 400 mM, 50 to 300 mM, 90 to 300 mM, 100 to 200 mM, or 100 to 150 mM.

[0057] In some embodiments, the wash solution and elution solution each contain concentrations of MgCl ₂ and NaCl in particular ranges. For example, in some embodiments, the wash solution includes 0-15 mM MgCl ₂ and 40-90 mM NaCl or 0-11 mM MgCl ₂ and 40-80 mM NaCl or 0-11 mM MgCl ₂ and 40-60 mM NaCl. In some embodiments, a concentration of MgCl ₂ in the wash solution is in the range of 0-11 mM and a concentration of NaCl in the wash solution is 45-80 mM. In some embodiments, a concentration of MgCl ₂ in the wash solution is in the range of 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, or 15 mM MgCl ₂ and a concentration of NaCl in the wash solution is in the range of 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53,

54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77 78, 79,

80, 81, 82, 83, 84, 85, 86, 87, 88, 89 or 90 mM.

[0058] In some embodiments, the wash solution and elution solution each contain concentrations of MgCl ₂ and NaCl in particular ranges. For example, in some embodiments, the wash solution includes 0-15 mM MgCl ₂ and 40-90 mM NaCl or 0-11 mM MgCl ₂ and 40-80 mM NaCl or 0-11 mM MgCl ₂ and 40-60 mM NaCl and the elution solution includes 0-5 mM MgCl ₂ and 80-140 mM NaCl or 0-5 mM MgCl ₂ and 90-130 mM NaCl or 0-2 mM MgCl ₂ and 100-120 mM NaCl or 1-2 mM MgCl ₂ and 100-110 mM NaCl or 1 mM MgCl ₂ and 110 mM NaCl. In some embodiments, a concentration of MgCl ₂ in the wash solution is in the range of 0-11 mM and a concentration of NaCl in the wash solution is 45-80 mM and a concentration of MgCl ₂ in the elution solution is in the range of 0-5 mM MgCl ₂ and 80-140 mM NaCl. In some

embodiments, a concentration of MgCl ₂ in the wash solution is in the range of 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, or 15 mM MgCl ₂ and a concentration of NaCl in the wash solution is in the range of 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77 78, 79, 80, 81, 82, 83, 84, 85, 86, 87,

88, 89 or 90 mM. In some embodiments, a concentration of MgCl ₂ in the elution solution is in the range of 0, 1, 2, 3, 4, or 5 mM and a concentration of NaCl in the wash solution is in the range of 80, 85, 90, 95, 100, 105, 110, 115, 120, 125, 130, 135, or 140 mM.

[0059] In some embodiments, the wash and elution solutions have a pH in the range of 6 to 10, for example in the range of 7 to 9, for example 6, 6.5, 7, 7.5, 8, 8.5, 9, 9.5, or 10. In some embodiments the pH of the wash and elution solutions are the same. In some embodiments, the pH of the wash and elution solutions are different. It will be appreciated that a variety of well- known buffers (e.g., Bis-Tris Propane) can be used to control the pH of the wash and elution solutions within these ranges or at any of these values.

[0060] As noted earlier, AAV capsid proteins have a greater UV absorbance at 280 nm than at 260 nm and the nucleic acid payload has greater UV absorbance at 260 nm than at 280 nm. Accordingly, as empty AAV particles lack in whole or in part the nucleic acid payload, they have a lower UV absorbance at 260 nm (A260) than full AAV particles (which include the nucleic acid payload). Thus, the A260/A280 ratio for an eluate that is enriched in empty AAV particles will be less than the A260/A280 ratio for an eluate that is enriched in full AAV particles. In some embodiments, the isocratic methods of the present disclosure preferentially release empty AAV particles from the anion-exchange medium with the first“wash” solution. In these embodiments, the A260/A280 ratio of the first eluate will be less than the second eluate which is released by the second“elution” solution (e.g., as shown in Figure 1). In some embodiments, the number and charge of the AAV capsid proteins and nucleic acid payload of a particular recombinant AAV may cause the full AAV particles to be preferentially released from the anion-exchange medium by the first“wash” solution. In these embodiments, the A260/A280 ratio of the first eluate will be greater than the second eluate.

[0061] In some embodiments, the methods of the present disclosure comprise a step of measuring UV absorbance (at 260 nm and 280 nm) of the eluate as it exits the anion-exchange medium during the collection step. In some embodiments, the measured A260 and A280 values (or A260/A280 ratio) are then used to determine which eluate(s) to discard (e.g., eluate(s) enriched for empty AAV particles) and which eluate(s) to keep (e.g., eluate(s) enriched for full AAV particles). In some embodiments, two or more“elution fractions” that are enriched for full AAV particles are pooled after they have been collected. In some embodiments, the methods of the present disclosure comprise a step of measuring the solution conductivity of the wash and elution fractions as they exit the anion-exchange medium during the collection step.

Isocratic elution with a single elution solution

[0062] In some embodiments the present disclosure provides an isocratic elution method that comprises (a) contacting an anion-exchange medium with a recombinant adeno-associated virus (AAV) preparation comprising full AAV particles and empty AAV particles; (b) applying to the anion-exchange medium an elution solution comprising a fixed concentration of chloride (MgCl ₂ ) and optionally sodium chloride (NaCl); and (c) collecting a series of elution fractions from the anion-exchange medium during the step applying the elution solution, wherein the series includes a first elution fraction and a second elution fraction collected after the first elution fraction, and wherein, (i) an A260/A280 ratio of the first eluate is less than an A260/A280 ratio of the second eluate, for example wherein an A260/A280 ratio of the first elution fraction is less than one and an A260/A280 ratio of the second elution fraction is greater than one, or (ii) an A260/A280 ratio of the first elution fraction is greater than an A260/A280 ratio of the second elution fraction, for example wherein an A260/A280 ratio of the first elution fraction is greater than one and an A260/A280 ratio of the second elution fraction is less than one.

[0063] As described herein, after an anion-exchange medium has been loaded with a recombinant AAV preparation, isocratic separation methods of the present disclosure involve applying an elution solution comprising a fixed concentration of magnesium chloride (MgCl ₂ ) and optionally sodium chloride (NaCl) to the anion-exchange medium. In some embodiments a series of elution fractions are collected while this elution solution is applied. In some embodiments, the series includes a first elution fraction, and a second elution fraction collected after the first elution fraction. In some embodiments, certain elution fractions from the series are subsequently pooled. In some embodiments, each elution fraction has approximately the same volume and the elution fractions are collected in a consecutive and continuous fashion. In some embodiments, each elution fraction corresponds to multiple column/membrane volumes. In some embodiments, each elution fraction corresponds to a column/membrane volume. In some embodiments, each elution fraction corresponds to a fraction of a column/membrane volume. It will be appreciated that the number of column/membrane volumes that are required to separate empty AAV particles from full AAV particles will vary and depend on the nature of the recombinant AAV preparation, the anion-exchange medium and the particular elution solution used. In some embodiments, at least 1 column/membrane volume, at least 5 column/membrane volumes, at least 10 column/membrane volumes, at least 15 column/membrane volumes, at least 20 column/membrane volumes, at least at least 25 column/membrane volumes, at least 30 column/membrane volumes, or more of the elution solution are applied to the column/membrane after it has been loaded with a recombinant AAV preparation.

[0064] In some embodiments, NaCl is not present in the elution solution. In some of these embodiments, the concentration of MgCl ₂ is within the range of 5 to 15 mM. In some embodiments, the concentration of MgCl ₂ in the elution solution is in the range of 7 to 11 mM. In some embodiments, the concentration of MgCl ₂ in the elution solution is in the range of 8 to 10 mM.

[0065] In some embodiments, NaCl is present in the elution solution. In some embodiments, the elution solution includes 5 to 15 mM MgCl ₂ and 25 to 75 mM NaCl. In some such embodiments, a concentration of MgCl ₂ in the elution solution is in the range of 7 to 11 mM, or 8 to 10 mM, and a concentration of NaCl in the elution solution is in the range of 25 to 75 mM, 40 to 60 mM, or 40 to 50 mM.

[0066] In some embodiments, the elution solution has a pH in the range of 6 to 10, for example in the range of 7 to 9, for example 6, 6.5, 7, 7.5, 8, 8.5, 9, 9.5, or 10. It will be appreciated that a variety of well-known buffers (e.g., Bis-Tris Propane) can be used to control the pH of the elution solution within these ranges or at any of these values.

[0067] As noted earlier, AAV capsid proteins have greater UV absorbance at 280 nm than at 260 nm and the nucleic acid payload has greater UV absorbance at 260 nm than at 280 nm. Accordingly, as empty AAV particles lack in whole or part the nucleic acid payload and therefore have a lower UV absorbance at 260 nm (A260) than full AAV particles (which include the nucleic acid payload). As a result, the A260/A280 ratio for an elution fraction that is enriched in empty AAV particles will be less than the A260/A280 ratio for an elution fraction that is enriched in full AAV particles. In some embodiments, the isocratic methods of the present disclosure preferentially release empty AAV particles from the anion-exchange medium at an earlier time point during the separation with the single elution solution than full AAV particles. In these embodiments, the A260/A280 ratio of the first elution fraction will be less than the second elution fraction, which is released following the first elution fraction (using the same elution solution) (e.g., as shown in Figure 5). In some embodiments, the number and charge of the AAV capsid proteins and nucleic acid payload of a particular recombinant AAV may cause the full AAV particles to be preferentially released from the anion-exchange medium at an earlier time point during the separation with the single elution solution. In these embodiments, the A260/A280 ratio of the first elution fraction will be greater than the second elution fraction. [0068] In some embodiments, the methods of the present disclosure comprise a step of measuring UV absorbance (at 260 nm and 280 nm) of the eluate as it exits the anion-exchange medium during the collection step. In some embodiments, the measured A260 and A280 values (or A260/A280 ratio) are then used to determine which elution fraction(s) to discard (e.g., fraction(s) enriched for empty AAV particles) and which elution fraction(s) to keep (e.g., fraction(s) enriched for full AAV particles). In some embodiments, two or more elution fractions that are enriched for full AAV particles are pooled after they have been collected. In some embodiments, the methods of the present disclosure comprise a step of measuring the solution conductivity of the eluate as it exits the anion-exchange medium during the collection step.

Quantifying Percentage Full AAV Particles

[0069] In some embodiments, the percentage of full AAV particles (vs. empty AAV particles) in a separated product (e.g., an elution fraction from an isocratic or gradient separation or a product generated by pooling several elution fractions from an isocratic or gradient separation) can be measured by an HPLC assay, e.g., as follows. A sample (e.g., elution fraction) is diluted at least 4-fold with 25 mM Bis-Tris Propane, 1 mM MgCl ₂ at pH 10 to promote binding to an anion-exchange HPLC column (BIA Separations CIMAC AAV full/empty-0.1 Analytical Column, Pores 1.3 pm). The column is equilibrated with the same dilution buffer as the sample. A small volume of diluted sample (e.g., about 50-200 pL) is injected into the equilibration buffer mobile phase, which is flowing through the column. AAV particles in the sample bind to the column. Subsequently, buffer is applied, whereby MgCl ₂ concentration of the buffer is held constant (e.g., 1 mM MgCl ₂ ), and NaCl concentration is gradually increased (e.g., from 0 to 250 mM). Empty AAV particles and full AAV particles partition/elute off of the HPLC column at different concentrations of NaCl, which correlates to actual time spent on the column by the AAV particles (“retention time”).

[0070] To decouple any effect of AAV capsid protein and AAV payload A280/A260 mutual absorbance, native florescence of AAV capsid protein is measured, at an emittance of 350 nm. The AAV payload (nucleic acid payload encapsidated by AAV particle) does not have significant florescence at emittance of 350 nm. As empty and full AAV particles desorb from the HPLC column, they flow past a fluorescence detector, and the measured fluorescence is recorded on a chromatogram. In some embodiments, empty AAV particles elute at lower NaCl concentrations than full AAV particles. Accordingly, in some such embodiments, a first peak is labeled“empty”; in these embodiments, full particles elute after empty particles, and a next peak is labeled“full.” It is to be understood that in some embodiments, a first peak may be labeled “full” and a second peak“empty”. It will be known by one of skill in the art that distinguishing between empty and full peaks may be facilitated using an AKTA gradient, which displays A280/A260 ratios, and informs order of HPLC column desorption (e.g., empty particles, then full or full particles, then empty).

[0071] Once peaks have been identified, they are integrated. The ratio of the integrated peaks (peak area) is representative of the ratio of empty AAV particles to full AAV particles (i.e., providing a“% Full AAV”).

[0072] In some embodiments, a sample (e.g., elution fraction) is enriched in full AAV particles and such a fraction comprises greater than 50% full AAV particles, as measured by an HPLC assay described herein. In some embodiments, a sample (e.g., elution fraction) comprises greater than 60, 70, 80, or 90% full AAV particles, as measured by the HPLC assay described herein.

EXAMPLES

[0073] The present disclosure exemplifies methods by which to separate full AAV particles and empty AAV particles within a recombinant AAV preparation which has been loaded onto an anion-exchange medium (e.g., column). The methods demonstrate that the separation can be achieved with high resolution using varying concentrations of NaCl and/or MgCl ₂ in the elution solutions. In the present examples, a preparation of affinity-purified recombinant AAV2 particles, which includes an empty to full ratio of about 2: 1, was diluted to a conductivity of below 4 mS/cm and to a final pH of 8.5 to 9.5. This preparation was then loaded onto a CIMMULTUS QA anion-exchanger monolith column equilibrated to the same conductivity and pH conditions as the preparation. Empty AAV particles and full AAV particles were then eluted from the CIMMULTUS QA column by using solutions with varying combinations and concentrations of NaCl and MgCl _2. Elutions were either gradient elutions or isocratic elutions.

[0074] The present disclosure also exemplifies methods demonstrating that the separation can be achieved with high resolution by using a single elution solution that includes MgCl ₂ and optionally NaCl. In the present example, a preparation of affinity-purified recombinant AAV2 particles, which includes an empty to full ratio in the range 1 :3 to 1 :1, was diluted to a conductivity of below 4.5 mS/cm and to a final pH of 8.5 to 9.5. This preparation was then loaded onto a CIMMULTUS QA anion-exchanger monolith equilibrated to the same conductivity and pH conditions as the preparation. Empty AAV particles and full AAV particles were then eluted from the CIMMULTUS QA column using a single elution solution that included MgCl ₂ and NaCl.

Example 1 : Gradient elution with varying NaCl concentration and constant MgCl concentration

[0075] In the present example, a gradient elution was performed over a CIMMULTUS

QA column, using 0 to 300 mM NaCl and a constant concentration (1 mM) MgCl ₂ , with a gradient volume of 80 column volumes at pH 9. As can be observed in Figure 2, empty AAV particles eluted at a lower concentration of NaCl than full AAV particles. The two

chromatograms in Figure 2 represent UV absorbance at wavelengths 260 nm and 280 nm. AAV capsid proteins have a maximum UV absorbance at 280 nm and the maximum absorbance for the nucleic acid payload occurs at 260 nm. The first elution peak (“early eluting”) peak in the chromatogram has an A260/A280 ratio of less than one, indicating that the fraction represented by the first elution peak is enriched in empty AAV particles. The A260/A280 ratio in the second elution peak (“later eluting”) peak is greater than one, which corresponds to a fraction that is enriched in full AAV particles.

Example 2: Gradient elution with varying NaCl and MgCl concentrations

[0076] In the present example, a gradient elution was performed over a CIMMULTUS

QA column, using solutions with increasing NaCl concentrations, from 0 to 300 mM, and increasing MgCl ₂ concentrations, from 0 to 40 mM, with a gradient volume of 80 column volumes at pH 9. Simultaneous increase in NaCl and MgCl ₂ concentrations resulted in an improved resolution between the empty AAV particles (first elution peak with A260/A280 < 1) and full AAV particles (second elution peak with A260/A280 > 1), as shown in Figure 3.

Example 3 : Isocratic elution with a first“wash” solution and a second“elution” solution

[0077] In the present example, we used the results from Examples 1 and 2 to design an isocratic elution with a first“wash” solution and second“elution” solution on a dMMEILTEiS QA column. An isocratic wash and elution were used to separate empty AAV particles and full AAV particles from one another. An isocratic wash at lower concentrations of NaCl and/or MgCl ₂ was used to remove a fraction enriched in empty AAV particles from the column. An isocratic elution at higher concentrations of NaCl and/or MgCl ₂ was then used to elute a fraction enriched in full AAV particles. Table 1 shows five different isocratic wash/elution combinations used in this example, as well as the percent (%) of full AAV particles (i.e., purity as measured by HPLC assay described herein) as measured in the eluates from the five different isocratic elutions. As can be seen in Table 1, increasing MgCl ₂ concentration in the first isocratic solution (wash) resulted in higher percentage (i.e., purity) of full AAV particles in the measured elution fractions.

Table 1: Separation of empty and full AAV particles by isocratic elution with a first “wash” solution and a second“elution” solution

[0078] Figure 4 shows the chromatograms that were obtained when a first“wash” solution with 10 mM MgCl ₂ , 53 mM NaCl, and pH 9 was applied to a CIMMULTUS QA column over 4 column volumes, followed by a second“elution” solution with 1 mM MgCl ₂ , 110 mM NaCl and pH 9 over 5 column volumes. Solution conductivity measurements are shown below the chromatograms. The first elution peak in the chromatogram (corresponding to eluate released from the column with the first“wash” solution) has an A260/A280 ratio of less than one, indicating that the fraction represented by the first elution peak is enriched in empty AAV particles. The A260/A280 ratio in the second elution peak (corresponding to eluate released from the column with the second“elution” solution) is greater than one, which corresponds to a fraction that is enriched in full AAV particles.

[0079] Percentage of full AAV particles (vs. empty AAV particles) was measured in samples (i.e., fractions identified in Table 1) by an HPLC assay as follows. A given sample was first diluted at least 4-fold with 25 mM Bis-Tris Propane, 1 mM MgCl ₂ at pH 10 to promote binding to an anion-exchange HPLC column (BIA Separations CIMAC AAV full/empty-0.1 Analytical Column, Pores 1.3 pm). The column was equilibrated with the same dilution buffer as the sample. A small volume of diluted sample (e.g., about 50-200 pL) was injected into the equilibration buffer mobile phase, which flows through the column. AAV particles in the sample bound to the column. Subsequently, buffer was applied, whereby MgCl ₂ concentration of the buffer was held constant (see Table 1), and NaCl concentration was gradually increased (see Table 1). Empty AAV particles and full AAV particles partitioned/eluted off of the HPLC column at different concentrations of NaCl, which correlates to actual time spent on the column by the AAV particles (“retention time”). [0080] To decouple any effect of AAV capsid protein and AAV payload A280/A260 mutual absorbance, native florescence of AAV capsid protein was measured, at an emittance of 350 nm. The AAV payload (nucleic acid payload encapsidated by AAV particle) does not have significant florescence at emittance of 350 nm. As empty and full AAV particles desorbed from the HPLC column, they flowed past a fluorescence detector, and the measured fluorescence is recorded on a chromatogram. Peaks were identified as either“empty” or“full” as described herein.

[0081] Once peaks were identified, they were integrated. The ratio of the integrated peaks (peak area) is representative of the ratio of empty AAV particles to full AAV particles (i.e., providing a“% Full AAV”), as shown in Table 1.

Example 4: Isocratic elution with a first“wash” solution and a second“elution” solution

[0082] In the present example, an isocratic wash and elution were again used to separate empty AAV particles and full AAV particles from one another on a CIMMULTUS QA column. An isocratic wash solution at lower concentrations of NaCl and/or MgCl ₂ and varying pH was used to remove a fraction enriched in empty AAV particles from the column. An isocratic elution at higher concentrations of NaCl and/or MgCl ₂ was then used to elute a fraction enriched in full AAV particles. Table 2 shows three different isocratic wash/elution combinations used in this example, as well as the percent (%) of full AAV particles (i.e., purity) as measured in the eluates from the three different isocratic elutions. As can be seen in Table 2, decreasing pH in the first isocratic solution (wash) resulted in higher percentage (i.e., purity) of full AAV particles in the measured elution fractions (for given NaCl and MgCl ₂ concentrations in the first isocratic solution (wash)). As also seen in Table 2, the maximum purity achieved was approximately 94% full AAV particles, and occurred in conditions where the first isocratic solution (wash) pH was 8.8 or 9.0. Table 2: Separation of empty and full AAV particles by isocratic elution with a first “wash” solution and a second“elution” solution

[0083] Percentage of full AAV particles (vs. empty AAV particles) was measured in samples (i.e., fractions identified in Table 2) by an HPLC assay as follows. A given sample was first diluted at least 4-fold with 25 mM Bis-Tris Propane, 1 mM MgCl ₂ at pH 10 to promote binding to an anion-exchange HPLC column (BIA Separations CIMAC AAV full/empty-0.1 Analytical Column, Pores 1.3 pm). The column was equilibrated with the same dilution buffer as the sample. A small volume of diluted sample (e.g., about 50-200 pL) was injected into the equilibration buffer mobile phase, which flows through the column. AAV particles in the sample bound to the column. Subsequently, buffer was applied, whereby MgCl ₂ concentration of the buffer was held constant (see Table 2), and NaCl concentration was gradually increased (see Table 2). Empty AAV particles and full AAV particles partitioned/eluted off of the HPLC column at different concentrations of NaCl, which correlates to actual time spent on the column by the AAV particles (“retention time”).

[0084] To decouple any effect of AAV capsid protein and AAV payload A280/A260 mutual absorbance, native florescence of AAV capsid protein was measured, at an emittance of 350 nm. The AAV payload (nucleic acid payload encapsidated by AAV particle) does not have significant florescence at emittance of 350 nm. As empty and full AAV particles desorbed from the HPLC column, they flowed past a fluorescence detector, and the measured fluorescence is recorded on a chromatogram. Peaks were identified as either“empty” or“full” as described herein. [0085] Once peaks were identified, they were integrated. The ratio of the integrated peaks (peak area) is representative of the ratio of empty AAV particles to full AAV particles (i.e., providing a“% Full AAV”), as shown in Table 2.

Example 5: Isocratic Separation with a single elution solution

[0086] In the present example, a separation of empty AAV particles and full AAV particles was performed over a CIMMULTUS QA column, using a single elution solution (14 column volumes) which included 9 mM MgCl ₂ and 50 mM NaCl at pH 9. Figure 5 shows the A260 and A280 chromatograms and conductivity trace of the eluate as it exits the column. As noted earlier, AAV capsid proteins have a greater UV absorbance at 280 nm than at 260 nm and the nucleic acid payload has greater UV absorbance at 260 nm than at 280 nm. Accordingly, as empty AAV particles lack in whole or part the nucleic acid payload, they have a lower UV absorbance at 260 nm (A260) than full AAV particles (which include the nucleic acid payload). The“early eluting” (i.e., first) peak in the chromatogram has an A260/A280 ratio of less than one, indicating that the elution fraction represented by the early eluting peak is enriched in empty AAV particles. The A260/A280 ratio in the“later eluting” (i.e., second) peak is greater than one, which corresponds to an elution fraction that is enriched in full AAV particles. Of note, we also observed a gradient in the conductivity trace which is shown below the chromatograms.

This gradient was unexpected since the elution solution applied to the anion-exchange medium during this time period was the same (i.e., there was no gradient in the salt concentration of the elution solution that was applied to the anion-exchange medium). Without being bound to any theory, we propose that the MgCl ₂ in the single elution solution interacts with AAV capsid proteins bound to the separation medium and does not exit during initial applications of the elution solution to the separation medium. As interaction sites get gradually saturated, the remaining MgCl ₂ being added to the separation medium starts to increase the concentration of “free” MgCl ₂ and hence the conductivity of the elution solution that exits the separation medium increases along a gradient (even though the elution solution being added to the column is not changing). In the experiments described herein, the empty AAV particles were less tightly bound to the anion-exchange medium and eluted earlier. The more strongly bound full AAV particles eluted later as the concentration of“free” MgCl ₂ increased. A second, elution solution with higher salt concentrations than the initial elution solution was subsequently applied to the column (as evidenced by the significant rise in the conductivity trace) but did not lead to release of any further AAV particles from the column (i.e., all the AAV particles had been released from the column by the initial elution solution).

[0087] As a control experiment, the same initial elution solution (i.e., 9 mM MgCl ₂ and

50 mM NaCl, pH 9) was applied to an unloaded column. As can be seen in Figure 6, when no AAV particles were present (as confirmed by A260 and A280 traces) we did not observe the same gradient in the conductivity trace during this time period.