Method for Purifying Biologically Active Peptide by Using Protein A Affinity Chromatography

Abstract

Provided is a method of purifying a mixture of Fc-containing bioactive peptides by using an affinity column including an affinity matrix containing a protein A ligand, wherein the mixture of Fc-containing bioactive peptides includes a first Fc-containing bioactive peptide and a second Fc-containing bioactive peptide, and the second Fc-containing bioactive peptide includes at least one more human VH3 domain, compared to the first Fc-containing bioactive peptide. According to the purification method, bioactive peptides having the same or similar structures can be precisely separated to the high level of purity while simplification of the process is achieved.

Claims

1. A method of purifying an Fc-containing bioactive peptide, the method comprising: (a) loading a mixture of Fc-containing bioactive peptides into a column including an affinity matrix containing protein A ligand, wherein the mixture of Fc-containing bioactive peptides includes a first Fc-containing bioactive peptide and a second Fc-containing bioactive peptide, and the second Fc-containing bioactive peptide includes at least one more human VH3 domain, compared to the first Fc-containing bioactive peptide; and (b) loading an eluate into the column to separate and elute the Fc-containing bioactive peptides at different pHs depending on the number of human VH3 domains included in each of the Fc-containing bioactive peptides that are included in the mixture of Fc-containing bioactive peptides.

2. The method of claim 1, wherein the mixture of Fc-containing bioactive peptides further comprises a third Fc-containing bioactive peptide which includes at least one more human VH3 domain, compared to the second Fc-containing bioactive peptide.

3. The method of claim 2, wherein in the mixture of Fc-containing bioactive peptide, the first Fc-containing bioactive peptide includes n human VH3 domains (n is an integer greater than or equal to 0), the second Fc-containing bioactive peptide includes n+1 human VH3 domains, and the third Fc-containing bioactive peptide includes n+2 human VH3 domains.

4. The method of claim 1, wherein the process (b) comprises: (b1) eluting the first Fc-containing bioactive peptide by loading an eluate having a first pH range into the column; and (b2) eluting the second Fc-containing bioactive peptide by loading an eluate having a second pH range lower than the first pH range into the column.

5. The method of claim 2, wherein the process (b) comprises: (b1) eluting the first Fc-containing bioactive peptide by loading an eluate having a first pH range into the column; (b2) eluting the second Fc-containing bioactive peptide by loading an eluate having a second pH range lower than the first pH range into the column; and (b2) eluting the third Fc-containing bioactive peptide by loading an eluate having a third pH range lower than the second pH range into the column.

6. The method of claim 1, wherein the Fc is an Fc to which a mutation affecting binding to wild-type protein A is not introduced.

7. The method of claim 1, wherein when the first Fc-containing bioactive peptide or the second Fc-containing bioactive peptide includes a VH3 domain, the first Fc-containing bioactive peptide and the second Fc-containing bioactive peptide each include a VH domain-containing variable region sequence selected from the following sequences: a heavy chain variable region sequence of SEQ ID No. 1 and a light chain variable region sequence of SEQ ID No. 2; a heavy chain variable region sequence of SEQ ID No. 3 and a light chain variable region sequence of SEQ ID No. 4; a heavy chain variable region sequence of SEQ ID No. 5 and a light chain variable region sequence of SEQ ID No. 6; a heavy chain variable region sequence of SEQ ID No. 7 and a light chain variable region sequence of SEQ ID No. 8; and a heavy chain variable region sequence of SEQ ID No. 9 and a light chain variable region sequence of SEQ ID No. 10.

8. The method of claim 2, wherein the third Fc-containing bioactive peptide includes a VH domain-containing variable region sequence selected from the following sequences: a heavy chain variable region sequence of SEQ ID No. 1 and a light chain variable region sequence of SEQ ID No. 2; a heavy chain variable region sequence of SEQ ID No. 3 and a light chain variable region sequence of SEQ ID No. 4; a heavy chain variable region sequence of SEQ ID No. 5 and a light chain variable region sequence of SEQ ID No. 6; a heavy chain variable region sequence of SEQ ID No. 7 and a light chain variable region sequence of SEQ ID No. 8; and a heavy chain variable region sequence of SEQ ID No. 9 and a light chain variable region sequence of SEQ ID No. 10.

9. The method of claim 1, wherein the Fc-containing bioactive peptide is an antibody containing IgG.

10. The method of claim 1, wherein the Fc-containing bioactive peptide is an IgG-scFv bispecific antibody.

11. The method of claim 1, wherein the Fc-containing bioactive peptide comprises a peptide drug bound to Fc.

12. The method of claim 11, wherein the peptide drug is selected from the group consisting of hormones, cytokines, enzymes, antibodies, growth factors, transcriptional regulators, blood factors, vaccines, structural proteins, ligand proteins, and receptors.

13. The method of claim 2, wherein the first Fc-containing bioactive peptide does not include scFv; the second Fc-containing bioactive peptide includes one scFv consisting of SEQ ID No. 11, connected to the C-terminus of one of two heavy chains of Fc; and the third Fc-containing bioactive peptide includes two scFv each consisting of SEQ ID No. 11, respectively connected to the C-termini of two heavy chains of Fc.

14. The method of claim 1, wherein the Fc-containing bioactive peptide is IgG, and a variable region of the IgG antibody comprises a human VH3 domain.

15. A purification method comprising: (a-1) loading an antibody mixture into a column including an affinity matrix containing protein A ligand, wherein the antibody mixture comprises: a monospecific antibody; a monovalent bispecific antibody in which one antigen-binding fragment including a human VH3 domain binds to the C-terminus of any one of two heavy chain constant regions of the monospecific antibody; and a bivalent bispecific antibody in which the antigen-binding fragment binds to the C-terminus of each of two heavy chain constant regions of the monospecific antibody; (b-1) eluting the monospecific antibody by loading an eluate having a first pH range into the column; (c-1) eluting the monovalent bispecific antibody by loading an eluate having a second pH range lower than the first pH range into the column; and (d-1) eluting the bivalent bispecific antibody by loading an eluate having a third pH range lower than the second pH range into the column.

16. The purification method of claim 15, wherein a variable region of the monospecific antibody comprises a human VH3 domain.

17. The purification method of claim 15, wherein a variable region of the monospecific antibody does not include a human VH3 domain.

18. The purification method of claim 15, wherein the antigen-binding fragment is scFv including an amino acid sequence of SEQ ID No. 11.

19. The purification method of claim 15, wherein the first pH range is 3.4 or more and 5.0 or less, the second pH range is 3.3 or more and 4.1 or less, and the third pH range is 3.0 or more and 4.0 or less.

Description

BRIEF DESCRIPTION OF DRAWINGS

[0085] FIG. 1 shows the purification profile of affinity chromatography using a MabSelet Sure resin applied on a bioactive peptide mixture including a monospecific antibody, a monovalent bispecific antibody, and a bivalent bispecific antibody.

[0086] FIG. 2 shows the purification profile of affinity chromatography using a MabSelect PrismA resin applied on a bioactive peptide mixture including a monospecific antibody, a monovalent bispecific antibody, and a bivalent bispecific antibody.

[0087] FIG. 3 shows the purification profile of affinity chromatography using a protein An FF resin applied on a bioactive peptide mixture including a monospecific antibody, a monovalent bispecific antibody, and a bivalent bispecific antibody.

[0088] FIG. 4 shows the purification profile of affinity chromatography using a MabSelect Xtra resin applied on a bioactive peptide mixture including a monospecific antibody, a monovalent bispecific antibody, and a bivalent bispecific antibody.

[0089] FIG. 5 shows SE-HPLC analysis of the purification results obtained by applying a MabSelect Xtra resin on a bioactive peptide mixture including a monospecific antibody, a monovalent bispecific antibody and a bivalent bispecific antibody.

[0090] FIG. 6 shows SDS-PAGE analysis of the purification results obtained by applying a MabSelect Xtra resin on a bioactive peptide mixture including a monospecific antibody, a monovalent bispecific antibody and a bivalent bispecific antibody.

[0091] FIG. 7 shows the purification profile of affinity chromatography using a protein A HP resin applied on a bioactive peptide mixture including a monospecific antibody, a monovalent bispecific antibody, and a bivalent bispecific antibody.

[0092] FIG. 8 shows SE-HPLC analysis of the purification results obtained by applying a protein A HP resin on a bioactive peptide mixture including a monospecific antibody, a monovalent bispecific antibody and a bivalent bispecific antibody.

[0093] FIG. 9 shows SDS-PAGE analysis of the purification results obtained by applying a protein A HP resin on a bioactive peptide mixture including a monospecific antibody, a monovalent bispecific antibody and a bivalent bispecific antibody.

[0094] FIG. 10 shows the purification profile of affinity chromatography using a POROS MabCapture A Select resin applied on a bioactive peptide mixture including a monospecific antibody, a monovalent bispecific antibody, and a bivalent bispecific antibody.

[0095] FIG. 11 shows SE-HPLC analysis of the purification results obtained by applying a POROS MabCapture A Select resin on a bioactive peptide mixture including a monospecific antibody, a monovalent bispecific antibody and a bivalent bispecific antibody.

[0096] FIG. 12 shows SDS-PAGE analysis of the purification results obtained by applying a POROS MabCapture A Select resin on a bioactive peptide mixture including a monospecific antibody, a monovalent bispecific antibody and a bivalent bispecific antibody.

[0097] FIG. 13 shows the purification profile of affinity chromatography using an AbSolute High Cap resin applied on a bioactive peptide mixture including a monospecific antibody, a monovalent bispecific antibody, and a bivalent bispecific antibody.

[0098] FIG. 14 shows SE-HPLC analysis of the purification results obtained by applying an AbSolute High Cap resin on a bioactive peptide mixture including a monospecific antibody, a monovalent bispecific antibody and a bivalent bispecific antibody.

[0099] FIG. 15 shows SDS-PAGE analysis of the purification results obtained by applying an AbSolute High Cap resin on a bioactive peptide mixture including a monospecific antibody, a monovalent bispecific antibody and a bivalent bispecific antibody.

[0100] FIG. 16 shows the analysis profile of each of the culture fluid produced by the transient method (Harvested Cell Culture Fluid) and the sample obtained by purifying the culture fluid with protein A, obtained using a POROS A 20 μm analysis column. Peak 1, peak 2, and peak 3 respectively indicate retention time of the purified monospecific antibody, the purified monovalent bispecific antibody, and the purified bivalent bispecific antibody after loaded into a POROS A 20 μm analysis column.

[0101] FIG. 17 shows the purification profile of affinity chromatography using POROS MabCapture A Select resin applied on a bioactive peptide mixture containing hu11F11 and hu11F11-F06.

[0102] FIG. 18 shows SE-HPLC analysis of the purification results obtained by applying a POROS MabCapture A Select resin on a bioactive peptide mixture including hu11F11 and hu11F11-F06.

[0103] FIG. 19 shows SDS-PAGE analysis of the purification results obtained by applying a POROS MabCapture A Select resin on a bioactive peptide mixture including hu11F11 and hu11F11-F06.

[0104] FIG. 20 shows the purification profile of affinity chromatography using POROS MabCapture A Select resin applied on a bioactive peptide mixture containing ch1E4 and ch1E4-F06.

[0105] FIG. 21 shows the purification profile of affinity chromatography using POROS MabCapture A Select resin applied on a bioactive peptide mixture containing anti-BACE and anti-BACE-F06.

[0106] FIG. 22 shows SE-HPLC analysis of the purification results obtained by applying a POROS MabCapture A Select resin on a bioactive peptide mixture including ch1 E4 and ch1E4-F06.

[0107] FIG. 23 shows SE-HPLC analysis of the purification results obtained by applying a POROS MabCapture A Select resin on a bioactive peptide mixture including anti-BACE and anti-BACE-F06.

[0108] FIG. 24 shows SE-HPLC analysis of the purification results obtained by applying a POROS MabCapture A Select resin on a bioactive peptide mixture including ch1E4 and ch1E4-F06.

[0109] FIG. 25 shows SDS-PAGE analysis of the purification results obtained by applying a POROS MabCapture A Select resin on a bioactive peptide mixture including anti-BACE and anti-BACE-F06.

[0110] FIG. 26 shows the purification profile of affinity chromatography using POROS MabCapture A Select resin applied on a bioactive peptide mixture containing Fc and Fc-F06.

[0111] FIG. 27 shows SE-HPLC analysis of the purification results obtained by applying a POROS MabCapture A Select resin on a bioactive peptide mixture including Fc and Fc-F06.

[0112] FIG. 28 shows SDS-PAGE analysis of the purification results obtained by applying a POROS MabCapture A Select resin on a bioactive peptide mixture including Fc and Fc-F06.

[0113] FIG. 29 shows the purification profile of affinity chromatography using POROS MabCapture A Select resin applied on a bioactive peptide mixture containing GLP-1-Fc and GLP-1-Fc-F06.

[0114] FIG. 30 shows SE-HPLC analysis of the purification results obtained by applying a POROS MabCapture A Select resin on a bioactive peptide mixture including GLP-1-Fc and GLP-1-Fc-F06.

[0115] FIG. 31 shows the purification profile of affinity chromatography using a POROS MabCapture A Select resin applied on a sample containing anti-BACE, anti-BACE-F06, hu11F11, and hu11F11-F06.

[0116] FIG. 32 shows SE-HPLC analysis of the purification results obtained by applying a POROS MabCapture A Select resin on a sample containing anti-BACE, anti-BACE-F06, hu11F11, and hu11F11-F06.

[0117] FIG. 33 shows SDS-PAGE analysis of the purification results obtained by applying a POROS MabCapture A Select resin on a sample containing anti-BACE, anti-BACE-F06, hu11F11, and hu11F11-F06.

[0118] FIG. 34 shows LC/MS analysis of the purification results obtained by applying a POROS MabCapture A Select resin on a sample containing anti-BACE, anti-BACE-F06, hu11F11, and hu11F11-F06.

[0119] FIG. 35 shows the purification profile of affinity chromatography using a POROS MabCapture A Select resin applied on a sample containing ch1E4, ch1E4-F06, hu11F11, and hu11F11-F06.

DETAILED DESCRIPTIOIN OF INVENTION

[0120] Hereinafter, the advantages and features of the present disclosure and methods for achieving the same will be described in detail with reference to the following examples. However, the present disclosure is not limited to the embodiments disclosed below, and may be implemented in various different forms. The present embodiments are provided to make the disclosure of the present invention complete, and to fully inform those of ordinary skill in the art to which the present invention belongs, the scope of the invention, and the present disclosure is defined by the claims only.

EXAMPLE 1: PREPARATION OF MONOVALENT BISPECIFIC ANTIBODIES

[0121] (1) Construction of Monovalent Bispecific Antibody Expression Vector

[0122] To construct a monovalent bispecific antibody expression vector, the nucleotide sequence of an antibody including the signal sequence was inserted into the multi cloning site (MCS) of the pcDNA3.4 (invitrogen) vector. The expression vector used herein was a monocistronic vector, and a heavy chain expression vector and a light chain expression vector were prepared, separately.

[0123] As the heavy chain expression vector, heavy chain expression vector 1 having a first heavy chain sequence and heavy chain expression vector 2 having a second heavy chain sequence were prepared.

[0124] In heavy chain expression vector 1, anti-IGF1R scFv (1564, F06, VH5, VH9, VH16, see Table 2) is connected as a linker to the C-terminus of immunoglobulin to which a human heavy chain constant region and a heavy chain variable region of an anti-α-synuclein antibody, such as ch11F11 (see the sequence list of WO2019/117684), ch1E4 (see heavy chain SEQ ID No. 18 and light chain SEQ ID No. 19), or hu11F11 (see the sequence list of WO2019/117684), or anti-BACE (see the sequence of hu2H8v29 in WO2019/094608) antibody, are connected, and a sequence substitution (hole mutation: T366S, L368A, Y406V) has occurred at a specific position in the heavy chain constant region.

[0125] In heavy chain expression vector 2, only the heavy chain variable region encoding the anti-α-synuclein antibody or the anti-BACE antibody and a human heavy chain constant region are connected, and a sequence substitution (knob mutation: T366W) has occurred at a specific position in the heavy chain constant region.

[0126] In the case of the light chain sequence inserted into the light chain expression vector, the light chain variable region encoding the anti-α-synuclein antibody or the anti-BACE antibody and the human light chain constant region are connected.

[0127] In Example 1, the nucleic acid sequence encoding the antibody in heavy chain expression vector 1, heavy chain expression vector 2, and the light chain expression vector is set forth in SEQ ID No. 20 to SEQ ID No. 22.

[0128] The amino acid sequence constituting hu11F11×F06, which is an example of the monovalent bispecific antibody according to an embodiment, is shown in Table 1 below.

TABLE-US-00001 TABLE 1 Double antibody heavy component Typical double Bispecific antibody Bispecific antibody structure Amino acid antibody light chain heavy component SEQ ID description sequence hu11F11 × F06 Hu11F11-VLv3 4c hu11F11(ver.2) SEQ ID hu11F11 (ver.2) WO2019/ (monovalent) (WO2019/117684) (M428L)- No. 12 (IgG) M428L, 117684 F06(de2)(StoP) HOLE MUTATION (HC 1, -hole) (G4S)3 SEQ ID No. 14 F06(de2)(StoP) VL SEQ ID No. 15 (G4S)4 SEQ ID No. 16 F06(de2)(StoP) VH SEQ ID No. 17 hu11F11(ver.2) SEQ ID hu11F11 (ver.2) WO2019/ (M428L) No. 13 (IgG) M428L, 117684 (HC 2, -knob) KNOB MUTATION

[0129] A monovalent bispecific antibody is a heterodimeric form in which an scFv is connected to the C terminus of one of the heavy chain Fc of an anti-α-synuclein antibody or an anti-BACE antibody and scFv is not connected to the other heavy chain Fc.

[0130] In general, a combination of knob-knob dimer antibody, a combination of knob-hole dimer antibody, and a combination of hole-hole dimer antibody are all produced in the production of monovalent bispecific antibodies. During the purification process for separation of a pure heterodimer (knob-hole dimer), for the purpose of purification resolution comparison and the use as a control material, a monospecific antibody (150 kd, which is the same size as the knob-knob dimer) and a bivalent bispecific antibody (200 kd, which is the same size as the hole-hole dimer) were additionally cloned.

[0131] For purification resolution comparison and the production of a control material required for analysis, in the case of the heavy chain sequence of a monospecific antibody, there are only a heavy chain variable region encoding an anti-α-synuclein antibody or an anti-BACE antibody and a human immunoglobulin heavy chain constant region, and in the case of the heavy chain sequence of the bivalent bispecific antibody, scFv is connected as a linker to the C terminus of the same heavy chain sequence as the single antibody, but no sequence substitution is made in the heavy chain constant region. The monospecific antibody, the bivalent bispecific antibody, and the monovalent bispecific antibody have the same light chain sequence region.

[0132] The heavy and light chain variable region sequences of the anti-IGF1R scFvs (1564, F06, VH5, VH9, VH16) used in this Example are shown in Table 2 below.

TABLE-US-00002 TABLE 2 Type of Antibody sequence Sequence 1564 Variable heavy EVQLLESGGGLVQPGGSLRLSCAASGFTFSSYDMSWVR chain region QAPGKCLEWVSAISYDNANTYYADSVKGRFTISRDNSKN sequence TLYLQMNSLRAEDTAVYYCAKGVLTTLMNWFDYWGQGT (SEQ ID No. 1) LVTVSS Variable light QSVLTQPPSASGTPGQRVTISCTGSSSNIGSNDVSWYQ chain region QLPGTAPKLLIYANVNRPSGVPDRFSGSKSGTSASLAISG sequence LRSEDEADYYCGAWDDSLNAYVFGCGTKLTVL (SEQ ID No. 2) F06 Variable heavy EVQLLESGGGLVQPGGSLRLSCAASGFTFSSYDMSWVR chain region QAPGKCLEWVSAISYDNANTYYADSVKGRFTISRDNSKN sequence TLYLQMNSLRAEDTAVYYCAKGVLTTLMNWFDYWGQGT (SEQ ID No. 3) LVTVSS Variable light QSVLTQPPSASGTPGQRVTISCTGSSSNIGSNDVSWYQ chain region QLPGTAPKLLIYANVNRPSGVPDRFSGSKSGTSASLAISG sequence LRSEDEADYYCGTWAGSLNAYVFGCGTKLTVL (SEQ ID No. 4) VH5 Variable heavy EVQLLESGGGLVQPGGSLRLSCAASGFTFSSYDMSWVR chain region QAPGKCLEWVSAISGDNASTYYADSVKGRFTISRDNSKN sequence TLYLQMNSLRAEDTAVYYCAKGVLTTLMNWFDSWGQGT (SEQ ID No. 5) LVTVSS Variable light QSVLTQPPSASGTPGQRVTISCTGSSSNIGSNDVSWYQ chain region QLPGTAPKLLIYANVNRPSGVPDRFSGSKSGTSASLAISG sequence LRSEDEADYYCGAWDDSLNAYVFGCGTKLTVL (SEQ ID No. 6) VH16 Variable heavy EVQLLESGGGLVQPGGSLRLSCAASGFTFSSYDMSWVR chain region QAPGKCLEWVSAISGSNANTYYADSVKGRFTISRDNSKN sequence TLYLQMNSLRAEDTAVYYCAKGVLTTLMNWFDYWGQGT (SEQ ID No. 7) LVTVSS Variable light QSVLTQPPSASGTPGQRVTISCTGSSSNIGSNDVSWYQ chain region QLPGTAPKLLIYANVNRPSGVPDRFSGSKSGTSASLAISG sequence LRSEDEADYYCGAWDDSLNAYVFGCGTKLTVL (SEQ ID No. 8) VH9 Variable heavy EVQLLESGGGLVQPGGSLRLSCAASGFTFSSYDMSWVR chain region QAPGKCLEWVSAISGDNGSTYYADSVKGRFTISRDNSK sequence NTLYLQMNSLRAEDTAVYYCAKGVLTTLMNWFDSWGQ (SEQ ID No. 9) GTLVTVSS Variable light QSVLTQPPSASGTPGQRVTISCTGSSSNIGSNDVSWYQ chain region QLPGTAPKLLIYANSNRPSGVSDRFSGSKSGTSASLAISG sequence LRSEDEADYYCGAWDDSLNGYVFGCGTKLTVL (SEQ ID No. 10)

[0133] (2) Transient Expression

[0134] The prepared vector was subjected to maxi-prep (Qiagen) to secure a large amount of plasmid DNA, and then introduced into the cell as follows. In the case of the production of the monovalent bispecific antibody, hole type heavy chain expression vector DNA, knob type heavy chain expression vector DNA, and light chain expression vector DNA were transduced at the ratio of 0.5 0.5:1, and in the case of the production of the monospecific antibody or the bivalent bispecific antibody, heavy chain expression vector DNA and light chain expression vector DNA were transduced at the ratio of 1:1.

[0135] The day before transfection, ExpiCHO™ (Gibco, Cat: A29127) cells were spread at the concentration of 3×10E6 to 4×10E6 viable cells/mL in ExpiCHO™ expression medium (Gibco, Cat: A29100-01), and then, cultured under the conditions of 8% CO.sub.2, 37° C., and 120 rpm for 1 day. On the day of DNA transfection, cells that had grown to have an amount of 7×10E6 to 10×10E6 viable cells/mL and the viability of 95% or more, were diluted using fresh medium to an amount of 6×10.sup.6 viable cells/mL.

[0136] For transfection into the prepared parental cells, ExpiFectamine™ CHO & plasmid DNA complex was prepared using the ExpiFectamine™ CHO transfection kit (Gibco, Cat: A29129). After dispensing cold OptiPRO™ SFM® (Gibco, Cat: 12309019) medium, DNA and ExpiFectamine™ CHO reagent prepared at appropriate concentrations were inoculated and then mixed and leave at room temperature for 5 minutes, and inoculated into parental cells to perform transfection, followed by culturing. The day after transfection, the enhancer included in the ExpiFectamine™ CHO transfection kit and feed were inoculated into the transfected cells, and 5 days later, the feed was additionally inoculated, followed by 10 days of culturing at the conditions of 8% CO.sup.2, 37° C., and 120 rpm.

[0137] (3) Culture Medium Harvest

[0138] In order to obtain a complete produced culture fluid, the culture fluid was transferred to a centrifugation bottle and centrifuged at 4° C., 6500 rpm for 30 minutes, followed by filtering with a filter having a size of 0.2 μm, and the culture fluid after the suspended matter was removed therefrom was obtained as a sample. Thereafter, the purification process was performed.

EXAMPLE 2: EXPERIMENTAL PREPARATION FOR CONFIRMATION OF PROCESS CONDITIONS FOR PURIFICATION OF MONOVALENT BISPECIFIC ANTIBODIES BY AFFINITY CHROMATOGRAPHY

[0139] A monovalent bispecific antibody was produced according to the method of Example 1. At this time, regarding the bivalent bispecific antibody, ch11F11 or hu11F11 clone was used as an anti-α-synuclein antibody, and 1564, VHS, VH9, VH16, or F06 clones was used as an anti-IGF-1R antibody, and both of which correspond to a VH3-family antibodies. Thus, each heavy chain variable region contained a VH3 domain.

[0140] In actual production, the target antibody, a monovalent bispecific antibody, is mainly produced by the knob-in-hole, and a bivalent bispecific antibody and a monospecific antibody, which are secondary products, are produced in relatively small amounts. However, in the present experiment, in order to more clearly prove the fact that the purification process according to the present disclosure can separate antibodies depending on the difference in affinity according to the number of VH3 domains, a composition with increased concentrations of the bivalent bispecific antibody and the monospecific antibody, which are considered as a by-product in the purification process, that is, an impurity was prepared (that is, by providing more unfavorable conditions for purification), and the purification process was performed on this composition.

[0141] Such a composition contains a monospecific antibody, a monovalent bispecific antibody, and a bivalent bispecific antibody at the ratio of 1:3:1 which are prepared in the same manner as in Example 1, except that a knob or a hole is not included. Such a composition includes antibodies isolated by a purification process as follows:

[0142] The affinity chromatography was performed using MabSelectSure (GE Healthcare, Cat. No. 17543803) having only the B (Z) domain as a resin to isolate purified substances having Fc.

[0143] After equilibration using an equilibration buffer (50 mM Tris-HCl pH 7.2, 100 mM NaCl), the recovered culture fluid was loaded into the column.

[0144] Once loading was completed, washing was performed using 3 column volume of an equilibration buffer, and then, elution was performed using 50 mM sodium citrate pH 3.4. The eluate was neutralized to pH 7.0 by adding 1 M Tris-HCl pH 9.0, followed by a 0.2 μm sterilization filter.

EXAMPLE 3: COMPARISON OF ANTIBODY PURIFICATION EFFECTS USING VARIOUS PROTEIN A RESINS

[0145] In order to select a protein A resin suitable for purification of a monovalent bispecific antibody, a sample including a ch11F11 monospecific antibody, a ch11F11-F06 monovalent bispecific antibody, and ch11F11-F06 bivalent bispecific antibody was purified using a commercially available protein A column having various specifications and domains as shown in Table 3 below, and the resulting effects were compared.

TABLE-US-00003 TABLE 3 Hitrap Hitrap POROS Absolute MabSelect MabSelect protein MabSelect protein MabCapture High Resin Sure PrismA A FF Xtra A HP A Select Cap Particle 85 60 90 75 34 45 35 Size (μm) Protein B(Z) B(Z) full full full full full A domain domain domain domain domain domain

[0146] The purification method using each resin was described in detail in Examples 3-1 to 3-7.

[0147] The purified sample was analyzed by size exclusion-high performance liquid chromatography (SE-HPLC), sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE), or liquid chromatography mass spectrometer (LC/MS) to confirm the purification results.

[0148] For SE-HPLC analysis, 40 mM sodium phosphate and 400 mM sodium perchlorate at pH 6.8 were used as a mobile phase, and an analysis sample in a volume of 20 μL was injected at an analysis sample concentration of 1 mg/mL, and the analysis was performed at a flow rate of 0.8 mL/min. Analytical peaks were determined by absorbance at a wavelength of 280 nm. For SDS-PAGE, NuPAGE 4-12% Bis-Tris gel and MOPS running buffer were used for analysis under a non-reducing condition and a reducing condition. In the case of the reducing condition, 5% 2-mercaptoethanol was added to the analysis sample, and the analysis was performed after heating at 70° C. for 10 minutes. For LC/MS analysis, a Waters Acquity Ultra High Performance Liquid Chromatography (UPLC) system and a Waters Synapt G2-S Quadrupole-Time of Flight (Q-TOF) mass spectrometer were used, and the analysis was performed using a PLRP-S 1000 Å BEH S-200 column. Data were analyzed using Waters BiopharmaLynx V.1.3. software.

EXAMPLE 3-1: MABSELECT SURE

[0149] The resolution of the sample was confirmed using HiTrap MabSelect Sure (GE Healthcare, Cat. No. 29-0486-84) resin. After equilibration using an equilibration buffer (25 mM citrate/25 mM sodium phosphate, pH 7.0), the sample prepared in Example 1 was loaded into the column. An elution buffer (25 mM citrate/25 mM sodium phosphate, pH 2.5) was used, and a residence time was 5 minutes, and wash conditions included 0%>50%/1 column volume of elution buffer and the range of pH of 7.0 to 5.0. The gradient condition was 50%>100%/30 column volume of elution buffer.

[0150] FIG. 1 shows a purification profile of affinity chromatography according to Example 3-1. In FIG. 1, the x-axis is the sample volume and the unit thereof is mL, and the y-axis is the ultra violet (UV) wavelength at 280 nm and the unit thereof is mAu. As shown in FIG. 1, Example 3-1 had only one elution peak. From this, it was confirmed that HiTrap MabSelect Sure resin showed no affinity difference according to the difference in the number of VH3 domains of the peptides present in the sample.

EXAMPLE 3-2: MABSELECT PRISMA

[0151] The resolution of the sample was confirmed using MabSelect PrismA (GE Healthcare, Cat. No. 17-5199-01). After equilibration using an equilibration buffer (25 mM citrate/25 mM sodium phosphate, pH 7.0), the sample prepared in Example 1 was loaded into the column. An elution buffer (25 mM citrate/25 mM sodium phosphate, pH 2.5) was used, and a residence time was 5 minutes, and wash conditions included 0%>50%/1 column volume elution buffer and the pH of 7.0 to 5.0. The gradient condition was 50%>100%/30 column volume of elution buffer.

[0152] FIG. 2 shows a purification profile of affinity chromatography according to Example 3-2. As shown in FIG. 2, Example 3-2 had only one elution peak. From this, it was confirmed that MabSelect PrismA resin showed no affinity difference according to the difference in the number of VH3 domains of the peptides present in the sample.

EXAMPLE 3-3: PROTEIN A FF

[0153] The resolution of the sample was confirmed using HiTrap protein A FF (GE Healthcare, Cat. No. 17-5079-01). After equilibration using an equilibration buffer (25 mM citrate/25 mM sodium phosphate, pH 7.0), the sample prepared in Example 1 was loaded into the column. An elution buffer (25 mM citrate/25 mM sodium phosphate, pH 2.5) was used, and a residence time was 5 minutes, and wash conditions included 0%>50%/1 column volume elution buffer and the pH of 7.0 to 5.0. The gradient condition was 50%>100%/30 column volume of elution buffer.

[0154] FIG. 3 shows a purification profile of affinity chromatography according to Example 3-3. As shown in FIG. 3, peaks of Example 3-3 were not clearly separated. From this, it was confirmed that HiTrap protein A FF resin showed no significant affinity difference according to the difference in the number of VH3 domains of the peptides present in the sample.

EXAMPLE 3-4: MABSELECT XTRA

[0155] The resolution of the sample was confirmed using MabSelect Xtra (GE Healthcare, Cat. No. 17-5269-07). After equilibration using an equilibration buffer (25 mM citrate/25 mM sodium phosphate, pH 7.0), the sample prepared in Example 1 was loaded into the column. An elution buffer (25 mM citrate/25 mM sodium phosphate, pH 2.5) was used, and a residence time was 5 minutes, and wash conditions included 0%>50%/1 column volume elution buffer and the pH of 7.0 to 5.0. The gradient condition was 50%>100%/30 column volume of elution buffer.

[0156] FIG. 4 shows the purification profile of affinity chromatography according to Example 3-4, and SE-HPLC and SDS-PAGE results are shown in FIGS. 5 and 6, respectively. As confirmed in FIG. 4, three peaks were identified in the pH range of the elution buffer. Specifically, at a relatively high pH, a monospecific antibody with a small number (two) of VH3 domains was eluted first (Peak 1), and as the pH was lowered, a monovalent bispecific antibody having three VH3 domains was eluted, and then, at further lower pH, the bivalent bispecific antibody with the largest number (four) of VH3 domains was eluted last. From this, it was confirmed that the MabSelect Xtra resin showed a difference in affinity according to the difference in the number of VH3 domains of the peptides present in the sample, and that ch11F11-F06 could be selectively purified according to the difference in affinity. Other clone combinations showed these results as well.

[0157] In addition, through the SE-HPLC and SDS-PAGE results according to FIGS. 5 and 6, it was reconfirmed that the peptides separated by affinity chromatography corresponded to the monospecific antibody, the monovalent bispecific antibody, and the bivalent bispecific antibody, respectively.

EXAMPLE 3-5: PROTEIN A HP

[0158] The resolution of the sample was confirmed using HiTrap protein A HP (GE Healthcare, Cat. No. 17-0402-01). After equilibration using an equilibration buffer (25 mM citrate/25 mM sodium phosphate, pH 7.0), the sample prepared in Example 1 was loaded into the column. An elution buffer (25 mM citrate/25 mM sodium phosphate, pH 2.5) was used, and a residence time was 5 minutes, and wash conditions included 0%>50%/1 column volume elution buffer and the pH of 7.0 to 5.0. The gradient condition was 50%>100%/30 column volume of elution buffer.

[0159] FIG. 7 shows the purification profile of affinity chromatography according to Example 3-5, and SE-HPLC and SDS-PAGE results are shown in FIGS. 8 and 9, respectively. As confirmed in FIG. 7, three peaks were identified in the pH range of the elution buffer. Specifically, at a relatively high pH, a monospecific antibody with a small number (two) of VH3 domains was eluted first (Peak 1), and as the pH was lowered, a monovalent bispecific antibody having three VH3 domains was eluted, and then, at further lower pH, the bivalent bispecific antibody with the largest number (four) of VH3 domains was eluted last. From this, it was confirmed that the HiTrap protein A HP resin showed a difference in affinity according to the difference in the number of VH3 domains of the peptides present in the sample, and that ch11F11-F06 could be selectively purified according to the difference in affinity. Other clone combinations showed these results as well.

[0160] In addition, through the SE-HPLC and SDS-PAGE results according to FIGS. 8 and 9, it was reconfirmed that the peptides separated by affinity chromatography corresponded to the monospecific antibody, the monovalent bispecific antibody, and the bivalent bispecific antibody, respectively.

EXAMPLE 3-6: POROS MABCAPTURE A SELECT

[0161] The resolution of the sample was confirmed using POROS MabCapture A Select (Thermo Fisher SCIENTIFIC, Cat. No. A26458). After equilibration using an equilibration buffer (25 mM citrate/25 mM sodium phosphate, pH 7.0), the sample prepared in Example 1 was loaded into the column. An elution buffer (25 mM citrate/25 mM sodium phosphate, pH 2.5) was used, and a residence time was 5 minutes, and wash conditions included 0%>50%/1 column volume elution buffer and the pH of 7.0 to 5.0. The gradient condition was 50%>100%/30 column volume of elution buffer.

[0162] FIG. 10 shows the purification profile of affinity chromatography according to Example 3-6, and SE-HPLC and SDS-PAGE results are shown in FIGS. 11 and 12, respectively. As confirmed in FIG. 10, three peaks were identified in the pH range of the elution buffer. Specifically, at a relatively high pH, a monospecific antibody with a small number (two) of VH3 domains was eluted first (Peak 1), and as the pH was lowered, a monovalent bispecific antibody having three VH3 domains was eluted, and then, at further lower pH, the bivalent bispecific antibody with the largest number (four) of VH3 domains was eluted last. From this, it was confirmed that the POROS MabCapture A Sele resin showed a difference in affinity according to the difference in the number of VH3 domains of the peptides present in the sample, and that ch11F11-F06 could be selectively purified according to the difference in affinity. Other clone combinations showed these results as well.

[0163] In addition, through the SE-HPLC and SDS-PAGE results according to FIGS. 11 and 12, it was reconfirmed that the peptides separated by affinity chromatography corresponded to the monospecific antibody, the monovalent bispecific antibody, and the bivalent bispecific antibody, respectively.

EXAMPLE 3-7: ABSOLUTE HIGH CAP

[0164] The resolution of the sample was confirmed using AbSolute High Cap (AGC). After equilibration using an equilibration buffer (25 mM citrate/25 mM sodium phosphate, pH 7.0), the sample prepared in Example 1 was loaded into the column. An elution buffer (25 mM citrate/25 mM sodium phosphate, pH 2.5) was used, and a residence time was 5 minutes, and wash conditions included 0%>50%/1 column volume elution buffer and the pH of 7.0 to 5.0. The gradient condition was 50%>100%/30 column volume of elution buffer.

[0165] FIG. 13 shows the purification profile of affinity chromatography according to Example 3-6, and SE-HPLC and SDS-PAGE results are shown in FIGS. 14 and 15, respectively. As confirmed in FIG. 13, three peaks were identified in the pH range of the elution buffer. Specifically, at a relatively high pH, a monospecific antibody with a small number (two) of VH3 domains was eluted first (Peak 1), and as the pH was lowered, a monovalent bispecific antibody having three VH3 domains was eluted, and then, at further lower pH, the bivalent bispecific antibody with the largest number (four) of VH3 domains was eluted last. From this, it was confirmed that the AbSolute High Cap resin showed a difference in affinity according to the difference in the number of VH3 domains of the peptides present in the sample, and that ch11F11-F06 could be selectively purified according to the difference in affinity. Other clone combinations showed these results as well.

[0166] In addition, through the SE-HPLC and SDS-PAGE results according to FIGS. 14 and 15, it was reconfirmed that the peptides separated by affinity chromatography corresponded to the monospecific antibody, the monovalent bispecific antibody, and the bivalent bispecific antibody, respectively.

SUMMARY OF PURIFICATION RESULTS OF EXAMPLES 3-1 TO 3-7

[0167] The pH concentration gradients of the elution buffers at which peaks of Examples 3-4 to 3-7 appeared are as shown in Table 4.

TABLE-US-00004 TABLE 4 Peak 1 Peak 1 Peak 2 Peak 3 Example Start pH End pH End pH End pH Example 3-4 3.86 3.58 3.39 3.26 Example 3-5 3.99 3.77 3.54 3.36 Example 3-6 3.97 3.73 3.53 3.36 Example 3-7 3.85 3.59 3.38 3.21

[0168] The purification results of Examples 3-1 to 3-7 are summarized and described in detail in Table 5.

TABLE-US-00005 TABLE 5 Hitrap Hitrap POROS Absolute MabSelect MabSelect Protein MabSelect Protein A MabCapture High Resin Sure PrismA A FF Xtra HP A Select Cap Resolution* I I I II II III IV [I-IV] (1.08/0.62) (1.01/0.91) (1.30/1.08) (1.53/1.21) (Peak 1-2/ Peak 2-3) Alkaline IV IV II II I III IV stability [I-IV] SE-HPLC Not Not Not 85.7 94.9 95.2 97.7 Purity (%) applicable applicable applicable Recovery 99 100 96 99 (%)** Monovalent 73.3 79.3 76.5 79.2 Yield (%)*** Scalability Yes Yes Yes Yes No Yes Yes

[0169] The resolution (Rs), recovery yield, and monovalent antibody yield of each example were calculated as follows.

[00001]$\mathrm{Rs}=1.18\frac{\left(t_{R_2}-t_{R_1}\right)}{\left(W_{1/2_2}+W_{1/2_1}\right)}*$

[0170] tR1: retention time of peak 1.

[0171] tR2: retention time of peak 2

[0172] W1/2.sub.1: Width at half height of peak 1

[0173] W1/2.sub.2: Width at half height of peak 2

Recovery (%)=Output (mg)/Input(mg)×100   **

Monovalent Yield (%)=Monovalent Output (mg)×Purity (%)/Monovalent Input (mg)×Purity (%)×100   ***

[0174] The experiments showed that only the columns used in Examples 3-4 to 3-7 showed affinity differences according to the number of VH3 domains included in the peptide. As a result, it was confirmed that the monospecific antibody, the monovalent bispecific antibody, and the bivalent bispecific antibody were effectively separated according to the pH concentration gradient. Specifically, the antibodies isolated in Examples 3-4 to 3-7 have 2, 3, and 4 VH3 domains which vary depending on the presence and number of anti-IGF1R scFv connected to the C-terminus of Fc of the monospecific antibody having a VH3 domain in a variable region thereof, and in the case of the columns used in Examples 3-4 to 3-7, since the protein A resin has a full domain, the difference in the number of VH3 domains of the antibodies is recognized precisely and thus antibodies were effectively separated. Compared to improved columns such as MabSelect Sure and MabSelect PrismA, which are generally more preferred due to the use of specific domain B(Z) alone to improve Fc binding affinity and alkaline stability, according to the present disclosure, a resin containing protein A of the related art (a special mutation is not introduced) is effective for the separation of peptides according to the number of VH3 domains from a peptide mixture containing a VH3 domain.

EXAMPLE 4: ANALYSIS OF ANTIBODY PURITY FROM CULTURE FLUID

[0175] Using POROS A 20 μm (Thermo fisher, Cat. No. 1-5022-26), it was confirmed whether the target Fc-containing bioactive peptide could be analyzed even when the culture fluid was directly used without the purification of protein A. After equilibration using an equilibration buffer (25 mM citrate/25 mM sodium phosphate pH 7.0), the culture fluid (Harvested Cell Culture Fluid) was directly loaded. Unbound wash was performed using 2 column volume of equilibration buffer, an elution buffer (25 mM citrate/25 mM sodium phosphate pH 2.5) was used, the residence time was 1 minute, and the gradient condition was used where the elution buffer was 50%>100%/30 column volume. The same method was used to analyze a sample which had been pre-purified with protein A, and the obtained results was compared.

[0176] FIG. 16 shows an analysis profile of affinity chromatography according to Example 4. As shown in FIG. 16, it was confirmed that the results of directly analyzing the culture fluid without going through the protein A column and the results of analyzing the sample pre-purified with protein A were very similar to each other as 90.3% and 90.5%, respectively, and the elution profile and the resolution were also similar to each other.

[0177] From these results, it was confirmed that, according to the separation method of an embodiment, peptides can be separated and analyzed with excellent resolution even without primary purification of the transient culture with protein A. Therefore, according to the separation method according to an embodiment, even when the number of candidate cell lines is large in the initial stage of cell line development, the culture fluid can be directly used to identify the ratio and purity of an antibody without the primary purification process.

EXAMPLE 5: PURIFICATION OF VARIOUS ASYMMETRIC POLYPEPTIDES

EXAMPLE 5-1: IGG(VH3(+))−SCFV(VH3+)

[0178] The resolution of the sample was confirmed using the protein A resin used in Example 3. All protein A resins having the confirmed resolution shown in Table 5 may be used as the protein A resin. As an example, POROS MabCapture A Select (Thremo Fisher SCIENTIFIC, Cat. No. A26458) was used for purification. As a sample for purification, a mixture including a monospecific antibody having two VH3 domains (hu11F11) and a monovalent bispecific antibody having three VH3 domains (hu11F11-F06) generated according to the method of Example 1 was used.

[0179] After equilibration using an equilibration buffer (25 mM citrate/25 mM sodium phosphate, pH 7.0), the sample was loaded into the column. An elution buffer (25 mM citrate/25 mM sodium phosphate, pH 2.5) was used, and a residence time was 6 minutes, and wash conditions included 0%>50%/1 column volume elution buffer and the pH of 7.0 to 5.0. The gradient condition was 50%>100%/30 column volume of elution buffer. FIG. 17 shows the purification profile of affinity chromatography, and SE-HPLC and SDS-PAGE results are shown in FIGS. 18 and 19, respectively.

[0180] As a result of applying the pH gradient of the elution buffer, two peaks appeared as shown in FIG. 17. Specifically, at a relatively high pH, a monospecific antibody with a small number of VH3 domains (two) was eluted first (Peak 1), and as the pH was lowered, a monovalent bispecific antibody having three VH3 domains was eluted (Peak 2). These results were reconfirmed from the SE-HPLC and SDS-PAGE results shown in FIGS. 18 and 19.

EXAMPLE 5-2. IGG(VH3(−))−SCFV(VH3+)

[0181] The resolution of the sample was confirmed using the protein A resin used in Example 3. All protein A resins having the confirmed resolution shown in Table 5 may be used as the protein A resin. As an example, POROS MabCapture A Select (Thremo Fisher SCIENTIFIC, Cat. No. A26458) was used for purification. As a sample for purification, a mixture including a monospecific antibody having no VH3 domain (ch1E4 or anti-BACE) and a monovalent bispecific antibody having one VH3 domain (ch1E4-F06 or anti-BACE-F06), both of which are prepared by the method of Example 1 except using anti-BACE antibody that does not contain VH3 domain in the variable region of IgG, was used.

[0182] After equilibration using an equilibration buffer (25 mM citrate/25 mM sodium phosphate, pH 7.0), the sample was loaded into the column. An elution buffer (25 mM citrate/25 mM sodium phosphate, pH 2.5) was used, and a residence time was 6 minutes, and wash conditions included 0%>50%/1 column volume elution buffer and the pH of 7.0 to 5.0. The gradient condition was 50%>100%/30 column volume of elution buffer. The elution peaks of affinity chromatography are shown in FIGS. 20 and 21, the SE-HPLC results are shown in FIGS. 22 and 23, and the SDS-PAGE results are shown in FIGS. 24 and 25.

[0183] As a result of applying the pH gradient of the elution buffer, two peaks appeared as shown in FIGS. 20 and 21. Specifically, at a relatively high pH, a monospecific antibody having no VH3 domains (zero) was eluted first (Peak 1), and as the pH was lowered, a monovalent bispecific antibody having one VH3 domains was eluted (Peak 2). These results were reconfirmed from the SE-HPLC and SDS-PAGE results shown in FIGS. 22 to 25.

EXAMPLe 5-3. FC−SCFV (VH3+)

[0184] The resolution of the sample was confirmed using the protein A resin used in Example 3. All protein A resins having the confirmed resolution shown in Table 5 may be used as the protein A resin. As an example, POROS MabCapture A Select (Thremo Fisher SCIENTIFIC, Cat. No. A26458) was used for purification. As a sample for purification, Fc and Fc-scFv (F06) from which Fab had been removed by treating a monospecific antibody (hu11F11) and a monovalent bispecific antibody (hu11F11-F06) with papain were prepared. Specifically, papain was diluted with digestion buffer (10 mM PBS, 20 mM EDTA, 10 mM cysteine-HCl, and pH 7.4), and mixed with the prepared sample at the ratio of 1:100 (papain:antibody), and the mixture was caused to react at a temperature of 37° C. for 4 hours.

[0185] After equilibration using an equilibration buffer (25 mM citrate/25 mM sodium phosphate, pH 7.0), the sample was loaded into the column. An elution buffer (25 mM citrate/25 mM sodium phosphate, pH 2.5) was used, and a residence time was 6 minutes, and wash conditions included 0%>50%/1 column volume elution buffer and the pH of 7.0 to 5.0. The gradient condition was 50%>100%/30 column volume of elution buffer. FIG. 26 shows the purification profile of affinity chromatography, and SE-HPLC and SDS-PAGE results are shown in FIGS. 27 and 28, respectively.

[0186] As a result of applying the pH gradient of the elution buffer, two peaks appeared as shown in FIG. 26. A typical antibody has Fab connected to the N-terminus of Fc, whereas the sample separated in this example does not contain Fab at the N-terminus of Fc. Nevertheless, according to FIG. 26, at a relatively high pH, Fc without a VH3 domain (zero) was eluted first (Peak 1), and as the pH decreased, Fc-scFv (F06) having one VH3 domain was eluted (Peak 2). These results were reconfirmed from the SE-HPLC and SDS-PAGE results shown in FIGS. 27 and 28. This result suggests that the purification method according to the present disclosure is used not only for the purification of an antibody, but also for the purification of various Fc-containing peptides having a difference in the number of VH3 domains, for example, various peptides containing Fc-scFv (for example, Fc-scFv connected to a drug or a functional peptide).

EXAMPLE 5-4. GLP−1−FC−SCFV (VH3+)

[0187] The resolution of the sample was confirmed using the protein A resin used in Example 3. All protein A resins having the confirmed resolution shown in Table 5 may be used as the protein A resin. As an example, POROS MabCapture A Select (Thremo Fisher SCIENTIFIC, Cat. No. A26458) was used for purification. As a sample for purification, a mixture including GLP-1-Fc (F06) and GLP-1-Fc-scFv (F06) having one VH3 domain was prepared in the same manner as used in Example 1, except that instead of Fab, the known sequence of GLP-1 (Glucagon-like Peptide-1) was connected to Fc. GLP-1 used in this example is a kind of peptide drug, and research and clinical studies thereinto are being actively conducted in relation to diseases such as diabetes and Alzheimer's. In the present example, it was tested whether separation according to the number of VH3 domains can be made even when such a peptide drug is included.

[0188] After equilibration using an equilibration buffer (25 mM citrate/25 mM sodium phosphate, pH 7.0), the sample was loaded into the column. An elution buffer (25 mM citrate/25 mM sodium phosphate, pH 2.5) was used, and a residence time was 6 minutes, and wash conditions included 0%>50%/1 column volume elution buffer and the pH of 7.0 to 5.0. The gradient condition was 50%>100%/30 column volume of elution buffer. FIG. 29 shows the purification profile of affinity chromatography, and FIG. 30 shows SE-HPLC results.

[0189] As a result of applying the pH gradient of the elution buffer, two peaks appeared as shown in FIG. 29. A typical antibody has Fab connected to the N-terminus of Fc, whereas the sample separated in this example is a fusion protein and contains GLP-1, instead of Fab, at the N-terminus of Fc. Nevertheless, according to FIG. 29, at a relatively high pH, GLP-1-Fc without a VH3 domain (zero) was eluted first (Peak 1), and as the pH decreased, GLP-1-Fc-scFv (F06) having one VH3 domain was eluted (Peak 2). These results were reconfirmed from the SE-HPLC results shown in FIG. 30. This result suggests that the purification method according to the present disclosure is used not only for the purification of an antibody, but also for the separation of various Fc-containing peptides having a difference in the number of VH3 domains, for example, bioactive peptides including, in addition to the components of an antibody, other active ingredients.

[0190] The pH concentration gradient of the elution buffer in Examples 5-1 to 5-4 is as shown in Table 6.

TABLE-US-00006 TABLE 6 Peak 1 Peak 1 Peak 2 Peak 2 Example start pH End pH start pH End pH Example 5-1 3.51 3.41 3.41 3.32 (hu11F11 & hu11F11-scFv) Example 5-2 4.23 4.08 3.96 3.76 (ch1E4 & ch1E4-scFv) Example 5-2 4.05 3.89 3.73 3.57 (anti-BACE & Anti-BACE-scFv) Example 5-3 (Fc & 3.98 3.78 3.78 3.53 Fc-scFv) Example 5-4 (GLP- 4.17 3.95 3.83 3.61 1-Fc & GLP-1-Fc- scFv)

EXAMPLE 5-5. IGG(VH3(±))−SCFV(VH3+)

[0191] The resolution of the sample was confirmed using the protein A resin used in Example 3. All protein A resins having the confirmed resolution shown in Table 5 may be used as the protein A resin. As an example, POROS MabCapture A Select (Thremo Fisher SCIENTIFIC, Cat. No. A26458) was used for purification. As a sample for the purification, a mixture comprising 4 types of antibodies, including monospecific antibody having two VH3 domains (hu11F11) and monovalent bispecific antibody having three VH3 domains (hu11F11-F06), both of which are prepared by the method of Example 1, and monospecific antibody having no VH3 domain (anti-BACE) and monovalent bispecific antibody having one VH3 domain (anti-BACE-F06), both of which are prepared by the same method of Example 1 except using anti-BACE antibody that does not contain VH3 domain in the variable region of IgG, was used.

[0192] After equilibration using an equilibration buffer (25 mM citrate/25 mM sodium phosphate, pH 7.0), the sample was loaded into the column. An elution buffer (25 mM citrate/25 mM sodium phosphate, pH 2.5) was used, and a residence time was 6 minutes, and wash conditions included 0%>50%/1 column volume elution buffer and the pH of 7.0 to 5.0. The gradient condition was 50%>100%/30 column volume of elution buffer. FIG. 31 shows the purification profile of affinity chromatography, and SE-HPLC, SDS-PAGE, and LC/MS results are shown in FIGS. 32 to 34, respectively.

[0193] As a result of applying the pH gradient of the elution buffer, four peaks appeared as shown in FIG. 31, and the peaks of the peptides (anti-BACE antibody: 0, anti-BACE-F06: 1, The peaks of hu11F11 antibody: 2, hu11F11-F06: 3) were clearly separated, confirming that a target peptide could be selectively purified. In addition, these results were reconfirmed from the SE-HPLC, SDS-PAGE and LC/MS results of FIGS. 32 to 34. From this, it was confirmed that the separation method of the present application can purify peptides very precisely according to the number of VH3 domains included in each peptide.

EXAMPLE 5-6. Igg(VH3(±))−SCFV(VH3+)

[0194] The resolution of the sample was confirmed using the protein A resin used in Example 3. All protein A resins having the confirmed resolution shown in Table 5 may be used as the protein A resin. As an example, POROS MabCapture A Select (Thremo Fisher SCIENTIFIC, Cat. No. A26458) was used for purification. As a sample for the purification, a mixture comprising 4 types of antibodies, including monospecific antibody having two VH3 domains (hu11F11) and monovalent bispecific antibody having three VH3 domains (hu11F11-F06), both of which are prepared by the method of Example 1, and monospecific antibody having no VH3 domain (ch1E4) and monovalent bispecific antibody having one VH3 domain (ch1E4-F06), both of which are prepared by the same method of Example 1 except using ch1E4 antibody that does not contain VH3 domain in the variable region of IgG, was used.

[0195] After equilibration using an equilibration buffer (25 mM citrate/25 mM sodium phosphate, pH 7.0), the sample was loaded into the column. An elution buffer (25 mM citrate/25 mM sodium phosphate, pH 2.5) was used, and a residence time was 6 minutes, and wash conditions included 0%>50%/1 column volume elution buffer and the pH of 7.0 to 5.0. The gradient condition was 50%>100%/30 column volume of elution buffer. FIG. 35 shows the purification profile of affinity chromatography.

[0196] As a result of applying the pH gradient of the elution buffer, four peaks appeared as shown in FIG. 35, and the peaks of the peptides (ch1E4: 0, ch1E4-F06: 1; hu11F11: 2, hu11F11-F06: 3) were clearly separated, confirming that a target peptide could be selectively purified according to the number of VH3 domains.