Patent Application
20230272445
Incorporated by reference in its entirety is a computer-readable nucleotide/amino acid sequence listing submitted concurrently herewith and identified as follows: ASCII (text) file named “55685A_Seqlisting.XML,” 88,169 bytes, created on Nov. 14, 2022.
The present invention relates to the finding of methods to shift the glycosylation profile of recombinant produced serum glycoproteins to the predominant bi-antennary form found in human plasma. This is accomplished by providing a mammalian cell line according to the invention with a series of knock outs and/or knock in's that facilitate this shift.
Recombinant glycoproteins and in particular human serum proteins, such as the glycoproteins from the family of serpins are produced for a range of applications. This included Alpha-1-antitrypsin (AAT), Plasma protease C1 inhibitor (C1Inh), Antithrombin-III (ATIII), Monocyte neutrophil elastase inhibitor (Serpin B1), Plasminogen activator inhibitor I (PAI1) that are produced for therapeutic applications in humans.
Alpha-1-antitrypsin (AAT) is used for treatment of people with AAT-deficiency. Such deficiency may result in lethal lung disease and liver disease. Over one million people have been estimated to be deficient of AAT globally. Currently, AAT is purified from human plasma.
This treatment regimen is both expensive (USD 52,000 per year per patient), and not optimal with regards to safety, as possible pathogens present in plasma may not be efficiently cleared.
Many approaches have been pursued to produce recombinant human AAT. Efforts of producing AAT in non-mammalian cells such as E.coli, yeast and plants have resulted in either non-glycosylated AAT or non-human glycosylation patterns. Production of AAT in transgenic animals such as sheep has also been reported. However, an immune response to endogenous (sheep) AAT in the purified product was later observed. This clearly demonstrates one of the major challenges that transgenic animal-derived therapeutics is facing. Finally, AAT has also been produced in CHO and human cells with an aberrant glycoprofile.
Thus, there is a need in the art for glycoproteins with a more native human glycoprofile.
It is an object of embodiments of the invention to provide methods and tools for producing recombinant proteins with a glycan profile found naturally in humans.
It has been found by the present inventor(s) that by specific modification of a mammalian host cell with the downregulation or inactivation of a series of genes in combination with the insertion of other specific gene(s), this mammalian host cell is made into a cell that will produce glycoproteins with a glycosylation profile that more resemble the glycosylation profile found for the same glycoproteins naturally in humans, such as in human plasma.
So, in a first aspect the present invention relates to a recombinant mammalian cell line having a) one or more of the endogenous genes Mgat4A, Mgat4B, Mgat5, St3Gal3, St3Gal4, St3Gal6, SPPL3, and FUT8 inactivated and/or downregulated; and b) optionally a gene encoding Beta-galactoside alpha-2,6-sialyltransferase 1 inserted.
In some embodiments the endogenous gene Mgat4A is inactivated and/or downregulated.
In some embodiments the endogenous gene Mgat4B is inactivated and/or downregulated.
In some embodiments the endogenous gene Mgat5 is inactivated and/or downregulated.
In some embodiments the endogenous gene St3Gal3 is inactivated and/or downregulated.
In some embodiments the endogenous gene St3Gal4 is inactivated and/or downregulated.
In some embodiments the endogenous gene St3Gal6 is inactivated and/or downregulated.
In some embodiments the endogenous gene SPPL3 inactivated and/or downregulated.
In some embodiments the endogenous gene B3GNT2 is inactivated and/or downregulated.
In some embodiments the endogenous gene GLUL is inactivated and/or downregulated.
In some embodiments the endogenous gene FUT8 is inactivated and/or downregulated.
In some embodiments, two, three, four, five, six, seven, or all eight of the endogenous genes Mgat4A, Mgat4B, Mgat5, St3Gal3, St3Gal4, St3Gal6, SPPL3, and FUT8 are inactivated and/or downregulated.
In some embodiments, two, three, four, five, six, seven, eight, or all nine of the endogenous genes Mgat4A, Mgat4B, Mgat5, St3Gal3, St3Gal4, St3Gal6, SPPL3, B3GNT2, and FUT8 are inactivated and/or downregulated.
In some embodiments, two, three, four, five, six, seven, eight, or all nine of the endogenous genes Mgat4A, Mgat4B, Mgat5, St3Gal3, St3Gal4, St3Gal6, SPPL3, GLUL and FUT8 are inactivated and/or downregulated.
In some embodiments, two, three, four, five, six, seven, eight, nine, or all ten of the endogenous genes Mgat4A, Mgat4B, Mgat5, St3Gal3, St3Gal4, St3Gal6, SPPL3, B3GNT2, GLUL and FUT8 are inactivated and/or downregulated.
In some embodiments two genes selected from Mgat4A and Mgat4B;
In some embodiments three genes selected from
In some embodiments four genes selected from Mgat4A, Mgat4B, Mgat5, and St3Gal3;
In some embodiments five genes selected from Mgat4A, Mgat4B, Mgat5, St3Gal3, and St3Gal4; Mgat4A, Mgat4B, Mgat5, St3Gal3, and St3Gal6; Mgat4A, Mgat4B, Mgat5, St3Gal3, and SPPL3; Mgat4A, Mgat4B, Mgat5, St3Gal3, and FUT8; Mgat4A, Mgat4B, Mgat5, St3Gal4, and St3Gal6; Mgat4A, Mgat4B, Mgat5, St3Gal4, and SPPL3; Mgat4A, Mgat4B, Mgat5, St3Gal4, and FUT8; Mgat4A, Mgat4B, Mgat5, St3Gal6, and SPPL3; Mgat4A, Mgat4B, Mgat5, St3Gal6, and FUT8; Mgat4A, Mgat4B, Mgat5, SPPL3, and FUT8; Mgat4A, Mgat5, St3Gal3, St3Gal4, and St3Gal6; Mgat4A, Mgat5, St3Gal3, St3Gal4, and SPPL3; Mgat4A, Mgat5, St3Gal3, St3Gal4, and FUT8; Mgat4A, Mgat5, St3Gal3, St3Gal6, and SPPL3; Mgat4A, Mgat5, St3Gal3, St3Gal6, and FUT8; Mgat4A, Mgat5, St3Gal3, SPPL3, and FUT8; Mgat4A, St3Gal3, St3Gal4, St3Gal6, and SPPL3; Mgat4A, St3Gal3, St3Gal4, St3Gal6, and FUT8; Mgat4A, St3Gal3, St3Gal4, SPPL3, and FUT8; Mgat4A, St3Gal4, St3Gal6, SPPL3, and FUT8;
GLUL, Mgat4B, Mgat5, St3Gal3, and St3Gal4; GLUL, Mgat4B, Mgat5, St3Gal3, and St3Gal6;
In some embodiments six genes selected from Mgat5, St3Gal3, St3Gal4, St3Gal6, SPPL3, and FUT8; Mgat4B, St3Gal3, St3Gal4, St3Gal6, SPPL3, and FUT8; Mgat4B, Mgat5, St3Gal4, St3Gal6, SPPL3, and FUT8; Mgat4B, Mgat5, St3Gal3, St3Gal6, SPPL3, and FUT8; Mgat4B, Mgat5, St3Gal3, St3Gal4, SPPL3, and FUT8; Mgat4B, Mgat5, St3Gal3, St3Gal4, St3Gal6, FUT8; Mgat4B, Mgat5, St3Gal3, St3Gal4, St3Gal6, SPPL3; Mgat4A, St3Gal3, St3Gal4, St3Gal6, SPPL3, and FUT8; Mgat4A, Mgat5, St3Gal4, St3Gal6, SPPL3, and FUT8; Mgat4A, Mgat5, St3Gal3, St3Gal6, SPPL3, and FUT8; Mgat4A, Mgat5, St3Gal3, St3Gal4, SPPL3, and FUT8; Mgat4A, Mgat5, St3Gal3, St3Gal4, St3Gal6, and FUT8; Mgat4A, Mgat5, St3Gal3, St3Gal4, St3Gal6, and SPPL3; Mgat4A, Mgat4B, St3Gal4, St3Gal6, SPPL3, and FUT8;
St3Gal6; GLUL, Mgat4A, Mgat4B, Mgat5, St3Gal3, and SPPL3; GLUL, Mgat4A, Mgat4B, Mgat5, St3Gal3, and FUT8; GLUL, Mgat4A, Mgat4B, Mgat5, St3Gal4, and St3Gal6; GLUL, Mgat4A, Mgat4B, Mgat5, St3Gal4, and SPPL3; GLUL, Mgat4A, Mgat4B, Mgat5, St3Gal4, and FUT8; GLUL, Mgat4A, Mgat4B, Mgat5, St3Gal6, and SPPL3; GLUL, Mgat4A, Mgat4B, Mgat5, St3Gal6, and FUT8; GLUL, Mgat4A, Mgat4B, Mgat5, SPPL3, and FUT8; GLUL, Mgat4A, Mgat4B, St3Gal3, St3Gal4, and St3Gal6; GLUL, Mgat4A, Mgat4B, St3Gal3, St3Gal4, and SPPL3; GLUL, Mgat4A, Mgat4B, St3Gal3, St3Gal4, and FUT8; GLUL, Mgat4A, Mgat4B, St3Gal3, St3Gal6, and SPPL3; GLUL, Mgat4A, Mgat4B, St3Gal3, St3Gal6, and FUT8; GLUL, Mgat4A, Mgat4B, St3Gal3, SPPL3, and FUT8; GLUL, Mgat4A, Mgat4B, St3Gal4, St3Gal6, and SPPL3; GLUL, Mgat4A, Mgat4B, St3Gal4, St3Gal6, and FUT8; GLUL, Mgat4A, Mgat4B, St3Gal4, SPPL3, and FUT8; GLUL, Mgat4A, Mgat4B, St3Gal6, SPPL3, and FUT8; GLUL, Mgat4A, Mgat4B, St3Gal6, SPPL3, and FUT8; GLUL, Mgat4A, Mgat5, St3Gal3, St3Gal4, and St3Gal6; GLUL, Mgat4A, Mgat5, St3Gal3, St3Gal4, and SPPL3; GLUL, Mgat4A, Mgat5, St3Gal3, St3Gal4, and FUT8; GLUL, Mgat4A, Mgat5, St3Gal3, St3Gal6, and SPPL3; GLUL, Mgat4A, Mgat5, St3Gal3, St3Gal6, and FUT8; GLUL, Mgat4A, Mgat5, St3Gal3, SPPL3, and FUT8; GLUL, Mgat4A, Mgat5, St3Gal4, St3Gal6, and SPPL3; GLUL, Mgat4A, Mgat5, St3Gal4, St3Gal6, and FUT8; GLUL, Mgat4A, Mgat5, St3Gal4, SPPL3, and FUT8; GLUL, Mgat4A, Mgat5, St3Gal6, SPPL3, and FUT8; GLUL, Mgat4A, St3Gal3, St3Gal4, St3Gal6, and SPPL3; St3Gal3, St3Gal4, St3Gal6, and FUT8; GLUL, Mgat4A, St3Gal3, St3Gal4, SPPL3, and FUT8; St3Gal3, St3Gal6, SPPL3, and FUT8; and GLUL, Mgat4A, St3Gal4, St3Gal6, SPPL3, and FUT8 are inactivated and/or downregulated.
In some embodiments seven genes selected from Mgat4A, Mgat4B, Mgat5, St3Gal3, St3Gal4, St3Gal6, and SPPL3: Mgat4A, Mgat4B, Mgat5, St3Gal3, St3Gal4, St3Gal6, and FUT8;
In some embodiments the gene encoding Beta-galactoside alpha-2,6-sialyltransferase 1 is inserted.
In a second aspect, the present invention relates to a method for the production of a recombinant protein of interest, the method comprising the steps of: a) culturing a population of recombinant mammalian cells according to any one of claims 4-8 in a suitable cell culture medium; and b) harvesting said human protein of interest from the cell culture or cell culture medium. In some embodiments the protein of interest is produced with a glycan structure similar or identical to the glycan profile of said glycoprotein of interest found in human plasma.
In a third aspect the present invention relates to a recombinant human glycoprotein of interest produced according to the method of the invention.
Glyco-analysis of human AAT from various sources revealed that plasma-purified AAT is glycosylated with a fully sialylated bi-antennary structure without core fucosylation (
The inventors of the present invention have found that a shift of the glycosylation profile of recombinant produced serum glycoproteins towards the predominant bi-antennary form found in human plasma, may be accomplished by knocking out or in any other way downregulating a selected a set of glycosylating enzymes. This will result in a 6, 7, 8, 9, 10 or 8-9 double knock out clone in which, glycoproteins, such as human serum proteins, such as human AAT are expressed. The following targets have been selected for this cell line:
With these modifications, it would be accomplished to shift in the glycosylation profile to the predominant bi-antennary form found in human plasma of recombinant produced serum glycoproteins, such as human serum proteins, such as human AAT, such as in CHO cells.
A host cell with these modifications may then be modified by insertion of a gene expressing an exogenous human glycoprotein of interest, such as a therapeutic human protein, such as a human serum protein, such as Plasma protease C1 inhibitor (C1Inh), Antithrombin-III (ATIII) or Human alpha-1-antitrypsin (AAT).
In some embodiments the mammalian cells used according to the present inventions is selected from the group consisting of a Chinese Hamster Ovarian (CHO) cells, such as CHO-K1; Baby Hamster Kidney (BHK) cell; COS cell; HEK293; NSO; SP2/0; YB2/0; HUVEC; HKB; PER-C6; or derivatives of any of these cells.
In some embodiments the cell line according to the present invention is modified to express a gene expressing an exogenous human glycoprotein of interest, such as a human serum protein selected from any one human serpin of table 1:
In particular the present inventors aimed to produce rhA1AT and rhC1INH in CHO-S with N-glycan profiles similar to pIAlAT and pIC1INH. First, the heterogeneous N-glycan profile of CHO-S WT cells was changed to more homogeneous profiles in bespoke cell lines with predominant A2G2 N-glycan structures. Disrupting nine N-glycosylation-related genes increased the A2G2 proportion on total secreted protein from 3.5% in CHO-S WT-derived cells to ˜80% in 10×KO cell lines. This supports the strategy to decrease N-glycan branching and alpha-2,3-sialylation by disrupting MGAT4A, MGAT4B, MGAT5, ST3GAL3, ST3GAL4 and ST3GAL6. The impact of gene disruptions on cell culture performance was assessed in batch cultures. Furthermore, the monoclonal cell lines with disruption in ten gene targets showed enhanced growth characteristics compared to CHO-S WT cells. This included a boosted cell growth in the GLUL-lacking 10×KO cell lines in L-glutamine-supplemented medium.
In contrast to the production platforms previously described, rhAlAT and rhC1INH produced in the 10×KO cell lines described herein are not only exceeding sialylation levels of pIAlAT and pIC1INH but also reveal human-like alpha-2,6-sialylation instead of alpha-2,3-sialylation. The increased sialylation of rhAlAT had no impact on in vitro activity.
The present inventors describes a strategy to successfully engineer the heterogeneous N-glycosylation profile of in particular CHO-S WT cells towards the specific A2G2S2 N-glycan structure with the purpose of producing serpins, such as rhA1AT and rhC1INH with N-glycan profiles similar to human plasma-derived products. Thus, the present invention shows the promise and potential of replacing cost-intensive and possibly unsafe plasma-derived augmentation therapy for AATD and C1INH-HAE patients by CHO- produced rhAlAT and rhC1INH. This strategy is in compliance with the Medical and Scientific Advisory Council (MASAC) recommendation of replacing plasma-derived products with recombinant products for treatment of diseases.
Alpha-1-antitrypsin (A1AT or AAT) refers to the protein identified as UniProtKB-P01009 (A1AT_HUMAN).
Plasma protease Cl inhibitor (ClInh) refers to the protein identified as UniProtKB-P05155 (IC1_HUMAN)
Antithrombin-III (ATIII) refers to the protein identified as UniProtKB-P01008 (ANT3_HUMAN)
The term “inactivated and/or downregulated” refers to a modification of a mammalian host cell, wherein some specific genes are either knocked out, downregulated, or completely or partially inactivated in any other way, such as by miRNA post translational silencing. Preferably this inactivation is a complete inactivation with no measurable sign of expression of this particular gene being inactivated. Suitable techniques to silence/knockout are very well described in the art and known to the person skilled in the art, e.g. as described in WO2015092737. In one specific embodiment, “inactivated and/or downregulated” refers to a gene knockout of the relevant gene.
The term “MGAT4A” as used herein refers to the gene encoding Mannosyl (Alpha-1,3-)-Glycoprotein Beta-1,4-N-Acetylglucosaminyltransferase, Isozyme A. This gene may also be referred to as
The term “MGAT4B” as used herein refers to the gene encoding Mannosyl (Alpha-1,3-)-Glycoprotein Beta-1,4-N-Acetylglucosaminyltransferase, Isozyme B. This gene may also be referred to as
The term “MGATS” as used herein refers to the gene encoding Mannosyl (Alpha-1,6-)-Glycoprotein Beta-1,6-N-Acetyl-Glucosaminyltransferase. This gene may also be referred to as
The term “ST3GAL3” as used herein refers to the gene encoding ST3 Beta-Galactoside Alpha-2,3-Sialyltransferase 3. This gene may also be referred to as ST3Gal III;
The term “ST3GAL4” as used herein refers to the gene encoding ST3 Beta-Galactoside Alpha-2,3-Sialyltransferase 4. This gene may also be referred to as Sialyltransferase 4C (Beta-Galactosidase Alpha-2,3-Sialytransferase);
The term “ST3GAL6” as used herein refers to the gene encoding ST3 Beta-Galactoside Alpha-10 2,3-Sialyltransferase 6. This gene may also be referred to as CMP-NeuAc:Beta-Galactoside Alpha-2,3-Sialyltransferase VI;
Alpha2,3-Sialyltransferase ST3Gal VI;
The term “B3GNT2” as used herein refers to the gene encoding UDP-GlcNAc:BetaGal Beta-1,3-N-Acetylglucosaminyltransferase 2. This gene may also be referred to as UDP-GIcNAc:BetaGal Beta-1,3-N-Acetylglucosaminyltransferase;
The term “GLUL” as used herein refers to the gene encoding glutamate-ammonia ligase also referred to as:
Glutamate Decarboxylase;
EC 4.1.1.15;
The term “SPPL3” as used herein refers to the gene encoding Signal Peptide Peptidase Like 3. This gene may also be referred to as
The term “FUT8” as used herein refers to the gene encoding Fucosyltransferase 8. This gene may also be referred to as
The term “ST6Gal1” as used herein refers to the gene encoding ST6 Beta-Galactoside Alpha-2,6-Sialyltransferase 1. This gene may also be referred to as ST6Gal I;
As detailed above in a first aspect the present invention relates to a recombinant mammalian cell line having a) one or more of the endogenous genes Mgat4A, Mgat4B, Mgat5, St3Gal3, St3Gal4, St3Gal6, SPPL3, and FUT8 inactivated and/or downregulated; and b) optionally a gene encoding Beta-galactoside alpha-2,6-sialyltransferase 1 inserted.
In some embodiments of the mammalian cell according to present invention the endogenous genes Mgat4A, Mgat4B, Mgat5, St3Gal3, St3Gal4, St3Gal6, SPPL3, and FUT8 are inactivated and/or downregulated; and the gene encoding ST6Gal1 is inserted.
In some embodiments the mammalian cell according to present invention has the endogenous genes Mgat4A, Mgat4B, Mgat5, St3Gal4, St3Gal6, and FUT8 inactivated and/or downregulated.
In some embodiments the mammalian cell according to present invention has the endogenous genes Mgat4A, Mgat4B, Mgat5, St3Gal4, St3Gal6, SPPL3, and FUT8 inactivated and/or downregulated.
In some embodiments of the mammalian cell according to present invention the endogenous gene B3GNT2 is present.
In some embodiments the mammalian cell according to present invention further has the endogenous B3GNT2 is inactivated and/or downregulated. It is to be understood that this may be in addition to any combination of other genes being inactivated and/or downregulated.
In some embodiments the mammalian cell according to present invention further has the endogenous GLUL is inactivated and/or downregulated. It is to be understood that this may be in addition to any combination of other genes being inactivated and/or downregulated.
In some embodiments the mammalian cell according to present invention is an in vitro cell 25 line, such as any one selected from the group consisting of a Chinese Hamster Ovarian (CHO) cells, such as CHO-K1, CHO-S, DG44; Baby Hamster Kidney (BHK) cell; COS cell; HEK293; NSO; SP2/0; YB2/0; HUVEC; HKB; PER-C6; NSO; or derivatives of any of these cells.
In some embodiments the mammalian cell according to present invention has been further modified to express an exogenous human glycoprotein of interest, such as a therapeutic human protein. In some embodiments said exogenous human glycoprotein of interest is a human serum protein, such as a human serpin, such as human serpin selected from the list consisting of SERPINA1, SERPINA2, SERPINA3, SERPINA4, SERPINA5, SERPINA6, SERPINA7, SERPINA8, SERPINA9, SERPINA10, SERPINA11, SERPINA12, SERPINA13, SERPINB1, SERPINB2, SERPINB3, SERPINB4, SERPINB5, SERPINB6, SERPINB7, SERPINB8, SERPINB9, SERPINB10, SERPINB11, SERPINB12, SERPINB13, SERPINC1, SERPIND1, SERPINE1, SERPINE2, SERPINE3, SERPINF1, SERPINF2, SERPING1, SERPINH1, SERPINI1, and SERPINI2.
In some embodiments the mammalian cell according to present invention is a cell line producing said glycoprotein of interest with a primary n-glycan structure that is a fully sialylated bi-antennary structure without core fucosylation, such as with more than 80%, such as 82%, such as 84%, such as 86%, such as 88%, such as 90% of the glycoproteins of interest produced being in with a fully sialylated bi-antennary structure without core fucosylation.
In some embodiments the mammalian cell according to present invention has a glycan structure according to the structure A2G2S2 with the following pictorial representations:
Such as according to the structure:
In some embodiments the mammalian cell according to present invention has been further modified to express an exogenous human glycoprotein of interest, which exogenous human glycoprotein of interest is selected from Plasma protease C1 inhibitor (C1Inh) glycosylated at one or more positions selected from Asn3, Asn47, Asn59, Asn216, Asn231, Asn250, and Asn330; Antithrombin-III (ATIII) glycosylated at one or more positions selected from Asn96, Asn135, Asn155 and Asn192; and Human alpha-1-antitrypsin (AAT) glycosylated at one or more, such as two or three of the positions Asn46, Asn83, and Asn247.
We knocked out 9 genes in the CHO-S cell line employing CRISPR/Cas9: FUT8, MGAT4a, MGAT4b, MGAT5, ST3GAL3, ST3GAL4, ST3GAL6, B3gnt2, Sppl3. Furthermore, we introduced the human gene ST6GAL1 to introduce human type branching of sialic acids. The human genes SERPING or SERPINA were then introduced in this host cell line to achieve expression of the human serum proteins Plasma protease Cl inhibitor (ClInh) or Human alpha-1-antitrypsin (AAT), respectively. With these modifications, it is accomplished to shift in the glycosylation profile to the predominant bi-antennary, non-core-fucosylated, and α2-6 linked sialic acid form found in human plasma of recombinant produced serum glycoproteins (
Plasmids 2632 (GFP_2A_Cas9) and 5920 (FUT8_681494) are described in Gray, L. M., Lee, 3. S., Gerling, S., Kallehauge, T. B., Hansen, A, H., Kol, S., Lee, G. M., Pedersen, L. E. and Kildegaard, H. F. (2015), One-step generation of triple knockout CHO cell lines using CRISPR/Cas9 and fluorescent enrichment. Biotechnology Journal, 10: 1446-1456. doi10.1002/biot.201500027.
Plasmids 2928 (MGAT4A_411545), 2933 (MGAT4B_1280368), 2937 (MGAT5_327084), 2940 (ST3GAL4_964386), 2943 (ST3GAL6_1812502), 4408 (B3gnt2 NW 003613880.1_1273293), 4412 (St3_gal3_NW_003613906.1_244730) and 4424 (Sppl3_NW_003613978.1_213040) were constructed as described in Ronda, C., Pedersen, L. E, Hansen, H. G., Kallehauge, T. B, et al., Accelerating genome editing in CHO cells using CRISPR/Cas9 and CRISPy, a web-based target finding tool, Biotechnol. Bioeng. 2014, 111, 1604-4616 with the following modification: sgRNA plasmid sgRNA1_C described in Ronda et al was used as template in the PCR reaction to generate the backbone of gRNA plasmids.
St6Gal1 vector map is shown in
SerpinA vector map is shown in
SerpinG vector map is shown in
SerpinC1 vector map is shown in
We knocked out 10 genes in the CHO-S cell line employing CRISPR/Cas9: FUT8, MGAT4a, MGAT4b, MGAT5, ST3GAL3, ST3GAL4, ST3GAL6, B3gnt2, Sppl3 and GLUL. We have constructed plasmids harbouring both the human ST6GAL1, and SERPING or SERPINA genes to simultaneously introduce human type branching of sialic acids and achieve expression of Plasma protease C1 inhibitor (ClInh), or Human alpha-1-antitrypsin (AAT), respectively.
Combined St6Gal1/SerpinA vector map is shown in
Combined St6Gall/SerpinG vector map is shown in
N-glycan analysis was performed with GlycoWorks RapiFluor-MS N-Glycan Kit (Waters, Milford, Mass.) according to the manufacturer's instruction. In this case 12 μl of 10×concentrated (MWCO filtered, Amicon Ultra-15, Merck, Darmstadt, Germany) secretome or purified protein sample were used for each. Labeled N-Glycans were analyzed by a LC-MS system using a Thermo Ultimate 3000 HPLC with fluorescence detector coupled on-line to a Thermo Velos Pro Iontrap MS. Separation gradient 30% to 43% buffer and MS was run in positive mode.
sgRNA, GFP_2A_Cas9 and A1AT/C1INH_St6gal1_GLUL Plasmid Design.
GFP_2A_Cas9 and single guide RNA (sgRNA) plasmids were constructed as previously described (Gray, L.M. et al., One-step generation of triple knockout CHO cell lines using CRISPR/Cas9 and fluorescent enrichment. Biotechnol. J. 2015, 10, 1446-1456). The sgRNA target design for MGAT4A, MGAT4B, MGAT5, ST3GAL3, ST3GAL4, ST3GAL6, B3GNT2, FUT8, SPPL3 and GLUL was performed using “CRISPy” (Ronda, C. et al., Accelerating genome editing in CHO cells using CRISPR Cas9 and CRISPy, a web-based target finding tool. Biotechnol. Bioeng. 2014, 111, 1604-1616). The target sites for the mentioned genes and the oligos for sgRNA cloning are listed in Table 2 and Table 3, respectively.
Plasmids for co-expression of A1AT/C1INH and St6gal1 were constructed with uracil-specific excision reagent cloning method as previously described (Pristovšek, N. et al., Using Titer and Titer Normalized to Confluence Are Complementary Strategies for Obtaining Chinese Hamster Ovary Cell Lines with High Volumetric Productivity of Etanercept. Biotechnol. J. 2018, 13; and Lund, A. M. et al., A Versatile System for USER Cloning-Based Assembly of Expression Vectors for Mammalian Cell Engineering. PLOS ONE 2014, 9(5): e96693). The DNA sequences of the plasmids are listed in
Cell Cultivation and Transfection for Genome Editing.
CHO-S suspension cells were incubated in a humidified incubator at 120 rpm, 37° C., 5% CO2, passaged to 2-3×105 cells/mL every 2-3 days and transfected in 6-well plates (BD Biosciences, San Jose, Calif.) as described previously (Gray, L. M. et al., One-step generation of triple knockout CHO cell lines using CRISPR/Cas9 and fluorescent enrichment. Biotechnol. J. 2015, 10, 1446-1456). The GFP_2A_Cas9/sgRNA plasmid ratios for each transfection was 1:1 of which the plasmid load of sgRNA was divided equally by the amount of different sgRNAs used per transfection (Table 5). To measure FACS sorting efficiency, pmaxGFP® vector (Lonza, Basel, Switzerland) transfection was performed as well. Cells were harvested for fluorescence-activated cell sorting (FACS) 48 h post transfection.
Single Cell Cloning of Genome Edited Cells Using FACS.
Before FACS, cells were filtered through a 40 μm cell strainer into a FACS-compatible tube.
Single fluorescent-positive (GFP) cells were sorted into 384-well plates (Corning, New York, N.Y.) containing 30 μL CD CHO medium supplemented with 8 mM L-glutamine, 1.5% HEPES buffer and 1% Antibiotic-Antimycotic (Gibco, Waltham, Mass.) per well as described previously (Hansen, H. G. et al, Case study on human alpha1 -antitrypsin: Recombinant protein titers obtained by commercial ELISA kits are inaccurate. Biotechnol. J. 2016, 11, 1648-1656). For cell sorting, fluorescent-positive cell populations were gated based on non-transfected WT CHO-S cells. Two weeks after cell sorting cell colonies were moved to 96-well flat-bottom plates (BD Biosciences) and expanded for deep sequencing analysis and batch cultivation.
Deep Sequencing Analysis.
Confluent colonies from 96-well flat-bottom replicate plates were harvested for genomic DNA extraction. DNA extraction was performed using QuickExtract DNA extraction solution (Epicentre, Illumina, Madison, Wis. according to the manufacturer's instruction. The library preparation was based on Illumina 16S Metagenomic Sequencing Library Preparation and deep sequencing was carried out on a MiSeq Benchtop Sequencer (Illumina, San Diego, Calif.). The protocol for amplifying the targeted genomic sequences, amplicon purification, adapter-PCR and following quality analysis was based on previously published work (Gray, L. M. et al., One-step generation of triple knockout CHO cell lines using CRISPR/Cas9 and fluorescent enrichment. Biotechnol. J. 2015, 10, 1446-1456). PCR primers are presented in Table 6.
Transfection and expression in polyclonal cell lines by applying MSX-selection Cells were seeded in 250 mL Corning vent cap shake flasks (Sigma-Aldrich) as duplicates with cell densities ˜1×106 cells/mL in 60 mL CD CHO medium supplemented with 8 mM L-glutamine (Life Technologies) and transfected with 75 μg of A1AT-GLUL-St6gal plasmid or 75 μg of C1INH-GLUL-St6gal1 plasmid using FreeStyleTM MAX reagent together with OptiPRO
SFM medium (Life Technologies) according to the manufacturer's recommendations. 1 μL/mL anti-clumping agent was added 24 h after transfection. pmaxGFP® vector (Lonza) transfection was performed to measure transfection efficiencies. Two days after transfection, cells were transferred into 60 mL CD CHO medium lacking L-glutamine (Life Technologies) and supplemented with 1 μL/mL anti-clumping agent and 0 μM, 10 μM, 30 μM or 50 μM MSX (EMD Millipore, Billerica, Mass.).
Cell densities and viabilities were determined once per day using the NucleoCounter NC-250 Cell Counter (ChemoMetec). The cells were passaged in fresh selection medium every 2-3 days until viability and doubling time reached stable values. Polyclonal cell lines (pools) were seeded in duplicates at ˜1×106 cells/mL with corresponding MSX concentrations. Cell densities and viabilities were determined once per day and supernatants of the pools were harvested three days after seeding and pooled within duplicates for purification of rhAlAT and rhC1INH.
Single Cell Cloning of Cells from Polyclonal Cell Pools Using FACS
Non-stained single cells were sorted from pools as described above. For cell sorting, all viable cells were gated for sorting into 384-well plates with L-glutamine-free medium. Two weeks after cell sorting the clones were moved to 96-well flat-bottom plates (BD Biosciences) and expanded to shake flask format in CD CHO medium supplemented with 1pL/mL anti-clumping agent, 25 μM MSX and lacking L-glutamine.
Screening cell pools and single cell clones for human-like a-2,6-sialic acid linkage formation with lectin staining.
For lectin staining of cells, triplicates of 10,000 cells per sample were diluted in 200 pL of 0.22 μm pore size filtered CD CHO medium (Life Technologies) supplemented with 5 μg/mL Hoechst 33342 (Merck, Darmstadt, Germany) and 1 μg/mL Fluorescein isothiocyanate (FITC) labeled Sambucus nigra agglutinin (SNA) lectin (Biomol, Hamburg, Germany). After 60 min incubation in the dark at 37° C. and 5% CO2 the cells were washed with 200 μL CD CHO medium and then washed twice with 200 μL phosphate buffered saline (PBS) (300 g, 5 min, RT). The samples were resuspended in 200 ul PBS and transferred to 96-well plate for final centrifugation at 300 g for one minute. Percentage of FITC SNA positive cells was determined in a 96-well optical-bottom microplate (Greiner Bio-One, Frickenhausen, Germany) using a Celigo Imaging Cell Cytometer (Nexcelom Bioscience, Lawrence, Mass.). Cells were identified using the blue channel (Hoechst-positive cells), and the green channel (FITC SNA-positive cells) was used to detect cells with alpha-2,6-sialic acid linkage. A Hoechst/FITC SNA-stained CHO-S WT sample was gated to distinguish between FITC-positive and FITC-negative cells.
Batch Cultivation: Cell Growth Analysis and N-glycosylation Profiling.
For batch cultivation and N-glycan analysis, cells were seeded at 0.4×106 cells/mL in 250 mL Corning vent cap shake flasks (Sigma-Aldrich, St. Louis, Mich.) as duplicates in 60 mL CD CHO medium supplemented with 1 μL/mL anti-clumping agent (Life Technologies). CHO-S WT and non-producing parental 10×KO cell lines were additionally supplemented with 8 mM L-glutamine. rhA1AT/rhC1INH producing clones were cultivated in L-glutamine-free medium at all times and passaged in medium containing 25 μM MSX until the batch cultivation was initiated. Cell densities and viabilities were determined once per day using the NucleoCounter NC-250 Cell Counter (ChemoMetec) until the viability was <70%, at which point the culture was terminated. Supernatant samples with total secreted protein (secretome) from CHO-S WT and parental, non-producing 10×KO cell lines were taken five days after seeding and pooled within biological replicates. The volume for secretome samples was calculated to harbor 20×106 cells. For all shake flasks, additional supernatant samples were taken by centrifuging 1 mL of cell suspension for 5 minutes at 1000 g and storage of supernatant at −80° C. until further analysis.
rhA1AT and rhC1INH Purification
rhA1AT and rhC1INH were purified using CaptureSelect affinity resins (Thermo Fisher Scientific) according to the manufacturer's instructions. rhA1AT was further purified by size exclusion chromatography on a Superdex 200 increase 10/300 GL column (GE Healthcare) equilibrated in PBS.
Titer Assessment of rhA1AT/rhC1INH Producing Clones
rhA1AT and rhC1INH titers were determined using biolayer interferometry on an Octet RED96 (Pall, Menlo Park, Calif., USA) as described previously for A1AT (Noh, S. M. et al., Reduction of ammonia and lactate through the coupling of glutamine synthetase selection and downregulation of lactate dehydrogenase-A in CHO cells. Appl. Microbiol. Biotechnol. 2017, 101, 1035-1045). After hydration in PBS, streptavidin biosensors (18-5021, Fortebio, Pall) were functionalized with CaptureSelect biotin anti-A1AT conjugate or CaptureSelect biotin anti-C1INH conjugate (Thermo Fisher Scientific) at 5 μg/mL in PBS, and blocked in PBS containing 1 μg/mL biocytin (600 and 300 s incubation steps, respectively). Standards were prepared in spent CHO-S medium using plasma-derived A1AT (Athens Research & Technology) at 100, 50, 25, 12.5, 6.3, 3.1 and 1.6 μg/mL or C1INH (R&D systems) at 40, 20, 10, 5, 2.5, 1.25 and 0.625 μg/mL. Samples and standards were diluted two-fold and contained 0.1% BSA w/v, 0.1% tween-20 v/v, and 500 mM NaCl. When needed, samples were further diluted to fall within the range of the standard dilution series. After equilibration in spent CHO-S medium (120 s), samples and standards were measured for 300 s with a shaking speed of 1000 rpm at 30° C. Regeneration was performed with 50 mM TRIS, 2 M MgCl2, pH 7.5. Assays were performed in 96-well black microplates (Greiner Bio-One, Kremsmünster, Austria). Octet System Data Analysis 7.1 software was used to calculate binding rates and absolute A1AT and C1INH concentrations.
SDS-PAGE, Isoelectric Focusing and PNGase Treatment
SDS-PAGE was performed on Novex 4-12% Tris-Glycine mini gels and isoelectric focusing (IEF) was performed on Novex pH 3-10 IEF gels (Thermo Fisher Scientific) as per the manufacturer's instructions. Deglycosylation with PNGase F was performed according to the manufacturer's instructions (New England Biolabs, Ipswich, Mass.).
Activity Sssays
A1AT inhibitory activity was determined using the EnzChek Elastase Assay Kit (Molecular
Probes, Eugene, Oreg.) according to the manufacturer's instructions. In short, A1AT (8.0, 4.0, 2.0, 1.0, 0.5, 0.25, 0.13, and 0.06 μM) was incubated with purified active porcine pancreatic elastase and fluorescently labelled substrate (DQ-elastin). Measurement of fluorescence was performed after 45 min at room temperature (Excitation: 485 nm, slit width 9.0 nm; Emission: 530 nm, slit width 13.5 nm).
C1INH inhibitory activity was determined using the Technochrom C1INH Assay Kit (TechnoClone, Vienna, Austria). In short, plasma containing C1INH activity (120%, 60%, 30%) and samples (˜0.25 μM) were incubated with substrate-buffer mixture for 3 min at room temperature, after which 50% acetic acid was added. Extinction was measured at 405 nm.
N-Glycan Analysis
N-glycans were derivatized with GlycoWorks RapiFluor-MS N-Glycan Kit (Waters, Milford, Mass.) according to the manufacturer's instruction. Briefly; 12 μg purified protein or 12 μl of 10×concentrated (Amicon Ultra-15, Merck) secretome sample were used for each sample. Labeled N-Glycans were analyzed by LC-MS as described previously (Gray, L. M. et al., One-step generation of triple knockout CHO cell lines using CRISPR/Cas9 and fluorescent enrichment. Biotechnol. J. 2015, 10, 1446-1456) Separation gradient from 30% to 43% 50 mM ammonium formate buffer and MS were run in positive mode. Amount of N-Glycan was measured by integrating the peaks with Thermo Xcalibur software (Thermo Fisher Scientific, Waltham, Mass.) giving the normalized, relative amount of the glycans.
Results
Growth Profile and N-glycan Profile of Clonal 10×KO cell lines
The aim of our study was to produce rhAlAT and rhC1INH in CHO cells with N-glycan profiles similar to human pIAlAT and pIC1INH. Our approach was to engineer the heterogeneous N-glycan profile of CHO-S WT cells towards a homogeneous A2G2S2 N-glycan structure, which is the predominant N-glycan on plA1AT/plC1INH. To this end, we generated out-of-frame insertions or deletions (indels) in eight glycosyltransferases (MGAT4A, MGAT4B, MGAT5, ST3GAL3, ST3GAL4, ST3GAL6, B3GNT2, FUT8) as well as in the genes SPPL3 and GLUL (Table 5) over four successive rounds of multiplexed CRISPR/Cas9 gene editing. Two clones with indels in the targeted genes were subjected to growth analysis and N-glycan profiling.
Two clones (10×KO A and 10×KO B) with out-of-frame indels in all ten gene targets were obtained and both showed a pronounced increase in batch culture longevity when compared to the parental CHO-S WT cell line (
CHO-S WT reached maximal viable cell density of —6×106 cells/mL on day five and cell viability declined rapidly to <50% on day 6. In contrast, the 10×KO A and 10x KO B clones had cell viabilities >75% until day 10 of the batch cultivation and reached higher maximal viable cell density than CHO-S WT.
N-glycan analysis of the CHO-S WT secretome resulted in more than 25 N-glycan structures (
After disruption of the targeted genes, the proportion of A2G2 within N-glycan structures of total secreted proteins was increased from 3.5% (CHO-S WT) to 79% in both 10×KO clones (
Introducing Human-Like Sialylation in 10×KO Cell Lines
On the basis of A2G2 secretome N-glycan structures of clone 10×KO B, we aimed to develop clonal cell lines expressing St6gall and rhC1INH or St6gall and rhAlAT. We envisioned that such cell lines are capable to produce rhA1AT or rhC1INH with predominant A2G2S2 N-glycan structures as found on pIA1AT and pIC1INH. The functional GLUL-KO selection system was confirmed by MSX-dosage dependent recovery times of cell viabilities from transfected cell pools. Passaging of the different transfection pools was performed until viability and doubling times were stable. We then conducted FACS-based single cell cloning with the 50 μM MSX-selected cells. During the expansion of the generated clones, only clones exhibiting predominant FITC-SNA staining and detectable levels of rhA1AT/rhC1INH in supernatants on coomassie-stained SDS-PAGE gels were selected. Based on these criteria, two rhA1AT (A1-1 and A1-2) and two rhC1INH (C1-1 and C1-2) producing clones were selected for further characterization.
SNA lectins are reported to bind predominantly to sialic acids of N-glycans linked to the galactose residue in a human-like alpha-2,6-sialylation. Analyzing FITC-SNA-stained CHO-S WT, we found relatively low levels of alpha-2,6-sialylation (
SDS-PAGE gel analysis revealed that purified rhAlAT and rhCiINH produced in the four clones seem to have hydrodynamic volumes (molecular weight) similar to pIAlAT and pIC1INH without detectable impurities as seen in pIC1INH (
To further characterize the CHO-produced rhAlAT and rhC1INH, we performed IEF gel analysis (
IEF gel analysis of rhC1INH produced in a CHO-S WT background resulted in isoforms with pI ranging from pH ˜4-5. A high degree of heterogeneity was also found in purified rhClINH produced in clone C1-1. However, rhC1INH produced in clone C1-2 was less heterogeneous with pI at pH ˜3.5 similar to pIC1INH.
In N-glycan analysis of purified rhA1AT and rhC1INH from CHO-S WT cells we detected a higher degree of heterogeneity compared to N-glycan structures on rhA1AT and rhC1INH from polyclonal 10×KO cell pools. The polyclonal cell lines revealed two predominant sugar structures on both proteins (A2G2 and A2G2S2 N-glycans), whereas we could not detect the A2G2S2 structure on products from CHO-S WT. Moreover, the amount of predominant N-glycan structures on rhAlAT and rhClINH was decreased from two (polyclonal pools) to one (monoclonal producers), identified as A2G2S2 N-glycan.
All four 10×KO-derived monoclonal cell lines produced rhAlAT and rhClINH with higher proportion of A2G2S2 structures than pIA1AT and pIC1INH (
Finally, we investigated the activity of purified rhA1AT and rhCl1NH. rhA1AT activity was determined by its inhibitory function of elastase activity (
1 engineered CHO Cell line with KO of the Sppl3 gene and CHO-S wt cells were both transiently transfected using chemical transfection with plasmids encoding either a Erythropoietin or C1inhibitor gene fused to a HPC4-affinity purification tag. The transfected cells were grown for 72 hours in CD CHO+8 mM L-gln using standard conditions as described previously, after which the supernatant was harvested, sterile filtered and stored at −80° C.
For protein purification, the supernatants were thawed and purified by affinity chromatography using a 1-mL anti-protein C affinity column for EPO and Clinhibitor, and the fractions containing the EPO and Clinhibitor respectively were pooled.
N-glycan analysis was performed on the purified samples, with GlycoWorks RapiFluor-MS N-Glycan Kit (Waters, Milford, Mass.) according to the manufacturer's instruction. In this case 12 μl of purified protein sample were used for each. Labeled N-Glycans were analyzed by a LC-MS system using a Thermo Ultimate 3000 HPLC with fluorescence detector coupled on-line to a Thermo Velos Pro Iontrap MS. Separation gradient 30% to 43% buffer and MS was run in positive mode.
| Number | Date | Country | Kind |
|---|---|---|---|
| 17204071.9 | Nov 2017 | EP | regional |
| 18182948.2 | Jul 2018 | EP | regional |
This application is a divisional of U.S. patent application Ser. No. 16/767,531, filed May 27, 2020, which in turn is a U.S. national stage of International Patent Application No. PCT/EP2018/081616, filed Nov. 16, 2018, which claims the benefit of European Patent Application No. 17204071.9, filed Nov. 28, 2017, and European Patent Application No. 18182948.2, filed Jul. 11, 2018, each of which are incorporated by reference in their entireties.
| Number | Date | Country | |
|---|---|---|---|
| Parent | 16767531 | May 2020 | US |
| Child | 18056126 | US |
