Reduction of formation of amidated amino acids in cell lines for protein expression

Abstract

The present invention is related to a method to reduce peptide amidation activity in a given cell line, cell lines with reduced peptide amidation activity, and uses thereof.

Claims

1. A cell transformed to express a heterologous protein having reduced peptide amidation, wherein the expression of a gene or translation of a gene encoding an enzyme catalyzing peptide a-amidation in the cell is inhibited or reduced.

2. The cell of claim 1, wherein the cell is a eukaryotic cell.

3. The cell of claim 1, wherein the cell is an animal cell and/or a plant cell.

4. The cell of claim 1, wherein the cell is a mammalian cell.

5. The cell of claim 1, wherein the cell is a recombinant cell.

6. The cell of claim 1, wherein the cell is a Baby hamster Kidney cell, a Chinese hamster ovary cell, a mouse myeloma cell, a human embryonic kidney cell, a human-retina-derived cell, or an amniocyte cell.

7. The cell of claim 1, wherein the enzyme catalysing peptide amidation is peptidylglycine alpha-amidating monooxygenase (PAM).

8. The cell of claim 1, wherein the enzyme catalysing peptide a-amidation catalyses the formation of C-terminal proline amide residues.

9. The cell of claim 1, wherein the translation of the gene encoding for said enzyme catalysing peptide a-amidation is inhibited, or reduced, by means of RNA interference (RNAi).

10. The cell of claim 1, wherein the cell comprises an siRNA or shRNA molecule comprising nucleotide sequences that are complementary to a portion of a peptidylglycine alpha-amidating monoxygenase (PAM) gene.

11. The cell of claim 10, wherein the shRNA nucleotide sequences are SEQ ID NOs: 17 and 19 or SEQ ID NOs: 18 and 20.

12. The cell of claim 10, wherein the siRNA or shRNA is in an expression vector.

13. A method for the recombinant production of a heterologous protein having reduced peptide amidation, wherein said method comprises: a) transforming a cell express a heterologous protein, wherein the expression of a gene or translation of a gene encoding an enzyme catalyzing peptide a-amidation in the cell is inhibited or reduced, and b) culturing said cell under conditions to express said heterologous protein.

14. The method of claim 13, wherein the cell is a eukaryotic cell.

15. The method of claim 13, wherein the cell is an animal cell and/or a plant cell.

16. The method of claim 13, wherein the cell is a mammalian cell.

17. The method of claim 13, wherein the cell is a recombinant cell.

18. The method of claim 13, wherein the cell is a Baby hamster Kidney cell, a Chinese hamster ovary cell, a mouse myeloma cell, a human embryonic kidney cell, a human-retina-derived cell, or an amniocyte cell.

19. The method of claim 13, wherein the enzyme catalysing peptide amidation is peptidylglycine alpha-amidating monooxygenase (PAM).

20. The method of claim 13, wherein the enzyme catalysing peptide a-amidation catalyses the formation of C-terminal proline amide residues.

21. The method of claim 13, wherein the translation of the gene encoding for said enzyme catalysing peptide a-amidation is inhibited, or reduced, by means of RNA interference (RNAi).

22. The method of claim 13, wherein the cell comprises an siRNA or shRNA molecule comprising nucleotide sequences that are complementary to a portion of a peptidylglycine alpha-amidating monoxygenase (PAM) gene.

23. The cell of claim 22, wherein the shRNA nucleotide sequences are SEQ ID NOs: 17 and 19 or SEQ ID NOs: 18 and 20.

24. The cell of claim 22, wherein the siRNA or shRNA is in an expression vector.

Description

BRIEF DESCRIPTION OF THE EXAMPLES AND DRAWINGS

(1) Additional details, features, characteristics and advantages of the object of the invention are disclosed in the subclaims, and the following description of the respective figures and examples, which, in an exemplary fashion, show preferred embodiments of the present invention. However, these drawings should by no means be understood as to limit the scope of the invention.

DRAWINGS

(2) In all figures, error bars show standard deviation. All shown results were further analysed with Student's T-test (see below).

(3) FIG. 1: Silencing of PAM mRNA using siRNA (Invitrogen) on CHO K1 PD cells. /1: parallel 1, /2: parallel 2, K: nontransfected CHO K1 PD cells, +K: CHO K1 PD cells transfected with scrambled siRNA (Ambion).

(4) FIG. 2: Silencing of PAM mRNA using siRNA (Ambion) on CHO K1 PD cells. /1: parallel 1, /2: parallel 2, K: nontransfected CHO K1 PD cells, +K: CHO K1 PD cells transfected with scrambled siRNA (Ambion).

(5) The experiments shown in FIGS. 1 and 2 were designed to evaluate the silencing effect on the PAM gene by siRNA provided by Invitrogen or Ambion. CHO K1 PD cells were transfected by siRNAs to determine the sequence with the most potent silencing effect. This was also proven by calculating % decrease in expression of PAM mRNA (Table 9) and using Student's t-Test (Table 10).

(6) FIG. 3: Silencing of PAM mRNA using shRNA on clone K25. /1: parallel 1, /2: parallel 2, K: nontransfected K25 cells.

(7) FIG. 4: Silencing of PAM mRNA using shRNA on clone K62. /1: parallel 1, /2: parallel 2, K: nontransfected K62 cells.

(8) FIG. 5: Silencing of PAM mRNA using shRNA on the CHO K1 PD parental cell line. /1: parallel 1, /2: parallel 2, K: nontransfected CHO K1 PD cells.

(9) FIG. 6: Silencing of PAM mRNA using shRNA on CHO SSF3 parental cell line. /1: parallel 1, /2: parallel 2, K: nontransfected CHO SSF3 cells.

(10) The experiments shown in FIGS. 3-6 were designed to evaluate the mRNA expression level of PAM using different shRNAs and different concentrations of puromycin (PURO) on tested cell lines. The highest silencing effect was observed using sh6 (if used alone or in the mix with sh5) on all cell lines respectively. This was also proven by calculating % decrease in expression of PAM mRNA (Table 11) and using Student's t-Test (Table 12). Only a minor reduction of the expression level was observed using 5 g/ml puromycin. Silencing of PAM was determined on mRNA and protein level respectively.

(11) FIG. 7: Correlation between PAM mRNA expression level and mAb proline amide content. /1: parallel 1, /2: parallel 2, K: nontransfected K25 and K62 cells.

(12) FIG. 8: Chart of pSilencer 2.1-U6 puro vector

(13) FIG. 9: Reaction catalysed by peptidylglycine alpha-amidating monooxygenase (PAM). PAM is a multifunctional protein containing two enzymatic activities that act sequentially to catalyse the C-terminal truncation and alpha-amidation of peptides. peptidylglycine alpha-hydroxylating monooxygenase (PHM) catalyses the first step of the reaction (peptidylglycine (C-terminal).fwdarw.peptidyl--hydrooxiglycine) and is dependent on copper (Cu), or copper ions, ascorbate, and molecular oxygen. The zinc dependent peptidylamido-glycolate lyase (PAL) catalyses the second step of the reaction (peptidyl--hydrooxiglycine.fwdarw.peptide--amide and glyoxylate). Figure taken from Prigge et al., Science (1997) 278, 1300-1305.

EXPERIMENTS

1. siRNA Based Gene Silencing

(14) Chinese hamster ovary (CHO) cells are a cell line derived from the ovary of the Chinese hamster (Cricetulus griseus). They are often used in biological and medical research and commercially in the production of therapeutic proteins. For this reason, it was experimentally shown to reduce peptide amidation activity in a CHO cell line. The exact sequence of the enzyme responsible for proline amide formation in CHO cells is not published in the literature or public databases. The potential nucleotide sequences were extracted from a proprietary CHO EST database. On the basis of this sequence information, siRNAs were constructed, and their silencing effect was evaluated. Short hairpin RNAs (shRNAs) were constructed on the basis of siRNA results, and the gene suppression was evaluated on the mRNA and protein level.

(15) After determination of the C. griseus peptidylglycine alpha-amidating monooxygenase nucleotide sequence (Table 1), 9 and 6 siRNA sequences were designed. The silencing effect was tested after transfection and cultivation of two CHO parental cell lines and two mAb producing CHO cell lines, one producing a product containing high and one with a low proline amide level, respectively. After cultivation, the PAM mRNA level was determined by qPCR and two silencers with the most potent effect were selected for shRNA design. After transfection of cells using shRNA and subsequent cell cultivation, both the mRNA and PAM level were analysed. Experimental details are presented below.

(16) TABLE-US-00001 TABLE1 PAMgenesequenceextractedfromaproprietary CHOESTdatabase.PAMgenesequenceextracted fromaproprietaryCHOESTdatabaseused forsiRNAconstruction(SEQIDNo1): GGGAGTGCTCCTAAGCCAGGCCAGTTCAGTGTTCCTCACAGTTTGGCCCT TGTGCCTCATTTGGACCAGTTGTGTGTGGCAGACAGGGAAAATGGCCGGA TCCAATGTTTCAGAACTGACACCAAAGAATTTGTGAGAGAGATTAAACAT GCGTCATTTGGGAGAAATGTATTCGCAATTTCATATATATCAGGTTTGCT CTTTGCAGTAAATGGGAAGCCTTACTTTGGAGACCATGAACCTGTGCAAG GCTTTGTGATGAACTTTTCCAGTGGGGAAATTATAGATGTCTTCAAGCCA GTACGCAAGCACTTTGACATGCCTCACGATGTGGTTGCCTCTGACGATGG GAATGTGTACATTGGAGACGCACACACGAACACGGTGTGGAAGTTCACCC TGACTGAAAAAATGGAGCATCGATCGGTTAAAAAGGCAGGCATTGAGGCT CAGGAAATCAAAGAAACCGAGGCAGTTGTTGAATCCAAAATGGAGAACAA ACCCACCTCCTCAGAATTGCAGAAGATGCAAGAGAAACAGAAACTGATCA AAGAGCCAGGTTCGGGAGTGCCCGTGGTTCTCATTACAACCCTTCTGGTT ATTCCTGTGGTTGTCCTGCTGGCCATTGTCATGTTTATTCGGTGGAAAAA ATCAAGGGCCTTTGGAGGAAAA
siRNA and shRNA Design and Preparation

(17) Respective design tools were utilised to design siRNA sequences against the peptidylglycine alpha-amidating monooxygenase (PAM) gene. On the basis of gene sequence siRNAs of different sequences and lengths were designed (9 by Invitrogen and 6 by Ambion) (Table 2). After siRNA evaluation, shRNAs were designed using Ambion's online design tool. Two complementary oligonucleotides for each shRNA were synthesised by Methabion (Table 3) and then annealed to generate double stranded oligonucleotides in house. Subsequently, the annealed oligonucleotides were cloned into the pSilencer 2.1-U6 puro vector (FIG. 8). DNA sequencing was performed to verify the sequence of the oligonucleotide insert.

(18) TABLE-US-00002 TABLE2 NucleotidesequencesofsiRNAs Name siRNA_Oligonucleotide SEQIDNo siRNA_1 CAGUUGUGUGUGGCAGACAGGGAAA 2 siRNA_2 CGGAUCCAAUGUUUCAGAACUGACA 3 siRNA_3 CCAAUGUUUCAGAACUGACACCAAA 4 siRNA_4 GAGAGAGAUUAAACAUGCGUCAUUU 5 siRNA_5 CAUGCGUCAUUUGGGAGAAAUGUAU 6 siRNA_6 UGGGAGAAAUGUAUUCGCAAUUUCA 7 siRNA_7 GGGAGAAAUGUAUUCGCAAUUUCAU 8 siRNA_8 CACACGAACACGGUGUGGAAGUUCA 9 siRNA_9 CAAAGAAACCGAGGCAGUUGUUGAA 10 siRNA_1 CAGUAAAUGGGAAGCCUUATT 11 siRNA_2 AGGCAGUUGUUGAAUCCAATT 12 siRNA_3 AACAGAAACUGAUCAAAGATT 13 siRNA_4 CAAGAGAAACAGAAACUGATT 14 siRNA_5 GAACUGACACCAAAGAAUUTT 15 siRNA_6 UUUCAGAACUGACACCAAATT 16

(19) TABLE-US-00003 TABLE3 NucleotidesequencesofshRNAs shRNA_ SEQ shRNA_ SEQ SenseStrand ID AntisenseStrand ID Name Oligonucleotide No Oligonucleotide No SH5_HLPAM 5GATCCGAACTGACA 17 5AGCTTTTCCAAAAA 19 CCAAAGAATTCTCAAG ACAGAACTGACACCAA AGAAATTCTTTGGTGT AGAATTTCTCTTGAGA CAGTTCTGTTTTTTGG ATTCTTTGGTGTCAGT AAA-3 TCG-3 SH6_HLPAM 5GATCCGTTTCAGAA 18 5AGCTTTTCCAAAAA 20 CTGACACCAAACTCAA ATGTTTCAGAACTGAC GAGATTTGGTGTCAGT ACCAAATCTCTTGAGT TCTGAAACATTTTTTG TTGGTGTCAGTTCTGA GAAA-3 AACG-3
Reconstitution of siRNA siRNAs were reconstituted in DEPC water to final 40 M concentration. 30 pmol of siRNA were used for each parallel nucleofection.
Cloning Hairpin siRNA Inserts into pSilencer Vector

(20) Two complementary oligonucleotides for each shRNA were annealed to generate double stranded oligonucleotides. Subsequently, annealed oligonucleotides were cloned into the pSilencer 2.1-U6 puro vector using BamHI and HindIII restriction sites. The whole procedure was performed as follows:

(21) 1. Dissolve the hairpin siRNA template oligonucleotides in approximately 100 l of nuclease-free water.

(22) 2. Dilute the oligonucleotides to approx. 1 g/l in TE.

(23) 3. Assemble the 50 l annealing mixture as follows (Table 4):

(24) TABLE-US-00004 TABLE 4 Annealing of siRNAs Amount Component 2 l sense siRNA template oligonucleotide 2 l antisense siRNA template oligonucleotide 46 l 1 DNA Annealing solution

(25) 4. Heat mixture to 90 C. for 3 min, then place in a 37 C. incubator and incubate 1 hr.

(26) 5. Dilute 5 l of the annealed hairpin siRNA template insert with 45 l nuclease free water to a final concentration of 8 ng/l.

(27) 6. Set up two 10 l ligation reactions; a plus insert ligation and a minus insert negative control.

(28) To each tube, add the following reagents (Table 5):

(29) TABLE-US-00005 TABLE 5 Ligation reactions Plus insert Minus insert Component 1 l Diluted annealed siRNA insert 1 l 1 DNA Annealing solution 6 l 6 l Nuclease free water 1 l 1 l 10 T4 DNA ligase buffer 1 l 1 l pSilencer vector 1 l 1 l T4 DNA ligase (5 U/l)

(30) 7. Incubate at 16 C. overnight.

(31) 8. For the transformation use pGEM-T Easy Vector System (Promega, Cat. No.: A13801): a) Place the E. coli JM109 competent cells in an ice bath until thawed. b) Transfer 50 l of cells to the ligation reaction tubes and add 3 l of ligation reaction to each tube. Gently flick the tubes and incubate for 20 min. c) Heat-shock the cells for 50 s in water bath at 42 C. Immediately return the tubes to ice for 2 min. d) Add 950 l of LB medium to the transformation reactions and incubate for 1.5 hr at 37 C. with shaking (225 rpm). e) Plate 100 l of each transformation culture onto LB/ampicillin plates and incubate overnight at 37 C. f) To identify clones with the siRNA template insert pick clones, isolate plasmid DNA, and digest with BamHI and HindIII, to confirm the presence of the 65 bp siRNA template insert.

(32) 9. Sequence the insert using following sequencing primers (Table 6):

(33) TABLE-US-00006 TABLE6 Sequencingprimers Forwardsequencing Reversesequencing primer(SEQIDNo21) primer(SEQIDNo22) 5-AGGCGATTAAGTTGGGTA-3 5-TAATACGACTCACTATAGGG-3

(34) After isolation and verification of the shRNA expression constructs, they were linearised using the single cutter restriction endonuclease Sspl (AAT/ATT). Maximal 50 g of plasmid DNA per reaction was digested using the Sspl enzyme at 3 U/g DNA (New England Biolabs, Cat. No.: R0132L). An appropriate amount of 10 reaction buffer and H.sub.2O was added to the reaction. The reaction was incubated at 37 C. for 3 hours. After digestion, DNA precipitation was performed under aseptic conditions in a laminar air-flow cabinet as described in the protocol below:

(35) 1. Add 1 volume of isopropanol (300 l)

(36) 2. Vortex thoroughly

(37) 3. Centrifuge 30 min at 21,000 g at 4 C.

(38) 4. Discard the supernatant

(39) 5. Add carefully 1 volume of sterile, ice cold 70% ethanol

(40) 6. Centrifuge 1 min at max speed, at 4 C.

(41) 7. Discard the supernatant

(42) 8. Air dry pellet at RT for 5-30 min. (in laminar)

(43) 9. Resuspend DNA in 50 l sterile water.

(44) Next, purity (O.D. 260/280 nm) and the concentration of linear DNA were determined using the NanoDrop ND-1000.

(45) Host Cells

(46) Four different cell lines were used during the study (parental CHO K1 PD and SSF3 cell lines and two mAb producing clones, K25 and K62). The CHO K1 PD cell line is a subpopulation of the CHO K1 cell line which originates from ATCC (Cat. No. CCL-61.3). The original cell line was adapted to serum free suspension culture and underwent 3 successive rounds of selection at increasingly dilute seeding densities to improve the frequency of serum-free subcloning in DM122 medium. The CHO SSF3 cell line is a serum free adapted cell line from DUKXB1. DUKXB1 was derived from CHO K1 cells. Both functional dhfr alleles were sequentially inactivated in CHO K1. However, the results showed that one of alleles is not inactivated irreversibly. Continuous serum free culture unexpectedly induced expression of low dihydrofolate reductase activity in the originally dihydrofolate reductase deficient (dhfr) CHO cells.

(47) K25 and K62 were prepared by transfection of the SSF3 parental cells with the pBW2017 plasmid vector. It was shown in previous experiments that both clones are expressing a mAb product which contains undesired proline amide structures. K25 and K62 were included in silencing experiments since the respective mAb products contained two extreme values of proline amide, i.e. a low proline amide content on the mAb produced by K25 (4%) and a high proline amide content on the mAb produced by K62 (14%).

(48) Nucleofection

(49) The Amaxa nucleofection system was used for cell transfection (Nucleofector kit V, Cat. No.: VCA-1003). Not more than 5 pools are transfected at once, to enable sufficient time for all necessary cell manipulations. A detailed protocol is described below: 1. At the time of transfection, cells should be up to 2E6/ml with viability90%. 2. 5E6 cells are used per nucleofection. 3. Count cells and centrifuge at 90g, 10 min, RT in 50 ml centrifuge tube. 4. Carefully remove the rest of the medium and resuspend the cell pellet in solution V (100 l per transfection) 5. Add DNA (30 pmol siRNA/nucleofection or 3 g/nucleofection shRNA) and mix gently. 6. Add 100 l cell suspension mixed with DNA into the transfection cuvette, place it in an Amaxa Nucleofector device. 7. Transfect the cells via nucleofection using Amaxa program U 23 8. Add some growth medium into the cuvette and transfer the cells carefully in a 125 ml shake flask with 20 ml medium. Rinse the cuvette 1-2 with fresh medium and add it to the shake flask. Incubate the cells for 24-48 h in a shaker (120 rpm), at 37 C., 10% CO.sub.2.
Growth Medium

(50) CHO K1 PD cells were cultivated in a suitable medium for culturing mammalian cells, such as DM122 growth medium supplemented with 8 mM L-glutamine (Sigma, Cat. No.: G7513). CHO SSF3 cells were cultivated in DM122 growth medium supplemented with 8 mM L-glutamine (Sigma, Cat. No.: G7513) and 1 mg/L insulin (Millipore, Cat. No.: 10131-027). K25 and K62 were cultivated in DM122 growth medium supplemented with 8 mM L-glutamine (Sigma, Cat. No.: G7513), 1 mg/L insulin (Millipore, Cat. No.: 10131-027) and 150 nM methotrexate (methotrexate hydrate, Sigma, Cat. No.: M8407). Cell selection steps were performed in the same medium additionally supplemented with 3 g/ml and subsequently 5 g/ml of puromycin (Gibco, Cat. No.: A11138-02).

(51) Thawing/Freezing of Cells

(52) Vials were thawed in 70% ethanol at 37 C. Cells were drop-wise inoculated directly into 250 ml shake flasks containing 50 ml pre-warmed medium at an initial cell density of cca. 1E5 viable cells per ml. Cells were cultured at 37 C., 10% CO.sub.2, 120 rpm. Cells were frozen in exponential growth phase at a viability>90%. 5-10E6 viable cells per vial were frozen in conditioned medium containing 7.5% DMSO. First the cell culture was centrifuged at 180 g, 5 min, RT, redundant supernatant was discarded. Subsequently DMSO was added to a final concentration of 7.5%. Cell pellets were gently resuspended. Cryo-vials were filled with 1 ml of cell suspension and transferred into a 80 C. deep freezer in a Mr. Frosty cryo box. Within 1 month the frozen vials were transferred into a liquid nitrogen container.

(53) Culture and Handling of Cells

(54) For siRNA experiments CHO K1 PD cells were transfected with siRNAs using nucleofection and cultivated for four days. On day four, cell pellets were collected for qPCR analysis. For shRNA experiments all four cell lines (CHO K1 PD, CHO SSF3, K25 and K62) were transfected with shRNAs using nucleofection. Cells were split on a 2-2-3-day schedule at 2-3E5 cells per ml in the appropriate pre-warmed medium to maintain exponential cell growth. After reaching the appropriate cell density and viability cells were divided and further processed in four separate steps: 1. samples were collected for qPCR (cell pellets) 2. a 10 day batch containing 3 g/ml of puromycin was inoculated (after 10 days supernatants were collected for CEX analysis) 3. 3 cell vials of each cell culture were frozen 4. cells were further cultivated in the medium containing 5 g/ml of puromycin.

(55) After reaching the appropriate cell density and viability using 5 g/ml of puromycin steps 1, 2, and 3 were repeated.

(56) Cells were cultivated in 125 ml shake flasks. Incubation conditions: 37 C., 90-110 rpm for 125 and 250 ml shake flasks and 10% CO.sub.2 for DM122 medium

(57) Puromycin Selection

(58) Antibiotic selection using puromycin was the first selection step after transfection. All transfected pools were selected using puromycin at a final concentration of 3 mg/ml. Puromycin was added to the cell culture 2 days after transfection when cell viability exceeded 60%. After each pool has reached at least 85% cell viability we proceeded with the selection using 5 mg/ml of puromycin.

(59) RNA Isolation and cDNA Synthesis

(60) 10 ng of luciferase RNA (Promega, Cat. No.: L4561) was added to 5E6 cells prior to RNA isolation. Total RNA (totRNA) was isolated using RNeasy Mini Kit (Qiagen, Cat. No.: 74104) on the automated workstation QIAcube. After isolation, the totRNA concentration was measured on NanoDrop. Subsequently, DNase I (Ambion, Cat. No.: AM1906) was added to 5 g of totRNA (Table 7) and incubated (25 min 37 C., 10 min 75 C.). After DNase treatment RNA was transcribed into cDNA using SuperScript VILO kit (Invitrogen, Cat. No.: 11754-050).

(61) TABLE-US-00007 TABLE 7 DNase I treatment and cDNA synthesis DNase treatment cDNA synthesis 5 g totRNA X L DNase treated totRNA 5 L 10 DNaseI Buffer 5 L 5 VILO reaction mix 4 L DNaseI 5 g 10 superscript enzyme 2 L NF water up to 50 L DEPC water 9 L
qPcr

(62) A qPCR method based on TaqMan chemistry was used for mRNA level determination (TaqMan MasterMix, Applied Biosystems, Cat. No.: 4326708 and Assay by design, Applied Biosystems, Cat. No.: 4331348, see Table 8). PAM mRNA expression level was calculated using absolute quantification and was expressed as the number of mRNA transcripts per cell as well as per reference gene ACTB (-actin). In case of calculation per ACTB gene a standard curve was constructed using isolated genomic DNA and was used for determination of ACTB mRNA copy number. The ratio between mRNA of PAM and ACTB was determined. When the mRNA copy number was calculated per cell, a standard curve was constructed using luciferase DNA and the mRNA copy number for luciferase was determined. The ratio between the mRNA of PAM and LUC was then calculated, and the mRNA level of PAM per cell was determined (see FIGS. 1-6)

(63) TABLE-US-00008 TABLE8 NucleotidesequencesofqPCRprimersandprobes SEQ SEQ SEQ Forward ID Reverse ID ID primer No primer No Probe No PAM GGCCGGAT 23 TCCCAAATGA 26 FAM- 29 CCAATGTT CGCATGTTTA CTGACACCAA TCAGAA ATCTCT AGAATTT ACTB AGCCACGC 24 CATCCTGCGT 27 FAM- 30 TCGGTCAG CTGGACCT CCGGGACCTG ACAGACT LUC CTGATTTT 25 GAGTTGTGTT 28 FAM- 31 TCTTGCGT TGTGGACGAA TCCGGTAAGA CGAGTTT GTAC CCTTTCG
Cation Exchange Chromatography (CEX)

(64) Protein A purified mAbs were analysed by CEX using an analytical HPLC chromatographic system. Using this method Lys and proline amide are eluted in the same peak. The amount of proline amide was further determined by product C-terminus treatment with carboxypeptidase. Followed by the same CEX analysis, the remaining peak presents the amount of proline amide.

(65) Experimental Results

(66) The goal of this study was to evaluate the silencing effect on PAM gene by siRNA and shRNA. Silencing of PAM was determined on mRNA and protein level respectively. CHO K1 PD cells were transfected by siRNAs to determine the sequence with the most potent silencing effect (FIGS. 1-2 and Tables 9-10).

(67) TABLE-US-00009 TABLE 9 Silencing effect in % difference when calculated per ACTB or per LUC. Silencing [%] Silencing [%] per ACTB per LUC siRNA Invitrogen si1 60.5 50.2 si2 79.7 78.4 si3 85.7 83.8 si4 68.0 64.7 si5 78.0 73.3 si6 89.6 89.1 si7 87.9 85.6 si8 85.5 83.9 si9 55.0 51.8 siRNA Ambion si1 31.9 32.1 si2 56.1 62.4 si3 45.2 50.6 si4 52.0 56.5 si5 73.2 74.1 si6 71.9 74.0

(68) TABLE-US-00010 TABLE 10 Student's t-Test p-value p-value siRNA Invitrogen si1 0.11 0.09 si2 0.06 0.11 si3 0.06 0.03 si4 0.09 0.06 si5 0.07 0.04 si6 0.05 0.03 si7 0.05 0.03 si8 0.06 0.03 si9 0.13 0.08 siRNA Ambion si1 0.58 0.16 si2 0.32 0.04 si3 0.44 0.03 si4 0.36 0.03 si5 0.22 0.01 si6 0.23 0.02

(69) From the results shown above (FIGS. 1-2 and Table 9) we can conclude that there is up to 90% decrease in expression of PAM mRNA using Invitrogen siRNAs and up to 75% decrease using Ambion siRNAs. Student's t-Test (Table 10) was performed to determine which siRNA's differ significantly from the negative control. The results show the same observations as obtained after calculation percentual difference (lower p-value presents greater difference in expression in comparison to negative control). On the basis of these results two siRNA sequences were selected and shRNA vectors were constructed (si5, si6 for PAM). Due to the shRNA design limitations only sequences from Ambion's siRNAs were used for shRNA construction. To evaluate the silencing of PAM two parental cell lines CHO K1 PD and SSF3 and two mAb producing clones (K25 with low and K62 with high content of proline amide on the product) were transfected with each of the two shRNAs respectively and with the mix of both. Selection of the transfected cells was performed with two different concentrations of puromycin consecutively (FIGS. 3-6). Silencing of PAM was further on evaluated on the protein level by CEX analysis of the produced mAb and the correlation to the mRNA level was determined (FIG. 7).

(70) The results on FIGS. 3-6 show the mRNA expression level of PAM using different shRNAs and different concentrations of puromycin (PURO) on tested cell lines. The highest silencing effect was observed using sh6 (if used alone or in the mix with sh5) on all cell lines respectively. This was also proven by calculating % decrease in expression of PAM mRNA (Table 11) and using Student's t-Test (Table 12). Only a minor reduction of the expression level was observed using 5 g/ml puromycin.

(71) TABLE-US-00011 TABLE 11 Silencing effect in % difference when calculated per ACTB or pre LUC Silencing [%] per ACTB Silencing [%] per LUC K25 3 g sh5/1 50.8 18.8 PURO sh6/1 58.6 56.9 sh5 + 6/1 45.2 48.6 K25 5 g sh5/1 16.2 47.0 PURO sh6/1 63.7 76.6 sh5 + 6/1 27.0 45.9 PD 3 g sh5 32.1 37.7 PURO sh6 63.2 65.1 PD 5 g sh5 48.9 55.7 PURO sh6 80.3 83.1 K62 3 g sh6 44.9 38.8 PURO sh5 + 6 55.1 66.5 K62 5 g sh6 38.8 52.0 PURO sh5 + 6 41.9 47.3 SSF3 3 g sh5 2.0 18.5 PURO sh6 30.5 20.8 SSF3 5 g sh5 38.8 55.4 PURO sh6 56.8 67.9

(72) TABLE-US-00012 TABLE 12 Student's t-Test Silencing [%] Silencing [%] per ACTB per LUC K25 3 g sh5/1 0.06 0.67 PURO sh6/1 0.04 0.27 sh5 + 6/1 0.04 0.33 K25 5 g sh5/1 0.56 0.35 PURO sh6/1 0.04 0.18 sh5 + 6/1 0.17 0.35 PD 3 g sh5 0.17 0.01 PURO sh6 0.03 0.00 PD 5 g sh5 0.07 0.00 PURO sh6 0.01 0.00 K62 3 g sh5 0.02 0.27 PURO sh6 0.01 0.01 K62 5 g sh5 0.09 0.24 PURO sh6 0.02 0.05 SSF3 3 g sh5 0.94 0.68 PURO sh6 0.51 0.70 SSF3 5 g sh5 0.06 0.12 PURO sh6 0.17 0.16

(73) The results in FIG. 7 show a correlation between mRNA and PAM modified mAb. This correlation vas also shown by calculating Pearson's function (Pearson's correlation coefficient was determined to be 0.55).

2. Targeted Gene Knockout in CHO Cells by Using Zinc Finger Nucleases (ZFNs)

(74) ZFNs can be designed to target a chosen locus with high specificity. Upon transient expression of these nucleases, the target gene is first cleaved by the ZFNs and then repaired by a naturalbut imperfectDNA repair process, nonhomologous end joining. This often results in the generation of mutant (null) alleles. Such approach is for example described in Santiago et al., 2008 (Targeted gene knockout in mammalian cells by using engineered zinc-finger nucleases, PNAS Apr. 15, 2008 vol. 105 no. 15).

(75) Site-specific zinc-finger nucleases which target the PAM gene locus are designed and screened in vitro for DNA binding to their target sites. The nuclease function of ZFNs is conferred by the catalytic domain of the endonuclease FokI, which is linked to the DNA-binding zinc-finger proteins.

(76) Plasmids expressing each pair of ZFNs are transfected into CHO cells. The frequency of ZFN-mediated disruption at the target site in each pool of cells is determined by using a CEL-I nuclease.

(77) PAM.sup./ cell lines are generated by transfecting CHO cells with a ZFN pair and then performing a cloning step (e.g., by limiting dilution, ClonePix [Molecular Devices Ltd., UK] or flow cytometry sorting) to obtain single-cell derived PAM-deficient cell lines. After cloning, isolates are analyzed for PAM gene disruption, using the CEL-I assay or qPCR analysis. The exact sequence of the mutant alleles in each cell line, and thus the genotype, is determined by PCR-amplifying the target locus and cloning the PCR product, or by using one of the available second generation sequencing technologies.

3. Gene Targeting with TALENs

(78) TALENs are novel fusion proteins that consist of assembled DNA-binding motifs coupled to FokI nuclease. The DNA-binding motifs come from proteins secreted by plant pathogens in the bacterial genus Xanthomonas.

(79) Assembly of a custom TALEN, or TAL effector construct, is described, e.g., in Cermak et al., 2011 (Efficient design and assembly of custom TALEN and other TAL effector-based constructs for DNA targeting; Nucl. Acids Res. 39 (12)), and involves two steps: (i) assembly of repeat modules into intermediary arrays of 1-10 repeats and (ii) joining of the intermediary arrays into a backbone to make the final construct. Details of this process are described in Cermak et al. 2011

(80) Software to design TALENs is available for use as an online tool (TAL Effector-Nucleotide Targeter, TALE-NT; http://boglabx.plp.iastate.edu/TALENT/). The tool provides a window to input DNA sequences of the gene of interest to be targeted, e.g., the PAM gene. The software identifies sets of TALEN recognition sites between 15 and 30 bp in length and separated by a spacer. The default spacer lengths are 15 bp and 18-30 bp, but other lengths can be specified by the user. In addition, buttons allow users to exclude design guidelines individually.

(81) One of the pairs of TALENs targeting the PAM gene is subcloned into the mammalian expression vector pCDNA3.1() (Invitrogen) using Xhol and AflII. These enzymes excise the entire TALEN from pTAL3 or pTAL4 and place the coding sequence under control of the CMV (cytomegalovirus) promoter. The resulting plasmids are introduced into HEK293T cells by transfection (e.g. by using Lipofectamine 2000 (Invitrogen) following the manufacturer's protocol). Cells are collected 72 h after transfection and genomic DNA isolated and digested with Hpy1881, which cuts in the spacer sequence of the TALEN target site. After digestion, a chromosomal fragment encompassing the target site is amplified by PCR. Subsequently, the PCR products are digested with Hpy1881 and cloned into a TOPO TA vector (Invitrogen). Independent clones containing the full-length PCR product are sequenced to evaluate mutations at the cleavage site.