A study on enhanced O-glycosylation strategy for improved production of recombinant human chorionic gonadotropin in Chinese hamster ovary cells

Journal Pre-proof A study on enhanced O-glycosylation strategy for improved production of recombinant human chorionic gonadotropin in Chinese hamster ovary cells Zhe Deng, Xiaoping Yi, Ju Chu, Yingping Zhuang

PII:

S0168-1656(19)30886-7

DOI:

https://doi.org/10.1016/j.jbiotec.2019.10.006

Reference:

BIOTEC 8523

To appear in:

Journal of Biotechnology

Received Date:

19 July 2019

Revised Date:

24 September 2019

Accepted Date:

7 October 2019

Please cite this article as: Deng Z, Yi X, Chu J, Zhuang Y, A study on enhanced O-glycosylation strategy for improved production of recombinant human chorionic gonadotropin in Chinese hamster ovary cells, Journal of Biotechnology (2019), doi: https://doi.org/10.1016/j.jbiotec.2019.10.006

This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2019 Published by Elsevier.

Title: A study on enhanced O-glycosylation strategy for improved production of recombinant human chorionic gonadotropin in Chinese hamster ovary cells Author names and affiliations Zhe Deng; Xiaoping Yi*; Ju Chu; Yingping Zhuang State Key Laboratory of Bioreactor Engineering, East China University of Science and Technology, Shanghai, Postcode: 200137, China *Corresponding

author: (X. Yi).

ro of

E-mail address: [email protected] Tel.: +86 21 64253337.

re

The expression of rhCG protein in CHO cells was limited by its O-glycosylation. GalNAc was the key precursor in O-glycan synthesis process of rhCG protein. Ugp2 and Galnt1 were the limiting genes in O-glycan synthesis stage of rhCG protein. The expression level of rhCG protein was improved by 3.68 times than the control. A strategy was provided to increase the expression of rhCG protein in CHO cells.

Abstract

na

lP

    

-p

Highlights

ur

Human chorionic gonadotropin (hCG) is a glycoprotein hormone that exists as a heterodimer comprised of an α subunit and β subunit linked with disulfide bridges. The β subunit contains four

Jo

O-glycosylation sites. Previous studies have found that the translation of mRNA to polypeptides of the β subunit was a severely limiting step for the expression of recombinant hCG protein in Chinese hamster ovary (CHO) cells. The effects of O-glycosylation on recombinant hCG protein expression were assessed by adding O-glycan precursors and overexpressing and knocking down key regulatory genes of O-glycan precursor synthesis and O-glycan sugar chain synthesis or hydrolases. The results indicated that O-glycosylation was indeed limiting in the expression of recombinant hCG protein, and N-acetylgalactosamine (GalNAc) was the major limiting precursor.

Glutamine-fructose-6-phosphate transaminase 2 (Gfat2) and Uridine diphosphate-glucose pyrophosphorylase 2 (Ugp2), key regulatory genes of O-glycan precursor synthesis, were overexpressed. Ugp2 overexpression significantly increased the recombinant hCG protein level by 1.92 times compared to that of the control. The LC-MS/MS analysis and Phaseolus vulgaris leucoagglutinin (PHA-L) lectin blot analysis showed that Ugp2 overexpression significantly increased the total galactosylation levels of intracellular proteins and the O-glycosylation of recombinant hCG protein. The stability of the hCG protein to trypsin digestion was also enhanced. Ugp2 is the major limiting enzyme of the O-glycan precursor synthesis in recombinant hCG

ro of

protein production. Furthermore, the effects and mechanisms of the key genes of O-glycan sugar chain synthesis and hydrolases such as polypeptide N-acetylgalactosaminyltransferase1 (Galnt1), Core 1 synthase, glycoprotein-N-acetylgalactosamine 3-beta-galactosyltransferase (C1galt1),

O-linked N-acetylglucosamine transferase (Ogt) and Hexosaminidase (Hex), were evaluated. The

-p

results indicated that Galnt1 overexpression increased the recombinant hCG protein level by 1.57 times and improved the total galactosylation of intracellular proteins, O-glycosylation and the

re

stability of recombinant hCG protein. Galnt1 is the major limiting enzyme of O-glycan sugar chain synthesis. Overexpression of Ugp2 and Galnt1 simultaneously improved the recombinant

lP

hCG protein level by 2.44 times, and both had synergistic effects. Based on the results of overexpression of Galnt1, the major limiting gene of O-Glycan chain synthesis, the precursors

na

GalNAc and Gal were added and increased the recombinant hCG protein level by 3.68 times. This study revealed the major limiting factors of O-glycosylation of recombinant hCG protein in CHO

ur

cells and proposed an effective expression regulation strategy.

Keyword Recombinant human chorionic gonadotropin; CHO cell; O-glycosylation; GalNAc;

Jo

Ugp2; Galnt1

1. Introduction Glycosylation is one of the most common and important post-translational modifications of proteins in vivo. This modification is ubiquitous in cell membrane proteins and secretory proteins and has important biological functions (Goochee et al., 1991; Morten Thaysen-Andersen et al., 2012; Varki, 1993). Glycoprotein biologics are the fastest-growing class of therapeutics (Walsh,

2014). The galactosylation of recombinant IgG was enhanced by the addition of galactose (Gal) (Kildegaard et al., 2016). The N-glycan antenna structure of recombinant EPO protein was regulated by transferring genes such as Human sialyltransferase, UDP-N-acetylglucosamine: α-1,3-D-mannoside β-1,4-N-acetylglucosaminyltransferase (GntIV) and UDP-N-acetylglucosamine: α-1,6-D-mannoside β-1,6-N-acetylglucosaminyltransferase (GntV) into CHO cells (Yin et al., 2015). The N-glycosylation structure of recombinant protein was controlled and engineered by knocking out the various N-glycosylation-related genes in CHO cells (Yang et al., 2015). These studies have shown that addition of glycosylation precursors and

effectively regulate glycosylation of recombinant proteins.

ro of

genetic engineering to modulate key genes of glycan precursors and glycan chain synthesis can

The common glycosylation modifications of proteins include N-glycosylation and

O-glycosylation, while O-linked N-acetylgalactosamine (O-GalNAc) is an important form of

-p

O-glycosylation (Chia, 2016). O-glycosylation can affect the biological activities of organisms in many ways (Gill et al., 2011). Some known O-glycosylation sites are located in the

re

"proline-glutamate-serine-threonine (PEST)" region of the protein. These sequences are thought to be involved in protein degradation, and O-glycosylation may play a critical role in protein stability

lP

and degradation (Hart et al., 2007).

Human chorionic gonadotropin (hCG) is a glycoprotein hormone that consists of an α subunit

na

and a β subunit and is secreted by the uterus during pregnancy. hCG is used to treat many reproductive-related diseases, including cryptorchidism and hypogonadotropism (Gemzell, 1965). The traditional method of producing hCG involves extraction from the urine of pregnant women,

ur

which has the disadvantages of difficulty in collection, low yield, and large differences between batches (Abdalla et al., 1987). The production of recombinant hCG protein in CHO cells was

Jo

shown to be an ideal alternative method. However, due to the complex structure of the hCG protein, the expression level of recombinant hCG protein is low. Previous studies found that the expression of recombinant hCG protein was significantly limited by polypeptide synthesis of the β subunit (Liu et al., 2015). A structural analysis of the recombinant hCG protein revealed that the C-terminus of the β subunit has a special structure containing four O-glycosylation sites (S121/S127/S132/S138) (Morgan et al., 1975). Limited O-glycosylation may be the major factor limiting the polypeptide synthesis of the β subunit and ultimately affecting the expression of the

recombinant hCG protein. The O-glycosylation structure of the hCG protein is a mucin-type structure consisting of three monosaccharides (Figure 1A). The O-glycan precursor and O-glycan sugar chain synthesis pathways of the hCG protein are shown in Figure 1B. The O-glycan precursor is mainly derived from glucose (Glc) in culture medium. Then, a series of O-glycan precursors are synthesized by the Glc backbone, including uridine diphosphate-glucose (UDP-Glc), uridine diphosphate-N-acetylglucosamine (UDP-GlcNAc), uridine diphosphate-galactose (UDP-Gal), uridine diphosphate-N-acetyl galactosamine (UDP-GalNAc), and cytidine

ro of

monophosphate-N-acetylneuraminic acid (CMP-NeuAc). Glutamine-fructose-6-phosphate transaminase 2 (Gfat2) and UDP-glucose pyrophosphorylase 2 (Ugp2) are two major regulatory enzymes in the synthesis of O-glycan precursors (Keppler et al., 1969; Kleczkowski, 1994; McKnight et al., 1992; Zhou et al., 1995). In the O-glycan sugar chain synthesis stage, the

-p

polypeptide N-acetylgalactosaminyltransferase (Galnt) family forms the O-GalNAc initiation site of the mucin-type structure (Elhammer et al., 1999), and the peptide chain shows specific

re

recognition due to different Galnts (Gerken et al., 2011). Core 1 synthase,

glycoprotein-N-acetylgalactosamine 3-beta-galactosyltransferase (C1galt1), also known as Cosmc,

lP

connects UDP-Gal to GalNAc residues to extend the O-glycan sugar chain (Ju et al., 2002). Hexosaminidase (Hex) is a key hydrolase of O-glycan sugar chains (Andersson et al., 2005). The

na

O-linked N-acetylglucosamine transferase (Ogt) forms the O-linked N-acetylglucosamine (O-GlcNAc) initiation site (Hart, 1997). These enzymes play an important regulatory role in the O-glycosylation of proteins.

ur

We investigated the effects of O-glycosylation on recombinant hCG protein expression in CHO cells by adding O-glycan precursors, overexpressing and knocking down key regulatory

Jo

genes of O-glycan precursors and sugar chain synthesis or hydrolysis, and identifying the major bottleneck in the process of O-glycosylation. The results of this study will help effectively promote protein production.

2. Materials and Methods 2.1 Construction of the vectors The sequences of overexpressed genes, Gfat2 (GenBank: XM_003505372.2), Ugp2 (GenBank: AF004368.1), Galnt1 (GenBank: Gene ID: 2589), C1galt1 (GenBank:

XM_003516051.3), and Ogt (GenBank: NM_181672.2), were synthesized by reverse transcription PCR. Hamster Gfat2, Ugp2 and C1galt1 gene sequences were obtained using cDNAs of CHO cells as a template, with the primer pairs gGfat2, gUgp2 and gC1galt1 (Table 1). Human Galnt1 and Ogt gene sequences were obtained in the same manner using cDNAs of HEK293 cells as a template, with the primer pairs gGalnt1 and gOgt (Table 1). These genes were subcloned into the pCI-neo vector (Promega, USA). The XhoI/XbaI (Gfat2), XhoI/XbaI (Ugp2), XhoI/XbaI (C1galt1), XhoI/SmaI (Galnt1), and XhoI/XbaI (Ogt) restriction sites were used for subcloning. Galnt1 and C1galt1 genes were connected by a 2A peptide derived from porcine teschovirus-1

ro of

(T2A) linker (Doronina et al., 2008) to construct the double gene overexpression vector. The sequence Galnt1-T2A was obtained by high-fidelity PCR using pCI-Galnt1 as a template, with the primer pair gGalnt1-T2A (Table 1), and the sequence T2A-C1galt1 was obtained in the same

manner using pCI-C1galt1 as a template, with the primer pair gT2A-C1galt1 (Table 1). Then, the

-p

sequence Galnt1-T2A-C1galt1 (G1C1) was obtained by overlap PCR. The XhoI/SmaI restriction sites were used for subcloning in the pCI-neo vector. The constructed pCI-neo vectors were

re

renamed pCI-Gfat2, pCI-Ugp2, pCI-Galnt1, pCI-C1galt1 and pCI-G1C1, with the gene under the control of the CMV promoter. The CRISPR/Cas9 gene editing technique was used for the pX458

lP

plasmid. A 600 bp Hex (GenBank: XM_007607736.2) genomic sequence (exon 1) downstream of the transcriptional start site was selected to design the Cas9 incision position. Two sgRNA

na

sequences were designed using the sgRNA design website for CHO cells (http://staff.biosustain.dtu.dk/laeb/crispy/) (Ronda et al., 2014). Finally, Hex-sgRNA3 (5'-ccgccgtggaattcggaccg-3') and Hex-sgRNA4 (5'-ccatgcgaccgtccccgcat-3') were selected. The

ur

construction of the PX458-Hex-sg3 and PX458-Hex-sg4 vectors was performed as described in the protocol (Ran et al., 2013). The two vectors were separately transfected into CHO cells, and

Jo

green fluorescence was observed 48 h after transfection. 2.2 Cell culture and transfection CHO-K1 cells coexpressing the hCG α subunit and β subunit were provided by the author's

laboratory and renamed CHO-hCG. The CHO cells were grown in D/F=1:1 medium (Thermo Fisher Scientific) for adherent culture containing 10% newborn calf serum (Thermo Fisher Scientific) in a 37 °C incubator with 5% CO2. Cells were seeded onto a 6-well plate at appropriate densities and transfected 16 h later using Lipofectamine 3000 Reagent (Thermo Fisher Scientific).

Monoclonal cell lines were screened by a limiting dilution method with 700 µg/L G418 (Geneticin, Inalco) in medium. The CHO-hCG cell line transfected pCI-Gfat2, pCI-Ugp2, pCI-Galnt1, and pCI-C1galt1 were renamed CHO-hCG-GFAT2, CHO-hCG-UGP2, CHO-hCG-GALNT1, CHO-hCG-C1GALT1 and CHO-hCG-G1C1, respectively. Then the monoclonal cell lines were adapted to serum free media for suspension cultivation. The suspension cells were grown in CHOGROW CD1 medium (Basalmedia, China) added 4 mM glutamine (Merck) in a 37 °C incubator with 5% CO2 and 120 rpm. In the precursor additive experiment, the sterile precursors O-linked N-acetylglucosamine (GlcNAc), GalNAc, and Gal (Aladdin, China) were dissolved in

2.3 Detection of overexpression gene mRNA level by RT-qPCR

ro of

PBS buffer and added to the medium to a final concentration after the cell seeding.

Transcription levels of the overexpressed genes were quantified by quantitative real-time qPCR. The cDNA of each cell line was synthesized by reverse transcription. The respective

-p

cDNAs were used as templates, and qPCR primers used were as follows: qPCR-Gfat2,

qPCR-Ugp2, qPCR-C1galt1, qPCR-Galnt1, qPCR-Ogt (Table 1). Hamster β actin was used as the

re

housekeeping gene with the primer pair qPCR-β actin (Table 1). The CHO-hCG cell line transfected with empty vector was used as a control group, and the relative mRNA expression fold

lP

was calculated using the 2-ΔΔ method (Livak et al., 2001). In all the real-time PCR experiments, the data were normalized to the expression of housekeeping genes.

na

2.4 Verification of Crispr/Cas9 gene knockout efficiency and selection of positive clones After 48 h of cotransfection with PX459-Hex-sg vectors at isomolar concentrations, the efficiency of genome cleavage was assessed by detecting mismatched DNA. The transfected cell

ur

pool was extracted from genomic DNA, which was used as a template, using the pair Hex600 (Table 1) for high-fidelity PCR. The PCR products were slowly annealed (program: 95 °C for 5

Jo

min, 95 °C to 25 °C, 0.2 °C/s) to generate the mutated DNA mismatch with the non-mutated DNA. Annealed DNA product (300 ng) was treated with T7 endonuclease (NEB, USA) at 37 °C for 1 h. Then, the enzyme-digested products were loaded into 2% agarose gel and separated by electrophoresis. Successfully cleaved DNA would have more than one band after electrophoresis and staining. After confirmation that cleavage occurred, 1% of the positive cells with the highest fluorescence intensity were sorted into 96-well plates using a flow cytometer, 1 cell per well. The Hex600 sequences of 15 monoclonal cell lines were TA subcloned into the PMD20 vector

(TaKaRa, Japan) to determine the final mutational genotype using DNA sequencing. The Hex gene of CHO-hCG cell line, CHO-hCG-Galnt1 cell line and CHO-hCG-C1G1 cell line were renamed CHO-hCG-HEX-KD, CHO-hCG-G1-HEX-KD and CHO-hCG-G1C1-HEX-KD, respectively. 2.5 Immunoblotting Cells were lysed using RIPA cell lysate (Beyotime, China), and the total concentration of the protein lysate was quantified using a BCA protein quantification assay kit (Thermo Fisher Scientific). Protein was separated by 12.5% (w/v) SDS-PAGE with loading 20 µg samples per

ro of

lanes and then transferred to a PVDF membrane (Merck Millipore, Germany). Samples were determined with the appropriate dilution of the primary antibodies. For Western blot analysis, the primary antibodies against Gfat2 (mouse anti-Gfat2), Ugp2 (mouse anti-Ugp2), Galnt1 (mouse

anti-Galnt1), C1galt1 (mouse anti-C1galt1)were obtained from Santa Cruz Biotechnology (USA).

-p

The primary antibodies against Ogt (rabbit anti-Ogt) was obtained from Cell Signaling

Technology (USA). The primary antibodies against HEX (rabbit anti-Hex) was obtained from

re

Abgent (China). Horseradish peroxidase-linked anti-mouse IgG or anti-rabbit IgG obtained from Santa Cruz Biotechnology (USA) was used as secondary antibodies individually. For the lectin

lP

blot, biotin-labeled Phaseolus vulgaris leucoagglutinin (PHA-L) (Vectorlabs, USA), a lectin reacting to galactose residue in glycoproteins(Gerfen et al., 1984), was used to incubate the

na

membranes, followed by incubation in horseradish peroxidase-avidin (Solarbio, China) as a secondary antibody. All horseradish peroxidase-linked secondary antibodies were detected by the enhanced chemiluminescent kit (Beyotime, China), according to the the manufacturer’s

ur

instructions

2.6 Intact recombinant hCG protein expression analysis

Jo

The expression level of intact recombinant hCG protein in the cell culture supernatant was

determined with a hCG ELISA kit (DPG, Germany) for quantification. The cell culture medium at 96 h was centrifuged at 200 g for 5 min to harvest the supernatant and then measured according to the ELISA manufacturer’s instructions. 2.7 TMT10plex mass tag labeled LC-MS/MS assay In quantification of glycopeptides, a Tandem Mass Tag (TMT) labeling technique (Thermo Fisher Scientific) by liquid chromatography-tandem mass spectrometry (LC-MS/MS) was used to

semi-quantify the abundance of various glycopeptides in different samples relative to that of the control group. With this method, all TMT labeled samples can be mixed together to detect the absence of peptides in the sample by one injection to avoid systematic errors caused by multiple injections. Labeled samples then are analyzed by high resolution Orbitrap LC-MS/MS before data analysis to identify peptides and quantify reporter ion relative abundance. For MS/MS fragmentation, higher energy collisional dissociation (HCD) were acquired for O-linked glycopeptides for the purpose of quantification. However, HCD can not provide enough information for determination of O-glycosylation sites. Therefore, the known glycosylation

ro of

structure was added as a dynamic modification in the ProteomeDiscovery@ software (Thermo Fisher Scientific) analysis process to semi-quantify the content of intact O-glycan in different samples to obtain more comprehensive O-glycosylation information by HCD MS/MS

fragmentation. For more accurate analysis, the hCG samples were purified by cation exchange

-p

chromatography to remove the host protein. Digestion and TMT isotope labeling were performed using TMT10plex Mass Tag Labeling Kits and Reagents (Thermo Fisher Scientific). Three

re

replicate hCG protein samples, each 50 µg in equal amounts from 10 clones, were labeled by the 10 different TMT mass tags (TMT-126, 127N, 127C, 128N, 128C, 129N, 129C, 130N, 130C, 131 )

lP

respectively. Each protein sample was pretreated according to the TMT10plex instructions. An Orbitrap Fusion™ Tribrid™ mass spectrometer (Thermo Scientific, Germany) coupled with a

na

Thermo Scientific™ Dionex™ Ultimate™ 3000 UHPLC instrument was used for LC-MS/MS analysis. Isobarically labeled digests were separated with a Waters SunFire C18 column (2.1×150 mm, 3.5 μm). The LC-MS/MS experimental parameters and method were previously described (H.

ur

Zhu et al., 2017).

2.8 Intact recombinant hCG protein stability analysis

Jo

The stability of recombinant hCG proteins that regulates the O-glycosylation levels before and

after genetic engineering was determined using trypsin digestion by ELISAs. The CHO-hCG cell line transfected with empty vector supernatant was used as a control, and the same amount of hCG protein from genetically modified cell supernatant was diluted with PBS to the same concentration as the control. Then, ELISAs were used to detect the residual hCG protein concentration at 37 °C after trypsin-free and trypsin treatments (Thermo Fisher Scientific) for 1, 2, 3, 4, and 5 h. Trypsin was added at 2.5 µg per 100 µl supernatant.

2.9 Data analysis LC-MS/MS data were analyzed using ProteomeDiscovery@ software (version 2.1). Samples were prepared and injected in triplicate for LC-MS/MS analysis. The analysis parameters are as follows: precursor mass tolerance=10 ppm, HCD&ETD, fragment mass tolerance (HCD)=40 ppm, fragment mass tolerance (ETD)=0.7 Da; modifications: TMT6plex/+229.162932 @ K and N-term (Fixed), Carbamidomethyl/+57.021464 @C (fixed), Oxidation/+15.994915@M (common1), HexHexNAcNeuAc/+672.5946, Dehydrated/-18.010565@S, T (Rare2), (De) Carbamidomethyl/-57.021464 @ C (rare 1), (De)TMT6plex/-229.162932 @ N-term (rare 1);

ro of

database: CGB (Human)+CGA (Human)+uniprot-Cricetulus griseus. The spectrum-level false discovery rate (FDR) was set as 1% cut, and peptide spectra matches (PSMs) with a manual score cut at 200 were used. Statistical analysis was analyzed using SPSS11 software. The statistical significance of variables was evaluated by applying the analysis of variance (ANOVA) using

-p

Fisher F test. Statistical significance was accepted at the p<0.05 level.

3. Results

re

3.1 Effect of O-glycan precursor addition on the recombinant hCG protein expression In vitro, O-glycan precursors are mainly derived from Glc in culture medium. A series of

lP

O-glycan precursors are synthesized from Glc, including UDP-Glc, UDP-GlcNAc, UDP-Gal and UDP-GalNAc. The highest expression level of recombinant hCG protein in CHO-hCG cell line as

na

control for suspension culture was 5.1 mg/L in this study. For ease of analysis, relative concentrations were used for showing all recombinant hCG protein expressions level. The addition of 4 mM GalNAc to the culture medium had the most significant effect on the expression

ur

level of recombinant hCG protein (Figure 2). The effect of the simultaneous addition of 4 mM GalNAc and 4 mM Gal or 4 mM GlcNAc on the recombinant hCG protein level was similar to

Jo

that of 4 mM GalNAc alone. These results indicated that GalNAc is the major limiting precursor of O-glycosylation of recombinant hCG protein. 3.2 Effects of overexpressing the key genes of O-glycan precursor synthesis on the O-glycosylation of recombinant hCG protein Gfat2 and Ugp2 are two key enzymes in the O-glycan precursor synthesis stage. To study the effects of these two enzymes on O-glycosylation of recombinant hCG protein, we overexpressed the Gfat2 and Ugp2 genes in the CHO-hCG cell line. The two recombinant monoclonal cell lines

were named CHO-hCG-GFAT2 and CHO-hCG-UGP2. The mRNA levels of the overexpressed genes were detected by RT-qPCR (Figure 3A). Using the CHO-hCG cell line transfected with empty vector as the control group, we found that the mRNA transcription levels of the two genes in the recombinant monoclonal cell lines were both significantly higher than those of the control. The intracellular protein levels of Gfat2 in the CHO-hCG-GFAT2 cell line and Ugp2 in the CHO-hCG-UGP2 cell line were determined by Western blot analyses, and the results indicated that the expression levels of the two proteins were significantly increased (Figure 3B). The concentrations of recombinant hCG protein in the culture supernatants of the

ro of

CHO-hCG-GFAT2 cell line and the CHO-hCG-UGP2 cell line were detected by ELISAs (Figure 4A). The results showed that there was no significant difference between the CHO-hCG-GFAT2

cell line and the control cell line. However, the concentration of recombinant hCG protein in the CHO-hCG-UGP2 cell culture supernatant was 1.92 times higher than that in the control. The

-p

results indicated that Ugp2 may be the main limiting enzyme of the synthesis of O-glycan precursors.

re

PHA-L was specifically used to identify modification of galactosylation. The levels of total galactosylation of intracellular proteins can indirectly indicate O-glycosylation of recombinant

lP

hCG protein. The total galactosylation levels of intracellular proteins in recombinant cells were examined by a PHA-L lectin blot analysis (Figure 4B). The total galactosylation level of

na

intracellular proteins in the CHO-hCG-GFAT2 cell line was almost the same as that in the control. The total galactosylation level of intracellular proteins in the CHO-hCG-UGP2 cell line was slight increased than that in the control.

ur

A LC-MS/MS technology analysis was used to detect the O-glycosylation level of recombinant hCG protein. The LC-MS/MS method has been widely used to study N-glycosylation

Jo

and O-glycosylation structures and functions (Huang et al., 2014; M. Thaysen-Andersen et al., 2014; Zhang et al., 2018; Z. Zhu et al., 2014). The N-/O-glycosylation characteristics of commercial hCG proteins from different sources and batches were analyzed using this method (H. Zhu et al., 2017). This strategy is the most effective O-glycosylation structure analysis method to date. The MS/MS spectra from multiple dissociation techniques (HCD) of the O-linked glycopeptide in β hCG subunit was shown in Supplementary Figure 1. All relative abundance of O-linked glycopeptide of recombinant hCG protein form 10 clones samples were calculated by

ProteomeDiscovery@ software. The O-glycopeptide of recombinant hCG protein produced from the CHO-hCG-UGP2 cell line was significantly higher than that from the control, but there was no significant difference between the CHO-hCG-Gfat2 cell line and the control line (Figure 4C). These results indicated that overexpression of Ugp2 improved the O-glycosylation of recombinant hCG protein. Trypsin stability experiments were performed to further investigate the effect of Ugp2 overexpression on the O-glycosylation of recombinant hCG protein. Using the same concentrations of recombinant hCG proteins produced from the CHO-hCG-UGP2 cell line and the

ro of

control, we detected the contents of residual recombinant hCG proteins by ELISAs in an hourly reaction system (Figure 4D). The results showed that the recombinant hCG protein contents were

almost the same after 5 h in the trypsin-free group. After addition of trypsin, the recombinant hCG protein contents in the two groups decreased significantly after 5 h, but the residual amount of

-p

recombinant hCG protein produced from CHO-hCG-UGP2 cells was higher than that from the control. The results further supported that Ugp2 overexpression can effectively improve the

re

O-glycosylation of recombinant hCG protein, and thus, the stability of the protein was eventually enhanced.

lP

3.3 Effects of regulating the key genes of O-glycan sugar chain synthesis and hydrolysis on the O-glycosylation of recombinant hCG protein

na

The key regulatory genes Ogt, Galnts, C1galt1 and Hex of O-glycan sugar chain synthesis and hydrolysis were systematically studied. There are 20 enzymes in the Galnt family. To determine the key enzymes regulating the O-glycosylation of recombinant hCG protein, we cloned

ur

six Galnt family genes from the placenta. The recombinant hCG protein contents in the culture supernatant were detected at 72 h after transient transfection. The results (Supplementary Figure 2)

Jo

showed that overexpressing the Galnt1 gene resulted in the highest expression of recombinant hCG protein. Galnt1 may be the key enzyme regulating O-glycosylation of recombinant hCG protein in the Galnt family. The human Galnt1 gene was used in this study. Stable monoclonal cell lines overexpressing the Ogt/Galnt1/C1galt1 genes were named the CHO-hCG-OGT cell line, the CHO-hCG-GALNT1 cell line and the CHO-hCG-C1GALT1 cell line, respectively. Analysis of the mRNA levels (Figure 5A) and protein levels (Figure 5B) indicated that the Ogt/Galnt1/C1galt1 genes were successfully overexpressed.

The O-glycan key hydrolase gene Hex was knocked down by Crispr/Cas9 technology. In this study, two cotransfected sgRNAs were used to knock down the Hex gene. The Hex sequencing results showed that the expected Hex gene sequence was deleted, and the size of the deletion fragment was 152 bp. Because the number of deletion fragments was not a multiple of 3, a frameshift mutation had occurred. Western blot analysis (Figure 5C) indicated that Hex expression decreased significantly and that the Hex gene was successfully knocked down. The knockdown cell line was named the CHO-hCG-HEX-KD cell line. The concentrations of recombinant hCG protein from the culture supernatants of the

ro of

CHO-hCG-OGT, CHO-hCG-GALNT1, CHO-hCG-C1GALT1, and CHO-hCG-HEX-KD cell lines and the control at 96 h were measured by ELISAs (Figure 6A). The results showed that the expression level of the CHO-hCG-GALNT1 cell line was significantly higher than that of the

control, which was increased by 1.57 times, and the levels in the other three recombinant cell lines

-p

did not change noticeably.

The results of the LC-MS/MS TMT analysis (Figure 6B) and the PHA-L lectin blot analysis

re

(Figure 6C) showed that Galnt1 gene overexpression significantly increased the total galactosylation of intracellular proteins and the O-glycosylation of recombinant hCG protein. The

lP

stability of the protein to trypsin digestion was also enhanced (Figure 6D). Galnt1 was confirmed to be the major limiting enzyme for O-glycan sugar chain synthesis of recombinant hCG protein.

na

The CHO-hCG-G1-HEX-KD cell line overexpressed the Galnt1 gene and silenced the Hex gene. The CHO-hCG-G1C1 cell line simultaneously overexpressed the Galnt1 and C1galt1 genes. The CHO-hCG-G1C1-HEX-KD cell line simultaneously overexpressed the Galnt1 and C1galt1

ur

genes and silenced the Hex gene. The mRNA and protein levels of each recombinant cell lines are shown in Figure 7A and 7B. Although the expression levels of these proteins were different, they

Jo

were successfully overexpressed or knocked down. The ELISA results (Figure 8A) and the LC-MS/MS TMT (Figure 8B) and PHA-L lectin blot

analyses (Figure 8C) showed that simultaneously overexpressing the Galnt1 and C1galt1 genes and knocking down the Hex gene could not further improve the O-glycosylation of recombinant hCG protein compared with overexpressing the Galnt1 gene alone. We further confirmed that Galnt1 is the major limiting enzyme of O-glycan sugar chain synthesis of the recombinant hCG protein.

3.4 Regulation of recombinant hCG protein expression Based on previous studies, GalNAc is the main limiting precursor, Ugp2 and Galnt1 are the two main limiting enzymes of the O-glycosylation of recombinant hCG protein, and a set of strategies was proposed to improve recombinant hCG protein expression (Table 2). The results showed that simultaneously overexpressing the Galnt1 and Ugp2 genes compared with those of the control increased the concentration of recombinant hCG protein by 2.44 times, and both had a synergistic promoting effect. Based on the results of Galnt1 gene overexpression, the O-glycan precursors GalNAc and Gal were added individually or simultaneously, and the concentration of

ro of

recombinant hCG protein was further increased by 2.87 times and 3.68 times.

4 Discussion

Previous studies found that the expression of recombinant hCG protein was obviously

limited by the mRNA translation to polypeptides for the β subunit in CHO cells(Yang et al.,

-p

2015). Dysfunctional O-glycosylation may be the main factor limiting the polypeptide synthesis of the β subunit and ultimately affecting the expression of recombinant hCG protein. To identify

re

the limiting factors of β subunit polypeptide synthesis and propose an effective regulation strategy,

recombinant hCG protein.

lP

this study systematically researched the regulatory mechanism of the O-glycosylation of

The O-glycosylation of proteins is mainly divided into two stages: O-glycan precursor

na

synthesis and O-glycan sugar chain synthesis. The synthesis of O-glycan precursors is an important aspect of O-glycosylation regulation. An adequate supply of O-glycan precursors is a prerequisite for efficient O-glycan synthesis. By adding O-glycan precursors and modulating the

ur

key genes of O-glycan precursors and sugar chain synthesis using genetic engineering, we found that O-glycosylation of recombinant proteins can be effectively regulated.

Jo

The precursor addition studies showed that O-glycan precursor synthesis indeed limits

recombinant hCG protein production in CHO cells. Among the various O-glycan precursors, GalNAc had the most significant impact on the expression of recombinant hCG protein and was the major limiting precursor. Analysis of the O-glycan precursor synthesis pathway and the addition of GlcNAc results indicated that there may be some restrictions in the synthesis of Glc to N-acetylglucosamine 6-phosphate (GlcNAc-6-P), in which Gfat2 is a key limiting enzyme. The addition of Gal showed that there may also be some limitations in the synthesis from Glc to

UDP-Gal, in which Ugp2 is a key regulatory enzyme. The effects of simultaneous addition of 4 mM GalNAc and 4 mM Gal or 4 mM GlcNAc on the expression of recombinant hCG protein were similar to that of 4 mM GalNAc alone, indicating that the effect of the GalNAc precursor on O-glycosylation is greater than that of the Gal precursor. Indirectly, these findings confirmed that the O-glycosylation of recombinant hCG protein is mainly a mucin-type structure in which O-GalNAc is the initiation site. Increasing the expression of Ugp2 significantly increased the expression level of recombinant hCG protein, indicating that Ugp2 may be the main limiting enzyme in the synthesis pathway

ro of

from Glc to UDP-Gal of O-glycan precursors. Overexpressing Ugp2 may improve O-glycan precursor synthesis, thus increasing O-glycosylation and recombinant hCG protein expression. There was no significant difference in the expression of recombinant hCG protein between the Gfat2 overexpression cell line and the control line. The Gfat2 enzyme may not be the limiting

-p

factor in the synthetic pathway of Glc to N-acetylglucosamine 1-phosphate (GlcNAc-1-P).

Nevertheless, in addition to the GlcNAc precursor, the expression level of recombinant hCG

re

protein was also increased. This result suggests that there may be other restrictions in the synthesis pathway from Glc to GlcNAc-1-P due to the complicated reaction.

lP

PHA-L lectin blot analysis showed when the expression of Ugp2 increased, the improvement in total galactosylation of intracellular proteins indicated that cells synthesize more Gal precursors;

na

thus, the O-glycosylation of intracellular proteins, including recombinant hCG protein, was increased. The same results indicated that Ugp2 was the major limiting enzyme for the synthesis of O-glycan precursors of recombinant hCG protein.

ur

The LC-MS/MS TMT analysis is the most effective O-glycosylation structure analysis method at present. The O-glycosylation of recombinant hCG protein produced from the Ugp2

Jo

overexpression cell line was significantly higher than that of the control; however, there was no obvious difference between the Gfat2 overexpression cell line and the control cell line. Overexpressing the Ugp2 gene effectively improved the O-glycosylation level of the recombinant hCG protein. O-glycosylation can mask the protease hydrolysis site and enhance the stability of the protein to proteolytic enzymes. Trypsin stability experiments can indirectly characterize the O-glycosylation level of recombinant hCG protein. This research further confirmed that Ugp2

gene overexpression can effectively improve the O-glycosylation of recombinant hCG protein, and thus, the stability of the protein was eventually enhanced. In general, all of the above studies showed that Ugp2 is the major limiting enzyme for the synthesis of O-glycan precursors of recombinant hCG protein. On the stage of O-glycan sugar chain synthesis regulation, the expression level of recombinant hCG protein of the Galnt1 overexpression cell line was higher than that of the control, which was increased by 1.57 times, and the levels in the other three recombinant cell lines did not change significantly. The results of the LC-MS/MS TMT analysis and the PHA-L lectin blot

ro of

analysis showed that Galnt1 gene overexpression increased the contents of total galactose residues of intracellular protein and the O-glycosylation of recombinant hCG protein. At the same time, the protein stability in trypsin digestion experiments was also enhanced. These studies indicated that Galnt1 is the main limiting enzyme for O-glycan sugar chain synthesis of recombinant hCG

-p

protein. Simultaneously, overexpressing the Galnt1 and C1galt1 genes and knocking down the

Hex gene did not further improve the O-glycosylation of recombinant hCG protein compared with

re

overexpressing the Galnt1 gene alone. We further confirmed that Galnt1 is the major limiting enzyme of O-glycan sugar chain synthesis of recombinant hCG protein.

lP

Simultaneously overexpressing the Galnt1 and Ugp2 genes further increased the recombinant hCG protein by 2.44-fold, and both had a synergistic promoting effect. Based on the Galnt1 gene

na

overexpression results, the O-glycan precursors GalNAc and Gal were added individually or simultaneously, and the expression levels of recombinant hCG protein were further increased by 2.87 times and 3.68 times.

ur

Taken together, the above studies showed that there are significant limitations in the synthesis of O-glycan precursors and sugar chains of the O-glycosylation of recombinant hCG protein. In

Jo

the mucin-type structure of the O-glycan of recombinant hCG protein, the main precursor GalNAc is significantly restricted. There are some restrictions in the synthesis pathway from Glc to UDP-GalNAc due to the complicated reactions; CHO cells prefer to effectively utilize exogenous GalNAc. In addition, because this pathway involves the synthesis of NeuAc precursors, GalNAc addition not only improves the supply of GalNAc precursors but also avoids substrate competition for NeuAc precursor synthesis, thereby improving the supply of NeuAc precursors. The structure information of NeuAc residue were added in LC-MS/MS analysis process, it increased as well as

GalNAc-Gal structure. Therefore, the synthesis of NeuAc is not limited in synthesis pathway of O-glycan sugar chain. Ugp2 and Galnt1 are the two main limiting enzymes for the O-glycosylation of recombinant hCG protein. The overexpression of the major limiting enzyme Ugp2 in the synthesis from Glc to UDP-Gal of O-glycan showed that CHO cells tend to utilize the exogenous Gal to effectively supply the Gal precursor. This study reveals the major limiting factors of O-glycosylation and provides an effective regulatory strategy to improve the expression of recombinant hCG protein in CHO cells (Figure 9) and proposes an effective method for the

ro of

study and regulation of O-glycosylation of recombinant proteins.

Declaration of interests

The authors declare that they have no known competing financial interests or personal

-p

relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

lP

re

The authors declare the following financial interests/personal relationships which may be considered as potential competing interests:

This work was supported by grants from the Chinese National Programs for High

Jo

ur

na

Technology Research and Development (2015AA020801)

References: Abdalla, H. I., Ah-Moye, M., Brinsden, P., Howe, D. L., Okonofua, F., & Craft, I. 1987. The effect of the dose of human chorionic gonadotropin and the type of gonadotropin stimulation on oocyte recovery rates in an in vitro fertilization program. Fertility and

ro of

Sterility, 48(6), 958-963.https://doi.org/10.1016/S0015-0282(16)59591-0 Andersson, S. V., Sjögren, E. C., Magnusson, C., & Gierow, J. P. 2005. Sequencing,

expression, and enzymatic characterization of β-hexosaminidase in rabbit lacrimal

-p

gland and primary cultured acinar cells. Glycobiology, 15(3), 211-220.https://doi.org/10.1093/glycob/cwi006

re

Chia, J. 2016. Short O-GalNAc glycans: regulation and role in tumor development and clinical

lP

perspectives. Biochimica et Biophysica Acta (BBA) - General Subjects, 1860(8), 1623-1639.https://doi.org/https://doi.org/10.1016/j.bbagen.2016.03.008

na

Doronina, V. A., Wu, C., de Felipe, P., Sachs, M. S., Ryan, M. D., & Brown, J. D. 2008. Site-Specific Release of Nascent Chains from Ribosomes at a Sense Codon. 28(13),

ur

4227-4239.https://doi.org/10.1128/MCB.00421-08 %J Molecular and Cellular Biology

Jo

Elhammer, A. P., Kézdy, F. J., & Kurosaka, A. 1999. The acceptor specificity of UDP-GalNAc:polypeptide N-acetylgalactosaminyltransferases. Glycoconjugate

journal, 16(2), 171-180.https://doi.org/10.1023/a:1026465232149

Gemzell, C. 1965. Induction of ovulation with human gonadotropins. Recent Prog Horm Res,

21, 179–204 Gerfen, C. R., & Sawchenko, P. E. 1984. An anterograde neuroanatomical tracing method that

shows the detailed morphology of neurons, their axons and terminals: immunohistochemical localization of an axonally transported plant lectin, Phaseolus vulgaris leucoagglutinin (PHA-L). Brain research, 290(2), 219-238.https://doi.org/10.1016/0006-8993(84)90940-5 Gerken, T. A., Jamison, O., Perrine, C. L., Collette, J. C., Moinova, H., Ravi, L., Markowitz, S. D., Shen, W., Patel, H., & Tabak, L. A. 2011. Emerging paradigms for the initiation of

ro of

mucin type protein O-glycosylation by the polypeptide GalNAc transferase (ppGalNAc T) family of glycosyltransferases.https://doi.org/10.1074/jbc.M111.218701

Gill, D. J., Clausen, H., & Bard, F. 2011. Location, location, location: new insights into

-p

O-GalNAc protein glycosylation. Trends in Cell Biology, 21(3),

re

149-158.https://doi.org/https://doi.org/10.1016/j.tcb.2010.11.004

Goochee, C. F., Gramer, M. J., Andersen, D. C., Bahr, J. B., & Rasmussen, J. R. 1991. The

lP

Oligosaccharides of Glycoproteins: Bioprocess Factors Affecting Oligosaccharide Structure and their Effect on Glycoprotein Properties. Bio/Technology, 9,

na

1347.https://doi.org/10.1038/nbt1291-1347

ur

Hart, G. W. 1997. Dynamic O-Linked Glycosylation of Nuclear and Cytoskeletal Proteins. 66(1), 315-335.https://doi.org/10.1146/annurev.biochem.66.1.315

Jo

Hart, G. W., Housley, M. P., & Slawson, C. 2007. Cycling of O-linked β-N-acetylglucosamine on nucleocytoplasmic proteins. Nature, 446, 1017.https://doi.org/10.1038/nature05815 Huang, L.-J., Lin, J.-H., Tsai, J.-H., Chu, Y.-Y., Chen, Y.-W., Chen, S.-L., & Chen, S.-H. 2014. Identification of protein O-glycosylation site and corresponding glycans using liquid

chromatography–tandem mass spectrometry via mapping accurate mass and retention time shift. Journal of Chromatography A, 1371, 136-145.https://doi.org/https://doi.org/10.1016/j.chroma.2014.10.046 Ju, T., Brewer, K., D'Souza, A., Cummings, R. D., & Canfield, W. M. 2002. Cloning and Expression of Human Core 1 β1,3-Galactosyltransferase. 277(1), 178-186.https://doi.org/10.1074/jbc.M109060200

ro of

Keppler, D., & Decker, K. 1969. Studies on the Mechanism of Galactosamine Hepatitis:

Accumulation of Galactosamine-1-Phosphate and its Inhibition of UDP-Glucose Pyrophosphorylase. 10(2),

-p

219-225.https://doi.org/doi:10.1111/j.1432-1033.1969.tb00677.x

re

Kildegaard, H. F., Fan, Y., Sen, J. W., Larsen, B., & Andersen, M. R. 2016. Glycoprofiling effects of media additives on IgG produced by CHO cells in fed-batch bioreactors.

lP

113(2), 359-366.https://doi.org/doi:10.1002/bit.25715

Kleczkowski, L. A. 1994. Glucose activation and metabolism through UDP-glucose

na

pyrophosphorylase in plants. Phytochemistry, 37(6),

ur

1507-1515.https://doi.org/https://doi.org/10.1016/S0031-9422(00)89568-0 Liu, Y., Yi, X., Zhuang, Y., & Zhang, S. 2015. Limitations in the process of transcription and

Jo

translation inhibit recombinant human chorionic gonadotropin expression in CHO cells.

J Biotechnol, 204, 63-69.https://doi.org/10.1016/j.jbiotec.2014.12.005

Livak, K. J., & Schmittgen, T. D. 2001. Analysis of Relative Gene Expression Data Using Real-Time Quantitative PCR and the 2−ΔΔCT Method. Methods, 25(4), 402-408.https://doi.org/https://doi.org/10.1006/meth.2001.1262

McKnight, G. L., Mudri, S. L., Mathewes, S. L., Traxinger, R. R., Marshall, S., Sheppard, P. O., & O'Hara, P. J. 1992. Molecular cloning, cDNA sequence, and bacterial expression of human glutamine:fructose-6-phosphate amidotransferase. 267(35), 25208-25212 Morgan, F. J., Birken, S., & Canfield, R. E. 1975. The amino acid sequence of human chorionic gonadotropin. The alpha subunit and beta subunit. 250(13), 5247-5258 Ran, F. A., Hsu, P. D., Wright, J., Agarwala, V., Scott, D. A., & Zhang, F. 2013. Genome

ro of

engineering using the CRISPR-Cas9 system. Nature Protocols, 8, 2281.https://doi.org/10.1038/nprot.2013.143

https://www.nature.com/articles/nprot.2013.143#supplementary-information

-p

Ronda, C., Pedersen, L. E., Hansen, H. G., Kallehauge, T. B., Betenbaugh, M. J., Nielsen, A.

re

T., & Kildegaard, H. F. 2014. Accelerating genome editing in CHO cells using CRISPR Cas9 and CRISPy, a web-based target finding tool. 111(8),

lP

1604-1616.https://doi.org/doi:10.1002/bit.25233

Thaysen-Andersen, M., & Packer, N. H. 2012. Site-specific glycoproteomics confirms that

na

protein structure dictates formation of N-glycan type, core fucosylation and branching.

ur

Glycobiology, 22(11), 1440-1452.https://doi.org/10.1093/glycob/cws110 Thaysen-Andersen, M., & Packer, N. H. 2014. Advances in LC-MS/MS-based glycoproteomics:

Jo

getting closer to system-wide site-specific mapping of the N- and O-glycoproteome.

Biochim Biophys Acta, 1844(9), 1437-1452.https://doi.org/10.1016/j.bbapap.2014.05.002

Varki, A. 1993. Biological roles of oligosaccharides: all of the theories are correct.

Glycobiology, 3(2), 97-130.https://doi.org/10.1093/glycob/3.2.97

Walsh, G. 2014. Biopharmaceutical benchmarks 2014. Nature Biotechnology, 32, 992.https://doi.org/10.1038/nbt.3040 https://www.nature.com/articles/nbt.3040#supplementary-information Yang, Z., Wang, S., Halim, A., Schulz, M. A., Frodin, M., Rahman, S. H., Vester-Christensen, M. B., Behrens, C., Kristensen, C., Vakhrushev, S. Y., Bennett, E. P., Wandall, H. H., & Clausen, H. 2015. Engineered CHO cells for production of diverse, homogeneous

ro of

glycoproteins. Nature Biotechnology, 33, 842.https://doi.org/10.1038/nbt.3280 https://www.nature.com/articles/nbt.3280#supplementary-information

Yin, B., Gao, Y., Chung, C.-y., Yang, S., Blake, E., Stuczynski, M. C., Tang, J., Kildegaard, H.

-p

F., Andersen, M. R., Zhang, H., & Betenbaugh, M. J. 2015. Glycoengineering of

re

Chinese hamster ovary cells for enhanced erythropoietin N-glycan branching and sialylation. 112(11), 2343-2351.https://doi.org/doi:10.1002/bit.25650

lP

Zhang, Y., Xie, X., Zhao, X., Tian, F., Lv, J., Ying, W., & Qian, X. 2018. Systems analysis of singly and multiply O-glycosylated peptides in the human serum glycoproteome via

na

EThcD and HCD mass spectrometry. Journal of proteomics, 170,

ur

14-27.https://doi.org/10.1016/j.jprot.2017.09.014 Zhou, J., Neidigh, J. L., Espinosa, R., LeBeau, M. M., & McClain, D. A. J. H. G. 1995. Human

Jo

glutamine: fructose-6-phosphate amidotransferase: characterization of mRNA and chromosomal assignment to 2p13. 96(1), 99-101.https://doi.org/10.1007/bf00214194

Zhu, H., Qiu, C., Ruth, A. C., Keire, D. A., & Ye, H. J. T. A. J. 2017. A LC-MS All-in-One Workflow for Site-Specific Location, Identification and Quantification of N-/OGlycosylation in Human Chorionic Gonadotropin Drug Products. 19(3),

846-855.https://doi.org/10.1208/s12248-017-0062-z Zhu, Z., Su, X., Go, E. P., & Desaire, H. 2014. New Glycoproteomics Software, GlycoPep Evaluator, Generates Decoy Glycopeptides de Novo and Enables Accurate False Discovery Rate Analysis for Small Data Sets. Analytical Chemistry, 86(18),

Jo

ur

na

lP

re

-p

ro of

9212-9219.https://doi.org/10.1021/ac502176n

Fig.2

ro of

-p

re

lP

na

ur

Jo Fig.1

ro of

-p

re

lP

na

ur

Jo Fig.3

ro of

-p

re

lP

na

ur

Jo Fig.4

ro of

-p

re

lP

na

ur

Jo Fig.5

ro of

-p

re

lP

na

ur

Jo Fig.6

Fig.8

ro of

-p

re

lP

na

ur

Jo Fig.7

ro of

-p

re

lP

na

ur

Jo Fig.9

ro of

-p

re

lP

na

ur

Jo

Table 1. Primer sequences used in this study Forward primer

Reverse primer

gGfat2

5’-tatctcgagaggcacagagagtcgtctga-3’

5’-cgctctagatgctccaatccccgtgtgt-3’

gUgp2

5’-atactcgagatgtctcaagatggtgc-3’

5’-atatctagatcagtggtccaagatgcg-3’

gC1galt1

5’-taactcgagatggaaattgagtatctg-3’

5’-cgctctagatcagtcattatcagaacc-3’

gGalnt1

5’-cgcctcgagatgagaaaatttgcatactg-3’

5’-atacccgggatttggtctcagaatatttc-3’

gOgt

5’-atactcgagatggcgtcttccgtgggcaa-3’

5’-cgctctagattatgctgactcagtgactt-3’

gGalnt1-T2A

5’-cgcctcgagatgagaaaatttgcatactg-3’

5’-tcaccgcatgttagcagacttcctctg

ro of

Name

ccctctccactgccgaatatttctggcagggt-3’

5’-ccagggtcgtccaatcatcc-3’

qPCR-Ugp2

5’-agccgttttctgcctgtcaa-3’

5’-ggaaaatgcggaagacgctg-3’

na

qPCR-C1galt1

lP

qPCR-Gfat2

re

aatcctggcccatagggaaattgagtatctg-3’

5’-cgccccgggtcagtcattatcagaacc-3’

-p

gT2A-C1galt1 5’-gtctgctaacatgcggtgacgtcgaggag

5’- atagcctcggagtacagcca-3’ 5’-cctgaaactgtgaggtgatc-3’ 5’-catgtccaaaggccctaagc-3’

5’-gccaccacctaaaaggcaac-3’

5’-aagccactgctgggacc-3’

qPCR-Ogt

5’- tggcaattaaacagaatcccc-3’

5’-tgtcacctgctgctaccaa-3’

ur

qPCR-Galnt1

Jo

qPCR-β actin Hex600

5’-gagggaaattgtgcgtgac-3’

5'-caggaaggaaggctggaa-3'

5’-gggcaccaaagcttctttcc-3’

5’-gagccaccttctcctcccta-3’

Table 2. ELISAs of the expression level of intact rhCG after transient transfection of Upg2 in CHO-hCG and CHO-hCG-GALNT1 cells at 96 h after addition of key precursors. Data are the mean and SD for 6 replicate samples. Values are the mean±SD.

add hCG concentration relative fold

none

add GalNAc GalNAc+Gal 1.4±0.04***

1.42±0.03**

1.57±0.08***

2.87±0.69***&###

3.68±0.6***&###

2.44±0.25***&###

2.1±0.25***&###

2.18±0.25***&###

1±0.05 (control)

CHO-hCG CHO-hCG-GALNT1

ro of

CHO-hCG-GALNT1+Ugp2

Jo

ur

na

lP

re

-p

***p<0.001 vs CHO-hCG control, ###p<0.001 vs CHO-hCG-GALNT1