Analysis and metabolic engineering of lipid-linked oligosaccharides in glycosylation-deficient CHO cells

Abstract

Glycosylation-deficient Chinese Hamster Ovary (CHO) cell lines can be used to expand our understanding of N-glycosylation pathways and to study Congenital Disorders of Glycosylation, diseases caused by defects in the synthesis of N-glycans. The mammalian N-glycosylation pathway involves the step-wise assembly of sugars onto a dolichol phosphate (P-Dol) carrier, forming a lipid-linked oligosaccharide (LLO), followed by the transfer of the completed oligosaccharide onto the protein of interest. In order to better understand how deficiencies in this pathway affect the availability of the completed LLO donor for use in N-glycosylation, we used a non-radioactive, HPLC-based assay to examine the intermediates in the LLO synthesis pathway for CHO-K1 cells and for three different glycosylation-deficient CHO cell lines. B4-2-1 cells, which have a mutation in the dolichol phosphate-mannose synthase (DPM2) gene, accumulated LLO with the structure Man(5)GlcNAc(2)-P-P-Dol, while MI8-5 cells, which lack glucosyltransferase I (ALG6) activity, accumulated Man(9)GlcNAc(2)-P-P-Dol. CHO-K1 and MI5-4 cells both produced primarily the complete LLO, Glc(3)Man(9)GlcNAc(2)-P-P-Dol, though the relative quantity was lower in MI5-4. MI5-4 cells have reduced hexokinase activity which could affect the availability of many of the substrates required for LLO synthesis and, consequently, impair production of the final LLO donor. Increasing hexokinase activity by overexpressing hexokinase II in MI5-4 caused a decrease in the relative quantities of the incomplete LLO intermediates from Man(5)GlcNAc(2)-PP-Dol through Glc(1)Man(9)GlcNAc(2)-PP-Dol, and an increase in the relative quantity of the final LLO donor, Glc(3)Man(9)GlcNAc(2)-P-P-Dol. This study suggests that metabolic engineering may be a useful strategy for improving LLO availability for use in N-glycosylation.

Figure 1. The LLO synthesis pathway

On the cytoplasmic face of the ER, dolichol phosphate is modified by glycosyltransferases, which utilize nucleotide-activated sugar donors, to form Man₅GlcNAc₂-PP-Dol. This structure is then flipped across the ER membrane to the luminal face, where additional glycosyltransferases use lipid-activated monosaccharides as sugar donors in order to form the final LLO donor: Glc₃Man₉GlcNAc₂-PP-Dol. The oligosaccharide is then transferred from the dolichol donor onto the polypeptide by the OST enzyme complex. The affected enzymes for B4-2-1 (DPM2), MI8-5 (ALG6), and MI5-4 (hexokinase) are indicated in gray italics.

Figure 2. HPLC profiles of LLO distributions for parental and mutant CHO cell lines

LLO were purified from (A) CHO-K1, (B) B4-2-1, (C) MI8-5, and (D) MI5-4 and then the oligosaccharides were cleaved from the lipids, fluorescently labeled, and detected via HPLC.

Figure 3. Western blot confirming expression of hexokinase II in MI5-4-hexoII

293F cell lysate was used as a positive control, untransfected MI5-4 cell lysate was used as a negative control, and β-actin was used to control for protein loading.

Figure 4. Comparison of relative LLO levels in MI5-4 vs. MI5-4-hexoII

Relative LLO levels (mean ± standard error, n=3) were quantified using the HPLC data by integrating the area under each peak. M5 = Man₅GlcNac₂, G3M9 = Glc₃Man₉GlcNac₂. Significance: *, p<0.05; **, p<0.005; ***, p<0.001.