A Chinese hamster ovary cell mutant F2A8 utilizes polyprenol rather than dolichol for its lipid-dependent asparagine-linked glycosylation reactions.

J Stoll,A G Rosenwald,S S Krag

doi:10.1016/s0021-9258(18)38038-4

Abstract

Previous results suggested that F2A8, a glycosylation mutant of Chinese hamster ovary cells, had a lower amount of dolichyl phosphate available for asparagine-linked glycosylation reactions relative to parental cells. The steady-state amounts and identities of polyisoprenoid lipids were determined by incubating F2A8, its parental cell line B4-2-1, and wild-type Chinese hamster ovary cells for 24 h with [2-3H]mevalonate. The neutral lipids, ubiquinone, cholesterol, and cholesteryl esters, which were the most highly labeled from [3H]mevalonate, were labeled equally in all three cell types. In wild-type and B4-2-1 cells, mevalonate incorporation into the anionic glycosylated and phosphorylated derivatives of dolichol was 10-fold higher than into the neutral free dolichol and dolichyl esters. In contrast, in F2A8 cells, label accumulated in neutral polyisoprenol lipids, so that the ratio of neutral to anionic lipids was 1:1 rather than 1:10. In wild-type and B4-2-1 cells, the polyisoprenoid found as free alcohol and in phosphorylated and glycosylated forms was shown by high pressure liquid chromatography using a silica column to be primarily dolichol, a polyisoprenol that has a saturated terminal isoprene unit. In contrast, in F2A8 cells the polyisoprenoid found primarily as the free alcohol and in phosphorylated and glycosylated forms appeared to be completely unsaturated polyprenol. The distribution of chain lengths of the labeled polyisoprenols of F2A8, B4-2-1, and wild-type cells was the same as determined by high pressure liquid chromatography using a reverse-phase column, with the predominant chain length being 19 isoprene units. These results combined with our previous studies on the phenotype of the F2A8 mutant indicate that the unsaturated polyprenyl phosphate derivatives do not function as well as dolichyl phosphate derivatives in cellular glycosylation reactions.

Highlights

Previous results suggested that F2A8, a glycosyla- the most abundant polyisoprenol found in mammalian tissues tion mutant of Chinese hamster ovarycells, had a lower
Numerous comparisons of the chemical and contrast, in F2A8 cells, label accumulated in neutral chromatographic properties of glycosylatedendogenous lipids polyisoprenol lipids, so that the ratio of neutral to to those of glycosylated dolichyl phosphate were made, and anionic lipids was 1:l rather than 1 : l O
In wild-type the results substantiated the role of dolichol derivatives in and B4-2-1cells, the polyisoprenoid found as free al- glycosylation

Summary

EXPERIMENTAL PROCEDURES

Chemicals and Radiochemic~ls-[2-~H]Mevalonolactone(1.28 Ci/ mmol) was obtained from Amersham Corp. ['4C]Mannosylphosphoryldolichol was prepared as described previously [43]. The mevinolin was prepared as a stock of 4 mg/ml in 0.1 N NaOH by heating for 2 h at 50 "C This particular ratio of mevinolin to mevalonate gave the maximum incorporation into dolichol while maintaining the viability of the cells.' Following the incubation, the labeling medium wasremoved from the dishes, and thecells were washed once with phosphate-buffered saline, pH 7.4. When the labeled amphipathic lipids extracted into chloroform:methanol:water (10103) were analyzed, the column was washed with 5 column volumes of glacial acetic acid, methanol, chloroform:methanol (1:1), and ch1oroform:methanol:water (1010:3). The reaction was terminated by adding 2 ml of ch1oroform:methanol [21]; radioactivity that partitioned into the chloroform phase after addition of 0.9% saline was purified by gel filtration chromatography on Fractogel resin as described below. Chromatographic Procedures-Samples dissolved in chloroform:methanol(2:1) were applied to a 0.9 X 43-cm column of Fractogel HW-4OF (EM Science, Cherry Hill, NJ), equilibrated, and runin ch1oroform:methanol (2:l)containing 50 mM ammonium acetate. Cells were incubated with [2-3H]mevalonafoter 24 h and extracted as described under “Experimental Procedures.” The molecules par-

RESULTS

I-ilI1

Findings

DISCUSSION