MANNOLIPID DONOR SPECIFICITY OF GLYCOSYLPHOSPHATIDYLINOSITOL MANNOSYLTRANSFERASE-I (GPIMT-I) DETERMINED WITH AN ASSAY SYSTEM UTILIZING MUTANT CHO-K1 CELLS
Aw. Deluca et al., MANNOLIPID DONOR SPECIFICITY OF GLYCOSYLPHOSPHATIDYLINOSITOL MANNOSYLTRANSFERASE-I (GPIMT-I) DETERMINED WITH AN ASSAY SYSTEM UTILIZING MUTANT CHO-K1 CELLS, Glycobiology, 4(6), 1994, pp. 909-915
The microsomal enzyme glycosylphosphatidylinositol mannosyltransferase
I (GPLMT-I) catalyses the transfer of a mannosyl residue from beta-ma
nnosylphosphoryldolichol (beta-Man-P-Dol) to glucosamine-alpha(1,6)(ac
yl)phosphatidylinositol (GlcN-aPI) to form Man alpha(1,4)GlcN-aPI (Man
GlcN-aPI), an intermediate in glycosylphosphatidylinositol (GPI) synth
esis, While the transfer of [H-3]mannosyl units to endogenous GlcN-aPI
was not seen when membrane fractions from normal Chinese hamster ovar
y (CHO) K1 cells were incubated with exogenous [H-3]Man-P-Dol, GPIMT-I
activity could be characterized with an in vitro enzyme assay system
employing membrane fractions from Lec15 or Lec35 cells, These CHO cell
mutants apparently contain elevated levels of endogenous GlcN-aPI due
to the inability to synthesize (Lec15) or utilize (Lec35) beta-Man-P-
Dol in vivo. The presence of a saturated a-isoprene unit in the dolich
yl moiety is required for optimal GPIMT-I activity since beta-mannosyl
phosphorylpolyprenol (beta-Man-P-Poly), which contains a fully unsatur
ated polyisoprenyl chain, was only 50% as effective as beta-[H-3]Man-P
-Dol as a mannosyl donor, When beta-[H-3]Man-P-Dol and alpha-[H-3]Man-
P-Dol were compared as substrates, GPIMT-I exhibited a strict stereosp
ecificity for the mannolipid containing the beta-mannosyl-phosphoryl l
inkage. beta-[H-3]Man-P-dolichols containing 11 or 19 isoprenyl units
were equally effective substrates for GPIMT-I. Membrane fractions from
Lec 9, a CHO mutant that apparently lacks polyprenol reductase activi
ty and synthesizes very little beta-Man-P-Dol, but accumulates beta-Ma
n-P-Poly, synthesized no detectable Man-GlcN-aPI when incubated with b
eta-[H-3]Man-P-Dol in vitro. This indirect assay suggests that GlcN-aP
I does not accumulate in Lec 9 cells, possibly because it is mannosyla
ted via beta-Man-P-Poly, or perhaps the small amount of Man-P-Dol form
ed by the mutant in vivo. These experiments demonstrate that: (i) memb
rane fractions from the CHO mutants, Lec15 and Lec35, provide a useful
system for the characterization of GPIMT-I activity; (ii) GPIMT-I uti
lizes Man-P-Dol or Man-P-Poly as direct mannosyl donors for Man-GlcN-a
PI synthesis, although Man-P-Poly is used less efficiently; and (iii)
the transfer of mannosyl residues from Man-P-Dol to GlcN-aPI is stereo
specific for mannolipid substrates containing mannosyl-phosphoryl link
ages of the beta-configuration.