Glycobiology vol. 8 no. 12 pp. 1195–1205, 1998
Transbilayer movement of Glc-P-dolichol and its function as a glucosyl donor:
protein-mediated transport of a water-soluble analog into sealed ER vesicles from pig brain
Jeffrey S.Rush, Klaus van Leyen1, Ouathek Ouerfelli1,
Beata Wolucka2 and Charles J.Waechter3
Department of Biochemistry, University of Kentucky College of Medicine,
Lexington, KY 40536, USA, 1Cellular Biochemistry and Biophysics,
Sloan-Kettering Institute, New York, NY USA and 2Department of
Applied Chemistry and Bioindustries, University of Louvain,
Louvain-la-Neuve, Belgium
Received on March 19, 1998; revised on May 14, 1998;
accepted on May 14, 1998
3To
whom correspondence should be addressed
The results described in the accompanying article support the
model in which glucosylphosphoryldolichol (Glc-P-Dol) is
synthesized on the cytoplasmic face of the ER, and functions
as a glucosyl donor for three Glc-P-Dol:Glc0–2Man9-GlcNAc2P-P-Dol glucosyltransferases (GlcTases) in the lumenal compartment. In this study, the enzymatic synthesis and structural
characterization by NMR and electrospray-ionization tandem
mass spectrometry of a series of water-soluble β-Glc-P-Dol
analogs containing 2–4 isoprene units with either the cis- or
trans-stereoconfiguration in the β-position are described. The
water-soluble analogs were (1) used to examine the stereospecificity of the Glc-P-Dol:Glc0–2Man9GlcNAc2-P-P-Dol
glucosyltransferases (GlcTases) and (2) tested as potential substrates for a membrane protein(s) mediating the transbilayer
movement of Glc-P-Dol in sealed ER vesicles from rat liver
and pig brain. The Glc-P-Dol–mediated GlcTases in pig brain
microsomes utilized [3H]Glc-labeled Glc-P-Dol10, Glc-P(ω,c)Dol15, Glc-P(ω,t,t)Dol20, and Glc-P-(ω,t,c)Dol20 as glucosyl donors with [3H]Glc3Man9GlcNAc2-P-P-Dol the major
product labeled in vitro. A preference was exhibited for
C15–20 substrates containing an internal cis-isoprene unit in
the β-position. In addition, the water-soluble analog, GlcP-Dol10, was shown to enter the lumenal compartment of
sealed microsomal vesicles from rat liver and pig brain via a
protein-mediated transport system enriched in the ER. The
properties of the ER transport system have been characterized.
Glc-P-Dol10 was not transported into or adsorbed by synthetic
PC-liposomes or bovine erythrocytes. The results of these
studies indicate that (1) the internal cis-isoprene units are
important for the utilization of Glc-P-Dol as a glucosyl donor
and (2) the transport of the water-soluble analog may provide
an experimental approach to assay the hypothetical
“flippase proposed to mediate the transbilayer movement of
Glc-P-Dol from the cytoplasmic face of the ER to the lumenal
monolayer.
Key words: Glc-P-Dol synthesis/ER/“flippase
 1998 Oxford University Press
Introduction
Glucosylphosphoryldolichol (Glc-P-Dol) functions as the direct
glucosyl donor for the synthesis of the triglucosyl cap of
Glc3Man9GlcNAc2-P-P-Dol, the precursor oligosaccharide
donor in the eukaryotic N-glycosylation pathway (Kornfeld and
Kornfeld, 1985; Hirschberg and Snider, 1987; Waechter, 1989;
Cummings, 1992). The results of the preceding study (Rush and
Waechter, 1998) support the topological model proposed by
Hirschberg and Snider (1987) in which Glc-P-Dol is synthesized
on the cytosolic face of the ER and then functions as the glucosyl
donor for three lipid-mediated glucosylation reactions forming
the terminal triglucosyl cap of the dolichol-bound precursor
oligosaccharide on the lumenal side of the ER in pig brain.
In view of evidence obtained from different experimental
approaches indicating that the unassisted transbilayer movement
of polyisoprenol-linked sugars in synthetic liposomes is extremely
slow (Hanover and Lennarz, 1978; McCloskey and Troy, 1980),
it is quite possible that the transverse diffusion of
Man5GlcNAc2-P-P-Dol (Flippase I), Man-P-Dol (Flippase II),
and Glc-P-Dol (Flippase III) from the site of synthesis on the
cytoplasmic face to the lumenal leaflet is mediated by ER
membrane proteins (Figure 1). We have previously used mannosylphosphorylcitronellol (Man-P-Dol10), a water-soluble analog
of mannosylphosphoryldolichol (Man-P-Dol), to demonstrate the
presence of a stereoselective ER protein(s) that facilitated the
transbilayer movement of the analog in sealed microsomal
vesicles from liver (Rush and Waechter, 1995).
To extend this approach to investigate membrane proteins that
might play a role in the transbilayer movement of Glc-P-Dol in
the ER, we have enzymatically synthesized a series of water-soluble
analogs of Glc-P-Dol, by incubating hen oviduct microsomes
with UDP-glucose and stereochemically defined, short-chain
dolichyl monophosphates (Dol-P). The short chain Dol-Ps were
synthesized chemically by the procedures developed by Jaenicke
and colleagues (Jaenicke and Siegmund, 1986, 1989; Jaenicke et al.,
1991; Berendes and Jaenicke, 1992).
To evaluate their potential to be recognized by a membrane
protein facilitating the transverse diffusion of Glc-P-Dol, the
water-soluble analogs were tested as substrates for the pig brain
Glc-P-Dol:Glc0–2Man9GlcNAc2-P-P-Dol glucosyltransferases
(GlcTases). These reactions are believed to be catalyzed by three
separate glucosyltransferases (Runge et al., 1984; Runge and
Robbins, 1986; D’Souza-Schorey and Elbein, 1993; Stagljar et al.,
1994; Zufferey et al., 1995). This enzymological comparison was
conducted to determine if they were effective glucosyl donors,
and to determine if the cis-isoprene unit in the β-position, as well
as the saturated α-isoprene unit (D’Souza-Schorey et al., 1994),
was recognized by the GlcTases. All of the water-soluble
Glc-P-Dol analogs were found to be utilized as substrates by the
Glc-P-Dol-mediated GlcTases with a slight preference shown for
the presence of a cis-isoprene unit in the β-position.
1195
J.S.Rush et al.
Fig. 1. Hypothetical ER proteins proposed to mediate the transbilayer
movement of Man5GlcNAc2-P-P-dolichol (Flippase I), Man-P-dolichol
(Flippase II), and Glc-P-dolichol (Flippase III).
Since the water-soluble analogs were utilized as glucosyl
donors by the GlcTases, it was plausible that their structural
resemblance to Glc-P-Dol may also be sufficient for recognition
by the hypothetical Glc-P-Dol flippase (III) proposed in Figure 1.
To address this question, the transport of Glc-P-Dol10 by sealed
microsomal vesicles from rat liver and pig brain was assessed.
The results of the transport studies document the presence of a
membrane protein(s) that mediates the transbilayer movement of
Glc-P-Dol10 into sealed microsomal vesicles from rat liver and
pig brain. The potential use of this transport system as an in vitro
assay for a membrane protein involved in the “flip-flopping of
Glc-P-Dol in the ER is discussed. Parts of this study were reported
in preliminary form (Rush et al., 1997a)
Results
Enzymatic synthesis and structural characterization of
short-chain, water-soluble analogs of Glc-P-Dol
The results in the preceding article (Rush and Waechter, 1998) are
consistent with the topological model of Hirschberg and Snider
(1987) in which Glc-P-Dol functions as a glucosyl donor for the
synthesis of Glc3Man9GlcNAc2-P-P-Dol in the lumenal compartment of the ER after diffusing transversely (“flip-flopping) from the
site of synthesis on the cytoplasmic face. It is therefore possible that
the transbilayer movement of the polar headgroup of the glucolipid
is mediated by an ER protein (Flippase III, Figure 1).
In a previous study, Man-P-Dol10, was used as a substrate for
transport studies to detect ER protein(s) mediating the transmembrane movement of Man-P-Dol (Rush and Waechter, 1995).
To extend the use of water-soluble analogs as a potential approach
to assay the hypothetical “flippases, [3H]Glc-P-citronellol
(Glc-P-Dol10) and other short-chain stereoisomers of Glc-P-Dol
have been synthesized enzymatically by incubating hen oviduct
microsomes with UDP-[3H]glucose and a series of stereochemically
defined short-chain isoprenyl monophosphates. The enzymatic
glucosylation of the isoprenyl phosphates is presumably catalyzed
by Glc-P-Dol synthase associated with the oviduct microsomes.
The water-soluble analogs have been purified and characterized
as β-Glc-P-dolichols by several chemical criteria. First, the synthesis
of the enzymatic products was dependent on the addition of each
exogenous isoprenyl phosphate to the reaction mixture. Following
chromatography on Silica Gel G thin layer plates, all of the
1196
Fig. 2. Negative-ion electrospray-ionization tandem mass spectrometry of
synthetic Glc-P-Dol10. The product ion spectrum of the [M - H]–
deprotonated molecule.
enzymatically synthesized, water-soluble Glc-P-dolichols stained
positively with the phospholipid spray reagent of Dittmer and Lester
(1964), and were detected as yellowish-green spots characteristic of
isoprenoid compounds by an anisaldehyde spray reagent (Dunphy
et al., 1967). Also consistent with the structure of Glc-P-Dols,
[3H]glucose and the appropriate Dol-P were liberated by mild acid
hydrolysis (0.05 N HCl in 50% isopropanol, 50C, 60 min).
Finally, the predicted molecular weights of the analogs and
anomeric configuration of the glucosyl 1-phosphate bonds were
confirmed by electrospray-ionization tandem mass spectrometry
(Figure 2, Table I). The product ion spectrum of the deprotonated [M
- H]– molecule at m/z 397 for Glc-P-Dol10 contained a prominent
fragment ion at m/z 235 (relative intensity, 64%) corresponding to
[Dol10-HPO4]–. The major fragment ion at m/z 235 arises from a
fragmentation pathway characteristic of glycosyl-P-polyisoprenols
with trans-hydroxyls at the 1 and 2 carbons of the glycosyl residue
(Wolucka et al., 1996, 1998). Hexosyl-P-dolichols with cis-hydroxyls at the 1 and 2 positions yield prominent fragment ions
corresponding to [Dol-PO4-(C2H3O)]– generated by cleavage
across the hexose ring and to [M - H2O -H]– dehydration products.
The absence of the [Dol10-PO4-(C2H3O)]– ion (m/z 277) and of the
[M - H2O - H]– (relative intensity 0.5%), dehydration product (m/z
379) (Figure 2, Table I), confirms the structure of the enzymatic
product as β-Glc-P-Dol10.
The deprotonated [M - H]– molecules observed in the
electrospray-ionization spectra of the water-soluble Glc-P-dolichols
were m/z 465 for Glc-P-(ω,c)Dol15 and m/z 533 for
Glc-P-(ω,t,t)Dol20 and Glc-P-(ω,t,c)Dol20 (Table I). The only
prominent product ions observed were m/z 303 (relative intensity,
100%) for Glc-P-(ω,c)Dol15 and m/z 371 for Glc-P-(ω,t,t)Dol20 and
Glc-P-(ω,t,c)Dol20 (relative intensity, 100%), and corresponded
to the respective dolichyl phosphates. The relatively minor
product ions arising from the respective [Dol-PO4-(C2H3O)]–
(relative intensities 5%) and the absence of the [M - H2O - H]–
dehydration products in the tandem mass spectrometry of the
water-soluble Glc-P-Dols, confirm that the compounds are βstereoisomers.
The structural relationships between Glc-P-Dol95 and the
water-soluble analogs, Glc-P-Dol10, Glc-P-(ω,c)Dol15,
Glc-P-(ω,t,t)Dol20 and Glc-P-(ω,t,c)Dol20, are illustrated in
Figure 3. The water-soluble analogs contain an ω-isoprene unit
and the reduced α-isoprene, characteristic of dolichols, but are
considerably less hydrophobic due to lacking 15–17 of the
intervening isoprene units found in mammalian C95-dolichol.
Transbilayer movement of Glc-P-dolichol
Table I. Negative-ion electrospray-ionization tandem mass spectrometry of water-soluble Glc-P-dolichols
Compound (MW)
[M - H2O -
H]–
Relative intensitiesa
[Dol-HPO4]–
[Dol-PO4-(C2H3O)]–
[PO3]–
Glc-P-Dol10
–
64
–
7
(398)
(379)
(235)
(277)
(79)
(ω,c)Glc-P-Dol15
–
100
5
25
(466)
(447)
(303)
(345)
(79)
(ω,t,t)Glc-P-Dol20
–
100
4
14
(534)
(515)
(371)
(413)
(79)
(ω,t,c)Glc-P-Dol20
–
100
4
13
(534)
(515)
(371)
(413)
(79)
Negative-ion electrospray-ionization tandem mass spectra were collected as described in Materials and methods. Only
signals of relative abundances greater than 2% are given. MW, Molecular weight of each analog is given in parentheses.
aRelative intensities of fragments (m/z) obtained from the [M - H]– deprotonated molecule at 20 V collision-offset voltage, %
Fig. 3. Structural relationship between short-chain water-soluble analogs and
Glc-P-Dol95.
Enzymatic transfer of [ 3H]glucose from the water-soluble
analogs of Glc-P-Dol into [ 3H]Glc3 Man9 GlcNAc2 -P-P-Dol in
pig brain microsomes
To verify that the structural features of the water-soluble analogs
were recognized by proteins or enzymes interacting with
Glc-P-Dol, their ability to serve as glucosyl donors for the pig
brain microsomal GlcTase(s) catalyzing the addition of the three
terminal glucose units in Glc3Man9GlcNAc2-P-P-Dol synthesis
was tested. When [3H]Glc-labeled Glc-P-Dol10, Glc-P-(ω,c)Dol15,
Glc-P-(ω,t,c)Dol20, and Glc-P-(ω,t,t)Dol20 were incubated with
pig brain microsomes, [3H]glucose was transferred to endogenous
Glc0–2Man9GlcNAc2-P-P-Dol acceptor substrates (Figure 4). The
enzymatic properties of the GlcTase(s) utilizing the water-soluble
Glc-P-Dols as substrates were identical to the properties observed
for the activities utilizing the natural long-chain (C95) glucosyl
donor (Waechter and Scher, 1978).
A comparison using the various Glc-P-Dols as substrates
indicated that the rate of glucosylation and the affinity of the
Glc-P-Dol–mediated GlcTases for the glucosyl donors increased
when the isoprenoid chain was increased in length from 2 to 3
isoprene units (Figure 4). Increasing the chain length of the
isoprenyl moiety by the addition of a fourth isoprene unit had only
a minor effect on both the maximal velocity and the affinity of the
GlcTases for the glucosyl donors. Furthermore, Glc-P-(ω,c)Dol15
and Glc-P-(ω,t,c)Dol20 (Figure 4, solid circles and open
triangles), which contain internal cis-isoprene units in the
β-position, were slightly better substrates than Glc-P-(ω,t,t)Dol20
(Figure 4, open squares).
Fig. 4. Comparison of short chain, water-soluble analogs of Glc-P-Dol as
substrates for Glc-P-Dol:Glc0–2Man9GlcNAc2-P-P-Dol GlcTase activity in
pig brain microsomes. Reaction mixtures for the assay of the lipid-mediated
GlcTases contained 50 mM Tris–HCl (pH 7.4), 0.075 mM sucrose, 10 mM
EDTA, 0.75 mg of pig brain microsomal protein and the indicated
concentration of [3H]Glc-P-Dol10 (open circles), [3H]Glc-P-(ω,c)Dol15
(solid circles), [3H]Glc-P-(ω,t,t)Dol20 (open squares), or
[3H]Glc-P-(ω,t,c)Dol20 (open triangles). Initial enzymatic rates were
determined as described in Materials and methods and the data were
analyzed by linear regression using the Sigma Plot Scientific Graph System
4.1 (Jandel Scientific, Corte Madera, CA).
The lipid-bound oligosaccharides synthesized by pig brain
microsomes during incubation with the water-soluble [3H]GlcP-Dols were sensitive to mild acid hydrolysis (Lucas et al., 1975)
releasing [3H]oligosaccharides with chromatographic properties
identical to authentic Glc1–3Man9GlcNAc2. The major enzymatic
product formed in the Glc-P-Dol:Glc0–2Man9GlcNAc2-P-P-Dol
1197
J.S.Rush et al.
Fig. 6. Experimental strategy for assaying the transport of Glc-P-Dol10 in
sealed ER vesicles.
Fig. 5. Characterization of lipid-linked [3H-Glc]oligosaccharides synthesized
in GlcTase reactions. The free [3H]oligosaccharides were released from the
enzymatic products formed with either Glc-P-Dol10 (A), Glc-P-(ω,c)Dol15
(B), Glc-P-(ω,t,c)Dol20 (C), or Glc-P-(ω,t,t)Dol20 (D) serving as the glucosyl
donor and analyzed by high pH anion-exchange chromatography as
described in Materials and methods. The sodium acetate gradient is depicted
with the dotted trace in (A).
GlcTase reactions using water-soluble Glc-P-Dols as substrates
coeluted with Glc3Man9GlcNAc2 by high pH anion-exchange
chromatography on a PA1 strong anion exchange column
(Figure 5a–d). Minor amounts of oligosaccharide products with
1–2 glucosyl residues were also present. These results show that
at least the Glc-P-Dol-mediated GlcTase adding the third
glucosyl unit, and probably all three, utilize the water-soluble
analogs as glucosyl donors.
Sealed microsomal vesicles from rat liver and pig brain contain
a transport system that facilitates the uptake of Glc-P-Dol10
Since the water-soluble analogs were utilized as substrates by the
brain GlcTase(s), it seemed reasonable that the same structural
features might be recognized by a protein involved in the
translocation of the Glc-P headgroup from the cytosolic face of
the ER to the lumenal compartment. Basically, the same
experimental strategy reported for Man-P-Dol10 (Rush and
Waechter, 1995) was followed as illustrated in Figure 6. As seen
1198
in Figure 7, β-[3H]Glc-P-Dol10 was transported into sealed
microsomal vesicles from rat liver and pig brain. Uptake of the
water-soluble analog reached a maximum value within 2 min at
19C with ∼3% of the radiolabeled Glc-P-Dol10 internalized.
Assuming that Glc-P-Dol10 transport occurs by facilitated
diffusion, an intravesicular volume of ∼3 µl/mg membrane
protein for rat liver microsomes can be calculated. This estimate
is similar to a previous value calculated by Bishop and Bell (1985)
based on diC4PC transport.
Due to the relatively higher rate of uptake observed with rat
liver microsomes, these preparations were used to characterize
the transport system more extensively. Several properties of the
Glc-P-Dol10 uptake system were found to be similar to the
Man-P-Dol10 transporter system (Rush and Waechter, 1995). First,
Glc-P-Dol10 uptake requires an intact permeability barrier. The
transport of the water-soluble analog (Figure 8, solid circles) is lost
in close parallel with Man 6-P phosphatase latency (Figure 8, open
circles). Approximately 50% of Glc-P-Dol10 transport activity was
lost when intact rat liver vesicles were incubated with 0.1% (wt/vol)
Triton X-100. In addition, uptake was saturable with respect to
Glc-P-Dol10 concentration (Figure 9) with transport reaching a
half-maximal rate at ∼1 mM Glc-P-Dol10. Vesicular integrity was
unaffected by this concentration of Glc-P-Dol10.
Uptake of the water-soluble analog is also apparently reversible
since, approximately 85% of the internalized Glc-P-Dol10 is lost
in a time-dependent manner when preloaded rat liver vesicles
were diluted with 25 volumes of isotonic buffer (10 mM
Tris-HCl, pH 7.4, 0.25 M sucrose) at 20C (data not included).
The effects of various ionophores were tested to determine if
Glc-P-Dol10 transport required an electrochemical gradient, but
no effects on the uptake system were seen when valinomycin,
monensin, FCCP or CCCP were present. The transport of
Glc-P-Dol10 was stimulated significantly (∼2–3 fold) by the
addition of 5–10 mM ATP, but not by the addition of the
nonhydrolyzable analog, γ-S-ATP (data not included). Other
nucleoside triphosphates as well as AMP and ADP were
significantly less effective in stimulating Glc-P-Dol10.
The specificity of the Glc-P-Dol10 transport system was
investigated by testing a variety of structurally related compounds
as potential competitive inhibitors of uptake. β-Methyl glucose,
p-nitrophenyl-β-glucoside, UDP-glucose, glucose 1-P, glucose
6-P, glucose, and glucose 1,6-diphosphate had no appreciable
effect on Glc-P-Dol10 uptake when added in a 100-fold molar
excess. The failure of an excess of UDP-Glc and Glc 6-P to
compete for transport is good evidence that Glc-P-Dol10 is not
Transbilayer movement of Glc-P-dolichol
Fig. 7. Time course for the uptake of β-[3H]Glc-P-Dol10 by sealed
microsomal vesicles from rat liver and pig brain. Transport assay mixtures
contained 10 mM Tris-HCl (pH 7.4), 0.25 M sucrose, liver, or brain
microsomal vesicles (93 µg of membrane protein) and 0.1 mM
β-[3H]Glc-P-Dol10 (25 c.p.m./pmol) in a total volume of 0.01 ml. Following
incubation at 19C for the indicated periods of time, the amount of
radiolabeled Glc-P-Dol10 transported into either rat liver (solid circles) or pig
brain (open circles) microsomal vesicles was determined as described in
Materials and methods.
Fig. 9. β-[3H]Glc-P-Dol10 transport system in sealed microsomal vesicles is
saturable. Glc-P-Dol10 transport assays contained 10 mM Tris-HCl (pH 7.4),
0.25 M sucrose, rat liver ER vesicles (93 µg membrane protein), and the
indicated concentration of [3H]Glc-P-Dol10 (5.2 c.p.m./pmol) in a total
volume of 0.01 ml. Following incubation for 30 s at 19C, the uptake of
[3H]Glc-P-Dol10 was determined as described in Materials and methods. An
apparent Km for uptake of 1 mM was calculated from the double-reciprocal
plot in the inset. The data were analyzed by second order nonlinear
regression using the Marquardt-Levenberg algorithm provided by the
Sigmaplot Scientific Graph System (Jandel Scientific). The data in the inset
were analyzed by linear regression using the Sigma Plot Scientific Graph
System 4.1 (Jandel Scientific).
being nonspecifically transported by the transporters for these
glucose-containing compounds. When the other short-chain
Glc-P-Dol analogs used in the enzymological study described
above were surveyed in the transport assay, only very modest
differences were found in the rate of uptake. β-Man-P-Dol10
when added at a 100-fold excess reduced Glc-P-Dol10 uptake by
less than 30%, suggesting that Flippases II and III (Figure 1)
could be separate ER proteins.
Glc-P-Dol10 transport system is protein-mediated and enriched
in ER vesicles
Fig. 8. Uptake of β-[3H]Glc-P-Dol10 by sealed microsomes requires an
intact permeability barrier. Transport assay mixtures contained 10 mM
Tris-HCl (pH 7.4), 0.25 M sucrose, rat liver ER vesicles (290 µg membrane
protein), 3 µM [3H]Glc-P-Dol10 (962 c.p.m./pmol), and the indicated
concentration of Triton X-100 in a total volume of 0.02 ml. Man 6-P
phosphatase reaction mixtures were identical to the transport reactions
except that [3H]Glc-P-Dol10 was replaced by 1 mM [3H] Man 6-P. The
fraction of latent Man 6-P phosphatase activity (open circles) at each
detergent concentration, and the amount of radiolabeled Glc-P-Dol10 (solid
circles) transported into microsomal vesicles during a 30 s incubation at
23C was determined as described in Materials and methods. The data were
analyzed by nonlinear regression using the Marquardt-Levenberg algorithm
provided by the Sigmaplot Scientific Graph System (Jandel Scientific).
To determine if Glc-P-Dol10 uptake by rat liver ER microsomes
involves a membrane protein, the effect of trypsin treatment on
transport activity was evaluated. Glc-P-Dol10 transport activity
decreased in a time-dependent manner during incubation with
trypsin added at 2.5 mg/ml (Figure 10, solid circles). Glc-P-Dol10
uptake was not affected during control incubations in the absence
of trypsin (open circles) or in which protease activity was blocked
by a 10-fold excess of trypsin inhibitor (soy bean) (open
triangles), excluding the possibility that the loss of uptake was due
to nonspecific inactivation of the transport system. In addition,
Man 6-P phosphatase latency (solid triangles) was unaffected
during the incubation period at 19C, indicating that the vesicular
permeability barrier remained intact during the experiment.
The requirement for a membrane protein(s) in Glc-P-Dol10
uptake was examined further by testing the ability of the analog
to associate with other bilayer systems. Under conditions that
were optimal for uptake by liver microsomes, Glc-P-Dol10 was
actively transported by microsomal vesicles, but not by bovine
1199
J.S.Rush et al.
Fig. 10. Transport of β-[3H]Glc-P-Dol10 into sealed microsomes is
inactivated by exposure to trypsin. Glc-P-Dol10 transport assays contained
10 mM Tris-HCl (pH 7.4), 0.25 M sucrose, rat liver ER vesicles (170 µg
membrane protein), 3 µM [3H]Glc-P-Dol10 (962 c.p.m./pmol) in a total
volume of 0.02 ml. Microsomal vesicles were preincubated without (open
circles) or with trypsin (2.5 mg/ml) (solid circles) or the combination of
trypsin and soybean trypsin inhibitor (25 mg/ml) (open triangles) for the
indicated periods of time at 19C. Following the preincubation period,
Glc-P-Dol10 uptake was assayed for 30 s at 19C as described in Materials
and methods. Vesicular integrity was unaffected by incubation with trypsin
for up to 240 min (solid triangles).
erythrocytes (Figure 11). The analog did not adsorb to or enter
synthetic PC-liposomes, indicating that it did not simply adsorb
to the outer leaflet of the microsomal vesicles. Glucose was
actively transported under the same incubation conditions by
bovine erythrocytes, but not microsomal vesicles or PC liposomes
(Figure 11B), confirming the integrity of the RBC transport
system.
When Glc-P-Dol10 transport activity was compared in a
limited number of rat liver subcellular fractions, transport activity
was found to be enriched in the ER, the site of lipid intermediate
synthesis, relative to Golgi vesicles and intact mitochondria
(Table II). Although the Golgi and mitochondrial preparations
take up a measurable amount of Glc-P-Dol10, the transport
activity present in these fractions can be accounted for by
cross-contamination with ER vesicles, assessed by the Man 6-P
phosphatase activity. The ER-enriched fraction, on the other
hand, contains very low amounts of the other two marker enzyme
activities, indicating that it is relatively free of Golgi vesicles or
mitochondria. All of these results provide evidence for the
presence of a membrane protein that facilitates the transbilayer
movement of Glc-P-Dol10 from the outside to the lumenal
compartment of sealed ER vesicles.
Transfer of [ 3H]glucose from β-[ 3H]Glc-P-Dol20 to
[ 3H]Glc3 Man9 GlcNAc2 -P-P-Dol in sealed microsomal vesicles
from pig brain
To prove more conclusively that β-[3H]Glc-P-Dol20 was transported
into the lumenal compartment of brain microsomal vesicles and
not simply adsorbed or intercalated into the outer leaflet, an
experiment was conducted to see if [3H]glucose was incorporated
into [3H]Glc3Man9GlcNAc2-P-P-Dol when sealed vesicles were
1200
Fig. 11. β-[3H]Glc-P-Dol10 does not freely enter PC-liposomes or bovine red
blood cells. Glc-P-Cit transport assay mixtures (A) contained 10 mM
Tris-HCI (pH 7.4), 0.25 M sucrose, 10 µM [3H]Glc-P-Dol10
(20 c.p.m./pmol), and the indicated amount of either rat liver microsomes
(solid circles), bovine erythrocytes (open circles) or synthetic PC-liposomes
(open triangles) in a total volume of 0.05 ml. Glucose transport assay
mixtures (B) were identical except that 80 µM [14C]glucose (52 c.p.m./pmol)
was added instead of [3H]Glc-P-Dol10. Following incubation at 19C for
1 min, the uptake of [3H]Glc-P-Dol10 or glucose was assayed as described in
Materials and methods .
incubated with the water-soluble analog. To establish that the pig
brain microsomes remained intact during the experiment, latency
of the deoxynojirimycin-sensitive, processing glucosidase I/II
activities was monitored as described in the preceding report
(Rush and Waechter, 1998).
The data in Table III show that a significant amount of
[3H]glucose was incorporated into [3H]Glc3Man9GlcNAc2P-P-Dol when at least 97% of the vesicles are sealed. The
enzymatically labeled oligosaccharide-lipid represented ∼5% of
the total amount of radioactivity associated with the sealed
vesicles after incubation with β-[3H]Glc-P-Dol20.
Transbilayer movement of Glc-P-dolichol
Table II. β-[3H]Glc-P-Dol10 transport system is enriched in ER membrane fractions
Subcellular fraction
Man 6-Pase activity
(nmol/min/mg)
Gal-Tase activity
(pmol/min/mg)
SDH activity
(nmol/min/mg)
Glc-P-Dol10 uptake
(pmol/min/mg)
ER
22
0.1
19.2
16.8
Golgi
12.6
1.8
24.9
6.6
Mitochondria
3.7
0.1
153
2.4
The details of the isolation of each fraction and the marker enzyme assays are presented in Materials and methods.
Table III. Incorporation of [3H]glucose into
[3H]Glc3Man9GlcNAc2-P-P-Dol when β-[3H]Glc-P-Dol20 is incubated with
sealed vesicles from pig brain
Additions
[3H]Glc incorporated
into [3H]oligo-P-P-Dol
(pmol/mg)
Glucosidase I/II
activity
(% of total activity)
None
3.8
0.03
0.05% Triton X-100
2.8
85.4
Note: GlcTase and glucosidase I/II activities were measured in intact and
Triton X-100 disrupted pig brain microsomes as described in Materials and
methods. Glucosidase I/II activity is expressed as a percentage of the activity
detected in pig brain vesicles in which the permeability barrier has been
completely disrupted by the addition of 0.2% Triton X-100 .
When 0.05% Triton X-100 was added, over 80% of the
vesicles were unsealed, as judged by the loss of latent glucosidase
I/II activity, but there was virtually no increase in the labeling of
[3H]Glc3Man9GlcNAc2-P-P-Dol, as would be expected if the
labeling in the control experiment were due to the reaction of the
water-soluble substrate with enzymes on the lumenal surface
exposed in a small fraction of disrupted vesicles. Although the
absolute rates of incorporation were slightly lower, very similar
results were obtained with [3H]Glc-P-Dol10. These results
provide additional evidence that the water-soluble analogs are,
indeed, transported into the lumenal compartment of sealed
microsomal vesicles.
Discussion
This article describes the enzymatic synthesis of water-soluble
analogs of Glc-P-Dol containing short chain, stereochemically
defined isoprenyl moieties. The structures of the water-soluble
analogs have been thoroughly characterized by electrosprayionization tandem mass spectrometry and analyses of their
chemical and chromatographic properties. When [3H]Glc-P-Dol10,
[3H]Glc-P-(ω,c)Dol15, [3H]Glc-P-(ω,t,t)Dol20, and [3H]Glc-P(ω,t,c)Dol20 were tested as substrates for the three GlcTases
catalyzing the addition of the three glucosyl units in the
triglucosyl cap of Glc3Man9GlcNAc2-P-P-Dol to evaluate the
recognition of the stereoconfiguration of the β-isoprene unit,
the pig brain GlcTases actively transferred glucosyl units from
each of the water-soluble analogs to endogenous
Glc0–2Man9GlcNAc2-P-P-Dol acceptors. Even considering the
limitations of this pseudokinetic comparison of the three glucosyltransferase activities with the stereochemically defined substrates, it
appears that the GlcTases recognize and prefer glucosyl donors
containing a cis-isoprene unit in the β-position of the isoprenyl
moiety as found in the natural substrate.
Discrimination between polyisoprenyl moieties containing
saturated (eukaryotes) or unsaturated (prokaryotes) α-isoprene
units by enzymes synthesizing glycosyl-P-polyisoprenols or
utilizing the glycolipids as glycosyl donors is well-documented
(Rush et al., 1993; DeLuca et al., 1994; D’Souza-Schorey et al.,
1994; Kean et al., 1994; Dotson et al., 1995; Rush et al., 1997b,
and numerous references cited therein). The preference for
Glc-P-Dols containing β-isoprene units of the cis-stereochemical
configuration by the GlcTases extends the information on the
structural features of the dolichyl moieties recognized by enzymes
participating in polyisoprenoid lipid intermediate pathways. In a
related study, all trans Dol7-P was found to be inactive as a
substrate for rat liver Glc-P-Dol synthase, whereas (ω,t3,c2)Dol7-P
and (ω,t2,c3)Dol7-P are both efficiently glucosylated (Jaenicke et al.,
1991). Conversely, other studies report that rat liver Glc-P-Dol
synthase glucosylates α-dihydro undecaprenyl-P (ω,t2,c7-) and
α-dihydro solanosyl-P (ω,t6-) at virtually equal rates (Mankowski
et al., 1975, 1977). In yeast, Man-P-Dol synthase utilizes
α-dihydro undecaprenyl-P (ω,t2,c7-) and α-dihydro solanosyl-P
(ω,t6-) as substrates and the mannolipid products serve as
mannosyl donors for PMT1 (Pless and Palamarczyk, 1978;
Palamarczyk et al., 1980). However, GPT 1 exhibited a strong
preference for substrates with a cis-isoprene unit in the β-position.
The enzymes that catalyze the synthesis of the pyrophosphorylpolyprenyl-linked trisaccharide of the S.anatum O-specific
polysaccharide also strongly prefer a polyisoprenyl phosphate
acceptor with a cis-isoprene unit in the β-position (Danilov et al.,
1989).
The results in the preceding article (Rush and Waechter, 1998)
are consistent with the topological model of Hirschberg and
Snider (1987), in which Glc-P-Dol functions as a glucosyl donor
for the synthesis of Glc3Man9GlcNAc2-P-P-Dol in the lumenal
compartment of the ER after diffusing transversely (“flip-flopping)
from the site of synthesis on the cytoplasmic face. Thus, the
transbilayer movement of Glc-P-Dol could be mediated by a
“flippase as proposed for phosphatidylcholine, phosphatidylserine,
glucosylceramide, glycosyl-phosphatidylinositol anchor precursors,
and other dolichol-linked intermediates of the N-glycosylation
pathway (Hirschberg and Snider, 1987; Trotter and Voelker,
1994; Menon, 1996). Prokaryotic flippases are also thought to
play a role in the assembly of the bacterial cell wall (Bugg and
Brandish, 1994) and many other bacterial exopolysaccharides
(McGrath and Osborn, 1991; Liu et al., 1996).
The utilization of the water-soluble analogs as substrates by
the Glc-P-Dol-mediated GlcTases suggested that their structural
resemblance to Glc-P-Dol might also be recognized by the
hypothetical membrane protein involved in facilitating the
transverse diffusion of the glucolipid (Figure 1). In order to
develop an assay for this putative “flippase activity, the transport
of Glc-P-Dol10 was investigated in sealed ER-enriched vesicles
from rat liver and pig brain. The results of this study demonstrate that
Glc-P-Dol10 is rapidly transported into the lumenal compartment of
intact rat microsomal vesicles from liver and pig brain. Extensive
studies with rat liver vesicles indicate that the water-soluble
1201
J.S.Rush et al.
analog is transported by a mechanism with the properties
expected for a protein-mediated process. Transport of the
water-soluble analog was found to: (1) be time-dependent, (2)
require an intact permeability barrier, (3) be saturable, (4) be
inactivated by trypsin, and (5) be enriched in ER vesicles relative
to Golgi vesicles and intact mitochondria. In addition, Glc-P-Dol10
was utilized as a glucosyl donor by the microsomal GlcTase(s)
that transfer glucosyl residues from Glc-P-Dol to endogenous
Glc0–2Man9GlcNAc2-P-P-Dol acceptors in sealed pig brain
vesicles, indicating that the water-soluble analog is transported
into the lumenal compartment, and is not simply adsorbed to the
outer leaflet of the microsomal vesicle.
Although the protein-mediated transport systems for the
water-soluble analogs of Man-P-Dol and Glc-P-Dol are enriched
in the ER where the “flippases would be expected to be located,
considerably more work will be required to prove conclusively
that the analogs are, in fact, being transported by the same
proteins facilitating the transbilayer movement of the lipid
intermediates. So far, attempts to identify a mutant in yeast or
mammalian cells to provide a genetic correlation between the
transport protein and the Man-P-Dol or Glc-P-Dol “flippases
have been unsuccessful. The results to date are nevertheless
encouraging, and efforts to isolate a mutant and to define
biochemical correlates between the transport systems and factors
required for the movement of the dolichol-linked sugars from the
cytoplasmic leaflet to the lumenal compartment are in progress.
Materials and methods
Materials
S-Citronellol, DE-52 cellulose, monensin, valinomycin, ionomycin,
carbonylcyanide p-trifluoromethoxy-phenylhydrazone (FCCP),
carbonylcyanide m-chlorophenylhydrazone (CCCP), ATP, and
γ-S-ATP were purchased from Sigma Chemical Co. (St. Louis,
MO). Phosphorus trichloride oxide was obtained from Alfa
Products (Danvers, MA). Trichloroacetonitrile was obtained
from Aldrich Chemical Co. (Milwaukee, WI). [U-14C]Glucose
(360 mCi/mmol) was obtained from American Radiolabelled
Chemicals (St. Louis, MO). UPD-[1–3H]glucose was synthesized
enzymatically as described previously (Rush and Waechter,
1998). [2–3H]Man 6-P was synthesized enzymatically and
purified as described previously (Rush and Waechter, 1992).
Millipore HA (0.45 µm) filter disks were purchased from Baxter
Scientific Products (Obetz, OH). Whatman GF/C glass microfiber
filter disks were obtained from American Scientific Products
(McGaw Park, IL). Econosafe Liquid Scintillation counting
cocktail is a product of Research Products International Corp.
(Mount Prospect, IL). All other chemicals and reagents were
purchased from standard commercial sources.
Synthesis of (ω,c)Dol15 , (ω,t,t)Dol20 , and (ω,t,c)Dol20
(ω,t,t)Dol20 and (ω,t,c)Dol20 were prepared as described previously
(Jaenicke and Siegmund, 1989) using geranyl-4-tolyl sulfone and
either 8-chloro-(6-c)-citronellylbenzyl ether or 8-chloro-(6-t)citronellylbenzyl ether as building blocks for the synthesis. For the
synthesis of (ω,c) Dol15, 1 g of (6-benzyloxy-4-methyl-(c)4-hexen-1-yl)triphenylphosphonium iodide, prepared according
to Sato et al. (1983), was dissolved in methanol and shaken
overnight at room temperature under a 35 psi hydrogen atmosphere
in the presence of 100 mg PtO2. After filtration over Celite, the
filtrate was concentrated under reduced pressure to yield 1g of a
yellowish oil that was directly used for the following Wittig
1202
reaction. The reaction product, (6-benzyloxy-4-methyl1-hexyl)-triphenyl-phosphonium iodide was dissolved in 5 ml
anhydrous tetrahydrofuran and stirred for 10 min at -78C after
the addition of 1.16 ml (1.85 mmol, 1.1 eq.) of 1.6 M
n-butyl-lithium in hexane. After 10 min, 0.235 g (1.1 eq.) of
6-methyl-5-hepten-2-one (Aldrich) was added dropwise and the
reaction was stirred overnight at room temperature. The reaction
mixture turned light brown, and TLC analysis indicated that no
starting material was left. After the addition of saturated
ammonium chloride solution, the mixture was decanted and
extracted twice with ether. The combined organic layers were
washed once with brine, then dried briefly over MgSO4 and
concentrated under reduced pressure to give an oily solid, which
was purified on a short path silica gel column equilibrated in
hexane and eluted stepwise with 2%, then 3% ethyl acetate to
yield 0.27 g of the benzyl ether of (ω,c)Dol15 (51%, based on
phosphonium salt). The benzyl group was removed by adding
0.3 g (0.95 mmol) of the benzyl ether in 5 ml anhydrous
tetrahydrofuran to a solution of 110 mg (4.77 mmol, 5 eq.) Na in
NH3 at -78C. After 30 min at -33C (refluxing NH3), solid
ammonium chloride was added until the blue color disappeared
and ammonia was allowed to evaporate over a period of ∼45 min.
The residue was evaporated under reduced pressure and purified
over a short path silica gel column with ethyl acetate/hexane (1:3)
to yield 177 mg (82.7%) of a colorless oil. Progress of reactions
was monitored by thin layer chromatography on silica gel in
either ethyl acetate/hexane (1:9) for the Wittig reaction or ethyl
acetate/hexane (1:2) for the removal of the benzyl group.
Reaction products were analyzed by 1H-NMR and the trans/cis
ratios of the double bonds were established by integration of the
vinylic methyl shifts at 1.6 ppm according to Jaenicke and
Siegmund (1989). (ω,c)Dol15, (ω,t,t)Dol20, and (ω,t,c)Dol20 were
chemically phosphorylated as described below for citronellol.
Synthesis and purification of short chain, water-soluble
analogs of Glc-P-Dol
Citronellyl phosphate was synthesized by phosphorylation of
citronellol using trichloroacetonitrile and phosphorus trichloride
oxide (Danilov and Chojnacki, 1981) and purified as described
previously (Rush et al., 1993). Enzymatic reactions for the
synthesis of Glc-P-Dol10 contained 10 mM MgCl2, 50 mM
Tris-HCl (pH 7.5), 0.5 mM EDTA, 2 mM Cit-P, 5 mM
2-mercaptoethanol, 1 mM sodium orthovanadate, 20–80 µM
UDP-[3H]Glc (1–500 c.p.m./pmol), and hen oviduct microsomes
(0.75 mg protein) in a total volume of 0.5 ml. Following
incubation for 1 h at 37C, the reaction was centrifuged
(100,000 × g, 10 min) in a Beckman TL-100 tabletop ultracentrifuge
and the supernatant removed. The resulting membrane pellet was
washed two times with 0.25 ml of water. The supernatants were
combined and loaded onto an 8 ml column of DE-52 cellulose.
After rinsing the ion-exchange column with 5 column volumes of
distilled water, Glc-P-Dol10 was eluted with a 50 ml gradient
(0–0.5 M) of NH4HCO3. The fractions containing Glc-P-Dol10
were combined, concentrated by rotary evaporation under
reduced pressure at 30C and desalted by gel-filtration chromatography on a Bio-Gel P-2 column (1.5 cm × 40 cm) eluted with
distilled water. The fractions containing Glc-P-Dol10 were again
combined, concentrated and stored at -20C until use.
Enzymatic reactions for the synthesis of (ω,c)-Glc-P-Dol15,
(ω,t,t)-Glc-P-Dol20, and (ω,t,c)-Glc-P-Dol20 contained 10 mM
MgCl2, 50 mM Tris-HCl (pH 7.5), 0.5 mM EDTA, 5 mM
2-mercaptoethanol, 1 mM sodium orthovanadate, 20–80 µM
UDP-[1–3H]Glc (5–500 c.p.m./pmol), hen oviduct microsomes
Transbilayer movement of Glc-P-dolichol
(1–10 mg protein) and the appropriate isoprenyl monophosphate
substrate (0.2 mM) in a total volume of 0.5–4 ml. Following
incubation for 2 h at 37C, the reaction was stopped by the
addition of 5 ml of CH3OH. The precipitated protein was
sedimented by centrifugation (1000 × g, 5 min) and the
supernatant removed and placed on ice. The pellet was then
resuspended twice with 2 ml of CHCl3/CH3OH/H2O (10:10:3)
and recentrifuged, and the supernatants were combined with the
CH3OH layer. The organic extract was then supplemented with
additional CHCl3 to give a final composition of
CHCl3/CH3OH/H2O (3:2:1) and mixed vigorously. The two
phases were separated by a brief centrifugation and the aqueous
(upper) layer was removed. The organic (lower) layer was then
reextracted with 5 ml of CHCl3/CH3OH/H2O (3:48:47). The
aqueous layers were combined, evaporated to dryness under
reduced pressure at 35C by rotary evaporation, redissolved in 1 ml
water, and transferred with two rinses to a glass centrifuge tube.
This solution was then mixed with 1 ml of water-saturated butanol
and centrifuged briefly. The butanol phase was collected and the
aqueous phase was re-extracted with 1 ml of water-saturated
butanol. The butanol phases were combined, washed once with
butanol-saturated water, and dried under a stream of N2.
Contaminating phospholipids were next destroyed by deacylation
at 0C, 30 min in toluene/CH3OH (1:1) containing 0.1 M KOH.
The reaction mixture was neutralized with glacial acetic acid and
dried under a stream of N2. The deacylated mixture was desalted by
partitioning with water-saturated butanol, as described previously,
and the Glc-P-Dols were purified by ion-exchange chromatography
on DE-52 cellulose as described by Waechter and Scher (1981).
Following purification by DE-52 cellulose chromatography,
Glc-P-Dols were again desalted by butanol/water partitioning and
purified further by preparative thin layer chromatography on
Silica Gel G developed in CHCl3/CH3OH/H2O/NH4OH
(65:35:4:1). Zones containing radioactive Glc-P-Dols were
located with a Bioscan 200 Imaging System, excised from the thin
layer plate and eluted from the silica gel with CHCl3/CH3OH/H2O
(10:10:3). Purified water soluble Glc-P-Dolichols were then dried,
dissolved in water and stored at -20C until use. The concentration
of various water-soluble analogs was determined by phosphate
analysis as described previously (Bartlett, 1959).
Electrospray-ionization tandem mass spectrometry of
water-soluble Glc-P-Dols
Negative-ion electrospray-ionization mass spectrometry was
performed with a Finnigan-MAT (San Jose, CA) TSQ 7000 triple
quadrupole instrument using direct injection as previously described
(Woluka et al., 1998). Samples were dissolved in acetonitrile-H2O
(1:1) and infused into the electrospray source at a flow rate of
3 µl/min. In tandem mass spectrometry experiments, the pressure
of the target gas (xenon) was 0.6 mTorr and the collision energy
was maintained at 15 or 20 eV.
Preparation of microsomes from rat liver and pig brain
For transport studies with the water-soluble analog of Glc-P-Dol,
ER-enriched microsomal vesicles were prepared from rat liver
essentially as described by Coleman and Bell (1978). The
membrane vesicles were judged to be >95% intact as assessed by
mannose 6-phosphate (Man 6-P) phosphatase latency (see
below). Pig brain microsomal vesicles were prepared as described
in the preceding publication (Rush and Waechter, 1998).
Preparation of microsomal fractions and mitochondria from rat
liver
Light (Golgi-enriched) and heavy (ER-enriched) microsomes
and mitochondria were prepared from rat liver by minor
modifications of the methods of Leelavathi et al. (1970) and
Fleischer and Kervina (1974) exactly as described previously
(Rush and Waechter, 1995). Cross-contamination of the subcellular
fractions from rat liver was assessed by measuring the established
marker enzyme activities: ER/glucose 6-phosphatase (Rush and
Waechter, 1992); Golgi/galactosyltransferase (Fleischer and Smigel,
1978) and mitochondria/succinate dehydrogenase (Pennington,
1961).
Isolation of bovine red blood cells
Bovine red blood cells were collected into 5 mM EDTA and
sedimented by centrifugation at 1000 × g, 10 min 4C. The buffy
coat (white cells) was removed by aspiration and the red cells
were resuspended in Dulbecco’s phosphate-buffered saline (PBS)
and stored on ice until used for transport assays.
Assay of Glc-P-Dol10 transport by microsomal fractions from
rat liver or pig brain
Assay mixtures for the measurement of [3H]Glc-P-Dol10 uptake
contained 10 mM Tris-Cl (pH 7.4), 0.25 M sucrose, the
appropriate concentration of [3H]Glc-P-Dol10 (5–500 c.p.m./
pmol) and brain or liver microsomes (150–500 µg membrane
protein) in a total volume of 10–20 µl. Following incubation at
19–23C for 30 sec, [3H]Glc-P-Dol10 transport was stopped by
the addition of 0.5 ml of ice-cold 0.25 M sucrose, 10 mM Tris-HCl
(pH 7.4), and the amount of [3H]Glc-P-Dol10 transported was
determined by a filtration assay as described by Bishop and Bell
(1985). The diluted assay mixtures were quickly transferred to a
chilled filtration manifold equipped with a Millipore HA (0.45 µm)
filter disk and suction-filtered. The disks were then washed with
an additional 10 ml of ice-cold 0.25 M sucrose, 10 mM Tris-HCl
(pH 7.4). Filtration was usually completed in less than 30 sec. The
Millipore filter was then transferred to a 20 ml scintillation vial
and the amount of radioactivity retained on the filter determined
by scintillation spectrometry in a Packard 2100TR Scintillation
Spectrometer after the addition of 1.0 ml of 1% SDS and 10 ml
of Econosafe liquid scintillation cocktail.
Measurement of glucose transport by erythrocytes and liver
ER-enriched microsomes
Glucose transport assay mixtures contained 10 mM Tris-HCl
(pH 7.4), 0.25 M sucrose, 0.5 mM [14C]glucose (52 c.p.m./pmol)
and the indicated amount of either bovine erythrocytes or microsomal vesicles from rat liver (0–120 nmol membrane phospholipid)
in a total volume of 50 µl. Following incubation at 19–23C for
1 min, transport reactions were diluted with 0.5 ml of ice-cold
10 mM Tris-HCl (pH 7.4), 0.25 M sucrose. Glucose transport was
determined exactly as described above for Glc-P-Dol10 except that
Whatman GF/C glass-fiber filter disks were used for the
measurement of transport by human erythrocytes.
Assessment of the integrity of ER-enriched and Golgi-enriched
vesicles and mitochondria
The integrity of liver ER microsomes was determined by
measuring Man 6-P phosphatase latency using [3H]Man 6-P as
substrate (Rush and Waechter, 1992). The intactness of pig brain
1203
J.S.Rush et al.
microsomes was assessed by assaying for glucosidase I/II latency
(Rush and Waechter, 1998). The integrity of rat liver Golgi
vesicles was determined by measuring the latency of rat liver
galactosyltransferase to proteolytic digestion with trypsin. Mitochondrial integrity was estimated by measuring the sensitivity of
succinate dependent reduction of potassium ferricyanide to
inhibition by antimycin A as described by Klingenberg (1979).
standards and an exogenously added internal standard with the
composition Man9GlcNAc1.
Analytic methods
Protein concentrations were determined by the method of
Rodriquez-Vico et al. (1989) using a protein assay reagent (BCA,
Pierce, Rockford, IL). Lipid-phosphorus was determined by the
method of Bartlett (1959).
Preparation of synthetic PC-liposomes
Synthetic liposomes were prepared using liposome kit No.
L-4012 (Sigma Chemical Co.) containing egg phosphatidylcholine,
dicetylphosphate, and cholesterol in a molar ratio of 7:2:1.
Liposomes were formed by sonication in 10 mM Tris-HCl
(pH 7.4), 0.25 M sucrose and dialyzed against the same solution
overnight before use as described previously (Rush and Waechter,
1995).
Measurement of uptake of [ 3H]Glc-P-Cit by synthetic
PC-liposomes.
Uptake of [3H]Glc-P-Cit by synthetic PC-liposomes was determined by gel filtration on a column (0.5 × 20 cm) of Sephacryl
S-300 equilibrated in 10 mM Tris-HCl pH 8.0, 0.25 M sucrose as
described previously (Rush and Waechter, 1995).
Assay of Glc-P-Dol10–20 :Glc0–2 Man9 GlcNAc2 -P-P-Dol
GlcTase activity in sealed pig brain vesicles
Incubation mixtures contained 50 mM Tris-HCl (pH 8), 0.2 M
sucrose, 10 mM EDTA, 2 mg/ml Triton X-100, 0.75 mg of pig
brain microsomal protein, and the indicated concentration of
[3H]Glc-P-Dol10–20 (500 c.p.m./pmol) in a total volume of
0.1 ml. Following incubation for 2 min at 37C the incorporation
of [3H]glucose into Glc1–3Man9GlcNAc2-P-P-Dol was determined
by a multiple extraction procedure (Waechter and Scher, 1978).
For the study described in Table III, reaction mixtures containing
50 mM Tris–HCl (pH 8), 0.2 M sucrose, 10 mM EDTA, 0.75 mg
of pig brain microsomal protein, and 0.2 µM [3H]Glc-P-Dol20
(500 c.p.m./pmol) in a total volume of 0.1 ml were incubated for
3 min at 37C in the presence or absence of Triton X-100
(0.5 mg/ml).
Characterization of lipid-linked [ 3H]oligosaccharide products
formed in Glc-P-Dol–mediated GlcTase reactions
The lipid-bound oligosaccharide products of the enzymatic
reactions were released by mild acid hydrolysis at 50C, 30 min,
in 80% tetrahydrofuran containing 0.1 M HCl (Lucas et al., 1975).
The hydrolysate was neutralized by the addition of 1 N NaOH, dried
under N2, and desalted by gel-filtration chromatography on a
Sephadex G-10 column (1.5 X 30 cm) equilibrated in distilled
water. The desalted oligosaccharides were characterized by high
pH anion-exchange chromatography using a Dionex BioLC
HPAEC system equipped with an analytical CarboPac PA1
column (0.4 × 25 cm) equilibrated in 0.25 M NaOH and eluted
with a sodium acetate gradient (0–200 mM, 35 min) at a flow rate
of 1 ml/min (Cooper and Rohrer, 1994). Fractions of 0.5 ml were
collected, neutralized with glacial acetic acid and analyzed for
radioactivity by scintillation spectrometry in a Packard Tri-Carb
2100TR liquid scintillation spectrometer. The chain lengths of the
enzymatically labeled oligosaccharides were determined by
comparing their elution positions with defined oligosaccharide
1204
Acknowledgments
We thank Dr. Austin H.Cantor and Dr. Mike Ford, University of
Kentucky Department of Animal Sciences, for their help in
obtaining fresh hen oviducts and Mr. James R.May, University of
Kentucky Department of Animal Sciences, for providing us with
fresh pig brains. This work was supported by NIH Grant
GM36065 awarded to C.J.W. We also thank Professor E.de
Hoffmann, University of Louvain, for his assistance with the
mass spectrometry measurements.
Abbreviations
Glc-P-Dol, glucosylphosphoryldolichol; Man-P-Dol, mannosylphosphoryldolichol; Man-P-Dol10, mannosylphosphorylcitronellol;
PBS, phosphate-buffered saline; GlcTase, Glc-P-Dol:Glc0–2Man9GlcNAc2-P-P-Dol glucosyltransferase; ER, endoplasmic reticulum;
Dol-P, dolichyl phosphate; FCCP, carbonylcyanide p-trifluoromethoxy-phenylhydrazone; CCCP, carbonylcyanide m-chlorophenylhydrazone.
References
Bartlett,G.R. (1959) Phosphate assay in column chromatography. J. Biol. Chem.,
234, 466–468.
Berendes,R. and Jaenicke,L. (1992) Short-chain dolichols of defined chain length
as cofactors in reactions of the microsomal dolichyl-phosphate cycle and
transglycosylations. Biol. Chem. Hoppe-Seyler, 373, 35–42.
Bishop,W.R. and Bell,R.M. (1985) Assembly of the endoplasmic reticulum
phospholipid bilayer: the phosphatidylcholine transporter. Cell, 42, 51–60.
Bugg,T.D.H. and Brandish,P.E. (1994) From peptidoglycan to glycoprotein.
Common features of lipid-linked oligosaccharide biosynthesis. FEMS Microbiol.
Lett., 119, 255–262.
Coleman,R. and Bell,R.M. (1978) Evidence that biosynthesis of phosphatidylethanolamine, phosphatidylcholine, and triacylglycerol occurs on the
cytoplasmic side of microsomal vesicles. J. Cell Biol., 76, 245–253.
Cooper,G.A. and Rohrer,J.S. (1994) Separation of neutral asparagine-linked
oligosaccharides by high-pH anion exchange chromatography with pulsed
amperometric detection. Anal. Biochem., 226, 182–184.
Cummings,R.D. (1992) Synthesis of asparagine-linked oligosaccharides: pathways,
genetics and metabolic regulation. In Allen,H.J. and Kisailus,E.C. (eds.),
Glycoconjugates: Composition Structure and Function. Marcel Dekker,
New York, pp. 333–360.
Danilov,L.L. and Chojnacki,T. (1981) A simple procedure for preparing dolichyl
monophosphate by the use of POCl3. FEBS Lett., 131, 310–312.
Danilov,L.L., Druzhinina,T.N., Kalinchuk,N.A., Maltsev,S.D. and Shibaev,V.N.
(1989) Polyprenyl phosphates: synthesis and structure-activity relationship
for a biosynthetic system of Salmonella anatum O-specific polysaccharide.
Chem. Phys. Lipids, 51, 191–203.
DeLuca,A.W., Rush,J.S., Lehrman,M.A. and Waechter,C.J. (1994) Mannolipid
donor specificity of glycosylphosphatidylinositol mannosyltransferase-I
(GPIMT-I) determined with an assay system utilizing CHO-K1 cells.
Glycobiology, 4, 909–916.
Dittmer,J.C. and Lester,R.L. (1964) A simple, specific spray for the detection of
phospholipids on thin-layer chromatograms. J. Lipid Res., 5, 126–127.
Dotson,S.B., Rush,J.S., Ricketts,A.D. and Waechter,C.J. (1995) Mannosylphosphoryldolichol-mediated O-mannosylation of yeast glycoproteins:
stereospecificity and recognition of the α-isoprene unit by a purified
mannosyltransferase. Arch. Biochem. Biophys., 316, 773–779.
D’Souza-Schorey,C. and Elbein,A.D. (1993) Partial purification and properties of
a glucosyltransferase that synthesizes Glc1Man9(GlcNAc)2-pyrophosphoryldolichol. J. Biol. Chem., 268, 4720–4727.
Transbilayer movement of Glc-P-dolichol
D’Souza-Schorey, C., McLachlan, K. R., Krag, S. S., and Elbein, A. D. (1994)
Mammalian glycosyltransfeerases prefer glycosyl phosphoryl dolichols
rather than glycosyl phosphoryl polyprenols as substrates for oligosaccharyl
synthesis. Arch. Biochem. Biophys., 308, 497–503.
Dunphy,P.J., Kerr,J.D., Pennock,J.F., Whittle,K.J. and Feeney,J. (1967) The
plurality of long chain isoprenoid alcohols (polyprenols) from natural sources.
Biochem. Biophys. Acta, 136, 136–147.
Fleischer,S. and Kervina,M. (1974) Subcellular fractionation of rat liver. Methods
Enzymol., 31, 6–41.
Fleischer,B. and Smigel,M. (1978) Solubilization and properties of galactosyltransferase and sulfotransferase activites of Golgi membranes in Triton
X-100. J. Biol. Chem., 253, 1632–1638.
Hanover,J.A. and Lennarz,W.J. (1978) The topological orientation of N,N′-diacetylchitobiosylpyrophosphoryldolichol in artificial and natural membranes.
J. Biol. Chem., 254, 9237–9246.
Hirschberg,C.B. and Snider,M.D. (1987) Topography of glycosylation in the rough
endoplasmic reticulum and golgi apparatus. Annu. Rev. Biochem., 54, 63–87.
Jaenicke,L. and Siegmund,H.-U. (1986) Total synthesis of cahin-length-uniform
dolichyl phosphates and their fitness to accept hexoses in the enzymatic
formation of lipoglycans. Biol. Chem. Hoppe-Seyler, 376, 787–795.
Jaenicke,L. and Siegmund,H.-U. (1989) Synthesis and characterization of
dolichols and polyprenols of designed geometry and chain length. Chem.
Physics Lipids, 51, 159–170.
Jaenicke,L., van Leyen,K. and Siegmund,H.-U. (1991) Dolichyl phosphatedependent glycosyltransferases utilize truncated cofactors. Biol. Chem.
Hoppe-Seyler, 372, 1021–1026.
Kean,E.L., Rush,J.S. and Waechter,C.J. (1994) Activation of GlcNAc-P-P-dolichol
synthesis by mannosylphosphoryldolichol is stereospecific and requires a
saturated α-isoprene unit. Biochemistry, 33, 10508–10512.
Klingenberg,M. (1979) The ferricyanide method for elucidating the sidedness of
membrane-bound dehydrogenases. Methods Enzymol., 56, 229–233.
Kornfeld,R. and Kornfeld,S. (1985) Assembly of asparagine-linked oligosaccharides.
Annu. Rev. Biochem., 54, 631–664.
Leelavathi,D.E., Estes,L.W., Feingold,D.S. and Lombardi,B. (1970) Isolation of a
Golgi-rich fraction from rat liver. Biochim. Biophys. Acta, 211, 124–138.
Liu,D., Cole,R.A. and Reeves,P.R. (1996) An O-antigen processing function for
Wzx (RfbX): a promising candidate for O-unit flippase. J. Bacteriol., 178,
2102–2107.
Lucas,J.J., Waechter,C.J. and LennarzW.J. (1975) The participation of lipid-linked
oligosaccharide in the synthesis of membrane glycoproteins. J. Biol. Chem.,
250, 1992–2002
Mankowski,T., Sasak,W. and Chojnacki,T. (1975) Hydrogenated polyprenol
phosphates- exogenous lipid acceptors of glucose from UDP-glucose in rat
liver microsomes. Biochem. Biophys. Res. Commun., 65, 1292–1297.
Mankowski,T., Sasak,W., Janczura,E. and Chojnacki,T. (1977) Specificity of
polyprenyl phosphates in the in vitro formation of lipid-linked sugars. Arch.
Biochem. Biophys., 181, 393–401.
McCloskey,M.A. and Troy,F.A. (1980) Paramagnetic isoprenoid carrier lipids. 1.
Chemical synthesis and incorporation into model membranes. Biochemistry,
19, 2061–2066.
McGrath,B.C. and Osborn,M.J. (1991) Localization of the terminal steps of
O-antigen synthesis in salmonella typhimurium. J. Bacteriol., 173, 649–654.
Menon,A.K. (1995) Flippases. Trends in Cell Biol., 5, 355–360.
Palamarczyk,G., Lehle,L., Mankowski,T., Chojnacki,T. and Tanner,W. (1980)
Specificity of solubilized yeast glycosyl transferases for polyprenyl derivatives.
Eur. J. Biochem., 105, 517–523.
Pennington,R.J. (1961) Biochemistry of dystrophic muscle. Mitochondrial
succinate-tetrazolium reductase and adenosinetriphosphatase. Biochem. J., 80,
649–654.
Pless,D.D. and Palamarczyk,G. (1978) Comparison of polyprenyl derivatives in
yeast glycosyl transfer reactions. Biochim. Biophys. Acta, 529, 21–28.
Rodriquez-Vico,F., Martinez-Cayeula,M., Garcia-Peregrin,E. and Ramirez,H.
(1989) A procedure for eliminating interferences in the Lowry method of
protein determination. Anal. Biochem., 183, 275–278.
Runge,K.W. and Robbins,P.W. (1986) A new yeast mutation in the glucosylation
steps of the asparagine-linked glycosylation pathway. J. Biol. Chem., 261,
15582–15590.
Runge,K.W., Huffaker,T.C. and Robbins,P.W. (1984) Two yeast mutations in
glucosylation steps of the asparagine glycosylation pathway. J. Biol. Chem.,
259, 412–417.
Rush,J.S. and Waechter,C.J. (1992) An anion-exchange radioassay for glucose
6-phosphate phosphatase: use in topological studies with endoplasmic
reticulum vesicles. Anal. Biochem., 206, 328–333.
Rush,J.S. and Waechter,C.J. (1995) Transmembrane movement of a water-soluble
analoque of mannosylphosphoryldolichol is mediated by an endoplasmic
reticulum protein. J. Cell Biol., 130, 529–536.
Rush,J.S. and Waechter,C.J. (1998) Topological studies on the enzymes
catalyzing the biosynthesis of Glc-P-dolichol and the triglucosyl cap of
Glc3Man9GlcNAc2-P-P-dolichol in microsomal vesicles from pig brain: use of
the processing glucosidases I/II as latency markers. Glycobiology, 8, 1207–1213.
Rush,J.S., Shelling,J.G., Zingg,N.S., Ray,P.H. and Waechter,C.J. (1993) Mannosylphosphoryldolichol-mediated reactions in oligosaccharide-P-P-dolichol
biosynthesis: recognition of the saturated α-isoprene unit of the mannosyl
donor by pig brain mannosyltransferases. J. Biol. Chem., 268, 13110–13117.
Rush,J.S., van Leyen,K., Ouerfelli,O., Wolucka,B. and Waechter,C.J. (1997a) Use
of water-soluble analogs to study the transbilayer movement of glucosylphosphoryldolichol and its function as a glucosyl donor in brain. Glycobiology, 7,
abstract 90.
Rush,J.S., Rick,P.D. and Waechter,C.J. (1997b) Polyisoprenyl phosphate specificity
of UDP-GlcNAc:undecaprenyl phosphate N-acetylglucosaminyl 1-P transferase
from E.coli. Glycobiology, 7, 315–322.
Sato,K., Miyamoto,O., Inoue,S., Furusawa,F. and Matsuhashi,Y. (1983) Stereoselective synthesis of a cisoid C-10 isoprenoid building block and some all-cis
polyprenols. Chem. Lett., 5, 725–728.
Stagljar,I., Te Heesen,S. and Aebi,M. (1994) New phenotype of mutations
deficient in glucosylation of the lipid-linked oligosaccharide: cloning of the
ALG-8 locus. Proc. Natl. Acad. Sci. USA, 91, 5977–5981.
Trotter,P.J. and Voelker,D.R. (1994) Lipid transport processes in eukaryotic cells.
Biochim. Biophys. Acta, 1213, 241–262.
Waechter,C.J. (1989) Glycoprotein biosynthesis. In Margolis,R.U. and Margolis,R.K.
(eds,), Neurobiology of Glycoconjugates. Plenum Press, New York, pp. 127–149.
Waechter,C.J. and Scher,M.G. (1978) Glucosylphosphoryldolichol: role as a
glucosyl donor in the biosynthesis of an oligosaccharide lipid intermediate by
calf brain membranes. Arch. Biochem. Biophys., 188, 385–393.
Waechter,C.J. and Scher,M.G. (1981) Methods for studying lipid-mediated
glycosyltransferases involved in the assembly of glycoproteins in nervous tissue.
In Marks,N. and Rodnight,R. (eds.), Research Methods in Neurochemistry.
Plenum Press, New York, pp. 201–231.
Wolucka,B.A., de Hoffmann,E., Rush,J.S. and Waechter,C.J. (1996) Determination
of the anomeric configuration of glycosyl esters of nucleoside pyrophosphates
and polyisoprenyl phosphates by fast-atom bombardment tandem mass
spectrometry. J. Am. Soc. Mass Spec., 7, 541–549.
Wolucka,B.A., Rush,J.S., Waechter,C.J., Shibaev,V.N. and de Hoffmann,E.
(1998) An electrospray-ionization tandem mass spectrometry method for
determination of the anomeric configuration of glycosyl 1-phosphate derivatives. Anal. Biochem., 255, 244–251.
Zuffrey,R., Knauer,R. , Burda,P., Stagljar,I., te Heesen,S., Lehle,L. and Aebi,M.
(1995) STT3, a highly conserved protein required for yeast oligosaccharyl
transferase activity in vivo. EMBO J., 14, 4949–4960.
1205