THE JOURN.,L OF Bmmcrcn~ C,m~,smv Vol. 252, No. 7, Issue of April 10, pp. 2271-2277, 1977

Printed in U.S A.

Glucosyltransferase Activity in Calf Pancreas Microsomes FORMATION OF DOLICHYL D-[‘~C]GLUCOSYL PHOSPHATE AND 14C-LABELED LIPID-LINKED OLIGOSACCHARIDES FROM UDP-D-[%]GLUCOSE*

(Received for publication, November 29, 1976)

ANNETTE HERSCOVICS, BIRGITTE BUGGE, AND ROGER W. JEANLOZ

From the Laboratory for Carbohydrate Research, Departments ofBiological Chemistry and Medicine, Harvard Medical School at the Massachusetts General Hospital, Boston, Massachusetts 02114

Calf pancreas microsomes incorporated radioactively la- beled n-glucose from DDP-o-glucose into products extracted with chloroform/methanol (2:1, v/v), chloroform/methanol/ water (10:10:2.5, v/v), and into the residual precipitate, with a pH optimum in Trislmaleate buffer of about 5.3.

The chloroform/methanol extract contained a single 14C- labeled acidic product, which was identified as dolichyl P-D-

glucosyl phosphate. It was stable to mild alkali, yielded D-

[‘4Clglucose upon mild acid hydrolysis, and a W-labeled compound with the chromatographic mobility of 1,6-anhy- dro-/I-a-ghicopyranosyl upon hot alkali treatment. The [*4Clglucolipid had the same chromatographic mobility as dolichyl /3-W4Clmannosyl phosphate, and its formation was stimulated by exogenous dolichyl phosphate.

The chloroform/methanol/water extract contained radio- active lipid-bound oligosaccharides which were retained on DEAE-cellulose more strongly than dolichyl o-[Wlglucosyl phosphate. They were stable to mild alkali, but labile to acid and hot alkali. Acid treatment yielded a o-glucose-labeled oligosaccharide fraction which was shown by gel filtration to be slightly larger than most of the D-mannose-labeled oligosaccharides. About 80% of the radioactive o-glucose residues could be removed with cu-glucosidase, but not with /3-glucosidase.

Pancreatic dolichyl p-o-[Wlglucosyl phosphate incu- bated with calf pancreas microsomes served as direct donor of o-glucosyl residues to lipid-bound oligosaccharides and to the precipitate. These oligosaccharides had the same size as those labeled from DDP-o-[14C]ghrcose, and the D-

[Wlglucose residues could also be removed with cy-glucosi- dase.

In a previous paper, we have shown that calf pancreas microsomes contain lipid-bound oligosaccharides consisting of n-mannose, n-glucose, and N-acetyl-n-glucosamine linked to

* This work was supported by Research Grant AM-03564 from the National Institute of Arthritis, Metabolism, and Digestive Diseases, National Institutes of Health, United States Public Health Service. Lipid Intermediates of Complex Polysaccharide Biosynthesis XIII. For paper XII see Ref. 1. This is Publication 723 of the Robert W. Lovett Memorial Group for the Study of Diseases Causing Deformi- ties, Harvard Medical School at the Massachusetts General Hospi- tal, Boston, Massachusetts.

dolichol through a pyrophosphate group attached to n-glucosa- mine (1). The oligosaccharide moiety appeared heterogeneous with variable proportions of the three sugars. Similar lipid- bound oligosaccharides containing labeled n-glucose have been found in other tissues (2-51, and, in liver microsomes, were demonstrated to donate their oligosaccharide moiety to endog- enous protein acceptor (6). Since, besides collagen, glucose- containing glycoproteins are rare and have not been described in the pancreas, a study of glucosyltransferase activity in this tissue was initiated. The aim is to further establish the pres- ence of n-glucose in the dolichol-linked oligosaccharides of pancreas, and to eventually clarify their role in glycoprotein biosynthesis.

The results presented demonstrate that calf pancreas micro- somes incubated with labeled UDP-n-glucose synthesize a la- beled glucolipid with the properties of dolichyl n-glucosyl phos- phate, and n-glucose-labeled lipid-bound oligosaccharides in which the oligosaccharides are slightly larger than the D-

mannose-labeled oligosaccharides described in the previous paper (1).

EXPERIMENTAL PROCEDURES

Materials -The source of chemicals is described in the preceding publication (1). In addition, UDP-n-[l-SHlglucose (4.85 Ci/mmol) and UDP-n-lU-14Clglucose (139 mCilmmo1) were purchased from New England Nuclear, Boston, Mass. P-Glucosidase from almond emul- sin was obtained from Miles Laboratories, Inc., Elkhart, Ind., and yeast a-glucosidase, type I from Sigma Chemical Co., St. Louis, MO.

Analytical Methods -The methods used are described in the pre- ceding paper (1). The solvent systems utilized for thin layer chroma- tography were: A, chloroform/methanol/water (10:10:3); B, l-pro- panobwater (7:3); C, chloroform/methanol/water (60:25:4); D, chloro- form/methanol/l5 M NH,OH/water (65:35:4:4); and E, 2,6-dimethyl- 4-heptanone/acetic acid/water (20:15:2); and for paper chromatogra- phy; F, ethyl acetate/pyridine/water (8:2:1); G, n-butyl alcohobpyri- dine/water (6:4:3); and H, isobutyric acid/l5 M NH,OH/water (59:4:39).

Preparation of Microsomes -Calf pancreas microsomes were pre- pared using the centrifugation scheme described previously (7). However, two methods of homogenization were used. In Method 1, pancreas was homogenized as previously described (7). In Method 2, the tissue was first minced with scissors, and then immediately homogenized with a Polytron PT 20 ST homogenizer (Brinkmann Instruments, Inc., Westbury, N. Y.) with three 30-s treatments at the lowest speed. Use of the Polytron homogenizer allowed the preparation of larger quantities of microsomes in a relatively short time. In addition, these microsomes were much more active in the utilization of labeled UDP-n-glucose/mg of protein than those pre- pared by the previous method.

2271

by guest on March 10, 2019

http://ww

w.jbc.org/

Dow

nloaded from

2272 [14CIGlucose-labeled Lipid-linked Oligosaccharides in Calf Pancreas

Incubation with Labeled UDP-D-GLUCOSE -For standard condi- tions, microsomes were incubated for 30 min at 30” in a total volume of 0.5 ml containing 4 to 6 mg of protein/ml, TrisJmaleate buffer, pH 5.3 (40 mM), MnCl, (10 mM1, and UDP-n-[U-14C]glucose (0.05 &i/ ml), unless otherwise indicated. In some cases, for preparative purposes, the volume of the incubation mixtures was increased B-fold keeping the same concentrations of constituents, and sometimes UDP-n-[1-“HIglucose (1 &i/ml) was the precursor.

Incubation with Pancreatic Dolichyl D-[~4CIGlucosyl Phosphate - The chloroform/methanol extract obtained from incubation of total microsomes with UDP-n-[“C]glucose was used. To remove any possi- ble contaminating UDP-D-[‘4C]glucose, the chloroform/methanol ex- tract was mixed with 0.2 volume of chloroform/methanol/water (3:48:47, v/v). After centrifugation, most of the upper phase was removed, and the lower phase was washed again with 0.2 volume of the same mixture. Most of the upper phase was discarded, and a few drops of methanol were added until a single phase was obtained. This extract will be shown later to contain a single radioactive product identified as dolichyl D-[‘4c]ghCOSyl phosphate. For stan- dard incubations, “washed” chloroform/methanol extract containing around 2500 cpm was mixed with 250 ~1 of 1% sodium taurocholate in chloroform/methanol (2:1, v/v). The mixture was dried under N,, and total microsomes (about 5 mg/ml of protein) suspended in 500 ~1 of 40 rnM Tris/maleate buffer, pH 6.3, were added. Incubations were per- formed at 30” for 30 min unless otherwise indicated. In some cases, for preparative purposes the volume of the incubation mixtures was increased 5-fold using the same final concentrations of constituents.

Extraction of Radioactively Labeled Products -At the end of the incubation, the tubes were cooled to 4”, and 5 volumes of chloroform/ methanol (2:1, v/v) were added. The suspension was mixed with a Vortex mixer, kept at room temperature for 10 to 20 min, mixed again, and centrifuged. First, the lower chloroform/methanol ex- tract, and then the upper aqueous phase were removed with a Pasteur pipette. The insoluble material originally present at the interphase was washed 2 to 3 times with 4 volumes of butanol, and then twice with 4 volumes of water. In some cases, it was necessary to let the mixture stand overnight at 4” during the first water extraction to allow sedimentation of the precipitate. This procedure removed any contaminating UDP-n-[‘4C]glucose. The residue was then extracted with 2 volumes of chloroform/methanol/water (10:10:2.5, v/v) at room temperature for 15 to 20 min. After centrifu- gation, the supernatant was collected, and a second extraction with 2 volumes of chloroform/methanol/water (10:10:2.5, v/v) was carried out. Both chloroform/methanol/water extracts were pooled. The pre- cipitate was allowed to dry at room temperature, and for estimation of radioactivity and protein concentration, it was dissolved in 0.5 M NaOH at 90” for 30 min.

Assay of Glycosidases - Glycosidase activity was assayed with the appropriate p-nitrophenyl glycosides as described in the preceding publication (1). For the P-glucosidase preparation, assays were done in 50 mM sodium acetate, pH 5, containing 0.1 M NaCl, 0.1 mM ZnSO,, and 100 pg/ml of bovine serum albumin, whereas for the a- glucosidase preparation the buffer was 100 mM potassium phosphate at pH 6.8. Under these conditions, the p-glucosidase preparation contained a-mannosidase (0.34 unitlmg), p-mannosidase (0.01 unit/ mg), a-glucosidase (0.02 unit/mg), p-glucosidase (3.2 units/mg), and P-N-acetylhexosaminidase (0.09 unit/mg), whereas the a-glucosi- dase preparation contained insignificant a- and p-mannosidase, p- N-acetylhexosaminidase, and a- and P-galactosidase; it had a-glu- cosidase activity (0.55 unit/mgl and p-glucosidase activity (0.02 unit/ mg).

RESULTS

Glucosyltransferase Activity of Microsomes -Incubation of calf pancreas microsomes with UDP-n-[‘4C]glucose resulted in the incorporation of radioactivity into endogenous acceptors. The labeled products were fractionated according to their solu- bility as described previously (1). There was a rapid initial incorporation of radioactivity into the chloroform/methanol extract which reached a maximum within 2 min and then decreased (Fig. 1). There was also an initial high rate of incorporation into the chloroform/methanol/water extract; the rate was linear for about 5 min and then decreased until maximum incorporation was achieved at about 30 min. The incorporation into the precipitate occurred at a slower rate

during the first 5 to 7 min of incubation, and the amount of radioactivity remained constant thereafter. About 75% of the incorporated radioactivity was found in the chloroform/metha- nol/water extract after 30 min of incubation. The optimum pH for incorporation of radioactivity into all three fractions was about 5.3, although there was a second broader peak, particu- larly significant for the chloroform/methanol extract and the precipitate, around pH 6.8 (Fig. 2). This second peak was not observed when rough microsomes were used.

Strong acid hydrolysis of the fractions labeled from UDP- n-[14C]glucose followed by paper chromatography in Solvent F showed that all the radioactivity in the chloroform/methanol extract and in the chloroform/methanol/water extract, and most of the radioactivity in the precipitate were recovered in a compound with the mobility of n-glucose (Fig. 31.

FIG. 1 (left). Incorporation of radioactivity from UDP-D- lY!lglucose into endogenous acceptors. Total microsomes prepared by method 2 were incubated under standard conditions for different periods of time. Radioactivity: A, chloroform/methanol extract; 0, chloroform/methanol/water extract; 0, precipitate.

FIG. 2 (right). Effect of pH on the utilization of UDP-D- [14C]glucose. Total microsomes prepared by method 2 were incubated under standard conditions at the indicated pH. The protein concen- tration was about 2 mg/ml. Radioactivity: A, chloroform/methanol extract; 0, chloroform/methanol/water extract; 0, precipitate.

i i , . 3- - - - -y ; ,

FIG. 3. Paper chromatography of n-[‘lC]glucose-labeled products. Chloroform/methanol extract (A), chloroform/methanol/water ex- tract (B), and precipitate (C) obtained from incubation of total microsomes (method 1) with UDP-n-[14C]glucose were dried and hydrolyzed for 8 h at 105” in 1 N H,SO,. The hydrolysates were then passed through coupled columns of AG 5OW-X8 (H+ form, 200 to 400 mesh), and AGl-X8 (formate form, 200 to 400 mesh). The resins were eluted with water, and the eluates were concentrated and chromato- graphed with standards on Whatman No. 1 paper in Solvent F for 32 h. The papers were scanned for radioactivity. Between 77 and 82% of the original radioactivity was recovered.

by guest on March 10, 2019

http://ww

w.jbc.org/

Dow

nloaded from

Properties of [WIGlucolipid in ChloroformlMethanol Ex- tract -The chloroform/methanol extract obtained from either rough or total microsomes contained a single major labeled product when examined by radioautography after thin layer chromatography in Solvents C (Fig. 41, D, and E. Radioautog- raphy showed that products with the same mobility in Solvent C were formed at both pH 5.3 and 6.8. When the chloroform/ methanol extract labeled from UDP-n-[14C]glucose was chro- matographed in Solvents C, D, and E with an extract contain- ing pancreatic dolichyl /3-n-[‘4C1mannosyl phosphate, the D- mannose and the n-glucose-labeled lipids had identical mobili- ties in all three solvent systems (Fig. 4).

[W]Glucose-labeled Lipid-linked Oligosaccharides in Calf Pancreas 2273

alkali treatment at both temperatures; it migrated during paper chromatography in Solvent G with a standard of 1,6- anhydro-P-n-glucopyranose (Rplueose = 1.61.

The addition of dolichyl phosphate stimulated the incorpora- tion of radioactivity from UDP-n-[‘4Clglucose into the chloro- form/methanol extract about 3-fold (Fig. 7). No stimulation was observed in the absence of Triton X-100. The chloroform/ methanol extract obtained from incubation with exogenous dolichyl phosphate contained a single radioactive product which had the same chromatographic mobility as the glucoli- pid formed from endogenous lipid.

Upon DEAE-cellulose chromatography of the chloroform/ methanol extract, the labeled glucolipid was retained on the column, and was subsequently eluted with 1 mM ammonium acetate (Fig. 5). The glucolipid was extremely labile to mild acid with a half-life in 0.01 M HCl at 90” of about 2 min (Fig. 61 which is similar to that previously obtained for dolichyl p-n- [Ylmannosyl phosphate (7). The water-soluble phase ob- tained after this acid treatment contained a “‘C-labeled prod- uct with the mobility of n-glucose in Solvent F. The labeled glucolipid was stable to mild alkali treatment at 37”, but labile to hot alkali. With 0.1 M NaOH it was completely degraded to ‘V-labeled water-soluble products within 10 min at go”, and 30 min at 65” (Fig. 6). One labeled product was obtained after hot

Properties of Glucose-labeled Lipid-bound Oligosaccha- rides -Thin layer chromatography of the chloroformlmetha- nol/water extract labeled from UDP-n-[*4C]glucose showed one major radioactive product with R, around 0.5 (Solvents A and B). A minor radioactive product migrating slightly faster than the major one was also observed. When the n-[‘4C1glucose- labeled lipid-bound oligosaccharides were compared with D-

[‘4C]mannose-labeled lipid-bound oligosaccharides (11, it was apparent that there were fewer n-glucose-labeled products and that their mobility in both Solvents A and B corresponded to that of the slower n-[14C]mannose-labeled products.

-F

DEAE-cellulose chromatography of the chloroformlmetha- nobwater extract showed that the n-[‘4C]glucose-labeled lipid- bound oligosaccharides were retained on the column and were then eluted with 50 mM ammonium acetate (Fig. 8). The behavior on DEAE-cellulose was similar to that observed for the n-[‘*Clmannose-labeled lipid-bound oligosaccharides, and the salt concentration required for elution was higher than that required for elution of the [14C]glucolipid in the chloro- form/methanol extract.

c-0

The n-glucose-labeled lipid-bound oligosaccharides were sta- ble to mild alkali, and labile to mild acid (Fig. 9) with a half- life similar to that observed for the n-mannose-labeled lipid- bound oligosaccharides (1). Gel filtration of the water-soluble products obtained after acid treatment showed the formation of a single major fraction (Fig. 10). In this experiment incuba- tion was performed with UDP-[3Hlglucose to compare with the n-[YJlmannose oligosaccharides. The n-[3H]glucose oligosac- charide fraction was eluted a littler earlier than the P [Ylmannose oligosaccharides although there was some over- lap. The n-[3H]glucose oligosaccharide fraction appeared more homogeneous than the n-[‘4C]mannose-labeled one. From the behavior on gel filtration it can be estimated using the method described in the preceding paper (1) that the minimum size of the n-glucose-labeled oligosaccharides is approximately 10 monosaccharide units, i.e. 2 units larger than the minimum size of the nmannose-labeled oligosaccharides.

Treatment of the n-glucose-labeled oligosaccharide fraction

FIG. 4. Thin layer chromatography of chloroform/methanol ex- tract labeled from UDP-o-[14Clglucose and from GDP-D- l’4Clmannose. Chloroform/methanol extract obtained from total mi- crosomes incubated with UDP-D-[14Clglucose under standard condi- tions, and a similar extract obtained from incubation of total micro- somes with GDP-n-1’4Clmannose as described in the preceding paper (1) were chromatographed in Solvent C. Radioautography: 1, chloro- form/methanol extract labeled from GDP-o-[14Clmannose; 2, chloro- form/methanol extract labeled from UDP-o-[L4Glglucose combined with chloroform/methanol extract labeled from GDP-D- [Vlmannose; 3, chloroform/methanol extract labeled from UDP-D- [14C]glucose. 0, origin; F, solvent front. When the o-[14Clmannose- labeled extract is prepared from incubation of total microsomes, it contains a small amount of labeled lipid-bound oligosaccharides near the origin, in addition to dolichyl p-o-[14C]mannosyl phosphate.

with cY-glucosidase caused the release of at least 80% of the radioactivity as n-glucose (Fig. 111, whereas incubation with p-glucosidase had no effect. These results indicate that at least 80% of the labeled n-glucose residues were a-linked at the nonreducing terminus of the oligosaccharide fraction. Fur- thermore, about 78% of the n-glucose-labeled oligosaccharide was retained on Con-A Sepharose (but not on Sepharose 4B1, and subsequently eluted with methyl cy-n-mannopyranoside (Fig. 12).

Utilization of Pancreatic Dolichyl D-S4C]Glucosyl Phos- phate -Incubation of calf pancreas microsomes with dolichyl n-[‘4C]glucosyl phosphate using conditions similar to those described for the utilization of dolichyl p-n-[14C1mannosyl phosphate (1) resulted in its rapid disappearance from the chloroform/methanol extract (Fig. 13). There was a rapid lin-

by guest on March 10, 2019

http://ww

w.jbc.org/

Dow

nloaded from

2274 [14C]Glucose-labeled Lipid-linked Oligosaccharides in Calf Pancreas

FRACTION NUM8ER

I / I

0 10 ’ ‘60 20 30

MINU i-ES

FIG. 5 (left). DEAE-cellulose chromatography of chloroform/ methanol extract. A chloroform/methanol extract obtained from standard incubation with UDP-D-[i4Clglucose was chromatographed on DEAE-cellulose in the acetate form (0.6 cm x 2.5 cm) equilibrated with chloroform/methanol/water (10:10:3, v/v). Sequential elution with chloroform/methanol/water (10:10:3, v/v) and chloroform/metha- nol/l rnM aqueous ammonium acetate (10:10:3, v/v) was carried out, and fractions of 1 ml were collected and assayed for radioactivity. The arrow indicates the time of addition of ammonium acetate. Of the original radioactivity, 97% was recovered.

FIG. 6 (center). Stability of the [14Clglucolipid. Equal volumes of radioactive chloroform/methanol extract (1900 dpm) obtained from a standard incubation with UDP-n-[**Clglucose were dried under N,. In mild acid hydrolysis, to each residue cooled to 4”, ice-cold 0.01 M (0) HCI (0.2 ml) was added and the mixtures were incubated at 90 for various periods of time; the samples were then cooled to 4”, neutralized with ice-cold 0.01 M NaOH (0.2 ml). In mild alkali treatment each residue was dissolved in 0.5 ml of chloroformlmetha- no1 (1:4, v/v), cooled to 4”, and 1.0 M NaOH (0.05 ml) was added (A). The mixtures were incubated for the times indicated at 37”, cooled to 4”, and neutralized with ice-cold 1.0 M acetic acid (0.05 ml). The solutions were dried under N,, and water (0.5 ml) was added. In hot

250r

2cQ-

50 -

O- I I , I I I I I 0 4 8 12 i6 20 24 28 32 36

FRAU/ON NlM8ER

FIG. 8. DEAE-cellulose chromatography of D-[14Clglucose-labeled lipid-bound oligosaccharides. A chloroform/methanol/water extract obtained from incubation of total microsomes prepared by method 1 with UDP-D-[14Clglucose under standard conditions was chromato- graphed on DEAE-cellulose in the acetate form (0.6 x 2.5 cm) equili- brated with chloroform/methanol/water (10:10:3). Sequential elution with chloroform/methanol/water (10:10:3, by volume), chloroform/ methanol/IO mM aqueous ammonium acetate (10:10:3, by volume), and chloroform/methanol/50 mM aqueous ammonium acetate (10:10:3, by volume) was carried out, and fractions of 1 ml were collected for radioactivity measurements. Of the radioactivity, 80% was recovered.

ear incorporation of radioactivity into the chloroform/metha- nol/water extract and into the precipitate which reached a maximum at about 30 min. At this time 20 to 55% of the radioactivity in the original dolichyl D-[‘4ClglucoSyl phosphate was recovered in these two fractions (Fig. 13, Table I). There was also a continuous increase of radioactivity in the aqueous phase during the course of the incubation. With some micro- somal preparations, the amount of radioactivity recovered in

0 10 20 30 40 50 60

DOLICHYL PHOSPHATE t,uMl

alkali treatment each residue was suspended in 1-propanol (0.1 ml), cooled to 4”, and 1 M NaOH (0.01 ml) was added. The mixtures were heated at 65” (A) or 90” (x) for various periods of time, then cooled and neutralized with 2 M HCI (0.005 ml). The solutions were dried under N,, and water (0.5 ml) was added. In all cases, 5 volumes of chloroform/methanol (2:1, v/v) were added, and after mixing and centrifugation, the radioactivity in the upper aqueous phase was measured and expressed as percentage of the original radioactivity in the chloroform/methanol extract.

FIG. 7 (right). Effect of dolichyl phosphate concentration on the incorporation of radioactivity from UDP-n-[i4Clglucose into the chlo- roform/methanol extract. Varying amounts of a solution of dolichyl phosphate were mixed with 10 ~1 of a solution containing equal parts of 0.2 M MnCl* and 0.1 M sodium EDTA, pH 5.3, to yield the final concentrations indicated. Each mixture was dried under N, and suspended in 20 ~1 of 2.5% (v/v) Triton X-100. Tris/maleate buffer, pH 5.3 (40 mM), total microsomes prepared by method 2 (5 mg of protein/ml), MnCl, (10 m&, and UDP-n-114Clglucose (0.05 &i/ml) were added in a final volume of 0.5 ml. Incubation was at 30” for 5 min. 0, Triton X-100 included in the incubation mixture; 0, incuba- tion performed with dolichyl phosphate in the absence of Triton X- 100.

L 100

80

60

40

20

0

I 1 I I I 0 i0 20 30 60

MIMI rE.s

FIG. 9. Stability of the n-l’4Clglucose-labeled products of the chlo- roform/methanol/water extract. The chloroform/methanol/water ex- tract (1000 dpm) obtained from a standard incubation with UDP-D- [14Clglucose was treated with 0.1 M HCl at 90” (0) or with 0.1 M NaOH at 37” (A) as described in Fig. 6.

the aqueous phase was much higher than that observed in Fig. 13, and the labeling of the chloroform/methanol/water extract and the precipitate was lower (Table I).

The addition of UDP to the incubation medium, but not of UMP, greatly stimulated the formation of UDP-D-[‘4C]glucose from dolichyl D-[i4C]glucosyl phosphate (Fig. 14). The presence of UDP inhibited the incorporation of radioactivity from doli- chyl D-[i4C]glucosyl phosphate into the chloroform/methanol/ water extract and into the precipitate (Table I), thereby dem-

by guest on March 10, 2019

http://ww

w.jbc.org/

Dow

nloaded from

I-

b 160 x

g 120

60

40

0 E

[‘4CIGlucose-labeled Lipid-linked Oligosaccharides in Calf Pancreas 2275

1 I I t I

20 30 40 50 60 70

FRAC?-IOff NUMBER

FIG. 10. Gel filtration of n-[“Hlglucose- and n-[14Clmannose-la- beled oligosaccharides. Chloroform/methanol/water extracts from separate incubations of rough microsomes prepared by method 1 with UDP-n-13Hlglucose and with GDP-n-I’*Clmannose were dried under N, and hydrolyzed with 0.1 M HCl at 90” for 30 min. After neutralization with 0.1 ml of M NaOH, the hydrolysates were com- bined and chromatographed on a column of Bio-Gel P-6, 200 to 400 mesh (1 x 120 cm), with bovine serum albumin to determine V, and n-mannose. Elution was with 0.1 M pyridine acetate, pH 5. Fractions of approximately 1 ml were collected and assayed for both 3H and 14C radioactivity. Appropriate corrections for separation of the isotopes were used.

on&rating that the increased availability of UDP-D- [Wlglucose did not cause an increased labeling of these frac- tions. Under these conditions of incubation, i.e. in the pres- ence of detergent and in the absence of divalent cation, there was negligible incorporation of radioactivity using UDP-D- [Wlglucose as precursor. It can be concluded therefore, that dolichyl D-[14c]ghCOSyl phosphate was acting as donor of D-

glucose residues without being first converted to UDP-n-glu- case.

After acid hydrolysis, the lipid-bound [W]oligosaccharide fraction labeled from dolichyl D-[‘4c]ghCOSyl phosphate had the same elution volume on Bio-Gel P-6 as that obtained after incubation with UDP-D-[“Clglucose. Similarly, most of the radioactivity could be released from the oligosaccharides as D-

[Wlglucose with cr-glucosidase, but not with p-glucosidase.

DISCUSSION

These results demonstrate that calf pancreas microsomes incubated with labeled UDP-n-glucose synthesize a labeled glucolipid with the properties expected of dolichyl n-glucosyl phosphate. This glucolipid is extremely labile to mild acid with a half-life similar to that of dolichyl /3-n-[Wlmannosyl phosphate, and exhibits the same stability to mild alkali and the same chromatographic behavior on DEAE-cellulose as dolichyl P-n-[W]mannosyl phosphate. Its formation is stimu- lated by exogenous dolichyl phosphate, and its chromato- graphic mobility on silica gel is similar to that of pancreatic dolichyl P-n-[“Qmannosyl phosphate. During hot alkali treatment, a product which chromatographs with 1,6-anhydro- P-n-glucopyranose is formed; it seems likely therefore that the n-glucose residue is present in the p-anomeric configuration as was suggested for the dolichyl D-ghCOSy1 phosphate formed in liver microsomes (8). The enzymic synthesis of dolichyl D-

16 /

I,-@

I! 6-

@ GLUCOSE

16 -

14 - “0

‘*- I

h

10 -

b s- Y o . 6-

4-

70 - FRACTlOh’ MM9ER

FIG. 11. Effect of a-glucosidase on n-[3Hlglucose-labeled oligosac- charides. Equal volumes of n-[3Hlglucose-labeled oligosaccharides prepared from a chloroform/methanol/water extract as described in Fig. 10 were incubated at 37” for 18 h in a total volume of 140 ~1 containing 50 ~1 of 0.1 M potassium phosphate, pH 6.8, and 50 ~1 of cycloheximide/chloramphenicol (500 pg/ml each). After incubation, the samples were chromatographed on Bio-Gel P-6 as described in Fig. 10. A was from a control incubation, and B from an incubation mixture containing 2.4 units of a-glucosidase. Similarly, D-

[“HIglucose-labeled hydrolysates were incubated with fi-glucosidase (2.4 units) under the same conditons except that the buffer was 0.05 M sodium acetate, pH 5, containing 0.1 M NaCl and 0.1 rnM ZnSG,, and the medium included 10 ~1 of bovine serum albumin (1 mgrml). Both control and P-glucosidase-treated samples showed the same profile as A after gel filtration. Similar results were obtained with the [Wloligosaccharides obtained from incubation with dolichyl D-

[‘4Clglucosyl phosphate as described in Fig. 13.

glucosyl phosphate was shown to be reversible in the presence of UDP, but not UMP: dolichyl phosphate + UDP-D- [Wlglucose = dolichyl ,&D-[‘4c]ghCOSyl phosphate + UDP. This enzyme activity has been demonstrated in liver (8) and preliminary reports indicate its occurrence in several other animal tissues (9-12).

Calf pancreas microsomes incubated with labeled UDP-glu- case also synthesize labeled lipid-bound oligosaccharides. These products have the same properties with respect to acid and alkali lability as the n-mannose-labeled lipid-bound oligo- saccharides described previously (l), and behave on DEAE- cellulose like polyprenyl derivatives containing a pyrophos- phate group since they are eluted at a higher salt concentra- tion than the dolichyl n-glucosyl phosphate.

There is a smaller number of n-glucose-labeled than of D-

by guest on March 10, 2019

http://ww

w.jbc.org/

Dow

nloaded from

2276 S4CIGlucose-labeled Lipid-linked Oligosaccharides in Calf Pancreas

FIG. 12 (left). Con-A Sepharose column chromatography of D- [3H]glucose-labeled oligosaccharides. D-[3HlGlucose-labeled oligo- saccharides obtained as described in Fig. 10 were dissolved in a solution containing 1 m&f each of MnCl,, MgCl,, and CaCl,, and then passed through a column of Con-A Sepharose (0.6 x 2 cm) previously equilibrated with the same solution. Elution was performed with the same salt solution. The arrow indicates the time of addition of 0.1 M methyl-a-o-mannopyranoside; fractions of 1 ml were collected and assayed for radioactivity. About 86% of the applied radioactivity was recovered.

FIG. 13 (right). Utilization of pancreatic dolichyl D-[‘4Clglucosyl phosphate. Total microsomes prepared by method 2 were incubated for different times with 2600 cpm of dolichyl D-[‘4Clglucosyl phos- phate under standard conditions. A, chloroform/methanol extract; 0. chloroform/methanol/water extract; 0, precipitate; A, aqueous phase.

TABLE I

Effect of UMP and UDP on pancreatic dolichyl D-[‘4c~gkLcosy~ phosphate utilization

Standard conditions described under “Experimental Procedures” were used, with 5 mM of the uridine nucleotides where indicated.

Additions Aqueous phase Precipitate Chloroform/

methanol/water extract

CPm CPm CPm

None 1570 310 240 UMP 1520 390 340 UDP 2050 100 160

mannose-labeled lipid-bound oligosaccharides formed in the microsomes, as may be expected from the smaller number of D- glucose residues obtained by gas-liquid chromatographic anal- ysis (l), and from their larger apparent size. The average D- glucose-labeled oligosaccharide moiety appears to be larger than the average D-mannose-labeled products, suggesting that D-glucose residues are incorporated into the oligosaccharides closer to the nonreducing terminus than most of the D-man- nose residues. Since the labeled D-glucose can be released by a-glucosidase, the incorporated D-glucose residues must be (Y- linked at the nonreducing end of the oligosaccharides. These results do not necessarily imply that D-glucose is the terminal sugar in the oligosaccharides which accumulate in the tissue, since the labeled products may only represent stages in the formation of these compounds.

Dolichyl D-[‘4Clglucosyl phosphate was demonstrated to act as an efficient donor of D-ghCOSe residues to the lipid-bound oligosaccharides and to the products in the precipitate, under conditions in which utilization of UDP-D-[‘4C]glucose itself was negligible. It seems therefore that dolichyl D-glUCOSy1 phosphate acts as an intermediate in the transfer of D-glucose with a similar inversion of configuration as that observed previously for the transfer of D-mannose residues (1). The (Y- linked D-glucose residue in UDP-D-glucose is a precursor of p- linked D-glucose in dolichyl D-ghCOSy1 phosphate, which in turn can act as donor of a-linked D-glucosyl residues in the

FIG. 14. Effect of UMP and UDP on the water-soluble products formed from dolichyl D-114Clglucosyl phosphate. Paper chromatogra- phy in Solvent H of the aqueous phases obtained from the experi- ment described in Table I. A, radioactive scan of the aqueous phase from the control incubation without uridine nucleotides. A similar scan was obtained from incubation mixtures containing UMP. B, radioactive scan of the aqueous phase from incubation with UDP.

oligosaccharides. The role of dolichyl D-glucosyl phosphate as intermediate in the formation of lipid-linked oligosaccharides observed in the pancreas is analogous to that previously re- ported for liver by Behrens et al. (2).

The fact that calf pancreas microsomes contain enzyme activities which result in the incorporation of D-glucose into acidic lipid-bound oligosaccharides is in good agreement with our previous work (1). Analysis of the unlabeled lipid-bound oligosaccharides associated with calf pancreas microsomal membranes showed that the oligosaccharide moiety contained N-acetyl-D-glucosamine/D-mannose/D-glucose varying in pro- portion from about 2:5:1 to 2:10:3, the larger oligosaccharides being the most abundant (1).

Recently, Spiro et al. (13) described an oligosaccharide-lipid containing 2 N-acetyl-D-glucosamine, 11 D-mannose, and 1 or 2 D-glucose residues in thyroid. Studies in thyroid slices showed that this compound behaved like an intermediate in glycopro- tein biosynthesis (4). Similar lipid-bound oligosaccharides were also formed in slices from several other tissues, except that the compound extracted from calf pancreas slices was smaller and deficient in D-glucose (5). It seems possible that the difference in the results obtained with pancreas slices compared to those obtained with microsomes may be due to the release and action of some glycosidases during incubation of the slices.

Acknowledgments -We thank Dr. C. D. Warren for the dolichyl phosphate, and Dr. Evelyn Walker-Nasir for the 1,6- anhydro-/3-D-glucopyranose.

Note Added in Proof-The o(- and P-anomers of dolichyl D-

glucopyranosyl phosphate have since been synthesized chemi- cally. The ol-anomer was stable to treatment with 0.1 M NaOH at 65-90” for 30 min, whereas the p-anomer was labile under these conditions. 1 These observations support our conclusion that dolichyl p-D-glucosyl phosphate is formed in the pancreas.

1 C. D. Warren, unpublished observations.

by guest on March 10, 2019

http://ww

w.jbc.org/

Dow

nloaded from

1.

2.

3.

4.

5.

6.

[X’IGlucose-labeled Lipid-linked Oligosaccharides in Calf Pancreas 2277

REFERENCES 7.

Herscovics, A., Golovtchenko, A. M., Warren, C. D., Bugge, B., and Jeanloz, R. W. (1977) J. Biol. Chem. 252, 224-234

6,

Behrens. N. H.. Parodi. A. J.. and Leloir. L. F. (1971)Proc. N&l. - Acad. ‘Sci. lJ: S. A. 68, 2867-2860 ’ Y.

Parodi. A. J.. Staneloni. R.. Cantarella, A. I., Leloir. L. F.. . _ Behrens, N.’ H., Carminatti, H., and Levy, J. A. (1973) Car& I”. hydr. Res. 26, 393-400

Spiro, M. J., Spiro, R. G., and Bhoyroo, V. D. (1976) J. Biol. “’ Chem. 251, 6400-6408

Spiro. M. J.. Spiro. R. G.. and Bhovroo. V. D. (1976) J. Biol. 12’ -Chdm. 251;6i20-6425 ”

Parodi. A. J.. Behrens. N. H.. Leloir. L. F.. and Carminatti. H. 13’ (197i) Proc: Natl. A&d. Scl. U. S,‘A. 69; 3268-3272

Tkacz, J. S., Herscovics, A., Warren, C. D., and Jeanloz, R. W. (1974) J. Biol. Chem. 249, 6372-6381

Behrens, N. H., and Leloir, L. F. (1970) Proc. Natl. Acad. Sci. U. S. A. 66, 153-159

Jankowski, W., and Chojnacki, T. (1972)Biochim. Biophys. Acta 260, 93-97

Breckenridge, W. C., and Wolfe, L. S. (1973) FEBS Lett. 29, 66- 68

Wedgwood, J. F., Warren, C. D., and Strominger, J. L. (197415. Biol. Chem. 249, 6316-6324

Waechter. C. J.. Lucas. J. J.. and Lennarz. W. J. (1974) Bio- them. Ekophys: Res. don&n. 56, 343-350’

Spiro, R. G., Spiro, M. J., and Bhoyroo, V. D. (1976) J. Biol. Chem. 251, 6409-6419

by guest on March 10, 2019

http://ww

w.jbc.org/

Dow

nloaded from

A Herscovics, B Bugge and R W JeanlozUDP-D-[14C]glucose.

D[14C]glucosyl phosphate and 14C-labeled lipid-linked oligosaccharides from Glucosyltransferase activity in calf pancreas microsomes. Formation of dolichyl

1977, 252:2271-2277.J. Biol. Chem. 

  http://www.jbc.org/content/252/7/2271Access the most updated version of this article at

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

to choose from all of JBC's e-mail alertsClick here

  http://www.jbc.org/content/252/7/2271.full.html#ref-list-1

This article cites 0 references, 0 of which can be accessed free at

by guest on March 10, 2019

http://ww

w.jbc.org/

Dow

nloaded from