Evidence for uniformity of the carbohydrate chains in individual glycoprotein molecular variants

Vol.

95, No.

July

31, 1980

BIOCHEMICAL

2, 1980

AND

BIOPHYSICAL

RESEARCH

COMMUNICATIONS

Pages 777-754

EVIDENCE

FOR UNIFORMITY OF THE CARBOHYDRATE CHAINS INDIVIDUAL GLYCOPROTEIN MOLECULAR VARIANTS

IN

Bernard

IiERCKAERT

Institut de Recherches sur Ze Cancer de Lille, 124 de Z'INSERM, BP 311, 59020 Lille Cddex,

Unite' Received

BAYARD and Jean-Pierre

France

June 16,198O

SUMMARY.- Pooled and alkylated al-acid glycoprotein was fractionated on a Con A-Sepharose column into two fractions : Con A-non reactive and Con A-reactive. The carbohydrate moiety from the al-acid glycoprotein Con A-reactive variant, obtained by hydrazinolysis and quantitative re-N-acetylation, contains only identical two-branched oligosaccharide chains. From the present work on al-acid glycoprotein and from previous studies on aI-fetoprotein, one can assume that glycoprotein glycosylation occurs uniformly along each polypeptide chain giving it identical oligosaccharide units at each glycosylation site. INTRODUCTION Most of glycoproteins ty experiments,

indicating

However,

structural

generally

using

are characterized determine

(i)

each

individual

(ii)

whether

for whether

was obtained

ve,

carbohydrate

site

variant

glycan of the

individual

enzymic

digestion

of

of the glycoprotein

and several

glycan

glycoprotein,

units

are

it

unevenly

glycoprotein

molecular

(1).

have been performed

by exhaustive

microheterogeneous

different

moieties

glycans

molecular populations

previous studies which

favours

the

two identical

reactive

two-branched

on the

the

upon lectin-affini-

or,

structures remains

to

distributed

on the

variants

stu-

carries

on

contrary, only

one type

chain.

contains

Con A-non

obtained

a given

each of these

From our

in their

one glycosylation

molecular

of carbohydrate

riant

glycopeptides

more than

heterogeneity

of asparagine-linked

heterogeneous

Then,if

molecular

variations

studies the

unfractionated, died.

exhibit

AFP variant glycan

units

on rat

alpha-fetoprotein

second

possibility

Con A-reactive, possesses with

(AFP)

3),

: the Con A-reactive two-branched

two other

an additional

(2,

identical

glycan

units

Con A-non

N-acetylglucosamine

evidence AFP vaand the reactiresidue

B-mannose. 0006-291X/80/140777-08$01.00/0 717

Copyright 0 1980 by Academic Press, Inc. All rrghfs of reproduction in an.)’ form reserved.

Vol.

95, No.

2, 1980

In order proteins

than

sent

work

tein

contains

tes

(4,

with

BIOCHEMICAL

to extend rat

this

the study about

5) as bi,

only

variant

one type

der the concept variant

in the

to less

glycan

of glycan

unit.

fondamental

process

MATERIAL by a method

These

glyco-

in the preThis

glycopro-

glycosylation (6).

glycans

variant

si-

Fractionation of the Con A-

at least

contains

lead

us to consi-

results

of individual

glycoprotein

molecular

glycosylation.

AND METHODS

Glycaphv&Ln in o.tUiun. plasma

this

of protein

five

structures of the

that

glycan-uniformity

(AGP).

onto

and characterization

(two-branched) of the

we have dealt

distributed

indicated

COMMUNICATIONS

heterogeneous

glycoprotein

and tetra-branched

of AGP have

RESEARCH

simple

content),

of human al-acid

43 % carbohydrate tri

BIOPHYSICAL

observation

AFP (6 % carbohydrate

of AGP on Con A-Sepharose reactive

AND

Human AGP was isolated described by Biirgi and Schmid

from (7).

a pool

of normal

human

Reduction and a.tkyLaaXon 06 Lhe glycoptLoLtein. Human AGP was reduced and alkylated with 0.1 mCi of iodo [I4Cl acetamide (specific activity 53 mCi/mmole) according to the method of Crestfield et aZ. (8). A6&iniZq ch4omatvghaphy. al-acid glycoprotein was chromatographed on Con A(9). Some commercial Sepharose column as described for rat a1 -fetoprotein Con A-linked agarose or Sepharose display different glycoprotein affinity patterns (10). We used commercial Con A coupled with Sepharose 4B previously activated according to the CNBr method of March et aZ. (11). The al-acid glycoprotein elution profile was obtained by rocket immuno-electrophoresis of each fraction on anti-AGP impregnated agarose plates (12) (anti-AGP was purchased from Immuno, Diagnostic Division, Belgium). Glycoproteins were also analysed by crossed-affino-immuno-electrophoresis with free Con A in the first dimension gel as previously described (13, 14). al-acid glycoprotein or its txhatitivive hy&azinoLyb-ih 06 Rhe gLywptloAein. Concanavalin A-affinity variants were subjected to hydrazinolysis as previously described (15, 16). Human al-acid glycoprotein (1 to 5 mg) was treated with distilled hydrazine (0.5 ml) in screw-cap vials at 100°C for 30 hours. The reaction mixture was then evaporated under a nitrogen stream. The dry residue was dissolved in 1 ml of 20 % acetic acid previously cooled to 4°C. The mixture was then desalted on a Biogel P 6 column ( 5 x 30 cm). and purified de-Nanhydride (17). acetylated glycans were re-N-acetylated with 14 C acetic

Thin .laye,t chnvmcctog4aphy. A thin the

separation

of glycoprotein-derived

layer

chromatographic oligosaccharides

system was used for (2).

of monosaccharides was estimated by Cahbohydute ana.Qbi.&. The molar ratio gas liquid chromatography according to Zanetta et ~2. (18) after methanolysis with methanol-O.5 M HCl at 80" for 20 hours (19). Total carbohydrates were determined by classical calorimetric methods (20).

778

Vol.

95, No.

2, 1980

BIOCHEMICAL

b

AND

BIOPHYSICAL

RESEARCH

COMMUNICATIONS

a

-26

-0Qc

ELECTROPHORESIS

Figure l.Right Peaks a (fractions Tris-Kl (nH 7.6) 0.15 M methyl-a-t)-glucose fraction c. Left the chromZtojF$Yhic

.

CHROMATOQRAPWY

: Elution pattern of human AGP 12-20) and b (fractions 26-60) 1 M NaCl/l fi MgC12/1 mM MnC12/1 resulted in the elution : Affino-imnuno-electrophoresis fractions eluted from the Con

on Con A-Sepharose column. were eluted with 0.05 M mM CaC12. Addition of of the Con A-reactive sequential analysis of A-Sepharose column.

RESULTS al-Acid

g~ycopm&in

chromatography three

on

fractions

reactive.

lecular

A-Sepharose,

(a)

Con -a and

desorbed

1 the

resis

Con

Fractions

specifically re

:

same

using form

Con free is

he&tiogmziXy

by

A in

approximately

al-acid

A-non

reactive,

-b are

eluted

addition

A affinity Con

on

of

pattern the

first : peak

Cm

A-Sephtion~.

glycoprotein (b) with

0.15 is

Con the

can A-weakly

be

dimension a 40.6

gel. %,

peak

resolved

buffer

M methylglucoside. by

affinity

reactive,

equilibration

obtained

119

Upon

As

shown

into (5)

Con

and

-c is

in

figu-

affino-immuno-electrophoThe

percentage

b 43.7

%,

of and

peak

each

mo-

c 15.6%.

A-

Vol.

95, No.

2, 1980

BIOCHEMICAL

AND

BIOPHYSICAL

RESEARCH

COMMUNICATIONS

E B 0"

Figure mated tailed As for

2.-

Affinity al-acid in figure 1.

rat

tions

chromatography glycoprotein.

aI-fetoprotein,

the

different

degrees

reflect

S-carboxymethylation non reactive

and the

completely

fraction

is due to hydrophobic backbone.

mannose

studied

of the

hydrazinolysis

the labeled

were

glycoprotein

carbohydrate

followed

glycoprotein

the presence

to be very

different

whereas

of the of AGP

Con A-non

re-

basis of three (Table

a molar

I).

The

composition

the Con A-reactive

frac-

composition.

Con A variants

chains were

re-N-acetylation.

780

conformation

on the

possess

The carbohydrate

by quantitative

of this

interactions

calculated

chain

the Con A-

molar ratios of the two

variants

oligosaccharide

hda-thn.

al-acid

found

After

reactive

S-carboxymethylated

The carbohydrate

glycans,

21,

Con A-weakly

special

frac-

Con A (9).

(Fig.

non sugar-specific

per

a two-branched

; the

and/or

and reactive),

0LLgonacchanLdtu

xymethylated

peak

reactive

tetra-branched

exhibits

only

al-acid

are obtained

interaction these

Con A affinity with

So, we can assume that

AGP variants.

reactive

glycoprotein

of interaction

(non

residues

Con A-non

or types

I

To avoid

and reactive

Con A fractions

tion

disappears

Con A we have

active

al-acid

Con A-reactive

fraction

with

three

of AGP on ly two peaks

peak,

polypeptide

on Con A-Sepharose of the T14C] S-carboElution conditions were the same as de-

from

obtained

the

two S-carbo-

by exhaustive

The hydrazinolysis

Vol.

95, No.

Table

I.-

BIOCHEMICAL

2. 1980

Molar

ratios

and in

AND

of monosaccharides

their

derived

present

A non Peak

Con

in

RESEARCH

COMMUNICATIONS

S-carboxymethylated

AGP Con A fractions

glycans.

S-carboxymethylated Monosaccharides

BIOPHYSICAL

Glycans

AGP

reactive 1

Con A reactive Peak 2

Peak

from Peak

1

Mana

3.0

3.0

3.0

3.0

Gal

4.3

2.4

4.0

2.1

GlcNAc

6.3

4.4

6.4

3.9

Fuc

0.8

0

0.7

0

NANA

4.3

2.0

4.1

1.7

aMolar

ratio

on the

procedure releases

basis

breaks

of 3 moles

the N-glycosyl

oligosaccharide

were

of low

by thin

quantities

ponding

layer

al-acid

(6).

thin

layer

of glycans

ving

glycan

as the unique

coprotein

fraction. in figure

ched oligosaccharide

chain.

and quantitatively

treatment

acetamido

residues

are also

acid

with

reveals

reactive

used,

E14C1 acetic

linkages cleaved.

oligosaccharides anhydride

is

(Fig.

one in the

chain

of rat

3) a very

chains

and characte-

This pattern

fast

AFP possess

oligosaccharide an identical

described

the fastest

component

of the Con A-reactive

moving

781

for

is,

migration

by patmo-

in fact,

al-acid

and the

unfrac-

corres-

the same complex

except moving

from

pattern

chains

presents

fast

obtained

complex

oligosaccharide

glycoprotein

absent.

3, this

of the glycans

AGP fraction

unfractionated

3) which

recovered

As shown

linkage

this

generally

of the different

as the

(Fig.

During

chromatography

glycoprotein

The Con A-non

tern

amide

side

chromatography.

to the mixture

others

of material

re-N-acetylated

As expected, tionated

(21).

carbohydrate

and N-acetylneuraminic

quantitatively

rized

per

asparagine

units

of N-acetylglucosamine Because

of mannose

gly-

two-branrate.

2

Vol.

95. No.

2, 1980

BIOCHEMICAL

AND

BIOPHYSICAL

RESEARCH

COMMUNICATIONS

A

B

C

0

Figure 3.Thin layer chromatography of glycans which derived xymethylated AGP (A) and their 2 Con A fractions : non reactive reactive (C). (D) Rat AFP two-branched glycan having the same the human AGP two-branched oligosaccharide chain (6).

the

Moreover

the

basis

carbohydrate

molar

of 3 mannose

ratio

residues

made on the Con A affinity

of this

per glycan,

variant

glycan

of intact

I),

(Table

confirmed

from S-carbo(B) and structure than

calculated

the carbohydrate

al-acid

on analysis

glycoprotein.

DISCUSSION In order catalyse

to determine

the successive

glycoproteins,

it

glycans

and their

unique

two-branched

Con A reactive

steps

of glycosyltransferase

of the biosynthesis to establish

localization

at each glycosylation

oligosaccharide

al-acid

glycoprotein

in the Con A-non

lowing

general

concept

asparagine

linked

reactive

the

precise

chemical site.

at each carbohydrate

variant

and the absence was additional

: each glycoprotein oligosaccharide

variant chains

structure

site

contains

of of an

for

type

of

the

fol-

identical

at each glycosylation

complex site

of

the molecule.

It is now established ride

"first

involves

the

that

biosynthesis

en b&c

transfer

782

of asparagine of a high

mannose

linked

of

of the

of this

support

which

chains

The presence

chain

molecule

system

of the carbohydrate

is necessary

glycan

type

the specificity

oligosaccha-

oligosaccharide

Vol.

95, No.

chain

2, 1980

from

plasmic

BIOCHEMICAL

a lipid

carrier

reticulum,

oligosaccharide review

of the

as that

plest

view

tion

that

described

each

possible tems. lyse

to find It

the

still

either

in well

types

of cells

ferases

defined

steps

and tissues

a

preliminary chain

could

was not

be explained

by

due to the folding

N. and Loucheux-Lefebvre, could

complex

explain

type

the

coexistence

asparagine

that

uniformity This

of the

it

along concept

linked

oli-

Since

these

within

as has been demonstrated

allowsthe

sim-

and assumpbe processed

molecular

microheterogethat

it

will

be

glycosyltransferase enzymic

of monosaccharide the same Golgi

the polypeptide

would

can be speculated

different

glycoprotein

chain

chain

microheterogeneous if

other

carbohydrate

system.

exists,

of addition

areas

for

glycosylation

accessible

of the polypeptide

chains

unknown

successive

less

endo-

of the

(22-24,

oligosaccharide

one can expect

glycosy ltransferase

remains

chain"

chain,

al-fetoprotein.

site

corresponding

rough

(27).

o f maturation

the

in the

Yet recent

Helbecque,

with

here,

rat

carbohydrate

in oligosaccharide

mannose

same oligosaccharide for

of the process

by the same defined neity

the

(26).

observation

chain

presented

may possess

chain

This

COMMUNICATIONS

by the processing

carbohydrate

become

in glycoproteins

results

chain

area

J.P.,

oligosaccharide

chains

From the variants

(Aubert,

polypeptide

carbohydrate

B-turns

communication).

mannose

type

RESEARCH

in most glycoproteins,

a high

complex

chain

type

that

that

BIOPHYSICAL

apparatus

in B-turn

some particular

personal

gosaccharide

complex

the fact

polypeptide

of high

the Golgi

occurs

a mature

that

nascent

has been shown

glycans

into

the fact

M.H.,

It

concerning

processed

in

to the mature

of N-asparagine results,

to the

followed

see 25).

AND

systems units

apparatus for

are

which

syscata-

present

or in different

specific

sialyltrans-

(25).

ACKNOWLEDGMENTS The authors wish to thank Professor G. Biserte for helpful discussion and encouragement in this work and Dr. Carol Smith for reviewing this paper. The skillful assistance of Mrs. D. Roux and S. Quief is acknowledged. This work was supported by the Delegation G&i&t-ale a la Recherche Scientifique et Technique (Grant no 78.7.2643) and by the Institut National de la Sante et de la Recherche Medicale (C.R.L. no 78.5.172.2).

783

Vol.

95, No.

2, 1980

BIOCHEMICAL

AND

BIOPHYSICAL

RESEARCH

COMMUNICATIONS

REFERENCES

:: 3. 4. 5.

6. 7. 8.

Smith, C. and Kelleher, P. (1980) Biochim. Biophys. Acta 605, l-32. G. (19v in "Protides Bayard, B., Kerckaert, J.P., Roux, D. and Strecker, of Biological Fluids" (Peeters, H., ed.),Vol. 27, pp. 153-156, Pergamon Press, Oxford. Bayard, B. and Kerckaert J.P. Eur. J. Biochem. (in press). Ikenaka, T., Ishiguro, M., Emura, J., Kaufmann, H., Isemura, S., Bauer, W. and Schmid, K. (1972),Biochemistry 11, 3817-3829. S., Bauer,., Emura, J., Motoyama, T., Schmid, K., Kaufmann, H., Isemura, Ishiguro, M. and Nanno, S. (1973) Biochemistry 12, 2711-2724. Fournet, B., Montreuil, J., Strecker, G., Dorland, L., Hzerkamp, J., Vliegenthart, J., Binette, P. and Schmid, K. (1978) Biochemistry 17, 5206-5214. BUrgi, W.%d Schmid, K. (1961) J. Biol. Chem. 236, 1066-1074. Crestfield, A.M., Moore, S. and Stein, W.H. (19w J. Biol. Chem. -’238

622-627.

10.

B. and Kerckaert, J.P. (1977) Biochem. 489-495. Kerckaert, J.P. and Bayard, B. (1980) Biochem.

11.

March,

12. 13.

Laurell, Nilsson,

9.

14. 15. 16. 17. 18.

Bayard,

95-101.

Res.

Commun. -77,

Biophys.

Res.

Commun. -’ 92

Parikh, I. and Cuatrecasas, P. (1974) Anal. Biochem. 149-152. C.B. (1966) Anal. Biochem. 15, 45-52. M. and Bdg-Hansen, T.C. (19m) in "Protides of Biological

S.C.,

60,

Fluids" Pergamon Press, Oxford. (Peeters, H., ed.) Vol. 27, pp. 599-602, Kerckaert, J.P., Bayard, B. and Biserte, G. (1979) Biochim. Biophys. Acta -'576 99-108. Bayard, B. and Fournet, B. (1976) Carbohyd. Res. 46, 75-86. Bayard, B. and Roux, D. (1975) FEBS-Lett. 55, 206711. Reading, C.L., Penhoet, E. and Ballou, C. p978) J. Biol. Chem. -253, 56005612. Breckenridge, W.C. and Vincendon, G. (1972) J. Chromatogr. Zanetta, J.P., Clamp,

27.

Biophys.

69, 291-304. J.K, Bhatti, T. and Chambers,

R.E. (1972) In "Glycoproteins", pp. 300-339, Elsevier Publishing Co., New York. Montreuil, J. and Spik, G. (1963) "Methodes colorimetriques de dosage des glucides totaux, Lab. Chim. Fat. Sci., Lille, ed. Isemura, M. and Schmid, K. (1971) Biochem. J., 124, 591-596. Li, E., Tabas, I. and Kornfeld, S. (1978) J. Six Chem. 253, 7762-7770. Chem. 253, 7771-7778. Kornfeld, S., Li, E. and Tabas, I. (1978) J. Biol. Tabas, I. and Kornfeld, S. (1975) J. Biol. Chem. 253, 777m86. Parodi, A. and Leloir, L. (1979) Biochim. Biophys.Acta 259, l-37. M.H. (1Vm) Arch. Biochem. Aubert, J.P., Biserte, G. and Loucheux-Lefebvre, Biophys. 174, 410-418. Van den Eijnden, DTStoffyn, P., Stoffyn, A. and Schiphorst, W. (1977) Eur. J. Biochem. -81, l-7.