Heterogeneity in high performance liquid chromatography of a variant surface glycoprotein of Trypanosoma brucei

BIOCHEMICAL

Vol. 144, No. 3, 1987

AND BIOPHYSICAL

RESEARCH COMMUNICATIONS Pages 1222-1228

May 14, 1987

HETEROGENEITY IN HIGH PERFORMANCE LIQUID CHROMATOGRAPHY OF A VARIANT SURFACE ELYCOPROTEIN OF TRYPANOSOMA BRUCEI’ Daniel Dominique Unite Received

TETAERT,

Veronique

GOMES, Guillemette

HUET-DUVILLIER

DEMEYER, Marylene HUBLART, Arnold BOERSMA and Pierre DEGAND* INSERM, No16 Place de Verdun, 59045 LILLE CEDEX (France)

March

2, 1987

High performance liquid chomatography (HPLC) procedures have been used to analyze a preparation of the variant surface glycoprotein AnTat l.lA of The native preparation gives several peaks with a high Tr anosoma brucei. -5-r repro uci ility both by reverse-phase (RP-) and gel permeation (GP-) HPLC. Under RP-HPLC conditions, nine fractions are fully resolved. The RP-HPLC fractions migrate with the same molecular weight VSG band on polyacrylamide slab gel electrophoresis and no significant differences are observed in amino acid composition among these fractions. The RP-HPLC resolution is found to be related to the ability of the VSG to polymerize as shown using GP-HPLC. These results suggest the existence of a microheterogeneity of the AnTat l.lA VSG in relation to post-translational modification of the VSG preparation 0 1987 Academic Press, Inc. molecule. The immune tion

parasitic

system is

by antigenic

the

distinct

protozoan

sequential

variant

the

entire

the

membrane

contains during

to the

of

*

RP-HPLC

(5), liquid on C

3

(1,2).

which

are

The is

have

has

as

of

the

for

proven

should

to

be fast

coat

(3).

at

the

occurs

purification (6),

(7).

of

The

VSG purification.

: mf VSG

time

of

the

spontaneously enzyme VSG :

reverse

Particularly, for

covering

been characterized

(sVSG)

which

varia-

immunologically

a dense

have

host

antigenic

of an endogeneous

chromatography (RP-HPLC)

of

diplyceride

described

mammalian

encoding form

form

sVSG,

action

immunoaffinity

column

genes,

soluble

released

the

basis

which

due to the

been

evades

Two forms

the

release

chromatography

To whom correspondence

separate (VSG)

VSG) and

disruption, methods

of

parasite

VSG (mf

acids

chromatography performance

the

brucei

The molecular

glycoproteins

sVSG (4).

cell

Several

expression of

form

fatty

conversion

variation.

surface

surface

Trypanosoma

phase the

(3). lectin high

procedure Further-

be sent.

Abbreviations : VSG, variant surface glycoprotein ; sVSG, soluble form VSG ; mf VSG, membrane form VSG ; AnTat, Antwerpen Trypanozoon-antigenic-type ; RP-HPLC, reverse phase high performance liquid chromatography ; GP-HPLC, gel permeation high performance liquid chromatography ; TFA, trifluoroacetic acid. SDS-PAGE, sodium dodecyl sulfate polyacrylamide gel electrophoresis. 1This work was supported by grants from the Commission of the European Communities (TSD-146-F (MR)), the U.E.R. de Medecine, Universite de Lille II and the Fondation pour la Recherche Medicale Francaise. 0006-291X/87 Copyright All rights

$1.50 0 I987 by Academic Press, of reproduction in any form

Inc. reserved.

1222

BlOCHEMlCALANDBlOPHYSlCALRESEARCHCOMMUNlCATlONS

Vol. 144. No. 3, 1987 more,

VSG is

obtained

in

high

work,

the

yield

and purity,

in

the

absence

of

detectable

proteolysis. In

the

present

(GP-HPLC)

has

the

1.1A

AnTat

been

combination

developed

VSG purified

to

of

investigate

according

to

MATERIAL

RP-HPLC the

the

and gel

permeation

chromatographic

conditions

of

HPLC

behaviour

Baltz

--et

al

of

(5).

AND METHODS

Growth and isolation of the bloodstream forms brucei AnTat l.lA and purification by Concanavalin chromatography were essentially as described previously

of

Trypanosoma A-Sepharose (8).

brucei affinity

HPLC methods were developed using Varian 5000 and Waters 660 systems. For RP-HPLC, a Beckman RPSC-Ultrapore C3 (300 w pore size, 4.6 x 75 mm) column was used. Eluates were detected with an UV absorbance detector LKB 2238 Uvicord S II and measurements were made at 206 nm in the sensitivity of 0.2 absorption unit full scale. The samples (loo-150 ug) were introduced into RP-HPLC column and eluted at a flow rate of 1 ml/min. The following gradient systems were employed : (i) conditions described according to Clarke M TFA in water --et al (7) : 0.0013 (solvent A) to 60 % 2-propanol (solvent 6) ; O-30%8, 10 min ; 30%B, 5 min ; 30-45%B, 40 min ; 45%B, 10 min ; 45-lOO%B, 10 min, (ii) in the same gradient elution conditions, but the solvent B is replaced by 1-propanol ; (iii) ammonium formate in water (solvent C) as described by Grandier-Vazeille and Tetaert (9) to 60 % 1-propanol (solvent D) : 33%D, 10 min ; 33-45%D, 40 min ; 45%D, 10 min ; 45 -lOO%D, 10 min. The buffers were RS for HPLC (Farmitalia Carlo Erba) and were filtered through a 0.22 urn Millipore membrane, before the use. For GP-HPLC, a LKB G 4000 TSK SW column (75 x 600 mm) was used. The samples (20 Ug) were eluted with 0.1 M phosphate buffer pH 7.0 containing 0.1 % SDS at a flow rate of 0.7 m?/min and the effluent was monitored at 220 nm (LDC Spectra Monitor III model 1204 A). The separation characteristics of GP-HPLC was determined with marker proteins : thyroglobulin (670 K) ; ferritin (440 K) ; catalase (232 K) ; lactate deshydrngenase (140 K) ; bovine serum albumin (67 K) and ovalbumin (44 K). Analytical methods, sodium dodecyl sulfate pnlyacrylamide gel electrophoresis (SDS-PAGE) with samples previouslv reduced in presence of 2-mercaptoethanol followed by western blotting using rabbit serum anti VSG AnTat l.lA and amino acid compositions were performed as previously described (8).

RESULTS Reverse

phase

HPLC (RP-HPLC)

The procedure better

resolution

1).

The

elution

and

3-9.

This

applied

onto

Concanavalin Several

of

Clarke

of

the

profile profile the

column,

--et al (7) was used VSG preparation. Nine

indicated was

two

faithfully

and also

with

with

slight

fractions

clusters

of

obtained

whatever

VSG preparations

modifications are

fractions

isolated

(Fig.

respectively the

l-2

VSG preparation

directly

form) as well as dialyzed and/or A column ("native" experiments advised us in the choice of our elution 1223

for

obtained lyophilized. conditions.

from

Vol.

144,

No.

3, 1987

0.2

BIOCHEMICAL

D.O.

206

AND

BIOPHYSICAL

RESEARCH

COMMUNICATIONS

nm

2

I---’ 60% ::: : ::: : ::: : ::: 0’ :: : /---------’ Z Q

h

0.1

/

i a

33%

I

Firstly,

using

shoulders.

The better

theless, 0.0013

TFA

of

resolution

: three

formate

system

different

RP-HPLC

reducing

conditions.

molecular

weight

protein

l.lA

gave

led

a band

with

a dimer

a

l-propanol

(more

shoulders of

major

peak

apolar

were VSG is

with

solvent)

detected.

Never-

degraded

in

us to use

the

neutral

resolved

into

nine

the

counterion,

peaks with

a

1). fractions

(l-9)

fraction

showed

60 K as determ;'ned in weight. However,

of

obtained

because

VSG was

of

molecular existence

Each

and

obtained

I.lA

on RP-HPLC C column using gradient sysgem (solvent C

only by

peaks

observation

AnTat (Fig.

we

Z-propanol was

This

: the

gradient The

the

Z-propanol,

replacement

M TFA solution.

ammonium

AnTat

and

no reproducibility

similar

min

1 . Profile of AnTat 1.1 A VSG preparation 33 to 60 % ammonium formate-I-propanol and D).

Figure

enabled

w

I 50

25

analyzed

a single

by

band

SDS-PAGE,

with

an

by comparison with standards the absence of reducing agent

an apparent with

were

molecular

a disulfide

weight bridge

of linkage

under

apparent of known the sVSG

120 K, suggesting (manuscript

in

preparation). The fractions.

western blots In addition

were also fairly to the 60 K band,

1224

similar higher

for the molecular

different weight

RP-HPLC components

Vol. 144, No. 3, 1987

BIOCHEMICAL

T

1

AND BIOPHYSICAL

234

RESEARCH COMMUNICATIONS

56709

-94K

Figure

2

.-Western different

blotting RP-HPLC

tion by

(T).

with rabbit fractions

Scale

Coomassie

indicating

blue

immunoreacted

existence

of

The amine,

of

of

for

could the

Gel

be really

glucosamine

permeation

was

corresponding

of

weight to

amount

Arrows

indicate

and the high polymeri-

l.lA

AnTat

indicating

the

2). composition

VSG,

in

found

revealed

each

RP-HPLC

among

the

GP-HPLC

the

the

presence

fraction

and

fractions

with

a major

peak

higher

molecular components

(d)

column

was performed

670 K and 44 K. In our

weight

(60 K by SDS-PAGE) large

the

has been determined

0)

anti-V%

acid

between by

molecular

1.1 A VSC- of the VSG prepara-

proteins.

appears

antiserum

K

of

only

ethanol-

very

a careful

small

examina-

HPLC (GP-HPLC)

characterized

These

the

-45

AnTat from

weights,

reference

K

content.

The calibration of molecular

of

VSG typically

(Fig.

amino

characteristic

difference tion

with

VSG-polymers

results

molecular

staining

position at which zed forms (w).

slightly

antiserum anti 1 to 9 obtained

-67

120

(M.W.

sizes

(240

using

protein

conditions,

K)

and

by

the three

markers native

minor

K ; 480 K and M.W. above

compared

to

indicated

the

existence

of

; tetramer

(c)

; octamer

(b)

the

apparent

different and

M.W.

polymers higher

VSG peaks

700 K).

of

monomer

: dimer,

polymer

(a)

in (Fig.

3A). We have not

potential

released applied

checked source

of

the

presence

artifacts

of

from

high

polymers

purification

rather

than

procedure,

dimer

because

the

was VSG

from intact trypanosomes, with the pH 5.5 buffer, was directly onto the GP-HPLC column and the high polymers were formely present.

Therefore, system

that

(Fig.

VSG-polymer

the 3B). repartition.

main The

RP-HPLC

fractions

chromatograms Table

I

(1 to

evidenced gave

the

1225

7)

were

significant relative

amount

analyzed

in

variations of

the

GP-HPLC

in different

the

BIOCHEMICAL

Vol. 144, No. 3, 1987

A

DO 220 0.7 nWmin

-

AND BIOPHYSICAL

RESEARCH COMMUNICATIONS

d: 120k

0.05 c: 240k

i

h-‘--x-

,‘i/ 0

“L

----j-j i0

2b

lb

m/n

t

0

IO

;0

3’0

RP-HPLC

6

RP-HPLC min

7

C

-2

0

Figure

3 . GP-HPLC on G-4000 (A) "native" VSG different RP-HPLC

Percentage

of reverse permeation

\

I

min

30

20

10

TSK column profiles of the AnTat ; (B) VSG reduced by 2-mercaptoethanol fractions (I to 7).

TABLE I HPLC fractions chromatography

(l-7) (a,

1

2

3

4

a

16 + 3

21 + 2

17 -+ 2

13 -+ 1

b

11+1

10 _+ 2

10 + 1

C

16 -+ 1

15 + 1

d

58 +- 4

53 + 4

b,

1.1

distributed c, d,)

in

gel

6

7

31 + 5

44 -+ 5

48 + 5

10 + 1

15

16 -+ 1

17 -+ 1

15+1-

16 _+ 2

23+1-

24 +_ 3

21 + 3

57 -+ 1

60 + 1

31 +- 6

19 + 8

19 + 8

1226

5

A VSG : ; (C) the

BIOCHEMICAL

Vol. 144, No. 3, 1987

oligomers

(fractions

percentage

of

high

a-d)

VSG dimer

of

polymerized

GP-HPLC

SDS-PAGE

with

60

polymerized

was

a reduced

eluted

VSG-forms

fractions

1 to

in

allow

a quantification

fractions

that

the

4 whereas

5 to

the

7.

The

small

but

the

high

by GP-HPLC.

VSG preparation

results

from

(Fig.

the

We observed

found

detected

blotting

RESEARCH COMMUNICATIONS

fraction,

in

not

still

of

western

K M.W.

9 did

were

profile

and

RP-HPLC

essentially

8 and

sVSG-forms

The the

were

fractions

each

was predominant

polymerized-forms

amount

for

AND BIOPHYSICAL

the

was

in

: essentially, column

agreement

a monomer

coming

with

VSG-form

with

slight

amount

of

that

VSG from

African

a charge

microhe-

3C).

DISCUSSION There

is

good

trypanosomes

shows

terogeneity

has

equiperdum

and

both

antigens

(14)

have

is of

that protein

oligosaccharide can also Our

aim

obtained tion the

the has

using

of

with

high

been

Nine

the

confirms

the

polymerized

been

of

: MITat

which

arise,

on

carbohydrate

Presper

and

Heath

a heterogeneous

two

asparagine-linked

hydrophobic

nature

a VSG preparation

affinity

quite

GP-HPLC

were

has

degree

affords

useful ; (ii)

and

;

been

of

(AnTat

chromatography. because

the

VSG

: (i)

l.lA)

The combinathe

molecular

weights

a small

amount

[iii)

identified

shown

according

that

the

of polymerization

to

gain

results

integrity are of

of

measured sample

is

in

which

in

sensitivity are

to

hydrophobic our GP-HPLC

compared

discerned

to

by

different

hydrophobic

behaviour

of

the

conditions.

the

The western

SDS-PAGE

GP-HPLC

:

VSG is

analyzes

existence

of

and high-

VSG forms.

Heterogeneity understanding

and the

precise

identity

We can suggest involved

by

or

the

a

two methods.

and it to

nature

by RP-HPLC

VSG fractions

blotting

the

A-Sepharose has

bands,

may comprise

one

that

glycosylation

non-immunogenic

1.2

documented

shown

(8,ll).

examine

unaltered

in the

properties

the

to

resolution

required

related

polymerization

RP- and GP-HPLC

VSG remains

as

of

Trypanosoma

also

Furthermore,

containing

As well

Concanavalin

the

VSG IaTat

molecules

chains.

cause

in

for

was

SDS-PAGE

(10).

mature

It

degree

multiple

heterogeneity

the

focusing

(13).

various

from

VSG produced from

Firstly,

isoelectric

congolense

tunicamycin-sensitive

reported

population

(1,10,13).

using

may result

1.21

laboratories

diversity

detected

studied,

addition

many

Trypanosoma

polymorphism

VSG and ILTat

whose

from

biochemical been

(12)

molecular 1.7

evidence

in

the

polymerization

organization of

of

the

various

an heterogeneity

of

formation

of

propertie the

of

trypanosome

VSG components the

1227

The

be

surface

coat.

remains

to

oligosaccharides

polymers.

VSG may

hydrophobic

chains

useful At

in

present,

be established. which

interactions

could

be

could

BIOCHEMICAL

Vol. 144, No. 3, 1987 presumably currently

influence under

the

AND BIOPHYSICAL

polymerization

(8).

These

RESEARCH COMMUNICATIONS

different

aspects

are

investigation.

ACKNOWLEDGMENTS We are groateful l.lA VSG antiserum. critical reading of

to Doctor T. Vervoort We thank Dr L. Mendonca the manuscript.

for the Previato

preparation for helpful

of anti Antat comments and

REFERENCES

:: 3. 4. 5. 6. 7. 8. 9.

Gross, G.A.M. (1984) Phil. Trans. R. Sot. Lond. 307, 3-12. Turner, M.J. (1985) British Medical Bulletin 41, 137-143. Cardoso de Almeida, M.L. and Turner, M.J. (1983) Nature 302, 349-352. Ferguson, M.A.J., Haldar, K. and Cross, G.A.M. (1985) J. Biol. Chem. 260, 4963-4968. Baltz, T., Baltz, D. and Pautrizel, R. (1976) Ann. Immunol. (Institut Pasteur) 127C, 761-774. Pearson, T.W. and Anderson, N.L. (1980) Anal. Biochem. 142, 360-368. Clarke, M.W., Olafson, R.W. and Pearson, T.W. (1985) Mol. Biochem. Parasitol. 17, 19-34. GomPs, V., Huet-Duvillier, I;., Aubert, J.P., Dirat, J., Tetaert, D., Moncany, M.L.J., Richet, C., Vervoort, T., Pays, E. and Degand, P. (1986) Arch. Biochem. Biophys. 249, 427-436. Grandier-Vazeille, X. and Tetaert, D. (1984) J. of Chromatngr. 296, 301-

308. 10.

McConnel,

J.,

Turner,

M.J.

and Rovis,

L.

(1983) Mol.

Biochem.

Parasitol.

8, 119-135. 11. 12. 13. 14.

Auffret, C.A. and Turner, M.J. (1981) Biochem. J. 193, 643-650. Duvillier, G., Aubert, J.P., Baltz, T., Richet, C. and Degand, P. (1983) Biochem. Biophys. Res. Comm. 110, 491-498. Onedera, M., Rosen, N.L., Lifter, J., Hotez, P.J., Boguki, M.S., Davis, Patton, C.L., Konigsberg, W.H. and Richards, F.F. 11981) Exp. G Pa;asitol. 52, 427-439. Presper, K.A. and Heath, E.C. (1986) Arch. Biochem. Riophys. 246, 460-

468.

1228