Ganglioside biosynthesis in mouse cells: Glycosyltransferase activities in normal and virally-transformed lines

Vol. 48, No. 1, 1972

BIOCHEMICAL

AND BIOPHYSICAL RESEARCH COMMUNICATIONS

GANGLIOSIDEBIOSYNTHESISIN MOUSECELLS: GLYCOSYLTRANSFERASE ACTIVITIES IN NORMALANDVIRALLY-TRANSFORMED LINES Peter H. Fishman*, Vivian W. McFarlandt; Peter T. Morat, and Roscoe 0. Brady* *National Institute of Neurological Disease and Stroke and tNationa1 Cancer Institute, National Institutes of Health, Bethesda, Maryland 20014 Received

May 22,

1972

of four glycosyltransferases involved in ganglioSummary- The activities side biosynthesis were measured in two different established mousecell lines and in the SV40 and polyoma transformed variants of these lines. The only consistent change observed was the very reduced activity of UDP-GalNAc: hematoside N-acetyl-galactosaminyltransferase in all of the transformed cell lines (15-20% of normal cells). There was no significant differences in any of the glycosyltransferases with increased cell density and cell-to-cell contact. Cells in culture growth properties changes (3-5). lipid

transformed by oncogenic DNAviruses have altered

(1) loss of contact inhibition Several laboratories

and glycoprotein

(2) and cell surface

have reported differences

compositions in such transformants

Studies with established mousecells

indicate

in glyco-

(6-9).

a less complex ganglioside

composition in SV40 and polyoma transformed lines.

Whereas the major

GM1' and GDla' generally the virally-transformed cells have mainly GM3 (6,10-12). The absence of these higher gangliosides are GM3’ ,

GM2’

gangliosides appears to be due to a decrease in the enzyme UDP-N-Acetylgalactosamine:

hematoside N-acetylgalactosaminyltransferase

Since the biosynthesis

(13,14).

of gangliosides appears to proceed by the

sequential addition of monosaccharide units from sugar nucleotide

1 The abbreviations used are: AcNeu, N-acetyl-neuraminic acid; cer, ceramide (N-acylsphingosine). For glycosphingolipids containing sialic acid the symbols proposed by Svennerholm (17) are used. G NAc, N-acetylneuraminylgalactosylglucosylceramide (hematoside); GM2, z3 acetylgalactosaminyl-[N-acetylneuraminyl]-galactosylglucosylcer~ide; galactosyl-N-acetylgalactosaminyl-[N-acetylneuraminyl]-galactosyliy&;osylceramide; G N-acetylneuraminyl-galactosyl-N-acetylgalactosaminyl-[N-acetylneur~~~yl]-galactosylglucosylceramide.

Copyright All rights

0 19 72 by Academic Press, Inc. in any form reserved.

of reproduction

48

Vol. 48, No. 1, 1972

BIOCHEMICAL

FIGURE

1.

AND BIOPHYSICAL RESEARCH COMMUNICATIONS

BIOSYNTIIETIC

PATHWAY

Lactosylceramide

FOR GANGLIOSIDES

Cer-glc-

gal

CMP-AcNeu

1

Sialyltransferase

I

CMP cer-glc-gal

‘M3

A!Neu

UDP-CalNAc 2

N-acetylgalactosaminyltransferase

UDP 4 cer-glc-gal-galNAc I AcNeu

GM2 UDP-Cal

3

Galactosyltransferase

UDP

G

cer-glc-gifl-galNAc-gal Ml

CMP-AcNeu 4

Sialyltransferase

II

CMP cer-glc-gal-galh’Ac-gal G

Dla

AC t.!eu

donors to the elongating glycolipid biosynthetic

acceptor (Fig. l),

enzymes are presumed to exist

it was important to determine if viral decrease in one specific thetic

pathway is altered.

AcNeu ’

(lS),

and the

as a multi-enzyme complex (15,16),

transformation

glycosyltransferase

results in a

or whether the entire biosyn-

The present communication compares the levels of

these enzymes in normal and virally-transformed

cells.

EXPERIMENTAL PROCEDURE Cells and cell culture:

The various established mousecell lines and

conditions used in these experiments have been described previously

(6,14).

The normal (N-AL/N) and transformed (SVS-AL/N and PY-AL/N) are non-clonal cell

lines while the three Swiss 3T3 cell

49

lines (normal Swiss 3T3 and

Vol. 48, No. 1, 1972

BIDCHEMICAL

transformed SVlOl and PYll) just prior

AND BIOPHYSICAL RESEARCH COMMUNICATIONS

are clonal derivatives.

Cells were harvested

to confluency unless otherwise noted.

Preparation of cell homogenates: The cells were harvested as previously described (14).

The washed packed cells were suspended in nine volumes of

0.25 M sucrose containing thawing four

0.1% 2-mercaptoethanol and disrupted by freeze-

in a dry ice-ethanol

times

bath.

Enzymatic activities

obtained with this procedure were comparable to those obtained with the previous technique of rupturing Glycolipid

acceptors and sugar nucleotide

dog erythrocytes

and further

from the brain of a patient gift

cells by sudden decompression (13)2.

purified

donors:

GM3NAcwas prepared from

and analyzed (13).

GM2was prepared

with Tay-Sachs disease (18) and was a generous

from Dr. Andrew Gal, NINDS. GM1was prepared from bovine brain ganglio-

sides (19).

Lactosylceramide was obtained from Miles Laboratories.

nucleotides were purchased from the following g-galactosamine (14 UDP-galactose-l-

mCi

sources:

per mmole) from International

Sugar

UDP-N-Acetyl-l- 14cChemical and Nuclear;

14C (254 mCi per mmole) and CMP-N-acetylneuraminic acid14 C] (223 mCi per mmole) from New England Nuclear;

[sialic-4,5,6,7,8,9-

UDP-galactose from Sigma (used to dilute

UDP-galactose- 14C to 4 mCi per

mmole). Enzyme Assays:

The reaction

various glycosyltransferase incubation,

mixtures and incubation

conditions

assays are described in Table 1.

for the

After

the reactions were terminated by adding 20 volumes of chloro-

form-methanol (2:l v/v) (60:30:4.5 v/v/v)

(CM) and 40 volumes of chloroform-methanol-water

(CMW). These mixtures were passed through 1 x 2.5

colums of Sephadex G-25 superfine

(lg) previously

equilibrated

cm

with CMW.

The columns were then washed with 5 ml of CMW(4 ml for N-acetylgalactosaminyl-transferase radioactive

assays).

glycolipid

This procedure effectively

products from the radioactive

separates the

sugar nucleotide

precursors (13). The eluants were dried and the radioactivity 2 Unpublished observations of P. H. Fishman and R. W. Smith.

50

determined

donor

50 )11

for

Glycosyltransferase

50 !.I1

CF-54 50 ug Tween - - 20 ul

2x10' cpm UDP-Gal 5 umoles pH 7.2 100 ug Triton

10 nmoles GM2

80

Assays

50 u1

- - 20 lJ1

50 pg Tween

10' cpm CMP-AcNeu 5 umoles pH 6.5 100 ug Triton

GM1

50 nmoles

Sialyltra sferase II a

a) Assay conditions same as ref. 20; 16; d) Assay conditions same as ref. b) Assay 21.

conditions

similar

to ref.

13; c) Assay

conditions

80

same as ref.

Glycolipid acceptors were added to the reaction tubes in CM and taken to dryness. The remaining components were then added and the reactions initiated by addition of the cell homogenates (4-6 mg protein per ml) followed by vigorous swirling. After incubating at 37OC for 2 hrs. the reactions were terminated by adding 1 ml CM and 2 ml CMW. The radioactive ganglioside products there then isolated as described under "EXPERIMENTAL PROCEDURE."

50 ?ll

- - 10 ?ll

x-100

cutscum

50 nmoles GM3NAc 10' cpm UDP-GalNAc 2.5 umoles pH 7.2 200 ug Triton

Final

volume

Conditions

Glycosyltransferase N-acetylgalactos-b Galactosylaminyltransferase transferaseC

and Incubation

10' cpm CMP-AcNeu 5 umoles pH 6.35 750 vg

50 nmoles cer-glc-gal

100 1J.g 20 ul

(14C)

acceptor

Cardiolipin Cell homogenate

Na Cacodylate buffer Detergents

Sugar

Glycolipid

Component

Mixtures

Sialyltrans ferase Ia

Reaction

TABLE I

CF-54

z F r z

?I

F 6 5

g 2 F

Vol. 48, No. 1, 1972

by liquid activity

BIOCHEMICAL

scintillation

AND BIOPHYSICAL RESEARCH COMMUNICATIONS

Each determination of enzyme

spectrometry.

represents net incorporation

of radioactive

acceptor and was calculated by subtracting incorporation

sugar onto exogenous

endogenous incorporation

from

with added acceptor. RESULTSANDDISCUSSION

Factors influencing conditions

enzyme activities:

of linearity

acceptor concentrations activation

in terms of

Each enzyme assay was run under and protein

time

were saturating

or inhibition

for all

assays.

Sialyltransferase

lines was stimulated 3-7 fold by cardiolipin.

had been reported for this sameenzyme activity (BHK) cells

(20).

No

Glycolipid significant

was observed when homogenates from normal and

transformed lines were assayed together. the cell

content.

Cardiolipin

I in all of

Similar findings

in baby hamster kidney

at 100 ug per incubation mixture produced

maximumactivity. Effect

of Cell Density on Glycosyltransferase

been reported that quantity

of certain

cell contact in somecontact-inhibited

Activities:

glycolipids cell

Since it has

increase upon cell-to-

lines

(22, 23), the influence

of cell density on the ganglioside biosynthesizing

enzymes was investigated.

Swiss 3T3 cells were plated in inocula of 5 x 104, lo', cells per plate.

The cells were harvested after

5 x lo5 and lo6

3 days by gentle trypsini-

zation in order to obtain suspensions for cell counting. procedure had no effect ferases exhibited

on the transferase

a slight

activitiesA.

decrease in specific

increased whereas the other two transferases

The trypsinization Both sialytrans-

activity

fluctuated

as cell density slightly

(Fig. 2).

For comparative purposes, cells were routinely

harvested at preconfluency

while still

Similar experiments with

in a logarithmic

clonal derivatives

phase of growth.

of Balb 3T3 cells

indicated

changes in N-acetylgalactosominyltransferase These cell lines were:

that there were no substantial activity

with cell density

OA31, a normal Balb 3T3; SO-IIA41 and llA8,

transformed Balb 3T3; and 10 A 7, a "flat

52

revertant"

(24).

SV40

SV40 transformed Balb

Vol. 48, No 1,1972

BIOCHEMICAL

AND BIOPHYSICAL RESEARCH COMMUNICATIONS

.

SIALYLTRANSFERASE

A

N-ACETYLGALACTOSAMINYL-

A

GALACTOSYLTRANSFERASE

0

SIALYLTRANSFERASE

I

TRANSFERASE

SPARSE

SUBCONFLUENT

CELL

CONFLUENT

DENSITY

AT

FIGURE

Effect

II

POSTCONFLUENT

HARVEST

2

of cell density on glycosyltransferases.

Swiss 3T3 cells were

plated in inocula of 5x104, 105, 5X105, and lo6 cells per plate. cells were harvested after determined. confluent,

three days and the number of cells were

Cell densities

(cells per cm2) were:

3x104, confluent,

1~10~; post-confluent,

syltransferase

activities

PROCEDURE."The activities

concentrations

(22, 23, 26).

in confluent

mousecells,

in hamster cells

cells.

reported to occur in hamster cells

it recently

(27).

in culture

were not determined in

correlated

Whether these different

53

changes

has been reported that cell

density dependent changes in glycosyltransferases glycolipids

Glyco-

to the cell density-dependant

Although ganglioside concentrations

sparse and confluent

1.5~10~.

of each transferase have been normalized

The results are in contrast

in glycolipid

sparse, 1~10~; sub-

were determined as described under "EXPERIMENTAL

and are given as the percent activity

3T3 (25).

The

with changes in observations

Vol. 48, No. 1, 1972

BIOCHEMICAL

are due to differences class of glycolipid

in cell

AND BIOPHYSICAL RESEARCH COMMUNICATIONS

type, or in clones (cf.

has yet to be determined.

ent changes in glycolipid

concentrations

ref.

26, 28), or in

However, cell density-depend-

does not appear to be a general

phenomenon. Enzyme levels in normal and transformed cells: glycosyltransferases of Swiss 3T3 cells

in normal and virally-transformed are given in Table II.

galactosaminyltransferase.

experi-

viral

transformation

in

lines was the marked decrease in N-acetylThe altered

uncloned and cloned virally-transformed laboratory

of similar

lines are presented in Table III.

change in enzyme levels after

both Swiss 3T3 and AL/N cell

clonal derivatives

The results

ments with the various non-clonal AL/N cell The only consistant

Levels of the ganglioside

ganglioside composition of both cell

lines investigated

is in agreement with this observation

(6, 10, 11).

in this Sialtrans-

ferase I is reduced in a polyoma transformed BHKclonal cell

line (20) and

was reported to be lower in SV40 transformed Swiss 3T3 (29).

This enzyme

activity

was elevated in SVlOl and PY AL/N, unaffected

moderately diminished in SVS AL/N.

Sialytransferase

decreased in SV40 transformed Swiss 3T3 cells normal was found by us in the SVlOl cells.

,in PYll and II was reported to be

(29); lower activity

However, this enzyme activity

was higher in both PY AL/N and SVS AL/N than in the control The results after

suggest that the alteration

transformation

effects results,

N AL/N cell line.

in ganglioside biosynthesis

of established mousecells by oncogenic DNAviruses

are complex when considering the change in activities glycosyltransferases.

than

of the various

SV40 and polyoma appear to induce somewhatdifferent

and the host cell may influence

any virally

mediated changes.

The

however, continue to support a commonbiochemical change in SV40

and polyoma virus-transformed N-acetylgalactosaminyltransferase.

mousecells;

reduced activity

Whether this consistantly

54

of a specific observed

Vol. 48, No. 1, 1972

BIOCHEMICAL

AND BIOPHYSICAL RESEARCH COMMUNICATIONS

55

Vol. 48, No. 1, 1972

BIOCHEMICAL

AND BIOPHYSICAL

56

RESEARCH COMMUNICATIONS

Vol. 48, No. 1, 1972

change is directly

BIOCHEMICAL

AND BIOPHYSICAL RESEARCH COMMUNICATIONS

related to viral

transformation

quence of selection during the transformation

or a secondary conse-

process remains to be

determined and cloning experiments are underway to explore these alternatives.

REFERENCES 1. 2.

3. 4. 5. 6. 7. a. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29.

22: 391 (1968). Black, P. H., Ann. Rev. Microbial., Abercrombie. M.. Eur. J. Cancer. 6: F(1970). Burger, M.M: and Goldberg, A.D.; Froc. Natl: Acad. Sci., -57: 359 (1967). Tevethia, S.S., Katz, M. and Rapp, F., Proc. Sot. Exp. Biol. Med., 119: 896 (1965). Hakomori, S., Teather, C. and Andrews, H., Biochem. Biophysic. Res. commun., 33: 563 (1968). Mora, P.T., Brady, R.O., Bradley, R.M. and McFarland, V.W. Proc. Natl. Acad. Sci., 63: 1290 (1969). Hakomori, S. and Murxami, W.T., Proc. Natl. Acad. Sci., -59: 254 (1969). Wu, H.C., Meezan, E., Black, P.H. and Robbins, P.W., Biochem., 8: 2509 (1969). Buck, C.A., Glick, M.C. and Warren, L., Biochem., 9: 4567 (1970). Brady, R.O. and Mora, P.T., Biochem. Biophysic. Acta., 218: 308 (1970). Dijong, I., Mora, P.T. and Brady, R.O., Biochem., 10: 4039 (1971). Sheinen, R., Onodera, K., Yogeeswaran, G., and Murray, R.K., The Biology of Oncogenic Viruses, IInd Le Petit Symposium, Milan Le Petit, p. 274 (1970). Cumar, F.A., Brady, R.O., Kolodny, E.H., McFarland, V.W. and Mora, P.T., Proc. Natl. Acad. Sci., 67: 757 (1970). Mora, P.T., Cumar, F.A. and Brady, R.O., Virology, 46: 60 (1971). Roseman,S., Chem. Phys. Lipids, 2: 270 (1970). Cumar, F.A., Fishman, P.H. and Brady, R.O., J. Biol. Chem., 246: 5075 (1971). Svennerholm, L., J. Neurochem., 10: 613 (1963). Gatt, S. and Berman, E.R., J. Neurochem., 10: 43 (1963). Kolodny, E.H., Brady, R.O., Quirk, J.M. anrKanfer, J.N., J. Lipid Research, 11: 144 (1970). Den, H., Schztz, A.M., Basu, M. and Roseman,S., J. Biol. Chem. 246: 2721 (1971). Kasan, B., Basu, S., and Roseman,S., J. Biol. Chem., 243: 5804 (1968). Hakomori, S., Eroc. Natl. Acad. Sci., 67: 1741 (1970). Robbins, P.W., and McPherson, I.A., PrG. Roy. Sot., -177: 49 (1971); also Nature, -229: 569 (1971). Cumar, F., and Mora, P.T., Unpublished observations. Smith, H.S., Scher, C.D. and Todaro, G.J., Virology, 44: 359 (1971). Sakiyama, H., Gross, S.K. and Robbins, P.W., Proc. Nax. Acad. Sci., 69: 872 (1972). Kijmoto, S. and Hakomori, S., Biochem. Biophysic. Res. Commun.,-44:557 (1971). Yogeeswaran, G., Schimmer, B.P., Sheinin, R., and Murray, R.K., Fed. Proc., &, 438 Abs. Grimes, W.J., Sasaki, T., and Robbins, P.W., Fed. Proc., E:, 1185 Abs. (1971).

57