Magnesium-dependent sphingomyelinase

Vol. 68, No. 1, 1976

BIOCHEMICAL

AND BIOPHYSICAL RESEARCH COMMUNICATIONS

MAGNESIUM-DEPENDENT SHIMON of Neurochemistry,

Laboratory Hebrew Received

SPHINGOMYELINASE

University-Hadassah

November

10,

GATT Department Medical

of Biochemistry,

School,

Jerusalem,

ISRAEL

1975

SUMMARY: An enzyme which requires divalent metals and hydrolyses sphingomyelin to ceramide and phosphorylcholine is present in rat and human brain and practically absent from other organs. The greatest activity is associated with the microsomal fraction. It had an optimal pH at about 7.4, required magnesium or manganese ions and was completely inhibited by EDTA. Triton X-100 was required for optimal activity and this detergent could also be used to partly solubilize the enzyme Lecithin was hydrolyzed at only 2% of the from rat brain microsomes. corresponding rate of hydrolysis of sphingomyelin. Sphingomyelinase 3.1.4.12)

is a hydrolase

sues it is of lysosomal animal liver

(Sphingomyelin

tissues, (3,4)

be several ions for

isoenzymes activity

origin

C activity

known

(l, 7,8)

et al.

Copyright All rig&s

enzyme

have reported

pH at about paper

and a phospholipase

of phosphatidylethanolamme

of the latter

(11).

specific

tissues moiety

for

have been

splits

has recently

but this

communication, 235

In

that there

may metal

lecithin

enzymes

with

or phos-

An enzyme

which

has long been

off the phosphorylethanolamine been described; of EDTA C, with

was retracted Quinn,

pH5. (2),

of the phosphoinositides C which

tis-

brain

reported.

of a phospholipase

0 1976 by Academic Press, Inc. of reproduction in any form reserved.

from

Other

the presence

lir a recent

purified

sphingomyelin;

by the addition

(lo),

at about

did not require

was increased

5 in rat brain

In animal

activity

preparations

not hydrolyzed.

in animal

of the phosphoinositol

portion

strictly

EC

(1).

it has been suggested

The above

were

splits

and partially

Recently

(6)0

and were

C family

and has an optimal isolated

(5).

phosphatidylethanolamine pholipase

phosphohydrolase,

of the phospholipase

it has been

and spleen

choline

while

the activity (9).

Roitman

an optimal

in a subsequent describing

BIOCHEMICAL

Vol. 68, No. 1, 1976

a phosphodiesterase not find This depends little

disease dependent sis

evidence paper

for

which

phospholipase

describes

on divalent

hydrolysis

presence

in brain

ions,

sphingomyelinase

of glycerophosphatides

C activity

Schneider

and Loyter

by the latter

in rat and human

pH between

of patients

et al.

in the stroma

(12) did

on lecithin.

and Kennedy

in spleen

RESEARCH COMMUNICATIONS

on glycerophosphorylcholine

has an optimal

enzyme

(5) and Hirshfeld

acts

a sphingomyelinase,

of lecithin.

of a similar

AND BIOPHYSICAL

two

enzymes

which

7-8 and catalyzes have suggested with

have described of avian

brain

only

the

Niemann-Pick’s a magnesium-

erythrocytes

(13).

Hydroly-

has not been tested.

MATERIALS AND METHODS: Substrates Two samples of tritium-labelled The first was labelled in the ceramide moiety by sphingomyelin were used. catalytic hydrogenation with tritium gas in the presence of palladium on charcoal (14). The second was labelled in the choline moiety by condensing 3H3-methyliodide (Amersham) with ceramide phosphoryldimethyletbanolamine (demethylated sphingomyelin, a generous gift of Prof. W. Stoffel) according to the method of Stoffel (15). Choline-labelled lecithin was a gif of, Drs. Greenzaid and Heller, it was prepared according to ref. 16. 2- i 3H\ oleyl lecithin and 1-13H] palmitoyl phosphatidylethanolamine were prepared according to reference 17. Triton X-100, sodium dodecylsulfate and EDTA were purchased from BDH; sodium taurocholate and taurodeoxycholate from CalBiochem; EGTA and Cetavlon from Sigma. Preparation of brain microsomes and subcellular fractions. Brains of 14 day old rats were homogenized and subcellular fractions were prepared according to ref. 18. The microsomal fraction of adult (about 120g) rat brain or of Brain was homogenized in 9 ~01s. of human brain were prepared as follows: 0.25M sucrose. Debris was removed for 10 min. at 7OOxg, the mitochondrial fraction was sedimented for 10 min. at 25,OOOxg and microsomes for 1 hr, at 100, OOOxg. The microsomes were suspended g 0.25M sucrose (lml per grm equivalent of brain tissue) and stored at -20 D Incubation mixtures in volumes of 0.2ml Assay of the enzymatic reaction: contained 20 pmoles of Tris, pH7.4, 0.05 pmoles of sphingomyelin, 0.4mg After 1 hr. of Triton X-100, 0.2 pmole of magnesium chloride and enzyme. at 37’ the reaction was terminated and the reaction products were isolated as follows: 1. Using sphingomyelin labelled in the ceramide moiety the reaction was terminated with Dole% reagent (2) and the ceramide released was isolated and counted using the procedure employed in the hydrolysis of lecithin 07). 2. Using choline-labelled sphingomyelin or lecithin: The reaction was terminated by the addition of 0.8ml of a mixture of chloroAn aliquot of the upper phase was form and methanol, 2:l (v/v). mixed with 3.5ml of Instagel (Packard) and counted in a liquid scintillation spectrometer. Identification of the products of the reaction: 1. Identification of ceramide: sphingomyelin, labelled with tritium in the ceramide moiety was used, the reaction was terminated with Dole’s reagent (Z), and the heptane phase was 236

Vol. 68, No. 1, 1976

BIOCHEMICAL

AND BIOPHYSICAL RESEARCH COMMUNICATIONS

evaporated with nitrogen. The residue was dissolved in chloroform-methanol, 2 :l and spotted on thin layer plates of silica gel. The plates were developed in the following three, separate solvent systems : a. Chloroform-methanolacetic acid, 94:4:2 (19) b. Chloroform-methanol-ammonia, 95:5:0.8 (20) and cm Chloroform-methanol-acetone-acetic acid-water, 80:2.5:15:2.5:2 (20) ~ N-stearoyl-DL-dihydrosphmgomyelin was used as reference compound in each case. The radioactivity on the plate was located with a Berthold radioscanner and further verified by scraping the silica gel, mixing well with lml of Triton X-lOO-absolute ethanol, 1:l and adding lOm1 of scintillation fluid. The ceramide was located in duplicate lanes by spray$ the plate with 10% sulfuric acid in methanol, followed by charring at 180 . 2. Identification of phosphorylcholme: The reaction mixture, containing 3H-choline-labelled sphingomyelin was terminated with 4 volumes of chloroform-methanol, 2:l. The upper phase was evaporated, dissolved in methanol and applied to a thin layer plate of silica gel which was developed in chloroform-methanol-3M trichloroacetic acid-water, 40:60:20:12.5 (21). Phosphorylcholine, choline and glycerophosphorylcholme were used as reference compounds. The radioactivity on the plate was located using a Berthold radioscanner and the spots were visualized by spraying according to Doizaki et al. (21) with a mixture of 70% HClO -5N HCl-S$&ammonium molybdate-3M TCA-H20, 2:4:8:5:21. After drying at 10 4 % the plates were further sprayed with a mixture of Dragendor’s reagent - Bi(NO3)3.5H20 @.7g in lOOm1 of 20% acetic acid) - 40% KI (w/v)H20, 4:1:20. Identification of diglyceride: Incubation mixtures containing 2- 3H-oleyl phosphatidylcholine were terminated with Dole’s reagent (2), the heptane was evaporated and the residue was spotted on thin layer plates of silica gel. The plates were developed in petroleum ether (40-600)-ether-acetic acid (22), 60:40:1 (22) or in chloroform-methanol-acetic acid, 98:2:1 (22). RESULTS: liver,

kidney

Distribution and spleen

in rat organs were

and subcellular

homogenized

ENZYME

in 0.25M

fractions. sucrose

Rat brain, and debris

was

imgl

Fig. 1. Effect of the concentrations of the homogenate minus debris and microsomes on the rate of hydrolysis of sphingomyelin. Standard assay conditions were used.

237

Vol. 68, No. 1, 1976

removed those

BIOCHEMICAL

for 10 min.

of the other

linearity were

at 700xg.

tissues

(see Fig.

1).

The quantitative

of control

tubes at pH7.4

liver

kidney

100,

showed

the relative

activities

deviation

crosomal

protein

obtained)

at higher

EDTA

were

Properties reaction

only

2-4s

EDTA

Triton

negligible

which

microsomes

dependence followed required hydrolyzed

0.2%

observed

without

activity.

The

on magnesium by dialysis ions.

Snmoles

238

levels

of rat of mi-

inhibition

was

but with

in the absence

of EDTA.

was always

of this H2M

also be used to preparation

However,

Tris

In a typical

of sphingomyelin

to the

or Miranol

microsomal ions.

added

in the absence

could

against

sediment

magnesium

dodecylsulfate X-100

pellet,

100:42:26:7.

(or some

was observed

Triton

of rat brain

at low

(2-3mg[ml)

sodium

divalent

were

obtained

activity

Cetavlon,

of the enzymatic

or EGTA,

X-100

4400,

in the microsomal,

the 100,OOOxg

The controls

acti-

of substrate Brain

equivalent)

was reached

of the activity

relative

in the microsomal

fractions

was

activity

the residual

were:

resided

done using linearity

the Triton.

only a small

a preparation treated

from

a small

fractionation

one gram

1) and a plateau

about

not replace

showed

were

Taurocholate,

part

of the activity

concentrations.

mixture,

solubilize

A subcellular

for

in the various

as nmoles

of tissue

from

of the curves

The approximate

equivalent

exceeded

deviated

portions

represents

expressed

and lysosomal

of the enzyme:

detergent. could

(Fig.

only

debris,

(calculated

studies

E curves

by subtracting

this

enzyme.

25.

that most

further Again,

brain.

spleen

homogenates

of the enzymes

obtained

was added;

minus

synaptosomal

Therefore,

were

by a gram

200,

homogenate

mitochondrial,

The v versus

of the lysosomal

in an hour

of brain

activities

values

of the homogenates

RESEARCH COMMUNICATIONS

only the ascending

to whichFDTA

hydrolyzed

with

manyfold.

the relative

tissues.

vities

The activity

Therefore

used to calculate

activity,

AND BIOPHYSICAL

buffer

treatment resulted

experiment

in EDTA-

in the absence

of

Vol. 68, No. 1, 1976

BIOCHEMICAL

lvlg, 32 in the presence

of 1mM

pH of the reaction

was between

(identified

separate

Free

in three

choline

the rate

Mg or Mn and 12 at 1mM 7 and 8.

solvents,

was not detected.

of hydrolysis

was verified

chromatography

showed

the presence

be stored

for

preparations

under

at least which

5 months

Similar

spleen

had only little

to the findings activity

6700 and 1800 nmoles of a newborn somal mg

-1 hr-l

for

66 and 68 for DISCUSSION: residues

the senile Animal

brain

metal

origin,

abound

of a divalent

activities

after

liver,

but

2 weeks

kidney

at

and

of human

brain

minus

were

debris

their

respective

were

156 and 345 nmole

of the young

in enzymes

enzyme i.e.,

which

(phospholipase

belongs

and lecithin

ions for

of activity;

of homogenates

and microsomes

tissues

the very brief

of lysosomal require

specific

C has not been firmly

erythrocyte

human

could

human

micro-

brain

and

brain.

sphingomyelin

for

loss

activity

and 2500 and 250 for

the homogenate

sphingomyelin

chicken

brain

lost

Microsomes

The activities

per gm equivalent

of the glycerophosphatides

hydrolyzes

Except

-1

only little

rat organs,

thin layer

Phosphatidylethanolamine

with

EDTA

2% of

a residual

lecithin;

.

conditions.

at pH 7.4.

The respective

of phospholipase

both

hr

and senile

fractions.

with

at only about

3H-oleyl

assay

with

ceramide

is indeed

of diglyceride

at -20’

The optimal

and phosphorylcholme.

this

by using

the standard

had been treated

-2oO.

That

Ca.

were

was hydrolyzed

of sphingomyelin.

activity

The products

see Methods]

Lecithin

phospholipase-C

was not hydrolyzed

AND BIOPHYSICAL RESEARCH COMMUNICATIONS

by Schneider

paper

sphingomyelinase 239

which It utilizes

C group.

and Kennedy

sphingomyelinases

This

the presence

enzyme

(l) but not in animal

they have an optimal (1).

off fatty acid

However, The

established.

in bacteria

(13) all

metal-dependent

A).

to the phospholipase

mentioning

activity

split

of animal

pH at about reports with

tissues. (5) and the tissues

are

5 and do not

the presence, a slightly

in alkaline

Vol. 68, No. 1, 1976

optimal pH.

BIOCHEMICAL

AND BIOPHYSICAL RESEARCH COMMUNICATIONS

The rate of hydrolysis by homogenates (less debris) of rat brain

at pH 7.4, with magnesium was about 5 pmoles x hr This value is greater than the corresponding activity

-1

per gram equivalent.

of the lysosomal enzyme,

as assayed at pH 5 (10). The microsomal sphingomyelmase probably has no general phospholipase C activity.

Lecithin was hydrolyzed at about one

fiftieth of the rate of hydrolysis of sphingomyelin. was not hydrolyzed even at this low rate. from the recently-described

Phosphatidylethanolamine

The enzyme here described differs

phospholipase C type enzyme which hydrolyses

phosphatidylethanolamine and is also localized in the microsomes, since the latter is stimulated by EDTA (9).

The relation of the metal-dependent

sphingomyelinase to the phosphatidylinositide splitting enzymes cannot be assessed to date.

While the sphingomyelinase is membranal, the latter en-

zymes are present partly in microsomes but even moreso in the soluble fraction and require calcium for their activity.

Experiments are planned to

solubilize the microsomal sphingomyelinase, purify it and then determine its substrate specificity. ACKNOWLEDGEMENTS: This work was supported in part by NM grant NS02967. The expert technical assistance of Mr. Avigdor Herzl is acknowledged. Thanks are due to Dr. A. Loyter for making his manuscript on the sphingomyelinase of chicken erythrocytes available to us prior to its publication. To Dr. Stoffel for a sample of the dimethyl precursor of sphingomyelin and for details of the method of its methylation. To Drs. P. Greenzaid and M. Heller for a sample of 3H-choline-labelled lecithin and Dr. D. Shapiro for synthetic ceramide . REFERENCES: 1. Gatt, S. (1973) in Metabolic Inhibitors, Vol. 4, Hochster, R. M., Kate@,M. and Quastel, J.H., Academic Press, pp. 350-387. 2. Barenholz , Y. , Roitman, A. and Gatt, S. Q966), J. Biol. Chem. 24l, 3731-3737. 3. HeIeo; y.Nand Shapiro, B. 0966)) Bi ochem. J. 98, 763-769. 4. . Young, 0. M., Shapiro, D. and Brady, R.O. (1966), J. Bio;. Ch&. E, 1081-1084. 5. Schneider, P.B. and Kennedy, E.P. (l967), J. Lipid Res., S, 202209. 6. Calahan, J. W., Khalil, M. and Gerrie, J. (l974), Biochem. Biophys. Res. Comm. 58, 384-390. 240

Vol. 68, No. 1, 1976

7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20.

21. 22.

BIOCHEMICAL

AND BIOPHYSICAL RESEARCH COMMUNICATIONS

Keough, K.M.W. and Thompson, W. Q97a,), J. Neurochem. 17, l-11. Thompson, W. and Keough, K. M. W. (l972), Biochim. Biophys. Aeta, 270, 324-3:s. Williams, D. J., Spanner, S. and Ansell, G.B. (1973), Biochem. Sot. Trans. 1, 466-7. Roitman, A. and Gatt, S. Q963), Israel J. Chem. 1, 190. Gatt, S. Q970), Chem. Physics Lipids, 5, 235-249. Abra, R.M. and Quinn, P.J. (1975), Biochim. Biophys. Acta, -380 ’ 436-441. Hirshfeld, D. and Loyter, A. (1975), Arch. Biochem. Biophys. 167, 186-192. Gatt, S., Herzl, A. and Barenholz, Y. (l973), FEBS Letters, 30, 281-285. Stoffel, W. (l975), Methods in Enzymol. Vol. 35, Lowenstein, J.M,, Press, pp. 533- 541. ed., Academic Ascher, Y., Heller, M. and Becker, Y. (1969), J. Gen. Virol. 4, 65-76. Gatt, S. @968), Biochim. Biophys. Acta, 2, 304-316. Gottesdiner, T. and Gatt, S. (l975), J. Neurochem., in press. Gatt, S. 0966)) J. Biol. Chem. 24, 3724-3730. Rouser, C., Kritchevsky, G., Yamamoto, A., Simon, G., Gall, C., and Bauman, A.J. (l969), in Methods Enzymol. vol. 14, Lowenstein, J.M., ed. Academic Press, pp. 272-317. Doizaki, N.M. and Zieve, L. (1963), Proc. Sot. Exp. Biol. Med. 113, 91-94. J. Biol. Chem. 245, 3047-3058. %&le, R.C. and Snyder, F. (l970),

241