|
|
|
|||||||
|
|||||||||
(Received for publication, October 25,
1994; and in revised form, December 28, 1994) From the
We have previously reported the presence of an endogenous
inhibitory activity in bovine brain for the ADP-ribosylation of
GTP-binding proteins catalyzed by pertussis toxin (PT) (Hara-Yokoyama,
M., and Furuyama, S.(1989) Biochem. Biophys. Res. Commun. 160,
67-71). In the present study, we identified the inhibitor as a
ganglioside. The screening of various gangliosides revealed that
G ADP-ribosylation is a post-translational modification of
proteins whereby the ADP-ribose moiety of NAD Several mono(ADP-ribosyl)transferases produced by bacteria
specifically modify target proteins in eukaryotic cells. Pertussis
toxin (PT), ( To elucidate how ADP-ribosylation is
regulated, we aimed to characterize the endogenous modulator molecules.
PT- and the C3 exoenzyme-catalyzed ADP-ribosylation was used as an
assay system to detect modulatory activity for ADP-ribosylation.
Previously, we reported the presence of an endogenous inhibitory
activity of PT-catalyzed ADP-ribosylation in rat liver (26) and
in bovine brain (27) . In the present study, we found that the
inhibitory activity was present in the crude ganglioside fraction of
bovine brain. Consequently, various species gangliosides, including
newly identified species(28, 29, 30) , were
examined for their effects on the ADP-ribosylation. We found that
G
The cholate
extract was heated in boiling water for 5 min, dialyzed overnight
against water, and concentrated using an Amicon YM5 membrane (extract
A). Extract A was mixed with 40 ml of chloroform/methanol (2:1, v/v)
and centrifuged at 2000
The upper layer fraction was fractionated by gel permeation column
chromatography using 45% acetonitrile, 0.1% trifluoroacetic acid. The
activity emerged from the column just after 2,4-dinitrophenol-alanine
(molecular mass of 260 Da), suggesting that the inhibitor is not a high
molecular mass species (data not shown). No proteins or peptides were
detected in the peak fraction by silver staining. On the other hand,
the inhibitory activity in the cholate extract was concentrated after
ultrafiltration using an Amicon YM5 unit (molecular mass cutoff about
5,000 Da). Similar results were obtained when the bovine brain
membranes were extracted by the nonionic detergent
nonanoyl-N-methylglucamide (MEGA9) or the ampholytic detergent
CHAPS (data not shown). The inhibitor behaves as either a low or high
molecular mass species depending on the conditions. These results
suggest that the inhibitor is a ganglioside. The change in the apparent
molecular mass may be explained by micelle formation, which is a
characteristic property of gangliosides.
Figure 1:
Effect of the ganglioside fraction from
bovine brain membranes on the PT and C3 exoenzyme-catalyzed
ADP-ribosylation. A, The PT and C3 exoenzyme-catalyzed
ADP-ribosylation was measured as described under ``Experimental
Procedures'' in the presence of the indicated amount of the
ganglioside fraction. Various volumes of the ganglioside fractions were
separately placed in glass tubes and evaporated, and the reaction
mixtures were added. The concentrations of G
Figure 2:
Effect of individual gangliosides on the
PT-catalyzed ADP-ribosylation. The method for the assay is described
under ``Experimental Procedures.'' A, the reaction
mixture contained either G
Figure 3:
Effect of NAD
Figure 4:
Thin layer chromatograms of the
water/methanol layer of the cholate extract and the crude ganglioside
fraction from bovine brain membranes. The upper layer of the cholate
extract and the crude ganglioside fraction were prepared as described
under ``Experimental Procedures.'' After dialysis, the
samples were spotted on precoated high performance thin layer
chromatography plates (Silica Gel 60; E. Merck, Darmstadt, Federal
Republic of Germany) and were developed with chloroform/methanol/water
(5:5:1, v/v/v). Lane 1, standard ganglioside mixture
(G
Figure 5:
Effect of gangliosides on the
ADP-ribosylation catalyzed by the S1 subunit of PT. The
ADP-ribosylation catalyzed by the S1 subunit of PT was assayed as
described under ``Experimental Procedures'' in the presence
of 0.1 mg/ml G
Figure 6:
Effect of gangliosides on the C3
exoenzyme-catalyzed ADP-ribosylation. The method is described under
``Experimental Procedures.'' The reaction mixture contained
G
To investigate whether the gangliosides
directly interact on the C3 exoenzyme, we examined the effect of
gangliosides on the NAD
Figure 7:
Effect of gangliosides on the
NAD
G
The inhibitory
effect of gangliosides on the activities of PT and the C3 exoenzyme is
probably not due to their interaction with the carbohydrate recognition
domain of the proteins, based on the following reasons. (i) While the
specific binding of G PT-catalyzed ADP-ribosylation was
inhibited by G
Figure 8:
Space-filling model of
NAD
Gangliosides reside in the cells
as a component of plasma membranes, organella(65) , or in
association with cytoskeleton(64, 66, 67) .
Gangliosides may be in the sites where the ADP-ribosylation should be
regulated. Indeed, several substrate proteins of the endogenous
ADP-ribosylation are the membrane proteins (4, 6, 7, 23, 25) and a
cytoskeletal protein(68) . Involvement of brefeldin A-sensitive
ADP-ribosylation in membrane transport was also suggested(8) .
The inhibitory action of gangliosides implies the importance of
membrane or cytoskeleton-targeted ADP-ribosylation in the cells.
Volume 270,
Number 14,
Issue of April 7, 1995 pp. 8115-8121
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
FOOTNOTES
ACKNOWLEDGEMENTS
REFERENCES
most effectively inhibited the
ADP-ribosyltransferase activities of both the holoenzyme and the
catalytic subunit of PT. G
is a ganglioside newly
identified as one of the antigens recognized by the cholinergic
neuron-specific antibody, anti-Chol-1
(Hirabayashi, Y., Nakao, T.,
Irie, F., Whittaker, V.P., Kon, K., and Ando, S.(1992) J. Biol.
Chem. 267, 12973-12978). G
also inhibited
the PT-catalyzed NAD
glycohydrolysis. Unlike PT
activity, the ADP-ribosylation and the NAD glycohydrolysis catalyzed by the C3 exoenzyme from Clostridium botulinum type C were inhibited by G
and G
. The ADP-ribosylation catalyzed by either PT
or the C3 exoenzyme was not inhibited by ceramide, galactocerebroside,
or sialic acid. In addition to the inhibitory action of gangliosides on
ADP-ribosylation, the importance of gangliosides as regulators of
NAD metabolism is discussed.
is
transferred to proteins as a monomer, oligomer, or polymer.
Poly(ADP-ribosyl)ation occurs mostly in nuclei, and accumulating
evidence indicates a close relationship between poly(ADP-ribosyl)ation
and the regulation of chromatin
activities(1, 2, 3) . While the physiological
roles of oligo- or mono(ADP-ribosyl)ation in the cytoplasm, organella,
or membranes in eukaryotic cells have not been established, the
involvement of mono(ADP-ribosyl)ation in the regulation of signaling
pathways or membrane traffic has been
suggested(4, 5, 6, 7, 8) . 
)an exotoxin produced by Bordetella
pertussis(9, 10) , ADP-ribosylates the
subunit of several GTP-binding proteins (G-proteins), which act as
central signal transducers(11, 12) , and causes the
functional uncoupling of the G-proteins from the
receptors(13, 14) . The C3 exoenzyme from Clostridium botulinum types C and
D(15, 16, 17, 18) ADP-ribosylates
the Rho family of low molecular weight GTP-binding proteins in
eukaryotic cells(19) , which are required for the organization
of the microfilament network(20, 21) .
ADP-ribosylation by the C3 exoenzyme interrupts the interaction of the
Rho proteins with the downstream effector molecules(22) .
Interestingly, ADP-ribosyltransferases comparable to PT and the C3
exoenzyme have been found in human erythrocytes (23) and in
bovine brain(24) , respectively. These endogenous
ADP-ribosyltransferases may be involved in the essential cellular
functions(25) .
, a minor ganglioside species recognized by a
cholinergic neuron-specific antibody(30) , has a strong
inhibitory activity for PT. On the other hand, gangliosides G and G had inhibitory activity for the C3 exoenzyme.
This is the first report that demonstrates the inhibitory effect of
gangliosides on ADP-ribosylation. We propose that gangliosides are
involved in the cellular NAD metabolism.
Materials
[adenylate-
P]NAD (800 Ci/mmol) and
[carbonyl-
C]NAD (25-40 Ci/mmol) were purchased from DuPont NEN.
Neuraminidases from Arthrobacter ureafaciens and C.
perfringens were from Boehringer Mannheim GmbH (Germany). Ceramide
from bovine brain sphingomyelin (type III), galactocerebroside from
bovine brain (type II), N-acetylneuraminic acid,
G
, and G
were purchased from Sigma.
Oligomers of N-acetylneuraminic acid were from Nakarai
(Japan). PT was from Seikagaku Kogyo (Japan). To activate, 500 nM PT was preincubated for 15 min at 30 °C with 50 mM dithiothreitol (31) and 1 mM ATP(32) .
G-proteins were partially purified from bovine brain membranes as
described previously (33) by successive chromatographies on
columns of DEAE-Sephacel, ULTROGEL AcA34 (IBF Biotechnics, France), and
heptylamine-Sepharose.C3 Exoenzyme
The C3 exoenzyme was purified from a
culture supernatant of C. botulinum type C, strain 6813, as
described previously(18) .Preparation of the Bovine Brain Membranes and the Cholate
Extract
Bovine brain was homogenized in 3 volumes of Buffer A
(10 mM Tris-HCl, pH 8.0, 1 mM EDTA, 1 mM dithiothreitol, 1 mM phenylmethylsulfonyl fluoride). The
homogenate was centrifuged at 100,000 
g for 60 min.
The pellet was washed twice and suspended in the same buffer. This
suspension was used as the membrane fraction. To make cholate extract,
the membrane fraction (70 mg of protein) was suspended in Buffer A (10
ml) containing 1% sodium cholate, stirred on ice for 60 min, and
centrifuged at 100,000 g for 60 min. g for 15 min. The
chloroform/methanol layer (lower), water/methanol layer (upper), and
the interface were separated. After evaporation, each fraction was
suspended in 10 ml of chloroform/methanol (2:1, v/v).Preparation of the Ganglioside Fraction from Bovine Brain
Membranes
The ganglioside fraction was prepared essentially
according to Folch et al.(34) . The membrane fraction
(300 mg of protein in 5 ml) was suspended in 100 ml of
chloroform/methanol (2:1, v/v) and was centrifuged at 2,000 g for 15 min. The pellet was resuspended in 100 ml of
chloroform/methanol (1:2, v/v) and was centrifuged. The first and the
second extracts were combined, evaporated, and suspended in
chloroform/methanol (2:1, v/v). After filtration, one-fourth volume of
water was added and the suspension was centrifuged. The upper layer (6
ml) was used as the crude ganglioside fraction.Preparation of Individual Gangliosides
Total
bovine brain gangliosides were applied to a Q-Sepharose column and
fractionated into 23 fractions as described previously(28) .
G
and G
were purified as
described previously(29, 30) .PT-catalyzed ADP-ribosylation
The typical reaction
mixture (50 µl) contained 50 nM activated PT (5 µl),
2.5 µl of partially purified G-protein (0.1 mg/ml), and 1
µM [adenylate-P]NAD (1 µCi/assay) in Buffer B (50 mM Tris-HCl, pH 8.0,
20 mM dithiothreitol, 1 mML--dimyristoylphosphatidylcholine). The reaction was
started by the addition of PT. After a 30-min incubation at 30 °C,
the reaction was stopped by the addition of 4% SDS and 16%
trichloroacetic acid. Precipitated proteins were collected on a
nitrocellulose filter (Schleicher & Schuell BA85) and were washed
with 20 ml of 6% trichloroacetic acid. The filters were dried and the
retained radioactivity was measured. The incorporation of the
radioactivity into the protein was linear versus time within
the 30-min incubation. To obtain the background level, PT was omitted
from the reaction mixture. The incorporation in the absence of other
additions was used as the control. After subtraction of the background
value due to nonspecific binding, each value was expressed as a
percentage of the control value. For the kinetic measurement, the
reaction mixture (50 µl) contained 74 nM G-protein,
0.25-4.0 µM [adenylate-P]NAD (50 Ci/mmol), and 10 nM activated PT in Buffer B. After
the reaction was started by adding PT, aliquots (10 µl) were
withdrawn at 20, 40, and 60 s.Preparation of the S1 Subunit of PT and Measurement of
ADP-ribosylation Catalyzed by the S1 Subunit
The S1 subunit of
PT was prepared using a haptoglobin-Sepharose
column(35, 36) . The flow-through fraction contained
only the S1 subunit. The fractions containing the S1 subunit were
combined and dialyzed against 10 mM Tris-HCl, pH 8.0. The S1
subunit (10 µl) was mixed with 10 µl of 25 mM dithiothreitol and 0.5 mM ATP and was incubated for 15
min at 30 °C for activation. The activated S1 subunit solution (20
µl) was used for the ADP-ribosyltransferase activity as described
above. The final concentration of the S1 subunit in the reaction
mixture was about 30 nM.C3 Exoenzyme-catalyzed ADP-ribosylation
The
reaction mixture (50 µl) contained 10 nM C3 exoenzyme, 0.5
µl of the cholate extract of bovine brain membranes (3.0 mg/ml),
and 0.06 µM [adenylate-P]NAD (0.5 µCi/assay) in Buffer C (10 mM Tris-HCl, pH 7.0,
10 mM dithiothreitol, 1 mM MgCl
). After
incubation for 15 min at 20 °C, the incorporation of the
radioactivity into the protein was measured. The linearity of the
incorporation versus time was confirmed.NAD
The reaction
mixture (20 µl) contained 46 µM [carbonyl- GlycohydrolysisC]NAD (0.1 µCi/assay) and 50 nM activated PT or C3
exoenzyme in Buffer B or Buffer C, respectively. After a 2.5
h-incubation at 30 °C, the reaction mixture was mixed with 5 µl
of 50 mM NAD and 10 mM nicotinamide
and was spotted onto Whatman No. 3MM paper. The paper was developed by
1 M AcONH
(pH 5.0), 95% EtOH (3:7, v/v) as
described previously(37) , and NAD and
nicotinamide were detected under UV light. The radioactivities in these
spots were measured. The release of nicotinamide from NAD was linear up to 2.5 h of incubation.Neuraminidase Treatment
The ganglioside fraction
from bovine brain (120 µl) was dried under vacuum and redissolved
in 50 µl of 0.1 M acetate buffer, pH 5.0, and 10 µl of A. ureafaciens neuraminidase (0.1 unit) or HO.
Alternatively, the ganglioside fraction from bovine brain (30 µl)
was dried up and redissolved in 6 µl of 1 M acetate
buffer, pH 5.0, and either 54 µl of C. perfringens neuraminidase (0.8 unit) in 10 mM EDTA or 54 µl of 10
mM EDTA. After incubation for 24 h at 40 °C, the fraction
was boiled for 5 min, and their aliquots (10 µl) were used to
examine the inhibitory activity.
Inhibitory Activity of PT-catalyzed ADP-ribosylation
Found in the Cholate Extract of Bovine Brain Membranes
We have
previously reported the presence of a heat-stable inhibitory activity
of PT-catalyzed ADP-ribosylation in the cholate extract of bovine brain
membranes(27) . As the inhibitory activity was also stable
after treatment with chloroform (data not shown), lipids were extracted
and fractionated from the cholate extract by the method of Folch et
al.(34) . The inhibitory activity in the upper layer
(gangliosides) was higher than that in either the lower layer
(phospholipids) or the interface (proteins) fraction (data not shown). The Inhibitory Activity of PT in the Crude Ganglioside
Fraction of Bovine Brain Membranes
To examine the involvement of
a ganglioside, a crude ganglioside fraction was prepared from bovine
brain membranes, according to Folch et al.(34) .
Addition of the crude gangliosides inhibited the PT-catalyzed
ADP-ribosylation in a dose-dependent manner (Fig. 1A).
The inhibitory activity of the ganglioside fraction was about one-fifth
that of the cholate extract obtained from the same amount of membranes.
About 50% of the inhibitory activity was relieved by treatment with
neuraminidase from A. ureafaciens (Fig. 1B).
The treatment with neuraminidase from C. perfringens led to a
30% attenuation of the inhibitory activity (data not shown). Therefore,
ganglioside species are probably involved in the inhibition.
and
G in the fraction were about 0.08 and 0.13 mg/ml,
respectively. Values are means ± S.D. from duplicate assays. B and C, the ganglioside fraction was treated with
neuraminidase as described under ``Experimental Procedures.''
The ADP-ribosylation was measured in the presence of ganglioside
without (1) or with (2) neuraminidase treatment.
Values are means ± S.D. from duplicate
assays.
Effect of Individual Gangliosides on the PT-catalyzed
ADP-ribosylation
For the screening of ganglioside species, crude
ganglioside fraction from bovine brain was fractionated by Q-Sepharose
column chromatography(28) . All fractions except for the
flow-through fraction inhibited the PT-catalyzed ADP-ribosylation (data
not shown). The inhibitory activity was highest in the final fraction
containing G
and G
(Fraction 23).
A similar result was obtained when the gangliosides were treated in
alkaline conditions prior to the separation by Q-Sepharose column
chromatography (data not shown). As shown in Fig. 2A,
PT-catalyzed ADP-ribosylation was inhibited by purified G
with an IC
value of about 0.1 mg/ml (40
mM), but not by G
. Dose-dependent
inhibition by various purified gangliosides was shown in Fig. 2B. The inhibition was 95% when G
was added at 0.25 mg/ml, whereas the inhibition did not exceed 50
to 80% when either G, G, G,
or G was added up to 0.9 mg/ml. The inhibition depends on
the concentration of NAD or L--dimyristoylphosphatidylcholine. The kinetic analysis
suggests that G
inhibits the PT-catalyzed
ADP-ribosylation in a competitive manner versus NAD (Fig. 3). The apparent inhibitory activity of
G
in the absence of L--dimyristoylphosphatidylcholine was higher than that in
the presence of L--dimyristoylphosphatidylcholine (data
not shown). G
was detected, by monoclonal antibody
GGR-41, in the upper layer of the cholate extract as well as in the
crude ganglioside fraction from bovine brain membranes (Fig. 4).
The amount of G
in the cholate extract was larger
than that in the crude ganglioside fraction.
(
) or
G
(
). B, the reaction mixture
contained G (
), G (
), G (
), G (
), or G
(). Values are means from duplicate assays. C,
structures of ganglioside species.
concentration on the inhibition of PT-catalyzed ADP-ribosylation by
G
. The PT-catalyzed ADP-ribosylation was measured in
the absence () or presence () of 0.1 mg/ml G
as described under ``Experimental Procedures.'' Data
are presented as Lineweaver-Burk plots. Essentially the same result was
obtained in a separate experiment.
, G
, G, G,
G
, G, and G in panel A and G
and G
in panel
B); lane 2, the water/methanol layer of the cholate
extract (100 µl); lane 3, the crude ganglioside fraction
(40 µl). The gangliosides were visualized by the resorcinol/HCl
reagent (A) or by immunostaining with the monoclonal antibody
GGR-41 (B).
Effect of Gangliosides on the ADP-ribosylation Catalyzed
by the S1 Subunit of PT
PT is a hexameric protein with an A-B
architecture (38) . The A protomer is composed of a single S1
subunit containing the catalytic site of ADP-ribosyltransferase. The B
oligomer is made up of five subunits (S2, S3, S4, and S5 in an 1:1:2:1
ratio) and is required for the binding of this toxin to the membranes
of target cells(35, 38) . The S2 and S3 subunits of
the B oligomer contain carbohydrate recognition domains (39) and bind to glycoproteins(40, 41) . Even
in the presence of ATP, which promotes dissociation of the PT subunits (42) , more than 80% of the S1 subunit is associated with the B
oligomer(31) . To examine whether the inhibition by
gangliosides is due to the direct interaction with the S1 subunit or
with the carbohydrate recognition domains of the B oligomer, the S1
subunit of PT was isolated by a haptoglobin-Sepharose column
chromatography. As shown in Fig. 5, gangliosides inhibited the
S1 subunit-catalyzed ADP-ribosylation. G
was more
effective than either G, G,
G, G, or G was. Therefore,
the site of ganglioside action is identified as the S1 subunit.
(1), G (2), G (3), G (4), G (5), and G
(6). Values are means ± S.D. from duplicate
assays.
Inhibitory Activity of C3 Exoenzyme-catalyzed
ADP-ribosylation in the Bovine Brain Membranes
As shown in Fig. 1A, the ganglioside fraction from the bovine brain
membranes inhibited the C3 exoenzyme-catalyzed ADP-ribosylation. The
inhibitory activity was completely relieved by treatment with
neuraminidase from C. perfringens (Fig. 1C).Effects of Individual Gangliosides on C3
Exoenzyme-catalyzed ADP-ribosylation
The effect of gangliosides
separated by Q-Sepharose column chromatography on the C3
exoenzyme-catalyzed ADP-ribosylation was studied. In contrast to the
effect on the PT-catalyzed ADP-ribosylation, the C3-exoenzyme catalyzed
ADP-ribosylation was inhibited by the fraction containing
G, G, G, or G,
but not by G
(Fraction 23, data not shown). As shown
in Fig. 6, G, G, G,
and G inhibited the ADP-ribosylation in a dose-dependent
manner. The effects of G and G were
slightly greater than those of G and G. The
amount of inhibitory activity of G
(Fraction 23) was
smaller than that of G, G, G,
and G. The inhibition approached 80% by the addition of
up to 0.9 mg/ml of the gangliosides.
(), G (), G (), G (), and Fraction 23 from the
Q-Sepharose column chromatography (). Values are means from
duplicate assays.
Effects of Gangliosides on NAD
The inhibition of ADP-ribosylation can be
explained by the interaction of gangliosides with
ADP-ribosyltransferase or by the interaction with the substrate
proteins. As the heterotrimeric form of G-protein is the preferred
substrate of PT(43) , the ADP-ribosylation can be inhibited by
the dissociation of heterotrimeric G-proteins. The C3
exoenzyme-catalyzed ADP-ribosylation is inhibited by the formation of
the GTP-bound form of the Rho/Rac proteins with guanine
nucleotides(44) . Glycohydrolysis glycohydrolysis activity of
the C3 exoenzyme. As shown in Fig. 7, G and
G inhibited the C3 exoenzyme-catalyzed NAD glycohydrolysis, indicating that these species act directly on
the C3 exoenzyme. On the other hand, the addition of either
G, G, or G
(Fraction 23)
had no effect. The lack of inhibition of NAD glycohydrolysis with G
is consistent with its
effect on the ADP-ribosylation. As for G and
G, our data could not discriminate whether they act
directly on the C3 exoenzyme. The crude ganglioside fraction also
inhibited the C3 exoenzyme-catalyzed NAD glycohydrolysis. PT-catalyzed NAD hydrolysis was
inhibited by the crude ganglioside fraction as well as by
G
. Accordingly, G/G and
G
inhibit the ADP-ribosylation by interacting with
the C3 exoenzyme and with PT, respectively.
glycohydrolysis. A, the C3
exoenzyme-catalyzed NAD glycohydrolysis was measured
as described under ``Experimental Procedures'' in the
presence of G (1), G (2),
G (3), G (4), or the
Q-Sepharose Fraction 23(5) , at final concentrations of 0.5
mg/ml, and in the presence of 4 µl of the crude ganglioside
fraction(6) . B, the PT-catalyzed NAD glycohydrolysis was measured as described under
``Experimental Procedures'' in the presence of G
at a final concentration of 0.1 mg/ml (1) and 4 µl
of the crude ganglioside fraction (2). To obtain the
background level, the enzymes were omitted from the reaction mixture.
The hydrolysis in the absence of other additions was used as the
control. After subtraction of the background, each value was expressed
as a percentage of the control value. Values are means ± S.D.
from duplicate assays.
Effects of Ceramide, Galactocerebroside, and Sialic
Acid
To investigate the structural requirement for the
inhibitory effect of gangliosides, the effects of ceramide,
galactocerebroside, and sialic acid were studied. Ceramide and
galactocerebroside at 0.1-0.5 mg/ml did not inhibit the PT and C3
exoenzyme-catalyzed ADP-ribosylation (data not shown). Sialic acid
species (monomer, dimer, trimer, tetramer, pentamer, and hexamer at
0.1-0.7 mg/ml) also did not inhibit either ADP-ribosylation (data
not shown). Thus, both the lipid components and the sialic acid
residues in the gangliosides are required for their inhibitory effect.
Inhibition of the PT- and C3 Exoenzyme-catalyzed
ADP-ribosylation by Gangliosides
In the present study, we
demonstrated that gangliosides inhibit the ADP-ribosyltransferases of
PT and the C3 exoenzyme. PT and the C3 exoenzyme modify different
target proteins with distinct amino acid specificities, a cysteine
residue (11, 12) and an asparagine
residue(22) , respectively. The C3 exoenzyme is a basic
protein(18) , whereas the catalytic S1 subunit of PT is an
acidic protein(35) , and no significant homology was found
between PT (45) and the C3 exoenzyme(46) . Therefore,
the inhibition of both ADP-ribosyltransferases by gangliosides suggests
that gangliosides interact with an essential structure of the enzymes
that is required in common for both types of ADP-ribosylation.G
PT has
been used to demonstrate the involvement of PT-sensitive G-proteins
upstream of the inhibition of adenylate cyclase, the activation of
phospholipase A
as an Inhibitor of PT, phospholipase C, or ion
channels(47, 48) . Sensitivity to PT has also been
used to P-label the G-proteins with
[adenylate-P]NAD. PT
is part of the standard criteria for the characterization of the
signaling processes. However, there is an ambiguity in the
interpretation of PT-insensitive results. The negative results can be
caused either by the presence of PT-insensitive G-proteins or the
signaling pathway, or by the presence of endogenous inhibitors.
Although the presence of an endogenous inhibitor has been previously
suggested(26, 27, 49, 50) , the
chemical nature of the inhibitor has never been characterized. Since we
found in this investigation that G
inhibits PT, some
of previous PT-insensitive results may be due to endogenous
G
. For example, the decrease in the endogenous
inhibitory activity of PT after K-ras transformation in
thyroid cells (50) is possibly related to the change in the
amount of G
.
is a minor
ganglioside, identified as the antigen of the cholinergic
neuron-specific antibody, anti-Chol-1, and represents about 0.03%
of total gangliosides(30) . G
was found to
exist in the central nervous system tissues as well as in rat
liver(51) . Information on ganglioside metabolism in these
tissues will be helpful to evaluate the PT-insensitive results.Mechanism of Inhibition
Gangliosides modulate the
activities of various enzymes(52) . G partly
mimics the effect of Ca
/calmodulin and modulates the
Ca/calmodulin-dependent kinase activity(53) .
G, G, and G bind to
calmodulin or a calmodulin-like binding site of enzymes and affect the
activity of the calmodulin-dependent nucleotide
phosphodiesterase(54, 55) . G inhibits
the kinase activity associated with the epidermal growth factor and
platelet-derived growth factor receptors by preventing
receptor-receptor aggregation, which is required for the subsequent
activation of the kinase domain(56, 57) . However,
these mechanisms do not explain the inhibitory effect on the
ADP-ribosylation catalyzed by PT or the C3 exoenzyme. to the B oligomer of PT was
reported(58) , the inhibitory effect was not specific for
G. (ii) The ADP-ribosyltransferase activity of the
catalytic S1 subunit of PT, which does not have the carbohydrate
recognition domain, was inhibited by gangliosides, particularly by
G
. (iii) Inherently, the C3 exoenzyme has no
carbohydrate recognition domain. The interaction of gangliosides with
either PT or the C3 exoenzyme may occur at different sites other than
carbohydrate binding domains.
, but not by G
.
This observation suggests that tandem sialic acid residues linked to
the internal galactose residue are involved in the inhibition. Tandem
sialic acid residues are also present in the molecules of G and G, which inhibit the C3 exoenzyme. As
G
acts as a competitive inhibitor versus NAD, a structural similarity may exist between
the molecules of G
and NAD. It is
possible to locate one carboxyl group of one sialic acid close to the
other by rotating the C-C or C-O bonds of the glycerol group connecting
the two sialic acids (Fig. 8). The distance between the two
carboxyl groups is similar to that between the two phosphate groups of
NAD. A negative charge cluster, formed by the two
carboxyl groups in the tandem sialic acid residues of the b-series
gangliosides, can mimic the diphosphate moiety of NAD.
As the dimer of sialic acid did not inhibit the ADP-ribosyltransferase
activity of PT and the C3 exoenzyme, another component of the
ganglioside molecule, such as the gangliotetraose core structure, may
also be required. A structural study of the conformation of
NAD bound to PT or the C3 exoenzyme will be necessary
to elucidate how gangliosides inhibit the activity of these
ADP-ribosyltransferases.
(A) and a sialic acid dimer (B).
The conformation of the nicotinamide-ribose bond and the adenine-ribose
bond are assumed to be anti, according to the results of
NAD bound to L-lactate
dehydrogenase(80) . The conformation of the pyranose ring of
sialic acid residue is C(81) .
Phosphorus, carbon, oxygen, and hydrogen atoms were colored in yellow, blue, red, and white,
respectively.
Involvement of Gangliosides in the Inhibition of
ADP-ribosylation
Benzamide and its derivatives are commonly used
as inhibitors of ADP-ribosyltransferases. Recently, hydrophobic
molecules, such as long-chain fatty acids and vitamin K,
were also reported to be inhibitors(59, 60) . As the
intracellular localization of gangliosides was
reported(61, 62, 63, 64) , they can
be potential members of the endogenous inhibitors of
ADP-ribosyltransferases in cells.Possible Action of Gangliosides as Regulators of
NAD
In this investigation,
we found that gangliosides inhibit the NAD Metabolism in Cells glycohydrolase activities of PT and the C3 exoenzyme. As PT and
the C3 exoenzyme are entirely different types of
ADP-ribosyltransferases, the crucial target of ganglioside for the
inhibition may be the NAD glycohydrolase. The presence
of an ectoenzyme of NAD glycohydrolase (69) suggests that gangliosides regulate the NAD metabolism on the surface of the cells. An ectoenzyme of
NAD glycohydrolase is induced during retinoic
acid-induced differentiation in human leukemic HL-60
cells(70) , and this enzyme was identified as the human
leukocyte cell surface antigen CD38(71) . CD38 has a hyaluronic
acid binding motif and probably interacts with the extracellular matrix
molecules(72) . Also, the rat T-cell alloantigen RT6.2 was
found to be a glycosylphosphatidylinositol-anchored NAD glycohydrolase (73) and is homologous to a
glycosylphosphatidylinositol-anchored mono(ADP-ribosyl)transferase from
rabbit skeletal muscle(74) . Although the physiological
significance of the NAD metabolism outside the cells
is elusive at present, we speculate that gangliosides mediate signals
from the cell surface to the cell interior by regulating the
NAD metabolism. An implicated result is that CD38 also
has an activity to generate cyclic
ADP-ribose(75, 76) , a newly identified second
messenger for Ca mobilization(77) . The
concept of the regulation of NAD metabolism by
gangliosides should be tested in future studies.
)
We thank Prof. Yasuo Kishimoto, at the University of
California, for the critical reading of the manuscript and helpful
discussions. We also thank Dr. Yutaka Ito, at the Institute of Physical
and Chemical Research (Riken), and Tohru Terada, at the University of
Tokyo, for the graphic drawings, Dr. Katsumi Nogimori, at the Kaken
Pharmaceutical Co., for valuable advice for the isolation of the S1
subunit of PT, and Dr. Zhang Gu, at the Institute of Physical and
Chemical Research (Riken), for the preparation of G
.
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  M. Hara-Yokoyama, I. Kukimoto, H. Nishina, K. Kontani, Y. Hirabayashi, F. Irie, H. Sugiya, S. Furuyama, and T. Katada Inhibition of NAD+ Glycohydrolase and ADP-ribosyl Cyclase Activities of Leukocyte Cell Surface Antigen CD38 by Gangliosides J. Biol. Chem., May 31, 1996; 271(22): 12951 - 12955. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Nakayama, M. N. Fukuda, Y. Hirabayashi, A. Kanamori, K. Sasaki, T. Nishi, and M. Fukuda Expression Cloning of a Human G[IMAGE] Synthase J. Biol. Chem., February 16, 1996; 271(7): 3684 - 3691. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| All ASBMB Journals | Molecular and Cellular Proteomics |
| Journal of Lipid Research | ASBMB Today |