|
|
|
|||||||
|
|||||||||
(Received for publication, December 6, 1994; and in revised form, January 6,
1995) From the
Monoclonal antibody, LAS-2, directed against the
Heterotrimeric guanine nucleotide-binding proteins (G proteins), ( The amino-terminal domain of G Cotranslational acylation of the
amino-terminal glycine with myristate occurs in several of the
G
Murine monoclonal antibodies against purified bovine
G
Figure 1:
Inhibition by LAS-2 of
G
To define the LAS-1 and -2
epitopes better, G
Figure 2:
Immunoreactivity of LAS antibodies. A, reaction of LAS-2 with submaxillary Arg-C-digested
G
Further mapping was done using recombinant proteins expressed in E. coli. G
Figure 3:
Requirement for amino-terminal
myristoylation of G
The LAS-2 antibody was very specific for
G
Figure 4:
Immunoreactivity of LAS antibodies. A, specificity of LAS-2 for G
The LAS-1 antibody shown
in Fig. 1reacted in a manner identical to LAS-2 on all
immunoblots including those performed with proteolytic digests of
G It appears that monoclonal antibody LAS-2, a
G Upon
exchange of GDP for GTP, G
Volume 270,
Number 12,
Issue of March 24, 1995 pp. 6436-6439
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
Subunits (*)
subunit
of transducin (G
), inhibited
G
-dependent, pertussis toxin-catalyzed ADP
ribosylation of G
and was specific for
G
. Immunoblotting studies on proteolytic fragments of
G
were consistent with an amino-terminal epitope. To
define the antibody recognition site, recombinant G
was synthesized in Escherichia coli cotransfected with
or without yeast N-myristoyltransferase. Amino-terminal fatty
acylation of G
, verified by use of radiolabeled fatty
acid, was required for immunoreactivity. LAS-2 did not react with a
chimeric protein consisting of residues 1-9 of G
and the remainder G
, regardless of its
myristoylation. Immunoreactivity was observed when amino acids
1-17 of G
were present in a G
chimera and the protein was amino-terminally myristoylated; there
was no reactivity without myristoylation. It appears that the LAS-2
epitope requires both G
-specific sequence in amino
acids 10-17 and a fatty acyl group in proximity to these
residues. These results are consistent with the hypothesis that the
myristoyl group is essential for protein structure; conceivably it
``folds back'' on and stabilizes the amino-terminal structure
of G
, as opposed to protruding from an amino-terminal
-helix and serving as an amino-terminal membrane anchor.
)consisting of , 
, and 
subunits, couple
cell surface receptors to effectors(1) . GTP/GDP exchange and
GTP to GDP hydrolysis, which regulate G protein activity, occur on the
subunit (G
) and require the presence of receptor
and the subunit
(G
)(2, 3, 4, 5) . 
is important for
interaction with G. Removal of 21 amino acids from
the amino terminus of the subunit of transducin
(G
), using Staphylococcal V8 protease, inhibited
-dependent reactions including GTP hydrolysis, GTP/GDP
exchange, and binding of G
to rhodopsin(2) .
Trypsin, which removes the first 18 amino acids of G
,
destroyed the ability of to stimulate pertussis
toxin-catalyzed ADP-ribosylation of G
(6) as
did amino-terminal deletions in a recombinant
G
(7) . Amino acids 7-10 of recombinant
G
were found to be critical for its interaction with
G(8) .
subunits (9, 10, 11, 12) and facilitates
their membrane attachment. Removal of the myristoylated amino terminus
of G
by tryptic digestion produced a soluble protein
from one that was originally membrane bound(13) . G
modified with hydrophilic myristate analogues was less likely
than native G
to associate with membranes (9) . Myristoylation also appears to play a role in
protein-protein interaction(14, 15) . Both
-stimulated ADP-ribosylation and affinity for
-Sepharose were significantly greater with myristoylated than
nonmyristoylated G
(8, 16) . N-Fatty acylation of G
was found important
for its interaction with G(17) . Based on
these data, it has been postulated that the myristoyl moiety may be
involved in stabilizing the structure of the amino terminus. We
describe here an anti-G
monoclonal antibody, LAS-2,
whose epitope involves the G-binding domain of
G
and requires amino-terminal fatty acylation as well
as specific G
amino-terminal sequence.
Materials
[
P]NAD (30 Ci/mmol), GA/1 antibody,
RM/1 antibody and AS/7 antibody were purchased from DuPont NEN;
TPCK-treated trypsin was from Worthington. Glutathione-agarose, NAD,
ATP, and soybean trypsin inhibitor were from Sigma, and blue Sepharose
CL6B was from Pharmacia Biotech Inc.
Isopropyl-1-thio--galactopyranoside was from U. S. Biochemical
Corp; pertussis toxin was from List Biological Laboratories. Goat
anti-rabbit IgG horseradish peroxidase-coupled antibody was from
Pierce, and goat anti-mouse IgG horseradish peroxidase-coupled antibody
was from Promega. LB medium was from Digene, ampicillin sodium and
kanamycin were from Amresco; and molecular weight standards were from
Life Technologies, Inc. SDS-PAGE and Western blotting were done in a
NOVEX Minicell apparatus or Bio-Rad Trans-Blot apparatus. Protein
concentration was determined using Bio-Rad protein assay with bovine
serum albumin as the standard.Methods
Preparation of Proteins
Transducin was purified
from bovine rod outer segments(18) . Blue-Sepharose
chromatography was used to separate the G
and
G subunits.(19) . Rhodopsin was prepared
by method of Hong and Hubbell(20) . G
and
G
were isolated from bovine brain (21) .
G
was a generous gift from Dr. Victor Rebois of NINDS,
National Institutes of Health.Proteolytic Cleavage of Transducin
G
was digested with trypsin (22) or Arg-C(23) .
Cleavage of G
with hydroxylamine was based
on the method of Borstein and Balian(24) .Immunoblots
Intact proteins or proteolytic
fragments were resolved by SDS-PAGE(25) ; transfer to
nitrocellulose was performed in 25 mM Tris, 192 mM glycine buffer with 20% methanol (26) either in a NOVEX
transfer module (30 V for 2 h) or a Bio-Rad Trans-Blot apparatus (100 V
for 16 h). Blots were processed as described(27) .ADP-Ribosylation of
G
G
(1 µg) and
G (1 µg) were incubated in the presence of
LAS-1, LAS-2, or control mouse IgG at 4 °C for 1 h before pertussis
toxin-catalyzed ADP-ribosylation, which was carried out at 30 °C
for 30 min in 50 mM potassium phosphate (pH 7.5), 10
µM [P]NAD (2 µCi/assay), 1
mM ATP with 2 µg of activated pertussis toxin in total
volume of 100 µl. Pertussis toxin was activated for 10 min with
dithiothreitol immediately before use(6) . Proteins were
precipitated with 7.5% trichloroacetic acid and subjected to
electrophoresis in 12% SDS-polyacrylamide gels, followed by
autoradiography with Kodak XAR film and an intensifying screen at
-70 °C for 30 min.Construction of Coding Regions for Recombinant
Proteins
Bovine retinal G
cDNA was used as
template for PCR amplification of the rG
coding
region. The sense primer was
5`-ACGACGACGCATATGGGGGCTGGGGCCAGCGCTGAGGAGA-3`, and the antisense
primer was 5`-GCGCCACGGTGATCATCAGAAGAGCCCGCAGTCTTT-3`. G
cDNA from bovine brain was used as template for PCR preparation
of coding regions for rG
and the
G
-G
chimeric proteins. The sense
primer for the rG
PCR was
5`-ACGACGACGCATATGGGATGTACTCTGAGCGCAGAGGAG-3`, and the antisense primer
was 5`-GCGCCACGGTGATCATCAGTACAAGCCGCAGCCCCGGAG-3`. For the
G
9-G
chimeric protein, which consists
of G
sequence in the first 9 amino acids followed by
G
sequence, the sense primer was
5`-ACGACGACGCATATGGGGGCTGGGGCCAGCGCAGAGGAGCGAGCCGCCCT-3`, and the
antisense primer was the same as that used in the rG
reactions. To synthesize the G
17-G
chimera, which contains G
sequence in amino
acids 1-17 and the remainder G
, an initial PCR
was carried out using
5`-ACGACGACGCATATGGGGGCTGGGGCCAGCGCTGAGGAGAAGCACTCAAGGGAGCTGGAGAAAAACCTCAAAGAG-3`
as the sense primer and the G
antisense primer used in
other reactions. The product of this PCR was used as template for a
second round of PCR amplification using the same 5` primer as in the
G
amplification and the 3` primer used for
G
amplifications. PCR products were purified with a
PCR purification spin kit (Qiagen), digested with restriction enzymes NdeI and BclI, and gel-purified. The NdeI
restriction site at the 5` end and BclI site at the 3` end
allowed directional cloning.Preparation of Recombinant Proteins
The
gel-purified, restriction-digested PCR products were inserted into a
modified pGEM vector, pT7/Nde, under control of a T7
promoter(28) . Competent DH5 Escherichia coli were transfected and screened on LB/ampicillin plates. The
presence of insert in the positive clones was verified by PCR. An
appropriate colony was selected and used for plasmid isolation and
transfection into BL21(DE3) E. coli (Novagen) or competent
BL21(DE3) E. coli that had previously been tranfected with the
pACYC177/ET3d/yNMT vector(29) , which contains the yeast N-myristoyltransferase coding region under control of a T7
promoter and a kanamycin resistance gene for coselection. Small scale
preparations were done by diluting fresh overnight cultures 1:10 and
growing at 37 °C in the presence of appropriate antibiotics for 1
h. Isopropyl-1-thio--D-galactopyranoside (final
concentration of 0.2 mM) and [
H]myristic
acid (final concentration of 5 µM with a specific activity
of 50 Ci/mmol) were added. Growth was continued an additional 3 h
before cells were pelleted and suspended in SDS-PAGE sample buffer.
Proteins were separated by SDS-PAGE and transferred to nitrocellulose.
Presence of incorporated fatty acid was verified by exposure of the gel
to Kodak XAR film at -70 °C after treatment with Pro-Mote
(Integrated Separation Systems).Antibodies
Hybridomas LAS-1 and LAS-2 and were
grown in serum-free medium (Ultradoma supplemented with glutamine) or
RPMI 1640 with glutamine and 5 or 10% fetal bovine serum
(BioWhittaker). G
polyclonal antibodies (pcGo) were
prepared as previously reported(30) .
were screened for functional effects on G
and G interaction. One clone, LAS-2,
inhibited G-stimulated ADP-ribosylation of
G
in a concentration-dependent manner; control
immunoglobulin of the same isotype (IgG
) and LAS-1 did not
inhibit the reaction (Fig. 1).
-dependent pertussis toxin-catalyzed
ADP-ribosylation of G
. Transducin was incubated with
LAS-1, LAS-2, or mouse IgG at 4 °C for 1 h before
initiation of pertussis toxin-catalyzed ADP-ribosylation, which was
carried out at 30 °C for 30 min in 50 mM potassium
phosphate (pH 7.5), 10 µM [P]NAD (2
µCi/assay), 1 mM ATP with 2 µg of activated pertussis
toxin in total volume of 100 µl. Pertussis toxin was activated for
10 min with dithiothreitol immediately before use(6) . Proteins
were precipitated with 7.5% trichloroacetic acid and separated by
SDS-PAGE in 12% polyacrylamide gels followed by autoradiography with
Kodak XAR film and an intensifying screen at -70 °C for 30
min. Lane1, 3 µg of indicated antibody; lane2, 10 µg of indicated antibody; lane3, 25 µg of indicated antibody. Cont.Ig,
control Ig.
was cleaved with endoprotease
Arg-C(23) , hydroxylamine at alkaline pH (24) , or
trypsin(22) ; the fragments were separated by SDS-PAGE,
transferred to nitrocellulose, and reacted with antibodies LAS-2 and
AS/7. Endoprotease Arg-C digestion generated LAS-2-reactive fragments
of 34, 23, and 15 kDa, all of which were refractory to microsequencing,
consistent with the presence of a blocked amino terminus (Fig. 2A). Hydroxylamine at alkaline pH cleaves
primarily at Asn-Gly bonds(24) ; cleavage of G
(Asn
-Gly
) produced 32-kDa
amino-terminal and 7-kDa carboxyl-terminal fragments. Both LAS-1 and -2
reacted with the 32-kDa piece, whereas AS/7 did not, confirming that
the LAS-reactive fragment lacked the carboxyl terminus (Fig. 2B, data not shown). Tryptic digestion of
G
initially removes a 2-kDa amino-terminal
fragment(22) ; this modification abolished immunoreactivity
with LAS-2. The presence of the remaining 37-kDa fragment was verified
by reactivity with the carboxyl-directed polyclonal antibody AS/7 (Fig. 2C). These findings all suggested that the LAS-2
epitope was located within the 18 amino-terminal amino acids removed by
trypsin, in agreement with the functional data obtained with
G(2, 3, 4, 5) .
. G
(5 µg) in buffer A (20
mM Tris (pH 7.5), 0.5 mM MgCl, 0.05
mM EGTA, 0.5 mM NaN, 1 mM dithiothreitol) was incubated with or without 0.625 µg of
Arg-C for 4 h at 37 °C; the reaction was stopped by the addition of
SDS-sample buffer and boiling for 10 min. Proteins and proteolytic
products were separated in 16% polyacrylamide gels and transferred to
nitrocellulose, which was then stained with Ponceau S or reacted with
LAS-2. Lane1, purified G
stained
with Ponceau S; lane2, Arg-C-treated G
stained with Ponceau S; lane3, Arg-C-treated
G
reacted with LAS-2. a-c indicate
LAS-2-reactive peptides subjected to sequencing by Edman degradation
(see text). Positions of protein standards (kDa) are on the left. B, reaction of antibodies with
hydroxylamine-treated transducin. 25 µg of G
in 50 µl of buffer A plus 58 µl of 3.75 M hydroxylamine, pH 9.25, were incubated at 37 °C for 3 h.
15-µl samples of this mixture were subjected to SDS-PAGE in
4-20% NOVEX gradient gels and then were stained with Coomassie
blue or used to transfer separated fragments to nitrocellulose for
reaction with antibodies. Positions of standard proteins are on the left; G
, G
,
G
, and 32 kDa are indicated in the center. Lane-, 2.5 µg of G
; lane+HA, hydroxylamine treatment, both stained
with Coomassie blue; laneLAS1, blot of
hydroxylamine-treated G
reacted with LAS-1; laneAS7, blot of hydroxylamine-treated
G
reacted with AS7. C, reaction of
LAS-2 with G
after tryptic digestion. G
(4 µg) in 40 µl of buffer A was incubated with or without
TPCK-treated trypsin (4 µg/ml) for 45 min at 30 °C,
precipitated with 10% trichloroacetic acid, subjected to
electrophoresis in 10-20% Tricine gradient gel, and transferred
to nitrocellulose. Transferred proteins and fragments were stained with
Amido Black (PRO) or reacted with LAS-2 or AS7 antibodies.
Presence or absence of trypsin is indicated by a (+) or (-),
respectively. Positions of G
, and 37-, 21-, 16-, and
5-kDa fragments are indicated at right.
cDNA inserted into a pGEX-2T
expression vector (Pharmacia) yielded high levels of rG
in a fusion protein with glutathione S-transferase(31) . Despite reacting with AS/7, this
fusion protein failed to react with LAS-2 both before and after
cleavage of the thrombin-sensitive link between glutathione S-transferase and G
(data not shown). An
amino-terminally myristoylated rG
was synthesized in E. coli by coexpressing G
and yeast N-myristoyltransferase(32) . Myristoylation of
rG
was verified by detection of H-labeled
G
after addition of [H]myristic
acid during protein synthesis. This myristoylated rG
reacted with LAS-1 and -2 on immunoblots. rG
synthesized without N-myristoyltransferase was not
labeled with [H]myristate and did not react with
LAS-2, although it did react with AS/7 and another transducin
monoclonal antibody, 6E4 (27) (Fig. 3). Fidelity of
expression and lack of proteolytic modification of the unmyristoylated
rG
was verified by Edman degradation though five
cycles, yielding the expected Gly-Ala-Gly-Ala-Ser sequence (Harvard
MicroChem).
for LAS reactivity. Transducin and
recombinant proteins, prepared as under ``Methods'' were
suspended in sample buffer, subjected to SDS-PAGE in 12%
polyacrylamide, transferred to nitrocellulose, and reacted with LAS or
6E4 antibodies. Amounts of G
(
4 µg) in the
three lanes were similar based on Coomassie blue staining. Lane1, G
; lane2, nonmyristoylated recombinant G
; lane3, myristoylated recombinant
G
.
. It reacted with only a single band from a bovine
retinal homogenate (Fig. 4A) and did not react with
G
, G
, or G
on
immunoblots (Fig. 4B). A recombinant
G
, expressed in E. coli, did not react with
LAS-2 either in a nonmyristoylated state or when coexpressed with N-myristoyltransferase and myristoylated (Fig. 4C, lanes1 and 2). A
chimeric protein, G
9-G
, consisting of
G
sequence in amino acids 1-9, and the
remainder G
sequence was constructed by replacing
Cys, Thr
, and Leu
of G
with Ala, Gly, Ala. This
protein was myristoylated when coexpressed with N-myristoyltransferase but did not react with LAS-2 either
with or without myristoylation (Fig. 4C, lanes3 and 4). A chimeric protein in which the first
21 amino acids of G
were replaced with the first 17
amino acids of G
(G
17-G
) did react with LAS-2,
but only when myristoylated; the nonmyristoylated
G
17-G
did not react (Fig. 4, lanes5 and 6).
among bovine
retinal proteins. Bovine retinas were homogenized in 0.25 M sucrose, 10 mM Tris-HCL (pH 7.5), 10 mM EDTA.
The homogenate was centrifuged (30 min/14,000 
g).
Supernatant proteins (100 µg) were precipitated with 7.5%
trichloroacetic acid, suspended in 20 mM phosphate buffer (pH
7.5), subjected to SDS-PAGE (16% polyacrylamide), transferred to
nitrocellulose, and stained with Ponceau S (lane1)
or reacted with LAS-2 (lane2). B,
specificity of LAS-2 monoclonal antibody. G
and
G were prepared as under ``Methods.''
G
and G
were isolated from bovine
brain(21) . G
was a generous gift from Dr.
Victor Rebois of NINDS, National Institutes of Health. 1 µg of each
of the proteins was subjected to SDS-PAGE (12% polyacrylamide),
transferred to nitrocellulose, and reacted with LAS-2. Protein transfer
was verified with 6E4 anti-G
(27) , pcGo
anti-G
(30) , and RM/1 anti-G
(DuPont NEN) antibodies. Lane1, bovine
G
; lane2, bovine
G; lane3, bovine brain
G
; lane4, bovine brain
G
; lane5, G
. C, reactivity of LAS-2 with rG
and
G
-G
chimeric proteins. rG
and chimeric G
-G
in whole
bacterial lysates were subjected to SDS-PAGE, transferred to
nitrocellulose, and reacted with LAS-2. There was approximately 1
µg of each recombinant protein in each lane. Lane1, myristoylated G
; lane2, rG
; lane3,
myristoylated G
9-G
; lane4, G
9-G
; lane5, myristoylated G
17-G
; lane6, G
17-G
; lane7, protein standards (97, 68, 43, 29, and 18
kDa).
, native G
proteins, and recombinant
proteins. It did not, however, interfere functionally with
G
-G interaction as judged by
inhibition of -dependent ADP-ribosylation. Why two antibodies
with seemingly identical mapping properties differ functionally is
unclear, but it may have to do with differing affinities for
G
.
-specific monoclonal antibody, which interferes with
G
-G interaction, has an epitope
that includes specific G
amino acid sequence within
positions 10-17 and requires a fatty acyl group in proximity to
these residues. Just as the first 9 amino acids are not specific for
G binding(17) , they do not encompass the
LAS-2 epitope, which lies further downstream. The amino terminus of
bovine G
is modified in a heterogeneous manner by
lauroyl (C:12), myristoyl (C:14), (cis-
-5)-tetradecaenoyl
(C:14:1), or (cis,cis-5,8)-tetradecadienoyl (C:14:2) fatty
acids(17, 33) . Kokame et al.(17) reported that the length of the amino-terminal fatty
acid was important for the ability of nonapeptides to inhibit the
interaction of G
and G, although
no requirement for specific amino acid sequence was observed.
disassociates from
G, and a conformational difference in crystal
structures of these two nucleotide-bound states has been documented (34, 35) . The proteins crystallized, however, lacked
the amino-terminal 25 amino acids and thus lacked the
G-binding domain and the LAS epitope studied
here. It has been postulated that myristate may act to stabilize an
amino-terminal amphipathic helix and participate in protein
function(38) . Recoverin, a retinal protein that is
heterogeneously fatty acylated in a manner similar to
G
, exposed hydrophobic acyl groups in response to
Ca
binding, thus behaving as a
Ca-myristoyl protein
switch(36, 37) . ADP ribosylation factor, a
myristoylated 
20-kDa GTP-binding protein involved in cellular
protein transport may utilize a GTP-myristoyl switch; the myristate may
participate in the reversible GTP-dependent association of ADP
ribosylation factor with membranes(37, 38) . The fatty
acyl group in the N-myristoylated G
proteins
may play a similar role, depending on whether GTP or GDP is bound. Our
findings suggest that within the G
amino-terminal
domain the fatty acid may ``fold back'' on and stabilize
peptide structure, consistent with the hypothesis that the fatty acid
does more than merely project from the protein into the lipid bilayer
to serve as an amino-terminal membrane anchor.
)
, transducin;
G
, subunit of transducin;
G, subunit of transducin;
G
, subunit of the abundant G protein obtained
from bovine brain; G
, subunit of the inhibitory
G protein of the adenylyl cyclase system; G
,
subunit of the stimulatory G protein of the adenylyl cyclase system;
PCR, polymerase chain reaction; rG, recombinant G protein.
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
ABSTRACT
This article has been cited by other articles:
|
|
 |
|
  E. Kostenis, F.-Y. Zeng, and J. Wess Functional Characterization of a Series of Mutant G Protein alpha q Subunits Displaying Promiscuous Receptor Coupling Properties J. Biol. Chem., July 10, 1998; 273(28): 17886 - 17892. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. M. Justice, J. J. Murtagh , Jr., J. Moss, and M. Vaughan Hydrophobicity and Subunit Interactions of Rod Outer Segment Proteins Investigated Using Triton X-114 Phase Partitioning J. Biol. Chem., July 28, 1995; 270(30): 17970 - 17976. [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 |