Biochem.
J.
(1973)
135,
569-572
Printed
in
Great
Britain
Phosphorylation
of
Myelin
Basic
Protein
by
an
Adenosine
3':
5'-Cycic
Monophosphate-Dependent
Protein
Kinase
By
P.
R.
CARNEGIE,*
B. E.
KEMP,t
P.
R.
DUNKLEY*
and
A.
W.
MuRRAYt
*Russell
Grimwade
School
ofBiochemistry,
University
ofMelbourne,
Parkville,
Vic.
3052,
and
tSchool
of
Biological
Sciences,
Flinders
University,
Bedford
Park,
S.
Austral.
5042,
Australia
(Received
20
August
1973)
Myelin
basic
protein
was
shown
to
be
a
substrate
for
protein
kinase
from
rabbit
muscle.
One
of
the
major
sites
of
phosphorylation
was
the
serine
residue
in
the
sequence
Gly-Arg-Gly-Leu-Ser-Leu.
The
arginine
residue
in
this
sequence
is
known
to
be
a
substrate
for
a
protein
methylase.
Cyclic
AMP-dependent
protein
kinases
are
exten-
sively
distributed
in
animal
tissues
(Krebs,
1972).
These
enzymes
are
frequently
assayed
by
using
histone
as
an
exogenous
substrate,
although
phosphorylation
of
a
limited
number
of
other
proteins
has
been
reported
(Krebs,
1972;
Johnson
et
al.,
1971;
Traugh
et
al.,
1973;
Weller
&
Rodnight,
1973).
Myelin
contains
a
protein
that
has
a
number
of
properties
in
common
with
histones.
This
protein
is
small
and
highly
basic,
is
acetylated
at
the
N-
terminal
residue,
is
methylated
at
an
arginine
residue
and
appears
to
lack
tertiary
structure
in
vitro
(Carnegie,
1971).
In
the
present
paper
myelin
basic
protein
is
shown
to
be
a
substrate
for
a
rabbit
muscle
protein
kinase.
Materials
and
methods
Rabbit
muscle
protein
kinase
was
prepared
by
the
procedure
of
Walsh
et
al.
(1968).
Myelin
basic
proteins
were
isolated
from
rat
and
human
brains
(Dunkley
&
Carnegie,
1973;
Deibler
et
al.,
1972).
[y-32P]ATP
was
prepared
as
described
by
Glynn
&
Chappell
(1964).
Other
materials
and
methods
were
as
described
by
Kemp
et
al.
(1973)
or
Carnegie
(1971).
Phosphorylation
of
the
human
myelin
basic
protein.
Standard
assay
mixtures
contained
5,umol
of
Hepes
buffer
[2-(N-2-hydroxyethylpiperazin-N'-yl)ethane-
sulphonic
acid,
potassium
salt,
pH7.5],
1
,umol
of
MgCl2,
52.03
nmol
of
[y-32P]ATP
(specific
radio-
activity
63.5
d.p.m./pmol)
and
100,ug
of
protein
in
a
final
volume
of
100,u.
In
assays
containing
cyclic
AMP
a
final
concentration
of
2,4M
was
used.
Assays,
done
in
duplicate,
were
carried
out
at
37°C
for
10min.
Termination
of
the
reaction,
acid
washing
and
liquid-
scintillation
counting
were
all
as
described
by
Kemp
et
al.
(1973).
Vol.
135
Results
and
discussion
There
was
no
detectable
phosphorylation
of
the
myelin
basic
protein
in
the
absence
of
protein
kinase
from
rabbit
muscle.
In
reaction
mixtures
containing
protein
kinase
and
myelin
basic
protein
most
of
the
radioactivity
of
the
phosphorylated
product
was
alkali-labile
(96%;
0.097M-KOH,
1000C,
15min)
and
stable
in
acid
(100%;
0.1
M-HCl,
100°C,
l5min).
After
partial
acid
hydrolysis
(2M-HCI,
100°C,
2h)
and
electrophoresis
in
formic
acid-acetic
acid-water
(1:4:45,
by
vol.;
pH2),
61
%
of
the
phosphorylated
product
was
not
hydrolysed
and
remained
at
the
origin
and
16%
migrated
with
the
O-phosphoserine
marker.
Only
a
small
amount
of
radioactivity
(3
%)
was
detected
in
the
O-phosphothreonine
position.
The
remaining
(20%)
radioactivity
migrated
with
the
Pi
marker.
No
correction
was
made
for
hydrolysis
of
O-phospho-
serine
or
O-phosphothreonine.
The
phosphorylated
product
bound
tightly
to
Sephadex
CM-25
and
was
not
eluted
with
0.1
M-NaHCO3.
The
phosphorylated
product
migrated
towards
the
cathode
during
electro-
phoresis
in
polyacrylamide
gel
(18%
acrylamide,
pH2.9).
The
radioactivity,
in
the
gel,
associated
with
the
basic
protein
accounted
for
91
%
of
the
total
radioactivity
precipitated
with
trichloroacetic
acid.
Under
these
electrophoresis
conditions
none
of
the
proteins
derived
from
the
rabbit
muscle
protein
kinase
migrated
towards
the
cathode.
Dephosphorylated
casein
(Kemp
et
al.,
1973),
whole
calf
thymus
histone
type
IIA
(Sigma
Chemical
Co.,
St.
Louis,
Mo.,
U.S.A.)
and
basic
protein
of
human
myelin
were
compared
as
substrates
for
the
rabbit
muscle
protein
kinase.
The
concentrations
of
dephosphorylated
casein,
histone
and
myelin
basic
protein
(mg/ml)
required
to
give
half-maximum
velocities,
in
the
presence
of
2,iM-cyclic
AMP,
were
3.7±0.5,
1.4±0.1
and
2.7±0.5
respectively.
The
corresponding
maximum
velocities
(pmol
of
569
P.
R.
CARNEGIE,
B.
E.
KEMP,
P.
R.
DUNKLEY
AND
A.
W.
MURRAY
Origin
t
Lys
,p..........
0
s...
Leu
Fig.
1.
Peptide
'map'
of
tryptic
digest
ofphosphorylated
basic
protein
of
human
myelin
Electrophoresis
was
at
3kV
for
33min
at
pH6.5
and
chromatography
was
in
butan-1-ol-pyridine-acetic
acid-
water
(15:10:
3:12,
by
vol.).
Conditions
and
numbering
of
the
peptides
1-30
were
as
described
by
Carnegie
(1971).
Spots
(major,
----;
minor,
.
j)
detected
by
radioautography
are
labelled
A
to
P;
*of
those
spots
only
A,
B
and
C
showed
a
distinct
ninhydrin
colour.
Lysine
and
leucine
were
used
as
markers
for
the
electrophoresis
and
chromatography.
Pi
transferred/min
per
mg
ofenzyme
protein)
obtained
with
these
substrates
were
31
+2,
956+53
and
540+
75.
Thus
the
myelin
basic
protein
was
an
effective
substrate
for
the
kinase.
Cyclic
AMP
caused
a
2.4-fold
increase
in
the
maximum
velocity
of
rabbit
muscle
protein
kinase
with
this
protein
as
substrate.
Tryptic
digestion
of
the
phosphorylated
basic
protein.
Preparation
of
phosphorylated
protein
for
sequence
studies
was
done
in
a
solution
containing
125,umol
of
Hepes
buffer
(potassium
salt,
pH7.5),
25,umol
of
MgCl2,
426.33
nmol
of
[y-32P]ATP
(specific
radioactivity
64.13
d.p.m./pmol),
5,umol
of
cyclic
AMP,
1Omg
of
human
basic
protein
and
rabbit
muscle
protein
kinase
(1.32mg)
in
a
final
volume
of
2.55
ml.
The
reaction
mixture
was
incubated
at
37°C
for
5h.
The
phosphorylated
basic
protein
was
purified
by
fractionation
on
Sephadex
CM-25
(Carnegie,
1971;
Dunkley
&
Carnegie,
1973),
and
examined
by
gel
electrophoresis
in
formic
acid-
acetic
acid-urea.
The
stained
gel
was
sliced
into
2mm
sections
and
of
the
radioactivity
measured
74%
was
recovered
in
a
single
band
corresponding
to
the
myelin
basic
protein.
The
remainder
was
associated
with
minor
components
which
were
possibly
breakdown
products
produced
by
traces
of
proteinases
present
during
the
prolonged
incubation.
The
phosphorylated
basic
protein
(0.28,umol;
0.37mol
of
PJ/mol)
was
digested
with
trypsin
(100,ug,
2h)
and
15
%
of
the
digest
was
fractionated
by
paper
electrophoresis
and
chromatography
(Carnegie,
1971).
The
peptide
'map'
of
the
digest
is
shown
in
Fig.
1.
Three
spots
A,
B
and
C,
not
normally
observed
in
tryptic
'maps'
of
the
basic
protein,
were
present.
Spot
A
was
predominant
and
spot
C
had
a
yellow
colour
with
ninhydrin.
Phosphorylated
peptides
were
located
by
radioautography
for
24h
and
the
radioactivity
of
the
peptides
was
determined
by
using
a
Nuclear-Chicago
gas-flow
counter
(efficiency
30.8%)
(Table
1).
The
peptide
spots
were
washed
with
acetone,
eluted
with
6M-HCI
and
heated
in
sealed
tubes
in
a
boiling-water
bath
for
lih.
The
1973
570
SHORT
COMMUNICATIONS
57
Table
1.
Recovery
of
radioactivity
from
a
tryptic
digest
ofphosphorylatedmyelin
basic
protein
For
position
of
spots
A
to
H
see
Fig.
1.
Spots
I
to
P
from
the
human
protein
had
a
total
of
8
%
of
the
radioactivity
of
the
digest.
Two
additional
spots,
which
were
present
only
in
the
digest
of
the
rat
protein,
had
1.8
and
6.9
%
of
the
radioactivity.
'Unknown'
denotes
an
unidentified
spot
with
an
electrophoretic
mobility,
at
pH
3.5,
1.61
times
that
of
O-phosphoserine.
Spot
A
B
C
D
E
F
G
H
Total
recovery
Total
d.p.m.
taken
Recovery
of
radioactivity
(Y.)
Human
27.4
11.8
15.2
7.6
4.4
5.2
8.5
2.6
90
0.47x
106
Rat
22.5
10.1
27.8
9.8
Absent
3.9
3.9
1.2
84.6
1.52
x
106
32P-labelled
amino
acid
in
hydrolysate
of
human
and
rat
proteins
Ser
Thr
Ser
Ser
Ser
and
unknown
Ser
Ser
Ser
partial
hydrolysates
were
examined
for
O-phospho-
serine
by
electrophoresis
at
pH3.5
and
radioauto-
graphy
(Table
1)
(Murray
&
Milstein,
1967).
Rat
myelin
contains
two
basic
proteins,
one
which
is
similar
to
the
human
protein
and
one
which
has,
when
compared
with
the
human
protein,
a
linear
deletion
of
40
amino
acids
in
the
C-terminal
half
of
the
molecule
(P.
R.
Dunkley
&
P.
R.
Carnegie,
unpublished
work).
The
smaller
protein
from
rats
was
phosphorylated
as
described
above
and
a
tryptic
digest
examined
(Table
1).
Most
of
the
spots
on
the
peptide
'map'
were
identical
with
those
found
in
the
digest
of
the
human
protein,
which
would
sug-
gest
that
phosphorylation
was
occurring
in
the
regions
of
invariant
amino
acid
sequence.
Identification
of
spots.
Spot
A
material
was
isolated
from
the
remainder
of
the
digest
of
the
human
protein.
It
had
the
composition
His
(0.9),
Gly
(0.9),
Ser
(1.0)
and
Lys
(1.2).
Histidine
was
shown
to
be
N-terminal
by
Edman
degradation
followed
by
amino
acid
analysis.
This
peptide
has
a
unique
composition
in
the
protein,
and
is
derived
from
peptide
T28,
His-Gly-Ser-Lys,
which
represents
amino
acid
residues
10
to
13
(Carnegie,
1971).
Spot
C
material
was
more
heavily
phosphorylated
in
the
rat
protein
than
in
the
human
protein.
From
its
yellow
colour
with
ninhydrin,
high
RF
value
and
amino
acid
composition
spot
C
material
was
identified
as
being
derived
from
Gly-Leu-Ser-Leu-
Ser-Arg.
Electrophoretic
analysis
of
a
thermolysin
digest
of
spot
C
material
showed
that
the
first
serine
residue
(i.e.
Leu-Ser-Leu)
rather
than
the
second
was
phosphorylated.
This
peptide
results
from
tryptic
Vol.
135
digestion
at
arginine-107
and
arginine-113
in
the
human
protein.
Spot
F
material
also
appears
to
be
derived
from
this
region
of
the
protein,
as
it
con-
tained
the
monomethylarginine
residue
(no.
197),
which
is
resistant
to
tryptic
digestion
(Baldwin
&
Carnegie,
1971).
The
myelin
basic
proteins
are
heterogeneous
with
regard
to
methylation
of
arginine-
107.
For
example
the
rat
protein
has
12
times
as
much
unmethylated
arginine
at
this
position
as
has
the
human
protein
(P.
R.
Dunkley
&
P.
R.
Carnegie,
unpublished
work).
It
was
surprising
to
find
no
trace
of
dimethylarginine
in
spot
F,
as
there
is
1.6
times
as
much
of
the
dimethyl
form
in
the
human
protein.
The
mono-
and
di-methyl
forms
would
be
expected
to
be
in
the
same
position
on
the
peptide
'map'.
Perhaps
the
extent
of
methylation
influences
the
degree
of
phosphorylation.
The
remaining
spots
were
present
in
too
low
a
yield
to
enable
definitive
identification
and
the
minor
spots
may
have
come
from
degradation
products
of
the
protein.
Spot
E
contained,
in
addition
to
a
small
amount
of
phosphoserine,
an
unidentified
acidic
phosphate
which
was
still
present
after
3
h
hydrolysis
with
acid.
Thus,
although
the
myelin
basic
proteins
have
a
high
content
of
serine
and
threonine,
only
a
few
resi-
dues
appear
to
be
phosphorylated
to
any
marked
extent
by
the
rabbit
muscle
protein
kinase.
There
is
no
obvious
similarity
in
amino
acid
sequence
in
the
regions
of
the
protein
containing
peptides
A
and
C.
This
would
suggest
that
either
the
kinase
recognizes
some
feature
of
ordered
structure
in
the
protein,
or
the
kinase
preparation
is
a
mixture
of
enzymes
with
571.
572
P.
R.
CARNEGIE,
B.
E.
KEMP,
P.
R.
DUNKLEY
AND
A.
W.
MURRAY
subtle
differences
in
their
substrate
specificities,
as
has
been
suggested
by
Bitensky
&
Gorman
(1973).
Johnson
et
al.
(1971)
found
that
myelin
had
some
endogenous
protein
kinase
activity
and
also
had
a
large
proportion
of
the
total
substrate
in
brain
for
a
cyclic
AMP-dependent
protein
kinase.
In
preliminary
experiments
we
have
found
that,
when
myelin
was
incubated
with
[y-32P]ATP,
the
endogenous
kinase
phosphorylated
the
basic
protein
but
not
the
proteo-
lipid,
which
is
the
major
protein
in
myelin.
Since
phosphorylation
of
cerebral
proteins
can
be
readily
demonstrated
in
vivo
(Rodnight,
1971)
it
will
be
important
to
assess
the
extent
of
phosphorylation
of
the
myelin
basic
protein,
in
different
physiological
and
behavioural
states.
We
thank
the
Australian
Research
Grants
Committee
and
the
National
Multiple
Sclerosis
Society,
New
York,
U.S.A.,
for
financial
support
and
Mr.
G.
Luxford
for
preparing
the
basic
proteins.
This
work
was
presented
in
part
at
the
4th
Meeting
of
the
International
Society
for
Neurochemistry,
August
1973.
Baldwin,
G.
S.
&
Carnegie,
P.
R.
(1971)
Science
171,
579-580
Bitensky,
M.
W.
&
Gorman,
R.
E.
(1973)
Progr.
Biophys.
Mol.
Biol.
26,
411461
Carnegie,
P.
R.
(1971)
Biochem.
J.
123,
57-67
Deibler,
G.
E.,
Martenson,
R.
E.
&
Kies,
M.
W.
(1972)
Prep.
Biochem.
2,
139-165
Dunkley,
P.
R.
&
Carnegie,
P.
R.
(1973)
in
Research
Methods
in
Neurochemistry
(Rodnight,
R.
&
Marks,
N.,
eds.),
Plenum
Press,
New
York,
in
the
press
Glynn,
I.
M.
&
Chappell,
J.
B.
(1964)
Biochem.
J.
90,
147-149
Johnson,
E.
M.,
Maeno,
H.
&
Greengard,
P.
(1971)
J.
Biol.
Chem.
246,
7731-7739
Kemp,
B.
E.,
Froscio,
M.
&
Murray,
A.
W.
(1973)
Biochem.
J.
131,
271-274
Krebs,
E.
G.
(1972)
Curr.
Top.
Cell.
Regul.
5,
99-133
Murray,
K.
&
Milstein,
C.
(1967)
Biochem.
J.
105,491
495
Rodnight,
R.
(1971)
Handb.
Neurochem.
5,141-163
Traugh,
J.
A.,
Mumby,
M.
&
Traut,
R.
R.
(1973)
Proc.
Nat.
Acad.
Sci.
U.S.
70,
373-376
Walsh,
D.
A.,
Perkins,
J.
P.
&
Krebs,
E.
G.
(1968)
J.
Biol.
Chem.
243,
3763-3774
Weller,
M.
&
Rodnight,
R.
(1973)
Biochem.
J.
132,
483-492
1973