Nudeotide sequence of the BsuRI restriction-modification system

Abstract

The genes of the 5′-GGCC specific BsuRI restriction-modification system of Bacillus subtills have been cloned and expressed in E. coll and their nucleotide sequence has been determined. The restriction and modification genes code for polypeptides with calculated
molecular weights of 66,314 and 49,642, respectively. Both enzymes are coded by the same DNA strand. The restriction gene
is upstream of the methylase gene and the coding regions are separated by 780 bp. Analysis of the RNA transcripts by S1-nuclease mapping indicates that the restriction and modification genes are transcribed from different promoters. Comparison
of the amino acid sequences revealed no homology between the BsuRI restriction and modification enzymes. There are, however, regions of homology between the BspRI SPR methylase and two other GGCC specific modification enzymes, the BspRI and SPR methylases.

Full text

Vlum
13
Nubr1
95NcecAisRsac
Nucleotide
sequence
of
the
BsuRI
restriction-modification
system
Antal
Kiss,
Gyorgy
Posfai+,
Christopher
C.Keller,
Pal
Venetianer+
and
Richard
J.Roberts
Cold
Spring
Harbor
Laboratory,
P.O.
Box
100,
Cold
Spring
Harbor,
NY
11724,
USA,
and
+Institute
of
Biochemistry,
Biological
Research
Center
of
the
Hungarian
Academy
of
Sciences,
P.O.
Box
521,
Szeged,
Hungary
Received
20
August
1985;
Accepted
27
August
1985
ABSTRACT
The
genes
of
the
5'-GGCC
specific
BsuRI
restriction-modification
system
of
Bacillus
subtilis
have
been
cloned
and
expressed
in
E.
coit
and
their
nucleotide
sequence
has
been
determined.
The
restriction
and
modification
genes
code
for
polypeptides
with
calculated
molecular
weights
of
66,314
and
49,642,
respectively.
Both
enzymes
are
coded
by
the
same
DNA
strand.
The
restriction
gene
is
upstream
of
the
methylase
gene
and
the
coding
regions
are
separated
by
780
bp.
Analysis
of
the
RNA
transcripts
by
S1-nuclease
mapping
indicates
that
the
restriction
and
modification
genes
are
transcribed
from
different
promoters.
Comparison
of
the
amino
acid
sequences
revealed
no
homology
between
the
BsuRI
restriction
and
modification
enzymes.
There
are,
however,
regions
of
homology
between
the
BsuRI
methylase
and
two
other
GGCC
specific
modification
enzymes,
the
BspRI
and
SPR
methylases.
INTRODUCTION
Type
II
restriction
and
modification
enzymes
are
promising
as
model
systems
for
the
study
of
sequence-specific
DNA-protein
interactions.
The
development
of
molecular
cloning,
DNA
sequencing
and
In
vitro
mutagenesis
techniques,
together
with
more
classical
biochemical
methods,
provide
the
tools
that
are
necessary
to
study
the
detailed
molecular
mechanism
of
sequence-specific
DNA
recognition.
One
of
the
attractions
of
the
restriction
and
modification
enzymes
is
that
for
most
recognition
sequences
different
enzymes
recognizing
the
same
sequence
are
available.
Comparison
of
these
enzymes
may
help
to
elucidate
some
general
rules
for
DNA
sequence
recognition.
Although
a
number
of
type
II
restriction
and
modification
genes
have
been
cloned,
only
the
EcoRI
(1,2),
HhaII
(3),
PstI
(4)
and
EcoRV
(5)
systems
have
been
sequenced.
In
addition
to
this,
the
nucleotide
sequences
of
the
E.
coit
dam
(6),
BspRI
(7),
SPR
(8,9)
and
T4
dam
(10)
methylases
have
also
been
determined.
The
restriction-modification
system
BsuRI
of
B.
subtilis
R
was
discovered
and
genetically
mapped
on
the
bacterial
chromosome
by
T.
Trautner
and
coworkers
(11).
The
BsuRI
nuclease
(12)
cleaves
the
recognition
sequence
©
I
RL
Press
Limited,
Oxford,
England.
Nucleic
Acids
Research
Volume
13
Number
18
1985
6403

Nucleic
Acids
Research
5'-GGCC
between
the
central
G
and
C
(13)
and
the
modification
enzyme
methylates
the
inner
C
(14).
The
BsuRI
restriction
and
modification
enzymes
have
been
purified
to
homogeneity
and
characterized
biochemically
(15-17).
We
reported
the
cloning
and
expression
in
E.
coli
of
the
gene
that
codes
for
the
BsuRI
methylase
(18).
Here
we
report
the
cloning
of
the
BsuRI
endonuclease
gene
and
the
nucleotide
sequence
of
the
complete
BsuRI
system.
MATERIAL-S
AND
METHTODS
Strains
and
media
Bacillus
subtilis
R
(11)
was
provided
by
Dr.
T.
Trautner.
The
E.
coli
strains
HB101
(19)
and
RR1
(20)
were
used
as
hosts
in
plasmid
cloning
experiments
and
JM107
(21)
was
used
as
host
for
cloning
in
phage
M13.
B.
subtilis
and
E.
coli
were
grown
in
LB
medium
(22),
at
37
0C.
Enzymes
and
chemicals
Restriction
endonucleases
were
either
prepared
in
the
Biol.
Res.
Cent.
of
the
Hung.
Acad.
Sci.
according
to
published
protocols
(23)
or
were
purchased
from
New
England
Biolabs.
DNA
polymerase
I
large
(Klenow)
fragment
was
from
BRL,
SI
nuclease
from
Boeringer
Mannheim
and
polynucleotide
kinase
from
New
England
Biolabs.
Synthetic
oligonucleotides
were
prepared
by
M.
Zoller
using
an
Applied
Biosystem
Synthesizer.
All
other
chemicals
were
reagent
grade
commercial
products.
Clnning
methods
Preparation
of
plasmids,
transformation
of
E.
coli,
restriction
mapping,
agarose
gel
electrophoresis
and
subcloning
of
DNA
fragments
were
done
by
standard
procedures
(22).
Cloning
of
the
BsuRT
endonuclease
gene
DNA
purified
from
B.
subtlUs
R
(18)
was
digested
with
SphI
and
SaII
and
ligated
to
pBR322
(20)
cleaved
with
the
same
enzymes.
The
ligated
DNA
was
transformed
in
E.
coit
RR1.
AmpR
transformants
(approximately
8,000
Tets
recombinants)
were
growm
to
saturation
in
200
ml
LB
containing
100
pg/ml
ampicillin
and
the
cells
were
used
for
preparation
of
plasmid
DNA.
This
plasmid
DNA
was
digested
with
HaeIII
and
transformed
in
E.
colt
RR1.
AmpR
transformants
were
selected.
HaeIII
(an
isoschizomer
of
BsuRI)
could
be
used
to
select
for
the
BsuRI
methylase
gene
because
the
BsuRI-specific
methylation
protects
the
DNA
against
HaeIII
cleavage
(14).
Detection
of
BsuRI
endonuclease
activity
in
the
clones
Cells
from
100
ml
saturated
cultures
of
the
E.
coit
clones
carrying
the
cloned
BsuRI
genes
were
sedimented
by
centrifugation,
washed
in
20
mM
6404

Nucleic
Acids
Research
TrisHCl
pH
8.0,
resuspended
in
1
ml
of
50
mM
TrisHCl
pH
8.0,
5
mM
2-mercaptoethanol,
0.1
mM
EDTA
and
disrupted
by
sonication.
After
adding
NaCl
to
IM,
the
homogenate
was
centrifuged
at
15,000
g
for
1
hr
at
4
0C.
The
supernatant
was
purified
further
by
gel
filtration
on
a
2.5
x
43
cm
Bio-Gel
A-0.Sm
column
in
10
mM
TrisHCl
pH
8.0,
10
mM
2-mercaptoethanol
and
1
M
NaCl.
1.5
/A
aliquots
of the
fractions
were
assayed
for
BsuRI
endonuclease
activity
in
25
pA
reaction
mixtures
containing
1
g
pBR322
DNA,
10
mM
TrisHCl
pH7.5,
10
mM
MgCl2,
50
mM
NaCl
and
1
mM
dithiotreitol.
The
samples
were
incubated
for
1
hour
at
37
°C
and
analyzed
on
1%
agarose
gels.
Measurement
of
in
vivo
restriction
Nonmodified
and
modified
Xvir
and
Sk80c
phages
were
prepared
by
growing
the
phage
on
either
E.
coil
HB11
(pBR322)
or
E.
coil
HB101
carrying
the
plasmid
pSUl
which
codes
for
the
BsuRI
methylase.
In
vivo
restriction
was
measured
by
determining
the
restriction
ratio
i.e.
the
ratio
between
the
titers
of
a
nonmodified
phage
on
the
host
investigated
and
on
a
nonrestricting
host.
The
plating
efficiency
was
determined
by
standard
methods
(24).
Determination
of
the
nucleotide
sequence
Part
of
the
DNA
sequence
(nucleotides
1-2904)
was
determined
by
the
chain
termination
method
(25)
using
M13
clones
as
templates, the
other
part
(nucleotides
2616-4253)
by
the
chemical
cleavage
method
(26).
Chain
termination
method:
Specific
fragments
were
cloned
in
the
M13
vectors
mpl8
and
mpl9
(27)
and
sequenced
(28)
using
the
synthetic
oligonucleotide
5'-GTAAAACGACGGCCAGT
as
universal
primer
and
35S-a-dATP
as
label
(29).
The
sequencing
products
were
run
on
40
x
20
x
0.04
cm
6
or
8%
polyacrylamide
gels
containing
8
M
urea
in
100
mM
Tris-borate
pH
8.3.
After
the
run
the
gels
were
fixed
in
10%
methanol,
10%
acetic
acid
for
10
min,
dried
and
exposed
to
autoradiographic
film
at
room
temperature
for
1
-
2
days.
Sequence
data
from
the
autoradiograms
were
entered
directly
into
the
computer
using
a
digitizing
tablet
(30).
Chemical
cleavage
method:
Labeling
of
restriction
fragments
by
polynucleotide
kinase,
sequencing
reactions
and
gel
electrophoresis
were
carried
out
as
described
in
(26).
Mapping
of
the
transcripts
Preparation
of
single-stranded
probes:
Two
synthetic
oligonucleotides,
An3
and
An4,
complementary
to
nucleotides
314-334
and
2837-2857
(Fig.3.)
were
used.
Following
synthesis,
oligonucleotides
were
fractionated
on
a
16%
polyacrylamide
gel
containing
7
M
urea
and,
after
elution,
were
purified
6405

Nucleic
Acids
Research
further
on
a
small
Sep-Pak
column
(31).
0.2
pmol
of
An3
and
An4
labeled
with
32P
at
the
5'-end
were
annealed
in
10
p1
containing
10
mM
TrisHCl
pH
7.5,
10
mM
MgCl2
to
approximately
1
ug
of
the
single-stranded
M13
templates
20/1
and
9/6,
respectively.
The
M13
clone
20/1
contains
the
494
bp
EcoRI-BglII
fragment
covering
the
5'-end
of
the
restriction
gene
(nucleotides
1-494)
and
clone
9/6
contains
the
434
bp
EcoRI-BglII
fragment
covering
the
5'-end
of
the
modification
gene
(nucleotides
2466-2899),
both
cloned
between
the
EcoRI
and
BamHI
sites
in
the
vector
mpl8.
After
annealing,
7
p1
of
10
mM
TrisHCl
pH
7.9
and
6
p1
of
a
solution
containing
0.5
mM
of
the
four
deoxynucleotide
triphosphates
were
added
and
the
primers
were
extended
with
1
p1
(3
U)
of
DNA
polymerase
Klenow
fragment
for
20
min
at
room
temperature.
The
reaction
was
stopped
by
the
addition
of
40
mM
EDTA
pH
8.0,
the
DNA
was
extracted
with
phenol/chloroform,
twice
with
chloroform
and
precipitated
with
ethanol.
The
precipitated
DNA
was
dissolved
and
digested
with
EcoRI,
then
extracted
with
phenol
and
chloroform
and
precipitated
with
ethanol.
The
terminally
labeled
329
nucleotide
(probe
R)
and
387
nucleotide
(probe
M)
long
single-stranded
fragments
(Fig.
3.)
were
isolated
(26)
from
a
6%
polyacrylamide
gel
containing
8
M
urea.
S1
maDping.Q
10-14
jug
total
bacterial
RNA
isolated
(32)
from
B.
subtilis
R
or
E.
colt
HB101
carrying
pSU13
was
hybridized
to
probes
R
or
M
(3-10
x
103
cpm)
in
50%
formamide,
4
x
SSC
at
37
°C
for
14
hrs
then
the
hybridization
mixture
(15
p1)
was
added
to
300
p1
SI
buffer
(0.045
M
Na-acetate
pH
4.6,
0.28
M
NaCl,
0.005
M
ZnSO
4)
and
digested
with
5000
U
(Boeringer)
of
S1
nuclease
for
1
hour
at
30
0C.
After
adding
10
p1
of
0.5
M
EDTA
pH
8.0
and
10
pug
tRNA
the
reaction
mixtures
were
extracted
with
phenol
and
chloroform,
then
the
S
-resistant
hybrids
were
precipitated
with
ethanol.
The
fragments
protected
from
S1-digestion
were
analyzed
by
running
them
side-by-side,
on
6%
polyacrylamide/8
M
urea
gels,
with
sequencing
products
obtained
from
sequencing
reactions
on
templates
20/1
and
9/6
using
the
kinase-labeled
An3
and
An4
oligonucleotides
as
sequencing
primers.
Positions
of
the
S
-resistant
fragments
determined
the
transcription
initiation
sites.
Computer
analysis
of
the
sequence
Handling
and
analysis
of
nucleic
acid
sequence
data
were
done
with
computer
programs
developed
at
the
Cold
Spring
Harbor
Laboratory
(30)
and
at
the
Biological
Research
Center
of
the
Hungarian
Academy
of
Sciences
(J.
6406

Nucleic
Acids
Research
Posfai,
unpubl.).
Protein
sequences
were
compared
using
the
programs
DIAGON
(33)
or
SEQHP
(34).
RESUT-TS
Cloning
of
the
R.BsuRI
gene
Clones
pSUl
and
pSUll,
two
overlapping
clones
isolated
independently,
cover
an
approximately
4.7
kb
region
of
B.
subtills
DNA
(18).
Both
clones
express
the
BsuRI
methylase
but
they
do
not
code
for
the
BsuRI
endonuclease
(18).
We
knew
from
genetic
data
(11)
that
the
BsuRI
genes
were
linked.
Therefore,
it
seemed
possible
that
the
endonuclease
and
methylase
genes
could
be
cloned
together
on
a
larger
fragment.
The
1.3
kb
HindIII-BgiII
fragment
containing
the
5'-end
of
the
methylase
gene
and
the
5'-flanking
sequences
(Fig.
1)
was
isolated
from
pSUl
and
used
as
hybridization
probe
to
identify,
by
Southern
blotting,
larger
fragments
of
the
B.
subtills
chromosomal
DNA
which
would
carry
the
methylase
gene
and
would
extend
into
the
neighbouring
regions
(data
not
showm).
One
such
fragment
was
a
9.5
kb
SphI-SalI
fragment.
Cloning
in
E.
coil
RR1
of
this
fragment
was
carried
out
using
selection
for
the
methylase
as
described
in
Materials
and
Methods.
The
resulting
clone
(pSU12)
carried
the
9.5
kb
SphI-SalI
fragment.
DNA
isolated
from
this
clone
was
resistant
to
HaeIII
showing
that
it
carried
and
expressed
the
BsuRI
methylase
gene.
(BsuRI
methylation
protects
the
DNA
against
HaeIII
cleavage;
ref.
14).
To
test
whether
the
plasmid
also
coded
for
the
restriction
enzyme,
we
prepared
a
cell-free
extract
from
the
clone
and
fractionated
it
on
a
Bio-Gel
A-O.5m
column.
Some
of
the
fractions
showed
BsuRI
activity indicating
the
presence
and
expression
of
the
BsuRI
endonuclease
gene
(Fig.
2).
Colonies
of
the
clone
RRI(pSU12)
were
slightly
heterogeneous
in
size
on
LB
agar
plate,
therefore
the
plasmid
pSU12
from
the
original
isolate
was
transformed
in
E.
colt
HB101.
HB1O1(pSU12)
always
gave
homogeneous
cultures.
The
observed
difference
between
the
RR1
and
HB101
clones
was
not
further
investigated.
In
the
rest
of
the
work
HB101
was
used
as
host.
To
test
whether
the
clone
showed
in
vivo
restriction,
the
phages
vir
and
5k80c
grown
either
on
the
nonmodifying
host
E.
colt
HB1O1(pBR322)
or
on
the
modifying
host
E.
coli
HB1O1(pSUl)
were
used
to
infect
the
E.
colt
clone
carrying
pSU12.
The
modified
phage
plated
with
an
e.o.p.
of
1,
whereas
the
non-modified
phage
was
restricted
(e.o.p.:
10
-
10
,
data
not
shown).
It
was
known
from
previous
data
(18)
that
the
methylase
gene
was
close
to
the
Sall
site
of
the
insert
in
pSU12
(Fig.
1).
To
localize
the
restriction
6407

Nucleic
Acids Research
A.
-I
0
2
3
4
5
6
7
(kb)
Hl
L
H
M H
H
IMM
Z
0
C0
0
L
J
a
l
Z
R
M
l
pSU13(r+
m+)
[Xho
I]
+
lxho
i]
pSUI9
(r
m+)
[Xho
I]
S
I
pSU17(r-
mi
)
I
pSUII(r
-m+)
pSUI(r
m+)
I
~~~~~~~~~pSU15(r
m+)
l
pSU127(r
m)
-l
lpSU128(r
m)
B.
Bam
HI
R
M
pSU13
SalI
|
Bam
HI
Nru
I
pBR
322
Fig.l.
Schematic
map
of
the
DNA
region
coding
for
the
BsuRI
enzymes.
A)
The
sequenced
region
is
indicated
by
the
thick
horizontal
bar
and
the
location
and
orientation
of
the
BsuRI
genes
by
thick
arrows.
Thin
horizontal
bars
represent
fragments
cloned
in
different
recombinant
plasmids.
Symbols
r
m+,
r
m
etc.
refer
to
the
restriction-modification
phenotype
of
the
clones.
Only
selected
restriction
sites
are
shown.
FXhoI:
XhoI
site
produced
with
XhoI
linker.
B)
Circular
map
of
pSUl3.
Designations
are
as
on
panel
A.
The
vector
is
indicated
by
the
dotted
segment.
gene,
shorter
subfragments
(Fig.
1)
of
the
original
9.5
kb
SphI-SalI
insert
were
cloned
in
pBR322
and
tested
for
restriction.
The
shortest
derivative
still
showing
in
vivo
restriction
was
pSUl9
(Fig.
1).
Nucleotide
sequence
of
the
gtenes
We
determined
the
sequence
of
4253
nucleotides
starting
at
the
nearest
EcoRI
site
upstream
of
the
endonuclease
gene
(Fig.
1
and
3).
The
sequence
contains
only
two
large
open
reading
frames,
both
on
the
same
strand:
the
first
(nucleotides
297-2027)
contains
576
amino
acids,
while
the
second
(2808-4118)
contains
436
amino
acids
(Fig.
3).
On
the
basis
of
the
restriction-modification
phenotype
of
the
clones
carrying
different
portions
of
6408

Nucleic
Acids
Research
A.
Bio-Gel
fractions
H
N
r4l)
I;j_
d)
w
rl-
OD
0)0-
N_
P-O't
0
(10
r~-
OD
)
a3I
B.
M
-S_
cx
C
C3
I:-
C
-
r'O
IQ_
0
m~ci
Fig.
2.
Endonuclease
assay
of
fractions
obtained
from
Bio-Gel
chromatography
of
a
cell-free
extract
prepared
from
the
B.
colt
clone
carrying
pSU12.
Panel
A:
Bio-Gel
fractions.
Panel
B:
Time
course
of
digestion.
pBR322
DNA
was
digested
for
different
lengths
of
time
with
1
gi
of
fraction
10
shown
in
panel
A.
Lanes
HaeIII:
pBR322
DNA
digested
with
HaeIII.
the
cloned
B.
subtlits
DNA
(Fig.
1),
the
first
reading
frame
was
assigned
to
the
endonuclease
and
the
second
to
the
methylase.
The
two
reading
frames
are
separated
by
780
bp.
This
intergenic
region
is
longer
than
any
other
so
far
found
in
type
II
restriction-modification
systems
(1-5).
The
endonuclease
gene
codes
for
a
66,314
dalton
protein
and
the
methylase
gene
for
a
49,642
dalton
protein.
The
sequencing
led
to
the
unexpected
observation
that
the
BsuRI
methylase
coded
by
pSUl
and
by
its
derivative
pSU1S
(18,
Fig.
1)
is
not
the
complete
protein.
These
plasmids
lack
the
sequence
downstream
of
the
HlndIII
site
at
position
4077
(Fig.
3),
and
which
contains
the
extreme
3'-end
of
the
6409

Nucleic
Acids
Research
GAATTCCCTTTATTCCCGCCTTCATAAATTGCATCTATACGGAGAGGGGCATTTTCTAAACTTTCAAATGGGTAAAT
77
TGAATTAAGTGTGTTTGTTTCTTTCAACTTCTGAACTCTCCTTCAGTTAGCATGTTAACATTTACAATTTATATT
152
ATAAGCTTTTACGAGAACCACTTGGAAGCTTTGGTTTTTAGGAAATCATAACTTATTTAGATTACTTGTGG
ACT
227
vv
~~~~~~~~~~~M
0
ATTTACTTAGTTATCTTTTCCACTTTTTCAAGAAAACATAAGAGAAATAGAAAACAGAGGGTGTATACAATGGGG
302
K
N
S
K
A
I
G
N
N
H
V
K
S
V
Y
Q
A
L
L
Q
S
L
K
S
K
AAAAATTCAAAGGCTATCGGTAACAATCATGTTAAATCAGTATACCAAGCATTATTGCAATCTTTAAAATCAAAA
377
S
V
N
G
F
S
K
I
T
I
E
T
I
S
F
I
K
N
L
Y
P
E
I
D
S
TCTGTAAACGGTTTTTCTAAGATAACAATAGAGACAATATCATTTATTAAAAATTTATATCCAGAAATTGATTCT
452
V
T
S
K
F
D
N
S
R P
D
Q
S
K
D
L
T
L
Y
L
K
S
G
E
T
GTAACTTCAAAGTTCGATAATTCACGTCCTGATCAATCAAAAGATCTTACTTTATATTTGAAAAGTGGCGAAACT
527
I
S
L
N
L
F
L
I
K
K
G
R
R
I
Q
P
K
N
A
G
A
K
S
F
L
ATCTCGTTGAATCTATTTCTGATTAAAAAAGGCCGACGCATTCAGCCTAAAAATGCTGGCGCGAAGAGTTTTTTA
602
E
K
Y
F
L
S
A
E
M
Q
K
I
F
N
K
E
F
E R
Y
Y
L
D
Y
L
GAAAAGTACTTT
TTATCAGCAGAAATGCAAAAGATTTTTAACAAAGAATTTGAAAGGTACTATTTAGACTACTTA
677
K
E
V V
E
H
K K
G
T
H Y
I
T
D
K
R
E
L
K
R
L
V
S
S
AAAGAAGTAGTGGAGCATAAAAAAGGAACACACTACATAACAGATAAAAGAGAGTTAAAAAGACTTGTGTCAAGC
752
H
F
P
K
F T
E E
I
N
L
Y
R D
KF
L
F
N
L
R
E
T
C
F
CATTTTCCAAAATTCACAGAAGAAATTAATTTATATAGAGACAAGTTTCTTTTCAACTTGCGTGAAACTTGTTTT
827
T
L
L
Q Q
F
Y
N
E
K
N
I
G
F
T
H
A
F
N
V
F
F
M
V
N
ACGTTGTTGCAGCAATTTTATAATGAGAAGAATATAGGATTTACCCATGCCTTTAATGTCTTCTTCATGGTCAAT
902
D
T
N
I
I
T
S
Y G
K
D
E
N
D
V
K
V
E
K
F
A
P
A
S
P
GATACAAATATCATTACAAGTTACGGCAAAGATGAAAATGATGTCAAAGTTGAAAAGTTTGCACCTGCATCCCCA
977
S
L
K
D
I
E
L
Y
K
T
G
K
S
T
VG
I
K
F
G
E
V
G
L
T
TCTTTGAAGGATATTGAACTTTACAAAACAGGAAAGAGTACAGTTGGGATTAAATTTGGAGAAGTGGGACTCACC
1052
L
R
F
K
F
E
S
D
P
W
K
S
I
K
L
A
T
G
Y H
E
F
P
K
E
CTAAGATTTAAATTTGAAAGTGATCCTTGGAAATCGATTAAACTTGCTACAGGTTACCATGAATTCCCTAAAGAA
1127
KE
R
V
N
V
N
L
K
T
M
R
R
M
E
K
L
L
N
K
H
E
Y
A
K
AAAGAGAGAGTGAATGTCAACTTAAAAACAATGAGGAGAATGGAAAAACTATTGAATAAACATGAGTACGCTAAA
1202
T
S
N
N S
N
A
I
G
K
C
H
E
A
W
T
Y
Y
Y
F
L
K
A
F
P
ACATCGAATAACAGTAACGCAATAGGCAAATGCCATGAAGCATGGACATATTATTATTTCTTAAAGGCGTTTCCC
1277
D
V
I
0
V
D
P
K
Q
C
V
E
L
I
N
T
Y
F S
S
I
N
Q
N
T
GACGTTATACAGGTAGATCCAAAGCAATGTGTTGAATTAATTAACACATATTTTTCTAGTATAAATCAAAATACA
1352
L
K
K
L
Y
S
S
T
S
T
I
V
D
A
I
T
E
K
L
R
Q
K
Y
H
D
CTAAAAAAACTATATAGCTCAACTTCTACAATAGTAGATGCTATTACAGAGAAACTAAGACAAAAATATCATGAC
1427
Y
I
I
E
S
I
E
L
I
P
D
A
Y
I
K
D
R
L
D T
G
D
L
I
L
TATATTATAGAAAGTATCGAATTAATCCCGGATGCATATATAAAAGATAGGCTTGATACAGGAGATCTTCAGTTA
1502
V
L
K
V
NH
N
I
I
V
E
N
I
S
L
K
A
L
A
K
R
N
S
K
I
GTCTTAAAAGTAAATAACAATATTATTGTTGAGAATATTTCTCTAAAAGCTTTAGCAAAAAGGAATAGCAAAATT
1577
T
T
K
N
P
G
0
G
S
I
L
G
P
T
Y
F
N
M
G
S
M
E
S
V
I
ACTACAAAGAATCCGGGTATGGGAAGCATTCTTGGACCAACATATTTTAATATGGGAAGCATGGAATCTGTTATT
1652
N
E
V
K
N
KF
T
I
G
E
F
N
H
R
K
S
LE
I
L
S
Y
E
F
AATGAAGTGAAAAACAAGTTTACCATAGGGGAATTTAATCATAGAAAAAGTTTAGAAATACTCTCTTATGAGTTC
1727
G
M
K
L
D
S
A
T
Q
E
Q
L
R
R
G
I
H
N
L
L
G
K
A
M
I
GGAATGAAACTTGACAGTGCAACTCAGGAACAATTAAGAAGAGGAATTCACAATTTATTAGGAAAAGCAATGATA
1802
A
I
T
I
Y
G
E
G
I
S
F
C
K
E
P
S
E
ID
G
E
V
K
V
H
GCTATTACTATTTATGGTGAGGGAATTAGCTTTTGCAAAGAGCCTTCTGAAATTGACGGTGAAGTAAAAGTGCAT
1877
V
N
V
P
S
A
I
Q
N
T
L
T
W
N
N
E
L
E
S
I
S
L
R
A
K
GTTAATGTGCCTTCTGCTATACAAAATACCTTAACATGGAATAATGAGTTAGAGTCAATTAGTTTACGTGCAAAA
1952
F
S
K
S
Q
K
H
G
W
S
S
I
K
L
T
S
E
C
Q
L
E
S
R
K
*
TTTAGTAAAAGTCAAAAGCACGGCTGGTCTTCTATTAAATTAACATCAGAATGCCAGTTGGAAAGCAGGAAATAA
2027
ACAATGTCCAATGATTATATGCCAATAACCTCTTTTGTTTTTTCTATCTTTTATGAGGGAATATTATAAATGATT
2102
CCAAAGAAAGGTTGGGGCAAATGGACAAGATGTCAAAAGAGTCTAGAAGCAATGTTATGAAATCCATTAAATCA
2177
GTTTCCCAATTAGAAAATCTGGTTGCAAGTGCATTGTGGAATCGGGGCTATCGGTTCCGAAGAAACACCAAAAGC
2252
CTTTTTGGGAAACCAGATTTGTCTATAAAAAAGTATAAGGTGGTCATTTTTATCGATTCTTGTTTCTGGCATTTC
2327
6410

Nucleic
Acids
Research
TGCCCTGTTCACGGCAGGATCCCTAAAAGCAATACAGATTATTGGAATGCTAAATATATAAAAAATAAAACCCGA
2402
GATGAAGAAGTCAACACATTTTATAGGGAGAATAATTGGAATATACTTCGTGTATGGGAACATGAATTCAAGGAA
2477
GATTTTGATTTTGCCATTGACACGATAGCTAATTTTATTGAACAATCTAAGAGAAAATAAAGAGCAAAATTAGAT
2552
TAAAAAGAACTTTCTTGTGGAATGTTTTGAACAAATCCATCTTATATTATGACCTCGCTTTCCTCCATTAAATCA
2627
ATGGCTAGCAATGTTTCTAATTTAATTCAATTTCCCAATTAACAGAATAACACAAAATAACCATTAAATAATAAT
2702
GTAAACATATAAGGAAAATCATTATATTAAGGACACTGGCGCTGGCCTTTTGAATTGAAATTTGA
TCA
2777
v
M
T
L
K
I
D
I
K
G
R
GK
Y
K P
TAAGAGGTGCAACAGGAGGTTGTTTAAAATATGACTTTAAAAATTGATATCAAAGGTAGAGGCAAATATAAGCCG
2852
A
S
D
Y
S
I
D
D
V
K
N
V
L
M
E
K
I
F
E
E
S
S
R
II
GCGTCTGATTATTCTATAGATGATGTAAAGAATGTACTAATGGAGAAGATCTTTGAGGAATCTTCAAGAATTATT
2927
N
S
D
D
D
L
E
I
I
E
K
V
D
F
R
T
D
K
I
N
V
L
S
L
F
AATTCTGATGATGATCTAGAAATCATTGAAAAGGTTGATTTTCGCACTGATAAAATAAATGTGCTTAGCCTGTTT
3002
S
G C
G
G
L
D
L
G
F
E
L
A
G
L
A
A
V
I
G
E
0
A
A
M
TCGGGCTGCGGCGGACTCGACCTTGGATTTGAATTAGCAGGATTGGCTGCAGTAATCGGTGAACAAGCAGCTATG
3077
E
A
F
K
D
K
D
R
F
N
E
L
R
N
K
S
I
F
H
T
I
Y
T
N
D
GAAGCATTTAAAGATAAAGACCGATTTAATGAACTAAGGAATAAAAGCATCTTCCATACCATATATACAAATGAC
3152
L
F
K
E
A
N
O
T
Y
K
T
N
F
P
G
H
V
I
O
H
E
K
D
I
R
TTGTTTAAAGAAGCCAATCAAACATACAAAACAAATTTTCCAGGTCATGTTATACAGCACGAAAAGGATATAAGA
3227
Q
V
K
Y
F
P K
C
N
L
I
L
G G
F
P
C
P
G
F
S
E
A
G
P
CAAGTTAAATATTTCCCAAAATGCAACCTTATCCTTGGAGGATTCCCCTGCCCTGGATTTAGTGAAGCTGGCCCA
3302
R
L
I
D
D
D
R
N
F
L
Y
L
H
F
I
R
S
L
I
O
A
Q
P
E
I
CGTTTGATAGATGACGATCGTAACTTCTTATATTTACATTTTATTAGAAGCCTGATACAAGCACAGCCAGAAATA
3377
F
V A
E
N
V
K
G
U
M
T
L
G
K
G
E
V
L
N
Q
I
I
E
D
F
TTTGTAGCAGAGAATGTTAAAGGCATGATGACTCTTGGAAAAGGCGAGGTTTTAAATCAGATTATCGAAGATTTT
3452
A
S
A
G
Y
R
V
O
F
K
L
L
N
A
R
D
Y
G
V
P
Q
L
R
E R
GCTTCAGCTGGCTATAGGGTTCAGTTTAAGCTATTAAATGCAAGGGACTATGGAGTTCCGCAACTCAGAGAGCGT
3527
V
I
I
E
G
V
R
K
D
I
S
F
N
Y
K
Y
P
S P
T
H
G
E
E
T
GTTATAATTGAAGGCGTCAGAAAAGACATTAGCTTCAATTACAAATATCCATCTCCAACCCATGGTGAAGAAACG
3602
G
L
K
P
F
K
T
L
R
D
S
I
G
D
L
V
T
D
P
G
P
Y
F T
G
GGCCTAAAGCCATTCAAAACGCTGAGAGATTCTATAGGAGATTTAGTGACTGATCCAGGACCGTACTTTACGGGG
3677
S
Y
S
S
I
Y
M
S R
N
R
K
K
S
W
D
E
O
S
F
T
I
G
A
S
TCATATTCTTCTATTTATATGTCTCGTAACAGAAAGAAAAGCTGGGACGAGCAAAGCTTTACCATCCAAGCTTCA
3752
G
R
Q
A
P
L
H
P
G
G
L
S
M
K
K
I
G
K
D
K
W
V
F
P
D
GGGAGACAGGCCCCCCTCCATCCAGGTGGCTTATCTATGAAAAAAATAGGAAAAGATAAGTGGGTTTTCCCTGAT
3827
G
E
E
NH
R
R
L
S
V
K
E
I
A
R
V
Q
T
F
P
D
W
F
O
F
GGGGAAGAAAACCATAGAAGGTTGTCTGTAAAGGAAATTGCCAGGGTGCAAACTTTCCCAGATTGGTTTCAATTT
3902
S
Q
G
T
N
S
Q
T
S
I
N
N R
L
D
K
Q
Y
K
0
1
G
N
A
V
AGTCAAGGAACAAACAGCCAGACTTCAATTAACAATAGACTAGACAAACAATACAAGCAAATAGGGAATGCTGTG
3977
P
V
L
L
A
K
A
V A
S
P
I
A
N
W
A
I
N Y
L
E
S
SPN
CCAGTTTTGCTGGCTAAGGCAGTTGCTTCTCCTATTGCAAATTGGGCAATAAATTATCTCGAAAGCTCTCCAAAT
4052
N K
I
K
N
R
E
R
K
L
S
I
R
T
F
L
R
I
K
T
S
*
AATAAAATAAAGAACAGAGAACGCAAGCTTTCAATTAGGACTTTTTTAAGAATCAAAACCAGTTAAATTGAATGC
4127
TTTCCTTTAATTAACTAGCCATCCCTAGCAAAATAAGATGGCTAGTTTTTTTTGTAAAGCTAGCTTTTGACAAGA
4202
4
-
AGGAGAACATACTTAAATATGTTTGCCAAAAAGAAGGATGGGCTTTATACT
4253
Fig.
3.
Nucleotide
sequence
of
the
BsuRI
genes
and
the
deduced
amino
acid
sequence
of
the
proteins.
Shine-Dalgarno
sequences
are
indicated
by
closed
circles
under
the
sequence,
transcriptional
initiation
sites
by
arrow-heads
and
the
self-complementary
sequence
of
the
putative
transcriptional
termination
site
by
horizontal
arrows.
The
-10
sequences
of
the
promoters
are
boxed.
Regions
complementary
to
probes
R
and
M
used
in
the
transcription
mapping
are
underlined
with
a
solid
line
(corresponding
to
oligonucleotide
An3
and
An4)
or
with
a
dashed
line
(corresponding
to
the
part
of
the
probes
made
by
primer
extension).
6411

_
Ut
S
_
a
S
a
..
a
_ra
S
a
a-£
_
S
S
S
a.
a^
S
S
S
U
a
am
am
e
4
....
_f
_
Oa
_S
_
_
_
Ce
4
s
2m
IN
U
-
a0
4S
a
a
Sm
0
Fig.
4.
S
-nuclease
mapping
of
the
BsuRI
transcripts.
Gel
electrophoresis
of
fragments
o1
the
restriction
gene
specific
(R)
and
modification
gene
specific
(M)
probes
protected
from
S
digestion
by
RNA
isolated
from
B.
cold
(pSU13)
or
B.
subtUis
R.
20/1
and
9/a:
sequencing
products
obtained
using
template
20/1
and
9/6.
methylase
gene.
In
these
constructs
the
pBR322
via
this
HindIII
site,
in
the
Examination
of
the
sequence
around
the
truncated
methylase
gene
is
linked
to
same
orientation
as
the
TetR
gene.
junction
shows
that
translation
of
the
Nucleic
Acids
Research
6412
-VW
4z
4m
4.
..W
4m
-
-
46-
qw

Nucleic
Acids
Research
methylase
ends
at
an
in-phase
UAA
codon
immediately
after
the
HindIII
site:
HindIII
stop
methylase
......
GAA
CGC
AAG
CTT
TAA
TGC
......
pBR322
Thus
both
pSUl
and
pSU15
code
for
a
functional
BsuRI
methylase
which
lacks
the
11
amino
acids
from
the
C-terminus
of
the
wild-type
protein.
Transcription
of
the
genes
As
the
two
genes
are
in
the
same
orientation,
in
principle
they
could
be
transcribed
as a
single
mRNA,
starting
from
the
endonuclease
promoter.
However,
this
seemed
unlikely
since
the
methylase
gene
was
active
in
several
recombinant
plasmids
where
the
endonuclease
gene
was
not
present
(Fig.
1.).
The
transcriptional
initiation
points
were
determined
by
S1
mapping
using
RNA
isolated
from
B.
subtilis
and
E.
coi
HB1O1(pSU13)
cells.
Fig.
4.
shows
that
there
are
separate
initiation
sites
for
the
two
genes.
The
transcription
of
the
endonuclease
starts
with
U
and
A
at
positions
235
and
236,
respectively,
and
the
methylase
transcript
starts
with
A
at
2779.
It
can
also
be
seen
that
the
major
initiation
sites
are
identical
in
B.
subtilis
and
E.
coll.
The
sequences
preceding
the
initiation
sites
were
searched
for
promoter-like
structures.
B.
subtilis
is
known
to
have
several
different
forms
of
RNA
polymerase
which
can
use
different
promoter
sequences
(35).
The
consensus
sequence
of
the
promoters
recognized
by
the
major
form
of
RNA
polymerase
55
present
in
vegetative
cells
(y
)
is
identical
with
the
consensus
sequence
for
E.
coli
promoters:
TTGACA
for
the
-35
and
TATAAT
for
the
-10
region
(36,37).
Examination
of
the
sequence
preceding
the
initiation
sites
of
the
BsuRI
nuclease
and
methylase
transcripts
reveals
the
presence
of
hexanucleotides
(Fig.
3)
showing
homology
with
the
canonical
-10
structure.
The
promoters
of
the
BsuRI
genes
seem,
however,
atypical
because
they
lack
an
appropriately
spaced
-35
region.
The
sequences
in
the
promoter
regions
do
not
resemble
promoters
utilized
by
minor
forms
of
B.
subtilis
RNA
polymerase
(35).
The
transcriptional
termination
of
the
BsuRI
genes
has not
been
investigated
experimentally.
In
E.
coit
(36)
and
probably
also
in
B.
subtilis
(38)
transcriptional
terminators
are
characterized
by
a
self-complementary
structure
followed
by
several
T
residues.
We
can
find
a
similar
sequence
dowmstream
of
the
methylase
stop
codon
(Fig.
3),
nucleotides
4140-4150
and
4164-4174)
and
we
would
propose
that
this
potential
stem-and-loop
structure
probably
functions
as
a
transcriptional
terminator.
No
similar
terminator-like
6413

Nucleic
Acids
Research
sequence
can
be
found
downstream
of
the
endonuclease
gene
(Fig.3.).
To
test
whether
the
endonuclease
gene
is
expressed
in
the
absence
of
the
methylase
we
tried
to
delete
the
methylase
gene
from
the
plasmid
pSU13.
This
was
done
by
digesting
the
plasmid
with
BamHI
(Fig.
IB)
and
then
ligating
it
at
low
DNA
concentration.
Aliquots
of
the
ligation
mixture
were
used
to
transform
HB101
and
HB1O1(pSU184-11)
cells.
pSU184-11
is
a
pACYC184
derivative
carrying
the
functional
BsuRI
methylase
gene
on
the
3.7
kb
EcoRI
fragment
originally
cloned
in
pSUll
(Fig.
1).
Equal
amounts
of
ligated
DNA
yielded
approximately
100
times
fewer
transformants
in
HB101
than
in
HB1O1(pSU184-11).
R-,striction
analysis
of
six
clones
obtained
in
the
HB101
host
revealed
that
all
retained
the
2.4
kb
BamHI
fragment
coding
for the
methylase.
Two
clones
contained
this
BamHI
fragment
in
the
original
orientation,
(i.e.
they
were
identical
with
pSU13),
four
clones
contained
it
in
the
opposite
orientation.
Inverting
the
methylase
gene
did
not
seem
to
influence
the
r-m
phenotype:
the
DNA
was
fully
resistant
to
HaeIII
and
the
clone
showed
the
same
level
of
restriction
in
vivo
as
pSU13.
As
expected
from
the
difference
in
the
transformation
efficiencies,
most
of
the
TcR
AmpR
clones
obtained
with
the
HB1O1(pSU184-11)
host
did
not
contain
the
2.4
kb
BamHI
fragment;
the
BsuRI
methylase
gene
on
the
compatible
replicon
compensated
for
the
deleted
gene
of
pSU13.
Phenotypically
(methylation
of
DNA
and
in
vivo
restriction)
these
clones
were
indistinguishable
from
pSU13.
From
these
experiments
we
conclude
that
expression
of
the
endonuclease
does
not
require
the
presence
of
active
methylase
and
the
endonuclease
is
lethal
without
the
methylase.
Although
we
did
not
try
to
quantitate
the
level
of
methylase
in
those
derivatives
where
the
methylase
gene
has
been
turned
around
or
put
on
another
compatible
replicon,
these
observations
confirm
the
conclusion
drawn
from
transcript
mapping,
that
the
two
genes
do
not
constitute
an
operon.
They
are
transcribed
independently.
Translational
signals
The
AUG
start
codons
of
the
BsuRI
genes
are
preceded
by
sequences
showing
complementarity
with
the
3'-end
of
B.
subtilis
16
S
RNA
(ref.
39,
Fig.
3).
The
ribosomal
binding
site
of
the
methylase
allows
for
a
much
stronger
Shine-Dalgarno
interaction
(calculated
free
energy
-18.8
kcal/mol;
ref.
40)
than
that
of
the
nuclease
(-9.4
kcal/mol).
Base
composition
and
codon
usage
The
base
composition
of
the
BsuRI
genes
is
characterized
by
a
high
A
+
T
content:
68.4%
for
the
nuclease
and
61.9%
for
the
methylase.
This
is
6414

Nucleic
Acids
Research
Table
1.
Codon
usage
of
the
BsuRI
genes.
higher
than
the
average
A
+
T
content
of
B.
subtilis
DNA
(57%,
ref.
41).
The
base
composition
is
reflected
in
the
codon
usage:
A
and
U
are
strongly
preferred
nucleotides
in
the
third
position
or,
whenever
possible,
in
the
first
position
of
codons
(Table
1).
Comparisons
of
protein
sequences
In
addition
to
the
BsuRI
enzymes,
the
amino
acid
sequences
of
two
other
proteins
recognizing
the
tetranucleotide
GGCC,
are
known.
These
are
the
6415

Nucleic
Acids
Research
H
(n
QL
H
U,
m
R.BsuRI
M.
BsuRI
0r
0-
cn
n:
M.BspRI
M.BsuRI
Fig.
5.
Sequence
comparisons
between
the
M.BsuRI,
M.BspRI,
M.SPR
and
R.BsuRI
enzymes.
Dot
matrix
outputs
were
generated
by
a
computer
program
(DIAGON,
ref.
32)
which
utilizes
Dayhoff's
similarity
scores
between
amino
acids.
Diagonal
stretches
represent
regions
of
homology.
BspRI
methylase
of
Bacillus
sphaericus
(7)
and
the
SPR
methylase
of
the
B.
subtilis
phage
SPR
(8,9).
The
M.BspRI
and
M.SPR
enzymes
were
found
to
share
partial
sequence
homology
(8,9).
We
performed
a
computer
search
to
test
whether
there
is
any
sequence
similarity
between
the
BsuRI
endonuclease
and
methylase
or
between
the
BspRI
and
SPR
methylases
and
the
BsuRI
enzymes.
No
significant
homology
was
detected
between
the
BsuRI
endonuclease
and
any
of
the
three
methylases.
There
are,
however,
regions
of
homology
between
the
BsuRI
methylase
and
the
two
other
modification
enzymes
(Fig.
5).
The
similarity
of
the
amino
acid
sequences
is
especially
strong
between
the
BsuRI
and
BspRI
methylases,
the
homology
extends
over
almost
the
whole
molecule
(Fig.
5
and
6).
Weaker,
but
significant
homology
exists
between
the
SPR
methylase
and
the
two
other
methylases.
These
sequence
similarities
were
found
in
two
blocks
(Fig.
6).
In
these
regions
many
amino
acids
are
conserved
in
all
three
enzymes.
The
enzymes
are
also
very
similar
in
size,
the
BspRI
methylase
consists
of
424,
the
BsuRI
methylase
436
and
the
SPR
methylase
of
439
amino
acids.
6416
I

Nucleic
Acids
Research
M.
Bsu
MNLKIDIKGRGKYKPASDYSIDDVXNVLMEKIFEESSRI
INSDDDLEI
IEKVDFRTDKINVLSLFSGCGGLDLGFELAGL
80
M.
Bsp
MA
IKINEKGRGKFKPAPTYEKEEVRQLLMEKINEEMEAVATATSDI
SN
DEIQYKSDKFNVLSLFCOAGGLDLGFELAGL
79
M.DZu
M.Ump
M.
SPR
M.Bhu
M.Bsp
M.
SPR
M.Dau
M.Bap
M.
SPR
M.Bsu
M.BaD
AAVIGEQAAMEAFKDKDRFNELNKS
I
FHTIYTNDLFKEANQTYKTNFPGHVIQHEKDfIlQNY{fL
I
PC
EQSLGTDKALEAFKDIDVYNAI
RHESVFHTVYANDI
FSEALQTYEDNMPNHVF
IHEKDIRKI.ElAL
VI
PC
FGDVSK
I
DIC
"
[1lEFD
VXSP
S
S
E
AG
P_LV
IDD_
ELYHFIRC
L
IQXEI
FV
QI
ASLQ
S
HGF
D
T31IFQYVE
I
E
IDL
KF
VII
SFNYKYPSPTHGEETOLKPFKTLRDS
IGDLVTDPOPYFTGSYSSI
ISRNRXSWDEQSFTIQASGRQAP
VI:
DFNYEYPEITOGNEEGLKPYVTLEEAIGDLSLDPOPYFTGSYST
I
SRNUKKKWTDQSFTIQASGIQAP
L
IIe13
17
LHPOOLS
IKGKDKWVFPDGEEN
VKIA
SQGTNSQTSI
V
LAAVAS
IHLGOLPM1KVDKNKWI
FPDGEEN
VI
ISDOGNMKVSVLTRAVAK
160
59
81
240
2
39
62
320
31
9
400
399
M.
SPR
. C
E4KFA
5J
:I
419
M.Bsu
PIANWAINYL
ESSPNN
416
M.
Bsp
S
IAQFAADYLKDNHPHE
416
Fig.
6.
Sequence
homologies
between
the
BsuRI,
BspRI
and
SPR
methylases.
These
homologies
were
found
using
the
program
SEQHP
(33).
Homologies
are
indicated
by
colons.
Amino
acids
which
are
conserved
in
all
three
methylases
are
boxed.
DISCUSSTON
The
potential
lethality
of
r-m
systems
is
of
primary
concern
in
all
attempts
aimed
at
cloning
these
genes.
Although
several
complete
r-m
systems
have
been
cloned
(1-5
and
G.
Wllson,
pers.
comm.),
in
at
least
as
many
other
cases
the
cloning
attempts
have
failed.
Cloning
of
the
BsuRI
genes
has
been
easily
accomplished
in
the
host
E.
coil
RR1
and
the
system
was
found
to
be
stable
in
E.
colU
HB101.
On
the
other
hand,
we
were
unable
to
transform
several
other
E.
coil
strains
(e.g.
DH1)
even
with
plasmids
coding
only
for
the
methylase.
We
found
that
plasmids
coding
for
the
BspRI
or
SPR
methylases
behaved
in
the
same
way
and
the
inability
to
transform
certain
E.
coil
strains
was
due
to
the
methylase
function.
Other
investigators
have
made
similar
observations
with
genes
coding
for
different
modification
enzymes
that
methylate
cytosine
(R.
Blumenthal,
T.
Trautner
and
G.
Wilson,
pers.
comm.).
In
addition
to
the
BsuRI
system,
the
organization
of
five
other
type
II
r-m
systems
CEcoRI,
HaII,
PstI,
EcoRV
and
PaeR7)
has
been
determined
and
very
different
gene
arrangements
have
been
found
(1-5
and
J.
Brooks,
pers.
comm.).
The
arrangement
of
the
BsuRI
genes
is
similar
to
the
EcoRI
genes
in
6417

Nucleic
Acids
Research
the
sense
that
they
are
tandemly
arranged
and
the
restriction
gene
is
upstream
of
the
methylase
gene,
but
there
is
considerable
difference
in
the
length
of
the
intergenic
region
(21
bp
between
the
EcoRI
genes
versus
780
bp
between
the
BsuRI
genes).
One
intriguing
observation
about
cloned
r-m
systems
is
that
plasmids
coding
for
both
enzymes
transform
E.
coli
with
a
frequency
similar
to
that
of
other
plasmids
of
similar
size.
Clearly,
some
mechanism
must
exist
which
ensures
that
expression
of
the
endonuclease
is
delayed
compared
to
the
methylase.
Sequential
transcription
(3),
difference
between
the
synthesis
rates
of
the
nuclease
and
the
methylase,
time
needed
for
the
nuclease
to
assemble
in
active
dimer
form
(4)
and
inhibition
of
the
nuclease
translation
by
a
potential
higher
structure
of
the
nuclease
mRNA
(5)
have been
suggested
as
control
mechanisms.
In
the
BsuRI
system
the
endonuclease
gene
is
upstream
of
the
methylase
gene,
therefore
sequential
transcription,
as
was
suggested
for
the
HhaII
system
(3),
cannot
play
a
role
in
regulating
the
expression
of
the
nuclease.
The
BsuRI
nuclease
is
thought
to
be
a
monomeric
enzyme
(15)
so
dimerization
cannot
be
a
regulating
factor.
A
search
of
the
sequence,
that
is
likely
to
correspond
to
the
BsuRI
endonuclease
mRNA,
for
self-complementary
regions
failed
to
reveal
any
that
could
form
a
secondary
structure
similar
to
that
predicted
for the
EcoRV
endonuclease
mRNA
(5).
At
present
the
only
feature
of
the
BsuRI
system
that
might
suggest
a
control
mechanism
is
the
difference
in
the
ribosomal
binding
sites.
The
Shine-Dalgarno
sequence
preceding
the
methylase
start
codon
is
much
stronger
than
that
preceding
the
nuclease
start
codon
and
this
difference
might
be
important
in
the
regulation
of
the
relative
amounts
of
methylase
and
endonuclease.
The
calculated
molecular
weight
of
the
endonuclease
(66,314)
is
in
good
agreement
with
the
value
observed
experimentally
(68,000;
ref.
15).
The
evaluation
of
the
molecular
weight
calculated
for
the
methylase
(49,642)
is
more
problematic.
Gunthert
et.
al.
(16,17)
found
two
methylases
associated
with
the
BsuRI
system.
The
enzymes
differed
slightly
in
their
enzymological
characteristics
and
in
their
molecular
weights.
Depending
upon
the
method
used
these
were
37-43
kd
for
M.BsuRIa
and
39-43
kd
for
M.BsuRIb.
It
was
not
determined
whether
the
two
enzymes
differ
as
a
result
of
post-translational
modification
or
whether
they
are
coded
by
different
genes
(16,
17).
Considering
the
similarity
of
the
two
enzymes
and
the
genetic
data
available
for
the
BsuRI
system
(11)
the
first
possibility
seems
more
likely.
We
assume
that
the
gene
we
have
characterized
in
this
paper
codes
for
a
precursor
that
would
undergo
post-translational
processing
to
give
rise
to
the
two
enzyme
6418

Nucleic
Acids
Research
forms.
The
same
mechanism
was
suggested
to
explain
the
existence
of
two
methylases
in
the
HpaII
system
(42).
Obviously,
further
studies
are
needed
to
understand
this
phenomenon
and
the
functioning
of
the
BsuRI
system
in
general,
but
the
availability
of
the
cloned
genes
and
their
sequence
should
prove
very
useful
for
future
work.
Our
main
purpose
with
the
study
of
restriction-modification
systems
is
to
learn
how
these
enzymes
recognize
a
specific
sequence
in
the
DNA.
The
approach
we
have
taken
is
to
clone
and
sequence
genes
of
enzymes
recognizing
the
same
DNA
sequence.
It
seemed
likely
that
comparison
of
these
proteins
may
help
find
the
common
structural
determinants
of
the
sequence-specific
interaction.
At
present
there
are
four
GGCC-specific
enzymes
(BspRI,
BsuRI
and
SPR
methylases
and
the
BsuRI
nuclease)
for
which
the
protein
sequence
(derived
from
the
DNA
sequence)
is
known
(7,8,9
and
this
paper).
Comparison
at
the
amino
acid
level
of
the
BsuRI
methylase
and
nuclease
did
not
reveal
any
significant
degree
of
homology.
Investigation
of
two
other
RM
systems
(EcoRI,
PstI)
led
to
similar
conclusions
(1,2,4).
Similarly,
in
a
computer
search
of
the
amino
acid
sequences
published
for
the
HhaII
and
EcoRV
enzymes
(3,5)
we
failed
to
detect
any
homology
between
the
endonuclease
and
methylase
of
the
same
system.
Although
it is
possible
that
analysis
of
the
three-dimensional
structure
may
reveal
common
structural
elements
in
the
nuclease
and
methylase
belonging
to
the
same
system,
it
now
seems
more
likely
that
the
lack
of
amino
acid
sequence
homology
reflects
the
different
nature
of
the
molecular
mechanisms
by
which
the
nuclease
and
the
methylase
interact
with
their
target
sequence.
Enzymological
studies
support
this
conclusion
(43).
Comparison
of
the
BspRI
and
SPR
methylases
detected
partial
homology
between
the
amino
acid
sequences
(8,9).
In
this
paper
we
show
that
there
is
a
similar
degree
of
homology
between
the
SPR
and
BsuRI
methylases
and
much
higher
homology
between
the
BspRI
and
BsuRI
methylases.
We
interpret
the
homologies
detected
as
an
indication
of
evolutionary
relatedness.
It
remains
to
be
seen,
however,
whether
these
homologies
are
related
to
the
enzymatic
function.
ACKNOWLEFDGEMENTS
We
thank
Dr.
T.
Trautner
for
the
strain
B.
subtills
R,
Dr.
M.
Zoller
for
synthesis
of the
oligonucleotides,
Dr.
J.
Posfai
for
doing
some
of
the
computer
work,
Drs.
T.
Trautner,
R.
Blumenthal,
G.
Wilson
and
J.
Brooks
for
6419

Nucleic
Acids
Research
communicating
their
results
before
publication,
Dr.
Ashok
Bhagwat
for
his
helpful
comments
on
the
manuscript
and
E.
Csorba
for the
skillful
technical
assistance.
Part
of
this
work
was
supported
by
NSF
grant
DMB
8217553.
A.
K.
was
supported
under
acollaborative
agreement
between
Cold
Spring
Harbor
Laboratory
and
Exxon
Research
and
Engineering.
REFERENCES
1.
Greene,
P.J.,
Gupta,
M.,
Boyer,
H.W.,
Brown,
W.E.
and
Rosenberg,
J.M.
(1981)
J.
Biol.
Chem.
256,
2143-2153.
2.
Newman,
A.K.,
Rubin,
R.A.,
Kim,
S.
and
Modrich,
P.
(1981)
J.
Biol.
Chem.
256,
2131-2139.
3.
Schoner,
B.,
Kelly,
S.
and
Smith,
H.O.
(1983)
Gene
24
227-236.
4.
Walder,
R.Y.,
Walder,
J.A.
and
Donelson,
J.E.
(1984)
J.
Biol.
Chem.
259,
8015-8026.
5.
Bougueleret,
L.,
Schwarzstein,
M.,
Tsugita,
A.
and
Zabeau,
M.
(1984)
Nucl.
Acids
Res.
12,
3659-3676.
6.
Brooks,
J.E.,
Blumenthal,
R.M.
and
Gingeras,
T.R.
(1983)
Nucl.
Acids
Res.
11
837-851.
7.
Posfai,
G.,
Kiss,
A.,
Erdei,
S.,
Posfai,
J.
and
Venetianer,
P.
(1983)
J.
Mol.
Biol.
170
597-610.
8.
Buhk,
H.J.,
Behrens,
B.,
Tailor,
R.,
Wilke,
K.,
Prada,
J.J.,
Gunthert,
U.,
Noyer-Weidner,
M.,
Jentsch,
S.
and
Trautner,
T.A.
(1984)
Gene
29
51-61.
9.
Posfai,
G.,
Baldauf,
F.,
Erdei,
S.,
Posfai,
J.,
Venetianer,
P.
and
Kiss,
A.
(1984)
Nucl.
Acids
Res.
12
9039-9049.
10.
Macdonald,
P.M.
and
Mosig,
G.
(1984)
EMBO
J.
3
2863-2871.
11.
Trautner,
T.A.,
Pawlek,
B.,
Bron,
S.
and
Anagnostopoulos,
C.
(1974)
Mol.
Gen.
Genet.
131
181-191.
12.
Bron,
S.,
Murray,
K.
and
Trautner,
T.A.
(1975)
Mol.
Gen.
Genet.
143,
13-23.
13.
Bron,
S.
and
Murray,
K.
(1975)
Mol.
Gen.
Genet.
143,
25-33.
14.
Gunthert,
U.,
Storm,
K.
and
Bald,
R.
(1978)
Eur.
J.
Biochem.
90,
581-583.
15.
Bron,
S.
and
Horz,
W.
(1980)
in
Methods
in
Enzymology,
Grossman,
L.,
Moldave,
K.,
ed.,
Vol.
65
pp.
112-132,
Academic
Press,
New
York.
16.
Gunthert,
U.,
Freund,
M.
and
Trautner,
T.A.
(1981)
J.
Biol.
Chem.
256
9340-9345.
17.
Gunthert,
U.,
Jentsch,
S.
and
Freund,
M.
(1981)
J.
Biol.
Chem.
256
9346-9351.
18.
Kiss,
A.
and
Baldauf,
F.
(1983)
Gene
21,
111-119.
19.
Boyer,
H.W.
and
Roulland-Dussoix,
D.
(1969)
J.
Mol.
Biol.
41
459-472.
20.
Bolivar,
F.,
Rodriguez,
R.L.,
Greene,
P.J.,
Betlach,
M.C.,
Heynecker,
H.L.,
Boyer,
H.W.,
Cross,
J.H.
and
Falkow,
S.
(1977)
Gene
2
95-113.
21.
Celeste,
Y.-P.,
Vieira,
J.
and
Messing,
J.
(1985)
Gene
33
103-119.
22.
Maniatis,
T.,
Fritsch,
E.F.
and
Sambrook,
J.
(1982)
in
Molecular
Cloning:
A
laboratory
manual,
Cold
Spring
Harbor
Laboratory,
Cold
Spring
Harbor,
New
York.
23.
Roberts,
R.J.
(1985)
Nucl.
Acids
Res.
13
r165-r200.
24.
Davis,
R.W.,
Botstein,
D.
and
Roth,
J.R.
(1980)
in
Advanced
Bacterial
Genetics.
A
Manual
for
Genetic
Engineering.
Cold
Spring
Harbor
Laboratory,
Cold
Spring
Harbor,
New
York.
25.
Sanger,
F.,
Nicklen,
S.
and
Coulson,
A.R.
(1977)
Proc.
Natl.
Acad.
Sci.
USA
74
5463-5467.
6420

Nucleic
Acids
Research
26.
Maxam,
A.M.
and
Gilbert,
W.
(1980)
in
Methods
in
Enzymology,
Grossman,
L.
Moldave,
K.,
ed.,
Vol
65
pp.
499-560,
Part
I.,
Academic
Press,
New
York.
27.
Norrander,
J.,
Kempe,
T.
and
Messing,
J.
(1983)
Gene
26
101-106.
28.
Messing,
J.
(1983)
Methods
Enzymol.
101
20-79.
29.
Biggin,
M.D.,
Gibson,
T.J.
and
Hong,
G.F.
(1983)
Proc.
Natl.
Acad.
Sci.
USA
80
3963-3965.
30.
Keller,
C.,
Corcoran,
M.
and
Roberts,
R.J.
(1984)
Nucl.
Acids
Res.
12
379-386.
31.
Lo,
K.-M.,
Jones,
S.S.,
Hackett,
N.R.
and
Khorana,
H.G.
(1984)
Proc.
Natl.
Acad.
Sci.
USA
81
2285-2289.
32.
Gilman,
M.Z.
and
Chamberlin,
M.J.
(1983)
Cell
35
285-293.
33.
Staden,
R.
(1982)
Nucl.
Acids
Res.
10
2951-2961.
34.
Goad,
W.B.
and
Kanehisa,
M.I.
(1982)
10
247-263.
35.
Johnson,
W.C.,
Moran,
C.P.
and
Losick,
R.
(1983)
Nature
302
800-804.
36.
Rosenberg,
M.
and
Court,
D.
(1979)
Ann.
Rev.
Genet.
13
319-353.
37.
Moran,
C.P.Jr,
Lang,
N.,
LeGrice,
S.F.J.,
Lee,
G.,
Stephens,
M.,
Sonenshein,
A.L.,
Pero,
J.
and
Losick,
R.
(1982)
Mol.
Gen.
Genet.
186,
339-346.
38.
Lehtovaara,
P.,
Ulmanen,
I.
and
Palva,
I.
(1984)
Gene
30
11-16.
39.
McLaughlin,
J.R.,
Murray,
C.L.
and
Rabinowitz,
J.C.
(1981)
J.
Biol.
Chem.
256,
11283-11291.
40.
Tinoco,
I.Jr,
Borer,
P.N.,
Dengler,
B.,
Levine,
M.D.,
Uhlenbeck,
O.C.,
Crothers,
D.M.
and
Gralla,
J.
(1973)
Nature
New
Biol.
246,
40-41.
41.
Marmur,
J.,
Seaman,
E.
and
Levine,
J.
(1963)
J.
Bacteriol.
85
461-467.
42.
Yoo,
O.J.
and
Agarwal,
K.L.
(1980)
J.
Biol.
Chem.
255
6445-6449.
43.
Modrich,
P.
and
Roberts,
R.J.
(1982)
in
Nucleases,
Linn, S.M.,
Roberts,
R.J.
ed.,
pp.
109-154,
Cold
Spring
Harbor
Laboratory,
Cold
Spring
Harbor,
New
York.
6421