Download Mechanist’s of DNA Replication and Recombination and more Lecture notes Molecular biology in PDF only on Docsity!
In: “Mechanist’s
of DNA Replication
and Recombination”
UCLA
Symposium on
Molecular
and Cellular
Biology,
New
Series
(I. Kelly and R McMacken,
eds.)
Alan R. Liss,
New York,
in press
(1981).
STUDIES
ON THE T BACTERIOPHAGE
DNA
REPLICATION
SYSTEM
Harold
E. Selick,
Jack (^) Barry,
Tai-An
Cha, (^) Maureen
Muon,
Mikiye
Nakanishi,
Mci Lie Wong and Bruce H. Alberts
Department
of Biochemistry
and Biophysics
University
of California,
San Francisco,
CA
ABSTRACT
The study of DNA replication
has (^) been
greatly
facilitated
by the development
of in vitro systems
that
render (^) the process
much more amenable
to study (^) than
it
is within
the living
cell.
The bacteriophage
T
in vitro
system (^) has
proven
particularly
valuable
for
understanding the
detailed
events
that take (^) place
at^ (^) the
moving
replication
fork.
The current
model
of the T
replication
complex
envisions
asophisticated
assemblage
of proteins,
in which
leading
and lagging
strand
DNA
syntheses
are (^) coupled
by means
of a true (^) “replication
machine”.
In this article,
we address
the (^) following
questions
with regard (^) to this repli canon
machine:
What
determines
(^) the (^) length
oLthe
Okazaki
fragments
and what^
are the template
sequences
required
(^) to begin
their
synthesis?
Which protein
is the actual
RNA primase
that
is required
(^) for Okazaki
fragment
synthesis?
Does the
helix-destabilizing
protein
the primosome
and the
two DNA polymerase
(^) molecules
(^) the repli
cation
complex?
What is the architecture
of the DNA
polymerase
holoenzyme
on primer-templatea
(^) junction
and
why is this enzyme
so complicated?
Finally,
what addi
tional
proteins
may be requirod
to^ (^) t•econstLtute
(^) the
origin-specific
(^) initiation
of T DNA replication
forks
in vitro?
INTRODUCTI
ON
Approximately
(^) twenty
(^) genes (^) have (^) been
shown (^) dv genetic
azayss
c b€ insoitad
1, 4 IO n1uat
on H
1 aboratones
(^) hne
been (^) us (^) ng tn:s genot 1 r
toal si as toola
for (^) reconstructing
(^) the (^) biochemis
try of the T DNA replication
process
in vitro
(2). Eleven
gene products
that are believed
to be directly
involved
(^) in the formation
and propagation
(^) of
DNA
replication
forks have thus far been isolated
and puri
fied (^) to near homogeneity
in our^
laboratory
and are listed
in
Table 1 (3-7).
The products
of genes (^) 43, 44/62,
and 45
comprise
the T DNA polymerase
holoenzvme.
By itself,
the T
gene 43 protein
(DNA
polymerase)
is capable
(^) of elongating
pre-existing
primers
on single-stranded
DNA
templates
while
its 35i exonuclease
activity
provides
aproofreading
function
(9).^
The products
(^) of genes 44/
and 45 comprise
the polymerase
accessory
proteins
and display
aDNA-dependent
ATPase
activity
On a primed
single-stranded
DNA
template,
(^) the polymerase
accessory
proteins
interact
with (^) the
DNA
polymerase
in areaction
that requires
ATP
hydrolysis
by
the 44/ complex.
This interaction
can result
in adramatic
increase
in both
the rate and processivity
of DNA synthesis
by the polymerase
(^) molecule
(l2-l5)
The activity
of the
accessory
proteins
(^) is most consistent
with their formation
(^) of
a complex
with (^) the
polymerase,
which
acts as a ttsliding
clamp”
that keeps (^) each polymerase molecule
at the 3’ end of a
growing
DNA chain (^) for many cycles
of synthesis
(^) (15,16).
The gene 32 protein
is^ ahelix
(^) destabilizing
protein
(or
single-stranded
DNA
binding
protein),
and it is required
(^) to
allow
the polymerase
hoioenzyme
to synthesize
DNA
on a
double-stranded
DNA
template
(17,18).^
This protein
binds
Table (^) L Properties
Bacteriophageof
(^) T (^) Replication
(^) Proteins
Molecular
Current
Type of
weight
punty
protein
T4 gene
(x
(%)
Activitiest
DNA polymerase
43
1015
99
5’ exonoclease5*polymerase;
Polytnerase
44/62:
352/2L4;
99;
(^55) DNAuerminC
accessory proteins
dependent
ATPase,
dAlPase
Helixdestabi1izing
32
34i
99
cooperative
(^) binding
protein
to 55 DNA
RNA-priming
(^) proteins
41
53i
99
long 55 DNA
dependent (^) GTPase.
ATPase
61
95
9 binds (^) DNA
Type-Il
39/52/
572/5ft6/
99 DS DSA strand^
topoisomerase
passage: DS^ DNA
depcndent (^) ATPase
DNA belicase
ida
(^47)
(^95)
ATPase;^55 DN4deoenden; (^) DNA
unwinding
155 indicates single-stranded;
(^) DS (^) indicates
(^) double-stranded,
The products
of genes 39, (^) 32,
and 60 encode
the three
subunits
of the T type II DNA topoisomerase
(6,33L
Nuta
tions
in these genes
seem to affect
the initiation
of DNA
replication
rather
than the rate of fork movement
(36L
However,
the precise
(^) role
of this topoisomerase
(^) in replica
tion is• as yet unclear.
RESULTS
Template
Restrictions
on RNAPrimed
DNA
Chain Starts.
Based upon
our previous
results,
we proposed
amodel
for
replication
fork movement
in which
lagging
strand
DNA synthe
sis is coupled
to that on the leading
strand
(30).
The
mechanism
permits
efficient
replication
of the (^) lagging
strand
by preventing
(^) the
dissociation
of its polymerase
(^) molecule
from the replication
complex,
which
allows
this lagging
strand
DNA polymerase
to be recycled
after the completion^
(^) of
each successive
Okazaki
fragment.
Implicit
in this “trombone
model”
of DNA synthesis
(^) is (^) the (^) notion
that the selection
of
successive
RNA
primer
sites for Okazaki
fragment
synthesis
is
triggered
by the completion
of the synthesis
of the (^) previous
Okazaki
fragment
on an individual
lagging
strand.
However,
as suggested
by the sequence
analysis
of the RNA primers
themselves
primer
site selection
will also be
constrained
by the template
DNA
sequence.
In order (^) to
determine
(^) the full extent
of this (^) constraint,
we have per
formed
an analysis
of the (^) template
sequence
reqnirements
for
primer
synthesis
on a simple
single-stranded
DNA
template,
using an approach
similar
to that previously^
employed
in the
T
bacteriophage
system
(37L
In Figure
1 we present
a
representative
(^) selection
from a compilation
(^) of thirty-two
different
RNA
primer
sites (^) that
have been^ mapped (^) on an Nl
single-stranded
template
In these
experiments,^
both^ •the
primosome
and the DNA polymerase
holoenzyne
were
present,
and the (^) “primers”
shown are all oligoribonucleotides
that were (^) found
to prime DNA synthesis
The earlier
(^) direct
sequencing
of^ the pentaribonucleo
tides (^) synthesized
by the 41/ (^) primosome
in vitro
revealed
prinurs
with (^) the sequences
ppp4pCp\p,
cdtson
c sea
of sequences
starting
with pppG that were less (^) well
charac
termed
(^) (25-27).
Figure
(^1) shows
that template
DNA
sequences
that (^) are (^) complementary
(^) to (^) the first two residues
(^) of pppApC
and pppGpC-start
(^) primers
(^) are (^) recognized
by the pri.mosome,
(^) as
expected.
In addition,
(^) the (^) data (^) reveals
(^) that (^) there
is^ a
48 GCCTAJTTCGTITAAAAA
60 A A A A AlT G A
S
C 1 T
C
A
T
T
T
105 A T A T TIA A C S T T T A C A A
124 AAATAITTTGCITTATAC
141 A T C T TIC C T S
TIT
T
TT
G
C
150 T T T T
TIC
0 0
CIT
T
T
T C T
A
T
T
C
AIC
A
T
CITS
A
C
T
193 A C A TGIC
T A S
TIT
T
T
A
C
C
C
A
T
TIA
C
C
S
T;T
C
A
T
C
223 A T T C T C T
ST
TIT
T
G
C
T
C
227 T C T T GT T T S
CIT
C
C
A
G
A
A
T
T
T
A 1 T
C
A
S
CIT
A
G
A
A
C
316 C T A 0 A 1 A C 0
S
TIT
C
A
A
T
A
442 ATCCTjGTGAAAT
RNA
primer is
complementary
to
the (^) boxed sequences
3’—
5
RNA
FIGURE
The template
(^) sequence
requirements^
for RNA
primer
synthesis.
DNA
sequences
corresponding
(^) to (^) several
different
RNA
primer
sites
which (^) have
been^ mapped
on a
singlestranded
Ml DNA template
are presented.
The
numbers
refer
to the distance
from the primer
start (^) site
to
the unique
Aval site on the M bacteriophage
chromosome.
conserved
T
residue
at^ (^) the
l position,
analogous
to the
conserved
C
residue
(^) at (^) the l position
Oi T bacteriophage
primer
start (^) sites
(37).
Amore
detailed
(^) analysis
shows that
the trinucleotide
sequences
0Th or OCT are (^) both necessary
and
sufficient
to specify
(^) the efficient
synthesis
of RNA primers
by the 41/ primosome complex
In order (^) to determine
the^ minimum
length
required^
(^) for a
functional
(^) primer,
(^) the (^) primer
activities
of truncated
oligo
ribonucleotides
resulting
from the omission
(^) of each^ one of
the io.r (^) r\Ts
rrom the (^) reaction
mix
determ,’ed
It
was thereby
found
that trinucleotides^
are completely^
inac
tive, while tetranucieotides
can pride synthesis^
as efficient
ly as the iulI-length
(^) 7entaribonncleotides
(29).^
Under
conditions
(^) designed
(^) to mimic
in vivo replication
as cbs ely as possible,
(^) thcre (^) is an absolute
(^) requirement
(^) for
both 41 and 61 proteins
(^) to observe primer^
synthesis^
(^) in vitro.
zlowcser,
when the (^) concentration
(^) of 61 protein
(^) is
ncreased
4babase rs
co&sç srarei
FIGURE
The distribution
of RNA primer
start
sites
within
a sequenced
(^5) kb region
of the T genome.
The
positions
of the sequence
(^) Gfl (^) in (^) the vicinity
of genes
and 61 on the T chromosome
are presented.
These
sites
correspond
to all of the potential
primer
sites
and do not
necessarily
reflect
those sites
that are actually
utilized
in vivo.
exposed
on the (^) lagging
strand (^) to trigger
RNA primer
synthesis
In any trombone
type (^) of model,
the (^) lagging
strand
DNA
polymerase
cannot^
initiate
synthesis
of a new Okazaki
fragment
until
it completes
the (^) synthesis
of the (^) previous
one. Moreover,
the very efficient
(^) utilization
of RNA primers
which
we have observed
suggests
that the release
of the
lagging
strand
template
by its polymerase
is the (^) signal
that
activates
the primosome
to synthesize
a new primer
at the
next available
GTE
site,
Using
this primer,
the lagging
strand
polymerase
then^ restarts
another
(^) cycle
of Okazaki
fragment
synthesis.
If the rates
of polymerization
on the leading
and
lagging
strands
were identical,
the length^ (^) of each successive
Okazaki
(^) fragment
on any particular
template
molecule
would (^) be
set equal (^) to (^) the l.ength
of the previouslysynthesized
frag
ment (30).
In this view,
the (^) polydisperse
size^ range (^) of
Okazaki
fragments
observed
(see Figure
35, below)
would
reflect
a corresponding
difference
in the size
of the first
Ukazaki
(^) fragment
synthesized
on each template
DNA
molecule
in vitro.
An attractive
aspect (^) of this proposal
(^) is that the
total
amount
of single-stranded
DNA
on the (^) lagging
(^) strand
would (^) remain
constant,
so that it would (^) be possible
for^ (^) the
many molecules
(^) of ‘ protein
at^ (^) the
fork. to be recycled
within (^) the rnp 1 icat’r
roonlex
as rb erved c t 5 e lag n
strand
DNA polymerase molecule
(3D).
We have tested:
whet.her
such^ a templating
mechanism
determines
(^) Ukazaki
(^) fragment
size in the (^) experiment
(^) schema
tized
in Figure
3A. By severely
(^) limiting
(^) the rNTPs required
for (^) lagging
(^) strand
(^) primer
(^) synthesis
(^) in (^) the (^) in vitro
system,
/©\
High //
Low
rNTPJ,/
\IrNTP I
HIgh IrNTP]
High [rNTP]^
labeied dNTPa
jlabeled dNTPs
=:=;:o
FIGURE
3A.
Is Okazaki
fragment
length (^) templated
at a
replication
fork?
Two possible
outcomes
are (^) presented
of an
experiment
(^) designed
to determine
whether^
the size of an
Okazaki
(^) fragment
is dependent
upon the size of the Okazaki
fragment
that was previously synthesized
at^ a fork.
Synthe
sis on anicked
double-stranded
DNA template
in the (^) presence
of non-limiting
(“high”)
concentrations
of each of, the ribo
nucleoside
triphosphates
(rNTPs)
yields
Okazaki
fragments
with an average
(^) length
(^) of l kb, (^) whereas
synthesis
in the
presence
(^) of limiting
(“low”)
rNTPs produces
abnormally^
(^) long
Okazaki
fragments
In this experiment,
once the long pattern
of Okazaki
(^) fragment
synthesis
was established,
the (^) concen
tration
of rNlPs was restored
to^ (^) the normal (^) “high”
(^) le.vels,
and labeled
dNTPs were (^) added
to allow the^ size
of the (^) newly
ssnthesized
fragments
(^) to be neacured
hs alkaline
ag2ros^
(^) eel
eIectrophores
(^) is foi, (^) Ioweo (^) by autoradiography.
The continued
synthesis
(^) of (^) the
long (^) fragments
would
he^ predicted
by a..
templating
mechanism,
while
a rapid
conversion^
to short
fragments
(^) would
(^) be suggestive
(^) of some other
mechanism
for
determining
(^) Okazaki
(^) fragment
size.
fewer
primers
are
synthesized
and
unusually
long
Okazaki
fragments
are
produced.
After
first
establishing
such
a
pattern
of
long
Okazaki
fragment
synthesis,
normal
concentra
tions
of
rNTPs
can
be
restored
along
with
labeled
dNTPs,
and
the
size
of
the
subsequently
synthesized
Okazaki
fragments
can
be
assessed
by
alkaline
agarose
gel
electrophoresis.
A
continued
synthesis
of
long
fragments
would
be
predicted
by
a
simple
length
templating
mechanism.
As
can
be
seen
in
the
data
of
Figure
SB,
however,
the
size
of
the
subsequent
Okazaki
fragments
does
not
appear
to
be
influenced
by
the
size
of
the
previously^
synthesized
fragment.
In
fact,
the
bottom
lane
shows
that
conversion
to
the
normal
kb
length
Okazaki
fragments
is
complete
within
1 mm
of
restoration
of
the
normal
concentrations
of
rNTPs.
The
results
in
Figure
suggest
that
the
original
model
of
coupled
leading
and
lagging
strand
DNA
synthesis
should
be
modified
as
depicted
in
Figure
In
this
view,
because
the
DNA
helix
has
already
been
opened
to
expose
the
single-
stranded
lagging
strand
template,
the
lagging
strand
DNA
polymerase
is
free
to
move
at
ratea
that
is
faster
than
that
of
the
leading
strand
polymerase
molecule
(or
rather,
to
pull
its
template
past
it
more
rapidly)
though
these
two
polymerase
molecules
are
physically
linked
together.
How
ever,
once
it
reaches
the
previously
synthesized primer, the^
lagging
strand
polymerase
is
forced
to
pause
before
releasing
its
DNA
template.
During
this
pause,
the
leading^
strand
polymerase
continues
its
translocation,
displacing^
an
addi
tional
amount
of
single-stranded
template
for
the
next
round
of lagging
strand
synthesis.
Finally,
the
release
of
the
DNA
by
the
lagging
strand
polymerase
signals
the^
associated
primosome
to
synthesize
an
RNA
primer
at
the
next^
available
primer
site
(providing
that
the
requisite
rNTPs
are
present),
thereby
restarting
the
next
cycle of
Okazaki
fragment
synthe
5 is.
There
are
two
predictions
of
the
model
in Figure
4 that
are
readily
tested.
First,
because
the
leading
and
lagging
strand
polymerases
move
at
different
rates,
the
total
amount
of
single-stranded
DNA
on
the
lagging
strand
will
change
throughout
each
cycle
of
Okazaki
fragment^
synthesis,
ruling
out
the
possibility
that
all
of
the
protein recycles.
In
order
to
test
for
such
recycling,
we
have
used
a
synthetic
RNA
molecule
that
binds
free
protein
tightly^
(polyribo
I;
ref.
as
atrap
for
free
protein
during
DNAa
synthesis
reaction.
The
addition
of
polyribo
I was
found
to
arrest
the
progression
of
replication
forks
nearly
immediately^
during
a
DNA
synthesis
reaction
that
requires
protein,
whereas
DNA
Ii
Q HeIixdestabiiizing
proten )gp32)
( D-nsn
(^) sn -rze
DEETh Pnrnosome(qp6T4l)
LASDIND STRAND
flfl
LEADING STRAND
W TERMINATE
SYNTHESIS
ON LAGGINO STRAND
‘ThN
CONTINUE
SYNTHESIS
ON LEADING STRAND
RELEASE
OF LAGGING
STRAND DNA
e RESTART
mmm:nm:r7DDIEtrxztzrRrn
FIGURE
A model
for DNA synthesis
at a replication
fork.
This model
of coupled
leading
and legging
strand
DNA
synthesis
differs
from our previous
one (2,30)
in assuming
that the rate of legging
strand
DNA synthesis
is^ faster
than
the rate
of leading
(^) strand
DNA synthesis
This (^) refinement
incorporates
a prolonged
(^) pause
hy the (^) lagging
(^) strand
DNA
ool’merase
molcle
hefnrn
Tt rnleAses
its DRI tnmolDtc
strand,
and it demands
that the total
amount
of singleS
stranded
DNA exposed
on the lagging^
(^) strand
(^) template
(^) changes
during
each cycle
of Okazaki
fragment
synthesis,
being
greatest-
when the (^) fragment
is initiated
and least (^) at (^) the
moment (^) when the fragment^
is completed.
(FIGURE
cont’d.)
molecules,
representative
of those
in^
which
the lagging
strand
polymerase
was captured
in the
process
of transiocation,
comprise
(^) only (^) about
of the
total.
The remaining
have type B structures,
which
represent
molecules
in which
the (^) lagging
strand
polymerase
has paused
after
completing
the snthesis
of an Okazaki
predominantfragment.
intermediate
if the speeds
of the (^) leading
and
lagging
strand
polymerase
molecules
are (^) the same.
When the
lagging
srrand
polymerase
has completed
synthesis
of its
current
Okazaki
fragment
and is paused
adjacent
to the
previously
synthesized
(^) primer,
molecules
like those
in^ the
bottom
panel
of Figure
5 are produced.
The fraction
of
molecules
(^) of this type (^) should
(^) approximate
the fraction
of the
cycle time that the lagging
strand
polymerase
(^) spends
waiting.
Our results
indicate
that the lagging
strand
polymera
se^
spends
only about 25% of its time moving (^) and 75% of its time
pausing
between
successive
rounds
of Okazaki
fragment
synthe
sis (Figure
(^) 5). Thus, the (^) lagging
strand
molecule
(^) appears
to
move along
its template
at more than (^) twice
the rate of its
leading
strand
counterpart.
Consistent
with
this rate
difference
on the two sides
of the (^) fork
s the (^) observation
that the measured
rates
of polymerase
translocation
are
faster
on single-stranded
than on double-stranded
DNA
tem
plates
in vitro (15,16).
Since the selection
of ENA primer
sites
is independent
of the (^) length
(^) of (^) the previously-synthesized
Okazaki
fragment,
some other mechanism
(^) must
be involved
(^) in (^) primer
site selec
tion.
According
to the model
in Figure
4, the primosome
becomes
“activated”
as a consequence
of the release
of the
DNA
by the (^) lagging
strand
polymerase
(^) molecule,
Since
the
potential
&TT
primer
sites will be encountered
with^ a fre
quency
of about
once in every
oO nucleotides,
primer
site
selection
by such an activated
primosome
should
be a very
fast event (^) in the (^) synthesis
(^) reaction.
If we therefore
ignore.
the brief
time required
(^) to (^) find
a primer
site,
the (^) major
parameters
that (^) will (^) influence
(^) the (^) frequency
(^) of primed
DNA
starts
and (^) hence
Okazaki^
(^) fragment
size (^) will
be the rates of
movement
of the leading^
and lagging
(^) strand
noiym:erases
(RI
nc RE ree
ev ant
I”gtn
ni
tbat
lagging
(^) strand
polymerase
(^) pauses
(^) before
(^) releasing
the lagging^
strand (^) template.
For the (^) synthesis
(^) of any particular
(^) pair
of
successive
Okazaki
(^) fragments,
these^ (^) narameters
(^) are (^) related
(^) to
the lengths^
of the two fragments
by the (^) foliowing equation:
1
[i
(n+l)
eq.
where
L(n)
=length
of
current
Okazaki
fragment
(nucleotides)
=length
of
next
Okazaki
fragment
(nucleot
ides)
T
=duration
of
lagging
strand
polymerase
pause
(sec)
H 1
=rate
of
leading
strand
polymerase
trans
location
(nucleotides/sec)
H 2
=rate
of
lagging
strand
polymerase
trans
location
(nucleot
ides/sec)
Here
we
have
assumed
that
T
is
aconstant,
whereas
in
reality
adistribution
of
pause
times
is
expected,
reflecting
afirst
order
rate
constant
for
the
dissociation
of
the
polymerase
from
its
template.
Thus,
a
broad
distribution
of
Okazaki
lengths
is
expected,
as
observed
(see
Figure
SB,
above).
From
the
above
equation,
L(n)
is
related
to
the
length
of
the
first
Okazaki
fragment,
L(o),
by
the
equation:
/R
\fl
-l
L(n) =k”)
L( 0 ) +T(R 1 )
(R 1 \
(eq.2)
R2)
where
L( 0 )
= length
of
the
first
Okazaki
fragment
(nucleot
ides)
L(n)
= length
of
the
nth consecutive
Okazaki
fragment
(nucleotides)
If
we
assume
that
Rl
and
R
do
not
change
with
time,
that
Rl
<
R2,
and
that
T
is
a
constant,
then
the
sizes
of
successively
synthesized
Okazaki
fragments
will
rapidly
converge
to
auniform
length,
L(n),^
that
is
independent
of
the
length of
the
initial
fragment,
L(o):^
R 1 R 2
as
ne
, L 1
R.-
(T)
(eq.3)
H
(^00) 0 m H H^0 0 H
CD rt H
- H 0 H H I H m m 0)H H U
Ox00)H HOH HI 0 0) HcUD HOH Ha)fr—D 00)H= H- (^) Q
- m HIBm n (^0) B ci ‘Im ciciH- HmH (^00) H
a’
CA t A^ ArA CIA CCCA^ CC CA^ AA (^) A A (^) rCA A (^) AACC^ (^) CA ACA (^) CAA (^) A (^) ClAY (^) CA (^) CACiA (^) AA ACt (^) r CAACC^ (^) TCrArC
Cjçjçn
AAAA ,CAC AAt TCAACACCCCCAA (^) A CC^ )
Send
A-C
0’ (^) 0’S
54
FIGURE
The sequence
and structure
of the 177/
nucleotide-long
DNA
primer-template
molecule
(^) used for the DNA
footprinting
analyses.
The construction
of this molecule,
derived
from a longer
P mp Haelll
fragment,
is descrihed
in ref.
The molecule
has (^) been
constructed
with the two
different
termini
shown,
in order (^) to distinguish
sequence-
specific
from structure-specific
protein-DNA
interactions.
The positions
of the T replication
proteins
on the
primer-template
molecule
were assessed
by determining
(^) the
extent
of protection
they afford
to cleavage
by several
different
cleaving
agents
(40)
The results
can be briefly
summarized
(^) as follows:
(^) Polymerase
alone (^) binds
the DNA substrate
adjacent
(^) to
the 3’ terminus,
protecting
a segment
of the duplex
region
from both
neocarzinostatin
and DNase I cleavage.
The
presence
of 32 protein
on the single-stranded
region
lowers
the Kd of the polymerase
for this DNA site by a factor
of
thirty,
presumably
as aconsequence
of the direct (^) interaction
observed
between
these two proteins
(^) (21,41).
The addition
(^) of 45 protein
extends
the region
of
DNase (^) I protection
(^) into (^) the duplex
several
(^) basepairs
beyond
•that conferred
by 43 alone.
In the absence
(^) of (^) polymerase
however,
the 45 protein
does not protect
any region
of the
primer 1 ’ternplate
molecule,
(^) with or
(^) without
protein
(^) present.
The polymerase
accessory
protein
complex,
in the
presence
(^) of the (^) nonhydrolyzable
ATP
analog,
ATPIS,
protects
basepairs
of duplex
and 8 nucleotides
of single-stranded
template
DNA
abutting
aprimer-template
(^) junction
from DNase (^) I
cleavage.
Addition
(^) of 32- protein
reduces
(^) the (^) concentrations
of 44/
and 45 proteins
(^) that
are (^) required to
(^) achieve
this
protection
by a factor
of
The binding
of^ the (^) non-hydro
e 4
ngse s 0’enatt
tot tic (^) fcunaor
at ttis
DNA-prote.in
comole:x;
no footprint
can be detected
in the
presence
of ATE, indicating
(^) that
the (^) hydrolysis
(^) of ATE
greatly
weakens
the complex.
In order (^) to (^) carry
(^) out (^) the
footprint
analysis^
on the
entire
holoenzyme
complex,
footprinting
experiments
were
performed
(^) using
a mixture
(^) of (^) the (^) accessory
proteins,^
FIGURE
A
model
of the moving
T DNA polymerase
holoenzyme
complex
on aprimer-template
junction.
This very
1 ot resolator
new (^) of tbe bo’oenzyrnc
DNA
tenplate
roncn
complex
(^) has (^) been (^) deduced
trom our DNA tootprint
anaiyses,
as
well as from a comparison
with similar
studies^
of the Klenow
fragment
(^) of F. coli DNA polymerase
Essential
to the
stability
of this presumed
structure
is the active
transloca
tion of the holoenzyme
(^) complex,
which^
is postulated
(^) to be
required
(^) to (^) maintain
(^) the
polymerase
accessory
proteins
in
their
“energized”
(°°ADPPi-bound)^
active
clamp
conforma
The tion. resulting
accessory
protein
complex,
(^) which
we propose
remains
hound to ADP Fi,± (^) represents
ahigh (^) energy
state and
constitutes
the active
polymerase
(^) clamp.
In the absence
of a
rapidly
(^) transiocating
polymerase,
(^) the decay
of this complex
to release
ADF
(^) to its dissociation
from the DNA
molecule,
explaining
our inability
to detect
it in cur
footprint
(^) analyses.
A
simple (^) diagram
depicting
the^ coupling,
of polymerase
trôns Ostin
to tee (^) Ca ntn (^) Pe n a tgr-Iearc
hs 1 c-
enzyme
complex
is depicted
in Figure
The accessory
protein
component
of the holoenzyme
(^) complex
is represented
(^) in
St eras 3 ,erg
states
s 1 t” F ,
and Th rep-escita
several
(^) activated
(^) conformations
(^) of (^) the (^) accessory
protein
complex,
(^) each
of^ (^) which
(^) forms
a stable
clamp
for (^) the
DNA
P
44/
Adwndanaf*Dc
Dissociation
of clamp
and DNA polymerase
FIGURE
Diagram
of
an
energy-driven
clock
to
measure
polymerase
pausing.
In
this
simple
model,
the
activated
holoenzyme complexes
denoted
E***,
E**,
and^
E*
are
maintained
by
their
coupling
to
the
polymerase
trans
location
cycle.
In
the
absence
of
polymerase
translocation,
the
accessory
protein
complex
decays
through
a
series
of
intermediate
conformations,
each
of
which
is
characterized
by
aparticular
half-life.
Decay
to
the
lowest
energy
state,
E,
allows
the
dissociation
of
the
accessory
protein
clamp
and
the
release
of
the
DNA
polymerase.
polymerase.
E
represents
the
lowest
energy
state
of
the
accessory
protein
complex
which,
having
released
its
bound
ADP
and
Pi,
has
lost
its
ability
to
clamp
the
polymerase
to
the
template.
In
this
view,
active
polymerase
translocation
causes
the
periodic
reactivation
of
the
accessory
protein
complex,
thereby
keeping
it
“pumped
up”
to
the
high
energy
states
that
are
required
to
keep
the
polymerase
tightly
bound.
As
soon
as
the
polymerase
pauses,
however,
the
activated
accessory
protein
complex
begins
to
decay
through
a
series
of
intermediate
energy
states,
each
characterized
by
a
particular
half-life.^
As
long
as
the
energy
state
of
the
polymerase
accessory
protein
complex
remains
at
or
above
E*
(that
is,
so
long
as
the
pause
time
is
less
than
the
sum
of
the
half-lives
for
each
decay
step),
the
accessory
protein
complex
can
be
recycled
to
its
activated
conformation
E*t
via
the
coupling
of^
the
reaction
E*
E***
to
polymerase
translocation.
For
longer
pauses,
however
--
such
as
when
