"mum." -

unansgggfi

gr:

ms":

unmufl" mam

 

 

 

 

 

FWAWR . 3.5....“

-.; flung 33.. .

gummy ‘mm . . fish
Uwunnmmwfifium$t
flu ‘ .... {EH

.1 :.
:h..fi$. :1
2

‘ .. g...

‘ 5...:

595‘.
1.56.0...
g!!! 1 u.
41......

Hatfikh

. 2 K‘ ‘3ng V.

tituiib..- «huh.
. 11;:

m.
3.13“"

 

 

“we LIBRARY
5 Michigan State

:02) University

This is to certify that the
dissertation entitled

DOWNSTREAM NTP EFFECTS ON HUMAN RNA POLYMERASE ll
TRANSCRIPTION ELONGATION

presented by
YALIN XIONG

has been accepted towards fulfillment
of the requirements for the

PhD. degree in Biochemistry and Molecular

Biology

 

Gm W» [a

) lMajor Projes‘sor’s Signature

a ‘l {233/ 0%

Date

 

 

MSU is an affinnative-action, equal-opportunity employer

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5/08 K lProflAcc&Pres/ClRC/DateDue indd

DOWNSTREAM NTP EFFECTS ON HUMAN RNA POLYMERASE ll
TRANSCRIPTION ELONGATION
By

YALIN XIONG

A DISSERTATION
Submitted to
Michigan State University

in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Biochemistry and Molecular Biology
2008

ABSTRACT
DOWNSTREAM NTP EFFECTS ON HUMAN RNA POLYMERASE ll
TRANSCRIPTION ELONGATION
By

YALIN XIONG

RNA polymerase II is the 12-Subunit enzyme that synthesizes
pre-messenger RNA in human cells. RNA polymerase II is among the largest and
most dynamic enzymes in the human biosphere, and it is the target of many
regulators. Transcription Factor llF (TFIIF), TFIIS, and the mushroom toxin
a-amanitin are used in this work as dynamic probes of the human RNA
polymerase II mechanism, using millisecond phase kinetic analyses and related
experimental strategies. The overall aim of this work is to determine the core
mechanism of elongation catalyzed by human RNA polymerase II and to
understand regulation by transcription factors.

The mushroom toxin a-amanitin is utilized as a transient state inhibitor
of human RNA polymerase II. These studies indicate that NTP substrates bind to
downstream DNA sites and apparently generate translocation force against a
translocation block induced by a—amanitin. In response to the block, RNA
polymerase II stalls in the core of the bond synthesis mechanism. It appears that
the set point of the stall can be adjusted by Changing the concentration of
downstream NTPS and the number of downstream Sites that can be accurately

occupied by NTPS. These experiments Show the value of transient state inhibition

experiments to analyze the internal mechanism of elongation catalyzed by human

RNA polymerase II and give insight into the core mechanism of RNA synthesis.

 

To my family

iv

 

‘F‘h‘fi

ACKNOWLEDGEMENTS

I would like to express the deepest appreciation to my advisor, Dr.

Zachary F. Burton, who has the attitude and the substance of a great mentor: he

 

continually and convincingly conveyed a spirit of adventure in regard to research r
and scholarship. Without his guidance and persistent help this dissertation would

not have been possible.

 

TA"!_!.-.. n . .

I would like to thank my guidance committee members, Dr. Honggao
Yan, Dr. Michael Feig, Dr. Gregg Zeikus and Dr. Jim Geiger for their contributions
during these years. I would also like to thank Dr. William Wedemeyer for his kind

help on bioinfomtatics.

I thank all my colleagues in the Burton laboratory for Sharing their

expertise, reagents, discussions, and laughs with me.

TABLE OF CONTENTS

List of Tables ................................................................................. viii
List of Figures ................................................................................................ ix
Key to Symbols and Abbreviations .............................................................. xi
Chapter One
The mechanism of elongation catalyzed by multi-subunit RNA polymerases .. 1
1. Transcription catalyzed by multi-subunit RNA polymerases .............. 1
2. a-amanitin: the specific and potent inhibitor of RNAP ll .................. 5
3. Translocation models ........................................................................ 7
4. The NTP-driven translocation model ............................................... 18
4.1 The secondary pore NTP loading hypothesis .......................... 18
4.2 The NTP-driven translocation hypothesis ................................ 21
5. NTP loading into the RNA polymerase II active Site through main
enzyme Channel ............................................................................... 28
6. The rock and roll model for RNA polymerase II translocation .......... 32
7. Fidelity and efficiency ...................................................................... 57
8. Conclusions ..................................................................................... 63
References .......................................................................................... 70
Chapter Two
NTP-driven translocation and regulation of downstream template opening by
multi-subunit RNA polymerases ..................................................................... 76
Abstract ............................................................................................... 76
1. Translocation models ............................................................ 78
2. Rate-limiting steps during elongation .............................................. 79
3. NTP-driven translocation ................................................................. 83
4. NTP loading into the RNA polymerase active Site ........................... 87
5. Evolutionary conservation of residues proposed to participate in
NTP-driven translocation .............................................................. 103
6. A structural model for NTP-driven translocation ............................ 107
7. Mushrooms and antimicrobials ............................................... 110
8. Fidelity and efficiency .................................................................... 111
9. Conclusions ................................................................................... 1 16
References ........................................................................................ 1 18
Chapter Three

A tunable ratchet driving human RNA polymerase II translocation adjusted by
accurately templated nucleoside triphosphates loaded at downstream sites and

by elongation factors .................................................................................... 122
Abstract ............................................................................................. 122
1. Introduction ......................................................................... 125

vi

2. Materials and Methods .................................................................. 130

2.1 Cell culture, extracts, and proteins .......................................... 130
2.2 NTP stocks .............................................................................. 131
2.3 Preparation of elongation complexes ...................................... 131
2.4 Running start, two-bond, double quench protocol ................... 133
2.5 Quality of RNAP elongation complexes .................................. 134
2.6 lsomerization reversal experiments ........................................ 135
2.7 Molecular images .................................................................... 136

3. Results .......................................................................................... 137
3.1 Robust isomerization reversal ................................................. 137
3.2 NTPS occupy template at least to i+4 ...................................... 148
3.3 The core mechanism of RNA synthesis .................................. 159

4. Discussion ..................................................................................... 173
4.1 NTP-driven translocation ......................................................... 173
4.2 lsomerization reversal and the “trigger loop” hypothesis ......... 174
4.3 A tunable ratchet driving translocation .................................... 176
4.4 NTP-driven translocation and fidelity ...................................... 184
References ........................................................................................ 186

Chapter Four

Summary and future plan ............................................................................. 190
Summary and future plan .................................................................. 190
References ........................................................................................ 204

vii

LIST OF TABLES

Table "-1. Yeast RNAP ll residues proposed to be important in NTP-driven
translocation ........................................................................... 1 06

viii

Figure l-1 .
Figure l-2.
Figure I-3.
Figure I-4.
Figure I-5.
Figure l-6.
Figure l-7.

Figure l-8.

Figure l-9.

Figure l-10.
Figure M 1.

Figure l-12.

Figure 1-13.

Figure l-14.

Figure "-1.

Figure "-2.

Figure "-3.

LIST OF FIGURES

The secondary pore NTP loading hypothesis ........................ 9
The NTP-driven translocation model .................................... 22
Schematic diagram of rock and roll ...................................... 33

The rocking assembly rocks, and the rolling assembly rolls. 42

The rolling assembly ............................................................ 44
Functional domains of the S. cerevisiae pr1 subunit ......... 46
Functional domains of the S. cerevisiae pr2 subunit ......... 50

The two-metal catalytic mechanism for RNA polymerase II
and the trigger loop hypothesis ............................................ 53

a-amanitin is a cyclic octapeptide that blocks rocking
of the rocking assembly and inhibits rolling of the

rolling assembly ................................................................... 55
The proposed i+2 NTP interaction Site ................................. 59
The proposed i+3 NTP interaction site ................................. 61

Six active site loops act as sensors of the phase of the bond
addition cycle and respond to i+1 NTP tightening ............... 64

Conformational coupling between the i+1 NTP, the
rocking assembly, the piston assembly, and the
proposed i+3 NTP site ......................................................... 66

Connections of the i+2 NTP interaction site on the rolling
assembly with the rocking assembly and active site ........... 68

Nucleotide trip-hosphate (NTP)-driven translocation

for multi-subunit RNA polymerases (RNAPS)

and (d)NTP-driven translocation for Simple DNA (DNAPS)
and RNA polymerases (proposed) ....................................... 80
Three NTP binding Sites in RNA polymerase II ................... 88

NTP-driven translocation by human RNA polymerase II ..... 93

ix

Figure "-4.

Figure Ill-1.

Figure Ill-2.

Figure Ill-3.

Figure Ill-4.

Figure Ill-5.

Figure Ill-6.

a-amanitin blocks translocation ........................................... 97

Robust isomerization reversal by human RNA polymerase II
induced by accurately templated downstream NTPS ......... 138

NTPS that are accurately templated at adjacent downstream
positions (i+2, i+3, and i+4, at a minimum) are required for
robust isomerization reversal at i+1 (G44) ......................... 150

UTP, templated at downstream sites, suppresses synthesis
of C45 and C46 ................................................................. 155

Evidence for a tunable ratchet driving translocation adjusted
by altering the concentrations of accurately templated
downstream NTPS ..................................................... 161

Evidence for a tunable ratchet driving translocation that is
regulated by TFIIS ............................................................. 171

A model for a tunable ratchet driving translocation by human
RNA polymerase II ............................................................ 177

 

KEY TO SYMBOLS AND ABBREVIATIONS

TF Transcription factor

RNAP RNA polymerase

TBP TATA-binding protein

mRNA Messenger RNA

CTD Carboxy—terminal domain

NTP Nucleoside triphosphate

NMP Nucleoside monophosphate
dNMP Deoxy nucleoside monophosphate
dNTP Deoxy nucleoside triphosphate
NDP Nucleoside diphosphate

EC Elongation complex

IR lsomerization reversal

PPi Pyrophosphate

xi

 

CHAPTER ONE

THE MECHANISM OF ELONGATION CATALYZED BY

MULTl-SUBUNIT RNA POLYMERASES

1. Transcription catalyzed by RNA polymerases

The regulation of gene expression is fundamental to normal growth,
development and survival of an organism. Gene regulation is executed mostly at
the level of transcription, the synthesis of RNA from a DNA template. This RNA
transcript is often processed during transcription to produce a mature RNA that
can itself be functional as a regulatory element, but for most genes is a
messenger (mRNA) for translation into protein. Transcription is carried out by the
enzyme RNA polymerase (RNAP). Numerous factors regulate transcription by
influencing the ability of the RNAP to access, bind and transcribe Specific genes
in response to appropriate Signals. The general transcription factors are involved
in various stages including recognition of promoter sequences, the response to
regulatory factors and conformational changes essential to the activity of RNAP
during the transcription cycle '(Conaway and Conaway, 1997; Kuhlman et al.,
1999; Ohkuma, 1997; Orphanides et al., 1996; Reinberg et al., 1998; Roeder,
T996; Sawadogo and Sentenac, 1990). Advances made over the past decades
have revealed several of the key general transcription factors (Borukhov and

Nudler, 2003; Murakami and Darst, 2003; Woychik and Hampsey, 2002), and

 

most recently, structures and models of RNAP ll interacting with general
transcription factors (Bushnell et al., 2004; Chen and Hahn, 2003; Chung et al.,
2003). Combined with biochemical and genetic studies, these structures provide
emerging views on the mechanism of the transcription machinery, the dynamic
nature of protein-protein and protein-DNA interactions involved, and the
mechanism of transcriptional regulation.

Although the transcription machinery of eukaryotes is much more
complex than that of prokaryotes or archaea, the general principles of
transcription and its regulation are conserved. Bacteria and archaea have only
one RNAP, whereas eukaryotes use three nuclear enzymes, RNAP H", to
synthesize different Classes of RNA. The nuclear RNAPS share five common
subunits, with the remainder showing strong similarity among the eukaryotic and
archaeal enzymes (Bell and Jackson, 1998; Lee and Young, 2000). Although
these enzymes have many more subunits than bacterial RNAP, subunits that
make up most of RNAP II are homologous to subunits from all cellular RNAPS,

suggesting that all these enzymes have the same or similar basic structure and

mechanism (Ebright, 2000). In bacteria, the a subunit is the sole general

transcription factor-like polypeptide. o recognizes promoter sequences,

promotes conformational changes in the RNAP-DNA complex upon initiation and

interacts directly with some transcription activators. In eukaryotes, a factor

function has been replaced by a much larger set of transcription factors, with

each of the three forms of RNAP having their own set of associated general

transcription factors (Grummt, 2003; Lee and Young, 2000; Schramm and
Hernandez, 2002). The RNAP ll transcription machinery is the most complex,
with a total of nearly 60 prOteinS, only a few of which are required for transcription
by the other nuclear RNAPS. In contrast, archaea use a simplified version of a
RNAP ll - RNAP Ill-like system, relying on only two essential general factors,
TBP (T ATA-binding protein) and TFB (related to the RNAP II and RNAP lll
general factors TFIIB and Brf1)(Bell and Jackson, 1998).

The three main phases of the transcription cycle are known as
initiation, elongation and termination, and each stage is subject to regulation.
During transcription initiation, a transcription-competent RNA polymerase
complex forms at the promoter and the DNA template is aligned in the active Site
of the polymerase. The active Site is where nucleotides are paired with the
template and are incorporated processively during elongation to produce the
RNA transcript. Termination of transcription involves release of the RNA
transcript and the dissociation of the transcription complex from the DNA
template.

Recent studies have Challenged the once commonly held view that
transcription is predominantly regulated at the level of RNA polymerase
recruitment to the promoter and it is now Clear that regulation of transcription at
the level of elongation is also important. In this study, I focus on RNA polymerase
II (RNAP ll) transcription elongation, which can be divided into three distinct
stages: promoter escape, promoter-proximal pausing, and productive elongation.

Each of these stages is defined by a marked difference in the stability and

behavior of the transcription complex, as well as a distinct group of factors that
associate with it.

RNAP II is the 12-subunit enzyme that synthesizes all mRNA-encoding
genes and some other genes that encode structural or regulatory RNAs. A
striking feature that distinguishes RNAP II from the other two eukaryotic RNAPS
is the extended carboxy-terminal domain (CTD) of pr1, the largest RNAP ll
subunit. The RNAP II CTD consists of tandem heptapeptide repeats of the
consensus sequence, YSPTSPS (Prelich, 2002). The CTD is thought to extend
from the core of the polymerase, and is subject to modification during the entire
transcription cycle. Modification of the CTD markedly affects its conformation and
ability to associate with factors that are involved in transcription elongation, RNA
processing and termination. Therefore, modification of the CTD is important for
the coordination of transcription events, and different modification states of the
CTD are Characteristic of different transcriptional stages.

A hallmark of RNAP Il during the transition from transcription initiation
to transcription elongation is a change from a hypophosphorylated (RNAP MA) to
a hyperphosphorylated (RNAP IIO) form (Sims et al., 2004). Phosphorylation of
the CTD occurs predominantly at Ser2 and Ser5 by CTD-modifying enzymes
(Meinhart et al., 2005). The level of Ser5 phosphorylation peaks early in the
transcription cycle and remains constant or decreases towards the 3' end of the
gene (Komamitsky et al., 2000). By contrast, Ser2 phosphorylation predominates
in the body and towards the 3' end of the gene and occurs concomitantly with

productive elongation (Komamitsky et al., 2000).

In eukaryotes, RNAP |l transcribes a DNA template that is associated
with a complex mixture of proteins, which is known as chromatin. The main
constituent of chromatin is the nucleosome, which consists of 147 base pairs of
DNA wrapped around an octamer of histone proteins. The mechanics of RNAP II
transcription requires marked perturbation of chromatin. Chromatin in transcribed
regions of the genome is more decondensed. Recent genomic analysis in
Saccharomyces cerevisiae have revealed that genes, especially highly
transcribed genes, have lower nucleosome occupancy than intergenic regions,
with promoter regions Showing evident nucleosome depletion (Pokholok et al.,
2005). Nevertheless, nucleosomes are usually present in coding regions and
pose a barrier to transcription elongation that must be overcome. Several protein
factors regulate Chromatin structure during transcription by covalently modifying
histones or by temporarily moving or disassembling and reassembling
nucleosomes (Henikoff and Ahmad, 2005). Several recent studies have revealed
important insights into the regulation of chromatin structure during transcription

elongation (Henikoff and Ahmad, 2005; Khorasanizadeh, 2004).

2. a-amanitin: the specific and potent inhibitor of RNAP II

a—amanitin selectively and specifically inhibits transcription by RNAPII
(Kedinger et al., 1970; Lindell et al., 1970) and RNAP lll (Weinmann and Roeder,
1974), RNAP ll being the most sensitive. a-Amanitin is a cyclic peptide which
binds to the pr1 subunit with high affinity (Kd ~1o-9 M) (Cochet-Meilhac and

Chambon, 1974; Lutter, 1982). After three decades of study, however, the mode

of this toxin action has not been resolved. Recently, an X-ray structure was
published of a-amanitin soaked into crystals of yeast RNAP II, indicating many of
the key atomic contacts that contribute to RNAP ll inactivation (Bushnell et al.,
2002) . Multiple strong interactions between a—amanitin and RNAP II are found in
the "funnel," the "cleft," and the key "bridge a-helix" regions of the pr1 subunit,
which constitute the RNAP ll "secondary pore" or "pore 1," a solvent accessible
channel into the RNAP ll active site. Contacts to the bridge helix are on the
opposite face from the RNAP ll active Site, indicating that a-amanitin is unlikely to
block catalysis directly. Bound a-amanitin partially occludes the pore, which has
been presumed to be a Channel for NTP substrate loading (Bushnell et al., 2002;
Cramer et al., 2001; Gnatt et al., 2001) but may be primarily the route of
pyrophosphate release (Nedialkov et al., 2003). Contacts of a-amanitin to the
bridge helix are intriguing because bending of the bridge a-helix is thought to
drive translocation of the RNA-DNA hybrid (Gnatt et al., 2001; Holmes and Erie,
2003; Nedialkov et al., 2003; Vassylyev et al., 2002). Translocation is essential
during each bond addition cycle to move the RNA-DNA hybrid away from the
space that must be Cleared to load the next substrate NTP into the active site.
Because a-amanitin binds to RNAP II with a relatively straight conformation of the
bridge helix and because of key contacts between the bridge helix and a-
amanitin, Kornberg and co-workers suggested that a-amanitin may block
translocation by blocking bridge helix bending (Bushnell et al., 2002). Several
other reports, however, indicate that RNAP ll may be capable of forming multiple

phosphodiester bonds without dissociation of a-amanitin (Chafin et al., 1995;

Rudd and Luse, 1996), so a-amanitin may allow occasional translocation without

dissociation from RNAP ll.

3. Translocation models

Mechanisms that describe translocation by multi-subunit RNA
polymerases include the power stroke model, the Brownian/thermal ratchet
model, the allosteric model, the pre-insertion model, the NTP-driven translocation
model, and the “trigger loop-trigger helices” model (Abbondanzieri et al., 2005;
Artsimovitch and Vassylyev, 2007; Bar-Nahum et al., 2005; Burton et al., 2005;
Epshtein et al., 2002; Foster et al., 2001; Gong et al., 2005; Holmes et al., 2003;
Landick, 2004; Oster, 2002; Sousa, 2003; Vassylyev and Artsimovitch, 2005;
Xiong and Burton, 2007).

Powerstroke and Brownian ratchet models have been invoked to
describe translocation by molecular motors. In a powerstroke mechanism,
translocation is tightly coupled to an energy generating reaction, such as NTP
hydrolysis. In many cases, however, molecular motors appear to keep separate
the power generating step from the translocation step. For RNA and DNA
polymerases, separating translocation from NTP hydrolysis may be essential to
ensure fidelity of NMP or dNMP incorporation (Burton et al., 2005; Gong et al.,
2005; Xiong and Burton, 2007). In a Brownian ratchet mechanism, translocation
is rapid and spontaneous, driven by thermal motions. After translocation, thermal
ratchet mechanisms require locking of the post-translocation state, is by NTP

binding. In general, molecular motors appear, however, to utilize binding of NTP

substrates as allosteric effectors to drive forward translocation. Hydrolysis Of the
NTP is dissociated from the translocation step, because linking NTP hydrolysis
directly to energy generation for translocation might constitute a low fidelity
mechanism. In multi-subunit RNA polymerase mechanisms, translocation is
directly linked to fidelity through allosteric coupling (Gong et al., 2005; Xiong and
Burton, 2007). Translocation may be a more deliberate and accurate process
than could be described either by a pure power stroke or thermal ratchet model,
lacking an allosteric component. The secondary pore NTP loading model
requires a rapidly reversible Brownian ratchet (Figure l-1, Images in this
dissertation are presented in color.), but dynamic studies of elongation may
not be consistent with rapid reversibility of translocation or spontaneous
pyrophosphate release (Burton et al., 2005; Gong et al., 2005; Xiong and Burton,
2007; Zhang and Burton, 2004; Zhang et al., 2005).

The NTP-driven translocation model has features of a power stroke,
thermal ratchet, and allosteric model (Burton et al., 2005; Gong et al., 2005;
Xiong and Burton, 2007). The energy generating reaction is NTP hydrolysis
during phosphodiester bond formation, and NTP hydrolysis is separated in the
mechanism from the primary translocation step, to ensure fidelity. In the NTP-
driven translocation mechanism, accurate NTP loading is coupled to
translocation, limiting the possibility of NTP loading errors. At a stall position, in
the absence Of the next NTP substrate, the thermal ratchet is revealed, although,

during rapid elongation, translocation appears to be NTP-driven. At a stall, once

Figure M. The secondary pore NTP loading hypothesis. The bridge a-helix is

blue (viewed from the C-terrninal end). The DNA template strand is green.
Phosphates are gold. The RNA strand is red. Consequences of the secondary

pore NTP loading model are indicated.

 

main enzyme channel G

bridge helix . -
A

e U

 
  

 

secondary pore

blocked . DNA

P G
pf AP RNA

pf PC Pre-translocated

aquSJaAaJ ‘pld 3
1914012.: ueiummg

Figure M.

10

Figure l-1 (continued).

6|C|ISJeAaJ ‘pidea
teqoieJ ueiuMOJa

 

main enzyme channel

 

   

P.pPG

Post-translocated

11

Figure l-1 (contn’d).

 

 

main enzyme channel

bridge helix .
T

.A

 

 

 

 

  
 
 

 

secondary pore

RNA
Post-translocated

12

 

Figure I-1 (continued).

Secondary pore NTP loading hypothesis

NTPS load at random to i+1 (active site)

NTPS usually mis-load (4 NTPS)

NTPS must load through the secondary pore

NTPS load only to the post-translocated EC

NTPS never load to the pre-translocated EC:
pore is blocked by RNA

“Entry” site: backwards NTP near active site

NTPS flip from the “E” site to the “A” site

Incorrectly loaded NTPS must be expelled and replaced
through the pore

Translocation is spontaneous, rapid, reversible
(not observed)

Competition by non-templated NTPS (not observed)

NTP loading must be rate-limiting (not observed)

Low fidelity mechanism

Fidelity through trigger loop closing?

13

 

pyrophosphate has been slowly but Spontaneously released, the elongation
complex is observed to ratchet between pre- and post-translocated states. Rates
of Spontaneous translocation, however, appear to be slow compared to NTP-
driven translocation (Burton et al., 2005; Gong et al., 2005; Nedialkov et al.,
2003; Xiong and Burton, 2007; Zhang and Burton, 2004; Zhang et al., 2003;
Zhang et al., 2005). During ongoing synthesis, accurate binding of NTP
substrates at the i+2 and i+3 downstream sites drives fonrvard translocation and
drives transfer of the i+2 NTP to the i+1 (substrate) position. In this manner, an
accurately paired NTP at i+2 becomes an allosteric effector for its transfer into
the active Site.

Erie has proposed an allosteric mechanism, which may be similar to
the NTP-driven translocation model, to describe elongation by E. coli RNA
polymerase (Foster et al., 2001; Holmes et al., 2003; Holmes et al., 2006). Erie
describes an “activated” state and an “unactivated” state for elongation. By
considering the activated state to be the post-translocated state and the
unactivated state to be the pre-translocated state, Erie’s model becomes a
version of the NTP-driven translocation model. Othenlvise, the Erie model
appears to require two templated NTP substrates (allosteric and substrate NTPS)
interacting Simultaneously at one template position, a model that does not appear
to make physical sense. So far, Erie and co-workers have reported kinetic
analyses of a single phosphodiester bond using a single quenching technique
(EDTA). Part of the difficulty in interpreting Erie kinetic experiments derives from

the lack of HCI quench data and data through formation of at least two

14

 

phosphodiester bonds. Availability of these data will Significantly Simplify the
comparison of elongation catalyzed by eukaryotic and prokaryotic multi-subunit
RNA polymerases. For human and yeast RNA polymerase II, the processive
transition (translocation linked to pyrophosphate release) is rate-limiting (Zhang
and Burton, 2004; Zhang et al., 2005). Monitoring only a Single bond, how can
one argue that a translocation event has been observed for the E. coli enzyme?
What if the E. coli enzyme initiates elongation from a stall position only in the
post-translocated state? At high NTP concentrations, E. coli appears to launch
the elongation reaction from a Single catalytic state (i.e. the post-translocated
state) (Foster et al., 2001), in contrast to human RNA polymerase II, which may
commence the reaction from both the pre- and post-translocated states (Zhang
and Burton, 2004; Zhang et al., 2005). At low to moderate NTP concentrations,
E. coli RNA polymerase can demonstrate bi-phasic kinetics, indicating that two
distinct elongation modes can be resolved (Foster et al., 2001).. Bi-phasic rate
curves at lower NTP concentrations indicate the “activated” and “unactivated”
pathways for elongation. Whether these faster and slower kinetic modes reflect
reversibility between post— and pre-translocation states or relate to a
characteristic of the EDTA quenching technique (i.e. Chelating an i+2 NTP-Mg2+
rather than the i+1 NTP-Mg”, which may be protected within the active site) is
not yet clear.

Cramer, Vassylyev, and Artsimovich have proposed the pre-insertion
model to describe elongation (Artsimovitch and Vassylyev, 2006; Artsimovitch et

al., 2005; Kettenberger et al., 2004; Vassylyev and Artsimovitch, 2005). This

15

model was initially based on an X-ray crystal structure from the Cramer
laboratory in which an NTP analogue assumes a position (“P site” for pre-
insertion Site) that may be inconsistent with NMP incorporation at i+1
(Kettenberger et al., 2004). Cramer therefore suggests that the NTP is held in a
pre-insertion site from which transfer occurs to the insertion A Site. From
crystallographic studies of the non-homologous bacteriophage T7 RNA
polymerase, strong evidence for a pre-insertion model was obtained (Landick,
2004; Tahirov et al., 2002; Yin and Steitz, 2002). For T7 RNA polymerase, the Y
helix can shift position to lock the pre-insertion NTP—dNMP base pair into the
active Site (insertion Site) for Chemistry.

Newer elongation complex structures from the Vassylyev (Vassylyev et
al., 2007; Vassylyev et al., 2007) and Kornberg laboratories (Wang et al., 2006)
indicate that conformational dynamics of the [3’ and pr1 trigger loop may
account for pre-insertion (also called pre-selection) and insertion site
intermediates. The pre-selection intermediate appears to have a relaxed trigger
loop structure. The insertion site structure is characterized by a Closed
conformation in which the trigger loop forms extended helices (“trigger helices”)
that enclose the NTP substrate within the active Site.

The trigger loop-trigger helices model is consistent with secondary
pore NTP loading, but does not necessarily require NTP loading by this route.
NTPS might load through the secondary pore, first to the E site. The NTP could
then flip into the P or PS (pre-insertion or pre-selection site). Tightening of the

trigger loop-trigger helices assembly would close and lock the active center for

16

chemistry. MiS-loading of an NTP would be expected to block Closing of the
trigger helices, resulting in release and ultimately accurate replacement of the
NTP. Loading of the accurately templated substrate NTP would permit formation
of the Closed A site structure. After NMP addition to the nascent RNA chain, the
trigger loop might be expected to relax. Relaxation of the active Site would allow
pyrophosphate release and release of the thermal ratchet for translocation. It
appears that an open trigger loop conformation would be required for NTP
loading through the secondary pore. In recent T. themiophilus RNA polymerase
structures, there does not appear to be sufficient space to load NTP substrates
through the main enzyme Channel. Also, the downstream transcription bubble is
closed, preventing NTP base-pairing to the DNA template at the i+2 and i+3
positions. The trigger loop-trigger helices model has aspects of secondary pore
NTP loading, thermal ratchet translocation, and induced fit to ensure
transcriptional fidelity. This model indicates that most NTP selection occurs in
and in close proximity to the active site.

NTPS may load through the secondary pore to the “entry" Site or “E”
Site before flipping into the pre-insertion site and/or the insertion site (Figure l-1)
(Bar-Nahum et al., 2005; Batada et al., 2004; Cramer et al., 2001; Gnatt et al.,
2001; Westover et al., 2004; Zhang et al., 1999). The secondary pore presents a
negative electrostatic barrier to NTP entry (Batada et al., 2004). NTPS can lead
through the secondary pore when the trigger loop-trigger helices assembly is in a
relaxed conformation. When the trigger helices are in the closed conformation,

there is not very much space to load an NTP through the secondary pore. In

17

recent T. thermophilus RNA polymerase structures (Vassylyev et al., 2007;
Vassylyev et al., 2007), there does not appear to be much space to load NTPS to
the active site through the main enzyme Channel, indicating that, at least for
bacteria, NTP loading most likely proceeds through the secondary pore.

Nudler and Ruckenstein have proposed a dual ratchet mechanism to
describe translocation (Bar-Nahum et al., 2005). Because the dual ratchet model
relies on a forward driving ratchet, which is suggested to be the pr1 bridge a-
helix driven by a conformational Change of the pr1 G-loop (trigger loop), this
model has elements of power stroke, ratchet, and allosteric Changes.
Experimental support for the dual ratchet model was primarily derived from
steady state experiments, so dynamic effects have been inferred from steady-
State approaches, including cross-linking experiments. Kinetic modeling was
based on data from the Erie laboratory for wild type RNA polymerase, and
comparative data were not reported for B’ I1134V and G1136S, the Nudler slow
and fast elongation mutants, which locate to the trigger loop (G-loop) region. The
dual ratchet mechanism relies on secondary pore NTP loading, a thermal ratchet
to support translocation, and dynamics of the trigger loop and bridge a-helix in

translocation and active Site tightening.

4. The NTP-driven translocation model
4.1. The secondary pore NTP loading hypothesis
The secondary pore NTP loading hypothesis and its predictions are

summarized in Figure l-1. According to this model, NTPS load to the RNA

18

polymerase II active site only through the secondary pore, never through the
main enzyme channel. Based on available structures, secondary pore NTP entry
appears to require loading to the post-translocated elongation complex. In the
pre-translocated elongation complex, the template DNA base faces the main
enzyme channel, not the secondary pore, so NTP loading to the pre-translocated
elongation complex appears impossible when NTPS load through the pore.
Single molecule elongation studies indicate that NTPS may lead to a pre—
translocated elongation complex (Abbondanzieri et al., 2005), potentially
contradicting the secondary pore NTP entry model.

Because NTPS are posited to load directly to the i+1 active Site, and
NTPS are selected at random from solution, miS-loading must be more frequent
than accurate NTP loading. Significant competition by non-templated NTPS is
expected, unless each selection of NTPS involves many NTP encounters.
Another consideration is that the pore is a 15 angstrom deep and 7 angstrom
wide Channel to the active Site from the “funnel” on the outside of the RNA
polymerase II structure (Batada et al., 2004). This Channel is only wide enough to
admit a single NTP at one time (NTPS have a 6 angstrom minimal diameter).
When an incorrect NTP is loaded (the usual case according to the secondary
pore NTP entry model), the NTP must be rejected at the active site and then
released through the pore, prior to loading another incorrect or correct NTP.
Bacteriophage T7 RNA polymerase, which is not a homologue of multi-subunit
RNA polymerases, appears to load NTPS through a deep and narrow secondary

pore.

19

Using X-ray crystallography, NTPS have been visualized in different
positions near the RNA polymerase II active Site: 1) the “E” site (for “entry” site);
and 2) the “A” Site for “acceptor” Site or insertion site (Kettenberger et al., 2004;
Wang et al., 2006; Westover et al., 2004). In this review, we consider a “P” Site
(for “pre-insertion” Site) structure to be a variant of the “A” Site structure
(Kettenberger et al., 2004). Closed trigger loop-trigger helices conformations are
probably the best approximations to “A" site structures. Relaxed trigger loop
conformations Characterize “P” site structures.

In E site structures, base-pairing functions of the NTP base face away
from the DNA template strand, so base-pairing to template is impossible in the E
site (Wang et al., 2006; Westover et al., 2004) (Figure M). An NTP in the E Site,
therefore, cannot recognize whether or not it is accurately templated. The E site
structure has been obtained: 1) using NTPS that are not accurately templated at
i+1; or 2) using the templated NTP in the presence of elevated Mg2+ (80 mM).
The E site may represent an “exit” Site in addition to an “entry” site (Wang et al.,
2006; Westover et al., 2004). Closed trigger loop-trigger helices structures have
an accurately paired NTP (or analogue) near the active site.

The secondary pore NTP loading hypothesis indicates that non-
templated NTPS compete for binding to the E and P Sites. Such competition is
not observed (Gong et al., 2005; Xiong and Burton, 2007), indicating that multiple
NTPS can be tested at the active site in each NTP loading through the pore.
NTPS that are accurately templated at downstream sites, but not at the i+1 site,

stimulate transcription upon i+1 NTP addition, rather than compete for NTP entry

20

(Gong et al., 2005; Xiong and Burton, 2007). This result does not appear
consistent with the secondary pore NTP loading hypothesis. According to the
secondary pore NTP loading hypothesis, translocation must be rapidly reversible
and pyrophosphate release must be rapid and Spontaneous. Kinetic studies,
however, indicate that translocation coupled to pyrophosphate release may be a
rate-limiting step during RNA synthesis (Zhang and Burton, 2004; Zhang et al.,
2005). Because of secondary pore dimensions and electrostatics, the secondary
pore NTP loading hypothesis appears to require that the rate of NTP loading be
rate-limiting (Batada et al., 2004). The rate for NTP loading to the i+1 active site
of human RNA polymerase II, however, is at least 1450 +l- 330 s", so NTP
loading is not rate-limiting (Zhang and Burton, 2004). The rate limiting steps in
RNA synthesis are approximately 20-30 5". Dynamic studies, therefore, appear

inconsistent with the secondary pore being the sole route for NTP loading.

4.2. The NTP-driven translocation hypothesis

A simpler model to describe NTP loading and translocation has
recently been hypothesized (Figure l-2). The NTP-driven translocation model
posits that NTPS interact at downstream positions i+2, i+3, and possibly i+4. NTP
substrates are sorted at a downstream site, is. i+3 or possibly i+4, so little
competition for NTP loading is observed at the i+1 active site (Burton et al., 2005;
Gong et al., 2005; Nedialkov et al., 2003; Xiong and Burton, 2007; Zhang and
Burton, 2004; Zhang et al., 2005). NTPS enter the i+1 active site by transfer

during translocation paired with their cognate dNMP base. Pyrophosphate

21

Figure l-2. The NTP-driven translocation model. Colors are as in Figure l-1.

Consequences of the NTP-driven translocation model are indicated.

22

 

 

 

M9
main enzyme channel
bridge helix
0
DNA
. P secondary pore RNA
p
Post-translocated

Figure l-2.

23

Figure l-2 (continued).

NTP-driven translocation model

Little competition by non-templated NTPS

NTPS at downstream sites (confirmed)

Templated pre-loading and pre-sorting of NTPS

Translocation is coupled to PPi release
(supported by kinetic studies)

PPi release through secondary pore

Main channel NTP loading

Translocation coupled to fidelity (confirmed)

Transcription errors sensed as translocation blocks
(confirmed)

High fidelity mechanism

24

release is coupled to NTP-driven translocation (Burton et al., 2005; Gong et al.,
2005; Xiong and Burton, 2007; Zhang et al., 2005). Pyrophosphate is eliminated
through the secondary pore, consistent with the constrictive dimensions and
repulsive electrostatics of the pore (Batada et al., 2004). Flow of NTPS into the
enzyme and excretion of pyrophosphate through the pore are maintained in a
consistent direction.

If the i+2 NTP can be shown to affect incorporation of the NTP at i+1,
this supports the NTP-driven translocation model and appears less consistent
with the secondary pore NTP loading model. Gong et al. (2005) have indicated
that, in the presence of the q-amanitin translocation block, i+2 and i+3 NTPS
appear to force rejection of a large fraction of stably loaded i+1 NTP (Gong et al.,
2005; Xiong and Burton, 2007). NTPS that are not templated at the i+2 and i+3
positions do not induce i+1 NTP rejection, demonstrating base-pairing specificity
at i+2 and i+3 Sites. Furthermore, dNTPS will not substitute for NTPS at i+2 and
i+3 downstream positions, demonstrating Chemical selectivity at downstream
Sites (Gong et al., 2005; Xiong and Burton, 2007). These experiments appear to
support the NTP-driven translocation model.

NTP-driven translocation was initially hypothesized based on dynamic
studies of RNA polymerase II elongation (Nedialkov et al., 2003; Zhang and
Burton, 2004; Zhang et al., 2005). Using the method of rapid Chemical quench
flow with millisecond reaction timing, the Burton laboratory has observed rapid
elongation by multi-subunit RNA polymerases through formation of multiple

bonds. Using a combination of reaction quenching (stopping) techniques, stable

25

NTP loading (EDTA quench) can be tracked independently from phosphodiester
bond formation (HCI quench) (Arnold and Cameron, 2004; Arnold et al., 2004;
Gong et al., 2004; Gong et al., 2005; Xiong and Burton, 2007; Zhang and Burton,
2004; Zhang et al., 2005). NTP-Mg2+ loading to the active Site can be tracked by
EDTA quenching because free NTPS are disabled by Mg2+ chelation.
Phosphodiester bond formation is tracked by HCI quenching, which stops the
reaction instantly. Interestingly, EDTA quench and HCI quench curves are well
separated, Showing that NTP-Mg2+ can be stably sequestered in the active site
prior to formation of a new bond (Zhang and Burton, 2004; Zhang et al., 2005).
After formation of the bond, another delay is observed before the next NTP-Mg2+
substrate is stably loaded into the active Site and becomes resistant to EDTA
quenching. These experiments Clearly reveal two rate-determining steps in each
bond addition cycle during ongoing RNA synthesis. There is a rate-limiting step
between stable NTP-Mg2+ loading in the active Site and phosphodiester bond
formation. It is likely that this delay represents an “induced fit” step in which the
NTP-Mg2+ is tightened prior to incorporation of NMP into the growing RNA chain.
Substrate tightening is likely to represent a fidelity Check to ensure accurate NMP
incorporation, because a mis-paired NTP would not support tightening. The
second rate-limiting step during elongation occurs after phosphodiester bond
formation but before the next NTP-Mg2+ is stably loaded (Zhang and Burton,
2004; Zhang et al., 2005). Based on the RNA polymerase mechanism, this step
must be translocation, pyrophosphate release, a reverse isomerization (i.e.

relaxation of the trigger loop) or NTP loading. This step is unlikely to be NTP

26

loading, because this step can be very rapid (>1450 S-1). The secondary pore
NTP loading model demands that translocation be rapidly reversible and that
pyrophosphate release be rapid and spontaneous. Neither of these requirements
of the secondary pore NTP loading model is supported by RNA polymerase II
elongation kinetics (Zhang and Burton, 2004; Zhang et al., 2005). The secondary
pore NTP-loading model appears to require that NTP loading be rate-limiting
(Batada et al., 2004). Stable NTP-Mg2+ loading occurs with a rate constant of at
least 1450 +/- 330 s'1 (Zhang and Burton, 2004), at least 50 times faster than can
easily be accommodated by the secondary pore NTP loading hypothesis (Batada
et al., 2004).

NTP-driven translocation is expected to be a high fidelity mechanism.
The secondary pore NTP loading mechanism, by contrast, may be a low fidelity
mechanism. According to NTP-driven translocation, incorrect NTPS are rarely
loaded into the active Site because of NTP pre-screening at downstream Sites.
dNTPS are excluded at downstream sites (Gong et al., 2005; Xiong and Burton,
2007), ensuring that multi-subunit RNA polymerases synthesize RNA and not
DNA. Furthermore, NTP-driven translocation appears to be coupled to error
prevention and error correction mechanisms. Experiments with a-amanitin
indicate that transcription errors are sensed as translocation blocks and that
NTP-driven translocation against a translocation block can lead to dynamic error
prevention and error correction (Gong et al., 2005; Xiong and Burton, 2007).

In “isomerization reversal” experiments (Gong et al., 2005; Xiong and

Burton, 2007), using an a-amanitin block and an EDTA quench procedure,

27

u-.'"l.'7¢'l*. .-

 

translocation driven by downstream NTPS induces reversal of stable loading of
the i+1 NTP-M92”. If downstream i+2 and i+3 NTP-Mg” are disabled at an early
time (i.e. 0.002 s) by EDTA chelation, little i+1 NTP-Mg2+ is released from the
active site, resulting in robust bond synthesis. If downstream NTP-Mg2+ is
allowed to strain against the a-amanitin translocation block for a longer period of
time (i.e. 0.1 s), a significant percentage of i+1 NTP-Mg2+ is released (Gong et
al., 2005; Xiong and Burton, 2007). lsomerization reversal experiments Show that
incoming NTPS are used to sense a translocation block, which normally would
signal a transcription error. Incoming NTPS help to prevent or correct the error by
dislodging the i+1 NTP from the active site. The secondary pore NTP loading
hypothesis appears to be a low fidelity mechanism because it requires frequent
mis-loading of NTPS to the i+1 active Site (the usual case according to the
secondary pore NTP loading model). The secondary pore NTP loading
hypothesis cannot allow for coupling of translocation to transcriptional fidelity, as

in the NTP-driven translocation model (Figures M and 2).

5. NTP loading into the RNA polymerase II active site through main
enzyme channel
The active site of a multi-subunit RNAP is deeply buried within the
structure. Whether NTPS load through the main enzyme channel or the
secondary pore, some explanation Of how NTPS navigate the constrained Space

to the catalytic center will ultimately be necessary.

28

The NTP-driven translocation hypothesis posits that RNAP ll may lead
at least three NTP substrates simultaneously to templated Sites (Burton et al.,
2005; Gong et al., 2005; Nedialkov et al., 2003; Xiong and Burton, 2007; Zhang
and Burton, 2004; Zhang et al., 2005). Substrates loaded to the active site as
NTP-dNMP base pairs, and translocation is driven by protein conformational
Changes induced by accurate NTP loading. In support of the model, Burton and
colleagues have demonstrated sequence-specific NTP interactions with template
at i+2 and i+3 downstream sites while the i+1 active site is occupied (Burton et
al., 2005; Gong et al., 2005; Xiong and Burton, 2007).

If the NTP-driven translocation hypothesis is correct, then NTP
substrates have to be first loaded into the RNAP ll main enzyme channel and not
the secondary pore (pore 1). NTPS have been suggested to be loaded to the i+1
active site and the i+2, i+3, and, possibly, the i+4 downstream Sites (Burton et al.,
2005; Gong et al., 2005; Xiong and Burton, 2007). Binding strengths of NTPS are
likely to be tuned so that the i+1 Site has the highest affinity for an accurately
paired NTP, followed by the i+2 site and then the i+3 Site. During normal
elongation, in the presence of all four NTP substrates, NTPS are freely
exchanged at a downstream site, is. i+3 or beyond. Downstream template
access may be regulated by the non-template DNA strand, because closing of
the downstream bubble would disrupt pairing of downstream NTPS. At i+1, i+2,
and i+3 Sites, improperly paired NTPS are exchanged more rapidly than
accurately paired NTPS. Altering the order of affinities of these three sites (i.e.

through mutation or by mis-pairing the downstream DNA template) might-reduce

29

 

transcriptional fidelity, by forcing misloading of NTP substrates to the active Site.
For instance, enhancing NTP affinity at the i+3 Site through mutation would be
expected to lower discrimination between accurately and inaccurately paired
NTPS at the i+2 and i+1 sites, increasing misincorporation at i+1. This is a
possible interpretation of the yeast pr1 E1103G mutant RNAP II, which is a fast
mutant with defects in transcriptional fidelity. According to these views, NTP-
driven translocation is directly coupled to transcriptional fidelity mechanisms.

The NTP-driven translocation mechanism provides an explanation for
evolution of the vast, enclosed main enzyme Channel, which is a Singular
characteristic of multi-subunit RNAPS. According to this view, a major
consequence of the RNAP “crab claw” structure is to enhance transcriptional
fidelity by supporting and tuning NTP-driven translocation. During processive
synthesis, the enclosed main Channel space is occupied by at least two
templated NTP substrates, allowing verification and re-affirrnation of NTP identity
prior to active Site loading. Thus, multi-subunit RNAPS have evolved for high
fidelity and transcriptional efficiency. Multi-subunit RNAPS with different gene
Specificities (i.e. RNAPS l and III) or those derived from different organisms are
evolved to adjust error and elongation rates. Extensive conservation of multi-
subunit RNAPS through three kingdoms attests to the enduring adaptability of
this enzyme. Another reason for the enclosed Channel is to maintain high
processivity during elongation, by maintaining tight DNA and RNA contacts.

In some yeast RNAP ll elongation complex structures, the downstream

transcription bubble appears to be open to position i+4 (Gnatt et al., 2001;

30

 

Westover et al., 2004) or i+6 (Wang et al., 2006), consistent with NTP pairing to
the i+1, i+2 and i+3 DNA template Sites (Burton et al., 2005; Gong et al., 2005;
Xiong and Burton, 2007). After NMP incorporation at i+1, the NTP in the i+2
position is transferred to the i+1 Site during translocation. Movement of the i+3
NTP to the i+2 position vacates the i+3 site, which is filled by random exchange
of incoming NTPS. In support of this model, the Burton laboratory has
demonstrated that NTPS templated at the i+2 and i+3 sites can determine the
fate of the NTP loaded at the i+1 position, and incoming NTPS stimulate fonrvard
translocation and bond completion, which is pyrophosphate release (Burton et
al., 2005; Gong et al., 2005; Xiong and Burton, 2007; Zhang et al., 2005).

The RNAP ll active site iS buried deep within the enzyme structure,
limiting accessibility and free exchange of NTP substrates. We suggest that NTP
substrates may enter the enzyme through the main enzyme channel and not
necessarily through the secondary pore. In this manner, the negative
electrostatic potential of the pore (Batada et al., 2004) can be bypassed in NTP
loading. We posit that loading of NTP substrates to downstream DNA sites helps
to induce conformational Changes in the elongation complex that drive forward
translocation (Burton et al., 2005; Gong et al., 2005; Xiong and Burton, 2007)

and substrate NTP tightening.

31

 

6. The rock and roll model for RNA polymerase II translocation

The “rock and roll” model is proposed to describe the multi-subunit
RNAP catalytic mechanism in molecular details. The model, which might apply to
both eukaryotic RNAP II and bacterial RNAP, is shown schematically in Figure l-
3. “Rock and roll” refers to rocking of the “rocking assembly” (S. cerevisiae
rocking assembly I: prl 526 to 764, pr8, pr9 48 to 120; rocking assembly ll:
pr9 1 to 46, pr1 869 to 1063, pr1 1113 to 1308), stimulated by i+1 NTP

tightening and NTP-driven translocation. Motion of the rocking assemblies

induces reversible rolling of the “rolling assembly” (pr1 810 to 868 (bridge a-

helix and C-terminal cap), pr1 764 to 810, pr2 500 to 542 (“fork loop 2”
region), pr2 750 to 778, pr2 706 to 738, pr1 1384 to 1402, pr1 1063 to
1112) (Figures l-4, 5, 6 and 7). Briefly stated, we propose that at least three NTP
substrates can be on template simultaneously at i+3, i+2, and i+1 (the active
Site). We propose significant conformational coupling between the NTP substrate
at i+1 and downstream NTPS at i+2 and i+3. Essentially, forward translocation,
partly driven by i+2 and i+3 NTPS, is coupled to i+1 NTP tightening, which in turn
is coupled to conformational changes that drive stepped fonrvard translocation.
The rock and roll model unifies NTP-driven translocation and events at the active
site, explaining how NTP-driven translocation can be coupled to pyrophosphate
release (Gong et al., 2005; Xiong and Burton, 2007; Zhang et al., 2005). Close

resemblance of the schematic to the structure is indicated in Figure l-3F.

32

 

Figure l-3. Schematic diagram of rock and roll. A) Relaxed elongation
complex. B) Tightened elongation complex. C) Phosphodiester bond formation.
D) NTP-driven translocation coupled to switching DNA contacts and return of the
translocation arm. E) Relaxed elongation complex. F) Similarity between the
elongation complex structure and the schematic. Tightening of the i+1 NTP and
NTP-driven translocation cause the rocking assembly to rock forward and the

rolling assembly to roll forward. Tightening Of the i+1 NTP also draws the

translocation arm forward through a set Of anti-parallel B-sheets linked to active

Site loops. The translocation arm projects multiple basic groups to interact with
the DNA template. pr1 K752 is posited to be a major sensor of the stage of the

bond addition cycle.

33

     
   
 

rolling assembly

on arm

rocking

pr1
479-NADFDGD

pr1

836-ED-837

Relaxed EC

A

Figure l-3.

34

Figure l-3 (continued).

 

NTP Tightening

Figure I-3 (continued).

 

 

Figure l-3 (continued).

 

NTP-driven translocation,
switching of DNA contacts
return of translocation arm

D

37

Figure l-3 (continued).

 

Relaxed EC

38

 

Figure I-3 (continued).

\‘ ‘\\.

translocatéon arm

‘7

B-Sheets

 

39

Figure I-3A shows a relaxed elongation complex structure with an
ATP-Mg2+ loaded at i+1, a CTR—Mg2+ at i+2, and a CTP-Mgz” at i+3. Current x-
ray crystal structures with NTPS loaded at i+1 resemble the image shown in this
figure. The ATP-Mg2+ is in a relaxed conformation in the active Site. Mg-A is held
by pr1 479-NAQFQGQE-486. Mg-B is out of range of pr2 830-YSGYNQEQS-
838 (see Figure l-8; 2E2H structure). Figure I-3B depicts tightening
(“isomerization”) of the i+1 ATP-Mg”. Mg-A is held by pr1 479-NAQFQGQE—
486. Mg-B is held by pr2 830-YSGYNQ_EQS-838, as observed in the 2NVT
crystal structure (Figure l-8). pr2 R1020 and pr1 K752 are brought close to
the y-phosphate of the ATP. Contact of pr1 K752 with the y-phosphate of ATP
generates translocation force through movement of the helix 20-21-22-23
assembly (pr1 674 to 763), which includes K752 on the pr1 750-GSKG-753
active site loop. Figure l-QC indicates that further translocation force is generated
as the phosphodiester bond is formed. Furthermore, after Chemistry,
pyrophosphate is retained in the active site. This feature of the model is indicated
by kinetic studies, which Show that bonds can be formed and then reversed,
apparently through endogenous pyrophosphorolysis, which is detected as bond
reversal using a combination of EDTA quench and HCI quench procedures. By
coupling NTP-Mg2+ tightening, bond formation, and pyrophosphate extraction to
ongoing translocation, induced fit within the i+1 active site is coupled to
transcriptional fidelity. At any point in the catalytic cycle, failure to achieve optimal
geometry within the active site results in backwards winding of the ratchet and

substrate rejection. In Figures l-3A-3C, the translocation arm exerts forward

40

 

 

translocation pressure on the DNA template strand through lysine and arginine
contacts to DNA phosphates. Maximal translocation strain in the system occurs
after Chemistry but prior to the next NTP loading, which in this case requires
GTP-Mg2+ transfer from the i+2 downstream position to i+1. Figure l-BD shows a
feature of the model, which is NTP-driven translocation coupled to
pyrophosphate release. We have depicted the later stage of this transition as the
return of the rocking assembly and bridge helix ratchet to their relaxed positions.
Pyrophosphate is released, perhaps because Of return of the rocking assembly
and pr1 K752. Contacts between the translocation arm and the DNA template
must first be broken and then be re-set at the next downstream position.
Changing the bend point in the DNA template strand is likely to be an important
feature of re-setting phosphate contacts during return of the translocation arm. In
Figure l-3E, the relaxed elongation complex is Shown with GTP-Mgz“ in the i+1
active site. Figure l-3F shows a crystal structure that resembles the relaxed
complexes Shown in Figures l-3A and 3E.

Mutations in yeast RNAP ll residues that are defective in elongation
may support the rock and roll model. Remarkably, o-amanitin, a potent
translocation blocker (Gong et al., 2004), is poised to block interaction between
the rocking assembly and the rolling assembly, inhibiting rocking and preventing
rolling (Figure l-9). The structure of a-amanitin bound to RNA polymerase II and
the distribution and precise locations of mutant RNA polymerase II proteins
provide support for the rock and roll model. Trigger loop dynamics could be

incorporated into the “rock and roll” model.

41

Figure M. The rocking assembly rocks, and the rolling assembly rolls. The
left panel iS the view from the N-terminal end of the bridge a-helix (blue; pr1
810 to 845). The N-ten'ninal end of the bridge a-helix can be identified by the
“propeller” (P; pr1 813-FFFH-816). The right panel is the view from the C-
terminal end of the bridge a—helix. The i+1 NTP is orange. The i+2 NTP (placed
by modeling) is pink. The i+3 NTP (placed by modeling) is lime. The DNA
template strand is green. The RNA strand is red. The non-template DNA strand
is white. Phosphates along the DNA template strand are gold. M9 is green. Basic
amino acids are blue. pr1 K752 is proposed to be a sensor of the state of the
catalytic cycle. Hydrophobic amino acids are white. DNA Chains were extended

and disordered parts Of the non-template DNA strand were placed by modeling.

42

 

1.656 E.

43

Figure l-5. The rolling assembly. The image is the same as Shown in Figure l-
4 except the rolling assembly is shown in additional detail. The protein segment
that connects the rocking assembly to the bridge a-helix is shown in purple (pr1
764 to 809). The fork loop 2 region is Shown in cyan (pr2 500 to 542; proposed
i+2 NTP interaction region). The helix 36-trigger loop-helix 37 region is yellow

(pr1 1060 to 1112; proposed i+3 NTP interaction region).

44

 

Inc; I...

45

Figure l-6. Functional domains of the S. cerevisiae pr1 subunit. Active site
loops are indicated. In some cases, important residues are highlighted in red.
Relevant mutations are in red. Homologous residues and structures found in

bacterial RNA polymerase from T. thermophilus are indicated in green.

46

R326 (-)
K330 (-)

K332 (K610)
R337 (R615)
R344 (R622)
R350 (R628)

Translocation Arm

 

l
1 Active Site ]
1 a, .
( ) 347 /355 525
(625) (634) (778)
446-RQPSLHKMS-454
pr1 (684-RAPTLHRLG-692)

(6’ homolog) 479-NADFDGDE-486
(737-NADFDGDN-744)

Mg-A

 

Figure l-6.

47

 

Figure I-6 (continued).

pr5 interaction
ys35 (y1093) pr6 Interaction

R839 R1095 (co interaction
R840 IK1 097; C-terminal Cap on Aridge

K843 ('I Piston/Rolling Assembly
E846 (-)
819-GGREG-823
(1076-GARKG-108tl)
813-FFFH-816
(1070-YFI-1073)
Bridge Helix Ratchet
(Rolling Assembly)

 
   
 
  
    
 
   
  
 

 

pr8 interaction
673
(

 

  

Connector
(Rolling Asse bly)

Helix 20-21-22-23 Assembly
750-GSKG-753
(1 026-GARG-1029)
67300
$751 F
A759P
C764F

pr9 Interaction

48

Figure l-6 (continued).

 

 

 

 

 

 

 

 

  
  
 

i+3 NTP
P1099 (P1257)
R1100 (R1258)
K1102 (-)
E1103 E11036 (E1261)
N1106(-)
K1112 (K1269)
Helix 36-Trigger loop-Helix 37 R b5 interaction
(Rolling Assembly) RgbG interaction
(0) interaction)
1446
(1505)
,. _ _ _.-_. P'sm" “CTD
[ ,. I1327N
86 106 1 03 E1351K
(112 (12 0) b9 interac 'on M3979
E1351 (E1351)
""3 V1384 (V1424)
(1271) R1386 (K1426)
Rocking Assembly ll E11332 ((51429)
(Piston)
F1402 (F1440)
E1352A (E1219)
A1078V
S 095%:
R1231L

49

Figure l-7. Functional domains of the S. cerevisiae pr2 subunit. Colors are

as in Figure l-6.

50

I+2 NTP

     
   
   

R504 (R420)

pr2 0505 (E421)
K510 (-)

([3 homolog) R512 R512C (R428)

0513 (-)

Rolling Assembly

Upstream Lobe Fork LOOP;

43 212 409 452 4
542

(10) (136) (386 420)
(4 8)

    
      

213 5.223: 4“
(137) (420)
Downstream Lobe‘ ‘ \
pr9 interaction

Rolling Assembly
pr9 interaction

Linked Segments
pr2 1 to 42 and Fork LOOP 1
pr2 653 to 688 H

Figure l-7.

51

Figure l-7 (continued).

Upstream DNA interactions
pr12 interaction

/'

       
 
  
 

Active site , _ ,.»/
N\-terminal Cap 0'1 rind/0,9 , jfipbB interaction
//’ (to interaction)
\ ’// / TranslocationArm
\ TIA / / .

  

'\

R1129 (R1031)
R1135 (-)
D1136 (-)

(537) 6 ('33) 1103 1106
Active Site Loops (1005)(1oos)
pr3-pr11 interaction
(or2 interaction)
pr10 interaction

830-YSGYNQEDS-838
(679-FDGYNFEDA-687)

Mg-B?

979-KFASRHGQKG-988
(838-KLANRHGNKG-847)

1013-NPHAIPSRMT-1022
(872-N PLGVPSRMN-881)

A753V

H761Y

A777T

R983G

A1016T

P1018$

R983-E1028
(R842-E887)

52

Figure l-8. The two-metal catalytic mechanism for RNA polymerase ii and
the trigger loop hypothesis. Three identical views from Protein Data Bank

2NVT (upper panels) and 2E2H (lower panels) crystal structures are shown. The

2NVT structure has a straighter conformation of the bridge a-helix (blue), a more

ordered helix 36 (H36) (yellow), an “open” conformation of the trigger loop

(yellow), and a wider spacing of Mg-A and Mg-B (green; A and B). The 2E2H

structure has a bent conformation of the bridge a-hellx, a disordered helix 36, a

“Closed" conformation of the trigger loop, and little separation of Mg-A and Mg-B.
Acidic amino acids are red (pr1 D481, D483, D485 and pr2 E836, 0837).
The i+2 (pink) and i+3 (lime) NTPS were placed by modeling. Colors are the

same as in Figures l-4 and 5.

53

'8-I amfitd

54

’ffiéae

fi‘sd

 

Figure l-9. a-amanitin (red) is a cyclic octapeptide that blocks rocking of the
rocking assembly (mauve) and inhibits rolling of the rolling assembly. The
helix 36-trigger loop-helix 37 region (pr1 1060 to 1112) is yellow. The
connecting segment between the rocking assembly and the bridge a-helix (pr1
764 to 809) is cyan. The positions of three Clustered SIT mutants in pr1 rocking
assembly I (G730D, A759P, and C764F) border the a-amanitin binding site. i+2
and i+3 NTPS and portions of the non-template DNA strand and template DNA

strand were placed by modeling. Other colors are as in Figures I4 and 5.

55

 

Figure l-9.

56

7. Fidelity and efficiency

NTP-driven translocation coupled to rock and roll provides a new view
of the efficiency and fidelity of RNA synthesis (Burton et al., 2005; Gong et al.,
2005; Xiong and Burton, 2007). Pre-loading and pre-aligning of NTP substrates
results in highly efficient polymerization, causing RNA polymerase to resemble a
miniature industrial assembly line for building an RNA Chain. The RNA assembly
line metaphor also suggests mechanisms for sophisticated quality control in RNA
polymerization, some of which have been demonstrated to be novel error
prevention and error correction mechanisms (Gong et al., 2005; Xiong and
Burton, 2007). NTPS are tested for accurate base pairing at downstream i+3 and
i+2 sites, before loading into the i+1 active site, which is the only position at
which bond formation and hence transcription errors can occur. The triphosphate
tails of incoming NTPS appear to be scanned at the i+3 Site (yeast pr1 K1102,
N1106, and K1112) and again at the i+2 site (pr2 R504, R512, and 0513)
(Figures I-10 and 11). For RNA polymerase II, the 2’-Ol-l of the ribose ring of the
incoming NTP may be monitored at the i+2 position by an aspartic acid (pr2
D505) (Burton et al., 2005). An aspartate residue provides a similar function in
the pre-insertion site of RNA-dependent RNAP from (T ellez et al., 2006;
Thompson and Peersen, 2004). The space for NTP-dNMP passage from the i+2
to the i+1 Site is limited, so translocation through a constrained space is
proposed to provide a fidelity check, in which inappropriate base pairs are

disrupted. In the active Site, induced fit is utilized to confirm the accuracy of NTP

57

loading. Prior to pyrophosphate release, there is a mechanism for reversing
phosphodiester bond synthesis in response to a block to translocation (Gong et
al., 2005; Xiong and Burton, 2007). RNAP ll, therefore, senses a transcription
error as a translocation block. The purpose of the bond reversal mechanism is to
remove a 3’-terminal, improperly loaded or incorporated base from the RNA

chain prior to pyrophosphate release and essentially irreversible elongation.

58

Figure MO. The proposed i+2 NTP interaction site. pr2 R504, D505, R512,

and 0513 on fork loop 2 (cyan; pr2 504 to 513) are posited to make contacts to

the i+2 NTP. The bridge u-helix is yellow. NTPS and amino acids are colored by

Chemistry. Mg-A and Mg-B are purple. The white arrow indicates the 90 degree
bend in the DNA template over the bridge a-helix. The i+2 NTP was placed by

modeling.

59

.+._

 

10.

Figure l

60

Figure H1. The proposed i+3 NTP interaction site. pr1 K1102, N1106, and
K1112 are posited to make important contacts to the i+3 NTP. pr1 R840 on the
bridge a-helix (yellow) makes main chain contacts to pr1 N1106. The helix 36-
trigger loop-helix 37 region (pr1 1063 to 1112) is mauve. The fork loop 2 region

(pr2 500 to 542) is cyan. The i+2 and i+3 NTPS were placed by modeling.

61

 

l-11.

Figure

62

8. Conclusions

A model is proposed to unify kinetic, structural, evolutionary, and
genetic data on RNAP II. In addition to aiding in the interpretation of a vast
collection of available data, the model makes extensive and precise predictions
for future experiments. Based on evolutionary comparisons (Figures l-12, 13, 14,
6 and 7), the rock and roll model may be applicable to all multi-subunit RNAPS.
Multi-subunit RNAPS, found in eubacteria, archaea, and eukarya, appear to
represent a Significant evolutionary advance allowing development of complex
life on earth. Eubacteria may have existed on earth for 3.2 to 3.5 billion years,
indicating the age of multi-subunit RNAPS. The enclosed main enzyme Channel
may have evolved to support NTP-driven translocation coupled to rock and roll in
order to: 1) enhance transcriptional fidelity and efficiency; 2) support processivity;
and 3) enforce Single base stepping during elongation to prohibit unrestrained
sliding. Alternatively, the trigger loop-trigger helices model posits that NTP
selection and induced fit occur only within the enzyme active site and do not

involve downstream NTP interactions.

63

Figure 142. Six active site loops act as sensors of the phase of the bond
addition cycle and respond to i+1 NTP tightening. The active site is
conformationally coupled to the rocking and rolling assemblies. Black rectangles
indicate ion pairs. Ovals indicate adjacent residues. In this figure, dotted lines
indicate coordination to M9”. A = archaea, B = eubacteria, I, II, III = RNA
polymerases I, II, and III, to indicate conservation of residues (Similarity or

identity). Relevant mutations are indicated in red.

64

26.... x3:

.33 I mam» meow.» mo=§u>uecaz<

ZkSmfim >. m. _. =. _= >. w. _. =

we”... semiovmrz—AZmL—ms
>35 «.8 Son 86:80..

>342
masses
<302>m>mv>

38;:
”WW.“ _. New.“ gas 258 we...» aee.<mo<zcmom.8a
> m _ __ =_ > . = _= .. ,, seems, >35 28 Son 3:qu
4336—0638: we”: 3o.z>0_ncoomt8m was» mean
>3: >33 «:6 .oou 32.3.. 33 - - - :93 I - - mam
Zea/x >. w. _. =. _= >. m. _. =. _=
I I
was» was» was» . was.
2030 ”new I men ”33 m .+._ z._._u I qun . main
>.w._.__.___ >.m._.__._= Pm...__.___ 3:86.58 >.w._.=._=
3.5 33mm 55.. Hahn—6.43
recon—598 32> I .33 >353 >oa<c 9.3 .02. «250..
>. m. _. =. =_ wen—.59 >ueoBE<
m on was» 3..w.2_u_._>=umw§._..3~n 380
u u .
.1 2.6 I .63 >326 use .00: uosooq >386
>. m. _. =. .= 03%

”can ouc..am>mm:oox.oa.~
>226 was .00.. 333..

Figure l-12.

65

Figure l-13. Conformational coupling between the i+1 NTP, the rocking
assembly, the piston assembly, and the proposed i+3 NTP site. Symbols are
as in Figure l-12. The black bar indicates a proposed 35 angstrom Charge relay.
Parentheses indicate that one ion pair interaction (pr5 E148-pr1 K1350) is
Slightly out of range (6.5 angstroms) in available structures. Dotted lines indicate

structural proximity and probable coupled movement.

66

 

new.

38“.. <33>a>§>
>.h. __ msa~>
we... x2: was. 153: >mmoBE< >322
.58 I mace I wane 0 .332 .9566 was: .
__ . P..... >. _.. =. 5 Tu 21.583020: madam $0.55
. _ I mfioao
wee... was. use. use. was. wee. >335:

.. me; 3 x38 0 mad. I :38

mdwméx

, r

_. ., >693 «=0 52. , , ,
.. .. mi z._._u seamen ,
. . uuobmxohmu
. ., we...
.. mum: 0 .43» I .1 2.2.
. . . >. a. _. =. =_
was. was. new. .
033 I an: I m3».
>. _. =. =_ >. .. =. _= >. :4. =1
ENE—r Quwoo
>~moc
0.32”
moo—:3 Roe—=6; - , - , - - .

I micw I Zionlré z...“v

_. =_ _= >.. .4. =. =1 >. w. _. =. =_ >. a. _. =. _= >. w. _. =. =_ _. =. =_ Susana—58

, , ,. , - mug
, , . .3...»
w. =

we... been. we...

03;» I ”no: I 0.5.3
>. _. =. _= _. =. _= >. _. =. _=

I
was. 158.6
aim
>. _. =. _=
- - . 15335330?

Figure M3.

67

Figure l-14. Connections of the i+2 NTP interaction site on the rolling
assembly with the rocking assembly and active site. Symbols are as in

Figures l-12 and 13.

68

r.» 2:0 r.» 2:.
Eczema—.98 NRC...
I I
use» use»
>31 we? 0 case
wee» use» >. P... =. 2b.... __
5.2. I muua
Pm...__.=_ Pm...=.=_ wee». as!
u .33 I ma.) -
__ _. =
mean .4» 2:6 a» 2:.
:32 0 0.39 «6.32598 Egon—598
>. m. _. __ I I
>wa< I ”EON NEON
use» mine 03“.. O we.»
38 2...: = .. Pa.........
=4 m2: ”Sun
use» 28» use.
38 28a :3.
__.I maeboxmcbne
sec: 2.03:2...
use» >. m. _. =. =_
998

means 883.62

__ wanna :2...

ing
s in

- - ”005.5 >336?

Figure M4.

69

References:

Abbondanzieri, E. A., Greenleaf, W. J., Shaevitz, J. W., Landick, R., and Block, S.
M. (2005). Direct observation of base-pair stepping by RNA polymerase. Nature
438, 460-465.

Arnold, J. J., and Cameron, C. E. (2004). Poliovirus RNA-dependent RNA
polymerase (3Dpol): pre-steady-state kinetic analysis of ribonucleotide
incorporation in the presence of MgZ+. Biochemistry 43, 5126-5137.

Arnold, J. J., Gohara, D. W., and Cameron, C. E. (2004). Poliovirus RNA-
dependent RNA polymerase (3Dpol): pre-steady-state kinetic analysis of
ribonucleotide incorporation in the presence of Mn2+. Biochemistry 43, 5138-
5148.

Artsimovitch, l., and Vassylyev, D. G. (2006). Is it easy to stop RNA polymerase?
Cell Cycle 5, 399-404.

Artsimovitch, l., and Vassylyev, D. G. (2007). Merging the RNA and DNA worlds.
Nat Struct Mol Biol 14, 1122-1123.

Artsimovitch, l., Vassylyeva, M. N., Svetlov, D., Svetlov, V., Perederina, A.,
lgarashi, N., Matsugaki, N., Wakatsuki, S., Tahirov, T. H., and Vassylyev, D. G.
(2005). Allosteric modulation of the RNA polymerase catalytic reaction is an
essential component of transcription control by rifamycins. Cell 122, 351-363.

Bar-Nahum, G., Epshtein, V., Ruckenstein, A. E., Rafikov, R., Mustaev, A., and
Nudler, E. (2005). A ratchet mechanism of transcription elongation and its control.
Cell 120, 183-193.

Batada, N. N., Westover, K. D., Bushnell, D. A., Levitt, M., and Kornberg, R. D.
(2004). Diffusion of nucleoside triphosphates and role of the entry site to the RNA
polymerase II active center. Proc Natl Acad Sci U S A 101, 17361-17364.

Bell, S. D., and Jackson, S. P. (1998). Transcription and translation in Archaea: a
mosaic of eukaryal and bacterial features. Trends Microbiol 6, 222-228.

Borukhov, S., and Nudler, E. (2003). RNA polymerase holoenzyme: structure,
function and biological implications. Curr Opin Microbiol 6, 93—100.

Burton, 2. F., Feig, M., Gong, X. 0., Zhang, C., Nedialkov, Y. A., and Xiong, Y.
(2005). NTP-driven translocation and regulation of downstream template opening
by multi-subunit RNA polymerases. Biochem Cell Biol 83, 486-496.

70

Bushnell, D. A., Cramer, P., and Kornberg, R. D. (2002). Structural basis of
transcription: alpha-amanitin-RNA polymerase II cocrystal at 2.8 A resolution.
Proc Natl Acad Sci U S A 99, 1218-1222.

Bushnell, D. A., Westover, K. D., Davis, R. E., and Kornberg, R. D. (2004).
Structural basis of transcription: an RNA polymerase ll-TFIIB cocrystal at 4.5
Angstroms. Science 303, 983-988.

Chafin, D. R., Guo, H., and Price, D. H. (1995). Action of alpha-amanitin during
pyrophosphorolysis and elongation by RNA polymerase II. J Biol Chem 270,
19114-19119.

Chen, H. T., and Hahn, S. (2003). Binding of TFllB to RNA polymerase II:
Mapping the binding site for the TFllB zinc ribbon domain within the preinitiation
complex. Mol Cell 12, 437-447.

Chung, W. H., Craighead, J. L., Chang, W. H., Ezeokonkwo, C., Bareket-Samish,
A., Kornberg, R. D., and Asturias, F. J. (2003). RNA polymerase llfl'FllF
structure and conserved organization of the initiation complex. Mol Cell 12, 1003-
1013.

Cochet-Meilhac, M., and Chambon, P. (1974). Animal DNA-dependent RNA
polymerases. 11. Mechanism of the inhibition of RNA polymerases B by
amatoxins. Biochim Biophys Acta 353, 160-184.

Conaway, R. C., and Conaway, J. W. (1997). General transcription factors for
RNA polymerase II. Prog Nucleic Acid Res Mol Biol 56, 327-346.

Cramer, P., Bushnell, D. A., and Kornberg, R. D. (2001). Structural basis of
transcription: RNA polymerase II at 2.8 angstrom resolution. Science 292, 1863-
1876.

Ebright, R. H. (2000). RNA polymerase: structural similarities between bacterial
RNA polymerase and eukaryotic RNA polymerase II. J Mol Biol 304, 687-698.

Epshtein, V., Mustaev, A., Markovtsov, V., Bereshchenko, 0., Nik'rforov, V., and
Goldfarb, A. (2002). Swing-gate model of nucleotide entry into the RNA
polymerase active center. Mol Cell 10, 623-634.

Foster, J. E., Holmes, S. F., and Erie, D. A. (2001). Allosteric binding of
nucleoside triphosphates to RNA polymerase regulates transcription elongation.
Cell 106, 243-252.

Gnatt, A. L., Cramer, P., Fu, J., Bushnell, D. A., and Kornberg, R. D. (2001).
Structural basis of transcription: an RNA polymerase II elongation complex at 3.3
A resolution. Science 292, 1876-1882.

71

_ I-w.b‘fi

 

Gong, X. 0., Nedialkov, Y. A., and Burton, 2. F. (2004). Alpha-amanitin blocks
translocation by human RNA polymerase ll. J Biol Chem 279, 27422-27427.

Gong, X. 0., Zhang, C., Feig, M., and Burton, 2. F. (2005). Dynamic error
correction and regulation of downstream bubble opening by human RNA
polymerase II. Mol Cell 18, 461-470.

Grummt, l. (2003). Life on a planet of its own: regulation of RNA polymerase I
transcription in the nucleolus. Genes Dev 17, 1691-1702.

Henikoff, S., and Ahmad, K. (2005). Assembly of variant histones into chromatin.
Annu Rev Cell Dev Biol 21, 133-153.

Holmes, S. F., and Erie, D. A. (2003). Downstream DNA sequence effects on
transcription elongation. Allosteric binding of nucleoside triphosphates facilitates
translocation via a ratchet motion. J Biol Chem 278, 35597-35608.

Holmes, S. F., Foster, J. E., and Erie, D. A. (2003). Kinetics of multisubunit RNA
polymerases: experimental methods and data analysis. Methods Enzymol 371,
71-81.

Holmes, S. F., Santangelo, T. J., Cunningham, C. K., Roberts, J. W., and Erie, D.
A. (2006). Kinetic investigation of Escherichia coli RNA polymerase mutants that
influence nucleotide discrimination and transcription fidelity. J Biol Chem 281,
18677-1 8683.

Kedinger, C., Gniazdowski, M., Mandel, J. L., Jr., Gissinger, F., and Chambon, P.
(1970). Alpha-amanitin: a specific inhibitor of one of two DNA-pendent RNA
polymerase activities from calf thymus. Biochem Biophys Res Commun 38, 165-

171.

Kettenberger, H., Armache, K. J., and Cramer, P. (2004). Complete RNA
polymerase II elongation complex structure and its interactions with NTP and
TFIIS. Mol Cell 16, 955-965.

Khorasanizadeh, S. (2004). The nucleosome: from genomic organization to
genomic regulation. Cell 116, 259-272.

Komarnitsky, P., Cho, E. J., and Buratowski, S. (2000). Different phosphorylated
forms of RNA polymerase II and associated mRNA processing factors during
transcription. Genes Dev 14, 2452-2460.

KUhlman, T. C., Cho, H., Reinberg, D., and Hernandez, N. (1999). The general
transcription factors IIA, "B, HF, and "E are required for RNA polymerase II
transcription from the human U1 small nuclear RNA promoter. Mol Cell Biol 19,

2130-2141.
Landick, R. (2004). Active-site dynamics in RNA polymerases. Cell 116, 351-353.

72

 

Lee, T. l., and Young, R. A. (2000). Transcription of eukaryotic protein-coding
genes. Annu Rev Genet 34, 77-137.

Lindell, T. J., Weinberg, F ., Morris, P. W., Roeder, R. G., and Rutter, W. J. (1970).
Specific inhibition of nuclear RNA polymerase II by alpha-amanitin. Science 170,
447-449.

Lutter, L. C. (1982). Photoreactivation of amanitin-inhibited RNA polymerase II. J
Biol Chem 257, 1577-1578.

Meinhart, A., Kamenski, T., Hoeppner, S., Baumli, S., and Cramer, P. (2005). A
structural perspective of CTD function. Genes Dev 19, 1401-1415.

Murakami, K. S., and Darst, S. A. (2003). Bacterial RNA polymerases: the wholo
story. Curr Opin Struct Biol 13, 31-39.

Nedialkov, Y. A., Gong, X. 0., Hovde, S. L., Yamaguchi, Y., Handa, H., Geiger, J.
H., Yan, H., and Burton, Z. F. (2003). NTP-driven translocation by human RNA
polymerase II. J Biol Chem 278, 18303-18312.

Ohkuma, Y. (1997). Multiple functions of general transcription factors TFllE and
TFllH in transcription: possible points of regulation by trans-acting factors. J
Biochem 122, 481-489.

Orphanides, G., Lagrange, T., and Reinberg, D. (1996). The general transcription
factors of RNA polymerase II. Genes Dev 10, 2657-2683.

Oster, G. (2002). Brownian ratchets: Darwin's motors. Nature 417, 25.

Pokholok, D. K., Harbison, C. T., Levine, S., Cole, M., Hannett, N. M., Lee, T. l.,
Bell, G. W., Walker, K., Rolfe, P. A., Herbolsheimer, E, et al. (2005). Genome-
wide map of nucleosome acetylation and methylation in yeast. Cell 122, 517-527.

Prelich, G. (2002). RNA polymerase II carboxy-terminal domain kinases:
emerging clues to their function. Eukaryot Cell 1, 153-162.

Reinberg, D., Orphanides, G., Ebright, R., Akoulitchev, S., Carcamo, J., Cho, H.,
Cortes, P., Drapkin, R., Flores, 0., Ha, l., et al. (1998). The RNA polymerase II
general transcription factors: past, present, and future. Cold Spring Harb Symp
Quant Biol 63, 83-103.

Roeder, R. G. (1996). Nuclear RNA polymerases: role of general initiation factors
and cofactors in eukaryotic transcription. Methods Enzymol 273, 165-171.

Rudd, M. D., and Luse, D. S. (1996). Amanitin greatly reduces the rate of
transcription by RNA polymerase II ternary complexes but fails to inhibit some
transcript cleavage modes. J Biol Chem 271, 21549-21558.

73

 

(20
Spe

Dilly
Sci

Sawadogo, M., and Sentenac, A. (1990). RNA polymerase B (II) and general
transcription factors. Annu Rev Biochem 59, 711-754.

Schramm, L., and Hernandez, N. (2002). Recruitment of RNA polymerase III to
its target promoters. Genes Dev 16, 2593-2620.

Sims, R. J., 3rd, Belotserkovskaya, R., and Reinberg, D. (2004). Elongation by
RNA polymerase II: the short and long of it. Genes Dev 18, 2437-2468.

Sousa, R. (2003). On models and methods for studying polymerase translocation.
Methods Enzymol 371 , 3-13.

Tahirov, T. H., Temiakov, D., Anikin, M., Patlan, V., McAllister, W. T., Vassylyev,
D. G., and Yokoyama, S. (2002). Structure of a T7 RNA polymerase elongation
complex at 2.9 A resolution. Nature 420, 43-50.

Tellez, A. B., Crowder, S., Spagnolo, J. F., Thompson, A. A., Peersen, O. B.,
Brutlag, D. L., and Kirkegaard, K. (2006). Nucleotide channel of RNA-dependent
RNA polymerase used for intermolecular uridylylation of protein primer. J Mol
Biol 357, 665-675.

Thompson, A. A., and Peersen, O. B. (2004). Structural basis for proteolysis-
dependent activation of the poliovirus RNA-dependent RNA polymerase. Embo J
23, 3462-3471.

Vassylyev, D. G., and Artsimovitch, l. (2005). Tracking RNA polymerase, one
step at a time. Cell 123, 977-979.

Vassylyev, D. G., Sekine, S., Laptenko, 0., Lee, J., Vassylyeva, M. N., Borukhov,
S., and Yokoyama, S. (2002). Crystal structure of a bacterial RNA polymerase
holoenzyme at 2.6 A resolution. Nature 417, 712-719.

Vassylyev, D. G., Vassylyeva, M. N., Perederina, A., Tahirov, T. H., and
Artsimovitch, I. (2007). Structural basis for transcription elongation by bacterial
RNA polymerase. Nature 448, 157-162.

Vassylyev, D. G., Vassylyeva, M. M, Zhang, J., Palangat, M., Artsimovitch, l.,
and Landick, R. (2007). Structural basis for substrate loading in bacterial RNA
polymerase. Nature 448, 163-168.

Wang, D., Bushnell, D. A., Westover, K. D., Kaplan, C. D., and Kornberg, R. D.
(2006). Structural basis of transcription: role of the trigger loop in substrate
specificity and catalysis. Cell 127, 941-954.

Weinmann, R., and Roeder, R. G. (1974). Role of DNA-dependent RNA
polymerase 3 in the transcription of the tRNA and 58 RNA genes. Proc Natl Acad
Sci U S A 71, 1790-1794.

74

Westover, K. D., Bushnell, D. A., and Kornberg, R. D. (2004). Structural basis of
transcription: nucleotide selection by rotation in the RNA polymerase II active
center. Cell 119, 481-489.

Westover, K. D., Bushnell, D. A., and Kornberg, R. D. (2004). Structural basis of
transcription: separation of RNA from DNA by RNA polymerase II. Science 303,
1014-1016.

Woychik, N. A., and Hampsey, M. (2002). The RNA polymerase II machinery:
structure illuminates function. Cell 108, 453-463.

Xiong, Y., and Burton, Z. F. (2007). A tunable ratchet driving human RNA
polymerase ll translocation adjusted by accurately templated nucleoside

triphosphates loaded at downstream sites and by elongation factors. J Biol Chem
282, 36582-36592.

Yin, Y. W., and Steitz, T. A. (2002). Structural basis for the transition from
initiation to elongation transcription in T7 RNA polymerase. Science 298, 1387-
1395.

Zhang, C., and Burton, Z. F. (2004). Transcription factors IIF and IIS and
nucleoside triphosphate substrates as dynamic probes of the human RNA
polymerase II mechanism. J Mol Biol 342, 1085-1099.

Zhang, C., Yan, H., and Burton, Z. F. (2003). Combinatorial control of human
RNA polymerase II (RNAP ll) pausing and transcript cleavage by transcription
factor "F, hepatitis delta antigen, and stimulatory factor II. J Biol Chem 278,
50101-50111.

Zhang, C., Zobeck, K. L., and Burton, Z. F. (2005). Human RNA polymerase II
elongation in slow motion: role of the TFllF RAP74 alpha1 helix in nucleoside
triphosphate-driven translocation. Mol Cell Biol 25, 3583-3595.

Zhang, G., Campbell, E. A., Minakhin, L., Richter, C., Severinov, K., and Darst, S.
A. (1999). Crystal structure of Therrnus aquaticus core RNA polymerase at 3.3 A
resolution. Cell 98, 811-824.

75

 

CHAPTER TWO

Burton, Z. F., Feig, M., Gong, X. 0., Zhang, C., Nedialkov, Y. A., and Xiong, Y

(2005). NTP-driven translocation and regulation of downstream template opening
by multi-subunit RNA polymerases. Biochem Cell Biol 83, 486-496.

76

“all?”

 

liar
9‘01

Enz

low:

CHAPTER TWO

NTP-DRIVEN TRANSLOCATION AND REGULATION OF
DOWNSTREAM TEMPLATE OPENING BY MULTl-SUBUNIT

RNA POLYMERASES

Abstract

Multi-subunit RNA polymerases bind nucleotide triphosphate (NTP) substrates in
the pre-translocated state and carry the dNMP—NTP base pair into the active site
for phosphoryl transfer. NTP-driven translocation requires that NTP substrates
enter the main-enzyme channel before loading into the active site. Based on this
model, a new view of fidelity and efficiency of RNA synthesis is proposed. The
model predicts that, during processive elongation, NTP-driven translocation is
coupled to a protein conformational change that allows pyrophosphate release:
coupling the end of one bond-addition cycle to substrate loading and
translocation for the next. We present a detailed model of the RNA polymerase II
elongation complex based on 2 low-affinity NTP binding sites located in the main-
enzyme channel. This model posits that NTP substrates, elongation factors, and
the conserved pr2 subunit fork loop 2 cooperate to regulate opening of the

downstream transcription bubble.

77

 

 

1. Translocation models

The virtue of transient-state (pre-steady-state) kinetic analysis is that
functional dynamic changes in the elongation complex can be tracked through
the formation of specific bonds on a millisecond time scale (Johnson 1992,
1995). As such, the relationship between kinetic analyses and elongation
mechanism becomes apparent. Based on transient-state studies of elongation
catalyzed by human RNA polymerase ll, we have proposed the nucleotide
triphosphate (NTP) - driven translocation model (Nedialkov et al. 20033, 2003b;
Zhang et al. 2003, 2005; Gong et al. 2004; Zhang and Burton 2004; Gong et al.
2005). Competing models to describe elongation and translocation include the
allosteric model, posited by Erie and colleagues (Foster et al. 2001; Holmes and
Erie 2003), and the thermal or Brownian-ratchet model, championed by Sousa,
Nudler, and others (Guajardo and Sousa 1997; Oster 2002; Wang and Oster
2002; Sousa 2003, 2005; Landick 2004; Bar-Nahum et al. 2005). We find fault
with these models, but describe the merits of disparate views. The purpose of
this review is to consider NTP-driven translocation in the context of structural
data and alternative models. Because much of the evidence for our model has
been published, we will concentrate on qualitative descriptions of our model and

related models.

78

2. Rate-limiting steps during elongation

During processive RNA synthesis, human RNA polymerase II has 2
primary rate—limiting steps: NTP-driven translocation coupled to pyrophosphate
release and phosphodiester bond synthesis (or a strongly coupled but unknown
conformational change) (Figure ll-1A) (Nedialkov et al. 2003a, 2003b; Zhang et
al. 2003, 2005; Gong et al. 2004; Zhang and Burton 2004). These 2 rate-limiting
steps are sufficient to explain overall rates of RNA polymerase ll elongation (15
to 30 nucleotides per second). NTP loading is rapid but is initially low-affinity, as
expected for an interaction mediated primarily through base-pairing, recognition
of the triphosphate (i.e., through basic side chain contacts), and scanning of the
ribose ring (to exclude dNTP loading). For human RNA polymerase II,
isomerization is rapid, followed by relatively slow phosphodiester-bond synthesis.
During a primary isomerization step, the incoming substrate NTP-Mg2+ is
tightened in the RNA polymerase ll active site, where the metal is sequestered
from chelation by EDTA (Zhang and Burton 2004; Zhang et al. 2005). Despite
tight binding of the incoming NTP-Mg”, subsequent phosphoryl transfer is
delayed. This is a somewhat confusing result, because the phosphoryl transfer
reaction is expected to proceed very rapidly (Patel et al. 1991). One way to
explain this apparent contradiction is to speculate that, after the NTP-Mg2+ is
initially tightened in the RNA polymerase II active site, the substrate must be
further distorted (i.e., puckering the ribose sugar and (or) aligning the

triphosphate) before bond formation can occur (Castro et al. 2005).

79

Figure "-1. Nucleotide trip-hosphate (NTP)-driven translocation for (A)
multi-subunit RNA polymerases (RNAPs) and (d)NTP-driven translocation
for (B) simple DNA (DNAPs) and RNA polymerases (proposed). Pre-
translocation (pre) and post-translocation (post) elongation complexes (ECs) are
indicated. Bold arrows indicate rate-limiting steps in the human RNA polymerase
II mechanism. Rate-limiting steps may be different for some multi-subunit RNA
polymerases. Bold type with an asterisk indicates an isomerized (closed)
elongation complex. Elongation is shown for advancement from RNA or DNA
length n to length n+2. Question marks indicate uncertainty in simple enzyme
mechanisms (B). Continuous preloading of NTPs is indicated in the multi-subunit
RNA polymerase mechanism (see subsequent figures). Elongation complexes
that relate to published crystal structures are labeled: (a) Gnatt et al. 2001; (b)
Westover et al. 2004a, 2004b; (0) Kettenberger et al. 2004; (d) Westover et al.
2004b; (9) Johnson et al. 2003; Johnson and Beese 2004; (f) Yin and Steitz
2002; Johnson et al. 2003; Johnson and Beese 2004; (g) Temiakov et al. 2004;
(h) Yin and Steitz 2004; and (I) Yin and Steitz 2004. Active site (i+1) and

downstream (i+2 and i+3) positions are indicated.

80

 

 

>. 2:38:33: mz>bm
miss mo

US$326 m0

.. .
d _ d d d
w m u m ? .6 m m m 7 .m w
m ... m _ m. .n. a m n a m m a m
O n .0 II :I _ I: u h 0 n o e um fi II .I m 0 .
a we m w. _ m at a. ems m. u m m .s. H
n .fl M a _ a .l. W n .m w u a . a . W. n ,,
m m, d: ...u m .m _ ...u w .mn m dim u m u .n. u m .
.9 mm a... m_ mm mn ... mm: 3.3 m 3.. mn ... .
e n. o. l m o m m
mm. mm mm? m H “mm. mm P mm» W m We hum P 2
I2._.? I _ .24.? IZ._.? .24.? 02...? IZ._.? m
.zdU 2:“. o .2? a... .21V .. .2:u 2:. .24.". .. .2:U .. .24?
was #94! nos... 2:". luv“ seaside Iv 13:3. 22 *0 6851.2? luv v8.2.2.2? m|_.v v3.1». Eu. .
w_0<< _ mai ??m 92 H
Zfiuw L—x _ Amodp- 3.0me 992 .2363 AmUfi? 3.03.:
o Tu Z._._uw o Tu _ 3m.w83 mow «3633 mg
m. ITMIIV U. o I.» _ A9933
“aw: "n yoga. 77* u
n
_
_

l Brownia
. ratchet

l
l
i

Inc—.6 =-._.

 

332%»... . oz»...

10:3 .3 Anoszzcmnv.

.. m m .m 3..
C v o 0 o lo.- m C
. d. u - .m “mm « ..... w m m
w m m c _ m. .n. n . m a m .m m m. n
t n a E n _ a s m PIN u e .... m
m n e n C n o C e on. T s r or m t .M In.
m m mm .m. .m a u m m m m Wmm m o m o m
m m .4 a w m n . m. n I m. P WtP P h P B P
..... m m we ... m. l m _ m .m w u .623...
w... m m. Wm % M. m. m u w m m m 6333.23 )|.w pomf+132duIv602333.2qu .._u3_.....~._e.u.m
mm. H m. ML“. % mu m. b _ h m M.‘ B 23 mod? 03me
.Aavzdu m. "3. 3&4? .V o 32...? “swag..." m9 A v
, ?3: $ ?omn:oAavZ._.?lIV“ I?Om»:oAQvZ._.? IV .I?..0:+._o??_4v ?om~:+.‘o Ann—Oman:
,2 v . " Amos? 382: on. 333
,avzdu A3216 _ 3%st mg m
w m. OTN L m. . _ w a .
nan 4|m 685:: u x m 308326 mo
mm " m m
m m _ m w
e n _

82

Interesting similarities have been observed in the kinetics of RNA
polymerase II (12 subunits) and simpler DNA and RNA polymerases, which are
generally single subunit enzymes and are not homologous to multi-subunit
enzymes. For the RNA-dependent RNA polymerase of poliovirus, analyzed in the
presence of Mn”, isomerization is rapid, and is followed by rate-limiting bond
formation (Arnold and Cameron 2004; Arnold et al. 2004; Castro et al. 2005). In
this case, isomerization is defined by sequestration of NTP-Mn2+ in the active
site, so that the metal is no longer available to react with EDTA as the reaction-
quenching agent. On the other hand, with Mn2+ as the metal cofactor, for
poliovirus RNA polymerase, isomerization (resistance to EDTA quench) is
apparently coincident with bond formation (acid quench). This result indicates
that, in the Mn2+-supported reaction, after initial tight sequestration of the NTP-
Mn2+ in the active site, another isomerization step might be necessary to
complete bond synthesis. Of significant interest is the startling similarity of the
poliovirus RNA polymerase elongation kinetics in the presence of an+ to the
human RNA polymerase ll elongation reaction in the presence of the

physiological metal Mg”.

3. NTP-driven translocation

Two relatively rapid kinetic phases are clearly resolved for human RNA
polymerase ll elongation from a stall position (Nedialkov et al. 2003a, 2003b;

Zhang et al. 2003, 2005; Gong et al. 2004; Zhang and Burton 2004) (Figure II-

83

 

1A). These phases have been interpreted as elongation from the pre- and post-
translocation states of the elongation complex. As expected, elongation from the
post-translocation state is much faster than from the pre-translocation state.
Occupancy of the 2 phases is changed in the presence of different sets of
elongation factors, showing that accessory factors shift RNA polymerase II
between functional modes (translocation states), as expected. The pre-
translocation state is highly sensitive to a-amanitin inhibition, but the post-
translocation state is resistant; the distinct rates of elongation clearly distinguish
the 2 phases (Gong et al. 2004). The mushroom toxin a-amanitin has been
shown to block translocation. Both pre- and post-translocation states are
responsive to NTP substrates, showing that the pre-translocated elongation
complex binds a templated NTP substrate.

The binding of an NTP to the pre-translocated elongation complex is
consistent with the NTP-driven translocation model and the allosteric model, but
not the Brownian-ratchet model. The Brownian-ratchet model has little meaning if
the NTP substrate binds prior to translocation and drives translocation forward
(Wang and Oster 2002; Landick 2004). The Brownian-ratchet mechanism
requires that thermal motion drive translocation and that the NTP act as a pawl
(as on a mechanical ratchet) to fix the elongation complex in the post-
translocation state (Bar-Nahum et al. 2005; Sousa 2005). The observed binding
of the substrate NTP to the pre-translocation state of the elongation complex
converts the Brownian-ratchet model into the NTP—driven translocation model.

The elongation mechanism for human RNA polymerase II, however, has

84

characteristics of both NTP-driven translocation (the favored pathway) and a
Brownian-ratchet (the default pathway in the absence of NTP substrates) (Figure
ll-1A). When NTPs are not present, the elongation complex fractionates into both
the pre- and post-translocation states, so RNA polymerases can function as
Brownian ratchets, at least in the absence of NTPs or during transcriptional
stalling.

The allosteric model is based on millisecond-phase kinetic studies of
Escherichia coli RNA polymerase elongation (Foster et al. 2001; Holmes and
Erie 2003). The model describes an allosteric transition between an unactivated
state and an activated state of the RNA polymerase elongation complex,
apparently requiring multiple interactions with the next incoming NTP substrate.
The difficulty with the allosteric model is that, because the allosteric site is
templated, allostery appears to require 2 NTP substrates simultaneously base-
paired to the same DNA template base. As far as we can determine, there is no
satisfactory explanation for this conundrum. We do, however, support the
following predictions of the allosteric model. First, an NTP substrate bound in the
pre-translocation state acts as an allosteric effector to drive translocation forward
from the unactivated state (pre-translocation state) to the activated state (post-
translocation state) (Zhang and Burton 2004; Zhang et al. 2005). As far as we
can discern, the allosteric model reduces to the NTP-driven translocation
mechanism, and no other view of the allosteric model makes physical sense.
Second, our laboratory has demonstrated other allosteric effects of the incoming

NTP substrate during transfer from the pre- to the post-translocation positions

85

(Gong et al. 2005). Specifically, the fate of a templated NTP tightened in the RNA
polymerase II active site is linked to the presence of NTPs paired at the next 2
downstream template positions. Because these experiments demonstrate
simultaneous occupancy of the pre- and post-translocation states of the
elongation complex, they demonstrate the major tenet of the NTP-driven
translocation model, which requires that both template positions be occupied by
NTP substrates. So the allosteric model is verified to this extent: the incoming
NTP substrate acts in a manner similar to an allosteric effector to drive
translocation forward.

Another prediction of our model is that, during processive RNA
synthesis, NTP-driven translocation is coupled to pyrophosphate release from
the previous bond-addition cycle (Zhang and Burton 2004; Zhang et al. 2005).
We posit that, during processive synthesis, pyrophosphate remains locked in a
tightened (closed) conformation of the elongation complex. For pyrophosphate to
be released, the active site must first be relaxed to the open conformation. NTP-
driven translocation induces the closed—>open conformational change in the
elongation complex that results in pyrophosphate release (Figure Il-1A).

This feature of our model is demonstrated when human RNA
polymerase II elongation in the presence of TFIIF is compared with TFIIF-mutant
proteins defective in stimulating elongation (Zhang et al. 2005). We identified
TFIIF-mutant proteins that fail to fully support NTP-driven translocation. As a
result, the processive transition between phosphodiester bonds, which requires

release of pyrophosphate, becomes very slow. Because bond completion

86

or
be

inc

pol
20l
liar
ass
Don
cha

Dore

(pyrophosphate release) relies on NTP-driven translocation, this step is highly
dependent on the concentration of the incoming NTP. This NTP-driven step is
facilitated by wild-type TFIIF. In the presence of the mutant TFIIF protein, RNA
polymerase ll escapes from a transcriptional stall by a mechanism that appears
to be NTP-independent. Stalled RNA polymerase ll lacks tightly bound
pyrophosphate. Pyrophosphate release results in the acceleration of the NTP-
independent translocation pathway, because relaxation of the active site during
stalling frees the Brownian ratchet (Figure ll-1A). During a processive transition,
only the NTP-driven pathway is available until pyrophosphate can spontaneously
be released, a slow process unless this step is facilitated by TFllF and (or) the

incoming NTP.

4. NTP loading into the RNA polymerase active site

Elongation complex structures have been published for yeast RNA
polymerase II (Gnatt et al. 2001; Kettenberger et al. 2004; Westover et al. 2004a,
2004b) (Figures ll- 2-4), and these structures are consistent with the NTP-driven
translocation model. These structures, however, indicate that the previously
assumed route of NTP entry into the RNA polymerase active site, the secondary
pore, is not the main route of NTP entry. The secondary pore presents a narrow
channel with a highly negative electrostatic potential (Batada et al. 2004), so the
pore does not appear to be a favorable NTP-loading route. Rather, the main

entry port for NTP substrates must pass through the main enzyme channel. For a

87

Figure "-2. Three NTP binding sites in RNA polymerase II. (A and B) Two
views of the yeast RNA polymerase II elongation complex structure (Gnatt et al.
2001; Westover et al. 2004b). (C and D) Simplified views of main channel (i+2
and i+3) and active site (i+1) NTP-loading sites. (C) Stick figures with labeled
amino acids thought to be involved in NTP binding and NTP-driven translocation.
Fork loop 2 (pr2 region Asn499 to His515) is a light blue ribbon. (D) Space-
filling representation, indicating the open space between the i+2 and i+1 sites.
Minor movements of pr2 fork loop 2 (light blue; amino acids pr2 500—540 are
shown) is necessary for NTP passage from i+2 to i+1. DNA-template strand is
red (from crystal structures) or pink (modeled, upstream DNA). The non-template
strand is white (from crystal structures) or silver (modeled, bubble and upstream
DNA). The modeled open complex is 18 bases. RNA is green. The bridge d-helix
is a blue cylinder. The i+1 (active site) (purple), i+2 (yellow), and i+3 (orange)
NTPs are indicated. pr2 Ar9504, pr2 ArgS12, pr1 Lys1102, and pr1
Lysl112 (blue) are indicated. pr2 Asp505 and pr2 G|u529 (red) and pr1

Al3828 (brown) are shown.

88

 

Figure "-2.

89

Figure "-2 (continued).

 

9O

Figure "-2 (continued).

9

4.

Bridge helix

 

91

Figure "-2 (continued).

 

92

Figure "-3. NTP-driven translocation by human RNA polymerase II. (A) Pre-
translocation elongation complex. Colors are the same as those indicated in
Figure "-2. (B) Translocation intermediate. (C) Post-translocation elongation
complex. Space is available to load an NTP (yellow) from the i+2 main-channel

site to the i+1 active site. An active site, Mg2+ (white), is shown.

93

,,’s:¢§.
s 5 a.
o ‘7 f 3 I

. c
m'ghannel

I+2'-~

Figure "-3.

 

Figure "-3 (continued).

 

Figure "-3 (continued).

 

96

Figure "-4. a-amanitin blocks translocation. (A) Structural model of a-amanitin
(pink) bound to the post-translocation elongation complex with 3 substrate NTPs
bound. Colors are the same as those indicated in Figure "-2. (B) Structural
comparison of d-amanitin (left panel) and Microcin J25 (right panel) (Bayro et al.
2003; Rosengren et al. 2003; Wilson et al. 2003; Gong et al. 2004). For (8)

similar amino acids are yellow.

97

Pore )’, .
“:3 ‘ -. u-amanltln
- . ‘

v ,0
94“

 

Figure "-4.

98

Inca =5 Ana-.3353.

3.392430

99

3305730..
.4

   

9.2335: .5331: gnu

stalled elongation complex in the post-translocation state, the secondary pore is
the likely route of NTP entry (Kettenberger et al. 2004; Westover et al. 2004a,
2004b). In the pre-translocation state, however, the DNA template base is single
stranded and faces the main enzyme channel, not the secondary pore (Gnatt et
al. 2001). To the pre-translocation state, therefore, the main enzyme channel is
the likely route of NTP entry, and the secondary pore is excluded as a potential
loading route. Kinetic studies demonstrate that the pre-translocation state of the
elongation complex binds NTPs (Nedialkov et al. 2003a; Gong et al. 2004, 2005;
Zhang and Burton 2004; Zhang et al. 2005). During processive synthesis, NTPs
occupy the pre-translocation position, while the post-translocation position is
occupied by the substrate NTP; this means that, under normal elongation
conditions, NTPs enter the active site from the main enzyme channel. Only when
the RNA polymerase stalls, releases pyrophosphate, and then translocates, can
NTPs be loaded through the secondary pore, a situation that does not arise
during processive synthesis (Figure ll-1A). Sufficient space is available to load
NTPs from the main channel into the RNA polymerase II active site (Figure ll-
20).

What is the energetic driving force that propels the dNMP-NTP base
pair to advance from the i+2 position to the i+1 active site (Figure "-3)? From
observation of available elongation complex structures, there is no obvious
answer to this question. We assume that the accurately paired NTP acts in a
manner similar to an allosteric effector to induce a conformational change in RNA

polymerase ll that drives translocation of the RNA-DNA hybrid forward. Bending

100

of the bridge d-helix (Gnatt et al. 2001), rearrangement of the trigger loop
(Vassylyev et al. 2002; Bar-Nahum et al. 2005), and movement of other mobile
protein switches (Holmes and Erie 2003) may be involved in the NTP-driven
translocation mechanism.

Available elongation complex structures of yeast RNA polymerase ll
present different views of the number of single-stranded DNA template bases
available for NTP pairing in the main enzyme channel. A structure from the
Kornberg laboratory indicates at least 3 main channel single-stranded template
bases (i+2, i+3, and i+4; i+1 represents the position of an NTP in the active site
poised for bond formation) (Gnatt et al. 2001). More recent structures from the
same researchers have shown a similar position for bubble melting (i+4 is
unpaired). The nontemplate DNA strand in these structures, however, lacked
bases i+2 and i+3, inhibiting its capacity to anneal (Westover et al. 2004b). The
structure from the Cramer laboratory shows that the downstream bubble can
close, leaving (at most) a single base (i+2) unpaired in the main enzyme channel
(Kettenberger et al. 2004). The Cramer structure is not informative about the
unpaired i+2 base, because the i+2 base was selected to be unpaired. To
support the NTP-driven translocation mechanism, a single unpaired DNA base
(i+2) in the main-enzyme channel is minimally sufficient (Nedialkov et al. 2003a;
Zhang and Burton 2004), and none of the available structures contradict this
requirement.

Based on experiments with human RNA polymerase II, however, at

least 3 DNA bases appear to be simultaneously available for pairing NTP

101

substrates (i+1, i+2, and i+3), (Gong et al. 2005; Figure "-2). This result is in
apparent contradiction to the Cramer structure, but it appears to be consistent
with the Kornberg structures. We therefore propose that the Cramer elongation
complex suffers downstream bubble collapse. The Cramer elongation complex
has a 7- rather than an 8-base-pair RNA—DNA hybrid (Kireeva et al. 2000; Gnatt
et al. 2001), and the crystal is badly disordered upon addition of a substrate
analogue (GMPcPP), which falls to load accurately into the catalytic site
(Kettenberger et al. 2004). In the Cramer structure, the i+2, i+3, and i+4 DNA
base pairs are strained, as if they are likely to separate in a more favorable
conformation of the elongation complex.

There is no clear demonstration that the Cramer structure is highly
active, leading us to wonder whether the crystals would disorder upon addition of
a natural GTP substrate, as they appear to do attempting to load the GMPcPP
analogue. The Cramer elongation complex does not represent the only
configuration of the downstream elongation bubble; the Kornberg structures, with
3 unpaired DNA bases in the main enzyme channel (i+2, i+3, and i+4) (Gnatt et
al. 2001; Westover et al. 2004b), appear to be more accurate portrayals of a
functional RNA polymerase II elongation complex. An NTP substrate loaded in
the active site (i+1) senses the presence of at least 2 downstream templated
NTP bases (i+2 and i+3), using verified and functional human RNA polymerase II
elongation complexes (Gong et al. 2005). Because ours is a kinetic experiment,
we do not know the structural positions occupied by incoming NTPs, but we have

modeled possible positions in Figures "-20 and 2D. Our experiments are done

102

with human enzyme, and available crystal structures are of the yeast enzyme. It
is possible that there are differences in the regulation of the downstream
transcription bubble in different multi-subunit RNA polymerases, although we do
not believe that this is the case for RNA polymerase II from different species. We
guess that all multi-subunit RNA polymerases use similar NTP-driven

translocation mechanisms.

5. Evolutionary conservation of residues proposed to participate

in NTP-driven translocation

Multi-subunit RNA polymerases are conserved in all three living
kingdoms, eukarya, archaea, and eubacteria. Eukarya possess RNAPs I, II and
Ill, encoded by the Rpa, pr and Rpc genes, respectively. Multiple sequence
alignments show the regions that we propose to be important in the NTP-driven
translocation mechanism are either invariant or highly conserved (data not
shown). The sequence alignments were initially done using 3D-Coffee
(O'Sullivan et al., 2004; Poirot et al., 2004) and then edited by hand.

The proposed functions of particular amino acids are listed in Table II-
1. Rbp1 Ala828 is on the bridge a-helix. This residue is invariant in this alignment
except for the highly divergent vaccinia virus RNAP for which this residue is
valine. pr1 Lys1102 is conserved in most prs, Rpcs and Rpas (as arginine).
pr1 Lysl112 is conserved in all prs and most Rpas. pr1 Lys1102 and

Lys1112 are proposed to recognize the i+3 NTP triphosphate, so some multi-

103

subunit RNAPs may use a different strategy to recognize the i+3 NTP or may use
a simpler version of the NTP-driven translocation mechanism (i.e. pre-loading
only the i+2 NTP). pr2 Ar9504, pr2 Asp505, pr2 ArgS12 and pr2 Glu529
are highly conserved among prs. Ar9504 is conserved except in Rpas. Asp505
is found in prs and eubacteria. Asp505 is proposed to recognize the 2’- and 3’-
hydroxyls of the ribose ring of the i+2 NTP. RNA polymerases lacking a similar
residue at this position must use a somewhat different strategy to discriminate
ribose and deoxyribose sugars. pr2 Arg512 is invariant in this alignment.
Because of the position of this buried arginine, it is difficult to account for its high
conservation without considering the NTP-driven translocation mechanism.
Arg512 is not positioned to interact with DNA, RNA or an NTP at the active site.
pr2 Glu529 is also very highly conserved throughout evolution. Based on the
high conservation of the fork loop 2 region and particular amino acids, we
propose that most or all multi-subunit RNA polymerases utilize NTP-driven

translocation and pre-loading of the i+2 NTP.

104

Table "-1. Yeast RNAP ll residues proposed to be important in NTP-driven
translocation.

105

 

 

 

 

 

 

 

 

 

955.: >350 >nE .uqouomoa mczomo:
mun: >_mmmm romaim 9. TN z._._U 8 5m
Awaaom 193 2.1 moz<m mam
m2: 5.3 8» mmooosao: 9. 5m 1w
AIm=x ad 2._._u Snzomvsmfim
m2: 5mg 4 AN mmooosao: 0.. :6 Ta
z._._U Egomurmfi
moon >68... 28832.0: 0* :5 TN
cue} roon my z._._U 330%sz
moan >mnmom 2800328: 9. :6 N- man
Que} roon 8 w... OI 9. 3m 1m
z._._u 150mm mcomq
mncm >6m§ mmooosao: 9. 3m 1.»
cue} roou NV z._._U 330335
Zoom Gamma Ozmamao: 2 3m
330.836 2 5m

 

 

583:6 z._._U mccmqmfim
mm = mama Sm Ni mog<m

mzm L

 

 

106

 

6. A structural model for NTP-driven translocation

Based on our work and available crystal-structure x-rays, we present a
detailed model to describe the preloading of NTP substrates in the RNA
polymerase II main channel prior to translocation (Figure "-2). The non-template
strand is disordered in crystal structures, so its position was modeled in the
images (Figures ll-2A, 28). Our model posits 2 NTP- Mg2+ substrates (i+2 and
i+3) paired with cognate DNA bases in the main channel; the active site is
occupied with another NTP substrate (i+1). We do not believe a third NTP can be
accommodated in the main channel because, at the i+4 position, an NTP would
clash with the non-template DNA strand. We posit that main channel NTP
substrates help to keep the downstream transcription bubble open, explaining
why the bubble can collapse in the absence of appropriate incoming substrates
(Kettenberger et al. 2004). For the post-translocated elongation complex, in the
presence of substrates, we suggest bubble opening at the i+2, i+3 and i+4
positions, as indicated in the Kornberg structure (Gnatt et al. 2001). To create our
model, we placed some amino acids that were disordered in the Kornberg
structures (yeast pr2 fork loop 2 region, residues Gly503-Leu508) (Figures ll-
2C, 2D). The i+2 and i+3 NTPs pair to their main channel template sites. A
binding site forms around the NTPs, bordered by pr2 subunit fork loop 2
protein side chains. Available amino acids project the expected basic groups
(pr2 Arg504, pr2 Arg$12, pr1 Lys1102, and pr1 Lysl112) and 1 acidic

group (pr2 Asp505). It is proposed, therefore, that the binding of NTPs in the

107

main channel sculpts and maintains the downstream edge of the transcription
bubble.

This model provides a mechanism by which RNA polymerase II can
anticipate upcoming transcriptional stalls. As RNA polymerase II approaches a
stall, we have observed that the elongation complex enters the paused state
more rapidly than can be accommodated by rate constants into and out of the
paused state, indicating that RNA polymerase II senses an upcoming stall before
the active site is deprived of substrate (Zhang et al. 2003). According to our
current model, the downstream DNA template, pr2 fork loop 2, and the
nontemplate DNA strand encompass the i+2 and i+3 NTP- Mg2+ substrates to
regulate the opening of the downstream edge of the transcription bubble. When
NTP substrates are not available, the downstream bubble tends to collapse,
leading to enhanced transcriptional pausing. The binding of elongation factors is
also expected to influence the extent of downstream bubble opening and, hence,
the frequency of pausing.

As described above, the NTP-Mg2+ binding site in the main RNA
polymerase ll channel is expected to screen incoming NTPs for accurate base
pairing, for triphosphate tails, and for 2'- and 3'-hydroxyls on the ribose ring. Our
model allows for templating at main channel sites and explains why, during NTP
deprivation, the transcription bubble can collapse, as iobserved in the Cramer
structure (Kettenberger et al. 2004). Deprived of the i+2 and i+3 NTPs, RNA
polymerase II might close the downstream bubble as it approaches the stall. We

posit that yeast RNA polymerase II amino acids pr2 Arg504 and Arg512

108

 

(contacting the i+2 NTP triphosphate) and pr1 Lys1102 and Lys1112
(contacting the i+3 NTP triphosphate) are involved in screening triphosphate
tails. pr2 Asp505 is positioned to scan the 2'- 3'—cis diol of the i+2 ribose ring.
We suggest that this interaction may be mediated by a mobile Mg” atom. For T7
RNA polymerase, a Mg” is used in a similar manner to prescreen the 2'-hydroxyl
of the incoming NTP substrate in the pre-insertion site (T emiakov et al. 2004).
For RNA polymerase II, neither the NTP- Mg” binding pockets at the i+2 or i+3
positions of the transcription bubble are stable binding sites. NTPs are not
expected to loiter in the main channel, but rather to rapidly dock and then
advance toward the active site, paired with their cognate DNA base. As the DNA
template translocates, another NTP- Mg” occupies the i+3 position. (See
supplementary Figures ll-1A-1C and supplementary Table l for a summary of
predictions based on our model and for evolutionary conservation of relevant
protein regions.)

As the i+2 NTP- Mg” is transferred to the i+1 active site during
translocation, other RNA polymerase II residues become important (Figures ll-

20, ZD, and 3). Bridge helix pr1 Ala828 is encountered during NTP passage

past the bridge a-helix (Nedialkov et al. 2003a). pr2 Glu529 is expected to help

orient the tri-phosphate, through charge repulsion, during its passage into the

active site. Residues on the bridge q-helix and in the underlying trigger loop (Bar-

Nahum et al. 2005) may be involved in the translocation mechanism. Amino

acids interacting with the RNA—DNA hybrid are also expected to be important,

109

 

because mobilizing the hybrid likely facilitates translocation (Holmes and Erie

2003)

7. Mushrooms and antimicrobials

A deadly mushroom toxin, o-amanitin, blocks translocation by human
RNA polymerase II (Bushnell et al. 2002; Gong et al. 2004) (Figure ”-4). q-
amanitin is structurally similar to a bacterial antibiotic Microcin J25. d-amanitin is
a bicyclic, covalently crosslinked, and multiply modified octapeptide (Figure "-43,
left panel). Microcin J25 (with 21 amino acids in total) consists of a cyclic
octapeptide (amino acids 1—8) and peptide tail (amino acids 9—21), the C-
terminus of which can enter the octapeptide ring to form a bridge reminiscent of
the o-amanitin covalent crossbridge (Bayro et al. 2003; Rosengren et al. 2003;
Wilson et al. 2003) (Figure "-48, right panel). d-amanitin is the most potent and
specific known inhibitor of human RNA polymerase II, and Microcin J25 inhibits
homologous bacterial RNA polymerases (Yuzenkova et al. 2002). Both toxins
penetrate the secondary pore, and d-amanitin interacts strongly and specifically
with the bridge o-helix (Bushnell et al. 2002) (Figure ll-4A), thought to be
important in translocation mechanisms (Gnatt et al. 2001; Vassylyev et al. 2002;
Bar-Nahum et al. 2005). Similar to its structural analogue a-amanitin, Microcin
J25 most likely inhibits translocation. Suggestions that Microcin J25 blocks NTP
loading through the secondary pore to inhibit transcription (Adelman et al. 2004;

Mukhopadhyay et al. 2004) are not likely correct. Certainly, the cork-in-bottle

110

 

model for Microcin J25 action is contrary to the NTP-driven translocation model,
which posits an alternate route of NTP entry. This issue will be resolved when
transient-state kinetic analyses of Microcin J25 inhibition are reported. Single
molecule elongation studies have lacked sufficient resolution to discriminate
between an inhibitor that blocks substrate binding and one that blocks
translocation (Adelman et al. 2004). As a potent translocation inhibitor, 0-
amanitin has proven to be an incisive probe with which to unravel the human

RNA polymerase II elongation mechanism (Gong et al. 2004, 2005).

8. Fidelity and efficiency

Why did NTP-driven translocation develop as the dominant mechanism
for substrate NTP loading into the active sites of multi-subunit RNA
polymerases? The answer relates to the requirement for high fidelity and
efficiency of RNA synthesis (Nedialkov et al. 20033; Zhang and Burton 2004).
There lingers a textbook ethic that RNA polymerases lack the separate
exonuclease activities of DNA polymerases, because RNA is a throw-away copy
of the genetic material, and, therefore, errors in transcription are tolerated. Errors
in DNA replication, in contrast, are thought to have a more severe consequence,
because lesions can be fixed into genetic material as heritable mutations. Based
on kinetic studies of elongation, however, it appears that many RNA polymerases

are highly accurate enzymes, perhaps matching the fidelity of DNA polymerases,

111

at least during ongoing synthesis. In that case, how do multi-subunit RNA
polymerases maintain high fidelity and efficiency during elongation?

For both DNA and RNA polymerases, high fidelity must be achieved
with relatively low-affinity substrate binding, because these enzymes must
recognize 4 substrates with different shapes and chemical groups. Accurate
base-pairing must provide a primary initial screening of the incoming (d)NTP, but
it is also essential to screen for or against the 2'- OH, so that DNA polymerases
can exclude ribo-NTPs, and RNA polymerases can exclude deoxy-NTPs.
Exclusion of mono- and diphosphates is also important. Recent structural data of
simple RNA and DNA polymerases and kinetic studies of multi-subunit RNA
polymerases indicate that the screening of substrates begins outside the active
site of the enzyme, and that a series of checkpoints must be passed before the
substrate tightens into the active site for phosphoryl transfer. Pre-assessment of
the NTP substrate largely precludes replication and transcription errors, with low-
energy cost, so DNA and RNA polymerases can achieve high fidelity with
relatively low-affinity substrate recognition.

For simple RNA and DNA polymerases (i.e., T7 RNA polymerase and
Bacillus stearothermophilus large-fragment DNA polymerase I), template pre-
insertion, (d)NTP pre-insertion, and insertion (active site) positions of the dNMP
template have been defined (Johnson et al. 2003; Johnson and Beese 2004;
Temiakov et al. 2004; Yin and Steitz 2004) (Figure ll-1B). Substrate (d)NTPs are
known to associate with the (d)NTP pre-insertion and insertion (active site)

positions. Based on the NTP-driven translocation model for human RNA

112

polymerase II, we suggest that accurate (d)NTP binding may be the route
favored to drive the elongation complex between the template pre-insertion (pre-
translocated) and the (d)NTP pre-insertion (post-translocated) sites for some
simple polymerases. If this supposition is correct, then simple RNA and DNA
polymerases may use a (d)NTP-driven translocation mechanism similar to that
used by multi-subunit RNA polymerases. This mechanism would allow 2 low-
affinity pre-assessments of the substrate (d)NTP prior to loading it into the active
site: 1 at the template pre-insertion site (Johnson et al. 2003; Johnson and Beese
2004); and 1 at the (d)NTP pre-insertion site (Temiakov et al. 2004). Transfer
from the (d)NTP pre-insertion site to the active site requires a protein
conformational change before potential (d)NMP mis-incorporation (Temiakov et
al. 2004). Pre-screening is an effective fidelity check, because a (d)NTP that is
rejected outside the active site cannot result in (d)NMP mis-incorporation. The
template pre-insertion site for DNA polymerases is located on the opposite side
of the O d- helix (Johnson et al. 2003; Johnson and Beese 2004), far from the
active site, so initial pre-assessment of an incoming dNTP might occur far from
the active site. In multi-subunit RNA polymerases, initial NTP pre-assessment
occurs on the opposite side of the bridge d-helix (Nedialkov et al. 2003a; Zhang
and Burton 2004; Zhang et al. 2005) (Figure ll-2C), far from the active site,
suggesting that NTP loading and assessment strategies may be similar for non-
homologous simple and multi-subunit enzymes. T7 RNA polymerase opens the
downstream transcription bubble at the N-terminal tip of the O'-d-helix (at

Phe644). For T7 RNA polymerase, therefore, only a single template base is

113

available to pair with an NTP substrate. To use NTP-driven translocation requires
that an NTP pair with template as soon as its DNA base becomes single
stranded, during DNA passage past Phe644 and the O'-d-helix. T7 RNA
polymerase has been shown to load NTPs to an NTP pre-insertion site prior to a
conformational change (open to closed transition involving rotation of the O-o-
helix finger domain) that brings the substrate NTP into the insertion (active) site
(Temiakov et al. 2004).

Kinetic analyses of RNA-dependent RNA polymerase of poliovirus (in
the presence of Mn”) (Arnold and Cameron 2004; Arnold et al. 2004; Castro et
al. 2005) are similar to those for human RNA polymerase II (in the presence of
Mg”) (Nedialkov et al. 2003a; Zhang and Burton 2004). Poliovirus RNA
polymerase is a homologue of simple RNA and DNA polymerases. This result
leads us to believe that simple enzymes, like multi-subunit enzymes, may use
(d)NTP-driven translocation mechanisms.

For multi-subunit RNA polymerases, the NTP substrates are
prescreened in the main enzyme channel, which is separated from the active site
by the bridge o-helix (Nedialkov et al. 2003a; Zhang and Burton 2004) (Figures
”-20, 2D). The incoming NTP is first screened by base-pairing to template at the
i+3 position. It is screened again at the i+2 position. We believe that the 2'-OH of
the ribose ring is assessed at the i+2 position by pr2 Asp505, so dNTPs are
normally rejected well outside the RNA polymerase II active site. The space from
the main enzyme channel to the active site, which the substrate NTP must

traverse in transition from the pre- to the post-translocation position, is

114

constrained, so only accurately paired NTPs can successfully complete the
passage (Nedialkov et al. 2003a) (Figures "-20, 2D, and 3). Once beyond the
bridge o-helix, the incoming dNMP-NTP base pair is again challenged during an
isomerization step from an open to closed active site (Zhang and Burton 2004;
Zhang et al. 2005). This is an induced-fit step, similar to the conversion from the
pre-insertion (open conformation) to the insertion (closed conformation) site for
simple RNA and DNA polymerases (Johnson et al. 2003; Johnson and Beese
2004; Temiakov et al. 2004; Yin and Steitz 2004). In kinetic studies, the open to
closed transition is indicated by sequestration of active-site metal ions from
EDTA chelation (Zhang and Burton 2004; Zhang et al. 2005). For human RNA
polymerase ll, the apparently slow bond synthesis step may represent a second
induced-fit fidelity check within the active site. If the tightened NTP-Mg” cannot
be appropriately deformed to the catalytically competent state, this would signal
an error in substrate loading, inducing reversal of isomerization and NTP release.
According to the NTP-driven translocation model, pyrophosphate release (bond
completion) is coupled to the next NTP-driven translocation (the following bond
initiation). lf translocation is blocked (through NMP mis-incorporation), bond
synthesis might be reversed and the inaccurately loaded NTP expelled, providing
a fidelity check even after phosphodiester bond formation but prior to
pyrophosphate release. Thus, simple DNA and RNA polymerases and multi-
subunit RNA polymerases use a complex pathway of (d)NTP prescreening with
multiple fidelity checks and an induced-fit step (or steps) prior to bond formation.

These enzymes may run a final fidelity check after bond formation but before

115

pyrophosphate release. Release of pyrophosphate results in (essentially)
irreversible bond completion.

NTP-driven translocation is efficient because it allows for accurate pre-
alignment, prescreening, and presorting of NTP substrates (Nedialkov et al.
2003a; Zhang and Burton 2004). Furthermore, the end of one bond-addition
cycle (pyrophosphate release) is coupled to the beginning of the next
(translocation and NTP loading) (Zhang and Burton 2004; Zhang et al. 2005).
NTPs load through the main channel. In the NTP-driven translocation model, the
secondary pore is the route of pyrophosphate release. Improperly loaded and
rejected NTP (or dNTP) substrates also exit through the secondary pore in cases
in which they pass prescreening checkpoints. Only when the elongation complex
is stalled and ratcheted to the post-translocated position can the secondary pore

be a route for substrate NTP loading.

9. Conclusions

In summary, a new view of translocation mechanisms has been
developed, based on kinetic analyses of elongation by human RNA polymerase
II. NTP-driven translocation is consistent with available kinetic and structural
data, although, in the future, inconsistencies with the Cramer elongation complex
structure will have to be resolved. NTP-driven translocation posits that the RNA

polymerase secondary pore is not the only, nor is it the major, route of NTP entry

116

into the elongation complex. NTPs are loaded first in the enzyme main channel.
NTP-driven translocation provides a new understanding of fidelity and efficiency,
indicating that RNA synthesis is a higher-fidelity reaction than was previously
thought. Simple RNA and DNA polymerases may use a shorthand version of
(d)NTP-driven translocation, providing multiple fidelity checks of the incoming
substrate at a low-energy cost.

Inspection of available RNA polymerase II elongation complex
structures suggests that the opening and closing of the downstream transcription
bubble is regulated during elongation. We suggest that downstream bubble
opening is maintained through NTP- Mg” loading, and is regulated by
transcriptional elongation factors. Downstream bubble collapse is a signal for
RNA polymerase to pause transcription. Maintenance of the open bubble
configuration is the signal for continued rapid synthesis, and is supported by the
weak binding of templated NTP- Mg” substrates within the main RNA
polymerase channel.

Disparate models for elongation will be reconciled using a combination
of structural studies, transient-state kinetic analyses, and mutagenic studies.
Understanding the coupled mechanisms of translocation and NTP- Mg” pre-
screening is important to the understanding of the fidelity and efficiency of multi-

subunit RNA polymerases.

117

References

Adelman, K., Yuzenkova, J., La Porta, A., Zenkin, N., Lee, J., Lis, J.T. et al.
2004. Molecular mechanism of transcription inhibition by peptide antibiotic
Microcin J25. Mol. Cell, 14: 753-762.

Arnold, J.J., and Cameron, CE. 2004. Poliovirus RNA-dependent RNA
polymerase (3Dpol): pre-steady-state kinetic analysis of ribonucleotide
incorporation in the presence of M92+. Biochemistry, 43: 5126—5137.

Arnold, J.J., Gohara, D.W., and Cameron, CE. 2004. Poliovirus RNA-dependent
RNA polymerase (3Dpol): pre-steady-state kinetic analysis of ribonucleotide
incorporation in the presence of Mn2+. Biochemistry, 43: 5138-5148.

Bar-Nahum, G., Epshtein, V., Ruckenstein, A.E., Rafikov, R., Mustaev, A., and
Nudler, E. 2005. A ratchet mechanism of transcription elongation and its control.
Cell, 120: 183-193.

Batada, N.N., Westover, K.D., Bushnell, D.A., Levitt, M., and Kornberg, RD.
2004. Diffusion of nucleoside triphosphates and role of the entry site to the RNA
polymerase II active center. Proc. Natl. Acad. Sci. USA. 101: 17361-17364.

Bayro, M.J., Mukhopadhyay, J., Swapna, G.V., Huang, J.Y., Ma, L.C., Sineva, E.
et al. 2003. Structure of antibacterial peptide Microcin J25: a 21-residue lariat
protoknot. J. Am. Chem. Soc. 125: 12382—12383.

Bushnell, D.A., Cramer, P., and Kornberg, RD. 2002. Structural basis of
transcription: a-amanitin-RNA polymerase II cocrystal at 2.8 A resolution. Proc.
Natl. Acad. Sci. USA. 99: 1218—1222.

Castro, 0., Arnold, J.J., and Cameron, CE. 2005. Incorporation fidelity of the
viral RNA-dependent RNA polymerase: a kinetic, thermodynamic and structural
perspective. Virus Res. 107: 141-149.

Foster, J.E., Holmes, SF, and Erie, DA. 2001. Allosteric binding of nucleoside
triphosphates to RNA polymerase regulates transcription elongation. Cell, 106:
243-252.

Gnatt, A.L., Cramer, P., Fu, J., Bushnell, DA, and Kornberg, RD. 2001.
Structural basis of transcription: an RNA polymerase II elongation complex at 3.3
A resolution. Science (Wash. DC), 292: 1876—1882.

Gong, X.Q., Nedialkov, Y.A., and Burton, Z.F. 2004. d-amanitin blocks
translocation by human RNA polymerase II. J. Biol. Chem. 279: 27422-27427.

118

Gong, X.Q., Zhang, C., Feig, M., and Burton, Z.F. 2005. Dynamic error correction
and regulation of downstream bubble opening by human RNA polymerase II.
Mol. Cell, 18: 461-470.

Guajardo, R., and Sousa, R. 1997. A model for the mechanism of polymerase
translocation. J. Mol. Biol. 265: 8—19.

Holmes, SF, and Erie, DA. 2003. Downstream DNA sequence effects on
transcription elongation. Allosteric binding of nucleoside triphosphates facilitates
translocation via a ratchet motion. J. Biol. Chem. 278: 35597—35608.

Johnson, K.A. 1992. Transient-state kinetic analysis of enzyme reaction
pathways. Enzymes, 20: 1—61.

Johnson, K.A. 1995. Rapid quench kinetic analysis of polymerases,
adenosinetriphosphatases, and enzyme intermediates. Methods Enzymol. 249:
38—61.

Johnson, SJ, and Beese, LS. 2004. Structures of mismatch replication errors
observed in a DNA polymerase. Cell, 116: 803—816.

Johnson, S.J., Taylor, J.S., and Beese, LS. 2003. Processive DNA synthesis
observed in a polymerase crystal suggests a mechanism for the prevention of
frameshift mutations. Proc. Natl. Acad. Sci. USA. 100: 3895-3900.

Kettenberger, H., Armache, K.J., and Cramer, P. 2004. Complete RNA
polymerase II elongation complex structure and its interactions with NTP and
TFIIS. Mol. Cell, 16: 955—965.

Kireeva, M.L., Komissarova, M, Waugh, D.S., and Kashlev, M. 2000. The 8-
nucleotide-long RNA: DNA hybrid is a primary stability determinant of the RNA
polymerase II elongation complex. J. Biol. Chem. 275: 6530—6536.

Landick, R. 2004. Active-site dynamics in RNA polymerases. Cell, 116: 351-353.

Mukhopadhyay, J., Sineva, E., Knight, J., Levy, RM, and Ebright, RH. 2004.
Antibacterial peptide Microcin J25 inhibits transcription by binding within and
obstructing the RNA polymerase secondary channel. Mol. Cell, 14: 739—751.

Nedialkov, Y.A., Gong, X.Q., Hovde, S.L., Yamaguchi, Y., Handa, H., Geiger,
J.H. et al. 2003a. NTP-driven translocation by human RNA polymerase II. J. Biol.
Chem. 278: 18303—18312.

Nedialkov, Y.A., Gong, X.Q., Yamaguchi, Y., Handa, H., and Burton, Z.F. 2003b.
Transient state kinetics of RNA polymerase II elongation. Methods Enzymol. 371:
252—262.

119

Oster, G. 2002. Brownian ratchets: Darwin’s motors. Nature (London), 417: 25.
Patel, S.S., Wong, l., and Johnson, K.A. 1991. Pre-steady-state kinetic analysis
of processive DNA replication including complete characterization of an
exonuclease-deficient mutant. Biochemistry, 30: 511-525.

O'Sullivan, O., Suhre, K., Abergel, C., Higgins, D. G., and Notredame, C. (2004).
3DCoffee: combining protein sequences and structures within multiple sequence
alignments. J Mol Biol 340, 385-395.

Poirot, O., Suhre, K., Abergel, C., O'Toole, E., and Notredame, C. (2004).
3DCoffee@igs: a web server for combining sequences and structures into a
multiple sequence alignment. Nucleic Acids Res 32, W37-40.

Rosengren, K.J., Clark, R.J., Daly, N.L., Goransson, U., Jones, A., and Craik,
DJ. 2003. Microcin J25 has a threaded sidechain-tobackbone ring structure and
not a head-to-tail cyclized backbone. J. Am. Chem. Soc. 125: 12464—12474.

Sousa, R. 2003. On models and methods for studying polymerase translocation.
Methods Enzymol. 371: 3—13.

Sousa, R. 2005. Machinations of a maxwellian demon. Cell, 120: 155-156.

Temiakov, D., Patlan, V., Anikin, M., McAllister, W.T., Yokoyama, S., and
Vassylyev, D.G. 2004. Structural basis for substrate selection by T7 RNA
polymerase. Cell, 116: 381-391.

Vassylyev, D.G., Sekine, S., Laptenko, 0., Lee, J., Vassylyeva, M.N., Borukhov,
S., and Yokoyama, S. 2002. Crystal structure of a bacterial RNA polymerase
holoenzyme at 2.6 A resolution. Nature (London), 417: 712—719.

Wang, H., and Oster, G. 2002. Ratchets, power strokes, and molecular motors.
Appl. Phys. A 75: 315—323.

Westover, K.D., Bushnell, DA, and Kornberg, R.D. 2004a. Structural basis of
transcription: separation of RNA from DNA by RNA polymerase II. Science
(Wash. DC), 303: 1014—1016.

Westover, K.D., Bushnell, DA, and Kornberg, R.D. 2004b. Structural basis of
transcription; nucleotide selection by rotation in the RNA polymerase II active
center. Cell, 119: 481—489.

Wilson, K.A., Kalkum, M., Ottesen, J., Yuzenkova, J., Chait, B.T., Landick, R. et
al. 2003. Structure of Microcin J25, a peptide inhibitor of bacterial RNA
polymerase, is a lassoed tail. J. Am. Chem. Soc. 125: 12475-12483.

'Yin, Y.W., and Steitz, TA. 2002. Structural basis for the transition from initiation
to elongation transcription in T7 RNA polymerase. Science (Wash. DC), 298:
1387—1395.

120

Yin, Y.W., and Steitz, TA. 2004. The structural mechanism of translocation and
helicase activity in T7 RNA polymerase. Cell, 116: 393-404.

Yuzenkova, J., Delgado, M., Nechaev, S., Savalia, D., Epshtein, V., Artsimovitch,
I. et al. 2002. Mutations of bacterial RNA polymerase leading to resistance to
Microcin J25. J. Biol. Chem. 277: 50867-50875.

Zhang, C., and Burton, Z.F. .2004. Transcription factors HF and IIS and
nucleoside triphosphate substrates as dynamic probes of the human RNA
polymerase II mechanism. J. Mol. Biol. 342: 1085—1099.

Zhang, 0., Yan, H., and Burton, Z.F. 2003. Combinatorial control of human RNA
polymerase II (RNAP ll) pausing and transcript cleavage by transcription factor

NF, hepatitis 5 antigen, and stimulatory factor II. J. Biol. Chem. 278: 50101—
50111.

Zhang, C., Zobeck, K.L., and Burton, Z.F. 2005. Human RNA polymerase II
elongation in slow motion: role of the TFIIF RAP74 a1 helix in NTP-driven
translocation. Mol. Cell. Biol. 25: 3583—3595.

121

CHAPTER THREE

Xiong, Y., and Burton, 2. F. (2007). A tunable ratchet driving human RNA
polymerase II translocation adjusted by accurately templated nucleoside
triphosphates loaded at downstream sites and by elongation factors. J Biol Chem

282, 36582-36592. [Originally published In Press as doi:10.1074/ibc.M707014200 on
September 17, 2007]

122

CHAPTER THREE

A TUNABLE RATCHET DRIVING HUMAN RNA POLYMERASE II
TRANSLOCATION ADJUSTED BY ACCURATELY TEMPLATED
NUCLEOSIDE TRIPHOSPHATES LOADED AT DOWNSTREAM

SITES AND BY ELONGATION FACTORS

Abstract

When nucleoside triphosphate (NTP) substrates and a-amanitin are
added to a human RNA polymerase II elongation complex simultaneously, the
reaction becomes stalled in the core of the bond synthesis mechanism. The
mode of stalling is influenced by NTP substrates at the active site and at
downstream sites, and by Transcription Factor llF (TFIIF) and TFIIS. NTP
substrates templated at i+2, i+3, and i+4 downstream DNA sites can reverse the
previously stable binding of an NTP loaded at the i+1 substrate site. Deoxy-
(d)NTPs and NDPs (nucleoside diphosphates) do not substitute for NTPs at the
i+2 and i+3 positions (considered together) or the i+4, i+5, and i+6 positions
(considered together). The mode of stalling is altered by changing the number of
downstream template sites that are accurately occupied by NTPs and by

changing NTP concentration. In the presence of the translocation blocker q-

123

amanitin, a steady state condition is established in which RNA polymerase II
stably loads an NTP substrate at i+1 and forms a phosphodiester bond but
cannot rapidly complete bond synthesis by releasing pyrophosphate. These
observations support a role for incoming NTP substrates in stimulating
translocation; results appear inconsistent with the secondary pore being the sole
route of NTP entry for human RNA polymerase II; and results indicate
mechanisms of dynamic error avoidance and error correction during rapid RNA

synthesis.

124

1. Introduction

A goal of transient state kinetic analyses is to reveal the internal
mechanism of an enzyme reaction by observing synchronized millisecond events
on a millisecond time scale. In this work, the mushroom toxin d-amanitin is
utilized as a transient state inhibitor, locking human RNA polymerase II in the
core of the elongation mechanism. By altering the conditions of stalling and by
employing two distinct reaction quenching methods, details of the core
mechanism of human RNA polymerase II become apparent.

Recent x-ray crystal structures of T. thennophilus RNA polymerase
elongation complexes indicate a simple thermal ratchet mechanism for
elongation (Vassylyev et al., 2007; Vassylyev et al., 2007), with NTP substrates
loading through the secondary pore, a solvent accessible channel, to the active
site (designated i+1). Very little space is available in these structures to load
NTPs through the main enzyme channel, which holds the DNA duplex and RNA-
DNA hybrid. The downstream transcription bubble is closed at the i+2 position by

base-pairing, so it is difficult to imagine how NTP substrates could interact at i+2

or i+3 template sites. T. thermophilus (3 R422 makes a specific contact to the i+1

DNA template phosphate, and appears to provide a specific mechanism for
closing the downstream bubble at the i+2 position. This mechanism for closing

the downstream bubble is not conserved in human or yeast RNA polymerase II,

125

in which the corresponding residue to (3 R422 is pr2 G493 (human) and pr2

G506 (yeast). Furthermore, even if the downstream bubble were open in the T.
thermophi/us structure, little space is available to flip a dNMP-NTP base pair from
the i+2 to the i+1 active site. T. thennophilus structures further indicate that a
closed conformation of the “trigger loop-trigger helix” assembly may support the
catalytic intermediate, during phosphodiester bond formation.

According to a thermal ratchet mechanism, translocation and
pyrophosphate release should be rapid and spontaneous, and neither process
can be driven by an incoming, templated NTP substrate. In this paper, however,
we present data that appears inconsistent with a simple thermal ratchet
mechanism for elongation catalyzed by human RNA polymerase II. Furthermore,
the data are most consistent with an open configuration of the trigger loop
supporting phosphodiester bond addition. Iata indicate that the secondary pore
is unlikely to be the sole route of NTP entry into human RNA polymerase II.
These experiments indicate a ratchet driving fonrvard translocation that is
regulated by incoming NTP substrates and transcriptional elongation factors.
Elongation catalyzed by human RNA polymerase II, therefore, indicates features
that are not apparent in images of the T. thermophilus elongation complex.

Two models have been invoked to describe NTP loading by multi-
subunit RNA polymerases. The secondary pore NTP loading hypothesis posits
that NTPs load only to the i+1 active site and that loading to i+2, i+3, and i+4
downstream sites is impossible (Bar-Nahum et al., 2005; Batada et al., 2004;

Cramer et al., 2001; Gnatt et al., 2001; Landick, 2005; Temiakov et al., 2005;

126

Zhang et al., 1999). According to the secondary pore NTP loading hypothesis,
pyrophosphate release and translocation are expected to be rapid and
spontaneous, and NTP loading is likely to be rate-limiting (Batada et al., 2004).
The NTP-driven translocation hypothesis, by contrast, posits NTP loading at the
i+1 and i+2 sites and potentially at sites further downstream (Burton et al., 2005;
Gong et al., 2004; Gong et al., 2005; Nedialkov et al., 2003; Zhang and Burton,
2004; Zhang et al., 2005). The secondary pore is a deep and narrow channel
extending from the “funnel” on the surface of the RNA polymerase II molecule to
the deeply buried active site (Batada et al., 2004; Landick, 2005). The pore
appears to be too narrow to exchange two NTPs, so, misloading of an NTP
substrate (the usual case according to the secondary pore NTP loading
hypothesis), requires release of the incorrect NTP followed by loading of the
correct (or another incorrect) NTP (or NDP or other small molecule). In addition
to being about 15 angstroms deep and having a minimum diameter of about 7
angstroms (NTPs have a minimum diameter of about 6 angstroms), the
secondary pore has significant negative electrostatic potential (Batada et al.,
2004). Because of the negative electrostatics of the pore, Kornberg and
colleagues estimated that NTP loading might be rate-limiting (20 to 30 s“) for
NTPs loaded through the pore, even at physiological NTP concentrations: i.e. 3.1
mM ATP, 0.47 mM GTP, 0.28 mM CTP, and 0.57 mM UTP (average values for
mammalian cells) (Traut, 1994). NTP loading, however, is not rate-limiting for
human RNA polymerase II. Stable NTP loading has been determined to be 1450

+/- 330 5'1 (Zhang and Burton, 2004). The secondary pore NTP loading

127

hypothesis requires substantial competition between templated and non-
templated NTPs at i+1, but, for the most part, such competition is not observed.
The secondary pore NTP loading hypothesis does not appear to be
consistent with millisecond phase kinetic studies of RNA synthesis by human
RNA polymerase II. Kinetic analyses show that RNA polymerase II has two rate-
limiting steps in each bond addition cycle, and neither slow step corresponds to
NTP loading (Burton et al., 2005; Gong et al., 2004; Gong et al., 2005; Nedialkov
et al., 2003; Zhang and Burton, 2004; Zhang et al., 2005). One rate-limiting step
appears to include a conformational change associated with phosphodiester
bond formation. This step occurs between rapid and stable NTP binding and slow
chemistry. The other rate-limiting step occurs within the interval between
phosphodiester bond synthesis and the next stable NTP loading. This step
includes pyrophosphate release, translocation, and NTP-Mg” loading. Because
stable NTP-Mg” loading can be very rapid, the rate-limiting component of this
interval has been interpreted as translocation coupled to pyrophosphate release.
As indicated above, the T. thennophilus RNA polymerase elongation complex
structures appear most consistent with the secondary pore NTP loading model
for bacterial RNA polymerase (Vassylyev et al., 2007; Vassylyev et al., 2007).
The NTP-driven translocation hypothesis requires NTP loading to
downstream template sites, and, therefore, posits an unpaired downstream
bubble. Based on available x-ray crystal structure data, however, the extent of
opening of the downstream bubble remains controversial (Gnatt et al., 2001;

Kettenberger et al., 2004; Vassylyev et al., 2007; Vassylyev et al., 2007; Wang et

128

al., 2006; Westover et al., 2004). All available yeast RNA polymerase II
elongation complex structures are open in the downstream region. Some of
these structures are open to the i+5 or i+6 downstream positions (Wang et al.,
2006), and base pairs that have the capacity to form in the downstream region
are not observed to form (Gnatt et al., 2001; Wang et al., 2006; Westover et al.,
2004). A T. thennophilus RNA polymerase elongation complex was observed to

close in the downstream region (closed at i+2), although a key residue for

closure, (3 R422 (corresponding to S. cerevisiae pr2 G506), is not conserved in

RNA polymerase II (Vassylyev et al., 2007). If RNA polymerase II has the
capacity to close the downstream bubble, therefore, RNA polymerase II uses
different contacts for closure than T. therrnophilus RNA polymerase. Some
chemical reactivity studies support a model of variable downstream template
opening for multi-subunit RNA polymerases (Kahl et al., 2000; Zaychikov et al.,
1995), and functional studies indicate that the downstream bubble of human RNA
polymerase II can interact with NTP substrates (Gong et al., 2005).

According to the NTP-driven translocation model, an NTP loads to the
active site by transfer as a dNMP-NTP base pair from the i+2 to the i+1 site.
NTP-driven translocation appears to be coupled to pyrophosphate release,
linking initiation of each new bond addition cycle to the end of the previous cycle.
According to this model, incorrect NTPs rarely load to the active site because
NTPs are pre-sorted on the DNA template prior to active site loading, enhancing
fidelity. In this model, the secondary pore is the route of pyrophosphate release

and may be a route for expulsion of NTP substrates rejected at the active site.

129

Furthermore, NTP-driven translocation can be utilized as a means to dislodge an
NTP from the active site in response to a translocation block, indicating a role for
downstream NTPs in dynamic error avoidance and error correction (Burton et al.,
2005; Gong et al., 2005). Nomally, a translocation block signals a transcription
error.

In this paper, we show templated effects of i+2, i+3, and i+4 NTPs on
the fate of an NTP substrate loaded at i+1. This result demonstrates the major
tenet of the NTP-driven translocation model: that NTPs load to downstream
template sites while the active site is occupied. It appears from these studies that
translocation pressure can be tuned by limiting the number of downstream sites
that are accurately filled by NTPs and by altering the concentrations of accurately
templated downstream NTPs. Elongation factors TFIIF and TFIIS also appear to
adjust translocation pressure. We present a model for a tunable ratchet driving
translocation that responds to accurately templated downstream NTPs and
elongation factors, and that couples NTP-driven translocation to NTP substrate

tightening at the i+1 active site.

2. Materials and Methods

2.1 Cell culture, extracts and proteins
HeLa cells were purchased from the National Cell Culture Center
(Minneapolis, MN). Extracts of HeLa cell nuclei were prepared as described

(Shapiro et al., 1988). Recombinant TFIIF was prepared as described (Wang et

130

al., 1993; Wang et al., 1994). Recombinant human TFIIS was purified by
phosphocellulose chromatography followed by MonoS chromatography (our

unpublished procedure).

2.2 NTP stocks

Ultra pure NTP sets were purchased from Amersham Phannacia Inc.
Based on our experience using these reagents, CTP and UTP stocks appear to
be free of detectable ATP and GTP contamination. The ATP stock is
substantially free of GTP contamination. The GTP preparation appears to be
lightly contaminated with ATP, which does not complicate our experiment,
because GTP is the final addition to the reaction, after prior addition of ATP. In
reactions containing TFIIF, we observe some tendency to incorporate AMP or
GMP, when none has been deliberately added to the reaction. As we show here,
this observation results from stimulation of RNAP ll elongation by TFIIF. The
probable major source of ATP and GTP contamination in these reactions is the
HeLa extract from which we initially derive elongation complexes (ECs) and not
the NTP stocks. When TFIIS is added to the reaction in addition to TFllF,
inappropriate AMP and GMP incorporation is suppressed, indicating that the

TFIIF preparation is not likely to be a source of ATP or GTP contamination.
2.3 Preparation of Elongation Complexes

32P-labeled C40 (40 nucleotide RNA ending in a 3'-CMP) RNAP u ECs

were formed on MagneSphere (Promega, Madison, WI) metal bead-immobilized

131

templates. The adenovirus major late promoter DNA was synthesized by
polymerase chain reaction using a 5'-biotinylated upstream primer and
immobilized on streptavidin-coated beads. Factors for initiation were derived from
a HeLa extract. RNA synthesis was done in transcription buffer (12 mM HEPES,
[pH 7.9], 12% vlv glycerol, 60 mM KCI, 0.12 mM EDTA, 0.12 mM EGTA, 1.2 mM
dithiothreitol, and 0.003% IGEPAL CA-630 [Sigma-Aldrich]) containing 8 mM
MgCl2. All steps were at 25°C. For these studies, the adenovirus major late
promoter had a modified downstream sequence (+1-
ACTCTC'I'I'CCCCTTCTC'ITI'CCTTCTCTTCCCTCTCCTCC-+40-AAAGCCTI'T-

+49). The purpose of the 39 nucleotide CT cassette was to synthesize C40 with
addition of 300 pM Apc dinucleotide, 10 pM dATP, 5 pCi per reaction a-3zP-CTP
(800 Ci per mmol), and 20 uM UTP (10 min), bypassing the requirement for
addition of ATP and GTP. 20 (M CTP was then added for 10 min. C40 ECs were
washed with 1% Sarkosyl and 0.5 M KCl buffer to dissociate initiation,
elongation, pausing, and termination factors, contributed by the HeLa extract,
and re-equilibrated with transcription buffer containing 8 mM MgCl2 and 20 pM
CTP and UTP (except where noted). Functionally saturating concentrations of
TFIIF (10 pmol) and TFIIS (3 pmol per reaction) were added as indicated. TFIIF
and TFIIS were added to reactions to maximize the fraction of the post-
translocated EC at A43 (Zhang and Burton, 2004) and to maximize isomerization
reversal. TFIIF was added during the pre-incubation. TFIIS was added during the
ATP pulse. Steps after ATP addition for A43 synthesis were performed by using

the Kintek Rapid Chemical Quench-Flow (RQF-3) instrument at 25°C.

132

2.4 Running start, two-bond, double-quench protocol

The running start, two-bond protocol has been described (Funk et al.,
2002; Nedialkov et al., 2003; Nedialkov et al., 2003; Zhang et al., 2003). Here,
we improve the procedure by reporting data from experiments quenched with
EDTA or with HCI (Arnold and Cameron, 2004; Arnold et al., 2004). Rapid
quench experiments were done using the Kintek Rapid Chemical Quench-Flow
(RQF-3) instrument. All steps were done at 25 °C. Elongation was through the
sequence 40-CAAAGGCC-47. C40 ECs were incubated with 20 uM CTP and
UTP. 10 or 100 pM ATP (depending on the protocol) was added to bead-
immobilized C40 ECs on the bench top, to extend the EC to the A43 position.
During the next 30—120 seconds (depending on the protocol), ECs were loaded
into the left sample port of the RQF-3 instrument. GTP (in transcription buffer) at
twice its working concentration was loaded into the RQF-3 right sample port.
Programmed, equal volume mixing in the RQF-3 first combined ECs with GTP
substrate. Reactions were then quenched with 0.5 M EDTA or 1 M HCI, after a
precisely timed delay (>0.002 second). For EDTA-quenched reactions, ECs on
beads were collected with a magnetic particle separator and processed by
electrophoresis, as described (Funk et al., 2002; Nedialkov et al., 2003;
Nedialkov et al., 2003; Zhang et al., 2003). HCI quenching dissociates the
transcript from the RNAP ll EC. HCI-quenched reactions were delivered into
collection tubes containing a sufficient volume of 1 M KOH and 300 mM Tris-

base (~70 pl), to neutralize the pH of the solution. Beads were removed using a

133

magnetic particle separator. The solution was extracted with an equal volume of
phenol-chloroform. The aqueous phase was adjusted to 0.3 M sodium acetate
containing 20 pg of glycogen carrier and ethanol-precipitated. After vacuum
drying, RNA samples were analyzed by electrophoresis. Gel bands were
quantified using a Molecular Dynamics Phosphorimager. At the A43 stall point,
the EC fractionates into multiple conformational states, which are revealed by
their distinct elongation kinetics to G44. Formation of the G44 bond, therefore,
provides detailed insight into the mechanism and informs about recovery from
the transcriptional stall at A43. The transition from G44—tG45 provides

information about processive elongation.

2.5 Quality of RNAP ll elongation complexes

The complex kinetics we report demonstrate multiple conformers of
A43 EC at the time of GTP addition in the running start, two-bond protocol.
Because ECs were isolated on bead templates from HeLa nuclear extracts, it is
reasonable to consider whether A43 ECs differ in their kinetic properties because
of experimental treatments or because of damage to a subset of ECs during
preparation. However, A43 conformational states are determined by treatments
that occur after EC purification. The initial conformational states detected at A43
are different in the presence of TFllF or in the absence of an elongation factor,
showing that protein factors drive RNAP ll between functional modes.
Furthermore, increasing GTP concentration blurs the distinction between

different kinetic states, indicating that RNAP ll changes conformation through

134

interactions with substrate, as expected for an RNAP. Also, the distribution of
A43 states is dependent on the time of stalling at A43, demonstrating reversibility
between states (Nedialkov et al., 2003). A43 conformational states, therefore, are
selected based on treatments (protein factors, substrate and time of incubation)
after purification and are not an artifact of preparation. In the purification scheme,
RNAP II molecules are selected for the ability to accurately initiate transcription in
concert with the general initiation factors, and all C40 and A43 complexes are
active in elongation. Sarkosyl and salt treatment appears to strip all
contaminating transcription factors and complicating activities from the EC (Lei et

aL,1999)

2.6 lsomerization reversal experiments
lsomerization reversal experiments and quantification of data were
done essentially as described (Gong et al., 2005). Briefly, human C40 (40
nucleotide RNA ending in 3’-CMP) RNA polymerase II elongation complexes
were formed by accurate initiation from the adenovirus major late promoter, using
an ApC dinucleotide primer. The transcribed sequence was modified (Figure Ill-
1) to allow C40 synthesis without substantial overrun. 5’-biotinylated DNA

templates were immobilized on streptavidin beads. All reactions were done at 25

°C. C40 complexes were radiolabeled with d-aZP-CTP. C40 complexes were

extended to A43 by addition of 100 pM ATP for 30 seconds (s). During the 30 s

incubation, A43 complexes were transferred into the left sample port of the

135

KinTek RQF-3 Rapid Chemical Quench Flow instrument and then extended by
computer-controlled rapid mixing. Reactions were quenched with EDTA or HCI,
as indicated in individual protocols.

To reactions in which TFIIS was added during the ATP running start,

20 pM CTP and UTP were added during the pre-incubation to maintain the C40

stall position. In this protocol, CTP and UTP are diluted to a working

concentration of 5 uM by two equal volume mixings. When TFIIS was not present

during the pre-incubation, CTP and UTP were also omitted (except as specified
in individual protocols). To compensate for inconsistencies in sample recovery or
gel loading, transcripts were quantified as a ratio of relevant bands within each
gel lane. Gel lanes were analyzed independently for percent of signal present in
A43 plus all longer transcripts. The Molecular Dynamics Phosphorimager was
calibrated and found to give a nearly linear response over a range of at least
20,000-fold 32F exposure above a detectable background. A linear response over

a 100-fold exposure range is sufficient to detect all relevant transcripts.

2.7 Molecular images
Molecular images were prepared using the graphics program Visual
Molecular Dynamics (Humphrey et al., 1996). Structures were modified (i+2 and
i+3 NTPs placed by modeling, DNA strands extended, non-template DNA

placed), as described previously (Burton et al., 2005).

136

3. Results

3.1 Robust isomerization reversal.

In Figure Ill-1, we show an isomerization reversal experiment modeled on those
of Gong et al. (2005) (Gong et al., 2005). Interestingly, the reaction is observed
to advance and then retreat, particularly when accurately templated downstream
NTPs are present (Figure Ill-1A). In the absence of accurately templated
downstream NTPs, reversal is diminished and transient (Figure Ill-1 B).
Phosphorimager quantification of the experiment is shown in Figure Ill-1C, with
the data represented on two time scales (1 and 0.2 seconds (3)). Normally,
enzymatic reactions do not advance and then retreat. So how can such an
unusual observation be understood?

“lsomerization reversal” refers to reversal of stable NTP loading to the

i+1 active site of RNA polymerase II in response to: 1) a translocation block (a-

amanitin); and 2) accurately templated downstream NTPs. Neither d-amanitin nor

downstream NTPs induce strong reversal by themselves.

The reaction design is shown at the top of Figure Ill-1. Human RNA
polymerase II is used to synthesize C40 elongation complexes (a 40 nucleotide
RNA ending in 3’-CMP). The encoded RNA sequence downstream of C40 is 40-

CAAAGCCUUU-49. Complexes are combined with elongation factor TFIIF, which

stimulates forward translocation. On the bench top, 100 (M ATP and elongation

factor TFIIS, which stimulates RNA dinucleotide cleavage and re-start, are added

137

Figure Ill-1. Robust isomerization reversal by human RNA polymerase II

induced by accurately templated downstream NTPs. The RNA sequence and

reaction protocol are shown at the top of the figure. DA indicates a-amanitin. A) Gel

data for the reaction that includes 2.5 mM CTP and UTP, which are accurately

templated at downstream sites. No transcripts longer than C45 are detected above

background. 5 uM CTP and UTP were included in all reactions to maintain the C40

elongation complex. 0* indicates that ATP and chase NTPs were not added to C40
elongation complexes. 0 indicates that the reaction was stopped after the 30 5 ATP
pulse, with no NTP chase. Chase times are in seconds (0.002 to 1 s). The A41
transcript is generated by TFIIS-mediated dinucleotide RNA cleavage from the A43
stall position. B) Gel data for the reaction lacking accurately templated downstream
NTPs adjacent to G44 (5 mM ATP in place of 2.5 mM CTP and UTP). C)
Phosphorimager quantification of the gels shown in A and B. Two time scales (1 and
0.2 s) are shown. G44+ % indicates G44 plus all longer transcripts, indicated as

percent of total (A43 and longer transcripts). Synthesis of the G44 bond is strongly

and stably reversed in response to both q-amanitin and accurately templated

downstream NTPs, using the EDTA quench procedure.

138

 

d #0 ha eh hm

>ocococcooooccococccooccococcooococooconifi.00C

Nm 3: Odo. O._._u. so
2 Nm 3: O._._u. ab 32. >._._u

so 2 ad . a: 9.. mo;
+ fuzm man

 

Oho + 4.3:“

3%.... :3.

139

 

«6:3 :3 303553.

. 5.
O .1.
A055

1
O.
O.
‘fiiifiifii iiiummm

""."' Aim

>fivrlsiiia‘lltleitlligiriii A>fi

08"

 

140

mac—.6 _=.._ 503553.

257
0001257 5
* 0.000.000.1125.
00000000000001
5‘03
93‘ 5555555nfl5t A55
EV 555555555

{ABEL
>fivfli55555555555

050v.

Figure Ill-1 (continued).

      
  

 

 

 

40 - __ ___________________________
30 -
A IR=58%
e:
+ 20-l
g - _ _ _____________ |R=100°/o
10 -
-I- 2.5 mM GTP, CTP, UTP
0 -_ ~r~25mv_9le_§-_QmMAT_E’____
I I I I I I
0.0 0.2 0.4 0.6 0.8 1.0

Time (s)

142

Figure III-1 (continued).

 

    
 

 

 

 

 

 

é |R=69%
g 20 - lR=100%
10 _,_.__ ___________
-I- 2.5 mM GTP, CTP, UTP
-®— 2.5 mM GTP, 5.0 mM ATP y
o “—1- ————————————————————————————————
I I I I I
0 O 005 0.1 015 02
Time (s)

143

for 30 s to advance the RNA to A43. The combination of TFIIF and TFIIS
promotes fonivard translocation (T FIIF) and increases the fraction of elongation
complexes on the active synthesis pathway (TFIIF and TFIIS), increasing the
signal for the experiment. During the 30 s stall at A43, the elongation complex
preparation is transferred to the sample port of the KinTek RQF-3 rapid chemical

quench flow instrument. Using the rapid mixing device, the sample is combined

with 2.5 mM GTP, 2.5 mM CTP, 2.5 mM UTP, and 1 mM a-amanitin for various

times (0.002 to 1 s) before quenching with 0.5 mM EDTA (Figure lll-1A). The
G44 signal is highest at the 0.002 and 0.005 s time points and then decreases.
Despite the abundance of CTP substrate, a very small amount of C45 is

detected, starting at 0.05 s. Little C45 is synthesized, and no longer transcripts

are detected above background, because d-amanitin is added to the reaction,

and d-amanitin blocks translocation (Bushnell et al., 2002; Gong et al., 2004;

Gong et al., 2005). d-amanitin appears to allow synthesis of G44 primarily from

those elongation complexes that were initially post-translocated at the A43 stall

position. After G44 synthesis, C45 synthesis requires translocation, so little C45
and no longer transcripts are observed, indicating that a-amantin effectively
blocks translocation. While GTP for G44 synthesis occupies the i+1 active site,
CTP can potentially occupy the i+2 and i+3 downstream template sites, and UTP

can potentially occupy the i+4, i+5, and i+6 downstream sites, depending on the

availability of downstream DNA for specific NTP interactions.

144

This experiment demonstrates 69 % transient (0.2 s) and 58 %
sustained (1 s) isomerization reversal with added CTP and UTP, calculated as
100 % - ((G44+ % (0.2 or 1 s))/(G44+ % (burst; 0.002 or 0.005 s)) x 100 %).
G44+ indicates G44 plus all longer transcripts (in this case, C45). We observe
greater isomerization reversal (IR = 69% or 58% in this experiment) than

observed in similar experiments using the same template in Gong et al. (2005)

(Gong et al., 2005) (IR = 25%) because we used 1 mM d-amanitin in this

experiment and 0.5 mM o-amanitin in comparable experiments in Gong et al.

(2005). Also, UTP, which stimulates reversal, was omitted in the experiments of
Gong et al. (2005) (Gong et al., 2005).

When 2.5 mM CTP and UTP are substituted with 5 mM ATP (an NTP
that is not accurately templated at adjacent downstream sites), isomerization
reversal is significantly reduced and is overcome with longer times of incubation
(IR = 25% at 0.2 s and IR = 0% at 1 s). Because the extent of reversal with
negative control samples varies with the time of EDTA quench addition, reversal

appears to be primarily determined by the activity of GTP- and/or ATP-Mg”.

Because of prior addition, all reactions in Figure Ill-1 contain 5 pM CTP and UTP

(added to maintain the C40 stall position in the presence of the RNA cleavage

factor TFIIS). 5 uM CTP and UTP may be sufficient to contribute weakly to

reversal. We suggest that NTPs that are not accurately templated may interact

weakly at downstream sites and promote transient reversal. When accurately

145

templated downstream NTPs are limiting, GTP-Mg” becomes stabilized in the
i+1 active site and becomes increasingly resistant to reversal with the passage of
time.

The observation of reversal with control samples is of interest, because
it shows isomerization reversal in response to a translocation block. Furthermore,
a mechanism for exchange of the i+1 GTP substrate is indicated that responds to
free pools of NTPs. The data indicate that free NTP-Mg” is necessary to
displace the i+1 GTP substrate, because little or no reversal is detected with the
earliest times of EDTA addition (0.002 to 0.005 s), a treatment that eliminates
free NTP-Mg” but not GTP-Mg” tightened in the active site. lsomerization
reversal, therefore, (1) requires free NTP-Mg”, (2) is strongly stimulated by
accurately templated downstream NTP-Mg”, and (3) appears to be a more
dramatic phenomenon than can be fully scored using the current reaction design
because of rapid GTP-Mg” re-Ioading after release, prior to EDTA addition.

Because the experiment in Figure Ill-1 indicates the activity of
accurately templated CTP and possibly UTP at downstream sites, i.e. i+2, i+3,
and i+4, on the fate of the GTP loaded and initially tightened in the active site
(i+1), this experiment indicates NTP loading at downstream sites while the active
site is occupied. EDTA quenching is expected to chelate Mg” from downstream
CTP- and UTP-Mg” very rapidly. The i+1 GTP-Mg”, however, can be protected
within the active site from EDTA quenching, and can proceed to form the G44

phosphodiester bond after EDTA addition. We posit that, because of NTP-driven

translocation, CTP and UTP strain against the a-amanitin translocation block,

146

resulting in expulsion of a large fraction of the i+1 GTP-Mg”, which had
previously been fated for G44 bond synthesis. Thus, RNA polymerase II could
utilize incoming NTPs in dynamic error avoidance and error correction, to relieve

a translocation block (Gong et al., 2005). Essentially, RNA polymerase II appears

to react to the o-amanitin translocation block by weakening binding of the i+1

GTP in the active site. Because GTP is the accurately templated substrate for
G44 synthesis, RNA polymerase II cannot easily reject GTP, limiting the extent of
isomerization reversal. As a result, only 69 or 58 % reversal is achieved.
Chelation of Mg” at early times (0.002 and 0.005 s) results in robust
G44 synthesis, because of rapid inactivation of CTP- and UTP-Mg”. Chelation of
Mg” at a later time (i.e. 0.02 s) results in dramatically reduced G44 synthesis,
apparently because of the activities of CTP and UTP acting at downstream
template positions. In Supplementary Material, a similar isomerization reversal
experiment is shown in which CTP and UTP are replaced by CTP and ATP.
Because substituting UTP with ATP slightly reduces isomerization reversal, it

appears that i+4 UTP contributes to reversal.

An important feature of the isomerization reversal experiment is that q-

amanitin is used on a millisecond time scale as a transient state inhibitor. In this

reaction design, d-amanitin is added together with accurately templated substrate

and downstream NTPs, and a-amanitin exerts its effects within milliseconds.

RNA polymerases catalyze an elongation step on about a 5 to 50 millisecond

147

time scale, underscoring the need for millisecond time resolution in analysis of
single bond additions (Gong et al., 2004; Gong et al., 2005; Nedialkov et al.,

2003; Zhang and Burton, 2004; Zhang et al., 2005). In isomerization reversal

experiments, conditions for forward synthesis and inhibition by a-amanitin are

established together, and what results is a glimpse of the internal mechanism of
RNA synthesis, as the reaction is driven forward by substrate and blocked at the

translocation step by inhibitor.

3.2 NTPs occupy template at least to i+4.

In previous work, accurate NTP interactions were detected at the i+2
and i+3 downstream sites, while the i+1 substrate site was occupied (Gong et al.,
2005). Here, we report templated interactions of NTPs at least to the i+4 position,
while the i+1 site is occupied. The assays shown in Figure Ill-2 differ from those
shown in Figure Ill-1 in that TFIIS was omitted, reducing the magnitude of the

G44 transcription signal, but, perhaps, increasing the translocation force

generated by TFIIF (in the absence of TFIIS). Furthermore, a-amanitin was

added at 0.5 mM, instead of 1 mM. Because we wished to test many conditions

and many time points, we needed to reduce the cost incurred from adding the

higher concentration of a-amanitin. Also, by slightly slowing the rate of a-amanitin

binding by limiting concentration, some elongation complexes may be able to

further advance through the bond addition mechanism before inhibition is fully

148

established. One goal of the transient state inhibition experiment is to reveal the
inner mechanism of RNA polymerase II elongation.

Figure III-2 shows isomerization reversal experiments with addition of
different downstream NTPs and analogues to test the requirements for occupying
the i+2 and i+3 CTP sites (considered together) and the i+4, i+5, and i+6 UTP
sites (considered together), while the i+1 active site is occupied by GTP.
Precisely how many downstream DNA template sites are bound by NTP
substrates has not been determined, so we wondered whether the i+4 site is
accurately occupied. For this experiment, concentrations of downstream NTPs
were reduced 5-fold compared to Figure Ill-1, from 2.5 to 0.5 mM, close to
physiological levels, to show that physiological concentrations of downstream
NTPs are sufficient to induce isomerization reversal. Reducing the concentration
of CTP (i+2 and i+3) in the reversal experiment is also expected to accentuate
any effects of UTP (i+4, i+5 and i+6). Because TFIIS was omitted from these
reactions, there was no need to include traces of CTP and UTP to maintain the
C40 stall position, before the addition of substrate NTPs, so CTP and UTP were
not added to these reactions during the pre-incubation.

Depending on the combinations of NTPs added, reversal curves separate
into 3 classes (Figure Ill-2). In every case, 2.5 mM GTP is included as the

substrate for G44 synthesis because this concentration allows very rapid binding

of GTP at i+1. a-amanitin is added at 0.5 mM. Downstream NTPs are added at

0.5 mM. Reactions are stopped by addition of EDTA. The combination of GTP,

CTP, and UTP induces the most robust reversal (IR = 50% at 0.2 s). Omission of

149

Figure Ill-2. NTPs that are accurately templated at adjacent downstream
positions (i+2, i+3, and i+4, at a minimum) are required for robust
isomerization reversal at i+1 (G44). The protocol is shown at the top of the figure.
RNA sequence is the same as shown in Figure Ill-1. TFIIS was omitted from
reactions. CTP is required to observe robust G44=>A43 isomerization reversal. UTP
(in the presence of CTP) strongly stimulates isomerization reversal. Close analogues
of CTP and UTP do not contribute to isomerization reversal. IR% is calculated
between 0.002 or 0.005 s and 0.2 5. Error bars indicate standard deviation of three
independent experiments performed on different days. For some data points, error
bars are obscured by the graph symbol. G, A, C, U, (10, dT, dU indicates the NTP or

dNTP. 2A, 2U indicates 1 mM ATP or UTP.

150

Na 32. 0...?
i- Pm 32. 04?
t- .32.? 2 20? mU...>

‘00 ES but—U o.m 32. 8? ”OD

 
 

0.5 + 3:” 8 e 2
III 0.0.: AI 99:0?
iv. 0.0 .f 9%.
ifii 0.019 iOi QNC
ibi 0.9a... lot 0.50.:
Lqr 0.9a: IT 900?}.

15:3 5.».

151

 

locum E.» 30:233..

1... ..I..l|||||l in 33°. - o i- c
l 1 in fihfix. o- c
4/ in 3.x: 02HV + c

N Ln_flflmc..\o
0+:

 

 

 

M

..‘

152

 

 

 

m
A k B5.Q_<.§,mDDCCCL$
o.

.TII-IIIII—IIIJI-IIIJIII—l
ob 0.» oh ob ob ._.o

3.8 @

 

CTP results in almost no sustained reversal, because the i+2 and i+3 sites must
be accurately occupied by CTP to strongly stimulate reversal. Reactions lacking
CTP, therefore, result in limited, transient reversal, and serve as controls for the
importance of downstream, accurately templated NTPs.

GTP, UTP, ATP, and dCTP will not substitute for CTP, at the i+2 and
i+3 positions (IR = 8-10 % at 0.2 s) (Figure Ill-2). CDP added in place of CTP
also does not stimulate robust reversal, although the curve does not precisely
overlay with the other control curves lacking CTP (IR = 14% at 0.2 3; green stars;
the curve has a reduced burst height). The slight, transient reversal observed in

control samples lacking CTP can be attributed to effects of non-templated NTP-

Mg2+ and a-amanitin. Probably, inaccurately templated NTPs contribute to

reversal in control experiments, because different times of EDTA addition result
in different responses, indicating that the mechanism for reversal involves EDTA
chelating Mg2+ from NTP-Mg”. It is not possible to exclude non-templated NTPs
from the reaction, because i+1 substrate GTP must be added for G44 bond
synthesis.

Surprisingly, UTP, which is accurately templated at the i+4, i+5, and
i+6 downstream sites, stimulates isomerization reversal at G44 (Figure Ill-2). In
the presence of CTP, which is required for reversal, withholding UTP from the
reaction reduces reversal. This result indicates that NTPs can simultaneously
and accurately occupy the i+1 (GTP; active site), i+2 (CTP), i+3 (CTP), and i+4
(UTP) sites, at a minimum. Whether NTPs can interact at the i+5 (UTP) and i+6

(UTP) sites has not been determined. Because dTTP and dUTP (2’-deoxy UTP)

153

do not stimulate reversal (in the presence of CTP), dTTP and dUTP cannot
substitute for UTP at the i+4, i+5, and i+6 positions (considered together). GTP,
ATP, CTP, and UDP also fail to substitute for UTP in the reversal experiment.

These results demonstrate the high selectivity of downstream NTP
binding. Binding is accurately templated, but the capacity for accurate base-
pairing to template is not sufficient. NTPs cannot be substituted with NDPs or
dNTPs, even when these analogues have comparable base-pairing specificity.
This result demonstrates selectivity of NTP interactions at downstream sites,
consistent with the high fidelity of RNA synthesis catalyzed by multi-subunit RNA
polymerases.

In Figure Ill-2, we demonstrate strong suppressive effects of UTP on

synthesis of G44, in the presence of CTP and d-amanitin. As predicted by the

NTP-driven translocation hypothesis, UTP also suppresses synthesis of C45 and

C46 in the presence of the a-amanitin translocation block (Figure Ill-3). In order

to provide a fair comparison between those samples that contain CTP and UTP
to those samples that contain CTP but lack UTP, the synthesis of C45 and C46
was normalized, respectively, to synthesis of the previous G44 and C45 bonds.
Only those samples containing CTP are considered, because these are the only
samples that can synthesize C45 and C46. Remarkably, inclusion of UTP in the
reaction strongly suppresses C45 synthesis and (within detection) abolishes C46
synthesis. For C45 synthesis, UTP is accurately templated at the i+3, i+4, and

i+5 downstream positions. For C46 synthesis, UTP is accurately templated at the

154

Figure Ill-3. UTP, templated at downstream sites, suppresses synthesis of
C45 and C46. C45+IG44+ % (A) and C46+IC45+ % (B) were plotted versus
time. In this way, C45 and C46 synthesis rates were normalized for available
G44 and C45 substrate, respectively. The protocol is shown at the top of the
figure. dUTP, dTTP, UDP, ATP, GTP, and CTP do not substitute for UTP for
suppression of C45 and C46 synthesis. UTP strongly suppresses detection of the

C46 transcript.

155

 

Nm 3: mju
9m 3: o._._u

Pm 3_<_ avzju 2 ZU_U mU._.>
Pm 3: 8°,

QOLEfiL 8.. L 2 @

Ill LQI IDI |an 1&1
99¢ 99% 0.0;: 99> 0.0.5.”.

 

mac—o ___.w.

156

C45+IG44+ (%)

Figure Ill-3 (continued).

 

4:.
CP

 

 

 

N00
00

 

...\
O

4o-CAAA§_QC_UUU.49

 

O
l

 

4'_l I I I F I
0.0 0.2 0.4 0.6 0.8 1.0

Time (s)

157

Figure Ill-3 (continued).

c}: 30 - f. I7 //

3 20

+ 10 1

§ 40-CAAAQQQUUU-49
+

 

 

J

 

I I I I #—
0.0 0.2 0.4 0.6 0.8 1.0
Time (s)

158

 

i+2, i+3, and i+4 positions. ATP, GTP, CTP, dUTP, TTP, and UDP fail to replace
UTP in reversal reactions. Once again, this result is most consistent with the
NTP-driven translocation model. This result does not appear consistent with the

secondary pore NTP loading hypothesis.
3.3 The core mechanism of RNA synthesis.

In order to gain more insight into the reversal experiment, we
compared EDTA quenched reactions to reactions quenched with HCI. In addition,
to gain insight into the effects of accurately templated downstream NTPs, the
concentrations of CTP and UTP were varied (Figure Ill-4). Stopping the reaction
with acid shows the timing of synthesis of the phosphodiester bond (Arnold and
Cameron, 2004; Arnold et al., 2004; Gong et al., 2005; Zhang and Burton, 2004;
Zhang et al., 2005). Bond synthesis, however, may remain reversible until
pyrophosphate is released. If the phosphodiester bond has been formed,
however, the bond cannot reverse after acid addition. If the bond has not yet
formed, the bond cannot be formed after HCI addition. By contrast, after EDTA
quench addition, two outcomes are possible: 1) phosphodiester bond formation;
and 2) phosphodiester bond reversal. Because an NTP-Mg” can be sequestered
from EDTA chelation within the i+1 active site, the NTP-Mg” can progress to
synthesize a phosphodiester bond after EDTA addition. Alternatively, the NTP-
Mg2+ can potentially be destabilized in the active site after EDTA addition, as in
the isomerization reversal experiment.

An ongoing issue in RNA synthesis is the dynamics of the

pyrophosphate release step and its link to translocation. Based on transient state

159

kinetic analyses of human RNA polymerase II elongation, the Burton laboratory
posits that pyrophosphate release is coupled to NTP-driven translocation,
explaining the surprisingly high NTP dependence of the rate-limiting step during
elongation that corresponds to both translocation and pyrophosphate release
(Gong et al., 2005; Zhang and Burton, 2004; Zhang et al., 2005). By contrast, the
secondary pore NTP loading hypothesis seems to require that pyrophosphate
release and translocation be rapid, spontaneous and NTP-independent. Without
these features in a secondary pore NTP loading model, RNA polymerase would
pause after synthesis of every bond, independent of substrate NTP
concentration. According to estimates of the Kornberg laboratory, for NTPs to
enter the active site through the secondary pore by diffusion, NTP loading would
likely be rate-limiting for RNA synthesis (20 to 30 nucleotides s'1) (Batada et al.,
2004). So far as we can discern from rapid kinetic studies, NTP loading is not
rate-limiting for RNA synthesis (>1450 +/- 330 s") and pyrophosphate release is
not rapid and spontaneous (about 30 s'1 and coupled to the NTP-driven

translocation step) (Zhang and Burton, 2004).

In the presence of a-amanitin, RNA polymerase is stalled at the core of

the bond addition mechanism, and the position of stalling is dependent on
accurately templated downstream NTPs (Figure Ill-4). Stalling is indicated by the
failure of the HCI quench (phosphodiester bond formation) and EDTA quench
(stable GTP-Mg2+ loading) curves to converge. With addition of GTP, CTP, and
UTP to the reaction, a long-lived steady state is established in which the HCI

quench curve rises above the EDTA quench curve, but does not rise to the level

160

 

Figure Ill-4. Evidence for a tunable ratchet driving translocation adjusted by
altering the concentrations of accurately templated downstream NTPs. A
comparison of HCI quench (red circles) and EDTA quench (black squares) curves is
shown for reactions done at different CTP and UTP concentrations (0, 0.25, 0.5, and
2.5 mM). A) 2.5 mM CTP and UTP. The gray line indicates the negative control
experiment with 0 mM CTP and UTP (also shown for reference in B and C). B) 0.5
mM CTP and UTP. Error bars indicate standard deviation of three independent
experiments done on different days. Some error bars are obscured by the graph
symbols. C) 0.25 mM CTP and UTP. D) 0 mM CTP and UTP. ln A-C, in which
downstream NTPs are present, the HCI quench curve rises above the EDTA quench
curve. In D, in which adjacent downstream NTPs are absent, the EDTA quench

curve rises above the HCI quench curve.

161

Oho + ._._u:_u

Nm 32. ®._._u
o-m.m 3_<_ O._._u. C._._u

Pm 32. 8s, mU._.> ow IO.

 

mom

so-o>>>nwObEccse

mass ___.s.

162

Figure III-4 (continued).

 

 

 

5 . +2.5 mM CTP, UTP, EDTA
+2.5 mM CTP, UTP, HCI
+ No CTP, UTP, EDTA

 

0:0 0:2 0:4 0:6 0:8 1:0
Time (s)

163

Figure Ill-4 (continued).

 

 

 

+0.5 mM CTP, UTP, EDTA
"' -*-Z>- 0.5 mM CTP, UTP, HCI
I + No CTP, UTP, EDTA

 

0.0 0.2 0.4 0.6 0.8 1.0
Time (s)

164

 

Figure Ill-4 (continued).

 

 

+ 0.25 mM CTP, UTP, EDTA
5 _ ...:.. 025 mM CTP, UTP, HCI
+ No CTP, UTP, EDTA

 

 

070 0:2 074 0:6 0:8 170
Time (s)

165

Figure Ill-4 (continued).

20- Eff/t;

$315q /

v

10 '
g , + No CTP, UTP, EDTA
5 ' . *3- No CTP, UTP, HCI

0- II

 

 

 

0.0 0.2 0.4 0.6 0.8 1.0
Time (s)

166

of control samples (gray symbols) lacking CTP and UTP (Figures lll-4A-D).
Higher concentrations of CTP and UTP appear to develop increased
translocation strain on the elongation complex, and therefore increase the
persistent separation of the HCI and EDTA quench curves (compare Figures lll-
4A-C). Long-lived elevation of the HCI quench curve above the EDTA quench
curve indicates that the G44 phosphodiester bond has been formed, but

pyrophosphate release has been prevented by the translocation block. In the

absence of o-amanitin, convergence of HCI and EDTA quench curves indicates

that pyrophosphate has been released and that bond synthesis is complete.
Having the HCl quench curve rise above the EDTA quench curve indicates that,
after EDTA addition, synthesis of the G44 phosphodiester bond is reversed by
endogenous pyrophosphorolysis (G44.PPi<=>A43.GTP=>A43 + GTP).

Persistent straining of accurately templated downstream NTPs against

the a-amanitin translocation block appears to force RNA polymerase into a more

highly strained conformation than that assumed with addition of non-templated
NTPs. When EDTA is added to strained complexes, and Mg2+ is chelated,
disabling downstream CTP- and UTP-Mg”, it appears that phosphodiester bond
reversal through endogenous pyrophosphorolysis becomes a dominant event.
When accurately templated downstream NTPs are withheld from the reaction,
this strained conformation of the elongation complex is not as strongly adopted,

and robust and sustained bond reversal is not observed (Figure Ill-4D).

167

Another way of viewing the somewhat unexpected observation of HCI
quench curves rising above EDTA quench curves is this. At the time of EDTA
addition, the extent of phosphodiester bond synthesis is that shown by the HCI
quench curve. After EDTA addition, the reaction reverses (in response to EDTA
addition) to the position indicated by the EDTA quench curve. Loss of G44 signal
requires reversal of phosphodiester bond synthesis, in this case, after addition of
the EDTA quench. The bond reversal reaction requires endogenous
pyrophosphate, once again indicating the difficulty in releasing pyrophosphate

and also indicating coupling of pyrophosphate release to NTP-driven

translocation, a step that is effectively blocked by o-amanitin. Addition of 10 mM

exogenous pyrophosphate has no effect on isomerization reversal reactions
(data not shown), and pyrophosphate was not an added component of these
reactions. Only very small amounts of pyrophosphate are generated from
elongation reactions, so exogenous pyrophosphorolysis does not contribute to
reversal.

Having HCl quench curves rise above EDTA quench curves provides
further evidence of phosphodiester bond reversal in response to addition of
EDTA. According to our interpretation of isomerization reversal, G44.PPi
transcripts, which score as G44 transcripts by HCI quenching, revert to A43, in
response to EDTA addition. Because G44 transcripts are detected by HCI
quenching and lost after EDTA quenching, G44 transcripts are not lost in the
quantification by extension to C45 (or beyond) and, thus, escaping detection.

Bond reversal in response to EDTA addition indicates that chelation of CTP- and

168

 

UTP-Mg2+ may inhibit accurately templated NTP-driven translocation, which is
required to maintain formation of the G44 bond. Furthermore, accurately
templated NTP—driven translocation appears to be coupled to phosphodiester
bond synthesis and pyrophosphate release. Comparing Figures Ill-4A-C shows
that reducing CTP- and UTP-Mg” concentration reduces separation of the HCI

and EDTA quench curves, indicating that lower concentrations of incoming NTPs

may generate reduced translocation pressure against the a -amanitin

translocation block.

The experiment shown in Figure Ill-4, in which HCl quench and EDTA
quench curves fail to converge within 1 s, was done in the presence of TFIIF and
the absence of TFIIS. In previously published experiments (Gong et al., 2005),
which were done in the presence of both TFIIF and TFllS, HCI quench and EDTA
quench curves converged within 0.2 to 0.4 s when accurately templated
downstream NTPs were present. By contrast, in the absence of accurately
templated downstream NTPs, convergence required about 2 3. Slow
convergence of HCl and EDTA quench curves was explained because accurately
templated NTP—driven translocation was necessary for bond completion or bond
reversal, particularly in the presence of a translocation block. Comparing the
results in Figure Ill—4 with previous results indicates that TFIIS will interact with
elongation complexes during active synthesis and help in some way to cause
either bond completion (pyrophosphate release) or isomerization reversal (or

both).

169

To confirm the previous result, we did the experiment shown in Figure
Ill-5, in which we compare the convergence of HCI and EDTA quench curves in
the presence and absence of TFIIS. As expected, TFIIS causes rapid
convergence of the HCI quench and EDTA quench curves (within about 0.1 s)
(Figure Ill-5). In Figure Ill-5, reactions that include TFIIS are identical to those

that lack TFIIS, except that, in this case, TFIIS was added with GTP, CTP, UTP,

and a-amanitin. If TFIIS is added to the reaction during the ATP pulse, the results

are similar to those previously observed in Gong et al. (2005) (Gong et al., 2005).
Convergence of HCI quench and EDTA quench curves, in the presence of TFIIS,
demonstrates a previously unknown function of TFIIS. TFIIS interacts with stalled
elongation complexes that have loaded an NTP into the active site and are
engaged in forming a phosphodiester bond. TFIIS affects isomerization reversal,
and reversal is an apparent fidelity mechnanism, involving rejection of the
substrate NTP in response to a translocation block. It was previously unknown
that TFIIS could interact with actively transcribing complexes. Furthermore,
because TFIIS contributes to convergence of HCI and EDTA quench curves,
separation of HCI and EDTA quench curves observed in Figures Ill-4 and 5 is not
due to an artifact of quantification. The observation of phosphodiester bond
reversal in Gong et al. (2005) (Gong et al., 2005) is confirmed, indicating that
accurately templated NTP-driven translocation is coupled to isomerization

reversal, bond formation and pyrophosphate release.

170

 

 

Figure Ill-5. Evidence for a tunable ratchet driving translocation that is
regulated by TFIIS. TFIIS helps to resolve translocation blocked RNA polymerase II
elongation complexes, without inducing RNA dinucleotide cleavage. In the absence
of TFIIS, EDTA and HCl quench curves fail to converge within 1 s (red symbols). In
the presence of TFIIS (black symbols), EDTA and HCl quench curves converge

rapidly (within 0.1 to 0.2 s).

171

2.5mM GTP

 

 

 

 

2.5 mM CTP
2.5 mM UTP
0.5 mM 0A
100 “M ATP +/-TF|IS EDTétgerHCI
C40 +TFlIF : 1 30s 1 At i
25' 40-CAAAQQQLluu-49
20"
A fl
°\° 15" I
+ "0
g 10-
5-
O -

 

 

 

0.0 0.2 0.4 0.6 0.8 1.0

Figure Ill-5.

172

4. Discussion
4.1 NTP driven translocation.

In the presence of a translocation block (o-amanitin), we demonstrate

robust effects of NTPs that are accurately templated at adjacent downstream
positions (i+2, i+3, and i+4, at a minimum) on the fate of a substrate NTP loaded
and initially tightened in the i+1 active site of human RNA polymerase II. Neither
improperly templated NTPs, nor accurately templated NDPs, nor accurately
templated dNTPs show these effects, revealing chemical selectivity at multiple
downstream positions. Because these results demonstrate simultaneous NTP
occupancy of the active site and downstream sites, these results support the
NTP-driven translocation model and appear inconsistent with the secondary pore
being the sole route of NTP loading for human RNA polymerase II. Based on
available structures (Gnatt et al., 2001; Kettenberger et al., 2004; Wang et al.,
2006; Westover et al., 2004; Westover et al., 2004), it does not appear that NTPs
can access the i+2, i+3, and i+4 sites through the secondary pore.

Recent x-ray crystal structures of T. thennophilus RNA polymerase
elongation complexes indicate that NTP-driven translocation may not be general
for all multi-subunit RNA polymerases. Because T. thennophilus structures have
a closed downstream bubble (closed at i+2), this raises the question of whether
bacterial RNA polymerase can bind an NTP substrate at i+2. It should be noted
that merely because T. therrnophilus has a closed downstream bubble in some

elongation complex structures, this does not show that the downstream bubble is

173

 

always closed. For instance, allosteric effects of substrate NTPs with E. coli RNA
polymerase may indicate interactions of an NTP substrate at or near the i+2 site
(Foster et al., 2001; Holmes and Erie, 2003). Single molecule studies of
elongation catalyzed by E. coli RNA polymerase are also most consistent with

NTP substrates binding at or near i+2 (Abbondanzieri et al., 2005).

4.2 lsomerization reversal and the “trigger loop” hypothesis.

In X-ray crystal structures, multiple placements of the trigger loop
(pr1 1081 to 1093, S. cerevisiae residues) have been observed, including
“open” and “closed” conformations (Vassylyev et al., 2007; Vassylyev et al.,
2007; Wang et al., 2006). The trigger loop appears to close over the i+1 NTP
substrate as part of an induced fit mechanism, tightening the catalytic
intermediate to promote catalysis and ensure transcriptional fidelity. The trigger
loop, however, is not required for phosphodiester bond formation by multi-subunit
RNA polymerases. A radical deletion of the E. coli RNA polymerase trigger loop
is slow for elongation, but it remains active (Toulokhonov et al., 2007). A four
amino acid substitution mutant designed to block trigger loop dynamics is also

very slow in elongation (Vassylyev et al., 2007). Based on the T. themiophilus
elongation complex structures, E. coli B’ M932, R933, and H936 are predicted to
be key residues to support elongation and to contribute to transcriptional fidelity.

Although mutant proteins in these residues were tested for pausing half-life,

activities in elongation compared to wild type RNA polymerase were not reported

(Toulokhonov et al., 2007). E. coli (3’ T934M, found in this critical region, is rapid

174

in elongation compared to wild type (Weilbaecher et al., 1994). Mutations in the
trigger loop increase the duration of pauses at an RNA hairpin-dependent pause
site from the histidine operon, demonstrating that the trigger loop has an

important role in regulating transcriptional pausing (Toulokhonov et al., 2007).

Trigger loop closing, however, appears inconsistent with o-amanitin

(Bushnell et al., 2002; Wang et al., 2006) and with TFIIS (Kettenberger et al.,

2003; Kettenberger et al., 2004) binding to the active elongation complex.

Because the data presented in this paper indicate that q-amanitin and TFIIS

affect NTP-Mg2+ isomerization and pyrophosphate release, it appears that a-

amanitin and TFIIS can interact with human RNA polymerase II during catalysis.
Results in the human RNA polymerase II system, therefore, are most
understandable if an open trigger loop conformation supports catalysis. At this
time, we do not have a clear explanation for this apparent discrepancy with the T.
thermophilus elongation complex structure. Yeast RNA polymerase II elongation

complexes with closed trigger loop conformations do not appear to be precise

catalytic intermediates because of a bend in the bridge q-helix and distortions in

base pairs at the i and i+1 positions (Vassylyev et al., 2007; Wang et al., 2006).
To obtain x-ray crystal structures that approach a catalytic intermediate requires
blocking NMP incorporation, so non-incorporatable NTP analogues or a 3’-deoxy
chain terminator at the RNA 3’ end are used. Structures with a closed trigger loop

conformation, therefore, could represent a response to the block to NMP

175

incorporation. Critical mutations in the trigger loop are so far characterized most
completely for transcriptional pausing, rather than for catalysis or transcriptional
fidelity, so it is not yet clear whether a closed trigger loop conformation is

normally utilized for bond synthesis (Toulokhonov et al., 2007).

4.3 A tunable ratchet driving translocation.

To explain the effects of a-amanitin, TFIIF, TFIIS, and downstream NTPs in

adjusting translocation pressure, we propose a model for a tunable ratchet

driving translocation (Figure Ill-6). Figure 6A shows a schematic of a tunable

translocation ratchet, with the bridge a-helix forming its core. Figure Ill-GB shows

a structural image of the bridge o-helix and a proposed conformational switch

(the “funnel”; S. cerevisiae pr1 673 to 763), proposed to help roll the ratchet
domain fonlvard and back. A primary feature of the tunable ratchet model is that it
posits conformational coupling between the RNA polymerase II active site (i+1)

and the NTP-driven translocation mechanism, which involves templated NTP

binding to downstream sites (i+2, i+3, and i+4). In this model, the bridge a-helix

forms the center of a cylindrical domain that rolls fonrvard to translocate the DNA
and then slides back relative to DNA, during each bond addition step.
Tightening of the i+1 NTP substrate is posited to place strain on the

DNA template stimulating fonNard translocation. A right angle bend in the

template DNA over the bridge d-helix is suggested to link catalysis to

176

 

Figure Ill-6. A model for a tunable ratchet driving translocation by human RNA

polymerase II. A) The bridge a-helix (yellow) is depicted as forming the core of a

cylindrical domain (light blue) that rolls forward and then slides back, relative to the
DNA template (green; phosphates indicated as yellow dots) during each
translocation step. The “funnel” (pink) connects active site and rolling ratchet
functions. pr2 R1020 and pr1 K752 are proposed to be sensors of progression
of each bond addition. RNA and NTP substrates are red. Mg2+ atoms are green
dots. Tilting of the funnel (pink), gray lines and letters indicate rotation of the ratchet
domain. pr1 Y836 (S. cerevisiae residue number) is purple. Y836 is posited to be

important in switching the bend point in the DNA template strand as it passes over

the bridge a-helix. a) relaxed elongation complex; b) isomerized (tightened)

elongation complex; c) complex after phosphodiester bond formation; d and e)

maximally strained elongation complexes, in which NTP-driven translocation is
coupled to pyrophosphate release; and f) relaxed elongation complex. q-amanitin (a
A) primarily blocks the dominant translocation step, but limited translocation is
suggested to be associated with multiple steps in the mechanism. B) Structural
model of the rolling ratchet mechanism. Top panels: three views of the bridge o-helix
(pr1 810 to 845) (yellow) and C-terminal cap (pr1 846 to 868) (yellow) and the
funnel region (pr1 526 to 763) (mauve) are shown in relation to the transcription

bubble. The proposed cylindrical ratchet domain location is indicated by white circles

(left and right images) and a white rectangle (center image). Bottom panel: a detail

177

‘1“

of the right angle bend in the DNA template as it passes over the bridge a-helix is

shown with basic residue contacts to phosphates. NTP substrates are shown at the
i+1 (orange), i+2 (pink), and i+3 (lime) positions. The i+2 and i+3 NTPs were placed
by modeling. Mg2+ atoms are green; the DNA template strand is green; the non-
template DNA strand is white; RNA is red; phosphorous atoms are indicated as tan
spheres; basic amino acids (pr1 R326, K330, K332, R337, K752, R839, R1386,

R1391, and pr2 R1129) are blue; pr1 Y836 is purple; phenylalanine residues

found at the N-terminal end of the bridge a-helix (pr1 813 to 815) are white. DNA

missing from available structures was placed by modeling (Burton et al., 2005).
Multiple crystal structures were used to generate this figure (Gnatt et al., 2001;

Westover et al., 2004).

178

 

         
 

at: s
<30 0.20 an 30:50

.3 5003013¢0= E

179

       

     

4030.08 00:0
01.33:?
0.2. .0223

ll

4300.00.20...
3.. 3.0000

240.0103:
0.0-3.00020:

i

9)

«65.0 ___.a.

”gum:

305.0 ___.m 303.3003.

_+A

/I

 

180

translocation (Bar-Nahum et al., 2005; Gnatt et al., 2001). We posit that the

bridge a-helix is at the core of a domain that rolls fonNard, stimulating fonNard

translocation. Tightening of the i+1 NTP is posited to increase contact between
the triphosphate and basic residues pr2 R1020 and pr1 K752 (residue
numbers are for S. cerevisiae). Located on the highly conserved 750-GSKG-753
loop, which is linked to the “funnel” (pr1 663 to 763), pr1 K752 is posited to
be a major sensor of the state of the catalytic cycle (Wang et al., 2006). As
catalysis progresses, a rocking motion is developed in the funnel that stimulates
rolling of the rolling ratchet domain. Movement of accurately templated
downstream NTPs contributes to translocation force. NTP-driven translocation is
thought to be coupled to pyrophosphate release, so bond completion would be

coupled to accurate NTP driven translocation for the next bond addition. As the

next NTP is transferred into the RNA polymerase II active site, the bridge a-helix

domain rolls back, sliding relative to the DNA template (Figure Ill-6A).

Kornberg and colleagues suggested that the bridge a-helix might cycle

between bending and straightening to drive fonlvard translocation (Gnatt et al.,
2001; Wang et al., 2006). Nudler, Ruckenstein and colleagues suggested a

similar dual ratchet mechanism for elongation, involving conformational changes

of the bridge a-helix and associated “G” or “trigger” loop (Bar-Nahum et al.,

2005). We favor a mechanism in which the bridge a-helix lies at the core of a

larger domain that rolls (relative to template) more than bends to drive forward

181

5’

 

i1 '

translocation. Rolling of the bridge a-helix forward causes pr1 Y836 to clash

with template DNA, helping to induce a change in the location of the 90 degree
bend in the DNA template strand over the bridge (Figure Ill-6A). Changing the
position of the bend in the DNA is suggested to be an important part of a
switching mechanism to alter contacts between basic groups (pr1 R326, K330,
K332, R337, R839, R1386, and pr2 R1129) and DNA phosphates (Figure Ill-
68, lower panel). Altering the DNA bend point is posited to cause first releasing
and then re-setting phosphate contacts to the next downstream position,
supporting stepped translocation of the DNA.

Breaking phosphate contacts results in a reverse roll of the ratchet

domain, causing the DNA template to slide relative to the bridge a-helix. In this

way, the bridge a-helix is at the core of a molecular ratchet driving forward

translocation. We propose that larger motions of RNA polymerase II may support

the ratchet roll of the bridge domain. For instance, the funnel may rock toward

the bridge a-helix, to enhance the fonrvard roll, and then rock back during

template sliding, during the reverse ratchet roll (Figure Ill-6). Because the funnel
includes pr1 K752, which is poised to interact with the i+1 NTP triphosphate,
this establishes a connection that is mediated through the funnel structure

between the i+1 active site and translocation driven by accurately templated

NTPs at i+2, i+3, and i+4. A functional link between the funnel and the bridge a-

helix is indicated by the structure of o-amanitin bound to RNA polymerase II,

182

because translocation is blocked when a-amanitin is wedged between the funnel

and the bridge a-helix, where the toxin would be expected to inhibit funnel

rocking and bridge helix rolling (Bushnell et al., 2002). Available genetics is
consistent with the proposed “rock and roll” mechanism for RNA polymerase II
translocation, because mutagenic analyses demonstrate the importance of the
funnel, the cleft (proposed to be important for i+3 NTP binding), and fork loop 2
region (proposed to be important for i+2 NTP binding) in translocation (T rinh et
al,2006)

A reversibly rolling ratchet that is tightened by induced fit at the i+1
active site and by NTP-driven translocation at i+2, i+3, and i+4 downstream sites
fulfills the requirements for a mechanism in which translocation tension can be
adjusted. The ratchet is wound forward by the i+1 substrate NTP as it tightens in
the active site. lnaccurately templated substrates fail to tighten at i+1 and,
therefore, are expected to be rapidly replaced before they can engage in
chemistry. The ratchet is also proposed to be driven by accurately templated
NTP substrates bound at downstream sites, explaining how NTP-driven
translocation can be coupled to pyrophosphate release. Elongation factors
appear to increase (TFIIF) or decrease (T FIIS) tension on the ratchet through
allosteric interactions with RNA polymerase II. Whether or not bacterial RNA
polymerases utilize NTP-driven translocation, we posit that an induced fit, “rock
and roll” conformational change may be a feature of translocation by all multi-

subunit RNA polymerases.

183

4.4 NTP-driven translocation and fidelity.

The rock and roll mechanism supports very deliberate, stepwise
translocation of the DNA template, limited to single-base pair steps. Free
reversibility of RNA polymerase II between pre- and post-translocated elongation
complexes has not been observed in millisecond phase kinetic studies, although
a rapidly reversible thermal ratchet and spontaneous pyrophosphate release are
predicted features of the secondary pore NTP loading model and the dual ratchet
model. In this paper, we show that human RNA polymerase II does not release
pyrophosphate spontaneously. lsomerization reversal experiments indicate that
pyrophosphate release is linked to translocation, rendering pyrophosphate
release coupled to accurate NTP-driven translocation a potential fidelity check

point in each step of RNA synthesis. In the presence of the translocation blocker

o-amanitin, the reaction can be halted after phosphodiester bond synthesis but

prior to pyrophosphate release. Accurately templated downstream NTPs seem to
drive fonrvard translocation, because they appear to adjust the set point of the
ratchet. These results support the NTP—driven translocation mechanism. No
model requiring either rapidly reversible translocation or secondary pore NTP
loading appears consistent with these data.

The primary issue in considering NTP-driven translocation is not
whether NTPs enter the structure through the secondary pore, as has been
generally assumed, or through the main enzyme channel, as appears to be the

case for human RNA polymerase II. The essential issue is the fidelity of RNA

184

synthesis. NTP-driven translocation induced by accurately templated
downstream NTPs is a high fidelity mechanism, because it provides a
mechanism for pre-sorting and repeated confirmation of accurately loaded NTP
substrates. For human RNA polymerase II, if NTPs are retained in the queue,
NTP substrates are confirmed versus template at the i+4, i+3, i+2, and i+1 sites,
prior to NMP incorporation at i+1. Remarkably, although necessary, the capacity
for accurate base-pairing is not sufficient for retention in the queue. Because they
fail to support isomerization reversal, very close NTP analogues such as dNTPs
and NDPs appear to be rejected downstream of the active site, even when they
are accurately templated. In addition to preventing inaccurate loading of NTPs to
the active site, accurately templated downstream NTPs appear to generate
translocation force that is used either: 1) to complete synthesis of correct bonds;
or 2) to prevent and reverse transcription errors in progress. In this manner,
accurately templated NTP-driven translocation appears to be coupled to fidelity

mechanisms.

185

References

Abbondanzieri, E. A., Greenleaf, W. J., Shaevitz, J. W., Landick, R., and Block,
S. M. (2005). Direct observation of base-pair stepping by RNA polymerase.
Nature 438, 460-465.

Arnold, J. J., and Cameron, C. E. (2004). Poliovirus RNA-dependent RNA
polymerase (3Dpol): pre-steady-state kinetic analysis of ribonucleotide
incorporation in the presence of MgZ+. Biochemistry 43, 5126-5137.

Arnold, J. J., Gohara, D. W., and Cameron, C. E. (2004). Poliovirus RNA-
dependent RNA polymerase (3Dpol): pre-steady-state kinetic analysis of
ribonucleotide incorporation in the presence of Mn2+. Biochemistry 43, 5138-
5148.

Bar-Nahum, G., Epshtein, V., Ruckenstein, A. E., Rafikov, R., Mustaev, A., and
Nudler, E. (2005). A ratchet mechanism of transcription elongation and its
control. Cell 120, 183-193.

Batada, N. N., Westover, K. D., Bushnell, D. A., Levitt, M., and Kornberg, R. D.
(2004). Diffusion of nucleoside triphosphates and role of the entry site to the RNA
polymerase II active center. Proc Natl Acad Sci U S A 101, 17361-17364.

Burton, 2. F., Feig, M., Gong, X. 0., Zhang, C., Nedialkov, Y. A., and Xiong, Y.
(2005). NTP driven translocation and regulation of downstream bubble opening
by multi-subunit RNA polymerases. Biochem Cell Biol 83, 486-496.

Bushnell, D. A., Cramer, P., and Kornberg, R. D. (2002). Structural basis of
transcription: alpha-amanitin-RNA polymerase II cocrystal at 2.8 A resolution.
Proc Natl Acad Sci U S A 99, 1218-1222.

Cramer, P., Bushnell, D. A., and Kornberg, R. D. (2001). Structural basis of
transcription: RNA polymerase II at 2.8 angstrom resolution. Science 292, 1863-
1876.

Foster, J. E., Holmes, S. F., and Erie, D. A. (2001). Allosteric binding of
nucleoside triphosphates to RNA polymerase regulates transcription elongation.
Cell 106, 243-252.

Funk, J. D., Nedialkov, Y. A., Xu, D., and Burton, 2. F. (2002). A key role for the

alpha 1 helix of human RAP74 in the initiation and elongation of RNA chains. J
Biol Chem 277, 46998-47003.

186

 

Gnatt, A. L., Cramer, P., Fu, J., Bushnell, D. A., and Kornberg, R. D. (2001).
Structural basis of transcription: an RNA polymerase II elongation complex at 3.3
A resolution. Science 292, 1876-1882.

Gong, X. 0., Nedialkov, Y. A., and Burton, Z. F. (2004). Alpha-amanitin blocks
translocation by human RNA polymerase II. J Biol Chem 279, 27422-27427.

Gong, X. 0., Zhang, C., Feig, M., and Burton, Z. F. (2005). Dynamic Error
Correction and Regulation of Downstream Bubble Opening by Human RNA
Polymerase ll. Mol Cell 18, 461-470.

Holmes, S. F., and Erie, D. A. (2003). Downstream DNA sequence effects on
transcription elongation. Allosteric binding of nucleoside triphosphates facilitates
translocation via a ratchet motion. J Biol Chem 278, 35597-35608.

Humphrey, W., Dalke, A., and Schulten, K. (1996). VMD: visual molecular
dynamics. J Mol Graph 14, 33-38, 27-38.

Kahl, B. F., Li, H., and Paule, M. R. (2000). DNA melting and promoter clearance
by eukaryotic RNA polymerase I. J Mol Biol 299, 75-89.

Kettenberger, H., Armache, K. J., and Cramer, P. (2003). Architecture of the
RNA polymerase ll-TFIIS complex and implications for mRNA cleavage. Cell
114, 347-357.

Kettenberger, H., Armache, K. J., and Cramer, P. (2004). Complete RNA
Polymerase II Elongation Complex Structure and Its Interactions with NTP and
TFIIS. Mol Cell 16, 955-965.

Landick, R. (2005). NTP-entry routes in multi-subunit RNA polymerases. Trends
Biochem Sci 30, 651-654.

Lei, L., Ren, D., and Burton, 2. F. (1999). The RAP74 subunit of human
transcription factor "F has similar roles in initiation and elongation. Mol Cell Biol
19, 8372-8382.

Nedialkov, Y. A., Gong, X. 0., Hovde, S. L., Yamaguchi, Y., Handa, H., Geiger,
J. H., Yan, H., and Burton, 2. F. (2003). NTP-driven translocation by human RNA
polymerase II. J Biol Chem 278, 18303-18312.

Nedialkov, Y. A., Gong, X. 0., Yamaguchi, Y., Handa, H., and Burton, 2. F.
(2003). Assay of transient state kinetics of RNA polymerase II elongation.
Methods Enzymol 371, 252-264.

Shapiro, D. J., Sharp, P. A., Wahli, W. W., and Keller, M. J. (1988). A high-
efficiency HeLa cell nuclear transcription extract. Dna 7, 47-55.

187

Temiakov, D., Zenkin, N., Vassylyeva, M. N., Perederina, A., Tahirov, T. H.,
Kashkina, E., Savkina, M., Zorov, S., Nikiforov, V., lgarashi, N., et al. (2005).
Structural basis of transcription inhibition by antibiotic streptolydigin. Mol Cell 19,
655-666.

Toulokhonov, L, Zhang, J., Palangat, M., and Landick, R. (2007). A central role of
the RNA polymerase trigger loop in active-site rearrangement during
transcriptional pausing. Mol Cell 27, 406-419.

Traut, T. W. (1994). Physiological concentrations of purines and pyrimidines. Mol
Cell Biochem 140, 1-22.

Trinh, V., Langelier, M. F., Archambault, J., and Coulombe, B. (2006). Structural
perspective on mutations affecting the function of multisubunit RNA polymerases.
Microbiol Mol Biol Rev 70, 12-36.

Vassylyev, D. G., Vassylyeva, M. N., Perederina, A., Tahirov, T. H., and
Artsimovitch, I. (2007). Structural basis for transcription elongation by bacterial
RNA polymerase. Nature 448, 157-162.

Vassylyev, D. G., Vassylyeva, M. N., Zhang, J., Palangat, M., Artsimovitch, I.,
and Landick, R. (2007). Structural basis for substrate loading in bacterial RNA
polymerase. Nature 448, 163-168.

Wang, B. Q., Kostrub, C. F., Finkelstein, A., and Burton, Z. F. (1993). Production
of human RAP30 and RAP74 in bacterial cells. Protein Expr Purif 4, 207-214.

Wang, B. Q., Lei, L., and Burton, 2. F. (1994). Importance of codon preference
for production of human RAP74 and reconstitution of the RAP30/74 complex.
Protein Expr Purif 5, 476-485.

Wang, D., Bushnell, D. A., Westover, K. D., Kaplan, C. D., and Kornberg, R. D.
(2006). Structural basis of transcription: role of the trigger loop in substrate
specificity and catalysis. Cell 127, 941-954.

Weilbaecher, R., Hebron, C., Feng, G., and Landick, R. (1994). Tennination-
altering amino acid substitutions in the beta' subunit of Escherichia coli RNA

polymerase identify regions involved in RNA chain elongation. Genes Dev 8,
2913-2927.

Westover, K. D., Bushnell, D. A., and Kornberg, R. D. (2004). Structural Basis of
Transcription: Separation of RNA from DNA by RNA Polymerase II. Science 303,
1014-1016.

Westover, K. D., Bushnell, D. A., and Kornberg, R. D. (2004). Structural Basis of
Transcription; Nucleotide Selection by Rotation in the RNA Polymerase II Active
Center. Cell 119, 481-489.

188

Zaychikov, E., Denissova, L., and Heumann, H. (1995). Translocation of the
Escherichia coli transcription complex observed in the registers 11 to 20:
"jumping" of RNA polymerase and asymmetric expansion and contraction of the
"transcription bubble". Proc Natl Acad Sci U S A 92, 1739-1743.

Zhang, C., and Burton, Z. F. (2004). Transcription Factors HF and IIS and
Nucleoside Triphosphate Substrates as Dynamic Probes of the Human RNA
Polymerase ll Mechanism. J Mol Biol 342, 1085-1099.

Zhang, 0., Yan, H., and Burton, Z. F. (2003). Combinatorial control of human
RNA polymerase II (RNAP ll) pausing and transcript cleavage by transcription
factor "F, hepatitis delta antigen, and stimulatory factor II. J Biol Chem 278,
50101-50111.

Zhang, C., Zobeck, K. L., and Burton, Z. F. (2005). Human RNA Polymerase ll
Elongation in Slow Motion: Role of the TFIIF RAP74 {alpha}1 Helix in Nucleoside
Triphosphate-Driven Translocation. Mol Cell Biol 25, 3583-3595.

Zhang, 6., Campbell, E. A., Minakhin, L., Richter, C., Severinov, K., and Darst,
S. A. (1999). Crystal structure of Thermus aquaticus core RNA polymerase at 3.3
A resolution. Cell 98, 81 1-824.

189

CHAPTER FOUR

SUMMARY AND FUTURE PLAN

Multi-subunit RNAPs are among the largest and most dynamic
enzymes known. RNAPs represent a unique class of enzyme with a
characteristic subunit structure conserved in eubacteria, archaea, eukarya, and
some viruses. Persistence of this structure through evolution is evidence of its
importance in supporting complex life. While transcribing a DNA template, a
multi-subunit RNAP behaves as a molecular motor with a translocation step that
can be largely sequestered from external forces exerted on the DNA or RNA
(Abbondanzieri et al., 2005; Adelman et al., 2002; Dalal et al., 2006; Davenport
et al., 2000; Greenleaf and Block, 2006; Herbert et al., 2006; Shaevitz et al.,
2003; Vassylyev and Artsimovitch, 2005; Wang et al., 1998; Yin et al., 1995).
Yeast RNAP II is a close homologue of human RNAP II, and elongation complex
structures are available for the yeast enzyme (Gnatt et al., 2001; Kettenberger et
al., 2004; Wang et al., 2006; Westover et al., 2004; Westover et al., 2004). Many
yeast mutant RNAPs with defects in elongation are known, providing potential
insight into the transcriptional mechanism (Trinh et al., 2006). Bacterial RNAP is
a more distant evolutionary homologue of the yeast and human enzymes.

Relevant X-ray crystal structures are available for bacterial RNAPs (Vassylyev at

190

al., 2007; Vassylyev et al., 2007), and many mutant proteins have been identified
in the bacterial system (T rinh et al., 2006). In studies of human RNAP II, the
Burton laboratory has developed evidence for previously unknown fidelity
mechanisms in transcriptional elongation related to translocation mechanisms
(Gong et al., 2005; Xiong and Burton, 2007).

X-ray crystal structures have recently become available for yeast
RNAP II and bacterial T. thermophilus RNAP elongation complexes (Gnatt et al.,
2001; Kettenberger et al., 2004; Vassylyev et al., 2007; Vassylyev et al., 2007;
Wang et al., 2006; Westover et al., 2004; Westover et al., 2004). In particular, T.
thermophilus structures indicate a possible mechanism for bacterial RNAPs,
although the predicted mechanism has not yet been fully vetted by functional
dynamic studies. At this time, we council caution in the interpretation of these
static X-ray images, particularly for such a large and dynamic protein as RNAP.
Compared to other enzymes, RNAPs contain a large proportion of water, which
in some cases is removed by dehydration of crystals. T. thermophilus is a hyper
thermophile, and its RNAP is not functional at the temperatures at which
structures are obtained. Some aspects of the crystal structure, therefore, may not
fully represent the functional enzyme. Because of high homology among multi-
subunit RNAPs, the core mechanisms of eukaryotic and prokaryotic RNAPs are
expected to be very similar. Because our kinetic analysis show that human
RNAP ll utilizes an NTP-driven translocation mechanism (Burton et al., 2005;
Gong et al., 2005; Nedialkov et al., 2003; Xiong and Burton, 2007; Zhang and

Burton, 2004; Zhang et al., 2003), we wonder whether bacterial enzymes also

191

utilize this mechanism, despite some features of structures that might be invoked
to argue against this conclusion.

The inferred bacterial RNAP model could be described as a reversible
isomerization, thermal ratchet translocation mechanism. “lsomerization”, or
substrate tightening, appears to involve closing of the trigger loop over the i+1
NTP. Based on structures, translocation and pyrophosphate release are
expected to be rapid and spontaneous. NTPs have been hypothesized to load to
the deeply buried active site past the “funnel”, and through the “secondary pore”,
a solvent accessible channel. In order for an NTP to load through the secondary
pore, the “trigger loop” is likely to be in an open conformation, because a closed
trigger loop conformation appears to occlude the pore (Vassylyev et al., 2007;
Vassylyev et al., 2007). In available T. thermophilus RNAP elongation complexes
structures, there does not appear to be space to load an NTP substrate through
the main enzyme channel, so the pore has been thought to be the sole route of
NTP entry. Furthermore, the downstream DNA template is within a double helix
with the non-template DNA strand in T. thermophilus structures, and a specific
contact of 8 R422 to the i+1 template strand DNA phosphate maintains bubble
closure (Vassylyev et al., 2007; Vassylyev et al., 2007). In T. thermophilus,
therefore, loading of NTP substrates to downstream template sites, as postulated
for human RNAP II, seemed unlikely. The i+1 substrate NTP, therefore, might be
expected to load only to the post-translocated elongation complex, not the pre-
translocated complex. The NTP may initially bind in a pre-insertion state that may

have an open trigger loop conformation (Vassylyev et al., 2007; Vassylyev et al.,

192

2007). Closing of the trigger loop tightens the NTP within the active site. This
transition is also described as conversion of a partially disordered trigger loop to
a trigger helix configuration. Closing of the trigger loop might provide an induced
fit fidelity mechanism to confirm the identity of the dNMP-NTP base pair in the
active site. Mis-pairs are expected to fail to tighten, resulting in release of the
rejected NTP and replacement of the NTP in the active site. Replacement of
incorrectly loaded NTPs must be very rapid, because little competition by non-
templated NTPs is observed for multi-subunit RNAPs, and stable NTP acquisition
is surprisingly fast (Gong et al., 2005; Xiong and Burton, 2007; Zhang and
Burton, 2004; Zhang et al., 2005).

Yeast RNAP Il elongation complex structures indicate that NTPs can
occupy either an “E” (for “entry”) site position or an “A” (for “acceptor”) site
position (Wang et al., 2006; Westover et al., 2004). Because NTPs in the E site
are oriented so that they cannot form accurate base pairs to the DNA template, it
is hypothesized that NTPs may flip from the E site into the A site as they enter
the active site through the secondary pore. Because of the static nature of X-ray
structures, however, it is difficult to be certain that an NTP is entering the active
site from the E site.

Based on our current understanding, we believe some aspects of the
currently dominant model for elongation might not be correct. We posit that
NTPs load through the main enzyme channel, not the secondary pore, for human
and bacterial RNAPs. The secondary pore has been shown to have negative

electrostatic potential, and, based on electrostatics, it was calculated that NTP

193

loading through the secondary pore would be expected to be a rate-limiting step
in elongation. NTP loading, however, is not rate-limiting for multi-subunit RNAPs.
Trigger loop closing appears to be important for catalysis, but, if NTPs load
through the main enzyme channel to pre-translocated template positions, the
trigger loop need not open after each bond addition to admit the next NTP.
According to our kinetic model, trigger loop opening would tend to occur in
response to NTP mis-loading in the active site. In this way, trigger loop opening
would initiate transcriptional pausing.

None of the yeast RNAP ll structures obtained to date appear to
precisely recapitulate a catalytic intermediate. Structures with a closed trigger
loop conformation (2E2H and 2NVZ) (Wang et al., 2006; Westover et al., 2004)
have a distinct bend in the bridge or-helix, which induces base pair distortions at
the i (RNA 3’ end) and i+1 (substrate NTP) sites. Because precise geometry is
essential to accurately catalyze addition of four different substrates, these
distortions do not appear to be consistent with phosphodiester bond formation
(Vassylyev et al., 2007). Furthermore, these structures do not closely resemble
the T. thermophilus elongation complex structure, with a closed’ trigger loop,
which has characteristics expected of a catalytic intermediate. Unlike the yeast
structures, the T. thermophilus structure has a straight conformation of the bridge
or-helix and reasonable geometry of the DNA template, RNA and the NTP-Mg2+
for catalysis. This suggests that a closed trigger loop and a straight conformation
of the bridge or-helix may be required to support catalysis. Poor geometry in

yeast elongation complex structures may be due to the block to catalysis, which,

194

although necessary to obtain a crystal structure with a loaded NTP substrate or
NTP analogue, may be sensed by yeast RNAP II as a block to isomerization (i.e.
trigger loop closing) and translocation. For human RNAP ll, blocks to
isomerization and/or translocation appear to signal transcription errors (Gong et
al., 2005; Xiong and Burton, 2007).

Yeast RNAP ll elongation complex structures also appear to differ from
T. thermophilus in the configuration of the downstream transcription bubble. All
yeast RNAP ll elongation complexes obtained to date have unpaired bases in the
downstream region, but opening is variable (Gnatt et al., 2001; Kettenberger et
al., 2004; Wang et al., 2006; Westover et al., 2004; Westover et al., 2004). One
structure from the Cramer laboratory is closed at i+3. The structure is open at i+2
because it was forced open by mis-pairing the DNA template (Kettenberger et al.,
2004). Other structures appear to be open to i+4 or to i+6, whether or not the
non-template strand has a greater capacity to pair (Gnatt et al., 2001; Wang et
al., 2006; Westover et al., 2004). At this time, the capacity of the downstream
bubble to open or close has not clearly been determined for eukaryotic RNAP II.
T. thermophilus relies on 8 R422 located on “fork loop 2” to close the
downstream bubble (Vassylyev et al., 2007; Vassylyev et al., 2007). 8 R422
aligns with G493 in human and G506 in yeast RNAP ll, showing that, if the
downstream bubble closes for human and yeast enzymes, it utilizes distinct (and
unknown) amino acid contacts. An open downstream bubble appears necessary

to support NTP-driven translocation, for which there is mounting evidence in the

195

human system (Gong et al., 2005; Xiong and Burton, 2007; Zhang and Burton,
2004; Zhang et al., 2005).

Mutagenic studies are incomplete for analysis of the role of the trigger
loop in catalysis. A recent paper describes the effects of a large number of critical
mutations in the E. coli trigger loop that affect pause site recognition without
clearly defining their activities in elongation (Toulokhonov et al., 2007). A radical
deletion of the trigger loop and also a four amino acid substitution support
elongation at very slow rates, showing that the trigger loop has a role in catalysis,
but it is dispensable. To date, it has not been reported whether trigger loop
residues predicted to contact the i+1 active site NTP (i.e. E. coli 8’ M932, R933,
and H936) have an effect either on catalysis or transcriptional fidelity, as
predicted if the closed trigger loop normally accompanies catalysis. This
omission was unexpected because elongation is necessary to obtain data on
pausing, so data on elongation rates are available, but not reported. [3’ T934M,
from this critical region, supports wild type elongation rates, but has defects in
pause and terminator recognition (Weilbaecher et al., 1994). So far, published
data, therefore, supports a role of the trigger loop in transcriptional pausing, but
not a direct role in catalysis. At this time, therefore, a complete picture of trigger
loop function is not available. We would suggest that the trigger loop may be in a
closed conformation during NTP substrate loading through the main enzyme
channel. Unless a mis—pair is loaded, trigger loop contacts to the substrate NTP

do not become very important for bond formation.

196

Based on functional dynamic studies from the Burton laboratory, the
trigger loop hypothesis presents some difficulties. For human RNAP II, it appears
that an open trigger loop conformation may be consistent with phosphodiester
bond synthesis (the forward reaction) and endogenous pyrophosphorolysis (the
reverse reaction) (Gong et al., 2005; Xiong and Burton, 2007). From overlaying
structures, it does not appear that or-amanitin can bind to a structure with a
closed trigger loop conformation. In experiments in which a-amanitin and NTP
substrates are added to the reaction simultaneously, however, events unfold that
are most consistent with a-amanitin binding to active transcription complexes that
first form a bond and then may reverse the newly formed bond.

In “isomerization reversal” experiments, a-amanitin and NTP
substrates are added to RNAP ll elongation complexes simultaneously. RNAP Il
forms a single bond and then appears to be blocked during the following
translocation step (Gong et al., 2005; Xiong and Burton, 2007). Very little second
bond is formed, which was interpreted to mean that a-amanitin effectively blocks
translocation. The presence of a-amanitin does not appear to strongly inhibit
NTP loading, isomerization (NTP tightening), or bond synthesis, although all
subsequent events are strongly affected, including bond completion, which
requires pyrophosphate release. If isomerization requires trigger loop closing,
then a-amanitin could not bind to the catalytic complex, because the closed
trigger loop would block a-amanitin binding. If or-amanitin is excluded from
binding, however, how does or-amanitin interfere with bond completion

(pyrophosphate release)? Opening of the trigger loop after chemistry would be

197

expected to induce pyrophosphate release, because opening of the trigger loop
would disrupt the enclosure of the active site. Furthermore, elongation complexes
reverse isomerization and phosphodiester bond formation in response to the or-
amanitin block (Gong et al., 2005; Xiong and Burton, 2007). If a-amanitin can
only bind to an elongation complex with an open trigger loop structure, such a
structure appears to support endogenous pyrophosphorolysis, which is reverse
chemistry. TFIIS-mediated dinucleotide cleavage must proceed in the presence
of an open trigger loop conformation, because TFIIS and a closed trigger loop
would clash (Kettenberger et al., 2003; Kettenberger et al., 2004; Wang et al.,
2006). This is another example of RNA polymerase II active site catalytic function
in the presence of an open trigger loop. The other example is the trigger loop
deletion, which is active in elongation. At this time, it is difficult to fully reconcile
rapid chemical quench flow data obtained with human RNAP II with the
mechanism most strongly indicated by T. thermophilus elongation complex
structures. At a minimum, isomerization reversal experiments with human RNAP
ll appear to place constraints on the trigger loop hypothesis that will require
further investigation to resolve.

Based on rapid chemical quench flow studies of E. coli RNAP and
human RNAP ll, done in the absence of inhibitors, further constraints are placed
on the trigger loop hypothesis. For instance, isomerization must be very rapid,
because NTP-Mgz+ becomes resistant to EDTA chelation faster than can be
accurately measured using rapid chemical quench flow (<0.002 s; rate constant

>1000 s“) (Nedialkov et al., 2003; Zhang and Burton, 2004; Zhang et al., 2003;

198

Zhang et al., 2005). If isomerization is trigger loop closing, this step must be very
rapid and essentially irreversible, because EDTA quench curves (isomerization)
remain above HCI quench curves (chemistry) (Xiong and Burton, 2007). If
isomerization were readily reversible, EDTA and HCI quench curves must
coincide, because stable isomerization cannot be maintained until chemistry is
complete. For an example, compare studies of elongation catalyzed by RNA-
dependent RNAP from poliovirus, done in the presence of Mg2+ or Mn2+ as the
metal co-factor. In the presence of Mg2+ EDTA quench and HCI quench curves
are indistinguishable (isomerization is readily reversible). In the presence of
an“, the EDTA quench curve rises above the HCI quench curve (NTP-Mg2+
isomerization is stable prior to chemistry) (Arnold and Cameron, 2004; Arnold et
al., 2004; Gohara et al., 2004).

If isomerization is rapid and essentially irreversible and represents
trigger loop closing, then additional rate-limiting steps detected during elongation
require explanation (Zhang and Burton, 2004). There are two rate-limiting steps
identified in each bond synthesis by human RNA polymerase II: 1)
phosphodiester bond synthesis (or an associated conformational change); and 2)
the interval that includes translocation, pyrophosphate release, and stable NTP-
Mg2+ loading. But NTP-Mg2+ loading cannot be simultaneously slow and
reversible and fast and irreversible, and it is clear from EDTA quench
experiments that NTP-Mg2+ loading and isomerization can be fast and essentially
irreversible (Nedialkov et al., 2003; Zhang and Burton, 2004; Zhang et al., 2003;

Zhang et al., 2005). Translocation coupled to pyrophosphate release, therefore,

199

appears to be a rate-limiting step in human RNA synthesis. Furthermore, this
step is highly NTP concentration dependent, indicating that human RNAP Il
utilizes an NTP-dependent translocation mechanism. The T. thermophilus RNAP
structure, by contrast, has been interpreted to indicate a simple thermal ratchet
mechanism with NTP loading only through the secondary pore (Vassylyev et al.,
2007). These conclusions are not consistent with the functional dynamics of
elongation catalyzed by human RNAP II.

In a thermal ratchet mechanism, every elongation complex passes
through a pre-translocated intermediate in each bond addition cycle (EC.PPi
(--) EC(pre)+PPi). By contrast, in an NTP-driven translocation mechanism,
product complex (EC.PPin) must load the next templated NTP substrate
(EC.NTPn+1.PPin) in order to advance (EC.NTPn+1.PPin -)EC.NTPn+1+ PPi).
Because pyrophosphate release and translocation are coupled in the NTP-driven
translocation mechanism, the pre-translocated elongation complex (EC(pre))
without bound pyrophosphate (EC.PPi) does not form, unless the elongation
complex is delayed (i.e. by NTP limitation).

Exogenous pyrophosphate, therefore, is a probe to discriminate the
NTP-driven translocation mechanism from the thermal ratchet mechanism.
Exogenous pyrophosphate binds to the pre-translocated elongation complex but
not the post-translocated elongation complex. In a thermal ratchet mechanism,
NTP substrates bind only to the post-translocated elongation complex. Therefore,
in a thermal ratchet mechanism, exogenous pyrophosphate should halt or delay

elongation at each bond formation, essentially independent of incoming NTP

200

concentration. In the NTP-driven translocation mechanism, by contrast, the
elongation complex is protected from inhibition by exogenous pyrophosphate
through formation of multiple bonds. Addition of exogenous pyrophosphate to
elongation reactions, therefore, appears to discriminate between different models
for elongation. Because an RNAP can elongate by a mixed mechanism, i.e. by
NTP-driven translocation at high NTP and by a thermal ratchet at limiting NTP,
pyrophosphate may permit robust elongation at high NTP concentrations and
inhibit elongation at low NTP concentrations.

Using or-amanitin as a transient translocation blocker, we found that
NTPs interact accurately with templated downstream DNA sites (i+2, i+3, and
i+4) and result in “isomerization reversal”: release of the i+1 NTP-M92+ substrate
from the active site in response to the presence of the translocation block and
accurately templated downstream NTPs (Gong et al., 2005; Xiong and Burton,
2007). In the absence of downstream templated NTPs, the i+1 NTP substrate
remains largely committed to engage in chemistry. This experiment appears to
be a simple proof of the NTP-driven translocation model, and, for human RNAP
ll, appears to falsify the secondary pore NTP loading hypothesis: that NTPs can
only enter the active site to the post-translocated elongation complex through the
secondary pore. For human RNAP Il, NTPs appear to load primarily through the
main enzyme channel to downstream DNA sites.

We have initially concentrated effort on the DNA template 5’-40-
CAAAGCCTTT-49-3’ (non-template strand), observing the influence of CTP and

UTP on incorporation of GMP at the G44 position, utilizing both EDTA and HCI

201

quenching (Xiong and Burton, 2007). Both CTP (i+2 and i+3) and UTP (i+4, i+5,
i+6) appear to stimulate isomerization reversal at G44 (Xiong and Burton, 2007).
Effects of UTP on isomerization reversal are completely dependent on the
presence of CTP, as expected from the NTP driven translocation model. We
were surprised to observe UTP effects on synthesis of G44, because this
indicates templated NTP binding at least to the i+4 downstream DNA position
(Xiong and Burton, 2007). This observation will be confirmed using
pyrophosphate block experiments, described above.

Next, we will do a similar experiment with a template such as 40-
CAAAGCT'IT-48 (non-template strand), to specifically discriminate effects of
CTP (i+2) and UTP (i+3, i+4, i+5) on GMP incorporation at G44 (GTP is
templated at i+1). The advantage of substituting this template is that i+2 and i+3
NTPs can be varied independently. We wish to determine the concentration
dependence of CTP and UTP required to stimulate isomerization reversal at
G44. In principle, this experiment provides information about the dissociation
constant for templated NTP binding at downstream sites. We wish to determine
what CTP and UTP analogues can substitute for the natural bases at the i+2 and
i+3 positions. Using the 40-CAAAGCC'lTl'-49 template, only CTP and UTP
influenced incorporation of GMP at G44 (Xiong and Burton, 2007). No analogue,
however similar, would suffice to replace either CTP (templated at i+2 and i+3) or
UTP (templated at i+4, i+5, and i+6) (Xiong and Burton, 2007).

In summary, the Burton laboratory has developed the most potent

approaches available to elucidate the functional dynamics of human RNAP II.

202

Our studies have revealed details of the internal mechanism of RNA synthesis
catalyzed by human RNAP ll. These studies do not appear to be consistent with
the idea that human RNAP II must function identically to T. thermophilus RNAP,
which has been inferred from structural studies to utilize a simple secondary pore
NTP loading, trigger loop opening and closing, thermal ratchet mechanism for
elongation. Based on functional dynamic studies, however, human RNAP lI might
not utilize this mechanism. The Burton laboratory has developed pyrophosphate
as a potent probe of the RNAP II mechanism. This approach demonstrates the
NTP-driven translocation mechanism for human RNAP II and will be used in the

future to further define the mechanism of elongation.

203

References

Abbondanzieri, E. A., Greenleaf, W. J., Shaevitz, J. W., Landick, R., and Block,
S. M. (2005). Direct observation of base-pair stepping by RNA polymerase.
Nature 438, 460-465. _

Adelman, K., La Porta, A., Santangelo, T. J., Lis, J. T., Roberts, J. W., and
Wang, M. D. (2002). Single molecule analysis of RNA polymerase elongation
reveals uniform kinetic behavior. Proc Natl Acad Sci U S A 99, 13538-13543.

Arnold, J. J., and Cameron, C. E. (2004). Poliovirus RNA-dependent RNA
polymerase (3Dpol): pre-steady-state kinetic analysis of ribonucleotide
incorporation in the presence of M92+. Biochemistry 43, 5126-5137.

Arnold, J. J., Gohara, D. W., and Cameron, C. E. (2004). Poliovirus RNA-
dependent RNA polymerase (3Dpol): pre-steady-state kinetic analysis of
ribonucleotide incorporation in the presence of Mn2+. Biochemistry 43, 5138-
5148.

Burton, Z. F., Feig, M., Gong, X. 0., Zhang, C., Nedialkov, Y. A., and Xiong, Y.
(2005). NTP-driven translocation and regulation of downstream template opening
by multi-subunit RNA polymerases. Biochem Cell Biol 83, 486-496.

Dalal, R. V., Larson, M. H., Neuman, K. C., Gelles, J., Landick, R., and Block, S.
M. (2006). Pulling on the nascent RNA during transcription does not alter kinetics
of elongation or ubiquitous pausing. Mol Cell 23, 231-239.

Davenport, R. J., Wuite, G. J., Landick, R., and Bustamante, C. (2000). Single-
molecule study of transcriptional pausing and arrest by E. coli RNA polymerase.
Science 287, 2497-2500.

Gnatt, A. L., Cramer, P., Fu, J., Bushnell, D. A., and Kornberg, R. D. (2001).
Structural basis of transcription: an RNA polymerase II elongation complex at 3.3
A resolution. Science 292, 1876-1882.

Gohara, D. W., Arnold, J. J., and Cameron, C. E. (2004). Poliovirus RNA-
dependent RNA polymerase (3Dpol): kinetic, thermodynamic, and structural
analysis of ribonucleotide selection. Biochemistry 43, 5149-5158.

Gong, X. 0., Zhang, C., Feig, M., and Burton, Z. F. (2005). Dynamic error
correction and regulation of downstream bubble opening by human RNA
polymerase II. Mol Cell 18, 461-470.

204

Greenleaf, W. J., and Block, S. M. (2006). Single-molecule, motion-based DNA
sequencing using RNA polymerase. Science 313, 801.

Herbert, K. M., La Porta, A., Wong, B. J., Mooney, R. A., Neuman, K. C.,
Landick, R., and Block, S. M. (2006). Sequence-resolved detection of pausing by
single RNA polymerase molecules. Cell 125, 1083-1094.

Kettenberger, H., Armache, K. J., and Cramer, P. (2003). Architecture of the
RNA polymerase ll-TFIIS complex and implications for mRNA cleavage. Cell
114, 347-357.

Kettenberger, H., Armache, K. J., and Cramer, P. (2004). Complete RNA
polymerase II elongation complex structure and its interactions with NTP and
TFIIS. Mol Cell 16, 955-965.

Nedialkov, Y. A., Gong, X. 0., Hovde, S. L., Yamaguchi, Y., Handa, H., Geiger,
J. H., Yan, H., and Burton, Z. F. (2003). NTP-driven translocation by human RNA
polymerase II. J Biol Chem 278, 18303-18312.

Shaevitz, J. W., Abbondanzieri, E. A., Landick, R., and Block, S. M. (2003).
Backtracking by single RNA polymerase molecules observed at near-base-pair
resolution. Nature 426, 684-687.

Toulokhonov, l., Zhang, J., Palangat, M., and Landick, R. (2007). A central role of
the RNA polymerase trigger loop in active-site rearrangement during
transcriptional pausing. Mol Cell 27, 406-419.

Trinh, V., Langelier, M. F., Archambault, J., and Coulombe, B. (2006). Structural
perspective on mutations affecting the function of multisubunit RNA polymerases.
Microbiol Mol Biol Rev 70, 12-36.

Vassylyev, D. G., and Artsimovitch, l. (2005). Tracking RNA polymerase, one
step at a time. Cell 123, 977-979.

Vassylyev, D. G., Vassylyeva, M. N., Perederina, A., Tahirov, T. H., and
Artsimovitch, l. (2007). Structural basis for transcription elongation by bacterial
RNA polymerase. Nature 448, 157-162.

Vassylyev, D. G., Vassylyeva, M. M, Zhang, J., Palangat, M., Artsimovitch, l.,
and Landick, R. (2007). Structural basis for substrate loading in bacterial RNA
polymerase. Nature 448, 163-168.

Wang, D., Bushnell, D. A., Westover, K. D., Kaplan, C. D., and Kornberg, R. D.
(2006). Structural basis of transcription: role of the trigger loop in substrate
specificity and catalysis. Cell 127, 941-954.

Wang, H. Y., Elston, T., Mogilner, A., and Oster, G. (1998). Force generation in
RNA polymerase. Biophys J 74, 1186-1202.

205

Weilbaecher, R., Hebron, C., Feng, G., and Landick, R. (1994). Terrnination-
altering amino acid substitutions in the beta' subunit of Escherichia coli RNA
polymerase identify regions involved in RNA chain elongation. Genes Dev 8,
2913-2927.

Westover, K. D., Bushnell, D. A., and Kornberg, R. D. (2004). Structural basis of
transcription: nucleotide selection by rotation in the RNA polymerase II active
center. Cell 119, 481-489.

Westover, K. D., Bushnell, D. A., and Kornberg, R. D. (2004). Structural basis of
transcription: separation of RNA from DNA by RNA polymerase II. Science 303,
1014-1016.

Xiong, Y., and Burton, Z. F. (2007). A tunable ratchet driving human RNA
polymerase II translocation adjusted by accurately templated nucleoside
triphosphates loaded at downstream sites and by elongation factors. J Biol
Chem.

Yin, H., Wang, M. D., Svoboda, K., Landick, R., Block, S. M., and Gelles, J.
(1995). Transcription against an applied force. Science 270, 1653-1657.

Zhang, C., and Burton, Z. F. (2004). Transcription factors HF and IIS and
nucleoside triphosphate substrates as dynamic probes of the human RNA
polymerase II mechanism. J Mol Biol 342, 1085-1099.

Zhang, C., Yan, H., and Burton, Z. F. (2003). Combinatorial control of human
RNA polymerase II (RNAP ll) pausing and transcript cleavage by transcription
factor "F, hepatitis delta antigen, and stimulatory factor II. J Biol Chem 278,
50101-50111.

Zhang, C., Zobeck, K. L., and Burton, Z. F. (2005). Human RNA polymerase II
elongation in slow motion: role of the TFIIF RAP74 alpha1 helix in nucleoside
triphosphate-driven translocation. Mol Cell Biol 25, 3583-3595.

206

11311111111111leIlljlljllil