How RNA Unfolds and Refoldshome.uchicago.edu/~jvieregg/pubs/annrev.pdf · ANRV345-BI77-05 ARI 1 May...

ANRV345-BI77-05 ARI 1 May 2008 11:17

How RNA Unfoldsand RefoldsPan T.X. Li,1 Jeffrey Vieregg,2

and Ignacio Tinoco, Jr.31Department of Biological Sciences, University at Albany, State Universityof New York, Albany, New York 12222; email: [email protected] of Physics, 3Department of Chemistry, University of California,Berkeley, California 94720; email: [email protected], [email protected]

Annu. Rev. Biochem. 2008. 77:77–100

The Annual Review of Biochemistry is online atbiochem.annualreviews.org

This article’s doi:10.1146/annurev.biochem.77.061206.174353

Copyright c© 2008 by Annual Reviews.All rights reserved

0066-4154/08/0707-0077$20.00

Key Words

RNA folding kinetics, single molecule, force unfolding, FRET,helicase, translation

AbstractUnderstanding how RNA folds and what causes it to unfold has be-come more important as knowledge of the diverse functions of RNAhas increased. Here we review the contributions of single-moleculeexperiments to providing answers to questions such as: How muchenergy is required to unfold a secondary or tertiary structure? Howfast is the process? How do helicases unwind double helices? Arethe unwinding activities of RNA-dependent RNA polymerases andof ribosomes different from other helicases? We discuss the use ofoptical tweezers to monitor the unfolding activities of helicases, poly-merases, and ribosomes, and to apply force to unfold RNAs directly.We also review the applications of fluorescence and fluorescenceresonance energy transfer to measure RNA dynamics.

77

Click here for quick links to Annual Reviews content online, including:

• Other articles in this volume• Top cited articles• Top downloaded articles• Our comprehensive search

FurtherANNUALREVIEWS

Ann

u. R

ev. B

ioch

em. 2

008.

77:7

7-10

0. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by C

alif

orni

a In

stitu

te o

f T

echn

olog

y on

09/

26/1

2. F

or p

erso

nal u

se o

nly.

ANRV345-BI77-05 ARI 1 May 2008 11:17

Contents

INTRODUCTION. . . . . . . . . . . . . . . . . 78SECONDARY STRUCTURE

FOLDING . . . . . . . . . . . . . . . . . . . . . . 78TERTIARY STRUCTRURE

FOLDING . . . . . . . . . . . . . . . . . . . . . . 81SINGLE MOLECULE STUDIES

OF RNA ENZYMES . . . . . . . . . . . . 83SALT EFFECTS ON

MECHANICAL UNFOLDINGOF RNA. . . . . . . . . . . . . . . . . . . . . . . . . 85Advantages of Mechanical

Unfolding. . . . . . . . . . . . . . . . . . . . . 85Cation Binding in Mechanical

Unfolding. . . . . . . . . . . . . . . . . . . . . 86Salt Effects on Secondary

Structures . . . . . . . . . . . . . . . . . . . . . 86Salt Effects on Tertiary Structure . . 87

LIGAND AND PROTEINBINDING TO RNA . . . . . . . . . . . . . 88

ENZYMES THAT UNFOLDRNA . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89

HELICASES . . . . . . . . . . . . . . . . . . . . . . . 90RNA-DEPENDENT RNA

POLYMERASES . . . . . . . . . . . . . . . . 91RIBOSOME . . . . . . . . . . . . . . . . . . . . . . . . 92

Translation . . . . . . . . . . . . . . . . . . . . . . . 92Mechanism of Elongation. . . . . . . . . 92

SINGLE-MOLECULETRANSLATION . . . . . . . . . . . . . . . . 93

INTRODUCTION

In 1999 one of us wrote a review article en-titled “How RNA Folds” (1), which describesthe RNA folding problem and contrasts itwith the much more difficult protein fold-ing problem. Atomic force microscopy wasconsidered as a potential method to unfoldRNA in physiological conditions of temper-ature and solvent; neither high temperaturesnor denaturants would be needed. Since thenlaser (or optical) tweezers have emerged as thepreferred method to apply force to unfold sin-gle molecules of RNA, because the techniqueallows direct observation of their unfolding

and refolding. As an RNA molecule unfoldsfrom tertiary structures to secondary struc-tures to single strands (and refolds in reverseorder), force and end-to-end distance of theRNA are measured; changes in the extensionof the molecule indicate structural transitionsin the RNA. Combining the force and dis-tance measurements provides the work nec-essary to unfold the RNA, and the work ob-tained when the RNA refolds. If the processis reversible, the work done (or obtained) isthe Gibbs free energy change. Kinetics aredetermined from the time dependence of theprocesses. A molecule that exists in two con-formations can hop back and forth betweenthe two states at equilibrium; the mean life-time in each state and their distributions char-acterize the kinetics. Also, the force can bequickly jumped or dropped to a new value,and the lifetime of a conformation at thenew value measured. Furthermore, measure-ments of fluorescence resonance energy trans-fer (FRET) on single molecules have revealeddetails of the kinetics of RNA folding and con-formational changes.

In biological cells helicases exist to unwindspecific RNAs; other enzymes, such as RNA-dependent RNA polymerases (RdRps) and ri-bosomes, need to unwind the RNA beforethey can perform their biological functions—transcribe or translate the sequence.

We summarize recent progress in un-folding RNA by force and by enzymes. Wealso discuss the application of single-moleculeFRET to obtain novel kinetic and structuralinformation about RNA reactions. Finally, wespeculate about the contributions of single-molecule methods in the future.

SECONDARY STRUCTUREFOLDING

Experimental investigations of RNA sec-ondary structure have been constrained bythe stability of the folded state. With as fewas three base pairs sufficient for duplex sta-bility at room temperature, larger structuresrequire high temperatures and/or chemical

78 Li · Vieregg · Tinoco

Ann

u. R

ev. B

ioch

em. 2

008.

77:7

7-10

0. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by C

alif

orni

a In

stitu

te o

f T

echn

olog

y on

09/

26/1

2. F

or p

erso

nal u

se o

nly.

ANRV345-BI77-05 ARI 1 May 2008 11:17

denaturants to unfold. However, thermalmelting data of secondary structure have beeninterpreted by a nearest-neighbor model (2)and form the basis for widely used structureprediction algorithms that predict secondarystructure with reasonable accuracy (3–5). Us-ing optical tweezers, an RNA structure canbe unfolded into an extended single strandby mechanical force in physiological buffersand temperatures; structural transitions areindicated by changes in the extension of themolecule (6). Several RNA and DNA hairpins,derived from ribozyme, viral, ribosomal, andsiRNA sequences (6–11), have been studiedusing this approach. Correction for the effectof force on the unfolded single strand is re-quired, but this is measurable and can be alsomodeled (12). Recent investigations have alsoexamined single-stranded homopolymers (13,14). In addition, atomic force microscopy hasbeen employed to examine the dissociation ofRNA duplexes (15).

Force can be applied in several ways tostudy unfolding and refolding of single RNAmolecules (9) (Figure 1a). Increasing or de-creasing the force at a constant rate (forceramp) generates a force-extension (F-X) curve(analogous to the pressure-volume curve for agas) that characterizes the RNA’s mechanicalbehavior (Figure 1b). If the folding/unfoldingtransitions are assumed to be two-state tran-sitions, it is possible to measure the ratesfrom the resulting distribution of transitionforces (16). Integration of the F-X curve yieldsthe mechanical work done during the pro-cess; for a reversible process the mechanicalwork equals the Gibbs free energy change.Typically, the molecule does not unfold re-versibly, which results in hysteresis in theF-X curve. Recent developments in nonequi-librium statistical mechanics, the Jarzynski(17) and Crooks relations (18), enable recov-ery of the reversible work, and thus the zero-force free energy of folding, from nonequilib-rium experimental data (8, 19). Of these, theCrooks Fluctuation Theorem is particularlyuseful, as it is unbiased and converges morerapidly. Using this method, the effect (10 kcal

mol−1) of a single mutation on the foldingfree energy of the S15 three-helix junctionwas recovered (8). Theoretical investigations(20, 21) indicate that it should be possible torecover the complete free energy landscapealong the force axis using this technique, evenfor folding that occurs away from equilibrium.This provides valuable information for under-standing folding, but it remains to be seenwhether experimental limitations make thistechnique possible in practice.

To directly measure the kinetics of unfold-ing or refolding, it is convenient to hold theforce constant and measure the lifetimes ofthe folded or unfolded states (Figure 1c). Therates k(F ) can be described by the follow-ing equation, (T: temperature in Kelvin, k:Boltzmann’s constant):

k(F ) = k0eF X‡kT . 1.

The derivative of ln(k) versus force yieldsthe distance to the transition state X‡ forthe folding or unfolding reaction. The pre-exponential k0 is the apparent rate constantat zero force, but it cannot be compared withzero force kinetics measured by other meth-ods, because the mechanism of unfolding orrefolding will be different (6, 23, 24). Somehairpins unfold and refold rapidly near theirequilibrium force. If many unfolding and re-folding transitions are observed, the foldingfree energy landscape can be deconvolvedfrom the probability distribution function ofthe extension of the molecule at a constantforce, as demonstrated for several DNA hair-pins (10, 11). This technique requires botha rapidly hopping molecule and a very sta-ble instrument as many thousands of transi-tions must be observed to adequately samplerarely occupied areas of the potential. Moreoften RNA molecules, especially those con-taining tertiary structure, have slow kinetics attheir equilibrium forces. We have developeda force-jump technique (9) to rapidly changeforce between values such that folding and un-folding rates can be measured separately atdifferent forces (Figure 1d ).

www.annualreviews.org • How RNA Unfolds and Refolds 79

Ann

u. R

ev. B

ioch

em. 2

008.

77:7

7-10

0. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by C

alif

orni

a In

stitu

te o

f T

echn

olog

y on

09/

26/1

2. F

or p

erso

nal u

se o

nly.

ANRV345-BI77-05 ARI 1 May 2008 11:17

RNAHandle A

Handle B

Lasertrapa

c

12.4 pN

12.7 pN

13.0 pN

12.1 pN

11.9 pN

11.6 pN

Ext

ensi

on

Time5 s

0

2

4

6

8

10

12

14

16

18

20

Fo

rce

(pN

)

b

10 nm

Extension (nm)

Rev

ersi

ble

wor

k

20 n

m

0 5 10 15 20 25

Time (s)

0

50

100

150

200

Ex

ten

sio

n (

nm

)

d

18 nm

tunfolded

2 pN 2 pN

HP

SS SS

HP

Unfolding Refolding

tfolded

Micropipette

Figure 1Mechanical unfolding of a hairpin. (a) A single RNA molecule consisting of a hairpin flanked bydouble-stranded DNA/RNA handles is tethered between two micron-size beads. One bead is held by anoptical trap and the other is mounted on the tip of a micropipette. By moving the micropipette, the RNAis stretched and relaxed (22). (b) Force-extension curve of a pulling (blue) and relaxation (red ) cycle.Un/refolding of the hairpin is characterized by a “rip” on the curve displaying negative slope. When theprocess is reversible, the area under the rip equals the un/refolding free energy. (c) When force is heldconstant near the equilibrium force, the extension of the molecule switches between two valuescorresponding to the folded and unfolded states. The lifetimes of the two states are force dependent.(d ) Extension traces of a force-jump experiment. Unfolding and refolding kinetics of a hairpin can bemonitored at different forces. Observations of lifetimes of folded and unfolded structures are collected tocompute unfolding and refolding kinetics, respectively.

RNA can adopt alternative conformationswith similar stability, and misfolding is a com-mon phenomenon for RNA (25, 26). To reachthe functional fold, the RNA must discrimi-nate against nonnative conformations, someof which may be nearly as stable as the nativestate. The mechanism of this conformationalsearch remains unclear. To understand therugged folding energy landscape of RNA, it

is indispensable to characterize alternativestructures and their folding pathways. Byvarying the rate of force relaxation, theHIV-1 TAR sequence was steered to foldinto the native hairpin in either one step orthrough multiple intermediates (27). Whenthe force was relaxed quickly, the RNAmisfolded, forming conformations withlonger end-to-end extensions and decreased


Ann

u. R

ev. B

ioch

em. 2

008.

77:7

7-10

0. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by C

alif

orni

a In

stitu

te o

f T

echn

olog

y on

09/

26/1

2. F

or p

erso

nal u

se o

nly.

ANRV345-BI77-05 ARI 1 May 2008 11:17

stability relative to the native state. Whenforce was increased subsequently, the RNArefolded into the native fold after thenonnative interactions were disrupted.Single-molecule manipulations, in combina-tion with modeling of the folding pathwaysand intermediates (23, 24, 28–32), providenew opportunities to explore the energylandscape of RNA folding.

TERTIARY STRUCTRUREFOLDING

Tertiary interactions between distal domainsare responsible for forming the compactthree-dimensional structures required formany of RNA’s catalytic and regulatory func-tions. Unfortunately, the predictive thermo-dynamic models that exist for secondary struc-ture have not, to this point, been developedfor tertiary structure. Whether such a modelcan be developed is one of the pressing ques-tions for the RNA folding field. Many ter-tiary interactions are sufficiently weak thatthey can be disrupted using temperatures orsolutions not too far from physiological con-ditions. The combination of clear compari-son with function and greater accessibility hasresulted in a much larger number of single-molecule studies of tertiary folding comparedto secondary structure, especially by fluores-cence techniques.

The majority of single-molecule investi-gations of tertiary folding have utilized theFRET technique [reviewed by Ha (34)] tomonitor the distance between two dye-labelednucleotides. This allows observation of theformation and dissociation of specific tertiarymotifs in real time, a capability that comple-ments the data obtained by force techniques.In addition to distinguishing discrete states(folded versus unfolded versus intermediates),the FRET signal quantitatively measuresthe distance between the labeled nucleotides.The FRET signals, though complicated bythe need to account for variations in fluo-rophore characteristics and orientation, pro-vide valuable constraints for structural mod-

eling. Ha et al. (33) first demonstrated theutility of FRET for studying RNA foldingby measuring conformational changes in athree-helix junction upon binding of Mg2+

or of a ribosomal protein. Subsequent ex-periments can generally be divided into twotypes: those which explore the folding ofa specific RNA molecule (typically an en-zyme) in detail, and those which primarilyaim to investigate the folding dynamics of spe-cific tertiary motifs. In this section, we focuson the latter, postponing discussion of ri-bozymes and ligand-binding RNAs for latersections.

One important motif that has been studiedin this manner is the tetraloop-receptor inter-action, in which a 4-nt hairpin loop (GNRAmotif ) docks with an asymmetric internalloop elsewhere in the molecule. This motif ispresent in many large folded RNAs and hasbeen studied extensively by both structuraland biochemical means (35). Tetraloop-receptor interactions have also been used toconstruct nanoscale synthetic RNA “buildingblocks” (36). A single-molecule FRET study(37) measured kinetics and equilibrium fordocking of a GAAA tetraloop for an RNA inwhich the hairpin and receptor-containingduplex were linked by a short strand of polyA(Figure 2a). The docking rate increased12-fold as the concentration of Mg2+ wasincreased from 0 to10 mM (Figure 2b),consistent with observations in bulk studies.The undocking rate was found to decreaseby a factor of three over the same rangeof Mg2+ concentration. A follow-on studyexplored the effect of changing the linkersequence and/or length (38), concluding thatthe linker acted solely to increase the effectiveconcentration of the loop and receptor anddid not change the equilibrium constant. Inthese studies, considerable heterogeneity wasobserved in molecular behavior, both kinet-ically and in the equilibrium species present.Such heterogeneity would complicate datainterpretation in bulk studies.

When single-molecule measurements arecombined with traditional molecular biology


Ann

u. R

ev. B

ioch

em. 2

008.

77:7

7-10

0. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by C

alif

orni

a In

stitu

te o

f T

echn

olog

y on

09/

26/1

2. F

or p

erso

nal u

se o

nly.

ANRV345-BI77-05 ARI 1 May 2008 11:17

AAA

GAAAtetraloop

CGGCU

UGCUGUGC

3'a

G

UAA

CC

Biotin5'

Undocked

G

CGCACTCGTCTGA

3'

C

C

GCGUGAGCAGACU

C

3'

Tetraloopreceptor

CCGA

ACGACACG

GGU

UA

Cy5

G

AA

b60

45

30

15

00 2 4 6 8 10

[Mg2+] (mM)

k do

ck (

s–1)

C

CCGA

ACGACACG

C

GCGUGAGCAGACU

C

3'

GG

U

UA

CGGCU

G AA

A

3'

U

GAA

CC

CUGUGC

A

AA

AAAA

G

CGCACTCGTCTGA

5'

3'

Docked

Biotin

AAAA

Cy35'

GA

C

C

CCGG

G5'Cy5

kdock

5'

kundock

CC

GG

Cy3

G

5' UG

G

10

8

6

4

2

0

ku

nd

ock (s

–1)

Figure 2Single-molecule measurement of a tetraloop-receptor tertiary interaction.(a) Three strands of RNA were annealed to form a duplex that containsthe receptor (blue text) linked to a short hairpin containing a GAAAtetraloop (red ) by a flexible (A)7 tether. Fluorophores Cy3 and Cy5 wereplaced at the ends of the hairpin and duplex, resulting in fluorescenceresonance energy transfer (FRET) efficiencies of 0.68 and 0.22 for thedocked and undocked states, respectively. (b) Docking and undockingkinetics as a function of [Mg2+]. Adapted from Reference 37.

techniques such as mutagenesis, it is possibleto understand folding processes in even moredetail, as demonstrated in a study of the du-plex docking interaction that forms the coreof the Tetrahymena ribozyme (39). The P1

duplex is a five-base-pair helix which, whendocked, forms a minor groove triplex with twosingle-strand sections of the ribozyme core.Both the duplex and its docking target arepreformed. The docking dynamics were mea-sured by a FRET signal of fluorophores onthe duplex and the receptor. All 8 nts of P1that make tertiary contacts were successivelymutated and the docking and undocking rateswere measured. Each mutation caused a sig-nificant change in the equilibrium constantof the reaction. Surprisingly, the docking ratewas largely unaffected by addition of urea ormutations, suggesting that the transition statefor docking does not resemble the dockedstate. Together with bulk chemical protectiondata (40), these results suggest that the rate-limiting step of duplex docking may be es-cape from a kinetic trap, possibly a misfoldedstructure of the J8/7 docking target. Recentreseach finds that combining cyclic repeatedMg2+ jumps with single-molecule FRET re-veals slow degrees of freedom in folding ofthe catalytic domain of RNase P that were notapparent from typical salt-jump experiments(Xiaohui Qu, personal communication). Bothof these studies illustrate the power of com-bining single-molecule techniques with tradi-tional biochemical methods to reveal featuresinaccessible to either alone.

A kissing interaction is the base pairingformed by complementary sequences betweentwo hairpin loops (41). Using optical tweez-ers, we studied an intramolecular kissing com-plex formed by two hairpins linked by A30; thisminimal kissing complex contains only twoG·C base pairs (42). By increasing force, theRNA was unfolded into four different confor-mations in order: kissing complex, two linkedhairpins, one hairpin, and single strand. Asforce is decreased, the single strand was re-folded into the kissing complex in the re-verse order. Kinetics of steps in the unfoldingand refolding were measured. In contrast tothe hairpins the unfolding rate for the kiss-ing interaction was found to be relatively in-sensitive to force. The distance to the transi-tion state for breaking the kissing interaction


Ann

u. R

ev. B

ioch

em. 2

008.

77:7

7-10

0. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by C

alif

orni

a In

stitu

te o

f T

echn

olog

y on

09/

26/1

2. F

or p

erso

nal u

se o

nly.

ANRV345-BI77-05 ARI 1 May 2008 11:17

(X‡ = 0.7 nm) indicates that both kissing basepairs were broken at the transition state. Thekissing loop was also found to lock the twohairpins in place at forces significantly higherthan those required to unfold the hairpinsalone. This phenomenon is likely due to theorientation of the kissed hairpins relative tothe force axis (42).

Pseudoknots, also a common motif inRNA structure, are critical in programmedribosomal frameshifting (43), telomere struc-ture (44), and ribozyme activity (45). Fold-ing free energies of various pseudoknots havebeen measured in bulk (43). A substantial ef-fort is underway to codify their thermody-namics and kinetics in an analogous mannerto the nearest-neighbor model of secondarystructure (46–49). Experimentally, two groupshave used optical tweezers to investigate pseu-doknot folding at the single molecule level.Hansen et al. measured the force required tounfold two molecules patterned after the in-fectious bronchitis virus (IBV) frameshiftingpseudoknot (50). Independently, we also stud-ied the mechanical unfolding of the wild-typeIBV pseudoknot and two mutants as well asthe constituent hairpins (51). The folding ki-netics and free energies for these moleculeswere measured. Together, these measure-ments cover a variety of different stem lengthsand nucleotide compositions, which should behelpful in constructing a more accurate pic-ture of the energy landscape and folding path-way for this important tertiary interaction.

The primary challenge in understandingRNA tertiary structure is to uncover the gen-eral principles that govern its folding and dy-namics. However, a rigorous folding algo-rithm for prediction of tertiary structure isnot yet available (52). Single-molecule stud-ies, together with more ensemble measure-ments, can provide important constraints tobe utilized in computation and to serve astest cases for modeling. Moreover, most func-tions of RNAs, such as catalytic activities, de-pend on the dynamics of the molecules. Inproteins, molecular dynamics (MD) simula-tion has been extensively employed to explain

observation of single-molecule dynamics atthe atomic level (53). Similar work in RNAis lamentably rare. One exception is a recentpaper by Rhodes et al. (54), in which MD sim-ulation is used to model the water moleculestrapped in the catalytic core of a ribozyme.The simulation predicted changes in hydro-gen networking caused by specific mutations.A surprising linear correlation was found be-tween the loss of water-mediated hydrogenbonds and the reduction of docking free en-ergy relative to the wild-type enzyme. Collab-oration between experimentalists and model-ers at the design stage is particularly useful,as data and simulation can then inform eachother synergistically rather than after the fact.

SINGLE MOLECULE STUDIESOF RNA ENZYMES

Ribozymes have been a preferred target forsingle-molecule studies of RNA for nearly asmany years as the techniques have existed (55).FRET has been the preferred method, andfolding is typically studied by adding Mg2+

ions to molecules with preformed secondarystructure and observing the resultant confor-mational changes. Ribozyme folding, and sin-gle molecule studies thereof, were the subjectof two recent reviews (56, 57). In this review,therefore, we focus on several recent resultsthat may point the way toward future progressin the field.

The first generation of single-molecule ri-bozyme experiments mostly focused on thefolding of the enzyme from secondary to ter-tiary structure. Although interesting as a studyof RNA folding, these results do not directlybear on the main function of ribozymes, whichis catalysis. Working with inactive enzymesavoids the problem of losing one’s sample viaself-cleavage, but makes it hard to determinethe relationship between observed conforma-tional dynamics and the catalytic cycle. Nahaset al. (58) were able to overcome this diffi-culty for the self-cleaving hairpin ribozymeby adding a single-strand RNA that resembledthe cleavage product, but was extended with


Ann

u. R

ev. B

ioch

em. 2

008.

77:7

7-10

0. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by C

alif

orni

a In

stitu

te o

f T

echn

olog

y on

09/

26/1

2. F

or p

erso

nal u

se o

nly.

ANRV345-BI77-05 ARI 1 May 2008 11:17

a

b

c

FR

ET

0

0.5

1

Cleavage

Ligation

0 20 40 60 80 100

UL

(undocked, ligated)DL

(docked, ligated)

FRET 0.2–0.3 0.8 0.8 0.2–0.3 0.2–0.3

DC

(docked, cleaved)PR

(3'P released)UC

(undocked, cleaved)

kL dock

kL undock k lig

k cleav

+

kC undock

kC dock k on

k off

>

>

Time (s)

Figure 3Observing active hairpin ribozymes with single molecule fluorescence resonance energy transfer(FRET). (a) Extended product strand ( purple) remains bound after cleavage (site marked with ˆ ),enabling a multiple turnover assay. (b) FRET trace of a single ribozyme undergoing multiple rounds ofcleavage and ligation, as described in (a). The cleaved enzyme ( purple bars) switches rapidly between twoFRET values. The lower value represents the undocked state and the higher the docked state. Theuncleaved RNA has a stable FRET efficiency of about 0.7. (c) Reaction mechanism of a variant hairpinribozyme. States with similar FRET values can be distinguished by changing Mg2+ and productconcentrations. Panels a and b adapted from (58) and c from (59).

nucleotides complementary to the ribozyme.Once the ribozyme cleaved itself, the newstrand remained bound to the ribozyme andwas religated by the reverse reaction. This ap-proach, which can be applicable to other en-zymes, enabled them to observe multiple cy-cles of cleavage and ligation (Figure 3a,b).The researchers were able to correlate thedocking/undocking with catalytic rates of theenzyme in a pH-dependent manner. Recently,Zhuang and coworkers reported an alternatescheme to dissect the reaction pathway of avariant of the hairpin ribozyme (59). A series

of Mg2+ pulse-chase experiments were per-formed to identify kinetic “fingerprints” of thevarious enzymatic states (undocked, docked,docked and cleaved, cleaved and undocked,and product released, Figure 3c) in a singleenzymatic cycle. Different states with similarFRET signals, such as cleaved and uncleavedRNA at docked conformations, could be dis-tinguished, revealing a detailed kinetic mech-anism of the hairpin ribozyme.

Among the exciting discoveries in re-cent years is the expansion of the types ofchemistry that RNA enzymes are capable of


Ann

u. R

ev. B

ioch

em. 2

008.

77:7

7-10

0. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by C

alif

orni

a In

stitu

te o

f T

echn

olog

y on

09/

26/1

2. F

or p

erso

nal u

se o

nly.

ANRV345-BI77-05 ARI 1 May 2008 11:17

performing. In addition to the naturallyoccurring phosphodiester cleavage/ligationreaction, in vitro selection approaches haverevealed ribozymes that catalyze a diverse ar-ray of chemical processes, including oxida-tion/reduction, nucleotide synthesis, and pep-tide bond formation (60). This has addedsupport for the “RNA World” hypothesis ofbiochemical origins (61). Although a fruitfultool for discovering new activities, in vitro se-lection is largely silent regarding the mech-anism of the enzymes that are discovered.Single-molecule techniques, with their capa-bility to resolve rare and short-lived inter-mediates, should be very valuable in deter-mining how these new ribozymes work. Arecent example of this is a study of the Diels-Alder ribozyme carried out by Kobitski et al.(62). Working with a truncated, but func-tional, form of the enzyme, they were able toobserve that the enzyme undergoes intercon-version between two states at equilibrium at arate of ∼20 s−1. This conformational switch-ing appears to resolve a conflict between thesolved crystal structure and bulk chemical as-say results: The crystal structures indicatedthat the product was trapped in the activesite. Future studies on this and other novelribozymes should further increase our knowl-edge of the various chemistries that RNA canparticipate in, particularly if combined withfunctional assays and MD simulation.

SALT EFFECTS ONMECHANICAL UNFOLDINGOF RNA

RNA is a polyelectrolyte with one negativecharge per phosphate. Consequently, RNAdepends critically on ionic conditions for itsstructure, stability, reactivity, and ability tobind to ligands and proteins. Like other poly-electrolytes, RNA attracts cations to forma conterion atmosphere around it. Theseclosely associated cations show similar self-diffusion coefficients to the much larger poly-mer (63–65). However, RNA cannot be sim-ply treated as a cylindrical polyelectrolyte with

infinite length (66–71). RNAs have a wide va-riety of sizes and shapes; in addition to dif-fuse metal ion binding, RNA tertiary structureoften contains specific binding sites for diva-lent metal ions like Mg2+. Both diffuse andspecific cation binding are critical for stabil-ity of RNA secondary and tertiary structures(72, 73), with tertiary structure particularlydependent on specific binding. Definitions ofdiffuse and specific binding are somewhat ar-bitrary; there is no clear boundary to distin-guish the two types of binding. Cation bind-ing to RNA is relatively weak; for instance,the Kd of specific Mg2+ binding is often in themillimolar range.

Salt effects on nucleic acids have beenextensively studied and reviewed elsewhere(66–70, 74). Here, we discuss how mechanicalunfolding techniques can be employed toprobe some perennial problems in under-standing salt effects on RNA structure andfolding.

Advantages of Mechanical Unfolding

Mechanical unfolding offers several advan-tages over bulk methods in studying ion bind-ing to RNA. First, force applied by opticaltweezers (<150 pN) changes only the non-covalent interactions in the RNA. Unfold-ing, especially of secondary structure, can bestudied at physiological temperatures regard-less of ionic conditions. This is a big advan-tage over thermal denaturation, during whichRNA degrades at high temperature, especiallyin the presence of Mg2+. Second, thermody-namic interpretation of mechanical unfoldingresults is straightforward because locally ap-plied force changes RNA structure, but notactivities of ions and water molecules. More-over, RNA is so dilute in single-molecule ex-periments that the reaction chamber is almosta solution of salt, thereby avoiding compli-cated treatment of colligative properties ofconcentrated RNA solutions (63). Third, de-velopment of force manipulation techniqueprovides precise control of RNA conforma-tions, allowing direct evaluation of disruption


Ann

u. R

ev. B

ioch

em. 2

008.

77:7

7-10

0. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by C

alif

orni

a In

stitu

te o

f T

echn

olog

y on

09/

26/1

2. F

or p

erso

nal u

se o

nly.

ANRV345-BI77-05 ARI 1 May 2008 11:17

or formation of a particular interaction (42).Hence, ionic effects on folding secondary andtertiary structures can be disentangled.

Cation Binding in MechanicalUnfolding

How is mechanical unfolding of RNA affectedby force and ionic conditions? Here we de-scribe a simple thermodynamic scheme for anunfolding reaction,

Folded + n1 M ↔ Unfolded + n2 M,

in which M refers to a metal ion; and n1 andn2 are numbers of metal ions associated withfolded and unfolded conformations. The ef-fect of salt on unfolding can be written as

∂ ln Keq

∂ ln [M ]= �n, 2.

in which Keq is the equilibrium constant (ratioof unfolded to folded conformation), �n =n1 − n2 is the net change in metal ions inthe unfolding, and [M ] is the total concentra-tion of metal ions. We can describe the effectof metal ions on the mechanical unfolding ofRNA as

ln Keq,F = F�X/kB T + ln K0 + �n ln[M],3.

in which Keq,F is the equilibrium constant atforce F, and K0 is the apparent equilibriumconstant extrapolated to zero force and zerosalt. Equation 3 can be directly applied toRNA folding in a monovalent cation solution.However, RNA folding, particularly that oftertiary structure, is often studied by varyingthe concentration of Mg2+ ions while holdingthe monovalent concentration at a constantvalue of typically several hundred millimolar.In these conditions, the monovalent salt canbe treated as part of the solvent; and the �nln[M] term can be used to describe the effectof Mg2+. We note that �X (and X‡) dependson the force (75); when free energy and kinet-ics are extrapolated to forces far away from thetransition forces, �X can vary significantly. Itis possible that �n also varies with force.

Salt not only affects the folded and un-folded states, but also the height of the ki-netic barrier. Salt effects on thermodynam-ics and kinetics are usually parameterized by�Gionic and �G‡

ionic, respectively. For a two-state unfolding reaction with a single transi-tion state (Figure 4a), cations raise the heightof the kinetic barrier but do not change theposition of the transition state along the re-action coordinate for unfolding. On the lnk versus force plot, the slope of the curve(X ‡/kBT ) is constant but the y-intercept de-creases as cation concentration increases. Thefirst-order un/refolding kinetics can be writ-ten as

ln k = F X ‡/kB T + ln k0 + �n‡ ln[M], 4.

in which k is the rate constant, k0 is rate con-stant extrapolated to zero force and zero [M],and �n‡ reflects change in the number ofcations bound to RNA at the transition state.

It is tempting to interpret �n and �n‡ asthe effective numbers of salt bridges formed,or effective numbers of counterions released,upon RNA structural transitions or ligandbinding. This simplified view does not takeinto account the effect of counterion con-densation and shapes of RNA molecules, butmore thermodynamically rigorous treatmentsfor mono- and divalent metal ion bindingsfor RNA have been developed (76–79). How-ever, �n and �n‡ can be conveniently used todescribe the ionic effect on the un/refoldingkinetics.

Salt Effects on Secondary Structures

The effect of Mg2+ on the stability of sev-eral RNA hairpins and a three-helix junction(6, 8) was studied. In general, RNA hairpinsunfold and refold at higher forces as salt con-centration increases; and Mg2+ shows a muchstronger effect than monovalent cations. Anexample of such an effect on a TAR hair-pin (P.T.X. Li, unpublished data) is shown inFigure 4b. The critical force, F1/2, at whichunfolding and refolding rates are equal, in-creased by ∼6.5 pN upon addition of 10 mM


Ann

u. R

ev. B

ioch

em. 2

008.

77:7

7-10

0. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by C

alif

orni

a In

stitu

te o

f T

echn

olog

y on

09/

26/1

2. F

or p

erso

nal u

se o

nly.

ANRV345-BI77-05 ARI 1 May 2008 11:17

Mg2+, whereas �X of unfolding the hair-pin changed less than 2 nm. We have variedconcentrations of monovalent cations andconcentration of Mg2+ up to 30 mM in thepresence of 100 mM KCl. The reversible me-chanical work required to unfold the hair-pin and �G(0 pN)22◦C appears to increase lin-early with the logarithm of concentration ofcations. This linearity is consistent with bulkobservations (80).

The stabilizing effect of metal ions onRNA hairpins results from both a decreasein the unfolding rate and an increase in thefolding rate. X ‡ of unfolding and refoldingremained largely unchanged by type and con-centration of cations, and �n‡ appeared tobe constant for each type of cation. Hence,Equations 3 and 4 are suitable for descriptionof salt-dependent thermodynamics and kinet-ics of folding secondary structure.

Interestingly, metal ions have a strongereffect on unfolding than refolding (6, 8). Inthe F-X curves, hysteresis between the unfold-ing and refolding trajectories increases withsalt concentration. This effect is also shownby reduced overlap between the unfoldingand refolding force distributions at high saltconditions (Figure 4b). The weaker salt de-pendence of refolding kinetics probably re-sults from its reaction mechanism. The rate-limiting step of hairpin folding is formation ofthe loop-closing base pair, which must com-pete with alternative combinations of basepairs (80, 81). Cations facilitate this processby offsetting the negative charges on the phos-phates but are unlikely to discriminate againstnonnative base pairs.

Cations, particularly monovalent cations,bind diffusely to a simple hairpin. But Mg2+

displays specificity to bulges and internalloops and affects dynamics of their neighbor-ing domains (82, 83). These unpaired regionsin the RNA are also interesting because manyof them bind small molecule ligands and canbe used as drug targets, a process that requiresdisplacement of counterions (84, 85). Salt ef-fects on mechanical unfolding of these struc-tures remain unexplored.

‡

Occ

urr

ence

a

b Refolding Unfolding

0

20

40

60

0

20

40

8 10 12 14 16 18 20

Force (pN)

no M+

M+

F

U

100 mM K+

100 mM K+

plus10 mM Mg2+

Figure 4Salt effects on force unfolding of RNA structure. (a) Effect of a cation, M+,on a simple two-state unfolding reaction. The cation stabilizes the foldedstructure, effectively raising the kinetic barrier for unfolding. Salt effects onthe unfolded state are less than that on the folded state. The position of thetransition state along the reaction coordinate is not affected by the cation.(b) Force distributions of TAR RNA in 100 mM KCl (top) and with anadditional 10 mM Mg2+ (bottom) (P.T.X. Li, unpublished data).

Salt Effects on Tertiary Structure

Folding and stability of tertiary structure arecritically dependent on ionic conditions, es-pecially Mg2+. Consequently, metal ions havemore pronounced effects on tertiary struc-ture than on secondary structure. Kineticsof breaking tertiary interactions are greatly


Ann

u. R

ev. B

ioch

em. 2

008.

77:7

7-10

0. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by C

alif

orni

a In

stitu

te o

f T

echn

olog

y on

09/

26/1

2. F

or p

erso

nal u

se o

nly.

ANRV345-BI77-05 ARI 1 May 2008 11:17

slowed by Mg2+, but folding rates of tertiarystructure are only moderately dependent onmetal ions.

Intron ribozyme. The first hint of Mg2+ ef-fects on mechanical unfolding of RNA tertiarystructure came from a study by Onoa and col-leagues (86). Mechanical unfolding of the L-21 ribozyme in 10 mM MgCl2 was character-ized by distinct rips, each of which representsunfolding of a structural domain. In the ab-sence of Mg2+, the RNA populated collapsedconformations as observed in bulk studies(87), and its unfolding trajectories showed noclear rips. As the concentration of Mg2+ wasincreased, unfolding rips gradually appearedat forces significantly higher than those re-quired to disrupt the collapsed forms, indi-cating differential Mg2+ dependence of indi-vidual interactions and domains (supportingmaterials in Reference 86).

Pseudoknots. Pseudoknots adopt compactstructures (43), and their stability dependson bound Mg2+ (73, 88). In one of the re-cent mechanical studies on pseudoknots (51),Mg2+ not only raised the rip force but alsodecreased Xunfolding

‡, as evidenced by changesin the force-dependent unfolding rates.

Loop-loop interactions. Using force ma-nipulation, formation and disruption of an in-tramolecular kissing complex can be directlyobserved and distinguished from un/refoldingof secondary structures (42). The salt de-pendence of kissing interactions depends onthe sequence of kissing base pairs. A well-studied case is two variants of the DIS kiss-ing complexes derived from HIV-1, in whichkissing sequences are GUGCAC (Mal) andGCGCGC (Lai) (41, 89, 90). Mg2+ has astrong stabilizing effect on both kissing com-plexes. Addition of 1 mM Mg2+ raised themean rip force to break the kissing interac-tion by over 20 pN (P.T.X. Li, unpublisheddata). Yet, unfolding of the two kissing com-plexes displayed different force and salt de-pendences, as evidenced by different values of

Xunfolding‡ and �nunfolding

‡. Kinetics of breakingthe kissing interactions can be well describedby Equation 4.

Ongoing efforts to understand salt effectson RNA tertiary interactions by force un-folding can yield key thermodynamic and ki-netic parameters. These new parameters needto be explained structurally. For instance, is�nMg2+ ‡ the number of Mg2+ released at theunfolding transition state? Are these Mg2+

ions bound to specific sites or diffusely associ-ated with the RNA? Because unfolding kinet-ics directly reflect the folded RNA complexedwith cations, unfolding kinetics may be moreuseful than the folding free energy in inter-pretation of cation binding.

LIGAND AND PROTEINBINDING TO RNA

A DNA or RNA molecule can be mechan-ically unfolded in the absence and presenceof ligands and proteins; differences in stabil-ity reflect the binding of ligands. A DNA he-lix can be mechanically perturbed either byshearing (pulling on opposite ends of the he-lix) or by unzipping (pulling the two strandsapart from the same end of the helix). In theshearing mode, ligand and protein binding areindicated by global changes in the molecule’sF-X curve (91); the binding affinity at bothzero force and high forces can be quantified(92). For the unzipping experiments, displace-ment of specifically bound proteins results ina clear rip (abrupt change in extension) whenthe unfolding fork displaces a bound protein(93–95). This approach can be used to mapthe protein binding site along the DNA du-plex. Applying this strategy to RNA, we foundthat argininamide, but not arginine, binds andstabilizes the TAR hairpin, confirming thespecificity of argininamide for this RNA (96).We expect to see more mechanical studies onligand- and protein-RNA interactions in thenear future.

Single-molecule fluorescence tech-niques have demonstrated great potentialin studying assembly and dynamics of large


Ann

u. R

ev. B

ioch

em. 2

008.

77:7

7-10

0. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by C

alif

orni

a In

stitu

te o

f T

echn

olog

y on

09/

26/1

2. F

or p

erso

nal u

se o

nly.

ANRV345-BI77-05 ARI 1 May 2008 11:17

ribonucleoproteins (RNP). Because many ex-tensive reviews are available (57, 97), we onlybriefly introduce three very recent studies.A straightforward but especially importantapplication of single-molecule fluorescencetechniques is counting the numbers ofsubunits in an RNP, an often difficult taskfor a large multicomponent complex. Forinstance, it has been deduced from electroncryomicroscopy and crystallography (98) thateach bacteriophage phi29 DNA-packagingmotor has six packaging RNA (pRNA)molecules. In a recent experiment (99),pRNAs, each labeled with a fluorophore,were incorporated into the packaging motor.By sequentially photobleaching fluorophoreson single motors, stoichiometry of six pRNAper motor has been confirmed.

Annealing of two hairpins with comple-mentary sequences into a duplex requiresunfolding of both hairpins. In HIV, TARRNA is annealed to its complementary DNA(cTAR) sequence with the help of the nu-cleocapsid protein during the critical minus-strand transfer step (100). The role of theprotein in this process remains unclear.Combining microfluidic techniques withsingle-molecule FRET and fluorescence cor-relation spectroscopy, Barbara and colleaguesobserved many intermediates in the annealingof TAR and cTAR, revealing both the anneal-ing pathway and the chaperone activity of thenucleocapsid protein (101, 102).

Assembly of RNPs is often sequential, sug-gesting that early protein binding events in-duce a conformational change in RNA thatallows subsequent associations. By labeling apair of fluorophores at various positions of thetelomeric RNA, Stone and colleagues havedemonstrated that binding of p65 protein in-duces a conformational change in the telo-mere RNA that facilitates the binding of thetelomerase reverse transcriptase (103).

ENZYMES THAT UNFOLD RNA

In physiological conditions RNA folds spon-taneously, but requires energy to unfold. A

single base pair closing a loop in RNA isnot stable, because base-base hydrogen bondsare only slightly more stable than base-waterhydrogen bonds, and there is a loss of en-tropy on constraining the RNA to form aloop. The free energy change is positive.However, adding successive base pairs soonproduces a stable structure with a nega-tive free energy change relative to the sin-gle strand. At 37◦C three G·Cs closing atetraloop (GCG[UUCG]CGC) are stable by–2.4 kcal mol−1 in 1 M NaCl; it takes fiveA·Us (AUAUA[UUCG]UAUAU) to becomestable with a �G = –1.9 kcal mol−1 (104).Each additional base pair—depending onthe sequence—decreases the free energy by–1.0 to –3.4 kcal mol−1. In ionic environ-ments more closely resembling those of a cell(100 mM univalent, 10 mM divalent ions), thefree energies are expected to be similar. Thestandard free energy of hydrolysis of ATP at37◦C is –7.4 kcal mol−1, therefore the cellmust burn one ATP to unfold three to fourRNA base pairs. An alternative way to un-fold RNA is to bind the single strand by asingle-strand specific protein. The binding tothe single strand must be strong enough sothat it can compete with base pair formation.An average �G of +2 kcal mol−1 for breakinga base pair means that the dissociation con-stant for single-strand specific protein is in-creased by a factor of 25 (e�G/RT) for eachbase pair that is broken before binding canoccur. Thus, if four base pairs must be un-folded to form a single-strand binding site fora protein, the dissociation constant for bind-ing the single strand increases by two orders ofmagnitude.

RNA must be single stranded for its se-quence to be interpreted during viral RNAreplication and translation, so it must first beunfolded by the RNA-dependent RNA poly-merase or ribosome, or by a helicase thatassists the unfolding. The enzymes use thechemical energy available in the hydrolysis ofnucleoside triphosphates to unfold their RNAsubstrates. The functions of these molecularmotors require them to work on RNAs of


Ann

u. R

ev. B

ioch

em. 2

008.

77:7

7-10

0. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by C

alif

orni

a In

stitu

te o

f T

echn

olog

y on

09/

26/1

2. F

or p

erso

nal u

se o

nly.

ANRV345-BI77-05 ARI 1 May 2008 11:17

widely differing sequences. They are promis-cuous; however, initiation of their activity of-ten involves specific sequences and structures.

HELICASES

Helicases are molecular motors that convertthe chemical energy of ATP hydroly-sis into removing bound proteins andmechanically separating the strands ofdouble-stranded nucleic acids (105). TheDEAD-box family of RNA helicases belongsto protein superfamily II and is by far thelargest family of RNA helicases—over 500sequences are known (106). Several crystalstructures have been published, includingthe structure of the NS3 hepatitis C helicasebound to DNA (107), and the structureof a thermophilic RNA helicase bound toATP (108). The many biochemical functionsof helicases have been reviewed (106, 109,110). We concentrate on their mechanismof action, particularly by single-moleculestudies (111).

Two RNA helicases essential for viral repli-cation have been extensively studied: NPH-IIhelicase from vaccinia virus (112, 113), andNS3 helicase from hepatitis C virus (114–117). Both load on an overhanging 3′-singlestrand and move in a 3′-to-5′ direction. Un-winding begins when the helicase arrives atthe junction of the double strand and singlestrand; the helicase then unwinds the RNA ina series of bursts and pauses. During the pausethe helicase can dissociate from the RNA orbegin another step of unwinding. The numberand size of unwinding steps that occurs deter-mines the processivity of the process (118).The fundamental characteristics of these he-licases (step sizes, efficiency, processivity) arenot well established. It is not even certain howmany helicases are actually working on eachRNA in different experiments.

Single-molecule studies of the hepatitis Cvirus RNA helicase were done by attaching theends of an RNA containing a hairpin to twobeads and monitoring the increase in end-to-end distance of the RNA as a helicase unfolded

the molecule at constant force (116). The ex-tension versus time showed discrete unwind-ing steps (extension increased with time) fol-lowed by pauses (extension did not changeduring the pause). The translocation stepswere 11 bp with substeps of 3 to 4 bp, in-dependent of ATP concentration. This re-sult contrasts with an earlier ensemble ex-periment, which found an average step sizeof 18 bp (114). The difference in measuredstep size between single-molecule and ensem-ble experiments were attributed to differencesin the oligomeric state of the active helicase;monomeric in the former and dimeric in thelatter. However, in neither experiment wasthe actual number of RNA-bound helicasesmeasured.

The single-molecule experiments allowedmeasurement of the effects of experimentalvariables (ATP, RNA sequence, force) on eachstep in the helicase mechanism. IncreasingATP concentration increases the rate of re-action, as expected, but has no effect on pro-cessivity. Increasing ATP concentration de-creases both the length of each pause andthe time for each unwinding step. Analysisof the dependence of the pauses, unwindingsteps and substeps on ATP concentration indi-cates that at least one ATP is involved in eachpause and substep. The unfolding of 11 bpclearly requires more than one ATP to be hy-drolyzed per cycle of pause and translocation.

Single-molecule fluorescence studies (119)of the hepatitis C NS3 helicase unwindingDNA have found steps of 3–4 bp, with evi-dence of 1-bp substeps and 1 ATP hydrolyzedper base pair. How much of the differences isdue to DNA versus RNA substrates or otherdifferences in the experiments (such as forceassistance) requires more work.

A question that is often posed for molec-ular motors is whether they take advantageof thermal energy to move—are they passivemotors or do they actively destabilize basepairs (120)? To test whether the HCV heli-case acts like a passive Brownian motor thatwaits for the double-stranded region to openspontaneously before moving forward, the


Ann

u. R

ev. B

ioch

em. 2

008.

77:7

7-10

0. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by C

alif

orni

a In

stitu

te o

f T

echn

olog

y on

09/

26/1

2. F

or p

erso

nal u

se o

nly.

ANRV345-BI77-05 ARI 1 May 2008 11:17

kinetics were compared with the probabil-ity of thermal base-pair opening of differ-ent RNA sequences. Two 60-bp hairpins werestudied; one contained 30 A·Us followed by30 G·Cs; the other had the two sequences in-terchanged so that the NS3 helicase encoun-tered the G·Cs first (117). The kinetics ofunwinding a helix by a pure Brownian mech-anism should vary with e−�G◦/RT , the expo-nential of the free energy of base pair opening(�G◦) compared to thermal energy, RT. At37◦C, �G◦ for opening a single base pair inthe sequence AA

UU is 1.0 kcal mol−1, it is 3.3kcal mol−1 for opening a base pair in the se-quence GG

CC (104). The difference in free en-ergy corresponds to a 42-fold faster rate forpassive opening of one base pair at a time fromthe A·U sequence compared to the G·C se-quence. If the helicase opens two base pairs ata time, the calculated rate increase for A·Usover G·Cs becomes 1750. The observed ratesof helix unwinding changed much less withsequence. NS3 pauses 10 times longer andtranslocates 3 times slower on G·C base pairsthan A·U base pairs. This result shows thatthe HCV helicase is not a passive Brownianmotor, but actively opens base pairs. A ring-shaped helicase from T7 that unwinds DNAhas also been found to be an active motor(121).

The processivity of NS3 also depends onthe G·C content; strong barriers accelerateNS3 detachment from RNA before the basepairs are unwound. The probability that thehelicase dissociates from the RNA increasesas it approaches a G·C sequence; this increaseoccurs up to 6 bp ahead of the opening fork(117).

Force applied to the hairpin changed nei-ther the helicase’s step size, nor its rate of un-winding. However, increasing force did in-crease processivity. This implies that forcedestabilizes the double strand, decreasing thebarrier to unfolding and favoring the enzymeremaining bound to the RNA during pauses.

A mechanistic picture consistent with allthese results (116, 117) is that the helicase hasone site that binds the single-stranded RNA

Repeat

Duplex destabilization

Substeps

Figure 5An inch worm model of the motion of NS3 helicase adapted fromReference 116. In each translocation step, the helicase reaches ahead anddestabilizes the double strand; the single-strand binding site then inchesforward unwinding the duplex.

and another that binds and destabilizes thedouble-stranded RNA ahead of the unwind-ing fork. The single-strand binding site inchesforward unwinding base pairs until it reachesthe double-strand binding site. The double-strand binding site releases the RNA, the heli-case stretches, and the site binds and destabi-lizes the RNA duplex ahead to repeat the cycle(Figure 5). The probability that the helicasewill release from the RNA, and thus decreasethe processivity, depends on the stability ofthe helix. A helix with high G·C content willdecrease binding and destabilization of theduplex by the double-strand binding domainand may allow the single-strand site to re-lease the helicase. Force on the hairpin willdestabilize the helix and thus increase theprocessivity.

RNA-DEPENDENT RNAPOLYMERASES

RNA viruses—other than retroviruses—arereplicated by their viral-encoded RdRps.These polymerases must synthesize both plusstrand and minus strand chains. They can ini-tiate from primers or by de novo initiationusually starting with a GTP. Non-viral RdRpsalso exist in many organisms. They synthesizeRNAs from RNA templates and are involvedin many cellular functions, including RNAsilencing (122).

Although many single-molecule studies ofDNA-dependent RNA polymerases (DdRps)have been done (123–126), RdRp studies have


Ann

u. R

ev. B

ioch

em. 2

008.

77:7

7-10

0. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by C

alif

orni

a In

stitu

te o

f T

echn

olog

y on

09/

26/1

2. F

or p

erso

nal u

se o

nly.

ANRV345-BI77-05 ARI 1 May 2008 11:17

not yet appeared. The experiment could beexactly analogous to the helicase study de-scribed here, or follow the DdRp-type exper-iments. There the enzyme is attached to onebead and the distance between the enzyme and5′ end of the RNA is monitored.

RIBOSOME

The ribosome is a very complex molecularmachine that catalyzes the hydrolysis of GTPto translate messenger RNAs into polypep-tides (127, 128). High resolution structuresare only available for prokaryotic ribosomes,and single-molecule experiments have notbeen done on eukaryotic ribosomes, thereforewe limit ourselves here to the prokaryotes.The function of ribosomes is so important tolife that it is the objective of many researchersto establish a detailed motion picture of theaction of a ribosome during translation. Thegoal is a mechanism of translation that charac-terizes the conformation of the ribosome andthe positions of the mRNA, the tRNAs, andthe translation factors at each step in the pro-cess. The energy at each step, and the barri-ers between steps—the thermodynamics andkinetics of the process—are also necessary tounderstand the operation of the ribosome. Wesummarize the progress here with emphasison single-molecule studies of the motion ofthe mRNA and tRNAs.

Translation

Initiation occurs with the binding of anmRNA, formyl methionine tRNA (fMet-tRNAfMet), and initiation factors IF1,IF2·GTP, and IF3 to the 30S subunit of theribosome. Next the 50S subunit is recruited,GTP is hydrolyzed, and the initiation factorsare released, leaving the fMet-tRNAfMet

bound at the P-site of the ribosome. Elon-gation starts when EF-Tu·tRNA·GTP bindsto the ribosome, which has a peptidyl-tRNAor fMet-tRNA in the P-site. Interactionof the EF-Tu with the ribosome catalyzesthe hydrolysis of the GTP, the correct

aminoacylated-tRNA is left at the A-site, andEF-Tu·GDP is released. A new peptide bondis formed when the α-amino group of theA-site amino acid attacks the carboxyl carbonof the amino acid linked to the P-site tRNA.As a peptide bond replaces an ester linkage,this step in the the reaction is spontaneous,and free energy decreases. Once the peptidechain has been transferred to the A-sitetRNA, a hybrid state is favored in whichthe 3′ end of the P-site tRNA moves to theexit site (E-site) and the 3′ end of the A-sitetRNA moves to the P-site. Now EF-G·GTPbinds and translocation occurs as GTP ishydrolyzed by interaction with the ribosome.Translocation involves the movement ofthe mRNA with the two tRNAs by 3 ntsto position the next codon at the A-site ofthe 30S subunit. Translocation can occureven in the absence of EF-G, but it is veryslow. Translation stops when a stop codonis reached. A release factor recognizes thestop codon and hydrolyzes the completedpolypeptide from the P-site RNA.

During translation the ribosome must beable to process 61 different codons and 3 dif-ferent stop signals. It interacts with a mini-mum of 20 tRNAs, but it can interact withup to nearly 60 different tRNAs, dependingon the organism. It must unfold base-pairedsecondary structures, as well as tertiary struc-tures, such as pseudoknots, in the mRNAs. Allthis is done rapidly (∼1–10 codons s−1) withhigh fidelity. The ribosome plus its translationfactors EF-Tu and EF-G constitute a mar-velous machine that has great flexibility forits RNA reactants, but very high precision forthe product protein it makes. A detailed mech-anism will help in revealing when and whyribosomes make mistakes, such as frameshift-ing, incorporating the wrong amino acid, orprematurely terminating.

Mechanism of Elongation

The kinetics of elongation have been exten-sively studied by the Rodnina-Wintermeyergroup using stopped-flow fluorescence and


Ann

u. R

ev. B

ioch

em. 2

008.

77:7

7-10

0. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by C

alif

orni

a In

stitu

te o

f T

echn

olog

y on

09/

26/1

2. F

or p

erso

nal u

se o

nly.

ANRV345-BI77-05 ARI 1 May 2008 11:17

quench flow methods (129–132). They haveobtained values for many of the rate con-stants in the reaction that can be used to cal-culate the time dependence of each species.Figure 6 shows the decrease in fraction ofempty A-sites and increase in fraction of A-sites containing an aminoacyl tRNAaa as afunction of time. Two intermediates of thereaction are also shown (EF·Tu·GTP·tRNAaa

bound and GTP hydrolyzed to form boundEF-Tu·GDP·tRNAaa = I1). Once the cor-rect tRNAaa is accommodated in the A-site,peptide bond formation is fast and not ratelimiting. Next, formation of the hybrid stateoccurs (discussed in the next section, Single-Molecule Translation), and translocation fin-ishes one step of the cycle.

SINGLE-MOLECULETRANSLATION

Figure 6 shows the fundamental limitationof ensemble kinetics; reactions occur asyn-chronously. This means that throughout thereaction reactants and products are bothpresent, and intermediates will appear and dis-appear as the reaction proceeds. This mixtureobviously complicates the problem of identi-fying and characterizing each species. In con-trast, in single-molecule experiments there isonly one species present at one time. Whenthe reactant reacts, it is replaced by an inter-mediate, which is replaced by the next inter-mediate, which eventually is transformed toproduct. Instead of rate constants, the life-time of each species is measured. The single-molecule reaction is repeated many times,each time yielding a different set of life-times for the species—because of the ran-dom nature of kinetics. The signal fromtwo species in a stable equilibrium will ex-hibit “hopping” with the signal switching intime, and corresponding alternately to eachstate.

The distribution of lifetime values can re-veal hidden intermediates between the mea-sured species. For a reaction with N steps (N-1

A-siteA-site•tRNAaa

I1

A-site•EF-Tu•GTP•tRNAaa

0.2

0.4

0.6

0.8

1

Fra

ctio

n

00.1 0.2 0.3 0.4 0.50

Time (s)

Figure 6Time-dependence of the incorporation of a tRNA by EF-Tu into the A-siteof a ribosome. The kinetic steps of the reaction are: A-site + EF − Tu·GTP·tRNAaa

k1�k−1

A-site·EF-Tu·GTP·tRNAaa →k2

I1 →k3

A-site·tRNAaa +EF-Tu·GDP, I1 = GTP hydrolyzed to form bound EF-Tu·GDP·tRNAaa.The fraction of A-site reacted is calculated for k1 = 100 μM−1 s−1;k−1 = 25 s−1; k2 = 35 s−1; k3 = 7 s−1 for a constant EF-Tuconcentration of 0.1 μM. Data from Reference 129.

intermediates) with equal rate constants, k,the distribution of lifetimes, τ , is a Poissonequation.

dP (τ ) =(

kNτ N−1

(N − 1)!

)(e−kτ )dτ 5.

Here dP(τ )/dτ is the probability density ofmeasuring a lifetime between τ and τ + dτ . Ifthere are no intermediates (N = 1), the dis-tribution is a simple exponential with a max-imum at τ = 0, and the mean value of thelifetimes, 〈τ 〉 = 1/k. For any value of N, thedistribution is at a maximum at τ = (N-1)/kwith mean value 〈τ 〉 = N/k. For hidden inter-mediates with unequal rate constants, the dis-tribution will depend on differences of expo-nentials. Thus, any difference from an expo-nential distribution of lifetimes indicates hid-den intermediates.

Blanchard et al. (133, 134) used FRETfrom fluorophore-labeled tRNAs to study thekinetics of tRNA selection and of transloca-tion in single-molecule experiments. Prior totranslocation, the tRNAs move to a hybridstate from the classical state (one tRNA is


Ann

u. R

ev. B

ioch

em. 2

008.

77:7

7-10

0. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by C

alif

orni

a In

stitu

te o

f T

echn

olog

y on

09/

26/1

2. F

or p

erso

nal u

se o

nly.

ANRV345-BI77-05 ARI 1 May 2008 11:17

in the P-site of both ribosomal subunits, andthe other is in the A-site of both subunits).The process is a dynamic equilibrium with theclassical state tRNApeptide in the A-site hav-ing a lifetime of 0.2 s. The hybrid statetRNApeptide shows two lifetimes with nearlyequal probabilities for returning to the

classical state (0.08 s, 0.39 s). These lifetimes(133) correspond to values of khybrid→classic

= 5 s−1 and kclassic→hybrid = 12.5 s−1 and2.5 s−1. The ensemble results of the Rodnina-Wintermeyer group and the single-moleculeresults of the Puglisi-Chu group are con-sistent, which is encouraging. The review

0

1

2

3

4

5

6b

Co

un

ts (

x104 )

0 2.7 5.4 8.1 10.8 13.5 16.2 18.9 0 0.05 0.1 0.15 0.2 0.25

Translocation time (s)Pairwise distance (nm)

a

21

20

19Fo

rce

(pN

)E

xten

sio

n

146 148 150 152 154 156 158 160 162 164 166

20 nm

1 nm

0.1 s

c

10

12

8

6

4

2

0

Pro

bab

ility

den

sity

(s–1

)

Time (s)

Figure 7Single ribosome translation. (a) A trajectory of extension of mRNA versus time for the translation of anRNA hairpin by a single ribosome; an enlarged view of a single translocation step is shown as an inset.(b) Pairwise distribution of points showing that translocation occurs with steps of constant increase inextension of 2.7 nm, corresponding to translocation of 3 nts (1 codon) per step. (c) The distribution oftranslocation lifetimes fits a Poisson equation with three substeps corresponding to 23 ms per substep.The purple curve is the best fit to a one-substep reaction, blue is for two substeps, and red is for threesubsteps. Reprinted with permission from Reference 136.


Ann

u. R

ev. B

ioch

em. 2

008.

77:7

7-10

0. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by C

alif

orni

a In

stitu

te o

f T

echn

olog

y on

09/

26/1

2. F

or p

erso

nal u

se o

nly.

ANRV345-BI77-05 ARI 1 May 2008 11:17

by Puglisi et al. (135) in this volumegives much more details.

We (136) have measured the increase inextension of a single mRNA molecule as aribosome translates the message. The methodis identical to the one used in our single-molecule, laser tweezers studies of HCV NS3RNA helicase (116). Beads are attached tothe ends of a hairpin RNA and are held ata force below that necessary to unwind thehairpin. As the ribosome translates the 5′-sideof the hairpin, the double helix is opened,releasing 1 nt on the 3′-side of the hairpinfor each nucleotide translated. We thus mea-sure a combination of the translocation andhelicase activity of the ribosome. Figure 7a

shows a time trace of the translation of a 60-bphairpin containing only valine (GUN) andglutamic acid (GAPu) codons. Characteristiccycles of a long (1–2 s) pause followed by ashort (0.1 s) translocation step are seen. Theincrease in end-to-end distance of the mRNAis 2.7 nm per step (Figure 7b), which cor-responds at 20-pN force to 6 nts unfoldedper step. Thus at each step of translation, acodon of 3 nts is traversed by the ribosome,and another 3 nts are released from the duplexof the hairpin. The distribution of translo-cation lifetimes fits a Poisson equation withthree identical substeps with mean lifetimes of

23 ms (Figure 7c), corresponding to a meantranslocation time of 69 ms. The measuredmean translocation time was 78 ± 47 ms.The distribution of pause times indicates twosubsteps with different rate constants. Al-though the pause times can differ widelyfrom ribosome to ribosome, and depend onthe force applied to the RNA, the translo-cation times are independent of force andribosome.

Single-molecule studies of translation arejust beginning. The processes and states de-termined by FRET and by laser tweezers arenot easily compared. During the pauses seenin the force experiments, the binding of EF-Tu, the recognition and insertion of the cor-rect tRNA, and the binding of EF-G takeplace. Variation of concentrations of factorsand substrates, and better time and distanceresolution, are needed to identify and char-acterize each of the intermediates involved inone cycle of translation of one codon. In theFRET experiments, fluorophores placed onthe ribosomal RNAs and proteins, the mRNAand the tRNAs can show relative motions ofall the actors. Of course, ensemble studies willcontinue to provide important information. Ina few years the goal of a step-by-step motionpicture of the prokaryotic ribosome in actionmay be realized.

SUMMARY POINTS

1. Mechanical force can be applied to unfold single RNA molecules. Structural transi-tions, indicated by changes in the extension of the molecule, are monitored in realtime. The free energy changes can be obtained from the mechanical work done tounfold the structure.

2. Single-molecule force manipulation can be used to control the structure and foldingpathways of large RNAs, allowing characterization of sequential un/refolding steps.Force can also induce RNA misfolding.

3. Single-molecule fluorescence techniques are powerful in studying tertiary folding anddomain dynamics of RNA structures, as well as in characterizing detailed reactionmechanisms of ribozymes.

4. Single-molecule force and fluorescence studies have elucidated the mechanisms bywhich helicases unwind DNA or RNA duplexes and convert chemical energy intomechanical work.


Ann

u. R

ev. B

ioch

em. 2

008.

77:7

7-10

0. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by C

alif

orni

a In

stitu

te o

f T

echn

olog

y on

09/

26/1

2. F

or p

erso

nal u

se o

nly.

ANRV345-BI77-05 ARI 1 May 2008 11:17

5. Translation by a single ribosome on a single RNA occurs by successive cycles oftranslocation steps and pauses. Each translocation step involves a motion of the RNAby 3 nts—one codon. Ribosome possesses helicase activity that can unwind secondarystructures in mRNA during translation.

FUTURE ISSUES

1. Temperature-controlled optical tweezers should be applied to directly measure en-thalpy and entropy changes of RNA folding at physiological temperatures. Kineticsof folding as a function of temperature are also needed.

2. More theoretical effort is needed to explain force unfolding kinetics, especially forintermediates and misfolded structures.

3. It remains unclear how RNA binding proteins, particularly chaperones, change theRNA structure. Assembly processes of large ribonucleoproteins, such as the spliceo-some and telomere, are also not clear. Single-molecule techniques may solve some ofthese puzzles.

4. There are many different helicases and nucleic acid translocases. Single-moleculeassays can provide detailed mechanisms for the activities of these molecular motorswith different functions.

5. Single-ribosome translation assays can answer perennial questions on the effects ofmessenger RNA structure on translation. A single example is: What is the mechanismof pseudoknot-induced translational frameshifting?

DISCLOSURE STATEMENT

The authors are not aware of any biases that might be perceived as affecting the objectivity ofthis review.

ACKNOWLEDGMENTS

We thank our colleagues who provided high-resolution figures (Figures 2, 3, 4–7). This workwas supported by National Institutes of Health grant GM-10840 (I.T.), a grant from Universityof Albany (P.T.X.L.), and a Graduate Research and Education in Adaptive bio-Technology(GREAT) Fellowship ( J.V.).

LITERATURE CITED1. Tinoco I Jr, Bustamante C. 1999. J. Mol. Biol. 293:271–812. Lu ZJ, Turner DH, Mathews DH. 2006. Nucleic Acids Res. 34:4912–243. Zuker M. 2003. Nucleic Acids Res. 31:3406–154. Hofacker IL. 2003. Nucleic Acids Res. 31:3429–315. Dirks RM, Bois JS, Schaeffer JM, Winfree E, Pierce NA. 2007. SIAM Rev. 49:65–886. Liphardt J, Onoa B, Smith SB, Tinoco I Jr, Bustamante C. 2001. Science 292:733–377. Mangeol P, Cote D, Bizebard T, Legrand O, Bockelmann U. 2006. Eur. Phys. J. E Soft

Matter 19:311–17


Ann

u. R

ev. B

ioch

em. 2

008.

77:7

7-10

0. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by C

alif

orni

a In

stitu

te o

f T

echn

olog

y on

09/

26/1

2. F

or p

erso

nal u

se o

nly.

ANRV345-BI77-05 ARI 1 May 2008 11:17

8. Collin D, Ritort F, Jarzynski C, Smith SB, Tinoco I Jr, Bustamante C. 2005. Nature437:231–34

9. Li PTX, Collin D, Smith SB, Bustamante C, Tinoco I Jr. 2006. Biophys. J. 90:250–6010. Woodside MT, Anthony PC, Behnke-Parks WM, Larizadeh K, Herschlag D, Block SM.

2006. Science 314:1001–411. Woodside MT, Behnke-Parks WM, Larizadeh K, Travers K, Herschlag D, Block SM.

2006. Proc. Natl. Acad. Sci. USA 103:6190–9512. Tinoco I Jr. 2004. Annu. Rev. Biophys. Biomol. Struct. 33:363–8513. Seol Y, Skinner GM, Visscher K. 2004. Phys. Rev. Lett. 93:11810214. Seol Y, Skinner GM, Visscher K, Buhot A, Halperin A. 2007. Phys. Rev. Lett. 98:15810315. Green NH, Williams PM, Wahab O, Davies MC, Roberts CJ, et al. 2004. Biophys. J.

86:3811–2116. Evans E, Ritchie K. 1997. Biophys. J. 72:1541–5517. Jarzynski C. 1997. Phys. Rev. Lett. 78:2690–9318. Crooks GE. 1999. Phys. Rev. E 60:2721–2619. Liphardt J, Dumont S, Smith SB, Tinoco I Jr, Bustamante C. 2002. Science 296:1832–3520. Hummer G, Szabo A. 2005. Acc. Chem. Res. 38:504–1321. Minh DD. 2006. Phys. Rev. E 74:06112022. Smith S, Cui Y, Bustamante C. 2003. Methods Enzymol. 361:134–6223. Manosas M, Wen JD, Li PTX, Smith SB, Bustamante C, et al. 2007. Biophys. J. 92:3010–

2124. Wen JD, Manosas M, Li PTX, Smith SB, Bustamante C, et al. 2007. Biophys. J. 92:2996–

300925. Gralla J, DeLisi C. 1974. Nature 248:330–3226. Lindahl T, Adams A, Fresco JR. 1966. Proc. Natl. Acad. Sci. USA 55:941–4827. Li PTX, Bustamante C, Tinoco I Jr. 2007. Proc. Natl. Acad. Sci. USA 104:7039–4428. Hyeon C, Thirumalai D. 2005. Proc. Natl. Acad. Sci. USA 102:6789–9429. Hyeon C, Thirumalai D. 2006. Biophys. J. 90:3410–2730. Hyeon C, Thirumalai D. 2007. Biophys. J. 92:731–4331. Manosas M, Ritort F. 2005. Biophys. J. 88:3224–4232. Manosas M, Ritort F. 2006. Phys. Rev. Lett. 96:21830133. Ha T, Zhuang X, Kim HD, Orr JW, Williamson JR, Chu S. 1999. Proc. Natl. Acad. Sci.

USA 96:9077–8234. Joo C, Balci H, Ishitsuka Y, Buranachai C, Ha T. 2008. Annu. Rev. Biochem. 77:51–7635. Cate JH, Gooding AR, Podell E, Zhou K, Golden BL, et al. 1996. Science 273:1678–8536. Jaeger L, Westhof E, Leontis NB. 2001. Nucleic Acids Res. 29:455–6337. Hodak JH, Downey CD, Fiore JL, Pardi A, Nesbitt DJ. 2005. Proc. Natl. Acad. Sci. USA

102:10505–1038. Downey CD, Fiore JL, Stoddard CD, Hodak JH, Nesbitt DJ, Pardi A. 2006. Biochemistry

45:3664–7339. Bartley LE, Zhuang X, Das R, Chu S, Herschlag D. 2003. J. Mol. Biol. 328:1011–2640. Latham JA, Cech TR. 1989. Science 245:276–8241. Brunel C, Marquet R, Romby P, Ehresmann C. 2002. Biochimie 84:925–4442. Li PTX, Bustamante C, Tinoco I Jr. 2006. Proc. Natl. Acad. Sci. USA 103:15847–5243. Giedroc DP, Theimer CA, Nixon PL. 2000. J. Mol. Biol. 298:167–8544. Theimer CA, Feigon J. 2006. Curr. Opin. Struct. Biol. 16:307–1845. Shih IH, Been MD. 2002. Annu. Rev. Biochem. 71:887–91746. Rivas E, Eddy SR. 1999. J. Mol. Biol. 285:2053–68


Ann

u. R

ev. B

ioch

em. 2

008.

77:7

7-10

0. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by C

alif

orni

a In

stitu

te o

f T

echn

olog

y on

09/

26/1

2. F

or p

erso

nal u

se o

nly.

ANRV345-BI77-05 ARI 1 May 2008 11:17

47. Isambert H, Siggia ED. 2000. Proc. Natl. Acad. Sci. USA 97:6515–2048. Huang X, Ali H. 2007. Nucleic Acids Res. 35:656–6349. Cao S, Chen SJ. 2007. J. Mol. Biol. 367:909–2450. Hansen TM, Reihani SN, Oddershede LB, Sorensen MA. 2007. Proc. Natl. Acad. Sci.

USA 104:5830–3551. Green L, Kim CH, Bustamante C, Tinoco I Jr. 2008. J. Mol. Biol. 375:511–2852. Shapiro BA, Yingling YG, Kasprzak W, Bindewald E. 2007. Curr. Opin. Struct. Biol.

17:157–6553. Sotomayor M, Schulten K. 2007. Science 316:1144–4854. Rhodes MM, Reblova K, Sponer J, Walter NG. 2006. Proc. Natl. Acad. Sci. USA

103:13380–8555. Zhuang X, Bartley LE, Babcock HP, Russell R, Ha T, et al. 2000. Science 288:2048–5156. Lilley DM. 2005. Curr. Opin. Struct. Biol. 15:313–2357. Zhuang X. 2005. Annu. Rev. Biophys. Biomol. Struct. 34:399–41458. Nahas MK, Wilson TJ, Hohng S, Javie K, Lilley DM, Ha T. 2004. Nat. Struct. Mol. Biol.

11:1107–1359. Liu S, Bokinsky G, Walter NG, Zhuang X. 2007. Proc. Natl. Acad. Sci. USA 104:12634–3960. Fiammengo R, Jaschke A. 2005. Curr. Opin. Biotechnol. 16:614–2161. Gilbert W. 1986. Nature 319:61862. Kobitski AY, Nierth A, Helm M, Jaschke A, Nienhaus GU. 2007. Nucleic Acids Res.

35:2047–5963. Manning GS. 1969. J. Chem. Phys. 51:924–3364. Magdelenat H, Turq P, Chemla M. 1974. Biopolymers 13:1535–4865. Manning GS. 1979. Accts. Chem. Res. 12:443–4966. De Rose VJ. 2003. Curr. Opin. Struct. Biol. 13:317–2467. Draper DE, Grilley D, Soto AM. 2005. Annu. Rev. Biophys. Biomol. Struct. 34:221–4368. Pyle AM, Fedorova O, Waldsich C. 2007. Trends Biochem. Sci. 32:138–4569. Sharp KA, Honig B. 1995. Curr. Opin. Struct. Biol. 5:323–2870. Woodson SA. 2005. Curr. Opin. Chem. Biol. 9:104–971. Sigel RK, Pyle AM. 2007. Chem. Rev. 107:97–11372. Rueda D, Wick K, McDowell SE, Walter NG. 2003. Biochemistry 42:9924–3673. Soto AM, Misra V, Draper DE. 2007. Biochemistry 46:2973–8374. Anderson CF, Record MT Jr. 1995. Annu. Rev. Phys. Chem. 46:657–70075. Tinoco I Jr, Li PTX, Bustamante C. 2006. Q. Rev. Biophys. 39:325–6076. Record MT Jr, Anderson CF. 1993. J. Phys. Chem. 97:7116–2677. Sharp KA, Friedman RA, Misra V, Hecht J, Honig B. 1995. Biopolymers 36:245–6278. Zacharias M, Luty BA, Davis ME, McCammon JA. 1992. Biophys. J. 63:1280–8579. Misra VK, Draper DE. 2001. Proc. Natl. Acad. Sci. USA 98:12456–6180. Bloomfield VA, Crothers DM, Tinoco I Jr. 2000. See Ref. 137, pp. 475–53481. Cantor CR, Schimmel PR. 1980. In Biophysical Chemistry, pp. 1109–77. New York:

Freeman82. Sun X, Zhang Q, Al-Hashimi HM. 2007. Nucleic Acids Res. 35:1698–71383. Zhang Q, Sun X, Watt ED, Al-Hashimi HM. 2006. Science 311:653–5684. Foloppe N, Matassova N, Aboul-Ela F. 2006. Drug Discov. Today 11:1019–2785. Sucheck SJ, Wong CH. 2000. Curr. Opin. Chem. Biol. 4:678–8686. Onoa B, Dumont S, Liphardt J, Smith SB, Tinoco I Jr, Bustamante C. 2003. Science

299:1892–9587. Doherty EA, Doudna JA. 2000. Annu. Rev. Biochem. 69:597–615


Ann

u. R

ev. B

ioch

em. 2

008.

77:7

7-10

0. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by C

alif

orni

a In

stitu

te o

f T

echn

olog

y on

09/

26/1

2. F

or p

erso

nal u

se o

nly.

ANRV345-BI77-05 ARI 1 May 2008 11:17

88. Cochrane JC, Lipchock SV, Strobel SA. 2007. Chem. Biol. 14:97–10589. Lorenz C, Piganeau N, Schroeder R. 2006. Nucleic Acids Res. 34:334–4290. Weixlbaumer A, Werner A, Flamm C, Westhof E, Schroeder R. 2004. Nucleic Acids Res.

32:5126–3391. McCauley MJ, Williams MC. 2007. Biopolymers 85:154–6892. Vladescu ID, McCauley MJN, Nunez ME, Rouzina I, Williams MC. 2007. Nat. Methods

4:517–2293. Koch SJ, Shundrovsky A, Jantzen BC, Wang MD. 2002. Biophys. J. 83:1098–10594. Koch SJ, Wang MD. 2003. Phys. Rev. Lett. 91:02810395. Jiang J, Bai L, Surtees JA, Gemici Z, Wang MD, Alani E. 2005. Mol. Cell 20:771–8196. Tinoco I Jr, Collin D, Li PTX. 2004. Biochem. Soc. Trans. 32:757–6097. Bokinsky G, Zhuang X. 2005. Acc. Chem. Res. 38:566–7398. Johnson JE, Chiu W. 2007. Curr. Opin. Struct. Biol. 17:237–4399. Shu D, Zhang H, Jin J, Guo P. 2007. EMBO J. 26:527–37

100. Levin JG, Guo J, Rouzina I, Musier-Forsyth K. 2005. Prog. Nucleic Acid Res. Mol. Biol.80:217–86

101. Liu HW, Zeng Y, Landes CF, Kim YJ, Zhu Y, et al. 2007. Proc. Natl. Acad. Sci. USA104:5261–67

102. Zeng Y, Liu HW, Landes CF, Kim YJ, Ma X, et al. 2007. Proc. Natl. Acad. Sci. USA104:12651–56

103. Stone MD, Mihalusova M, O’Connor CM, Prathapam R, Collins K, Zhuang X. 2007.Nature 446:458–61

104. Turner DH. 2000. See Ref. 137, pp. 259–334105. Singleton MR, Dillingham MS, Wigley DB. 2007. Annu. Rev. Biochem. 76:23–50106. Rocak S, Linder P. 2004. Nat. Rev. Mol. Cell Biol. 5:232–41107. Kim JL, Morgenstern KA, Griffith JP, Dwyer MD, Thomson JA, et al. 1998. Structure

6:89–100108. Rudolph MG, Heissmann R, Wittmann JG, Klostermeier D. 2006. J. Mol. Biol. 361:731–

43109. Delagoutte E, von Hippel PH. 2002. Q. Rev. Biophys. 35:431–78110. Delagoutte E, von Hippel PH. 2003. Q. Rev. Biophys. 36:1–69111. Rasnik I, Myong S, Ha T. 2006. Nucleic Acids Res. 34:4225–31112. Jankowsky E, Gross CH, Shuman S, Pyle AM. 2000. Nature 403:447–51113. Kawaoka J, Jankowsky E, Pyle AM. 2004. Nat. Struct. Mol. Biol. 11:526–30114. Serebrov V, Pyle AM. 2004. Nature 430:476–80115. Levin MK, Gurjar M, Patel SS. 2005. Nat. Struct. Mol. Biol. 12:429–35116. Dumont S, Cheng W, Serebrov V, Beran RK, Tinoco I Jr, et al. 2006. Nature 439:105–8117. Cheng W, Dumont S, Tinoco I Jr, Bustamante C. 2007. Proc. Natl. Acad. Sci. USA

104:13954–59118. Lucius AL, Maluf NK, Fischer CJ, Lohman TM. 2003. Biophys. J. 85:2224–39119. Myong S, Bruno MM, Pyle AM, Ha T. 2007. Science 317:513–16120. Astumian RD, Hanggi P. 2002. Physics Today 55:33–39121. Johnson DS, Bai L, Smith BY, Patel SS, Wang MD. 2007. Cell 129:1299–309122. Ahlquist P. 2002. Science 296:1270–73123. Landick R. 2006. Biochem. Soc. Trans. 34:1062–66124. Steitz TA. 2006. EMBO J. 25:3458–68125. Bai L, Santangelo TJ, Wang MD. 2006. Annu. Rev. Biophys. Biomol. Struct. 35:343–60


Ann

u. R

ev. B

ioch

em. 2

008.

77:7

7-10

0. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by C

alif

orni

a In

stitu

te o

f T

echn

olog

y on

09/

26/1

2. F

or p

erso

nal u

se o

nly.

ANRV345-BI77-05 ARI 1 May 2008 11:17

126. Galburt EA, Grill SW, Wiedmann A, Lubkowska L, Choy J, et al. 2007. Nature 446:820–23

127. Spirin AS. 2000. Ribosomes. New York: Springer. 422 pp.128. Liljas A. 2004. Structural Aspects of Protein Synthesis. Singapore: World Sci.129. Pape T, Wintermeyer W, Rodnina MV. 1998. EMBO J. 17:7490–97130. Wilden B, Savelsbergh A, Rodnina MV, Wintermeyer W. 2006. Proc. Natl. Acad. Sci.

USA 103:13670–75131. Savelsbergh A, Katunin VI, Mohr D, Peske F, Rodnina MV, Wintermeyer W. 2003. Mol.

Cell. 11:1517–23132. Wintermeyer W, Peske F, Beringer M, Gromadski KB, Savelsbergh A, Rodnina MV.

2004. Biochem. Soc. Trans 32:733–37133. Blanchard SC, Gonzalez RL, Kim HD, Chu S, Puglisi JD. 2004. Nat. Struct. Mol. Biol.

11:1008–14134. Blanchard SC, Kim HD, Gonzalez RL Jr, Puglisi JD, Chu S. 2004. Proc. Natl. Acad. Sci.

USA 101:12893–98135. Marshall RA, Aitken CE, Dorywalska M, Puglisi JD. 2007. Annu. Rev. Biochem. 77:177–

203136. Wen J-D, Lancaster L, Hodges C, Zeri A-C, Yoshimura SH, et al. 2008. Nature. 452:598–

603137. Bloomfield VA, Crothers DM, Tinoco I Jr. 2000. Nucleic Acids: Structure, Properties and

Functions. Sausalito, CA: Univ. Sci. Books


Ann

u. R

ev. B

ioch

em. 2

008.

77:7

7-10

0. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by C

alif

orni

a In

stitu

te o

f T

echn

olog

y on

09/

26/1

2. F

or p

erso

nal u

se o

nly.

AR345-FM ARI 7 May 2008 14:43

Annual Review ofBiochemistry

Volume 77, 2008Contents

Prefatory Chapters

Discovery of G Protein SignalingZvi Selinger � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �1

Moments of DiscoveryPaul Berg � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 14

Single-Molecule Theme

In singulo Biochemistry: When Less Is MoreCarlos Bustamante � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 45

Advances in Single-Molecule Fluorescence Methodsfor Molecular BiologyChirlmin Joo, Hamza Balci, Yuji Ishitsuka, Chittanon Buranachai,and Taekjip Ha � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 51

How RNA Unfolds and RefoldsPan T.X. Li, Jeffrey Vieregg, and Ignacio Tinoco, Jr. � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 77

Single-Molecule Studies of Protein FoldingAlessandro Borgia, Philip M. Williams, and Jane Clarke � � � � � � � � � � � � � � � � � � � � � � � � � � � � �101

Structure and Mechanics of Membrane ProteinsAndreas Engel and Hermann E. Gaub � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �127

Single-Molecule Studies of RNA Polymerase: Motoring AlongKristina M. Herbert, William J. Greenleaf, and Steven M. Block � � � � � � � � � � � � � � � � � � � �149

Translation at the Single-Molecule LevelR. Andrew Marshall, Colin Echeverría Aitken, Magdalena Dorywalska,and Joseph D. Puglisi � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �177

Recent Advances in Optical TweezersJeffrey R. Moffitt, Yann R. Chemla, Steven B. Smith, and Carlos Bustamante � � � � � �205

Recent Advances in Biochemistry

Mechanism of Eukaryotic Homologous RecombinationJoseph San Filippo, Patrick Sung, and Hannah Klein � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �229

v

Ann

u. R

ev. B

ioch

em. 2

008.

77:7

7-10

0. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by C

alif

orni

a In

stitu

te o

f T

echn

olog

y on

09/

26/1

2. F

or p

erso

nal u

se o

nly.

AR345-FM ARI 7 May 2008 14:43

Structural and Functional Relationships of the XPF/MUS81Family of ProteinsAlberto Ciccia, Neil McDonald, and Stephen C. West � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �259

Fat and Beyond: The Diverse Biology of PPARγ

Peter Tontonoz and Bruce M. Spiegelman � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �289

Eukaryotic DNA Ligases: Structural and Functional InsightsTom Ellenberger and Alan E. Tomkinson � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �313

Structure and Energetics of the Hydrogen-Bonded Backbonein Protein FoldingD. Wayne Bolen and George D. Rose � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �339

Macromolecular Modeling with RosettaRhiju Das and David Baker � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �363

Activity-Based Protein Profiling: From Enzyme Chemistryto Proteomic ChemistryBenjamin F. Cravatt, Aaron T. Wright, and John W. Kozarich � � � � � � � � � � � � � � � � � � � � � �383

Analyzing Protein Interaction Networks Using Structural InformationChristina Kiel, Pedro Beltrao, and Luis Serrano � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �415

Integrating Diverse Data for Structure Determinationof Macromolecular AssembliesFrank Alber, Friedrich Förster, Dmitry Korkin, Maya Topf, and Andrej Sali � � � � � � � �443

From the Determination of Complex Reaction Mechanismsto Systems BiologyJohn Ross � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �479

Biochemistry and Physiology of Mammalian SecretedPhospholipases A2

Gerard Lambeau and Michael H. Gelb � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �495

Glycosyltransferases: Structures, Functions, and MechanismsL.L. Lairson, B. Henrissat, G.J. Davies, and S.G. Withers � � � � � � � � � � � � � � � � � � � � � � � � � � �521

Structural Biology of the Tumor Suppressor p53Andreas C. Joerger and Alan R. Fersht � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �557

Toward a Biomechanical Understanding of Whole Bacterial CellsDylan M. Morris and Grant J. Jensen � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �583

How Does Synaptotagmin Trigger Neurotransmitter Release?Edwin R. Chapman � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �615

Protein Translocation Across the Bacterial Cytoplasmic MembraneArnold J.M. Driessen and Nico Nouwen � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �643

vi Contents

Ann

u. R

ev. B

ioch

em. 2

008.

77:7

7-10

0. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by C

alif

orni

a In

stitu

te o

f T

echn

olog

y on

09/

26/1

2. F

or p

erso

nal u

se o

nly.

AR345-FM ARI 7 May 2008 14:43

Maturation of Iron-Sulfur Proteins in Eukaryotes: Mechanisms,Connected Processes, and DiseasesRoland Lill and Ulrich Mühlenhoff � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �669

CFTR Function and Prospects for TherapyJohn R. Riordan � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �701

Aging and Survival: The Genetics of Life Span Extensionby Dietary RestrictionWilliam Mair and Andrew Dillin � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �727

Cellular Defenses against Superoxide and Hydrogen PeroxideJames A. Imlay � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �755

Toward a Control Theory Analysis of AgingMichael P. Murphy and Linda Partridge � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �777

Indexes

Cumulative Index of Contributing Authors, Volumes 73–77 � � � � � � � � � � � � � � � � � � � � � � � �799

Cumulative Index of Chapter Titles, Volumes 73–77 � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �803

Errata

An online log of corrections to Annual Review of Biochemistry articles may be foundat http://biochem.annualreviews.org/errata.shtml

Contents vii

Ann

u. R

ev. B

ioch

em. 2

008.

77:7

7-10

0. D

ownl

oade

d fr

om w

ww

.ann

ualr

evie

ws.

org

by C

alif

orni

a In

stitu

te o

f T

echn

olog

y on

09/

26/1

2. F

or p

erso

nal u

se o

nly.

How RNA Unfolds and Refoldshome.uchicago.edu/~jvieregg/pubs/annrev.pdf · ANRV345-BI77-05 ARI 1 May...

Documents

Transcript of How RNA Unfolds and Refoldshome.uchicago.edu/~jvieregg/pubs/annrev.pdf · ANRV345-BI77-05 ARI 1 May...