Lewis acid catalysis of phosphoryl transfer from a copper(II)-NTP complex in a kinase ribozyme.

0.88 MB·English

by  Biondi, Elisa

Published online 28 January 2013 Nucleic Acids Research, 2013, Vol. 41, No. 5 3327–3338
doi:10.1093/nar/gkt039
Lewis acid catalysis of phosphoryl transfer from a
copper(II)-NTP complex in a kinase ribozyme
Elisa Biondi1,2, Raghav R. Poudyal2, Joshua C. Forgy2, Andrew W. Sawyer2,
Adam W. R. Maxwell2 and Donald H. Burke1,2,*
1Department of Molecular Microbiology and Immunology, Bond Life Sciences Center, University of Missouri
School of Medicine, Columbia, MO 65211, USA and 2Department of Biochemistry, Bond Life Sciences Center,
University of Missouri School of Medicine, Columbia, MO 65211, USA
Received May 26, 2012; Revised January 4, 2013; Accepted January 8, 2013
ABSTRACT thechemicalstrategiesusedbythegreatmajorityofthem
are poorly understood. Revealing how they mediate these
The chemical strategies used by ribozymes to
transformationsiscrucialforassessingthefeasibilityofan
enhance reaction rates are revealed in part from
RNA-based metabolism, for evaluating RNA world
their metal ion and pH requirements. We find that theories of early evolution, for engineering artificial
kinase ribozyme K28(1-77)C, in contrast with previ- enzymesandothertoolsforsyntheticbiology,andforbio-
ouslycharacterizedkinaseribozymes,requiresCu2+ medical applications of ribozymes. Those ribozymes that
foroptimalcatalysisofthiophosphoryltransferfrom havebeenstudiedindetailuseseveralofthesamecatalytic
GTPcS. Phosphoryl transfer from GTP is greatly strategiesasproteinenzymes,suchasLewisacidcatalysis,
reduced in the absence of Cu2+, indicating a proton transfer, precise positioning and desolvation of
substrates and allosteric regulation, and a small handful
specific catalytic role independent of any potential
of natural and artificial nucleic acid catalysts exploit the
interactions with the GTPcS thiophosphoryl group.
chemical reactivity of bound organic cofactors (1–4).
In-line probing and ATPcS competition both argue
Metal ion cofactors accelerate ribozyme catalysis by
against direct Cu2+ binding by RNA; rather, these
polarizingandacidifyinginnerspherewater,byincreasing
data establish that Cu2+ enters the active site
electrophilicityofphosphatesandcarbonyls,bystabilizing
within a Cu2+(cid:2)GTPcS or Cu2+(cid:2)GTP chelation com- negative charges that develop on transition states, inter-
plex, and that Cu2+(cid:2)nucleobase interactions further mediates and products, by assisting in folding and by
enforce Cu2+selectivity and position the metal ion other mechanisms [reviewed in (5)]. Although Mg2+ is
for Lewis acid catalysis. Replacing Mg2+ with the dominant bioavailable divalent cation, the transition
[Co(NH ) ]3+ significantly reduced product yield, metalionsinthefirstrowoftheperiodictableprovidean
36
but not k , indicating that the role of inner-sphere array of unique chemical capacities, such as variations in
obs
Mg2+coordinationisstructuralratherthancatalytic. Lewis acidity, charge density, preferred bond lengths
Replacing Mg2+ with alkaline earths of increasing and coordination geometries. Some metal ions retain
their inner hydration sphere and interact via bound
ionic radii (Ca2+, Sr2+ and Ba2+) gave lower yields
water molecules. For others, one or more waters are
and approximately linear rates of product accumu-
replaced by direct contact with the ribozyme
lation. Finally, we observe that reaction rates
or substrate(s). The relative stabilities of inner-sphere
increased with pH in log-linear fashion with an complexes involving these ions tend to follow the Irving–
apparent pKa=8.0±0.1, indicating deprotonation in Williams series (6): Mn2+<Fe2+<Co2+<Ni2+<
the rate-limiting step. Cu2+>Zn2+. A different series is observed for the
affinities of these same ions in complexes with nucleotide
mono-, di- and tri phosphates (7) and phosphate mono-
INTRODUCTION
esters (5), for which the affinities follow the order
Artificialribozymeshavebeenisolatedbyinvitroselection Mn2+>Fe2+>Co2+&Ni2+<<Cu2+>>Zn2+<Cd2+.
to catalyse a wide range of chemical reactions, although Interestingly, Fe2+ has recently emerged as a potential
*To whom correspondence should be addressed. Tel:[phone]; Fax:[phone]; Email: [email protected]
Present address:
Elisa Biondi, Foundation for Applied Molecular Evolution, 720 SW 2nd Ave, Suite 207, Gainesville, FL 32601, USA.
The authors wish to be known that, in their opinion, the first two authors should be regarded as joint First Authors.
(cid:2)TheAuthor(s)2013.PublishedbyOxfordUniversityPress.
ThisisanOpenAccessarticledistributedunderthetermsoftheCreativeCommonsAttributionNon-CommercialLicense(http://creativecommons.org/licenses/
by-nc/3.0/),whichpermitsunrestrictednon-commercialuse,distribution,andreproductioninanymedium,providedtheoriginalworkisproperlycited.
3328 NucleicAcidsResearch,2013,Vol.41,No.5
prebioticmetalion,basedonitslikelyhighconcentrations hydrogen bond to the developing transition state. These
inaqueous solutionon theearly Earthand on itspositive observations for natural ribozymes contrast with studies
impact on ribozyme folding and catalysis (8). of two previously studied kinase ribozymes (Kin.46 and
Ribozymes have been identified that use most of these 2PT3.2min) (50,51) and by a kinase DNAzyme (Dk1)
ions. For example, the hammerhead (HH) ribozyme is (25). For all three catalysts, (thio)phosphoryl transfer
active in low concentrations of Mn2+, Co2+, Ni2+, Zn2+ rates were shown to be independent of pH over the
and Cd2+(9–14), and RNaseP functions with Zn2+as the range pH 6.0–8.0 (Dk1 lost activity above 8.0).
sole divalent cation (15). From in vitro selections, RNAs The presentwork details the metalion andpH require-
have been isolated that assemble Ni2+ or Pt2+ into a ments for phosphoryl transfer by ribozyme K28(1-77)C.
coordinate-covalent RNA-amino acid complex (16) and This 58nt RNA is a variant of ribozyme K28, which was
that interact with immobilized Ni2+ or Zn2+ on affinity originally selected as a 126nt species for thiophosphoryl
matrices (16,17). Use of divalent copper is highly transfer activity using GTPgS as donor (52). Ribozyme
unusual among structured nucleic acids—in particular K28(1-77)C folds into a compact pseudoknot and trans-
among structured RNA—although a few examples have fers a phosphoryl group from GTP (or a thiophosphoryl
beennotedamongcatalyticDNAs.ADNAzymehasbeen groupfromGTPgS)ontoitselfattwodifferentsitesinthe
describedthatinsertsCu2+andothertransitionmetalions primary sequence (Figure 1A) (53). The original selection
into a mesoporphyrin ring (18,19), and DNAzymes have and functional analysis of ribozyme K28 and its deriva-
been identified that have strict dependence on Cu2+ for tives were performed in the presence of alkaline earth
their catalysis of DNA ligation (20), DNA cleavage metal ions (Mg2+ and Ca2+), transition metal ions
(21,22) and DNA-capping activities (23). Several (Mn2+and Cu2+) and monovalent cations (K+and Na+)
50-self-phosphorylating DNAzymes have been shown to and was buffered to near neutrality with hydroxyethyl-
use Cu2+ either as one among many functional cations piperazine ethane sulphonate (HEPES, pH 7.5) (52). In
[Dk5, (24)] or as a strict requirement [Cu1, Cu4 and Cu7 the present work, we sought to determine the contribu-
(25)].Divalentionsarenotalwaysrequired,however,and tions of each of these components to RNA-catalyzed
the hairpin (HP), HH and Varkud Satellite (VS) ribo- phosphoryl transfer. We find that both outer-sphere and
zymes remain functional in the absence of divalent inner-sphere interactions with hydrated Mg2+ play im-
metals at high concentrations of monovalent ions, portant structural roles. More surprisingly, K28(1-77)C
relying exclusively on nucleotide functional groups for is also the first kinase ribozyme to be fully dependent on
catalytic chemistry (26–30). Cu2+for optimal activity, and the first for which the rate
Proton transfer is integral to the mechanisms of ribo- exhibits a log-linear pH dependence. Other ribozymes
zymes such as the hepatitis delta virus (HDV), HP, HH, from the selection that gave rise to K28 and its derivative
glmS, VS and others (31–37). In the HDV ribozyme, for K28(1-77)CdidnotshowthisdependenceonCu2+.In-line
example, a bound metal ion hydroxide is in position to probing(ILP)andcompetitionstudieswithadenosinetri-
abstract a proton from the 20OH nucleophile. The active phosphate(ATP)andATPgSshownoevidenceofadirect
site cytosine C75 is ‘histidine-like’, in that its apparent Cu2+-RNA complex and instead reveal that Cu2+enters
pKa is perturbed approximately three pH units from its into the ribozyme as part of a Cu2+(cid:2)GTP chelation
normal value of 4.2 to neutrality (32), and its N3 is in complex, with additional stabilization likely to come
position to stabilize the leaving group by donating a from interactions with nucleobase nitrogens in the RNA.
protontotheribose50OHofG1(32–34).Forself-cleavage TheproposedbindingmodeexplainstheCu2+dependence
by the HP ribozyme, the apparent pKa is (cid:3)6.0, and the and positions the bound Cu2+to participate in the cata-
deprotonated state of residue A38 has been shown to be lytic step of the reaction by serving as a Lewis acid
important in catalysis (27,38,39). Intriguingly, in molecu- catalyst.
lardynamicssimulations, theN1ofA38movesinto close
proximity of the active site 20OH, where it would be in
positiontoactasgeneralbase(40).Fortertiary-stabilized MATERIALS AND METHODS
HH ribozymes, X-ray crystallography show both G8 and Materials
G12 near the scissile bond, with G12 in position to act as
Oligodeoxynucleotides were purchased from Integrated
general base (36,41), consistent with earlier predictions
DNA Technologies (Coralville, IA). RNA was transcribed
based on pH rate profiles for site-specifically substituted
in vitro using phage T7 RNA polymerase, which was
ribozymes (35,42). The potential role of G12 as general
overproduced in bacteria and purified in the laboratory.
baseisfurthersupportedbylabellingofitsN1vianucleo-
philic attack on 20-bromoacetamide (43). Nucleobase- GTPgS and ATPgS were purchased from Sigma (St.
Louis). Radiolabelled nucleotides for labelling internal
metal ion interactions also appear to contribute to
phosphates or internal 20OH positions ([a32P]-CTP and
shifting nucleobase pKa values within the active sites of
[g32P]-GTP, respectively) were purchased from Perkin-
some extended HH ribozymes (9,44), and a suitably pos-
itioned Mn2+ ion has been observed in HH ribozyme Elmer (Waltham, MA). N-acryloyl-aminophenylmercuric
(APM) chloride was prepared as described (53–55).
crystals (45). For the glmS ribozyme, an active-site
guanine is in position to extract a proton from the 20OH
Self-thiokinase reactions
nucleophile (46–48), although it appears from studies of
pHdependenceofthereactiontoparticipateincatalysisin Ribozyme kinetic analyses were carried out essentially
its neutral, protonated state (49), perhaps by donating a as described (53), in most cases using tri-layer
NucleicAcidsResearch,2013,Vol.41,No.5 3329
A 0.5 to 1mM and moving the reaction mixtures to 32(cid:4)C.
Except where noted, reactions were quenched at various
timesinstopbuffer(95%formamide,15mMethylenedia-
minetetraacetic acid and trace amounts of xylene cyanol
and bromophenol blue as tracking dyes). ‘Zero’ time
points were collected immediately after all components
werepresentin thereactionmix.Products wereseparated
on 8% denaturing tri-layered organomercurial gels in
which the middle layer contained 100mg/mL APM
[(N-acryloylamino)phenyl]mercuric chloride (52,53,56).
Autoradiographs were obtained with a FLA-5000
phosphorimager (FujiFilm) and analysed with
MultiGaugesoftware.ThefractionoftheRNAconverted
toproductatagiventime[f(t)]wascalculatedbydividing
B theintensitiesofRNAretainedattheAPMinterfaceinto
the sum of all bands within a given lane. First-order ex-
ponential rate constants (k ) and extrapolated plateaus
obs
values(f )ofmostreactionswereobtainedbyfittingthe
max
data to a first-order rate equation using KaleidaGraph
(Synergy Software): f(t)=(f )[1(cid:6)exp(-kobst)]. Data
max
points in figure 5 are the mean values of at least two rep-
licas for concentrations in which activity was detectable.
Datapointsfortheextrapolation ofk inFigures 1and
obs
5 were the mean values of at least three replicas for con-
centrationsinwhichactivitywasdetectableandwerefitto
a linear equation.
Self-32P-kinase reactions for Cu2+/sulphur dependence
Ribozyme K28(1-77)C was transcribed without radiolabel,
C gel purified, unfolded in water at 85(cid:4)C for 5min, then
refolded on ice for 5min by addition of 5(cid:5) SP buffer or
5(cid:5) SP buffer lacking Cu2+. Kinase reactions used 1mM
RNA and were initiated by adding [g-32P]GTP to a final
concentrationof0.6mMandmovingthereactionsmixtures
to 32(cid:4)C. After the indicated times (0, 0.5, 1,2, 3,and 9h),
samplesweremoved toice and ethanolprecipitatedimme-
Figure 1. Cu2+ dependence of SP. (A) Secondary structure of diately by addition of NaOAc, glycogen and ethanol to
K28(1-77)C. Circled sP indicate the two self(thio)phosphorylation remove excess non-incorporated radiolabelled GTP. After
sites; nucleotide G1 represents the major site (52). (B) Thiophosphoryl
transfer fromGTPgS wasmonitored while titrating Cu2+from0.01 to resuspension in 1(cid:5) gel loading buffer, samples were
2mM(0,0.01,0.1,0.5,1,2,5,10,50,100mM;fulldatasetto100mMis analysed on 8%, 8M urea denaturing polyacrylamide gels
shown in inset.) The ‘zero-Cu2+’ data point was omitted from the plot as aforementioned.
to allow the log plot. Maximal rate at saturation was 0.00151±
0.00019min(cid:6)1. Fitting the initial rates to the quadratic form of a Self-thiokinase reactions with competitors
standard 2-state binding equation for a 1:1 complex yielded an
apparent dissociation constant for Cu2+ (KdCu2+) of 0.9±0.4mM. K28(1-77)C was unfolded and refolded as aforemen-
(C)SPreactions(nosulphur)wereperformedwith[g-32P]GTPasphos- tioned.ATPorATPgSwasaddedtoafinalconcentration
phoryl donor in SP buffer in the absence or presence of 10mM Cu2+.
of 2mM and kept in ice for 5min. Donor GTPgS was
Reaction times in hours are given above the lanes.
added to final concentration of 0.5mM, followed where
indicatedbyadditionofsupplementalCu2+toafinalcon-
organomercurial gels to analyse product formation by in- centration of 30mM.
ternally radiolabelled transcripts (52,53,56). In brief, in-
ILP analysis
ternally radiolabelled K28(1-77)C RNA and other RNA
moleculesweregelpurified,unfoldedinwaterat85(cid:4)Cfor Ribozyme K28(1-77)C was 50-end labelled using
5min, then refolded on ice for 5min by addition of the [g-32P]ATP, unfolded at 85(cid:4)C for 5min, and then
same 5(cid:5) self-phosphorylation (SP) buffer used originally refolded by addition of SP buffer with or without CuCl ,
2
toselectthisribozymeinvitro(52)(1(cid:5)SPbuffer=6mM and modified to contain 25mM TRIS (2-Amino-
MgCl , 0.2mM CaCl , 0.5mM MnCl , 10mM CuCl , 2-hydroxymethyl-propane-1,3-diol) (pH 8.0) in place of
2 2 2 2
200mM KCl, 15mM NaCl, 25mM HEPES, pH 7.5) or 25mMHEPES(pH7.5).Competitors,donorandsupple-
modifications of this buffer as detailed in the text. Kinase mental Cu2+were added in the same order as described
reactionsused1mMRNA(50000–200000cpm)andwere earlier in the text. ILP reactions were carried out at room
initiated by adding GTPgS to a final concentration of temperature and stopped after 10h by addition of half
3330 NucleicAcidsResearch,2013,Vol.41,No.5
a volume of 95% formamide and 50mM ethylene- non-radiolabelled K28(1-77)C RNA was incubated with
diaminetetraacetic acid. Digestion products were then [g-32P]GTP in SP buffer with and without Cu2+.
separatedonan8%denaturingpolyacrylamidegelelectro- AccumulationofradioactivityintheRNAunderthesecon-
phoresis gel, and autoradiographs were analysed as ditions indicates sulphur-independent SP by K28(1-77)C.
aforementioned. As with the GTPgS donor, product formation from the
GTP donor occurred much more vigorously when Cu2+
Self-thiokinase reactions for pH dependence was present, even though there is no sulphur in the GTP
donor(Figure1C).Therefore,theobservedrequirementfor
SP buffer was modified as follows: 25mM [2-(N-
Cu2+inbothGTP-dependentand GTPgS-dependentreac-
morpholino)ethanesulphonic acid] (MES) was used in
tionsindicatesaspecificfunctionofthemetalionandnota
place of HEPES for pH 5.5, 6.0. 6.5, 6.8; HEPES was
spurious consequence of having used GTPgS during the
used for pH 6.8, 7.0, 7.2, 7.5, 7.8, 8.0; and TRIS was
initial selection. Therefore, all further reactions included
used at pH 8.0, 8.2, 8.5, 8.8, 9.0, 9.5. Owing to precipita-
10mM Cu2+unless otherwise noted.
tion of manganese at elevated pH, this cation was
eliminated in all SP buffers used in the pH study.
Uniqueness of the Cu2+dependence of kinase
Control reactions performed over the pH range 5.5–8.0
ribozyme K28
in HEPES verified that activity remains the same with
and without Mn2+ (data not shown), thereby validating We next sought to determine whether other divalent ions
this approach. Very low product accumulation at the could substitute for Cu2+ and whether other ribozymes
lowest pH values (MES buffer) precluded fitting to a from the same selection as K28 displayed a similar de-
first-order exponential rate equation. Rates for these low pendence on Cu2+. Little or no product was observed in
pHvalueswereobtainedfromtheslopeofthelineofbest overnight reactions when 10mM Cu2+was replaced with
fit through these data, adjusted to an assumed plateau of this same concentration of transition metal ions (Cr3+,
66% (plateau values for the HEPES and TRIS data are Co2+, Ni2+, Cd2+, Mn2+ and Zn2+) or alkaline earth
66±7% conversion to product). The observed rate con- metals (Mg2+, Ca2+, Sr2+ and Ba2+), in the presence of
stants were then fit to a standard equation for a the other buffer components (Figure 2, data not shown).
one-proton transfer: kobs=kmax/[1+10(pKa-pH)]. Kinetic These other metal ions are therefore poor substitutes for
reactions were performed at least in duplicate. 10mM Cu2+. Because Cu2+binds phosphate more tightly
than do these other metal ions (5,58), we next monitored
product yields as a function of metal ion concentration.
RESULTS
There is a modest increase in product formation when
A Cu2+-dependent kinase ribozyme 10mM Cu2+ is replaced with at 100–500mM Mn2+ or
Ni2+, but strong inhibition is observed for Zn2+, Co2+
Ribozyme K28, which is the 126-nt parent form of the
and potentially Cd2+ at (cid:7)100mM concentrations,
58-nt ribozyme K28(1-77)C, was originally selected for
perhaps because of non-specific binding to the RNA.
thiophosphoryl transfer activity in SP buffer, which
Therefore, the role played by Cu2+ cannot be fully
contains monovalent ions (200mM K+ and 15mM
provided by these other divalent metal ions, which can
Na+), divalent alkaline earth ions (6mM Mg2+ and
only support sub-optimal activity (at best) even at 10- to
0.2mM Ca2+) and divalent transition metal ions
100-fold higher concentrations.
(0.5mM Mn2+and 10mM Cu2+). To begin to understand
The in vitro selection that gave rise to kinase ribozyme
the contributions of each of these ions to thiophosphoryl
K28 also produced several other ribozymes that could be
transfer activity by ribozyme K28(1-77)C, reactions were
classified into at least three different families based on
carried out in SP buffer in which individual components
their secondary structures (52). Because all of these ribo-
were omitted. Surprisingly, the ribozyme was essentially
zymes were selected under identical ionic conditions, two
inactive without Cu2+. When product formation was
ribozymes from each structural family were tested for
monitored, as Cu2+was titrated from 0.01 to 100mM in
Cu2+ dependence by performing the self-thiphosphory-
the presence of all of the other buffer components, initial
lation reaction in the absence and presence of 10mM
rates increased quickly at sub-micromolar concentrations
Cu2+. Although ribozymes K28(1-77)C and K28 showed
andreachedaplateauneartheconcentrationoftheRNA
theexpecteddependenceonCu2+,allsixoftheotherribo-
(1mM),withlittleadditionalchangeatstillhigherconcen-
zymes produced at least as much product in the Cu2+-
trations (Figure 1B). Fitting the initial rates to the quad-
depleted SP buffer as they did in the complete SP buffer,
ratic form of a standard two-state binding equation
and K5 and K6 yielded slightly more product when Cu2+
yielded a good fit for a 1:1 stoichiometry and an
apparent dissociation constant for Cu2+ (KdCu2+) of wasomitted(Figure3).Cu2+dependenceisthereforenota
widespread characteristic among the ribozymes that were
0.9±0.4mM, which was close to the concentration of
selectedinthepresenceofCu2+andisuniqueforK28and
RNA used in the assay.
Cu2+ is considered to be ‘borderline’ on the Pearson its derivatives.
scale of hard and soft metal ions, and it associates
ILP identifies RNA binding by GTP and GTP-Cu2+, but
well with ‘soft’ ligands such as nitrogen and sulphur (57).
not by Cu2+alone
To determine whether the observed Cu2+ requirement
is the result of essential interactions between the Cu2+ ILP assays were used to determine whether ribozyme
and the thiophosphoryl group of the GTPgS donor, K28(1-77)C uses Cu2+ to fold into its active structure.
NucleicAcidsResearch,2013,Vol.41,No.5 3331
A B
80 80
edPercent Product Form 712345600000000 10µM 0 10 M1n002+ 500 1000 medPercent Product For 123456700000000 10µM 0 10 1N00i2+ 500 1000
Cu2+ Cu2+
C D
80
ercent Product Formed 45678123000000000 Zn2+ oduct FormedPercent Pr 456712300000000 Cd2+
P 1100 µµMM No0 1100 110000 550000 11000000 1100µ µMM N0o 1100 110000 550000 11000000
E CoCpupe2+rCopper CCopup2+erCopper
80
ed 70
m
For 5600 Co2+
uct 40
d
o 30
Pr
nt 20
e 10
c
re 0
P
1100 µµMM N0o 1100 110000 550000 11000000
CoCpup2e+rCopper
Figure 2. Substitutions for Cu2+reduce or eliminate activity. Product yields were plotted for 18h self-kinase reactions that were carried out in SP
buffersthatweremodifiedbyreplacing10mMCuCl withMnCl,NiCl ,ZnCl ,CdCl orCoCl attheindicatedconcentrations(greybars).Yield
2 2 2 2 2 2
from a parallel reaction carried out in normalSP buffer (‘10mM Cu2+’) isplotted for comparison (black). Average values for four sets of reaction
(except for Cd2+, for which N=2) are plotted. Error bars represent the standard error of the mean.
ILP gives information on conformationally mobile substrateanalogs(52).GTP,30dGTPandafewotherana-
phosphoester bonds that increase the fraction of time in logues were recognized by the RNA and occupied the
which the 20OH samples conformations appropriate for active site in a manner that prevented GTPgS use. ATP
in-line attack on the adjacent phosphate and release of exhibited no such competition and is therefore not
the downstream 50OH (59). No new ILP cleavage was believed to interact with the ribozyme. Consistent with
observed when Cu2+was added to the reaction buffer at this interpretation, the intensity of ILP cleavage at A32
a concentration of either 10 or 30mM (Figure 4A, lanes inthepresenceof2mMATPand0.5mMGTPwasiden-
1–3). In contrast, addition of 0.5mM GTP induced a tical to those observed in reactions without ATP, and
strong cleavage after A32, and this cleavage was further addition of excess Cu2+had no effect on the cleavage in-
sensitized by addition of Cu2+(Figure 4A lanes 4–6). No tensity (Figure 4A, lanes 7 and 8). However, we reasoned
changesareseenatotherpositionsinresponsetoaddition that a different mode of competition could result from
of Cu2+ or GTP. These data support direct binding of sequestration of free Cu2+by the triphosphates of other
GTP in the absence of Cu2+ and enhanced binding of nucleotides, thereby reducing both GTP-induced ILP
GTP when Cu2+ is present, but they do not provide cleavage and self-thiophosphorylation yield from the
evidence of a direct interaction between RNA and Cu2+ GTPgS donor. Furthermore, although Cu2+dependence
in the absence of GTP. Instead, we interpret the stimula-
doesnotrequirethesulphurmoietyofGTPgS(Figure2),
tion as indicating that Cu2+ binds the triphosphate of
the presence of the sulphur can be safely assumed to
GTP in solution, and that the metal-bound form of the increase affinity of the NTP for Cu2+. Indeed, when ILP
GTP makes additional RNA-Cu2+contacts.
was carried out in the presence of 2mM ATPgS, the
cleavage signal at A32 owing to GTP was diminished.
Cu2+interacts with the triphosphate region of the
Signal strength was partially restoredon supplementation
phosphoryl donor with additional Cu2+ (Figure 4A, lanes 9 and 10).
Previous studies measured competition for access to the Quantification of signal strength at position A32
ribozyme active site between GTPgS and a series of shows that ATPgS reduced total cleavage by (cid:3)4.6-fold
3332 NucleicAcidsResearch,2013,Vol.41,No.5
A(cid:129)G YG(cid:129)A G(cid:129)AAAAA UAGG(cid:129)GA(cid:129)AA
(cid:129)G NAG Ac/aGCGU AUCCAA
8800
d 70
e
m
r 60
o
F
t 50
c
u
d 40
o
rPP 3300
t
n
e 20
c
r
e 10
P
0
28CK28 K5 K6 K11 K20 K22 K37
Figure 3. Stimulation by Cu2+is unique to ribozyme K28. Percentage
of the indicated ribozymes that were converted to
self-thiophosphorylated product is shown for 18-h reaction that was
performed in the presence (black) or absence (grey) of 10mM Cu2+.
The secondary structural features shared between each pair of ribo-
zymes are indicated above the graph [detailed in ref. (52)]. Asterisks
indicate sites of self-thiophosphorylation; ‘28C’ indicates ribozyme
K28(1-77)C.
(Figure 4B). When GTP was omitted from the ILP
reaction, neither ATP nor ATPgS induced RNA cleavage
at A32, irrespective of Cu2+ concentration (Figure 4A,
lanes 11–14). These results rule out the possibility of
direct RNA–ATP or RNA–ATPgS interactions and
suggest that competition by ATPgS is indirect, as would
beexpectedfromsequestrationoffreeCu2+byATPgS.In
thepreviousanaloguestudies,GMPcompetedonlyweakly
with GTPgS, indicating that the beta and gamma phos-
phates are important for efficient recognition. ILP reac-
tionsusing GMP inplace of GTP gavea Cu2+-dependent
cleavage signal at A32 that was much weaker than the
GTP-dependent signal, along with faint bands at C39,
A46 and C47, all of which were suppressed by ATPgS
(SupplementaryFigureS1).
The effect of ATPgS on the ribozyme’s ability to
interact with GTPgS was also examined by monitoring
self-thiophosphorylation. Ribozyme K28(1-77)C was
incubated overnight with 0.5mM GTPgS and 2.0mM
ATP or ATPgS. As observed previously (52), excess
ATP did not compete and had no effect on reaction Figure 4. Triphosphate interactions revealed by ILP. ILP analysis was
used to determine Cu2+-induced structural changes. (A) ILP reactions
yield (Figure 4C). In contrast, ATPgS reduced the
were performed for 10h in SP buffer modified to contain 25mM TRIS
observed self-thiophophorylated product by (cid:3)3.8-fold,
(pH8.0).ATPorATPgSwasaddedbeforeGTPwhenbothnucleotides
and the addition of supplemental Cu2+restored the yield were present. For lanes 3, 6, 8 and 10, Cu2+concentration was initially
of thiophosphorylated product. These data establish that 10mM and was adjusted to 30mM after addition of ATP/ATPgS and
Cu2+ does interact with the triphosphate region of the GTP. (B) Integrated ILP cleavage signal at position A32 under the
indicatedconditions.(C)Chelation-mediatedcompetitioninself-thiopho-
nucleotide donor, and that the effects of ATPgS on ILP
sphorylationreactionswasevaluatedbyquantifyingproductformedafter
cleavage and thiophosphoryl transfer is due to chelation 18h reactions under the indicated conditions. As aforementioned, com-
of Cu2+by the thiophosphate of ATPgS. petitorswereaddedbeforeadditionofGTP,andsupplementedCu2+was
added after the additionof competitors and donor.
Non-specific Mg2+interactions
Apart from Cu2+, the major divalent ion present during K28(1-77)C, as Mg2+ concentration was titrated from
theinitial selection was Mg2+.Therefore, all divalent ions 0.01 to 15mM. Thiophosphoryl transfer rates increased
except for Cu2+ were omitted from SP buffer, and linearly from 1 to 15mM Mg2+, with no indication of
self-thiophosphorylation was monitored for ribozyme approaching saturation over this range (Figure 5A).
NucleicAcidsResearch,2013,Vol.41,No.5 3333
A B
50
d + Copper
-30) 3.0 orme 40 -Copper
1 F
x ct 30
-1n 2.0 du
i(m oPro 20
kobs1.0 ent
rc 10
e
P
0 0
0 2 4 6 8 10 12 14 16 6 15 30 60 100
[Mg2+] (mM) [Mg2+] (mM)
D
C 60 Cobalt 60 Alkaline Earths Mg2+
med 50 Hexammine med 50 Metal Ions Ca2+
oduct For 4300 6 mMMg2+ 12 mM oduct For 4300 Sr2+
Prnt P 20 CCoHHex nt rPP 20
Perce 10 C6o HmeMx Perce 10 Ba2+
0 0
0 400 800 1200 1600 0 200 400 600
Time (min) Time (min)
Figure 5. Outer-sphereandinner-sphereMg2+interactionsforribozymeK28(1-77)C.(A)ReactionswerecarriedoutinthepresenceofvariousMg2+
concentrations(0,0.01,0.05,0.1,0.2,0.5,1,3,6,and15mM).Observedkineticconstants(k )areplottedversusMg2+concentrationforreactions
obs
in which product was detected. Reactions in which Mg2+was <1mM yielded insufficient product for accurate determination of k and were not
obs
plotted. (B) Accumulation of thiophosphorylated product in the presence of the indicated Mg2+concentrations after short (3h) reactions in the
presence(black)orabsence(grey)of10mMCu2+.(C)Toevaluatepotentialinner-sphereMg2+interactions,kinetictraceswereplottedforreactions
in which the 6mM Mg2+component of SP buffer was replaced with 6 or 12mM CoHex. Uncertainty reported for the rate constants reflects the
uncertaintiesofthefit.(D)SimilartopartC,exceptthatMg2+wasreplacedwithotheralkalineearthmetalions.Dashedlinesindicatelinearfitsto
these data.
TitrationwithMn2+yieldedasimilarpatternastheMg2+ black bars). Interestingly, product yield also increased to
titration to (cid:3)6 mM but was inhibitory at still higher con- wellabovebackgroundintheabsenceofCu2+atelevated
centrations (data not shown). Little or no product was Mg2+, although the yield reached a plateau at (cid:3)30mM
observed with <1mM Mg2+ or Mn2+, and no product Mg2+ and did not approach the yield observed in the
was detected when 6mM Co2+ or Ni2+ was used in presence of Cu2+(Figure 5B, grey bars). Thus, although
place of 6mM Mg2+ (data not shown). Thus, Mg2+ or Mg2+maypartiallysubstituteforCu2+athighconcentra-
Mn2+(or potentially other divalent ions) is required for tions (for example, by forming a Mg2+(cid:2)GTPgS complex),
phosphoryl transfer by K28(1-77)C via low-affinity inter- this partial rescue is only observed when Mg2+ is least
actions,butneitherdivalentcationisspecificallyrequired. 3000 times higher than the Cu2+concentration and >15
Naturalkinaseandpolymeraseenzymesmakeextensive times higher than the normal Mg2+concentration in SP
use of Mg2+(cid:2)NTP and Mg2+(cid:2)dNTP complexes [see buffer. It would therefore be premature to infer that the
(60–62) and references therein]. However, Cu2+ has a precise contribution of Mg2+ at high concentrations is
higher affinity for phosphates than do Mg2+ and other fully interchangeable with the role of 10mM Cu2+.
divalent metal ions [>30-fold higher for NMP-metal ion Overall, the observations from Mg2+ titrations suggest
complexes (5)]. Formation of Cu2+(cid:2)GTPgS is therefore that low-affinity, non-specifically bound Mg2+ions serve
expected to be strongly favoured over formation of primarily to aidfolding the ribozyme into its activestruc-
Mg2+(cid:2)GTPgS when both ions are present. To test ture, which can then use the Cu2+(cid:2)GTPgS complex for
whether Cu2+dependency at relatively low Mg2+concen- optimal catalysis.
trations could be overcome at higher Mg2+ concentra-
Evidence for inner sphere interactions with Mg2+
tions, self-thiophosphorylation yield for relatively short
reaction times (3h) was monitored at Mg2+ as high as Mg2+ can contribute to ribozyme activity by contacting
100mM in both the presence and absence of 10mM ribozymes and substrates directly (inner sphere), and via
Cu2+. As aforementioned, product yield in the presence ligandssuch as water that are bound directly to themetal
of Cu2+ continued to increase with Mg2+ concentration ion (outer sphere). Cobalt hexammine {[Co(NH ) ]3+,
3 6
with no sign of approaching saturation (Figure 5B, CoHex} is a reasonably good analogue for outer sphere
3334 NucleicAcidsResearch,2013,Vol.41,No.5
interactions with hydrated Mg2+owing to its similar size, (MES,HEPESandTRIS),includingseparatereactionsin
geometry and charge, but kinetic stability of the coord- MESorHEPES(pH6.8)andinHEPESorTRIS(pH8.0)
inated NH ligands precludes significant inner-sphere at the two overlapping pH values to control for
3
interactions. CoHex supports catalysis as well as, or buffer-specific effects. In contrast to previously analysed
nearly as well as, Mg2+ for ribozymes that do not kinase ribozymes (50,51) and DNAzymes (25), for which
require direct contact with the metal ion, such as the HP the rates showed no pH dependence, thiophosphoryl
ribozyme(26,27,63)andtheinvitroselectedATRib/AT02 transfer by K28(1-77)C was strongly dependent on pH
acyltransferase ribozymes (64,65). In contrast, CoHex (Figure 6). The observed first-order rate constants (k )
obs
does not support catalysis at all, or not nearly as well as increased (cid:3)1000-fold from pH 5.5 to 8.5, indicating one
Mg2+, for ribozymes that exploit direct contacts with the proton-transfer event in the rate-limiting step. Fitting
metal ion, such as the HH (44,66), VS (67), RNaseP (68), these data to a single-proton equation (see ‘Materials
HDV (69,70) ribozymes and the in vitro selected 2PT3.2 and Methods’ section) yielded an apparent pKa value of
kinase ribozyme (51), and CoHex even induces aberrant 7.99±0.08, and a maximal rate constant
folding of RNaseP (15). (0.049±0.002min(cid:6)1) that is >15-fold higher than that
When the 6mM Mg2+component of SP was replaced observed in normal SP buffer. The rate constants for re-
with 6mM CoHex, self-thiophosphoryl transfer was actions carried out in HEPES were 3–5-fold lower than
greatly reduced ((cid:3)5% yield after 16h) (Figure 5C) those for reactions carried out in either MES or TRIS at
relative to Mg2+at the same concentration, supporting a the two overlapping pH values, and the overall increase
potential role for at least one inner-sphere contact with with pH was less steep in the presence of HEPES relative
Mg2+. Doubling the CoHex concentration increased to MES or TRIS, indicating buffer-specific effects that
yield after 16h (18%), but the calculated kobs values at make HEPES a suboptimal buffer for ribozyme
these two CoHex concentrations are similar (0.0034± K28(1-77)C.
0.0006 and 0.0026±0.0003min(cid:6)1, respectively). Increas-
ing CoHex therefore stimulates the amount of RNA that
can self-thiophosphorylate without altering the intrinsic
DISCUSSION
rate. Importantly, the k values measured in CoHex/
obs
Cu2+are not significantly different from those measured We demonstrate here that phosphoryl transfer by the
inMg2+/Cu2+(0.0032±0.0001min(cid:6)1),eventhoughmuch K28(1-77)C kinase ribozyme strongly depends on the
more product is formed in Mg2+/Cu2+. Thus, Mg2+plays presence of Cu2+, that the reaction also requires Mg2+
a role in forming the active structure of ribozyme or similar divalent cations and that the rate-limiting step
K28(1-77)C but does not play a catalytic role in ofthereactionincludesaprotontransferevent.Verylittle
thiophosphoryl transfer, even for the Mg2+ ions that product was formed in the absence of Cu2+, independent
appear to make inner sphere contacts. of whether the donor was GTPgS or [g-32P]GTP, estab-
lishing that the Cu2+requirement persists apart from any
Probing the dimensions of the Mg2+-binding site(s) potential interaction with the donor sulphur. Fitting the
titration data to a binding isotherm indicated a 1:1 stoi-
To determine whether divalent alkaline earth ions larger
chiometry, with KdCu2+ of 0.9±0.4mM. However, ILP
than Mg2+ could also support catalysis, thiophosphoryl
analysis argues against interpreting these data in terms
transfer was monitored for reactions in which Ca2+, Sr2+
of a simple direct metal ion-RNA interaction, as there
orBa2+wastheonlydivalentionapartfrom10mMCu2+.
Replacing 6mM Mg2+with 6mM Ca2+or Sr2+produced
amodestreductioninoverallrateandyieldduringthe11-
-1
hreaction,andmuchlessproductformedin6mMBa2+as
the major divalent cation (Figure 5D). Interestingly, the
rate of product accumulation for these ions is approxi-
mately linear rather than fitting well to a first-order
kinetic best-fit curve, potentially indicating a shift in the -1n)] -2
rate-limiting step (e.g. slow folding). These three trends mi
(reductions in rate and yield, and deviation from simple (
bs
first-order kinetics) all correlate with the ionic radius of o
k
these alkaline earth metal ions. Specifically, thiopho- [g -3
o
sphoryl transfer activity decreased in the order L
Mg2+>Ca2+>Sr2+(cid:8)Ba2+, for which the ionic radii of
their hexacoordinate forms are 0.86A˚ , 1.14A˚ , 1.32A˚
and 1.49A˚ , respectively (71).
-4
pH-dependence 5 6 7 8 9 10
pH
To determine the potential role of proton transfer in the
Figure 6. pH rate profile of K28(1-77)C. k values are plotted as a
reactivity of ribozyme K28(1-77)C, apparent rate con- obs
function of pH. Because HEPES (triangles) exhibited sub-optimal,
stants were measured, as pH was varied from pH 5.5 to
buffer-specific effects, only the data from reaction carried out in
9.5. Three different buffers were used to span this range MES (circles) and TRIS (squares) were used in determining the fit.
NucleicAcidsResearch,2013,Vol.41,No.5 3335
b o z y Co2+,Ni2+,Cu2+,Cd2+andZn2+wasinitiallyreportedas
i m
R e requiringCu2+(72,73).TheyinterpretedtheCu2+require-
mentintermsofcatalyticCu2+Lewisacidsites,consistent
Cu2+ Mg2+ with Lewis acid-catalysed Diels–Alder reactions in water.
However,oneofthoseDiels–Alderaseribozymeswaslater
shown to function in the absence of Cu2+, given elevated
concentrations of Mg2+, thereby establishing that Cu2+
plays a non-essential role in that ribozyme (74).
Nevertheless, Lewis acid catalysis figures prominently in
ribozyme reactions related to phosphoryl transfer. For
example, a recent crystal structure of the HDV ribozyme
shows inner-sphere ligand interactions between an active
Figure 7. Schematic representation of metal ion interactions in siteMg2+ionandthepro-R(P)oxygenofthescissilephos-
ribozyme K28(1-77)C. Cu2+(blue) is proposed to serve as Lewis acid phate and the 20-hydroxyl nucleophile (75). Similarly,
catalyst by chelating both the nucleotide triphosphate and nucleobase
nitrogens within an azophilic binding pocket in the RNA, while Mg2+ polymerase and pyrophosphatase enzymes (which
(red-orange) contributes to structure through both inner sphere and formally catalyse phosphoryl transfer to a polynucleotide
outer sphere interactions. Postulated RNA-GTP interactions through chain or to water) use two Mg2+ions as Lewis acid cata-
the guanine Watson-Crick face, 30OH and phosphates are based on
lysts to shield charges and lower the pKa of the leaving
previous analysis using analog competition (52).
group (62).
The reduction in overall thiophosphoryl transfer
activity by ribozyme K28(1-77)C on replacing Mg2+with
are no new RNA cleavages associated with addition of
the non-exchanging CoHex indicates that the ribozyme
Cu2+ alone. Instead Cu2+ stimulates GTP-dependent takes advantage of inner sphere interactions with Mg2+;
RNA cleavage and partially alleviates the inhibitory however,theprimaryroleofMg2+istohelpestablishthe
effects of competitor ATPgS on RNA cleavage during
active structure, as evidenced by the insensitivity of the
ILPandonproductformationduringself-thiophosphory-
first-order rate constants to the CoHex substitution.
lation reactions. All of these observations suggest that Larger alkaline earth metal ions also promoted
Cu2+ enters the ribozyme as a Cu2+(cid:2)GTPgS chelation self-thiophosphorylation, but they did so less well and
complex, wherein it is additionally chelated by a induced an intriguing and increasingly prominent devi-
nitrogen-rich site in the RNA (Figure 7). Cu2+is highly ation from first-order kinetic behaviour. The correlation
azaphilicandpronetoJahn-Tellergeometrydistortionsin of these trends with ionic radius may reflect simple steric
certain binding environments. As such, it may be particu- constraints (upper size limit on ions that can successfully
larly well suited to bind within a nitrogenous pocket that replace Mg2+in at least one specific binding site), or they
is unable to form strong interactions with other metal may arise from chemical features of the ions that also
ions. Such interactions would ideally position the metal correlatewithionicradius,suchasincreasedpolarizability
ion to promote the chemical step of the reaction by with- or reduced charge density.
drawing electrons from the phosphate (Lewis acid cataly- In addition to its unusual metal ion requirements,
sis), making it more susceptible to nucleophilic attack by K28(1-77)C is also the first kinase ribozyme to demon-
the acceptor oxygen. In addition, Cu2+binds phosphate strate a dependence on pH. Prior work with ribozymes
more tightly than do the other metal ions assayed here Kin.46 (50) and 2PT3.2min (51) observed no change in
(5,58), and phosphodiester cleavage rates by HH ribo- reaction rates, as pH were varied over several pH units.
zymes in various metal ions correlate with binding These observations suggested that deprotonation and ac-
affinities of those metal ions with phosphate monoesters tivation of the 20OH acceptor nucleophile was not rate
(5). However, phosphate affinity alone does not explain limiting for those ribozymes and potentially suggested
metal ion selectivity, as these other metals cannot fully an S 1-like dissociative mechanism in which release of
N
substitute for Cu2+even at 100-fold (for transition metal the gamma (thio)phosphate from the donor precedes for-
ions) to >1000-fold (for Mg2+) higher concentrations. mation of the acceptor-phosphate bond (50). The
Partial rescue in the absence of Cu2+at very high Mg2+ log-linear increase in the pH rate profiles observed here
concentrations suggests that the Mg2+(cid:2)GTPgS complex indicatesthatdeprotonationofaspecieswithpKaof7.99
can also be used. However, the yield of thiophosphory- ± 0.08 is required for maximal thiophosphoryl transfer
latedproductinthepresenceofCu2+continuestoincrease activity.Theribose20OHatthetwomodificationsitesare
well above the Mg2+concentration where the yield satur- logical candidates for deprotonation to increase their
ates in the absence of Cu2+, indicating that the nucleophilicity for attack on the gamma phosphate.
Mg2+(cid:2)GTPgS is not used by the ribozyme as well as the However, the pKa for a ribose 20OH is in the range of
Cu2+(cid:2)GTPgS complex. We therefore propose that Cu2+ 12.2–13.7 (52,76,77)and would need to be perturbed four
ion selectivity results from a combination of coordination tofivepHunitstoaccountforthedatahere.ThepKafor
geometry, azaphilicity, geometric constraints of the active hydrated Mg2+ (11.4) is closer to the observed kinetic
site and affinity for phosphates. pKa, but the essentially unperturbed k in CoHex
obs
ApartfromK28(1-77)C,therearefewornootherCu2+- argues against a catalytic role for Mg2+. The pKa values
dependent ribozymes available for comparison. A Diels– for free guanosine (N1)H and uridine (N3)H are each
AlderaseribozymeselectedinthepresenceofMn2+,Fe2+, (cid:3)9.2 and could be shifted to near 8.0 within the context
3336 NucleicAcidsResearch,2013,Vol.41,No.5
of the ribozyme active site, for example, via metal ion and D.H.B. designed experiments, interpreted data and
interactions (9,44). Intriguingly, the pKa of Cu2+-bound wrote the manuscript. E.B., R.R.P, J.F., A.W.S. and
waterisnear7.5(71,78)andisexpectedtoshiftupwardin A.W.R.M. performed the experiments.
complexes with ligands that donate electrons to the metal
ioncentre.Deprotonationofoneofthewatersonabound
FUNDING
(inner sphere, partially dehydrated) catalytic copper
hydrate or Cu2+-nucleobase interactions that modulate Funding for open access charge: National Science
the pKa of the nucleobase would provide an intriguing Foundation Chemistry of Life Processes program [grant
functional linkage between the Cu2+ requirement and number CHE-1057506].
pH sensitivity of this kinase ribozyme.
Although HEPES is generally considered to be a Conflict of interest statement. None declared.
non-chelating buffer with respect to hard metal ions
such as Mg2+ (79), interactions between Cu2+ and
REFERENCES
HEPES have been observed by isothermal titration calor-
imetry (80), electron paramagnetic resonance spectros- 1.Roth,A. and Breaker,R.R. (1998) An amino acid as a cofactor
copy (81) and potentiometric titrations (81,82). for a catalytic polynucleotide. Proc. Natl Acad. Sci. USA, 95,
However, the suboptimal performance of HEPES buffer 6027–6031.
2.Tsukiji,S., Pattnaik,S. and Suga,H. (2003) An alcohol
isnotfullyexplainedbyasimplecompetitionbetweenthe
dehydrogenase ribozyme. Nat. Struct. Biol., 10, 713–717.
bufferandtheribozyme/donorforaccesstothemetalion; 3.Tsukiji,S., Pattnaik,S. and Suga,H. (2004) Reduction of an
TRISbufferformsTRIS(cid:2)Cu2+chelateswithaffinitiesthat aldehyde by a NADH/Zn2+-dependent redox active ribozyme.
are similar to those of the HEPES(cid:2)Cu2+complex (58,81) J. Am. Chem. Soc., 126, 5044–5045.
4.Gong,B., Klein,D., Ferre´-D’Amare´,A. and Carey,P. (2011) The
andyetiswellbehavedwithrespecttoitsinfluenceonthe
glmS ribozyme tunes the catalytically critical pK(a) of its
pHdependenceofthereactionrate.Buffer-specific effects
coenzyme glucosamine-6-phosphate. J. Am. Chem. Soc., 133,
were not further explored. 14188–14191.
Inconclusion,thechemicalmechanismoftheribozyme 5.Schnabl,J. and Sigel,R. (2010) Controlling ribozyme activity by
K28(1-77)C stands in sharp contrast with previously metal ions. Curr. Opin. Chem. Biol., 14, 269–275.
6.Irving,H. and Williams,R. (1953) The stability of transition-metal
characterized kinase ribozymes, using a tightly bound
complexes. J. Chem. Soc., 3192–3210.
Cu2+ion and also using a species that deprotonates with 7.Sigel,H., Hofstetter,F., Martin,R.B., Milburn,R.M., Scheller-
pKa&8.0, in addition to specifically bound (potentially Krattiger,V. and Schellerla,K.H. (1984) General considerations on
innersphere)andnon-specificcationsrequiredforfolding. transphosphorylations: mechanism of the metal ion facilitated
The contributions of Cu2+ to the catalytic mechanism dephosphorylation of nucleoside 50-triphosphates, including
promotion of atp dephosphorylation by addition of adenosine
include unique chemical features that cannot be 50-monophosphate. J. Am. Chem. Soc., 106, 7935–7946.
provided by other first-row transition metal ions and 8.Athavale,S., Petrov,A., Hsiao,C., Watkins,D., Prickett,C.,
that appear to implicate Lewis acid catalysis through Gossett,J., Lie,L., Bowman,J., O’Neill,E., Bernier,C. et al. (2012)
Cu2+chelation by the incoming donor triphosphate in a RNA folding and catalysis mediated by iron (II). PLoS One, 7,
Cu2+(cid:2)GTP or Cu2+(cid:2)GTPgS complex. The chemical e38024.
9.Roychowdhury-Saha,M. and Burke,D. (2006) Extraordinary
strategies that ribozymes K28 and its derivatives use to cleavage rates in transition metal ion mediated catalysis by a
accomplish (thio)phosphoryl transfer are clearly distinct tertiary stabilized hammerhead ribozyme. RNA, 12, 1846–1852.
from those used by other kinase ribozymes to perform 10.Horton,T., Clardy,D. and DeRose,V. (1998) Electron
paramagnetic resonance spectroscopic measurement of Mn2+
the same reaction, including those that were co-selected
binding affinities to the hammerhead ribozyme and correlation
along with K28 under identical ionic conditions. An im-
with cleavage activity. Biochemistry, 37, 18094–18101.
plication for RNA world theories is that the parallel evo- 11.Hunsicker,L. and DeRose,V. (2000) Activities and relative
lutionofstrikinglydiversemetalionandpHrequirements affinities of divalent metals in unmodified and
phosphorothioate-substituted hammerhead ribozymes. J. Inorg.
amongribozymeswithinasinglepopulationhighlightsthe
Biochem., 80, 271–281.
ability of RNA to evolve structures that capitalize on a
12.Long,D., LaRiviere,F. and Uhlenbeck,O. (1995) Divalent metal
diversesuiteofmechanismsandlocalresourcestoaccom- ions and the internal equilibrium of the hammhead ribozyme.
plish the same chemical objectives. Biochemistry, 34, 14435–14440.
13.Peracchi,A., Beigelman,L., Scott,E., Uhlenbeck,O. and
Herschlag,D. (1997) Involvement of a specific metal ion in the
transition of the hammerhead ribozyme to its catalytic
SUPPLEMENTARY DATA
conformation. J. Biol. Chem., 272, 26822–26826.
14.Markley,J.C., Godde,F. and Sigurdsson,S.T. (2001) Identification
Supplementary Data are available at NAR Online:
and characterization of a divalent metal ion-dependent cleavage
Supplementary Figure 1.
site in the hammerhead ribozyme. Biochemistry, 40, 13849–13856.
15.Cuzic,S. and Hartmann,R. (2005) Studies on Escherichia coli
RNase P RNA with Zn2+as the catalytic cofactor. Nucleic Acids
ACKNOWLEDGEMENTS Res., 33, 2464–2474.
16.Hati,S., Boles,A.R., Zaborske,J.M., Bergman,B., Posto,A.L. and
The authors thank Tim Glass, Frank Schmidt, George Burke,D.H. (2003) Nickel2+-mediated assembly of an RNA-amino
Smith, Stefan Sarafianos and members of the Burke la- acid complex. Chem. Biol., 10, 1129–1137.
17.Ciesiolka,J., Gorski,J. and Yarus,M. (1995) Selection of an RNA
boratory and MU Combinatorial Chemistry Club for
domain that binds Zn2+. RNA, 1, 538–550.
stimulating discussions, and Sanchita Hati and Tracey
18.Li,Y. and Sen,D. (1998) The modus operandi of a DNA enzyme:
Nevitt for comments on the manuscript. E.B., R.R.P. enhancement of substrate basicity. Chem. Biol., 5, 1–12.