Title: RNA splicing: group I intron crystal structures reveal the basis of splice site selection and metal
1RNA splicing group I intron crystal
structures reveal the basis of splice site
selection and metal ion catalysis Mary
R Stahley and Scott A Strobel Current
Opinion in Structural Biology 2006, 1618
Zhang Xiaorong
2??????
- ??1/3??????????????
- (1)??? ????????????????? ?Fe 3 Fe 2 Cu 2 Zn
2 Mn 2 Co 2 - (2)????? ?????????????,?????,????????????????????
?2??Na 1 K 1 Ca 2 Mg 2
3- The group I intron has served as a model for
RNA catalysis since its discovery 25 years ago.
Four recently determined high-resolution crystal
structures complement extensive biochemical
studies on this system. Structures of the
Azoarcus, Tetrahymena and bacteriophage Twort
group I introns mimic different states of the
splicing or ribozyme reaction pathway and provide
information on splice site selection and metal
ion catalysis.
4-
- The 5splice site is selected by formation of a
conserved GU wobble pair between the 5exon
terminus and the intron. - The 3splice site is identified through
stacking of three base triples, in which the
middle triple contains the conserved terminal
nucleotide of the intron, OG. -
5- The structures support a two-metal-ion
mechanism for group I intron splicing that might
have corollaries to group II intron and pre-mRNA
splicing by the spliceosome. -
- The two-metal-ion mechanism for catalytic RNA
was first elucidated by Steitz and Steitz in
1993 which published on PNAS
6Since a lot of protein phosphoryltransfer enzymes
used two-metal-ion mechanism and they does not
require that protein side chains of the enzyme
participate directly in the chemistry of the
reaction. Thus, the mechanism can be easily
accommodated by an enzyme composed of RNA.
By Steitz and Steitz in 1993
7contents
- Introduction
- Capturing reaction states
- Splice site selection
- Active site metals
- Metal ion catalysis in RNA splicing
- conclusions
8Introduction
- RNA enzymes, or ribozymes, catalyze a variety
of reactions including protein synthesis,
self-splicing, self cleavage and tRNA 5end
formation . RNAs active role in the cell was
first illuminated with the self-splicing activity
of the Tetrahymena thermophila pre-rRNA . - But only recently have these efforts yielded
molecular resolution structural information that
includes the complete - active site.
9- This review discusses the four group I intron
crystalstructures published in 2004 and 2005 - two group I intronexon complexes from
pre-tRNAIle in Azoarcussp. BH72 , a Tetrahymena
pre-rRNA apo enzyme and a bacteriophage Twort
ribozyme-product complex from a pre-mRNA .
10- The focus of this review is the structural
basis of splice site selection and metal ion
catalysis as observed in these crystal
structures. The metal binding pocket of the group
I intron active site will be compared to RNA,
protein and ribonucleoprotein complexes that
perform similar phosphoryl transfer reactions.
11Capturing reaction states
- Group I introns self-splice via two
consecutive phosphoryl transfer reactions In the
presence of Mg2, a bound exogenous G(aG)
initiates splicing with attack of its O3 on the
5splice site. A conformational change then
brings the terminal nucleotide of the intron,OG,
into the active site, displacing the aG and
positioning the 3exon for exon ligation.
12- aG attack its O3 on the 5splice site
- Conformation change displace aG with OG
- Exon ligation and intron release are
accomplished by the O3 of the 5exon attacking
the 3-splice site.
13- In this RNA, the 3 and 5ends of the intron
are removed ,and an aG cleaves, in trans, an
oligonucleotide at the 5splice site. The
reaction is analogous to the first step of
splicing -
14- In order to mimicking different states of the
reaction pathway, crystal of different intron
subclasses and different reaction states are
represented. In each case ,modifications or
deletions were made to promote crystallization
and capture a particular reaction intermediate. - (1) The first Azoarcus structure has four
2deoxy substitutions in the active site,
including a substitution at the last nucleotide
of the 5exon (U-1). Mimic pre-1S - (2)The second Azoarcus structure restores three
of the four riboses, including OG, but retains a
single U-1 substitution. Mimic pre-2S
15- (3)The Tetrahymena group I intron structure (at
3.8 A resolution) is comparable to the apo form
that lacks both exons and the complementary - sequence in the intron to which they bind
(the internal guide sequence, IGS), but it does
include the OG . Mimic the 1st step of splicing - (4)The bacteriophage Twort group I intron 50-exon
complex (at3.6 A resolution) is similar to the
product of the ribozyme reaction, although it
also has the OG. The Twort structure omits the
scissile phosphate, the 3exon and the portion of
the IGS complementary to the 3exon.mimic the
product.
16Azoarcus deoxy-wG pre-2S structure
17Mimic the first step
Tetrahymena apo enzyme structure
18Mimic product
Structure of Twort group I intron with P1 helix
19 OG
Group Iintron secondary structure with G binding
site
Superposition of three intron ribbon structure
Azoarcus (grey), Tetrahymena (gold) and Twort
(green)
20- Despite these differences in reaction state
and subclass, the four structures fold into a
single 3D structure within the conserved paired
and joining regions . Two key elements of the
group I intron self-splicing reaction can now be
structurally examined - How does the intron select the 5and 3splice
sites? - How does the RNA catalyze the splicing
reactions?
21Splice site selection
- Accurate splicing requires selecting two
specific phosphates among hundreds for cleavage
and ligation reactions. How group I introns
achieve this selection using secondary and
tertiary interactions is visualized in these
structures.
22Group I intron 5splice site selection is based
on recognition of a conserved wobble pair in the
substrate helix formed between the terminal
nucleotide of the 5exon,U-1, and a conserved G
within the introns IGS .The substrate helix
makes tertiary contacts to two conserved regions
within the active site, J4/5 and J8/7.
- 5Splice site selection. The GU wobble pair
is recognized by stacked As in the J4/5 wobble
receptor. - Coordinates from Azoarcus ribo-OG structure.
U-1 o2 is modeled (grey).
23The substrate helix makes tertiary contacts to
two conserved regions within the active site,
J4/5 and J8/7.
243Splice site selection is achieved through
specific binding of the intron s conserved
terminal nucleotide, OG, which forms a triple
with a GC pair in the P7 helix . All four group
I intron structures contain the OG and all
demonstrate how the P7 region folds to form the G
binding site The OG triple is stacked between two
additional P7 triples, which suggests that the
introns terminal nucleotide is selected through
hydrogen bonding and stacking interactions.
25- 3Splice site selection. Coordinates from
Tetrahymena structure.molecule D. O G forms a
GCG triple at the center of the G binding site.
Two additional triples stack above and below to
complete the G binding through stacking
interactions
G binding to the group I intron occurs in two
stages an initial misaligned binding, followed
by a conformational change placing the G in an
active orientation .
26Twort
Tetrahymena
Azoarcus
27a, Side view of the active site oriented to
emphasize the stacking interactions that mediate
QdG206 binding. The two active-site metal ions
are shown as black spheres. b, Top-down view of
the base triple between the unpaired A129 and the
closing G128C178 base pair in the P7 helix. c,
Top-down view of the base triple between OdG206
and the G130C177 base pair
28Active site metals
- Group I intron splicing requires divalent
metal ions for proper folding and catalysis . Six
active site oxygens have been biochemically
implicated for direct metal ion coordination
through metal specificity switch experiments
These assays substitute oxygens with soft
atoms, such as sulfur or nitrogen, which are less
able to coordinate Mg 2, a hard metal. If the
addition of a soft cation (such as Mn 2 or Cd
2) rescues interference due to the
substitution,the oxygen is considered a metal
ligand.
29 Group I intron splicing requires divalent
metal ions for proper folding and catalysis. Six
active site oxygens have been biochemically
implicated for direct metal ion coordination
30- Mg2 coordination to substrate oxygens
activates the nucleophile(M1), stabilizes the
scissile phosphate and stabilizes the developing
charge on the leaving group(M2). - The ribo-OG Azoarcus group I intron crystal
- structure includes all these ligands . This
structure shows two Mg2 ions bound in the active
site with all biochemically defined metal ligands
satisfied, including all those in a previously
proposed three-metal-ion model -
315exon(U1) 3exon OG Mg2 ions K
This structure shows two Mg2 ions bound in the
active site with all biochemicallydefined metal
ligands satisfied, including all those in a
previously proposed three-metal-ion model.
32M1,which coordinates the nucleophile of the
second step, is equivalent to the biochemically
termed MA. M2 has the coordination of MB and MC
combined, as it includescoordination to both the
O3 and O2 of the OG
metal ion interactions with three substrate
atoms, the 3-oxygen of the oligonucleotide
substrate and the 3- and 2-moieties of the
guanosine nucleophile, are mediated by three
distinct metal ions.(Herschlag)
33This structure shows two Mg2 ions (referred to
as M1 and M2) bound in the active site with all
biochemically defined metal ligands satisfied,
including all those in a previously proposed
three-metal-ion model. (ribo-OG Azoarcus)
34The 2-OH is a metal ion binding ligand. Removal
of the 2-OH at the terminal nucleotide, causes a
structural rearrangement that switches the metal
specificity from a Mg2 in the ribo-wG structure
to a K bound in the deoxy-wG structure
35Metal ion catalysis in RNA splicing
- Steitz and Steitz proposed that group I
introns , group II introns and the spliceosome
catalyze phosphoryl transfer reactions with the
same two-metal-ion mechanism seen in protein
polymerases and phosphatases. The ribo-OG
Azoarcus group I intron crystal structure
provides the first structural evidence for this
model.
36 (a) Group I intron Azoarcus ribo-wG structure
and (b) T7 DNA polymerase.
The RNA intron uses its phosphate backbone to
form an active site metal binding pocket in the
same way that proteinaceous polymerases use
carboxylates of aspartate side chains.
37an open octahedral coordination position on the
leaving group stabilizing metal M2 for a
metal-coordinated water Although this water was
not visible at the resolution of the ribo-OG
structure, an intriguing possibility is that this
water could protonate the O3 OG leaving group
38- Group II introns and the spliceosome might
catalyze exon ligation using the same
two-metal-ion architecture observed in the group
I intron. Metal specificity switch experiments
have been performed on active site oxygens in all
three systems and the interference and metal
rescue data indicate analogous metal ion
coordination.
39Biochemically implicated catalytic metal ion
ligands in RNA-based splicing exon ligation.
Active site oxygens that show interference
(sulfur or amino substitution) followed by soft
metal rescue are shown in red .Oxygens that show
only interference are shown in orange .The ?
indicates a position that has not been S or N
substituted 41. Note that double sulfur
substitutions required for the scissile
phosphates pro-RP oxygen metal specificity
switch in the group I intron have not been
performed in the group II intron and spliceosome.
40Conclusions
- The four group I intron structures described
here have demonstrated the structural basis for
splice site selection and metal ion catalysis. A
two-metal-ion mechanism of group I intron
splicing is supported by these structures,
consistent with a large body of biochemical data.
Armed with this picture of RNA-catalyzed splicing
and biochemical analysis, similarities between
metal ion catalysis of exon ligation by the group
I intron, group II intron and - spliceosome are beginning to emerge.
41Thank you