THIS WEEK IN

R EPORTS
model is supported by evidence that Sir2p alters
chromatin structure within EXP (18). DSBs in
the IGS are expected to recruit cohesin (19),
countering the effect of transcription. However,
because not many DSBs form in the
rDNA repeats (20), this effect is likely to be
minor. Also, DSB formation at the RFB was
similar in the GAL1/10 strain when grown on
glucose/galactose (0.89/1.00, relative values).
Therefore, cohesin dissociation appears to be
the major role of E-pro transcription activity.
Transcription-induced cohesin dissociation
provides a potential mechanism for the
well-established link between transcription
and recombination (21), the molecular mechanism(s)
of which have remained controversial.
For instance, in immune cells, antibody
gene recombination requires the transcription of
flanking genes (22, 23), and this transcriptiondependent
recombination may be mediated
through cohesin dissociation. Given the large
amount of noncoding transcription recently
found in higher eukaryotes (24), some of these
transcripts may be involved in the regulation
of cohesin association, allowing cells to
regulate recombination.
References and Notes
1. L. H. Hartwell, M. B. Kastan, Science 266, 1821 (1994).
2. F. Fabre, A. Chan, W.-D. Heyer, S. Gangloff, Proc.
Natl. Acad. Sci. U.S.A. 99, 16887 (2002).
3. R. S. Hawley, C. H. Marcus, Annu. Rev. Genet. 23, 87
(1989).
4. G. P. Smith, Cold Spring Harbor Symp. Quant. Biol.
38, 507 (1973).
5. K. D. Rodland, P. J. Russell, Biochim. Biophys. Acta
697, 162 (1982).
6. T. Kobayashi, D. J. Heck, M. Nomura, T. Horiuchi,
Genes Dev. 12, 3821 (1998).
7. T. Kobayashi, M. Nomura, T. Horiuchi, Mol. Cell. Biol.
21, 136 (2001).
8. G. M. Santangelo, J. Tornow, C. S. McLaughlin, K.
Moldave, Mol. Cell. Biol. 8, 4217 (1988).
9. A. R. D. Ganley, K. Hayashi, T. Horiuchi, T. Kobayashi,
Proc. Natl. Acad. Sci. U.S.A. 102, 11787 (2005).
10. C. H. Haering, K. Nasmyth, Bioessays 25, 1178
(2003).
11. E. F. Glynn et al., PLoS Biol. 2, e259 (2004).
12. A. Lengronne et al., Nature 430, 573 (2004).
13. T. Kobayashi, T. Horiuchi, P. Tongaonkar, L. Vu, M.
Nomura, Cell 117, 441 (2004).
14. S. Laloraya, V. Guacci, D. Koshland, J. Cell Biol. 151,
1047 (2000).
15. A. V. Strunnikov, V. L. Larionov, D. Koshland, J. Cell
Biol. 123, 1635 (1993).
16. J. S. Smith, J. D. Boeke, Genes Dev. 11, 241 (1997).
17. M. Bryk et al., Genes Dev. 11, 255 (1997).
18. C. E. Fritze, K. Verschueren, R. Strich, R. E. Esposito,
EMBO J. 16, 6495 (1997).
19. E. Unal et al., Mol. Cell 16, 991 (2004).
20. H. Zou, R. Rothstein, Cell 90, 87 (1997).
21. A. Aguilera, EMBO J. 21, 195 (2002).
22. T. K. Blackwell et al., Nature 324, 585 (1986).
23. M. S. Schlissel, D. Baltimore, Cell 58, 1001 (1989).
24. P. Bertone et al., Science 306, 2242 (2004).
25. Materials and methods are available as supporting
material on Science Online.
26. We thank M. Nomura (University of California, Irvine)
and D. Koshland (Carnegie Institute) for strains and
M. Inagaki (National Institute for Basic Biology, Okazaki)
for technical support. This work was supported in
part by grants 13141205, 14380332, 17080010, and
17370065 from the Ministry of Education, Science
and Culture, Japan, and by a Human Frontier Science
Program grant to T.K.
Supporting Online Material
www.sciencemag.org/cgi/content/full/309/5740/1581/
DC1
Materials and Methods
Figs. S1 to S3
Table S1
References
14 June 2005; accepted 5 August 2005
10.1126/science.1116102
An mRNA Is Capped by a 2,5
Lariat Catalyzed by a
Group I–Like Ribozyme
Henrik Nielsen, 1,2 * Eric Westhof, 3 Steinar Johansen 2
Twin-ribozyme introns are formed by two ribozymes belonging to the group I
family and occur in some ribosomal RNA transcripts. The group I–like
ribozyme, GIR1, liberates the 5 end of a homing endonuclease messenger
RNA in the slime mold Didymium iridis. We demonstrate that this cleavage
occurs by a transesterification reaction with the joining of the first and the
third nucleotide of the messenger by a 2,5-phosphodiester linkage. Thus, a
group I–like ribozyme catalyzes an RNA branching reaction similar to the
first step of splicing in group II introns and spliceosomal introns. The resulting
short lariat, by forming a protective 5 cap, might have been useful in a
primitive RNA world.
1 Department of Medical Biochemistry and Genetics,
The Panum Institute, University of Copenhagen, DK-
2200N Copenhagen, Denmark.
2 Department of Molecular
Biotechnology-RNA Research Group, Institute
of Medical Biology, University of Tromsø, N-9037
Tromsø, Norway. 3 Institut de Biologie Moléculaire et
Cellulaire, CNRS, Université Louis Pasteur, 67084
Strasbourg Cedex, Strasbourg, France.
*To whom correspondence should be addressed.
E-mail: hamra@imbg.ku.dk
RNA splicing is found in most prokaryotic and
eukaryotic organisms and different RNA
splicing mechanisms have evolved for different
classes of genes (1, 2). Group I introns (3) carry
out splicing in a structurally and chemically
distinct way from that of group II introns and
the spliceosomal introns found widespread in
higher eukaryotes. The group I twin-ribozyme
intron found in the extrachromosomal ribosomal
DNA (rDNA) of the myxomycete Didymium
iridis (Dir.S956-1) consists of two self-catalytic
units, a conventional group I splicing ribozyme
(GIR2) and a group I–like cleavage ribozyme
(GIR1) (Fig. 1A). A homing endonuclease gene
(HEG) encoding the I-DirI mRNA is found
inserted downstream of GIR1 (4–6). The 5 end
of the I-DirI mRNAisformedbycleavage
catalyzed by the GIR1 ribozyme (7). Primer
extension analyses have led to the suggestion
of two cleavage sites located three nucleotides
apart (5, 8) referred to as IPS1 (internal
processing site 1), and IPS2, respectively
(Fig. 1B). Cleavage at IPS1 was shown to be
hydrolytic (5, 9). IPS2 has not been characterized
in detail. Primer extension analyses
of cellular RNAs exclusively show a stop at
IPS2 (10, 11), whereas the primer extension
stop at IPS1 is only observed in analysis of
full-length intron (7) or deletion constructs in
vitro.
We found the processing pattern to be
strongly dependent on sequences at both the 5
and 3 ends of the ribozyme and selected two
variants for a study of IPS2 cleavage (12). The
length variants 166.22 Eincluding 166 nt
(nucleotides) upstream and 22 nt downstream
of IPS1 Fig. 1, A and B^ and 157.22 have
comparable cleavage kinetics (fig. S1), but
primer extension analysis shows a distinct
difference in processing pattern. A primer
extension stop at IPS1 accumulates over time
in 166.22 and a stop at IPS2 accumulates in
157.22 (Fig. 1C). In a parallel cleavage analysis
with 3 end-labeled RNA (Fig. 1D) the 3
fragment that accumulates from cleavage of
both 166.22 and 157.22 is of the same length
(22 nt). This is inconsistent with cleavage at
IPS2, and we conclude that the observed primer
extension stop at IPS2 is a structural stop.
Incubation of a 22-nt 3 fragment isolated
from cleavage of 157.22 (IPS2) with the 166-nt
5 fragment results in a complete conversion of
the primer extension signal from IPS2 to IPS1
(Fig. 1E) because of ligation and recleavage by
hydrolysis. Ligation of the 22-nt fragment onto
the 3 end of the 5 fragment followed by
recleavage is shown in Fig. 1F. The ligation
reaction is fast and is dependent on the
presence of G229 because the removal of this
nucleotide by b-elimination inhibited the reaction
(fig. S2). The ligation experiments suggest
that the IPS2 modification conserves the
energy from the cleavage reaction.
The 5 ends of the two 22-nt RNAs were
analyzed by treatment of 3 end-labeled RNA
with modifying enzymes (Fig. 2A). Incubation
of the 3 fragment carrying the IPS2 modifica-
1584
2 SEPTEMBER 2005 VOL 309 SCIENCE www.sciencemag.org