DTD 5
Review
ARTICLE IN PRESS
TIGS 280
TRENDS in Genetics Vol.xx No.xx Monthxxxx
The natural history of group I introns
Peik Haugen, Dawn M. Simon and Debashish Bhattacharya
Department of Biological Sciences and Roy J. Carver Center for Comparative Genomics, University of Iowa, 312 Biology Building,
Iowa City, IA 52242-1324, USA
There are four major classes of introns: self-splicing
group I and group II introns, tRNA and/or archaeal
introns and spliceosomal introns in nuclear pre-mRNA.
Group I introns are widely distributed in protists, bacteria and bacteriophages. Group II introns are found in
fungal and land plant mitochondria, algal plastids,
bacteria and Archaea. Group II and spliceosomal introns
share a common splicing pathway and might be related
to each other. The tRNA and/or archaeal introns are
found in the nuclear tRNA of eukaryotes and in archaeal
tRNA, rRNA and mRNA. The mechanisms underlying the
self-splicing and mobility of a few model group I introns
are well understood. By contrast, the role of these highly
distinct processes in the evolution of the 1500 group I
introns found thus far in nature (e.g. in algae and fungi)
has only recently been clarified. The explosion of new
sequence data has facilitated the use of comparative
methods to understand group I intron evolution in a
broader context and to generate hypotheses about
intron insertion, splicing and spread that can be tested
experimentally.
Introduction
Mobile genetic elements have had a major impact on
eukaryotic genomes [1]. Group I introns are small RNAs
(they range in size from 250–500nt) and comprise one of
the four distinct types of introns (Box 1). These elements
are found in a wide variety of organisms (e.g. in fungi,
algae and in many other unicellular eukaryotes), genes
(i.e. protein, rRNA and tRNA coding genes) and genomes
throughout the tree of life. Group I introns spread efficiently at the DNA level into intronless cognate sites by a
process termed homing (see Glossary). The success of
group I introns in genomes is also a result of their ability
to self-splice from RNA transcripts, which potentially
renders them neutral to the host. The structure, folding
and autocatalysis of a few model group I introns have been
rigorously studied, and we now have a detailed understanding of how these elements act as catalytic RNAs
(ribozymes) during splicing of precursor RNA [2–4] (Box 2).
The mechanism of group I intron spread in DNA has also
been studied in detail. Homing endonuclease genes
(HEGs) that invade non-critical regions (i.e. terminal
loops) of group I introns promote intron mobility by
encoding highly site-specific homing endonucleases (HEs).
HEGs are themselves selfish genetic elements and the
vast majority of group I introns in nature do not contain a
Corresponding author: Bhattacharya, D. ([email protected]).
HEG (see the following section for intron life cycle). HEs
and putative HEs (encoded by HEG-like sequences) are
divided into four families based on the presence of
conserved amino acid motifs: these are the His-Cys box,
LAGLIDADG, GIY-YIG and HNH families. Differences in
the endonuclease sequences suggest independent origins
of the four families of HE and putative HE proteins [5].
Despite extensive knowledge about group I intron biochemistry, understanding the complex and often unpredictable evolutionary history of group I introns and their
associated HEGs has proven to be a formidable challenge.
The explosive growth of DNA sequence data in GenBank
(http://www.ncbi.nlm.nih.gov/GenBank/GenBankstats.
html) has, however, turned the tide enabling a more
Box 1. Types of introns
Four major classes of introns are recognized based on their splicing
mechanism. These are autocatalytic group I and group II introns,
spliceosomal introns, tRNA and/or archaeal introns. Group I introns
are widespread in protist nuclear rDNA genes, fungal mitochondria,
bacteria and bacteriophages. These pre-RNA insertions self-splice
by a well-characterized and distinctive two-step pathway that relies
on an external guanosine nucleotide as the cofactor. Group II introns
are found in bacterial and organellar genomes and self-splice through
a pathway that is different from group I introns but is similar to the
mechanism of spliceosomal intron removal. In these introns, rather
than using an external guanosine, the 2 0 -OH of an adenine residue
within the intron is the nucleophile. The shared splicing mechanism
suggests that the small nuclear (sn)RNA components of the spliceosome are derived from a group II intron. Spliceosomal introns are
the most common insertions found in nuclear pre-mRNA genes.
The tRNA introns are found in eukaryotic nuclei and in Archaea
and are enzymatically removed by a cut-and-rejoin mechanism that
requires ATP and an endonuclease. This pathway is completely
different from that of spliceosomal introns.
Glossary
Homing: the process by which an intron (or intein) spreads into the homologous position in an allele that lacks the intron.
Homing endonuclease: a highly specific endonuclease protein that is typically
encoded by self-splicing introns. They promote intron mobility through homing.
Introns: intervening sequences in genes that are removed from precursor RNA
in a process termed splicing.
Monophyletic: a group of organisms (or molecular sequences) which includes
the most recent common ancestor of all of its members and all of the descendants of that most recent common ancestor. Monophyletic groups are also
called clades.
Phylogeny: the evolutionary relationships among organisms or genes.
Reverse splicing: the reversal of the forward self-splicing reaction in which a
free intron integrates into another RNA molecule. This term is sometimes used
to describe the process by which a group I intron becomes stably integrated
into the genome following the reverse splicing reaction and subsequent reverse
transcription and reintegration into the genome.
Ribozymes: catalytic RNAs (RNA enzymes).
www.sciencedirect.com 0168-9525/$ - see front matter Q 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.tig.2004.12.007
ARTICLE IN PRESS
DTD 5
Review
2
TRENDS in Genetics Vol.xx No.xx Monthxxxx
Box 2. Group I intron secondary structure and splicing
The typical secondary structure of a group I intron consists of approximately ten paired elements (P1–P10, plus the optional P11–P17;
Figure Ia) that are organized into three domains at the tertiary
structure level [4]. The intron recognizes the 5 0 exon sequence by a
4–6nt base pairing [the internal guide sequence (IGS) interaction].
Approximately 100 nucleotides constitute the central catalytic core
of the intron RNA (the shaded regions in Figure Ia). Group I introns
are removed from precursor RNAs through a two-step splicing reaction
(Figure Ib). Many introns can self-splice in vitro as naked RNAs (i.e. no
proteins are required for intron splicing), whereas others need protein
factors to facilitate correct folding of the ribozyme core. A prerequisite
for splicing is the binding of an exogenous guanosine (exoG) cofactor
to a pocket in the catalytic core of the intron, called the G-binding site.
During the first step of splicing, the cofactor attacks the 5 0 splice site
(SS) and attaches to the intron resulting in the release of the upstream
exon. The exoG leaves the G-binding site and is replaced by the last
nucleotide of the intron, which is always a G (called uG). The second
step is initiated by an attack by the 3 0 end of the released exon on
the 3 0 SS, which results in ligation of the exons and release of the
intron RNA. Successful catalysis is dependent on the correct folding
of the intron.
(a)
P9.2
P9.1
P9b
P13′′
P9a
Group IC1 intron
P9.0
P10
5′SS
P1
P5a
G
IGS
3′SS
P5
P7
P5c
P4
P3
P14′′
P6
P5b
P8
P2
P2.1
P14′
(b)
P13′
5′SS
3′SS
5′
Intron
ωG
3′
ωG
3′
exoG
OH
first step
5′
Intron
OH + exoG
second step
Ligated exons
Free intron
exoG
5′
3′
+
Intron
ωG
TRENDS in Genetics
Figure I. Secondary structure and self-splicing of group I introns.
www.sciencedirect.com
TIGS 280
detailed picture to be drawn of intron distribution and
spread through populations and species. In this article, we
present an overview of recent advances in the understanding of group I intron distribution and evolution and
describe the driving forces that bring introns to their
genomic destinations.
Group I intron distribution in nature
The distribution of group I introns on the tree of life is
shown in Figure 1 (derived from http://www.rna.icmb.
utexas.edu/ [6]). We accounted for intron redundancy by
representing with single entries, sets of sequences from
taxa with identical names that represent population
samples (e.g. 59 introns at the same site in different
isolates of the green alga Closterium ehrenbergii) or
different cultivated strains. As of June 8 2004, a total of
w1400 group I introns were reported in eukaryotic
genomes. Of these, 800 are in the nucleus at 47 different
sites in small subunit (SSU) ribosomal (r)RNA and 44 sites
in large subunit (LSU) rRNA, 220 are in mitochondrial
genes and 370 are in plastid DNA. Nuclear group I introns
are limited to rDNA genes, whereas in the organelles, they
are found in rRNA, tRNA and protein coding regions.
Group I introns have a sporadic and highly biased distribution in nature and many microbial eukaryotes are
rich in these sequences (Figure 1). However, group I
introns appear to be absent in animals, except for three
cases. Two are found in the mitochondrial genome of the
sea anemone Metridium senile and one is in the mitochondrial gene cytochrome c oxidase subunit I of the
coral Acropora tenuis. Group I introns have not yet been
found in Archaea.
Lineages in the tree of life that are particularly intronrich are fungi, plants and the red and green algae. These
taxa contain w90% of all group I introns that have been
identified to date. However, DNA sequencing efforts
(in particular for organelle genomes) have been biased
towards particular groups of genes and organisms [e.g. in
animals and plants (of the 623 sequenced, or partially
sequenced, mitochondrial genomes, 555 are from animals;
http://www.ncbi.nlm.nih.gov/genomes/static/euk_o.html)],
and it is possible that larger pools of introns remain to be
discovered (e.g. in excavates, a group of potentially
anciently diverged flagellates that bear a unique cluster
of flagellae; http://microscope.mbl.edu/) as the taxonomic
breadth of sequencing programs increase. In contrast to
the situation in eukaryotes, group I introns are rare in
bacteria. Only fifty introns representing nine different
insertion sites occur in this prokaryotic lineage (i.e. four in
tRNA genes, four in rRNA genes and one in the recA gene
[6–8]). Group I introns also occur infrequently in viruses
[9] and phages [10,11] and are not shown in Figure 1.
Group I intron mobility
The evolutionary fates of group I introns
The sporadic distribution of group I introns suggests that
many of these mobile elements have undergone horizontal
transfer (through homing or reverse splicing) into different species and genes. The potential evolutionary fates
of laterally transferred group I introns are loss from the
lineage, vertical inheritance and/or movement within the
ARTICLE IN PRESS
DTD 5
Review
TIGS 280
TRENDS in Genetics Vol.xx No.xx Monthxxxx
3
Alveolates
E
N
yte
ph
ria
s
a
iol
ns
Ciliates
nio
d
Ra
uglyp
hid a
moe
bae
Foraminiferans
h
rac
cids
soe Stramenopiles
Oomycetes
Bico
Glaucophyte
algae
NM
Lobose
amoeba
Dictyostelid
e
slime molds
s
Hapto
phyte
Co
re j
Eubacteria
rid
ia
Fun
gi
po
os
M
N
N
Discicristates
Excavates
Archaea
N
ias
NM
s
n
ma
Opisthokonts
bid
ish
M
ako
s
Acra
Vah sid slime m
olds
lka
mp
fiid
am
Eu
oeb
gle
ae
Tr
no
yp
ids
an
os
om
es
Diatoms
N
Brown
algae
NP
Ot
he
rc
hl
a,c
alg
ae
NP
Le
tes
ella als
g
a
l
f
im
no
oa
An
s
ad
on
s
om salid
pl
Di raba
s
Pa r tamonad
Reto
Oxymonads
Ch
thulids
Cryptophytes
olds
Amoebozoa
Labyrin
Opalinid
l slime m
ia
Plasmod
icr
N
Rhizaria
rop
Alga hyte
e
P
M
ae
alg
M
Chlo
onads
a
lor
d
Re
NP
Str
ph eptoyte
s*
Plants
Cercom
N
P
Ch
MN
Din
o fl a
Ap
gel
l a te
ico
s
m
pl
ex
an
s
N
B
Figure 1. The distribution of group I introns on the tree of life. The presence of introns in nuclear (N), plastid (P), mitochondrial (M) or bacterial (B) genomes is indicated by
arrowheads and orange text and the relative number of introns is represented by the letter heights. The five different letter sizes indicate the presence of 1–25, 26–50, 51–100,
101–250 or w515 group I introns. Most (90%) introns are found in land plants plus chlorophyte (green) and rhodophyte (red) algae and the fungi (the presence of many introns
is indicated in larger text). Streptophytes (marked with an asterisk) represent both charophyte algae and land plants. There are no reported group I introns in the nuclear DNA
of land plants. The tree is a consensus of molecular and ultrastructural data (for more details, see Ref. [43]) and the broken lines represent the possible positions of the root
(the prokaryotic lineages) on the consensus phylogeny of eukaryotes. Adapted with permission from Ref. [43].
lineage with partial loss or fixation (i.e. stable inheritance
with no losses and no apparent mobility within the lineage).
An additional level of complexity is that many group I
introns appear to be mosaic genetic elements. Unlike their
counterparts in group II introns, HEGs in group I introns
are less likely to share the same evolutionary history as
the catalytic RNA [12]. In fact, a group I intron-encoded
HEG from one intron (e.g. intron one) can invade another
group I intron at the homologous (same) or heterologous
(other) position, independent of the group I ribozyme
encoded by intron one (see the following section).
The intron life-cycle and homing
In 1999 Goddard and Burt published a model of intron
evolution that involved cyclical gain and loss (Figure 2
[13]). Analysis of the yeast omega group I ribozyme
showed that a group I intron with a full-length HEG
invades an intron-minus population and spreads into all
homologous (same) sites through homing (see Ref. [5] for
mechanistic details of homing). Once the intron becomes
fixed in the population, the HEG no longer has a biological
function (i.e. the HEG function is to confer intron mobility),
accumulates mutations and is eventually inactivated or lost.
Unable to spread, the intron is also destined to be lost over
www.sciencedirect.com
time and will not reappear until the same intron with the
full-length HEG is reintroduced through gene flow.
To evaluate if the ‘omega’ life-cycle applies to other
HEG-associated group I introns, it is necessary to examine
many different introns from natural populations. If
multiple populations of an organism that carries a
HEG-containing intron are sampled, then one would
expect to find introns with full-length HEGs, degenerate
HEGs, introns lacking HEGs in addition to examples of
intron absence. This pattern fits well with what has been
reported previously [6,11,14–20], although every step in
the intron cycle is not always found in each case (probably
as a result of insufficient sampling). An extensive study
of SSU rDNA (S)516 and S1506 (the numbering reflects
the homologous position in Escherichia coli rDNA) group I
introns from geographically distinct populations of red
algae identified individuals lacking introns, those containing introns without a HEG and those containing introns
with various deletions and/or mutations in the HEG [17].
In summary, the Goddard and Burt intron life-cycle
model, which was established with the yeast omega
intron, is supported by the intron and HEG distribution
in red algae and is consistent with the intron distribution
in other naturally occurring populations. This suggests
that recurrent gain and loss is likely to be a general
ARTICLE IN PRESS
DTD 5
4
Review
TIGS 280
TRENDS in Genetics Vol.xx No.xx Monthxxxx
Precursor RNA
No intron
Intron invasion
and fixation
Precise loss
Intron loss
Homing
Functional mobile intron
Precursor RNA
Precursor RNA
Intron with nonfunctional HEG
HEG
or
HEG
HEG degeneration
Loss of HEG function
TRENDS in Genetics
Figure 2. This model was proposed by Goddard and Burt [13] and is consistent with other analyses of HEG-associated group I introns. In this scheme, a mobile intron with a
functional HEG invades an intron-less population and becomes fixed in all homologous (same) insertion sites through homing. Because the HEG no longer can promote
intron spread (all sites are occupied) the HEG becomes redundant and eventually lost. The intron life-cycle is completed by the precise loss of the intronmodel for the
recurrent gain and loss of the omega group I intron.
feature of HEG-associated group I introns. Introns do however manage to ‘escape’ from the cycle by inserting into a
new genic position (discussed in the following section).
Introns in the same insertion site are evolutionarily
related
Reconstruction of group I intron phylogeny is made possible by the existence of the conserved RNA secondary
structures (P1–P9) that can be used to guide the alignment of these sequences. The alignments are used as input
to computer programs (e.g. PAUP*, PHYLIP, MrBayes
and MEGA) to infer evolutionary trees. The core region
used in these analyses is short in length (100–250 nt) but,
as shown in Figure 3, it still encodes sufficient phylogenetic signal to resolve many nodes in the intron trees.
An important insight into group I intron evolution that
has come from phylogenetic analyses is that sequences
from homologous genic sites are closely related (monophyletic) even though they can reside in distantly related
organisms (e.g. eukaryotes and prokaryotes) [21]. It is
unlikely that this result is explained by a single origin of
each intron before the eukaryotic radiation (or the split of
bacteria and eukaryotes) followed by widespread loss in
different lineages. Introns are highly divergent sequences
and it is therefore unlikely that they remain conserved
across the millions or billions of years that separate taxa
such as prokaryotes and eukaryotes. A more parsimonious
scenario is that group I introns are laterally transferred
across the tree of life resulting in their sporadic and highly
biased distribution. If the second idea is correct, then the
observation that intron movement nearly always occurs to
the homologous site in different organisms suggests that
strong restrictions exist to guide introns to their target
(i.e. high sequence specificity in homing).
Intron movement through reverse splicing
Group I introns do however occasionally escape to heterologous genic sites, but the mechanism(s) responsible for
www.sciencedirect.com
these invasions have proven elusive. Because of the
limited sequence requirement (4–6nt) for group I ribozymes to locate their target region, reverse splicing of free
intron RNAs into target RNAs provides a more plausible
model for intron spread into ectopic sites than homing,
which relies on a 15–45nt recognition sequence for the
intron-encoded homing endonuclease. Whether this pathway has an important role in group I intron spread,
however, is unknown because of the lack of any direct
evidence of reverse-splicing-mediated intron movement
from genetic crosses. Furthermore, in contrast to homing
that has been shown to be highly efficient, it is likely that
reverse splicing, with its reliance on chance integration
followed by two additional steps (i.e. reverse-transcription
and then recombination), would be less effective in
promoting intron movement. Rare reverse splicing events
are therefore most likely to be recognized in the context of
a broadly sampled intron phylogeny. A final consideration
is that rDNA exists as a multi-copy gene family necessitating that alleles containing transferred introns must
rise to high frequency (presumably through concerted
evolution or less parsimoniously, repeated reverse splicing
events) in individuals and in populations to ensure
survival and fixation.
In spite of these constraints, conclusive experimental
evidence exists that group I introns can fulfill one crucial
step in the reverse splicing mobility pathway, full integration into foreign RNA in vivo [22,23]. In one analysis,
19 variants of the Tetrahymena intron, all with different
internal guide sequences were expressed in E. coli [23].
Partial reverse splicing at 69 sites and complete integration at one novel site in the 23S rRNA were found with
this approach. Phylogenetic analyses provide support for
another prediction of the reverse splicing pathway,
statistical support for the close relationship of introns at
ectopic rDNA sites (e.g. S1046–S1052, S289–S1199 [18]).
In these cases, sequence similarity is limited to the proximal 5 0 flanking exon in each pair of related introns. This
ARTICLE IN PRESS
DTD 5
Review
TIGS 280
TRENDS in Genetics Vol.xx No.xx Monthxxxx
(a)
5
(b)
66
S516
S788
73/68
98
97
80/67
L2563
S1199
L2066
S516
S516
S516
S516
L1926
L1923
88
63
87/73
91/84
S1506
S516
94/86
S1516
95/76
S989
90/62
95/77
S1210
IC1
S1199
66
100/94
100/99
S1199
66/65/-
S943
100
S1199
77
S1199
S516
IE
L2449
S516
0.1 substitutions per site
0.2 substitutions per site
TRENDS in Genetics
Figure 3. The phylogeny of nuclear rDNA group I introns. A summary of the phylogenetic relationships (a) between introns in algae and fungi, and (b) between introns in
ascomycete fungi are shown. In both trees, introns at homologous sites are generally more closely related to each other than to introns at ectopic sites. Statistical support for
clades using different phylogenetic methods is shown above and below the branches (bootstrap values R60%) or as thick lines (bayesian posterior probabilities R0.95). Both
trees were built using the neighbor-joining method with evolutionary distances calculated according to the (a) Kimura two-parameter or (b) the Jukes-Cantor model. For more
details about the phylogenetic methods used in (a), see Ref. [21] and for those used in (b), see Ref. [37]. Branch lengths are proportional to the number of substitutions per site
and the size of the triangles represents the number of sequences in each clade.
finding is consistent with the 4–6nt sequence requirement
in the 5 0 exon that is predicted for the RNA-based reverse
splicing and integration pathway. In addition, HEGs have
not been found within the introns at these insertion sites
making it less likely that endonucleases mediated the
invasions of new sites at the DNA level. Although we still
lack conclusive proof of reverse splicing-mediated intron
mobility, accumulating experimental and comparative
data are in favor of this mechanism having contributed
to the spread of group I introns.
Intron-independent HEG mobility
It has been known for some time that sequences encoding
HEGs insert into the peripheral regions of ribozymes
and mobilize them through homing. We now have a more
detailed picture of the intron-HEG relationship and the
evolutionary history of the two partners turns out to be
more complicated than previously thought. The most
direct approach to identify intron-independent HEG movement is through the comparison of intron and HEG trees.
Unfortunately, this approach is stymied by the rare
occurrence of intact HEGs in nature (i.e. they decay rapidly
following intron fixation [20]). HEG mobility can, however,
leave clear evolutionary footprints (Figure 4). For example,
www.sciencedirect.com
HEGs that are closely related to the His-Cys box family
are inserted into different peripheral loops or encoded
on different strands of homologous group I introns
(e.g. introns inserted after positions 516, 943 and 1506
in the gene that encodes the nuclear SSU rRNA [20]).
Movement to different peripheral loop-regions within the
intron provides evidence for HEG mobility within a single
ribozyme lineage. Furthermore, closely related HEGs are
found in phylogenetically distantly related group I introns
that are located at adjacent rDNA sites (sites L1923,
L1925 and L1926), suggesting that local HEG mobility into
heterologous introns occurs [20]. A study of rDNA introns
that encode HEGs of the LAGLIDADG family also identified
related HEGs in adjacent introns, in some cases, resulting
from HEG mobility alone or, in other cases, as a result of the
mobility of the intron–HEG element as a unit [24].
Did a gene duplication event trigger the spread of
LAGLIDADG HEGs?
Homing endonucleases that contain the conserved
LAGLIDADG motif represent the largest family of HEGs
and are widespread in nature, occurring in several different types of mobile elements (i.e. group I and II introns,
inteins and archaeal introns) and as freestanding open
ARTICLE IN PRESS
DTD 5
Review
6
(a)
Intron homing
TIGS 280
TRENDS in Genetics Vol.xx No.xx Monthxxxx
HEG
Allele "a"
Allele "A"
Intron
DSBR
HEG
HEG
Allele "A"
Intron
Intron
Allele "a"
Intron
or
Intron
Allele "a"
Both alleles are protected against ENase activity
(b)
HEG invades
homologous
intron or intron in
adjacent site
HEG
Allele "A"
Intron
Allele "a"
DSB (in intron)
HEG
Illegitimate
recombination
Allele "A"
Intron
HEG
HEG
Intron
or
Intron
Intron
or
Intron
HEG
or HEG
HEG
or HEG
Intron
Intron
or
or
Intron
Intron
HEG
Allele "a"
Allele "a"
Allele "a"
Allele "a"
HEG
TRENDS in Genetics
Figure 4. A model for the movement of homing endonuclease genes. Homing endonuclease genes (HEGs) have the potential to move with the group I intron as (a) a unit by
homing or (b) independent of the ribozyme. Homing results in the transfer of the intron-HEG element plus flanking exon sequences (co-conversion) into the homologous
genic position. The HE recognition sequence is shown in red. Ribozyme-independent HEG transfer is likely to result in the insertion of the HEG into homologous HEG-minus
introns (both shown in orange), or heterologous introns that are located at an adjacent site (shown in orange and blue) with respect to the donor intron. HEGs can insert into
different strands of the intron encoding sequence (indicated by arrows above and below the intron), and into different peripheral regions of the ribozyme structure (indicated
by the different locations of the HEG on the schematic representation of the intron).
reading frames (ORFs) [25]. Comparative studies suggest
that the successful spread of LAGLIDADG HEGs was
driven by a gene duplication and fusion event followed by
the rapid evolution of the DNA-binding domain [24,26,27].
Approximately 40 group I intron-associated LAGLIDADG
HEGs contain one copy of the motif (i.e. LAGLIDADG
motif), whereas the remainder contain two copies (O200
cases in total [5,24]). Interestingly, the single-motif HEGs
are restricted to introns located at six insertion sites in
LSU rDNA; five out of six of these are clustered between
L1917 and L1951. By contrast, the double-motif HEGs are
more widely distributed in rDNA introns and in introns in
other types of genes.
The difference in the success of these HEGs in
spreading into novel introns probably reflects how they
recognize target sequences. The single-motif HEs function
www.sciencedirect.com
as homo-dimers and, therefore, need a high degree of
symmetry in the DNA target sequence, whereas the
double-motif HEs function as monomers. The N- and
C-terminal halves of the monomeric HEs (i.e. each half
corresponds to one protein in the dimers of single-motif
HEs) have accumulated many amino acid substitutions
and might have a less stringent requirement for DNA
target symmetry [26]. Most double-motif LAGLIDADG
HEGs in rDNA can be traced to one of two lineages [24].
One of these (designated ‘Clade 1’ in Ref. [24]) appears to
be derived from a duplication and fusion event involving a
single-motif HEG in an intron located between the rDNA
positions L1917–L1951, whereas the origin of the other
group of double-motif HEGs (‘Clade 2’) is unclear. In
summary, evolution has found an effective solution to
target site restriction in LAGLIDADG HEGs. Duplication
ARTICLE IN PRESS
DTD 5
Review
TIGS 280
TRENDS in Genetics Vol.xx No.xx Monthxxxx
and divergence of the ancestral single-motif HEGs to form
a two-motif unit appears to have freed them to explore a
larger set of host introns and genic sites.
7
structure). It is therefore likely that the maturase activity
evolved as a secondary adaptation. In summary, there
appears to be a dynamic relationship between ribozyme,
HE and maturase activities in HEG-associated group I
introns. The relative evolutionary importance of these
processes reflects primarily the potential for intron spread
in populations or species and the extent of dependence on
the maturase for intron splicing.
The role of proteins in group I intron splicing
HEGs can avoid loss by gaining maturase activity
A fascinating aspect of the group I intron-HEG relationship is that some endonucleases also function as maturases (i.e. they assist in intron folding [28]). The existence
of these bifunctional proteins suggests that the gain of
maturase activity has provided the HEs and/or their associated introns with an evolutionary advantage. So what
advantage could maturase activity offer? One hypothesis
is that the gain of a HEG-encoded DNA (typically w1kb in
size) in the group I intron might lower the efficiency of selfsplicing. The maturase activity conferred by the HEG could
therefore be selected for because it rescues or sufficiently
improves splicing ability to compensate for the presence
of the ORF [28]. Another intriguing possibility is that
maturase activity evolves (or is retained) because it gives
the HEG an opportunity to escape inactivation or loss [29]
(i.e. endonuclease function will be lost if the intron is fixed
in the population, whereas maturase activity might be
retained to ensure correct intron excision). This scenario
appears to have played out recently in Saccharomyces
cerevisiae in which two non-adjacent amino acid changes
activate endonuclease activity in an intron-encoded
LAGLIDADG maturase [30,31]. Other important insights
have come from structural and biochemical studies of the
I-AniI HE protein [29,32,33]. In this case, the RNA and
DNA-binding domains of the protein are non-overlapping
(i.e. they are located at different places in the protein
The long-term evolution of self-splicing ability
Insights into the evolution of RNA structure and autocatalysis can be gained by comparing group I introns that
have been vertically inherited for millions of years. These
sequences provide potential examples of intron-host
co-evolution and are candidates for introns that might
have evolved a function that is beneficial to the host. Two
outstanding examples of long-term vertical inheritance
are the ancient group I intron inserted in the tRNA-Leu
gene of cyanobacteria and plastids [34–36] and the S788
intron in the ascomycete fungi [37]. The tRNA-Leu intron
has probably been vertically inherited for more than a
billion years in plastids [38] and two-to-three billion years
in the cyanobacterial ancestors of these organelles [34],
whereas the S788 intron was probably acquired in ascomycetes w400–600 million years ago [37]. Although there
are some obvious differences in the evolutionary history of
these two group I introns, there are also striking similarities. Most importantly, when self-splicing activity is
mapped on the intron phylogeny (which generally recapitulates the host tree as a result of intron vertical
inheritance), in both cases the early diverging ribozymes
self-splice efficiently as naked RNAs. By contrast, the later
Euphyllophytina
65
1st
−
74
+
−
Chlorophyta 1st
63
75
Stramenopiles
Glaucophyta 1st
Cyanobacteria
+
+
+
+
+
1st
Charophytes
Bryophytes
Lycophytina
97
69
(b)
−
−
−
−
−
−
−
−
−
−
−
70
Streptophyta
Streptophyta
Euphyllophytina
Coleochaetales
Charales
Zygnematales
Chlorokybales
Chlorophyta
Stramenopiles
Glaucophyta
Cyanobacteria
Charophyta
(a)
− P5abcd
100
77
−
66
83
84
+
+
+
+
+
−
−
−
+ P5abcd
−
Figure 5. Analysis of the phylogeny and in vitro splicing of two vertically inherited group I introns. Summary phylogenies and the in vitro splicing ability of introns inserted
(a) into the tRNA-Leu gene in plastids and in cyanobacteria and (b) after position 788 in the SSU rRNA of ascomycete fungi are shown. Introns that splice efficiently (C),
undergo the first step of splicing (1st), or show no activity in vitro (K), are indicated on the tree. The boxed cladogram in (a) shows the host relationships of taxa containing
the tRNA-Leu intron. The host and intron trees are in general agreement. Two distinct clades are indicated in the S788 intron, an ancestral pool that contains the P5 extensions
(CP5abcd) and a later diverging clade that lacks this domain (-P5abcd). The trees are modified with permission from Ref. [36] and Ref. [37].
www.sciencedirect.com
DTD 5
8
Review
ARTICLE IN PRESS
TRENDS in Genetics Vol.xx No.xx Monthxxxx
diverging introns can either only complete the first step of
splicing [i.e. attack of the 5 0 splice site by the exogenous
guanosine which becomes covalently attached to the
intron (the 5 0 exon is freed)] or lack autocatalytic activity
entirely (Figure 5). It is tempting to speculate that the
partial or complete loss of activity in both ribozymes is due
to their dependence over evolutionary time on particular
host proteins to facilitate splicing. Furthermore, the fixation of the tRNA-Leu intron in all plants combined with
their loss of self-splicing ability suggests that a reciprocal
dependence between the intron and the host might have
evolved in the common ancestor of this highly successful
lineage. Although unproven, this idea is supported by a
study that shows chloroplast ribonucleoproteins (cpRNPs)
associate with unspliced tRNA-Leu transcripts in Nicotiana
tabacum [39]. The cpRNP–tRNA complexes confer stability and ribonuclease resistance to the RNAs and potentially act as a scaffold for host-mediated splicing of the
intron-containing tRNAs [39].
The S788 group I intron is also of great interest because
derived members of this ribozyme lineage have undergone
major structural changes with many nucleotide substitutions in the central catalytic core, some of which are
crucial for intron folding, and the loss of the larger peripheral P5abcd domain (Figure 5). Two distinct clades are
identified in phylogenetic analyses of the S788 intron, an
ancestral pool that contains the P5 extensions (CP5abcd,
Figure 5) and a later diverging clade that lacks this domain
(KP5abcd) [37]. The loss of P5abcd is correlated with the
inability to self-splice in vitro. This example therefore provides a direct link between group I intron structure evolution and autocatalysis in a vertically inherited intron.
Concluding remarks
The detailed mechanisms of intron autocatalysis and mobility have been established through rigorous biochemical
analyses. Intron evolution is being clarified through the
analysis of intron phylogeny, distribution, secondary
structure and splicing in a diverse group of microbial
eukaryotes. Bringing together these two disparate pools of
information is essential to understand the natural history
of group I introns. Recent work has successfully integrated
structural and functional insights regarding group I introns
into a broader evolutionary context. Examples of important
contributions to our understanding of intron biology and
evolution are the compilation of introns into a comprehensive database [6], development of the cyclical model
for the gain and loss of HEG-associated introns [13],
novel insights into the dynamic intron-HEG relationship
[24,26,27,29] and the recognition that group I introns can
be vertically inherited over long periods of evolutionary
time and their interaction with the host cell can provide
important insights into intron biology [34–37,39]. It will
be important to investigate the role of host factors in
intron splicing across the tree of life {e.g. mitochondrial
tyrosyl-tRNA synthetase (Cyt-18, [31]), leucyl-tRNA
synthetase (LeuRS, [40]) and mitochondrial RNA splicing 1
(Mrs1, [41])} and the contribution of these factors to intron
fixation and mobility. It is likely that host proteins generally have a role in the splicing of many group I introns;
however, their broader impact on intron evolution is
www.sciencedirect.com
TIGS 280
currently unknown. Another unanswered question is how
group I introns are transferred between species. Viruses
have often been suggested as possible vectors but evidence is still lacking. Finally, the true biological role(s) for
group I intron RNAs remains to be clarified (for more
details, see Ref. [42]).
Acknowledgements
This work was generously supported by grants awarded to D.B. from the
National Science Foundation (MCB 0110252, DEB 0107754), a grant from
The Norwegian Research Council to P.H., and Avis E. Cone and Stanley
fellowships from the University of Iowa to D.S. We thank H. Joseph Runge
(Iowa) for helpful discussions.
References
1 Hurst, G.D. and Werren, J.H. (2001) The role of selfish genetic
elements in eukaryotic evolution. Nat. Rev. Genet. 2, 597–606
2 Cech, T.R. (2002) Ribozymes, the first 20 years. Biochem. Soc. Trans.
30, 1162–1166
3 Westhof, E. (2002) Group I introns and RNA folding. Biochem. Soc.
Trans. 30, 1149–1152
4 Adams, P.L. et al. (2004) Crystal structure of a self-splicing group I
intron with both exons. Nature 430, 45–50
5 Chevalier, B.S. and Stoddard, B.L. (2001) Homing endonucleases:
structural and functional insight into the catalysts of intron/intein
mobility. Nucleic Acids Res. 29, 3757–3774
6 Cannone, J.J. et al. (2002) The comparative RNA web (CRW) site: an
online database of comparative sequence and structure information
for ribosomal, intron, and other RNAs. BMC Bioinformatics 3, 2 doi:
10.1186/1471-2105-3-2 (http://www.biomedcentral.com/1471-2105/3/2)
7 Ko, M. et al. (2002) Group I self-splicing intron in the recA gene of
Bacillus anthracis. J. Bacteriol. 184, 3917–3922
8 Nesbø, C.L. and Doolittle, W.F. (2003) Active self-splicing group I
introns in 23S rRNA genes of hyperthermophilic bacteria, derived
from introns in eukaryotic organelles. Proc. Natl. Acad. Sci. U. S. A.
100, 10806–10811
9 Nishida, K. et al. (1998) Group I introns found in Chlorella viruses:
biological implications. Virology 242, 319–326
10 Edgell, D.R. et al. (2000) Barriers to intron promiscuity in bacteria.
J. Bacteriol. 182, 5281–5289
11 Sandegren, L. and Sjöberg, B.M. (2004) Distribution, sequence
homology, and homing of group I introns among T-even-like bacteriophages: evidence for recent transfer of old introns. J. Biol. Chem. 279,
22218–22227
12 Toor, N. et al. (2001) Coevolution of group II intron RNA structures
with their intron-encoded reverse transcriptases. RNA 7, 1142–1152
13 Goddard, M.R. and Burt, A. (1999) Recurrent invasion and extinction
of a selfish gene. Proc. Natl. Acad. Sci. U. S. A. 96, 13880–13885
14 Cho, Y. et al. (1998) Explosive invasion of plant mitochondria by a
group I intron. Proc. Natl. Acad. Sci. U. S. A. 95, 14244–14249
15 Haugen, P. et al. (1999) Complex group-I introns in nuclear SSU rDNA
of red and green algae: evidence of homing-endonuclease pseudogenes
in the Bangiophyceae. Curr. Genet. 36, 345–353
16 Foley, S. et al. (2000) Widespread distribution of a group I intron and
its three deletion derivatives in the lysin gene of Streptococcus
thermophilus bacteriophages. J. Virol. 74, 611–618
17 Müller, K.M. et al. (2001) A structural and phylogenetic analysis of the
group IC1 introns in the order Bangiales (Rhodophyta). Mol. Biol.
Evol. 18, 1654–1667
18 Bhattacharya, D. et al. (2002) Vertical evolution and intragenic spread
of lichen-fungal group I introns. J. Mol. Evol. 55, 74–84
19 Nozaki, H. et al. (2002) Evolution of rbcL group IA introns and intron
open reading frames within the colonial Volvocales (Chlorophyceae).
Mol. Phylogenet. Evol. 23, 326–338
20 Haugen, P. et al. (2004) The evolution of homing endonuclease genes
and group I introns in nuclear rDNA. Mol. Biol. Evol. 21, 129–140
21 Nikoh, N. and Fukatsu, T. (2001) Evolutionary dynamics of multiple
group I introns in nuclear ribosomal RNA genes of endoparasitic fungi
of the genus Cordyceps. Mol. Biol. Evol. 18, 1631–1642
DTD 5
Review
ARTICLE IN PRESS
TRENDS in Genetics Vol.xx No.xx Monthxxxx
22 Roman, J. and Woodson, S.A. (1998) Integration of the Tetrahymena
group I intron into bacterial rRNA by reverse splicing in vivo. Proc.
Natl. Acad. Sci. U. S. A. 95, 2134–2139
23 Roman, J. et al. (1999) Sequence specificity of in vivo reverse splicing
of the Tetrahymena group I intron. RNA 5, 1–13
24 Haugen, P. and Bhattacharya, D. (2004) The spread of LAGLIDADG
homing endonuclease genes in rDNA. Nucleic Acids Res. 32, 2049–2057
25 Dalgaard, J.Z. et al. (1997) Statistical modeling and analysis of the
LAGLIDADG family of site-specific endonucleases and identification
of an intein that encodes a site-specific endonuclease of the HNH
family. Nucleic Acids Res. 25, 4626–4638
26 Lucas, P. et al. (2001) Rapid evolution of the DNA-binding site in
LAGLIDADG homing endonucleases. Nucleic Acids Res. 29, 960–969
27 Chevalier, B. et al. (2003) Flexible DNA target site recognition by
divergent homing endonuclease isoschizomers I-CreI and I-MsoI.
J. Mol. Biol. 329, 253–269
28 Belfort, M. (2003) Two for the price of one: a bifunctional intronencoded DNA endonuclease-RNA maturase. Genes Dev. 17, 2860–2863
29 Chatterjee, P. et al. (2003) Functionally distinct nucleic acid binding
sites for a group I intron encoded RNA maturase/DNA homing endonuclease. J. Mol. Biol. 329, 239–251
30 Szczepanek, T. and Lazowska, J. (1996) Replacement of two nonadjacent amino acids in the S. cerevisiae bi2 intron-encoded RNA
maturase is sufficient to gain a homing-endonuclease activity. EMBO J.
15, 3758–3767
31 Lambowitz, A.M. et al. (1999) Group I and II ribozymes as RNPs:
Clues from the past and guides to the future. In The RNA World
(2nd edn) (Gesteland, R.F. et al., eds), pp. 451–485, Cold Spring Harbor
Laboratory Press
www.sciencedirect.com
TIGS 280
9
32 Bolduc, J.M. et al. (2003) Structural and biochemical analyses of DNA
and RNA binding by a bifunctional homing endonuclease and group I
intron splicing factor. Genes Dev. 17, 2875–2888
33 Geese, W.J. et al. (2003) In vitro analysis of the relationship between
endonuclease and maturase activities in the bi-functional group I
intron-encoded protein, I-AniI. Eur. J. Biochem. 270, 1543–1554
34 Kuhsel, M.G. et al. (1990) An ancient group-I intron shared by
eubacteria and chloroplasts. Science 250, 1570–1573
35 Xu, M.Q. et al. (1990) Bacterial origin of a chloroplast intron:
conserved self-splicing group-I introns in cyanobacteria. Science 250,
1566–1570
36 Simon, D. et al. (2003) Phylogeny and self-splicing ability of the plastid
tRNA-Leu group I Intron. J. Mol. Evol. 57, 710–720
37 Haugen, P. et al. (2004) Long-term evolution of the S788 fungal
nuclear small subunit rRNA group I introns. RNA 10, 1084–1096
38 Yoon, H.S. (2004) A molecular timeline for the origin of photosynthetic
eukaryotes. Mol. Biol. Evol. 21, 809–818
39 Nakamura, T. et al. (1999) Chloroplast ribonucleoproteins are associated with both mRNAs and intron-containing precursor tRNAs.
FEBS Lett. 460, 437–441
40 Rho, S.B. et al. (2002) An inserted region of leucyl-tRNA synthetase
plays a critical role in group I intron splicing. EMBO J. 21, 6874–6881
41 Bassi, G.S. et al. (2002) Recruitment of intron-encoded and co-opted
proteins in splicing of the bI3 group I intron RNA. Proc. Natl. Acad.
Sci. U. S. A. 99, 128–133
42 Nielsen, H. et al. (2003) The ability to form full-length intron RNA
circles is a general property of nuclear group I introns. RNA 9,
1464–1475
43 Baldauf, S.L. (2003) The deep roots of eukaryotes. Science 300,
1703–1706