A group-I intron in the mitochondrial small subunit ribosomal RNA gene of

Abstract

A 1 380-bp intervening sequence within the mitochondrial small subunit ribosomal RNA (mt SSU rRNA) gene of the fungus Sclerotinia sclerotiorum has been sequenced and identified as a group-I intron. This is the first report of an intron in the mt SSU rRNA gene. The intron shows close similarity in secondary structure to the subgroup-IC2 introns from Podospora (ND3i1, ND5i2, and COIi5) and Neurospora (ND5i1). The intron has an open reading frame (ORF) that encodes a putative protein of 420 amino acids which contains two copies of the LAGLI-DADG motif. The ORF belongs to a family of ORFs identified in Podospora (ND3i1, ND4Li1, ND4Li2, ND5i2, and COIi5) and Neurospora (ND5i1). The putative 420-aa polypeptide is also similar to a site-specific endonuclease in the chloroplast large subunit ribosomal RNA (LSU rRNA) gene of the green alga Chlamydomonas eugametos. In each clone of S. sclerotiorum examined, including several clones which were sampled over a 3-year period from geographically separated sites, all isolates either had the intron or lacked the intron within the mt SSU rRNA gene. Screening by means of Southern hybridization and PCR amplification detected the intron in the mt SSU rRNA genes of S. minor, S. trifoliorum and Sclerotium cepivorum, but not in other members of the Sclerotiniaceae, such as Botrytis anamorphs of Botryotinia spp., or in other ascomycetous and basidiomycetous fungi.

Full text

Curr Genet (1995) 27:166-176 9 Springer-Verlag 1995
Ignazio Carbone 9 James B. Anderson 9 Linda M. Kohn
A group-I intron in the mitochondrial small subunit ribosomal RNA gene
of Sclerotinia sclerotiorum
Received: 18 March 1994 / 14 July 1994
Abstract.
A 1 380-bp intervening sequence within the
mitochondrial small subunit ribosomal RNA (mt SSU
rRNA) gene of the fungus
Sclerotinia sclerotiorum
has
been sequenced and identified as a group-I intron. This is
the first report of an intron in the mt SSU rRNA gene. The
intron shows close similarity in secondary structure to the
subgroup-IC2 introns from
Podospora
(ND3il, ND5i2,
and COIi5) and
Neurospora
(ND5il). The intron has an
open reading frame (ORF) that encodes a putative protein
of 420 amino acids which contains two copies of the
LAGLI-DADG motif. The ORF belongs to a family of
ORFs identified in
Podospora
(ND3il, ND4Lil, ND4Li2,
ND5i2, and COIi5) and
Neurospora
(NDSil). The puta-
tive 420-aa polypeptide is also similar to a site-specific en-
donuclease in the chloroplast large subunit ribosomal RNA
(LSU rRNA) gene of the green alga
ChIamydomonas eu-
gametos.
In each clone of
S. scIerotiorum
examined, in-
cluding several clones which were sampled over a 3-year
period from geographically separated sites, all isolates ei-
ther had the intron or lacked the intron within the mt SSU
rRNA gene. Screening by means of Southern hybridiza-
tion and PCR amplification detected the intron in the
mt SSU rRNA genes of
S. minor, S. trifoliorum
and
Sclerotium cepivorum,
but not in other members of the
Sclerotiniaceae, such as
Botrytis
anamorphs of
Botryoti-
nia
spp., or in other ascomycetous and basidiomycetous
fungi.
Key words: Asexual reproduction 9 Clonality 9 Group-I
intron - Mitochondrial small subunit ribosomal RNA gene
Sclerotinia sclerotiorum
Ignazio Carbone (N) James B. Anderson 9 Linda M. Kohn
Department of Botany, University of Toronto, Erindale College,
Mississauga, Ontario L5L 1C6, Canada
Communicated by H. Bertrand
Introduction
Fungal mitochondrial introns have been classified in two
major groups, I and II, based on secondary structure char-
acteristics and the presence of short conserved sequences.
Group-I introns predominate in fungal mitochondrial ge-
homes and are characterized by the presence of an invari-
ant uridine (U) residue at the 3' end of the upstream exon,
an invariant guanosine (G) at the 3' end of the intron, four
highly conserved sequence elements denoted P, Q, R, and
S (Cech 1988), and nine base-paired regions denoted P1
through P9 (Davies et al. 1982; Michel et al. 1982; Burke
et al. 1987). On the basis of the presence or absence of
these base-paired structures and other conserved primary
and secondary structure features, group-I introns have been
subdivided into four major subgroups, designated as IA,
IB, IC, and ID (Michel and Westhof 1990). Introns within
each subgroup have been further classified into numerical
groups. The four conserved elements of group-I introns
base-pair with each other, P with Q to form P4, and R with
S to form P7. The last two bases of Q also base-pair with
the first two bases of R to form P6. These internal pairings
form a secondary core structure that is conserved among
all group-I introns and is important in splicing (Davies et
al. 1982; Michel et al. 1982). This core structure always
contains base-paired regions P3 to P9 and in most cases
also contains P2 (Waring and Davies 1984). In addition to
the conserved core elements, all group-I introns possess an
internal guide sequence (IGS) near their 5' splice site which
can base-pair with exon sequences adjacent to the 5' and
3' intron splice sites to align them precisely for subsequent
splicing and ligation reactions (Davies et al. 1982; Waring
and Davies 1984).
Variability has been reported among group-I introns.
One source of variability is in the size of loops that join
the 3" and 5' regions of base-paired regions, particularly
P1 and P5 (Waring and Davies 1984; Collins 1988). Vari-
ability is also observed in the presence or absence of an
ORF, the position of the ORF, as well as the putative pro-
tein encoded by the ORF. Some ORFs overlap the core

structure, but most are located in looped-out regions that
are external to the core.The majority of intron ORFs have
been shown to encode proteins which function either as
RNA maturases that promote splicing of their host intron,
or as site-specific DNA endonucleases that mediate group I
intron mobility (reviewed by Lambowitz and Belfort
1993). Many maturase-type proteins contain two copies of
a short 12-aa (LAGLI-DADG) dodecapeptide repeat
spaced approximately 115 aa apart (Hensgens et al. 1983;
Michel 1984; Michel and Cummings 1985; Cummings and
Domenico 1988). The intron-encoded endonucleases in-
clude proteins in two structural classes characterized by
different sequence motifs, as well as some proteins that do
not fall into either structural class. The larger class con-
tains endonucleases with one or two copies of the LAGLI-
DADG dodecapeptide motif. The second class contains en-
donucleases with the GIY-10/11 aa-YIG motif.
Many group-I introns have been reported from the nu-
clear genomes of fungi, the majority from the SSU rRNA
gene (Sogin and Edman 1989; DePriest and Been 1992; De
Wachter et al. 1992; Nishida et al. 1993). To-date, introns
within fungal mitochondrial genomes have been reported
from the LSU rRNA gene, from the cytochrome b gene,
from genes encoding cytochrome oxidase subunits (COI,
II and III), from NADH dehydrogenase subunits (ND1, 2,
3, 4, 4L, 5 and 6), and from ATPase subunits (ATPase 6, 8
and 9) (Michel and Cummings 1985; Cummings and Do-
menico 1988; Cummings et al. 1990; De Bi~vre and Du-
jon 1992). No introns have been reported from the mt SSU
rRNA gene. Group-I introns have also been observed in
the nuclear and organellar genomes of plants, green algae
(D~ivila-Aponte et al. 1991; Wilcox et al. 1992; Ragan et
al. 1993), amoebas, ciliates, and slime molds, as well as in
cyanobacteria and bacteriophage (for reviews see Burke
1988; Lambowitz and Belfort 1993). In this study, we re-
port the presence of a group-I intron in the mt SSU rRNA
gene of the fungus, Sclerotinia sclerotiorum.
S. sclerotiorum is an important ascomycetous fungal
pathogen of plants. Its life cycle is characterized by a hap-
loid mycelial somatic state, by asexual reproduction via
sclerotia, and by homothallic sexual reproduction with no
evidence of outbreeding observed to-date, all acting to
maintain a predominantly clonal population structure
(Kohli et al. 1992, in press; Kohn et al. 1991). A system
of mycelial incompatibility, possibly a component of the
general phenomenon of vegetative incompatibility (Leslie
1993), has been identified in this species (Kohn et al. 1991).
All isolates of a clone are mycelially compatible with other
isolates of the clone, i.e., able to anastomose and form one
colony, but they are mycelially incompatible with isolates
belonging to other clones. It is not known to what extent
mycelial incompatibility may impede nuclear or cytoplas-
mic exchange between clonal genotypes.
In a previous study (Kohn et al. 1991) of two field pop-
ulations of S. sclerotiorum on canola (Brassica napus L.),
we noted that PCR amplification of the mt S SU rRNA gene
with primers MS 1 and MS 2 (White et al. 1990) yielded
amplification products of either 2 kb or 0.6 kb. Two fea-
tures of this dimorphism were intriguing. First, in one field
167
in which six clones were identified from 33 isolates, the
large 2-kb amplification product predominated. In the sec-
ond field, in which 27 clones were identified from 30 iso-
lates, the small, 0.6-kb amplification product was recov-
ered in a higher percentage of isolates than in the first field.
There seemed to be an association between the presence
of the large product and frequently sampled clones. The
second noticeable feature was that isolates belonged to one
of two groups of clones. In one group of clones, all of the
isolates had the large amplification product, while in the
other group of clones all of the isolates had the small am-
plification product. No clone comprised some isolates with
the large product and some isolates with the small prod-
uct. Our tentative hypothesis was that the dimorphism in
the PCR amplification product was due to the presence
of an intron in the 2-kb large form that was stably
maintained through mitosis in some clonal lineages but ab-
sent in others. We also considered the possibility that the
intron was present primarily in clones sampled at high fre-
quency.
S. sclerotiorum is one of a group of closely related spe-
cies which we believe to share a recent common origin. In
a recent study of the internal transcribed spacer 1 in 40 iso-
lates of 27 species in the Sclerotiniaceae (Carbone and
Kohn 1993), we reported sequence similarity of > 98%
among S. sclerotiorum, S. minor, S. trifoliorum and the
asexual species, Sclerotium cepivorum. These species are
well-defined by a variety of morphological and biochem-
ical characteristics. If an intron in a novel site such as the
mt SSU gene was present in S. sclerotiorum, was it also
highly conserved and present in closely related species?
While the presence of the intron in closely related species
would be consistent with a hypothesis of inheritance from
a common ancestor, the equally interesting possibility of
lateral transfer could not be ruled out.
In this study we address several questions. A basic ques-
tion is whether the dimorphism observed in PCR amplifi-
cation products of the mt SSU rRNA gene in S. scleroti-
orum is due to an intron present in the 2-kb product, but
absent from the 0.6-kb product. Other questions pertain to
clonality and conservation of the intron in S. sclerotiorum,
such as whether the intron is conserved among different
clones and whether the intron is maintained within the same
site in clonal lineages. We also investigate whether this in-
tron, or one similar to it, is present in the same gene of
closely related species in Sclerotinia, in species of closely
related genera in the Sclerotiniaceae, and in other fungi.
The presence of this intron in a novel location in the nat
SSU rRNA gene of several fungi might support either in-
heritance of the intron through common ancestry, or alter-
natively, lateral transfer from closely or distantly related
taxa.
Materials and methods
Cultures and DNA extraction. Isolates of S. sclerotiorurn used in this
study are from previous studies of S. sclerotiorum from canola fields
in Ontario (LMK 198, LMK 212 from clone 2, LMK 205 from clone

168
3, LMK 213 from clone 7, LMK 225 from clone 9, and LMK 232
from clone 11; Kohn et al. 1991) and western Canada (RM 2, RM
144 from clone 67, RM 167, RM 172 from clone 2, FI-ll-L1, F1-
14-3-L2, F1-22-4-L1, F1-31-L2, F1-71-L3 from clone 2, F1-94-L1,
FI-100-L1, FI-lll-L1, FI-ll4-L1, Fl-123-L2 from clone 34, F1-
93-L3, F1-97-L2, FI-100-L2, FI-113-L1 from clone 39, F1-97-L1,
FI-102-1-LI, Fl-107-L3, F2-1 l-L3, F2-14-4-L3, F2-28-L1, F2-35-
L3 from clone 67, F1-95-L2 from clone 69, Fl-106-L2 from clone
70, F1-94-L3 from clone 71, FI-92-L3 from clone 72, F1-95-L1 from
clone 73, F1-107-L2 from clone 77, Fl-104-L2 from clone 78, F1-
118-L1 from clone 83, Fl-102-4-L1 from clone 95 and Fl-128-L1
from clone 258; RM isolates are from Kohli et al. 1992, F1 and F2
are from Kohli et al. in press). Isolates representing other species and
genera in the Sclerotiniaceae are those previously examined by Car-
bone and Kohn (1993). Additional cultures used in this study were:
S. minor (LMK 25, 45, 117), S. trifoliorum (LMK 104 and LMK 625,
626, 627, 628,629, 630, 631,632, 633,634, 635, and 636 from R.
G. Pratt, USDA, ARS, Crop Science Research Laboratory, Missis-
sippi State, Miss., USA), Fusarium soIani f.sp. xanthoxyli (NRRL
22163, with a large PCR amplification product in the MS 1 - MS 2
region, and 22277, with a small product in the same region), Nectria
illudens (NRRL 22090, with a large PCR amplification product in
the MS 1 - MS 2 region) and N. haematococca (NRRL 22180, with
a small PCR amplification product in the same region; all NRRL iso-
lates and information on them from K. O'Donnell, USDA, ARS, Mi-
crobial Properties Research Laboratory, Peoria, Ill., USA), Neuros-
pora crassa (74-OR23-1A from R. A. Collins, Department of Med-
ical Genetics, University of Toronto), Podospora anserina (JS162
from J. A. Scott, Department of Botany, University of Toronto),
Agaricus bitorquis (no. 8-2, with a large PCR amplification product
in the MS 1-MS 2 region and no. 9-3 with a small product in the
same region, from J. R Xu, Department of Botany, University of To-
ronto, Erindale College), and Armillaria gallica (nos. 671 and 690,
both with a large PCR amplification product in the MS 1 - MS 2 re-
gion and no. 722 with a small product in the same region, from J. B.
Anderson, Department of Botany, University of Toronto, Erindale
College).
Cultures were grown and maintained on potato-dextrose agar
(Difco Laboratories, Detroit, Mich., USA) in the dark and at room
temperature (20-220C). Mycelia were grown on liquid complete
yeast medium (Kohli et al. 1992), harvested after 2-4 days and vac-
uum-dried. Total genomic DNA was extracted from 15-20 mg of
vacuum-dried mycelia by the small-scale (mini-prep) method of Zo-
lan and Pukkila (1986) as modified by Kohn et al. (1991). DNA prep-
arations were stored at -20~
PCR amplification, DNA sequencing and analysis. The insertion-
containing segment of the mt SSU rRNA gene was amplified by PCR
with primers MS 1 and MS 2 (White et al. 1990). Within the inser-
tion, locations of additional priming sites for PCR amplifications and
DNA sequencing are indicated in Fig. 1. PCR mixtures (100 ~ll) were
prepared in 0.5-ml polypropylene microcentrifuge tubes by combin-
ing 50 ~tl of a 200-fold dilution of mini-prep DNA and 50 gl of a 2 •
mix of all other reaction components as previously described (Kohn
et al. 1991). Triton X-100 was not added to the reaction mixtures. A
DNA-free control tube was included to detect contamination. To re-
duce evaporation, reaction mixtures were covered with 50 gl of au-
toclaved mineral oil. Samples were then placed in a DNA thermal
cycler and subjected to a 25-cycle amplification (94~ 1 min; 50~
1 min; 72~ 1 min) with a 9-min extension at 72~ on the final cy-
cle. After completion, 10-gl aliqnots were removed and analyzed by
electrophoresis in 1.5% agarose gels buffered with 1 x Tris-acetate
EDTA (TAE). The presence of a single bright band in each lane in-
dicated a successful amplification. The remaining PCR reaction
product (90 gl) was purified using the Geneclean II kit (Bio 101, La
Jolla, Calif., USA) following the manufacturer's instructions and
then eluted in 24 gl of autoclaved distilled water. One microliter (ap-
proximately 50 fmol) of this purified PCR amplification product was
used as a template in subsequent sequencing reactions. Double-
stranded DNA templates were sequenced directly using the chain-
terminating dideoxynucleotide dsDNA Cycle Sequencing System
(Gibco BRL, Burlington, Ontario, Canada) and33P-labeled ATP
(3000 Ci/mmol, DuPont-NEN, Markham, Ontario, Canada) accord-
ing to the manufacturer's protocol. Sequencing reactions were elec-
trophoretically analyzed on polyacrylamide gels containing 6% acry-
lamide, 8 M urea in 1 x Tris-borate EDTA buffer (TBE) for approx-
imately 2.5 h (short run) and 5 h (long run). Between 300-350 nu-
cleotides were obtained from both runs. The MS 1 - MS 2 region of
the mt SSU rRNA gene was sequenced for six strains of S. scleroti-
orum, three with the small (0.6-kb) amplification product (LMK 213,
225,232) and three with the large (2-kb) product (LMK 198, 205,
212). Both complementary strands were sequenced in one strain con-
taining the insertion (LMK 198) in the mt SSU rRNA gene and in
one strain lacking the insertion (LMK 232) in this gene. DNA se-
quences were aligned and compared with the assistance of Gene-
Works (IntelliGenetics, Inc. Mountain View, Calif., USA). Gene-
Works progressively aligns sequences by performing all possible
pairwise comparisons using an alignment algorithm similar to FAS-
TA (Lipman and Pearson 1985). Putative intron-exon splice junc-
tions in the S. sclerotiorum mt SSU rRNA gene were predicted by
sequence comparison with mt SSU rRNA genes from isolates of
S. sclerotiorum lacking the insertion. Computer database searches for
sequence similarity, using the entire intron DNA sequence and the
intron ORF aa sequence as queries, were performed using the Basic
Local Alignment Search Tool algorithm (BLAST, National Center
for Biotechnology Information, Bethesda, Md., USA; Altschul et al.
1990). The intron ORF was translated into polypeptides (single-let-
ter code) using the universal genetic code, with the exception that
UGA encoded tryptophan. Putative intron-encoded proteins contain-
ing segments (50-200 aa long) showing >_ 50% aa similarity to the
putative protein encoded by the intron ORF were retrieved from Gen-
Pept (translated GenBank database). Entire amino-acid sequences
were then globally aligned with GeneWorks. The multiple alignment
was analyzed visually and gaps were inserted where appropriate to
obtain the best visual alignment of all sequences. Regions of con-
served secondary structure within the intron were also taken into con-
sideration in building the final multiple alignment. A secondary
structure model of the putative intron sequence was predicted ac-
cording to the structural conventions for group-I introns (Burke et
al. 1987). Secondary structures within loops were resolved with the
energy minimization protocol of the computer program MulFold ver-
sion 2.0 (Jaeger et al. 1989; Zuker 1989; Jaeger et al. 1990) and
viewed with LoopViewer (Gilbert t990). We amplified three sam-
ples of DNAs with primers MS 1 and MS 2 to screen for patterns of
intron inheritance in the mt SSU rRNA gene. The first group of iso-
lates was screened to determine, from a representative sample,
whether the intron is present in all isolates of a clone or absent in all
isolates of a clone, and to determine whether the intron is found main-
ly in clones with wide geographic dispersal and high frequency in
population sampling. We amplified the MS 1 - MS 2 region in a sam-
ple of seven isolates from clone 2, five isolates from clone 34, four
isolates from clone 39, and nine isolates from clone 67. The second
sample was screened to determine whether the intron is present in
clones occurring at low frequency and among clones detected only
once; we amplified the MS 1 - MS 2 region in a sample of one iso-
late from each of clones 1, 77, 78, 95 and 258. The third sample was
selected to determine whether the intron is present in other species;
we amplified the MS 1 - MS 2 region in a sample of 67 isolates.
When amplifications with MS 1 and MS 2 were unsuccessful, am-
plifications with other primer pairs complementary to highly con-
served regions upstream of and downstream from the intron inser-
tion site were attempted.
Restriction enzyme digestion and Southern hybridization. Southern
hybridizations were also used to screen for the presence of the in-
tron in the mt SSU rRNA gene of other species. Mini-prep DNA
(1-2 gg) was digested separately with each of two restriction en-
zymes, BamHI and PstI, which have no restriction sites within the
S. sclerotiorum intron sequence. Restriction enzyme-digested DNA
was analyzed by electrophoresis in 1% agarose gels buffered with
1 • EDTA (TAE) and transferred onto nylon membranes
(GeneScreen Plus Hybridization Transfer Membrane; DuPont-NEN,
Markham, Ontario, Canada). Membranes were prehybridized for
2-4 h at 65~ in 6xSSC containing 5xDenhardt's and 0.5% SDS.

169
Hybridization probes were made using PCR-amplified segments of P
the intron (Fig. 1). Probes were labeled with [alpha-3~P] dCTP Gcr ~ rm z~ a ...... A CAT[~ czm cTA ~a Lcra PccA N~aC sTcc ~ ~c aTc: Lcra ~0
(3 000 Ci/mmol; ICN Biomedicals, Inc., Mississauga, Ontario, Can- _ .......... ~_ .........
ada) by nick translation (Gibco BRL, Burlington, Ontario, Canada) ~ ~ c~ IaTA cca age ~ a~T cca ~G GCT aTa CGT mC ~ a~ ~ar TaC TCa CGT 320
S F S R K A I R G W S N Y S G
following the manufacturer's protocol. Hybridization was carried out
.......... 2f .........
for 16 h at 65~ in 6 x SSC containing 5 x Denhardt's, 0.5% SDS, zTr Tra ~ra Gar r~r ~ ~ra ~ ~r eta ara ca~ ~G Gra aca a~r CaG ~G a~ aTc ~80
10% dextran sulfate, and 100gg/ml of denatured salmon-sperm F L L D c w L K ~ L ~ v K v T s Q x M
DNA. Membranes were washed first in 2 • SSC at room tempera- c~ A~ Go~ ar~ ~T Q
TAT CGC GC~A TCT AAA TCA GTA ATA CCC CAA AAT ATT ACT GTA AAA 240
ture (RT), then in 2 x SSC - 1% SDS at 65~ and finally in 0.l x ~ R E N D Y R G S K S V I P Q N I T V K
SSC at RT. Membranes were covered with Saran Wrap and exposed R
(X OMAT-AR; USA) c~ cGo G~a cac c~ ~T ~ ~c ~r ~a~ cc~ ~rc cc~ am ~ac ~a~ ~ ~ra a~ 300
to X-ray film - Kodak, Rochester, N.Y., for ~ ~ v ~ ~ ~ w c z ~ p v p ~ ~ ~ ~ ~
l-6 h, or for 16 h at -70~ with an intensifying screen (Dupont --~ ......... s
Cronex Lightning Plus). Removal of the probe from the blot was car- ~ rra am ~ ac~ c~a a~ mc rrc ~ aca ~GT ra~ CAa arc aGO arc c~ ~cr ~ ~60
C L R C T L M G F E R S g Q I R I L S K
tied out by heating 0.1 x SSC - 1% SDS to boiling and washing mem-
branes in the hot elution buffer at RT and then rinsing in 0.1 x SSC ~ ~rn car ~ ~ GOT aca ac~ rzc rat Tcr act zca ~ a~ zac ~rr ~r ~ ~ar ~20
atRT. In order to determine if single or multiple copies of the intron ~ ~ ~ ~ ~ ~ a ~ r r s ~ s ~ i ~ v ~ ~
sequence were located elsewhere in the nuclear and mitochondrial ......... 2~ ..........
genomes of
Sclerotinia
and other fungal species, a dilution series of
I P ~ F L T G ~ I GUA3T ~E G~ cTGT TF~ AGOR A~PrI sTCT
the intron spanning two orders of magnitude above and below the dodecapeptide
gl AC~ a~a ~v: Act a~ ~c~ Ara m~ ~ AGO ~ caa cv: ~r T?c c~z ara ~c ~a Cac ~o
expected quantity of sin e-copy sequences on the blot was in- r ~ v z ~ a i e ~ ~ v e ~ ~ ~ e z ~ 5
cluded as a standard. The estimated quantities of single-copy se ......... ~ .............
quences were based on an estimate of the entire
Sclerotinia
genome
~G ~AG DGAT ATAI A ..... L L A ~A GA .... I RCGT GD ...... Y FTT GmC v~TA ~GA ATGM ..... I H T 800
of about 35 Mbp, comparable to the genomes of other fungi (Griffin
1992).
~
TCT GGG ACA AAT CTC GTC CAA TAT AGA ATA CAG ACT TTT GAC GAA TTA TCT ATT CTA 660
K S G T N L V Q Y R I Q T F D E L S I L
Results
Identification and characterization of the insertion
The three isolates (LMK 198,205,212) of
S. sclerotiorum
with a 2-kb PCR amplification product in the MS 1 - MS 2
region of the mt SSU rRNA gene contained a 1 380-bp
insertion. The DNA sequences flanking the insertion site
were of the same length (564 bp) and nucleotide sequence
as the MS 1-MS 2 sequences from the three isolates with
a 0.6-kb product. No sequence variability was observed
within the insertion. A BLAST database search, using the
564-bp sequence as the query, identified segments
(180-260 bp long) showing > 80% similarity with the mt
SSU rRNA genes of
P. anserina and Aspergillus nidulans.
A BLAST database search, using the 1 380-bp insertion
sequence as the query, identified long segments (200-360
bp) within the sequence showing > 60% similarity with
regions in the
P. anserina
ND3, ND4L, ND5 and COI
subunit genes, as well as with the
N. crassa
ND5 gene.
Subsequent retrieval of these sequences and align-
ment with the 1 380-bp sequence showed that the similar-
ity was confined to the group-I introns located in these
genes.
Translation of the 1 380-bp insertion in
all
six frames
identified five potential ORFs in frame 1 that encode pro-
teins (420 aa, 370 aa, 365 aa, 329 aa, and 322 aa, respec-
tively). Each of the five proteins contain two copies of the
LAGLI-DADG dodecapeptide sequence motif. With the
exception of the shortest ORF, all of the potential ORFs
overlap the core structure. As shown in Fig. 1, the longest
ORF (ORF1) starts at position 25 with an AUG initiation
codon and ends at position 1 287 with a UAA termination
codon. A putative Shine-Delgarno sequence (AGGA) is lo-
cated about ten nucleotides upstream from the translation
initiation codon of ORF1 (Fig. 1).
I L N D Y P V S Q K K W D F L F
....... 2d ............
ATA G~VI ' TCT CTT AAA GCA TCT T?A AAT TTA C-C,A CTA TCT GAA C~A TTG AAA TTA GCG TTT 840
I V S L K a S L N L G L S E A L K L A F
............ id ............
CCT AAT GTA AAG AAT GCC ACG AAA CYT ACA TCT AGT ACT GTA AGT ATT CCT GAT CCT CAT 900
P N V K N A T K L T S S T V S I P D P H
TGA TIT TCA GGG TTT ACT TCT GCT GAG GGT TGT TIT ATG GTA GGT AT? GCT A~ TCT AAA 960
W F S G F T S A E G C F M V G I A K S K
dodecapept ide
T T G Y Q Y L S F I V Q H
EGAA LTPA TTA TTA AAG TGT TTG ATT GAT TAT TTT AGT TGT GGA AGA TTA GCC AGA AAG AGA 1080
L L K C L I D Y F S C G R L A R K R
......... 2c ............
~AT V~ TAT GAG TAT CAG GTT sTCC KAAG T~ TCA GAT GTT GAA KAAG ~FT ATA GAT FTTT FYTC 1140
E Y Q V ~S D V E
....... le ..........
GOT AAA TAC CCT ATC CTT GGA GAA AAG TCC AAG GAC TAT TTA GOT TIT CGT ACG GTG TCA 1200
D K Y P I L G E K S K D Y L D F R T V S
........... 2b- ........
GAA ATA ATG AGA TCA AAA GAT CAT TTA ACT GAA GTA COT GTT CCC AAG GTA CGA ATA Aq~P 1260
E I M R S K D H L T E V G v A g V R I I
~AA ~G GGGT ATG AAT AGA GGT CGG TAA] TTT TAT TAT AGG CAA TYG GAT GAG ATT TTC TCT M N R G R 1320
CTT AAT TCT AAG ATA AAT TTG CTA GGT CAG CAT TTA CGC CTA TAG ATG CTG ATT GCC GTG 1380
Fig. 1 DNA and putative amino-acid sequence (sense strand) of the
1 380-bp group-I intron in the mt SSU rRNA gene of
S. sclerotior-
urn.
ORF1 boundaries are denoted by
square brackets.
Conserved
sequence elements and LAGLI-DADG dodecapeptide motifs are
underlined. Dashed lines
show the annealing sites of primers com-
plementary to the antisense strand (primers beginning with 1) and
complementary to the sense strand (primers beginning with 2).
Probes pR1 and pR2 were synthesized by PCR amplification with
primers 1 a, 2b and lc, 2c, respectively. The intron sequence has been
submitted to the GenBank data library and given the accession num-
ber U07553
Conserved sequence and structure elements
An alignment of the 564-bp sequence of
S. sclerotiorum
with the same region in the mt SSU rRNA gene ofA.
nid-
ulans
showed that the putative group-I intron in
S. sclerot-
iorum
is situated in U3, a region of universally conserved

170
A U U A
Gu_AA
GU-A
U-A
U-A
A-U
A U G- C
PSd
A U A-U
U-A U-A
A G-c U-A
A-U
PSe
A-UG_c U-AG A
U-A U-A U
GC-GGuuuU~A GGUAAC A
i~ [ illl A
GU G CAUAU-AACCAU G
C-G A-U
G-C G-C
PSb
GA-L/ U-A
PSa
A-U C-G
A-U U-A
U A A G C-GAUGAUCG A
G U
IoUGG cU C A A A A
- UC_GUAAAGAG
U-A C-G
G-C U-A
P5
U-A A-U
0-A~ (c u~
U U*t p~A A/Q
C-G*
*
"r G. U (
5' 1"
G'U [G-C[
sp Icesl~
~U'G G G
g-C
P1 ~/G-C
p4
g-C C-G
a-U
P3
C-G P6 AAAu
P~
?C C
5'--agug
................ AAAACUA iC ICAAACU-A-- UCU fAfUAAU~ ~ C
P8
UUGAUGGC A
P7
?GAGA GUCAUUAU A
A
Cc,O
............
I o
..... ....... *?
AU C U UCCGAAUUACG U A G R "~ AUG~A ~/~'~ po n
138(I
AA A A C A AGG A A UUAAUG U G U U U A A G GUGUACUCUAAUGG" UAUCAA-UGCCGUG u u
aaaggg
ua c cu-- 3'
"G.U A-U $
**
S P9
C-G U-A
!
U-A
C-G
3' spli~
site
U-A A-U
GC-~ AG-CG
AA A-U
U-A
Pg,1
uuCC-GAuA
u ,u
i
i
1005 nt
i
with
ORF
Fig. 2 Secondary structure model for the group-I intron in the mt
SSU rRNA gene of S. sclerotiorum. Conserved sequence elements
are underlined and conserved base-paired regions are indicated by
P1 to P9. The residues denoted by asterisks can base-pair to form
P10. The long-range P9.0 pairing is also indicated. Exon sequences
are shown in lower case letters and intron sequences in upper case
letters. Putative splice sites at the 5' and 3' exon-intron boundaries
are shown. The arrow designates the start of the putative intron
ORF1. The dot between bases indicates alternative base-pairings
primary sequence (Cummings et al. 1989a). Analysis of
the insertion DNA sequence (Fig. 1) and secondary struc-
ture model (Fig. 2) demonstrated that the insertion con-
tained four features that are known to be highly conserved
among group-I introns. First, the insertion contains the P,
Q, R, and S sequence elements that allow the formation of
the characteristic group-I core structure. Second, the last
upstream exon base preceding the 5' intron splice site is a
U, and the last downstream intron base preceding the 3' in-
tron splice site is a G. Third, the U at the 5' splice site is
paired with a G in the IGS forming P1. Exon bases flank-
ing the 3' splice site can potentially also pair with the IGS
and form a P10 base-paired region, which is present in
most, but not all, group-I introns (Michel and Westhof
1990). Lastly, as in all group-I introns, base-pairing of the
conserved sequence elements P, Q, R and S, and folding of
the entire 1 380-bp sequence (Fig. 2) showed that base-
paired regions P3 through P9 are present. Further analysis
of the intron in the mt SSU rRNA gene of S. sclerotiorum
(Ss.mtSSU intron) has shown that it shares a number of
secondary structure features typical of all introns within
23
22 i g -e
ia-u
',u-a
f ~ 3'
a
/
5'
/
866
39
24 3S
1
18
i
io~ 9
V9
b v2
V7
12
43
4~ V8
f
47
Fig. 3a, b Location of the Ss.mtSSU intron within the secondary
structure model of SSU rRNAs. Helix numbering is according to
Neefs et al. (1993). a Secondary structure model showing location
and orientation of intron in the overall secondary structure model.
Nucleotides in bold are part of the conserved SSU rRNA sequence
(U3). Nucleotides at positions 570 and 866 are involved in tertiary
interactions. Dashed lines indicate predicted hydrogen-bonding
interactions between the intron IGS and the 5' half of helix 22. b Sec-
ondary structure model for the mt SSU rRNA gene from A. nidulans;
bold lines indicate universally conserved secondary structure regions
of eukaryotic SSU rRNAs (Gutell et al. 1985). The arrow indicates
the location of the insertion site of the Ss.mtSSU intron. The most
variable regions in the secondary structure model are labeled V1
to V9
subgroup-IC2 (Michel and Westhof 1990). These include:
(1) the absence of base-paired region P2, (2) the unpaired
G in element Q, (3) the bulged C in element R, and (4) a
long P5 loop with the characteristic adenine-rich bulge on
the 3' side of P5a (Collins 1988).
Group-I introns in rRNA genes are inserted in highly
conserved regions of primary sequence and secondary
structure (Turmel et al. 1991). The Ss.mtSSU intron lies
within a region of highly conserved primary sequence
(U3), a region of highly conserved secondary structure
(between helices 22 and 23), and also a region of highly

Fig. 4 Amino-acid sequence
alignment of the putative pro-
teins encoded by ORFs of in-
trons in
P. anserina
and N.
crassa
with the putative 420 aa
protein encoded by ORF1 in
the Ss.mtSSU intron. The
brackets
indicate the conserved
LAGLI-DADG dodecapeptide
sequence motifs. Additional re-
peated sequences are denoted
as
r A
and r m Residues that are
common to at least four se-
quences appear in the consen-
sus sequence. The
boxed
resi-
dues are common to all se-
quences.
Dashes
are used to de-
note alignment gaps. Names of
introns are abbreviated as fol-
lows; the first two letters stand
for the organism
(Pa
is
P. anse-
rina; Nc
is
N. crassa; Ss
is
S. sclerotiorum),
subsequent
characters are for the gene that
is interrupted
(ND
and
CO
are
the NADH dehydrogenase and
cytochrome oxidase subunit
genes, respectively, and
mtSSU
is the mt SSU rRNA gene) and
the
number
following the letter
i indicates the rank of the intron
in genes with multiple introns
Ss.mtSSU
Pa.ND3il
Pa.CoIi5
Pa.ND5i2
Nc.ND5il
Pa.ND4Lil
Pa.ND4Li2
Consensus
Ss.mtSSU
Pa.ND3il
Pa.COIi5
Pa.NDSi2
Nc.NDSil
Pa.ND4Lil
Pa.ND4Li2
Consensus
Ss.mtSSU
Pa.ND3il
Pa. COIi5
Pa.ND5i2
Nc.ND5il
Pa.ND4Lil
Pa. ND4Li2
Consensus
......... MCLKLPNSGDILKLLIPSFSRKAIRGW~S~FLLDCWLKNLIDKVTSQKM 51
K-NWTGKSINWVKLSNSGKTLKLLILSINRKVNCGW~N~S~I ........... VISQKI 47
A-KWFRESLLWEKLSNSGEALKLLIPNLVWKYGGGR~N~S[M ............ VTSQKM 47
RGSVEGIRFYGGKLSNSGELLKLKVPSCIWKIISGW~S~M ........... VTSLKI 48
RGRVRGKRVYGDKLSNSGEVLKLKVPSCSWKTMSGW~S~T ............ VTSLKM 48
.............. INSPDK ............ NPR~N~S[SHFNLLPTISRRSRNLIQS 33
................ ASN ......... YTFSCKNT~N~S~CKTLNKVSDLKAKNRLLSV 35
.- .......... KLSNSG..LKL..PS...K...GW ~ .- ........... VTS.K. 60
IEREMDYRGSKSVIPQNI-TVKEQRVDGSWCIKPVPKN KELMCLRCTLMGFERSYQIR 108
YESIIGNRGSKSAL---L-SVKEQRVYGSCT ......... NNLV-LRCTLVGLEISYQAK 93
IEKKMGNRGSKSAL-NKN-TVKEQRVDGSCINSFLVKKFIKKGFMLRCTLMGSERNyQVK 105
CENK~IDNRGSKSTKLKVN-WKEQRVNGSWSAKSNLTN ....... LRCTLTGFEKNRGIN i00
SENKMDNRGSKSVVIDSNSTVKEQRVDGSWSIKSHLMD ....... LRCTLRGFERNRGIK i01
RNYSSL ......... SPRTSAV .................. CSVRS .......... EGER 55
RPKTFV .......... RNYS .................... HLI ............ QKRV 52
.E..M.NRGSKS..-..N-.VKEQRV.GS ......... --.~ ...... 120
ILSKELHKKDRRFYSTS ---N IYVNKN ........ I@FL@~ISLTKVT 156
IPSNSINFK-RNF-ST ....... LESK ......... L~P~YI~G[~E~F~LTIIKDN 135
ALSNQITNK- - RFYSSSLPKSVKLSKET ........ L~ P~ I'Tg~[ ~D~E~ ~F~ I R I RKN S 155
~~wi~vv~~p~THq~p~i~w~ ~0
Kzv .................................
L~FI~[~]9~rrILNNP S2
S ................................ L~P~FI~G~D~E~qF~IGL FI IGVRRDP 77
...N ............ S ................... L~_-~.~F. I....N. 180
I I
dodecapepdde
i71
Ss.mtSSU
Pa.ND3il
Pa. COIi5
Pa.ND5i2
Nc.ND5il
Pa.ND4Lil
Pa.ND4Li2
Consensus
R A ~--~VQL~
KYK~C~
KYK A\
TRK IEP}
SRK L}
9 . K~VQ..
I
~LGLH D
~ I KLH~KD~Aq]
~I. LH K~. ~ L~
{DIRDYFG-~IHK-SGTNLVQYRIQTFDE-LSIL 212
qKIKEFFD-~G~VFL-MTKDSAQYRVESLKG-LDLI 192
{SIKTFFGG~G[IKK-HGKGTFSYRIESSEQIMKFI 214
LQLQLFLGG~G~IYSARNGEFVNYIISSIKD-LNKL 219
{LLQQYLGG~IHLARNRDIVNYSIDSIED-LNKL 217
~LIQNYFYN~G~ISTLKGGSTVEFRVSDITSLNNII 142
<AIQKYFGG~VKI-RG-DKCTFIVSSLSQITKVI 135
..IQ.YFGG.~.I .... G...V.YRI.S .... L..I 240
Ss.mtSSU
Pa.ND3il
Pa. COIi5
Pa.ND5i2
Nc.ND5il
Pa.ND4Lil
Pa.ND4Li2
Consensus
INHLN~S~EL~AHELVKMNgNK~LK IVS L~--~E~LAF 272
INHFD~Y~L~T~K~D~KL~K~AHNLIKNK~H~TK~LELVAI~V~GL~N~L~IAF 252
IPFFD~Y~T~K~LL~K~WEMLNNK~H~TE~YKIVSL~GL~E~L~AAF 274
VVHLEI~I~I~FFI~IIIDLINN~TV~KIVNL;I~I~I~ILEF 279
IIHLE FL VKLVNN TV QIVNI SEL 277
IPHFE~L~ T.~KK ~I I~KK !VSLMSE!.~TL~!EILEY~.~L~ D~L~ ESF 202
IPHFD FL IIEIMKAK TL NEIVRM D TAF 195
I.HF. .L ..L..N T. .IV.. .AF 300
| !
rA
Ss.mtSSU
Pa.ND3il
Pa. CoIi5
Pa.ND5i2
NC.NDSil
Pa.ND4Lil
Pa.ND4Li2
Consensus
P ATKLT SS SIPD PJ{ I OIAK K S 0 VTO
PGINTI LRPD- TSLPQ I LN- - P~F~D~E~qF~VVVFKSKTSK~AV~L~F~LTQS 309
PQC I PVFRPT-VYNKI I PD-- PN~A~F~SqE~qF~S ILKKSS SVKV~QS ~L~F~VTQH 331
PGyQVMER P I FNC I EVAAN I S P~ ~ F~S~E~F~VRTPK- TNS K~RV~L~F~ I TQH 338
AGYTPVERPVINCDNVFLD- - P~W~ ~F~S~E~F~VRMPS - TNSKL~[RVqL~F~ I STL 334
PS IVPVKRVE- IEDNILSNLPS~W~GF~AqE~F~ ITI S ..... ~KV~L~F~ IAQD 254
AETFSTSA I IKR ILNKES I PHG~FrS~E~F[VNIFKS SHHK~Q I~L~F~ IAQH 255
P ...... RP ............ P.~. S~_~VSA~3.~/V...KS.. SK.~Y.V.~J.m. ITQH 360
dodecapeptide

172
Ss.mtSSU
Pa.ND3il
Pa. COIi5
Pa.ND5i2
Nc.ND5il
Pa.ND4Lil
Pa.ND4Li2
Consensus
VRDE~KCL IDY F S ~--~LA- RKRDV-y EYQVS KF SDV-EKFID~KYPI~L 386
NRDE~LFKSL IEYLG~TSLDPRGT- IDFKVTNFSS I KDI IVP~F~KY PL~K~D~T 368
ARDE~L~ESL I SYFK~G~ IEKDPRGPWLSY IVSNFTDIYTKI I P~FFQYN I ~G~K~D~N
391
VRD I ~L~EKLVQYLG~G~VYKy - TR SGVHL S IVDF S L I TNR I I PIF~EN PL~G~K~ ~D~E 397
RVN I~L~EK IVEYFG~G~ IYKYGGKSAVSLTIVDFTDITNILVP~F~KYp I ~G~K~D~L 394
SRDI~L~KSLVEFFN~G~ IAQYK~RKVCEF IVTKINDI I Iy I I P~F~KYKI ~G~K~fV 314
IRDE~L~LSFTEFFG?~ I SRYSSDM-VKFRCTKFSDI STI IVP~F~EYLP~HK~Q 314
.RDE. L~. SL. EYFGO~. I..y ......... V.. FSDI... IIPFt~. KYPI.~LIK~. N_D ~. 420
Fig. 4. (continued)
Ss.mtSSU
Pa.ND3il
Pa. COli5
Pa.NDSi2
Ne.NDSil
Pa.ND4Lil
Pa.ND4Li2
Consensus
DFRTVSE IHRSK~AKVR I I KQGbINRGR ................
DFCEVVRLMENK ~HL~K~GbDQ I KK I RNR~TNRK .............
DWCK IATLHQDK~HLT~G~NKI I S I KGGMNKGRLSSVSGAQAPHPTHSR
DWCK IHEL~VNR~HLT~ ~G~NF IQE I KLGMNRGR ........... SIETE
DWCKIHSLMINR~HL~U~r~G~NS I SLL- LGRRR ..................
NFKEAA I PI KNK~HLT~G~NK I IELKNSMNKND
................
DFCKVCELMENN?L~NQ IRE IKARMNSFRK
.............
DFCK. . . LN. NK. HI.H_~. I~N. I. . IK.GNN. .R ................
i !
r B
420
403
441
436
425
348
349
480
conserved tertiary structure (570/866 interaction) of the mt
SSU rRNA secondary structure model (Fig. 3; Gutell et al.
1985, 1986). The insertion sites of many group-I introns
have been assigned functional roles suggesting that the in-
trons lie in exposed regions of the ribosome (Stern et al.
1988; Turmel et ah 1993). The Ss.mtSSU intron is inserted
at a novel site within the SSU rRNA secondary structure
model at which no other introns have been reported and
for which no functional role has been described.
Analysis of the ORF
A BLAST database search, using the predicted amino-acid
sequence encoded by ORF1 as the query, identified seg-
ments (60-180 aa long) showing > 50% aa similarity with
the putative proteins encoded by the ORFs of IC2 introns
in P. anserina (ND3i 1, ND5i2, and COIi5) and in N. crassa
(ND5i 1), as well as with proteins encoded by ORFs of IC 1
introns in P. anserina (ND4Lil and ND4Li2). An align-
ment showing three areas of similarity among the amino-
acid sequences of the putative proteins encoded by ORF1
and the N. crassa ND5il and P. anserina ND3il, ND4Lil,
ND4Li2, ND5i2, and COIi50RFs is shown in Fig. 4. First,
all of the proteins contain two well-conserved copies of the
LAGLI-DADG dodecapeptide sequence motif, and a sec-
ond set of repeated sequences (denoted as r A and rB) lo-
cated at the same distance downstream from each dodec-
apeptide sequence. Second, regions downstream from each
dodecapeptide sequence are very similar to each other in
length and aa sequence. Third, upstream of the first dodec-
apeptide sequence, the putative protein encoded by the
ORF1 of the Ss.mtSSU intron is very similar in length and
aa sequence to the putative proteins encoded by the ORFs
ofP. anserina ND3il, ND5i2, COIi5 and N. crassa ND5il.
Although the other four putative proteins encoded by the
S s.mtS SU intron display two copies of the LAGLI-DADG
motif, the protein encoded by ORF1 is favoured as the po-
tential product because its 5' end is similar in length and
secondary structure to putative proteins encoded by the
ORFs of P. anserina ND3il, ND5i2, COIi5, and N. crassa
ND5il. The looping-out of these ORFs in P9.1 is highly
conserved among their secondary structure models. Fur-
thermore, all of these ORFs overlap the core structure and
therefore include the conserved P, Q, R and S sequence ele-
ments. The ORFs of P. anserina ND4Lil and ND4Li2 are
located in loops that are external to the core and therefore
do not contain the R Q, R and S elements. ORFs emerge
in Pl in the IC1 introns of P. anserina (ND4Lil and
ND4Li2) (Michel and Westhof 1990).
Association of intron occurrence with clone frequencies
in population samples
A further screening of S. sclerotiorum from field popula-
tions in western Canada found no obvious association
between clone frequencies in population samples and the
presence or absence of the intron (data not shown). The
2-kb amplification product was produced by all isolates
representing three clones of S. sclerotiorum occurring at
high frequency (seven isolates from clone 2, five isolates
from clone 34, and nine isolates from clone 67) in western
Canada. However, all of the isolates sampled from another
clone occurring at high frequency (four isolates from clone
39) produced the 0.6-kb amplification product. A 2-kb
product was also observed among clones occurring at low
frequency (one isolate from clone 78, one isolate from
clone 95) and among clones detected only once in our sam-
ples (clone 77 and clone 258). In all cases, all of the iso-
lates tested within a clone had either a 2-kb or a 0.6-kb am-
plification product, as previously observed (Kohn et al.
1991).
Screening for this intron in the mt SSU rRNA gene of
other fungal species
Whole-cell DNAs of isolates LMK 198, 205, and 212,
known to contain the intron in their mt SSU rRNA gene

("intron-containing isolates"), and LMK 213,225 and 232,
known to lack this intron ("intron-lacking isolates"), were
digested separately with
BamHI
and
PstI
and were then
probed with intron and mt SSU rRNA gene sequences in
Southern hybridizations. Probe pR1, containing an
1 154 bp sequence which includes the highly conserved Q,
R, and S elements, as well as the majority of the intron
ORF (Fig. 1), hybridized weakly to two small
BamHI
frag-
ments (either 4 kb and 5.2 kb or 5.2 kb and 6 kb) in all iso-
lates and strongly to a high-molecular-weight
BamHI
frag-
ment (23 kb) present only in the intron-containing isolates.
When the same blots were re-hybridized with another
probe, pR2, containing a smaller, less-conserved element,
only the 23-kb
BamHI
fragment in the intron-containing
isolates could be detected. This suggested that the smaller
BamHI
fragments common to these isolates may either
represent closely related group-I introns or artifacts of
nonspecific hybridization. Probes pR1, pR2, and pRint,
which contains 258 bp of the mt SSU rRNA gene starting
74 bp downstream from the 3' intron splice site, each hy-
bridized to a 23-kb
BamHI
fragment. This indicated that
the 23-kb
BamHI
fragment contained both the group-I in-
tron and the 3' end of the mt SSU rRNA gene of
S. scle-
rotiorum.
When genomic DNAs of intron-containing and intron-
lacking isolates of
S. sclerotiorum
were digested with
PstI,
pR1 hybridized to a single fragment, which was slightly
larger in the intron-containing isolates (15 kb) than in the
intron-lacking isolates (13 kb). These blots were then hy-
bridized with pRint to show that a
PstI
fragment of the
same electrophoretic mobility contained the 3' end of the
mt SSU rRNA gene. When the blots were probed with pR2,
only the larger
PstI
fragment (15 kb) of the intron-contain-
ing isolates was detected. This suggests that at least one
other group-I intron, hybridizing only with the highly con-
served core sequences of pR1 and not with the less con-
served intron sequences of pR2, is inserted in another site
within the mitochondrial genome of
S. sclerotiorum.
This
group-I intron is present in both Ss.mtSSU intron-contain-
ing and Ss.mtSSU intron-lacking isolates of
S. scleroti-
orum,
on the same
PstI
fragment as the mt SSU rRNA gene,
but on a different
BamHI
fragment. It was estimated that
all fragments hybridizing to the probe containing mt SSU
rRNA gene sequences were present at about ten fold the
quantity of single-copy nuclear DNA sequences. Results
from hybridization experiments demonstrate that in order
to screen whole-cell DNAs for a specific group-I intron,
smaller probes from less-conserved intronic regions, such
as pR2, are needed.
Results from hybridizations of either
BamHI-
or
PstI-
digested DNAs with pR2 were consistent with results from
PCR amplifications using primers MS 1 and MS 2, thereby
providing an additional screening technique for this
group I intron. The smaller, 0.6-kb product from PCR am-
plifications, indicative of the absence of the intron, was
observed in a sample of isolates from other species in the
Sclerotiniaceae. These included
S. trifoliorum
(LMK 36
and 47),
S. minor
(LMK 118),
Botrytis
spp. (LMK 18, 20,
21,521 and 523),
Myriosclerotinia dennisii
(LMK 0-7)
173
Fig. 5 Southern hybridizations of
PstI-digested
total DNAs with
probe pR2.
Lanes
1-5 are
S. sclerotiorum
(LMK 198, 213, 2, 44, and
57),
6-10
are
S. minor
(LMK 25, 45, 115, 117, and 118),
11-14
are
S. trifoliorum
(LMK 36, 47, 55, and 629), and
lane 15
is
Sclerotium
cepivorum
(LMK l).
HindIII-digested
bacteriophage lambda DNA
is in the leftmost lane
and
M. scirpicola
(LMK 0-2),
Verpatinia calthicola
(LMK
448),
Cristulariella moricola
(LMK 161),
Monilinia meg-
alospora
(LMK 415) and
M. fructigena
(LMK 431),
Lam-
bertella langei
(LMK 400), and
Ciboria caucus
(LMK
501). The larger, 2-kb amplification product, indicative of
the presence of the intron, was observed in
S. sclerotiorum
(LMK 2, 44, 47) and in one isolate of
S. trifoliorum
(LMK
55). The presence of the intron within the 2-kb amplifica-
tion product was confirmed both by subsequent PCR am-
plifications with intron-specific primers and by hybridiza-
tions with pR2 (data not shown). Amplifications with prim-
ers MS 1 and MS 2 and with other conserved primer pairs
that lie outside the intron insertion site failed to amplify
the potential intron-containing segment of the mt SSU
rRNA gene in some isolates of
S. minor
(LMK 3 and
115),
Sclerotium cepivorum
(LMK 1, 71 and 126), and in
many of the isolates representing other species in the Scle-
rotiniaceae. Amplifications using a variety of intron-spe-
cific primer pairs (Fig. 1) were also unsuccessful. The
failure to amplify may have been attributable to too
many mismatches between priming sites and template
DNAs.
Although amplifications failed in some isolates, hybrid-
izations using pR2 and
PstI-digested
total DNA" (Fig. 5)
were successful in detecting this intron in the mt SSU rRNA
gene of four isolates of
S. minor
(LMK 25, 45, 115, 117),
two isolates ofS.
cepivorum
(LMK 1,126) and the one iso-
late of
Rutstroemia petiolorum
(LMK 402).
R. petiolorum
showed two hybridizing bands (4 kb and 6 kb) of lower
molecular weight than the single hybridizing band ob-
served in
Sclerotinia
species and in
S. cepivorum.
Amplifi-
cations with intron-specific primers, and hybridizations of
whole-cell
PstI-digested
DNAs probed with pR2, did not
detect the Ss.mtSSU intron in
F. solani
(NRRL 22163), N.
illudens
(NRRL 22090),
A. bitorquis
(no. 8-2), andA.
gal-
lica
(nos. 671 and 690), and there was no evidence for the
presence of this group-I intron in the genomes of
S. trifol-
iorum
(LMK 104 and LMK 625-636),
F. solani
(NRRL
22277),
N. haematococca, N. crassa, A. gallica
(no. 722)
and
P. anserina,
using these screening techniques.

174
Discussion
The presence of the conserved R Q, R and S elements iden-
tifies the 1 380-bp insert in the mt SSU rRNA gene of
S. sclerotiorum
as a group-I intron. This is the first report
of an intron in the mt SSU rRNA gene. The Ss.mtSSU in-
tron lies at a novel site within SSU rRNAs and has an ORF.
In comparison, group-I introns within the nuclear SSU
rRNA gene range in size from about 200 to 500 bp and lack
ORFs, with the exception of the two group-I introns from
amoebas that are over 1 000 bp long and have ORFs (Em-
bley et al. 1992; De Jonckheere 1993).
The Ss.mtSSU intron encodes a putative protein of 420
aa which contains two copies of the LAGLI-DADG motif.
The close similarity between the first and the second half
of the putative protein encoded by ORF1 in the Ss.mtSSU
intron suggests that the two copies of the LAGLI-DADG
dodecapeptide motifs, and the aa sequence that extends
downstream from them, may have arisen from an ances-
tral monomeric protein via an internal duplication. The
Ss.mtSSU intron shows close similarity in secondary struc-
ture to the IC2 introns of
P. anserina
(ND3il, ND5i2, and
COIi5) and
N. crassa
(ND5il). The putative protein en-
coded by ORF1 is very similar to the putative proteins en-
coded by the ORFs of these IC2 introns, as well as to the
putative proteins encoded by the ORFs in the IC 1 introns,
ND4Lil and ND4Li2, in
Podospora
(Fig. 4; Cummings et
al. 1989b). All of these ORFs encode putative proteins that
share very similar aa sequences downstream from each do-
decapeptide motif. The second halves of the proteins are
also similar to the last 166 aa encoded by the ORF of the
IB4 intron in the chloroplast LSU rRNA gene of the green
alga,
C. eugametos
(Turmel et al. 1991). This group-I in-
tron has been shown to be genetically mobile and to code
for a site-specific endonuclease that could mediate its
transposition (Gauthier et al. 1991; Turmel et al. 1991).
Further studies of the self-splicing and mobility capabil-
ities of the Ss.mtSSU intron are required to determine if
its ORF encodes a maturase, an endonuclease, or both. The
similarity of the putative protein encoded by ORF1 of the
Ss.mtSSU intron to the site-specific endonuclease in
C. eu-
gametos
raises questions regarding its potential role in hor-
izontal intron transfer (transposition).
The Ss.mtSSU intron is located at a position in the mt
SSU rRNA gene where no group-I introns in nuclear SSU
rRNAs have previously been reported. Like the group-I in-
trons reported from the nuclear SSU rRNA gene, however,
the Ss.mtSSU intron lies within highly conserved regions
of primary sequence and potential secondary structure
(Fig. 3). The insertion site of the Ss.mtSSU intron is lo-
cated between helices 22 and 23, which are separated by
seven nucleotides. A comparable placement of an inser-
tion site between two helices has been observed with the
group-I intron in the nuclear SSU rRNA gene of
Ustilago
maydis,
which is located between helices 33 and 34 (De
Wachter et al. 1992). In
S. sclerotiorum,
the intron is lo-
cated in a region of the rRNA molecule that is structurally
constrained at the 5' end by secondary structure require-
ments (stem of helix 22) and at the 3' end by tertiary inter-
actions between positions 570 and 866. These interactions
are known to maintain the overall geometry of mature
rRNA (Gutell et al. 1985, 1986) and would compete for
hydrogen-bonding specific to the intron structure. Hydro-
gen-bonding is predicted to occur between the IGS and the
5' half of helix 22 (bold nucleotides in Fig. 3). Further co-
axial stacking within the intron could bring splice sites in
close proximity, conceivably facilitating splicing of the in-
tron (De Wachter et al. 1992).
On the basis of similarity in secondary structures and
ORFs (Figs. 2 and 4), the Ss.mtSSU intron is most closely
related to IC2 introns, specifically
P. anserina
ND3il,
ND5i2, COIi5, and
N. crassa
ND5il. The conserved loop-
ing-out of the ORFs of these IC2 introns in the terminal
helix of P9.1 suggests a possible functional significance
for this structure (Cummings et al. 1989b). Both base-
paired structures P9 and P9.1, located between the two
halves of P9.0, have been shown to influence 3' splice-site
selection (Van der Horst and Inoue 1993). The secondary
structures of the Ss.mtSSU intron and the IC2 introns of
Podospora
and
Neurospora
are very similar. The putative
protein encoded by ORF1 of the Ss.mtSSU intron is simi-
lar to other proteins in the IC2 subgroup. These proteins
are also similar to those coded by ORFs in the IC1 and IB4
intron subgroups. This suggests that the portion of the in-
tron containing the conserved sequence elements that al-
low the formation of the characteristic group-I intron core
structure (whether IC 1, IC2, or IB4) may have evolved in-
dependently from the protein-coding region. A comparable
lack of association between the intron secondary structure
and the ORF has also been noted by Mota and Collins
(1988), who found group-I introns located at the same po-
sition in the ND1 gene ofN.
crassa
and
N. intermedia
with
very similar secondary structures but completely dissimi-
lar ORFs.
Screening for the Ss.mtSSU intron in isolates from
clones recovered in population samples also raises ques-
tions about the evolution, transmission and inheritance of
this intron. Given that, in each clone, all isolates either have
the intron or all isolates lack the intron within the mt SSU
rRNA gene, does mycelial incompatibility between clones
preclude the possibility of transposition? Can the intron in-
fect mycelia that lack it by cytoplasmic exchange in anas-
tomosis ? Acquisition or loss of this intron was not detected
even in the many clones with widespread geographical dis-
persal. Events of loss or gain must have occurred when
clones originated or when clones were only locally dis-
persed; existing patterns of presence or absence cannot dis-
tinguish between gain or loss of this intron in the past. Both
in vitro and in planta pairing studies, as well as phyloge-
netic studies of clonal evolution, may resolve these issues.
Based on hybridization data (Fig. 5), the existence of
this group-I intron in the mt SSU rRNA gene seems to be
restricted to
S. scIerotiorum, S. minor, S. trifoliorum
and
Sclerotium cepivorum.
However, the ability to successfully
amplify the intron using intron-specific primers in
S. scle-
rotiorum
and
S. trifoliorum,
but not in
S. minor
and
Scle-
rotium cepivorum,
suggests that introns in the latter two

175
species, although similar, are not identical to the introns in
S. sclerotiorum
and
S. trifoliorum.
Neither the inability to
amplify with intron-specific primers, nor the absence of
hybridization signals with probes specific to the Ss.mtSSU
intron, necessarily rule out the possibility that introns are
present. Introns may be highly variable even in conserved
regions. In Southern hybridizations of
Sclerotinia
species,
the variability in signal intensity of fragments (Fig. 5) may
indicate some sequence divergence among introns; the var-
iability in fragment size among species may be due to point
mutations, insertions, deletions, or rearrangements within
their mitochondrial genomes. The existence of this group-
I intron in
S. sclerotiorum, S. trifoliorum,
and potentially
also in
S. minor
and
Scletvtium cepivorum,
is consistent
with results from previous studies (Carbone and Kohn
1993) which suggest that these species form a clade shar-
ing a common, recent evolutionary origin. Although a close
relationship among these well-defined fungal species is
supported by other biochemical and morphological char-
acters it is also possible that the intron has been inherited
from more-distantly related fungi or other organisms via
lateral transfer.
The highly similar secondary structure models of the
Ss.mtSSU intron, Pa.ND3il, Pa.ND5i2, Pa.COIi5 and
Nc.ND5il suggest any of three possible scenarios. First, it
is tempting to conclude that
ScIerotinia, Podospora
and
Neurospora
are related to each other, perhaps descended
from an ancestral filamentous ascomycete that had an IC2
intron within the mt genome, but because these introns are
in different sites the argument for common ancestry is ten-
uous. Another possibility is that some lineages of fungi,
for example some groups of filamentous ascomycetes, are
cytoplasmically receptive to this group of introns; introns
may have been acquired by lateral transfer, later in the ev-
olution of these fungi. Third, introns similar to these
may be more ubiquitous in fungal mitochondrial genomes
than is indicated by investigations of mitochondrial ge-
nomes in the relatively limited number of fungi studied to-
date.
Acknowledgements We thank Rick Collins for help with the sec-
ondary structure model and many helpful suggestions. We also thank
Ban'y Saville for a critical reading of the manuscript.
References
Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ (1990) Ba-
sic local alignment search tool. J Mol Biol 215:403-410
Burke JM (1988) Molecular genetics of group-I introns: RNA struc-
tures and protein factors required for splicing - a review. Gene
73:273-294
Burke JM, Belfort M, Cech TR, Davies RW, Schweyen RJ, Shub DA,
Szostak JW, Tabak HF (1987) Structural conventions for group-
I introns. Nucleic Acids Res 15:7217-7221
Carbone I, Kohn LM (1993) Ribosomal DNA sequence divergence
within internal transcribed spacer 1 of the Sclerotiniaceae. My-
cologia 85:415-427
Cech TR (1988) Conserved sequences and structures of group-I in-
trons: building an active site for RNA catalysis - a review. Gene
73:259-271
Collins RA (1988) Evidence of natural selection to maintain a func-
tional domain outside of the 'core' in a large subclass of group-
I introns. Nucleic Acids Res 16:2705-2715
Cummings D J, Domenico JM (1988) Sequence analysis of mitochon-
drial DNA from
Podospora anserina.
Pervasiveness of a class-I
intron in three separate genes. J Mol Biol 204:815-839
Cummings DJ, Domenico JM, Nelson J, Sogin ML (1989a) DNA se-
quence, structure, and phylogenetic relationship of the small sub-
unit rRNA coding region of mitochondrial DNA from
Podospo-
ra anserina.
J Mol Evot 28:232-241
Cummings D J, Michel F, McNally KL (1989b) DNA sequence anal-
ysis of the 24.5-kilobase pair cytochrome oxidase subunit I mit-
ochondrial gene from
Podospora anserina:
a gene with sixteen
introns. Curr Genet 16:381-406
Cummings DJ, McNally KL, Domenico JM, Matsuura ET (1990)
The complete DNA sequence of the mitochondrial genome of
Po-
dospora anserina.
Curr Genet 17:375-402
Davies WR, Waring RB, Ray JA, Brown TA, Scazzocchio C (1982)
Making ends meet: a model for RNA splicing in fungal mito-
chondria. Nature 300:719-724
Dfivila-Aponte JA, Huss VAR, Sogin ML, Cech TR (1991) A self-
splicing group-I intron in the nuclear pre-rRNA of the green al-
ga,
Ankisttvdesmus stipitatus.
Nucleic Acids Res 19: 4429-
4436
De Bi~vre C, Dujon B (1992) Mitochondrial DNA sequence analy-
sis of the cytochrome oxidase subunit I and II genes, the ATPase9
gene, the NADH dehydrogenase ND4L and ND5 gene complex,
and the glutaminyl, methionyl and arginyl tRNA genes from
Tri-
chophyton rubrum.
Curr Genet 22:229-234
De Jonckheere JF (1993) A group-I intron in the SSU rDNA of some
Naegleria
spp. demonstrated by polymerase chain reaction am-
plification. J Euk Microbiol 40:179-187
DePriest PT, Been MD (1992) Numerous group-I introns with vari-
able distributions in the ribosomal DNA of a lichen fungus. J Mol
Biol 228:315-321
De Wachter R, Neefs J-M, Goris A, Van de Peer Y (1992) The gene
coding for small ribosomal subunit RNA in the basidiomycete
Ustilago maydis
contains a group-I intron. Nucleic Acids Res 20:
1251-1257
Embley TM, Dyal P, Kilvington S (1992) A group-I intron in the
small subunit ribosomal RNA gene from
Naegleria andersoni
ssp.
andersoni
strain PPMFB-6. Nucleic Acids Res 20:6411
Gauthier A, Turmel M, Lemieux C (1991) A group-I intron in the
chloroplast large subunit rRNA gene of
Chlamydomonas euga-
rectos
encodes a double-strand endonuclease that cleaves the
homing site of this intron. Curr Genet 19:43-47
Gilbert DG (1990)
LoopViewer,
a Macintosh program for visualiz-
ing RNA secondary structure. Published electronically on the
Internet, available via anonymous ftp to iubio.bio.indiana.edu.
Griffin DH (1992) Nucleic acids and nucleotides. In: Arora DK,
Elander RR Mukerji KG (eds) Handbook of applied mycology,
vol 4: fungal biotechnology. Marcel Dekker, Inc., New York, pp
445-473
Gutell RR, Weiser B, Woese CR, Noller HF (1985) Comparative
anatomy of 16s-like ribosomal RNA. Prog Nucleic Acid Res Mol
Biol 32:155-216
Gutell RR, Noller HF, Woese CR (1986) Higher order structure in
ribosomal RNA. EMBO J 5:1111-1113
Hensgens LAM, Bonen L, De Haan M, Van der Horst G, Grivell LA
(1983) Two intron sequences in the yeast mitochondrial COX1
gene: homology among URF-containing introns and strain-de-
pendent variation in flanking exons. Cell 32:379-389
Jaeger JA, Turner DH, Zuker M (1989) Improved predictions of sec-
ondary structures for RNA. Proc Natl Acad Sci USA 86:
7706-7710
Jaeger JA, Turner DH, Zuker M (1990) Predicting optimal and sub-
optimal secondary structure for RNA. In: Doolittle RF (ed) Meth-
ods in enzymology. Academic Press, San Diego, California, pp
281-306
Kohli Y, Morrall RAA, Anderson JB, Kohn LM (1992) Local and
trans-Canadian clonal distribution of
Sclerotinia sclerotiorum
on
canola. Phytopathology 82:875-880

176
Kohli Y, Brunner LJ, Yoell H, Milgroom MG, Anderson JB, Morrall
RAA, Kohn LM (in press) Clonal dispersal and spatial mixing in
populations of the plant pathogenic fungus, Sclerotinia scleroti-
orum. Mol Ecol
Kohn LM, Stasovski E, Carbone I, Royer J, Anderson JB (1991) My-
celial incompatibility and molecular markers identify genetic var-
iability in field populations of Sclerotinia sclerotiorurn. Phytop-
athology 81:480-485
Lambowitz AM, Belfort M (1993) Introns as mobile genetic ele-
ments. Annu Rev Biochem 62:587-622
Leslie JF (1993) Fungal vegetative compatibility. Annu Rev Phytop-
athol 31:127-151
Lipman D J, Pearson WR (1985) Rapid and sensitive protein similar-
ity searches. Science 227:1435-1441
Michel F (1984) A maturase-like coding sequence downstream of
the OXI2 gene of yeast mitochondrial DNA is interrupted by two
GC clusters and a putative end-of-messenger signal. Curr Genet
8:307-317
Michel F, Cummings DJ (1985) Analysis of class-I introns in a mit-
ochondrial plasmid associated with senescence of Podospora an-
serina reveals extraordinary resemblance to the Tetrahymena ri-
bosomal intron. Curr Genet 10:69-79
Michel F, Westhof E (1990) Modeling of the three-dimensional ar-
chitecture of group-I catalytic introns based on comparative se-
quence analysis. J Mol Biol 216:585-610
Michel F, Jacquier A, Dujon B (1982) Comparison of fungal mito-
chondrial introns reveals extensive homologies in RNA secon-
dary structure. Biochimie 64:867-881
Mota EM, Collins RA (1988) Independent evolution of structural
and coding regions in a Neurospora mitochondrial intron. Nature
332:654-656
Neefs J-M, Van de Peer Y, De Rijk P, Chapelle S, De Wachter R
(1993) Compilation of small ribosomal subunit RNA structures.
Nucleic Acids Res 21:3025-3049
Nishida H, Blanz PA, Sugiyama J (1993) The higher fungus Proto-
myces inouyei has two group-I introns in the 18s rRNA gene. J
Mol Evol 37:25-28
Ragan MA, Bird CJ, Rice EL, Singh RK (1993) The nuclear 18s ri-
bosomal RNA gene of the red alga Hildenbrandia rubra contains
a group-I intron. Nucleic Acids Res 21:3898
Sogin ML, Edman JC (1989) A self-splicing intron in the small sub-
unit rRNA gene of Pneumocystis carinii. Nucleic Acids Res 17:
5349-5359
Stern S, Weiser B, Noller HF (1988) Model for the three-dimension-
al folding of 16s ribosomal RNA. J Mol Biol 204:447-481
Turmel M, Boulanger J, Schnare MN, Gray MW, Lemieux C (1991)
Six group-I introns and three internal transcribed spacers in the
chloroplast large subunit ribosomal RNA gene of the green alga
Chlamydomonas eugametos. J Mol Biol 218:293-311
Turmel M, Gutell RR, Mercier J-P, Otis C, Lemieux C (1993) Anal-
ysis of the chloroplast large subunit ribosomal RNA gene from
17 Chlamydomonas taxa: three internal transcribed spacers and
12 group-I intron insertion sites. J Mol Biol 232:446-467
Van der Horst G, Inoue T (1993) Requirements of a group-I intron
for reactions at the 3' splice site. J Mol Biol 229:685-694
Waring RB, Davies RW (1984) Assessment of a model for intron
RNA secondary structure relevant to RNA self-splicing - a re-
view. Gene 28:277-291
White TJ, Bruns T, Lee S, Taylor J (1990) Amplification and direct
sequencing of fungal ribosomal RNA genes for phylogenetics.
In: Innis MA, Gelfand DH, Sninsky JJ, White TJ (eds) PCR pro-
tocols: a guide to methods and applications. Academic Press, San
Diego, California, pp 315-322
Wilcox LW, Lewis LA, Fuerst PA, Floyd GL (1992) Group-I introns
within the nuclear-encoded small-subunit rRNA gene of three
green algae. Mol Biol Evol 9:1103-1118
Zolan ME, Pukkila PJ (1986) Inheritance of DNA methylation in Co-
prinus cinereus. Mol Cell Biol 6:195-200
Zuker M (1989) On finding all suboptimal foldings of an RNA mole-
cule. Science 244:48-52