www.nature.

com/scientificreports

OPEN Complex intron generation in the

yeast genus Lipomyces
Norbert Ág1, Napsugár Kavalecz1,2, Fruzsina Pénzes1,2, Levente Karaffa1,
Claudio Scazzocchio3,4, Michel Flipphi1 & Erzsébet Fekete   1*
In primary transcripts of eukaryotic nuclear genes, coding sequences are often interrupted by U2-type
introns. Such intervening sequences can constitute complex introns excised by consecutive splicing
reactions. The origin of spliceosomal introns is a vexing problem. Sequence variation existent across
fungal taxa provides means to study their structure and evolution. In one class of complex introns called
[D] stwintrons, an (internal) U2 intron is nested within the 5'-donor element of another (external) U2
intron. In the gene for a reticulon-like protein in species of the ascomycete yeast genus Lipomyces, the
most 5' terminal intron position is occupied by one of three complex intervening sequences consistent
of differently nested U2 intron units, as demonstrated in L. lipofer, L. suomiensis, and L. starkeyi. In
L. starkeyi, the donor elements of the constituent introns are abutting and the complex intervening
sequence can be excised alternatively either with one standard splicing reaction or, as a [D] stwintron,
by two consecutive reactions. Our work suggests how [D] stwintrons could emerge by the appearance
of new functional splice sites within an extant intron. The stepwise stwintronisation mechanism may
involve duplication of the functional intron donor element of the ancestor intron.

In the eukaryote nuclear genome, coding sequences are often interspersed with intervening sequences (introns).
Introns have to be precisely excised from the primary transcript to generate the open reading frame that translates
into the appropriate peptide. Excision of U2-type introns and splicing of the bordering exons is catalysed by the
major spliceosome1–3, a large supramolecular organelle within the nucleus. (NB. Few U12-type introns are excised
by the minor spliceosome). Differently from spliceosomal introns, group I and group II introns are ribozymes,
which may or not need accessory proteins for self splicing. In eukaryotes, group I and group II introns are primar-
ily restricted to mitochondrial and chloroplast DNAs.
Intervening sequences of any of the three main classes – group I, group II and spliceosomal – can be made up
of more than one functional intron unit4–6. Previously, we have described a particular class of nested U2 introns,
the stwintron, for which the excision of the internal intron is required to accomplish the subsequent excision
of the external intron because the former interrupts one of the three conserved intron splicing elements of the
latter, the 5'-donor, the sequence element around the lariat branch point adenosine, or the 3'-acceptor7–11 (Fig. 1).
We named stwintron types according to where the insertion of the internal intron occurred – ‘D', indicating
insertions within the donor sequence, ‘L’, within the conserved sequence element around the lariat branch point
adenosine, and ‘A’, within the acceptor – followed by the numbers of the successive nucleotides (nt) in the consen-
sus sequence separated by the internal intron. To date, we have demonstrated the existence of [D1,2], [D2,3] and
[D5,6] stwintrons (i.e. internal intron situated between the first and the second nt, the second and the third nt,
and the fifth and the sixth nt of the donor of the external intron, respectively) and an [A2,3] stwintron (i.e. inter-
nal intron splitting the acceptor of the external intron between its second and third nt). We defined stwintrons as
the spliceosomal analogues of those original group II/III twin introns (twintrons) in the Euglena gracilis plastid
genome12,13, in which the internal intron disrupts a sequence element essential for the excision of the external
intron. (NB. Group III introns are abbreviated versions of group II introns). Correct stwintron excision requires
intron definition whereby the nearest 5’- and 3'-splice sites are paired across the intron prior to splicing14–16. This
results in excision of the smallest possible U2 intron; in the case of a stwintron, this is thus the internal intron. In
fungi, spliceosomal introns are generally quite small – often <100 nucleotide (nt). Nevertheless, one of the fungal
stwintrons we have described is crucial to an authentic alternative splicing event11.

1
Dept. of Biochemical Engineering, Faculty of Science and Technology, University of Debrecen, Debrecen, 4032,
Hungary. 2Juhász-Nagy Pál Doctoral School of Biology and Environmental Sciences, University of Debrecen,
Debrecen, 4032, Hungary. 3Dept. of Microbiology, Imperial College London, SW7 2AZ, London, UK. 4Institut de
Biologie Intégrative de la Cellule, Centre National de la Recherche Scientifique – Unité Mixte de Recherche UMR
9198, Gif-sur-Yvette, 91190, France. *email: [email protected]

Scientific Reports | (2020) 10:6022 | https://doi.org/10.1038/s41598-020-63239-6 1

www.nature.com/scientificreports/ www.nature.com/scientificreports

[D] [A]
Di Li Ai Di Li Ai

upstream exon De Le Ae downstream exon

Di Li Ai
[L]

De (25,120) Le (4,24) Ae
External intron: GURWGYNNNNNNNNNNNNNNNDYURAYNNNNHAG

Di (25,120) Li (4,24) Ai
Internal intron: GURWGYNNNNNNNNNNNNNNNDYURAYNNNNHAG

Figure 1.  Definition of stwintrons: The tree classes of stwintron (spliceosomal twin intron). Stwintrons are
complex intervening sequences consistent of nested U2 introns. In a [D] stwintron, the internal intron is nested
within the 5' donor element (consensus sequence: 5'-GURWGY) of the external intron. In a [L] stwintron, the
lariat branch point sequence element (consensus sequence: 5'-DYURAY) of the external intron is disrupted,
while in an [A] stwintron, it interrupts the 3' acceptor (consensus sequence: 5'-HAG). Consequently, excision
of the internal intron is indispensable for the subsequent excision of the external intron. The external intron
is shown as a red bar and the internal intron as a blue bar. The three conserved intronic sequence elements
essential for U2 splicing are indicated as black boxes labelled as “D”, “L” or “A”: external and internal elements
are labelled by “e” and “i”, respectively. Underneath, consensus sequences of the 5'-donor, the lariat branch point
element and the 3'-acceptor in fungi18 and their relative positions in the internal and external U2 introns are
depicted.

Two [D] stwintrons conserved across whole classes of filamentous fungi are located at exactly the same intron
position as a canonical U2 intron present in early divergent Pezizomycotina taxa8. One possible mechanism by
which such a stwintron could emerge involves the stepwise formation of new 5'- and 3' splice sites within an
extant intron of sufficient size to evolve into two U2 introns (Fig. 2). We propose to call this mechanism “stwint-
ronisation” by analogy with intronisation, the formation of intronic sequences from exonic sequences17. The gen-
esis of a new 5' splice site within or next to the donor element of the host intron is necessary for the formation of a
[D] stwintron. A small duplication including part or the entirety of the canonical six-nt donor element (consensus
sequence: 5'-GURWGY)18 could occur during repair of an asymmetric DNA double-strand break (DSB) which
would leave a 3' overhang. In an independent event, mutations in the central region of the host intron could result
in functional 3'-splice sites. The new internal splice sites may pair to define the new U2 unit, whose subsequent
excision is necessary for the formation of a functional donor for the second U2 definition and standard splicing
reaction to properly remove all intervening sequences.
In a previous paper11, we identified a [D5,6] stwintron in the gene encoding a reticulon-like protein in the
Pezizomycotina, one of the three Ascomycota subphyla. The gene product is a “protein of unknown function
which associates with the endoplasmic reticulum” (protein family PF02453). While assessing the ancestry of this
[D5,6] stwintron, we encountered different complex intervening sequences (CIS) consistent of nested U2 introns
in the orthologous gene in seven species of Lipomyces, a taxon of the Saccharomycotina, another subphylum of the
Ascomycota. The parallel evolution of these three different CIS at the very same intron position in species of the
same fungal genus supports the stwintronisation mechanism of stwintron evolution.

Results
Evolution of the intron-exon structure of the transcript of the genes for a reticulon-like protein
in the genus Lipomyces.  The conservation of intron positions in multiple species of Taphrinomycotina
and in the family of the Lipomycetaceae (Saccharomycotina) is consistent with their reticulon-like genes being
genuine orthologues of the one in Aspergillus nidulans (rtnA) and in other Pezizomycotina11. Figure 3a shows the
monophyletic cluster of the Lipomyces genus (nine species) drawn from an extensive maximum likelihood tree
of the reticulon-like protein in all Ascomycota (see Methods section for details). Lipomycetaceae occupy a unique
phylogenetic position at the basis of the Saccharomycotina clade19. The species Lipomyces starkeyi harbours many
more ORF-interrupting introns than ascomycete yeasts outside the Lipomycetaceae20. The intron density of L.
starkeyi resembles that seen in Pezizomycotina rather than that apparent in other Saccharomycotina.
Exceptionally within the Saccharomycotina, the primary transcript of the gene for the reticulon-like protein
harbours multiple introns in eight genome-sequenced Lipomyces. However, the intron content varies considerably
from the maximum of five introns in L. starkeyi and three closely related species, as well as in L. lipofer (Fig. 3b).
L. japonicus lacks the most 3' of these five introns while in L. suomiensis, the intron positions 3 and 4 are not
occupied. Furthermore, L. doorenjongii only carries introns at the first two positions and L. oligophaga harbours
one lone intron corresponding with the most 5' intron in the eight other species. The four most 3' introns in L.
starkeyi interrupt the DNA coding for the conserved PF02453 domain and are positionally conserved in either
species of Pezizomycotina (position 2), or species of Taphrinomycotina (position 3), or species from both these
other subphyla (positions 4 and 5). The observed simplification of the intron-exon structure of the gene encoding

Scientific Reports | (2020) 10:6022 | https://doi.org/10.1038/s41598-020-63239-6 2

www.nature.com/scientificreports/ www.nature.com/scientificreports

Stwintronisation
upstream exon downstream exon
D Lp Ap L A

DSB
upstream exon downstream exon

D Lp Ap L A

upstream exon downstream exon

D D Lp Ap L A

upstream exon downstream exon

D D Lp Ap L A

upstream exon downstream exon

De Di Li Ai Le Ae

upstream exon downstream exon

De Di Li Ai Le Ae

Figure 2.  The stwintronisation mechanism of complex intron generation. Scheme of events that result in the
stepwise formation of new functional 5'- and 3'-splice sites within an extant U2 intron –stwintronisation –
illustrated for the emergence of a [D] stwintron. The original host intron is of a length such that it can be split
into two functional U2 introns and is indicated by the purple bar. Duplication of functional donor elements
may result from the repair of an asymmetric DNA double-strand break (DSB) leaving small 3' overhangs. These
overhangs are filled in (green) by polymerase activity associated with the canonical nonhomologous DNA end-
joining (cNHEJ) machinery before blunt end ligation. In a separate event, precursor sequences (“p”) resembling
lariat branch point (L) and associated acceptor (A) elements in the middle of the host intron morph into
functional 3' splice elements (“i”) by mutation. The newly formed 5'-donor- and lariat branch point elements
allow definition of a new internal intron disrupting the functional donor element of the new external intron.
Donor duplication and generation of functional 3' consensus sites are independent events and can take place in
any order.

the reticulon-like protein suggests that some early divergent Lipomyces species have lost introns rather than that
new introns have been gained in the other taxa, completely independent of the massive intron loss observed in all
other families of Saccharomycotina21.

Identification of different complex intervening sequences (CIS) at the same intron position in
the gene for the reticulon-like protein.  The first intervening sequence in the primary transcript of the
reticulon-like gene is at a conserved position, unique to the nine Lipomyces species. The second exon is either
162 or 165 nt (L. japonicus & L. lipofer) long. The extra codon is located upstream of the codon of the absolutely
conserved Trp of the PF02453 domain (Fig. 3b), within the DNA encoding the N-terminal extension of the
reticulon-like protein. In L. oligophaga and L. doorenjongii, the intervening sequences at the most 5' intron posi-
tion are short canonical U2 introns, respectively 78 and 46 nt in length. However, in the other seven species, the
corresponding sequences are considerably longer – from 140 nt in L. suomiensis to 320 nt in L. lipofer – and in
each case, appear to consist of two nested canonical U2 introns. We have detected three modes of organisation of
the CIS at the first intron position in the gene for the reticulon-like protein in different Lipomyces species, each
with the putative internal U2 intron disrupting the putative external U2 intron near the 5' extremity of the latter
at a slightly different position. This is illustrated in Fig. 4. We have experimentally verified the predicted internal
introns in three representative species by targeted RT-PCR of the respective stwintron splicing intermediates and
subsequent cDNA sequencing (see Methods section for details).
In L. lipofer, the CIS is 320 nt long. An 158-nt long internal intron (5' GUAAGU – 131-nt – ACUGAC – 12-nt
– UAG) is disrupting the 5'-donor element of an 162-nt long external intron (5' GUGA | GU – 137-nt – ACUAAC
– 10-nt – UAG) between A4 and G5. The splicing scheme is shown in Fig. 5a. The experimentally determined
sequence of the splicing intermediate of this [D4,5] stwintron, lacking the predicted internal intron and featur-
ing the uninterrupted 5' donor of the predicted external intron, is shown in Supplementary Fig. S1. Determined
sequences of the mature mRNA and the stwintron splicing intermediate were deposited at GenBank under
accession numbers MN689083 and MN689084, respectively. This L. lipofer stwintron is the first donor-disrupted

Scientific Reports | (2020) 10:6022 | https://doi.org/10.1038/s41598-020-63239-6 3

www.nature.com/scientificreports/ www.nature.com/scientificreports

a
1 2 3 4 5
Lipomyces japonicus [PPJV]
1 2 3 4 5 0.95 1 2 3 4 5
Lipomyces oligophaga [PPJR]
1 2 3 4 5
0.71 Lipomyces suomiensis [PPJQ]
1 2 3 4 5
0.91 Lipomyces doorenjongii [PPJU]
1 2 3 4 5
0.54 Lipomyces lipofer [PPJX]

0.8 Lipomyces kononenkoae [PPJW]

1 2 3 4 5
0.96 Lipomyces starkeyi [LSGR]
0.87
Lipomyces mesembrius [PPJS]
0.82
Lipomyces arxii [PPJT]

0.1

b
1 W 2 3 4 5 Y Lipomyces starkeyi [LSGR]
55 162 58 145 70 278

1 2 3 4 Lipomyces japonicus [PPJV]

37 165 58 142 396

1 2 5 Lipomyces suomiensis [PPJQ]

34 162 273 263

1 2 3 4 5 Lipomyces lipofer [PPJX]

37 165 58 142 70 305

1 2 Lipomyces doorenjongii [PPJU]

55 162 533

1 Lipomyces oligophaga [PPJR]

139 701
[1]
Neolecta irregularis [LXFE]
56 125 196 142 70 323
[2]
Pneumocystis murina [AFWA]
31 93 196 142 70 371
[1] [1] [2]
Tuber melanosporum [CABJ]
55 137 118 197 70 359
[1] [1]
Aspergillus nidulans [AACD]
43 24 117 32 118 197 450
[1] [1] [1] [0] [1] [0]
W (conserved) Y (conserved)
Reticulon domain

Figure 3.  Evolutionary relations between Lipomyces species and the intron-exon structure of their genes for
the reticulon-like protein. (a) Monophyletic clade of the Lipomyces genus drawn from an extensive maximum
likelihood tree, inferred from a trimmed alignment of 902 ascomycete reticulon-like proteins. Branch support is
estimated with approximate likelihood-ratio tests for branches32. The scale bar gives the evolutionary distance of
0.1 amino acid substitutions per site. The circles along some branches indicate the presence (green) or absence
(red) of introns at five conserved positions in the primary transcript(s) of the species at the end of that branch.
Species names are labelled with the unique four-letter combination of their respective Whole Genome Shotgun
Master Accessions between square brackets. (b) Schematic representation of deduced gene models of the genes
for the reticulon-like protein in Lipomyces species compared with selected species of Ascomycota. The size of
each exon is given in nucleotides (nt): for the first exon with coding sequences, the size is given from the start
codon to the first intron and for the last exon with coding sequences, from the last intron to the stop codon.
The intron positions between the exons are given by the uninterrupted vertical lines in the gene bars. The phase
of the introns is given underneath between square brackets: phase zero in red, phase one in purple and phase
two in cyan. CIS are represented by thick purple vertical lines. The sequences encoding the conserved PF02453
domain are roughly indicated by the red lined bar underneath the gene models; the strictly conserved Trp (W73
in A. nidulans) and Tyr (Y212 in A. nidulans) residues are situated near the N-terminus and C-terminus of the
PF02453 domain, respectively. The introns at conserved positions 2–5 are connected by dashed lines between
the gene bars.

Scientific Reports | (2020) 10:6022 | https://doi.org/10.1038/s41598-020-63239-6 4

www.nature.com/scientificreports/ www.nature.com/scientificreports

Lipomyces lipofer 158 nt

[D4,5] stwintron --TGCCTCgtgagtaagt--actgac--taggttcag-----actaac--tagCACCCC--

320 nt

Lipomyces suomiensis 61 nt

[D7,8] stwintron --TTGATCgtgagtggtaggt--tctaat--tagaatgcc--actgat--tagCTCCGT--

140 nt

Lipomyces japonicus 162 nt

[D7,8] stwintron --TGCCTTgtaagtggtaagt--gctaac--tagattgat--attaat--tagCACCTC--

286 nt

Lipomyces starkeyi
120 nt
[D6,7] stwintron --TGTCTCgtacgtgtaagt--tttaat--cagattgat---gctgac--tagCTCAAT--

252 nt

standard intron --TGTCTCGTACGTgtaagt----------------------gctgac--tagCTCAAT--

246 nt

Figure 4.  The three complex intervening sequences [D4,5], [D6,7], [D7,8] in the transcript of the gene
coding for the reticulon-like protein of selected Lipomyces species schematically aligned at their splice sites.
The alternative splice option in L. starkeyi, excision of one large canonical U2 intron instead of the [D6,7]
stwintron, is also depicted. The last uninterrupted codon up- and the first uninterrupted codon downstream
the intervening sequences is underlined. Intronic sequences are in lower case letter. The sequences of the
three conserved intronic sequence elements, the 5' donor (D), the lariat branch point element (L) and the 3'
acceptor (A), are color coded: Those of the internal introns are in blue, those of the external introns are in red
and those of the large alternatively spliced canonical U2 intron (L. starkeyi) are in purple. For convenience, the
underlying internal introns are highlighted by the light grey background and the external introns by the dark
grey background. The exact sizes of the internal introns and the complex intervening sequences are given for
each species.

stwintron of the [D4,5] type as well as the first ever stwintron (sensu stricto) described outside the Pezizomycotina
subphylum.
In two species, L. suomiensis and L. japonicus, we detected two potential canonical 5' donor sequences sepa-
rated by one nt at the 5' extremity of the predicted CIS. Furthermore, a fully canonical 3'-acceptor and associated
lariat branch point element could be identified within the CIS in both species. Sequence analysis of the mature
messenger in L. suomiensis (GenBank MN689081 & Fig. 5b) showed that the most 5' of these donor elements
(5'-GUGAGU) serves at the ultimate 5' splice site of the whole 140 nt-long CIS. Subsequent analysis of the splicing
intermediate (GenBank MN689082 & Supplementary Fig. S2) confirmed that the 3' donor copy (5'-GUAGGU) is
used to excise an 61-nt long internal U2 intron (5' GUAGGU –38-nt – UCUAAC – 8-nt – UAG) nested in a 79-nt
long canonical external U2 intron (5' GUGAGUG | A – 45-nt – ACUGAC – 17-nt – UAG) between G7 and A8
(Fig. 5b illustrates the splicing scheme in L. suomiensis). Similarly, in L. japonicus, we predict that an 162-nt long
canonical U2 intron (5' GUAAGU – 135-nt – ACUAAC – 12-nt – UAG) is disrupting an 124-nt long canonical
U2 intron (5' GUAAGUG | A – 97-nt – AUUAAU – 10-nt – UAG) between G7 and A8. The whole CIS is 286 nt
long and the two functional donor elements are identical in sequence (5'-GUAAGU) (see Fig. 4 for L. japonicus).
A difference between the two species is that the constituting U2 introns in L. suomiensis are half the size of those
in L. japonicus.
To excise these [D7,8] CIS, two standard splicing reactions are necessary. This intron nesting does not consti-
tute a [D] stwintron in sensu stricto as previously defined7, since the internal intron does not interrupt the donor
element of the external intron; the two canonical donor elements are actually one nt apart from each other. In the
primary transcript of the gene for the reticulon-like protein in L. suomiensis, the primary selected 5'- and 3'-splice
sites are clearly those that define the smallest possible internal intron (intron definition). We can therefore con-
sider such CIS of nested introns examples of a new class of fungal stwintron, the “stwintron sensu lato”.
In L. starkeyi and the three closely allied species (Fig. 3a) we detected two abutting potential 5' donor elements
at the 5' extremity of the predicted CIS in the gene for the reticulon-like protein, a second instance of stwintron
sensu lato, [D6,7] (Fig. 5c and Supplementary Fig. S3). The most 3' of the donor elements has the ideal consensus
sequence (5'-GUAAGU) while the one at 5' (5'-GUACGU) is somewhat less canonical. Given the position of
the intervening sequence within the DNA/RNA coding for the poorly conserved N-terminal extension of the
reticulon-like protein, we first tested in L. starkeyi (strain CCY 33-1-1: gene sequence, GenBank MN689085)

Scientific Reports | (2020) 10:6022 | https://doi.org/10.1038/s41598-020-63239-6 5

www.nature.com/scientificreports/ www.nature.com/scientificreports

a Lipomyces lipofer c Lipomyces starkeyi

Exon I [D4,5] stwintron Exon II Exon IAlt Exon II

--UGCCUCgugaguaagu--acugac--uaggu--acuaac--uagCACCCC-- --UGUCUCGUACGUCUCAAU--

First splicing
Splicing
Exon I External intron Exon II Exon IAlt standard intron Exon II
--UGCCUCgugagu--acuaac--uagCACCCC-- --UGUCUCGUACGUguaagu--uuuaau--cag--gcugac--uagCUCAAU--
stwintron splicing intermediate
Second splicing
Standard splicing

Exon I Exon II
--UGUCUCguacguguaagu--uuuaau--cag--gcugac--uagCUCAAU--
--UGCCUCCACCCC-- primary transcript
Stwintron splicing
b Lipomyces suomiensis
Exon I [D7,8] stwintron Exon II Exon I [D6,7] stwintron Exon II
--UUGAUCgugagugguaggu--ucuaau--uag--acugau--uagCUCCGU-- --UGUCUCguacguguaagu--uuuaau--cag--gcugac--uagCUCAAU--

First splicing First splicing

Exon I External intron Exon II Exon I External intron Exon II

--UUGAUCgugagug--ucuaau--uagCUCCGU-- --UGUCUCguacgu--gcugac--uagCUCAAU--
stwintron splicing intermediate stwintron splicing intermediate
Second splicing Second splicing

Exon I Exon II Exon I Exon II

--UUGAUCCUCCGU-- --UGUCUCCUCAAU--

Figure 5.  Splicing schemes for the different complex intervening sequences (CIS) in the orthologue gene
coding for the reticulon-like protein in three Lipomyces species. (a) Schematic representation of the structure of
the L. lipofer [D4,5] stwintron and its two-step excision. Intronic sequences are in lower case letters. The internal
intron is represented by the light grey bar and its 5′-donor-, lariat branch point domain- and 3′-acceptor
sequences are in blue lettering. The external intron is marked by the darker grey bar; its 5′-donor-, lariat branch
point domain- and 3′-acceptor sequences are in red lettering. Note that the internal intron is nested in the donor
element of the external intron between a4 and g5. (b) The structure of the L. suomiensis [D7,8] CIS (stwintron
sensu lato) and its two-step excision. The internal intron is nested in the external intron between g7 and a8. The
nt between the two functional 5'-donor elements (g7) is highlighted by the yellow background. (c) The structure
of the L. starkeyi [D6,7] CIS (stwintron sensu lato) and its two-step excision (from the primary transcript in
the centre of the Panel downwards). The internal intron is nested in the external intron between u6 and a7.
Alternatively, the intervening sequence is excised by one standard splicing reaction (from the primary transcript
in the centre of the Panel upwards). The large canonical U2 intron is marked by a darker grey bar; its 5′-donor-,
lariat branch point domain- and 3′-acceptor sequences are in purple lettering. Note that the most 5' of the
abutting donor sequences (5'-GUACGU) is exonic when the transcript is alternatively spliced with one standard
splicing reaction. As a consequence, the upstream exon (Exon I ALT) is six nt longer.

whether the intervening sequence could be alternatively excised using either the one or the other donor element.
Amongst the cDNA clones of matured mRNA generated, we indeed found two different ORFs, one of them con-
taining six nt (GTACGT) more than the other (GenBank MN689086 & MN689087), demonstrating alternative
splicing. Under our experimental conditions, fewer cDNA clones contained the GTACGT sequence. Subsequent
inspection of the intervening sequence suggested the presence of a coupled lariat branch point element and
3'-acceptor roughly in its centre, of which the lariat branch point element could be the imperfect 5'-AUUCAU or
the equally imperfect 5'-UUUAAU, associated with a 5'- CAG shortly downstream. If the internal splice signals
were recognised by the spliceosome, one would expect a complex [D6,7] intervening sequence of 252 nt, where
an 120-nt long internal U2 intron (5' GUAAGU – 75-nt – AUUCAU – 14-nt – UUUAAU – 10-nt – CAG) is dis-
rupting an 132-nt long external intron (5' GUACGU | A – 109-nt – GCUGAC – 7-nt – UAG) between U6 and A7
(see Fig. 5c for the splicing scheme and Supplementary Fig. S3). We sequenced the expected stwintron splicing
intermediates in the transcripts of the genes coding for the reticulon-like protein in L. starkeyi CCY 33-1-1 and
L. kononenkoae (GenBank MN689088–MN689090). However, if the spliceosome would ignore the imperfect

Scientific Reports | (2020) 10:6022 | https://doi.org/10.1038/s41598-020-63239-6 6

www.nature.com/scientificreports/ www.nature.com/scientificreports

internal 3' splice site halfway the intervening sequence but not the fully canonical proximal donor, one would
expect that a 246-nt long canonical U2 intron (5' GUAAGU – 224-nt – GCUGAC – 7-nt – UAG) is excised in one
standard splicing reaction (Fig. 5c: Alternative splicing option), giving rise to the 6-nt longer mRNA species (i.e.,
GenBank MN689086).
In L. starkeyi, a [D6,7] stwintron sensu lato, characterised by abutting 5' donor elements, is functionally
implicated in alternative splicing. The stwintron alternative employs both donor elements in subsequent
splicing reactions, the first of which uses the proximal, ideal consensus 5'-donor paired with the imperfect
3'-splice site halfway the CIS instead of the canonical 3' splice site at its end. The U2 spliceosome thus does
not “choose” between directly neighbouring functional 5'-donors but rather between available lariat branch
point elements situated at much longer distance from one another, of which the more proximal one is, cru-
cially, aberrant.

Discussion
In the genes for the reticulon-like protein of genome-sequenced species of the Lipomycetaceae, the earliest diver-
gent family of the Saccharomycotina subphylum, we identified and experimentally confirmed three different for-
mats of intron nesting in the most 5' intervening sequence, and therewith three new types of stwintron. Like for
the [D5,6] stwintron in Pezizomycotina11, the phase of the respective external introns is one and the [D4,5]-,
[D6,7]- and [D7,8] stwintrons are located in the DNA/RNA encoding the intrinsically disordered N-terminal
extension of the reticulon domain (PF02453), which is poorly conserved both in length and in sequence in asco-
mycete reticulon-like proteins.
The origin of spliceosomal introns is a vexing problem. Some commonly purported mechanisms involve
mobile elements like transposons or group II introns of mitochondrial or bacterial origin, while other proposed
mechanisms (tandem duplication, gene conversion or intronisation) are autonomous and have a strictly endoge-
nous origin22. Some modes of autonomous intron gain could involve repair of double-strand breaks (DSB) by an
end-joining mechanism23–26. The formation of a new stwintron from the sequences of an existing canonical intron
can occur similar to the intronisation mechanism17, by the formation of new 5'- and 3'-splice sites within the
parental intron at the chromosomal locus (see Fig. 2 for a schematic overview). The three position-conserved CIS
in seven Lipomyces primary transcripts of the gene encoding the reticulon-like protein each constitute two differ-
ently nested canonical U2 introns. Crucially, in L. starkeyi, the CIS can also be removed in one standard splicing
reaction, which suggests that the current Lipomyces [D4,5]-, [D6,7]-, and [D7,8] stwintrons found in other spe-
cies have evolved by stwintronisation of one long primordial intron. An aligned representation (Fig. 4) suggests
that the donor of the original intron has been duplicated – entirely or in part – to generate the 5' splice site of
the new internal intron. Such small tandem duplications (4–7 nt) may well be the result of repair of asymmetric
double-strand breaks with a limited 3' overhang by canonical nonhomologous end-joining (cNHEJ)27. However,
we cannot exclude the possibility that formation of one or more of the three different stwintrons involved point
mutations near the 5' end of the parental canonical intron to create the donor of the new internal intron, rather
than a small duplication. Secondary mutations within the parental intron are still necessary to generate functional
3' splice elements for the internal intron of the new stwintron. Where stwintronisation appears completed in L.
lipofer, L. suomiensis and L. japonicus, the situation in L. starkeyi ([D6,7]) could represent an intermediate stage in
which the CIS can be removed by one- as well as by two subsequent U2 splicing reactions, as the internal imper-
fect 3' lariat branch point element is apparently functionally weak.
Natural sequence variation existent across a fungal phylum provides means to study intron structure and
evolution. In this work, we describe stwintrons sensu lato, in which the internal intron is not nested within a con-
served intronic sequence element essential for the excision of the external intron, but for which the constituent
introns are still excised in “inside out” order by consecutive standard splicing reactions. The occurrence of three
different stwintrons at a strictly conserved intron position in seven species of Lipomyces, one of which embodies
an authentic alternative splice option, provides an insight into an evolutionary mechanism by which complex
intervening sequences consistent of nested intron units may emerge. We have called this stepwise mechanism
stwintronisation, where new functional 5' and 3' splice sites evolve within a pre-extant canonical intron.

Methods
Mining fungal genes for the reticulon-like protein.  In an effort to probe the ancestry of the
Pezizomycotina [D5,6] stwintron in the gene for the fungal reticulon-like protein, we searched the DNA data-
bases at the National Center of Biotechnology Information (NCBI) for homologous genes in other fungi.
Additional homologues were found in the Saccharomycotina and Taphrinomycotina subphyla of the Ascomycota,
the Blastocladiomycota, the Glomeromycota, the Mucoromycotina and the Mortierellomycotina (Supplementary
Table S1). Although most genome-sequenced ascomycete yeasts have intronless genes for the reticulon-like
protein (Rtn), quite some taxa, including species from the Cephaloascaceae, Alloascoideaceae, Pichiaceae and
Debaryomycetaceae families, retained one standard intron at the position corresponding to the fifth intron in the
orthologous Aspergillus nidulans rtnA gene11. This position conservation suggests that the Ascomycota genes are
phylogenetically related. Multiple genes were often encountered in Mucoromycotina and Mortierellomycotina, but
we could not find [D5,6] stwintrons beyond the Pezizomycotina subphylum of the Ascomycota.
Species of Lipomyces are the only Saccharomycotina that harbour multiple introns in their orthologous genes
for the reticulon-like protein. Of these, the second intron occurs at the position of the fifth A. nidulans intron.
We used all Rtn proteins we deduced from the encoding genes in Ascomycota to infer maximum likelihood trees
(not shown), to verify the status of the Lipomycetaceae as the earliest divergent family of Saccharomycotina19,20.
Most of these analyses yielded monophyletic clades for the nine sequenced Lipomyces: A representative max-
imum likelihood subtree is shown in Fig. 3a. The used Lipomyces genome sequence resources are: L. starkeyi
(GenBank Whole Genome Shotgun Master accession: LSGR00000000.120), L. mesembrius (PPJS00000000.219),

Scientific Reports | (2020) 10:6022 | https://doi.org/10.1038/s41598-020-63239-6 7

www.nature.com/scientificreports/ www.nature.com/scientificreports

L. arxii (PPJT00000000.219), L. kononenkoae (PPJW00000000.119), L. lipofer (PPJX00000000.119), L. doorenjongii

(PPJU00000000.119), L. suomiensis (PPJQ00000000.219), L. oligophaga (PPJR00000000.219) and L. japonicus
(PPJV00000000.119).

Maximum likelihood phylogenetic analysis of ascomycete reticulon-like proteins.  The phy-

logenetic tree at the basis of the monophyletic Lipomyces clade shown in Fig. 3, was inferred from a trimmed
multiple sequence alignment of 902 putative reticulon-like proteins. The proteins were first aligned with MAFFT
(Multiple Alignment using Fast Fourier Transform; version 7)28, applying the L-INS-i iterative refinement method
trained on recognising one conserved domain with long extensions – appropriate for the PF02453 domain – with
the BLOSUM80 scoring matrix. This MAFFT alignment was trimmed to 136 informative residues with Block
Mapping and Gathering with Entropy software (BMGE version 1.12)29, using the BLOSUM40 similarity matrix
with the block size set at 4. A maximum likelihood tree was then calculated from the trimmed alignment by
PhyML (version 3.0) with automatic substitution model selection30 in default mode. The Smart Model Selection
(SMS) software selected LG + G + I + F as the substitution model31 using the Akaike Information Criterion.
Branch support was estimated by approximate likelihood-ratio tests32 implemented by the PhyML webserver.
The inferred tree was drawn from the PhyML Newick output with FigTree (version 1.4.3: http://tree.bio.ed.ac.uk/
software/figtree). The Lipomyces subtree was selected in FigTree (select clade option) and saved in a separate file,
then exported as an interactive pdf. Final annotation was done in Adobe Illustrator.

Lipomyces strains, maintenance and cultivation conditions.  Lipomyces lipofer CBS 944 (National
Collection of Agricultural and Industrial Microorganisms, Budapest, Hungary: NCAIM Y.00351), Lipomyces
suomiensis CBS 7251 (Westerdijk Fungal Biodiversity Institute, Royal Netherlands Academy of Sciences, Utrecht,
The Netherlands), and Lipomyces starkeyi CCY 33-1-1 (Dept. of Microbiology and Biotechnology, Corvinus
University of Budapest, Hungary) were used to verify the existence of a particular CIS in the primary transcript of
the orthologous genes encoding a reticulon-like protein. Lipomyces kononenkoae CBS 2514 was also assessed in
support of results obtained for L. starkeyi. All yeasts were maintained and cultivated on rich growth media listed
in Supplementary Table S2 at 26 °C, as recommended by the respective stock centers supplying the reference
materials.
For isolation of nucleic acids, yeast cultures were grown in 500-mL Erlenmeyer flasks with 100 mL medium
in an orbital shaker (Infors HT Multitron) at 200 revolutions min–1 (rpm) for 16 h. The biomass was harvested
by centrifugation at 10,000 rpm for 5 min then transferred to tubes with ceramic beads (MagNA Lyser Green
Beads; Roche) and homogenised at 4000 rpm for 30 s with the help of a MagNA Lyser Instrument (Roche).
Macherey-Nagel NucleoSpin kits were used for the extraction of genomic DNA (NucleoSpin Plant II) and total
RNA (NucleoSpin RNA Plant) from the homogenate. Nucleic acids were quantified using NanoDrop technology
(Thermo Scientific).

Reverse transcription and polymerase chain reaction (RT-PCR) verification of the intron-exon
structures, mature RNAs and stwintron splicing intermediates.  Reverse transcription was primed
off 1 μg of total RNA with Oligo(dT) as the primer using the First Strand cDNA Synthesis Kit (Thermo Scientific).
PCR reactions were performed in a 25 μL volume containing 4 μL of single strand cDNA, using gene-specific oli-
gonucleotides (Supplementary Table S3) as primers and DreamTaq DNA Polymerase (Thermo Scientific). Cycling
conditions after initial denaturation at 95 °C (2 min) were: 40 cycles of 95 °C for 30 s, 56 °C for 1 min, and 72 °C
for 1 min, followed by one post-cyclic elongation at 72 °C for 5 min. Amplified fragments were resolved in native
agarose gels. All RT-PCR experiments were done in triplicate, starting with biomass from independent cultures.
Double-strand cDNA was gel-purified (NucleoSpin Gel & PCR Clean-up, Macherey-Nagel) and cloned
(pGEM-T Easy Vector System I, Promega). Plasmid DNA was isolated using the NucleoSpin Plasmid EasyPure
kit (Macherey-Nagel). Independent clones were sequenced over both strands using universal primers hybrid-
ising to the vector (Eurofins Genomics, Ebersberg, Germany). All sequences determined were deposited at
GenBank. For convenience, determined sequences of RNA stwintron splicing intermediates are also given in the
Supplementary Information.

Accession numbers.  Accession numbers for sequences determined during this study :
MN689081–MN689090.

Data availability
All datasets generated are included in the manuscript and/or the Supplementary Information. All sequences
determined in the course of this study were deposited at GenBank.

Received: 17 January 2020; Accepted: 27 March 2020;

Published: xx xx xxxx

References
1. Plaschka, C., Newman, A. J. & Nagai, K. Structural basis of nuclear pre-mRNA splicing: Lessons from yeast. Cold Spring Harbor
Perspect. Biol. 11, a032391, https://doi.org/10.1101/cshperspect.a032391 (2019).
2. Yan, C., Wan, R. & Shi, Y. Molecular mechanisms of pre-mRNA splicing through structural biology of the spliceosome. Cold Spring
Harbor Perspect. Biol. 11, a032409, https://doi.org/10.1101/cshperspect.a032409 (2019).
3. Kastner, B., Will, C. L., Stark, H. & Lührmann, R. Structural insights into nuclear pre-mRNA splicing in higher eukaryotes. Cold
Spring Harbor Perspect. Biol. 11, a032417, https://doi.org/10.1101/cshperspect.a032417 (2019).
4. Hafez, M. & Hausner, G. Convergent evolution of twintron-like configurations: One is never enough. RNA Biol. 12, 1275–1288,
https://doi.org/10.1080/15476286.2015.1103427 (2015).

Scientific Reports | (2020) 10:6022 | https://doi.org/10.1038/s41598-020-63239-6 8

www.nature.com/scientificreports/ www.nature.com/scientificreports

5. Georgomanolis, T., Sofiadis, K. & Papantonis, A. Cutting a long intron short: Recursive splicing and its implications. Front. Physiol.
7, 598, https://doi.org/10.3389/fphys.2016.00598 (2016).
6. Zumkeller, S., Gerke, P. & Knoop, V. A functional twintron, ‘zombie’ twintrons and a hypermobile group II intron invading itself in
plant mitochondria. Nucleic Acids Res., gkz1194, https://doi.org/10.1093/nar/gkz1194 (2020).
7. Flipphi, M., Fekete, E., Ág, N., Scazzocchio, C. & Karaffa, L. Spliceosome twin introns in fungal nuclear transcripts. Fungal Genet.
Biol. 57, 48–57 (2013).
8. Flipphi, M. et al. Emergence and loss of spliceosomal twin introns. Fungal Biol. Biotechnol. 4, 7, https://doi.org/10.1186/s40694-017-
0037-y (2017).
9. Ág, N., Flipphi, M., Karaffa, L., Scazzocchio, C. & Fekete, E. Alternatively spliced, spliceosomal twin introns in Helminthosporium
solani. Fungal Genet. Biol. 85, 7–13 (2015).
10. Fekete, E. et al. A mechanism for a single nucleotide intron shift. Nucleic Acids Res. 45, 9085–9092 (2017).
11. Kavalecz, N. et al. A spliceosomal twin intron (stwintron) participates in both exon skipping and evolutionary exon loss. Sci. Rep. 9,
9940, https://doi.org/10.1038/s41598-019-46435-xv (2019).
12. Copertino, D. W. & Hallick, R. B. Group II twintron: an intron within an intron in a chloroplast cytochrome b-559 gene. EMBO J.
10, 433–442 (1991).
13. Copertino, D. W. & Hallick, R. B. Group II and group III introns of twintrons: Potential relationships with nuclear pre-mRNA
introns. Trends Biochem. Sci. 18, 467–471 (1993).
14. De Conti, L., Baralle, M. & Buratti, E. Exon and intron definition in pre-mRNA splicing. Wiley Interdiscip. Rev.-RNA 4, 49–60,
https://doi.org/10.1002/wrna.1140 (2013).
15. Romfo, C. M., Alvarez, C. J., van Heeckeren, W. J., Webb, C. J. & Wise, J. A. Evidence for splice site pairing via intron definition in
Schizosaccharomyces pombe. Mol. Cell. Biol. 20, 7955–7970 (2000).
16. Shao, W., Kim, H.-S., Cao, Y., Xu, Y.-Z. & Query, C. C. A U1-U2 snRNP interaction network during intron definition. Mol. Cell. Biol.
32, 470–478 (2012).
17. Irimia, M. et al. Origin of introns by ‘intronization’ of exonic sequences. Trends Genet. 24, 378–381 (2008).
18. Kupfer, D. M. et al. Introns and splicing elements of five diverse fungi. Eukaryot. Cell 3, 1088–1100 (2004).
19. Shen, X. X. et al. Tempo and mode of genome evolution in the budding yeast subphylum. Cell 175, 1533–1545 (2018).
20. Riley, R. et al. Comparative genomics of biotechnologically important yeasts. Proc. Natl. Acad. Sci. USA 113, 9882–9887 (2016).
21. Stajich, J. E., Dietrich, F. S. & Roy, S. W. Comparative genomic analysis of fungal genomes reveals intron-rich ancestors. Genome Biol.
8, R223, https://doi.org/10.1186/gb-2007-8-10-r223 (2007).
22. Yenerall, P. & Zhou, L. Identifying the mechanisms of intron gain: Progress and trends. Biol. Direct 7, 29, https://doi.
org/10.1186/1745-6150-7-29 (2012).
23. Haviv-Chesner, A., Kobayashi, Y., Gabriel, A. & Kupiec, M. Capture of linear fragments at a double-strand break in yeast. Nucleic
Acids Res. 35, 5192–5202 (2007).
24. Li, W., Tucker, A. E., Sung, W., Thomas, W. K. & Lynch, M. Extensive, recent intron gains in Daphnia populations. Science 326,
1260–1262 (2009).
25. Ragg, H. Intron creation and DNA repair. Cell. Mol. Life Sci. 68, 235–242 (2011).
26. Li, W., Kuzoff, R., Wong, C. K., Tucker, A. & Lynch, M. Characterization of newly gained introns in Daphnia populations. Genome
Biol. Evol. 6, 2218–2234, https://doi.org/10.1093/gbe/evu174 (2014).
27. Schimmel, J., Kool, H., van Schendel, R. & Tijsterman, M. Mutational signatures of non-homologous and polymerase theta-
mediated end-joining in embryonic stem cells. EMBO J. 36, 3634–3649 (2017).
28. Katoh, K. & Standley, D. M. MAFFT multiple sequence alignment software version 7: Improvements in performance and usability.
Mol. Biol. Evol. 30, 772–780 (2013).
29. Criscuolo, A. & Gribaldo, S. BMGE (Block Mapping and Gathering with Entropy): A new software for selection of phylogenetic
informative regions from multiple sequence alignments. BMC Evol. Biol. 10, 210, https://doi.org/10.1186/1471-2148-10-210 (2010).
30. Lefort, V., Longueville, J.-E. & Gascuel, O. SMS: Smart Model Selection in PhyML. Mol. Biol. Evol. 34, 2422–2424 (2017).
31. Le, S. Q. & Gascuel, O. An improved general amino acid replacement matrix. Mol. Biol. Evol. 25, 1307–1320 (2008).
32. Anisimova, M. & Gascuel, O. Approximate likelihood-ratio test for branches: A fast, accurate, and powerful alternative. Syst. Biol.
55, 539–552 (2006).

Acknowledgements
E.F. received a János Bolyai Research Scholarship from the Hungarian Academy of Science (BO/00093/18/8).
N.K. and E.F. were supported by the New National Excellence Program of the Hungarian Ministry for Innovation
and Technology, grants ÚNKP-19-3 and ÚNKP-19-4, respectively. This work was supported by the European
Union and co-financed by the European Regional Development Fund under grant GINOP-2.3.2-15-2016-00008;
by the EFOP-3.6.1-16-2016-00022 project co-financed by the European Union and the European Social Fund;
and by the Hungarian National Research, Development and Innovation Fund, grants KH 129602 and NN128867,
to L.K.

Author contributions
E.F., M.F. and L.K. conceived this study. N.Á. and M.F. discovered the new complex intervening sequences in silico
and mined data. N.K. and F.P. performed bench experiments. E.F. supervised the experimental work. All Authors
analysed and discussed obtained data. M.F., C.S. and E.F. wrote the manuscript with contributions from N.Á. and
L.K. All Authors read and approved the submitted manuscript.

Competing interests
The authors declare no competing interests.

Additional information
Supplementary information is available for this paper at https://doi.org/10.1038/s41598-020-63239-6.
Correspondence and requests for materials should be addressed to E.F.
Reprints and permissions information is available at www.nature.com/reprints.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and
institutional affiliations.

Scientific Reports | (2020) 10:6022 | https://doi.org/10.1038/s41598-020-63239-6 9

www.nature.com/scientificreports/ www.nature.com/scientificreports

Open Access This article is licensed under a Creative Commons Attribution 4.0 International
License, which permits use, sharing, adaptation, distribution and reproduction in any medium or
format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Cre-
ative Commons license, and indicate if changes were made. The images or other third party material in this
article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the
material. If material is not included in the article’s Creative Commons license and your intended use is not per-
mitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the
copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.

© The Author(s) 2020

Scientific Reports | (2020) 10:6022 | https://doi.org/10.1038/s41598-020-63239-6 10