Available via license: CC BY 4.0
Content may be subject to copyright.
Ecology and Evolution. 2024;14:e10940.
|
1 of 19
https://doi.org/10.1002/ece3.10940
www.ecolevol.org
1 | INTRODUCTION
The repeated shift from outcrossing to selfing is a central topic in
plant evolution (Cutter, 2019; Stebbins, 1957). The significance of
this transition lies in its central role in altering the partitioning of
genetic diversity within and among populations, thus influencing
how populations respond to natural selection (Barrett, 2019; Wright
et al., 2013). Additionally, mating- system shifts can affect species
longevity, with outcrossing species being more resilient and selfing
species more prone to extinction (de Vos et al., 2014; Goldberg et al.,
Received:21July2023
|
Revised:2D ecemb er2023
|
Accepted :3Januar y2024
DOI:10.1002 /ece3.10940
RESEARCH ARTICLE
Genomic analyses elucidate S- locus evolution in response to
intra- specific losses of distyly in Primula vulgaris
E. Mora- Carrera1 | R. L. Stubbs1 | G. Potente1 | N. Yousefi1 | B. Keller1 |
J. M. de Vos2 | P. Szövényi1 | E. Conti1
This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium,
provided the original work is properly cited.
© 2024 The Authors. Ecology and Evolution published by John Wiley & Sons Ltd.
1Depar tmentofSyste maticand
Evolutionary Botany, University of Zurich,
Zurich, Switzerland
2Depar tmentofEnvironmentalS ciences
– Botany, University of Basel, Basel,
Switzerland
Correspondence
E.Mora-CarreraandE.C onti,Depa rtment
of Systematic and Evolutionary Botany,
University of Zurich, Zurich, Switzerland.
Email: emiliano.mora@systbot.uzh.ch and
elena.conti@systbot.uzh.ch
Funding information
Swiss Government Excellence Scholarship,
Grant/Award Number: 2018.0475;
Forschungskredit Candoc, Grant/Award
Number: K- 74202- 05- 01; Schweizerischer
Nationalfonds zur Förderung der
Wissenschaftlichen Forschung, Grant/
Award Number: 310030_185251, 3100-
061674.00/1 and 175556
Abstract
Distyly,afloraldimorphismthatpromotesoutcrossing,iscontrolledbyahemizygous
genomic region known as the S-locus . Disruptions of genes wi thin the S- locus are
responsible for the loss of distyly and the emergence of homostyly, a floral mono-
morphism that favors selfing. Using whole- genome resequencing data of distylous
and homostylous individuals from populations of Primula vulgaris and leveraging high-
quality reference genomes of Primula we tested, for the first time, predictions about
the evolutionary consequences of transitions to selfing on S-genes.Ourresultsreveal
a previously undetected structural rearrangement in CYPᵀ associated with the shift
to homostyly and confirm previously reported, homostyle- specific, loss- of- function
mutations in the exons of the S- gene CYPᵀ. We also discovered that the promoter
and intronic regions of CYPᵀ in distylous and homostylous individuals are conserved,
suggesting that down- regulation of CYPᵀ via mutations in its promoter and intronic
regions is not a cause of the shift to homostyly. Furthermore, we found that hemizy-
gosity is associated with reduced genetic diversity in S- genes compared with their
paralogs outside the S- locus. Additionally, the shift to homostyly lowers genetic di-
versity in both the S- genes and their paralogs, as expected in primarily selfing plants.
Finally, we tested, for the first time, long- standing theoretical models of changes in
S- locus genotypes during early stages of the transition to homostyly, supporting the
assumption that two copies of the S- locus might reduce homostyle fitness.
KEYWORDS
hemizygosity, heterostyly, mating- system transitions, Primula, S- locus
TAXONOMY CLASSIFICATION
Genetics
2 of 19
|
MORA-CARRERA et al.
2010). Previous studies used phenotypic traits typically associated
with selfing to estimate, for example, the number and tempo of tran-
sitions to selfing in phylogenies of ancestrally outcrossing taxa (de
Vos et al., 2014; Goldbe rg & Igic, 2012). However, missing knowl-
edge of the genes that control mating systems has hindered the
study of molecular processes associated with transitions to selfing
until recently, especially in non- model organisms. Current advances
in genomics now facilitate the identification of the genes and muta-
tions associated with mating- system shifts.
A prime model to investigate the transition from outcrossing
to selfing has been the shift from distyly to homostyly in Primula
(Barrett, 2019).Dist yl yischa ra cteri zed byt hec o-oc curre nceinp op-
ulations of two floral types (called the pin and thrum, respectively)
of self- incompatible individuals distinguished by the reciprocal ar-
rangement of male (anthers) and female (stigma) sexual organs in
their flowers (Figure 1a; Ganders, 1979; Keller et al., 2014; Lloyd &
Webb, 1992). Specifically, pins are characterized by having the
stigma positioned above t he anther level within the flowers, whereas
thrums have the anthers above the stigma level. This floral hetero-
morphism represents an adaptation for outcrossing reported in at
least 26 angiosperm families (Naiki, 2012). Conversely, homostyly
isafloralhomomorphismthatenablesselfing.Itischaracterizedby
self- compatible individuals bearing flowers with both stigma and an-
thers at the same level in the corolla tube (Figure 1a; Barrett, 2019).
Evidence supporting higher selfing in homostylous than distylous
plants has been reported both intra- and inter- specifically (Mora-
Carrera et al., 2021; Schoen et al., 1997; Zhao et al., 2023; Zhong
et al., 2019). Additionally, independent shifts from distyly to ho-
mostyly have been documented both within and between species
(Costa et al., 2019; de Vos et al., 2014; Kissling & Barrett, 2013; Ruiz-
Martín et al., 2018; Zhou et al., 2012).
IthaslongbeenknownthattheS- locus, a supergene, controls
distyly and the shift to homostyly (Lewis & Jones, 1992). However,
the molecular and functional characterization of the S- locus has
been achieved only recently. The breakthrough occurred in Primula,
where the S- locus comprises five genes known as S- genes (CCMT,
GLOT, CYPT, PUMT, and KFBT) and is hemizygous in thrums (S/0)
but absent in pins (0/0; Figure 1a; Li et al., 2016; Potente, Léveillé-
Bourret , et al., 2022). Four of the five S- genes (i.e., CCMT, GLOT, CYPT,
and KFBT) are the product of gene duplications, and their paralogs
(named CCM1, GLO1, CYP734 A51 , and KFB1) have been localized
in Primula veris, where they are scattered throughout the genome
(Potente, Léveillé-Bourret, et al., 2022). The function of S- genes has
been experimentally characterized only for GLOT and CYPT, which
were recently shown to control key traits in thrum flowers: GLOT
determines long anthers, while CYPT determines shor t stigma and fe-
male self- incompatibility (Huu et al., 2016, 2020, 2022). Specifically,
experimental silencing of GLOT in Primula forbesii thrums lowered
anther position, producing flowers with both anthers and stigma
in the middle of the corolla tube (i.e., short- homostyly). However,
self- incompatibility was retained, preventing self- fertilization in
short- homostyles (Huu et al., 2020). Conversely, silencing of CYPT
in P. ve ris was associated with both style elongation and loss of
self- incompatibility, thus turning self- incompatible thrum flow-
ers into self- compatible, homostylous flowers with both stigma
and anthers at the mouth of the corolla tube (i.e., long- homostyly;
Huu et al., 2016, 2022). Although both short- and long- homostyly
have been reported, long- homostyly is more common in Primula
(Charlesworth & Charlesworth, 1979; Lewis & Jones, 1992). The
likely explanation is that, as reported above, CYPT disruption causes
both reduced herkogamy and self- compatibility (Huu et al., 2016,
2020, 2022), thus enabling self- pollination and the production of
selfed seed in long- homostylous flowers (Piper et al., 1986), whereas
short- homostyles with disrupted GLOT, but functional CYPT, can-
not produce selfed seed despite their reduced herkogamy, for they
retain self- incompatibility (Huu et al., 2022). Given the higher fre-
quency of long- homostyles and the rarity of short homostyles in
Primula, for simplicity's sake, we hereafter refer to long- homostyly
as homostyly (Figure 1a).
The shift from distyly to homostyly has been intensely studied
in populations of P. vulgaris from Somerset, England, that display
frequency variation of thrums, pins, and homostyles (Crosby, 1940,
1949). Targeted Sanger sequencing of the five CYPT exons revealed
that all tested thrums from the mentioned region shared the same
functional CYPT allele (CYPT- 1; Figure 1b), whereas 21 homostyles
harbored six different CYPT alleles, each with a unique, potentially
disruptive mutation (CYPT- 2 to CYPT- 7 in Mora- Carrera et al., 2021;
Figure 1b). One pos sible explanatio n for the lack of shared CYPT
mutations among the homostyles is that homostyly evolved inde-
pendently multiple times. However, the same study also found that
six homostyles from two different populations had the same CYPT
allele as that of thrums (i.e., CYPT- 1). This result raised the possibility
that homostyly initially arose via CYPT silencing caused by either a
structural rearrangement (e.g., inversion or translocation) involving
any of the CYPT exons or an inactivating mutation in the CYPT pro-
moter or a mutation in one of the intronic regions disrupting exon
processing, subsequently followed by multiple, unique mutations in
CYPT exons, as those found in CYPT- 2 to CYPT- 7 (Charlesworth, 2022;
Mora- Carrera et al., 2021). The three types of mutations mentioned
above cannot be detected using data from Sanger sequencing of
individual CYPT exons but can be detected by analyzing data from
Whole Genome Resequencing (WGR). Furthermore, determining
whether homostyly in P. vulgaris arose multiple times via indepen-
dent mutations in CYPT exons or once through any of the three
mentioned types of mutations requires a high- quality annotation
of the S- genes in a reference genome from the same or a closely
related species and WGR data covering coding and non- coding re-
gions of CYPT. Both types of resources, that is, WGR and reference
genomes, are now available, enabling an expanded analysis of the
molecular mechanisms possibly causing the transition to homostyly
in P. vulgaris.
The recently published high- quality annotation of the five S-
genes and t heir four paralog s (Potente, Léveillé-Bo urret, et al., 2022),
combined with newly generated sequences of these nine genes ex-
tracted from WGR data of thrums and homostyles, provide an ideal
opportunity to test how the genetic architecture of the S- locus and
|
3 of 19
MORA-CARRER A et al.
the transition to homostyly affect genetic variation and selection
on S- genes and their paralogs. First, the thrum- specific segregation
of the hemizygous S- locus should cause a 3/4th reduction of effec-
tive population size (Ne), lowering genetic diversity in S- genes (Huu
et al., 2016). Moreover, hemizygosity could have contrasting effects
on the efficacy of purifying selection on S-genes.Ontheonehand,
the reduction of Ne in the S- genes should make purifying selection
less efficient when compared to their paralogs (Huu et al., 2016).On
the other hand, similarly to what happens in the Y sex- chromosome
(Gossmann et al., 2011), selection to maintain function of S- genes,
and the exposure of recessive deleterious mutations under hemizy-
gosity should enhance the efficacy of purifying selection when com-
pared to their paralogs. Second, the transition to homostyly should
reduce genetic diversity in both S- genes and their paralogs due to
FIGURE 1 DistylyandhomostylyinPrimula vulgaris. (a) Phenotypes (top) and genotypes (bottom) of distylous (thrum; pin) and
homostylous individuals in P. vulgaris. Short- styled flowers (thrums) have male sexual organs (anthers) above the short female sexual organ
(style), while long- styled flowers (pins) have anthers below the long style. The reciprocity of sexual organs promotes pollen transfer between
floralmorphs,whileself-andintra-morphself-incompatibilitypreventsselfing.DistylyiscontrolledbytheS- locus, which is hemizygous in
thrums [S/0] and absent in pins [0/0]. Homostyly arises primarily through the disruption of the CYPT gene located in the S- locus, causing
both stigma elongation and loss of self- incompatibility. Homostyles can have one [S*/0] or two copies [S*/S*] of the S- locus with a disrupted
CYPT; S- locus haplotypes with disrupted CYPT are designated with S*. Black arrows indicate compatible pollen transfers: distyly promotes
outcrossing, homostyly enables selfing. (b) Graphical representation of the seven CYPT alleles from natural populations of P. vulgaris in
Somerset, England, reported in Mora- Carrera et al. (2021): CYPT- 1 designates the functional copy in thrums and CYPT- 2 to 7 designate
alleles with deletions (red triangles), non- synonymous (yellow triangles), and synonymous (gray triangles) mutations in different exons of the
homostyles. The CYPT- 8 allele (gray box), first reported here, is characterized by the loss of exon 1. (c) Predictions of changing frequencies
of S- locus genotypes over generations assuming that homostyles with S*/S*- genotypes have 35% lower viability than thrums, pins, and
S*/0- homostyles. Equilibrium is reached when homostyles with S*/S*- vs. S*/0- genotypes segregate at roughly equal frequencies, while pin
genotypes (0/0) are maintained at low frequencies and thrum genotypes (S/0) disappear from the population (c). (d) Predictions of changing
frequencies of S- locus genotypes over generations assuming that S*/S*- homostyles have equal viability as thrums, pins, and S*/0- homostyles
(see Table S1). Equilibrium is reached when S*/S*- homostyles become fixed in the population. Plots in (c) and (d) were generated in R v3.6
using original equations of Crosby's model (Crosby, 1949). At the phenotypic level, the model predicts a sharp increase of homostyles over
generations at the expense of thrums, with pins remaining either at low frequencies (1c) or disappearing (1d). Calculating the actual counts
of S*/S*- and S*/0- genotypes in natural populations had been impossible until recently because discriminating between the two types of
homostylous genotypes requires knowledge of S- locus genes and ability to determine whether the S- locus is haploid or diploid (see Table 4).
4 of 19
|
MORA-CARRERA et al.
increased homozygosity in homostyles (Mora- Carrera et al., 2021).
However, the extent to which the efficacy of purifying selection dif-
fers between S- genes and their paralogs and whether the transition
to homostyly reduces genetic diversity in these genes remain poorly
understood (Potente, Léveillé-Bourret, et al., 2022).
Additionally, the WGR data newly presented here allowed us to
assess the ploidy level of specific genomic regions via the analysis of
read depth (Gutiérrez- Valencia et al., 2022), thus enabling the test-
ing of long- standing predictions, first proposed by Crosby (1949),
regarding the changing frequencies of S- locus genotypes during
the transition from distyly to homostyly. In a pioneering study,
Crosby (1949) proposed a model for how the frequencies of thrums,
pins, and homostyles change over time (Figure 1c,d). The model
developed by Crosby (1949) assumed that thrums were typically
heterozygous dominant at the S- locus, pins were homozygous re-
cessive, and homostyles stemmed from thrums via recombination
at the S- locus (Bateson & Gregory, 1905). Crosby assumed that the
viability (i.e., the proportion of germinated seeds that developed
into a seed- producing plant) of homozygous homostyles was either
lower than or equal to the viability of pins, thrums, and heterozygous
homostyles. Crosby's assumption of lower viability for homozygous
homostyles rested on previous crossing experiments by Mather and
Winton (1941) suggesting that homozygous dominant thrums (S/S)
had lower viability than heterozygous thrums. His model is appli-
cable also under the recently demonstrated hemizygosity of the S-
locus in Primula by assuming that the viability of homostyles with
a diploid disrupted S- locus (S*/S*- genotypes) is either 35% lower
than or equal to the viability of homostyles with a haploid S- locus
(S*/0- genotype), thrums (S/0- genotype), and pins (0/0- genotype;
see Figures 1c,d).
In P. vulgaris, repeated phenotypic surveys conducted in
Somerset, England, have shown that, when homostyles are present
at high freq uency, thrums tend to b e less frequent an d, in some cases ,
absent, compared with pins (Crosby, 1949; Curtis & Curtis, 1985;
Mora- Carrera et al., 2021). These findings align with the predic-
tions of Crosby's model under lower viability of S*/S*- homostyles
(Figure 1c). However, one Somerset population consisted exclusively
of homostyles (Curtis & Curtis, 1985; Mora- Carrera et al., 2021),
suggesting that the fixation of homostyly is possible, as expected
under the model with equal viability for S*/S*- and S*/0- homostyles.
Of note, previous phenotypic surveys could not discriminate be-
tween S*/S*- and S*/0- genotypes for homostyles due to the lack
of adequate sequencing tools and molecular knowledge of the
S- locus. Recently developed sequencing technologies enable the
estimation of sequencing depth at the S- locus (Gutiérrez- Valencia
et al., 2022), allowing us to determine whether the S- locus is haploid
or diploid in homostylous and thrum individuals (the S- locus being
absent from pins). Therefore, it is now possible to estimate whether
the observed frequencies of S*/S*- and S*/0- homostyles in natural
populations support the model assuming lower or equal viability for
S*/S*- genotypes in relation to the other genotypes, testing Crosby's
genotypic predictions for the first time.
Here, we analyze new WGR data from nine populations of P. v ul-
garis with var ying frequencies of pins, thrums, and homostyles, to an-
swerthef ollowingques tions:(1)Doallh omostylescarr yingdifferent
disrupted CYPT exons share the same mutation in its promoter region,
or in one of the CYPT introns, and/or a structural rearrangement in-
volving CYPT exons that might disrupt CYPT function or expression,
allowingforthepossibilityofasingleoriginofhomostyly?(2)Dothe
haploid S- genes have lower genetic diversity and ef ficacy of purif ying
selection than their diploid paralogs, which are located outside the S-
locus?(3)DoS- genes and their paralogs have lower genetic diversity
inhomostylesthan inthrums?(4)Doobser vedfrequencies ofS*/ 0-
and S*/S*- homostyles in natural populations better match genotypic
frequencies predicted under the assumptions of lower or equal viabil-
ity for S*/S*-homost yles?Ourstudyillustrateshowrecentlyacquired
knowledge of the genes controlling mating systems combined with
high- quality genomic resources can generate novel insights into the
genotypic changes and evolutionary consequences associated with
phenotypic transitions from outcrossing to selfing.
2 | MATERIALS AND METHODS
2.1 | Study species
Primula vulgaris (primrose) is an ancestrally heterostylous (de Vos
et al., 2014; Mast et al., 2006), diploid (2n = 22), perennial, rosette-
formingplantbloomingfromFebruarytoApril.Distylouspopulations
of P. vulgaris occur across Eurasia, inclu ding in Turkey, most of Western
Europe,largerMediterraneanIslands,andallBritishIsles(Jacquemyn
et al., 2009).Incontrast,populationswithvaryingfrequenciesofho-
mostyles have been discovered only in Somerset and Chiltern Hills,
England (Crosby, 1940, 1949), and one population in the Netherlands
(Barmentlo et al., 2017). Habitat fragmentation due to intensive pas-
toral and agricultural activities in these areas were suggested as the
potential selective pressure favoring the shift to self- fertilizing homo-
styles in these populations (Mora- Carrera et al., 2021).
2.2 | Sampling and sequencing of plant material
In spring 2 019, we collected l eaf tissue from 105 in dividuals of P.
vulgaris, comprising 74 heterostyles (37 pins and 37 thrums) and 31
homostyles. The samples were obtained from nine populations, in-
cluding: six dimorphic populations (comprising only pins and thrums)
fromTurkey(TR-D),Slovakia(SK-D),Switzerland(CH-D)andEngland
(EN1-D,EN2-D,andEN3-D);twotrimorphicpopulations(comprising
pins, thrums, and homostyles) from England (EN4- T and EN5- T); and
one monomorphic population (comprising only homostyles) from
England (EN6- M) (Table 1). Populations EN4- T, EN5- T, and EN6- M
corresponded to populations T04, T07, and M01, respectively,
from our previous study (Mora- Carrera et al., 2021). Moreover, the
11 homostyles from EN6- M analyzed here included three of the
|
5 of 19
MORA-CARRER A et al.
homostyles carrying CYPT- 1 (detected with Sanger sequencing) from
population M01 of Mora- Carrera et al. (2021; Figure 1b).
To investigate possible molecular mechanisms for the transition
to homostyly not detected before and the consequences of such
transition on S- genes and their paralogs in P. vulgaris, we gener-
ated WGR data for all the above 105 individuals, asfollows. DNA
extractions were performed using the Maxwell extraction method
(Promega, USA) at the Functional Genomics Center Zurich (Zurich,
Switzerland).Librarypreparationandpaired-endsequencing(150 bp)
were condu cted by RAP iD GENOMICS (G ainesville , Florida, USA )
using Illumina Novaseq6000 platform, generating 7,077,866,510
paired- end sequencing reads with an average sequencing depth of
18.9 (±SD = 3.08).
2.3 | Mapping and variant calling of WGR data to
CYPT of the P. v ulgaris genome
To determine the sequences of all five exons and four introns of
CYPT, along with it s up- and down- stream intergenic regions, we first
produced a high- quality reference S- locus assembly for P. vulgaris
by replacing all S- locus contigs (LH_v2_0002458, LH_v2_0067593,
LH_v2_0003915, and LH_v2_0000241) in the genome assembly
of P. vulgaris (Cocker et al., 2018) with a published, highly contigu-
ous 450 kb sequenceof the P. vulgaris S- locus (Li et al., 2016). We
then mapped WGR reads from all 105 individuals of P. vulgaris to
the S- locus assembly of the same species generated above. To de-
termine the position of CYPT in the reference S- locus assembly, we
aligned the coding sequence of CYPT against the reference using
the query function of blastn with default parameters, which is part
ofthe NCBIBLAST+ toolkit v2.6.0 (Camacho et al., 2009). The de-
scribed approach enabled accurate infra- specific alignment of WGR
reads to CYPT intronic and intergenic regions, which are expected
to be highly variable because they are under lower selection than
exonic regions. Prior to mapping, Illumina adapters were clipped
from raw reads with Trimmomatic v0.38 (Bolger et al., 2014) using
default parameters. Mapping was performed using BWA- mem v7.17
(Li & Durbi n, 2009) with default parameters. As negative control,
pin individuals (0/0) were included in the analysis and, as expected,
none of the sequencing reads from the 37 pins mapped to the S-
locus. Duplicatedreads weremarkedwiththeMarkDuplicatestool
included in Picard v2.18.4 (http:// broad insti tute. github. io/ picard/ ).
Variant calling of SNPs and of insertions and deletions (indels) for
the S- locus was conducted using HaplotypeCaller, implemented in
the Genome Analysis Toolkit (GATK) v4.1.2.0 (McKenna et al., 2010)
pipeline. Finally, SNP variants were filtered from the Variant Call
Format (VCF) file using the SelectVariants with the following filters:
q u a l i t y - b y - d e p t h ( Q D ) > 2.0;mappingquality(MQ) > 40.0;strandbias
(FS) < 60.0; mapping qualit y rank-sum test (MQRankSum) > −12.5;
a rank-sum test (ReadPositionRankSum) > −8.0; site read depth
(DP > ½X ) || (DP < 3X). Addit ionally, sites wit h fixed heter ozygosity
(i.e., InbreedingCoeff < −0.99), likely representing incorrect SNP
calling(O'Learyetal.,2018; Pavan et al., 2020), were filtered out.
2.4 | Identification of mutations putatively
disrupting CYPT function in P. v ulgaris
2.4.1 | DisruptivemutationsinCYPT coding regions
To identify potential homostyle- specific, loss- of- function CYPT mu-
tations, including non- synonymous mutations, insertions, and de-
letions, we compared the sequences of the five CYPT exons in 31
homostyles with the functional CYPT allele of the 37 thrums. For
this, we extracted the respective sequences of the five CYPT exons
from the S- locus VCF file using the intersect function included in
BEDtools v2.29. 2(Quinlan& Hall, 2010) and converted them into
a single FASTA file using vcf2phylip.py (https:// github. com/ edgar
TABLE 1 SummaryofsampledPrimula vulgaris populations, collected individuals, and frequencies of S- locus genotypes (S/0 or S/S, 0/0,
S*/0, and S*/S*).
Population Latitude Longitude Population type
Collected individuals
(thrum:pin:homostyle)
Thrum Pin Homostyle
S/0: S/S0/0 S*/0 S*/S*
T R - D 40 .74 39.55 D5:5:– 0.5 0.5 – –
S K - D 48.77 19.05 D5:5:– 0.5 0.5 – –
CH-D 46.44 6.91 D5:5:– 0.5 0.5 – –
E N 1 - D 52.14 − 0.13 D5:4:– 0.55 0.45 – –
E N 2 - D 51.35 −2. 33 D4:5:– 0.22:0.22 0.55 – –
EN3-D 51. 09 −2 .71 D5:5:– 0.5 0.5 – –
EN 4- T 51. 08 −2 .3 0 T3:4:10 0.17 0.24 0.24 0.35
EN5- T 51.13 −2. 42 T4:5:10 0.16:0.04 0.27 0.32 0.21
EN6- M 51.03 −2 .6 3 M–:–:11 – – – 1
Note: Ploidy level (haploid vs. diploid) of S- locus genotypes was identified by estimating the sequencing depth of S- locus genes relative to the
sequencing depth of genome- wide coding regions. Population abbreviations—geographic origin: CH, Switzerland; EN, England; SK, Slovakia; TR,
Turkey;digitfollowinggeographicacronymindicatespopulationnumber;capitalletterfollowingdashindicates:D,dimorphicpopulationwithpins
and thrums; M, monomorphic population consisting entirely of homostyles; T, trimorphic population with pins, thrums, and homostyles.
6 of 19
|
MORA-CARRERA et al.
domor tiz/ vcf2p hylip ). The sequence alignment of all exons and the
detection of putatively disruptive mutations in CYPT of homostyles
and thru ms were perform ed with MEGA X (Kumar et al ., 2018).
Finally, we compared the resulting sequence alignment with an
alignment of previously detected mutations in CYPT exons reported
by Mora- Carrera et al. (2021).
2.4.2 | MutationsinCYPT intronic regions
To determine whether the shift to homostyly is associated with
mutations involving CYPT introns that might affect intron splicing
(causing, e.g., inactivating reading frameshifts), we examined the
alignment of all four introns in the sampled thrums (37 individu-
als) and homostyles (31 individuals) to detect any single- nucleotide
polymorphism (SNP) fixed in the homostyles but absent in thrums.
2.4.3 | StructuralrearrangementsinCYPT
To determine whether the shift to homostyly is associated with
structural rearrangements involving CYPT exons, we examined the
paired- end sequencing reads mapped to the introns and exons of
CYPT in both thrums (as a reference) and homostyles using the
InteractiveGenomic Viewer (IGV)v2.8.6(Robinson etal.,2011).
Translocations can be identified by analyzing the mapped paired-
end reads, where one read is mapped to one position in the ge-
nome (e.g., CYPT in the S- locus), while its mate- pair is mapped to
a different position, either in the same or different chromosome.
Inversion s can be detec ted by compari ng the orient ation of the
mapped read- pairs to CYPTinthereferencegenome.Ifthereisa
small inversion, both mapped paired- end reads should be oriented
in the same direction (→→ or ←←), whereas in the absence of a
structural change normal mapped paired- end reads should be ori-
ented toward each other (→←). Finally, deletions are characterized
by drops in sequencing read coverage at specific positions in the
genome, in this case, within CYPT. We characterized a deletion as
a complete absence of mapping depth and a site read depth equal
tozero(DP = 0).
2.4.4 | DisruptivemutationsinCYPT
promoter region
To determine whether mutations in the promoter region are respon-
sible for CYPT loss of function in homostyles, we conducted an anal-
ysis to identify the putative CYPT transcription- factor binding- site
andsearcheditforhomostyle-specificmutations.Itisexpectedthat
the3 kbregionupstreamofageneofinterestcontainsitspromoter,
including the transcription- factor binding- site (Yu et al., 2016); thus,
wefirstextracted andalignedthe3 kbsequenceupstream ofCYPT
exon1from20thrumsindimorphicpopulationsTR-D,SK-D,CH-D,
and EN1-D of P. vulgaris. Considering that the transcription- factor
binding- site important for distyly should be relatively conserved
across sp ecies, we include d the 3 kb sequence u pstream of CYPT
from the reference genome of P. ve ris (Potente, Léveillé-Bourret,
et al., 2022) to our aligned sequence from the 20 individuals of P.
vulgaris. To identify motifs enriched for transcription- factor binding-
sites in the aligned dataset of relevant sequences from 20 thrums
of P. vulgaris and one P. veri s individual (see above), we employed
the Simple Enrichment Analysis (SEA) implemented in the Multiple
Em for Motif Elicitation (MEME) suite program v5.5.0 (McLeay &
Bailey, 2010). We used the Arabidopsis thalianaDAPmotifsdatabase
(O'Malleyetal.,2016) and default parameters to identify the afore-
mentioned enriched motifs. Finally, after identifying enriched motif
sequences in thrums, we compared all 37 thrums and 31 homostyles
byaligningtheenrichedmotifs usingMEGAX (Kumar etal.,2018)
and inspected the alignment to check whether any SNP was fixed
in the homostyles but absent in thrums. To determine the poten-
tial function of the identified enriched motifs, we consulted The
ArabidopsisInformationResource(TAIR)(http:// www. arabi dopsis.
org).
In addition to detecting transcription-factor binding-sites, we
searched for additional cis-regulator yelementsinthe3 kbregionup-
stream of CYPT using PlantCARE (Lescot et al., 2002) as performed in
Henning et al. (2023). For this analysis, we obtained the sequences
ofthe3 kbupstreamofCYPT from the reference S- locus genomes of
P. vulgaris (Cocker et al., 2018) and P. ve ris Potente, Léveillé-Bourret,
et al., 2022).Oncetheputativecis- regulatory elements were identi-
fied, we searched for mutations in these elements that were fixed in
homostyles but absent in thrums.
2.5 | Genetic variation in
S- genes and their paralogs
To quantify genetic variation in the five S- genes (CCMT, GLOT, CYPT,
PUMT, and KFBT) and their four paralogs (CCM1, GLO1, CY P73 4A 51,
and KFB1), we calculated nucleotide diversity (π) at synonymous
(πS) and non- synonymous (πN) sites. For this analysis, we mapped
the WGR reads of P. vulgaris to the chromosome- scale reference
genome of P. ver is (Potente, Léveillé-Bourret, et al., 2022) because
it provides a better annotation of all S- genes and their paralogs
outside the S- locus than the P. vulgaris reference genome. Prior to
mapping, we annotated sites at 4- and 0- fold degenerate sites as
synonymous and non- synonymous sites, respectively, for all nine
genes using the script NewAnnotateRef.py (https:// github. com/
fabby rob/ scien ce/ tree/ master/ pileup_ analy zers [last accessed on
March 25, 2022]) (Williamson et al., 2014). Mapping of sequencing
reads, variant calling, and filtering were performed with BWA- mem
v7.17(Li&Durbin,2009) and GATK v4.1.2.0 (McKenna et al., 2010)
as described in Section 2.3. Subsequently, we split the VCF files of
each of the nine genes into synonymous and non- synonymous sites
usingBEDtoolsv2.29.2(Quinlan&Hall,2010). Using these VCF files ,
we estimated πS and πN with pixy v.1.0.0 (Korunes & Samuk, 2021).
Specifically, we estimated πS and πN of S- locus paralogs in hetero-
styles (i.e., pins and thrums combined) and homostyles, while the
estimates of πS and πN for S- genes included thrums (since pins lack
|
7 of 19
MORA-CARRER A et al.
the S- locus) and homostyles. Finally, we estimated the strength of
purifying selection in all S- genes and their paralogs by calculating
the ratio of nucleotide diversity at non- synonymous vs. synonymous
sites (πN/πS).
2.6 | S- locus genotypes in natural populations of
P. v ulgaris
To determine whether homostyles and thrums have either a hap-
loid (i.e., S*/0 or S/0, respectively) or diploid (i.e., S*/S* or S/S, re-
spectively) S- locus, we calculated the relative sequencing depth of
the S- locus (RelS- locus depth), as follows. We first estimated the aver-
age site depth of the S- locus coding regions (depthS- locus) and of the
genome- wide coding regions (depthGenome- wide) from filtered VCF
files using BCFtools v1.9, as indicated in Section 2.5, then we cal-
culated RelS- locus depth as depthS- locus/depthGenome- wide. This normali-
zation allowed us to account for differences of sequencing depth
among individuals. A RelS- locus depthvalueofapproximately0.5 ± 0.25
indicates a haploid S- locus (S*/0 or S/0 for homostyles and thrums,
respec tively), whil e a value close to 1 ± 0. 25in dicates adiploid S-
locus (S*/S* or S/S for homostyles and thrums, respectively).
To determine whether the proportions of individuals carrying
0/0- , S/0- , S*/S*- , and S*/0- genotypes in natural populations sup-
port the model with lower or the one with equal viability of S*/S*-
homostyles (Figure 1c,d), we compared the observed frequencies of
the four genotypes in EN4- T, EN5- T, and EN6- M with the expected
genotype frequencies predicted by Crosby's model (Crosby, 1949;
Figure 1c,d).Inbrief,the modelestimates theproportionofpin(p),
thrum (q), and the two types of homostylous (r and s) genotypes
in a population at each generation using the proportion of p, q, r,
and s individuals of the previous generation (Table S1). Specifically,
p, q, r, and s at each generation are calculated by summing up the
expected offspring proportion of pins, thrums, and homostyles fol-
lowing each possible cross in the population (see Table S1 for all
possible crosses). All offspring is equally viable except for the S*/S*-
homostyles, whose viability can be either equal or 35% lower than
that of the other genotypes (viability whose fitness is determined
by the proportion of germinated seeds that develop into a seed-
producing plant; v in Table S1). Additionally, the original model as-
sumed that pins have a selfing rate of 0.10 (Crosby, 1949). However,
eliminating the assumption of selfing in pins does not change any of
the results of the model (de Jong & Klinkhammer, 2005). Crosby's
original model rested on the previous genetic model for the S- locus,
which assumed that thrums were typically heterozygous dominant
at the S- locus (i.e., S/s), pins homozygous recessive (i.e., s/s), and ho-
mostyles had one or two copies of a disrupted S- locus (i.e., s*/s and
s*/s*). The recently discovered hemizygosity of the S- locus does not
change any assumption of the simulation, which can be adjusted by
assuming that S/s, s/s, s*/s, and s*/s* correspond to S/0, 0/0, S*/0, and
S*/S*, respectively (Figures 1c,d).
Using Crosby's equations (1949), we first calculated the ex-
pected frequencies of all four genotypes (0/0- , S/0- , S*/ S*- , and
S*/0) at generations 10, 20, 30, and 40 after the onset of homostyly,
specifying either lower (v = 0.65) or equal (v = 1) via bility of S*/S*-
homostyles. Since natural frequencies of the four genotypes might
be compatible with levels of viability not included in Crosby's (1949)
model, we additionally estimated expected genotype frequencies
under v = 0.9,0.8,0.7,0.6,and0.5.Viability values below 0.5were
not used because phenotypic frequencies reflecting these condi-
tions (roughly equal frequencies of pins and homostyles and absence
of thrums) have never been reported in natural populations. Second,
we estimated the observed frequencies of all four genotypes as fol-
lows. We tallied the raw counts of pins (0/0- genotypes) and thrums
(S/0- genotypes) in EN4- T and EN5- T based on previous population
surveys (T04 and T07, respectively, in Mora- Carrera et al., 2021;
Section 2.2, here). Furthermore, we calculated the number of homo-
styles with S*/0- and S*/S*- genotypes in EN4- T, EN5- T, and EN6- M
by multiplying the proportion of S*/0- and S*/S*- genotypes reported
here (see Section 3.3) by the raw number of homostyles in each of
the three populations. Finally, to determine whether observed and
expected genotype frequencies are compatible with Crosby's model
under lower or equal viability of S*/S*- homostyles at 10, 20, 30, and
40 generations, we used chi- squared tests with Bonferroni cor-
rections. A significant difference between observed and expected
frequencies indicates that observed genotype frequencies are not
compatible with Crosby's model, while non- significant results indi-
cate that they are compatible with it.
3 | RESULTS
3.1 | Mutations putatively disrupting CYPT function
in P. v ulgaris
The use of newly generated WGR data enabled the identification
of novel mutations in the exonic, intronic, and upstream regions of
CYPT and exonic rearrangements in CYPT that could not be discov-
ered by using the exon- by- exon Sanger sequencing approach em-
ployed in a previous study (Mora- Carrera et al., 2021).
3.1.1 | DisruptivemutationsinCYPT coding region
We discovered four, never- before reported, non- synonymous CYPT
mutations in different exons of thrum individuals. Specifically, in the
37 thrums analyzed here, two non- synonymous mutations were de-
tected in exon 1 of four out of five thrums from the Slovakian popula-
tionSK-D;onenon-synonymousmutationwasfoundinexon3oftwo
thrums from the English population EN5- T; and one non- synonymous
mutatio n was identified i n exon 5 of one thrum from t he Swiss popula-
tionCH-D.Incontrasttothenon-synonymousmutationsdiscovered
in homostyles (see below), none of the above- mentioned mutations
found in thrums introduced a premature stop codon.
As expected (Huu et al., 2016; Li et al., 2016; Potente, Léveillé-
Bourret, et al., 2022), we detected putatively disruptive CYPT
8 of 19
|
MORA-CARRERA et al.
mutations exclusively in homostyles (Figure 2), confirming previous
results (Mora- Carrera et al., 2021; Figure 2). Specifically, we corrob-
orated a previously reported non- synonymous mutation [Serine to
Stop codon] in exon 2 (referred to as allele CYPT- 2 in Mora- Carrera
et al., 2021) and an 8 bp d eletion in exon 1 that sh ifts the open
reading frame of CYPT (referred to as allele CYPT- 6 in Mora- Carrera
et al., 2021; Figure 1b). Both previously reported mutations in
CYPT- 2 and CYPT- 6 introduce a premature stop codon, likely causing
incomplete CYPTtransl ation .Inth epres entWGRda taset ,CYPT- 2 oc-
curred in t wo homostyles of population EN4- T and all 10 homostyles
of EN5- T, while CYPT- 6 occurred in eight homostyles of EN4- T.
3.1.2 | MutationsinintronicregionsofCYPT
Our anal ysis of the four CYPT introns in 37 thrums and 31 homo-
styles discovered a total of 1131 SNPs along a total intronic length
of66,515 bps.Mostvariantsites[1043SNPs]weredetectedinnon-
English populations (TR, CH, and SK; Figure S2). A total of 24 SNPs
were present in homost yles, of which seven were exclusive to homo-
styles (i.e., absent from thrums; Table S2). However, only two out of
the seven intronic homostyle- specific SNPs segregated in more than
one homos tylous individual. Specif ically, the intronic SNP at position
259,566 was shared by all eight individuals carrying CYPT- 6, whereas
the intronic SNP at position 275,096 was specific to all 11 individuals
carrying CYPT- 1 ( Figure S1). Therefore, no intronic SNP was fixed in
all 31 homostyles.
3.1.3 | StructuralrearrangementsinCYPT
A drop in sequencing coverage compared to the rest of the ge-
nome revealed forthefirsttimealarge deletion(ca.2150 bps) en-
compassing CYPT exon 1 and its upstream and downstream regions
(Figure 3). This 2150 bp s deletion was de tected exclusi vely in the
11 homostyles from population EN6- M (Figures 2 and 3). Additional
analysesofWGRdata inIGVshowed thatasmall number ofreads
within this region had paired mates mapping to different chromo-
somes (not shown), suggesting a local deletion of exon 1 and its po-
tential translocation to another location in the genome. Notably, the
remaining four CYPT exons in these homostyles showed sequencing-
read coverage values comparable to average sequencing- read cover-
age across the genome and did not exhibit any additional disruptive
mutations. Apart from the local deletion of exon 1, we did not iden-
tify additional inversions or translocations involving the remaining
CYPT exons and introns. We designated this previously unreported
allele as CYPT- 8 (Figure 1b).
3.1.4 | Mutationsinputativepromoter
region of CYPT
We identified 20 conserved motifs in the region upstream of CYPT
that are potential binding sites for seven different transcription-
factor families, as summarized in Table 2. Among these mo-
tifs, 10 were associated with various subfamilies of the DREB
(Dehydration-ResponsiveElement-Bindingprotein)group within
the ERF/AP2 transcription- factor family, four were related to the
MYB (Myeloblastosis viral oncogene) transcription factor, two
were associated with the basic leucine- zipper transcription fac-
tor, and one each were linked to the telomere- repeat binding fac-
tor, basic pentacysteine, CHC (Clathrin Heavy Chain) protein, and
a FAR1 (Fatty Acyl- CoA Reductase 1)- related protein. All these
conserved motifs were found within a range of 1361–15 base
pairs upstream of CYPT exon 1. Since CYPT is a brassinosteroid-
degrading enzyme expressed exclusively in the style in P. vulgaris
(Huu et al., 2016), we focused on eight conserved motifs involved
in brassinosteroid regulation and associated with genes expressed
in the carpel of Arabidopsis thaliana,basedonthe TAIRdatabase
(Table 2). Inspection of the aligned 3 kb sequences upstream
of CYPT in 20 out of the 31 homostyles indicated that such se-
quences were identical to those of the 37 thrums with a func-
tional S- locus (Appendix S1). However, in the 11 homostyles with
the2150 bpdeletionaf fecting CYPT exon 1 and surrounding up-
stream and downstream regions (i.e., CYPT- 8 allele: see above), all
eightconservedmotifswereabsentfromthepromoterregion.In
summary, CYPT promoter regions are conserved and likely func-
tional in all analyzed homostyles except for the 11 homostyles
with a deleted CYPT exon 1 (CYPT- 8 allele), which also appear to
lack the promoter region.
Our anal ysis of the putat ive cis-regulat ory eleme nts upstrea m
of CYPT indicates that both P. vulgaris and P. veris have similar cis-
elements (Figure S2), as expected in closely related species. Further
examination of these elements in the 37 thrums and 31 homostyles
of P. vulgaris, and one individual of P. ve ri s indicates that they are
present in the homostyles and that their sequence is identical to that
FIGURE 2 AlignmentofCYPT exon 1 and 2 sequences containing homostyle- specific mutations likely to disrupt CYPT function. No
potential loss- of- function mutations were detected in exons 3, 4, 5, hence sequences of those exons are not shown. Sequences were
extracted from whole genome resequencing (WGR) data of 37 thrums and 31 homostyles from nine populations of Primula vulgaris in
Turkey (TR), Slovakia (SK), Switzerland (CH), and England (EN). (a) Sequence alignment of positions 1 to 20 of CYPTexon1(286 bplong).
EighthomostylesofEN4-Tshowan8 bp-deletiondisruptingCYPT reading frame. No homostyles of EN6- M showed sequencing reads
mapping to exon 1, implying its loss (see also Figure S2). (b) Sequence alignment of positions 364 to 383 of CYPTexon2(227 bplong).
A non- synonymous mutation in position 374 that introduces an early stop codon occurs in 12 homostyles of EN4- T and EN5- T and one
thrum of EN4- T. This latter thrum is heterozygous and carries one functional and one disrupted copy of CYPT. Homostyles are enclosed
in black squares. The mutations shown in (a) and (b) correspond to the CYPT- 6 and CYPT- 2 alleles in Figure 1b, respectively. Population
abbreviations as in Table 1.
|
9 of 19
MORA-CARRER A et al.
10 of 19
|
MORA-CARRERA et al.
of the thrums, implying that the putative cis- regulatory elements
identified here are still functional in the homostyles.
3.2 | Genetic variation in S- genes and their paralogs
Ourresultsshowedthat,asexpectedduetoS- locus hemizygosity, πS
was lower in S-genes ofthrums(0.0 012 ± 0.00 06[mean ± SE])than
in their pa ralogs in pins and t hrums (0.00 34 ± 0.000 8; Table 3A).
Moreover, πS in homostyles was zero for both S- genes and their par-
alogs, except for GLOT and C YP73 4A 51, where πS was extremely low
(0.0012 and 0.0008, respectively; Table 3B), supporting the predic-
tion that the shift to predominant selfing should be associated with
lower πS in homostyles than in heterostyles.
Furthermore, our results indicated that, on average, πN/πS values
were higher for S- genes of thrums than for their paralogs in pins and
thrums(1.01 ± 0.37vs.0.53 ± 0.29,respectively;Table 3A), implying
lower purifying selection in S- genes. Additionally, πN/πS was lower in
KFB1 than in KFBT (πN/πS = 0.23 and1.83, respec tively), but higher
in CCM1 than in CCMT (πN/πS = 1.5and0.91,respectively;Table 3A),
implying stronger and weaker purifying selection on the two par-
alogs than on their respective S- genes, respectively. Finally, we
found that, within the S- locus, πN/πS was higher for CCMT, KFBT, and
PUMT (πN/πS = 0.91,1.83,and10.36,respectively;Table 3B) than for
CYPT (πN/πS = 0.28), whereas πN/πS in GLOT was not calculated due
tothelackof variationat synonymoussitesin this gene. Inhomo-
styles, πN/πS was zero for most S- genes and their paralogs, except
for CY P734A51, due to the lack variation at synonymous and non-
synonymous sites (Table 3B).
3.3 | S- locus genotypes in natural populations of
P. v ulgaris
Relative S- locus sequencing depth (RelS- locus depth) allowed us to de-
termine S-locusploidyintheanalyzedthrumsandhomostyles.Of
the 37 thrums collected from six dimorphic (i.e., pins and thrums)
and two trimorphic (i.e., pins, thrums, and homostyles) populations
across our sampling range, 34 had a haploid S- locus (i.e., S/0) and
three had a diploid S- locus (S/S; Figure 4).Ofthesethreethrums,
FIGURE 3 IntegrativeGenomic
Viewer(IGV)plotshowingsequencing
reads mapping to exon 1 of CYPT and
its flanking regions to the right (intronic
region) and to the left (intergenic region)
in one thrum individual, two homostyles
with different CYPT alleles (see Figure 1),
and the 11 homostylous individuals of the
monomorphic population EN6- M showing
the absence of sequencing reads mapping
to a ~2150 bpdeletionencompassingexon
1(286 bplong)ofCYPT. No other thrum
or homostyle individuals showed this
deletion.
|
11 of 19
MORA-CARRER A et al.
two were homozygous for the functional copy of CYPT (CYPT-
1/CYPT- 1; i.e., S/S) and belonged to one trimorphic and one dimor-
phic population each, respectively (Table 1), while one thrum was
heterozygous and carried one functional and one disrupted copy
of CYPT (CYPT- 1/CYPT- 2; i.e., S/S*).Ofthe31homostyles,10(32%)
had a haploid S- locus (S*/0) and 21 (68%) had a diploid S- locus
(S*/S*) (Figure 4). Specifically, S*/0- homostyles represented 40%
and 60% of homostyles in the trimorphic populations EN4- T and
EN5- T, respectively, while all tested homostyles of the monomor-
phic population EN6- M had the S*/S*- geno typ e (Table 1; purple
triangles in Figure 4).
We also calculated expected and observed frequencies of 0/0,
S/0- , S*/0- , and S*/S*- genotypes under different assumptions for
viability of S*/S*- genotypes compared with the other three geno-
types and after different numbers of generations following the ori-
gin of homostyles (Table 4). The results of chi- squared tests showed
that non- significant differences were found only in seven cases, of
which four occurred in the monomorphic, homostylous population
EN6- M and three in the two trimorphic populations EN5- T and EN4-
T. Specifically, in EN6- M, observed frequencies matched expected
frequencies at generations 30 and 40 under the assumption of equal
and slightly lower viability (v = 1and0.9,respectively)fortheS*/S*-
homost yles. In EN5-T, observed frequencies matched expected
frequencies at 20 generations under lower viability (v = 0.8)ofS*/S*-
homostyles. Finally, in EN4- T, observed frequencies matched ex-
pected genotypic frequencies at generations 20 and 30 under levels
of viability for S*/S*- homostyles close or equal to those of the model
in Figure 1c (i.e., v = 0.70and0.65,respectively).
4 | DISCUSSION
We integrated whole- genome resequencing data from a com-
prehensive sampling of long- monitored, English populations and
from other European populations of P. vulgaris with knowledge
of the recently assembled genomes of P. vulgaris and P. v eris to
explain the causes and consequences of the transition from di-
styly to homostyly. We identified a novel loss- of- function struc-
tural rearrangement in CYPT associated with the transition to
homostyly that had remained undetected using exonic Sanger
sequencing (Mora- Carrera et al., 2021).Importantly,wefoundno
evidence for a potential single origin of homostyly in P. vulgaris
TABLE 2 Summaryofthe20putativetranscriptionfactor(TF)bindingsitesupstreamofCYPT in Primula vulgaris estimated from motif
enrichment analyses (see Section 2).
TAIR ID CODE TF FAMILY
Expressed in
carpel
Expressed with
BRs treatment
Distance from
exon 1 CYPT (in bp) Consensus sequence
AT1G 01250 DREB Yes No 1361 TGTCGGTGRHKDNGD
AT1G19210 DREB Yes Yes 917 DTGGWCG GTGRHG R
AT1G36060 DREB No No 460 DTKGKCGGTGGHGR
AT4G 28140 DREB No Ye s 177 KGTCGGTGGHGRNGD
AT5G65130 DREB No Ye s 177 HYNHHNHYDCCRCCGMCMW
AT3G50260 DREB No No 174 CCDYCDCCACCGMCA
AT5G67190 DREB Yes No 173 YCDYCDCCACCGACA
AT2G23340 DREB Yes Yes 173 HDWTGTCGGTGRHDNND
AT4G06746 DREB No –173 HNNNNNHDYCACCGACAWH
AT1G21910 DREB Yes Yes 171 YCNCCDYCDYCDCCACCGMC
AT5G42 520 BPC YesaNo 171 C TCTCTC TCTCTC TCTCTC TC
AT3G22780 CHC YesaYes 171 WWTTWAAAATTTAAA
AT3G23250 MYB No No 171 YHHHAHHWHHYYCACCAACCH
AT4G 01680 MYB No Yes 168 WGGT WGGTRRRNNDD
AT1G 095 40 MYB No –160 NYY YACCWACCWH
AT1G34670 MYB No No 80 NNWDBYYCACCWAMC
AT5G675 80 TRBF2 Yes Yes 64 WWWWHTWWRCCCTAAWTHH
AT4G 38170 FAR1- related Yes aYes 64 CTCTCTCTCTCTCTCTCTCTC
AT2G40620 bZip Yes Yes 16 NMCAGCTGKCA
AT1G 06850 bZip Yes Yes 16 DTGMCAGCTKGKHW
Note: Table includes “The ArabidopsisInformationResource”(TAIR)identificationcode,theTFfamilyassociatedwiththeconservedmotifs,whether
or not the genes associated with this conserved motif are expressed in the carpel and in the presence of brassinosteroids (BR) in Arabidopsis thaliana
basedontheTAIRdatabase,approximateupstreamdis tancefromthefirstexonofCYPT, and consensus motif sequence of the TF binding site.
Abbreviations:BPC ,basicpent acysteine;bZIP,basicleucinezippertranscriptionfactor;CHC ,Chlathrinheavychain;DREB,dehydration-responsive
element- binding; FAR1, Fatty Acyl- CoA Reductase 1; MYB, myeloblastosis viral oncogene; TRBF2, Telomore Repeat Binding Factor 2.
aHighly overexpressed.
12 of 19
|
MORA-CARRERA et al.
via mutations in the CYPT promoter, cis- regulatory, and intronic
regions, or structural mutations involving CYPT exons, thus the
previously supported hypothesis of multiple transitions to homo-
styly via independent loss- of- function mutations in CYPT exons
stands (Mora- Carrera et al., 2021). Furthermore, population ge-
netic analyses validated theoretical expectations of decreased ge-
netic diversity in S- genes due to hemizygosity. However, contrary
to expectations, purifying selection was not consistently lower in
all S- genes than in their paralogs. Finally, the genomic resources
newly available in the Primula system enabled, for the first time,
the testing of long- standing predictions on changing frequencies
of S- locus genotypes during intraspecific transitions from distyly
to homostyly, partially supporting the possible role of viability
(i.e., seed germination and successful development into seed-
bearing plants) differences between homostyles with haploid vs.
diploid S- locus genotypes in preventing the fixation of homostyly.
Altogether, our study provides a detailed overview of the early
molecular and population genetic causes and consequences of
mating- system transitions.
4.1 | Genetic basis of transitions from distyly to
homostyly in Primula vulgaris
Shifts from outcrossing to selfing are common in flowering plants
and can be caused by loss- of- function mutations in the genes of
interest, structural rearrangements of their exons, or mutations in
their promoters (Shimizu & Tsuchimatsu, 2015).One questioncon-
cerns whether disruptive mutations in the alleles that determine
outcrossi ng act domin antly or recess ively. In Brassic aceae, loss of
self- incompatibility often stems from mutations in dominant al-
leles of genes controlling this trait (Bachmann et al., 2019; Busch
et al., 2011; Nasrallah, 2017; Tsuchimatsu et al., 2012), although in
Arabidopsis lyrata loss of self- incompatibility caused by mutations
in recessive alleles has also been discovered (Mable et al., 2017).
In Primula, the S- locus controlling distyly is hemizygous in both P.
vulgaris and P. ver is (Huu et al., 2016, 2020), but previous models
assumed S- locus heterozygosity, with thrum phenotype associated
with the dominant S- locus allele (Bateson & Gregory, 1905). Based
on this model and greenhouse crossing experiments, Crosby (1949)
Gene nπSπNπN/πS
(A)
CYPT37a0.0026 0.0007 0.28
GLOT37a00.0015 NA
CCMT37a0.0018 0.0016 0.91
KFBT37a0.0004 0.0007 1.83
PUMT37a0.0001 0.0011 10.36
Average (±SE) 0.0010 (±0.0005) 0.0011 (±0.0002) 3.35 (±2.04)
0.0012 (±0.0006)c0.0011 (±0.0003)c1.01 (±0.37)c
CY P7 34A51 74b0.0040 0.0015 0.38
GLO1 74b0.0019 0NA
CCM1 74b0.0019 0.0029 1.50
KFB1 74b0.0056 0.0013 0.23
Average (±SE) 0.0034 (±0.0008) 0.0014 (±0.005) 0.53 (±0.29)
(B)
CYPT31 00.0005 NA
GLOT31 0.0012 0 0
CCMT31 0 0 0
KFBT31 0 0 0
PUMT31 0 0 0
Average (±SE) 0.0002 (±0.00) 0.0001 (±0.00) 0
CY P7 34A51 31 0.0008 0.0002 0.28
GLO1 31 0 0 0
CCM1 31 00.0003 NA
KFB1 31 00.0008 NA
Average (±SE) 0.0002 (±0.00) 0.0003 (±0.0001) 0.14 (±0 .14)
Note: n, sample size.
Abbreviation: NA, not available.
aThrums.
bPins and thrums.
cAverage without PUMT.
TABLE 3 Estimatesofnucleotide
diversity at synonymous (nS), non-
synonymous sites (nN), and the strength
of purifying selection (nN/nS) of all five S-
genes (CYPT, GLOT, CCMT, KFBT, and PUMT)
and their paralogs (CYP75341, GLO1,
CCM1, and KFB1) in (A) thrums (S- locus
genes) and heterostyles (paralogs) and
(B) homostyles of Primula vulgaris.
|
13 of 19
MORA-CARRER A et al.
assumed that S- locus alleles associated with homostyly should be re-
cessive. Previously, we documented seven CYPT haplotypes (CYPT- 2
to CYPT- 7) with putative loss- of- function mutations that occur ex-
clusively in P. vulgaris homostyles (Mora- Carrera et al., 2021). Two
of these haplotypes (CYPT- 2 and CYPT- 6) have an early stop codon
causing premature termination of translation. Our results indicate
that, whether hemizygous or homozygous, these disrupted CYPT
alleles lead to homostyly. However, CYPT- 2 behaves recessively in
the single heterozygous individual carrying one functional and one
disrupted copy of CYPT (i.e., CYPT- 1/CYPT- 2, represented as S/S* in
Figure 4), thus determining a thrum phenotype. This finding aligns
with results of crossing experiments between thrums and homo-
styles of Primula oreodoxa showing that S* is recessive to S when
the two alleles co- occur (Yuan et al., 2018), corroborating Crosby's
prediction (1949). Therefore, our results indicate that the presence
of a disrupted CYPT allele does not alter the thrum morph when
paired with a functional CYPT allele, hence the disrupted allele acts
recessively.
In addition to loss-of-function mutations in coding regions,
mating- system shifts can also stem from transcription silencing or
exonic rearrangements in pertinent genes (Chakraborty et al., 2023).
For instance, down- regulation caused by transposon- like insertions
in the promoter regions of the male- determining self- incompatibility
genes MGST and Bn SP11- 1 trigger the shift from self- incompatibility
to self- compatibility in Prunus avium and Brassica napus, respectively
(Gao et al., 2016; Ono et al., 2018). In Primula vulgaris, previous
Sanger sequencing of individual CYPT exons identified homostyles
with an apparently functional CYPT allele (CYPT- 1 ; Figure 1b), sug-
gesting that homostyles might also arise through CYPT silencing
caused by a disruptive mutation in its promoter or intronic regions
or via exonic rearrangements in CYPT not detectable via Sanger
sequencing (Mora- Carrera et al., 2021). However, 20 of the 31 ho-
mostyles analyzed in the present study did not share any homost yle-
specific SNP in the promoter or intronic region that was absent in
the 37 thrums, implying that the shift to homostyly in these plants
was likely caused by loss- of- function mutations in CYPT exons rather
than in its promoter or intronic region. The remaining 11 homostyles
werecharacterizedbyalarge2150 bpdeletionthateliminatedboth
CYPT exon 1 and its promoter region (Figure 3). Thus, our current
results do not support the conclusion that mutations in the promoter
region or exonic rearrangements in CYPT can alone cause the shift to
homostyly in P. vulgaris.
The evidence above also has implications for determining
whether homostyly arose once or multiple times in P. vulgaris. The
single origin of homostyly, followed by independent mutations in
CYPT, would have been supported if all studied homostyles had
shared the same promoter mutation or rearrangement in CYPT.
However, this is not the case, favoring the hypothesis of multiple or-
igins of homostyly via independent mutations in CYPT exons, as pre-
viously proposed (Mora- Carrera et al., 2021). Nevertheless, a study
of a single homostyle from Chiltern Hills, England (population not
included in our analyses), found reduced expression of CYPT when
compared to a thrum, suggesting that epigenetic silencing might play
a role in the shift to homostyly (Huu et al., 2016). The mentioned
study however did not provide sequences of CYPT exons, thus it re-
mains unknown whether they contained any potentially disruptive
mutations in the coding region. Therefore, transcriptome analyses
of homostylous flowers are necessary to conclusively discard the
possibility that disruptive promoter mutations causing reduced CYPT
expression might also cause the shift to homostyly.
Finally, it remains to be explained why the three homostyles
previously thought to have the functional CYPT- 1 allele based
on Sanger sequencing of the five CYPT exons (Mora- Carrera
et al., 2021) were here found to contain the 2150 bp deletion
including exon 1 (i.e., CYPT - 8 haplotype: see Figures 1b–3). A
possible explanation is that exon 1 was deleted from the S- locus
(causing CYPT loss of function, hence homostyly) and translocated
to a highly repetitive genomic region. The translocation could
have allowed targeted amplification and subsequent Sanger se-
quencing using exon- 1- specific PCR primers, while preventing
exon- 1 detection via next generation sequencing due to biases
arising, for example, during genomic DNA sonication used to
produce shortDNAfragmentspriortoshort-read librar yprepa-
ration (Garafutdinov et al., 2016; Jennings et al., 2017; Poptsova
et al., 2014). Notably, a few low- quality sequencing reads did map
to CYPT exon 1, suggesting this exon is indeed present in the ge-
nome of these homostyles but was not successfully sequenced
using short- read sequencing methodology. Long- read sequencing
FIGURE 4 Numberofindividualswitheitherhaploid(S/0 = 34;
S*/0 = 10)ordiploid(S/S = 2;S/S* = 1;S*/S* = 21)S- locus, as inferred
from the relative depth of sequencing reads mapping to the S- locus
(see Section 2). Thrums are represented as circles (n = 37)and
homostyles as triangles (n = 31).ThesamecolorsusedinFigure 1
were also used here to represent the different genotypes of
thefloralmorphs:Lightgreen = S/0;mediumgreen = S/S*; dark
green = S/S;turquoise = S*/0;purple = S*/S*. A total of 68 individuals
were sampled from dimorphic (i.e., only pins and thrums),
trimorphic (i.e., pins, thrums, and homostyles) and monomorphic
(i.e., only homostyles) populations of P. vulgaris in TR, SK, CH, and
EN (Table 1).
14 of 19
|
MORA-CARRERA et al.
TABLE 4 ComparisonofexpectedfrequenciesandobservedfrequenciesofS- locus genotypes (0/0- , S/0- , S*/0- , and S*/ S*) in two
trimorphic (EN4- T and EN5- T) and one monomorphic populations (EN6- M).
(A) Observed frequencies (0/0:S/0:S*/0:S*/S*)
Generation
v = 1
EN4- T E N 5 - T E N 6 - M
Expec ted freq
(0/0:S/0:S*/0:S*/S*)
10 0.50:0.42:0.05:0.03 0.36:0.16:0.29:0.19** 0.29:0.09:0.25:0.36** 0:0:0:1**
20 0.17:0.04:0.29:0.50 0.36:0.16:0.29:0.19** 0.29:0.09:0.25:0.36** 0:0:0:1**
30 0.02:0.00:0.06:0.92 0.36:0.16:0.29:0.19** 0.29:0.09:0.25:0.36** 0:0:0:1NS
40 0.00:0.00:0.01:0.99 0.36:0.16:0.29:0.19** 0.29:0.09:0.25:0.36** 0:0:0:1NS
(B) Observed frequencies (0/0:S/0:S*/0:S*/S*)
Generation
v = 0.9
EN4- T E N 5 - T E N 6 - M
expected freq
(0/0:S/0:S*/0:S*/S*)
10 0.46:0.34:0.14:0.07 0.36:0.16:0.29:0.19** 0.29:0.09:0.25:0.36** 0:0:0:1**
20 0.15:0.02:0.29:0.54 0.36:0.16:0.29:0.19** 0.29:0.09:0.25:0.36** 0:0:0:1**
30 0.04:0.00:0.11:0.85 0.36:0.16:0.29:0.19** 0.29:0.09:0.25:0.36** 0:0:0:1NS
40 0.01:0.00:0.04:0.95 0.36:0.16:0.29:0.19** 0.29:0.09:0.25:0.36** 0:0:0:1NS
(C) Observed frequencies (0/0:S/0:S*/0:S*/S*)
Generation
v = 0.8
EN4- T E N 5 - T E N 6 - M
Expec ted freq
(0/0:S/0:S*/0:S*/S*)
10 0.48:0.48:0.10:0.04 0.36:0.16:0.29:0.19** 0.29:0.09:0.25:0.36** 0:0:0:1**
20 0.25:0.05:0.37:0.33 0.36:0.16:0.29:0.19** 0.29:0.09:0.25:0.36NS 0:0:0:1**
30 0.12:0.00:0.27:0.62 0.36:0.16:0.29:0.19** 0.29:0.09:0.25:0.36** 0:0:0:1**
40 0.08:0.00:0.19:0.73 0.36:0.16:0.29:0.19** 0.29:0.09:0.25:0.36** 0:0:0:1**
(D) Observed frequencies (0/0:S/0:S*/0:S*/S*)
Generation
v = 0.7
EN4- T E N 5 - T E N 6 - M
Expec ted freq (0/0:S
/0:S*/0:S*/S*)
10 0.50:0.40:0.07:0.03 0.36:0.16:0.29:0.19** 0.29:0.09:0.25:0.36** 0:0:0:1**
20 0.35:0.13:0.35:0.17 0.36:0.16:0.29:0.19NS 0.29:0.09:0.25:0.36** 0:0:0:1**
30 0.21:0.01:0.40:0.38 0.36:0.16:0.29:0.19** 0.29:0.09:0.25:0.36** 0:0:0:1**
40 0.17:0.00:0.37:0.46 0.36:0.16:0.29:0.19** 0.29:0.09:0.25:0.36** 0:0:0:1**
(E) Observed frequencies (0/0:S/0:S*/0:S*/S*)
Generation
v = 0.65
EN4- T E N 5 - T E N 6 - M
Expec ted freq (0/0:S
/0:S*/0:S*/S*)
10 0.52:0.46:0.01:0.01 0.36:0.16:0.29:0.19** 0.29:0.09:0.25:0.36** 0:0:0:1**
20 0.48:0.37:0.12:0.03 0.36:0.16:0.29:0.19** 0.29:0.09:0.25:0.36** 0:0:0:1**
30 0.34:0.01:0.38:0.18 0.36:0.16:0.29:0.19NS 0.29:0.09:0.25:0.36** 0:0:0:1**
40 0.24:0.01:0.43:0.32 0.36:0.16:0.29:0.19** 0.29:0.09:0.25:0.36** 0:0:0:1**
Note: Expected frequencies of S- locus genotypes were estimated at 10, 20, 30, and 40 generations after the onset of homostyly with different
levels of viability (v) for S*/S*- homostyles relative to S*/0- homostyles, S/0- thrums, and 0/0- pins: (A) v = 1(seeFigure 1d), (B) v = 0.9,(C)v = 0.8,(D)
v = 0.7,and(E)v = 0.65(seeFigure 1c). Significance was estimated using chi- square tests with Bonferroni corrections. Non- significant differences
between expected frequencies and observed frequencies of S- locus genotypes (boldfaced) indicate the conditions under which observed genotypic
frequencies are compatible with Crosby's model predictions (Crosby, 1949).
** = p < .01;NS, non- significant differences.
|
15 of 19
MORA-CARRER A et al.
technologies capable of sequencing through repetitive regions
would be necessary to definitively resolve whether CYPT exon 1
was translocated to a highly repetitive genomic region in these ho-
mostyles. To summarize, our findings indicate that the homostyles
previously identified as having a functional CYPT allele in fact do
possess a disrupted CYPT allele due to exon 1 deletion from the
S- locus (designated CYPT- 8 allele: Figure 3).Overall,theseresults
emphasize that not only non- synonymous mutations or small de-
letions, but also structural rearrangements such as large deletions
and translocations can cause mating- system transitions.
4.2 | Population genetic consequences of
hemizygosity and transition to homosyly on S- genes
One of the mo st notable, re cent discoveries on th e S- locus is that
it is hemizygous and present only in thrums in all systems where its
genetic architecture has been investigated, including Primula (Huu
et al., 2016; Li et al., 2016), Turnera (Shore et al., 2019), Fagopyrum
(Fawcett et al., 2023; Matsui & Yasui, 2020), Linum (Gutiérrez- Valencia
et al., 2022), and Gelsemium (Zhao et al., 2023), representing differ-
ent families and orders of flowering plants. The hemizygosity of the
S- locus should affect patterns of molecular diversity. Specifically, tight
genetic linkage provided by recombination suppression in S- genes and
the fact that the S- locus is present only in thrum individuals are ex-
pected to cause a reduction of Ne and consequently a decrease of ge-
netic diversity inside the S- locus compared to other genomic regions
(Gutiérrez- Valencia et al., 2021). Our results demonstrate that the
mean πS of S- genes (CCMT, GLOT, CYPT, and KFBT; πS:0.0012 ± 0.0006)
is lower than that of their paralogs (CCM1, GLO1, CY P73 4A 51, and
KFB1; πS:0.0034 ± 0.0007)locatedelsewhereinthegenome(Table 3).
This result corroborates previous studies that found an overall de-
crease in genetic diversity between the S- locus and its flanking regions
in Primula (Potente, Léveillé-Bourret, et al., 2022), Linum (Gutiérrez-
Valencia et al., 2022), and Turnera (Henning et al., 2023). Thus, our
work confirms predictions that S- locus genomic architecture influ-
ences patterns of molecular evolution in S- genes.
The hemizygous, non- recombining nature of the S- locus also
affects its response to natural selection when compared to recom-
bining regions. Specifically, the increased linkage disequilibrium
caused by the absence of recombination should lead to a reduction
of Ne within the S- locus. This reduction of Ne should, in turn, lead to
less efficient purifying selection (Gossmann et al., 2011) on S- genes
compared to their paralogs outside the S- locus. Consequently, in-
creased degeneration due to accumulation of deleterious mutations
is expected in these genes (Charlesworth & Charlesworth, 2000;
Huu et al., 2016). Conversely, if selection to maintain function were
strong, purifying selection should be more efficient on S- genes than
on their paralogs due to the dominant nature of the hemizygous
S- locus (Gutiérrez- Valencia et al., 2021; Potente, Léveillé-Bourret,
et al., 2022). Rega rding the forme r hypothesis, a gr eater accumulati on
of transposable elements in S- locus non- coding regions compared
to the rest of the genome was detected, supporting the conclusion
that purifying selection on the S- locus might be relaxed (Potente,
Léveillé-Bourret, et al., 2022). However, whether the efficacy of
purifying selection differs between coding regions of S- genes and
their paralogs remains poorly understood (Potente, Léveillé-Bourret,
et al., 2022). Our results indicatethat,onaverage,S- genes exhibit
higher accumulation of non- synonymous mutations than their paral-
ogs, implying purifying selection is less effective on the former (πN/
πS = 1.01 ± 0.37 and 0.53 ± 0.25, respectively; Table 3), conformant
with predicted effects of reduced S- locus Ne. However, patterns
of selective constraints within and outside the S- locus vary among
gene duplicates. For example, the strength of purifying selection
is similar between CYPT and C YP734 A51, albeit slightly stronger in
the former (πN/πS = 0.28 and 0.38 , respective ly). Conversely, puri-
fying selection is less efficient in the S- locus for KFB (πN/πS = 1.83
[KFBT] and 0.23 [K FB1]), whereas CCM shows the opposite pattern
(πN/πS = 0.91[CCMT] and 1.50 [CCM1]; Table 3). Taken together, the
results imply that the effects of hemizygosity on purifying selection
vary among P. vulgaris S- genes, corroborating previous results in P.
veris (Potente, Léveillé-Bourret, et al., 2022).
A key question for the genetics of distyly concerns whether the
strength and nature of selection differ among the genes within S-
locus. Among the three, nine, and five protein- coding genes iden-
tified in the S- locus of Gelsemium, Linum, and Primula, respectively,
(Gutiérrez- Valencia et al., 2022; Li et al., 2016; Potente, Léveillé-
Bourret, et al., 2022; Zhao et al., 2023) only two, namely CYPT and
GLOT of Primula, have been functionally characterized, showing that
CYPT determines short styles and female self- incompatibility (Huu
et al., 2016, 2022), while GLOT determines high anthers in thrums
(Huu et al., 2020). However, it remains unclear whether CCMT, KFBT,
and PUMT play a role in Primula distyly. The markedly reduced and
non- floral specific expression of CCMT, KFBT, and PUMT compared
to CYPT and GLOT in both P. vulgaris and P. ve ris (Cocker et al., 2018;
Potente, Stubbs, et al., 2022) cast doubt on whether the former
threegenesare essentialfordistyly.InthedistylousGelsemium el-
egans (Gentianales), the homolog of Primula CCMT was absent from
the genome, while homologs of KFBT and PUMT were present but
did not localize to the putative S- locus and were expressed in both
pin and thrum flowers (Zhao et al., 2023). Taken together, previous
evidence suggests that CCMT, KFBT, and PUMT may not be essential
for the core traits of distyly (i.e., reciprocal placement of sexual or-
gans and self- incompatibility); hence, they might be under relaxed
purifyingselection.If this istrue,onemightexpectthrumstoex-
hibit higher accumulation of non- synonymous mutations in CCMT,
KFBT, and PUMT than in CYPT and GLOT.Ourr esult ss upportt hi spre-
diction, for we found that, within the S- locus, CCMT, KFBT, and PUMT
(πN/πS = 0.91,1.83,and10.36,respectively)comparedtoCYPT (πN/
πS = 0.28; Table 3A).Itis unlikely that theresults areexplained by
positive directional selection on advantageous non- synonymous
mutations of the three genes above in thrums, because positive
selection should cause rapid fixation of advantageous mutations,
hence the absence of polymorphism at non- synonymous sites
(Hahn, 2018), which is not what we found (Table 3A). To summa-
rize, in P. vulgaris purifying selection seems stronger on the only
16 of 19
|
MORA-CARRERA et al.
two S- genes for which a key function in distyly has been demon-
strated (namely, CYPT and GLOT) than on CCMT, KFBT, and PUMT.
Discoveringwhetherthelaterthreegenesabovemayplayarolein
controlling ancillary traits of distyly (e.g., pollen size and number,
and male incompatibility) requires additional functional studies in
Primula and other distylous taxa.
Comprehensive population genetic analyses of variability in S-
genes and their paralogs had never been performed until now, due
to missing knowledge of relevant genes, unavailability of sequences
from said genes, and inadequate population sampling across ex-
tensive geographic ranges. Here, we expanded on previous Sanger
sequencing analyses of CYPT in Somerset (England) populations
(Mora- Carrera et al., 2021) by analyzing also sequences of S- genes
and their paralogs extracted from WGR data of Slovakian, Swiss,
and Turkish populations of P. vulgaris. First, homostyles, found ex-
clusively in three Somerset populations, exhibited lower genetic di-
versity than thrums for both S- genes and their paralogs (Table 3),
corroborating previous reports of reduced genetic diversity in homo-
styles (Husband & Barrett, 1993; Ness et al., 2010; Yuan et al., 2017;
Zhong et al., 2019; Zhou et al., 2017). Second, both S- genes and their
paralogs have markedly lower genetic variation in English popula-
tions than in other Eurasian populations of P. vulgaris (Table S3). This
finding suggests a recent genetic bottleneck in English populations.
This bottleneck could be associated with colonization of England fol-
lowing glacial retreat during the Last Glacial Maximum (ca. 10,000–
12,00 0 yearsago),assuggestedforotherplantspecies(Birks,1989).
Future genomic and demographic investigations will determine
whether the signatures of genetic bottlenecks detected in S- genes
and their paralogs apply to the entire genome, thus helping to infer
the timing and mode of P. vulgariscolonizationoftheBritishIsles.
4.3 | Does lower viability of S*/S*- homostyles
prevent the fixation of homostyly in P. vu lg ar is?
Theoretical and experimental work suggests that, in the absence
of inbreeding depression and assuming that all individuals produce
equal num ber of seeds, once s elfing originate s, the selfing ph enotype
should increase in frequency and eventually become fixed over time
(Charlesworth et al., 1990; Fisher, 1941; Lande & Schemske, 1985).
In the transition from distyly to homostyly, Crosby's model
(Crosby, 1949) predicted that the rate of increase and ultimate fixa-
tion of homostyles in a population depends on whether homostyles
with diploid S- locus have lower or equal viability as the other geno-
types in the population (Figure 1c,d). The assumption of lower viabil-
ity for S*/S*- homostyles of P. vulgaris expanded upon prior evidence
from crossing experiments in P. sinensis suggesting that homozygous
dominant thrums had 30% lower viability than heterozygous thrums
(Mather & Winton, 1941). More recently, results of crossing experi-
ments in a Primula hybrid (Primula × tommasinii) were interpreted
as evidence of inviability for S/S- thrums (Kurian & Richards, 1997).
Furthermore, population surveys of pin- to- thrum ratios in P. o re o-
doxa indicated that thrums were overrepresented at the seed (~1:3)
but not adult stage (~1:1), implying that differences in viability could
occur during the life cycle (Yuan et al., 2018). However, genotyping
of thrums was not carried out, thus preventing the determination
of whether the decrease of thrums from seed to adult stage was
caused by lower viability of S/S-thrums.Ourobserved frequencies
of S*/0- and S*/S*- homostyles from the two trimorphic (i.e., pins,
thrums, and homostyles), English populations EN4- T and EN5- T of P.
vulgaris are consistent with Crosby's prediction of a recent transition
to homostyly (20–30 generations) under 30%–40% lower viability
of S*/S*- homostyles (Table 4), supporting the model that assumes
lower fitness for S*/S*- homostyles than S*/0- homostyles (Figure 1c).
Conversely, the occurrence of a monomorphic, homostylous
population of P. vulgaris in England, first reported by Curtis and
Curtis (1985) 38 years ago and recently sampled by Mora Carrera
et al. (2021 and present study) is congruent with the assumption of
equal viability for S*/S* homostyles. All 11 genotyped homostyles in
this popu lation (here name d EN6- M) c arry the S*/S*- gen ot y pe ( Table 1
and Figure 3); thus, EN6- M could represent a case in which homostyly
increased in frequency over time and became fixed in the population
by displacing pins and thrums, as predicted under the assumption of
equal viability for S*/S* homostyles (Figure 1d). Alternatively, EN6- M
could have been established by an S*/S*- homostyle stemming from a
nearby population, thus it might have been a monomorphic homosty-
louspopulation fromthebeginning.Indeed,CurtisandCurtis(1985)
reported that this monomorphic population was located only about
240 m away from a tr imorphic p opulatio n which might have s erved
as a source for the initial homostyle that gave origin to EN6- M.
Finally,EN6-M hadavery lowpopulationsize(n = 19;Mora-Carrera
et al., 2021), suggesting that stochasticity could have played a role in
the fixation of S*/S*- homostyles in this population and that homozy-
gosity of an S- locus with disrupted CYPT might have detrimental ef-
fects on population growth.
To summarize, our results suggest that a diploid S- locus with
inactivated CYPT* may not per se be incompatible with homostyle
viability. However, the occurrence of two copies of the remain-
ing S- genes (i.e., CCMT, GLOT, PUMT, and KFBT) in the genome of a
homostyle could have detrimental effects on viability at different
stages of the life cycle, possibly stemming from gene- dosage ef-
fects (Ascencio et al., 2021; Li et al., 2015; Rice & McLysaght, 2017;
Tasdighian et al., 2017). Future research combining S- locus genotyp-
ing and characterization of function and dosage effects of S- genes
at different life- cycle stages with fitness measurements in the field
and in greenhouse experiments is essential to address whether dif-
ferences in viability prevent the widespread fixation of homostyly
in P. vulgaris.
AUTHOR CONTRIBUTIONS
E. Mora- Carrera: Conceptualization (equal); data curation (equal);
formal analysis (equal); funding acquisition (equal); investigation
(equal); methodology (equal); visualization (equal); writing – origi-
nal draft (equal); writing – review and editing (equal). R. L. Stubbs:
Data curation(equal); formal analysis(equal);investigation(equal);
writing – review and editing (equal). G. Potente: Formal analysis
|
17 of 19
MORA-CARRER A et al.
(supporting); methodology (supporting); writing – review and edit-
ing (equal). N . Yousefi: Formal analysis (supporting); methodology
(supporting); resources (lead); software (equal); writing – review and
editing (equal). B. Keller:Investigation(equal);methodology(equal);
writing – r eview and editin g (equal). J . M. de Vo s: Investigation(equal);
supervision (equal); writing – review and editing (equal). P. Szövényi:
Methodology (supporting); software (equal); supervision (equal);
writing – review and editing (equal). E. Conti: Conceptualization
(equal); funding acquisition (lead); project administration (lead); su-
pervision (lead); writing – original draft (equal); writing – review and
editing (equal).
ACKNOWLEDGMENTS
We thank Natural England for permits and landowners for access
to populations; Ferhat Celep and Judita Kochjarová for assistance
during fieldwork collection. We thank the editor and two anony-
mous reviewers for their comments that improved this manu-
script. This research was supported by the Graduate Campus office
at the University of Zurich through a GRC- Travel Grant and the
Forschungskredit Candoc (grant no. K- 74202- 05- 01 to EM- C), by
the Swiss Government Excellence Scholarship (grant no. 2018.0475
to EM- C), and by the Swiss National Science Foundation (grant
no. 310030_185251 to JdV and grant nos. 3100- 061674.00/1 and
175556 to EC).
CONFLICT OF INTEREST STATEMENT
The authors declare no competing interests.
DATA AVAIL ABILI TY STATEMENT
Original sequencing reads have been uploaded to the NCBI re-
pository (https:// www. ncbi. nlm. nih. gov/ ) under the BioP roject ID:
PRJNA1066534. Sequence alignment of all exons of CYPT and align-
ment of CYPT promoter region, BAM files of S- locus of the individu-
als used detect structural rearrangements has been uploaded to
Dryad(https:// doi. org/ 10. 5061/ dryad. prr4x gxtf).
ORCID
E. Mora- Carrera https://orcid.org/0000-0001-8237-4265
R. L. Stubbs https://orcid.org/0000-0001-7386-2830
G. Potente https://orcid.org/0000-0002-4343-3952
N. Yousefi https://orcid.org/0000-0001-6292-1516
B. Keller https://orcid.org/0000-0002-7903-8938
J. M. de Vos https://orcid.org/0000-0001-6428-7774
P. Szövényi https://orcid.org/0000-0002-0324-4639
E. Conti https://orcid.org/0000-0003-1880-2071
REFERENCES
Ascencio,D., Diss,G.,Gagnon-Arsenault,I., Dubé, A.K.,DeLuna,A.,&
Landr y, C. R. (2021). Expression attenuation as a mechanism of
robustness against gene duplication. Proceedings of the National
Academy of Sciences of the United States of America, 118, 1–10 .
Bachmann,J.A.,Tedder, A., Laenen,B.,Fracassetti,M.,Désamoré, A.,
Lafon- Placette, C., Steige, K. A., Callot, C., Marande, W., Neuffer,
B., Bergès, H., Köhler, C., Castric, V., & Slotte, T. (2019). Genetic
basis and timing of a major mating system shift in Capsella. New
Phytologist, 224, 505–517.
Barmentlo, S. H., Meirmans, P. G., Luijten, S. H., Triest, L., &
Oos termeijer,J.G .B.(2017).Out breedingdep ress iona ndbr eed-
ing system evolution in small, remnant populations of Primula
vulgaris: Consequences for genetic rescue. Conservation Genetics,
19, 545–554.
Barrett, S. C. H. (2019). ‘A most complex marriage arrangement’: Recent
advanc es on heterost yly and unres olved questi ons. New Phytologist,
224, 1051–1067.
Bateson ,W.,&G rego ry,R .P.(1905).Ontheinh eritanc eofhe tero sty lism
in Primula. Proceedings of the Royal Society of London. Series B, 76 ,
581–586.
Birks, H. J. B. (1989). Holocene isochrone maps and patterns of tree-
spreadingintheBritishIsles.Journal of Biogeography, 16, 503–540.
Bolger, A. M., Lohse, M., & Usadel, B. (2014). Trimmomatic: A flexible
trimmerforIlluminasequencedata.Bioinformatics, 30, 2114–2120.
Busch, J . W.,J oly, S., & Schoen , D. J. (2011). Demogra phic signatu res
accompanying the evolution of selfing in Leavenworthia alabamica.
Molecular Biology and Evolution, 28, 1717–1729.
Camacho, C., Coulouris, G., Avagyan, V., Ma, N. , Papadopoulos, J., Bealer,
K., & Madden, T. L. (2009). BLAST+: Architecture and applications.
BMC Bioinformatics, 10, 1–9.
Chakraborty,S.,Dut ta,S.,&Das,M.(2023).Geneticsbehindsexualin-
compatibility in plants: How much we know and what more to un-
cover? Journal of Plant Growth Regulation, 42, 7164–7188.
Charlesworth, B ., & Charlesworth, D. (1979). The maintenance and
breakdown of distyly. The American Naturalist, 114, 499–513.
Charlesworth, B ., & Charleswor th, D. (2000). The degeneration of Y
chromosomes. Philosophical Transac tions of the Royal Society B:
Biological Sciences, 355, 1563–1572.
Charlesworth,D.(2022).Primrosehomostyles:Aclassiccaseofpossible
balancing selection revisited. Molecular Ecology, 32, 1–3 .
Charlesworth, D.,Mor gan,M.T.,&Charleswor th,B.(1990).Inbreeding
depression, genetic load, and the evolution of outcrossing rates in a
multilocus system with no linkage. Evolution, 44, 1469–1489.
Cocker,J.M.,Wright,J., Li, J.,Swarbreck,D., Dyer,S.,Caccamo, M.,&
Gilmartin, P. M. (2018). Primula vulgaris (primrose) genome assem-
bly, annotation and gene expression, with comparative genomics
on the heterostyly supergene. Scientific Reports, 8, 1–13 .
Costa, J., Torices, R., & Barrett, S. C. H. (2019). Evolutionary history
of the buildup and breakdown of the heterostylous syndrome in
Plumbaginaceae. New Phytologist, 224, 1278–1289.
Crosby, J. L. (1940). High proportions of homostyle plants in populations
of Primula vulgaris. Nature, 145, 672–673.
Crosby, J. L. (1949). Selection of an unfavourable gene- complex.
Evolution, 3, 212–230.
Curtis, J., & Curtis, C. F. (1985). Homostyle primroses re-visited. I.
Variation in time and space. Heredity, 54, 22 7–234 .
Cutter, A. D. (2019). Reproductive transitions in plant s and animals:
Selfing syndrome, sexual selection and speciation. New Phytologist,
224, 10 80–1094.
de Jong, T. J., & Klinkhammer, P. G. L. (2005). Evolutionary ecol ogy of plant
reproductive strategies. Cambridge University Press.
de Vos, J. M., Hughes, C. E., Schneeweiss, G. M., Moore, B. R., & Conti,
E. (2014). Heterostyly accelerates diversification via reduced ex-
tinct ion in Dar win's primro ses. Proceedings of the Royal Society B:
Biological Sciences, 281, 20140075.
Fawcett, J. A ., Takeshima, R., Kikuchi, S., Yazaki, E., Katsube- Tanaka,
T.,Dong,Y.,Li, M.,Hunt,H.V.,Jones,M. K.,Lister,D.L.,Ohsako,
T., Ogiso-Tanaka, E., Fujii, K., Hara, T., Matsui, K ., Mizuno, N.,
Nishimura, K., Nakazaki, T., Saito, H., … Yasui, Y. (2023). Genome
sequencing reveals the genetic architecture of heterostyly and
domestication history of common buckwheat. Nature Plants, 9,
1236–1251.
18 of 19
|
MORA-CARRERA et al.
Fisher, R. (1941). Average excess and average effect of a gene substitu-
tion. Annals of Eugenics, 11, 53–63.
Ganders, F. R. (1979). The biology of heterostyly. New Zealand Journal of
Botany, 17, 607–6 35.
Gao, C., Zhou, G., Ma, C., Zhai, W., Zhang, T., Liu, Z., Yang, Y., Wu, M.,
Yue,Y.,D uan,Z.,Li, Y.,Li,B.,Li,J.,S hen,J.,Tu,J., &Fu, T.(2016).
Helitron- like transposons contributed to the mating system transi-
tion from out- crossing to self- fertilizing in polyploid Brassica napus
L. Scientific Reports, 6, 1–10.
Garafutdinov, R. R., Galimova, A. A., Sakhabutdinova, A. R., &
Chemeris, A. V. (2016). PCR- based evaluation of sequence speci-
ficityofDNAfragment ationbyultrasound.Molecular Biology, 50,
236–241.
Goldberg,E.E.,&Igic,B.(2012).Tempoandmodeinplantbreedingsys-
tem evolution. Evolution, 66, 3701–3709.
Goldberg, E. E., Kohn, J. R., Lande, R., Robertson, K. A., Smith, S. A., &
Igić, B. (2010). Species selection maintains self-incompatibility.
Science, 330, 493–495.
Gossmann,T.I .,Woolf it,M.,& Eyre-Walker,A.(2011).Quantifying the
variation in the effective population size within a genome. Genetics,
189, 138 9–14 02.
Gutiérrez-Valencia,J.,Fracassetti,M.,Berdan,E .L.,Bunikis, I.,Soler,L.,
Dainat,J.,Kutschera ,V.E.,Losvik,A.,Désamoré,A.,Hughes,P.W.,
Foroozani, A., Laenen, B., Pesquet, E., Abdelaziz, M., Pettersson,
O. V.,N ystedt, B ., Brenna n, A. C., A rroyo, J., & Slo tte, T. (2022).
Genomi c analyses of the Linum dis tyly super gene reveal converge nt
evolutio n at the molecular l evel. Current Biology, 32, 4360–4371.e6.
Gutiérrez- Valencia, J., Hughes, P. W., Berdan, E. L., & Slotte, T. (2021).
The genomic architecture and evolutionary fates of supergenes.
Genome Biology and Evolution, 13, 1–19.
Hahn, M. W. (2018). Molecular population genetics. Sinauer Associates.
isbn:978- 0878939657.
Hennin g, P. M ., Roalson, E. H ., Mir, W., McCubbin, A . G., & Shore , J. S. (2023).
Annotation of the Turnera subulata (Passifloraceae) draft genome re-
veals the S- locus evolved after the divergence of Turneroideae from
Passifloroideae in a stepwise manner. Plants, 12, 286.
Husband, B. C., & Barrett, S. C. H. (1993). Multiple origins of self-
fertilization in tristylous Eichhornia paniculata (Pontederiaceae):
Inferences from style morph and isozyme variation. Journal of
Evolutionary Biology, 6, 591–608.
Huu, C. N., Kappel, C., Keller, B., Sicard, A., Takebayashi, Y., Breuninger,
H.,Nowak,M.D.,Bäurle, I.,Himmelbach,A., Burkar t,M.,Ebbing-
Lohaus, T., Sakakibara, H., Altschmied, L., Conti, E., & Lenhard, M.
(2016). Presen ce versus absence of C YP734A 50 under lies the style-
length dimorphism in primroses. eLife, 5, 1–15 .
Huu, C . N., Keller, B., Con ti, E., Kap pel, C., & Le nhard, M. (2 020). Super gene
evolution via stepwise duplications and neofunctionalization of a
floral- organ identity gene. Proceedings of the National Academy of
Sciences of the United States of America, 117, 23148–23157.
Huu, C. N., Plaschil, S., Himmelbach, A., Kappel, C., & Lenhard, M. (2022).
Female self- incompatibility type in heterostylous Primula is de-
termined by the brassinosteroid- inactivating cytochrome P450
CYP734A50. Current Biology, 32, 671–676.
Jacquemyn, H., Endels, P., Brys, R., Hermy, M., & Woodell, S. R. J. (2009).
BiologicalfloraoftheBritishIsles:Primula vulgaris Huds. (P. acaulis
(L.) Hill). Journal of Ecology, 97, 812–833.
Jenning s, L. J., Arcil a, M. E., Corles s, C., Kamel-Reid , S., Lubin, I. M. ,
Pfeifer, J., Temple- Smolkin, R. L., Voelkerding, K. V., & Nikiforova,
M. N. (2017). Guidelines for validation of next- generation sequenc-
ing–based oncology panels: A joint consensus recommendation of
the Association for Molecular Pathology and College of American
Pathologists. Journal of Molecular Diagnostics, 19, 3 41–365.
Keller,B.,Thomson,J.D.,&Conti,E. (2014).Heterostylypromotesdis-
assortativepollinationandreducessexualinterferenceinDarwin's
primroses: Evidence from experimental studies. Functional Ecology,
28, 1413–1425 .
Kissling, J., & Barrett, S. C. H. H. (2013). Variation and evolution of her-
kogamy in Exochaenium(Gentianaceae):Implicationsfortheevolu-
tion of distyly. Annals of Botany, 112, 95–102.
Korunes, K. L., & Samuk, K. (2021). pixy: Unbiased estimation of nu-
cleotide diversity and divergence in the presence of missing data.
Molecular Ecology Resources, 21, 1359–1368.
Kumar,S .,Stecher,G., Li, M., Knyaz, C., &Tamura,K. (2018). MEGAX :
Molecular evolutionary genetics analysis across computing plat-
forms. Molecular Biology and Evolution, 35, 1547–1549.
Kurian, V., & Richards, A. J. (1997). A new recombinant in the heteromor-
phy“S”supergeneinPrimula. Heredity, 78, 383–390.
Lande,R.,& Schemske,D. W.(1985).Theevolution ofself-fertilization
and inbreedingdepressionin plants. I. Genetic models. Evolution,
39, 24–40.
Lescot,M., Déhais, P.,Thijs, G., Marchal, K., Moreau, Y.,VandePeer,
Y., Rouzé, P., & Rombauts, S. (2002). PlantCARE, a database of
plant cis- acting regulatory elements and a portal to tools for in
silico analysis of promoter sequences. Nucleic Acids Research, 30,
325–327.
Lewis, D., & Jones, D. A. (1992). The genetics of heterostyly. In S.
Barrett (Ed.), The evolution and function of heterostyly (pp. 129–151).
Springer- Verlag.
Li,H.,&Durbin,R .(2009).Fastandaccurateshortreadalignmentwith
Burrows- Wheeler transform. Bioinformatics, 25, 1754–1760.
Li, J.,Cocker,J. M., Wright, J., Webster,M.A .,McMullan, M., Dyer,S.,
Swarbre ck, D., Cacc amo, M., Van Oo sterhou t, C., & Gil martin , P.
M. (2016). Genetic architecture and evolution of the S- locus super-
gene in Primula vulgaris. Nature Plants, 2, 1–7.
Li,Z.,Defoor t,J.,Tasdighian,S.,Maere,S.,VanDePeer,Y.,&DeSmet,R.
(2015). Gene duplicability of core genes is highly consistent across
all angiosperms. Plant Cell, 28, 326–344.
Lloyd, D. G. , & Webb, C. J. (1992). The sel ection of het erostyl y.In S .
Barrett (Ed.), The evolution and function of heterostyly (pp. 179–207).
Springer- Verlag Berlin.
Mable, B. K., Hagmann, J., Kim, S. T., Adam, A., Kilbride, E., Weigel,
D.,&Stift,M .(2017).Whatcausesmating systemshifts inplant s?
Arabidopsis lyrata as a case study. Heredity, 118, 52–63.
Mast, A . R., Kelso, S., & Conti, E. (2006). Are any primroses (Primula)
primitively monomorphic? New Phytologist, 171, 605–616.
Mather,K.,&Winton,D.D.E.(1941).Adaptationandcounter-adaptation
of the breeding system in Primula. Annals of Botany, 5, 297–311.
Matsui, K., & Yasui, Y. (2020). Genetic and genomic research for the
development of an efficient breeding system in heterostylous
self- incompatible common buckwheat (Fagopyrum esculentum).
Theoretical and Applied Genetics, 133, 1641–1653.
McKenna, A ., Hanna, M., Banks, E., Sivachenko, A., Cibulskis, K.,
Kernyt sky,A., Garimella, K., Altshuler,D., Gabriel, S., Daly, M., &
DePristo,M.A .(2010).TheGenomeAnalysisToolkit:AMapReduce
framework for analyzing next-generation DNA sequencing data.
Genome Research, 20, 1 29 7–1303 .
McLeay, R. C ., & Bailey, T. L. (2010). Motif enr ichment analysi s: A unified
framewo rk and an eva luation o n ChIP data . BMC Bioinformatics,
11, 1–11 .
Mora- Carrera, E., Stubbs, R. L., Keller, B., Léveillé- Bourret, É., de Vos, J.
M.,Szö vényi, P.,&Cont i,E .(2 021) .Di ffere ntmole cul arc han gesun-
derliethesamephenotypictransition:Originsandconsequencesof
independent shifts to homostyly within species. Molecular Ecology,
32, 1–18 .
Naiki, A . (2012). Heterostyly and the possibility of its breakdown by
polyploidization. Plant Species Biology, 27, 3–29.
Nasrallah, J. B. (2017). Plant mating systems: Self- incompatibility and
evolutionary transitions to self- fer tility in the mustard family.
Current Opinion in Genetics and Development, 47, 54–60.
Ness,R.W.,Wright,S.I.,&Barrett,S.C.H.(2010).Mating-systemvari-
ation, demographic history and patterns of nucleotide diversity in
the tristylous plant Eichhornia paniculata. Genetics, 184, 381–392.
|
19 of 19
MORA-CARRER A et al.
O'Lea ry, S. J., Purit z, J. B., Wil lis, S. C. , Hollenbe ck, C. M., & P ortnoy,
D.S.(2018).T hese aren't the loci you'e lookingfor: Principlesof
effective SNP filtering for molecular ecologists. Molecular Ecology,
27, 3193–3206.
O' Mal ley,R.C .,Huang,S.S .C. ,So ng,L. ,Lewse y,M.G., Bar tl ett,A .,N er y,
J. R., Galli, M., Gallavotti, A., & Ecker, J. R. (2016). Cistrome and epi-
cistromefeaturesshapetheregulatory DNAlandscape. Cell, 165,
1280–1292.
Ono,K.,Akagi,T.,Morimoto, T.,Wónsch,A., &Tao,R.(2018). Genome
re- sequencing of diverse sweet cherry (Prunus avium) individu-
als reveals a modifier gene mutation conferring pollen- part self-
compatibility. Plant and Cell Physiology, 59, 1265–1275.
Pavan,S.,Delvento,C., Ricciardi,L.,Lotti,C .,Ciani,E.,&D'Agostino, N.
(2020). Recommendations for choosing the genotyping method
and best practices for quality control in crop genome- wide associa-
tion studies. Frontiers in Genetics, 11, 4 47.
Piper,J.G., Charlesworth,B., &Charleswor th,D.(1986).Breeding sys-
tem evolution in Primula vulgaris and the role of reproductive assur-
ance. Heredity, 56, 207–217.
Poptsova,M.S.,Il'Icheva, I. A.,Nechipurenko,D. Y.,Panchenko, L. A.,
Khodikov, M. V., Oparina, N. Y., Polozov, R. V., Nechipurenko,
Y.D.,&Grokhovsky,S.L.(2014).Non-randomDNAfragmentation
in next- generation sequencing. Scientific Reports, 4, 1–6.
Potente, G., Léveillé-Bourret, É., Yousefi, N., Choudhury, R. R ., Keller, B.,
Diop,S.I., Duijsings,D.,Pirovano,W.,Lenhard, M.,Szövényi,P.,&
Conti, E . (2022). Comp arative Ge nomics Elu cidates the O rigin of
a Supergene Controlling Floral Heteromorphism. Molecular Biology
and Evolution, 39, msac035.
Potente, G., Stubbs, R.L., Yousefi, N., Pirovano, W., Szövényi, P., &
Conti, E., (2022). Comparative transcriptomics reveals commonali-
ties and differences in the genetic underpinnings of a floral. dimor-
phism Sci Rep, 12, 20771.
Quinlan,A.R.,&Hall,I.M.(2010).BEDTools:Aflexiblesuiteof utilities
for comparing genomic features. Bioinformatics, 26, 841–842.
Rice, A ., & McLysaght, A . (2017). Dos age-sen sitive genes i n evolution
and disease. BMC Biology, 15, 1–10 .
Robinson,J. T.,Thorvaldsdóttir,H., Winckler,W.,Gut tman, M.,L ander,
E.S.,Getz,G.,&Mesirov,J.P.(2011).Integrativegenomicsviewer.
Nature Biotechnology, 29, 24–26.
Ruiz- Martín, J., Santos- Gally, R., Escudero, M., Midgley, J. J., Pérez-
Barrales, R., & Arroyo, J. (2018). Style polymorphism in Linum
(Linaceae): A case of Mediterranean parallel evolution? Plant
Biology, 20, 100–111.
Schoen,D.J.,Johnston,M.O.,L'Heureux,A.-M.,Marsolais,J.V.,Heureux,
A.- M. L ., & Marsolais, J. V. (1997). Evolutionary history of the mat-
ing system in Amsinckia (Boraginaceae). Evolution, 51, 1090–1099.
Shimizu, K. K., & Tsuchimatsu, T. (2015). Evolution of selfing: Recurrent
patterns in molecular adaptation. Annual Review of Ecology,
Evolution, and Systematics, 46, 593–622 .
Shore, J.S.,Hamam,H.J., Chafe, P.D.J., Labonne,J. D.J.,Henning,P.
M., & McCubbin, A. G. (2019). The long and short of the S- locus in
Turnera (Passifloraceae). New Phytologist, 224, 1316–1329.
Stebbins, G. L. (1957). Self fertilization and population variability in the
higher plants. The American Naturalist, 91, 337–35 4.
Tasdighian,S.,VanBel,M.,Li,Z.,VanDePeer,Y.,Carretero-Paulet,L.,&
Maere, S. (2017). Reciprocally retained genes in the angiosperm
lineage show the hallmarks of dosage balance sensitivity. Plant Cell,
29, 2766–2785.
Tsuchimatsu, T., Kaiser, P., Yew, C. L., Bachelier, J. B., & Shimizu, K. K.
(2012). Recent loss of self- incompatibility by degradation of the
male component in allotetraploid Arabidopsis kamchatica. PLoS
Genetics, 8, 16–19.
Williamson, R. J., Josephs, E. B., Platts, A. E., Hazzouri, K. M., Haudry,
A.,Blanchet te,M., &Wright,S.I.(2014).Evidenceforwidespread
positive and negative selection in coding and conserved noncoding
regions of Capsella grandiflora. PLoS Genetics, 10, 1–12 .
Wright, S . I., Kalis z, S., & Slot te, T.( 2013). Evolutiona ry conse quences
of self- fertilization in plants. Proceedings of the Royal Society B:
Biological Sciences, 280, 20130133.
Yu, C. P., Lin, J. J., & Li, W. H. (2016). Positional distribution of transcrip-
tion factor binding sites in Arabidopsis thaliana. Scientific Reports,
6, 1–7.
Yuan,S.,Barrett,S.C.H.,Duan,T.,Qian,X.,Shi,M.,&Zhang,D.(2017).
Ecological correlates and genetic consequences of evolution-
ary transitions from distyly to homostyly. Annals of Botany, 120,
775–789.
Yuan,S.,Barrett,S.C .H.,Li,C.,Li,X.,Xie,K.,&Zhang,D.(2018).Genetics
of distyly and homostyly in a self- compatible Primula. Heredity, 122,
1–10 .
Zhao,Z.,Zhang,Y.,Shi,M.,Liu,Z.,Xu,Y.,Luo,Z.,Yuan,S.,Tu,T.,Sun,Z.,
Zhang, D., &Barrett, S. C.H.(2023).Genomicevidence suppor ts
the genetic convergence of a supergene controlling the distylous
floral syndrome. New Phytologist, 237, 6 01–614.
Zhong,L.,Barrett,S.C.H.,Wang,X.,Wu,Z.,Sun,H.,Li,D.,Wang,H.,&
Zhou, W. (2019). Phylogenomic analysis reveals multiple evolution-
ary origins of selfing from outcrossing in a lineage of heterostylous
plants. New Phytologist, 224, 1290 –130 3.
Zhou,W.,Barret t,S.C.H.,Li,H.D.,Wu,Z.K.,Wang,X.J.,Wang,H.,&Li,
D.Z.(2017).Phylogeographicinsightsontheevolutionarybreak-
down of heterostyly. New Phytologist, 214, 1368–1380.
Zh ou,W., Barre tt ,S .C.H. ,Wa ng,H. ,&L i,D.Z.(2 012 ).L ossofflo ra lpo ly-
morphism in heterostylous Luculia pinceana (Rubiaceae): A molecu-
lar phylo geographic pe rspective. Molecular Ecology, 21, 4631–4645.
SUPPORTING INFORMATION
Additional supporting information can be found online in the
Suppor tingInformationsectionattheendofthisarticle.
How to cite this article: Mora- Carrera, E., Stubbs, R. L.,
Potente, G., Yousefi, N., Keller, B., de Vos, J. M., Szövényi, P.,
& Conti, E. (2024). Genomic analyses elucidate S- locus
evolution in response to intra- specific losses of distyly in
Primula vulgaris. Ecology and Evolution, 14, e10940. ht t ps: //
doi.org/10.1002/ece3.10940
