Sex is determined by XY chromosomes across the radiation of dioecious Nepenthes pitcher plants

Abstract

Species with separate sexes (dioecy) are a minority among flowering plants, but dioecy has evolved multiple times independently in their history. The sex‐determination system and sex‐linked genomic regions are currently identified in a limited number of dioecious plants only. Here, we study the sex‐determination system in a genus of dioecious plants that lack heteromorphic sex chromosomes and are not amenable to controlled breeding: Nepenthes pitcher plants. We genotyped wild populations of flowering males and females of three Nepenthes taxa using ddRAD‐seq and sequenced a male inflorescence transcriptome. We developed a statistical tool (privacy rarefaction) to distinguish true sex specificity from stochastic noise in read coverage of sequencing data from wild populations and identified male‐specific loci and XY‐patterned single nucleotide polymorphsims (SNPs) in all three Nepenthes taxa, suggesting the presence of homomorphic XY sex chromosomes. The male‐specific region of the Y chromosome showed little conservation among the three taxa, except for the essential pollen development gene DYT1 that was confirmed as male specific by PCR in additional Nepenthes taxa. Hence, dioecy and part of the male‐specific region of the Nepenthes Y‐chromosomes likely have a single evolutionary origin.

Full text

LETTER
doi:10.1002/evl3.142
Sex is determined by XY chromosomes
across the radiation of dioecious Nepenthes
pitcher plants
Mathias Scharmann,1,2 ,3 T. Ulmar Grafe,4Faizah Metali,4and Alex Widmer1
1Institute of Integrative Biology, ETH Zurich, Z ¨
urich 8092, Switzerland
2Department of Ecology and Evolution, University of Lausanne, Lausanne 1015, Switzerland
3E-mail: mathias.scharmann@env.ethz.ch
4Faculty of Science, Universiti Brunei Darussalam, Gadong BE 1410, Brunei Darussalam
Received October 24, 2018
Accepted September 3, 2019
Species with separate sexes (dioecy) are a minority among flowering plants, but dioecy has evolved multiple times independently in
their history. The sex-determination system and sex-linked genomic regions are currently identified in a limited number of dioecious
plants only. Here, we study the sex-determination system in a genus of dioecious plants that lack heteromorphic sex chromosomes
and are not amenable to controlled breeding: Nepenthes pitcher plants. We genotyped wild populations of flowering males and
females of three Nepenthes taxa using ddRAD-seq and sequenced a male inflorescence transcriptome. We developed a statistical
tool (privacy rarefaction) to distinguish true sex specificity from stochastic noise in read coverage of sequencing data from wild
populations and identified male-specific loci and XY-patterned single nucleotide polymorphsims (SNPs) in all three Nepenthes
taxa, suggesting the presence of homomorphic XY sex chromosomes. The male-specific region of the Y chromosome showed little
conservation among the three taxa, except for the essential pollen development gene DYT1 that was confirmed as male specific
by PCR in additional Nepenthes taxa. Hence, dioecy and part of the male-specific region of the Nepenthes Y-chromosomes likely
have a single evolutionary origin.
KEY WORDS: Carnivorous plant, dioecy, molecular sexing, plant sex chromosome, privacy rarefaction, sex-determination, sex-
specific loci.
Impact Summary
One of the most striking polymorphisms observed in
organismal populations is the existence of male and
female individuals. In contrast to animals, where this
condition is common, plants are usually functional
hermaphrodites. Some plants, however, are dioecious,
that is, individuals are either of male or female sex.
Dioecy has evolved hundreds of times independently in
plants, which offers the potential for comparative studies
of sex chromosome evolution and for investigating the
genetic basis of transitions between hermaphroditism
and dioecy (Charlesworth 2015). Yet empirical data to
test hypotheses about why some species are dioecious
and others hermaphroditic, and how such transitions are
achieved, are lacking. Despite their potential, the sex-
determination mechanisms of most dioecious plants are
not known, and few new species have been investigated
since the seminal review by Westergaard (1958). Also,
markers for molecular sexing have important appli-
cations in agriculture, horticulture, and conservation.
Historically, the identification of sex-determination sys-
tems was limited to species that can be bred in controlled
1
C2019 The Author(s). Evolution Letters published by Wiley Periodicals, Inc. on behalf of Society for the Study of Evolution
(SSE) and European Society for Evolutionary Biology (ESEB).
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.
Evolution Letters

M. SCHARMANN ET AL.
conditions or have heteromorphic sex chromosomes.
This is now changing with genome-scale sequencing
technology. Here, we investigated the sex-determination
system of carnivorous pitcher plants in the genus Ne-
penthes. We surveyed wild populations of three species
by genotyping a large number of loci throughout their
genomes. Because such data are noisy, we propose
a solution to the common problem of distinguishing
true signal from noise in presence–absence data by
generating null distributions through permutations of
the observed data. We discovered loci that occur only in
males and reveal an XY sex chromosome system. One
gene on the Nepenthes Y chromosome is particularly
interesting, because it is presumably essential for pollen
development and present only in males, and thus can
be used to diagnose the sex of nonflowering plants
Although the majority of flowering plant species are
functional hermaphrodites, plant sexual systems and sex-
determination mechanisms are highly diverse (Charlesworth
2002; Bachtrog et al. 2014). Only 5–6% of species have fe-
male and male flowers on separate individuals (dioecy), but the
evolutionary transition to dioecy may have occurred as many
as 800 times independently in angiosperms (Renner 2014). In
contrast to outcrossing–selfing transitions due to loss of self-
incompatbility, for some of which the underlying genetic changes
have recently been uncovered (e.g., Shimizu and Tsuchimatsu
2015), relatively little is known about the genes involved in tran-
sitions from hermaphroditism to dioecy and in sex determination
in plants (Charlesworth 2016), although sex-determining genes
have been identified in three dioecious plant species: persim-
mon (Diospyros lotus, Akagi et al. 2014), Asparagus officinalis
(Harkess et al. 2017; Murase et al. 2017), and kiwifruit (Actinidia,
Akagi et al. 2018). The main hypotheses for the evolution of sep-
arate sexes in plants involve a combination of trade-offs between
the sex functions, plus disadvantage of inbreeding (Charlesworth
and Charlesworth 1978).
Many dioecious plants have genetic sex determination, which
may involve sex chromosomes. Sex chromosomes differ from au-
tosomes by having suppressed meiotic recombination around the
sex-determining genes. These form a fully sex-linked chromo-
somal region whose transmission is limited to one sex. When
this sex is male, the system is referred to as male heteroga-
mety (male genotype XY, female XX), and female heterogamety
when the fully sex-linked region is transmitted via females (male
ZZ, female ZW). The sex-specific, fully nonrecombining regions
(male-specific region of the Y, termed MSY, or female-specific
region of the W) show a number of special properties. First, an
MSY may contain sequences that are absent from the X, and thus
male specific (Y-hemizygous, transmitted only from fathers to
sons). As X and Y chromosomes are thought to evolve from a
pair of autosomes, the gain of male-specific sequences can be
explained by several mechanisms, such as the rise of a new male-
determining mutation, or the translocation of a male-determining
cassette (Tennessen et al. 2018), or the localized expansion of
repetitive sequences due to the lack of recombination. Second,
over evolutionary time, the MSY may undergo genetic degenera-
tion and lose functional genes that were initially shared with the
X chromosome (Bachtrog 2013). Sex chromosomes have evolved
independently many times in plants, and will therefore proba-
bly have diverse ages and levels of degeneration. Heteromorphic
sex chromosomes have diverged sufficiently in size or structure
to be distinguished optically with a microscope, whereas homo-
morphic sex chromosomes may have more subtle differences that
can be detected only by molecular genetic methods. Despite their
great potential for comparative studies, few plant sex chromo-
somes have been studied in detail (Ming et al. 2011; Harkess
and Leebens-Mack 2017; Muyle et al. 2017). Knowledge of sex-
determination systems and the identification of fully sex-linked
genetic markers are important for molecular sexing of juveniles
or nonflowering adults in agriculture, breeding, and conservation.
Cytogenetics and linkage analysis in families are established
methods to study sex determination and discover sex linkage of
genes (Charlesworth and Mank 2010). However, these strategies
fail in many dioecious organisms because their karyotypes are
homomorphic (Filatov 2015), or because controlled breeding is
difficult, since many dioecious plants are woody and reproduce
only after many years (Renner and Ricklefs 1995). Several next-
generation sequencing techniques have now greatly increased
knowledge about sex-linked genes (reviewed by Muyle et al.
2017). However, they require either prior knowledge of heteroga-
mety, controlled breeding, or whole-genome sequencing, which
remains expensive and time consuming. An alternative class of
strategies uses population polymorphism to infer sex linkage of
loci (reviewed by Muyle et al. 2017), even without pedigrees.
These strategies can potentially allow sex-linked regions to be
discovered by cheaper reduced-representation sequencing (RRS)
methods such as RAD-seq (Baird et al. 2008; Elshire et al. 2011;
Peterson et al. 2012), although the gained information will remain
incomplete, because typically only a few percent of a genome
is covered. Nevertheless, the discovery of sex-linked markers
by RRS has been successful in organisms such as Crustaceans
(Carmichael et al. 2013), Anolis lizards (Gamble and Zarkower
2014), geckos (Gamble et al. 2015a), and frogs (Brelsford et al.
2017; Jeffries et al. 2018).
A major problem faced by approaches that use popula-
tion polymorphism to infer sex-linkage, and sex specificity in
2EVOLUTION LETTERS 2019

PITCHER PLANT SEX CHROMOSOMES
Figure 1. Sexual dimorphism in Nepenthes inflorescences. Left:
male inflorescence of N. rafflesiana s.l. Right: Female inflorescence
of N. mirabilis var. globosa. Photos: M. Scharmann
particular, is error in the measurement of locus presence and ab-
sence (Text S4). Presence–absence error has long been recognized
as a problem in fragment length genotyping methods, but it is ex-
acerbated in RRS data, in which missing loci occur in a highly
stochastic manner (Mastretta-Yanes et al. 2015; Bresadola et al.
2019), and can make sex-specific sequences appear in both sexes
(Bewick et al. 2013; Gamble and Zarkower 2014; Heikrujam
et al. 2015; Brelsford et al. 2017), and probably represent false
positive results. One suggested solution to reduce the number of
false positives is to compare increasing numbers of males and fe-
males (Gamble and Zarkower 2014; Gamble et al. 2015b). Unfor-
tunately, in RAD data the number of shared loci decreases with in-
creasing sample numbers (Mastretta-Yanes et al. 2015). Beyond a
number that is unpredictable and specific to each dataset, true sex-
specific loci may be missed because they are too rarely sequenced.
We developed a statistical procedure to deal with this prob-
lem, and applied it to characterize the sex-determination system of
Nepenthes pitcher plants. Nepenthes (Nepenthaceae, Caryophyl-
lales) includes at least 160 species of perennial vines and shrubs
occurring mostly in Southeast Asia (Cheek and Jebb 2001; Clarke
et al. 2018). They are carnivorous plants that supplement their nu-
trition by killing and digesting animals in their modified pitcher
leaves (Juniper et al. 1989; Moran and Clarke 2010; Pavloviˇ
c
and Saganov´
a 2015). All Nepenthes are dioecious, whereas close
relatives (families Ancistrocladaceae, Dioncophyllaceae, Droser-
aceae, and Drosophyllaceae; Cu´
enoud et al. 2002; Renner and
Specht 2011; Walker et al. 2017) are hermaphroditic. The indi-
vidual male and female flowers (Fig. 1) are readily recognized be-
cause reproductive organs of the other sex abort early in develop-
ment (Subramanyam and Narayana 1971). We hypothesized that
sex in Nepenthes has a genetic basis, or is determined during early
life stages, because there are no reports of sexual plasticity (Clarke
2001), or functional hermaphroditism. Nepenthes karyotypes
(2n=80, Heubl and Wistuba 1997) do not suggest heteromorphic
sex chromosomes.
Here, we investigated the previously undescribed sex-
determination system of multiple Nepenthes species. Because
controlled breeding of these slow growing plants faces many
challenges, we sampled wild populations. We used Silene lati-
folia to test our method, as this species has well-studied hetero-
morphic sex chromosomes. Specifically, we asked the following
questions: (1) Are there sex-linked loci in Nepenthes? (2) Are the
same sex-linked loci shared among different Nepenthes species?
(3) Which expressed genes are sex linked? We discovered fully
male-specific and XY-patterned loci and developed a molecular
sexing assay for Nepenthes. The identified markers include two
candidate sex-determination genes, and these suggest that part of
the Y chromosome is ancestral in this genus.
Methods
SAMPLING, ddRAD-seq, AND GENOTYPING
Natural populations of Nepenthes were sampled in Brunei Darus-
salam (Borneo), Singapore, and the Seychelles. Fresh leaf material
was stored in a nucleic acid preserving buffer (Camacho-Sanchez
et al. 2013). The sexes of Nepenthes plants were recorded from
fresh or dry inflorescences. Scans for sex-linked loci were con-
ducted separately on three Nepenthes taxa: N. pervillei Blume,
N. gracilis Korth., and N. rafflesiana sensu lato (Table 1). We
extracted DNA from leaves using silica column kits (Nucleospin
Plant II; Macherey Nagel, D¨
uren, Germany) and prepared se-
quencing libraries following the ddRAD-seq protocol (Peterson
et al. 2012) using the enzymes EcoRI and TaqI. Library pools (84
or 96-plex) were sequenced on an Illumina HiSeq 2500. Bioin-
formatic data filtering, de novo assembly of reference contigs
(“RAD-tags,” very short contigs with a mean length of c. 96
bases), read mapping, genotype calling, and quality filtering fol-
lowed a modified dDocent pipeline (Puritz et al. 2014) and code
is deposited at https://github.com/mscharmann.
The exploration of sex-specific markers in N. rafflesiana s.l.
followed an iterative strategy with two rounds of sexing, geno-
typing, bioinformatic analysis, and PCR validation (Text S1). To
increase the phylogenetic range of our study and to validate molec-
ular sexing, we also included individuals of known sex for addi-
tional species (Table S2-1in Text S2). To validate our method for
detection of sex-specific loci, we also genotyped populations of a
species with a well-known, heteromorphic XY sex-determination
EVOLUTION LETTERS 2019 3

M. SCHARMANN ET AL.
Tab l e 1 . Sample sizes and origin for the taxa sequenced in this study.
Taxon Sampling location
Number of
males
Number of
females Sequencing method
Nepenthes pervillei Blume Seychelles, Mah´
e 28 22 ddRAD-seq (Peterson et al. 2012)
Nepenthes gracilis Korth. Brunei Darussalam, Borneo 10 10 ddRAD-seq (Peterson et al. 2012)
Nepenthes rafflesiana sensu lato,
here defined as:
39 22 ddRAD-seq (Peterson et al. 2012)
Nepenthes rafflesiana “typical
form” (Clarke 1992, 1997)
Brunei Darussalam, Borneo 13 7
Nepenthes rafflesiana “giant
form”(Clarke 1992, 1997)
Brunei Darussalam, Borneo 5 3
Nepenthes rafflesiana Jack Singapore 10 4
Nepenthes hemsleyana Macfarl. Brunei Darussalam, Borneo 11 8
Nepenthes khasiana Hook.f. cultivated/artificially prop. 1 – RNA-seq
Silene latifolia Poiret Switzerland 27 32 GBS (Elshire et al. 2011)
system, S. latifolia Poiret, using a single-digest GBS protocol
(Elshire et al. 2011). Details of the Silene samples and genotyp-
ing are provided in Text S3.
DETECTION OF SEX-LINKAGE
We distinguished between sex-linked loci showing sex-specificity,
meaning without homology between the two sex chromosomes
(e.g., Y-linked loci whose fully X-linked copy is absent or
undetectable by our methods), and loci that are present on both
sex chromosomes but whose allele frequencies diverged between
the sex chromosomes, called ZW- or XY-patterned variants
(Gammerdinger and Kocher 2018). We define sex-specific loci as
contigs to which sequencing reads can be aligned from only one of
the sexes (presence–absence polymorphism). However, this need
not reflect true presence–absence, because observed absence may
be due to methodological factors (deliberate or random) specific
to each dataset. Imperfect detection may be caused by many
underlying factors, including the number of male and female
individuals investigated, their genetic relatedness, the species’
genome size and structure, the library preparation method,
sequencing depth, and bioinformatic processing. Consider, for
example, a particular genetic marker scored “present” in nine out
of ten males and “absent” in five out of five females. True complete
male specificity cannot be distinguished from technical artifacts
due to random variation, for example, in sequencing depth or
sampling bias. To improve the inference of sex specificity and fa-
cilitate comparisons of datasets from different species, sampling
schemes, and sequencing runs, we used a resampling strategy that
evaluates the biological signal among dataset-specific artifacts
and uncertainties. The procedure is implemented in python and
named “privacy rarefaction” (https://github.com/mscharmann).
Further details, including a performance analysis with simula-
tions, are presented in Texts S4 and S5.
We tested for completely or partially XY- or ZW-patterned
SNPs, that is, ones with different allele and heterozygote
frequencies in the two sexes, using genotypes and PLINK
version 1.07 (Purcell et al. 2007). Associations between biallelic
SNPs and sex were analyzed by chi-squared tests, and candidate
SNPs were accepted as sex associated at a false discovery rate
of 5% (Benjamini and Hochberg 1995), and then classified as
XY-patterned if males were predominantly heterozygous, or as
ZW-patterned if females were predominantly heterozygous. To
perform this test on a reasonable number of SNPs, we allowed
data from up to 25% of individuals to be missing per SNP.
POPULATION GENETICS OF CANDIDATE SEX-LINKED
LOCI
We tested whether linkage disequilibrium (LD), that is, the non-
random association of alleles at separate loci, differed in male
samples between sex-linked regions and the genomic background
(represented by 15,000 randomly selected pairs of nonsex-linked
SNPs). Stringent quality filters were applied: singleton SNPs
were removed, SNPs in male-specific contigs were excluded if
any heterozygous genotypes were called in the males (because Y-
hemizygosity implies that heterozygosity is not true, so these were
probably paralogous sequences), and for non-sex-linked contigs
any excessively heterozygous SNPs (Hardy–Weinberg test with
significance level 5%) were excluded. Excessively heterozygous
SNPs were retained for XY-patterned contigs. LD (r2) was calcu-
lated exclusively for SNPs from different contigs using PLINK.
The same contrasts were made for nucleotide diversity π,which
was averaged per contig for all SNP-containing contigs (including
singleton SNPs) in VCFtools version 0.1.15 (Danecek et al. 2011)
and for contigs without SNPs taken directly from bam alignments.
The same filters were applied to both data (minimum read depth
3, maximum read depth 75, and minimum population presence
4EVOLUTION LETTERS 2019

PITCHER PLANT SEX CHROMOSOMES
0.75). The significance of differences of the means was evaluated
by randomization (10,000 rounds of re-sampling without replace-
ment from the two groups, randomizing group membership).
COMPARISON OF CANDIDATE LOCI TO A MALE
INFLORESCENCE TRANSCRIPTOME
We sequenced and assembled the transcriptome of a single de-
veloping male inflorescence of Nepenthes khasiana Hook.f. (Text
S6) to identify and annotate sex-linked candidate loci. Fresh inflo-
rescences of the species used for ddRAD-seq were not available
in cultivation. The transcriptome was searched (a) by BLAST
for similarity to candidate contigs (thresholds 90 aligned bases
and 75% identity) and (b) by repeating privacy rarefaction with
ddRAD-seq reads directly mapped to the transcriptome rather
than the RAD-tag reference (bwa mem; Li 2013; retaining multi-
ple mappings).
Candidate transcripts from both approaches were annotated
by BLAST search against the NCBI Genbank nucleotide col-
lection (November 7, 2016 version) and the nonredundant protein
collection (March 26, 2016 version). Transposable elements (TEs)
were detected using RepeatMasker 4.0.6 (Smit et al. 2013) ver-
sion 20150807 (eukaryota). Proteins with at least 50 amino acids
were predicted by TransDecoder (Trinity package) and annotated
against nr, UniProt Swiss-Prot (August 17, 2016 version), and
Arabidopsis thaliana proteins in UniProtKB (April 3, 2016 ver-
sion). PFAM domains were detected using hmmer 3.1b1 (Eddy
et al. 2016), accepting hits at e-value 10−5.
PCR VALIDATION
Candidate sex-specific contigs were chosen for PCR validation
based on a ranking of the privacy rarefaction results (using the
highest stringency level reached and the bootstrap support), tax-
onomic overlap, and the quality of annotation of matching tran-
scripts. PCR primers were designed in Geneious R6 (Biomatters
Ltd., Auckland, New Zealand), and tested according to the proto-
col described in Text S2.
Results
SEX-LINKED LOCI
We first searched for sex-specific contigs in the illustrative exam-
ple of S. latifolia GBS data using privacy rarefaction. When small
numbers of individuals of each sex were analyzed, the procedure
yielded similar numbers of male- and female-specific candidates
(Table S8; Fig. 2, top, dark gray zone), which decreased mono-
tonically with greater numbers of individuals, as expected. When
the numbers of males and females analyzed were increased, a
clear signal of male heterogamety emerged. With four or more
individuals of each sex analyzed, the proportion of male-specific
candidate loci increased and finally became significantly greater
than the number of female-specific candidates (Fig. 2, top, light
gray zone). At 11 and more individuals of each sex analyzed, the
number of female-specific candidates dropped to zero, whereas
the number of male-specific candidates remained high (Fig. 2,
top, white zone). Hence, these curves correctly diagnosed an
XY-system for S. latifolia, and rejected a ZW-system. Due to
this characteristic drop-out of false-positives, we also refer to the
numbers of analyzed individuals of each sex as “privacy rarefac-
tion stringency.” Some of the herein identified male-specific S.
latifolia contigs were previously reported to be sex-linked (Text
S3, Table S11).
Qualitatively consistent signatures of male-specific contigs
were detected independently in N. pervillei,N. gracilis,andN. raf-
flesiana s.l. (Fig. 2, Table S8). The proportion of male-specific loci
among all loci was, however, about 10-fold lower in Nepenthes
than S. latifolia. Estimates based on subsamples of 10 males and
10 females were only 0.02% for N. pervillei (11.8/51,002), 0.11%
for N. rafflesiana s.l. (43.7/40,508.7), and 0.06% for N. gracilis
(13/22,789), versus 1.52% for S. latifolia (586.4/38,455).
In N. pervillei and N. rafflesiana s.l., as well as in S. latifo-
lia, we also detected XY-patterned SNPs, but not in N. gracilis,
while none of the species yielded any ZW-patterned SNPs. The
XY-patterned SNPs for S. latifolia recovered 289 contigs already
known to be sex linked in that species, which, however, represent
only a small fraction of the known and theoretically expected
S. latifolia sex-linked sequences (c. 1/7 of the genome; Text S3,
Table S11). This low power was expected for a sequencing strat-
egy that targets only a small subset of the genome. Almost all
XY-patterned SNPs had an allele frequency close to 0.5 and
near-complete heterozygosity in males, but were homozygous
in females (Table S9). The proportions of XY-patterned SNPs
were also much lower in Nepenthes than in S. latifolia (2376/149,
311 =1.6%, similar to the proportion of male-specific loci): the
estimates were, respectively, 0.25% and 0.017% for N. pervillei
(97/38,783) and N. rafflesiana s.l. (37/222,188).
POPULATION GENETICS OF SEX-LINKED LOCI
Fully Y-linked loci experience no recombination, which should
lead to increased population LD between different male-specific
contigs. All three testable male-specific contigs of N. pervillei
showed perfect LD (r2=1, which was c. 0.7 units higher than
the genomic background; N=3 SNP pairs, P=0.06). Like-
wise, the male-specific contigs of N. rafflesiana s.l. (contain-
ing 82 SNP pairs) had an r2that was on average c. 0.4 above
the genomic mean (P<10−5); the median value was complete
LD (r2=1). Among contigs with XY-patterned SNPs, r2was
0.15 higher than the background mean in N. pervillei (P=10−5,
N=147), but was no different from the mean in N. rafflesiana
s.l. (P=0.56, N=11). Our observation of some low LD values
between sex-specific contigs is likely due to the discreteness of
EVOLUTION LETTERS 2019 5

M. SCHARMANN ET AL.
Figure 2. Evidence for male-specific loci and XY sex-determination systems in Silene latifolia and three Nepenthes spp. (privacy rarefac-
tion curves). Shown are counts of sex-specific contigs (y-axis) as a function of the number of individuals of each sex sampled to score
sex specificity (x-axis, stringency). Sex-specific contigs are defined as those to which sequencing reads from only one sex can be aligned.
Dots represent averages, and whiskers one standard deviation of 200 bootstrapped combinations of males and females. Note natural
log-scale of y-axis and hence undefined zero and negative values in the SD ranges. The background shading of the plots indicates three
relevant zones that are directly informative on the sex-determination system: a dark gray zone (low stringency) indicates no difference
between the sexes, the light gray zone (intermediate stringency) highlights where significant differences between sexes are found, and
white background (highest stringency) shows the biologically plausible zone where sex-specific markers are obtained in only one sex.
Male-specific candidates were found in all species up to the maximum possible stringency (the minimum number of male individuals and
female individuals), except in N. pervillei (asterisk).
allele frequency estimates from small sample sizes, and the allele
frequency dependence of LD metrics, whose maximum possible
value is frequently much less than 1.0 (VanLiere and Rosenberg
2008).
The mean nucleotide diversity πin male-specific (putatively
Y-linked) contigs tended to be lower than the genomic background
in all three taxa (Fig. 3), consistent with theoretical expectations
for Y-specific loci, whose effective population size is only 1/4
of that of autosomal genes, and can be reduced much further by
genetic hitchhiking and high variance among males in the number
of sired offspring (Wilson Sayres et al. 2014). This difference was
significant for N. rafflesiana s.l. (P<10−3), but not for N. pervillei
6EVOLUTION LETTERS 2019

PITCHER PLANT SEX CHROMOSOMES
Figure 3. Mean per-site nucleotide diversity πof contigs in male Nepenthes of three taxa for male-specific, XY-patterned, and random
nonsex-linked contigs. All contigs mapping 3–75 reads in 75% of males per population were included. The same sets of individuals are
considered in each category. No XY- or ZW-patterned contigs were found in N. gracilis.Median=white dot, box =25–75% quartiles,
whiskers =1.5∗interquartile range, violin =estimated kernel density.
or N. gracilis (P=0.13 and P=0.055, respectively). In contrast,
mean πof males in contigs with XY-patterned SNPs was higher
than the genomic mean for both N. pervillei (P=10−5,Fig.3)
and N. rafflesiana s.l. (P=0.002; Fig. 3), a consequence of high
male heterozygosity in the XY-patterned SNPs.
SHARED CANDIDATE LOCI BETWEEN SPECIES,
FUNCTIONAL ANNOTATIONS, AND PCR VALIDATION
Six candidate sex-specific contigs were found at privacy rarefac-
tion stringency level 5 that were shared between N. gracilis and
N. rafflesiana s.l. No sex-specific candidates were shared between
N. pervillei and the other species (Table S10). There was no over-
lap in XY-patterned SNPs between the Nepenthes species, and no
direct overlap between male-specific contigs and XY-patterned
ones. However, one male-specific contig of N. gracilis and one
XY-patterned contig of N. rafflesiana s.l. both matched (full
length alignment, e-value 1×10–19) to the same inflorescence
transcript containing a DUF4283 (domain of unknown function,
http://pfam.xfam.org/family/PF14111, November 9, 2016).
One male-specific contig of N. pervillei aligned to the tran-
script of a bHLH transcription factor, and the best matches in all
accessed databases were consistently to predicted orthologs of
the Arabidopsis gene DYSFUNCTIONAL TAPETUM1 (DYT1).
A further XY-patterned contig of N. pervillei matched a tran-
script annotated as A. thaliana SEPALLATA-1 (SEP1),which
aligned to the predicted 3-UTR of the putative N. pervillei
SEP1-ortholog, and contained two SNPs that were both homozy-
gous in 95% of females and heterozygous in 96% of males. No
estimate of SEP1 X–Y divergence was possible because the male
inflorescence transcriptome reads were not heterozygous. In N.
gracilis, a male-specific contig matched a long transcript similar
to a mitochondrial NADH-ubiquinone oxidoreductase from Beta
vulgaris (Swiss-Prot). This finding was unexpected and may rep-
resent either an unspecific match of the short (96 bp) contig to
the inflorescence transcript, or else a cyto-nuclear transfer to the
sex chromosomes. The occurrence of organellar genes on plant
sex chromosomes has been documented in other species (Steflova
et al. 2013).
All other candidate loci either included traces of TEs, or had
no known sequence motifs (Table S10). In particular, of the 38
sex-linked inflorescence transcripts identified here (by 41 match-
ing sex-linked contigs), 34 (89%) could not be annotated, or con-
tained TEs. TEs were commoner than in nonsex-linked transcripts
(χ2df =1=5.2, P=0.02). Nine out of 13 sex-linked TE-transcripts
annotated as gypsy-like retrotransposons, a significant overrepre-
sentation relative to nonsex-linked TE transcripts (χ2df =1=8.85,
P=0.003).
Complementary to the sex-specificity scan on the ddRAD-
de novo reference, we repeated privacy rarefaction by directly
mapping the ddRAD reads to the male inflorescence transcrip-
tome to find further annotated sex-linked genes. This identified
seven transcripts (Table S10) that map male-specific regions of
the genomes. We recorded only high-confidence male-specific
candidate transcripts present in at least four males and absent
in at least four females, and bootstrap support greater 0.5 in at
least one species. No female-specific transcripts (false positives)
EVOLUTION LETTERS 2019 7

M. SCHARMANN ET AL.
reached these support levels. A single transcript was male-specific
in N. rafflesiana s.l. but could not be annotated. Four close tran-
script “isoforms” (contigs that share sequence but differ slightly
in structure, as assembled by Trinity; Haas et al. 2013) were male
specific in both N. gracilis and N. rafflesiana s.l., but they lacked
similarity to any known motif except for one isoform similar to
a Jockey-1 Drh retrotransposon. However, two transcripts were
male specific in both N. pervillei and N. rafflesiana s.l., and one of
these also matched a N. pervillei male-specific contig (see above).
These two transcripts appear to be close isoforms (putative intron
presence–absence), and both annotated as DYT1 (see above).
We tested by PCR whether the putative DYT1-ortholog is
male specific in a broad range of Nepenthes species. A single
PCR product of approximately 290 bp length was observed exclu-
sively and consistently in phenotypically sexed male Nepenthes
but never in females (Text S2). Multiple males and females were
screened in eight taxa, and 1–2 individuals from 14 further taxa.
Presence–absence of the PCR product was fully consistent with
the phenotypic sex of all 56 individuals. Sanger sequencing of the
PCR product confirmed the identity of the target region. Hence,
this locus is male specific across a phylogenetically broad range
of Nepenthes species and can be used for molecular sexing.
PERFORMANCE OF THE RESAMPLING STRATEGY
We tested the performance of privacy rarefaction on simulated
datasets resembling typical RAD-seq experiments under a range
of missing data levels, sampling schemes, and sizes of the sex-
specific region (Text S5). The procedure correctly identified the
heterogametic sex in >90% of simulations when the proportion
of sex-specific contigs in the genome was at least one permil, and
virtually all contigs classified as sex specific were true positives.
A naive scoring method, in contrast, failed to detect the heteroga-
metic sex in most scenarios because sex-specific contigs appeared
in both sexes, and it typically reported many false positives. How-
ever, most of the true sex-specific contigs were not detected in
simulated RAD-seq data because of the missing data inherent to
this sequencing method (low sensitivity). Nevertheless, the rela-
tive size of the sex-specific regions was usually estimated to the
true order of magnitude (Table S5-1 in Text S5).
Discussion
THE NEPENTHES SEX-DETERMINATION SYSTEM
Our findings reveal that sex determination has a genetic basis
in Nepenthes and involves a nonrecombining region in males.
Nepenthes karyotypes suggest that the sex chromosomes are
homomorphic (Heubl and Wistuba 1997), consistent with the
lower proportions of sex-linked contigs in Nepenthes compared
to S. latifolia with its large and heteromorphic Y-chromosome.
The proportions of male-specific contigs allow us to hypothesize
Figure 4. Summary of results on the sex-determination sys-
tem for Nepenthes, annotated on a plastid phylogeny (after
Meimberg and Heubl 2006). The crown of the genus is c. 17.7 (CI
11.0-24.3) million year old (Text S7). It constrains the minimum age
at which dioecy evolved and DYT1 became a male-specific gene.
NA =not available/not tested. Genome sizes were quantified
by flow cytometry. The proportion of Y-specific contigs is given
at 10 individuals of each sex (stringency). Nepenthes rafflesiana
s.l. contains several entities, for which the PCRs were conducted
separately.
that the size of the MSY relative to the whole genome is about
10-fold smaller in Nepenthes than in S. latifolia, and that within
the genus, it is smallest in N. pervillei. However, we note that the
characterization of sex chromosomes via reduced-representation
sequencing methods necessarily remains incomplete (Text S3),
and very strict analyses, such as the resampling procedure we
propose here, are required to avoid false inferences (Text S5).
Furthermore, our sampling included several subpopulations in
S. latifolia and N. rafflesiana s.l., which may have impeded the
detection of deme-specific sex-linked loci.
The MSY of Nepenthes appears to contain a “core region”
that is conserved throughout the genus. The DYT1 gene was
male specific in both N. pervillei and N. rafflesiana s.l., and
part of it was consistently PCR amplified in known males but
never in females of 22 Nepenthes species (Text S2), representing
all major clades (Fig. 4; Mullins 2000; Meimberg et al. 2001;
Meimberg and Heubl 2006; Scharmann et al. unpubl. data). The
shared MSY locus therefore suggests a single origin of dioecy
in Nepenthes that most likely predates the most recent com-
mon ancestor of extant Nepenthes at 17.7 (CI 11.0–24.3) mil-
lion years ago but followed the split between Nepenthaceae and
8EVOLUTION LETTERS 2019

PITCHER PLANT SEX CHROMOSOMES
hermaphroditic Droseraceae at least 44.2 million years ago (av-
erage 71.1 with CI 44.2–98.0; Text S7). However, the age of the
shared MSY core does not necessarily reflect the age of the sex
chromosomes: their identity could have changed over time and
may also differ between Nepenthes species because the ancestral
MSY could have been translocated to other chromosomes in a
process called “sex-chromosome turnover” (Blaser et al. 2014;
Jeffries et al. 2018; Tennessen et al. 2018). Nevertheless, the Ne-
penthes MSY core is probably older than the heteromorphic S.
latifolia sex chromosomes (11 million years, Krasovec et al.
2018). These alternatives can be explored in future comparative
whole genome sequencing or mapping studies.
During the radiation of Nepenthes, the MSY has diverged
between species, as is expected over such long divergence times,
particularly for noncoding sequences. Only six out of 135 male-
specific contigs were shared between N. rafflesiana s.l. and N.
gracilis, and none were shared with the more distant N. pervillei
(Fig. 4). Male-specific loci shared between N. pervillei and N.
rafflesiana s.l. were only recovered with the help of longer, tran-
scriptome contigs to align ddRAD reads. Absence of shared male-
specific contigs should not, therefore, be interpreted as evidence
for independent origins of sex chromosomes, but rather reflects
sequence divergence between species. Further evidence for a com-
mon origin followed by interspecific divergence is found in a
DUF4283 transcript, which is male specific in N. gracilis but
XY-patterned in N. rafflesiana s.l., suggesting X and MSY alleles
(i.e., gametologs) have lost sequence similarity in the former but
not in the latter species.
NONCODING DNA AND SPECIAL SIGNIFICANCE OF
DYT1 AND SEP1
Nonrecombining regions of sex chromosomes accumulate repet-
itive, noncoding sequences and TEs in species with both hetero-
morphic or largely homomorphic sex chromosomes ( ˇ
Cerm´
ak et al.
2008; Wang et al. 2012). In Nepenthes, most sex-linked genomic
regions detected by our approach were noncoding sequences and
TEs and only a few genes with putative developmental functions
were identified. Of these, a Nepenthes homolog of DYT1 appears
to be located in the MSY of all Nepenthes species. DYT1 is es-
sential for tapetum development and thus pollen fertility in A.
thaliana (Zhang et al. 2006), rice (Jung et al. 2005; Wilson and
Zhang 2009; Cai et al. 2015), and tomato (Jeong et al. 2014).
Given this gene’s functional conservation in these distantly re-
lated Angiosperms, we speculate that its function is the same in
Nepenthes, and future ork could validate this hypothesis, for ex-
ample, via transient transformation of Nepenthes (Migueletal.
2019). Our analysis suggests that DYT1 is absent from Nepenthes
females and must thus be absent from the X chromosome. Such a
deletion of DYT1 from the X chromosome would constitute a re-
cessive male-sterility mutation, as required early in the evolution
of dioecy for the transition from a hermaphroditic to a gynodioe-
cious mating system (Charlesworth and Charlesworth 1978). It is
notable that in Arabidopsis, DYT1 directly regulates the expres-
sion of TDF1 (Gu et al. 2014), a gene that in dioecious asparagus
is essential for male fertility and, like DYT1 in Nepenthes,is
located in the MSY (Harkess et al. 2017; Murase et al. 2017).
Apparently, this pollen development pathway was involved twice
independently in the evolution of angiosperm XY chromosomes,
and possibly in the transition to dioecy.
The second Nepenthes gene of interest is a homolog of the
homeotic MADS box gene SEP1, an early-acting regulator of flo-
ral organ identity in A. thaliana (Pelaz et al. 2000), which was
XY-patterned in N. pervillei.TwoSEP1-linked SNPs were het-
erozygous in 27 of 28 males, whereas 21 of 22 females were
homozygous, consistent with the existence of strongly X- and Y-
linked copies. If SEP homologs in Nepenthes are involved in the
determination of floral organ identity (as in A. thaliana,Theißen
et al. 2016), the sex-linked Nepenthes SEP1 homolog could be
involved in unisexual flower development. In particular, sequence
differences between the Nepenthes SEP1 X- and Y-linked copies
might modify their functions such that they suppress the develop-
ment of either carpels or stamens. In S. latifolia,however,SEP1
homologs are not directly involved in sex determination and are
not located on the sex chromosomes (Matsunaga et al. 2004).
The possible roles of DYT1 and SEP1 in the origin of
dioecy in Nepenthes require further attention. Even if these are
not primary sex-determining genes in extant Nepenthes,they
might have been under sexually antagonistic selection during the
evolution of dioecy because loss of function of DYT1 or SEP1
alleles might abort nonfunctional organs at early developmental
stages, thus saving resources. The fully unisexual morphology
of extant Nepenthes flowers (Subramanyam and Narayana 1971)
implies further developmental genetic differences between males
and females.
Conclusion
This study reports the discovery of an XY sex-determination sys-
tem in dioecious pitcher plants (Nepenthes spp). The sex chromo-
somes are homomorphic with a small Y-specific region, which
has a relatively old core that is shared between distinct species.
The nonrecombining region is enriched for noncoding sequences
and TEs, but also contains several expressed genes with putative
developmental functions.
ACKNOWLEDGMENTS
We thank N. Zemp for data of Silene latifolia. We are indebted to H.
Luqman, C. K¨
uffer, J. Mougal, K. Beaver, C. Morel, E. Nancy, A. Street,
and T. Kropf for accessing N. pervillei. A.H.H. Tinggal and I. Daud with
family are thanked for hospitality in Brunei, and L.W. Ngai for assistance
EVOLUTION LETTERS 2019 9

M. SCHARMANN ET AL.
with field work in Singapore. The Seychelles Bureau of Standards, the
Seychelles Ministry of Environment and Energy, the Ministry of Industry
and Primary Resources of Brunei Darussalam, the Agri-Food and
Veterinary Authority, and the National Parks Board of Singapore kindly
granted permits for research, sampling, and export of plant material.
C. Michel is thanked for preparing sequencing libraries. S. Hartmeyer
and U. Zimmermann kindly donated samples of sexed Nepenthes from
their private collections. Sequencing and computation were carried out
with the Genetic Diversity Center Z¨
urich, the Quantitative Genomics
Facility Basel, and the Functional Genomics Center Z¨
urich. We thank the
numerous colleagues and anonymous reviewers for valuable comments
on earlier versions of the manuscript. This work was supported by the
ETH Z¨
urich and the R¨
ubel foundation. Keygene N.V. owns patents
and patent applications protecting its Sequence Based Genotyping
technologies. The authors declare no conflicts of interest.
AUTHOR CONTRIBUTIONS
MS performed the research including data collection and analysis, TUG
and FM contributed logistic support and materials, and MS and AW
designed and interpreted the research and wrote the manuscript.
DATA ARCHIVING
Sequencing reads of ddRAD-seq and RNA-seq are deposited at the Euro-
pean Nucleotide Archive under projects PRJEB20488 and PRJEB22838.
Bioinformatic and statistical code used for this study is available at
https://github.com/mscharmann. VCF files, data used for statistics, and
assembled Nepenthes contigs in fasta format including the sex-linked loci
are deposited at Dryad (https://doi.org/10.5061/dryad.8gs9388).
LITERATURE CITED
Akagi, T., I. M. Henry, R. Tao, and L. Comai. 2014. A Y-chromosome–
encoded small RNA acts as a sex determinant in persimmons. Science
346:646–650.
Akagi, T., I. M. Henry, H. Ohtani, T. Morimoto, K. Beppu, I. Kataoka, et al.
2018. A Y-encoded suppressor of feminization arose via lineage-specific
duplication of a cytokinin response regulator in kiwifruit. Plant Cell
30:780–795.
Bachtrog, D. 2013. Y-chromosome evolution: emerging insights into pro-
cesses of Y-chromosome degeneration. Nat. Rev. Genet. 14:113–124.
Bachtrog, D., J. E. Mank, C. L. Peichel, M. Kirkpatrick, S. P. Otto, T.-L.
Ashman, et al. 2014. Sex determination: why so many ways of doing it?
PLoS Biol. 12:e1001899.
Baird, N. A., P. D. Etter, T. S. Atwood, M. C. Currey, A. L. Shiver, Z. A.
Lewis, et al. 2008. Rapid SNP discovery and genetic mapping using
sequenced RAD markers. PLoS One 3:e3376.
Benjamini, Y., and Y. Hochberg. 1995. Controlling the false discovery rate:
a practical and powerful approach to multiple testing. J. R. Stat. Soc. B
Stat. Methodol. 57:289–300.
Bewick, A. J., F. J. J. Chain, L. B. Zimmerman, A. Sesay, M. J. Gilchrist,
N. D. L. Owens, et al. 2013. A large pseudoautosomal region on the
sex chromosomes of the frog Silurana tropicalis. Genome Biol. Evol.
5:1087–1098.
Blaser, O., S. Neuenschwander, N. Perrin, A. E. A. Gardner, and E. T.
Day. 2014. Sex-chromosome turnovers: the hot-potato model. Am. Nat.
183:140–146.
Brelsford, A., G. Lavanchy, R. Sermier, A. Rausch, and N. Perrin. 2017.
Identifying homomorphic sex chromosomes from wild-caught adults
with limited genomic resources. Mol. Ecol. Resour. 17:752–759.
Bresadola, L., V. Link, C. A. Buerkle, C. Lexer, and D. Wegmann. 2019. Esti-
mating and accounting for genotyping errors in RAD-seq experiments.
bioRxiv 587428.
Cai, C.-F., J. Zhu, Y. Lou, Z.-L. Guo, S.-X. Xiong, K. Wang, Y. Zhong-Nan.
2015. The functional analysis of OsTDF1 reveals a conserved genetic
pathway for tapetal development between rice and Arabidopsis. Sci.
Bull. 60:1073–1082.
Camacho-Sanchez, M., P. Burraco, I. Gomez-Mestre, and J. A. Leonard.
2013. Preservation of RNA and DNA from mammal samples under field
conditions. Mol. Ecol. Resour. 13:663–673.
Carmichael, S. N., M. Bekaert, J. B. Taggart, H. R. L. Christie, D. I. Bassett,
J. E. Bron, et al. 2013. Identification of a sex-linked SNP marker in the
Salmon Louse (Lepeophtheirus salmonis) using RAD sequencing. PLoS
One 8:e77832.
ˇ
Cerm´
ak, T., Z. Kub´
at, R. Hobza, A. Koblizkova, A. Widmer, J. Macas, et al.
2008. Survey of repetitive sequences in Silene latifolia with respect
to their distribution on sex chromosomes. Chromosome Res. 16:961–
976.
Charlesworth, B., and D. Charlesworth. 1978. A model for the evolution of
dioecy and gynodioecy. Am. Nat. 112:975–997.
Charlesworth, D. 2002. Plant sex determination and sex chromosomes.
Heredity 88:94–101.
———. 2015. Plant contributions to our understanding of sex chromosome
evolution. New Phytol., 208:52–65.
———. 2016. Plant sex chromosomes. Annu. Rev. Plant Biol. 67:397–420.
Charlesworth, D., and J. E. Mank. 2010. The birds and the bees and the flowers
and the trees: lessons from genetic mapping of sex determination in
plants and animals. Genetics 186:9–31.
Cheek, M. R., and M. H. P. Jebb. 2001. Nepenthaceae. Flora Malesiana 15:1–
157.
Clarke, C. M. 1992. The ecology of metazoan communities in Nepenthes
pitcher plants in Borneo, with special reference to the community of
Nepenthes bicalcarata Hook. f. PhD Thesis. University of New England,
Armidale, Australia.
———. 1997. Nepenthes of Borneo. Natural History Publications (Borneo),
Kota Kinabalu, Malaysia.
———. 2001. Nepenthes of Sumatra and Peninsular Malaysia. Natural His-
tory Publications (Borneo), Kota Kinabalu, Malaysia.
Clarke, C. M., J. Schlauer, J. A. Moran, and A. S. Robinson. 2018. Systematics
and evolution of Nepenthes. Pp. 58–69 in A. M. Ellison and L. Adamec,
eds. Carnivorous plants: physiology, ecology, and evolution. Oxford
Univ. Press, Oxford, U.K.
Cu´
enoud, P., V. Savolainen, L. W. Chatrou, M. Powell, R. J. Grayer, and M.
W. Chase. 2002. Molecular phylogenetics of Caryophyllales based on
nuclear 18S rDNA and plastid rbcL, atpB, and matK DNA sequences.
Am. J. Bot. 89:132–144.
Danecek, P., A. Auton, G. Abecasis, C. A. Albers, E. Banks, M. A. De-
Pristo, et al. 2011. The variant call format and VCFtools. Bioinformatics
27:2156–2158.
Eddy, S. R., T. J. Wheeler, and the HMMER development team. 2016. HM-
MER. http://hmmer.org.
Elshire, R. J., J. C. Glaubitz, Q. Sun, J. A. Poland, K. Kawamoto, E. S. Buckler,
et al. 2011. A robust, simple genotyping-by-sequencing (GBS) approach
for high diversity species. PLoS One 6:e19379.
Filatov, D. A. 2015. Homomorphic plant sex chromosomes are coming of age.
Mol. Ecol. 24:3217–3219.
Gamble, T., and D. Zarkower. 2014. Identification of sex-specific molecular
markers using restriction site-associated DNA sequencing. Mol. Ecol.
Resour. 14:902–913.
Gamble, T., J. Coryell, T. Ezaz, J. Lynch, D. P. Scantlebury, and D. Zarkower.
2015a. Restriction site-associated DNA sequencing (RAD-seq) reveals
10 EVOLUTION LETTERS 2019

PITCHER PLANT SEX CHROMOSOMES
an extraordinary number of transitions among gecko sex-determining
systems. Mol. Biol. Evol. 32:1296–1309.
Gamble, T., J. Coryell, T. Ezaz, J. Lynch, D. P. Scantlebury, and D. Zarkower.
2015b. Restriction site-associated DNA sequencing (RAD-seq) reveals
an extraordinary number of transitions among gecko sex-determining
systems. Mol. Biol. Evol. 32:1296–1309.
Gammerdinger, W. J., and T. D. Kocher. 2018. Unusual diversity of sex
chromosomes in African cichlid fishes. Genes 9:480.
Gu, J.-N., J. Zhu, Y. Yu, X.-D. Teng, Y. Lou, X.-F. Xu, et al. 2014. DYT1
directly regulates the expression of TDF1 for tapetum development and
pollen wall formation in Arabidopsis. Plant J. 80:1005–1013.
Haas, B. J., A. Papanicolaou, M. Yassour, M. Grabherr, P.D. Blood, J. Bowden,
et al. 2013. De novo transcript sequence reconstruction from RNA-Seq:
reference generation and analysis with Trinity. Nat. Protoc. 8:1494–
1512.
Harkess, A., and J. Leebens-Mack. 2017. A century of sex determination in
flowering plants. J. Hered. 108:69–77.
Harkess, A., J. Zhou, C. Xu, J. E. Bowers, R. Hulst, S. Ayyampalayam, et al.
2017. The asparagus genome sheds light on the origin and evolution of
a young Y chromosome. Nat. Commun. 8:1279.
Heikrujam, M., K. Sharma, M. Prasad, and V. Agrawal. 2015. Review on
different mechanisms of sex determination and sex-linked molecular
markers in dioecious crops: a current update. Euphytica 201:161–194.
Heubl, G., and A. Wistuba. 1997. A cytological study of the genus Nepenthes
L. (Nepenthaceae). Sendtnera 4:169–172.
Jeffries, D. L., G. Lavanchy, R. Sermier, M. J. Sredl, I. Miura, A. Borz´
ee,
et al. 2018. A rapid rate of sex-chromosome turnover and non-random
transitions in true frogs. Nat. Commun. 9:4088.
Jeong, H.-J., J.-H. Kang, M. Zhao, J.-K. Kwon, H.-S. Choi, J. H. Bae, et al.
2014. Tomato male sterile 1035 is essential for pollen development and
meiosis in anthers. J. Exp. Bot. 65:6693–6709.
Jung, K.-H., M.-J. Han, Y.-S. Lee, Y.-W. Kim, I. Hwang, M.-J. Kim, et al.
2005. Rice undeveloped Tapetum1 is a major regulator of early tapetum
development. Plant Cell 17:2705–2722.
Juniper, B. E., R. J. Robins, and D. M. Joel. 1989. The carnivorous plants.
Academic Press, London.
Krasovec, M., M. Chester, K. Ridout, and D. A. Filatov. 2018. The mutation
rate and the age of the sex chromosomes in Silene latifolia. Curr. Biol.
28:1832–1838.e4.
Li, H. 2013. Aligning sequence reads, clone sequences and assembly contigs
with BWA-MEM. arXiv:1303.3997 [q-bio].
Mastretta-Yanes, A., N. Arrigo, N. Alvarez, T. H. Jorgensen, D. Pi˜
nero, and B.
C. Emerson. 2015. Restriction site-associated DNA sequencing, geno-
typing error estimation and de novo assembly optimization for popula-
tion genetic inference. Mol. Ecol. Resour. 15:28–41.
Matsunaga, S., W. Uchida, E. Kejnovsky, E. Isono, F. Moneger, B. Vyskot,
et al. 2004. Characterization of two SEPALLATA MADS-box genes
from the dioecious plant Silene latifolia. Sex. Plant Reprod. 17:189–
193.
Meimberg, H., and G. Heubl. 2006. Introduction of a nuclear marker for
phylogenetic analysis of Nepenthaceae. Plant Biol. 8:831–840.
Meimberg, H., A. Wistuba, P. Dittrich, and G. Heubl. 2001. Molecular phy-
logeny of Nepenthaceae based on cladistic analysis of plastid trnK intron
sequence data. Plant Biol. 3:164–175.
Miguel, S., E. Nisse, F. Biteau, S. Rottloff, B. Mignard, E. Gontier, et al.
2019. Assessing carnivorous plants for the production of recombinant
proteins. Front. Plant Sci. 10:793.
Ming, R., A. Bendahmane, and S. S. Renner. 2011. Sex chromosomes in land
plants. Annu. Rev. Plant Biol. 62:485–514.
Moran, J. A., and C. M. Clarke. 2010. The carnivorous syndrome in Nepenthes
pitcher plants. Plant Signal. Behav. 5:644–648.
Mullins, J. T. 2000. Molecular systematics of Nepenthaceae. PhD Thesis.
University of Reading, Reading, U.K.
Murase, K., S. Shigenobu, S. Fujii, K. Ueda, T. Murata, A. Sakamoto, et al.
2017. MYB transcription factor gene involved in sex determination in
Asparagus officinalis. Genes Cells 22:115–123.
Muyle, A., R. Shearn, and G. A. Marais. 2017. The evolution of sex chromo-
somes and dosage compensation in plants. Genome Biol. Evol. 9:627–
645.
Pavloviˇ
c, A., and M. Saganov´
a. 2015. A novel insight into the cost–benefit
model for the evolution of botanical carnivory. Ann. Bot. 115:1075.
Pelaz, S., G. S. Ditta, E. Baumann, E. Wisman, and M. F. Yanofsky. 2000. B
and C floral organ identity functions require SEPALLATA MADS-box
genes. Nature 405:200–203.
Peterson, B. K., J. N. Weber, E. H. Kay, H. S. Fisher, and H. E. Hoekstra.
2012. Double digest RADseq: an inexpensive method for de novo SNP
discovery and genotyping in model and non-model species. PLoS One
7:e37135.
Purcell, S., B. Neale, K. Todd-Brown, L. Thomas, M. A. R. Ferreira, D.
Bender, et al. 2007. PLINK: a tool set for whole-genome association
and population-based linkage analyses. Am. J. Hum. Genet. 81:559–
575.
Puritz, J. B., C. M. Hollenbeck, and J. R. Gold. 2014. dDocent: a RADseq,
variant-calling pipeline designed for population genomics of non-model
organisms. PeerJ 2:e431.
Renner, S. 2014. The relative and absolute frequencies of angiosperm sexual
systems: dioecy, monoecy, gynodioecy, and an updated online database.
Am. J. Bot. 101:1588–1596.
Renner, S., and R. Ricklefs. 1995. Dioecy and its correlates in the flowering
plants. Am. J. Bot. 82:596–606.
Renner, T., and C. D. Specht. 2011. A sticky situation: assessing adaptations
for plant carnivory in the Caryophyllales by means of stochastic character
mapping. Int. J. Plant. Sci. 172:889–901.
Scharmann, M., and T. U. Grafe. 2013. Reinstatement of Nepenthes hems-
leyana (Nepenthaceae), an endemic pitcher plant from Borneo, with a
discussion of associated Nepenthes taxa. Blumea 58:8–12.
Shimizu, K. K., and T. Tsuchimatsu. 2015. Evolution of selfing: recurrent
patterns in molecular adaptation. Annu. Rev. Ecol. Evol. Syst. 46:593–
622.
Smit, A., R. Hubley, and P. Green. 2013. RepeatMasker Open-4.0. 2013–2016.
Steflova, P., R. Hobza, B. Vyskot, and E. Kejnovsky. 2013. Strong accumula-
tion of chloroplast DNA in the Y chromosomes of Rumex acetosa and
Silene latifolia. Cytogenet. Genome Res. 142:59–65.
Subramanyam, K., and L. L. Narayana. 1971. A contribution to the floral
anatomy of Nepenthes khasiana Hook f. Proc. Indian Acad. Sci. B
73:124–131.
Tennessen, J. A., N. Wei, S. C. K. Straub, R. Govindarajulu, A. Liston, and
T.-L. Ashman. 2018. Repeated translocation of a gene cassette drives
sex-chromosome turnover in strawberries. PLoS Biol. 16:e2006062.
Theißen, G., R. Melzer, and F. R¨
umpler. 2016. MADS-domain transcription
factors and the floral quartet model of flower development: linking plant
development and evolution. Development 143:3259–3271.
VanLiere, J. M., and N. A. Rosenberg. 2008. Mathematical properties of the
r2 measure of linkage disequilibrium. Theor. Popul. Biol. 74:130–137.
Walker, J. F., Y. Yang, M. J. Moore, J. Mikenas, A. Timoneda, S. F. Brock-
ington, et al. 2017. Widespread paleopolyploidy, gene tree conflict, and
recalcitrant relationships among the carnivorous Caryophyllales. Am. J.
Bot. 104:858–867.
Wang, J., J.-K. Na, Q. Yu, A. R. Gschwend, J. Han, F. Zeng, et al. 2012.
Sequencing papaya X and Yh chromosomes reveals molecular basis
of incipient sex chromosome evolution. Proc. Natl. Acad. Sci. USA
109:13710–13715.
EVOLUTION LETTERS 2019 11

M. SCHARMANN ET AL.
Westergaard, M. 1958. The mechanism of sex determination in dioecious
flowering plants. Pp. 217–281 in M. Demerec, ed. Advances in Genetics.
Academic Press, London.
Wilson Sayres, M. A., K. E. Lohmueller, and R. Nielsen. 2014. Natural
selection reduced diversity on human Y chromosomes. PLos Genet.
10:e1004064.
Wilson, Z. A., and D.-B. Zhang. 2009. From Arabidopsis to rice: pathways in
pollen development. J. Exp. Bot. 60:1479–1492.
Zhang, W., Y. Sun, L. Timofejeva, C. Chen, U. Grossniklaus, and H. Ma.
2006. Regulation of Arabidopsis tapetum development and function
by DYSFUNCTIONAL TAPETUM1 (DYT1) encoding a putative bHLH
transcription factor. Development 133:3085–3095.
Associate Editor: S. Wright
Supporting Information
Additional supporting information may be found online in the Supporting Information section at the end of the article.
Tex t S1 . Preliminary molecular sexing assay for Nepenthes rafflesiana s.l.
Tex t S2 . A molecular sexing assay for the genus Nepenthes.
Tex t S3 . Analyses of sex-linkage in Silene latifolia.
Tex t S4 . Privacy rarefaction.
Tex t S5 . Performance analysis of privacy rarefaction on simulated RAD data.
Tex t S6 . Male inflorescence transcriptome of N. khasiana.
Tex t S7 . Phylogenetic dating of Nepenthes.
Tab l e S8 . Result tables of privacy rarefaction for three Nepenthes and Silene.
Tab l e S9 . Result tables of association tests for SNPs with sex and heterozygosities.
Table S10. Summary table of sex-linked loci including overlap between species and annotations for RAD-tag and transcriptome references.
Table S11. Data table for S. latifolia sex-linkage comparison of our results to previous studies.
12 EVOLUTION LETTERS 2019