Content uploaded by Madeleine Carruthers
Author content
All content in this area was uploaded by Madeleine Carruthers on Sep 04, 2022
Content may be subject to copyright.
Articles
https://doi.org/10.1038/s41559-021-01555-4
1Institute of Biodiversity, Animal Health and Comparative Medicine, College of Medical, Veterinary & Life Sciences, University of Glasgow, Glasgow, UK.
2School of Geographical and Earth Sciences, University of Glasgow, Glasgow, UK. 3Present address: Biotechnical Faculty, University of Ljubljana, Ljubljana,
Slovenia. 4Present address: School of Biological Sciences, University of Bristol, Bristol, UK. 5Present address: Inserm U981, Gustave Roussy Cancer Campus,
Université Paris Saclay, Villejuif, France. ✉e-mail: Kathryn.Elmer@glasgow.ac.uk
Whether offspring hatch from calcified eggs after a
period of external incubation (oviparity) or emerge
from the mother via live birth (viviparity) is a funda-
mental and dichotomous reproductive strategy with wide-ranging
consequences for amniote ecology, development, physiology and
evolution1–4. The difference between these parity modes is com-
plex; however, major transitions from oviparity to viviparity have
evolved repeatedly2,3. All birds, most reptiles and a few mammals
retain the ancestral state of oviparity, while nearly all mammals and
~20% of reptiles are viviparous2. Mammals transitioned to vivipar-
ity once, while squamate reptiles have independently transitioned
perhaps >115 times2,5. Amniote viviparity differs in the type and
extent of maternal provisioning, ranging from simple (nourished
by yolk, lecithotrophy) to complex (nourished through placenta,
placentotrophy)1,2,4,6. Nonetheless, the reproductive physiology,
neuroendocrine and developmental pathways regulating viviparous
reproduction are thought to involve structurally homologous tis-
sues among amniotes1,7–10. Gathering evidence to test this hypothe-
sis has been impeded by the deep evolutionary divergences between
oviparous–viviparous sister taxa, which obscure inference and pre-
clude crossing experiments. First attempts to circumvent this issue
have been implemented using differential gene expression between
oviparous–viviparous sister taxa9,10, yet the genetic basis of parity
mode has not been resolved for any animal3,4,11.
Major adjustments of maternal traits related to embryo develop-
ment are necessary for the transition from oviparity to viviparity,
including (1) an increase in the internal gestation time, to the point
where the embryo is fully developed before parturition and (2) the
loss of a calcified eggshell12. Simultaneously, changes are necessary
for the mother and for the healthy development of the embryo1,7,
such as an active physiological exchange of gas, water and calcium
and an adjusted immune response for maternal–foetal communica-
tion1,13, which is facilitated by uterine angiogenesis and membrane
vascularization1,7. Advanced evolutionary stages of viviparity often
involve more complex placentation, with enhanced nutritional
transfer from mother to foetus1,2,6. To prevent abortion and pro-
mote foetal development, the mother’s immune response may be
modified; for example, via the downregulation of pro-inflammatory
cytokines1. However, the similarity in immune response during
pregnancy across viviparous squamates remains poorly investigated
and inconclusive at this point1. These developmental and physiolog-
ical changes need coordinated integration, making the transition a
step-change; intermediate phenotypes are considered a fitness val-
ley and most species are fixed for oviparity or viviparity14.
Here, we identify the functional genomic architecture of alternative
parity modes—oviparity and viviparity—in an exceptional natural
model. The Eurasian common lizard (Zootoca vivipara) is viviparous
across most of its distribution while some southern populations are
oviparous15. Oviparity is its ancestral state and the viviparous lineages
represent one of the youngest transitions to viviparity in amniotes16.
Viviparity is at an early evolutionary stage in Z. vivipara live-bearing
lineages, with no or very limited nutritional transfer from the mother
to the embryo4,17. In this reproductively bimodal species, alternative
parity modes are genetically fixed but can interbreed15,18,19. These liz-
ards are therefore uniquely informative to genetically map the bases
of these parity modes, demonstrate the functional genetic mecha-
nisms and targets of natural selection and thereby infer the molecular
underpinnings of egg-laying and live-bearing.
Egg-laying and live-bearing lizards hybridize
We intensively sampled a contact zone between oviparous and vivip-
arous common lizards with a divergence time of ~4 Ma (million
The functional genetic architecture of egg-laying
and live-bearing reproduction in common lizards
Hans Recknagel1,3, Madeleine Carruthers1,4, Andrey A. Yurchenko 1,5, Mohsen Nokhbatolfoghahai1,
Nicholas A. Kamenos 2, Maureen M. Bain1 and Kathryn R. Elmer 1 ✉
All amniotes reproduce either by egg-laying (oviparity), which is ancestral to vertebrates or by live-bearing (viviparity), which
has evolved many times independently. However, the genetic basis of these parity modes has never been resolved and, conse-
quently, its convergence across evolutionary scales is currently unknown. Here, we leveraged natural hybridizations between
oviparous and viviparous common lizards (Zootoca vivipara) to describe the functional genes and genetic architecture of parity
mode and its key traits, eggshell and gestation length, and compared our findings across vertebrates. In these lizards, parity
trait genes were associated with progesterone-binding functions and enriched for tissue remodelling and immune system path-
ways. Viviparity involved more genes and complex gene networks than did oviparity. Angiogenesis, vascular endothelial growth
and adrenoreceptor pathways were enriched in the viviparous female reproductive tissue, while pathways for transforming
growth factor were enriched in the oviparous. Natural selection on these parity mode genes was evident genome-wide. Our
comparison to seven independent origins of viviparity in mammals, squamates and fish showed that genes active in pregnancy
were related to immunity, tissue remodelling and blood vessel generation. Therefore, our results suggest that pre-established
regulatory networks are repeatedly recruited for viviparity and that these are shared at deep evolutionary scales.
NATURE ECOLOGY & EVOLUTION | www.nature.com/natecolevol
Articles NATurE EcOlOgy & EVOluTION
years ago)20 where hybridization had been speculated18,21 (Extended
Data Fig. 1). We quantified the parity mode traits for pregnant
females and their clutches (n = 606): (1) embryonic stage at oviposi-
tion/parturition, (2) number of days a clutch was externally incu-
bated, (3) calcium content (% Ca) within eggshells/membranes and
(4) eggshell/membrane thickness. In addition, we genotyped adults
(using double-digest RAD sequencing (ddRADSeq); ref. 15), result-
ing in 80,696 genome-mapped22 polymorphic loci (n = 831, includ-
ing the phenotyped females).
Oviparous and viviparous clutches differed substantially. Eggs
from oviparous females had a thick shell (49.8 ± 0.8 μm) covered
with a dense layer of calcium crystals (25.4 ± 1.3% Ca), encircling
embryos laid at an early developmental stage (stage 31.1 ± 0.1) and
requiring extended external incubation (35 ± 0.2 d) (Fig. 1a–d)
before they hatched at full development (neonate). Conversely,
offspring from viviparous females were delivered fully devel-
oped (as neonates), surrounded by a thin uncalcified membrane
(4.2 ± 0.1 μm; 1.6 ± 0.6% Ca) from which they emerged within a
short time (1.9 ± 0.1 d) (Fig. 1r–u and Extended Data Fig. 2). These
phenotyping results were consistent with previous work on gesta-
tion time and eggshell traits in common lizards17,23.
Despite these dramatic differences in reproduction traits, phe-
notypic and genotypic data revealed oviparous–viviparous hybrids.
We found that 14.0% of adults inherited at least 10% of their genome
from the alternative parity mode and that 6.1% of adults were
first-generation (F1) hybrids (Fig. 1v). F1 hybrid females (45–55%
genomic admixture) were intermediate between viviparous and
oviparous characteristics with respect to eggshell thickness (mean,
26.7 ± 2.3 μm; 13.0 ± 2.9% Ca), gestation time (mean, 25.8 ± 1.4 d)
and developmental stage at oviposition (mean, stage 33.9 ± 0.4)
(Figs. 1j–m,v). Therefore, here we provided definitive evidence of
contemporary natural hybridization in the wild between oviparous
and viviparous lizards.
The genetic basis of parity modes
This striking hybridization between parity modes presents a natural
genetic crossing experiment for determining the genetic architec-
tures of oviparity and viviparity. Using genotypic and phenotypic
data from females (n = 458 individuals; 80,696 single nucleotide
polymorphisms, SNPs), we performed admixture mapping on two
essential reproductive traits: (1) gestation time (a score combining
the number of external incubation days and the embryonic stage at
oviposition/parturition per clutch) and (2) eggshell traits (a score
calculated from the eggshell/membrane thickness and calcium con-
tent), using linear mixed models (Extended Data Figs. 2 and 3).
To identify relevant genes and biological processes, we estimated
the chromosome-wide linkage disequilibrium (LD) decay from a
whole-genome resequencing dataset (n = 65 admixed females; mean
genome coverage 4.1× each) (Supplementary Data 1). Genes within
the same chromosome-specific LD-window as the SNPs associated
with gestation time and/or eggshell traits (Extended Data Fig. 4)
were considered as candidate genes.
We found that eggshell traits and gestation time were both genet-
ically determined and that our experiment had the ability to resolve
these traits with high resolution, even in this complex natural popu-
lation. The phenotypic variance explained by the genotype data was
very high: gestation time, 0.97 (95% confidence interval (CI) = 0.96–
0.98) and eggshell traits, 0.98 (95% CI = 0.96–0.98) (Supplementary
Table 1). Genome-wide, 221 genetic variants (488 genes) were sig-
nificantly associated with reproductive traits (adjusted P < 0.01).
More genetic variants were associated with gestation time (210
SNPs and 439 candidate genes) than with eggshell traits (17 SNPs
and 38 genes), indicating that the genetic basis of embryo reten-
tion is more complex than that of the eggshell traits (Supplementary
Data 2). From the 17 SNPs associated with eggshell traits, an excess
was shared with gestation time (six when the expected shared SNPs
by chance is 0.05; P < 0.0001; 11 genes). These SNPs were located
within a locus on chromosome 7 (two SNPs within 100 base pairs
(bp)) and within a locus on chromosome 3 (two SNPs within
100 bp), while two SNPs were located on unplaced scaffolds. The
candidate genes (465 genes) were significantly enriched for eight
pathways, including the immune system response (Fas signalling
and T cell activation) and tissue remodelling (the cadherin signal-
ling pathway) (Supplementary Table 2).
For gestation time, significant associations were found on most
chromosomes (total accumulated span of 41.2 megabases (Mb);
Supplementary Data 2 and Extended Data Fig. 5), with an excess
on chromosome 8 (64 significant SNPs; total span of 19.4 Mb)
(Fig. 2a). The candidate genes were significantly enriched (P < 0.01)
for processes associated with cell growth, proliferation and death
(Supplementary Table 3), indicative of tissue remodelling. The three
SNPs most highly associated with gestation time were found close
to the LYPLA1 (chromosome 7), PGRMC1 (chromosome 8) and
SOCS2 (chromosome 14) genes, all of which have reproductive
relevance. LYPLA1 has been previously associated with pregnancy
in mammals24. PGRMC1 is a progesterone-binding membrane
receptor protein highly expressed in mammalian placentae25.
Progesterone levels are essential for embryonic implantation in
the uterus and control development and ultimately parturition of
the foetus12,26–28 and the introgression of Neanderthal variants into
modern humans is associated with higher fertility29. SOCS proteins
are negative regulators of pro-inflammatory cytokines and therefore
modulate the maternal immune response during pregnancy30.
For eggshell traits, significant associations were found on
chromosomes 1, 2, 3, 7, 10, 11 and 14 (total span of 3.3 Mb;
Supplementary Data 2, Extended Data Fig. 5 and Fig. 2b). Genes
were significantly enriched (P < 0.01) for processes related to cell
communication and the immune system (Supplementary Table 4).
The three loci that showed the strongest associations with eggshell
traits mapped closely to the LGMN (chromosome 3), LYPLA1
(chromosome 7) and CRTC1 (chromosome 2) genes. These genes
have clear functional relevance for amniote reproduction. LGMN
is upregulated in placenta and uterine tissues8,31 and is associated
with the cadherin pathway, which establishes the interconnectiv-
ity between the maternal–foetal membranes1 and LYPLA1 plays
a role in mammalian pregnancy24 and was highly associated with
gestation traits in our study. CRTC1 is a transcription co-activator
involved in calcium augmentation32 and fertility in mice33 and has
been correlated with egg productivity in birds34.
Developmental pathways regulating the parity modes
We compared gene expression between functional reproduc-
tive tissue (glandular uterus) from female oviparous and vivipa-
rous common lizards during reproductive (early pregnancy and
mid-pregnancy) and postreproductive (after oviposition/parturi-
tion, also referred to as ‘parition’) time periods using RNA-seq
(Supplementary Table 5) (n = 17). Early pregnancy represents the
period when oviparous and viviparous females have completed
vitellogenesis and started ovulation. Meanwhile, at mid-pregnancy,
ovulation is completed and, in oviparous females, the eggshell is
formed35. We found 2,610 genes that were significantly differen-
tially expressed between the parity modes at reproductive stages
(early pregnancy and/or mid-pregnancy) but not postpregnancy
(Extended Data Fig. 6 and Fig. 3a). These differentially expressed
genes (DEGs) were functionally involved in ten enriched biological
pathways, including significant enrichment (P < 0.05) for apoptosis
signalling and the Parkinson disease pathway (Supplementary Table 6).
We speculate that the former may regulate the tissue remodelling
occurring at the interphase of the embryonic membranes and uterine
tissue and that the latter includes genes involved in vesicular trans-
port; for example, in the exchange of inorganic molecules36. While
viviparous common lizards mainly provide embryonic nutrition
NATURE ECOLOGY & EVOLUTION | www.nature.com/natecolevol
Articles
NATurE EcOlOgy & EVOluTION
Q ancestry
0
0.2
0.4
0.6
0.8
1.0
Qoviparous = 0.9–1.0
(n = 321)
Qviviparous = 0.0–0.1
(n = 299)
Qadmixed = 0.1–0.9
(n = 97)
100 µm
100 µm
100 µm
100 µm
100 µm
50 µm
50 µm
10 µm
10 µm
10 µm
0.5 mm 2 mm1 mm 2 mm
5 mm5 mm 5 mm 5 mm 5 mm
2 mm
a e j n r
b f k o s
c g l p t
d
v
h m q u
Fig. 1 | Natural hybridization and backcrossing between oviparous and viviparous common lizards result in intermediate phenotypes and genotypes.
a–d, In clutches laid by oviparous females, the eggshell is more calcified and external incubation time is longer (~35 d) (a), embryo is at an early
developmental stage (stage ~31) (b), eggshell has fewer calcium crystals (~25%) (c) and eggshell is thick (~50 μm) (d). e–h, Hybrid females that
backcrossed with oviparous lizards have clutches with generally thinner eggshells (e,g,h) and embryos that are laid at a later developmental stage (f)
compared to oviparous females. j–m, F1 hybrids are intermediate in eggshell thickness (j,l,m) and embryonic development (k) between oviparous and
viviparous phenotypes. n–q, Females backcrossed with viviparous lizards have clutches with short incubation times (o) and thinner eggshells (n,p,q).
r–u, Offspring from viviparous females (r,s) emerge from thin membranes (s) within hours to a few days (~1.9 d) and are fully developed (> stage 40,
neonate). (In s, the dehydrated eggshell membrane is still evident and image contrast has been increased to facilitate distinguishing the neonate from the
background.) The membrane has no calcium crystals (~1.6%; t) and is much thinner (~4 μm; u). (Note the difference in scale between eggshells thickness
images taken from oviparous and viviparous individuals. In c, g, l, p: white arrowheads indicate calcium crystals on the outer surface of eggshells.)
v, The gradient in phenotypic traits is reflected in the genotypes of common lizards in the examined contact zone. Genome-wide ancestries (Q) differ
considerably and show varying proportions of ‘oviparous’ and ‘viviparous’ ancestral genomes in admixed individuals.
NATURE ECOLOGY & EVOLUTION | www.nature.com/natecolevol
Articles NATurE EcOlOgy & EVOluTION
via the yolk and do not generate complex placental structures,
exchange of inorganic molecules between mother and embryo has
been shown37. Moreover, we found a total of 507 genes differentially
expressed after pregnancy and these potentially reflect fundamental
differences between phylogeographic lineages or in uterine struc-
ture or activity between parity modes. These genes were enriched
for cell regulation, extracellular matrix organization, blood plasma
regulation and immune response (Supplementary Table 6).
We found a significant relationship between gene expression at
reproductive stages and the genes inferred from genetic mapping
of the parity mode traits (more overlap than expected by chance;
98 out of 488 candidate genes were also differentially expressed;
P < 0.0001) (Fig. 3b). This included LYPLA1, which was a top can-
didate from genetic mapping that was downregulated in the vivipa-
rous uterine tissue. Further, genome-wide P values for differential
expression at reproductive stages were correlated with genome-wide
differentiation (FST) (estimate = −0.034; P = 0.017), indicating that
DEGs were more likely to be located in genomic regions with higher
genetic differentiation between the parity modes (below).
Three gene co-expression modules were associated with parity
mode (false discovery rate (FDR) < 0.05). Significantly enriched
pathways associated with the two modules overexpressed in the
viviparous uterine tissue (total of 1,245 genes) included angio-
genesis, the vascular endothelial growth factor signalling path-
way and the beta 1 and 2 adrenergic receptor signalling pathways
(Supplementary Table 7). The module overexpressed in the ovipa-
rous uterine tissue (total of 203 genes) was significantly enriched
for the transforming growth factor beta signalling pathway
(Supplementary Table 7), which regulates growth and cell prolifera-
tion38. This is consistent with both viviparous and oviparous uterine
tissues undergoing changes during pregnancy; the oviparous expe-
riences more growth35, while the viviparous increases vasculariza-
tion and potentially maternofetal communication, for example to
provide gas exchange and facilitate embryonic retention12,35. These
changes involve more genes and pathways in viviparous reproduc-
tive tissue than in oviparous tissue.
To pinpoint key regulators, we conducted a hub gene search on
the basis of the three co-expression modules associated with par-
ity mode. We identified a total of 64 (51 annotated) differentially
expressed, highly connected hub genes (Supplementary Methods
and Supplementary Data 3), which were significantly enriched for
biological processes associated with cellular structure and extra-
cellular communication, such as the establishment of cell polarity,
epithelium development and biological adhesion (Supplementary
Table 8). Four hub genes overlapped with our candidate genes iden-
tified from the genetic mapping, which is more than expected by
chance (3.2× fold enrichment; P = 0.036), two of which (RAPGEF2
and RHOG) were located on chromosome 8. RAPGEF genes are
key regulators of cell adhesion, secretion, proliferation and differen-
tiation39,40 and blood formation in extra-embryonic membranes41.
RHOG is involved in cell migration42. The other two genes were
MAP1A (chromosome 11), for a protein involved in microtubule
formation43 and implicating a potential role in vesicular transport,
and KANK3 (chromosome 2). KANK genes are co-opted for ver-
tebrate vascular development44, which might be related to the vas-
cularization of uterine tissue required for viviparity. Therefore,
the functional genetic regulation of tissues involved in alterna-
tive reproductive modes identified here has confirmed roles in
vertebrate reproduction and is underpinned by some of the same
genes inferred from our admixture mapping analysis.
Divergent selection between oviparous and viviparous
lineages
Having identified relevant heritable genetic variations, we aimed to
infer the loci under divergent selection between the oviparous and
viviparous lineages by applying an evaluation based on principal
component analysis (PCAdapt approach)45 on females and males
(n = 717 adults from the sample locality, ranging from pure ovipa-
rous to pure viviparous and including hybrids, for 83,696 SNPs). This
made it possible to explore the relationship between the functional
genes identified from admixture mapping and the genome-wide
response to selection while holding the environment constant, which
0
2
4
6
8
10
12
14
Chromosome
Gestation time (–log
10
P)
10 11 12 13 14 15 16 17 18 191 3 4 5 6 7 8 92
a
0
2
4
6
8
10
Eggshell traits (–log
10
P)
10 11 12 13 14 15 16 17 18 19
Chromosome
1 3 4 5 6 7 8 92
b
Fig. 2 | Genetic architecture of parity mode. a,b, Admixture mapping reveals genetic variants associated with gestation time (a) and eggshell traits (b).
The genome-wide significance is indicated by the horizontal red dotted line, while the blue line indicates a P-value threshold of <0.01 using Benjamini–
Hochberg correction.
NATURE ECOLOGY & EVOLUTION | www.nature.com/natecolevol
Articles
NATurE EcOlOgy & EVOluTION
is shared at this site. In total, 1,050 SNPs showed significant signals
of response to selection (q < 0.01) and these SNPs were distributed
across the genome (Extended Data Fig. 7). We found a significant
overlap of SNPs under selection between oviparous and viviparous
populations and SNPs associated with the female reproductive traits
from the admixture mapping analysis (n = 67, 30.3% SNPs shared).
Using our LD-decay approach (Extended Data Fig. 4), we iden-
tified 1,621 candidate genes under divergent selection between
the oviparous and viviparous lineages. A functional link was also
apparent from the significant overlap (n = 312 overlapping genes)
between the genes differentially expressed in uterine tissue and the
genes under selection (7.8% of genes shared; P < 0.0001). Highly
enriched pathways for genes under selection included immune sys-
tem response genes (Fas signalling; P = 0.05), blood formation and
vascularization (angiogenesis) and tissue remodelling (the cadherin
signalling pathway and epidermal growth factor receptor signalling
pathway) (Supplementary Table 9). Of the top ten enriched path-
ways of SNPs showing a response selection (Supplementary Table 9)
and the top ten enriched pathways for SNPs associated with parity
mode (Supplementary Table 2), five pathways are shared. This indi-
cates that a large fraction of the genes under selection are function-
ally relevant for the alternative parity modes.
CSAD
NFE2L2
KRT6A
TMBIM6
SHMT1
KIF5A
FKBP11
ACVRL1
COQ10A
TUBA1A
PECAM1
APOH
SRP68
METRNL
GPX3
SPCS1
GATA2
PLXND1
EOGT
ADAMTS9 IL4I1
HIST1H110
TRIM39
KLF2 KCNN1
OTOR
PXDN
CDYL2
TMED6
DPEP1
RPL3
SLC7A9
EFHD2
PADI1 ANK1
NCAPH
PROM2
BEGAIN
SLC25A29
FAM180B
LGR4
NLRPL
SPTBN1
CRYBA2
PARP9
KMO
UGT1A1 TIGD5
AURKA
TUBB1
TULP1
FREM1
GPR88
TMEM63A
APOB
SDC1
PRR11
GALNT17
PSPH NXPE3
CNDP1
PRSS12
SLC34A2
SEMA7A
CLIC3
TBX3
RAMP2
ACE
NXPH3
HBM
HBZ
ADAMTS17
HBBR
FUT4
SLC7A1
ETS2
UXS1
CNTN1
GBA2
CDO1
ALDH18A1
ANO3
Unknown
VCAN
0
10
20
30
40
50
−10 −5 0 5 10
log2 fold-change
NS
log2 FC
P value
Total = 18,105 genes
b
1 3 4 5 6 7 10 11 12 13 14 15 16 17 18 192 8 9
−log10 P−log10 P
Chromosome
0
10
20
30
40
P value and log2 FC
a
Fig. 3 | Differential expression analyses between oviparous and viviparous uterine tissue during pregnancy. a, Differentially expressed genes (DEGs)
during the reproductive stages (pregnancy). Genes that show significant differential expression (adjusted P < 0.05) and a strong difference in expression
(log2-fold change threshold >1×) between the two parity modes are shown in red. Genes showing relatively higher expression in oviparous glandular
uterus have more negative log2-fold changes, while those overexpressed in viviparous glandular uterus have more positive log2-fold changes. Blue-coloured
genes show significant P values but relatively low differential expression (<1× log2-fold change), while beige-coloured genes show higher log2-fold changes
(>1×) but low P (>0.05). Grey-coloured genes show low differential expression (log2-fold change <1×; P > 0.05). b, DEGs during reproductive stages
across the genome. Candidate regions identified from admixture mapping are illustrated in red and differentially expressed genes (adjusted P < 0.05)
falling into these regions are also shown in red, with the gene density shown on the top in grey.
NATURE ECOLOGY & EVOLUTION | www.nature.com/natecolevol
Articles NATurE EcOlOgy & EVOluTION
Genomic regions showing a response to selection as inferred by
PCAdapt (q > 0.01) had high levels of differentiation (FST) between
parity modes, while nucleotide diversity within parity modes was
low in these regions (Extended Data Fig. 7). Significant correlation
of the selection scores with the divergence between parity modes
(Extended Data Fig. 8a), the high overall diversity (Extended Data
Fig. 8b) and the low nucleotide diversity within the parity modes
(Extended Data Figs. 8c,d) suggests that these regions are not only
under divergent selection between parity modes but also under
directional selection within parity modes, potentially because they
contain crucial genes controlling reproductive traits. While most
SNPs associated with gestation time in the admixture mapping
analyses had high selection scores and high FST and were fixed, or
nearly fixed, between parity modes (n = 60), another set of SNPs
(n = 23 SNPs) had lower selection scores and low FST (<0.5; Fig. 4a).
Most of these SNPs (total n = 19 out of 23 SNPs) were located on a
sex-linked region, chromosome 8 (Fig. 2 and Extended Data Fig. 7).
By contrast, SNPs associated with eggshell traits showed consis-
tently high FST values (>0.5; Fig. 4b). Because females are hetero-
gametic and therefore only carry a single copy of an allele on the
female sex chromosome, FST values of SNPs associated with parity
mode differ on these sex-associated regions compared to autosomal
regions46. Moreover, an autosome and the W sex chromosome have
fused in the evolutionary history of viviparous common lizards,
resulting in a larger W chromosome in some viviparous lineages
than in the ancestral oviparous lineage47. Sex-linked genes can lead
to rapid evolutionary change due to the restriction of recombina-
tion48. This prompts the interesting and open question of whether
sex chromosome evolution may be linked to parity mode evolution.
In summary, we found a large proportion of genes showing selec-
tion signals between the two parity modes, of which many were
associated with pathways in vertebrate reproduction as well as other
potential adaptive responses.
Shared genetic and developmental programmes of
viviparity across vertebrates
Having shown the functional genomic mechanisms of ancestral
oviparous and derived viviparous parity modes in common liz-
ards, we aimed to describe the extent to which these molecular
pathways are shared more broadly. Using available published data
(Supplementary Methods), we compiled two databases: (1) genes
differentially expressed in uterine tissue between squamate ovipa-
rous–viviparous sister lineages that diverged <30 Ma (refs. 20,49)
(n = 3 squamate genera) and (2) genes differentially expressed in
uterine tissue of viviparous vertebrates during pregnancy com-
pared to during non-pregnancy stages (n = 11, spanning five squa-
mate families/subfamilies, five orders of mammals (even-toed and
odd-toed ungulates, carnivores, marsupials and primates) and sea-
horse, in which embryos are incubated in the father’s pouch instead
of the mother’s uterine tract50,51). Across lineages, we explored the
overlap of genes that were differentially expressed in those published
studies. Gene symbols for the significantly differentially expressed
genes were extracted from each study and these gene symbol lists
were then compared. Due to the inherent experimental differences
across that breadth of studies (that is, variation in stages and tissues
analysed) and exclusion of novel lineage-specific variation, our
statistical analyses will be an underestimate and our results are
primarily descriptive and comparative.
We found that the genes differentially expressed across three
evolutionarily independent transitions from oviparity to viviparity
in squamates overlapped considerably. A total of 166 (out of 3,039)
genes were shared across two transitions; the overlap between com-
mon lizards and agamid lizards was high (n = 99 genes; in a sta-
tistical framework the exact t-test P < 0.001), while the other two
comparisons were not (Supplementary Table 10 and Fig. 5a). Even
though these are deeply divergent squamate lineages (with shared
origins ~180 Ma), we found that eight genes were shared across all
three oviparous–viviparous comparisons (exact t-test, P < 0.001;
Supplementary Table 10 and Fig. 5a).
Comparing pregnant and non-pregnant states for viviparous
squamates, we found that the DEG sets overlapped highly across
most species (22 out of 26 species comparisons had more overlapping
DEGs than expected by chance; Supplementary Table 11). Across
all squamates and the 9,057 genes involved, one gene (CDH5) was
shared between all five groups and another 31 genes were shared
across four groups (Supplementary Table 12). Sharing was higher in
our analysis focused on viviparous mammals, especially when
accounting for phylogenetic relatedness (Supplementary Data 4)
and all gene set comparisons between species overlapped more
than expected by chance (Extended Data Fig. 9 and Supplementary
Table 13). For example, out of the 8,509 DEGs we assessed, five genes
were shared across all five mammalian orders and 41 genes were
shared by four orders (Supplementary Table 14). These genes could
be considered a core developmental gene set for regulating uterine
tissue functionality and pregnancy in mammals; they are enriched
for the biological processes of the morphogenesis of a branching
structure, regulation of vascular development and urogenital sys-
tem development (Supplementary Table 15). The high degree of
sharing across viviparous mammals can be explained because vivi-
parity arose once in the common ancestor of therian mammals
~160 Ma (ref. 52). Conversely, the viviparous squamates included
here represent five independent transitions to viviparity from
0
3
6
9
12
0 0.25 0.50 0.75 1.00
0
3
6
9
12
0 0.25 0.50 0.75 1.00
Corrected F
ST
Selection score (–log10 P )Selection score (–log10 P)
a
b
Gestation time variants
Eggshell trait variants
Fig. 4 | Genome-wide analysis of selection score (-log10P from PCAdapt)
versus genetic differentiation (FST) between adult common lizards
spanning oviparity to viviparity (64,846 loci). a,b, Genetic variants
associated with reproductive traits from genetic mapping of gestation time
(a) and eggshell traits (b) are highlighted in red.
NATURE ECOLOGY & EVOLUTION | www.nature.com/natecolevol
Articles
NATurE EcOlOgy & EVOluTION
oviparous ancestors and occurring between 4 and 67 Ma (ref. 49).
As a consequence, the overlap we identified reveals marked con-
vergence in the developmental genes recruited for the evolution of
viviparity in squamates (Supplementary Table 11 and Fig. 5b).
We explored the gene set overlap more broadly and found sub-
stantial sharing of DEGs across a breadth of viviparous reproduc-
tion: mammals, squamates and seahorses. The ten largest gene sets
(those with the most shared genes) for pregnancy all consisted of
viviparous mammals and viviparous squamates (Fig. 5c). More
broadly, at least one squamate and one mammal were involved in
the gene sets significantly shared by at least three species (85 of the
132 (64.4%) possible combinations), revealing considerable shar-
ing across divergent amniote groups (Supplementary Data 5). In
all instances, the anamniote representative, seahorse, showed the
1,025
1,011
396
229227
131
99 85 64 60 52 48 42 36 27 17 15 13 10 654321 1
0
200
400
600
800
1,000
0 5025
362
2,599
6,602
458
2,150
Number of shared genes
P. entrecasteauxii
S. equalis
P. vlangalii
C. ocellatus
Z. vivipara
050100150200
P. entrecasteauxii
S. equalis
P. vlangalii
C. ocellat us
Z. vivipara
85
185
178
Time (Ma)
0
20
40
60
80
100
120
Number of shared genes
0 5025
311
2,150
458
6,602
2,599
362
2,152
5,095
342
1,439
2,310
0100200300400
M. domestica
H. sapiens
C. ocellat us
S. equalis
P. vlangalii
H. abdominalis
Z. vivipara
E. caballus
B. taurus
C. lupus
P. entrecasteauxii
79
185
312
435
77
97
159
78
178
85
M. domestica
H. sapiens
C. ocellatus
S. equalis
P. vlangalii
H. abdominalis
Z. vivipara
E. caballus
B. taurus
C. lupus
P. entrecasteauxii
Time (Ma)
b
c
99
68
15
8
0
20
40
60
80
100
Number of shared genes
–log
10
P–log
10
P
0 126
P. vlangalii
458
S. equalis
605
Z. vivipara
2,150
0
Z. vivipara
S. equalis
P. vlangalii
185
178
200 100
Time (Ma)
a
–log
10
P
79
Fig. 5 | Overlap between differentially expressed genes in pregnant and non-pregnant viviparous vertebrates. a, Overlap of differentially expressed
genes (DEGs) in reproductively bimodal squamate reptiles, with each lineage having a different viviparous ancestor. The pairwise overlaps between
S. equalis and the other two bimodal species are each not significant, while the Z. vivipara and P. vlangalii overlap and the overlaps between all three species
are significant. b, Overlap across squamate reptiles. All five species have different viviparous ancestors; that is, viviparity independently arose in all five
cases. Intersections across all groups are significant. c, Overlap across viviparous amniotes and a seahorse species, which is an anamniote and exhibits
a similar function of incubation or ‘pregnancy’, but in males. Only significant (P < 0.001) intersections with at least four species are shown and these are
sorted by significance, left to right. Sharing was extensive between all amniotes but less extensive with the seahorse (H. abdominalis). Divergence times
between species are derived from TimeTree52.
NATURE ECOLOGY & EVOLUTION | www.nature.com/natecolevol
Articles NATurE EcOlOgy & EVOluTION
lowest proportion of significant overlaps with gene lists from other
viviparous vertebrates (intersections that show significant gene list
overlaps across groups: mammals, 42.3% (588 out of 1391); squa-
mates, 40.0% (565 out of 1,414); seahorse, 15.2% (135 out of 888))
(Fig. 5c). Seahorses and amniotes do not share homologous repro-
ductive tissues to accommodate pregnancy; in seahorses, the male
develops a ‘brood pouch’ that performs functionally analogous tasks
to the uterine tissue in female amniotes50. Nonetheless, a relatively
high number of genes were also shared between seahorses and
amniotes (227 out of the 311 genes shared with at least one other
group; Supplementary Data 5). Overall, we find the strongest con-
vergence in pregnancy genes across amniotes and a lesser overlap
across vertebrates (Supplementary Data 4 and 5). Gene overlap was
generally higher if viviparous species shared a more recent com-
mon ancestor and if more parity mode transitions and species were
included in the analysis (Supplementary Table 16).
Discussion
Here we provided experimental evidence for the functional genetic
architecture of parity mode by leveraging natural hybridizations
and comparison of oviparous and viviparous lizards. Viviparity has
evolved from oviparity repeatedly across vertebrates53 but t he molec-
ular mechanisms and functional consequences have been challeng-
ing to infer. In this squamate model, several lines of evidence suggest
that eggshell characteristics are under simpler genetic control than
is gestation time. Because the different reproductive traits have dif-
ferent genetic architecture, we suggest that this may directly impact
the likelihood and mechanism of transitions from oviparity to vivi-
parity. Specifically, our findings suggest a strong polygenic genetic
basis for parity mode, compatible with the widespread unidirection-
ality of the transition to viviparity and the genetic and phenotypic
impediments to reversal10,54,55. Our results also suggest that the speed
of evolution may differ between gestation time and eggshell traits12;
eggshell has a simpler genetic basis that presumably enables more
rapid evolution. However, while gestation time is more complex, sex
chromosome linkage evolution may have facilitated and increased
the speed of evolutionary change for gestation time. Our results also
show that parity mode genes are under strong divergent selection.
The combination of strong selection and sex-linkage might there-
fore act to accelerate the evolution of viviparity and contribute to
a lack of intermediate forms in nature. Future work is required to
identify the minimum set of genetic differences determining parity
mode, to pinpoint the genetic basis of components of trait variation
within parity modes and to recapitulate its evolutionary steps. For
example, it would be intriguing to examine if the genetic architec-
ture of parity mode traits is similar in a different oviparous common
lizard lineage that presumably re-evolved oviparity15.
Our comparative assessment of vertebrate reproduction genes
suggests that the functional genetic architectures inferred from this
common lizard natural model are potentially broadly relevant. First,
there was high overlap between the genes that were consistently dif-
ferentially expressed in uterine tissues across divergent viviparous
vertebrates (two or more species) and the genes identified in our
common lizard genetic mapping experiment (208 out of 488 candi-
date genes from genetic mapping; overlapping more than expected
by chance). The highest proportion of gene sharing was found across
amniotes: the relative proportion of shared genes increased as we
compared more species that independently evolved viviparity, sug-
gesting that a core set of genes active in pregnancy is deeply shared
across amniotes (Supplementary Data 4 and Supplementary Table 16).
Of the 38 candidate genes that also showed overlapping differen-
tial expression across at least four viviparous vertebrates, the most
commonly shared genes were ASAH1, PLP1, SLC22A23 and EPAS1
(Supplementary Data 6 and 7). These genes act in transmembrane
transport (PLP1 and SLC22A23), cell differentiation (ASAH1) and
vascularization (EPAS1) and are involved in the establishment of
pregnancy, immune response and pre-eclampsia1,56,57. Cadherins
were found to be involved in the genetic basis of the eggshell traits
and showed molecular signals of response to selection and CDH5
was the only gene differentially expressed in all pregnant squa-
mates we compared; cadherins are calcium-dependent cell adhe-
sion proteins known to be key regulators of reproductive tissues58.
Other strongly implicated candidate genes extensively shared across
viviparous vertebrates included LYN, LYPLA1, LGMN and KANK3
(Supplementary Data 6 and 7).
Second, of the 51 annotated hub genes we found that differed
between oviparous and viviparous common lizards, 32 were differ-
entially expressed in at least two other viviparous vertebrates (1.95×
enrichment, P < 0.001). Three of these genes (RAPGEF2, RHOG
and KANK3) were also functional candidates that we inferred from
genetic mapping. RAPGEF genes are relevant for blood formation
in extra-embryonic membranes41,59 and each of the seven genes
in the family (RAPGEF1–6 and RAPGEFL1) was differentially
expressed during pregnancy in at least two viviparous species in our
comparative analysis of gene sets, indicating that this gene family is
crucial for live-bearing reproduction. RHOG is a gene involved in
cell migration42 that we found to be differentially expressed in four
viviparous squamates and mammals. KANK genes are key in verte-
brate vascular development44 and we speculate may play a role in the
well-vascularized uterine tissue required for viviparity in squamates
and mammals. Altogether, our results show that similar develop-
mental pathways controlling viviparous reproduction have evolved
or are involved repeatedly across vertebrates. The notable level of
convergence that we inferred suggests that pre-established regula-
tory networks are used7 and that these networks are often related
to immunity, tissue remodelling and blood vessel generation. Given
that fish, mammals and squamates have independently evolved
viviparity over the past ~450 Myr (ref. 60), this implies that similar
functional genes are recruited repeatedly during the evolution of
viviparity. To test whether similar genetic architectures have also
evolved repeatedly, the genetic basis of parity mode should be inves-
tigated in other amniotes and compared to the results reported here.
With more systems evaluated, this may also allow for the molecular
comparison of different types of viviparity, that is varying degrees
of placentation.
The evolution of key innovations, such as live-bearing reproduc-
tion, eyes, scales and feathers, offer unprecedented possibilities to
exploit new ecological niches and are associated with exceptional
diversification44. Viviparity is associated with increased specia-
tion rates, more complex parental behaviour and ecological niche
expansion1,2,45. Our study shows how this ecological and evolution-
ary innovation could be underpinned by a trait-specific genomic
architecture of retention time and eggshell characteristics, involving
particular and deeply shared genes and pathways, either retained
and re-used or convergently evolved. We speculate that other ori-
gins of viviparity and differences from oviparity will involve similar
key pathways, controlled by some common functional genes, coor-
dinated in a modular, trait-specific fashion.
Methods
Biological samples. Adult common lizards were sampled from April to
August during 2014–2017 in Carinthia, Austria15,21 under collection permit no.
HE3-NS-959/2013 (Supplementary Methods). Pregnant females were housed
individually in terraria (following ref. 61) and phenotyped (n = 480) for parity
mode: (1) number of external incubation days aer parition, (2) embryonic
stage at parition, (3) eggshell thickness and (4) calcium content in eggshells
(Supplementary Data 8; Extended Data Fig. 10). A small subset of reproductive
females was sampled for glandular uterus tissue, which was stored immediately in
RNAlater for subsequent RNA extraction (Supplementary Table 5 and Extended
Data Fig. 6). Clutches were weighed, staged and measured for hatching time
and success.
Eggshell/membrane thickness was measured by scanning electron microscopy.
Calcium composition of eggshell was measured by energy-dispersive X-ray
spectroscopy. Combining these, an ‘eggshell’ score was calculated for each female.
NATURE ECOLOGY & EVOLUTION | www.nature.com/natecolevol
Articles
NATurE EcOlOgy & EVOluTION
We calculated a ‘gestation time’ score for each female from the embryonic stage at
parition62 and the number of external incubation days.
Genomics. The ddRADSeq libraries were prepared with enzymes PstI and MspI
following ref. 15. Genome-referenced22 genotypes were inferred in STACKS v.1.44
(ref. 63) using draft genome version ‘JM_LG_L1_aftersplit’. Individual genomic
ancestry was inferred by ADMIXTURE v.1.3 (ref. 64) with two genetic clusters. For
each phenotyped female, the correlation coefficient with the genome-wide degree
of admixture (Q value) was estimated. A subset of hybrid, oviparous and viviparous
females was additionally selected for genotyping by whole-genome resequencing
(Supplementary Data 1).
Admixture mapping was performed in females with reproductive trait data and
ddRADSeq genotypes using GEMMA v.0.98 (ref. 65). First, a Bayesian sparse linear
mixed model was used to infer associations between genotypes and two phenotypic
scores: gestation time and eggshell characteristics. Second, a linear mixed model
was performed on all four reproductive phenotypes (external incubation days,
embryonic stage at parition, eggshell calcium, eggshell thickness) and the two
combined phenotype scores for gestation time and eggshell characteristics.
The whole-genome dataset was used to quantify regional and genome-wide
LD and its decay. We used the chromosome-specific half-life as linkage boundary
for each candidate SNP from admixture mapping to empirically determine
the regional candidate genes (Supplementary Data 9). We performed pathway
enrichment analyses using the PANTHER option in WebGestalt66 on this set of
genes, using the chicken genome RefSeq protein dataset (GCA_000002315.5) as a
reference.
Analyses of genome-wide divergence were based on the ddRADseq genotype
matrix between purely (Q > 99%) oviparous and viviparous individuals. Nucleotide
diversity was calculated in sliding windows of 50,000 sites67. Haplotype-based
FST values were extracted from STACKS. Outliers in nucleotide diversity were
identified by estimating the top 5% and lowest 5% quantiles across the genome of
each parity mode. FST outliers were identified as the top 5% quantiles. PCAdapt45
was used to identify candidate genomic loci under selection between parity
modes, with a threshold of q = 0.01. We identified the genes in linkage with outlier
loci using the LD block inference method from admixture mapping. We tested
for pathway enrichment of candidate genes and candidate SNPs (inferred by
association and selection) using PANTHER.
Gene expression. RNA-seq data from uterine tissue were aligned to the reference
genome22 using STAR v.2.5.2b (ref. 68) and normalized69. Count data were
log2-scaled using the rlog function in DESeq2. Then a principal component
analysis was conducted using the svd approach in the R package pcaMethods and a
redundancy analysis generated using the R package vegan. Reproductive stage was
specified as a ‘condition’.
To identify genes associated with parity mode, we performed separate
differential expression analyses with DESeq2 to identify sets of DEGs between
parity modes within and across stages and only considered genes that were
significantly differentially expressed (adjusted P = 0.01) during a reproductive
stage. Second, we used a Weighted Gene Co-Expression Network Analysis
(WGCNA) to identify modules of co-expressed genes associated with parity
mode70. Pearson’s correlations were calculated between module eigengenes
(the first principal component of the module’s expression profile) and trait
measurements (parity mode and reproductive stage). Positive correlations
represent upregulation in viviparous (versus oviparous) and negative correlations
represent upregulation in oviparous (versus viviparous). To identify key regulator
genes, we identified a set of hub genes for all modules significantly associated with
parity mode. Hub genes were selected on the basis of being within the top 10%
quantile of the module membership scores and gene significance scores.
We performed separate functional enrichment analyses on the
parity-associated gene sets identified using differential expression and WGCNA
analyses, using the PANTHER classification tool71. Gene Set Enrichment Analyses
(GSEAs)66 were used to identify significantly enriched pathways for DEG gene
sets using ranked expression scores. To identify pathways that were enriched
within parity-associated co-expression modules, we performed individual
Over-Representation Analyses (ORAs) for each module. The background gene set
for all GSEA and ORA analyses was specified as the full set of 21,187 genes in
the genome22.
Comparative analysis of reproduction genes. Using gene symbols from previous
studies, we assessed the overlap between DEGs in oviparous and viviparous
squamate uterine tissue from three lizard systems (Saiphos equalis10, Z. vivipara
and sister species in the genus Phrynocepahlus9) (Supplementary Table 17). We
assessed the overlap of the three gene symbol lists and if intersections exceeded
that expected by chance (on the basis of a total gene set of 20,000 genes typical
of vertebrates72) using the R package SuperExactTest73. Comparison of published
gene lists has the drawback that novel, unannotated genes cannot be identified
and that different studies use diverse methodological pipelines (Supplementary
Table 17). Here, we chose an inclusive approach to identify robust biological
pathways and key genes controlling viviparous pregnancy across developmental
stages and times. Transcriptomic datasets tend generally to show extreme variation
and this is challenging to control for in a multistudy comparison74. In the future,
an experimental approach that considers consistent timing, methodologies and
sequencing effort would be ideal to investigate the genetic basis of pregnancy
in a more controlled setting; for example, to identify genes that are expressed in
stage-specific ways.
To assess a breadth of vertebrates, we compared DEGs in uterine tissues of
pregnant viviparous species relative to their non-pregnant states (Supplementary
Table 17). This included: mammals Canis lupus (wolf), Equus caballus (horse),
Homo sapiens (human), Monodelphis domestica (opossum) and three closely
related even-toed ungulates (Bos taurus, Capra aegagrus and Sus scrofa)
combined into a single gene set; squamate bimodal S. equalis and Z. vivipara,
Phrynocepahlus vlangalii and two viviparous skinks Chalcides ocellatus and
Pseudemoia entrecasteauxii that evolved viviparity independently; and an
anamniote outgroup, the seahorse Hippocampus abdominalis. In contrast to
all viviparous amniotes, seahorses exhibit male ‘pregnancy’ or incubation of
embryos; therefore, the reproductive organs are not structurally homologous but
the pouch serves a function analogous to uterine tissue, including facilitating
exchange between parent and developing offspring50,51. For simplicity, we refer to
all viviparous amniotes and the seahorse exhibiting male pregnancy as viviparous
vertebrates, although seahorse pregnancy is not structurally homologous to
amniote viviparity50. The final dataset consisted of 11 gene sets (five mammalian,
five squamate and one fish) composed of 13 species (Supplementary Table 17)
and representing seven independent origins of viviparity across vertebrates53.
Intersections between gene sets were again assessed using SuperExactTest73 for:
(1) overlap of genes across mammalian gene sets, (2) overlap of genes across
viviparous squamate gene sets and (3) overlap of genes across all vertebrate viviparous
gene sets. Biological pathway enrichment analyses were performed in WebGestalt66
for the genes shared by more than four groups within mammals and squamates.
Reporting Summary. Further information on research design is available in the
Nature Research Reporting Summary linked to this article.
Data availability
The raw sequence data presented in this paper can be found on NCBI (NCBI
BioProject ID PRJNA657575; https://www.ncbi.nlm.nih.gov/bioproject/
PRJNA657575/). The genotype file and list of DEGs can be found at the University
of Glasgow Enlighten repository (https://doi.org/10.5525/gla.researchdata.1138)75.
Received: 9 October 2020; Accepted: 20 August 2021;
Published: xx xx xxxx
References
1. Van Dyke, J. U., Brandley, M. C. & ompson, M. B. e evolution of
viviparity: molecular and genomic data from squamate reptiles advance
understanding of live birth in amniotes. Reproduction 147, R15–R26 (2014).
2. Blackburn, D. G. Evolution of vertebrate viviparity and specializations for
fetal nutrition: a quantitative and qualitative analysis. J. Morphol. 276,
961–990 (2015).
3. Wright, A. M., Lyons, K. M., Brandley, M. C. & Hillis, D. M. Which came
rst: the lizard or the egg? Robustness in phylogenetic reconstruction of
ancestral states. J. Exp. Zool. B 324, 504–516 (2015).
4. Stewart, J. R. Fetal nutrition in lecithotrophic squamate reptiles: toward a
comprehensive model for evolution of viviparity and placentation. J. Morphol.
274, 824–843 (2013).
5. King, B. & Lee, M. S. Y. Ancestral state reconstruction, rate heterogeneity,
and the evolution of reptile viviparity. Syst. Biol. 64, 532–544 (2015).
6. Mossman, H. W. Comparative morphogenesis of the fetal membranes
and accessory uterine structures. Carnegie Inst. Contrib. Embryol. 26,
129–246 (1937).
7. Grith, O. W. & Wagner, G. P. e placenta as a model for understanding
the origin and evolution of vertebrate organs. Nat. Ecol. Evol. 1, 0072 (2017).
8. Brandley, M. C., Young, R. L., Warren, D. L., ompson, M. B. & Wagner, G. P.
Uterine gene expression in the live-bearing lizard, Chalcides ocellatus, reveals
convergence of squamate reptile and mammalian pregnancy mechanisms.
Genome Biol. Evol. 4, 394–411 (2012).
9. Gao, W. et al. Genomic and transcriptomic investigations of the evolutionary
transition from oviparity to viviparity. Proc. Natl Acad. Sci. USA 116,
3646–3655 (2019).
10. Foster, C. S. P., ompson, M. B., Van Dyke, J. U., Brandley, M. C. &
Whittington, C. M. Emergence of an evolutionary innovation: gene
expression dierences associated with the transition between oviparity and
viviparity. Mol. Ecol. 29, 1315–1327 (2020).
11. Wagner, G. P. & Zhang, J. e pleiotropic structure of the genotype–
phenotype map: the evolvability of complex organisms. Nat. Rev. Genet. 12,
204–213 (2011).
12. Guillette, L. J. e evolution of viviparity in lizards: ecological, anatomical
and physiological correlates lead to new hypothesis. Bioscience 43,
742–751 (1993).
NATURE ECOLOGY & EVOLUTION | www.nature.com/natecolevol
Articles NATurE EcOlOgy & EVOluTION
13. Moett, A. & Loke, C. Immunology of placentation in eutherian mammals.
Nat. Rev. Immunol. 6, 584–594 (2006).
14. Shine, R. Life-history evolution in reptiles. Annu. Rev. Ecol. Evol. Syst. 36,
23–46 (2005).
15. Recknagel, H., Kamenos, N. A. & Elmer, K. R. Common lizards break Dollo’s
law of irreversibility: genome-wide phylogenomics support a single origin of
viviparity and re-evolution of oviparity. Mol. Phylogenet. Evol. 127, 579–588
(2018).
16. Recknagel, H., Kamenos, N. A. & Elmer, K. R. Evolutionary origins of
viviparity consistent with palaeoclimate and lineage diversication. J. Evol.
Biol. 34, 1167–1176 (2021).
17. Stewart, J. R., Ecay, T. W. & Heulin, B. Calcium provision to oviparous and
viviparous embryos of the reproductively bimodal lizard Lacerta (Zootoca)
vivipara. J. Exp. Biol. 212, 2520–2524 (2009).
18. McLennan, D., Recknagel, H., Elmer, K. R. & Monaghan, P. Distinct telomere
dierences within a reproductively bimodal common lizard population.
Funct. Ecol. 33, 1917–1927 (2019).
19. Arrayago, M. J., Bea, A. & Heulin, B. Hybridization experiment between
oviparous and viviparous strains of Lacerta vivipara: a new insight into the
evolution of viviparity in reptiles. Herpetologica 52, 333–342 (1996).
20. Cornetti, L., Menegon, M., Giovine, G., Heulin, B. & Vernesi, C.
Mitochondrial and nuclear DNA survey of Zootoca vivipara across the
eastern Italian Alps: evolutionary relationships, historical demography and
conservation implications. PLoS ONE 9, e85912 (2014).
21. Lindtke, D., Mayer, W. & Böhme, W. Identication of a contact zone between
oviparous and viviparous common lizards (Zootoca vivipara) in central
Europe: reproductive strategies and natural hybridization. Salamandra 46,
73–82 (2010).
22. Yurchenko, A. A., Recknagel, H. & Elmer, K. R. Chromosome-level assembly
of the common lizard (Zootoca vivipara) genome. Genome Biol. Evol. 12,
1953–1960 (2020).
23. Heulin, B., Ghielmi, S., Vogrin, N., Surget-Groba, Y. & Guillaume, C. P.
Variation in eggshell characteristics and in intrauterine egg retention between
two oviparous clades of the lizard Lacerta vivipara: insight into the oviparity–
viviparity continuum in squamates. J. Morphol. 252, 255–262 (2002).
24. Júnior, G. A. O. et al. Genomic study and Medical Subject Headings
enrichment analysis of early pregnancy rate and antral follicle numbers in
Nelore heifers. J. Anim. Sci. 95, 4796–4812 (2017).
25. Zhang, L. et al. Expression of progesterone receptor membrane component 1
and its partner serpine 1 mRNA binding protein in uterine and placental
tissues of the mouse and human. Mol. Cell. Endocrinol. 287, 81–89 (2008).
26. Pru, J. K. & Clark, N. C. PGRMC1 and PGRMC2 in uterine physiology and
disease. Front. Neurosci. 7, 168 (2013).
27. Zachariades, E. et al. Changes in placental progesterone receptors in term and
preterm labour. Placenta 33, 367–372 (2012).
28. Xavier, F. Progesterone in the viviparous lizard Lacerta vivipara: ovarian
biosynthesis, plasma levels, and binding to transcortin-type protein during
the sexual cycle. Herpetologica 38, 62–70 (1982).
29. Zeberg, H., Kelso, J. & Pääbo, S. e Neandertal progesterone receptor.
Mol. Biol. Evol. 37, 2655–2660 (2020).
30. Fitzgerald, J. S., Toth, B., Jeschke, U., Schleussner, E. & Markert, U. R.
Knocking o the suppressors of cytokine signaling (SOCS): their roles in
mammalian pregnancy. J. Reprod. Immunol. 83, 117–123 (2009).
31. Shim, J. et al. Analysis of legumain and cystatin 6 expression at the
maternal–fetal interface in pigs. Mol. Reprod. Dev. 80, 570–580 (2013).
32. Ma, L. et al. in Advances in Applied Biotechnology (eds Zhang, T. &
Nakajima, M.) 173–178 (Springer, 2015).
33. Roa, J. & Tena-Sempere, M. Connecting metabolism and reproduction: roles
of central energy sensors and key molecular mediators. Mol. Cell. Endocrinol.
397, 4–14 (2014).
34. Zhao, Q. et al. Genome-wide association analysis reveals key genes responsible
for egg production of Lion Head goose. Front. Genet. 10, 1391 (2020).
35. Heulin, B. et al. Development of the uterine shell glands during the
preovulatory and early gestation periods in oviparous and viviparous Lacerta
vivipara. J. Morphol. 266, 80–93 (2005).
36. Stewart, J. R. Placental structure and nutritional provision to embryos in
predominantly lecithotrophic viviparous reptiles. Integr. Comp. Biol. 32,
303–312 (1992).
37. Stewart, J. R., Heulin, B. & Surget-Groba, Y. Extraembryonic membrane
development in a reproductively bimodal lizard, Lacerta (Zootoca) vivipara.
Zoology 107, 289–314 (2004).
38. Massagué, J. TGF-β signal transduction. Annu. Rev. Biochem. 67,
753–791 (1998).
39. Dubé, N. et al. e RapGEF PDZ-GEF2 is required for maturation of cell–cell
junctions. Cell. Signal. 20, 1608–1615 (2008).
40. Kusama, K. et al. Possible roles of the cAMP-mediators EPAC and RAP1 in
decidualization of rat uterus. Reproduction 147, 897–906 (2014).
41. Wei, P. et al. Defective vascular morphogenesis and mid-gestation embryonic
death in mice lacking RA-GEF-1. Biochem. Biophys. Res. Commun. 363,
106–112 (2007).
42. Katoh, H., Hiramoto, K. & Negishi, M. Activation of Rac1 by RhoG regulates
cell migration. J. Cell Sci. 119, 56–65 (2006).
43. Liu, Y., Lee, J. W. & Ackerman, S. L. Mutations in the microtubule-associated
protein 1A (Map1a) gene cause Purkinje cell degeneration. J. Neurosci. 35,
4587–4598 (2015).
44. Hensley, M. R. et al. Evolutionary and developmental analysis reveals KANK
genes were co-opted for vertebrate vascular development. Sci. Rep. 6, 27816
(2016).
45. Luu, K., Bazin, E. & Blum, M. G. B. pcadapt: an R package to perform
genome scans for selection based on principal component analysis. Mol. Ecol.
Resour. 17, 67–77 (2017).
46. Gammerdinger, W. J., Toups, M. A. & Vicoso, B. Disagreement in FST
estimators: a case study from sex chromosomes. Mol. Ecol. Resour. 20,
1517–1525 (2020).
47. Odierna, G., Aprea, G., Capriglione, T. & Puky, M. Chromosomal evidence
for the double origin of viviparity in the European common lizard, Lacerta
(Zootoca) vivipara. Herpetol. J. 14, 157–160 (2004).
48. Mank, J. E., Axelsson, E. & Ellegren, H. Fast-X on the Z: rapid evolution of
sex-linked genes in birds. Genome Res. 17, 618–624 (2007).
49. Zheng, Y. & Wiens, J. J. Combining phylogenomic and supermatrix
approaches, and a time-calibrated phylogeny for squamate reptiles (lizards
and snakes) based on 52 genes and 4162 species. Mol. Phylogenet. Evol. 94,
537–547 (2016).
50. Whittington, C. M. & Friesen, C. R. e evolution and physiology of male
pregnancy in syngnathid shes. Biol. Rev. 95, 1252–1272 (2020).
51. Blackburn, D. G. in Encyclopedia of Reproduction (ed. Skinner, M. K.)
443–449 (Academic Press, 2018).
52. Kumar, S., Stecher, G., Suleski, M. & Hedges, S. B. TimeTree: a resource
for timelines, timetrees, and divergence times. Mol. Biol. Evol. 34,
1812–1819 (2017).
53. Blackburn, D. G. Squamate reptiles as model organisms for the evolution of
viviparity. Herpetol. Monogr. 20, 131 (2006).
54. Grith, O. W. et al. Ancestral state reconstructions require biological
evidence to test evolutionary hypotheses: a case study examining the
evolution of reproductive mode in squamate reptiles. J. Exp. Zool. B 324,
493–503 (2015).
55. Fenwick, A. M., Greene, H. W. & Parkinson, C. L. e serpent and the egg:
unidirectional evolution of reproductive mode in vipers? J. Zool. Syst. Evol.
Res. 50, 59–66 (2012).
56. Chabrun, F. et al. Data-mining approach on transcriptomics and methylomics
placental analysis highlights genes in fetal growth restriction. Front. Genet.
10, 1292 (2020).
57. Wojcik-Baszko, D., Charkiewicz, K. & Laudanski, P. Role of dyslipidemia in
preeclampsia—a review of lipidomic analysis of blood, placenta,
syncytiotrophoblast microvesicles and umbilical cord artery from women
with preeclampsia. Prostaglandins Other Lipid Mediat. 139, 19–23 (2018).
58. Rowlands, T. M., Symonds, J. M., Farookhi, R. & Blaschuk, O. W. Cadherins:
crucial regulators of structure and function in reproductive tissues. Rev.
Reprod. 5, 53–61 (2000).
59. Satyanarayana, A. et al. RapGEF2 is essential for embryonic hematopoiesis
but dispensable for adult hematopoiesis. Blood 116, 2921–2931 (2010).
60. Irisarri, I. et al. Phylotranscriptomic consolidation of the jawed vertebrate
timetree. Nat. Ecol. Evol. 1, 1370–1378 (2017).
61. Recknagel, H. & Elmer, K. R. Dierential reproductive investment in
co-occurring oviparous and viviparous common lizards (Zootoca vivipara)
and implications for life-history trade-os with viviparity. Oecologia 190,
85–98 (2019).
62. Dufaure, J. P. & Hubert, J. Table de développement du lézard vivipare: Lacerta
(Zootoca) vivipara Jacquin. Arch. Anat. Microsc. Morphol. Exp. 50, 309–328
(1961).
63. Catchen, J. M., Amores, A., Hohenlohe, P., Cresko, W. & Postlethwait, J. H.
Stacks: building and genotyping loci de novo from short-read sequences.
Genes Genomes Genet. 1, 171–182 (2011).
64. Alexander, D. H., Novembre, J. & Lange, K. Fast model-based estimation of
ancestry in unrelated individuals. Genome Res. 19, 1655–1664 (2009).
65. Zhou, X. & Stephens, M. Genome-wide ecient mixed-model analysis for
association studies. Nat. Genet. 44, 821–824 (2012).
66. Wang, J., Vasaikar, S., Shi, Z., Greer, M. & Zhang, B. WebGestalt 2017: a more
comprehensive, powerful, exible and interactive gene set enrichment analysis
toolkit. Nucleic Acids Res. 45, W130–W137 (2017).
67. Martin, S. H., Davey, J. W. & Jiggins, C. D. Evaluating the use of ABBA–BABA
statistics to locate introgressed loci. Mol. Biol. Evol. 32, 244–257 (2015).
68. Dobin, A. et al. STAR: ultrafast universal RNA-seq aligner. Bioinformatics 29,
15–21 (2013).
69. Xiong, B., Yang, Y., Fineis, F. R. & Wang, J.-P. DegNorm: normalization of
generalized transcript degradation improves accuracy in RNA-seq analysis.
Genome Biol. 20, 75 (2019).
70. Langfelder, P. & Horvath, S. WGCNA: an R package for weighted correlation
network analysis. BMC Bioinf. 9, 559 (2008).
NATURE ECOLOGY & EVOLUTION | www.nature.com/natecolevol
Articles
NATurE EcOlOgy & EVOluTION
71. Mi, H., Muruganujan, A., Casagrande, J. T. & omas, P. D. Large-scale gene
function analysis with the PANTHER classication system. Nat. Protoc. 8,
1551–1566 (2013).
72. Hillier, L. W. et al. Sequence and comparative analysis of the chicken genome
provide unique perspectives on vertebrate evolution. Nature 423, 695–777
(2004).
73. Wang, M., Zhao, Y. & Zhang, B. Ecient test and visualization of multi-set
intersections. Sci. Rep. 5, 16923 (2015).
74. Eidem, H. R., Ackerman, W. E., McGary, K. L., Abbot, P. & Rokas, A.
Gestational tissue transcriptomics in term and preterm human pregnancies:
a systematic review and meta-analysis. BMC Med. Genomics 8, 27 (2015).
75. Recknagel, H. et al. Data from: e functional genetic architecture of
egg-laying and live-bearing reproduction in common lizards. University of
Glasgow https://doi.org/10.5525/gla.researchdata.1138 (2021).
Acknowledgements
We thank the late W. Mayer for his support and guidance in developing the project. For
technical assistance, we are grateful to A. Adams, M. Capstick, M. Mullin and J. Galbraith.
We thank K. Schneider for assistance with whole-genome resequencing libraries. We
thank H. Leitao, T. Caribe da Rocha, M. Mullin and J. Gallagher for help with eggshell
sample preparation and calcium content measurements. We thank R. Page for support
in project development, A. Jacobs for discussions and comments on data analysis and
T. Stevenson for comments on a draft. We thank G. Migiani for creating some illustrations
used in the figures. Finally, many thanks are due to the field assistants M. Andrews,
M. Capstick, R. Carey, K. Gallagher-Mackay, M. Lamorgese, N. Lawrie, M. Layton,
H. Leitão, J. McClelland, G. Migiani, M. Raske, J. Smout and M. Sutherland who helped
with sampling and husbandry. Sampling permission was issued by local authorities under
multiyear permit no. HE3-NS-959/2013. Research was funded by: a Genetics Society
Heredity Fieldwork grant to H.R.; a University of Glasgow Lord Kelvin/Adam Smith PhD
studentship to K.R.E., N.A.K. and H.R.; and NERC grants NE/ N003942/1 to K.R.E. and
M.M.B. with R.D.M. Page, NBAF964 to K.R.E. and NBAF1018 to K.R.E. and A.Y.
Author contributions
H.R., M.M.B., N.A.K. and K.R.E. designed and led the project. H.R. and
K.R.E carried out the fieldwork. H.R. conducted the phenotypic analyses,
generated and analysed the genomic data and genetic mapping, generated and
interpreted the transcriptome data, and compiled and analysed the comparative
analyses. M.N. staged the embryonic development. M.C. conducted transcriptome
bioinformatics. A.Y. assembled and annotated the reference genome and conducted
whole-genome bioinformatics. H.R. and M.M.B. conducted the eggshell microscopy.
H.R. and M.C. produced the figures. H.R. and K.R.E. wrote the manuscript with
contributions from M.C. and A.Y. All authors contributed to interpreting the results and
revising the manuscript.
Competing interests
The authors declare no competing interests.
Additional information
Extended data is available for this paper at https://doi.org/10.1038/s41559-021-01555-4.
Supplementary information The online version contains supplementary material
available at https://doi.org/10.1038/s41559-021-01555-4.
Correspondence and requests for materials should be addressed
to Kathryn R. Elmer.
Peer review information Nature Ecology & Evolution thanks Gunter Wagner and the
other, anonymous, reviewer(s) for their contribution to the peer review of this work.
Peer reviewer reports are available.
Reprints and permissions information is available at www.nature.com/reprints.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in
published maps and institutional affiliations.
© The Author(s), under exclusive licence to Springer Nature Limited 2021
NATURE ECOLOGY & EVOLUTION | www.nature.com/natecolevol
Articles NATurE EcOlOgy & EVOluTION
Articles NATurE EcOlOgy & EVOluTION
Extended Data Fig. 1 | Distribution of common lizards in Europe. a, The distribution of oviparous (dark brown) and viviparous (light brown) common
lizard lineages. b, The sampling area for this study, a contact zone in Austria with overlapping oviparous and viviparous common lizards, including hybrids
(Q-value 0.1–0.9). The map was drawn in R using Google Map data retrieved by API Key.
NATURE ECOLOGY & EVOLUTION | www.nature.com/natecolevol
Articles
NATurE EcOlOgy & EVOluTION Articles
NATurE EcOlOgy & EVOluTION
Extended Data Fig. 2 | Histograms of all reproductive phenotypes describing parity mode characteristics. a, The number of external incubation days was
measured for each clutch in the field, and b, one sample per clutch was taken to identify the embryonic stage at oviposition/parturition. c, From these two
phenotypes, a gestation time score was calculated, where 0 is more viviparous and 1 is more oviparous. To describe eggshell characteristics, d, calcium
content and e, eggshell thickness were measured for one eggshell per clutch. f, From these two phenotypes, an eggshell score was calculated for each
individual, where 0 is more viviparous and 1 is more oviparous.
NATURE ECOLOGY & EVOLUTION | www.nature.com/natecolevol
Articles NATurE EcOlOgy & EVOluTION
Articles NATurE EcOlOgy & EVOluTION
Extended Data Fig. 3 | Correlation between reproductive phenotypes and genome-wide ancestry. Correlations between reproductive traits and summary
scores for a–c, gestation time and d–f, eggshell characteristics are shown plotted against genome-wide Q-value for parity mode (0 = oviparous, 1 = viviparous)
inferred from ADMIXTURE.
NATURE ECOLOGY & EVOLUTION | www.nature.com/natecolevol
Articles
NATurE EcOlOgy & EVOluTION Articles
NATurE EcOlOgy & EVOluTION
Extended Data Fig. 4 | Decay of linkage disequilibrium (LD) in candidate regions for chromosomes containing SNPs significantly associated with
parity mode.
NATURE ECOLOGY & EVOLUTION | www.nature.com/natecolevol
Articles NATurE EcOlOgy & EVOluTION
Articles NATurE EcOlOgy & EVOluTION
Extended Data Fig. 5 | Genetic associations between reproductive phenotypes and 80,696 SNPs. Manhattan plots from admixture mapping analyses
on the four individually measured reproductive traits: a, incubation days, b, embryonic stage, c, calcium content, d, eggshell thickness. The genome-wide
significance is indicated by the horizontal red dotted line, while the blue line indicates a p-value threshold of <0.01 using the Benjamini–Hochberg
correction.
NATURE ECOLOGY & EVOLUTION | www.nature.com/natecolevol
Articles
NATurE EcOlOgy & EVOluTION Articles
NATurE EcOlOgy & EVOluTION
Extended Data Fig. 6 | RNA expression across 14,102 genes for oviparous and viviparous females at different reproductive stages in multi-dimensional
space. a, Schematic timeline of the reproductive season of oviparous and viviparous common lizards at the sampling location. Sampling times for RNA at
early, mid and after parition are indicated with stars. Sampling times reflect embryonic developmental stages (early: < stage 22; mid: stage 23–30). b, A
principal component analysis (PCA), and c, a redundancy analysis (RDA) from the RNAseq analysis. d, shows a Venn diagram of the overlap of differential
gene expression between the three stages ‘early’ in pregnancy, ‘mid’ pregnancy and ‘after’ pregnancy. Genes that were differentially expressed (DE)
either during ‘early’, ‘mid’ or both ‘early’ and ‘mid’ were considered as functionally important DE genes (noted with asterisks), excluding those that also
overlapped ‘after’ pregnancy.
NATURE ECOLOGY & EVOLUTION | www.nature.com/natecolevol
Articles NATurE EcOlOgy & EVOluTION
Articles NATurE EcOlOgy & EVOluTION
Extended Data Fig. 7 | Genome-wide selection scan and estimates of diversity and divergence. a, Out of a total of 80,696 SNPs, those highlighted in
blue and red (N = 1051 SNPs) showed significant signals of selection (Benjamini–Hochberg corrected p-value < 0.01). In addition, all SNPs in red (N = 128
SNPs) are those associated with reproductive traits via admixture mapping analyses. Nucleotide diversity estimates within b, oviparous and c, viviparous
common lizards. Smoothed means are shown in dark brown for oviparous and light brown for viviparous. d, shows FST across the genome. Candidate
regions are marked as red bars. Standard deviations of smoothed means are shown in light grey. e, Genome-wide difference in coverage between a male
and female common lizard, suggesting regions on chromosome 7 and 8 act as sex chromosomes where coverage in the female/heterogametic sex is lower.
NATURE ECOLOGY & EVOLUTION | www.nature.com/natecolevol
Articles
NATurE EcOlOgy & EVOluTION Articles
NATurE EcOlOgy & EVOluTION
Extended Data Fig. 8 | Correlation between selection scores (shown as negative Log10 P) and genetic differentiation and diversity measures. SNP
selection scores positively correlate with a, FST and with b, overall nucleotide diversity when all individuals are pooled. Within both c, oviparous and
d, viviparous parity modes, nucleotide diversity negatively correlates with selection scores.
NATURE ECOLOGY & EVOLUTION | www.nature.com/natecolevol
Articles NATurE EcOlOgy & EVOluTION
Articles NATurE EcOlOgy & EVOluTION
Extended Data Fig. 9 | Comparative assessment of functional genes. a, Overlap in differentially expressed genes between viviparous mammals during
pregnancy. Species comparisons in each histogram are shown in turquoise circles. A phylogenetic tree of the included viviparous mammals shows
estimated divergence times. Note that all species diverged from a common egg-laying ancestor around 160 million years ago. ***P < 0.001. Divergence
times were derived from TimeTree42. b, Correlation between unfiltered gene lists extracted from literature and gene lists filtered with the Database for
Annotation, Visualization and Integrated Discovery (DAVID). Gene lists filtered with DAVID were based on human and chicken gene symbols. While
some genes (on average 6.9%) are lost using the filtering, the fold enrichment of observed versus expected overlap of gene lists between all possible
intersections across viviparous vertebrates was highly correlated (R2 = 0.997).
NATURE ECOLOGY & EVOLUTION | www.nature.com/natecolevol
Articles
NATurE EcOlOgy & EVOluTION Articles
NATurE EcOlOgy & EVOluTION
Extended Data Fig. 10 | Correlation between reproductive phenotypes. Correlations between all reproductive phenotypes from females: combinations of
embryonic stage at parition, number of external incubation days from parition to hatching, eggshell thickness, and calcium content in eggshells.
NATURE ECOLOGY & EVOLUTION | www.nature.com/natecolevol
1
nature portfolio | reporting summary March 2021
Corresponding author(s): K. Elmer
Last updated by author(s): 2021-Aug-17
Reporting Summary
Nature Portfolio wishes to improve the reproducibility of the work that we publish. This form provides structure for consistency and transparency
in reporting. For further information on Nature Portfolio policies, see our Editorial Policies and the Editorial Policy Checklist.
Statistics
For all statistical analyses, confirm that the following items are present in the figure legend, table legend, main text, or Methods section.
n/a Confirmed
The exact sample size (n) for each experimental group/condition, given as a discrete number and unit of measurement
A statement on whether measurements were taken from distinct samples or whether the same sample was measured repeatedly
The statistical test(s) used AND whether they are one- or two-sided
Only common tests should be described solely by name; describe more complex techniques in the Methods section.
A description of all covariates tested
A description of any assumptions or corrections, such as tests of normality and adjustment for multiple comparisons
A full description of the statistical parameters including central tendency (e.g. means) or other basic estimates (e.g. regression coefficient)
AND variation (e.g. standard deviation) or associated estimates of uncertainty (e.g. confidence intervals)
For null hypothesis testing, the test statistic (e.g. F, t, r) with confidence intervals, effect sizes, degrees of freedom and P value noted
Give P values as exact values whenever suitable.
For Bayesian analysis, information on the choice of priors and Markov chain Monte Carlo settings
For hierarchical and complex designs, identification of the appropriate level for tests and full reporting of outcomes
Estimates of effect sizes (e.g. Cohen's d, Pearson's r), indicating how they were calculated
Our web collection on statistics for biologists contains articles on many of the points above.
Software and code
Policy information about availability of computer code
Data collection No software was used in data collection.
Data analysis All versions of published software and citations are provided in the Methods.
For manuscripts utilizing custom algorithms or software that are central to the research but not yet described in published literature, software must be made available to editors and
reviewers. We strongly encourage code deposition in a community repository (e.g. GitHub). See the Nature Portfolio guidelines for submitting code & software for further information.
Data
Policy information about availability of data
All manuscripts must include a data availability statement. This statement should provide the following information, where applicable:
- Accession codes, unique identifiers, or web links for publicly available datasets
- A description of any restrictions on data availability
- For clinical datasets or third party data, please ensure that the statement adheres to our policy
The raw sequence data presented in this paper can be found on NCBI (NCBI BioProject ID: SUB7928930 - PRJNA657575; BioSample ID: SUB7928962 -
SAMN15832663). The genotype file and list of differentially expressed genes can be found at the University of Glasgow Enlighten repository (http://
dx.doi.org/10.5525/gla.researchdata.1138).
2
nature portfolio | reporting summary March 2021
Field-specific reporting
Please select the one below that is the best fit for your research. If you are not sure, read the appropriate sections before making your selection.
Life sciences Behavioural & social sciences Ecological, evolutionary & environmental sciences
For a reference copy of the document with all sections, see nature.com/documents/nr-reporting-summary-flat.pdf
Ecological, evolutionary & environmental sciences study design
All studies must disclose on these points even when the disclosure is negative.
Study description Analysis of reproductive traits and their genetic basis in common lizards
Research sample Wild-caught Zootoca vivipara of both sexes. Most sampling non-lethal.
Sampling strategy For genetics, sample size based on encounters. For transcriptomics, based on biological variability and sampling permissions.
Data collection For male Z. vivipara: images, mass, SVL, and a tissue sample were taken in the field. Males were released immediately after.
For females: images, mass, SVL were taken on site. Pregnant females were taken to the field camp where they were housed
individually in tanks and kept until oviposition/parturition. All female reproductive data (clutch size, clutch mass, clutch survival, etc.)
was recorded collaboratively by the fieldwork team. All members were trained to record all data types. A tissue sample was taken
from each female, and both female and offspring were released to the (female's) point of capture. From each clutch, a single egg was
taken and the embryo removed for staging the development. In addition, the eggshell was used for analyses on calcium content and
eggshell thickness later in the laboratory. Eggshell measurements were done after training from experts in chemical analysis and
microscopy.
Timing and spatial scale Tissue samples and reproductive data: These were collected from 2014-2017. Sampling was repeated across years to increase
sampling size. Genetic data was generated in 2016 and 2017. Phenotype data was collected during time in the field, and eggshell
analyses were performed at Glasgow University 2018 and 2019 after all samples had been collected.
Data exclusions Individuals captured consecutively across years were combined where possible, or only one sample was included.
Reproducibility For eggshell data analyses a subset of samples were re-measured in order to quantify reproducibility (see Supplementary methods).
Randomization Samples were measured in a blind, randomized fashion to avoid observer bias.
All sequencing libraries contained individuals of both reproductive modes, while trying to balance them in equal proportions. Since
reproductive mode was not known a priori (e.g. for males), this was not possible to do 100%.
Blinding Blinding was used for eggshell measurements.
Did the study involve field work? Yes No
Field work, collection and transport
Field conditions Not recorded because not relevant.
Location Straniger Alm, Carinthia, Austria
Access & import/export Samples were collected with permission of the regional authorities (permit HE3-NS-959/2013)
Disturbance No disturbance was made.
Reporting for specific materials, systems and methods
We require information from authors about some types of materials, experimental systems and methods used in many studies. Here, indicate whether each material,
system or method listed is relevant to your study. If you are not sure if a list item applies to your research, read the appropriate section before selecting a response.
3
nature portfolio | reporting summary March 2021
Materials & experimental systems
n/a Involved in the study
Antibodies
Eukaryotic cell lines
Palaeontology and archaeology
Animals and other organisms
Human research participants
Clinical data
Dual use research of concern
Methods
n/a Involved in the study
ChIP-seq
Flow cytometry
MRI-based neuroimaging
Animals and other organisms
Policy information about studies involving animals; ARRIVE guidelines recommended for reporting animal research
Laboratory animals Did not involve laboratory organisms
Wild animals Male and female Zootoca vivipara ware caught in the field by hand. Females were housed in terraria until oviposition/ parturition and
then released with offspring at site of capture. Details are in the methods.
Field-collected samples NA
Ethics oversight No ethical oversight was required under Home Office legislation.
Note that full information on the approval of the study protocol must also be provided in the manuscript.
A preview of this full-text is provided by Springer Nature.
Content available from Nature Ecology & Evolution
This content is subject to copyright. Terms and conditions apply.