Journal of Heredity, 2024, 115, 11–18

https://doi.org/10.1093/jhered/esad067
Advance access publication 31 October 2023
Original Article

Original Article
Introduced house sparrows (Passer domesticus) have
greater variation in DNA methylation than native house
sparrows

Downloaded from https://academic.oup.com/jhered/article/115/1/11/7334204 by guest on 28 January 2025

M. Ellesse Lauer1,*, , Haley Kodak1,2, Tamer Albayrak3, , Marcos R. Lima4, , Daniella Ray1,
Emma Simpson-Wade1,5, David R. Tevs1, Elizabeth L. Sheldon6, , Lynn B. Martin6, and
Aaron W. Schrey1,
1
Department of Biology, Georgia Southern University, Statesboro and Savannah, GA 30458 and 31419, United States,
2
Department of Ecology and Evolutionary Biology, University of Tennessee, Knoxville, TN 37996, United States,
3
Department of Biology, Budur Mehmet Akif Ersoy University, Burdur, Turkey,
4
Laboratório de Ecologia Evolutiva e Conservação, Departamento de Biologia Animal e Vegetal, Centro de Ciências Biológicas, Universidade
Estadual de Londrina, Londrina, Paraná, Brazil,
5
Biomedical Science, University of Iowa, Iowa City, IA 52242, United States,
6
USF Global Health and Infectious Disease Research Center and USF Genomics Center, College of Public Health, University of South Florida,
Tampa, FL 33620, United States
*
Corresponding author: M. Ellesse Lauer, Department of Biology, Georgia Southern University, Statesboro and Savannah, Georgia, 30458 and 31419, USA.
Email: [email protected]
Corresponding Editor: Bridgett vonHoldt

Abstract
As a highly successful introduced species, house sparrows (Passer domesticus) respond rapidly to their new habitats, generating pheno-
typic patterns across their introduced range that resemble variation in native regions. Epigenetic mechanisms likely facilitate the success of
introduced house sparrows by aiding particular individuals to adjust their phenotypes plastically to novel conditions. Our objective here was to
investigate patterns of DNA methylation among populations of house sparrows at a broad geographic scale that included different introduction
histories: invading, established, and native. We defined the invading category as the locations with introductions less than 70 years ago and the
established category as the locations with greater than 70 years since introduction. We screened DNA methylation among individuals (n = 45)
by epiRADseq, expecting that variation in DNA methylation among individuals from invading populations would be higher when compared with
individuals from established and native populations. Invading house sparrows had the highest variance in DNA methylation of all three groups,
but established house sparrows also had higher variance than native ones. The highest number of differently methylated regions were detected
between invading and native populations of house sparrow. Additionally, DNA methylation was negatively correlated to time-since introduction,
which further suggests that DNA methylation had a role in the successful colonization’s of house sparrows.
Key words: epigenetic buffering, epigenetic potential, epiRADseq, phenotypic plasticity

Introduction the highly successful introduced house sparrows and may

The house sparrow (Passer domesticus) is a human-commensal be involved in the production of rapid phenotypic responses
songbird that has been introduced or expanded its range to challenges in the new environment (Schrey et al. 2012;
across all continents, except Antarctica, within the last 200 Liebl et al. 2013). House sparrows have phenotypic varia-
years (Anderson 2006; Liebl et al. 2015). Each location of in- tion among and within introduced populations (Johnston
troduction has a unique set of environmental conditions and and Selander 1973; Liebl and Martin 2012; Martin et al.
thus presents novel biotic and abiotic challenges, creating a 2017), that can also differ from native populations (Lima et
highly heterogeneous landscape to surmount (Carneiro and al. 2012). However, the rapidly generated latitudinal clines in
Lyko 2020), similar to populations of wild boar (Sus scrofa; phenotypes among house sparrows in the introduced range
both introduced and native), which differ in their food con- match those observed among native populations. Introduced
sumption among plant, animal, and fungi based on location populations of house sparrows are genetically differentiated
and season (Ballari and Barrios‐García 2014). DNA meth- from native populations and the most recently introduced
ylation has been associated to ecologically relevant traits in house sparrow population was the most different, having
lowest genetic diversity and the most private alleles (Schrey

Received June 22, 2023; Accepted October 27, 2023

© The Author(s) 2023. Published by Oxford University Press on behalf of The American Genetic Association. All rights reserved. For permissions, please
e-mail: [email protected]
12 Journal of Heredity, 2024, Vol. 115, No. 1

et al. 2011). These patterns strongly suggest 1) that plasticity differentiated than native ones (Schrey et al. 2011; Lima et
plays an important role in the development of these clines al. 2012; Sheldon et al. 2018a). The weight of evidence, so
and 2) that such plasticity is at least partially adaptive for far, suggests that plasticity via DNA methylation is associ-
introduced house sparrows. ated with phenotypic diversification in house sparrow colo-
Epigenetic mechanisms have been associated with the global nization of most new areas. The current study examines if
spread of introduced house sparrows, yet their functional this idea holds at a broader geographic scale than previously
role remains speculative and requires mechanistic testing. In tested.
introduced house sparrows, observed phenotypic variation The latent potential of an individual’s genome to have
among populations has happened so quickly that it is unlikely different epigenetic states, defined as epigenetic potential
to have arisen by rapid selection on standing genetic variation (Kilvitis et al. 2017), could also contribute to the success of
alone (Schrey et al. 2011; Liebl et al. 2015). DNA methylation introduced house sparrows. Epigenetic potential is variable
is the most well-studied epigenetic mechanism in vertebrates among species, individuals, and genes (Sheldon et al. 2023),
(Schrey et al. 2013), partly because it has been associated and individuals with more epigenetic potential are expected
with alterations in gene expression (Nätt et al. 2012) and to have greater capacity to use DNA methylation to fine-tune

Downloaded from https://academic.oup.com/jhered/article/115/1/11/7334204 by guest on 28 January 2025

thus might generate behavioral, physiological, and even or- their gene expression. In house sparrows, epigenetic potential
ganismal variation without changes to DNA sequence (Cubas varies among individuals and among genes within individuals
et al. 1999; Husby 2022). DNA methylation of the genome and is related to gene expression (Hanson et al. 2020a, 2021).
can happen more rapidly than DNA sequence can evolve, and Importantly, epigenetic potential is highest in house sparrows
moreover, it can be reversible and environmentally-induced from more recently colonized locations (Hanson et al. 2022),
(Jablonka and Lamb 1989, 1998; Hawes et al. 2018). As which suggests that in the most challenging locations, house
such, DNA methylation is likely important to introduced spe- sparrows benefit from high epigenetic potential. House
cies’ responses to their novel environments, as demonstrated sparrows with high epigenetic potential in the TLR4 pro-
in invasive ascidians (Didemnum vexillum; Hawes et al. moter are also more resistant to infection by a novel pathogen
2019), brown anoles (Anolis sagrei; Hu et al. 2019), cane (Salmonella enterica) than house sparrows with low epige-
toads (Rhinella marina; Sarma et al. 2021), pygmy mussels netic potential (Sheldon et al. 2023), which supports the idea
(Xenostrobus secures; Ardura et al. 2017), and Japanese that epigenetic potential is associated with at least one evolu-
knotweed (Fallopia japonica; Richards et al. 2012). Further, tionarily salient form of novelty in the house sparrow. Thus,
epigenetic diversity, measured as haplotype diversity (h), theoretically, epigenetic potential establishes the conditions
increased as genetic diversity, measured as observed heterozy- for plasticity, changes in DNA methylation states actualize the
gosity (Ho), decreased in recently introduced house sparrows plasticity, and plasticity facilitates the success of introduced
from Kenya (Liebl et al. 2013). Similar trends have been in house sparrows.
observed clonally reproducing Chrosomus eon-neogaeus Our objective was to assess how DNA methylation
from heterogeneous habitats (Massicotte et al. 2011) and varied among populations of house sparrows spanning
in an ancestrally bottlenecked population of freshwater four continents, including six introduced and two native
sticklebacks (Gasterosteus aculeatus; Ord et al 2023), which locations. We grouped populations of house sparrow into
suggest that greater epigenetic diversity can provide a source three introduction categories: invading, established, and
of variation to invading groups to facilitate their response to native, to account for the known differences in genetic
new environmental conditions. In house sparrows, there is a diversity and genetic differentiation among native and
significant negative correlation between DNA methylation at introduced populations (Schrey et al. 2011). We separated
transcription start sites and gene expression (Lundregan et the introduced house sparrows into invading and estab-
al. 2022). Also, DNA methylation in the putative promoter lished groups to account for the genetic differences detected
of Toll-like receptor 4 is negatively correlated with gene ex- among individuals in recently introduced areas (lowest ge-
pression (Kilvitis et al. 2019), suggesting that the expression netic diversity, most different genetic differentiation) and
of this immune response gene is at least partially affected by those from populations that have existed in the introduced
DNA methylation. These findings support the hypothesis that range for an extended period of time (genetic diversity
differences in DNA methylation can underlie changes in tran- lower than native, but not as low as invading; genetic dif-
scription, which may contribute to the generation of different ferentiation from invading and native; Schrey et al. 2011,
phenotypes between introduced and native house sparrows. 2012). We defined the invading category as the locations
Geographic patterns of variation in DNA methylation also with introductions less than 70 years ago, to account for
indicate that it is likely contributing to introduction success the largest genetic differences detected in the most recently
of house sparrows. House sparrows in more recently invaded introduced populations of house sparrows (Schrey et al.
locations have higher diversity in DNA methylation across 2011). We defined the established category as the locations
their genomes compared to individuals in more established with greater than 70 years since introduction, and native
locations (Liebl et al. 2013). Further, geographically and as birds from the native range. Previous studies of DNA
temporally separated introductions of house sparrows have methylation in house sparrows focused on single locations
different patterns of DNA methylation (Schrey et al. 2012; (Liebl et al. 2013; Kilvitis et al. 2019; Hanson et al. 2022;
Sheldon et al. 2018a), and DNA methylation varies within Siller 2022), comparisons between only two introductions
introduced populations (Liebl et al. 2013; Sheldon et al. (Schrey et al. 2012), or comparisons among introductions
2018a). The presence of variation in DNA methylation within on the same continent (Sheldon et al. 2018a). We expect
and among introduced house sparrow populations suggests that introduced populations of house sparrows will have
that it contributes to their post-introduction response. At greater variance in DNA methylation because they face the
the same time, introduced populations are less genetically most challenging environments.
Journal of Heredity, 2024, Vol. 115, No. 1 13

Methods 2005. In Kenya, a range of dates is provided for time of arrival,

as house sparrows originally were introduced to the port city
House sparrows screened
of Mombasa, then largely spread north-westward towards the
We screened DNA methylation in house sparrows (n = 45; border with Uganda: Mombasa 1950, Voi 1960, Nairobi 1990,
Table 1, Fig. 1) that were collected from two locations in Nakuru 2000, Garissa 2000, and Kakamega 2005 (Coon and
their native range, France (n = 4) and Turkey (n = 8), and six Martin 2014). The established locations were Florida, USA,
locations in their introduced range: British Columbia (n = 4), 1886 (Peña-Peniche et al. 2021), South Africa 1900 (Liebl et
Brazil (n = 3), Florida, USA (n = 4), Kenya (n = 14), Panama al. 2015), Brazil 1905 (Lima et al. 2012), and British Columbia
(n = 4), and South Africa (n = 4). We used the documented year 1915 (Hanson et al. 2020b). We extracted DNA from blood
of introduction to categorize introduced house sparrows as samples stored in 100% ethanol or dried on Whatman paper
invading (less than 70 years since introduction) or established using the DNeasy Kit (Qiagen, Valencia CA USA).
(more than 70 years since introduction). The invading locations
were Panama 1980 (Hanson et al. 2020b) and Kenya 1950 to
Next-generation sequencing

Downloaded from https://academic.oup.com/jhered/article/115/1/11/7334204 by guest on 28 January 2025

Table 1. House sparrow sample locations, grouped by introduction We used epiRADseq (Schield et al. 2016) to screen variation
category based on year of initial introduction (year) and those from the in DNA methylation among individuals on the Ion Torrent
native range, with sample sizes (n). PGM platform (Thermo Fisher Scientific, Waltham, MA).
epiRADseq is a ddRADseq protocol, developed for species
Location Year n Mean methylation without well-annotated genomes, that uses a DNA methyl-
(variance) ation sensitive restriction enzyme, HpaII, which fails to cut
when its CCGG restriction site is modified by DNA methyl-
Invading 0.864 (0.006)
ation at the internal CG. This generates a variable fragment
 Panama 1980 4 0.862 (0.0003)
library among individuals based on the DNA methylation
 Kenya 1950 to 2005 14 0.864 (0.0080) state of the HpaII restriction site. If the site is methylated, no
Established 0.913 (0.002) fragments are generated to be sequenced. Thus, variation in
 British Columbia 1915 4 0.864 (0.0004) DNA methylation is assayed as read count variation among
 Brazil 1905 3 0.870 (0.000005) individuals, which estimates the differences in DNA meth-
 South Africa 1900 4 0.949 (0.000006) ylation of the screened CCGG sites. Only variable sites are
 Tampa, FL, USA 1886 4 0.958 (0.00011)
meaningful to this analysis, as 100% methylated sites among
all individuals will not appear in the library and 0% meth-
Native 0.963 (0.0004)
ylated sites are expected to sequence in ratios predicted by
 France … 4 0.960 (0.0003)
the amount of sequence generated for each individual, thus
 Turkey … 8 0.964 (0.0004) not have different frequencies. epiRADseq generates data
in which a zero read count result for an individual is very
Mean methylation was estimated for each location. Variance in
methylation is included for both introduction category and sample meaningful. As such, we did not use cutoffs for differences
locations. in DNA methylation.

British Columbia
1915 France
Established Native
Turkey
Florida, USA Native
1886
Established

Panama Kenya
1980 Brazil 1950-2005
Invading 1905 Invading
Established South Africa
1900
Established

https://www.openstreetmap.org/#map=2/38.7/0.7

Fig. 1. House sparrow sampling locations, year of initial introduction, and defined introduction category (invading, established, native). Map data from
OpenStreetMap (© OpenStreetMap).
14 Journal of Heredity, 2024, Vol. 115, No. 1

The epiRADseq technique is a vast improvement on (invading, established, or native) to further understand how
MS-AFLP (Schrey et al. 2013), yet it maintains many of the populations of house sparrows respond to introduction with
benefits and limitations of MS-AFLP. The benefits are: it does DNA methylation. To characterize the pattern of change in
not require a reference genome, can be used among vastly dif- DNA methylation among introduction categories, we de-
ferent organisms, follows a simplified RNA-seq style analysis fined DMR that were shared in comparisons among mul-
approach, and is economical. The limitations are: it screens tiple categories, and DMR that were unique to a particular
anonymous CCGG sites, it focuses on sites that are variable comparison.
among the screened individuals, and it is not comparable to bi- For every house sparrow, we also calculated an estimate
sulfite sequencing-like approaches. epiRADseq generates data of total methylation, standardized by sequencing depth. We
in which zero read count result for an individual is very mean- divided the total number of binned reads by the total number
ingful, and therefore, we did not use cutoffs for differences of sequences observed for each individual, then subtracted
in methylation. Additionally, epiRADseq targets CCGG sites this ratio from one. We compared total methylation estimates
with variable DNA methylation among individuals. As such, among introduced and native birds, and by the different in-
it should be used to ask questions about variation in DNA troduction categories (i.e. invading, established, or native). We

Downloaded from https://academic.oup.com/jhered/article/115/1/11/7334204 by guest on 28 January 2025

methylation among experimental units, rather than specific then compared mean total methylation and variance in total
questions about the functional role of DNA methylation at methylation among populations and population categories
the molecular level (e.g. studies interested in the effect of using t-tests and f-tests, respectively. For introduced house
DNA methylation on gene expression, or characterizing DNA sparrows, we also tested for a relationship between total meth-
methylation patterns across genomic elements, etc.). ylation scores and the year of introduction among individuals
We followed a genotype-by-sequencing (GBS) protocol de- using Pearson’s correlations. Statistical tests of total methyl-
veloped for the Ion Torrent platform (Mascher et al. 2013), ation used alpha = 0.05 and were corrected by the sequential
substituting the DNA methylation sensitive restriction en- Bonferroni method when appropriate (Rice 1989).
zyme HpaII for MspI (New England Biolabs, Ipswich, MA)
to construct the epiRADseq library. After restriction diges-
tion, we ligated Ion Torrent IonXpress barcoded adaptors Results
and y-adapters. We ran emulsion polymerase chain reactions Screening DNA methylation using the epiRADseq method on
(PCR) following manufacturers protocols of the Ion PGM-Hi- the Ion Torrent PGM in house sparrows (n = 45) generated a
Q-View OT2-200 kit on the Ion Express OneTouch2 platform. pseudo-reference of 11,371,350 bases. The observed sequences
We sequenced resultant fragments following manufacturers represented 76 to 71,214 CCGG sites per individual.
protocols of the Ion PGM-Hi-Q-View Sequencing 200 Kit
using an Ion 316v2 BC Chips.
Introduction-based DMR
Thousands of differences in DNA methylation were present
Data analysis among all introduction categories, and the greatest DMRs
We demultiplexed runs and conducted quality control with were found between the introduction categories that were the
Torrent Suite version 4.4.3. We trimmed sequences to 150 bp. most different from each other (i.e. invading house sparrows
We performed a de novo assembly and constructed a pseudo- and native house sparrows). There were 8,734 DMR between
reference using Geneious Prime v. 2022.1.1 (Dotmatics). We introduced and native house sparrow groups. Further, there
mapped individual sequences with BWA Galaxy Version were 19,874 DMR among all birds after separating house
0.7.17.4 (Li and Durbin 2009, 2010). We used featureCounts sparrows into invading, established, and native categories (Fig.
Galaxy Version 1.6.4 + galaxy1 (Liao et al. 2014) to deter- 2). The highest number of DMR, 11,538, occurred between
mine read counts of fragments within 150 bp bins spanning invading and native group; 4,226 DMR occurred between
the pseudo-reference. The 150 bp bins were used to count invading and established house sparrows; and 4,110 DMR
fragments among individuals ultimately to represent variation occurred between established and native house sparrows
in DNA methylation among the CCGG sites screened. For a (Fig. 2). There were shared and unique DMR across all
fragment to be sequenced, it had to have a non-methylated combinations of comparisons among introduction categories
CCGG site. Counting matches to the bins across the pseudo- (Fig. 2). Invading house sparrows had the most DMR and the
reference equates to variation in DNA methylation among most unique DMR (Fig. 2). Yet, DMR were shared among all
the CCGG sites. As epiRADseq generates data with the zero comparisons.
read count result indicating DNA methylation, we used two
approaches to control for sequencing coverage differences.
First, we only analyzed individuals with 5,838 sequencing Introduction-based total methylation
reads or higher, and second, we only analyzed results from DNA methylation was related to introduction category (Fig.
the first 20,000 bins, and removed the other bins, as they had 3); introduced populations of house sparrows had a lower
progressively lower coverage. mean total methylation than those from the native range
We used edgeR, Galaxy Version 3.24.1 + galaxy1 (Robinson (t-test P = 0.00028; Table 1). In other words, a given CCGG
et al. 2010), to detect differently methylated regions (DMR), site in introduced birds was less likely to be methylated than
which indicate specific regions of the genome that had signifi- a given CCGG site in native birds. Mean total methylation
cantly different levels of DNA methylation among individuals, was also lower in invading (mean total methylation = 0.864;
with a False Discovery Rate (FDR) of 0.05. First, we tested Table 1) than established house sparrows (mean total meth-
for the presence of DMRs among individuals between all ylation = 0.913; t-test invading vs. established P = 0.020),
introduced versus native birds. We then tested for DMR lower in invading than native house sparrows (mean total
with edgeR in an analysis based on introduction categories methylation = 0.963; t-test invading vs. native P = 0.0001),
Journal of Heredity, 2024, Vol. 115, No. 1 15

Discussion
Introduction history explained the patterns of DNA meth-
ylation among populations of house sparrows across mul-
tiple continents. Invading populations had the lowest mean
Invading to Native total methylation, and highest variance in total methylation.
11538
Native 11538 Variance in DNA methylation also decreased with time-since
introduction in non-native populations. These findings signif-
icantly extend those of Schrey et al. (2012), which detected
differences in DNA methylation between population of house
3206 3415
sparrow from Tampa and Kenya, and those of Sheldon et
45 al. (2018a), which found epigenetic differentiation among
populations of house sparrow from separate introductions
Invading to Established to into Australia. Also, the higher variance in DNA methyla-
Established 256 Native tion among individuals in more recently introduced locations

Downloaded from https://academic.oup.com/jhered/article/115/1/11/7334204 by guest on 28 January 2025

4226
• 4226 4110 corresponds with increased epigenetic potential found in
• 4110 more recently introduced populations. Specifically, Liebl et
al. (2013) found that epigenetic diversity compensated for
decreased genetic diversity among populations of introduced
house sparrow in Kenya. Integrating our findings of greater
variance in DNA methylation among populations of house
Fig. 2. Venn diagram of the number of differently methylated regions, sparrows from more recently introduced locations with the
indicating different CCGG restriction sites, and the pattern of sharing
findings of higher epigenetic potential in those from more
among house sparrows grouped by comparisons among introduction
categories. Statistical significance for all differently methylated regions
recently introduced locations (Hanson et al. 2021, 2022)
was determined by an FDR of 0.05. suggests that house sparrows with more epigenetic potential
manifest greater diversity in epigenetic states, which might re-
flect the utility of plasticity for adjusting to novel challenges
in new areas.
The presence of both shared and unique changes in DNA
methylation among introduction categories indicates that
DNA methylation in house sparrows responds to introduc-
tion in at least three fundamental ways: one being a consistent
response to introduction, the second being location specific,
and the third being hypervariable. The consistent response
across categories suggests that house sparrows have a set of
loci that are consistently methylated differently in response to
introduction. For example, introduction likely presents a con-
sistent set of challenges that must be overcome in all cases. A
loose corollary situation may be seen in the rapid response to
domestication in chickens (Gallus gallus), which is facilitated
by DNA methylation (Nätt et al. 2012). The location specific
Fig. 3. Variance in DNA methylation is higher in the introduced house response (i.e. unique DNA methylations across populations)
sparrows (invading and established) compared to the native house
suggests that house sparrows methylate different loci in re-
sparrows. Variance among all categories was significant at alpha = 0.05.
sponse to the challenges associated with introduction into
different areas. For example, house sparrows in Kenya man-
and lower in established than native house sparrows (t-test ifest different patterns of DNA methylation compared to
P = 0.001). Finally, individual-level total methylation those from the United States or Australia (Schrey et al. 2012;
estimates were correlated to year of introduction (including Liebl et al. 2013; Sheldon et al. 2018a). The hypervariable
natives: r = −0.505, P = 0.0002, n = 45; excluding natives: response (i.e. the middle of the Venn diagram) indicates that
r = −0.614, P = 0.0001, n = 33). some loci vary in DNA methylation state among individuals
Variance in total methylation decreased with increasing across all categories. Also, we note that it is likely that DNA
time-since introduction in the introduced house sparrows, methylation responds to habitat variation at a finer scale than
with native house sparrows showing the lowest variance in was captured by our sampling design. For example, these loci
total methylation (Table 2). Variance in total methylation could respond to specific stimuli such as differences in temper-
ranged from 0.0039 to 62.96, with house sparrows from ature (Sheldon et al. 2020), nest habitat conditions (vonHoldt
introduced populations having higher variance in total meth- et al. 2023), brood size (Baker 1995; Sheldon et al. 2018b),
ylation than native house sparrows (f-test P = 0.00007; Table and reproductive behavior (Liebl et al. 2021).
1). Invading house sparrows had the highest variance in total The pattern of more recently introduced populations of
methylation (0.006) and native house sparrows had the lowest house sparrows having the highest variance in DNA meth-
variance in total methylation (0.0004; Table 1). Invading ylation may be explained by epigenetic buffering. Epigenetic
(f-test P = 0.00003) and established (f-test P = 0.005) house buffering is a phenomenon whereby a group of individuals
sparrows had higher variance in total methylation than native endures challenging conditions and avoids extirpation be-
house sparrows. cause individual organisms somehow manifest high epigenetic
 16 Journal of Heredity, 2024, Vol. 115, No. 1

variation (O’Dea et al. 2016; Hawes et al. 2018). For example,

0.0301
0.0815
0.0237
0.5758
0.0011
0.0219
0.0304
Brazil great tits (Parus major) in urban environments had more

…
variation in DNA methylation compared to those in rural

The P-values for the t-tests are presented below the diagonal and those for the f-tests are presented above the diagonal. Values in bold indicate statistical significance after sequential Bonferroni correction.
environments (Watson et al. 2021). Also, invasive mussels
(Xenostrobus securis) altered DNA methylation in response
to higher stress environments (Ardura et al. 2018). If epige-
netic buffering is important, variance in DNA methylation
Panama

0.2526
0.4248
0.8482
0.0253
0.0214
0.8627
0.9934

should be higher in more resource heterogeneous, or stressful,

…
locations (O’Dea et al. 2016). The reduction in variance of
total DNA methylation for the established house sparrows
may indicate directional selection occurs on this state (Gould
1988), which would suggest that the established populations
have existed in their introduced locations long enough for se-
Florida, USA

lection to specialize individuals to their new location and sta-

Downloaded from https://academic.oup.com/jhered/article/115/1/11/7334204 by guest on 28 January 2025

<0.0001
<0.0001
0.3285

bilize DNA methylation on the states conferring higher fitness.

0.1037
0.0048
0.2978
0.4293

Interestingly, higher epigenetic potential among individuals

…

from recently introduced areas (Hanson et al. 2020a, 2022)

would facilitate epigenetic buffering at the population level.
Here, epigenetic potential provides the necessary conditions
for epigenetic regulation of phenotypic plasticity and pheno-
typic accommodation (Baldwin 1896; Ghalambor et al. 2007;
British Columbia

Foster et al. 2015). DNA methylation would actualize epi-

genetic potential into a particular and presumably adaptive
plastic response. Thus, having more CpG sites increases the
0.3426
0.4300
<0.0001
0.0179
0.0303
0.9570
0.8418

potential for the gene of an individual to be regulated vari-

ably by DNA methylation; when a given gene is methylated
…

epigenetic potential becomes actualized into a particular phe-

notypic state. Ultimately, at the population level, higher epige-
netic potential in individual organisms would generate more
phenotypic variation among organisms, buffering the popula-
South Africa

tion against highly heterogeneous environments expected of

new areas.
<0.0001
<0.0001
0.0675
<0.0001
0.0002
0.0145
0.0257

In conclusion, we found more variance in DNA methyl-

…

ation in the most recently introduced populations of house

sparrows, possibly as evidence of epigenetic buffering via
selection on epigenetic potential. These patterns may sug-
Table 2. Comparisons of mean and variance in total methylation among house sparrows.

gest that introduced species are using DNA methylation to

fine-tune gene expression in their new environments. Here,
0.4619
0.4798
0.0281
0.4993
0.04140
0.0007
0.0211

we used anonymous markers to characterize the pattern of

Kenya

DNA methylation in house sparrows across the broadest

…

scale studied to date. We note that future studies would

benefit from identifying the DNA methylation patterns and
epigenetic potential in specific genes. We propose that ep-
igenetic buffering extends beyond DNA methylation to in-
<0.0001
0.0175
0.3200
<0.0001
0.0947
0.0030
0.8552

clude the other epigenetic mechanisms, in particular histone

Turkey

modifications. We also suggest that DNA methylation-based

…

epigenetic buffering is active in other introduced species

and in other cases of responses of species to heterogeneous
environments.
0.0002
0.0001
0.4577
0.0002
0.1369
0.0.0271
0.3664

Acknowledgments
France

Samples were collected with local permission and imported

…

with appropriate USDA permission and USF IACUC author-

ization. We thank Col. T. Brooks for Excel tutorials and N.
Schrey, K. Russell, and O. Ige for lab assistance.
British Columbia

Funding
Florida, USA
South Africa

This work was supported by the National Science Foundation

(NSF IOS 2027054 to L.B.M. and A.W.S.). We also thank
Panama
Turkey
France

Kenya

Brazil

Georgia Southern University’s Graduate Student Organization

Grants Program and the Jim Spence Ornithology scholarship
Journal of Heredity, 2024, Vol. 115, No. 1 17

for internal funding. The authors declare no conflicts of in- Jablonka E, Lamb MJ. The inheritance of acquired epigenetic varia-
terest. tions. J Theor Biol. 1989:139:69–83.
Jablonka E, Lamb MJ. Epigenetic inheritance in evolution. J Evol Biol.
1998:11:159–183.
Data Availability Johnston RF, Selander RK. Evolution in the house sparrow III Variation
in size and sexual dimorphism in Europe and North and South A-
We have deposited the raw data underlying these analyses to merica. Am Nat. 1973:107:373–390.
the SRA database (BioProject accession PRJNA1020895). Kilvitis HJ, Hanson H, Schrey AW, Martin LB. Epigenetic potential as a
mechanism of phenotypic plasticity in vertebrate range expansions.
Integr Comp Biol. 2017:57:385–395.
References Kilvitis HJ, Schrey AW, Ragsdale AK, Berrio A, Phelps SM, Martin
Anderson TR. Biology of the ubiquitous house sparrow: from genes LB. DNA methylation predicts immune gene expression in
to populations. New York, NY, USA: Oxford University Press; introduced house sparrows Passer domesticus. J Avian Biol.
2006. 2019:50:e01965.
Ardura A, Clusa L, Zaiko A, Garcia-Vazquez E, Miralles L. Stress re- Li H, Durbin R. Fast and accurate short read alignment with Burrows–
lated epigenetic changes may explain opportunistic success in bio- Wheeler transform. Bioinformatics. 2009:25:1754–1760.

Downloaded from https://academic.oup.com/jhered/article/115/1/11/7334204 by guest on 28 January 2025

logical invasions in Antipode mussels. Sci Rep. 2018:8:10793. Li H, Durbin R. Fast and accurate long-read alignment with Burrows–
Ardura A, Zaiko A, Morán P, Planes S, Garcia-Vazquez E. Epigenetic Wheeler transform. Bioinformatics. 2010:26:589–595.
signatures of invasive status in populations of marine invertebrates. Liao Y, Smyth GK, Shi W. featureCounts: an efficient general purpose
Sci Rep. 2017:7:42193. program for assigning sequence reads to genomic features. Bioin-
Baker M. Environmental component of latitudinal clutch-size variation formatics. 2014:30:923–930.
in house sparrows (Passer domesticus). Auk 1995:112:249–252. Liebl AL, Martin LB. Exploratory behaviour and stressor hyper-
Baldwin JM. A new factor in evolution. Am Nat. 1896:30:441–451. responsiveness facilitate range expansion of an introduced song-
Ballari SA, Barrios-García MN. A review of wild boar Sus scrofa diet bird. Proc Biol Sci. 2012:279:4375–4381.
and factors affecting food selection in native and introduced ranges. Liebl AL, Schrey AW, Andrew SC, Sheldon EL, Griffith SC. Invasion
Mamm Rev. 2014:44:124–134. genetics: lessons from a ubiquitous bird, the house sparrow Passer
Carneiro VC, Lyko F. Rapid epigenetic adaptation in animals and its domesticus. Curr Zool. 2015:61:465–476.
role in invasiveness. Integr Comp Biol. 2020:60:267–274. Liebl AL, Schrey AW, Richards CL, Martin LB. Patterns of DNA meth-
Coon CA, Martin LB. Patterns of haemosporidian prevalence along ylation throughout a range expansion of an introduced songbird.
a range expansion in introduced Kenyan house sparrows Passer Integr Comp Biol. 2013:53:351–358.
domesticus. J Avian Biol. 2014:45:34–42. Liebl AL, Wesner JS, Russell AF, Schrey AW. Methylation patterns at
Cubas P, Vincent C, Coen E. An epigenetic mutation responsible for fledging predict delayed dispersal in a cooperatively breeding bird.
natural variation in floral symmetry. Nature. 1999:401:157–161. PLoS One. 2021:16:e0252227.
Foster SA, Wund MA, Baker JA. Evolutionary influences of plastic be- Lima MR, Macedo RH, Martins TL, Schrey AW, Martin LB, Bensch
havioral responses upon environmental challenges in an adaptive S. Genetic and morphometric divergence of an invasive bird: the
radiation. Integr Comp Biol. 2015:55:406–417. introduced house sparrow (Passer domesticus) in Brazil. PLoS One.
Ghalambor CK, McKay JK, Carroll SP, Reznick DN. Adaptive versus 2012:7:e53332.
non-adaptive phenotypic plasticity and the potential for contem- Lundregan SL, Makinen H, Viitaniemi H, Ronning B, Jensen H, Husby
porary adaptation in new environments. Funct Ecol. 2007:21:394– A. 2022. DNA methylation regulates sex-biased gene expression in
407. the house sparrow. Ecol Evol. 12:e9539.
Gould SJ. Trends as changes in variance: a new slant on progress and Martin LB, Kilvitis HJ, Brace AJ, Cooper L, Haussmann MF, Mutati A,
directionality in evolution. J Paleontol. 1988:62:319–329. Fasanello V, O’Brien S, Ardia DR. Costs of immunity and their role
Hanson HE, Koussayer B, Kilvitis HJ, Schrey AW, Maddox JD, Martin in the range expansion of the house sparrow in Kenya. J Exp Biol.
LB. Epigenetic potential in native and introduced populations 2017:220:2228–2235.
of house sparrows (Passer domesticus). Integr Comp Biol. Mascher M, Wu S, Amand PS, Stein N, Poland J. Application of
2020a:60:1458–1468. genotyping-by-sequencing on semiconductor sequencing platforms:
Hanson HE, Zolik JE, Martin LB. House sparrow (Passer domesticus a comparison of genetic and reference-based marker ordering in
Linnaeus, 1758). In: Downs CT, Hart LA, editors. Invasive birds: barley. PLoS One. 2013:8:e76925.
global trends and impacts. Wallingford (UK): CABI; 2020b. p. Massicotte R, Whitelaw E, Angers B. DNA methylation: a source of
85–96. random variation in natural populations. Epigenetics. 2011:6:421–
Hanson HE, Wang C, Schrey AW, Liebl AL, Ravinet M, Jiang RH, 427.
Martin LB. Epigenetic potential and DNA methylation in an ongo- Nätt D, Rubin CJ, Wright D, Johnsson M, Beltéky J, Andersson L,
ing house sparrow (Passer domesticus) range expansion. Am Natu- Jensen P. Heritable genome-wide variation of gene expression and
ralist. 2022:200:662–674. promoter methylation between wild and domesticated chickens.
Hanson HE, Zimmer C, Koussayer B, Schrey AW, Maddox JD, Martin BMC Genom. 2012:13:1–12.
LB. Epigenetic potential affects immune gene expression in house O’Dea RE, Noble DW, Johnson SL, Hesselson D, Nakagawa S. The
sparrows. J Exp Biol. 2021:224:jeb238451. role of non-genetic inheritance in evolutionary rescue: epigenetic
Hawes NA, Amadoru A, Tremblay LA, Pochon X, Dunphy B, Fidler buffering, heritable bet hedging and epigenetic traps. Environ
AE, Smith KF. Epigenetic patterns associated with an ascidian in- Epigenet. 2016:2:dvv014.
vasion: a comparison of closely related clades in their native and Ord J, Gossman TI, Adrian-Kalchhauser I. High nucleotide diversity
introduced ranges. Sci Rep. 2019:9:14275. accompanies differential DNA methylation in naturally diverging
Hawes NA, Fidler AE, Tremblay LA, Pochon X, Dunphy BJ, Smith KF. populations. Mol Biol Evol. 2023:40:msad068.
Understanding the role of DNA methylation in successful biologi- Peña-Peniche A, Mota-Vargas C, García-Arroyo M, MacGregor-Fors
cal invasions: a review. Biol Invasions. 2018:20:2285–2300. I. On the North American invasion of the House Sparrow and its
Hu J, Askary AM, Thurman TJ, Spiller DA, Palmer TM, Pringle absence in the Yucatan Peninsula. Avian Conserv Ecol. 2021:16:18.
RM, Barrett RD. The epigenetic signature of colonizing new Rice WR. Analyzing tables of statistical tests. Evolution. 1989:43:223–
environments in Anolis lizards. Mol Biol Evol. 2019:36:2165– 225.
2170. Richards CL, Schrey AW, Pigliucci M. Invasion of diverse habitats by
Husby A. Wild epigenetics: insights from epigenetic studies on natural few Japanese knotweed genotypes is correlated with epigenetic dif-
populations. Proc Biol Sci. 2022:289:20211633. ferentiation. Ecol Lett. 2012:15:1016–1025.
18 Journal of Heredity, 2024, Vol. 115, No. 1

Robinson MD, McCarthy DJ, Smyth GK. edgeR: a Bioconductor pack- songbirds against Salmonella enterica infection. J Exp Biol.
age for differential expression analysis of digital gene expression 2023:226:jeb245475.
data. Bioinformatics. 2010:26:139–140. Sheldon EL, Schrey A, Andrew SC, Ragsdale A, Griffith SC. Epigenetic
Sarma RR, Crossland MR, Eyck HJF, DeVore JL, Edwards RJ, and genetic variation among three separate introductions of the
Cocomazzo M, Zhou J, Brown GP, Shine R, Rollins LA. Intergener- house sparrow (Passer domesticus) into Australia. R Soc Open Sci.
ational effects of manipulating DNA methylation in the early life of 2018a:5:172185.
an iconic invader. Philos Trans R Soc. 2021:376:20200125. Sheldon EL, Schrey AW, Ragsdale AK, Griffith SC. Brood size influences
Schield DR, Walsh MR, Card DC, Andrew AL, Adams RH, Castoe TA. patterns of DNA methylation in wild Zebra Finches (Taeniopygia
Epi RAD seq: scalable analysis of genomewide patterns of meth- guttata). Auk. 2018b:135:1113–1122.
ylation using next-generation sequencing. Methods Ecol Evol. Sheldon EL, Schrey AW, Hurley LL, Griffith SC. Dynamic changes in
2016:7:60–69. DNA methylation during postnatal development in zebra finches
Schrey AW, Alvarez M, Foust CM, Kilvitis HJ, Lee JD, Liebl AL, Martin Taeniopygia guttata exposed to different temperatures. J Avian
LB, Richards CL, Robertson M. Ecological epigenetics: beyond Biol. 2020:51:e02294.
MS-AFLP. Integr Comp Biol. 2013:53:340–350. Sheldon EL, Schrey AW, Lauer ME, Martin LB. Epigenetic potential: pro-
Schrey AW, Coon CAC, Grispo MT, Awad M, Imboma T, McCoy ED, moter CpG content positively covaries with lifespan and is dependent
Mushinsky HR, Richards CL, Martin LB. Epigenetic variation may on gene function among vertebrates. J Hered. 2023:114:207–218.

Downloaded from https://academic.oup.com/jhered/article/115/1/11/7334204 by guest on 28 January 2025

compensate for decreased genetic variation with introductions: Siller SJ. Epigenetic modification of the hypothalamic-pituitary-adrenal
a case study using house sparrows (Passer domesticus) on two axis during early life of the house sparrow (Passer domesticus).
continents. Genet Res Int. 2012:979751:1–7. New York, NY, USA: Columbia University; 2022.
Schrey AW, Grispo M, Awad M, Cook MB, McCoy ED, Mushinsky vonHoldt BM, Kartzinel RY, van Oers K, Verhoeven KJF, Ouyang JQ.
HR, Albayrak T, Bensch S, Burke T, Butler LK, et al. Broad‐scale Changes in the rearing environment cause reorganization of mo-
latitudinal patterns of genetic diversity among native European and lecular networks associated with DNA methylation. J Anim Ecol.
introduced house sparrow (Passer domesticus) populations. Mol 2023:92:648–664.
Ecol. 2011:20:1133–1143. Watson H, Powell D, Salmón P, Jacobs A, Isaksson C. Urbanization is
Sheldon E, Zimmer C, Hanson H, Koussayer B, Schrey A, Reece D, associated with modifications in DNA methylation in a small pas-
Wigley P, Wedley AL, Martin LB. High epigenetic potential protects serine bird. Evol Appl. 2021:14:85–98.