Phylogeographic structure of Italian Formica pratensis (Retzius 1783) populations in the framework of the species Eurasian range

Abstract

The phylogeography and demographical history of Italian Formica pratensis populations were examined and compared with the Eurasian-wide dataset available for this species and the other red wood ant species Formica lugubris . Forty-eight workers belonging to eight populations from both Alps and Apennines were analysed sequencing a 1.5-kilobase mitochondrial DNA fragment, including the cytochrome b gene and part of the NADH dehydrogenase subunit 6 gene. A total of 127 sequences were screened, scoring 53 different haplotypes amongst all specimens, with five new haplotypes discovered in the Italian populations. All the Italian haplotypes clustered in a monophyletic clade, underlining a clear phylogeographical separation of this group from the other Eurasian groups and suggesting a glacial separate forest refugia and different post-glacial colonisation patterns. The haplotypes from the Alps and the Apennines showed a high genetic proximity, pointing out an ancient (Pleistocene) wide distribution of this species across all these areas and common ancestral lineages. No shared haplotypes were scored between Northern and Central Apennine populations, but the low inter-population genetic distance indicated similar post-glacial selective processes acting on these groups. The diversity we recorded may be influenced by the actual fragmentation of F. pratensis populations across its entire Eurasian range, and by the limited geographical origin and sample dimension of the dataset analysed. Future studies with a more extensive sampling in the Alps and Eastern Europe are needed to confirm our result.

Full text

Vol.:(0123456789)
Insectes Sociaux
https://doi.org/10.1007/s00040-024-00999-8
RESEARCH ARTICLE
Insectes Sociaux
Phylogeographic structure ofItalian Formica pratensis (Retzius 1783)
populations intheframework ofthespecies Eurasian range
A.Masoni1 · F.Castellucci2 · A.Piccinini3 · C.Greco4 · P.Balzani5 · F.Frizzi1 · F.Mattucci4 ·
P.Giangregorio4 · E.Guariento6 · M.Zaccaroni1 · G.Santini1 · A.Luchetti2
Received: 18 June 2024 / Revised: 12 September 2024 / Accepted: 16 September 2024
© The Author(s) 2024
Abstract
The phylogeography and demographical history of Italian Formica pratensis populations were examined and compared
with the Eurasian-wide dataset available for this species and the other red wood ant species Formica lugubris. Forty-eight
workers belonging to eight populations from both Alps and Apennines were analysed sequencing a 1.5-kilobase mitochon-
drial DNA fragment, including the cytochrome b gene and part of the NADH dehydrogenase subunit 6 gene. A total of 127
sequences were screened, scoring 53 different haplotypes amongst all specimens, with five new haplotypes discovered in
the Italian populations. All the Italian haplotypes clustered in a monophyletic clade, underlining a clear phylogeographical
separation of this group from the other Eurasian groups and suggesting a glacial separate forest refugia and different post-
glacial colonisation patterns. The haplotypes from the Alps and the Apennines showed a high genetic proximity, pointing
out an ancient (Pleistocene) wide distribution of this species across all these areas and common ancestral lineages. No shared
haplotypes were scored between Northern and Central Apennine populations, but the low inter-population genetic distance
indicated similar post-glacial selective processes acting on these groups. The diversity we recorded may be influenced by the
actual fragmentation of F. pratensis populations across its entire Eurasian range, and by the limited geographical origin and
sample dimension of the dataset analysed. Future studies with a more extensive sampling in the Alps and Eastern Europe
are needed to confirm our result.
Keywords Red wood ants· mtDNA· Phylogeography· Cytochrome b gene· Haplotypes network
Introduction
Red wood ants (RWAs, Formica rufa group) are one of
the dominant ecosystem components in cool and temper-
ate coniferous woodlands of the Palearctic Region (Risch
etal. 2016). Fourteen morphologically similar species can
be found in this region with at least six species recorded in
Europe, where most of them can often be found in sympatry
and even generate hybrids (Seifert 2021). The distribution
of European RWA species is well-known, given the large
number of studies focussing on these taxa (Stockan etal.
2016). However, the frequent occurrence of hybrid colonies
makes it more difficult to understand the actual geographi-
cal range: for example, many RWA populations of Great
Alberto Masoni and Filippo Castellucci have been contributed
equally to this work.
* A. Masoni
alberto.masoni@unifi.it
1 Department ofBiology, University ofFlorence,
SestoFiorentino, Italy
2 Department ofBiological, Geological andEnvironmental
Sciences, University ofBologna, Via Selmi 3,
40126Bologna, Italy
3 Centre forEvolutionary Zoology, School ofBiological
Sciences, University ofWestern Australia, Crawley 6009,
Perth, Australia
4 Institute forEnvironmental Protection andResearch
(ISPRA), Laboratory ofGenetics, Via Cà Fornacetta 9,
40064OzzanoDell’Emilia(BO), Italy
5 Faculty ofFisheries andProtection ofWaters, South
Bohemian Research Center ofAquaculture andBiodiversity
ofHydrocenoses, University ofSouth Bohemia inČeské
Budějovice, Zátiší 728/II, 38925Vodňany, CzechRepublic
6 Institute forAlpine Environment, Eurac Research,
Drususallee 1, 39100Bolzano, Italy

A.Masoni et al.
Britain are supposed to consist of hybrids between F. rufa
and F. polyctena, since 95% of all samples collected pre-
sent intermediate phenotypic traits between the two species
(Seifert 2021).
The F. rufa group origin dates back to around 15 million
years ago when the group diverged from the other Formica
species. According to the mitochondrial DNA analysis, this
clade is monophyletic and is divided into a Nearctic and
a Palearctic sub-clade (Goropashnaya etal. 2012). In par-
ticular, the Palearctic species have been characterised by
a reticulate evolution largely driven by hybridisation and
introgression events (Seifert and Goropashnaya 2004; Trager
2016; Romiguier etal. 2018). Many of these hybridisation
events are currently ongoing between sympatric populations
of RWA species (Satokangas etal. 2023), but they are dif-
ficult to detect without a molecular screening approach due
to the extreme intraspecific polymorphism and occurrence
of local morphological variants (Seifert, 2021).
RWAs occupy a keystone position in the habitats they
colonise, influencing their nests and foraging activity impor-
tant biotic and abiotic ecosystem components (Frizzi etal.
2018; Di Nuzzo etal. 2022; Guariento and Fiedler 2021).
Some of their ecological functions, such as nutrient recy-
cling, trophobiosis, seed dispersal, soil aeration, and preda-
tion on other arthropods are fundamental for forest ecosys-
tem functioning (Frouz etal. 2016; Guariento etal. 2021;
Balzani etal. 2022a), furthermore, their nests host a huge
variety of arthropod species representing a real biodiver-
sity hot spot (Frizzi etal. 2020; Castellucci etal. 2022).
In the last 50years, local decline or even extinction events
have been recorded in many areas of Europe, with agricul-
tural activities, industrialization, and habitat fragmentation
representing the main threats, together with climate change
(Sorvari 2016; Çamlıtepe and Aksoy 2019). Therefore, in
1996, the International Union for Conservation of Nature
(IUCN), listed five RWA species [F. rufa, F. lugubris, F.
polyctena, F. aquilonia, and F. pratensis (plus F. uralen-
sis)] as Near Threatened at a global level. However, after
more than 20years, no further official assessment with spe-
cies status updating was carried out, nor any other RWA
species has been included in the Red List (IUCN Red List
2023). Although the standing recognition of their impor-
tance prompted these species for legal protection in several
European countries (Cherix etal. 2012), there is not a unique
reference framework; therefore, the development of wide-
scale conservation actions is needed (Sorvari 2016; Balzani
etal. 2022b).
In Italy, RWA species are widespread along the Alps
and only the more thermophilic Formica pratensis (Ret-
zius, 1783) naturally occurs at lower latitudes in the Apen-
nine mountains. In those areas additional RWA species (F.
lugubris, F. rufa, F. paralugubris, and F. aquilonia) may be
found, originating from past intentional introductions, with
F. paralugubris being the most common species (Masoni
etal. 2019), as biocontrol agents against the outbreaks of
forest insect pests (Ronchetti and Groppali 1995). Some of
these populations are still extant, are expanding and show
signs of genetic diversification from their source populations
(Masoni etal. 2022). Other populations, instead, underwent
rapid decline and went extinct in a few years (Ronchetti etal
1986; Frizzi etal. 2018).
Formica pratensis mainly lives in open grassland habi-
tats, often at the edge of coniferous or mixed coniferous-
deciduous forests and compared to the other RWAs, it occurs
at lower elevations, forming smaller colonies in open sunny
habitats (Véle etal.2009). Monogynous colonies of this
species predominate, although polygyny (Seifert 1991; Pirk
etal. 2001; Aksoy and Camlitepe 2018) and, rarely, superco-
loniality (Kiss and Kóbori 2010) may also occur. Polydomy
is also described in this species, which allows colonies to
create new nests without going through the single-queen nest
foundation (Seifert 1991; Ellis and Robinson 2014).
The distribution of F. pratensis populations along the
Italian peninsula is unknown, as are their ecology and con-
servation status. The scant information about the presence
and distribution of the species is limited to outdated pub-
lications (Pavan 1981, Fig.S1), where it was reported as
separated into two species (F. nigricans and F. pratensis,
see below), and some recent records on scientific blogs or
social networks (iNaturalist, https:// www. inatu ralist. org/;
Antwiki, https:// antwi ki. org/). At the beginning of the last
century, Emery (1909) described the Apennine populations
as belonging to F. nigricans, which differed from the Alpine
and European populations (F. pratensis) by having darker
head and thorax. However, morphological studies carried out
by Seifert since 1991 rejected this separation, and recently
Seifert (2021) confirmed that the former F. nigricans is to
be considered as an ecomorph of the F. pratensis originally
described by Retzius in 1783. Considering the rapid decline
of this species in recent decades across its entire European
range (e.g. Dekoninck etal. 2010; Çamlıtepe and Aksoy
2019), many Apennine populations may have experienced
the same fate, disappearing, decreasing in number, or form-
ing scattered isolated populations.
In this study, we analysed the geographical pattern of
mitochondrial cytochrome b gene (Cyt-b) variation amongst
eight populations of F. pratensis, seven located along the
Apennine and one in the Italian Alps. This mitochondrial
region was chosen given the high variability that was
observed in the genus Formica L. in previous phylogeog-
raphy and population structure studies (Goropashnaya etal.
2004a; Antonov and Bukin 2016) and a Eurasian-wide data-
set based on this marker is available for both F. lugubris and
F. pratensis (Goropashnaya etal. 2004b). We first analysed
the relationships between haplotypes of the Italian popula-
tions and their geographical distribution, then we compared

Phylogeographic structure ofItalian Formica pratensis (Retzius 1783) populations inthe…
the Italian populations with the Eurasian ones investigated
by Goropashnaya etal. (2004a) and stored in Gene Bank,
NCBI (http:// www. ncbl. nlm. nih. gov/). Our results will
improve the knowledge of the F. rufa group speciation and
spread. They can also help to turn the spotlight on the impor-
tance of the Italian populations of F. pratensis, prompting
urgent conservation plans.
Materials andmethods
Sample collection
Fieldwork was carried out between March 2021 and Sep-
tember 2023. We collected samples from eight F. pratensis
populations (See Table1) distributed across the Alps and
the Apennines. These populations strongly differ in eleva-
tion and habitat selection. For each population, 20 workers
were sampled from six different nests. To avoid sampling
ants from non-independent nests belonging to the same
polydomous colony as much as possible, we used only nests
located at least 30m away and not connected by trails of
workers. All ants from each nest were preserved in absolute
ethanol and divided into two groups, one stored at − 80°C
until DNA extraction, and the other one used for species
identification (Seifert 2021).
DNA Extraction, amplification andsequencing
Total genomic DNA was extracted from the leg tissue of
a single ant per nest, using a Nucleospin® DNA insect kit
(Macherey–Nagel) following the manufacturer’s instruc-
tions. To compare the newly sampled Italian populations
with the Eurasian dataset by Goropashnaya etal. (2004b),
a highly variable 1571bp mitochondrial DNA fragment
including the complete cytochrome b gene (Cyt-b, 1125bp
long) and the first part of the NADH dehydrogenase subunit
6 (ND-6, 446bp long) has been amplified via PCR and
sequenced. A microsatellite characterised by the repetition
of a TTA motif is found in the intragenic region between the
Cyt-b gene and the ND-6. Two primer couples were chosen
to amplify two overlapping fragments (Goro1 and Goro2, see
Suppl. Tab. S1), subsequently assembled to reconstruct the
whole 1571bp sequence. Amplification via PCR was carried
out with an initial denaturation step of 2min at 94°C, 45–50
PCR cycles (50s at 94°C, 45s at Ta, 50s at 72°C) and a
final extension step of 5min at 72°C. PCR products were
checked via electrophoresis on a 1% agarose gel and purified
using ExoSAP-IT Product Cleanup Reagent (Thermo Fisher
Scientific). Sanger sequencing of the forward and reverse
strands was performed by Macrogen Europe (Amsterdam,
The Netherlands). Chromatograms were inspected and
edited with the software SeqTrace v.0.9.0 (Stucky 2014).
The search for potential contaminants was carried out using
BLASTn (Zhang and Madden 1997) on the NCBI database.
The obtained sequences were submitted to NCBI GenBank
(accession numbers PP179335-PP179375).
Phylogenetic andhaplotype network analyses
We obtained sequences from a total of 48 workers. These
were, then, added to the 79 F. pratensis + F. lugubris
sequences from Goropashnaya etal. (2004b) Palearctic data-
set, stored in GenBank, NCBI (PopSet accession number:
46403874 and 45,934,296), to build the final dataset. Align-
ment was performed using the automatic detection of the
best-fit algorithm on MAFFT v7.503 (Katoh and Standley
2013). The microsatellite close to the 5’-end of the Cyt-b
gene was excluded from the analyses to facilitate compari-
son of the results. Sites with missing data or alignment gaps
were removed and haplotypes were retrieved running the
R software v 4.1.2 (R Core Team 2021), using the package
“haplotypes” (Aktas 2020). A single sequence for each of
the observed haplotypes was maintained in the final dataset.
Table 1 Populations of F. pratensis analysed in this work
Localities Pop ID Altitude Habitat N° nest Latitude Longitude
Alps
Bozen province (Trentino-Alto Adige) BOL 1200–1400 Open areas/Fir Forest > 10 46°49′12.01ʺN 11°9′27.00ʺE
Northern Apennine
La Consuma (Florence, Tuscany) CO 1000–1100 Open areas/Fir Forest < 10 43°45′57.65ʺN 11°35′18.35ʺE
Vallombrosa (Florence, Tuscany) VAL 1000–1100 Fir Forest < 10 43°73′38.34ʺN 11°55′70.38ʺE
La Verna (Arezzo, Tuscany) LV 1000–1200 Open areas/Fir Forest > 10 43°42′51.71ʺN 11°56′38.32ʺE
Orecchiella (Lucca, Tuscany) ORE 1100–1250 Fir Forest < 10 44°12′21.03ʺN 10°20′45.38ʺE
Ticchiano (Parma, Emilia-Romagna) TIC 950–1100 Fir Forest < 10 44°26′6.46ʺN 10° 6′7.42ʺE
Central Apennine
Barrea (Aquila, Abruzzo) BAR 1200–1300 Open areas < 10 41°46′2.43ʺN 13°59′40.56ʺE
Scanno (Aquila, Abruzzo) SCA 1100–1300 Open areas < 10 41°53′29.81ʺN 13°55′7.10ʺE

A.Masoni et al.
Model selection and maximum likelihood phylogenetic
reconstruction were carried out using IQ-TREE (Minh etal.
2013) and nodal support was estimated with 1000 replicates
of UltraFast bootstrap.
The same dataset used for phylogenetic reconstruc-
tion was used to build a haplotype network with PopART
(Leigh and Bryant 2015), implementing the Median Joining
method.
Genetic distances within and between F. pratensis clades
were computed with MEGA v.11.0.13 (Tamura etal. 2021)
using the Tamura and Nei (1993) substitution model.
Results
The final dataset consisted of 127 sequences of 1431bp,
excluding sites with missing data, alignment gaps and the
microsatellite found in the intergenic region of Cytb accord-
ing to Goropashnaya etal. (2004b). The TTA motif of this
microsatellite was repeated four to thirteen times in the Ital-
ian haplotypes and four to seven times in Eurasian ones,
underlining the first clear separation of the Italian popula-
tions. Overall, 53 different haplotypes were scored amongst
all specimens, with five new haplotypes discovered in the
Italian populations of F. pratensis (Table2).
The maximum likelihood tree obtained on the haplotypes
of both species (Fig.1) was comparable to the neighbour-
joining tree by Goropashnaya etal. (2004b) for the F. prat-
ensis clade. Two major, well-supported clades were identi-
fied (bootstrap > 90), separating haplotypes belonging to F.
pratensis and F. lugubris with few exceptions. As in Goro-
pashnaya etal. (2004b), few haplotypes (H2, H3, H4, and
H6) found in populations from the Urals and the Pyrenees,
morphologically identified as F. pratensis, clustered within
the F. lugubris clade, and one haplotype (H5) was found to
be shared between populations of both species. Concerning
the Italian populations, all haplotypes clustered in a strongly
supported clade (bootstrap = 100), which was found in sis-
ter relation to all other Eurasian F. pratensis haplotypes
(Fig.1). Within this clade, no clear distinctions could be
found between populations from the Alps (haplotypes H49
and H50) and the Apennines (haplotypes H49, H51, H52 and
H53), with haplotype H49 being shared between populations
sampled on the two mountain ranges.
The median-joining network showed a clear distinc-
tion between the F. pratensis and the F. lugubris haplotype
groups (Fig.2), differentiated by at least seven nucleotide
substitutions. The F. lugubris haplotypes were organised in
a star-like sub-network, with haplotype H5 in a central posi-
tion. This haplotype was scored in populations from Russia,
Sweden, and the Pyrenees, but also in F. pratensis popula-
tions from the Urals. Moreover, haplotypes H2, H3 and H4,
belonging to F. pratensis specimens from the Pyrenees, and
haplotype H6, belonging to F. pratensis specimens from
Urals, were nested into the F. lugubris group, as already
evidenced in the phylogenetic tree (Fig.1). The remaining
F. pratensis haplotypes, on the other hand, were included in
a sub-network showing a more branched pattern. Here, the
Italian haplotypes (coloured circles in Fig.2) appeared to
form a distinct group from the other Eurasian ones, being
differentiated by a high number of substitutions and several
missing haplotypes inferred by the median-joining method.
Surprisingly, Central Apennine populations seemed
closer to the Alpine populations than to the ones in the
Northern Apennines (Fig.3). Haplotype H49 was indeed
shared by the Central Apennine populations from Barrea and
Scanno and the population from the Alps (Fig.3). Regard-
ing the Northern Apennine populations, haplotype H51 was
shared by all the localities, whilst haplotypes H52 and H53
were private for the populations of Ticchiano and Vallom-
brosa, respectively. Haplotype H50 was found only in the
Alpine population.
We evaluated the level of divergence, considering the
genetic distances amongst the Eurasian F. pratensis haplo-
types and the Italian haplotypes group. The genetic distance
was higher in the European cluster (d = 0.00637) compared
to the Italian ones (d = 0.00126), as expected considering
the different number of haplotypes in the two groups and
the widespread geographical distribution. Interestingly,
the genetic distance between the two groups (d = 0.0104)
resulted higher than the within-group, underlining a diver-
gence of the Italian populations from all the others.
Discussion
The phylogeography and demographical history of Italian
Formica pratensis populations were examined and com-
pared with the Eurasian haplotypes dataset (Goropashnaya
etal. 2004b). We included in our analysis also the pub-
lished F. lugubris haplotypes, given the close relation-
ship between these two species, and the already reported
genetic evidence for past hybridization events (Goropash-
naya etal. 2004b). The analyses based on a 1.5-kilobase
mitochondrial DNA fragment, including the Cyt-b gene
and part of the ND-6 gene, suggested no evidence of past
or recent mitochondrial DNA exchange between Italian F.
pratensis and F. lugubris. The exchanges with other spe-
cies are difficult in the Apennines, where F. pratensis is
the only RWA species naturally occurring in these areas
since the last glaciation. Interactions with other introduced
RWAs (mainly F. paralugubris), although possible, have
never been documented and, in our opinion, are unlikely.
In the Foreste Casentinesi National Park (Northern Tus-
cany), where one of the largest introduction campaigns
was carried out in the past and there are several extant

Phylogeographic structure ofItalian Formica pratensis (Retzius 1783) populations inthe…
Table 2 List of the scored
haplotypes. Concerning
the Italian populations, the
number of individuals of each
population that own the given
haplotype is reported in brackets
Haplotype Species Localities Ref.seq database
H1 F. pratensis Finland AY584199.1
H2 F. pratensis Pyrenees AY584232.1
H3 F. pratensis Pyrenees AY584231.1
H4 F. pratensis Pyrenees AY584230.1
H5 F. pratensis-F.lugubris Urals + Sweden + Russia + Pyrenees AY573860.1
H6 F. pratensis Urals AY584227.1
H7 F. pratensis Sweden AY584226.1
H8 F. pratensis Sweden AY584225.1
H9 F. pratensis Sweden AY584224.1
H10 F. pratensis Romania AY584223.1
H11 F. pratensis Romania AY584222.1
H12 F. pratensis Russia AY584221.1
H13 F. pratensis Russia AY584220.1
H14 F. pratensis Sweden AY584219.1
H15 F. pratensis Urals AY584218.1
H16 F. pratensis Urals AY584217.1
H17 F. pratensis Finland AY584216.1
H18 F. pratensis Sweden AY584215.1
H19 F. pratensis Finland AY584214.1
H20 F. pratensis Finland + Russia AY584212.1
H21 F. pratensis Urals AY584211.1
H22 F. pratensis Russia AY584209.1
H23 F. pratensis Urals AY584208.1
H24 F. pratensis Urals AY584207.1
H25 F. pratensis Urals AY584206.1
H26 F. pratensis Russia AY584203.1
H27 F. pratensis Sweden AY584202.1
H28 F. pratensis Denmark AY584201.1
H29 F. pratensis Russia AY584198.1
H30 F. pratensis Finland AY584197.1
H31 F. pratensis Finland AY584196.1
H32 F. lugubris Russia + Urals AY573873.1
H33 F. lugubris Sweden + Switzerland AY573870.1
H34 F. lugubris Russia AY573866.1
H35 F. lugubris Pyrenees AY573874.1
H36 F. lugubris Russia AY573872.1
H37 F. lugubris Urals AY573871.1
H38 F. lugubris UK AY573869.1
H39 F. lugubris UK AY573868.1
H40 F. lugubris UK AY573867.1
H41 F. lugubris Urals AY573865.1
H42 F. lugubris Urals AY573864.1
H43 F. lugubris Russia AY573863.1
H44 F. lugubris Russia AY573862.1
H45 F. lugubris Sweden AY573861.1
H46 F. lugubris Sweden AY573859.1
H47 F. lugubris Sweden AY573858.1
H48 F. lugubris Pyrenees AY573856.1
H49 F. pratensis Italy: BAR (6), BOL (1), SCA (6) this study
H50 F. pratensis Italy: BOL(5) this study

A.Masoni et al.
F. paralugubris populations, repeated surveys of the area
found no evidence of contact between the two species.
The closest known F. pratensis population is the one in
La Verna (Province of Arezzo), more than 20km apart.
Moreover, the two species have different environmental
requirements, grasslands for F. pratensis, closed canopy
stands for F. paralugubris, which further limits the pos-
sibility of contact. None of the other F. pratensis popu-
lations used in this study have other RWAs populations
in their proximity. Considering the Alpine range, where
several RWA species live in sympatry and such events are
more likely to occur (Bernasconi etal. 2011), we sampled
only one population; therefore, future studies with a more
extensive sampling are needed to confirm this result.
An interesting outcome concerns the monophyletic
clade that characterised the Italian samples, supported by
both phylogenetic analyses, with maximum bootstrap sup-
port, and the haplotype network. Alpine and Apennine F.
pratensis haplotypes clustered in the same clade, and the
haplotype H49 was shared by the two groups. The phylo-
geographic separation of the Italian population from the
other Eurasian ones implied a different vicariant history,
with the divergence dating before the diversification of the
other Eurasian populations into the suggested western and
eastern clades, and in any case in the last glacial period
(Desalle etal. 1987). The current western distribution of F.
pratensis has been hypothesised to belong to populations
that survived the glaciations in the Carpathian Mountains
refugia (Sumegi & Krolopp 2002), but this theory is only
supported by the analysis of a few haplotypes from Roma-
nia and Sweden (Goropashnaya etal. 2004a). Conversely,
we did not find any genetic evidence (i.e. shared haplo-
type) that supports a potential proximity of the Italian
cluster with the Western one. Instead, the genetic distance
of the first group from the other Eurasian populations sug-
gested a possible different scenario: both the Alps, espe-
cially the south-western part, and Apennine areas may
represent other possible glacial refugia for this species as
it was reported for many animals (Schebeck etal. 2019;
Korábek etal. 2023) and plants (Záveská etal. 2021; Pari-
sod 2022). The continuous S-shaped mountain-hill system
comprising both mountain chains (37° to 48° of latitude)
encompasses the glacial refugia of the Italian Peninsula
(Dapporto etal., 2019) and mountain areas covered by ice
caps during glacial maxima (Menchetti etal. 2021). This
area is considered a single biogeographic unit for many
endemisms (Petit etal. 2003; Drovetski etal.2018; but
see Menchetti etal. 2021) and our results agreed with this
notion. Furthermore, the haplotypes from the Alps and the
Apennines showed high genetic proximity, with one shared
haplotype, suggesting an ancient (Pleistocene) wide dis-
tribution of this species across all these areas. Moreover,
after the last glaciation, the physical barrier of the Alps
may have limited the northward dispersal exchanges with
Central European populations (Hewitt 1999; Drovetski
Table 2 (continued) Haplotype Species Localities Ref.seq database
H51 F. pratensis Italy: CO (6), LV (6), ORE (6), TIC (5),
VAL (5)
this study
H52 F. pratensis Italy: TIC (1) this study
H53 F. pratensis Italy: VAL (1) this study
Fig. 1 Maximum likelihood tree of Formica pratensis and F. lugubris
built using IQ-TREE with 1000 ultrafast bootstrap replicates. BS boot-
strap values. Haplotypes in bold were detected in populations mor-
phologically identified as F. pratensis but clustered with haplotypes
of F. lugubris. The haplotype H5 marked with a * was shared by pop-
ulations belonging to both species

Phylogeographic structure ofItalian Formica pratensis (Retzius 1783) populations inthe…
etal. 2018), thus isolating the Italian F. pratensis popula-
tions from the other European ones, as suggested by the
lack of haplotypes shared between these groups.
Considering the genetic structure of Apennine samples
(Fig.3), no haplotypes were shared between Northern and
Central populations, but the haplotype H52 from Ticchiano
was genetically closer to the Central group than to its native
group, underlining the sharing of common ancestral line-
ages and the wide distribution along the inter-refugium areas
of these species. The low interpopulation genetic distance
amongst these groups (one missing haplotype) can suggest
possible similar post-glacial evolutionary history and selec-
tive process acting on these two groups, which can explain
the geographical patterns recorded. This partly contrasts
with the evolutionary dynamics recorded in these areas
for other cold-adapted insect species (Martín-Bravo etal
2010; Lecocq etal. 2013), which experienced more intense
intraspecific differentiation processes leading sometimes
to speciation events, as happened for Bombus monticola
mathildisand B. konradini (Martinet etal 2018).
In addition, the diversity we recorded is greatly influ-
enced by the actual fragmentation of F. pratensis popula-
tions because of the rapid decline that this species has been
experiencing for many decades in Italy (e.g. at the time we
are submitting this article, the population sampled in Val-
lombrosa has gone extinct, personal observation) and across
the entire Eurasian range (Kiss and Kóbori 2010; Stockan
etal. 2016; Çamlıtepe and Asksoy 2019). The same situation
was recorded in the last 50 years also for other RWA spe-
cies, with agricultural activities, industrialization and habitat
fragmentation representing the main threats (Mäki-Petäys
and Breen 2007; Dekoninck etal. 2010), together with cli-
mate change (Sorvari 2016). In 1996 the International Union
for Conservation of Nature (IUCN), listed five RWA spe-
cies (F. rufa, F. lugubris, F. polyctena, F. aquilonia, and F.
pratensis (plus F. uralensis)) as Near Threatened at a global
Fig. 2 Median-joining network of Formica pratensis and F. lugubris
built with popART. Italian haplotypes are coloured in different shades
of blue according to mountain range of origin. Black circles represent
inferred missing haplotypes, whilst the number of small transversal
dashes on the lines connecting two haplotypes represents the number
of nucleotide substitutions differentiating them

A.Masoni et al.
level. However, after more than 20 years, no further official
assessment with species status updating was carried out,
or any other RWA species listed (IUCN Red List 2023). It
is therefore necessary to take rigorous measures to protect
and facilitate the survival and dispersal of this and the other
RWA species (Balzani etal. 2022b).
In conclusion, the results of this study provided a descrip-
tion of the phylogeographic relationship amongst Italian F.
pratensis populations and those from Central and North
Europe. The results also generated several open-ended
questions, answering which will require the extension of
the sample size, with more populations from the Alps and
intermediate areas between Italy and Europe like Slovacchia,
Germany, Poland, but also a different approach. In particular
it might be useful to employ different genetic markers to
screen both mitochondrial and nuclear DNA to better detect
the genetic diversity amongst and between these popula-
tions, which is crucial for understanding species genetic
structure and defining evolutionary and management units
for future conservation planning.
Supplementary Information The online version contains supplemen-
tary material available at https:// doi. org/ 10. 1007/ s00040- 024- 00999-8.
Acknowledgements We sincerely thank Dr Davide Alberti and Mar.
Ord. RFPGiovanni Galipò for assisting us during the project. We
extend our gratitude to the Reparto Carabinieri Biodiversità di Pistoia
and the Reparto Carabinieri Biodiversità di Vallombrosa for their help
in the field activities. We gratefully acknowledge the Parco Nazionale
d’Abruzzo, Lazio e Molise, the Parco Nazionale Foreste Casentinesi,
Monte Falterona e Campigna, the Riserva Naturale Biogenetica di
Vallombrosa, the Riserva Naturale dello Stato dell’Orecchiella and
the Ente di Gestione per i Parchi e la Biodiversità Emilia Occidentale.
Funding Open access funding provided by Università degli Studi di
Firenze within the CRUI-CARE Agreement. This study was partly
funded by the Parco Nazionale Foreste Casentinesi, Monte Falterona
e Campigna. The study was also funded by the Italian Ministry of Uni-
versities and Research under the Biodiversa + framework (call 22–23,
project MonitAnt)and bythe National Recovery and Resilience Plan
(PNRR), which is part of the Next Generation EU (NGEU) program
established by the European Union.
Data availability The sequences generated and analysed in the present
study are available from NCBI database.
Declarations
Conflict of interest The authors have no relevant financial or non-fi-
nancial interest to disclose.
Open Access This article is licensed under a Creative Commons Attri-
bution 4.0 International License, which permits use, sharing, adapta-
tion, distribution and reproduction in any medium or format, as long
as you give appropriate credit to the original author(s) and the source,
provide a link to the Creative Commons licence, and indicate if changes
were made. The images or other third party material in this article are
included in the article’s Creative Commons licence, unless indicated
otherwise in a credit line to the material. If material is not included in
the article’s Creative Commons licence and your intended use is not
permitted by statutory regulation or exceeds the permitted use, you will
need to obtain permission directly from the copyright holder. To view a
copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
Fig. 3 Haplotypes distribution
maps and network of Italian
Formica pratensis populations.
The different colours represent
the haplotypes ID whilst the pie
chart portions their frequencies
(n = 6 individuals x populations)

Phylogeographic structure ofItalian Formica pratensis (Retzius 1783) populations inthe…
References
Aksoy V, Camlitepe Y (2018) Effects of genetic relatedness, spatial dis-
tance, and context on intraspecific aggression in the red wood ant
Formica pratensis (Hymenoptera: Formicidae). Turkish Journal
of Zoology 42(3):297–306. https:// doi. org/ 10. 3906/ zoo- 1708- 17
Aktas C (2020) Manipulating DNA Sequences and estimating unam-
biguous haplotype network with statistical parsimony. https://
CRAN.R- proje ct. org/ packa ge= haplo types
Antonov IA, Bukin YS (2016) Molecular phylogenetic analysis of the
ant genus Formica L. (Hymenoptera: Formicidae) from Palearctic
region. Russ J Genet 52:810–820. https:// doi. org/ 10. 1134/ S1022
79541 60800 20
Balzani P, Masoni A, Venturi S, Frizzi F, Bambi M, Fani R, San-
tini G (2022) CO2 biogeochemical investigation and microbial
characterization of red wood ant mounds in a Southern Europe
montane forest. Soil Biol Biochem 166:108536
Balzani P, Dekoninck W, Feldhaar H, Freitag A, Frizzi F, Frouz J,
Santini G (2022) Challenges and a call to action for protecting
European red wood ants. Cons. Biol 36(6):e13959. https:// doi.
org/ 10. 1111/ cobi. 13959
Bernasconi C, Cherix D, Seifert B, Pamilo P (2011) Molecular
taxonomy of the Formica rufa group (red wood ants) (Hyme-
noptera: Formicidae): a new cryptic species in the Swiss Alps.
Myrmecological News 14:37–47
Çamlıtepe Y, Aksoy V (2019) Distribution and conservation status of
the European Red Wood ant species Formica pratensis Retzius,
1783 (Hymenoptera, Formicidae) in (European) Turkey. J Ent
Res Soc 21(2):199–211
Castellucci F, Schifani E, Luchetti A, Scharff N (2022) New associa-
tion between red wood ant species (Formica rufa group) and the
myrmecophilic spiders Mastigusa arietina and Thyreosthenius
biovatus. Bull Insectology 75(2):231–238
Cherix D, Bernasconi C, Maeder A, Freitag A (2012) Wood ants in
Switzerland: situation and outlook of a monitoring. Schweiz
Z Forstwes 163(6):232–239. https:// doi. org/ 10. 3188/ szf. 2012.
0232
Dapporto L, Cini A, Vodă R, Dincă V, Wiemers M, Menchetti M,
Magini G, Talavera G, Shreeve T, Bonelli S, Casacci LP (2019)
Integrating three comprehensive data sets shows that mitochon-
drial DNA variation is linked to species traits and paleogeographic
events in European butterflies. Mol Ecol Res 19(6):1623–1636.
https:// doi. org/ 10. 1111/ 1755- 0998. 13059
Dekoninck W, Hendrickx F, Grootaert P, Maelfait JP (2010) Pre-
sent conservation status of red wood ants in north-western Bel-
gium: worse than previously, but not a lost cause. Eur J Entomol
107:209–218. https:// doi. org/ 10. 14411/ eje. 2010. 028
DeSalle R, Freedman T, Prager EM, Wilson AC (1987) Tempo and
mode of sequence evolution in mitochondrial DNA of Hawaiian
Drosophila. J Mol Evol 26:157–164
Di Nuzzo L, Masoni A, Frizzi F, Bianchi E, Castellani MB, Balzani P,
Morandi F, Sozzi Y, Vallese C, Santini G, Benesperi R (2022) Red
wood ants shape epiphytic lichen assemblages in montane silver
fir forests. iForest 15:71–76. https:// doi. org/ 10. 3832/ ifor3 897- 014
Drovetski SV, Fadeev IV, Raković M, Lopes RJ, Boano G, Pavia M,
Voelker G (2018) A test of the European Pleistocene refugial para-
digm, using a Western Palaearctic endemic bird species. Proc R
Soc B 285(1889):20181606. https:// doi. org/ 10. 1098/ rspb. 2018.
1606
Ellis S, Robinson EJH (2014) Polydomy in red wood ants. Insect
Sociaux 61:111–122. https:// doi. org/ 10. 1016/j. cois. 2014. 09. 002
Frizzi F, Masoni A, Quilghini G, Ciampelli P, Santini G (2018) Chroni-
cle of an impact foretold: the fate and effect of the introduced For-
mica paralugubris ant. Biol Invasions 20(12):3575–3589. https://
doi. org/ 10. 1007/ s10530- 018- 1797-x
Frizzi F, Masoni A, Migliorini M, Fanciulli PP, Cianferoni F, Balzani
P, Santini G (2020) A comparative study of the fauna associated
with nest mounds of native and introduced populations of the
red wood ant Formica paralugubris. Eur J Soil Biol 101:103241.
https:// doi. org/ 10. 1016/j. ejsobi. 2020. 103241
Frouz J, Jílková V, Sorvari J (2016) Contribution of Wood Ants to
Nutrient Cycling and Ecosystem Function. In: Stockan and Rob-
inson (Eds), Wood ant ecology and conservation. Cambridge Uni-
versity Press, Cambridge, UK, 207–220
Goropashnaya AV, Fedorov VB, Pamilo P (2004) Recent speciation in
the Formica rufa group ants (Hymenoptera-Formicidae): infer-
ence from mitochondrial DNA phylogeny. Mol Phylogenet Evol
32(1):198–206. https:// doi. org/ 10. 1016/j. ympev. 2003. 11. 016
Goropashnaya AV, Fedorov VB, Seifert B, Pamilo P (2004) Limited
phylogeographical structure across Eurasia in two red wood ant
species Formica pratensis and F. lugubris (Hymenoptera, For-
micidae). Mol Ecol 13(7):1849–1858. https:// doi. org/ 10. 1111/j.
1365- 294X. 2004. 02189
Goropashnaya AV, Fedorov VB, Seifert B, Pamilo P (2012) Phyloge-
netic relationships of Palaearctic Formica species (Hymenoptera,
Formicidae) based on mitochondrial cytochrome B sequences.
PLoS ONE 7(7):e41697. https:// doi. org/ 10. 1371/ journ al. pone.
00416 97
Guariento E, Fiedler K (2021) Ant diversity and community composi-
tion in alpine tree line ecotones. Insects 12(3):219. https:// doi. org/
10. 3390/ insec ts120 30219
Guariento E, Wanek W, Fiedler K (2021) Consistent shift in nutritional
ecology of ants reveals trophic flexibility across alpine tree-line
ecotones. Ecol Entomol 46(5):1082–1092. https:// doi. org/ 10.
1111/ een. 13052
Hewitt GH (1999) Post-glacial re-colonization of European biota. Biol
J Lin Soc 68:87–112. https:// doi. org/ 10. 1111/j. 1095- 8312. 1999.
tb011 60.x
IUCN (2023) The IUCN Red List of Threatened Species. Version
2023–1. https:// www. iucnr edlist. org. Accessed on [25/05/2024]
Jermiin LS, Crozier RH (1994) The cytochrome b region in the mito-
chondrial DNA of the ant Tetraponera rufoniger: sequence diver-
gence in Hymenoptera may be associated with nucleotide content.
J Mol Evol 38:282–294. https:// doi. org/ 10. 1007/ BF001 76090
Katoh K, Standley DM (2013) MAFFT multiple sequence alignment
software version 7: improvements in performance and usability.
Mol Biol Evol 30(4):772–780. https:// doi. org/ 10. 1093/ molbev/
mst010
Kiss K, Kóbori O (2010) Formica pratensis supercolony in the Hoia
Forest (Cluj Napoca, Romania). Acta Scientiarum Transylvanica
18:66–76
Korábek O, Adamcová T, Proćków M, Petrusek A, Hausdorf B,
Juřičková L (2023) In both directions: Expansions of European
land snails to the north and south from glacial refugia. J Biog
50(4):654–668. https:// doi. org/ 10. 1111/ jbi. 14531
Lecocq T, Dellicour S, Michez D (2013) Scent of a break-up: phy-
logeography and reproductive trait divergences in the red-tailed
bumblebee (Bombus lapidarius). BMC Evol Biol 13:263. https://
doi. org/ 10. 1186/ 1471- 2148- 13- 263
Leigh JW, Bryant D (2015) PopART: Full-feature software for haplo-
type network construction. Methods Ecol Evol 6(9):1110–1116
Liautard C, Keller L (2001) Restricted effective queen dispersal at a
microgeographic scale in polygynous populations of the ant For-
mica exsecta. Evolution 55:2484–2492
Mäki-Petäys H, Breen J (2007) Genetic vulnerability of a remnant ant
population. Cons Gen 8(2):427–435. https:// doi. org/ 10. 1007/
s10592- 006- 9182-1
Martín-Bravo S, Valcárcel V, Vargas P, Luceño M (2010) Geographi-
cal speciation related to Pleistocene range shifts in the western
Mediterranean mountains. Taxon 59:466–482. https:// doi. org/ 10.
2307/ 25677 605

A.Masoni et al.
Martinet B, Lecocq T, Brasero N, Biella P, Urbanová K, Valterová I,
Rasmont P (2018) Following the cold: geographical differentiation
between interglacial refugia and speciation in the arcto-alpine spe-
cies complex Bombus monticola (Hymenoptera: Apidae). System
Entomol 43(1):200–217. https:// doi. org/ 10. 1111/ syen. 12268
Masoni A, Frizzi F, Natali C, Bernasconi C, Ciofi C, Santini G (2019)
Molecular identification of imported red wood ant populations
in the Campigna biogenetic nature Reserve (foreste Casentinesi
national Park, Italy). Conservation Genet Resour 11(2):231–236.
https:// doi. org/ 10. 1007/ s12686- 018- 0982-7
Masoni A, Coppi A, Balzani P, Frizzi F, Fani R, Zaccaroni M, Santini
G (2022) Assessing molecular diversity in native and introduced
populations of red wood ant Formica paralugubris. Animals
12(22):3165. https:// doi. org/ 10. 3390/ ani12 223165
Menchetti M, Talavera G, Cini A, Salvati V, Dincă V, Platania L, Dap-
porto L (2021) Two ways to be endemic. Alps and Apennines
are different functional refugia during climatic cycles. Mol Ecol
30(5):1297–1310. https:// doi. org/ 10. 1111/ mec. 15795
Minh BQ, Nguyen MAT, von Haeseler A (2013) Ultrafast approxima-
tion for phylogenetic bootstrap. Mol Biol Evol 30(5):1188–1195.
https:// doi. org/ 10. 1093/ molbev/ mst024
Parisod C (2022) Plant speciation in the face of recurrent climate
changes in the Alps. Alp Botany 132(1):21–28. https:// doi. org/
10. 1007/ s00035- 021- 00259-6
Pavan M (1981) Utilità delle Formiche del gruppo Formica rufa. Col-
lana Verde n°57, Ministero dell’Agricoltura e delle Foreste, Italy,
pp. 99.
Petit R, Aguinagalde I, de Beaulieu JL, Bittkau C, Brewer S, Ched-
dadi R, Ennos R, Fineschi S, Grivet D, Lascoux M, Mohanty
A, Müller-Starck G, Demesure-Musch B, Palmé A, Martín JP,
Rendell S, Vendramin GG (2003) Glacial refugia: hotspots but not
melting pots of genetic diversity. Science 6:1563–1565. https://
doi. org/ 10. 1126/ scien ce. 10832 64
Pirk CWW, Neumann P, Moritz RFA, Pamilo P (2001) Intranest relat-
edness and nestmate recognition in the meadow ant Formica prat-
ensis (R.). Behav Ecol Soc 49:366–374. https:// doi. org/ 10. 1007/
s0026 50000 315
R Core Team (2021) A language and environment for statistical com-
puting. R Foundation for Statistical Computing, Vienna, Austria.
URL https:// www.R- proje ct. org/.
Risch AC, Ellis S and Wiswell H (2016) Where and why? Wood ant
population ecology. In: Stockan and Robinson (Eds.), Wood ant
ecology and conservation. Cambridge University Press, Cam-
bridge, UK
Romiguier J, Rolland J, Morandin C, Keller L (2018) Phylogenomics of
palearctic Formica species suggests a single origin of temporary
parasitism and gives insights to the evolutionary pathway toward
slave-making behaviour. BMC Evol Biol 18(1):1–8. https:// doi.
org/ 10. 1186/ s12862- 018- 1159-4
Ronchetti G and Groppali R (1995) Quarantacinque anni di protezi-
one forestale, con Formica lugubris Zett. (Hym. Formicidae).
L'esperienza di Monte d'Alpe (Appennino Ligure in provincia di
Pavia) Pavia. Istituto di Entomologia dell'Università: 161–162.
Ronchetti G, Mazzoldi P, Groppali R (1986) Venticinque anni di
osservazioni sui trapianti di Formica lugubris Zett. dalle Alpi
alle foreste Casentinesi (Italia centrale). Pubblicazioni dell’Istituto
di Entomologia dell’Università di Pavia, Pavia.
Satokangas I, Nouhaud P, Seifert B, Punttila P, Schultz R, Jones
MM, Kulmuni J (2023) Semipermeable species boundaries cre-
ate opportunities for gene flow and adaptive potential. Mol Ecol
31:4329–4347. https:// doi. org/ 10. 1111/ mec. 16992
Schebeck M, Schuler H, Einramhof B, Avtzis DN, Bertheau C (2019)
The Apennines as a cryptic Pleistocene refugium of the bark bee-
tle Pityogenes chalcographus (Coleoptera: Curculionidae). Biol
J Linn Soc 127:24–33. https:// doi. org/ 10. 1093/ bioli nnean/ blz012
Seifert B (1991) Formica nigricans Emery 1909 - an ecomorph of
Formica pratensis Retzius, 1783 (Hymenoptera, Formicidae).
Entomol Fennica 2:217–226
Seifert B (2021) A taxonomic revision of the Palaearctic members of
the Formica rufa group (Hymenoptera: Formicidae)–the famous
mound-building red wood ants. Myrmecol News 31:133–179.
https:// doi. org/ 10. 25849/ myrme col. news_ 031: 133
Seifert B, Goropashnaya AV (2004) Ideal phenotypes and mismatch-
ing haplotypes–errors of mtDNA treeing in ants (Hymenoptera:
Formicidae) detected by standardized morphometry. Org Divers
Evol 4(4):295–305. https:// doi. org/ 10. 1016/j. ode. 2004. 04. 005
Sorvari J (2016) Threats, conservation and management. In: Stockan
and Robinson (Eds.), Wood ant ecology and conservation. Cam-
bridge University Press, Cambridge, UK. pp. 264–287.
Stockan J, Robinson EJ, Trager JC, Yao I and Seifert B (2016) Intro-
ducing wood ants: Evolution, phylogeny, identification and dis-
tribution. In: Stockan and Robinson (Eds.), Wood ant ecology
and conservation. Cambridge University Press, Cambridge, UK.
pp. 1–36.
Stucky BJ (2014) SeqTrace: A graphical tool for rapidly processing
DNA sequencing chromatograms. J B T 23(3):90–93
Sumegi P, Krolopp E (2002) Quatermalacological analyses formod-
eling of the Upper Weichselian palaeoenvironment changesin the
Carpathian Basin. Quatern Int 91:53–63
Tamura K, Nei M (1993) Estimation of the number of nucleotide sub-
stitutions in the control region of mitchondrial DNA in humans
and chimpanzees. Mol Biol Evoi 10:512–526
Tamura K, Stecher G, Kumar S (2021) MEGA11: molecular evolution-
ary genetics analysis version 11. Mol Biol Evol 38(7):3022–3027.
https:// doi. org/ 10. 1093/ molbev/ msab1 20
Trager J (2016) Diversity, ecology and conservation of wood ants in
North America. – In: Stockan, JA and Robinson EJH (Eds.) Wood
ant ecology and conservation. Cambridge University Press, Cam-
bridge, UK, pp. 221–237.
Véle A, Holuša J, Frouz J (2009) Ecological requirements of some
ant species of the genus Formica (Hymenoptera, Formicidae) in
spruce forests. J For Sci 55(1):32–40
Záveská E, Kirschner P, Frajman B, Wessely J, Willner W, Gattringer
A, Schönswetter P (2021) Evidence for glacial refugia of the for-
est understorey species Helleborus niger (Ranunculaceae) in the
Southern as well as in the Northern Limestone Alps. Front Plant
Sci 12:683043. https:// doi. org/ 10. 3389/ fpls. 2021. 683043
Zhang J, Madden TL (1997) PowerBLAST: a new network BLAST
application for interactive or automated sequence analysis and
annotation. Genome Res 7(6):649–656. https:// doi. org/ 10. 1101/
gr.7. 6. 649