This is an author version of the contribution published on:
Questa è la versione dell’autore dell’opera:
Italian Journal of Zoology, Volume 81, Issue 2, 2014, 10.1080/11250003.2013.870240
The definitive version is available at:
La versione definitiva è disponibile alla URL:
The “Evolutionary Significant Unit” concept and its applicability in biological conservation
L. P. CASACCI, F. BARBERO*, & E. BALLETTO
Department of Life Sciences and Systems Biology, University of Turin, Italy
Although most conservationists claim to protect “species”, the conservation unit actually and practically managed is the individual population. As resources are not unlimited, we need to focus on a restricted number of populations. But, how can we select them? The Evolutionarily Significant Unit (ESU), first conceptualised by Ryder in 1986, may offer some answer. Several definitions have been proposed for the ESU, but all make reference to units “whose divergence can be measured or evaluated by putting differential emphasis on the role of evolutionary forces at varied temporal scales”. Thus, an ESU might be fully identical with a “species”, or a “species” could be composed of multiple ESUs. On the other hand, an ESU might comprise single/multiple populations exchanging a degree of gene flow, such as meta-populations. In an attempt to show strengths and weaknesses of ESU concepts, we present here, among several others, some case studies on the myrmecophilous butterflies of the genus Maculinea. In particular, we analyse the apparently everlasting debate about Maculinea alcon and M. rebeli, whose separation into separate species has been accepted by many authors, on mainly ecological criteria, but has not been fully supported by molecular analyses. We also discuss how the tight association with host ants may have driven selection for increasingly more strictly adapted Maculinea populations, arguably deserving specific taxonomic identity. Finally we discuss how current DNA analyses may fail to detect critical information on differences between taxa recently originated by the action of separate adaptive processes, which non-molecular studies can sometimes reveal. We conclude by discussing some current and often conflicting taxonomic trends, in their relationships with conservation policies.
Keywords: Evolutionary Significant Unit, butterflies, Maculinea, conservation, species concepts
*Correspondence: F. Barbero, Department of Life Sciences and Systems Biology, University of Turin, Turin 10123, Italy. Tel: +39 011 6704538. Fax: +39 0116704508. Email: email@example.com
Luca Pietro Casacci and Francesca Barbero contributed equally to this manuscript.
Luca Pietro Casacci, Department of Life Sciences and Systems Biology, University of Turin, Turin 10123, Italy. Tel: +39 011 6704538. Fax: +39 0116704508. Email: firstname.lastname@example.org
Emilio Balletto, Department of Life Sciences and Systems Biology, University of Turin, Turin 10123, Italy. Tel: +39 011 6704515. Fax: +39 0116704508. Email: email@example.com
Running head: ESUs in biological conservation
The conservationist’s dilemma: What should we protect?
It is widespread belief, in biological conservation, that what should be protected are species (Mace 2004). Even setting aside the many important theoretical issues and definition problems related to this concept (Hausdorf 2011; Simonetta in this issue), it remains that with a few though notable exceptions, the IUCN (2010), as well as, among the others, the European Union, with its “Birds” (2009/147/EC, 30 Nov 2009) and “Habitats” (92/43/EEC and later amendments) Directives, (almost) exclusively recognise this taxonomic rank. A number of species are deemed threatened at European level, even though what is a “species”, for a taxonomist or a conservationist based in a given Country or geographical area, may be a “subspecies” for another based elsewhere. Apart from this and other theoretical issues related to varying taxonomic philosophies, this approach is not necessarily the best, or the most pragmatic.
ESU and other related concepts
At global level, other Countries follow different approaches and several are trying to protect so called “Independent Conservation Units” (ICU), rather than species, at least as the latter are normally defined. In the USA, for instance, the “Endangered Species Act” (1973, see Waples 1991) makes reference to “species” including any number of “Distinct Population Segments” (DPS), the latter a concept having strongly pragmatic basis and often explicitly refusing any theoretical definition, at least in its application (Pennock & Dimmick 1997). The only prerequisite of a DPS is that it should be reproductively at least partially isolated by some physical barrier. Some authors or State Agencies make reference in this respect to ill-defined Ecologically Significant Units, such as the salmons living in a given lake, or “distinct” riverine system. Taken in this way, DPS can easily encompass all conservation needs, from protecting viable local populations of periodically harvested species (shooting, angling), to preventing local extinctions, on a State by State, or even County by County, basis.
Of course, in a man-dominated “Athropocene” (Settele & Spangenberg 2013), the “unit” actually and practically managed can only be the individual population. Ideally, in case resources were unlimited, we might try and protect all populations of any given species, but this scenario is often not realistic and we need to focus on a restricted number of local populations. But, then, how to select them?
The Evolutionarily Significant Unit (ESU), conceptualised by Ryder (1986) as a conservation unit below the species level, but theoretically applicable to a wide range of taxa, may offer some answer. Indeed, the ESU notion was conceived to provide a theoretical background for prioritising taxa for conservation purposes and in the face of economic constraints, as well as of the inability of taxonomy to reflect apparent genetic diversity (Moritz 1994a,b). On this basis, an ESU might be fully identical with a “species” (i.e. a species encompasses one evolutionary lineage), or a “species” could be composed of multiple ESUs. On the other hand, an ESU lineage might comprise single/multiple populations, as well as groups exchanging a degree of gene flow such as meta-populations, this being always dependent on specific life histories (Fraser & Bernatchez 2001). Similarly to the “species” concept, several and sometimes contrasting definitions have been proposed for ESU in the course of time (Ryder 1986; Waples 1991; Dizon et al. 1992; Moritz 1994a,b; Vogler & Desalle 1994; Fraser & Bernatchez 2001; de Guia & Saitoh 2007). All of these definitions, however, aim at defining an identical “entity” i.e. “segments of species [viz. evolutionary lineage] whose divergence can be measured or evaluated by putting differential emphasis on the role of evolutionary forces at varied temporal scales” (Fraser & Bernatchez 2001).
Some current definitions are summarized in the Table I.
[Insert Table 1 about here]
The ESU notion has gained scientific support and has been adapted or linked to various criteria and scenarios. Some authors, in fact, have suggested that obtaining a fixed and universal definition of ESU, valid across all species, may not be feasible (e.g. Fraser & Bernatchez 2001). Since all ESU definitions possess both strengths and bits of weakness, authors have argued that differing approaches may work more efficiently than others, depending on cases and circumstances. This implies that designating ESUs should be done flexibly, on a case-by-case basis (Fraser & Bernatchez 2001). Traditionally, judgements on how distinctive a population should be, before it becomes eligible for being recognised as an ESU, were based on ecological, as well as on variously measurable genetic information, thereby trying to take into account its effective evolutionary distinctness (see Ryder’s but also Crandall et al. 2000 definitions in Table I). Definitions by some other authors (Waples 1991; Dizon et al. 1992; Vogler & Desalle 1994; Bowen 1998) tend to overlap with subspecies concepts. Some recent works, taking into account the ever increasing availability of genetic data, have suggested, or sometimes even tended to force, the adoption of criteria exclusively based on molecular phylogenies, while largely ignoring all other otherwise measurable adaptive components (Avise 1994; Moritz 1994a,b). Researchers are often prompted to assess supposedly neutral genetic variation, more or less combined with as supposedly adaptive nuclear DNA variation. Here we have a theoretical issue, because although mainstream notions of speciation mechanisms include evolution of separate adaptations in allopatric or peripatric conditions (see Mayr 1963; Provine 2004 for reviews), the ESU concept, which implies demonstrable adaptation, does not necessarily overlap, or can be applied, to every separate segment in a phylogenetic tree, whose divergence may be a consequence of other genetic mechanisms, unless otherwise demonstrated. Strictly speaking, populations or taxa apparently characterised only by genetic divergence should better be considered Conservation Significant Units (see Yuan et al. 2011), rather than ESUs.
We will now analyse some case studies, trying to show some strengths and weaknesses of ESU concepts, which, since our scientific work has mainly dealt with Lepidoptera, will be drawn from this Insects’ Order.
The ESU: Some insights from butterflies studies
The field of insect conservation is littered with enormous challenges (Stewart et al. 2007). Among insects, butterflies possess well-known ecological preferences and respond to the action of drivers of change even more strongly and faster than other well-studied taxa, such as birds and vascular plants (Warren et al. 2001). Thus, butterflies represent a good indicator group for other insects taxa (Thomas 2005; van Swaay et al. 2010). More and more frequently, local extinctions have occurred even in nature reserves, where species are supposedly not facing any resources’ shortage (New et al. 1995; Bonelli et al. 2011), so that many early attempts to conserve declining butterfly species have failed because of our inadequate understanding of their biology and causes of decline (Thomas et al. 2009). Among Lepidoptera, many taxa of conservation concern include entities which significantly diverge for molecular, morphological (i.e. wing patterns), ecological (i.e. differences in phenology etc.) and/or behavioural features.
When taxonomic variability does not reflect biological diversity in butterflies
Current molecular studies have surely done a lot to improve our understanding of butterfly evolution, speciation, taxonomy and conservation priorities. A proportionally great number of ‘double’ (in one case triple) species have been identified by these methods and butterflies may be second only to Amphibians, or sometimes Mammals, in this respect. Cases such as those of Zerynthia polyxena ([Denis & Schiffermüller], 1775)/Z. cassandra Geyer, ; Pieris daplidice (Linné, 1758)/P. edusa (Fabricius, 1777), Leptidea sinapis (Linné, 1758)/L. juvernica Williams, 1946/L. reali Reissinger, ; Melitaea phoebe (Goeze, 1779)/M. ornata Christoph, 1893; Melitaea athalia (Rottemburg, 1775)/M. nevadensis Oberthür, 1904; Coenonympha pamphilus (Linné, 1758)/C. lyllus Esper, ; Polyommatus icarus (Rottemburg, 1775)/P. celinus Austaut, 1879 (e.g. Porter et al. 1997; Dapporto 2010; Dincă et al. 2011a,b; Tóth & Varga 2011; Zinetti et al. 2013), to cite some of the most recently demonstrated, do not fit well into the ESU paradigm. These species, irrespectively of their generally strong genetic differentiation, apparently show too scarce ecological distinctness, within the same pair, to allow us to classify them as separate ESUs. It seems almost as though two or more species may form a single ESU, at least on the basis of the more adaptively restrictive definitions, as well as of our current understanding of their biology.
The case of Maculinea butterflies
The five European Lycaenids of the genus Maculinea van Eecke, 1915 (M. arion (Linné, 1758), M. teleius (Bergsträsser, 1779), M. nausithous (Bergstrasser, 1779), M. alcon ([Denis & Schiffermüller], 1775), M. rebeli (Hirschke, 1905)) are among the most well studied myrmecophilous butterflies and have become a model system for studies in the field of evolutionary ecology (Thomas & Settele 2004; Barbero et al. 2009b; Settele & Kuhn 2009).
Maculinea butterflies are obligate social parasites, since their larval survival depends both on the presence of specific food plants and specific Myrmica Latreille, 1804ant species (Thomas 1980). After spending 10-15 days feeding on a species-specific food plant, Maculinea larvae drop to the ground and wait until they are found and carried into an ant nest by a Myrmica worker (Elmes et al. 1991a; Akino et al. 1999; Elmes et al. 2002; Thomas 2002). Adoption of the parasite caterpillars by the host ants is mediated by chemical deception (Akino et al. 1999; Schönrogge et al. 2004; Nash et al. 2008; Fürst et al. 2011). Once in the ant colony, butterfly larvae make use of different feeding strategies: Maculinea alcon and M. rebeli are called ‘cuckoo feeders’ because their larvae are fed directly by the ant workers by trophallaxis (Elmes et al. 1991b; Thomas & Elmes 1998), while M. arion and M. teleius are ‘predatory species’ and directly prey on ant brood. The alimentary strategy of Maculinea nausithous has not yet been fully clarified, but some authors suggest the coexistence of both “cuckoo” and “predatory” strategy or the predominance of the “cuckoo” behaviour (Thomas & Settele 2004; Patricelli et al. 2010). Irrespective of the species, Maculinea larvae spend 11 or 23 months inside their host colonies mimicking their host ants by both chemical and acoustical cues (Schönrogge et al. 2004; Barbero et al. 2009 a,b; Barbero et al. 2012; Witek et al. 2013).
In the past few decades, all Maculinea species have experienced severe declines over most of their ranges (Thomas 1995; Wynhoff 1998; Thomas & Settele 2004; Thomas et al. 2009). Consequently, they have attracted wide public attention, owing to their extraordinary life history and endangered status. At least partially as a consequence, Maculinea butterflies are mentioned in Annexes II & IV of the European Habitats Directive.
The Maculinea alcon-rebeli debate
Among cuckoo species, M. alcon and M. rebeli are sometimes considered one of the best examples of ‘ecological races’ in butterflies, since they inhabit very distinct biotopes and show distinct ecological preferences (Descimon & Mallet 2009).
Historically, these two types have been considered either as distinct species, i.e. Maculinea alcon and Maculinea rebeli, or as subspecies (or “Formenkreis”) of a single species (M. alcon alcon and M. alcon rebeli). Although their adult morphologies and genitalic characters are indistinguishable (Sibatani et al. 1994; Pech et al. 2004), their separation into two species has been accepted by many authors, mainly on ecological criteria (e.g. Munguira 1989; Thomas et al. 1989; Elmes et al. 1991a,b; Munguira & Martin 1999). The separation was made principally according to habitat characteristics, their initial larval food plant, and the host ant species. M. alcon occurs on wet meadows dominated by Molinia coerulea (L.) Moench, where females primarily oviposit on Gentiana pneumonanthe L., while M. rebeli’s main food plant is Gentiana cruciata L. and adults inhabit dry grasslands (Thomas 1995). Adaptation to different gentian species may also explain two important behavioural differences among populations of M. alcon and M. rebeli, i.e. the variation in caterpillars’ growth rate and in adult phenology. Sielezniew and Stankiewicz (2007) demonstrated that M. rebeli caterpillars using G. cruciata acquire about half of their final body mass before overwintering, while those of M. alcon adapted to G. pneumonanthe, gain most of their weight in the late spring of the following year. This is apparently tuned with the phenology of the two host plants. G. cruciata is in the appropriate phenological state for female oviposition a month earlier than G. pneumonanthe. As a consequence, M. alcon caterpillars have to accelerate their development after diapause to obtain the optimal timing of adult emergence (Sielezniew & Stankiewicz 2007).
All across their European distribution, M. alcon and M. rebeli populations use as hosts more than ten Myrmica species (see Thomas et al. 1989; Elmes et al. 1991a,b, 1994; Akino et al. 1999; Schlick-Steiner et al. 2004; Sielezniew & Stankiewicz 2004; Steiner et al. 2003; Tartally et al. 2008; Nowicki et al. 2009). Such a relatively large number of host switches, together with observations that individual populations are typically highly species-specific with respect to ant association, suggest that cuckoos may be undergoing rapid ecological divergence (Elmes et al. 1994; Meyer-Hozak 2000; Als et al. 2001, 2004; Steiner et al. 2003; Witek et al. 2006).
Our findings support the existence of a clear separation of the two population groups. Italian populations of M. rebeli and M. alcon are characterized by marked phenological differences, being respectively on the wing from mid June till mid July, and from end July to end of August. Choices made by adult butterfly during oviposition provide even stronger evidences of sharp ecological separation between M. alcon and M. rebeli, as well as their use of host ant species (Czekes et al. 2013) (see the following paragraph).
Of course, the matter boils down to whether or not these two population types form separate clades. Even though some unpublished data obtained by K. Schönrogge and L. P. Casacci from larval epicuticular hydrocarbons would support their differentiation, other molecular studies based on sequence data of nuclear and mtDNA, or on allozymes (Als et al. 2004; Pech et al. 2004; Thomas & Settele 2004; Bereczki et al. 2005; Fric et al. 2007; Pecsenye et al. 2007), have failed to find evidence for a clade-level separation between M. alcon and M. rebeli. An important consequence of the unresolved taxonomic status of these two groups of populations is that the Appendixes and Annexes to the Bern Convention and to the EU Habitats Directive do not list them among species threatened of extinction in Europe (see also Kudrna et al. 2011). Habitat patches of both population types are becoming more and more isolated because of recent landscape fragmentation, generally due to natural forestation, owing to the abandonment of previous extensive agricultural practices and light grazing (van Swaay & Warren 1999). M. alcon is more severely threatened by the southern limits of its range, and perhaps especially in Italy, due to climate change and habitat degradation concomitant to the sinking water table. In north-east Europe, in contrast, M. rebeli is more vulnerable than M. alcon and needs urgent conservation actions, which should obviously be different from those for M. alcon.
A recent molecular investigation on 16 M. alcon and M. rebeli populations along ca. 700 km of the north-eastern edge of their distribution in Poland and Lithuania (Sielezniew et al. 2012) reopened the debate on the conservational status of these two taxa. As for previous studies, a sequence analysis of the nuclear EF1-a gene was insufficient for establishing an exact taxonomic classification of M. alcon and M. rebeli, but some microsatellite data were consistent with ecological host races. Combining EF1-a results and microsatellite information, the authors suggested the existence of at least three evolutionary significant units (ESUs) as defined by Crandall et al. (2000), corresponding to the north-eastern populations of M. alcon and the two geographically separated host races of M. rebeli, each of which would deserve specific conservation measures.
Host ants’ diversification in Maculinea populations
In the case of an obligate myrmecophilous species, the tight association with its host ant species may have forced selection for more locally adapted populations.
Since they inhabit ants’ brood chambers and become highly integrated with their host society, cuckoo species show highly specific interactions with their host ants. They also receive frequent grooming and are fed by the nurse ants mainly by trophallaxis (cuckoo feeding). Individuals passing the initial period of integration usually survive well with any Myrmica ant species, so long as the colony remains well fed. However, if the colony experiences food shortages or any other similar stress, cuckoo species survive well only with their own specific host, while in non-host colonies, parasite larvae are killed (Elmes et al. 2004). The high level of host specificity is explained, therefore, by underlying integration mechanism.
On leaving its food plant, M. rebeli secretes a simple mixture of surface hydrocarbons that weakly mimic those of its host Myrmica ant, but it is sufficiently similar to all other Myrmica species for the larvae to be quickly retrieved by the first ant worker coming by. After adoption, the intruding larvae successfully integrate within colonies of the model host species, by synthesizing additional hydrocarbons that more precisely mimic their Myrmica host (Schönrogge et al. 2004; Witek et al. 2013). By contrast, caterpillars adopted within nests of other Myrmica species suppress their secretions and rely on the passive acquisition of their current host colony odour (Schönrogge et al. 2004). Acquired camouflage alone, however, is an insufficient mechanism to survive periods of stress or deprivation inside the colony, when worker ants become more discriminating (Elmes et al. 2002).
The main cost of the cuckoo lifestyle is that increased specialization restricts each social parasite to a smaller, regional part of its host range. Thus, the host specificity pattern observed in M. rebeli and M. alcon is extremely complex, as a consequence of local adaptations. For instance, studies on M. rebeli from the Pyrenees show that its populations restrictively exploit colonies of M. schencki Emery, 1894 while eastern M. rebeli populations (mainly in Poland) use both M. sabuleti Meinert, 1861 and M. scabrinodis Nylander, 1846 (Thomas et al. 2005b, 2013). Thomas et al. (2013) have more recently suggested that this host shift could be a trace of a major difference in the chemical profiles, enabling each social parasite to infiltrate and exploit even very different Myrmica host societies. Extreme specialization makes each population incompatible for the survival with another’s host species.
Analysing host specificity patterns in the Italian peninsula, we found that Myrmica schencki is the ant species most frequently used as “primary” host (sensu Thomas et al. 2005a) by all the M. rebeli populations investigated. In some cases, however, a shift towards ‘new’ Myrmica species was observed. Populations where the parasite is hanging in the balance between two host species may be interpreted as coevolutionary hot spots (Thompson 2005), where differentiation is in progress (Casacci & Barbero unpublished data; de Assis et al. 2012) (Figure 1).
[Insert Figure 1 about here]
A field study (Elmes et al. 2004) and an analysis of pre-adoption chemical profiles (Nash et al. 2008) suggested that similar differentiation may have evolved between the main European form of M. alcon, which exploits Myrmica scabrinodis, and that of Scandinavia and the Netherlands, which is adapted to Myrmica rubra/M. ruginodis.
Ants association is therefore a double-edged sword for the conservation of these lycaenid butterflies, since it has promoted rapid rates of diversification, thereby creating a mosaic of overlapping ants and plant hosts, and has produced small, isolated non inter-exchangeable populations. This condition can lead to speciation, but if fragmentation increases in the face of anthropogenic disturbance and habitat loss, the risk of local extinction may also dramatically increase. The case of Maculinea arion in the United Kingdom remains emblematic: the species went extinct in 1979 because modest changes in grazing regimes and vegetation structure caused the host ant to be replaced by unsuitable congener species, unable to support the parasite’s caterpillars (Thomas et al. 2009).
At molecular level, local host ant adaptations would be detectable only by markers linked with genes under selections by specific aspects of social parasitism. While waiting for geneticists to identify these markers, we are convinced that it is extremely important that myrmecophilous insect populations exploiting different host ants are recognised as separate Evolutionary Significant Units.