The status of the European crayfish represents a paradigmatic example of the worldwide freshwater biodiversity crisis. Currently, there are six native European crayfish species: the white-clawed crayfish (WCC) (Austropotamobius pallipes species complex, see below), the stone crayfish (SC) (Austropotamobius torrentium), the noble crayfish (NC) (Astacus astacus), the narrow-clawed crayfish (NCC) (Pontastacus leptodactylus), the thick-clawed crayfish (TCC) (Pontastacus pachypus), and the recently described Austropotamobius bihariensis (Pârvulescu, 2019). These crustaceans have experienced a rapid decimation due to overexploitation, water pollution, flow modification, habitat destruction, habitat and populations fragmentation, invasive species, and especially the emerging crayfish plague disease caused by the pathogen Aphanomyces astaci (Jussila et al., 2021 for review).

As a consequence of this decline, several crayfish species are included in one of the IUCN Red list categories (e.g., WCC and NC are listed as endangered and vulnerable, respectively) with a declining population trend (IUCN, 2021). Thus, there has been a recent interest in studying the patterns of genetic diversity in the European crayfish species to establish sound management and conservation plans (Schubart and Huber, 2006; Akhan et al., 2014; Jelić et al., 2016; Bláha et al., 2017; Schrimpf et al., 2017; Lovrenčić et al., 2020). This is especially true for the WCC in the Iberian Peninsula since this species has experienced a drastic decline for the last 45 years (Füreder et al., 2010). The taxonomic status of WCC, furthermore, is complex. Initial analyses based on nuclear DNA supported the existence of one species (Chiesa et al., 2011; Scalici and Bravi, 2012) while analyses based on mitochondrial DNA (mtDNA) and allozymes suggested that Austrapotamobius pallipes actually represents a species complex. This complex was proposed to comprise two species, A. pallipes sensu stricto and A. italicus, which in turn includes four subspecies: A. i. carsicus, A. i. meridionalis, A. i. italicus and A. i. carinthiacus (Fratini et al., 2005).

The WCC populations of the Iberian Peninsula have been referred as the westernmost part of the A. i. italicus lineage. They were thought as an introduction from the Italian Peninsula (Vedia and Miranda, 2013; Clavero et al., 2016), although they are now considered of autochthonous/natural origin (e.g., Beroiz et al., 2008; Diéguez-Uribeondo et al., 2008; Pedraza-Lara et al., 2010; Matallanas et al., 2011, 2016). Historically, its distributional range in the Iberian territory covered most part of this area until the introduction of two North American invasive crayfish, Procambarus clarkii and Pacifastacus leniusculus, which are chronic carriers of the crayfish plague pathogen (Alonso et al., 2000). Since the introduction of this pathogen, at least 80% of the original European populations have disappeared (Füreder et al., 2010), and the trend of the remaining Iberian populations is still decreasing (Aldabe et al., 1991; Temiño and Sáez-Royuela, 1998). The remaining Iberian WCC populations, and their genetic diversity, thrive in isolated mountainous creeks or in inaccessible brooks. Initial studies analyzing the genetic diversity of the Iberian WCC were based on mitochondrial regions, e.g., 16S rRNA and Cytochrome Oxidase Subunit I (COI). These investigations did not find significant levels of genetic diversity among the populations studied (Santucci et al., 1997; Grandjean et al., 2000, 2001, 2002; Trontelj et al., 2005). However, subsequent studies evidenced higher levels of genetic diversity and a perceptible genetic structure by increasing the number of populations and sample size (Beroiz et al., 2008; Diéguez-Uribeondo et al., 2008), especially when new markers were designed. These studies indicated a strong geographical structure and confirmed the existence of high genetic diversity, similar to the levels found in other European crayfish populations (Pedraza-Lara et al., 2010; Matallanas et al., 2011, 2016). In these studies, however, the distributional range of WCC in the Iberian Peninsula was not fully covered. Although this species does not appear to be naturally distributed in some areas of the Western Iberia peninsula (is absent or very rare in the acid rock areas of Portugal, Galicia, Extremadura and West of Andalucía) (Alonso et al., 2000), there was an overlooked gap in the Central-Western area, still under-sampled and unstudied. The fact that WCC has not been studied within its whole range of distribution, could bias the estimates for overall genetic patterns of diversity for the species in the Iberian Peninsula. Therefore, a study considering the whole range of this species in the Iberian Peninsula is in need, and will provide new insights that will also help clarifying previous claims of a non-native origin of this species in the Iberian Peninsula (Vedia and Miranda, 2013; Clavero et al., 2016).

Thus, the aim of this study was to unravel the potential hidden genetic diversity within the unexplored Iberian WCC populations and to identify patterns of the genetic variation and structure. For this purpose, we have significantly increased the number of sampled populations from previous studies along a wider distributional range in the Iberian Peninsula, applying two highly informative mitochondrial DNA regions, the 16S rRNA and COI genes. Approaching the true diversity and identifying patterns of genetic diversity of the native WCC in the Iberian Peninsula is crucial for maintaining the genetic pool of this endangered species. Moreover, these results will improve the design of conservation programs for the species.

As far as we know, this work represents the most complete and updated approximation of the WCC genetic diversity in the Iberian Peninsula. The results obtained complete and confirm the historical scenario proposed by previous studies on the genetic diversity of the WCC (Pedraza-Lara et al., 2010; Jelić et al., 2016; Matallanas et al., 2016), but also reveal new patterns of genetic diversity and phylogeographic structure. The phylogeographic approach used here, which incorporates samples from previously unexplored areas, has allowed us to find a higher genetic variation in the WCC than previously reported and also identifies new private haplotypes in the Iberian Peninsula. Moreover, the results regarding the origin of the genetic diversity and its phylogeographic structure do not support the hypothesis of an introduction from Italy in the 17th century (Vedia and Miranda, 2013; Clavero et al., 2016), and, instead, strongly suggest a native origin of the WCC in the Iberian Peninsula.

Genetic variation found within Iberian WCC populations is strongly structured geographically. The results from this phylogenetic approach, also supports the scenario proposed by Pedraza-Lara et al. (2010) and Jelić et al. (2016) for the Iberian A. i. italicus populations within the European area. In these scenarios, the Iberian lineages grouped together with populations from Austria, Italy and South France. This group is closely related to the A. i. carsicus, which occurs in Croatia and Italy, and phylogenetically separated from the Central European WCC populations. Following these previous studies and the results in here, the existing genetic differentiation between Central and Southern Europe WCC populations would be of Pleistocenic/Holocenic origin, and might be related to the climate oscillations and glaciations, including the phylogeographic and demographic effects of the ice sheet presence during and after the Last Glacial Maximum (LGM) (Hewitt, 2000, 2004). The fact that Central regions of Europe were glaciated during the LGM led many species to remain isolated in the Southern glacial refugia, i.e., the Iberian, Italian, and Balkan Peninsulas. Besides the presence of the ice itself, the LGM also entailed drastic changes in temperature, droughts, desertification, and large drops in the sea levels changes, that might have shaped the range area in WCC. The distribution of A. i. italicus follows a Circum-Mediterranean distribution. This distribution is also found in other freshwater species (Perea et al., 2010) associated with the isolation in glacial refugia during the LGM, and/or with ancient paleogeography events. For instance, the Alps orogenesis during the late Miocene until the Pleistocene had isolated the Iberian Peninsula from Central Europe and prevented most Mediterranean freshwater species, such as the European cyprinids, to move northward (Zardoya and Doadrio, 1999; Perea et al., 2010).

The phylogenetic analyses for the Iberian WCC populations showed a basal polytomy, indicating non-solved phylogenetic relationships among them. Although some well-supported grouping are presented, more molecular markers would be necessary to identify evolutionary lineages and their relationships. These results agree with the previous assignation of Pedraza-Lara et al. (2010) and Matallanas et al. (2016) for the Iberian populations. On the other hand, we confirmed the utility of these mitochondrial markers that revealed variability patterns not found in previous studies (Toon et al., 2009). The combination of a greater sampling effort and the use of these two regions revealed a total of 35 haplotypes, 16 of which were new for the Iberian Peninsula, while Matallanas et al. (2016) found 19 haplotypes using the same 16S and COI genes.

In addition, we found a strong phylogeographic structure in the Iberian populations. Previous studies designated two phylogeographic areas, while we found three genetically and geographically differentiated areas. Thus, we found two areas in the North (North-Central and North-Western) represented by two main haplotypes, H24 and H30, respectively. This matches with the differentiation and structure found in other terrestrial and aquatic species, such as Salamandra salamandra (García-París et al., 2003), Ichthyosaura alpestris (Recuero et al., 2014) or Lissotriton helveticus (Recuero and García-París, 2011). During the LGM, strong range shifts and bottlenecks occurred and they probably played a key role in Iberian WCC populations. Although the Northern regions of the Iberian Peninsula were mostly covered with ice, glaciers remained confined within the mountain systems and did not reach the surrounding lowlands (Oliva et al., 2019). As a consequence, a number of unglaciated habitats are believed to have acted as refugia for several species (Gómez and Lunt, 2007). The identification of private haplotypes in this area could indicate that these WCC populations were geographically limited and isolated in these regions, as it occurred to other species (Hewitt, 2000). A significant postglacial expansion of these endemic haplotypes might have occurred from several populations during favorable climate periods (Hewitt, 2004). Moreover, the genetic results and the shape of the haplotype network indicated an evidence of recent demographic expansions within these two groups. These results are consistent with those of Matallanas et al. (2016), in which molecular estimations dated last WCC population expansion back to Pleistocene.

The third phylogeographic area that comprises Central-Eastern and the westernmost Iberian populations, suggested another expansion event. The evidences from the significant results obtained in the genetic diversity analysis, as well as the star-like haplotype network showed another demographic expansion, and also reflect the possibility of another LGM-refugium. The Central-Eastern area is also represented by two main haplotypes, H1 and H16. During the LGM, these populations could have persisted in Southern areas of the Iberian Peninsula in absence of geographical or climate barriers. As evidenced by the analyses and the samples studied, these two main haplotypes could represent ancestral haplotypes: H1 (the most frequent), and H16 (only separated by one mutational step from H1). Besides, the H1 has been found in other European populations (Jelić et al., 2016; Matallanas et al., 2016). The Central-Eastern area could have had an earlier expansion than that of the Northern areas, since this area was not covered by ice during the LGM and thus, in absence of barriers, crayfish could have expanded over long distances to some extent (Robinson et al., 2000), and a better climate would have favored it. However, the unequal sampling of the specimens within the studied populations seem to have favored higher frequencies of the H1 and H16 haplotypes, so this statement must be taken with caution. Furthermore, the low genetic diversity found within the WCC southern areas could be explained by the introduction of the North American crayfish species P. clarkii and P. leniusculus during the 1970s (Alonso et al., 2000; Martín-Torrijos et al., 2019). These North American crayfish are natural carriers of the crayfish plague pathogen A. astaci (Martín-Torrijos et al., 2021) and responsible for the decline of the Iberian WCC until nowadays. In particular, by the end of the 1990s the Southern WCC populations almost had disappeared in the Iberian Peninsula (Alonso et al., 2000; Martín-Torrijos et al., 2019).

The current genetic diversity found in Iberian Peninsula may have been shaped by the LGM, as suggested above and in previous studies. The extensive sampling of this study has allowed us to find the greatest haplotype and nucleotide diversities so far reported for the Iberian WCC populations (Matallanas et al., 2016). We should point out, however, that 27 populations were monomorphic. These monomorphic populations were represented by the four most common haplotypes (H1, H16, H24, and H30). This agrees with the strong phylogeographic structure found. In contrast, the most diverse phylogeographic area was the Central-Eastern. This contains populations highly diverse that are located in the provinces of Burgos, Teruel and Huesca. This high genetic variation found within these populations coincides with the historical records of dense populations of WCC described (Alonso et al., 2000). This abundance evidences the importance that crayfish fisheries used to have in local economies of rural Iberian areas. By 1964, Spanish legislation already had regulated the size, amount of crayfish and the fishing gear allowed for the fishing activities (Torre Cervigón and Rodríguez Marqués, 1964). This regulation might have helped to avoid the overexploitation of the resource, maintaining most of the genetic diversity that remains nowadays. Although recent human translocations might have influenced the current Iberian crayfish distribution, there is, still, a strong phylogeographic structure. During the past years, several authors have suggested that the origin of the Iberian populations might have been the result of an Italian crayfish translocation during the 17th century (Vedia and Miranda, 2013; Clavero et al., 2016). However, the results obtained in the present study show a greater genetic diversity than that described in previous studies, and indicate that is highly structured and difficult to attribute to a 17th century translocation from Italy. Therefore, our results support the native origin of the WCC in the Iberian Peninsula. Moreover, the rapid impact of the crayfish plague during the 1970s’ dramatically reduced the number of Iberian WCC populations in less than 2 years (Alonso et al., 2000). These massive declines might have extinguished highly diverse WCC populations, and what we actually come across is a small fraction of its original genetic diversity. In addition, the enormous extinctions due to the crayfish plague may have driven Iberian WCC populations to suffer inbreeding, bottlenecks and genetic drift. Currently, the difficulty of obtaining samples due to the threatened status of the WCC made us to use different number of individuals from each of the analyzed populations. The reduced number of samples obtained from some of the populations revealed the appearance of rare private haplotypes in low frequencies. These rare haplotypes may represent a reflect of the remaining biodiversity within WCC populations.

In addition, the delimited phylogeographic areas seem to explain better the genetic diversity of the Iberian WCC populations. The AMOVA analysis showed more genetic differentiation among the three phylogeographic areas and a slight genetic differentiation among populations within these areas. This suggests substantial gene flow among populations from the same phylogeographic areas. On the other hand, the reduced population sites, the geographical distribution, and the pressure over the remaining Iberian WCC populations by the continuous threat by crayfish plague, for instance, might be favoring their isolation and hindering the gene flow between them. Thus, the unique genetic diversity represented by private haplotypes, which is found in low frequencies, would remain in the same populations instead of being transferred to proximal populations.

Current populations are a remain of what Iberian WCC populations used to be. The massive extinction events that Iberian WCC have been suffering during the past 45 years due to the introduction of North American invasive crayfish carrying the crayfish plague (Martín-Torrijos et al., 2019), and intensive harvesting, might have had irreversible effects on the Iberian WCC genetic diversity. Thus, the majority of the ancient Iberian WCC genetic variation might be already extinct. The reduction of the genetic diversity, and consequently their adaptive potential (Boulding, 2008; Jump et al., 2009), could increase the species extinction risk. To preserve the maximum genetic diversity, we recommend that current conservation and management programs for the WCC in the Iberian Peninsula should consider the patterns of genetic diversity found in this study. Thus, we propose that the three phylogeographic areas revealed in this study should be considered as essential management units to preserve the genetic diversity that characterized them. Furthermore, conservation actions that include breeding and re-stocking programs should consider for each specific area not only the most common but also the private haplotypes. This will certainly help to preserve the unique genetic pool from the endangered Iberian WCC populations.

The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found in the article/Supplementary Material.

LM-T contributed to the design, with the laboratory work, and wrote the manuscript. DB, ID, AM, and JD-U contributed to the supervision of the manuscript and wrote the manuscript. All authors contributed to the article and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.