Meta-analysis of spatial genetic patterns among European saproxylic beetles

The literature search was conducted according to PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines in the WoS (Web of Science) and Scopus databases. The search was done exactly as in a review by Kajtoch et al. (2022) but extended to recent publications (2022–2023, till 31.12.2023). The same criteria were adopted to newest paper search with use of the same keywords: (ALL = (Phylogeny OR phylogenetic OR phylogeography OR phylogeographic OR population genetic OR conservation genetic OR landscape genetic OR population genomic OR conservation genomic OR landscape genomic) ALL = (AND beetle OR Coleoptera AND saproxylic OR xylophagic OR saproxylophagic OR cambiophagic OR cambioxylophagic OR xylophagous OR saproxylophagous OR cambiophagous OR cambioxylophagous OR deadwood)). All relevant papers were retrieved according to the following criteria: i) concern species from Europe, ii) available data for numerous sites across the continent with georeferences, iii) available genetic polymorphism metrics (haplotype diversity (Hd) based on mitochondrial DNA markers, and/or heterozygosity (Ho) based on microsatellites). Heterozygosity (observed heterozygosity) was selected as more articles reported this metric than allelic richness (moreover allelic richness is more dependent on sample size). The same was true for haplotype diversity being more frequently available in articles than nucleotide diversity. Due to very few studies based on the genomic data (from next-generation sequencing) (Schebeck et al. 2018, 2019; Mykhailenko et al. (in press)), in this review, we refrained from presenting genetic metrics measured from single nucleotide polymorphisms.

Initially, we planned to analyze intraspecific variability in particular populations or their groups in Europe separately, but due to the large differences in several data sets (examined sites), the analysis was restricted to a species level. For some species with available mitochondrial or microsatellite data showing overall intraspecific diversity in Europe, the general picture of genetic population structure (number and distribution of clades or clusters) was presented. The same was done for the presence of identified refugia across defined regions of Europe and adjacent areas, namely: Scandinavian (Norway, Sweden, Denmark, Finland), North-Western (Ireland, Britain), Western (Benelux, lowland France, and Germany), Iberian (Portugal, Spain), Apennines (S Italy), Balkans (Balkan Peninsula), Alpine (Switzerland, Austria, Slovenia, S Germany, N Italy, SE France), Pannonian-Carpathian-Bohemian (Czech R., Slovakia, S Poland, SE Ukraine, inner Romania, Hungary), Pontic (E Romania, Moldova, S Ukraine, SW Russia), North-Eastern (N Poland, Baltic countries,), Eastern (W, N, central Russia Belarus), Caucasus (Georgia, Armenia, Azerbaijan, SE Russia), and Anatolia (Turkey).

The main challenge was the inconsistency in sampling design across different studies. Sample sizes varied widely, ranging from a few individuals to thousands, and the number of sampled individuals per site was often inconsistent, typically ranging from a few to a dozen. This imbalance was unavoidable, as many papers lacked precise information on the number of individuals sampled per site or population. Additionally, some studies did not clearly define whether a sampling site referred to a specific location or a population within a broader area, further complicating data comparisons. Moreover, sampling only covered a portion of each species' range. Species distribution range percentages were calculated separately for haplotype diversity and heterozygosity using QGIS by dividing the sampled area of each species by the overall distribution range, with data from the Global Biodiversity Information Facility (GBIF) and selected publications (Table 1). The analysis revealed that only 30–50% of each beetle species' range was included in studies, resulting in a similar bias level across all species with available genetic and geographic data. Despite these imbalances, the review and meta-analysis relied on the best available published information.

All statistical analyses were implemented using R (R Core Team 2020, R Studio Team 2015, Wickham, 2016). For genetic diversity (Hd and Ho), we ran separate models for Hd and Ho with species, longitude, latitude, and altitudes and their interactions as predictors. Subsequently, separate models for each saproxylic species separately were performed with longitude, latitude, and altitude as predictors. All models were performed using a generalized linear model (glm function) and Gaussian distribution. P-values were obtained using ANOVA implemented in the car package (Fox and Weisberg 2015). Selection of the best model was made based on the corrected Akaike’s information criteria for sample size (AICc) and weights. Then, we ran a specific model based on the output of the model selection analysis. In case AICc returns “Null model”, we chose the next best model with delta AIC < 2. Relationships between geographical dimensions (longitude, latitude, and altitude) with haplotype diversity or heterozygosity were assessed using linear associations (Pearson, 1896). All boxplots and scatterplots were incorporated using (ggplot2) for visualization. All maps were conducted using QGIS software (version: 3.36.3, 2019).

Thirty-three species of saproxylic beetles were considered for this review. Information on genetic diversity data was available for 10 species (eight with mitochondrial data and five with microsatellites). Moreover, for all 33 taxa general phylogeographic and/or population genetic information were available, which were used for descriptive elaboration of beetle history and distribution of evolutionary units. Among selected beetles were both threatened species (Osmoderma spp., Rosalia alpina, Lucanus cervus, Morimus funereus, Cucujus cinnaberinus, Pytho kolwensis), and common taxa, either of economic value (e.g. Dendroctonus micans, Ips typographus, Ips sexdentatus, Pityogenes chalcographus, Tomicus destruens, Tomicus piniperda, Monochamus galloprovincialis), or considered neutral in forestry (e.g. Cetonia aurata, Oxythyrea spp., Protaetia cuprea, Bolitophagus reticulatus, Rhagium inquisitor, Monochamus sartor, Morimus asper, Anastrangalia spp., Pytho depressus, Pytho abieticola) (Table 1).

Haplotype diversity was available for 12 beetle species. However, for four species, the number of independent sites was too low for any analyses (namely: Monochamus sartor, Monochamus galloprovincialis (Cerambycidae; Plewa et al. 2018a; Vondracek et al. 2018), and Osmoderma barnabita (Scarabaeidae; Melosik et al. 2020). For Bolitophagus reticulatus (Eberle et al. 2021) information about haplotype diversities was provided for groups of populations from a given country, therefore it was not possible to use these data in most analyses. Two pairs of sibling species: Tomicus piniperda and T. destruens (Curculionidae; Kerdelhué et al. 2002), were formerly considered as one, and belong to the species complex. The final list of examined species for haplotype diversities included nine taxa, these are: Lucanus cervus (Lucanidae; Solano et al. 2016), Rosalia alpina (Cerambycidae; Drag et al. 2018), Cucujus cinnaberinus (Cucujidae; Sikora et al. 2023a), Protaetia cuprea (Scarabaeidae; Vondráček et al. 2018), Monochamus sartor (Cerambycidae; Plewa et al. 2018b), Ips typographus (Curculionidae; Stauffer et al. 1999), Pityogenes chalcographus (Curculionidae; Schebeck et al. 2018), Tomicus destruens and T. piniperda (Curculionidae; Faccoli et al. 2005) (Fig. 1, Table 1, Table S3).

Among the identified articles that met the criteria presented above, only five taxa were available with heterozygosity values from many sites: Lucanus cervus (Lucanidae; Solano et al. 2023), Bolitophagus reticulatus (Eberle et al. 2021), Rosalia alpina (Cerambycidae; Drag et al. 2018), Monochamus galloprovincialis (Cerambycidae; Haran et al. 2018), and Cucujus cinnaberinus (Cucujidae; Sikora et al. 2023a) (Fig. 1, Table 1, Table S3).

The significant effect of species, latitude, longitude, altitude, and their interactions on genetic diversity (Hd and Ho) were tested. First, model selection analysis (AIC) was done to keep only the most relevant variables (Table 2, Table S4). The results showed a significant effect of the main variables of species, latitudes, and altitudes on haplotype diversity, and significant effects of species and altitudes along with their interaction in heterozygosity, revealing the leading influence of geographical distribution on genetic diversity, with the highest impact on haplotype diversity (Table 2, Table S4).

Although geographical distribution showed strong significant effects on genetic diversity, the geographical influences of altitude, latitude, and longitude differed between species (Fig. 2, Table 3).

In haplotype diversity, R. alpina expressed the lowest diversity values with strong significant effects of the geographical distribution of latitude, longitude, and altitude, in contrast to P. cuprea which showed the highest diversity values with a weak significant trend of altitude (Fig. 2, Table 3, Table S5). Geographical distribution had no significant effect on Ips typographus (Table S5).

In heterozygosity, M. galloprovincialis and R. alpina showed the lowest diversity values with the more significant geographical effects on M. galloprovincialis in contrast to B. reticulatus, C. cinnaberinus, and L.cervus which showed the highest diversity values (Fig. 2, Table 3), however, geographical distribution has no significant effect on B. reticulatus and Rosalia alpina (Table 3, Table S6).

The correlation scatter plots revealed varying patterns between species, with some exhibiting strong correlations between latitude with Hd, and altitude with Hd and Ho, while others showed no discernible trends.

Along latitude, the following species expressed a decrease of haplotype diversity to the north: L. cervus, R. alpina, C. cinnaberinus, I. typographus, P. chalcographus, and T. piniperda-destruens, although this trend was not significant for all bark beetles (Fig. 3, Fig. S1, S2). In heterozygosity, three species expressed a decrease in diversity values to the north, although significant only in the case of L. cervus. B. reticulatus and R. alpina have similar heterozygosity along latitude (Fig. 4, Fig. S1, S2). Along longitude, L. cervus and R. alpina showed a decrease in haplotype diversity to the west, whereas I. typographus, P. chalcographus, and T. pinniperda have opposite trends, although significant only for T. pinniperda (Fig. 3, Fig. S1, S2). However, for heterozygosity, results revealed either an insignificant decrease or increase in diversity values to the west in R. alpina (Fig. 4, Fig. S1, S2).

With increasing altitude only one species showed a decrease in haplotype diversity, which was P. cuprea. L. cervus, C. cinnaberinus, R. alpina, T. piniperda-destruens and P. chalcographus showed a significant positive correlation for altitude (Fig. 3, Fig. S1, S2). For heterozygosity, L. cervus and C. cinnaberinus showed a significant positive correlation with altitude, and M. galloprovicialis showed the same trend although not significant (Fig. 4, Fig. S1, S2).

Information about population structure with the presence of distinct genetic units (clades) was available for 10 species and 7 species complexes.

The largest number of these taxa have genetic information from following regions: Alpine (18), Balkans (16), Appennines (15), Scandinavia (15), Pannonian-Carpathian-Bohemian (13), Western (12), Eastern (11), North-Western (11), North-Eastern (9), Iberian (9), Anatolian (8), Pontic (6), and Caucasus (3) (Fig. 5a).

The highest number of evolutionary units was found in the Balkans (42 for 16 species), followed by Apennines (34 for 15 species) and Alpine (34 for 18 species). The lower number of clades was found in Pannonia-Carpathians-Bohemia (25 for 13 species), Scandinavia (24 for 15 species), Western (21 for 13 species), followed by Eastern (17 for 12 species), North-Eastern (16 for 9 species), North-Western (13 for 11 species), and Anatolia (11 for 8 species). The lowest number of clades was found in Iberia (11 for 9 species), Pontic (7 for 6 species), and Caucasus (3 for 3 species) (Fig. 5a).

Regarding unique evolutionary units (present in only one region), the most diverse occurred to be both Balkans (16 for 8 species) and Apennines (16 for 9 species), followed by Anatolia (7 for 5 species), Iberia (7 for 3 species), North-Eastern (4 for 1 species), Eastern (3 for 1 species). Single unique clades were found in Alpine, Pontic, and Caucasus (Fig. 5c).

The largest number of these taxa have genetic information from the following regions: Western, Apennines, Balkans, Alpine, and Carpathian (6 in each), followed by Scandinavian (5), North-Western, North-Eastern, Eastern (4 in each), Iberian and Pontic (3 in each), and Caucasus (1). No data was available for Anatolia (Fig. 5b).

The highest number of genetic clusters was found in the Balkans (15 for 6 species), followed by Alpine (14 for 6 species), Pannonia-Carpathians-Bohemia (13 for 6 species), Western (12 for 6 species), North-Western (11 for 4 species), Apennines (10 for 6 species), Scandinavia (9 for 5 species), North-Eastern (7 for 4 species), Eastern (5 for 4 species), Iberian (5 for 3 species), Pontic (4 for 4 species) and Caucasus (3 for 1 species) (Fig. 5b).

Regarding unique clusters (present in only one region), the only cases reported were from the Balkans (2 for 2 species), Apennines (2 for 2 species), Iberia (1), and Western (1) (Fig. 5d).

Regarding mtDNA, in the Balkans, there were identified 30 such refugia (for 16 species, in some taxa more than one), followed by Apennines (21 for 15 species). In Iberia we found 8 refugia (for 7 species in each), in Anatolia 7 refugia (for 8 species), and in Eastern 6 such areas (for 6 species). In Alpine, Caucasus and Pontic there were 2 refugia (for 2 species in each), and single ones in Pannonia-Carpathians-Bohemia and North-Eastern, whereas there were no refugia detected in Scandinavian, North-Western, and Western (Fig. 5c).

Regarding microsatellites, in the Balkans we identified 7 such refugia (for 6 species, in some taxa more than one), followed by Apennines (6 for 5 species), and Eastern (3 for 3 species). In the Alps and Iberia we identified 2 refugia (for 2 species in both), and a single refuge in Pontic and Western (Fig. 5d).

Determination of higher genetic diversity for south-European populations of the majority of saproxylic beetles is rather an obvious result considering the history of forest-dwelling species since glacial times (Hewitt et al. 1999; Hagge et al. 2019; Taberlet et al. 1998). This concerns mostly taxa associated with deciduous trees which were forced to retreat to the Mediterranean Basin during the Pleistocene and spread to the north during climate warming in the Holocene (Jalut et al. 2009). Not only is the higher genetic diversity observed in southern populations consistent with this pattern, but the presence of distinct clades or clusters and identified refugial areas in Mediterranean peninsulas further supports it. The Balkans have historically served as refugia for many saproxylic beetles, a role that continues to be prominent today (Taberlet et al.1998). Also, the Apennines played a substantial role in sheltering populations of wood-dwelling beetles during glaciations, although the Alps formed a barrier in the expansion of some of them to the north (Drag et al. 2018). Other areas acting as refugees, but on a lower scale, were Anatolia, Iberia, and Eastern Europe (Bilgin et al. 2011; Varga et al. 2009). This begs the question, could the fewer identified clades and refugia in areas such as Eastern Europe be attributed to limited genetic data available for these regions, or possibly to the extinction of local populations? This discrepancy may also be influenced by historical factors, such as extensive deforestation in regions like Iberia (García-Ruiz et al. 2016). Additionally, the absence of a clear south-north pattern in these areas contrasts with taxa originating from the east, particularly those inhabiting the boreal zone with isolated populations in European mountains (Varga et al. 2008). Unfortunately, some of the observed patterns could be biased due to insufficient data in available published research.

South to north decrease of genetic diversity in saproxylic beetles is not the only geographic pattern that can be observed. Relatively many wood-dwelling Coleoptera also express decreased genetic diversity toward their western populations. It is not the rule, as not all saproxylic beetles have ranges covering eastern Europe—this particularly concerns species associated with temperate deciduous trees having ranges not exceeding Central Europe. Surprisingly many taxa express higher genetic diversity in their central, eastern and south-eastern—European populations. One of the reasons for this pattern could be the presence of beetles belonging to various evolutionary units in this part of Europe as a result of contact between populations having refugia in the south-west and south-east Europe (a phenomenon commonly reported in European biogeography, (Hewitt et al. (2011). This is likely true for various tenebrionid beetles (Fattorini et al. 2012). Another explanation is that beetle populations in eastern and southeastern Europe are in better condition, being more widespread, abundant, and vital, due to inhabiting higher-quality forests (Karpinski et al. 2021). These forests are either natural or closely resemble natural forests, with a substantial amount of dead wood available (Reif and Walentowski, 2008). Indeed, this part of Europe is known for remnants of primeval forests, such as the Białowieża and Pripyat forests in the lowlands, and the Carpathians and the Balkans in the mountains. Additionally, forest management in eastern and southeastern Europe is generally less intensive than in western, southern, and northern Europe (Angelstam et al. 2011). There are known exceptions like dehesas in the Iberian Peninsula (Ferraz-de-Oliveira et al. 2016), but these are woodlands scattered over agricultural landscapes, and many European saproxylic beetles are not present there. Although there are forests in eastern and southeastern Europe that are planted and heavily logged, many areas are still covered by semi-natural mixed forests that support rich populations of various species, including saproxylic beetles. While remnants of such semi-natural forests also exist in western and southern Europe, they are mostly confined to small, highly isolated mountainous areas. The fragmentation and isolation of these natural, heterogeneous, and mixed forests likely contribute to the generally lower diversity of saproxylic beetle populations in western and southern Europe. Timber logging in these regions is extensive and most forests in these regions are artificially planted, resulting in a significant deficiency of dead wood, as it is actively removed from commercial forests (Frank et al. 2009; Horák & Rébl, 2013; Karpinski et al. 2021). The lack of significant genetic diversity observed in some species may be due to sampling biases, as the sampling effort was much more concentrated in western and southern Europe compared to the easternmost regions of their ranges. In C. cinnaberinus, heterozygosity showed no significant correlation with longitude, likely due to the inclusion of a distant Russian population with no or less polymorphism (Sikora et al. 2023a).

Beetle populations sampled at higher altitudes exhibited increased genetic diversity. This increased diversity is likely related to three factors: i) history (phylogeography), ii) altitudinal distribution of species in various regions, and iii) current conditions of forests. The first explanation corresponds to the high diversity of mountain populations in the Mediterranean peninsulas, where refugial populations exist, such as in the Apennines and the Balkans (Karpinski et al. 2021). Indeed, for many saproxylic species, the highest genetic diversity of populations was observed in south-European mountains or the Alps, where species spread after ice-sheet retreated and where evolutionary lineages expanded from southernmost refugia. The altitudinal distribution of species is different in various regions of Europe, and southern areas many saproxylic beetles live on higher elevations simply because forests are distributed higher there (lowland forests in the Mediterranean were cut in ancient times). On the other hand, this pattern was not entirely expected since populations at higher altitudes must endure more challenging conditions, particularly climatic, than those living in lowlands. Such conditions could produce bottlenecks that typically reduce intrapopulation genetic diversity. Here, altitude probably acts more than the past history of populations. The dead wood, a crucial microhabitat for saproxylic beetles, is typically absent except in areas inaccessible to logging. Higher altitudes are usually less affected by forest management due to harsh topography, and numerous protected areas designed in mountain ranges. Some of the most diverse beetle populations have been observed in regions such as the Carpathians and the Dinaric Mountains, although these are outside of the major refugia for forest-dwelling species in Europe (Lassauce et al. 2011). Again, these trends along elevation could be biased by unequal sampling or by low accuracy of determination of sampling sites in examined publications (too low resolution of coordinates).

One of the assumptions was that rare, threatened species have overall lower genetic diversity due to the isolation of their populations caused by forest fragmentation and lack of essential microhabitats such as deadwood, in commercial forests (De Groot et al. 2019). The genetic diversity of common species is generally assumed to remain unaffected by forest management. However, a comparison of available data reveals that this is not always the case, and no consistent patterns emerge among saproxylic beetle species (Edelmann et al. 2023). These differences could be explained by various histories of species formation and expansion, but also by recent global climatic and environmental changes that also affect common taxa (known decline of invertebrates, Rhodes 2019).

There are instances where rare species exhibit lower genetic diversity compared to most other taxa. This is particularly evident in species associated with specific tree species and habitats, such as Lucanus cervus, Cerambyx cerdo, and Osmoderma spp., which depend on old oaks, and Rosalia alpina, which is linked to old beeches, especially at the edges of their distribution ranges (Seibold et al. 2015). In contrast, Cucujus cinnaberinus, despite also being rare, has high genetic diversity. This species inhabits the bark of various tree species, likely enhancing its dispersal abilities (Vrezec et al. 2017). Common beetles typically exhibit relatively high genetic diversity, which reflects their widespread distribution and numerous populations, primarily in commercial forests such as pine or spruce woods. However, there are exceptions to this pattern. For instance, Ips typographus shows relatively low genetic variability within its populations. This species is known for undergoing cyclic outbreaks followed by population collapses, which tend to reduce its overall genetic diversity (Valeria et al., 2016). It is known that some species with large distribution ranges show less genetic diversity, possibly because their greater dispersal ability promotes greater genetic homogeneity. Such a species usually have high mobility which results in intensive gene flow and lack of geographic structure of their populations.

The patterns discussed above are also evident in other wood-inhabiting organisms. Among the best-known arboreal animals are woodpeckers (Picidae), whose European members display similar patterns to those of saproxylic beetles which are their primary prey (Fayt et al 2005). Generalist species like the great-spotted woodpecker (Dendrocopos major) and green woodpecker (Picus canus) exhibit high genetic diversity, but this diversity is unstructured across their European ranges (Myczko et al. 2014). In contrast, distinct evolutionary units in central and east Asia are likely separate species (Perktaş & Quintero 2013). Similarly, the Iberian woodpecker (P. sharpei) was previously considered a subspecies of the green woodpecker (Pons et al. 2011). This unstructured genetic diversity pattern is not observed in saproxylic beetles but may be present in other common, widespread generalist species yet to be examined. Rare specialists associated with deciduous trees, such as the white-backed woodpecker (Dendrocopos leucotos) (Pons et al. 2021) and the middle-spotted woodpecker (Dendrocoptes medius) (Kamp et al. 2019; Schweizer et al. 2022), have distinct subspecies in their Mediterranean and Middle Eastern ranges, differing from their widespread European mainland counterparts. This pattern is shared with saproxylic beetles like Rosalia alpina and Osmoderma spp. (Chiari et al. 2013) The three-toed woodpecker (Picoides tridactylus) shows the least genetic diversity (Zink et al. 2002), mirroring the genetic patterns of its prey, the bark beetle (I. typographus) (Stauffer et al. 1999).

A similar example is found in fungi associated with deadwood, such as the wood decay fungus Fomitopsis pinicola or Gloeoporus taxicola, which shows similar population structure and genetic diversity across different geographical areas (Högberg et al. 1999; Liu et al. 2021). Bolitophagus reticulatus shows similarities in the genetic diversification of populations with its host fungi Fomes fomentarius (Peintner et al. 2019). These similarities are expected, as it has been previously noted that beetles and fungi exhibit many congruences in their taxonomic, functional, and phylogenetic diversities (Thorn et al. 2018). Likely the same similarities could be expected for other saproxylic organisms like moths (Jaworski 2018) or lichens (Rinas et al. 2023), but so far there is a deficiency of phylogeographic and population genetic studies on other deadwood-dwelling species.

In addition to expanding our understanding of historical range changes and contemporary population variability of saproxylic beetles, genetic information about these species as shown here could have significant practical applications. Knowledge of the genetic diversity of “pests” can inform effective management strategies for their populations. This is especially relevant for understanding the causes, mechanisms, and biological consequences of outbreaks, which are cyclic events in the life cycles of economically important species such as bark beetles and certain longhorn or jewel beetles (Valeria et al., 2016; Haack et al. 2010; Kiran et al. 2019). Indeed, ongoing research in this area has shown promising results (https://​genomicsofoutbre​aks.​com/​).

Large-scale genomic data can also be used to trace the ancestral populations of invasive beetles and to identify species that are morphologically similar but genetically distinct (Cui et al. 2022). Such research has already been conducted for some saproxylic beetles that are invasive in Europe or for European species that have invaded other regions of the world (Wondafrash et al. 2016). Additionally, genetics could aid in the biological control of pest populations by identifying the microbial, biological or chemical factors that can sustainably reduce their numbers (Elnahal et al. 2022).

The most commonly used designation to study the genetics of rare and threatened taxa is by studying the distribution of distinct evolutionary units of conservation value, such as evolutionarily significant units (ESUs) (Casacci et al. 2014) and management units (MUs) (Moritz 1994). An ESU is a population of organisms that is considered distinct for purposes of conservation, it needs to be geographically isolated, express substantial reduction in gene flow with other ESUs and usually possess some distinct phenotypic traits. MU is central to the management of natural populations and is crucial for monitoring the effects of human activity on species abundance. ESUs and MUs are essential for identifying populations that require special protection, particularly those that must be preserved to maintain a substantial part of genetic diversity and potentially important adaptive characteristics vital for the survival of populations (Fraser et al. 2001; Moritz et al. 1994). Additionally, understanding gene flow among populations is important for planning the spatial organization of protected forest sites to sustain connectivity among beetle populations (Balkenhol et al. 2017). Knowing the genetic assignment of individuals to populations is necessary for translocations or reintroductions to avoid outbreeding or inbreeding depression in newly established populations (Maschinski et al. 2013).

Apart from species of economic or conservation value, the majority of saproxylic beetles are natural elements of forest ecosystems, playing crucial roles in carbon and nutrient cycling through their participation in the decay of deadwood (Lieutier et al. 2004a, b). Genetics can also be utilized in this context, for instance, to understand which environmental and spatial factors are responsible for sustaining genetic intra- and inter-population diversity. Tools such as landscape genetics are applicable here, although their use for saproxylic beetles in Europe is still relatively new (Bolliger et al. 2010; Manel et al. 2003).

The current study serves as a stepping stone for further studies that could potentially employ next-generation sequencing (NGS) technologies like restriction site-associated DNA sequencing or genotyping-by-sequencing (Elshire et al. 2011; Narum et al. 2013; Peterson et al. 2012) to study saproxylic beetles based on large datasets of genomic information (e.g. sets of single nucleotide polymorphisms) and at a larger scale, shedding light on their evolutionary history, adaptation mechanisms, and responses to environmental changes.

Next-generation sequencing (NGS) is already a widely used tool for determining cryptic taxa, including their larvae or pupae, and for delimiting undiscovered evolutionary units (including new taxa) through barcoding (Pante et al. 2015). Additionally, genomic data obtained through NGS can enhance our understanding of phylogenetic relationships among closely related species of saproxylic beetles (Hajibabaei et al. 2007; Lui et al. 2022).

As we have seen, some threatened species are associated with veteran trees (like Rosalia alpina with beeches, Lucanus cervus and Osmoderma spp. with oaks (Drag et al. 2015). So-called "pests" are usually specialists of particular tree species (e.g. conifers infested by Scolytinae, Avtzis et al. 2019). Therefore, one of the principal future directions should be to incorporate information related to the forest quality and structure into understanding the genetic diversity of saproxylic beetles – the topic that has not been addressed so far, although knowledge about the relations of forest environments with alpha and beta diversity of wood-dwelling beetles is widely studied (Müller, et al. 2015).

Genetic data also show substantial isolation of remote populations particularly in the case of rare taxa like Cucujus cinnaberinus and Elater ferrugineus (Sikora et al. 2023a; Oleksa et al. 2015), therefore suitable forest habitat continuity (or forest cover fragmentation) should be considered in future studies on gene flow among populations of saproxylic beetles.

Similarly, Ips typographus and other bark beetles, known for their cyclic population dynamics, need also be studied using NGS to examine the genetic diversity across different outbreak and non-outbreak phases (Mykhailenko et al. 2023). By sampling individuals from various outbreak regions and time points, we could elucidate how population fluctuations affect the species' genetic structure and adaptive potential.

This study opens up various avenues and gives direction for a comprehensive understanding of the genetic dynamics, ecological roles, and conservation needs in the face of ongoing environmental changes of these enigmatic species.