Advertisement

Hydrobiologia

, Volume 810, Issue 1, pp 177–189 | Cite as

Glacial perturbations shaped the genetic population structure of the endangered thick-shelled river mussel (Unio crassus, Philipsson 1788) in Central and Northern Europe

  • Sarah Feind
  • Juergen GeistEmail author
  • Ralph Kuehn
FRESHWATER BIVALVES

Abstract

The thick-shelled river mussel, Unio crassus Philipsson 1788, was formerly one of the most abundant and widely spread freshwater bivalves among the order of Unionida in Europe, but there is still a lack of information on its genetic diversity and spatial genetic structure. We characterised the genetic constitution of 18 U. crassus populations in Germany and Sweden originating from six major drainage systems using a set of nine microsatellite markers. U. crassus populations from Northern Germany and Sweden revealed similar genetic constitution with high genetic diversity, whereas populations from Southern Germany showed higher proportions of common ancestors and a stronger host-dependent diversification. The structured spatial genetic patterns suggest two glacial refugia and a strong north–south differentiation. The Elbe river basin was likely an important source for recolonisation northwards to Sweden, and populations from the low mountain range in the Danube–Rhine systems acted as a source for recolonisation southwards to the Alps after retreating glaciations. We identified three major conservation units for the analysed U. crassus populations which should be considered in future conservation efforts.

Keywords

Microsatellite Freshwater mollusk conservation Conservation prioritisation Supportive breeding Unionidae 

Introduction

Freshwater mussels represent one of the most severely threatened biota worldwide (Lydeard et al., 2004; Bogan, 2008; Lopes-Lima et al., 2016). Out of the 16 currently recognised Unionida species in Europe, 12 species are on category Threatened or Near Threatened based on the last IUCN Red List assessment. Especially, the nine species of the tribe Unionini are imperiled (Lopes-Lima et al., 2016). Within the Unionini, the species Unio crassus, once considered the most abundant unionid in Europe, declined dramatically in western and central Europe during the second half of the last century (Israel, 1913; Zwanziger, 1920; Hochwald & Bauer, 1990; Engel, 1990; Hochwald, 1997; Lechner, 1999). As a result, U. crassus is listed as Endangered in the IUCN Red List and ranked as Critically Endangered in some European countries (e.g. Germany, Austria, Switzerland). In Germany, the loss of populations was even estimated about 90% in the last few decades, with a total amount of remaining U. crassus specimens of only around 1 million (Zettler & Jueg, 2007). The dramatic declines result mainly from habitat degradation, water pollution and increased predation due to the introduction of invasive species as well as the decline or loss of host fish populations which are an essential component in the life cycle (Taeubert et al., 2012a, b; Stoeckl et al., 2015). In particular, host limitation is likely to play a major role in cases with a lack of juvenile recruitment (Stoeckl et al., 2015). Other reasons for population losses like the direct predation pressure by the invasive muskrat (Zahner-Meike & Hansons, 2001) or river dredging (Zajac & Zajac, 2011) have also been reported to be of local importance.

As filter feeders, freshwater mussels provide important ecosystem services by filtering out large amounts of particulate matter from the water column (Lummer et al., 2016), as well as by their effects on substrate properties due to bioturbation (Boeker et al., 2016; Richter et al., 2016). Their importance for the functioning of aquatic ecosystems makes them ideal target species for conservation of freshwater habitats (Howard & Cuffey, 2006; Geist, 2010; Lummer et al., 2016).

Besides deterministic factors directly influencing endangered species, small and isolated populations face additional risks of extinction such as inbreeding depression and loss of genetic diversity (Frankham et al., 2002). The loss of genetic diversity is likely to reduce the adaptation potential, as well as survival and reproduction of a population, and can thus result in reduced effective population size. In turn, demographic fluctuations, environmental variation and catastrophes can further impact the effective population size and enhance the formation of small and fragmented populations. These processes can ultimately lead to a downward spiral towards extinction, termed the ‘extinction vortex’ (Frankham et al., 2002).

Due to conservation planning in the 1970s and 1980s and the Habitats Directive in 1992, U. crassus came into the focus as a priority target species in freshwater conservation (Lopes-Lima, 2014). The thick-shelled river mussel holds a high legal conservation status and is a priority target species for conservation. Still, there is currently very limited knowledge of its genetic structure except for the first description of microsatellite markers in the species (Sell et al., 2013), and a study about host fish compatibility of U. crassus related to a genetic background (Douda et al., 2014). However, an understanding of spatial patterns of genetic diversity is generally crucial for the development of integrative conservation strategies in freshwater systems (Geist, 2010, 2011, 2015). In aquatic species, the genetic population structure is determined not solely by present-day dynamics, but also by geological history (Hewitt, 2000). Genetic structure of most biota in Europe has been strongly influenced throughout the past two million years when climatic oscillations, implementing several glacial periods, led to several areas of glacial refugia (Taberlet et al., 1998; Hewitt, 2000) and subsequent postglacial recolonisation processes. Putative glacial refugia are inferred to harbour a higher intraspecific genetic diversity in comparison to recolonised surrounding regions. It is assumed that populations surviving in the refugial areas accumulate genetic variation during changing glacial periods, while postglacial recolonisation processes may lead to a lower fraction of genetic diversity in leading edge populations (Hewitt, 1996).

The Danube system is a particularly interesting study region in this context, since it has been reported to have acted as an important refugium for many fish species during glaciation, for example, Bitterling (Rhodeus amarus), Grayling (Thymallus thymallus), European chub (Squalius cephalus) and the European bullhead (Cottus gobio) (Volckaert et al., 2002; Gum et al., 2005; Bryja et al., 2010; Seifertová et al., 2012). Some of these species have overlapping distribution areas with U. crassus and are regionally important host fishes of U. crassus, like the European chub (Taeubert et al., 2012b) or the European bullhead (Lamand et al., 2016). Since the dispersal capacity of unionid mussels is limited to the dispersal of specific host fishes, it is likely that spatial genetic patterns of unionid mussels are reflected in the colonisation history of their hosts. Coevolutionary relationships may lead to congruence of distributional range, especially between parasites and their hosts (Stewart et al., 2010).

Also, from the perspective of glacial refugia and postglacial colonisation pathways, the Danube drainage is considered highly subdivided and heterogeneous, as evident from the plethora of endemic subspecies of freshwater mussels that have been described for this particular region (Nesemann, 1993). The group of recent U. crassus was already represented in the Middle Oligocene in the Northern Alpine foreland (Modell, 1959). Northern Germany is also considered a special area where three different drainage systems (Elbe, Schlei-Trave, Eider) converge on a geographically small area representing a postglacial contact zone. In this area, the maximum extend of the Elster glaciation (400,000– 320,000 years ago) during the Middle Pleistocene reached until the low mountain range of Middle Germany, indicating that temperate-adapted freshwater mussels could not have survived in the whole glaciated North German Plain. For these recolonised regions in Central Europe and Scandinavia, it has been reported that colonising lineages have met several times and formed admixed secondary contact zones with unexpectedly high population genetic diversity (Havrdová et al., 2015). For example, higher genetic diversity in postglacial colonisation regions has been described for the freshwater pearl mussel in Northern Europe, which is likely the result of a high number of founders due to recolonisation from multiple source populations (Geist & Kuehn, 2008). For the primary host fish S. cephalus, the Elbe drainage was a contact zone, resulting in a mix of haplotypes from different European lineages (Durand et al., 1999). The same author suggests that this species survived in multiple separate refugia during glaciation, which remains to be tested for U. crassus.

Furthermore, for many aquatic species present-day population differentiation does not reflect present-day drainage system structure (Geist & Kuehn, 2005; Gum et al., 2005). A recent study on two genetically differentiated U. crassus populations suggests that population-level differentiation is related to differences in physiological host fish compatibility (Douda et al., 2014). Consequently, analyses into the genetic structure of U. crassus populations in Central and Northern Europe are likely to not only allow identification of priority populations for the conservation in this single species, but are likely to also contribute to the general understanding of postglacial colonisation patterns in aquatic species.

The overall objective of our study was to gain information about the genetic constitution of eighteen different thick-shelled river mussel (U. crassus) populations from various drainage systems in Germany and Sweden. Specifically, we hypothesised a strong genetic population structure between geographically isolated drainage systems and, analogously to other declining species of mussel such as Margaritifera margaritifera, a strong influence of genetic drift on population structuring.

Materials and methods

Sampling strategy, DNA isolation and species verification

We sampled a total of 336 specimens from eighteen thick-shelled river mussel populations originating from six major drainage systems of the Danube (five populations), Rhine (three populations), Elbe (four populations), Schlei–Trave (three populations) and Eider (two populations) in Germany and Kävlingeåns (one population) in Sweden, representing the central distribution area of U. crassus (Table 1). A description for all sampled populations including estimated census population sizes, drainage system and sampling site with GPS coordinates is provided in Table 1. Population census sizes are estimates for population sizes based on recent state monitoring results.
Table 1

Sampling sites with drainage and subdrainage system of thick-shelled river mussel populations in Germany and Sweden; GPS coordinates in decimal degree latitude and decimal degree longitude (WGS 84)

Pop ID

Drainage

Subdrainage

Population census size

Latitude

Longitude

S1

Kävlingeåns

Bråån

5,000

55.800

13.641

I1

Eider

Bollingstedter Au

50,000

54.398

09.387

I2

Eider

Mittlere Eider

15,000

54.291

09.998

T1

Schlei-Trave

Schwentine

100

54.195

10.307

T2

Schlei-Trave

Alte Schwentine

50,000

54.185

10.241

T3

Schlei-Trave

Trave

1,000

53.978

10.367

E1

Elbe

Alster

10,000

53.765

10.063

E2

Elbe

Bille

100

53.554

10.314

E3

Elbe

Ilmenau

10,000

53.148

10.467

E4

Elbe

Cederbach

5,000

53.003

12.010

R1

Rhine

Truppach

2,000

49.915

11.389

R2

Rhine

Ailsbach

5,000

49.802

11.331

R3

Rhine

Gießgraben

10,000

49.656

10.368

D1

Danube

Frankenohe

100

49.629

11.846

D2

Danube

Ischler Ache

10,000

47.948

12.415

D3

Danube

Götzinger Ache

500

47.941

12.833

D4

Danube

Haldenseebach

2,000

47.788

11.079

D5

Danube

Staffelseeach

6,000

47.753

11.113

We carried out DNA extraction using the noninvasive method of hemolymph extraction (Geist & Kuehn, 2005). For this method, 1-ml syringes attached to 0.80 × 50 mm 21 G × 2′′ Sterican needle (Braun, Melsungen, Germany) were used for collecting about 0.1–0.2 ml of hemolymph by sticking the needle carefully into the foot of the mussel. We then transferred hemolymph samples to 1.7-ml Eppendorf tubes, cooled at 5°C during the sampling trip and processed immediately in the laboratory. After centrifugation at 14,000×g for 5 min, the supernatant was discarded and DNA of the remaining pellet was extracted with the NucleoSpin® Tissue Kit (Macherey–Nagel), following the manufacturer’s instruction. Sample collection was carried out from 2010 to 2015.

We validated morphological species identification by genetically comparing the fragment length of the ITS region for each sampled mussel. PCR was performed according to Zieritz et al. (2012) with minor changes, i.e. annealing 60 s, elongation 60 s, 0.08 U/μl Taq Polymerase (FIREPol®, Solis BioDyne), 0.3 pmol/μl of ITS-1 primer (forward primer labelled with Cye5) in a total volume of 15 μL. PCR products were analysed on ALFexpress II DNA Analyser and scored with AlleleLinks 1.02 and ALFwin Sequence Analyser 2.1 software.

Microsatellite analysis

For genetic analysis, we applied a system of 9 microsatellite loci (Uc5, Uc7, Uc15, Uc16, Uc19, Uc25, Uc39, Uc69, Uc77) based on a novel set of microsatellite markers developed for U. crassus (Sell et al., 2013). Forward primers were fluorophore-labelled either with HEX, 6-FAM or TAMRA to run a multiplex system. PCR was performed in a total volume of 15 μl composed of 40 ng of genomic DNA, 0.2 pmol/μl of each primer, 0.2 mM dNTP (Solis BioDyne), 2.5 mM MgCl (Solis BioDyne), 1× Reaction Buffer BD (0.8 M Tris–HCl, 0.2 M (NH4)2SO4) and 0.08 U/μl Taq Polymerase (FIREPol®, Solis BioDyne). PCR was carried out on a MasterCycler Gradient Thermal cycler (Eppendorf) with cycling conditions as follows: initial denaturation at 94°C for 3 min; 35 cycles of 94°C for 30 s, 58°C for 30 s, and 72°C for 30 s, before final extension at 72°C for 3 min. Fragment sizes were determined by electrophoresis on 4.5% (w/v) denaturing 19:1 acrylamide:bisacrylamide gels on the ABI Prism™ 377 sequencer, using the GeneScan 2.0 software and a ROX-labelled commercial size standard as an internal standard (Applied Biosystems). In addition, two reference individuals were included in each ABI Prism™ 377 run in order to ensure exact scoring and to improve cross-referencing among the runs.

Genetic and statistical analyses

Microsatellite dataset was checked with MicroChecker v. 2.2.3 (Van Oosterhout et al., 2004) for null alleles, large allelic dropout and stuttering. We arranged microsatellite data with the Excel Microsatellite Toolkit 3.1.1. (Park, 2001) and converted the data into favoured file types. With FSTAT v. 2.9.3. (Goudet, 2001), allele frequencies, average allele numbers per locus (A), expected and observed heterozygosities (H e, H o), pairwise F ST values (Weir & Cockerham, 1984), F IS values and allelic richness (A R) as a standardised measure of the number of alleles corrected by the sample size were calculated. F values between populations were computed in the 2MOD programme (Ciofi & Bruford, 1999) applying a Markov Chain Monte Carlo simulation with 100,000 iterations. The first 10% of the iterations were discarded in order to avoid bias in parameter estimation resulting from the starting conditions. The F value gives information about relatedness between individuals based on the probability that two genes share a common ancestor within a population. We used Genepop on the Web v4.0.10 (Raymond & Rousset, 1995) to test genotypic distribution for conformance with Hardy–Weinberg (pHW) applying the probability test with the following Markov chain parameters: 1000 dememorisations, 100 batches and 1000 iterations per batch and a significance value of P < 0.05 after Bonferroni correction. We considered alleles as private alleles (A P) if they occurred only within a single population and at a frequency of more than 5%. Numbers of multilocus genotypes (MLG) per population were identified using the R package poppr (Kamvar et al., 2014) for testing clonal organisation (i.e., selfing). Analysis of molecular variance (AMOVA; Excoffier et al., 1992) as implemented in Arlequin 3.5.2.2 software (Excoffier et al., 2005) was used to hierarchically quantify genetic population structure. Presence of isolation by distance (IBD) was tested using Mantel’s test between the Nei’s DA genetic distance (Nei et al., 1983) and the geographic Euclidian distance among population sites. IBD was computed with a Monte Carlo randomisation test, based on 999 replicates, implemented in the R package ade4 (Thioulouse et al., 1997). We visualised the genetic structures by performing a Discriminant Analysis of Principal Components (DAPC) with the R package adegenet (Jombart, 2008) on individual and population levels. This approach can be processed without making any assumptions in terms of the population model, for instance the Hardy–Weinberg equilibrium. Each principal component is represented as a channel of the RGB colour transformation. These channels are mixed to create colours representing the genetic similarity of individuals or populations. We combined the results of the DAPC on population level with geographical data in a synthesis map in order to illustrate the genetic constitution in space using the R package mapdata and maps. We used STRUCTURE 2.3.4 software (Pritchard et al., 2000) to determine the number of genetic clusters (K) and to assign the probabilities of an individual to belong to each cluster. We tested the number of clusters from 1 to 20 with 10 iterations for each K (20,000 burn-ins, 200,000 Markov chain Monte Carlo replicates in each run) using the ‘No admixture’ model and assuming correlated allele frequencies to assess the convergence of the probability ln P(X|K). We determined the final number of clusters from ΔK, the rate of change in the log probability over all 10 iterations (Evanno et al., 2005), using STRUCTURE HARVESTER (Earl & von Holdt, 2012). We used CLUMPP software (Jakobsson & Rosenberg, 2007) applying the Greedy algorithm with 10,000 random input orders to find the optimal individual alignments of replicated cluster analyses.

Results

We verified all sampled freshwater mussels in this study as thick-shelled river mussel (U. crassus) by identification of ITS region fragment lengths of 372 bp, in conformance with the fragment length of U. crassus in Zieritz et al. (2012).

Genetic diversity

Evidence for genotyping errors in the microsatellite dataset due to null alleles, stuttering or large allelic dropout was not found. The average number of alleles per locus (A) and allelic richness (A R) varied between populations and averaged 6.5 and 4.3 over all loci and populations, respectively. Table 2 shows a summary of all microsatellite diversity indices. We found the highest (A = 9.33, A R = 5.19) allelic diversity in E3 population from the Elbe. We observed the lowest number of alleles (A = 4.11) in population T1 from the Schlei-Trave and the lowest allelic richness (A R = 3.19) in population D5 from the Danube system. Populations originating from the Danube drainage (D1–D5) and Rhine drainage (R1–R3) showed comparatively low levels of allelic richness (A R) to all other populations from the northern part of the study area. Expected and observed heterozygosities (H e and H o) averaged 0.69 and 0.59, respectively. Expected heterozygosities varied from 0.502 in D5 to 0.806 in E1, and observed heterozygosities ranged from 0.413 in D5 to 0.736 in I1. Lowest values of observed heterozygosity were found in the Danubian populations D4 and D5 and the Rhine population R2 (H o = 0.461, 0.413 and 0.421, respectively). We detected maxima of observed heterozygosity in populations I1, E3 and E1 (H o = 0.736, 0.734 and 0.731, respectively). In general, all values of observed and expected heterozygosity (H o) were relatively high and about in the same range without retaining a broad variance. We found the greatest inbreeding coefficient F IS in the populations R2, D4 and R1 (F IS = 0.390, 0.341 and 0.253, respectively) and lowest in the T3 population from the Schlei–Trave system (F IS = −0.036), indicating outbreeding in this population. F values displaying the proportion of common ancestors within each population were highest in populations D3, D4 and D5 of the Danube in Southern Germany and lowest in populations E1, E3 and E4 originating from the Elbe drainage system in Northern Germany. Also, large populations like I1 and I2 from the drainage Eider revealed relatively high F values. We found private alleles (A P) at six different loci in seven populations over all drainage systems. Thick-shelled river mussel populations were not in Hardy–Weinberg equilibrium, except for populations I2, T1 and T3. No identical multilocus genotypes were detected for U. crassus.
Table 2

Microsatellite diversity indices of all thick-shelled river mussel populations; population codes according Table 1 (Pop ID), sample size (N), number of distinct multilocus genotypes (MLG), average number of alleles per locus (A), mean allelic richness per population (A R), number of private alleles (A P), expected (H e) and observed (H o) heterozygosity, F IS, F value and result of Hardy–Weinberg probability test for deviation from expected Hardy–Weinberg proportions (P HW) with the number of alleles deviating from Hardy–Weinberg Equilibrium after Bonferroni correction (P < 0.0055)

Pop ID

N

MLG

A

A R

A P

H e

H o

F IS

F value

P HW

Loci not in P HW

S1

18

18

6.78

4.50

0

0.704

0.698

0.009

0.092

1

Uc69

I1

13

13

5.78

4.26

1

0.733

0.736

0.035

0.102

1

Uc39

I2

6

6

4.22

4.02

0

0.717

0.667

0.070

0.114

0

n.s.

T1

5

5

4.11

4.11

1

0.667

0.622

0.067

0.107

0

n.s.

T2

16

16

7.00

4.44

0

0.706

0.571

0.192

0.077

2

Uc5, Uc69

T3

5

5

4.33

4.33

0

0.686

0.711

−0.036

0.097

0

n.s.

E1

12

12

7.00

5.15

0

0.806

0.731

0.092

0.033

1

Uc5

E2

18

18

7.11

4.64

1

0.731

0.604

0.174

0.077

1

Uc5

E3

29

29

9.33

5.19

0

0.774

0.734

0.051

0.025

2

Uc5, Uc15

E4

20

20

8.67

4.86

0

0.736

0.594

0.193

0.035

1

Uc77

R1

20

20

6.44

4.00

2

0.662

0.494

0.253

0.148

5

Uc15, Uc25, Uc39 Uc69, Uc77

R2

30

30

7.56

4.19

0

0.689

0.421

0.390

0.124

7

Uc5, Uc7, Uc15, Uc16 Uc19, Uc25, Uc69

R3

29

29

7.22

3.99

0

0.670

0.525

0.216

0.132

3

Uc5, Uc25, Uc39

D1

12

12

5.56

4.12

1

0.704

0.583

0.172

0.159

3

Uc5, Uc15, Uc25

D2

28

28

7.44

4.04

0

0.686

0.546

0.205

0.133

3

Uc15, Uc16, Uc19

D3

20

20

5.89

3.69

1

0.582

0.531

0.088

0.195

2

Uc19, Uc39

D4

27

27

6.67

3.97

1

0.699

0.461

0.341

0.191

3

Uc5, Uc16, Uc39

D5

28

28

5.78

3.12

0

0.502

0.413

0.178

0.279

1

Uc5

Genetic differentiation

According to the analysis of variance (AMOVA) of hierarchical gene diversity, 84% of the genetic variation was found within populations, and only 9.1% of the variation was due to differences among drainage systems. 6.9% of the variation was explained by differences among populations within drainage systems. Even though the F IS value was high (F IS = 0.186, P < 0.001), most of the global deviation from HW proportions (F IT = 0.302, P < 0.001) was caused by interpopulation differentiation with a moderate and significant average F ST = 0.144 (P < 0.001). A differentiation of northern from southern populations was evident from the pairwise estimates of F ST. F ST values among U. crassus populations ranging from 0.001 between the two populations R2 (Rhine drainage) and D1 (Danube drainage) to 0.368 between the two populations T3 (Schlei-Trave drainage) and D5 (Danube drainage). To visualise the degree of differentiation between populations, we illustrated the matrix of pairwise F ST values through the heat map shown in Fig. 1. The means of F value and F IS for population from northern parts (F = 0.08, F IS = 0.08) were significantly lower (P < 0.001, one tailed t test) than the means in the southern region (F = 0.17, F IS = 0.23). In contrast, means of observed heterozygosity and allelic richness in northern populations (A R = 4.6, H O = 0.67) were significantly higher (P < 0.001, one tailed t test) than those from the southern area (A R = 3.9, H O = 0.49).
Fig. 1

Heat map of pairwise F ST value distance matrix among all eighteen thick-shelled river mussel populations; dark red colours reflect high genetic differentiation and lighter colours reflect low genetic differentiation

Generally, populations from the northern drainage systems Elbe, Schlei–Trave, Eider and Sweden showed relatively low F ST values compared to each other, indicating weak population differentiation. In contrast, populations from the Danube catchment revealed very strong genetic differentiation to the northern populations.

Populations D4 and D5 originating from the Danube drainage also showed high F ST values in comparison to D1, D2 and D3 (Danube system) despite all of them belonging to the same drainage system. Pairwise F ST values and Nei’s D A distances for each population are listed in Table 3 in the Supplementary material.

The synthesis maps of Germany and South Sweden combining geographical and genetic data using DAPC showed a significant spatial structure of three genetic clusters (Fig. 2). In the northern population, drainage dependence was slightly visible; however, admixture was the predominant genetic structure (i.e., the Swedish population S1 had a related genetic constitution to the populations from the contact zone in Northern Germany). The populations D4 and D5 from the Danube system were separated from the other Danubian populations, whereas D4 and D5, two geographically closely located and connected populations, showed genetic similarity.
Fig. 2

Synthesis map of Germany and South Sweden combining geographical and genetic data using the DAPC method according to Jombart (2008) (left side) with black dotted line indicating the approximate limits of ice maxima during Weichselian glaciation; major drainage systems are represented in different shades of grey; plots of the DAPC result (right side) on population and individual levels of eighteen U. crassus populations with a total amount of 336 individuals and a number of 42 retained principal components on axes 1–2. Location of the dots refer to the sampling site of the U. crassus populations; population codes are according to Table 1

Fig. 3

Population structure of eighteen U. crassus populations using model-based structure software and a dataset of 336 individuals for K = 3. Populations are classified according to their originating drainage systems from north to south. The results were averaged by CLUMPP software and plots were generated by Excel. Harvester was used to determine K. Population codes are according to Table 1; K Kävlingeåns, I  Eider, T Schlei-Trave

We identified two (ΔK = 1741), three (ΔK = 60) and six (ΔK = 33) genetic clusters of thick-shelled river mussel populations as being most likely for the dataset using ΔK and ln Pr(X|K) in the STRUCTURE analyses. Plots generated in STRUCTURE Harvester showing the mean log likelihood of K and variance per K value (Fig. 4A) and Evanno’s ΔK statistics (Fig. 4B) were computed. Figure 3 illustrates the individual proportion of the cluster membership (K = 3; barplot), also supporting a distinct north–south differentiation of populations. However, populations originating from the contact zone in Northern Germany revealed an admixed genetic structure. D4 and D5 from the Danube drainage revealed evidence of an isolated population based on their population divergence and diversity based on F ST, F IS and F values.
Fig. 4

Plots of mean log likelihood of K and variance per K value (A) and Plot of Delta K (=ΔK) values (B) given by STRUCTURE Harvester analysis of 336 Unio crassus samples using nine polymorphic microsatellite markers

Mantel’s test for isolation by distance among all populations was highly significant (simulated P = 0.001, correlation = 0.819). The two-dimensional kernel density estimation of the correlation of genetic and geographic distances in U. crassus revealed a continuous cline (separating the northern and southern populations). Mantel’s test for isolation by distance focused on the northern populations (Northern Germany and Sweden) revealed no significant correlation (P = 0.476, correlation = 0.014).

Discussion

The examined thick-shelled river mussel populations are strongly structured, but this finding cannot comprehensively be explained by present-day drainage systems. The isolation-by-distance model explains only the strong genetic divergence between populations from the northern and southern areas. In the within-population variability, a north–south difference with a higher genetic variability in the northern populations was evident, indicating a diversity hotspot in this area. This observation is in line with genetic results in the endangered freshwater pearl mussel (M. margaritifera) (Geist & Kuehn, 2005, 2008; Bouza et al., 2007; Geist et al., 2010) which also revealed highest genetic variability in northern areas. The great variability within the investigated populations (84%) was already observed for two thick-shelled river mussel populations in the Czech Elbe river basin in Central Europe (within variability 94.6%), stressing the importance of interpopulation variability in host-affiliate relationships (Douda et al., 2014).

We suggest that the low genetic differentiation (F ST = 0.052), the high genetic variability and admixed structure among the Northern populations are probably related to the Weichselian glacial period where whole Scandinavia and parts of northeastern Germany were covered with ice sheets and a land bridge connected Sweden with Central Europe (Bjoerk, 1995). It is likely that Northern German and Swedish populations were established at the same time from ancestral populations south of the glaciation zone after retreating ice sheets. The genetic diversity parameters of the population in the southern edge of the Elbe river basin indicate that these populations likely survived the last glaciation in a separate, second refugium and then acted as a source of colonisation of U. crassus. After retreating glaciation, the thick-shelled river mussel probably recolonised areas northwards to Sweden through a land bridge with associated freshwater connections between Sweden and Northern Germany. The higher genetic diversity of U. crassus in this area can be a result of recolonisation by multiple source populations, where a high number of founders could have admixed in several colonisation events and new recombinant genotypes arose in these populations, as also observed for the freshwater pearl mussel in its northern distribution range (Geist & Kuehn, 2008; Geist et al., 2010).

In line with the observations in the Iberian freshwater pearl mussel populations in their southernmost distribution area (Bouza et al., 2007), U. crassus populations from the Danube showed lower genetic variability, a high genetic differentiation and the greatest proportion in common ancestors towards their southern propagation limit on the edge of the alpine mountains. In contrast, populations in the northern margin of the Danube system showed slightly higher genetic diversity which leads to the assumption that this region of the low mountain range in the northern Danube system might also have served as an important refugium for the thick-shelled river mussel during glacial perturbations. It is likely that populations from the Danubian low mountain range expanded southwards towards their present-day distribution edge at the alpine mountains after melting of glaciers. However, the high genetic differentiation of U. crassus populations within the Danube system cannot be fully explained by its postglacial colonisation history alone, although the existence of multiple glacial refugia in its most important host fish, S. cephalus as suggested by Durand et al. (1999), may be reflected in U. crassus as well. We propose that the strong genetic divergence in the Danube might be a result of coevolution in the mussel–host relationships, since it is reported that U. crassus populations in the Danube drainage are adapted to highly different host fish community structures in relation to different abundances of single species (Taeubert et al., 2012b). The selection pressure of different and diverse host fish communities on U. crassus in this particular area might have altered the genetic constitution of populations in the Danube basin to the current observed degree of genetic divergence after recolonisation southwards to the Alps, as also previously observed in two Czech U. crassus populations (Douda et al., 2014). A host-dependent genetic structure and a host-dependent variation in freshwater mussels have already been observed in the freshwater pearl mussel where populations utilising different hosts showed large genetic differences, but populations utilising the same host showed small genetic differences, strengthening the fact that population structure is heavily dependent on host affiliation (Karlsson et al., 2014).

In summary, our data on the spatial genetic pattern of the thick-shelled river mussel (U. crassus) suggest that it was strongly shaped by Pleistocene glaciations with two different postglacial refugial areas acting as sources of recolonisation after glaciers melted. Especially for the Danubian populations, local adaptation to highly diverse host fish communities is likely to have additionally shaped the genetic structure.

In terms of applying this knowledge to practical conservation, we suggest that the three genetic conservation units (one northern and two southern) should be considered in conservation management strategies of the species.

One of the identified conservation units comprises the Danubian populations D4 and D5 at the southernmost distribution limit. They are connected via the river Ammer and build a specific and separate unit that should be considered as one management unit. The other populations from the Danube and Rhine system also form a separate genetic unit, having played a key role in postglacial colonisation of the Danube river basin. Populations from northern catchment areas all cluster into one large unit, which does not match the present-day drainage system structure. The three recognised genetic conservation units, regardless if resulting from historical isolation or from local adaption, are likely to have a distinct evolutionary potential, which needs to be taken into account for species conservation management that considers retaining a maximum of the genetic diversity within a species. In a next step, identification of possible links between genetic constitution and differences in host fish communities, as well as sampling of additional populations, particularly from the southern areas with the greatest spatial genetic structuring, is recommended.

In Europe, several conservation programmes for endangered mussel species are currently in place (Lopes-Lima et al., 2016), including captive breeding for restocking purposes in an increasing number of species (Gum et al., 2011), as well as habitat restoration (Geist & Hawkins, 2016). As previously demonstrated for the freshwater pearl mussel M. margaritifera, integrative conservation approaches (Geist, 2010) in U. crassus should simultaneously consider genetic information as well as the recently updated information on the habitat and host fish preferences of the species (e.g., Taeubert et al., 2012a, b; Denic et al., 2014; Stoeckl et al., 2015; Stoeckl & Geist, 2016). Populations of different genetic origins have previously been demonstrated to perform differently in a cross-exposure experiment of M. margaritifera (Denic et al., 2015) as well as in their host use in an experiment with two genetically distinct U. crassus populations (Douda et al., 2014). Thus, the possible local adaptation of the conservation units identified herein should be considered to ensure proper conservation management.

Notes

Acknowledgement

The study was made possible due to the generosity of the following colleagues providing U. crassus samples: R. Brinkmann, T. Berger, M. Österling and K. Stöckl. We would like to give special thanks to the Bavarian Mussel Coordination Office (K. Stöckl) funded by the Bavarian State Ministry of the Environment and Consumer Protection and to H. Bayerl for logistic support. We also acknowledge 'Faunistisch-Ökologische Arbeitsgemeinschaft' and the 'Landesamt für Landwirtschaft, Umwelt und ländliche Räume Schleswig–Holstein' for their financial contributions to this study. We also acknowledge the support from 'Regierung Oberbayern', 'Regierung Oberpfalz', 'Regierung Unterfranken' and 'Regierung Oberfranken' for providing the required sampling licences.

Supplementary material

10750_2017_3134_MOESM1_ESM.docx (16 kb)
Supplementary material 1 (DOCX 17 kb)

References

  1. Bjoerck, S., 1995. A review of the history of the Baltic Sea, 13.0–8.0 ka BP. Quaternary International 27: 19–40.CrossRefGoogle Scholar
  2. Boeker, C., T. Lueders, M. Mueller, J. Pander & J. Geist, 2016. Alteration of physico-chemical and microbial properties in freshwater substrates by burrowing invertebrates. Limnologica-Ecology and Management of Inland Waters 59: 131–139.CrossRefGoogle Scholar
  3. Bogan, A. E., 2008. Global diversity of freshwater mussels (Mollusca, Bivalvia) in freshwater. Hydrobiologia 595: 139–147.CrossRefGoogle Scholar
  4. Bouza, C., J. Castro, P. Martínez, R. Amaro, C. Fernández, P. Ondina, A. Outeiro & E. San Miguel, 2007. Threatened freshwater pearl mussel Margaritifera margaritifera L. in NW Spain: low and very structured genetic variation in southern peripheral populations assessed using microsatellite markers. Conservation Genetics 8: 937–948.CrossRefGoogle Scholar
  5. Bryja, J., C. Smith, A. Konečný & M. Reichard, 2010. Range-wide population genetic structure of the European bitterling (Rhodeus amarus) based on microsatellite and mitochondrial DNA analysis. Molecular Ecology 19: 4708–4722.CrossRefPubMedGoogle Scholar
  6. Ciofi, C. & M. W. Bruford, 1999. Genetic structure and gene flow among Komodo dragon populations inferred by microsatellite loci analysis. Molecular Ecology 8.Google Scholar
  7. Denic, M., M. Stoeckl, B. Gum & J. Geist, 2014. Physicochemical assessment of Unio crassus habitat quality in a small upland stream and implications for conservation. Hydrobiologia 735: 111–122.CrossRefGoogle Scholar
  8. Denic, M., J. E. Taeubert, M. Lange, F. Thielen, C. Scheder, C. Gumpinger & J. Geist, 2015. Influence of stock origin and environmental conditions on the survival and growth of juvenile freshwater pearl mussels (Margaritifera margaritifera) in a cross-exposure experiment. Limnologica-Ecology and Management of Inland Waters 50: 67–74.CrossRefGoogle Scholar
  9. Douda, K., Sell, J., Kubíková-Peláková, L., Horký, P., A. Kaczmarczyk & M. Mioduchowska, 2014. Host compatibility as a critical factor in management unit recognition: population-level differences in mussel–fish relationships. Journal of Applied Ecology 51: 1085–1095.CrossRefGoogle Scholar
  10. Durand, J. D., H. Persat & Y. Bouvet, 1999. Phylogeography and postglacial dispersion of the chub (Leuciscus cephalus) in Europe. Molecular Ecology 8: 989–997.CrossRefPubMedGoogle Scholar
  11. Earl, D. A. & B. M. von Holdt, 2012. STRUCTURE HARVESTER: a website and program for visualizing STRUCTURE output and implementing the Evanno method. Conservation Genetics Resources 4: 359–361.CrossRefGoogle Scholar
  12. Engel, H., 1990. Untersuchungen zur Autoekologie von Unio crassus (Philipsson) in Norddeutschland. PhD thesis, Tieraerztliche Hochschule Hannover, Germany.Google Scholar
  13. Evanno, G., S. Regnaut & J. Goudet, 2005. Detecting the number of clusters of individuals using the software STRUCTURE: a simulation study. Molecular Ecology 14: 2611–2620.CrossRefPubMedGoogle Scholar
  14. Excoffier, L., P. E. Smouse & J. M. Quattro, 1992. Analysis of molecular variance inferred from metric distances among DNA haplotypes: application to human mitochondrial DNA restriction data. Genetics 131: 479–491.PubMedPubMedCentralGoogle Scholar
  15. Excoffier, L., G. Laval & S. Schneider, 2005. Arlequin ver. 3.0: an integrated software package for population genetics data analysis. Evolutionary Bioinformatics Online 1: 47–50.Google Scholar
  16. Frankham, R., D. A. Briscoe & J. D. Ballou, 2002. Introduction to Conservation Genetics. Cambridge University Press, New York.CrossRefGoogle Scholar
  17. Geist, J., 2010. Strategies for the conservation of endangered freshwater pearl mussels (Margaritifera margaritifera L.): a synthesis of conservation genetics and ecology. Hydrobiologia 644: 69–88.CrossRefGoogle Scholar
  18. Geist, J., 2011. Integrative freshwater ecology and biodiversity conservation. Ecological Indicators 11: 1507–1516.CrossRefGoogle Scholar
  19. Geist, J., 2015. Seven steps towards improving freshwater conservation. Aquatic Conservation: Marine and Freshwater Ecosystems 25: 447–453.CrossRefGoogle Scholar
  20. Geist, J. & S. J. Hawkins, 2016. Habitat recovery and restoration in aquatic ecosystems: current progress and future challenges. Aquatic Conservation: Marine and Freshwater Ecosystems 26: 942–962.CrossRefGoogle Scholar
  21. Geist, J. & R. Kuehn, 2005. Genetic diversity and differentiation of central European freshwater pearl mussel (Margaritifera margaritifera L.) populations: implications for conservation and management. Molecular Ecology 14: 425–439.CrossRefPubMedGoogle Scholar
  22. Geist, J. & R. Kuehn, 2008. Host-parasite interactions in oligotrophic stream ecosystems: the roles of life history strategy and ecological niche. Molecular Ecology 17: 997–1008.CrossRefPubMedGoogle Scholar
  23. Geist, J., H. Söderberg, A. Karlberg & R. Kuehn, 2010. Drainage-independent genetic structure and high genetic diversity of endangered freshwater pearl mussels (Margaritifera margaritifera) in northern Europe. Conservation Genetics 11: 1339–1350.CrossRefGoogle Scholar
  24. Goudet, J., 2001. FSTAT, version 2.9.3: a program to estimate and test gene diversities and fixation indices. [Available at http://www.unil.ch/izea/softwares/fstat.html].
  25. Gum, B., R. Gross & R. Kuehn, 2005. Mitochondrial and nuclear DNA phylogeography of European grayling (Thymallus thymallus): evidence for secondary contact zones in central Europe. Molecular Ecology 14: 1707–1725.CrossRefPubMedGoogle Scholar
  26. Gum, B., M. Lange & J. Geist, 2011. A critical reflection on the success of rearing and culturing juvenile freshwater mussels with a focus on the endangered freshwater pearl mussel (Margaritifera margaritifera L.). Aquatic Conservation: Marine and Freshwater Ecosystems 21: 743–751.CrossRefGoogle Scholar
  27. Havrdová, A., J. Douda, K. Krak, P. Vít, V. Hadincová, P. Zákravský & B. Mandák, 2015. Higher genetic diversity in recolonized areas than in refugia of Alnus glutinosa triggered by continent-wide lineage admixture. Molecular Ecology 24: 4759–4777.CrossRefPubMedGoogle Scholar
  28. Hewitt, G., 1996. Some genetic consequences of ice ages, and their role in divergence and speciation. Biological Journal of the Linnean Society 58: 247–276.CrossRefGoogle Scholar
  29. Hewitt, G., 2000. The genetic legacy of the quaternary ice ages. Nature 405: 907–913.CrossRefPubMedGoogle Scholar
  30. Hochwald, S., 1997. Das Beziehungsgefuege innerhalb der Groeßenwachstums- und Fortpflanzungsparameter bayerischer Bachmuschelpopulationen (Unio crassus Phil. 1788) und dessen Abhaengigkeit von Umweltparametern. PhD thesis, Universitaet Bayreuth, Germany.Google Scholar
  31. Hochwald, S. & G. Bauer, 1990. Untersuchungen zur Populationsoekologie und Fortpflanzungs-biologie der Bachmuschel Unio crassus (PHIL.) 1788. Schriftenreihe Bayerisches Landesamt für Umweltschutz Heft 97: 31–49, Muenchen.Google Scholar
  32. Howard, J. K. & K. M. Cuffey, 2006. The functional role of native freshwater mussels in the fluvial benthic environment. Freshwater Biology 51: 460–474.CrossRefGoogle Scholar
  33. Israel, W., 1913. Biologie der europaeischen Sueßwassermuscheln. K. G. Lutz Verlag, Stuttgart.Google Scholar
  34. Jakobsson, M. & N. A. Rosenberg, 2007. CLUMPP: a cluster matching and permutation program for dealing with label switching and multimodality in analysis of population structure. Bioinformatics 23: 1801–1806.CrossRefPubMedGoogle Scholar
  35. Jombart, T., 2008. Adegenet: a R package for the multivariate analysis of genetic markers. Bioinformatics 24: 1403–1405.CrossRefPubMedGoogle Scholar
  36. Kamvar, Z.N., J.F. Tabima & N.J. Gruenwald, 2014. Poppr: an R package for genetic analysis of populations with clonal, partially clonal, and/or sexual reproduction.Google Scholar
  37. Karlsson, S., B. M. Larsen & K. Hindar, 2014. Host-dependent genetic variation in freshwater pearl mussel (Margaritifera margaritifera L.). Hydrobiologia 735: 179–190.CrossRefGoogle Scholar
  38. Lamand, F., K. Roche & J. N. Beisel, 2016. Glochidial infestation by the endangered mollusc Unio crassus in rivers of north-eastern France: Phoxinus phoxinus and Cottus gobio as primary fish hosts. Marine and Freshwater Ecosystems, Aquatic Conservation.Google Scholar
  39. Lechner, S., 1999. Biologie und aktuelle Bestandssituation der Bachmuschel (Unio crassus) in Hessen. In Fricke, W., B. Neugierg & C. Pietzke-Widdig (eds), Schutz bedrohter Tierarten in den Fließgewaessern Mittelhessens. NZH. Verlag, Wetzlar: 9–11.Google Scholar
  40. Lopes-Lima, M., A. Teixeira, E. Froufe, A. Lopes, S. Varandas & R. Sousa, 2014. Biology and conservation of freshwater bivalves: past, present and future perspectives. Hydrobiologia 735: 1–13.CrossRefGoogle Scholar
  41. Lopes-Lima, M., R. Sousa, J. Geist, D. C. Aldridge, R. Araujo, J. Bergengren, Y. Bespalaya, E. Bódis, L. Burlakova, D. Van Damme, K. Douda, E. Froufe, D. Georgiev, C. Gumpinger, A. Karatayev, Ü. Kebapçi, I. Killeen, J. Lajtner, B. M. Larsen, R. Lauceri, A. Legakis, S. Lois, S. Lundberg, E. Moorkens, G. Motte, K. O. Nagel, P. Ondina, A. Outeiro, M. Paunovic, V. Prié, T. von Proschwitz, N. Riccardi, M. Rudzite, M. Rudzitis, C. Scheder, M. Seddon, H. Sereflisan, V. Simic, S. Sokolova, K. Stoeckl, J. Taskinen, A. Teixeira, F. Thielen, T. Trichkova, S. Varandas, H. Vicentini, K. Zajac, T. Zajac & S. Zogaris, 2016. Conservation status of freshwater mussels in Europe: state of the art and future challenges. Biological Reviews 92: 572–607.CrossRefPubMedGoogle Scholar
  42. Lummer, E., K. Auerswald & J. Geist, 2016. Fine sediment as environmental stressor affecting freshwater mussel behavior and ecosystem services. Science of the Total Environment 571: 1340–1348.CrossRefPubMedGoogle Scholar
  43. Lydeard, C., R. H. Cowie, W. F. Ponder, A. E. Bogan, P. Bouchet, S. A. Clark, K. S. Cummings, T. J. Frest, O. Gargominy, D. G. Herbert, R. Hershler, K. E. Perez, B. Roth, M. Seddon, E. E. Strong & F. G. Thompson, 2004. The global decline of nonmarine molluscs. BioScience 54: 321–330.CrossRefGoogle Scholar
  44. MODELL, H., 1959. Die tertiären Najaden des ungarischen Beckens. Jahrbuch der Geologischen Bundesanstalt 1958, 75:197-250, Hannover.Google Scholar
  45. Nei, M., F. Tajima & Y. Tateno, 1983. Accuracy of genetic distances and phylogenetic trees from molecular data. Journal of Molecular Evolution 19: 153–170.CrossRefPubMedGoogle Scholar
  46. Nesemann, H., 1993. Zoogeographie und Taxonomie der Muschel-Gattungen Unio Philipsson 1788, Pseudanodonta Bourguignat 1877 und Pseudunio Haas 1910 im oberen und mittleren Donausystem (Bivalvia: Unionidae, Margaritiferidae) (mit Beschreibung von Unio pictorum tisianus n. ssp.). Nachrichtenblatt der Ersten Vorarlberger Malakologischen Gesellschaft 1: 20–40.Google Scholar
  47. Park, S.D.E., 2001. The excel Microsatellite-Toolkit (versions 3.1). Animal Genomics Laboratory, UCD, Ireland. [Available at http://animalgenomics.ucd.ie/sdepark/ms-tookit].
  48. Pritchard, J. K., M. Stephens & P. Donnelly, 2000. Inference of population structure using multilocus genotype data. Genetics 155: 945–959.PubMedPubMedCentralGoogle Scholar
  49. Raymond, M. & F. Rousset, 1995. GENEPOP version 3.4: population genetics software for exact tests and ecumenicism. Journal of Heredity 86: 248–249.CrossRefGoogle Scholar
  50. Richter, A., K. Stoeckl, M. Denic & J. Geist, 2016. Association between the occurrence of the thick-shelled river mussel (Unio crassus) and macroinvertebrate, microbial, and diatom communities. Freshwater Science 35: 922–933.CrossRefGoogle Scholar
  51. Seifertová, M., J. Bryja, M. Vyskočilová, N. Martínková & A. Šimková, 2012. Multiple Pleistocene refugia and post-glacial colonization in the European chub (Squalius cephalus) revealed by combined use of nuclear and mitochondrial markers. Journal of Biogeography 39: 1024–1040.CrossRefGoogle Scholar
  52. Sell, J., M. Mioduchowska, A. Kaczmarczyk & R. Szymańczak, 2013. Identification and characterization of the first microsatellite loci for the thick-shelled river mussel Unio crassus (Bivalvia: Unionidae). Journal of Experimental Zoology Part A: Ecological Genetics and Physiology 319: 113–116.CrossRefGoogle Scholar
  53. Stewart, J. R., A. M. Lister, I. Barnes & L. Dalén, 2010. Refugia revisited: individualistic responses of species in space and time. Proceedings of the Royal Society of London B: Biological Sciences 277: 661–671.CrossRefGoogle Scholar
  54. Stoeckl, K. & J. Geist, 2016. Hydrological and substrate requirements of the thick-shelled river mussel Unio crassus (PHILIPSSON 1788). Aquatic Conservation: Marine and Freshwater Ecosystems 26: 456–469.CrossRefGoogle Scholar
  55. Stoeckl, K., J. E. Taeubert & J. Geist, 2015. Fish species composition and host fish density in streams of the thick-shelled river mussel (Unio crassus) – implications for conservation. Aquatic Conservation: Marine and Freshwater Ecosystems 25: 276–287.CrossRefGoogle Scholar
  56. Taberlet, P., L. Fumagalli, A. G. Wust-Saucy & J. F. Gosson, 1998. Comparative phylogeography and postglacial colonization routes in Europe. Molecular Ecology 7: 453–464.CrossRefPubMedGoogle Scholar
  57. Taeubert, J. E., B. Gum & J. Geist, 2012a. Host-specificity of the endangered thick-shelled river mussel (Unio crassus, Philipsson 1788) and implications for conservation. Aquatic Conservation: Marine and Freshwater Ecosystems 22: 36–46.CrossRefGoogle Scholar
  58. Taeubert, J. E., A. M. P. Martinez, B. Gum & J. Geist, 2012b. The relationship between endangered thick-shelled river mussel (Unio crassus) and its host fishes. Biological Conservation 155: 94–103.CrossRefGoogle Scholar
  59. Thioulouse, J., D. Chessel, S. Doledec & J. M. Olivier, 1997. ADE-4: a multivariate analysis and graphical display software. Statistics and Computing 7: 75–83.CrossRefGoogle Scholar
  60. Van Oosterhout, C., W. F. Hutchinson, D. P. Wills & P. Shipley, 2004. MICRO-CHECKER: software for identifying and correcting genotyping errors in microsatellite data. Molecular Ecology Notes 4: 535–538.CrossRefGoogle Scholar
  61. Volckaert, F. A. M., B. Haenfling, B. Hellemans & G. R. Carvalho, 2002. Timing of the population dynamics of bullhead Cottus gobio (Teleostei: Cottidae) during the Pleistocene. Journal of Evolutionary Biology 15: 930–944.CrossRefGoogle Scholar
  62. Weir, B. S. & C. C. Cockerham, 1984. Estimating F-statistics for the analysis of population structure. Evolution 38: 1358–1370.PubMedGoogle Scholar
  63. Zahner-Meike, E. & J. M. Hanson, 2001. Effect of muskrat predation on naiads. In Bauer, G. & K. Wächtler (eds), Ecology and Evolution of the Freshwater Mussels Unionoida. Springer, Berlin: 163–184.CrossRefGoogle Scholar
  64. Zajac, K. & T. Zajac, 2011. The role of active individual movement in habitat selection in the endangered freshwater mussel Unio crassus Philipsson 1788. Journal of conchology 40: 446–461.Google Scholar
  65. Zettler, M. L. & U. Jueg, 2007. The situation of the freshwater mussel Unio crassus (Philipsson, 1788) in northeast Germany and its monitoring in terms of the EC Habitats Directive. Mollusca 25: 165–174.Google Scholar
  66. Zieritz, A., B. Gum, R. Kuehn & J. Geist, 2012. Identifying freshwater mussels (Unionoida) and parasitic glochidia larvae from host fish gills: a molecular key to the North and Central European species. Ecology and Evolution 2: 740–750.CrossRefPubMedPubMedCentralGoogle Scholar
  67. Zwanziger, G., 1920. Ueber die Verbreitung der Najaden im Gebiet der saechsischen Saale bei Hof. Archiv fuer Molluskenkunde 52: 14–33.Google Scholar

Copyright information

© Springer International Publishing Switzerland 2017

Authors and Affiliations

  1. 1.Aquatic Systems Biology Unit, Department of Ecology and Ecosystem ManagementTechnical University of MunichFreisingGermany
  2. 2.Unit of Molecular Zoology, Chair of Zoology, Department of Ecology and Ecosystem ManagementTechnical University of MunichFreisingGermany
  3. 3.Department of Fish, Wildlife and Conservation EcologyNew Mexico State UniversityLas CrucesUSA

Personalised recommendations