Genetic characterization of Atlantic sturgeon stocking material used in Lithuania to restore the Baltic sea population

Background. Recent genetic analyses of fish remains, obtained from archaeological sites, revealed that more than 2000 years ago the Baltic Sea was inhabited by sturgeons closely related to the Atlantic sturgeon, Acipenser oxyrinchus Mitchill, 1815. As some wildlife populations of the Atlantic sturgeon are still present in Canadian rivers, we decided to use those fish to restore the extinct Baltic population. Materials and methods. Fin clips of 50 A. oxyrinchus were collected from fry representing two hatcheries. At the hatchery No. 1 sturgeons were reared from fertilized eggs received from the Acadian Sturgeon and Caviar Inc., Canada. At the hatchery No. 2, larvae provided by the Regional Research Institute for Agriculture and Fisheries (Germany) were reared for several months until they were released into rivers. The molecular data of 22 and 28 specimens from hatcheries No. 1 and No. 2, respectively, were compared with homological D-loop sequences and some microsatellite loci derived from two museum specimens of sturgeons from the Tadas Ivanauskas Museum of Zoology, Kaunas, Lithuania. Signs of possible introgressive hybridization between Acipenser sturio Linnaeus, 1758 and A. oxyrinchus in two museum specimens were checked by comparison of alleles at loci AoxD161, AoxD188, AoxD297, AoxD242, and AoxC30. A possible affinity of between sturgeons released in Lithuanian rivers in 2015 and samples representing hatcheries No. 1 and No. 2 was estimated based on multi-locus genotyping data using the Structure software, version 2.3.4. Results. The same D-loop haplotype H1 characteristic for the museum specimens (MK637525 and MK637526) was also found as the only haplotype distributed among sturgeons reared in hatcheries and used for population reestablishment. A possible hybridisation between A. sturio and A. oxyrinchus was not confirmed for two museum specimens studied. The first results of genetic application undertaken in Lithuania also revealed the possibility to discriminate the individuals obtained from different hatcheries based on the set of 13 microsatellite loci Aox45, AoxD54, AoxD161, AoxD297, AoxD188, AoxD234, AoxD242, AoxC45, AoxC30, AoxD241, AoxD64, AoxD186, and Ls-68. The analysis of A. oxyrinchus specimens representing F1 generation of parental individuals reared in two separate hatcheries in Canada and Germany revealed significant differences in the allele composition and high probabilities for individuals released into the natural environment to be assigned to the hatchery of their origin. Conclusion. The results of the molecular analysis will be useful for identification of age, pedigree, and for initiating genetic monitoring of the restored population of the Atlantic sturgeon in the Baltic Sea.


INTRODUCTION
Sturgeons have a long lifespan, delayed maturation, and complex life cycle, involving anadromy (Sulak andRandall 2002, Williot et al. 2011) and these factors could have been beneficial while going through severe environmental changes in the past (Dudu et al. 2011). Nevertheless, today these features, along with extensive anthropogenic activity, have a strong impact on declines of sturgeon populations in their natural habitats all over the world (Billard andLecointre 2001, Bloesch et al. 2006). According to the IUCN Red List of Threatened Species, only two out of 25 species in the family Acipenseridae have the status of least concern. Whereas 16 species are critically endangered, two are considered DOI: 10.3750/AIEP/02557 endangered, three species are vulnerable, and two species are considered near threatened (Nelson et al. 2013).
Atlantic sturgeon, Acipenser oxyrinchus Mitchill, 1815 (see Birstein and Doukakis 2000), once was abundant in the Baltic Sea. It is thought, that A. oxyrinchus migrants had entered the Baltic Sea at least 4000 years ago and became a dominating species by replacing Acipenser sturio Linnaeus, 1758 approximately 2000 years ago , Chassaing et al. 2016. Seemingly, European waters had been colonized by migrants originating from North America, as studies of mtDNA D-loop revealed Baltic population to have had similar haplotypes A and B as St. Lawrence, St. John, and Hudson rivers populations in North America (Wirgin et al. 2000). However, the Baltic Sea population of A. oxyrinchus is now extinct in the wild as a consequence of intense anthropogenic activity, in particular, overfishing (Ludwig et al. 2002, Dadswell 2015. The last recorded individual in the Baltic Sea was caught off the coasts of Estonia in 1996 (Paaver 1999, which is considered as the year of the final extinction of the A. oxyrinchus population in Europe.
There are several restitution programs launched in Europe with the aim to re-establish local populations of A. oxyrinchus by breeding hatchery individuals and releasing them into local rivers. Currently, such restitution measures for sturgeons in western Europe are being carried out in Italy, France, and Germany . There are similar efforts in the Danube catchment area (Bloesch et al. 2006), Turkey (Memiş et al. 2006), and Poland (Kolman et al. 2011(Kolman et al. , 2014. Recently, a special program aimed to re-establish an extinct population of A. oxyrinchus has been approved by the Ministry of Agriculture of the Republic of Lithuania and increasing numbers of sturgeons are released into inland water bodies of Lithuania starting from 2011 and this program will continue until 2020.
The first results of the application of molecular markers used for estimation of genetic background of A. oxyrinchus stock released in natural environment during few years period of implementation of the restitution program undertaken in Lithuania and the analysis of two samples collected from museum specimens, which represent the extinct population of Atlantic sturgeon, serves as the basis for evaluation of genetic diversity among captivebred individuals, as well as a dataset for the future genetic monitoring of restored population.

MATERIAL AND METHODS
Fin clips of 50 A. oxyrinchus individuals were collected from fry representing two distinct hatcheries. At hatchery No. 1 * , located in Rusnė (Lithuania), sturgeons were reared from fertilized eggs received from Acadian Sturgeon and Caviar Inc., Canada. At hatchery No. 2, located in Simnas (Lithuania), larvae provided by Regional Research Institute for Agriculture and Fisheries (Germany) were reared for several months until being released in the Neris and Šventoji rivers. Tissue samples representing two different hatcheries consisted of 28 and 22 individuals, respectively. Samples were preserved in 70% ethanol and used for DNA analysis. Additionally, dry fin-clips of two museum specimens of sturgeons  were collected at the Tadas Ivanauskas Museum of Zoology in Kaunas, Lithuania. Both individuals were caught in 1960 in the Baltic Sea, near Nida (Lithuania), and previous investigators identified them as representatives of A. sturio. DNA from both museum specimens and fresh tissue samples were extracted using Aljanabi and Martinez (1997)

method.
A molecular investigation of representatives of museum specimens as well as two hatcheries was carried out by sequencing 196 base pairs (bp) D-loop fragment (mtDNA) and performing a fragment analysis of 13 microsatellite DNA loci.
The 13 microsatellite DNA loci (Aox45, AoxD54, AoxD161, AoxD297, AoxD188, AoxD234, AoxD242, AoxC45, AoxC30, AoxD241, AoxD64, AoxD186, Ls-68) were selected and amplified in accordance with the methodology provided by Panagiotopoulou et al. (2014). These loci were amplified using PCR reaction mixture of 25 µL containing 0.25 µg of DNA, 5 pmol of each primer (the 5′ ends of forward primers were fluorescently labelled), 0.2 mM of dNTPs, 3 mM of MgCl 2 , 1 × buffer with KCl, and 0.625 U of Taq Polymerase (Thermo Fisher Scientific Baltics, Lithuania). The PCR conditions were as follows: initial denaturation for 15 min at 95°C, 35 cycles for 30 s at 94°C, 30 s at 61°C, or 63°C, 30 s at 72°C and the final extension step for 30 min at 60°C. Both DNA sequencing and DNA fragment analysis were performed using Genetic Analyzer 3500 (Applied Biosystems) at the Laboratory of Molecular Ecology (Nature Research Centre, Vilnius, Lithuania).
Species-specific mtDNA D-loop sequences, representing the most frequent haplotypes of A. oxyrinchus and A. sturio accessed through GenBank and DNA sequencing data obtained during the presently reported study were aligned using CLC 6.8.1 software ** . The Neighbor-Joining dendrogram (Saitou and Nei 1987) was constructed using the same CLC 6.8.1 software. Genetic diversity parameters such as: total number of alleles per locus (N T ), number of effective alleles (N E ), observed heterozygosity (H O ), and expected heterozygosity (H E ) were calculated using GenAlEx software version 6.5 (Paekal and Smouse 2012).
Signs of possible introgressive hybridization between A. sturio and A. oxyrinchus in two museum specimens were checked by comparing alleles at loci AoxD161, AoxD188, AoxD297, AoxD242, and AoxC30, as the allele range sizes of Atlantic and European sturgeons did not overlap at these loci.
The genetic structure of the population of A. oxyrinchus released into inland water bodies of Lithuania in 2015 was assessed and possibility to assign each studied fish to one of the samples representing different hatchery was estimated based on multi-locus genotyping data using Structure software, version 2.3.4 (Pritchard et al. 2000).

RESULTS AND DISCUSSION
Despite the relatively old age of the museum specimens (No. PP-1146 and No. 153) a sufficient amount of good quality DNA was extracted from dry fin-clips and the targeted fragments of mtDNA control region were amplified successfully. Newly obtained 180 bp sequences of the two museum sturgeons were aligned with the homologous sequences of both A. sturio (GenBank accession numbers FN256383, AJ428274, FN256380) and A. oxyrinchus (GenBank accession numbers AF162716.1, AF162721.1, AF162729.1, AF162722.1, AF162726.1) and phylogenetic relations of the above-mentioned sequences inferred by NJ reconstruction revealed the two museum specimens of sturgeons to be attributed to haplotype A of A. oxyrinchus (Fig. 1).
A comparison of the mtDNA D-loop sequences of both museum specimens and additionally obtained homologous sequences of captive-bred sturgeons released into Lithuanian inland waters in 2015 revealed that all the studied individuals of sturgeons represented the same haplotype A (GenBank access number AF162716.1) known as the most frequent haplotype H1, detected among representatives of extinct population of A. oxyrinchus (see Popović et al. 2014).
Evaluation of introgressive hybridization between A. sturio and A. oxyrinchus, described earlier , has become possible after comparing allele sizes (bp) at loci AoxC30, AoxD297, AoxD242, AoxD188, and AoxD161 discriminating a wild population of A. oxyrinchus from A. sturio and a successful identification of alleles at three out of five latter microsatellite loci for two museum specimens (Table 1). Due to the fact that allele sizes at loci AoxD188, AoxD242, and AoxD297 fall into the range characteristic of A. oxyrinchus signals of possible hybridisation between A. sturio and A. oxyrinchus after examining two studied museum specimens No. PP-1146 and No. 153 could not be confirmed.
In order to evaluate the genetic diversity and the genetic structure of the sturgeon population which consists of sturgeons released into a natural habitat, two groups of young sturgeons obtained from hatcheries No. 1 and No. 2 were genotyped. Private alleles characteristic of each hatchery were identified at seven loci out of 13 microsatellite loci examined (Table 2). A smaller number of alleles detected at the majority of loci examined, as well as the number of private alleles, indicate that the population of sturgeons representing hatchery No. 1 was founded by a reduced number of parental individuals in comparison to those of the hatchery No. 2.
A comparison of genetic diversity parameters (the total number of alleles per locus (N T ), the number of effective alleles (N E ), observed heterozygosity (H O ), and expected heterozygosity (H E ) confirmed the diversity between the two studied samples representing hatcheries No. 1 and No. 2 but also it revealed the higher than expected heterozygosity characteristic of representatives of the hatchery No. 1 in the majority of studied loci except for AoxD161 and AoxC30 and for the representatives of the hatchery No. 2 at locus AoxD54 (Table 3). The majority of loci under study exhibited more than two alleles (N a ) in the two studied stocks of A. oxyrinchus and the genetic diversity between those groups of fish were identified on the basis of many private alleles. Subsequent restocking, based on different stocking material, can increase these indicators on a global scale and will have a positive impact on the population of A. oxyrinchus in the Baltic Sea basin.
A similar genetic analysis of A. oxyrinchus restocking material derived from Canada was conducted during the restitution program in Poland where the restocking material was released in the local Polish rivers of the Baltic Sea basin, e.g., , Drwęca and Drawa rivers (Kolman et al. 2014). The participation of different countries and the multistage restitution program can positively affect the genetic condition of the sturgeon population in the Baltic Sea. Although formation of parental generation is dependent on a small number of spawning fish collected from the wild population (in case of sturgeon farm located in Canada) or it directly depends on a limited number of A. oxyrinchus individuals reared artificially (farms situated in Germany and Poland) it does not lead to the production of F1 generation with deficiency of heterozygotes ( Table  3). Due to this phenomenon, good starting conditions and the possible formation of a highly variable and viable re-established population of A. oxyrinchus in the Baltic Sea Region could be predicted.
Although the number of alleles for artificially reared A. oxyrinchus population examined in Poland ) was considerably higher in comparison to the gene pool determined at the same loci studied in Lithuania the genetic diversity of the re-established population is expected to increase over time. Subsequently, a higher genetic diversity of the re-established population could be reached by adding new genetically unrelated female and male individuals participating in fertilization of eggs in hatcheries.
The process of restoration of the extinct population of A. oxyrinchus in the Baltic Sea region should be  controlled by evaluating the population genetic structure of A. oxyrinchus released into inland water bodies. Informative and promising results could be obtained with the help of such genetic tool as the program Structure (Pritchard et al. 2000). Testing of two stocks of sturgeons representing two hatcheries No. 1 and No. 2 revealed the possibility to assign each studied fish to the sample that covers individuals belonging to the same spawning pool (Fig. 2).
The results of the presently reported molecular analyses will be useful for the identification of age of fish released into the natural environment in case it is captured and genotyped in the future and the obtained molecular data will be compared with the accumulated genetic data of young sturgeons representing different hatcheries in different years. The age of captured sturgeons will be identified by attributing individual genotypes to one of the samples representing the released sturgeons that had been reared in different hatcheries during the period of the implementation of the program of the restoration of A. oxyrinchus in the Baltic Sea.

CONCLUSIONS
Two museum specimens No. PP-1146 and No. 153, representing the extinct Baltic population of sturgeons were characterized by the same haplotype of D-loop fragment that was identified in the representatives of A. oxyrinchus participating in the process of population re-establishment in Lithuania.
A comparison of the genetic diversity parameters and the genetic structure based on the 13 microsatellite DNA loci analysis, revealed significant differences in allele composition of representatives of hatcheries No. 1 and No. 2 and high observed heterozygosity of both studied sturgeon samples.
A microsatellite data analysis carried out with the Structure software revealed the possibility to discriminate representatives of hatcheries No. 1 and No. 2 by assignment each individual to one of two respective samples.
The results of the molecular analysis will be used for the future genetic monitoring of the population of A. oxyrinchus undergoing the process of re-establishment.