Hybridization of Russian Sturgeon (Acipenser gueldenstaedtii, Brandt and Ratzeberg, 1833) and American Paddlefish (Polyodon spathula, Walbaum 1792) and Evaluation of Their Progeny

Two species from the families Acipenseridae and Polyodontidae, Russian sturgeon (Acipenser gueldenstaedtii, Brandt and Ratzeberg, 1833; functional tetraploid) and American paddlefish (Polyodon spathula, Walbaum 1792, functional diploid) were hybridized. The hybridization was repeated using eggs from three sturgeon and sperm from four paddlefish individuals. Survival in all hybrid family groups ranged from 62% to 74% 30 days after hatching. This was the first successful hybridization between these two species and between members of the family Acipenseridae and Polyodontidae. Flow cytometry and chromosome analysis revealed two ploidy levels in hybrids. The chromosome numbers of the hybrids ranged between 156–184 and 300–310, in “functional” triploids and “functional” pentaploids, respectively. The hybrid origin and the ploidy levels were also confirmed by microsatellite analyses. In hybrids, the size and the number of dorsal and ventral scutes correlated with the ploidy levels as well as with the calculated ratio of the maternal and paternal chromosome sets. An extra haploid cell lineage was found in three hybrid individuals irrespective of the ploidy level, suggesting polyspermy. Although the growth performance showed high variance in hybrids (mean: 1.2 kg, SD: 0.55), many individuals reached a size of approximately 1 kg by the age of one year under intensive rearing conditions.

The breeders were injected with LHRH (luteinizing hormone-releasing hormone) analog Des-Gly 10 (D-Ala 6 ) LHRH-ethylamide (40 µg/kg for females and 20 µg/kg males) to induce ovulation and spermiation. Milt was stripped 48 h post injection, while eggs were stripped 48-52 h post injection using the method of Podushka, cited by Štěch et al. [21]. Immediately before fertilization, milt was diluted with water at a ratio of 1:200, and 2 mL of this diluted solution was added to egg samples weighing 300 g, according to the common aquaculture practice [22]. In the case of gynogenic fertilization, the milt was irradiated by 1 kGy dose of 60 Co γ ray. For retention of the second polar body, a 2 min 36 • C heat shock was applied 18 min post fertilization [12,23]. Fertilization rates at 48 h post fertilization and survival at hatching were determined by counting 3 × 100 eggs in each parent combination. Two thousand newly hatched fry from each of the treatment groups were stocked into 400 L volume tanks with trough-flow water for nursing. On the eighth day after hatching, fry received frozen brine shrimp nauplii, and mosquito larvae followed dry feed feeding from the twentieth day. When fish reached the size of 3 cm, they were transferred into larger tanks. When fingerlings reached 12-15 cm, individuals were labeled with passive induced transponder (PIT) tags (Loligo®Systems, Toldboden, Viborg, Danemark).

Microsatellite Marker Analysis
Fin clips were collected from the female Russian sturgeons (6F15; 174B; AG03) and male American paddlefish (596B; 6517; 2E75) parents and from 19 hybrid offspring. Genomic DNA was extracted using the standard protocol of the E.Z.N.A. Tissue Kit (Omega Bio-Tek, Doraville, CA, USA). Four pairs of microsatellite primers were used. Three microsatellite loci (Psp-28, Psp-29, Psp-32) were developed for American paddlefish [24], and the fourth locus (Spl_101) was developed for shovelnose sturgeon [25]. The universal fluorescent labeling method was used for capillary electrophoretic analysis of microsatellite alleles [26]. The forward primers were tailed by a 5 17 bp long sequence to provide an attachment site for fluorescent dye-labeled (fluorescent dyes: FAM, VIC, NED, PET) "universal" oligos. One tail-specific oligo was added to each reaction mixture. The reactions consisted of 100 ng genomic DNA, 10× Dream Taq PCR buffer (it contains 20 mM MgCl 2 ), 0.4 µL (0.2 mM of each dNTP), 0.5 µL forward and reverse primer, (10 µM), 0.5 µL fluorescent dye-labeled tail specific oligo (5 dye-ATTACCGCGGCTGCTGG-3 ) (10 µM), and 1.0 U DreamTaq DNA polymerase in a total volume of 20 µL. PCR conditions: Initial denaturing at 94 • C for 2 min; 35 cycles of 94 • C denaturing for 30 s, annealing temperature (Table 2) for 30 s, 72 • C extension for 1 min, and a final extension at 72 • C for 5 min. Amplified, labeled products were subjected to capillary electrophoresis on an ABI Prism 3130 Genetic Analyzer (Applied Biosystems, Waltham, MA, USA). GeneMapper 4.0 Software (Applied Biosystems, Waltham, MA, USA) was used to determine the fragment size of the detected alleles.

Flow Cytometry and DNA Content Analysis
The DNA content of red blood cells (RBCs) was measured by flow cytometry using propidium iodide (PI) staining. Blood cells were centrifuged at 500× g for 5 min and re-suspended in 500 µL PBS to set the cell concentration to 10 7 /mL. Then, by adding 4500 µL 70% ethanol (Scharlab Hungary, Budapest) to the samples, the cells were fixed and permeabilized for 30 min at room temperature [27]. Subsequently, the samples were washed twice in PBS (at 500× g for 5 min and re-suspended in 5000 µL PBS). At that point, the ethanol-fixed cells were stained with 40 µg/mL PI (Sigma-Aldrich) for 30 min. The samples were measured by using a Becton Dickinson FACS Array flow cytometer. A green 532 nm laser was used for the excitation of PI, and the fluorescence signal was detected at 585/42 nm. Cell debris was excluded from analysis, on the basis of forward scatter (FSC) and side scatter (SSC) signals. Flow cytometry data were analyzed using the Flowing Software (Flowing Software 2.5.1 (released 4.11.2013)), Cell Imaging Core, Turku Centre for Biotechnology (Turku, Finland).
Flow cytometry measurements were carried out on 98 hybrid fish, two Russian sturgeon (6F15, Ag03), and four male American paddlefish (2E75, 551F, 596B, 6517). Each measurement was repeated three times independently. The analysis was carried out in a blind manner except for the four male American paddlefish blood samples. Upon DNA content analysis, the mean PI fluorescence intensity of each progeny was normalized to the mean PI fluorescence intensity measured in the case of the three male American paddlefish and expressed in pg.
The genome size of each individual was calculated using a modified formula by Tiersch et al. [28]: DNA content (pg) = 3.9× (I/P), where I and P are the fluorescence intensities of hybrids and paternal individuals, respectively. The constant of 3.9 is the DNA content of paddlefish cells given in the picogram [28].

Chromosome Analysis
Chromosome analysis was carried out on tissue culture cells, modified after Tóth et al. [29], from eight ten-month-old hybrid individuals. The fibroblast culture was set up from the pectoral fin. The culture medium was changed twice a week. After 4-6 weeks, when the fibroblast monolayer was almost confluent in the flasks (Greiner 690175; 50 mL), 40 µL of KaryoMAX™ Colcemid™ Solution (Gibco 15212012) was added. Two hours later, cells were trypsinized, and after 7-min centrifugation (1500× g) were resuspended in 0.56% KCL as hypotonic treatment; this was followed by three changes of methanol/acetic acid (3:1) fixative. The cell suspension was spread on slides, dried at room temperature, and stained with 5% fresh Giemsa solution (in phosphate buffer pH 7.0) (Sigma 48900) for 7-8 min [30].
Five slides from each individual were prepared, and at least 30 metaphase spreads per individual were examined to determine the chromosome number. Chromosomes were classified using the nomenclature of Levan et al. [31].

Morphometry
After clove oil narcotization, we measured 21 morphometric and four meristic characters (Table 3) on 218 American paddlefish × Russian sturgeon hybrids, 49 American paddlefish, and 50 Russian sturgeon individuals with a digital caliper. The morphometric characters were determined according to Holcik et al. [32], Keszka and Krzykawski [33], and Hoover et al. [34]. In addition, we measured the width of the first three dorsal scutes. The meristic characters covered the number of dorsal (sDL), lateral (sLL), ventral scutes (sVL) and the number of barbels. The age of the examined fish varied between 228 and 258 days.
Morphometric data were standardized to eliminate the effect of body size, according to Elliott et al. [35]. Fishes were divided into five groups based on their origin and genome size: American paddlefish (PF), Russian sturgeon (RS), small genome size hybrids (SH), large genome size hybrids (LH), and hybrids without genome size data (NH). We ran a series of non-parametric univariate tests (Kruskal-Wallis) and pairwise comparisons (Mann-Whitney U test with Bonferroni correction) for each morphometric and meristic character to reveal differences between groups. Principal component analysis (PCA) was performed on the entire standardized dataset to evaluate the morphometric variation. Scute size was excluded from the multivariate test because it was absent in all paddlefishes and many hybrids. Statistical analyses were performed using PAST 1.95 [36]. Table 3. Mean morphometric (mm) and meristic characteristics among Russian sturgeon and American paddlefish and their hybrids. The numbers in parentheses are the coefficient of variation percentage values. The hybrids were grouped by genome size into small genome hybrids, large genome hybrids, and hybrids without genome size data. Different bold letters (a, b, c, d) indicate the significant differences (p > 0.05) based on Kruskal-Wallis tests and Mann-Whitney U pairwise comparisons on standardized data. The data were standardized on the total length; hence, this character was not available for pairwise comparison.

Fertilization, Survival, Growth of Hybrids
The fertilization rate varied from 86 to 93% and the hatching rate from 78 to 85% in the five parent combinations. Survival after hatching at 30 days ranged from 62 to 74% and at 180 days from 49 to 68%. Substantial differences in individual weights were observed within the groups. By the end of the first year, many hybrid individuals reached a bodyweight of over 1 kg (mean: 1.2 kg, SD: 0.55).

Microsatellite Analysis
There was no common allele between the two species. The Psp-28, Psp-32, and Spl-101 were disomic, while the Psp-29 was tetrasomic on the American paddlefish males. In the case of American paddlefish, all of the detected alleles of the four loci were different from the Russian sturgeon: five at the 2E75 males, six at the 596B males, and seven at 6517 males. Four alleles were common in both American paddlefish males (Table 4). The Psp-29 was disomic, while the Psp-28, Psp-32, and Spl-101 were tetrasomic on Russian sturgeon females. Two female genotypes did not differ from each other. Altogether twelve alleles were detected at the four loci, and none of them were present in American paddlefish. All the markers segregated in the offspring groups. All hybrid individuals inherited one paternal allele from the disomic (Psp-28, Psp-32, and Spl-101) and two alleles from the tetrasomic alleles loci. The SH individuals inherited one maternal allele from the Psp-28 or Psp-29 and two alleles from the tetrasomic Psp-32 and Spl-101 loci, while the LH hybrids inherited two, three, or four alleles in the case of tetrasomic markers, but at least one of the two maternal tetrasomic loci revealed the presence of two maternal genomes ( Table 4). None of the genotypes suggested the presence of polyspermy.

Flow Cytometry and DNA Content Analysis
Based on the DNA content analysis, we distinguished two main groups: A small genome size group (SH), with 7.22 ± 0.71 pg DNA content, and a large genome size group (LH), with 11.60 ± 0.84 pg DNA content. The hybrids were divided into five subgroups based on their parents. Estimated genome sizes of the hybrids are shown for the different parent combinations in Table 1. The majority of the tested progenies belonged to the small genome size (SH) group in the case of each parent combination. Although there was no statistically significant difference in mean DNA content among subgroups of the SH group, we identified some outlier individuals with statistically different DNA content, such as the D4AF3A individual (Table 5). Table 5. DNA content and chromosome number of the different hybrid types (SH/AP and LH/AAP). AP was an interspecies triploid hybrid with two maternal (4n) genomes and one paternal (2n) genome characterized by small genome size (SH). AAP was an interspecies pentaploid hybrid with four maternal genomes and one paternal genome characterized by large genome size (LH). PIT IDs are the individual numbers of internal tags for individual identification.

Chromosome Analysis and Ploidy Level in Different Cytometry Class Groups
The chromosome analysis of hybrid individuals confirmed the results of DNA content analysis. Two large groups could be distinguished by their chromosome number: inter-species triploid hybrids (SH) and inter-species pentaploid hybrids (LH). The triploid hybrids were formed by one maternal gamete (2n) of a functional tetraploid female Russian sturgeon and one paternal gamete (1n) from a functional diploid paddlefish. The chromosome numbers of SH varied between 156 ± 4 and 174 ± 10 ( Table 5) due to the different numbers of microchromosomes. The pentaploid LH group contained the entire maternal genome (4n = 250 ± 8) and the paternal haploid genome (n = 60). They had 304 ± 6 chromosomes ( Table 5). The morphology of microchromosomes was unclear because of their very small size. The paternal haploid cell lines described by Iegorova et al. [18] were also observed in several individuals (PIT ID D52B1B, D4B091, and D4D979) ( Table 5).
A detailed karyotype analysis was performed on one randomly chosen SH (PIT ID D52B1B; 3n = 176) and LH individual (PIT ID D4D979; 5n = 310). The karyotype of triploid individuals consisted of 70 meta-and submetacentric, 12 acrocentric, and 94 microchromosomes (Figure 1, Table 6). In the case of pentaploid individuals, the analysis showed the occurrence of 118 metacentric and submetacentric, 18 acrocentric, and 174 microchromosomes ( Figure 2, Table 6). In both cases, two large acrocentric chromosomes (Acipenseriformes marker chromosomes) were found (Figures 1 and 2). The reference data of karyotypes found in Table 6 regarding the number of middle-size acrocentric chromosomes were incomplete because references [3,8] did not classify them separately.

Morphometry
The most pronounced difference between American paddlefish and Russian sturgeon occurred in morphometric characters related to the rostrum length (rc, R, rr, C), which were higher in American paddlefish (Table 3). Russian sturgeons had longer upper caudal lobe (Cfa) and pectoral fin (IP). The rostrum length related characters of hybrids differed from the parent species', and further differences  2). The reference data of karyotypes found in Table 6 regarding the number of middle-size acrocentric chromosomes were incomplete because references [3,8] did not classify them separately.

Morphometry
The most pronounced difference between American paddlefish and Russian sturgeon occurred in morphometric characters related to the rostrum length (rc, R, rr, C), which were higher in American paddlefish (Table 3). Russian sturgeons had longer upper caudal lobe (Cfa) and pectoral fin (IP). The rostrum length related characters of hybrids differed from the parent species', and further differences

Morphometry
The most pronounced difference between American paddlefish and Russian sturgeon occurred in morphometric characters related to the rostrum length (rc, R, rr, C), which were higher in American paddlefish (Table 3). Russian sturgeons had longer upper caudal lobe (Cfa) and pectoral fin (IP). The rostrum length related characters of hybrids differed from the parent species', and further differences occurred between the LH and SH groups. The upper caudal lobe (Cfa) and pectoral fin (IP) lengths showed similar patterns, but in the latter character, the LHs differed just marginally from the Russian sturgeons. The size and number of scutes were higher in Russian sturgeon, lower in the SH group, and intermediate in the LH group. The number of lateral scutes was typically higher in hybrids.
The first principal component accounted for the vast majority of the total variance (84.43%); the subsequent component explained much less (10.38%; Figure 3). The ordination showed pronounced differentiation between the American paddlefish and Russian sturgeon. The hybrids were closer to the Russian sturgeon. The PCA plot supported the result of the univariate test and highlighted the importance of rostrum length-related characters (rr, R, rc, C) and upper caudal lobe (Cfa) and pectoral fin (IP) length in the separation of hybrids and parent species (Figures 3 and 4). Although we could not divide the hybrids into distinct subgroups, the LH individuals were projected closer to the Russian sturgeon.
Genes 2020, 11, x FOR PEER REVIEW 10 of 16 occurred between the LH and SH groups. The upper caudal lobe (Cfa) and pectoral fin (IP) lengths showed similar patterns, but in the latter character, the LHs differed just marginally from the Russian sturgeons. The size and number of scutes were higher in Russian sturgeon, lower in the SH group, and intermediate in the LH group. The number of lateral scutes was typically higher in hybrids. The first principal component accounted for the vast majority of the total variance (84.43%); the subsequent component explained much less (10.38%; Figure 3). The ordination showed pronounced differentiation between the American paddlefish and Russian sturgeon. The hybrids were closer to the Russian sturgeon. The PCA plot supported the result of the univariate test and highlighted the importance of rostrum length-related characters (rr, R, rc, C) and upper caudal lobe (Cfa) and pectoral fin (IP) length in the separation of hybrids and parent species (Figures 3 and 4). Although we could not divide the hybrids into distinct subgroups, the LH individuals were projected closer to the Russian sturgeon.

Discussion
Former experiences on the hybridization of acipenserids and paddlefishes [14,38] support the assumed inability of the hybridization of taxa with such a high phylogenetic distance separating them. Still, the cross of female Russian sturgeon and male American paddlefish, as a negative control for gynogenesis, resulted in viable hybrids. The fertilization, hatching, and survival rates of these hybrids were close to that of the pure maternal species. The progenies exhibited two ploidy levels (triploid and pentaploid) due to spontaneous polyploidization, relating to external morphological and meristic characteristics.
The reproductive success of these two distant species could be the combined consequence of the ancient genome duplication (180 Mya, that occurred in the common ancestor of the two families) and the slow evolution rate of acipenserids. The theory of slow evolution was reinforced by the "small" morphological differences of the living sturgeon species, the smaller substitution rates of mitochondrial or nuclear genes, and the slower protein evolution, compared to other fish species. This theory has been recently confirmed by a detailed evaluation of the sterlet genome [39]. Since the common sturgeon genome duplication, the sterlet genome has not formed back to a diploid state, and structural and functional elements have been found on the tetraploid level in unexpectedly high degree [39]. These phenomena could lead to a higher similarity, compatibility, and flexibility among the sturgeon genomes and allow the hybridization between Russian sturgeon and American paddlefish despite the large geographical, physiological, and morphological distances.
The intermediate appearance is common in interspecific hybrids [40,41]. A further increase of morphological variability occurs in polyploid hybrids, primarily due to the difference in gene dosage from parent species [42]. Hybrids showed the widest variability in characters, which differed the most parental species, i.e., rostrum-length-related characters and length of the upper caudal lobe. The LH individuals were more similar to the maternal species, presumably due to the larger gene dosage from Russian sturgeon overwhelming the paddlefish-like traits [43]. The SH hybrids showed a more diverse appearance since the ration of maternal and paternal chromosome sets is more balanced than LH hybrids [43].
Similar to the interspecific teleost hybrids, where a substantial difference has been observed in size and the number of scales between hybrids and parental species [40,43], the number of scutes differed substantially in hybrids in conjunction with ploidy level. The presence or absence of the scutes, as well as their modified forms in the hybrids, offer the opportunity to study the role of the fibroblast growth factor receptor gene (fgfr) as it influences the number and size of scales in the mirror, nude, and scaled forms of common carp and zebrafish [44]. Sub-functionalization of paralogues of this gene may have occurred following genome duplication during the evolutionary process [45].
The characteristics of karyotypes and the results of microsatellite analysis confirmed that the progenies were interspecies hybrids. As with Váradi et al. [46], chromosome analysis revealed the occurrence of two ploidy levels in hybrids: triploid hybrids (SH, AP) and pentaploid hybrids (LH, AAP). Slight variation in chromosome numbers was also observed within the groups. On the one hand, inconsistency in the published chromosome numbers within a pure-bred species could be the consequence of difficulties in the determination of microchromosome numbers [37]. On the other hand, microchromosome loss also occurs in interspecies hybrids of species characterized by different ploidy levels [47]. We observed haploid paternal cell lineages in some individuals in both groups, suggesting polyspermy [18]. In this case, a supernumerary spermatozoon began to develop as a single cell line, and triploid/pentaploid and haploid cell lines were present in the genome of individuals, causing mosaicism in the hybrids.
Investigations on the functional scale of the ploidy level [20] assume that the interspecific differences in ploidy levels (e.g., diploid American paddlefish and tetraploid Russian sturgeon) originated from whole-genome duplication events during the evolution of Acipenseriformes [8]. Furthermore, intraspecific polyploidization is also relatively frequent among the acipenseriform fishes. It has been detected in white sturgeon (Acipenser transmontanus) [48], Siberian sturgeon (Acipenser baerii) [20,49], kaluga (Huso dauricus) [14], sterlet [50], and Sakhalin sturgeon (Acipenser mikadoi) [14]. The spontaneous intraspecific polyploidization can result in variance in genome size. Havelka et al. [20] described the spontaneous duplication of the maternal chromosome set in Siberian sturgeon of different ploidy caused by either premeiotic endomitosis or the retention of the second polar body. Many other studies [51][52][53] have stated the retention that the second polar body can lead to polyploidization. The different ratios of triploid/pentaploid individuals among the parent combinations in our study also supported this hypothesis.
Microsatellites from both species and hybrids were amplified successfully. All the used microsatellite markers were suitable for distinguishing the two species and for verification of the hybridization. Previously, Zou et al. [14] used the Psp-28, Psp-29, Psp-32 markers on Polyodontidae and Acipenseridae fishes to verify the success of gynogenesis. They also found different allele ranges between the paddlefish and Amur sturgeon. However, the allele sizes could not be compared directly because different techniques were used for allele sizing (polyacrylamide gels and silver staining). All three Psp loci were disomic in that experiment, while the Psp-29 appeared to be tetrasomic in our experiment as well as in the description of Heist et al. [24]. The analyses proved the presence of both the Russian sturgeon and American paddlefish genomes in all examined hybrid individuals, discarding the possibility of spontaneous gynogenesis, which was observed in Siberian sturgeons [20,49] and sterlet [54]. The two tetrasomic markers enabled us to follow the segregation of the tetraploid Russian sturgeon genome and to analyze ploidy levels. These markers showed the presence of only one American paddlefish and two Russian sturgeon alleles in the SH genomes, while in the LH genomes, more alleles were found. This suggested the presence of spontaneous duplication of the maternal genomes in the individuals with high chromosome numbers (250 ± 8-304 ± 4) and/or nucleus size.
The polyploid status of the species within the Acipenserdae family, as well as their similar karyotype and genome, contribute to easy interspecific and intergeneric hybridization [3]. Based on our findings, successful interspecific hybridization is also supposed in other species combinations in the order Acipenseriformes. Consequently, American paddlefish may hybridize with other sturgeons independent of ploidy level.
The aquaculture potential of this hybrid is a subject for debate. Sturgeon hybrids are commonly used in aquaculture to exploit the advantages of heterosis [55]. It is estimated that hybrids account for approximately 35 and 20% percent of global sturgeon meat and caviar production, respectively [56]. If the planktivorous feeding habit of parent species of paddlefish is inherited to a certain extent, the new Genes 2020, 11, 753 13 of 15 hybrid can play an important role in adapting pond aquaculture to the challenges of climate change. Non-fed species have a lower carbon footprint than fed ones, and the co-culture of a filter feeder with relatively high market value as a supplementary species would strengthen pond aquaculture both from an ecological and environmental perspective [57].