Genome Duplication Events and Functional Reduction of Ploidy Levels in Sturgeon (Acipenser, Huso and Scaphirhynchus)

Corresponding author: Arne Ludwig, Institut für Gewässerökologie und Binnenfischerei, Müggelseedamm 310, 12561 Berlin, Germany., al{at}igb-berlin.de (E-mail)

Communicating editor: D. J. GRUNWALD

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Sturgeon (order Acipenserformes) provide an ideal taxonomic context for examination of genome duplication events. Multiple levels of ploidy exist among these fish. In a novel microsatellite approach, data from 962 fish from 20 sturgeon species were used for analysis of ploidy in sturgeon. Allele numbers in a sample of individuals were assessed at six microsatellite loci. Species with 120 chromosomes are classified as functional diploid species, species with 250 chromosomes as functional tetraploid species, and with 500 chromosomes as functional octaploids. A molecular phylogeny of the sturgeon was determined on the basis of sequences of the entire mitochondrial cytochrome b gene. By mapping the estimated levels of ploidy on this proposed phylogeny we demonstrate that (I) polyploidization events independently occurred in the acipenseriform radiation; (II) the process of functional genome reduction is nearly finished in species with 120 chromosomes and more active in species with 250 chromosomes and 500 chromosomes; and (III) species with 250 and 500 chromosomes arose more recently than those with 120 chromosomes. These results suggest that gene silencing, chromosomal rearrangements, and transposition events played an important role in the acipenseriform genome formation. Furthermore, this phylogeny is broadly consistent with previous hypotheses but reveals a highly supported oceanic (Atlantic-Pacific) subdivision within the Acipenser/Huso complex.

STURGEON have been fishes of great public interest for several centuries. They include the most valuable freshwater fish species worldwide, the producer of black caviar. Since the middle ages, Eurasian species (Acipenser gueldenstaedtii, A. stellatus, A. sturio, and Huso huso) have been used for production of caviar. Since the colonization of North America by European settlers, North American species (A. brevirostrum, A. oxyrinchus, and A. transmontanus) have also been used for production of caviar. This exploitation resulted in a drastic decline of all populations. All species are currently near extinction or strongly threatened (BIRSTEIN 1993A ).

All sturgeon species are members of the order Acipenserformes. Early debate about their classification arose from their unusual mixture of morphologic characters. For this reason, Acipenserformes occupy a special place in historical ideas about the classification and evolution of fishes (BEMIS et al. 1997 ). Sturgeon fossils are known from the Lower Jurassic (200 million years ago; BEMIS et al. 1997 ). All fossil and recent taxa arose in the Holarctic. Basic questions such as the chromosomal evolution or the phylogeny of these living fossils are still unresolved. Their evolutionary age, their special place in fish systematics, and their wide distribution in the Northern hemisphere, together with the large public interest, make sturgeon a good model for studies of molecular evolution in fishes.

Previous studies demonstrate that assigning an exact number of chromosomes can be somewhat arbitrary depending on the technique used that may permit observation of varying numbers of microchromosomes. For example, the number of chromosomes reported in A. transmontanus ranged from 230 (HEDRICK et al. 1991 ) to 248 ± 8 (FONTANA 1994 ) up to 276 (VAN EENENNAAM et al. 1998A ). Species of the genera Acipenser, Huso, Scaphirynchus, and Polyodon are separable into three different classes of chromosome numbers: (I) species with 120 chromosomes including all taxa with between 110 and 130 chromosomes; (II) species with 250 chromosomes including all taxa with between 220 and 276 chromosomes; and (III) species with 500 chromosomes (reviewed in BIRSTEIN et al. 1997 , Table 1). The ploidy levels of recent acipenseriform species are controversial in the literature.

Some authors believe that all species with 120 chromosomes are tetraploid (e.g., OHNO et al. 1969 ; DINGERKUS and HOWELL 1976 ; BIRSTEIN and VASILIEV 1987 ; BIRSTEIN et al. 1997 ); others call them functionally diploid (FONTANA 1994 ; FONTANA et al. 1998A , FONTANA et al. 1998B ; JENNECKENS 1999 ; TAGLIAVINI et al. 1999 ). Divergent opinions concerning ploidy levels also apply to species with higher chromosome numbers, even in studies that use the same general methods. For instance, on the basis of observed numbers of stained active nuclear organizer regions (NORs) in four different sturgeon species, BIRSTEIN and VASILIEV 1987 suggested tetraploidy for species with 120 and octaploidy for species with 250 chromosomes. However, using data on the chromosomal assignment of NORs, it was shown that these elements occurred on different chromosome pairs rather than on quadruplets (FONTANA 1994 ; FONTANA et al. 1998B ). Therefore, these authors suggested diploidy for sturgeon with 120 chromosomes and tetraploidy for sturgeon with 250 chromosomes. Nonuniform results concerning the ploidy level of acipenseriform species were also found among studies focused on expressed gene products. BIRSTEIN et al. 1997 observed duplicated protein loci in species with 120 chromosomes, which they attributed to tetraploidy. By contrast, KUZ'MIN 1996 assigned diploidy to species with 120 chromosomes in a densitometric study of serum albumins.

In our study we used highly polymorphic markers (microsatellites) that are randomly distributed in the nuclear genome and inherited in a Mendelian fashion (MAY et al. 1997 ) to investigate ploidy levels of 20 sturgeon species. Because of the large number of acipenseriform species and the large sample size we wished to examine, traditional methods for assessing ploidy levels and chromosome numbers were not feasible. The assumption of our work was that the number of alleles at a single locus should be a minimal reflection of the ploidy level. We examined a large sample of mostly wild-caught individuals to exclude interpretation difficulties that may result from inbreeding or small numbers of parents.

Hypotheses concerning the evolution of genome formation require testing in a phylogenetic context (PAGEL and HARVEY 1988 ). Several recent articles have addressed the phylogeny of Acipenseriformes (ARTYUKHIN 1995 ; BEMIS et al. 1997 ; BIRSTEIN and BEMIS 1997 ; BIRSTEIN and DESALLE 1998 ; KRIEGER et al. 2000 ; LUDWIG et al. 2000 ). Cladistic analyses indicate that Acipenser is a sister taxon to the monophyletic lineage formed by Scaphirhynchus and Pseudoscaphirhynchus, and together these three genera comprise the subfamily Acipenserinae with Huso as its sister taxon (FINDEIS 1997 ). MAYDEN and KUHAJDA 1996 supported the monophyly of the genus Acipenser with only one autapomorphy. BIRSTEIN and DESALLE 1998 and LUDWIG et al. 2000 called the validity of Huso as a separate taxonomic unit into question. Although both studies dealt with most of the extant sturgeon species (BIRSTEIN and DESALLE 1998 , 24 species; LUDWIG et al. 2000 , 19 species), they resulted in controversial topologies. LUDWIG et al. 2000 observed a separation of the Atlantic and the Pacific species, which was not found by BIRSTEIN and DESALLE 1998 . This separation was supported by other studies that focused only on North American Atlantic and Pacific species (BROWN et al. 1996 ; KRIEGER et al. 2000 ). In this study, we investigated ploidy level in light of molecular phylogeny. To accomplish this, we examined sequences of the entire mitochondrial cytochrome b gene. Three additional Asian Pacific species, A. schrenckii, A. sinensis, and H. dauricus, were added to the data of LUDWIG et al. 2000 with the aim of bringing more light to this postulated Atlantic-Pacific separation.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Fish samples and DNA extraction:
The acipenseriform species examined in this study and their geographical origins are given in Table 1. Sample sizes are listed in Table 2. In addition, data from 96 artificially reproduced juveniles of A. sturio resulting from the last successful artificial reproduction of this species in 1995 (WILLIOT et al. 1997 ) were used to examine the inheritance of two microsatellite loci. These fish are reared under aquaculture conditions for future restocking programs in France and Germany. The DNA was isolated from fin or blood samples using QIAamp blood and tissue kits (QIAGEN, Hilden, Germany).

Microsatellite analyses:
Allele number and lengths were determined for the loci amplified by the following primer pairs designed from lake sturgeon (A. fulvescens): Afu-19, Afu-34, Afu-39, Afu-54, Afu-57, and Afu-68 (MAY et al. 1997 ). Primers were labeled with 6-FAM, HEX, and TET (ABI, Foster City, CA). Amplifications were performed in a total volume of 25 µl containing 50 ng genomic DNA, 0.25 units Taq DNA polymerase (QIAGEN), 5 pmol of each primer, 0.10 mM Tris-HCl (pH 8.8 at 25°), 50 mM KCl, 1.5 mM MgCl₂, 0.1 µg/µl bovine serum albumin (BSA), 0.08% (v/v) Nonidet P40, and 100 µM of each dNTP. Amplification was performed with 30 cycles of the following steps: 30 sec at 94°, 30 sec at 52° up to 57° (Table 2), 30 sec at 72°, and a 5-min final extension at 72°. The size of alleles was determined using an ABI 310 DNA sequencer with internal standards. In principle, detection of two electropherogram peaks differing in size from one diploid individual would represent two different alleles (four different peaks would represent four alleles in a tetraploid species, etc.). For those fish exhibiting two or more peaks differing in size, the ratio of peak heights and areas was calculated with ABI GENSCAN 2.0 and these data were used to estimate gene dosages in the same manner as we performed in JENNECKENS et al. 2001 .

Amplification and sequencing of cytochrome b:
The entire cytochrome b gene (cyt-b) and some flanking sequences, comprising 1221 bp, were amplified using flanking tRNA primers L-14735 (WOLF et al. 1999 ) and cytb-rev1, as well as four additional primers to permit sequencing of overlapping fragments, as described in LUDWIG et al. 2000 . The following reaction conditions were used: 60 sec at 94°, 30 sec at 55°, 90 sec at 72° for 30 cycles, and a final extension at 72° for 5 min. Amplifications were carried out with 1 unit Taq DNA polymerase (QIAGEN), 10 pmol of each primer, 50 ng DNA, 0.10 mM Tris-HCl (pH 8.8 at 25°), 50 mM KCl, 1.5 mM MgCl₂, 0.1 µg/µl BSA, 0.08% (v/v) Nonidet P40, and 100 µM of each dNTP. PCR products were run on a 1.5% agarose gel at 150 V for 2 hr. The bands of interest were excised from the gel and extracted using the QIAquick gel extraction kit (QIAGEN). Direct sequencing was performed in both directions using the Big Dye RR Terminator cycle sequencing kit (ABI); sequencing products were visualized using an ABI 310.

Microsatellite data analysis:
The microsatellite-allele-counting approach used for investigation of ploidy levels is limited in that the number of alleles in an individual is an estimate of the ploidy level. The number of alleles observed in a single individual, within a sample of individuals, may not represent the maximum possible number of alleles at that locus for that species. Furthermore, the observation of four alleles in a single individual may indicate tetrasomic inheritance at that locus, or it may indicate the presence of two disomically inherited loci. Theoretically, low levels of polymorphism may be especially prevalent in small populations such that it would be highly unlikely to find an individual with the maximum possible number of alleles. This problem is more acute in 2n species than in 4n species or 8n species because the number of heterozygous individuals is larger in these systems (ALLENDORF et al. 1975 ; NADEAU and SANKOFF 1997 ). To increase our confidence in the evidence presented here, we examined allelic distributions of artificially produced A. sturio (120 chromosomes), demonstrating the inheritance pattern for the loci examined in this species.

Sequence data analysis:
We computed basic sequence statistics with the program DnaSP 2.2 (ROZAS and ROZAS 1997 ). Analysis of molecular variance (AMOVA) values were computed with the program ARLEQUIN 2.000 (SCHNEIDER et al. 2000 ). Tree calculations in this study were performed on the entire cyt-b gene (1141 bp). Phylogenetic relationships among taxa were estimated using maximum parsimony and maximum-likelihood algorithms, implemented by the PAUP* (SWOFFORD 1998 ) and PUZZLE v. 2.5.1. (STRIMMER and VON HAESELER 1996 ) program packages. Bootstrap analyses were performed with 1000 iterations. The consistency index (CI) and retention index (RI) were calculated in PAUP* (SWOFFORD 1998 ). Maximum-likelihood/quartet-puzzling support values are equivalent to bootstrap values.

Spectral analysis was conducted using the program SPECTRUM 2.0 (CHARLESTON 1996 ) to measure the amount of support for, and conflict against, a given phylogenetic arrangement. A spectrum is a representation of all inferred or estimated branch lengths of a phylogenetic tree, which is generated without the necessity of actually inferring a tree. For a given set of sequences of aligned characters, the support for a hypothesized partition (split) of the sequences into two parts is the sum of the sequence positions that have a unique character state in that partition. The conflict value for a split is the sum of all other splits that contradict the partitioning of sequences in the first split. A detailed description and application of this method is given in LENTO et al. 1995 and PITRA and VEITS 2000 . The main focus of this study is the Atlantic-Pacific separation.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Microsatellite data:
Table 2 summarizes the ranges of allele sizes and number of alleles observed at each locus within each species.

Table 3 gives an overview of the ploidy levels inferred from microsatellite analyses. Wherever a ">" precedes the allelic band patterns observed (e.g., 8n), the exact number of alleles was not detectable using the quantitative measurements obtained by GENESCAN, resulting from technical limitations. PCR amplification of microsatellites often generates stutter, that is, products that are one or a couple of motif sizes smaller or larger than the actual intended product size. These bands are less well amplified than the actual microsatellite product, but they create false peaks in an electropherogram, making gene dosage calculations often difficult. This problem is exacerbated if there are eight or more alleles that are similar in size. In addition, the strength of amplification depends not only on the number of DNA copies in the genome but also on the length of the fragment. Shorter fragments sometimes amplify more intensively. Both factors make an exact assessment of the allele number using gene dosage calculations difficult. In these cases, however, the number of single alleles was larger than the number shown. Some microsatellite patterns were not evaluated in some species because they did not amplify (indicated by a dash); amplifications at other loci produced a single allele (indicated by M). With few exceptions (see Table 3), the microsatellite loci examined in this study, Afu-19, Afu-34, Afu-39, Afu-54, Afu-57, and Afu-68, showed disomic allelic patterns in species with 120 chromosomes and tetrasomic allelic patterns in species with 250 chromosomes. In one group of species (A. brevirostrum, A. fulvescens, A. gueldenstaedtii, A. medirostris, A. mikadoi, and A. naccarii), microsatellite patterns indicated octosomic or greater allelic band patterns at a minimum of one locus. Allelic band patterns in an additional four species (A. baerii, A. persicus, A. sinensis, and A. transmontanus) showed evidence of possible octosomy (or greater than tetrasomy) at a minumum of one locus. Allelic band patterns greater than octosomic were observed in A. mikadoi at two loci.

None of the microsatellite loci examined in Acipenseridae amplified in P. spathula. Microsatellite data obtained in this study are used for discussion of ploidy levels and not for questions of population genetics. Aspects of population genetics or interspecific distribution of alleles are discussed in previous publications (MAY et al. 1997 ; JENNECKENS 1999 ; JENNECKENS et al. 2001 ; LUDWIG et al. 2000 ). Additional discussions of other aspects of population genetics will be published elsewhere.

Inheritance of microsatellites:
The sire and the dam used in creation of A. sturio showed allelic variability at two loci, Afu-19 (118/142:118/118) and Afu-54 (184/184:184/188) in one of the parents. Gene dosage assessments indicated that the allele ratios for the heterozygotic individuals were 1:1. Therefore, parental genotypes of the heterozygous individual had to be 118:142 if disomic or 118:118:142:142 if tetrasomic at Afu-19 and 184:188 if disomic or 184:184:188:188 if tetrasomic at Afu-54. The offspring were 52% homozygous for allele 118 and 48% heterozygous for alleles 118 and 142 at Afu-19. At Afu-54, 48% were homozygous for allele 184 bp and 52% were heterozygous for alleles 184 and 188. Approximately equal ratios of homozygous and heterozygous offspring confirm disomic inheritance at these loci for this species.

Cyt-b characteristics:
Sequences are archived in the EMBL GenBank (AJ245825-41, AJ249692-4, AJ251451, and AJ252186-7).

The 1221-bp sequences generated by single or multiple amplifications included the entire cyt-b gene and an additional 80 bp of related tRNA genes. No direct stop-codon was detected in any of the sequences obtained. The TAA stop signal of the cyt-b gene is completed by the addition of 3' A residues to the mRNA. The 34 bp (primer sequence included) upstream of the cyt-b gene was derived from tRNA^Glu. In most acipenseriform species, two nucleotides serve as a spacer between the tRNA^Glu gene that precedes the cyt-b gene and cyt-b itself; however, the spacer was composed of three nucleotides in A. nudiventris, A. medirostris, and A. mikadoi. In the acipenseriforms, cyt-b continues directly into the tRNA^Thr gene without any spacing nucleotides. Downstream of the cyt-b gene, inclusive of primer nucleotides, 46 nucleotides in the amplified fragment derived from the tRNA^Thr gene were amplified. The 1141 bases comprising the cyt-b gene sequences from 22 acipenseriform taxa included 324 variable characters (28.4% of the total cyt-b gene sequence), with most of the changes (79.9%) occurring at the third codon position. The majority of the changes (90.7%) were synonymous; those remaining resulted in 18 amino acid substitutions that were informative for parsimony analysis. Most amino acid substitutions involved exchanges between hydrophobic residues. Alignment of the 22 cyt-b sequences revealed no nucleotide insertions or deletions within sequences. The cyt-b sequences can be fully translated using the bovine mitochondrial code (ANDERSON et al. 1982 ) without nonsense or intervening stop codons. Finally, the alignment revealed neither an increased abundance of first and second codon position changes nor a shift in the typical mammalian mtDNA transition bias, both of which are known to occur in mtDNA sequences translocated to the nuclear genome (ZHANG and HEWITT 1996 ).

Molecular phylogeny:
An identical branching topology was best supported by both maximum parsimony (single-most-parsimonious tree L = 1101 steps, CI = 0.639, RI = 0.645) and maximum-likelihood (LnL = -6645.70; transition/transversion ratio, 2.7 ± 0.2) analyses. A representative maximum-likelihood tree is shown in Fig 1. Four robustly supported conclusions can be inferred from the trees: (I) Acipenser and Scaphirhynchus form a monophyletic assemblage, the Acipenserinae (maximum likelihood = 84%, maximum parsimony = 100%); (II) the A. oxyrinchus-A. sturio cluster (maximum likelihood = 100%, maximum parsimony = 100%) is basal for the clade; (III) the two species of Huso are embedded within the genus Acipenser—it appears that the genus Huso had a polyphyletic origin or that it is an inappropriate taxonomic unit; (IV) there are two monophyletic groups within the Acipenser/Huso assemblage. These include the Atlantic species A. baerii, A. brevirostrum, A. fulvescens, A. gueldenstaedtii, A. naccarii, A. nudiventris, A. persicus, A. ruthenus, A. stellatus, and H. huso as one group (maximum likelihood = 100%, maximum parsimony = 94%); and the Pacific species A. medirostris, A. mikadoi, A. schrenckii, A. sinensis, A. transmontanus, and H. dauricus as a second group (maximum likelihood = 96%, maximum parsimony = 71%). To further assess the reliability of this previously uncovered subdivision in Acipenser (LUDWIG et al. 2000 ), we used spectral analysis to measure the definitive phylogenetic signals in our data set, supporting or conflicting with this hypothesis (Fig 2). The resulting spectrum indicates that the second strongest split signal (see split 16,775) in the cyt-b data set supports the partition of the Acipenser/Huso complex independently of any conventional tree-construction method. It is interesting to note that 11 of the most frequent partitions also occur in the corresponding maximum-likelihood tree (Fig 1). Therefore, spectral analysis demonstrates both that there is a high level of phylogenetic signal in our cyt-b data set and that the partition of the Acipenser/Huso complex is well supported.

View larger version (21K): In this window In a new window Download PPT slide   Figure 1. Phylogenetic tree based on cytochrome-b gene sequences estimated by the maximum-likelihood algorithm using PUZZLE. The quartet-puzzling support values are shown at each node and were derived from 1000 puzzling steps. The tree was rooted using the corresponding outgroup sequences from Salmo salar and Polypteris ornatipinnis.

View larger version (40K): In this window In a new window Download PPT slide   Figure 2. Spectral analysis of the cytochrome-b DNA sequences from 16 species of the genera Acipenser and Huso. Bars above the x-axis represent frequency of support for each taxonomic grouping (split). Bars below the x-axis represent the sum of all conflicts against the corresponding split above the x-axis. Solid bars (both support and conflict) indicate the splits that were selected for inclusion in the optimal tree shown in the inset. Terminal taxa (shaded bars) are labeled by the corresponding taxon name, and internal splits (open bars) are labeled by their split number. There are 32,768 possible splits in this data set and the vast majority have little or no support, so only the strongest 27 are shown. The asterisk indicates the second highest-ranking split supporting the oceanic subdivision in the Acipenser/Huso complex.

Acipenserid phylogeography:
The primary phylogenetic subdivision in the Acipenser/Huso complex correlated with the current geographic distribution. The first monophyletic group contained 10 species inhabiting tributaries of the Atlantic Ocean. The second group included 6 species inhabiting tributaries of the Pacific Ocean. The phylogeographic bipartition involved 7.5 ± 1.2% Kimura2 corrected sequence divergence and 80.4 nucleotide differences. At the amino acid level, positions 230 (Phe Leu) and 371 (Trp Leu) are diagnostic for the oceanic divergence within the Acipenser/Huso complex. The AMOVA showed that 33.2% (P < 0.0001) of the total variance occurs between the two phylogeographic clades, and 23.4% (P < 0.0001) was distributed among species within clades.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Evolution of ploidy levels in sturgeon:
Analyses of karyotypes of 21 of the 24 or 25 extant acipenseriform species revealed that most species possessed either 120 or 250 chromosomes (reviewed in BIRSTEIN et al. 1997 , Table 3). On the basis of genome quantification estimated using flow cytometry calibrated against reference species with previously analyzed chromosome numbers, however, two sturgeon species with higher chromosome numbers were identified. Flow cytometry allows inference of ploidy levels from the ratios of DNA content among species, provided the basal ploidy level is known. It proved to be informative in developing hypotheses regarding genome evolution. For example, A. medirostris and A. mikadoi are almost indistinguishable morphologically and were considered to be one species by some authors (ARTYUKHIN and ANDRONOV 1990 ; ARTYUKHIN 1995 ). A taxonomic separation of the North American species (A. medirostris) from the East Asian Pacific coast species (A. mikadoi) was evident from measurements of DNA content. In A. mikadoi, the DNA content is approximately two times higher (14.4 pg) than in sturgeon species with 250 chromosomes (7.8–8.3 pg), including A. medirostris, and approximately four times higher than in species with 120 chromosomes (BIRSTEIN 1993B , BIRSTEIN 1993C ). On the basis of DNA content, therefore, BIRSTEIN 1993C and BIRSTEIN et al. 1997 inferred that A. mikadoi has 500 chromosomes. We propose that A. mikadoi is octaploid on the basis of our microsatellite data.

BLACKLIDGE and BIDWELL 1993 propose that A. brevirostrum is an allopolyploid species, a descendant of ancestral spontaneous triploids (12n = 360 chromosomes). A hybrid origin of A. brevirostrum between species with 250 and 500 chromosomes, as they proposed, may be possible, but no species with 500 chromosomes are known within the Atlantic group, which could be put forth as its likely ancestral form. Allopolyploidy is extremely rare in other acipenserids, and therefore BIRSTEIN et al. 1997 found it more probable that this is a 16n species; this proposal is in keeping with their inferred chromosome number of 500. On the basis of evidence at one microsatellite locus, we propose octaploidy in A. brevirostrum. A. brevirostrum is a U.S. federally endangered (since 1973) species and therefore we were able to obtain only six samples. The low sample size, and a recent decrease in population size that could have increased genetic drift, could reduce the total number of microsatellite alleles observed, which could reduce the ploidy number inferred from these data. Additional analyses of chromosome numbers of A. brevirostrum would be necessary to provide conclusive evidence regarding the ploidy level of this species.

The in vitro amplification of polymorphic nuclear markers, such as microsatellite loci, permits a direct although partial view of the genome. Observation of the maximum number of alleles per individual at a given locus provides evidence of ploidy level at that locus. For a diploid locus a maximum of two alleles differing in size (base pairs) is expected. For species with 120 chromosomes, one or two peaks were detected at most microsatellite loci within each individual. For a heterozygous specimen the two peaks observed showed an approximately equal peak area and size. On the basis of the assumption that species with 120 chromosomes are diploid, the presence of four alleles in one individual at a locus implies tetraploidy. The presence of no more than two alleles may indicate diploidy at that locus but does not contradict tetraploidy (or higher). Allelic band patterns inferred by methods such as these (including also isozyme data) indicate the maximum number of homologous forms a locus may have in a genome. However, evidence of a disomic band pattern at one locus precludes neither tetraploidy at another locus nor that the species in question is tetraploid. Rather, should a mixture of evidence be found in a species, it indicates a transitional ploidy level in which some loci have diverged sufficiently from their genomic duplicates such that they no longer amplify in polymerase chain reaction or are no longer expressed as protein products (ALLENDORF and THORGAARD 1984 ). This transition toward diploidy, or diploidization, is expected to occur in polyploid vertebrates until a species may be considered to be functionally diploid. Diploidization is completed when only divalents are formed during meiosis; this behavior, however, is extremely difficult to observe empirically, especially in species with many chromosomes, and thus ploidy levels are more commonly inferred indirectly. Even in fully diploidized species, residual evidence of polyploid ancestry is occasionally observed because of gene homology. Similarly, tandem duplications will create apparent polyploidy when those genes are examined.

Among 11 microsatellite loci studied by PYATSKOWIT et al. 2001 in A. fulvescens (250 chromosomes), octosomic allelic patterns were never observed; all loci showed di- or tetrasomic allelic patterns. In another study, 1 of these loci, Afu-39, showed no more than two alleles in 501 fish sampled from six sturgeon species with 120 chromosomes and four alleles in 265 samples from four species with 250 chromosomes (JENNECKENS et al. 2001 ). This supported di- and tetraploidy rather than tetra- and octaploidy in species with 120 and 250 chromosomes, respectively. However, although microsatellite loci Afu-19, Afu-34, Afu-39, and Afu-54 investigated in the present study support this pattern, loci Afu-57 and Afu-68 appear to be duplicated in species with 250 chromosomes (Table 3). No more than two alleles were observed in all individuals at all loci for all species with 120 chromosomes (with the exception of Afu-68 in H. dauricus, Table 3). This pattern strongly indicates a functional diploidy for species with 120 chromosomes. Additional functional diploidy in species with 120 chromosomes was also supported by the inheritance of two loci in A. sturio.

All phylogenetic trees (Fig 1) indicate that Polyodon spathula is basal to all other sturgeon examined in this study. This finding is consistent with common understanding of ancient divergence of the families Polyodontidae and Acipenseridae in acipenseriform evolution (e.g., BIRSTEIN and DESALLE 1998 ). At the next level, the three species of Scaphirhynchus and the A. sturio/A. oxyrinchus clade are basal within the Acipenseridae. All six of these species have 120 chromosomes; thus, it is likely that the common ancestor of all members of the family Acipenseridae had this chromosome number.

We agree with BIRSTEIN et al. 1997 that the common diploid ancestor of all Acipenseriformes had a karyotype of 60 chromosomes. A genome duplication event must have occurred in this ancestor to produce the Acipenserid lineage, whose basal species have 120 chromosomes. A reestablishment of the diploid condition, as postulated by FONTANA 1994 , before the radiation of this order seems plausible. The proposal of genome duplication and subsequent reduction in functional ploidy is not rare; even in the evolution of diploid vertebrates, genome quadruplication is discussed (LEIPOLDT 1983 ; SPRING 1997 ). Although the chromosome number of some Acipenseriformes, for example, Psephurus gladius (family Polyodontidae), is not known at this time, additional genome duplication events seem to have evolved more recently because they are observed in certain species within the genus Acipenser only. We found evidence for locus duplication (Afu-57 and Afu-68) only in species with 250 or 500 chromosomes but not in species with 120 chromosomes (Table 3 and Fig 3). This supports our proposal that the nearest ancestor to Acipenserids was functionally diploid with 120 chromosomes (tetraploid derived, which is reflected in DNA content) and that species with higher chromosome numbers are the result of more recent genome duplication events. In contrast to the proposed single ancient genome duplication in a common ancestor of all salmonid fishes (ALLENDORF and THORGAARD 1984 ), our studies of sturgeon molecular phylogeny showed that several independent genome duplication events happened during sturgeon evolution (Fig 4).

View larger version (21K): In this window In a new window Download PPT slide   Figure 3. Evidence of single gene duplication events associated with chromosome number and inferred ploidy levels in sturgeon species with 120, 250, and 500 chromosomes (n = 866 individuals from 20 species).

View larger version (29K): In this window In a new window Download PPT slide   Figure 4. Ploidy evolution in the Acipenseridae, including some related Acipenseriform taxa for comparison. The tree represents the topology, calculated using maximum parsimony, of the entire cytochrome-b gene. Polyploidization events are mapped on the tree using black boxes. The existence of heteroplasmy resulting from length variations of the mtD-loop is shown by a white box (data were taken from LUDWIG et al. 2000 ; A. schrenckii, A. sinensis, and H. dauricus; A. LUDWIG, unpublished data; ZHANG 1998 ). The number of chromosomes is reviewed in BIRSTEIN et al. 1997 . See also Table 3.

It is quite possible that the observation of low numbers of duplicated locus expression in the acipenseriforms, in spite of a genome duplication event, may be explained by subsequent gene silencing, as found for Salmonids and Catostomids (FERRIS and WHITT 1978 ; STONEKING et al. 1981 ) and for paddlefish (CARLSON et al. 1982 ; DANIELSON et al. 1999 ). A good example of gene silencing is the existence of two proopiomelanocortin (POMC) genes in paddlefish (P. spathula) and sturgeon (A. transmontanus). In both of the sturgeon POMCs, the N-terminal paired-basic proteolytic cleavage site that normally flanks -melanocyte-stimulating-hormone (MSH) is missing. In place of the paired-basic residues, the two sturgeon POMC genes have nonbasic residues (ALRUBAIAN et al. 1999 ; DANIELSON et al. 1999 ). A degeneration of the -MSH core sequences was reported in one of the duplicated sturgeon POMC sequences (DANIELSON et al. 1999 ). Therefore DANIELSON et al. 1999 concluded that the loss of the basic residues at the N-terminal cleavage site occurred prior to duplication of the ancestral POMC gene and that degeneration of -MSH core sequence was a subsequent event that occurred after POMC duplication. The loss of a -MSH proteolytic cleavage site in the ancestral chondrostean lineage served to functionally eliminate the -MSH region by preventing this sequence from being cleaved from the POMC precursor (DANIELSON et al. 1999 ). Like sturgeon and paddlefish, salmonids have two POMC genes (DANIELSON et al. 1999 ). Phylogenetic analysis suggests that duplication of the salmonid POMCs was an evolutionary event that occurred independently and long after the duplication of the chondrostean POMC sequences. Compared to the Chondrosteii, however, the rates of amino acid substitution between paralogously related salmonid POMC sequences are much higher in rainbow trout and sockeye salmon (DANIELSON et al. 1999 ). Furthermore, tetraploid loaches (Cobitidae), although arising more recently than salmonids, show a greater rate of loss of duplicated genes (BAILEY et al. 1978 ). This supports the theory of very slow molecular evolution in Chondrosteii (RICO et al. 1996 ) and demonstrates that gene and/or genome duplication events happened independently in several lineages. It is likely, therefore, that the process of gene silencing has lineage-specific patterns.

BLACKLIDGE and BIDWELL 1993 observed great differences in the mean genome sizes observed among isolated populations of A. transmontanus. Apart from gene silencing, chromosomal rearrangements may be assumed to play another important role in the acipenseriform genome formation. This is heavily supported for A. transmontanus by combining the data we obtained at different microsatellite loci with the outcome of the study of VAN EENENNAAM et al. 1998B . These authors showed no evidence of multivalent formation in synaptonemal complex analyses when they examined males of this taxon. This suggests a completion of the diploidization process (VAN EENENNAAM et al. 1998B ). However, allelic scores at microsatellite loci Afu-57 and Afu-68 indicated tetrasomic allelic band patterns at these loci.

The octosomic allelic band patterns observed at Afu-54 for A. brevirostrum and the tetrasomic allelic band patterns observed at Afu-68 for H. dauricus may be explained by tandem duplications of single genes rather than polyploidization events. Observation of single gene duplications may be more common in fishes than some other taxa. Analysis of zebrafish showed that these fish often have more multigene families than do mammals (POSTLETHWAIT et al. 1998 , POSTLETHWAIT et al. 2000 ). The additional genes in zebrafish could have arisen from either (I) a greater propensity of tandem gene duplications in the fish lineage, (II) less loss of paralogous gene copies in fish, or (III) additional independent chromosome duplication or tetraploidization in fish. Mapping data gave no evidence for the clustering of additional genes in zebrafish, thus reducing the likelihood that tandem duplication is a general mechanism for gene duplication (POSTLETHWAIT et al. 1998 ). A lower rate of molecular evolution in fish is often described (RICO et al. 1996 ), thus providing support for the second proposed explanation. Our data corroborate the proposal of additional independent chromosome duplication or tetraploidization because they show this effect in one of the most primitive fish lineages.

Within all 492 wild-caught specimens of species with 120 chromosomes, the number of alleles per locus ranged between 2.22 (Afu-39) and 6.71 (Afu-57). This range was lower than the range observed in 363 specimens of species with 250 chromosomes, between 5.25 (Afu-34) and 12.77 (Afu-68). From this evidence, we propose that higher levels of ploidy permit a greater flexibility in accommodating mutations that modify functions because of complementation among partially divergent gene family members. Mutations that are deleterious when only a single gene is present may be neutral if a gene duplicate compensates for the loss of function (NADEAU and SANKOFF 1997 ).

These data, taken as a whole, indicate strongly an ongoing process of functional reduction of ploidy level in sturgeon. The data are highly consistent in indicating the following functional ploidy levels: for species with 120 chromosomes, 2n; 250 chromosomes, 4n; and 500 chromosomes, 8n.

Molecular phylogeny and acipenserid phylogeography:
To investigate the consistency of the chromosomal acipenseriform evolution that we proposed in this study, the chromosome number for each species was superimposed on a reconstructed molecular phylogenetic tree based on the entire cyt-b gene sequence (Fig 4). All previous studies of molecular phylogeny of sturgeon resulted in the following agreements: Acipenser and Scaphirhynchus form a monophyletic assemblage, the A. oxyrinchus/A. sturio cluster is most distantly related to the other species of the genus Acipenser, and species of the genus Huso do not cluster together, but within separate clusters of the genus Acipenser (BROWN et al. 1996 ; BIRSTEIN and DESALLE 1998 ; KRIEGER et al. 2000 ; LUDWIG et al. 2000 ). We address a postulated oceanic subdivision of the order, Acipenseriformes, into Atlantic and Pacific lineages (LUDWIG et al. 2000 ). From our phylogenetic tree (Fig 1), a separation into two clades, Atlantic and Pacific, emerged. The spectral analysis indicated that the second strongest subdivision was an internal node (split 16,775; Fig 2), reflecting the same primary oceanic subdivision in Acipenser and Huso observed in the conventional phylogenetic reconstructions, independently corroborating this proposal.

The evolution of chromosomes and distribution of heteroplasmy (Fig 4), the structure of repeat units and general organization of the mtD-loop region (BUROKER et al. 1990 ; BROWN et al. 1992 , BROWN et al. 1996 ; LUDWIG et al. 2000 ), as well as geographic distribution (discussed in LUDWIG et al. 2000 ) strongly supported the molecular phylogeny constructed in this study (Fig 1 and Fig 4).

Concluding remarks and future perspectives:
Distribution of allele number within microsatellites shows an ongoing process of functional genome copy reduction in sturgeon. The number of alleles observed at Afu-57 and Afu-68 showed the process of diploidization to be more recent in species with 250 chromosomes and 500 chromosomes. Considering the outcomes of previous studies and this study, it appears that gene silencing, chromosomal rearrangements, and transposition events are playing important roles in the acipenseriform genome formation. We believe that the common diploid ancestor of Acipenseriformes had 60 chromosomes. A genome duplication event must have occurred in this ancestor. A genomic duplication event, followed by a reestablishment of a functional diploid condition before the radiation of this order, seemed to be plausible. Functional ploidy levels as indicated by our microsatellite approach are for species with 120 chromosomes = 2n, 250 chromosomes = 4n, and 500 chromosomes = 8n. In contrast to single genome duplication events in other fish taxa, our results indicated that genome duplication occurred repeatedly during sturgeon evolution, making sturgeon an ideal taxonomic group to address the selective pressures that favored extensive duplication of the genome.

	ACKNOWLEDGMENTS

The authors recognize numerous staff members and professional fishermen from Albania, Canada, France, Georgia, Germany, Iran, Netherlands, Romania, Russia, and U.S. game fish agencies for logistic support, obtaining specimens, and permitting catches during protected spawning periods (P. Anders, E. N. Artyukhin, V. Birstein, H. Bolig, P. Bronzi, M. Chebanov, L. Debus, B. Elvira, J. Fletcher, J. Gessner, O. Goemann, R. Gross, T. Gulyas, H. Kincaid, F. Kirschbaum, S. LaPan, M. Lepage, B. May, N. Patriche, M. Pourkazemi, E. Rochard, C. Tracy, and P. Williot). We thank J. Becker, C. Kaltwasser and D. Lieckfeld for technical assistance. We also thank B. May, D. J. Grunwald, and two anonymous reviewers for helpful comments on an earlier draft, as well as J. D. Pyatskowit, T. King, C. Grunwald, and all coauthors for providing us with their submitted manuscripts. In addition, we thank P. Williot for data about the brood stock of A. sturio. The studies were made possible by grants from the Deutsche Forschungsgemeinschaft (KI 189/11-1 and ME 712/12).

Manuscript received January 22, 2001; Accepted for publication April 24, 2001.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

ALLENDORF, F. W., and G. H. THORGAARD, 1984 Tetraploidy and the evolution of salmonid fishes, pp. 1–45 in Evolutionary Genetics of Fishes, edited by B. J. TURNER. Plenum Press, New York.

ALLENDORF, F. W., F. M. UTTER and B. P. MAY, 1975 Gene duplication within the family Salmonidae: II. Detection and determination of the genetic control of duplicate loci through inheritance studies and the examination of populations, pp. 415–432 in Isozymes IV: Genetics and Evolution, edited by C. L. MARKERT. Academic Press, New York.

ALRUBAIAN, J., P. DANIELSON, M. FITZPATRICK, C. SCHRECK, and R. M. DORES, 1999  Cloning of a second proopiomelanocortin cDNA from the pituitary of the sturgeon, Acipenser transmontanus. Peptides 20(4):431-436[Medline].

ANDERSON, S. M., H. L. DE BRUIN, A. R. COULSON, I. C. EPERON, and F. SANGER et al., 1982  Complete sequence of bovine mitochondrial genome. Nature 290:457-464.

ARTYUKHIN, E. N., 1995  On biogeography and relationships within the genus Acipenser. Sturgeon Q. 3:6-8.

ARTYUKHIN, E. N. and A. E. ANDRONOV, 1990  A morphological study of the green sturgeon, Acipenser medirostris (Chondrostei, Acipenseridea), from the Tumnin (Datta) River and some aspects of the ecology and zoogeography of Acipenseridae. Zool. Zhurn. 69:81-91.

BAILEY, G. S., T. M. RUSSEL, T. M. POULTER, and P. A. STOCKWELL, 1978  Gene duplication in tetraploid fish: model for gene silencing at unlinked duplicated loci. Proc. Natl. Acad. Sci. USA 75:5575-5579[Abstract/Free Full Text].

BEMIS, E., E. K. FINDEIS and L. GRANDE, 1997 An overview of Acipenseriformes, pp. 25–71 in Sturgeon Biodiversity and Conservation, edited by V. J. BIRSTEIN, J. R. WALDMAN and W. E. BEMIS. Kluwer Academic Publishing, Dordrecht, The Netherlands.

BIRSTEIN, V. J., 1993a  Sturgeons and paddlefishes: threatened fishes in need of conservation. Conserv. Biol. 7:773-787.

BIRSTEIN, V. J., 1993b  DNA content in three Eurasian sturgeon species determined by flow cytometry. Cytometry 14:377-383[Medline].

BIRSTEIN, V. J., 1993c  Is Acipenser medirostris one or two species? Sturgeon Q. 1:2.

BIRSTEIN, V. J., and E. BEMIS, 1997 How many species are there within the genus Acipenser? pp. 157–163 in Sturgeon Biodiversity and Conservation, edited by V. J. BIRSTEIN, J. R. WALDMAN and W. E. BEMIS. Kluwer Academic Publishing, Dordrecht, The Netherlands.

BIRSTEIN, V. J. and R. DESALLE, 1998  Molecular phylogeny of Acipenserinae. Mol. Phylogenet. Evol. 9:141-155[Medline].

BIRSTEIN, V. J. and V. P. VASILIEV, 1987  Tetraploid-octaploid relationships and karyological evolution in the order Acipenseriformes (Pisces). Karyotypes, nucleoli, and nucleolus-organizer region in four acipenserid species. Genetica 73:3-12.

BIRSTEIN, V. J., R. HANNER and R. DESALLE, 1997 Phylogeny of the Acipenseriformes: cytogenetic and molecular approaches, pp. 127–155 in Sturgeon Biodiversity and Conservation, edited by V. J. BIRSTEIN, J. R. WALDMAN and W. E. BEMIS. Kluwer Academic Publishing, Dordrecht, The Netherlands.

BLACKLIDGE, K. H. and C. A. BIDWELL, 1993  Three ploidy levels indicated by genome quantification in Acipenseriformes of North America. J. Hered. 84:427-430[Abstract/Free Full Text].

BROWN, J. R., A. T. BECKENBACH, and M. J. SMITH, 1992  Mitochondrial DNA length variation and heteroplasmy in populations of white sturgeons (Acipenser transmontanus). Genetics 132:221-228[Abstract].

BROWN, J. R., K. BECKENBACH, A. T. BECKENBACH, and M. J. SMITH, 1996  Length variation, heteroplasmy and sequence divergence in the mitochondrial DNA of four species of sturgeon (Acipenser). Genetics 142:525-535[Abstract].

BUROKER, N. E., J. R. BROWN, T. A. GILBERT, P. J. O'HARA, and A. T. BECKENBACH et al., 1990  Length heteroplasmy of sturgeon mitochondrial DNA: an illegitimate elongation model. Genetics 124:157-163[Abstract].

CARLSON, D. M., M. K. KETTLER, S. E. FISHER, and G. S. WHITT, 1982  Low genetic variability in paddlefish populations. Copeia 3:721-725.

CHARLESTON, M. A., 1996 Spectrum V. 1.0.5. University of Glasgow, Glasgow, Scotland, UK.

DANIELSON, P. B., J. ALRUBAIAN, M. MULLER, J. M. REDDING, and R. M. DORES, 1999  Duplication of the POMC gene in the paddlefish (Polyodon spathula): analysis of -MSH, ACTH, and ß-endorphin regions of ray-finned fish POMC. Gen. Comp. Endocrinol. 116:164-177[Medline].

DINGERKUS, G. and W. M. HOWELL, 1976  Karyotypic analysis and evidence of tetraploidy in the North American paddlefish, Polyodon spathula. Science 194:842-844[Abstract/Free Full Text].

FERRIS, S. D. and G. S. WHITT, 1978  Phylogeny of tetraploid catastomod fishes based on the loss of duplicate gene expression. Syst. Zool. 27:189-206.

FINDEIS, E. K., 1997 Comparative osteology and relationships of recent sturgeons, pp. 73–126 in Sturgeon Biodiversity and Conservation, edited by V. J. BIRSTEIN, J. R. WALDMAN and W. E. BEMIS. Kluwer Academic Publishing, Dordrecht, The Netherlands.

FONTANA, F., 1994  Chromosomal nucleolar organizer regions in four sturgeon species as markers of karyotype evolution in Acipenseriformes (Pisces). Genome 37:888-892.

FONTANA, F., M. LANFREDI, M. CHICCA, V. AIELLO, and R. ROSSI, 1998a  Localization of the repetitive telomeric sequence (TTAGGG)_n in four sturgeon species. Chromosome Res. 6:303-306[Medline].

FONTANA, F., J. TAGLIAVINI, L. CONGIU, M. LANFREDI, and M. CHICCA et al., 1998b  Karyotypic characterization of the great sturgeon, Huso huso, by multiple staining techniques and fluorescent in situ hybridization. Mar. Biol. 132:495-501.

HEDRICK, R. P., T. S. MCDOWELL, R. ROSEMARK, D. ARONSTEIN, and C. N. LANNAN, 1991  Two cell lines from white sturgeon. Trans. Am. Fish. Soc. 120:528-534.

JENNECKENS, I., 1999 Investigations on the suitability of biochemical and molecular markers for an identification of different sturgeon species as well as their hybrids. Ph.D. Thesis, University of Göttingen, Göttingen, Germany.

JENNECKENS, I., J.-N. MEYER, G. HÖRSTGEN-SCHWARK, B. MAY, and L. DEBUS et al., 2001  A fixed allele at microsatellite LS-39 is characteristic for the black caviar producer Acipenser stellatus. J. Appl. Icht. 17:39-42.

KRIEGER, J. P., A. FUERST, and T. CAVENDER, 2000  Phylogenetic relationships of North American sturgeons (order Acipenserformes) based on mitochondrial DNA sequences. Mol. Phylogenet. Evol. 16:64-72[Medline].

KUZ'MIN, E. V., 1996  Blood serum albumin system in Acipenseriformes during river period of life. Vopr. Ikhytiol. 36:101-108.

LEIPOLDT, M., 1983  Towards an understanding of the molecular mechanism regulating gene expression during diploidization in phylogenetically polyploid lower vertebrates. Hum. Genet. 65:11-18[Medline].

LENTO, G. M., R. E. HICKSON, G. K. CHAMBERS, and D. PENNY, 1995  Use of spectral analysis to test hypotheses on the origin of pinnipeds. Mol. Biol. Evol. 12:28-52[Abstract].

LUDWIG, A., B. MAY, L. DEBUS, and I. JENNECKENS, 2000  Heteroplasmy in the mtDNA control region of sturgeon (Acipenser, Huso and Scaphirhynchus). Genetics 156:1933-1947[Abstract/Free Full Text].

LUDWIG, A., I. JENNECKENS, L. DEBUS, A. LUDWIG, and J. BECKER et al., 2000  Genetic analyses of archival specimens of the European sturgeon Acipenser sturio. Boletin Instituto Espanol de Oceanografia 16:181-190.

MAY, B., C. C. KRUEGER, and H. L. KINCAID, 1997  Genetic variation at microsatellite loci in sturgeon: primer sequence homology in Acipenser and Scaphirhynchus. Can. J. Fish. Aquat. Sci. 54:1542-1547.

MAYDEN, R. L., and B. R. KUHAJDA, 1996 Systematics, taxonomy, and conservation status of the endangered Alabama sturgeon, Scaphirhynchus suttkusi Williams and Clemmer (Actinopterygii, Acipenseridae). Copeia 1996: 241–273.

NADEAU, J. H. and D. SANKOFF, 1997  Comparable rates of gene loss and functional divergence after genome duplications early in vertebrate evolution. Genetics 147:1259-1266[Abstract].

OHNO, S., J. MURAMOTO, C. STENIUS, L. CHRISTIAN, and W. A. KITTRELL et al., 1969  Microchromosomes in holocephalian, chondrostean and holostean fishes. Chromosoma 26:35-40[Medline].

PAGEL, M. D. and P. H. HARVEY, 1988  Recent developments in the analysis of comparative data. Q. Rev. Biol. 63:413-440[Medline].

PITRA, C. and J. VEITS, 2000  Use of mitochondrial DNA sequences to test the Ceratomorpha (Perissodactyla:Mammalia) hypothesis. J. Zool. Syst. Evol. Res. 38:65-72.

POSTLETHWAIT, J. H., Y.-L. YAN, M. A. GATES, S. HORN, and A. AMORES et al., 1998  Vertebrate genome evolution and the zebrafish gene map. Nat. Genet. 18:345-349[Medline].

POSTLETHWAIT, J. H., I. G. WOODS, P. NGO-HAZELETT, Y.-L. YAN, and P. D. KELLY et al., 2000  Zebrafish comparative genomics and the origins of vertebrate chromosomes. Genet. Res. 10:1890-1902.

PYATSKOWIT, J., H. C. KINCAID, C. C. KRUEGER, and B. MAY, 2001  Inheritance of microsatellite loci in polyploid lake sturgeon (Acipenser fulvescens). Genome 44:185-191[Medline].

RICO, C., I. RICO, and G. HEWITT, 1996  470 million years of conservation of microsatellite loci among fish species. Proc. R. Soc. Lond. Ser. B 263:549-557[Medline].

ROZAS, J. and R. ROZAS, 1997  DnaSP version 2.0: a novel software package for molecular population genetics analysis. Comput. Appl. Biosci. 13:307-311.

SCHNEIDER, S., D. ROESSLI and L. EXCOFFIER, 2000 Arlequin v. 2000: a software for population genetics analysis. Genetics and Biometry Laboratory, University of Geneva, Geneva.

SPRING, J., 1997  Vertebrate evolution by interspecific hybridisation—are we polyploid? FEBS Lett. 100:2-8.

STONEKING, M., B. MAY, and J. E. WRIGHT, 1981  Loss of duplicated gene expression in salmonids: evidence for a null allele polymorphism at the duplicate aspartate amino transferase loci in brook trout (Salvelinus fontinalis). Biochem. Genet. 19:1063-1077[Medline].

STRIMMER, K. and A. VON HAESELER, 1996  Quartet puzzling: a quartet maximum likelihood method for reconstructing tree topologies. Mol. Biol. Evol. 13:964-969.

SWOFFORD, D. L., 1998 PAUP* (Phylogenetic Analyses Using Parsimony): Preliminary Test Version. Sinauer, Sunderland, MA.

TAGLIAVINI, J., P. WILLIOT, L. CONGIUS, M. CHICCA, and M. LANFREDI et al., 1999  Molecular cytogenetic analysis of the karyotype of the European Atlantic sturgeon, Acipenser sturio. Heredity 83:520-525.

VAN EENENNAAM, A. L., J. D. MURRAY, and J. F. MEDRANO, 1998a  Mitotic analysis of the North American white sturgeon, Acipenser transmontanus Richardson (Pisces, Acipenseridae), a fish with a very high chromosome number. Genome 41:266-271.

VAN EENENNAAM, A. L., J. D. MURRAY, and J. F. MEDRANO, 1998b  Synaptonemal complex analysis in spermatocytes of white sturgeon, Acipenser transmontanus Richardson (Pisces, Acipenseridae), a fish with a very high chromosome number. Genome 41:51-61.

WILLIOT, P., E. ROCHARD, G. CASTELNAUD, T. ROUAULT, T. BRUN et al., 1997 Biological characteristics of European Atlantic sturgeon, Acipenser sturio, as the basis for a restoration program in France, pp. 359–370 in Sturgeon Biodiversity and Conservation, edited by V. J. BIRSTEIN, J. R. WALDMAN and W. E. BEMIS. Kluwer Academic Publishing, Dordrecht, The Netherlands.

WOLF, C., P. HÜBNER, and J. LÜTHY, 1999  Differentiation of sturgeon species by PCR-RFLP. Food Res. Int. 32:699-705.

ZHANG, D.-X. and G. M. HEWITT, 1996  Nuclear integrations: challenges for mitochondrial DNA markers. TREE 11:247-251.

ZHANG, S.-M., 1998 Molecular population genetics of Chinese sturgeon with reference to cellular and molecular evolution of some species of Acipenseriformes. Doctoral Dissertation, Chinese Academy of Sciences, Wuhan, China.

This article has been cited by other articles:

				H. Rompler, C. Staubert, D. Thor, A. Schulz, M. Hofreiter, and T. Schoneberg G Protein-Coupled Time Travel: Evolutionary Aspects of GPCR Research Mol. Interv., February 1, 2007; 7(1): 17 - 25. [Abstract] [Full Text] [PDF]

J. Krieger, A. K. Hett, P. A. Fuerst, V. J. Birstein, and A. Ludwig Unusual Intraindividual Variation of the Nuclear 18S rRNA Gene is Widespread Within the Acipenseridae J. Hered., May 1, 2006; 97(3): 218 - 225. [Abstract] [Full Text] [PDF]