Enrichment of tetranucleotide microsatellite loci from invertebrate species.

ABSTRACT An experimental procedure using biotin labelled probes and streptavidin bound magnetic beads was developed to produce microsatellite-enriched libraries in two bivalve species, geoduck and Japanese scallop, and one crustacean, Dungeness crab. Microsatellite libraries were enriched tar GAT[A.sub.4] and GAC[A.sub.4] by 75 84%. The magnetic bead protocol produced genomic inserts ranging in size from 347-449 up. with an average 5' and 3' flanking of 89-130 bp. Eight polymorphic loci were isolated from both geoduck dam and Dungeness crab. The heterozygosities ranged from 0.57-0.92 for Dungeness crab and 0.91-0.97 for geoduck. These novel polymorphic microsatellite loci will be useful in studying recruitment dynamics and genetic structure.

KEY WORDS: microsatellites, enrichment, magnetic beads, geoduck, crab, scallop 

INTRODUCTION

Marine invertebrate species with pelagic larval stages should have tremendous capacity for dispersal and mixing over long distances. However, numerous studies have shown that over ecological time scales (10's of years), dispersal in many species is primarily local (Swearer et al. 2002). Even on evolutionary time scales (100's to millions of years), while the majority of marine invertebrate species with pelagic larvae do not contain a high degree of spatial structure at the 10-100+ km scale, there are some that do (e.g. Jiang et al. 1995; Burton 1998). Further, many studies have noted fine-scale heterogeneity among cohorts (termed chaotic genetic patchiness in Hellberg et al. 2002) that suggests (1) that few adults contribute to each recruitment event (Hedgecock 1994), or (2) that there is selection on genetic diversity in the larval stage (Johnson and Black 1984). Because effective management units must incorporate information on recruitment dynamics and genetic structure, it is imperative that these parameters are understood for species of economic importance and those of conservation concern.

For genetic analyses of population structure, numerous genetic markers have been utilised, including allozymes, mtDNA, minisatellites, microsatellites, and other coding and non-coding loci. Microsatellite loci, which are simple sequence repeats with 2 to 10 bp motifs organised in tandem arrays, are currently considered the marker of choice for determining the level of population connectedness over the span of 100's to 1000's of years. Microsatellite loci are often highly polymorphic due to a high rate of mutation through replication slippage, resulting in the gain or loss repeat units. In addition, microsatellite loci are as vulnerable to point mutations as the rest of the genome, which tends to divide longer repeat stretches into smaller units, and hence decrease the rate at which slippage occurs (Bell and Jurka 1997; Kruglyak et al. 1998). This "slippage/point-mutation" theory suggests that the frequency distribution of microsatellite lengths is a balance of expansion due to slippage and contraction due to point mutation (Sibly et al. 2003).

To date, all prokaryotic and eukaryotic genomes have been found to contain microsatellite loci. However, among eukaryote species, microsatellite repeats are more abundant and longer in vertebrates than invertebrates (Chambers and MacAvoy 2000). As a result, many researchers have experienced difficulty in isolating microsatellite loci from marine invertebrate species, especially tetranucleotide repeat loci which am generally preferred for population studies due to their lack of stutter. In a study comparing the prevalence of di-, tri-, tetra- and hexanucleotide repeats in shrimp, Xu et al. (1999) found that only 9% of microsatellite loci were in the form of tetranucleotide repeals. Similarly, using traditional library screening methods, we probed red sea urchin (Strongelocentrotus franciscanus) libraries with [sup.32]p-labelled (GACA) and (GATA) oligonucleotides, and less than 10% of the microsatellite sequences obtained were tetranucleotide repeats (Miller et al.). Moreover, most of the tetranucleotide containing loci were not sufficiently polymorphic for use in population studies or pedigree analyses.

Numerous methods advancing the techniques used in the isolation of microsatellites have been developed. The first microsatellite enrichment protocol was described by Ostrander et al. (1992) and later expanded by Paetkau (1999). Under their protocol, the fractionated DNA was packaged into a phagemid or phage vector and an ssDNA library was obtained. The ssDNA was used as a template for PCR using the repeated oligonucleotides as the primers, thus creating double stranded product enriched for repeats. Both methods were reviewed in Zane et al. (2002), but it was noted that only five primer papers to date have been accredited to these approaches.

Fischer and Bachmann (1998) and Hamilton et al. (1999) described magnetic bead methods of microsatellite enrichment. Both groups used a linker ligated to the restricted DNA fragments as a primer for PCR reactions. The adapter-linked DNA was then used as the target for 5' biotinylated repeat oligonucleotides and paramagnetic streptavidin beads. Both groups used genomic:oligonucleotide hybrids as the target for streptavidin magnetic beads. Fischer and Bachmann reported a greater than 60% enrichment of microsatellites in the onion plant Allium cepa, while Hamilton and colleagues reported a 20-95% enrichment rate. In their review, Zane et al. (2002) presented their own magnetic approach to microsatellite isolation termed FIASCO (Fast Isolation by AFLP of sequences containing repeals). In principle, the DNA was digested simultaneously with the AFLP adaptor to achieve a one-step digestion-ligation reaction. As above, the adapter linked DNA was used in a PCR step, hybridised with biotin-[(AC).sub.17] and separated with streptavidin beads resulting in a 50-95% enrichment for dinucleotide repeat microsatellites.

The present study outlines the experimental procedure we developed to produce microsatellite-enriched libraries for marine invertebrate species. The magnetic bead enrichment approaches outlined above and in O'Reilly et al. (2000) form the basis of the methods developed, but substantial improvements were made to enhance the length of the core repeat and flanking sequence and the level of enrichment obtained. This procedure was applied in the development of highly polymorphic microsatellite loci from two gastropod species, geoduck (Panopea abrupta) and Japanese scallop (Patinopectin yessoensis), and one crustacean species, Dungeness crab (Cancer magister). Primers to microsatellite loci for geoduck clams and Dungeness crab are described herein.

METHODS

Microsatellite enriched libraries were produced using magnetic bead hybridisation selection (Fig. 1). Genomic DNA was extracted from geoduck mantle, crab muscle, and scallop adductor muscle using a Stratagene DNA Extraction Kit (Stratagene, La Jolla, CA). Genomic DNA (approximately 50 [micro]g) was partially digested with 10.8 U HaeIII for 10, 20, and 30 min at 37[degrees]C. DNA fragments of 600-2000 bp were size selected and dephosphorylated by incubating 53 [micro]l clean cut DNA with 10 U of CIP for 2 hr at 37[degrees]C. The CIP treated DNA was cleaned using spin columns (QIAquick PCR purification kit, Qiagen, Valencia, CA) and eluted to a final volume of 60 [micro]l in elution buffer.

Figure 1. Flow chart of the magnetic bead enrichment protocol. Each
major step is noted within the shaded boxes. The total elapsed time for
the library experiment is approximately 7-9 days. The 96 well format
greatly reduces the time and labour, thus offsetting the potential
increase in consumable costs. In practice, one could produce successful
library enrichment from a new species within two weeks.

Preparation of the Genomic DNA and Hybridization Template

Extraction
Restriction Enzyme Digestion
(600-2000bp) Size Selection
CIP Treatment of DNA Fragments
SNX Ligation and SNX PCR

Estimated Time: 2-3 days

Hybridization of Template and Oligonucleotide Probes

Screening Probes: GAC[A.sub.4], GAT[A.sub.4], G[A.sub.8], G[T.sub.8]
Prepare Streptavidin Magnetic Beads and Hybe to Oligo-biotin probes
Denature SNX PCR + competitor (Template)
Hybridize Bead:Oligo and Template; Wash Steps; Elution
SNX PCR of Eluted SSRs; Size Selection
TOPO Cloning and Plating

Estimate Time: 1-2 days

Sequencing and Designing SSR Loci

Pick Colonies from each Plate of Enriched Probe
(i.e. higher #'s of colonies = more SSRs)
QIAquick 96 Miniprep
M13F and M13R Sequencing Reactions
Run Sequencing Gel (96 Lane format)
Identify Potential SSRs and Design Primers

Estimated Time: 4 days

The SNX linkers were designed according to Hamilton et al., 1999. Primers included SNX forward: 5'CTAAGGCCTTGCTAGCAGAAGC3', and SNX reverse: 5'pGCTTCTGCTAGCAAGGCCTTAGAAAA3'. The SNX linkers were added by PCR cycling in the presence of the SNX linkers, ligase and XmnI. The ligation mixture contained 10 [micro]l of CIP-treated DNA fragments, 3.9 mM mixed SNX reverse and forward, 1x Ligase Buffer, 1 U XmnI and 2 U Ligase in a final volume of 30 [micro]l. Ligations were cycled lot 30' at 16[degrees]C, 10' at 37[degrees]C for 5-18 cycles, followed by 20' at 65[degrees]C to inactivate the enzymes, and then kept at 4[degrees]C. Different ligation cycling regimes were tested, and the optimum number of ligation cycles was empirically determined to be five. More than five cycles resulted in shorter fragments with multiple SNX linkers. The SNX ligation product was cleaned using spin columns and eluted to a final volume of 20 [micro]l in EB.

The SNX PCR cocktail contained 0.8 mM SNX f, 0.8 mM dNTPs, 5 U Qiagen Taq, 2.5 mM MgCl, 10 [micro]l SNX ligated DNA fragments, and 1x Qiagen PCR Buffer. The cycling protocol was 2' at 92[degrees]C followed by 40 cycles of 94[degrees]C/45", 62[degrees]C/1 ', 72[degrees]C/1 ', a final extension of 30'/72[degrees]C, and then held at 4[degrees]C. At this point, the PCR product can be immediately processed to the hybridisation step, size selected and concentrated, or cleaned with spin columns and retained for long-term storage. The ligation step can be tested by resolution on an agarose gel. A smear of products indicates a successful ligation and PCR.

All hybridisation steps were performed at 48[degrees]C in a controlled environment. All trays, tips, wash solutions and MPC (Molecular Particle Concentrator) were preheated to 48[degrees]C. Dynabead M-280 Streptavidin magnetic beads (Dynal ASA, Oslo, Norway) were washed 3 times with phosphate buffered saline with 0.1% sodium dodecyl sulphate using the Dynal MPC-S or -P (single or plate format) between washes. The beads were re-suspended in 5x SSC (20x SSC stock: 175.3g sodium chloride and 88.2g sodium citrate per litre; adjust pH to 7.0 with 10N hydrochloric acid). Previously, Hamilton et al. (1999) used genomic oligonucleotide hybrids as the target for streptavidin magnetic beads and O'Reilly et al. (2000) combined [(GATA).sub.7] oligonucleotides bound to streptavidin coated paramagnetic beads with genomic DNA. We found that SNX-ligated DNA was efficiently enriched by shorter biotin-[(GACA).sub.4] and [(GACA).sub.4] oligonucleotides bound to streptavidin beads. Notably, 100 [micro]l of magnetic beads were incubated with 1 [micro]l (200 pmol) biotinylated oligo probes for 15 min at room temperature. The 5' biotinylated oligonucleotide primers were obtained from commercial sources (University of Calgary, AB, Canada). The bead:oligo complex was washed 3 times with 5x SSC using the MPC between washes, resuspended in 35 [micro]l of 10x SSC, and held at 48[degrees]C for the hybridization steps. The target DNA was denatured by incubating 10 [micro]l of SNX PCR product, 10 [micro]l SNXf (competitor) (0.5 ng/[micro]l) and 55 [micro]l water at 95[degrees]C/15' then plunged into ice. The bead:oligo complex was combined with the denatured target (SNX PCR product+competitor) and incubated at 48[degrees]C for 60 min.

Some groups have reported successful enrichments at room temperature (RT) using magnetic beads (see Zane et al. 2002 and Fischer and Bachmann 1998). However, we did not obtain enrichment of GAC[A.sub.4] or GAT[A.sub.4] microsatellites in geoduck or eulachon (data not shown) at RT. All wash steps were performed without removing the sample from the 48[degrees]C environment. Using the MPC between washes, the three wash steps were performed as follows: 2xSSC+SNXf. 1xSSC+SNXf, 0.5xSSC+SNXf, using 200 [micro]g SNXf per 1 ml of SSC. The presence of the SNX competitor in the hybridization and wash steps ensured a high level of stringency, thus reducing non-specific probe binding. The microsatelliteenriched fraction was eluted by adding 50 [micro]l of TE at 48[degrees]C to the bead:oligo:DNA incubated at 95[degrees]C for 15'. Using the MPC, the enriched fraction was removed without disturbing the streptavidin beads. The enriched fraction was used as the template in the SNX PCR, as noted above. By resolving the SNX PCR and oligo-PCR products on agarose gels, using each fraction (ligations, hybridisation, washes, elutions) as a template, the technical success was assessed (Table 1). Ideally, the smear of products amplified by the SNX primer declined with each wash step.

Following the SNX PCR with the enriched fraction as the template, the fragments were size selected (600-2000 bp), cloned and transformed using a TOPO Cloning 5' PCR kit (Invitrogen, Carlsbad, CA). The transformations were plated, grown overnight at 37[degrees]C and single colonies minipreped using a QIAprep 96 Turbo Miniprep Kits (Qiagen, Valencia, CA). Clones were sequenced using Big Dye Primer M13F and M13R Sequencing Kits (ABI, Foster City, CA). In general, the plates with the highest number of colonies correlated to the most abundant repeat motif. For the identification of unique or rare microsatellites, extensive probing from plates with fewer colonies was required.

Sequencing gels were electrophoresed on an ABI 377 automated sequencer using ABI Prism 377-96 Collection software (ABI). Sequences were viewed and base-called using Sequencing Analysis 3.4.1 (ABI). Sequencher software (Gene Codes Corporation, Ann Arbor, MI) was used to remove the vector, align and group sequences and highlight microsatellite repeats. PCR primers were designed for each potential microsatellite locus.

Over the past decade, we have isolated and applied microsatellite loci to population genetics studies of more than 20 marine fish and shellfish species. The substantial empirical data gained from these studies guides our ranking of microsatellite sequences for primer design in new species. Our criteria follows seven general rules fin order of sequence preference): Primers are designed to (1) All perfect tetranucleotide repeats of 10 to 35, (2) All perfect dinucleotide repeats of 15-45 (noting that those with over 25 repeals can contain substantial stutter), (3) Slightly imperfect tetranucleotide repeats of 13 or more, and (4) Compound di-di, tetra-tetra, and di-tetra repeats with little or no imperfections. Primers are generally not designed to (5) di/tri, tri/tetra or sequences containing stretches of more than five single nt, as these o/ten result in 1-bp alleles, (6) Highly imperfect loci, as these often contain single base insertions or deletions, also resulting in 1-bp alleles, and (7) Tetranucleotide repeats containing three mononucleotides (e.g., AAAN). In general, if sufficient flanking sequence is available, three sense and three anti-sense primers are designed for each locus, which enables the detection of null alleles in the first round of screening and provides a variety of size ranges to choose from for the design of multiplexes. Because our population surveys generally contain sample sizes of 100 or more individuals, loci with high numbers of alleles are still considered potentially useful.

Polymerase chain reaction (PCR) amplifications were performed using 35 cycles of 94[degrees]C/30", 48-54[degrees]C/30", and 70[degrees]C/45". Each 8.0 [micro]L reaction contained 0.50 [micro]L of a 1:2 dilution of Chelex extracted DNA (approximately 0.01-0.03 [micro]g), 0.48 [micro]M of each primer, 0.80 [micro]M dNTP, 0.15 units of HotStarTag[TM] DNA polymerase (Qiagen, Valencia, CA) and 1x HotStarTag[TM] PCR Buffer containing Tris-HCl, KCl [(N[H.sub.4]).sub.2]S[O.sub.4], 1.5 mM Mg[Cl.sub.2], pH 8.7. Primers were initially optimised by amplification of 8 individuals size-fractionated on 10% non-denaturing polyacrylamide gels. Manual gels were stained with ethidium bromide, and sized against a 20-bp ladder using Phoretics[TM] ID version 5.10 software (Nonlinear Dynamics Ltd., Newcastle upon Tyne, UK). Multiple primers per microsatellite sequence were tested to aid in the identification of null alleles and to provide a range of product sizes for subsequent multiplexing. Null alleles and allele size ranges were assessed by amplifying 24 individuals for 2 to 3 primersets per polymorphic locus, size separated on manual gels. One sense primer for each polymorphic locus was then fluorescently labelled, and loci were incorporated into a multiplex for automated electrophoresis on an ABI 377 automated sequencer using 4.5% denaturing polyacrylamide gels. Allele sizes were determined with Genescan 3.1 and Genotyper 2.5 software (PE Biosystems, Foster City, CA), and the Genetic Data Analysis (GDA) program of Lewis and Zaykin (2001) was used to analyse allelic and genotypic frequency data.

RESULTS & DISCUSSION

The experimental time-table is outlined in Figure 1. The experimental design can be modified for high throughout, by employing a 96-well plate format MPC. This approach replaces the physically demanding, lengthy and potentially hazardous radioisotope method. The magnetic bead approach produced a high level of enrichment for tetranucleotide repeats in all three invertebrate species tested (Table 1). The microsatellite enrichment averaged 75% over all species/microsatellites surveyed. The level of enrichment of each microsatellite motif varied among species. In Dungeness crab, both GAC[A.sub.4] and GATA4 were highly enriched, at 85% and 82%, respectively. In geoduck clam, GACA enrichment was more successful (94% versus 45%) whereas in scallop, GATA was more highly enriched (92% versus 67%). Scallop was also highly enriched for C[T.sub.8], CCA[T.sub.4] and GAC[T.sub.4]repeats.

Toth et al. (2000) demonstrated that di- and tetranucleotide motifs are the most common repeat in vertebrates. Epplen et al. (1998) compiled a table of representative microsatellite DNA sequences deposited in the EMBL/GENBANK data bank. They indicated that GAAA, AAAT, GATA and GGAA were the most common tetranucleotide repeats while GACA and GACT were rarer. Likewise, GT was the most common dinucleotide repeat followed by AT and GA. However. GATA and GACA were the most frequent tetranuclcotide repeat in the bird, reptile and fish sequences surveyed (Toth et al. 2000). In human chromosomes, Katti et al. (2001) found tetranucleotide repeats were very frequent, with the most common type being [(AAAN).sub.n]. Indeed, Schable et al. (2002) obtained success in enriching for microsatellites from the dollar sunfish using [(AAAG).sub.6] and [(ACAG).sub.6]. However, because AAAG repeats often contain imperfections such as AAG or AAAAG which result in 1-bp alleles, we did not enrich for [(AAAN).sub.n]. microsatellites.

Herin, PCR protocols were optimised for tetranucleotide repeat containing microsatellites, which are generally found on the longer fragments. We note three steps in the enrichment process that were crucial to the obtainment of long fragments and sufficient flanking sequence. First, decreasing the SNX-ligation cycling steps from 18 to 5 cycles enhanced the length of flanking sequence. Second, increasing the number of cycles in the SNX-PCR from 30 to 40 enhanced the population of high molecular weight PCR fragments. Third, size selection of the 600-2000 bp fragments immediately following the restriction digest reduced self ligation of small fragments and bias towards smaller products in the PCR. Fourth, a second round of size selection of the 600-2000 bp fragments after the SNX-PCR was imperative because of the small insert bias of TA cloning. Using these protocols, we obtained an average fragment length of 347 bp (scallop) to 450 bp (crab) and average 5' and 3' flanking sequences ranging in length from 89 bp (scallop) to 130 bp (crab) (Table 1).

To directly assess the effectiveness of the enrichment protocol described herein, we compared the resolution of tetranucleotide repeat loci in geoduck clams obtained under the enrichment protocol to traditional library screening. Using traditional techniques, we screened over 3 x [10.sup.4] clones using [sup.32]P-end labelled (GAC[A.sub.9]) and (GAT[A.sub.9]) oligonucleotides, as in Miller et al. (2001). Sixty-six putative microsatellite-containing colonies were sequenced, 30 of which were round to contain microsatellite repeats. Only a single clone contained a simple tetranucleotide repeat. The remainder of the clones contained dinucleotide repeats, although four clones contained compound di-/trinucleotide repeats. Hence, although libraries were probed exclusively with tetranucleotide repeats, most microsatellites obtained were dinucleotide repeat sequences. Alternately, using the enrichment protocol, we sequenced 55 clones, 41 of which contained microsatellite repeats. Of these, 22 clones contained simple tetranucleotide repeats, 8 clones contained simple dinucleotide repeats, and the remaining 11 clones contained compound di/tetranucleotide repeats. Hence, 80% of the microsatellite sequences isolated under the enrichment protocol contained tetranucleotide repeats versus 16% using the traditional method.

Primers to thirteen microsatellite sequences isolated through magnetic bead enrichment of geoduck clam were designed and tested. In addition, primers to ten microsatellite sequences isolated through conventional radioactive library screening (described above) were designed. In all, one tetranucleotide (Pab 117) and three di-nucleotide loci (Pub 101f, Pab 132 and Pub 156) isolated through conventional screening were chosen for use in population studies, and four tetranucleotide repeat loci (Pab 101e, Pab 105e, Pab 106e, and Pub 112e) from the enriched method were chosen (Table 2). Loci were chosen by of degree of pnlymorphism, clarity of alleles, and absence of (demonstrated) null alleles defined in the original small survey of 24 individuals.

Polymorphism of the eight geoduck loci was evaluated over approximately 200 individuals collected from two sites within British Columbia, Canada, one located in the Strait of Georgia and the other off the Queen Charlotte Islands. The geoduck microsatellite loci were highly polymorphic, with numbers of alleles ranging from 21 (Pub 156) to 60 (Pub 132) and expected heterozygosities ranging from 0.91 (Pab 112e) to 0.97 (Pab 156). However, significant deviations from Hardy Weinberg (HWE) equilibrium caused by heterozygote deficits were found at all but three of the loci (Pab 101e, Pub 106e and Pab 112e). Heterozygote deficiencies were also observed at most loci isolated from geoduck clams in a previous study (Vadopalas and Bentzen 2001), and have been commonly observed in a variety of marine invertebrate species (e.g. Brown 1991; Miller et al. 2001; Perez-Losado et al. 2002; Addison and Hart 2004). Possible explanations for heterozygote deficits in marine invertebrate species include the presence of null alleles, a temporal Wahlund effect gained through Sweepstakes style recruitment (Hedgecock 1994), inbreeding, nonrandom mating, and selection.

Primers to eighteen microsatellite sequences isolated from Dungeness crab were designed and tested. Nine of these loci were chosen for use in population studies. Two loci contained dinucleotide repeat units (Cma 107 and Cma 118), six loci contained tetranucleotide repeats (Cma 102, Cma 103, Cma 108a, Cma 108b, Cma 114 and Cma 117), and one locus contained a compound di-tetranucleotide repeat (Cma 1). Polymorphism of the eight loci was evaluated over approximately 200 individuals collected from two sites within British Columbia, Canada: Hecate Strait and Port McNeil. The microsatellite loci isolated from Dungeness crab contained a moderate level of polymorphism. The number of alleles identified at each of the loci ranged from 6 (Cma 117) to 39 (Cma 107) with an average of 14.4 alleles per locus. Expected heterozygosity ranged from 0.57 (Cma 117) to 0.92 (Cma 107). Deviations from HWE equilibrium was observed for only one of the loci (Cma 107).

In summary, the magnetic bead enrichment protocol described herein was successfully employed to isolate tetranucleotide repeat loci from marine invertebrate species that have been notoriously difficult to work with using traditional isolation methods. The enrichment protocol can be performed in a few as 7 days, where as, standard methods can take months. The high throughput format enabled us to screen, clone and sequence hundreds of microsatellites, thus facilitating rapid isolation of new loci for multiplex population analysis. To date, we have successfully utilised this protocol to isolate microsatellite loci from six marine invertebrate (herein) and fish species (Kaukinen et al, in press and unpublished data).

TABLE 1.
Species used to test the enrichment protocol include geoduck (Pab),
Dungeness crab (Cma), and Japanese scallop (Pye). All species were
screened with GAC[A.sub.4] and GAT[A.sub.4] probes, with scallop
probed for a variety of additional core repeats. The total number
of clones sequenced, contained microsatellite repeats, and had
primers designed are presented for each species. The % enrichment
was determined by dividing the total number of
microsatellite-containing clones by the total number of clones
sequenced. Likewise, the % success was calculated by dividing the
total number of useful loci by the total number of designed loci
(primers for Pye have not yet been tested). Average insert length
and average 3' and 5' flanking were calculated for each species.

          Probe     Number      [mu]sat     Designed
Species   Repeat   Sequenced   Containing     Loci     % Enrichment

Pub        GACA        33          31          10         93.94
           GATA        22          10           3         45.45
Total                  55          41          13         74.55

Cma        GACA        87          74          18         85.06
           GATA        67          55           0         82.09
Total                 154         129          18         83.77

Pye        GACA        48          32           4         66.67
           GATA        48          44          10         91.67
            CT         16          13           1         81.25
            CA         16           9           2         56.25
           CCAT        16          13           3         81.25
           GACT        16          13           0         81.25
           AAC         16          11           0         68.75
           GGT         16          12           1         75.00
Total                 192         147          21         76.56

                                            Avg 5' & 3'
          Probe                Avg Insert    Flanking
Species   Repeat   % Success   Size (bp)       (bp)

Pub        GACA
           GATA
Total                30.77       401.2         103.3

Cma        GACA
           GATA
Total                50.00       448.9         129.6

Pye        GACA
           GATA
            CT
            CA
           CCAT
           GACT
           AAC
           GGT
Total                 N/A         347          88.75

TABLE 2.
Nine polymorphic microsatellite loci were developed from Dungeness
crab and eight for geoduck clam. Repeats denoted with (i) contain
imperfect repeated motifs. Locus-specific annealing temperatures
are shown under [T.sub.a] ([degrees]C). Levels of polymorphism
measured by number of alleles and expected ([H.sub.E]) and observed
([H.sub.O]) heterozygosity, and inbreeding coefficients ([F.sub.is])
were calculated from a survey of approximately 200 individuals
collected at two sites for each species. The asterix (*) under
[H.sub.O] denotes loci that contain a deficit of heterozygotes are
and are significantly out of Hardy-Weinberg Equilibrium. Accession
numbers for each locus are shown in the end column. The first four
geoduck clam loci were obtained through traditional library screening
methods in which [sup.32]P-labelled GACA and GATA probes were utilised.

                                     Primer Sequence
 Locus          Repeat                   (5'-3')

Dungeness Crab
Cma 102    [(GACA).sub.12]    F:TTCAGCTGCACTTCAGTGAT
                              R:CTGTAGTGAACTAAATTACTGTT
Cma 103    [(GACA).sub.13]    F:GTTCCAAATACAGTTGACC
                              R:GTCTTCCTATGTCCTCCTT
Cma 107    [(GT).sub.40i]     F:GCGTTCAAGGATATTACTGAGT
                              R:GTTTCCCCTGACTCATCCCCTC
Cma 108a   [(GACA).sub.13]    F:GCAGTAGGAACAGCAGCTGAT
                              R:GTTTATTTCGTCACCAGAGAGA
Cma 108b   [(GACA).sub.13]    F:CAGGTGTGGTTGTGTCCCTTTA
                              R:GTTCAGTTGAACCCAGAGTGACA
Cma 114    [(GACA).sub.12]    F:CAAGTAAGAGAATGGAATCGTATT
                              R:GTTTGCCAAAGAGCATCAGTGACAA
Cma 117    [(GACA).sub.9]     F:GTCTGAGACGAGCCAACATC
                              R:GTTTCAACAGGAAACATGAAATAGGAT
Cma 118    [(GT).sub.28]      F:GGAGAGGGAGCGACTGTC
                              R:GTTTGGTGTATTACAAAACAACCAGTAA

Geoduck Clam
Pab 101    [(GA).sub.16]      F:TGTTGAGATATAACCACTT
                              R:GTTTGTCTATGGTTTGCATTGTA
Pab 117    [(GACA).sub.29]    F:TGTTGAGATATAACCACTT
                              R:GTTTCGACCCAACAATAGTTGA
Pab 132    [(AG).sub.72i]     F:TCGCTCACTAACTCACTT
                              R:GTTTGCTATTGATAATTCTGAGA
Pab 156    [(GT).sup.21]      F:GAGTGACATAATGAGATACT
           [(CA).sub.19]      R:TTTCATCTCGTTACATATCAATATT
Pab 101e   [(GACA).sub.20]    F:GTACCTGATGGTGTTAATAGTA
                              R:TTGATCATTATATTTTGTCATAGAC
Pab 105e   [(GACA).sub.18]    F:CAACCATGGTGTCTCAAAGA
           [(GACA).sub.5]     R:GGCAGATGGGTCTATAGTTT
Pab 106e   [(GACA).sub.24i]   F:GGCAGTCAGACAGACCAG
                              R:ATGATCTCTCTATATCTGCTTCAAC
Pab 112e   (GCAC).sub.23]     F:GCGCTTAGAATACTGCGGAAT
                              R:GTTTACCATTACCATTGTCACGGTA

                                [T.sub.a      Size     No. of
 Locus          Repeat        ([degrees]C)    Range    Alleles

Dungeness Crab
Cma 102    [(GACA).sub.12]         50        136-175     10
Cma 103    [(GACA).sub.13]         48        205-226      9
Cma 107    [(GT).sub.40i]          50        145-220     39
Cma 108a   [(GACA).sub.13]         54        152 206     16
Cma 108b   [(GACA).sub.13]         54        116-137      8
Cma 114    [(GACA).sub.12]         52        233-257      7
Cma 117    [(GACA).sub.9]          54        286-314      6
Cma 118    [(GT).sub.28]           52        167-203     19

Geoduck Clam
Pab 101    [(GA).sub.16]           56         80-215     27
Pab 117    [(GACA).sub.29]         50        220-440     44
Pab 132    [(AG).sub.72i]          48        155 305     60
Pab 156    [(GT).sup.21]           50        100-250     21
           [(CA).sub.19]
Pab 101e   [(GACA).sub.20]         52        119-282     23
Pab 105e   [(GACA).sub.18]         50        128-214     21
           [(GACA).sub.5]
Pab 106e   [(GACA).sub.24i]        52        104 301     45
Pab 112e   (GCAC).sub.23]          50        109-380     59

 Locus          Repeat        [H.sub.E]   [H.sub.O]   [F.sub.is]

Dungeness Crab
Cma 102    [(GACA).sub.12]      0.75        0.77        -0.03
Cma 103    [(GACA).sub.13]      0.71        0.70         0.01
Cma 107    [(GT).sub.40i]       0.92        0.69 *       0.25
Cma 108a   [(GACA).sub.13]      0.70        0.70        -0.00
Cma 108b   [(GACA).sub.13]      0.71        0.72        -0.02
Cma 114    [(GACA).sub.12]      0.61        0.55         0.09
Cma 117    [(GACA).sub.9]       0.57        0.59        -0.05
Cma 118    [(GT).sub.28]        0.85        0.88        -0.03

Geoduck Clam
Pab 101    [(GA).sub.16]        0.94        0.52 *       0.45
Pab 117    [(GACA).sub.29]      0.97        0.69 *       0.28
Pab 132    [(AG).sub.72i]       0.96        0.77 *       0.19
Pab 156    [(GT).sup.21]        0.91        0.62 *       0.32
           [(CA).sub.19]
Pab 101e   [(GACA).sub.20]      0.93        0.86         0.08
Pab 105e   [(GACA).sub.18]      0.93        0.72 *       0.23
           [(GACA).sub.5]
Pab 106e   [(GACA).sub.24i]     0.96        0.93         0.03
Pab 112e   (GCAC).sub.23]       0.97        0.87         0.10

 Locus          Repeat        Accession no.

Dungeness Crab
Cma 102    [(GACA).sub.12]      AY521552
Cma 103    [(GACA).sub.13]      AY521553
Cma 107    [(GT).sub.40i]       AY521554
Cma 108a   [(GACA).sub.13]      AY521555
Cma 108b   [(GACA).sub.13]      AY521556
Cma 114    [(GACA).sub.12]      AY521557
Cma 117    [(GACA).sub.9]       AY521558
Cma 118    [(GT).sub.28]        AY521559

Geoduck Clam
Pab 101    [(GA).sub.16]        AY520562
Pab 117    [(GACA).sub.29]      AY520563
Pab 132    [(AG).sub.72i]       AY520564
Pab 156    [(GT).sup.21]        AY520565
           [(CA).sub.19]
Pab 101e   [(GACA).sub.20]      AY520566
Pab 105e   [(GACA).sub.18]      AY520567
           [(GACA).sub.5]
Pab 106e   [(GACA).sub.24i]     AY520568
Pab 112e   (GCAC).sub.23]       AY520569

ACKNOWLEDGMENTS

The authors acknowledge Tobi Ming, Karen Laberee, Shoarong Li and Brent Vadopalas for technical assistance. We thank Island Scallops for the Japanese Sea Scallops, and Johnstone Strait Dungeness Crab Fishery. Funding was provided by Fisheries & Oceans Canada and the Government of Canada, Canadian Biotechnology Strategy.

LITERATURE CITED

Addison, J. A. & M. W. Hart. 2003. Characterization of microsatellite loci in sea urchins (strongylocentrotus spp.). Mol. Ecol. Notes. 2:493-494.

Bell, G. I. & J. Jurka. 1997. The length distribution of perfect dimer repetitive DNA is consistent with its evolution by an unbiased single-step mutation process. J. Mol. Evol. 44:414-421.

Brown, L.D. 1991. Genetic Variation and population structure in the blacklip abalone, Haliotis rubra. Aust. J. Mar. Freshwat. Res. 42:27-90.

Burton, R. S. 1998. Intraspecific phylogeography across the Point Conception biogeographic boundary. Evol. 52:734-745.

Chambers, G. K. & E. S. MacAvoy. 2000. Microsatellies: consensus and controversy. Comp. Biochem. Physiol. B126:455-476.

Epplen, J. T., W. Maueler & E. J. M. Santos. 1998. On GATAGATA and other "junk" in the barren stretch of genomic desert. Cytogenet. Cell Genet. 80:75-82.

Fisher, D. & K. Bachmann. 1998. Microsatellite enrichment in organisms with large genomes (Allium. cepaL.). Biotechniques 24:796-802.

Jiang, L., W. L. Wu & P. C. Huang. 1995. The mitochondrial DNA of Taiwan abalone Haliotis diversicolor Reeve, 1846 (Gastropoda: Archaeogastropoda: Haliotidae). Molec. Mar. Biol. Biotech. 4:353-354.

Johnson, M. S. & R. Black. 1984. Pattern beneath the chaos: the effect of recruitment on genetic patchiness in an intertidal limpet. Evol. 38: 1371-1383.

Hamilton, M. B., E. L. Pincus, A. DiFiore & R. C. Fleischer. 1999. Universal linker and ligation procedures for construction of genomic DNA libraries enriched for microsatellites. BioTechniques 27:500-504.

Hedgecock, D. 1994. Does variance in reproductive success limit effective population sizes of marine organisms? In: Beaumont Ar, Editor. Genetics and evolution of aquatic organisms. London: Chapman and Hall. pp. 122-134.

Hellberg, M. E., R. S. Burton, J. E. Neigel & S. R. Palumbi. 2002. Genetics assessment of connectivity among marind populations. Bull. Marine Sci. 70:273-290.

Katti, M. V., P. K. Ranjekar & V. S. Gupta. 2001. Differential distribution of simple sequence repeats in eukaryotic genuine sequences. Mol. Biol. Evol. 181:1161-1167.

Kaukinen, K. H., K. J. Supernault & K. M. Miller, 2004. Development of mircrosatellite loci in Eulachon fish (Thaleichthys pacificus). Mol. Ecol. Notes, In press.

Kruglyak, S. R. T. Durrett. M. D. Schug & C. F. Aquadro. 1998. Equilibrium distributions of microsatellite repeat length resulting from a balance between slippage events and point mutations. Proc. Nut. Acad. Sci. 95:10774-10778.

Lewis, P. O. & D. Zaykin. 2001. Genetic data analysis: Computer program for the analysis of allelic data. Version 1.0 (d16c). Free program distributed by the authors over the internet from http://lewis.eeb.uconn. edu/leishome/software.html.

Miller, K. M., K. Laberee. K. H. Kaukinen, S. Li & R. E. Withler. 2001. Development of microsatellite loci in pinto abalone (haliotis kamtschatkana) Mol. Ecol. Notes 1:315-317.

Miller, K. M., K. H. Kaukinen, K. Laberee & K. J. Supernaut. 2004. Microsatellite from Red Sea Urchins (Strongylocentrotus franciscan). Molecular Ecology Notes (In Press).

O'Reilly, P. T., M. F. Canino, K. M. Bailey & P. Bentzen. 2000. Isolation of twenty low stutter di- and tetranucleotide microsatellites for population analyses of walleye pollock and other gadoids. J. Fish Biol. 56: 1074-1086.

Ostrander, E. A., P. M. Jong, J. Rine & G. Duyk. 1992. Construction of small insert genomic DNA libraries highly enriched for microsatellite repeat sequences, Proc. Natl. Acad. Sci. 89:3419-3423.

Paetkau, D. 1999. Microsatellites obtained using strand extension: An enrichment protocol. BioTechniques 26:690-694.

Perez-Losada, M. A. Guerra, G. R. Carvalho, A. Sanjuan & P. W. Shaw. 2002. Extensive population subdivision of the cuttlefish, Sepia officinalis (Mollusca: Cephalopoda) around the Iberian Peninsula indicated by microsatellite DNA variation. Hered. 89:417-424.

Schable, N. A., R. U. Fischer & T. C. Glenn. 2002. Tetranucleotide microsatellite DNA loci from the dollar sunfish (Lepomis Marginatus). Mol. Ecol. Notes 2:509-511.

Sibly, R. M., A. Meade, N. Boxall, M. J. Wilkinson, D. W. Come & J. C. Whittaker. 2003. The structure of interrupted human AC microsatellites. Mol. Biol. Evol. 20:453-59.

Swearer, S. E., J. S. Shima, M. E. Hellberg, S. R. Thorrold, G. P. Jones. D. R. Robertson, S. G. Morgan, K. A. Selkie, G. M. Ruiz & R. R. Warner. 2002. Evidence of self-recruitment in demersal marine populations. Bull. of Mar. Sci. 70:251-271.

Toth, G., Z. Gaspari & J. Jurka. 2000. Microsatellites in different eukaryotic genomes: survey and analysis. Gen. Res. 10:967-981.

Vadopalas, B. & P. Bentzen. 2001. Isolation and characterization of di- and tetranucleotide microsatellite loci in geoduck clams. Panopea abrupta. Mol. Ecol. 9:1433-1434.

Xu, Z., A. K. Dhar, I. Wyrzykowski & A. Alcivar-Warren. 1999. Identification of abundant and informative microsatellites from shrimp (Penaeus monodon) genuine. Anita. Genert. 30:150-156.

Zane, L., L. Bargelloni & T. Patarnello. 2002. Strategies for microsatellite isolation: a review. Mol. Ecol. 11:1-16.

K. H. KAUKINEN, * K. J. SUPERNAULT AND K. M. MILLER

Department of Fisheries and Oceans Canada, Science Branch, Pacific Biological Station, Nanaimo, B.C., Canada V9R 5K6

* Corresponding author. E-mail: [e-mail]