Transpecific microsatellites for hard pines

Theor Appl Genet (2002) 104:819–827 DOI 10.1007/s00122-001-0794-z M. Shepherd · M. Cross · T. L. Maguire M.J. Dieters · C.G. Williams · R.J. Henry Transpecific microsatellites for hard pines Received: 8 January 2001 / Accepted: 18 May 2001 / Published online: 6 February 2002 © Springer-Verlag 2002 Abstract Microsatellites are difficult to recover from large plant genomes so cross-specific utilisation is an important source of markers. Fifty microsatellites were tested for cross-specific amplification and polymorphism to two New World hard pine species, slash pine (Pinus elliottii var. elliottii) and Caribbean pine (P. caribaea var. hondurensis). Twenty-nine (58%) markers amplified in both hard pine species, and 23 of these 29 were polymorphic. Soft pine (subgenus Strobus) microsatellite markers did amplify, but none were polymorphic. Pinus elliottii var. elliottii and P. caribaea var. hondurensis showed mutational changes in the flanking regions and the repeat motif that were informative for Pinus spp. phylogenetic relationships. Most allele length variation could be attributed to variability in repeat unit number. There was no evidence for ascertainment bias. Keywords Conifers · Gymnosperms · Trans-specific polymorphism Communicated by P. Langridge M. Shepherd (✉) · M. Cross · R.J. Henry Cooperative Research Centre for Sustainable Production Forestry, Centre for Plant Conservation Genetics, Southern Cross University, P.O. Box 157 Lismore, NSW 2480, Australia e-mail: mshepher@scu.edu.au Fax: +61-2-66222080 T.L. Maguire Department of Botany, The University of Queensland, Brisbane, Queensland 4072, Australia M.J. Dieters Queensland Forestry Research Institute, M.S. 483 Fraser Rd, Gympie, Queensland 4570, Australia C.G. Williams Graduate Genetics Program, TAMU 2135, Texas A & M University, College Station, Texas, 77843-2125, USA Introduction Transferring microsatellites from related conifer species is appealing because it circumvents de novo microsatellite development. Developing pine microsatellites has proven difficult because the size (approx. 28,000 pg/C) and complexity (approx. 75–86% highly repetitive DNA) of the pine genome represent significant barriers (Echt et al. 1999; Smith and Devey 1994; Soranzo et al. 1998). Secondary screening of enriched libraries (Pfeiffer et al. 1997; Scott et al. 1999), eliminating repetitive regions of the genome (Smith and Devey 1994) or enriching for low-copy genomic sequences (Elsik et al. 2000) have also increased the efficacy of pine microsatellite development. The paucity of pine genomic sequences in the public sequence databases limits the value of this approach to finding microsatellites. Similarly, a survey of several thousand clones from a pine expressed sequence tag (EST) library yielded only eight polymorphic microsatellite markers (Echt and Burns 1999). Microsatellite transfer in pines is uncertain for several reasons. First, microsatellite transfer in flowering plants tends to be restricted to closely-related genera (Peakall et al. 1998). Similarly, transfer to other subgenera or nonPinus coniferae has been reported to be poor (Echt et al. 1996; Fisher et al. 1998), although optimisation of polymerase chain reaction (PCR) conditions increased transfer rates for some of these loci (Karhu et al. 2000). Kutil and Williams (2001) reported that tri-nucleotide microsatellites from low-copy genomic regions increased transfer rates within Pinaceae. Unlike previous investigators, they noted that both flanking regions and repeat motifs were highly conserved in soft and hard pines. Consequently, the transfer of pine microsatellites should be feasible in some cases. Identifying polymorphic microsatellites for Pinus caribaea Morelet and P. elliottii Little and Dorman (section Pinus; subsection Australes) has application worldwide. These species are indigenous to southern USA, Central America or the Caribbean and are cultivated throughout the world for timber and fibre (Lamb 1973; 820 Table 1 Results of transfer and PCR conditions for 50 Pinus spp. microsatellite loci Locus Source taxona Repeat motif Expected sizeb MgCl2 c (mM) Bufferd Transfer classe Referencef PtTX 2008 PtTX 2037 PtTX 2034 PtTX 2123 PtTX 2142 PtTX 2146 PtTX 3011 PtTX 3013 PtTX 3018 PtTX 3019 PtTX 3020 PtTX 3025 PtTX 3029 RPTest 01 RPTest 05 RPTest 08 RPTest 09 RPTest 11 RPTest 20 NZPR1 NZPR2 NZPR5 NZPR7 PtTX 2128 PtTX 3001 PtTX 3002 RPS105 RPS150 RPS61 PtTX 2164 PtTX 3034 PtTX 3021 PtTX 3023 PtTX 3027 PtTX 3035 PtTX 3037 RPTest 13 RPTest 15 NZPR3 NZPR6 PR-4.6 RPS3 RPS160 PtTX 3014 PtTX 3017 PtTX 3026 PtTX 3030 PtTX 3032 NZPR4 PR-9.3 PT PT PT PT PT PT PT PT PT PT PT PT PT PT PT PT PT PT PT PR PR PR PR PT PT PT PS PS PS PT PT PT PT PT PT PT PT PT PR PR PR PS PS PT PT PT PT PT PR PR (GAT)5 (GTGA)8 (GT)14 (TTTG)9 (AGC)8 (CTG)5 (GCT)21 (GAA)8 (GAT)18 (GAG)3 (GTT)10 (GAT)13 (GAA)11 (CAA)9 (CAA)10 (GCT)5 …(GCT)8 …(GCT)5 (ATA)7 (CAA)6 (CAA)5 (AAC)4 (GCG)6 (A)5 (TG)4 (GAG)5 (CAG)11 (GCA)7 (GCA)5 (CAT)7 (CAG)5 (AG)17 (AG)23 (AG)29 (AT)22 (AG)27 (GAC)8 (CAA)3 …(CAA)3 CAG(CAA)4 (CAA)6 …(GAG)4 AA(GAG)4 AC (GAG)4 AC (TCG)19 (TCA)16 (GT)10 (GA)13 (GTTTTT)4.(GTT)5.(GTT)5.(GTT)4 (CAA)4.(CAA)4 (CAT)10 (AGG)7.(AGG)2 CGG(AGG)3.(AGG)7.(AGG)3 (GA)9 (CAA)15 (CTG)5 (GTG)6 (TGG)4 (TGG)4 (T)5 (AG)14 (AGGG)3 (AG)25 (CA)21 (TA)6 N11 (TAA)9 (AC)19 (ACAG)3 AGGC(AGAC)3 (GTT)11 (GAT)9 (ACC)8 (ATC)22 (TA)8 (GGT)10 (GAT)42 (GAC)17 (AG)20 (CA)14 307 177 217 202 262 180 186 134 155 223 211 266 255 125 197 196 289 213 259 139 162 114 149 245 313 194 151 248 195 252 207 471 168 280 325 144 277 246 148 198 206–222 287 246 148 212 344 287 335 146 92–108 4.0 4.0 5.0 4.0 2.0 5.0 4.0 4.0 4.0 4.0 4.0 4.0 4.0 3.5 3.5 3.5 3.5 3.5 3.5 1.5 1.5 1.5 1.5 5.0 4.0 4.0 1.5 1.5 1.5 5.0 4.0 4.0 4.0 4.0 4.0 4.0 3.5 3.5 1.5 1.5 1.5 1.5 1.5 4.0 4.0 4.0 4.0 4.0 1.5 1.5 1 1 1 1 1 1 1 2 1 1 1 1 1 1 1 1 1 1 1 2* 2* 2 2* 1 1 1 2 2 2 1 1 1 1 1 1 1 1 1 2 2 2 2 2 1 1 1 1 1 2 2 1 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1* 1 1 1 1 1* 1* 1* 1 1* 1* 2* 2 2 2* 2* 2* 3 3 3 3 3 3 3 3 3 3 3 3 3 3 4 4 4 4 4 4 4 3 3 1 3 3 3 3 3 1 3 3 3 1 4 4 4 4 4 4 2 2 2 2 3 1 8 5 6 5 3 3 1 1 1 1 3 4 4 2 2 7 5 6 1 3 8 3 3 2 7 a b Taxa are: PT P. taeda ·PR P. radiata ·PS P. strobus Published expected PCR product size based on the sequence from the clone from the individual used to develop the library c Final MgCl concentration in PCR buffer 2 d See Methds and materials for buffer compositions. An asterisk (*) indicates AmpliTaq Gold DNA Polymerase (Perkin Elmer) was used in the PCR e See Methods and materials. An asterisk (*) indicates primer-pair was tested on a population of ten megagametophytes from one individual each of P. elliottii var. elliotti and P. caribaea var. hondurensis f References: 1 This paper, 2 Fisher et al. 1998, 3 Elsik et al. 2000, 4 Echt and Burns 1999, 5 Echt and May-Marquardt (unpublished) and Echt et al. 1999, 6 Echt et al. 1996, 7 Smith and Devey 1994, 8 Kutil and Willims (in press) Nelson et al. 1993). The option of creating de novo microsatellites for every pine species is daunting as pines constitute the largest genus in the Pinaceae family, with 100 species in the northern hemisphere (Little and Critchfield 1969). Transfer may be inefficient because the Pinaceae family is ancient, extending back to the late Triassic (more than 190 million years ago) (Miller 1977). Similarly, the phylogenetic distance between hard and soft pines is great; the fossil record indicates that the subgenera Pinus (hard pines) and Strobus (soft pines) 821 were formed by the late Cretaceous (approx. 130 million years ago) (Miller 1977). The study reported here tested microsatellite transfer to P. elliottii var. elliottii and P. caribaea var. hondurensis from two other hard pines and a soft pine. Criteria for successful transfer included clear amplification product and polymorphism. Focal species were defined as the source of de novo microsatellite loci and nonfocal taxa referred to the target taxa for transfer, P. elliottii var. elliottii and P. caribaea var. hondurensis. A nonfocal population was defined as the P. elliottii var. elliottii and P. caribaea var. hondurensis combined. Table 2 Primer-pairs for microsatellite loci obtained from P. taedu low-copy libraries Marker Primer sequences (5′ -3′) PtTX2034a TCTGAGGAGGAACATGTCATTTACT - F GCATGTCTGAATTATTGTGTTCTAT - R ATAAAGGCAGAGGATGAACA - F CCCAATTGTTATTTCTGATT - R TTGTTGTGCTCATAATTACTAGTGT - F CTCCTAAGCTTGCTCATGTG - R CCATTGACGCTCCCGTTACGTTACT - F CGAGAGACGTGCGGATACAAGACCT - R CCATTTATGAACCAGAGA - F ATTAAAACCATGAGACCTT - R TTCATCCTAGCTGCTTGCTTT - F CTCAGCGTCTACCCCATCAA - R CATCTAGTTACCAAAGTTAT - F ATTTATGAAAATGGTAAGT - R TCCATTTGAGAACTTTTT - F AGGAGCCACAACATAATA - R CTTGTTGCTGCTTCTGC - F AACAAAATAATATAAATGCTCTGC - R AGGAGGAGGAGTTGGAGTT - F ATCGCCCTAGCTGGTTTAT - R PtTX3001 PtTX3002 PtTX3014 PtTX3018 PtTX3021 Methods and materials Taxonomic classification and sources of DNA samples The study encompassed material from four hard pine (subgenus Pinus) taxa, P. radiata, P. taeda, P. caribaea var. hondurensis and P. elliottii var. elliottii, and a soft pine (subgenus Strobus), P. strobus. Pinus radiata belongs to the subsection Oocarpae (section Pinus), whereas the other three hard pines belong to the subsection Australes (section Pinus) (Little and Critchfield 1969). Pinus strobus is a member of the subsection Strobi (section Strobus). Four P. elliottii var. elliottii and five P. caribaea var. hondurensis individuals were randomly selected from the Queensland Forest Research Institute breeding populations as the nonfocal samples. Foliage material or DNA from a single individual from each of the focal species was obtained as controls. DNA was extracted from foliage tissue according to Graham et al. (1996). Additionally, ten seeds from an open-pollination of a P. elliottii var. elliottii tree (2PEE1-102) and a P. caribaea var. hondurensis tree (1PCH1063) provided megagametophyte tissue. Megagametophyte tissue was freeze-dried and ground prior to extraction using a DNeasy Kit (QIAGEN Valencia, Calif). Microsatellite transfer and optimisation Fifty microsatellite markers (Table 1) were examined for transfer: 28 markers from the Pinus taeda (PtTX) series were developed from total genomic or low-copy libraries (Elsik et al. 2000; Kutil and Williams 2001; Table 2); eight P. taeda (RPTest) markers were derived from ESTs (Echt and Burns 1999); two P. radiata (PR) markers were derived from a total genomic library of P. radiata (Smith and Devey 1994); seven P. radiata (NZPR) markers were derived from total genomic libraries (Fisher et al. 1998); five P. strobus (RPS) markers were derived from total genomic libraries (Echt et al. 1996). Microsatellite transfer was a stepwise process. First, microsatellite loci were evaluated for amplification in the nonfocal population and the focal taxa control using published conditions. PCR amplification was assessed on 3.5% agarose gels. If positive, then amplification products were tested for polymorphism using an ABI 310 Genetic Analyser (Perkin Elmer, Foster City, Calif.) and detected by fluorescence from dye-labelled dUTPs (R110, Perkin Elmer) or from a labelled primer (FAM, HEX or TET). In cases where primer-pairs gave low yield or complex banding patterns, PCR optimisation was attempted by testing a range of MgCl2 concentrations on the control focal species sample. Hot start PCR, using an antibody-inactivated Taq (AmpliTaq Gold, Perkin Elmer), was tested on nine recalcitrant microsatellites (Table 1). Amplifications were carried out using GeneAmp 9700 or a 9600 thermocycler (Perkin Elmer), using 25-u1 or 12.5u1 reactions with 20 ng of DNA template per reaction and final concentrations of 100 uM each dNTP and 200 nM each primer. One of two PCR buffers was used. Buffer 1 consisted of 10 × PCR reaction buffer, 100 mM Tris-HC1, l5 mM MgCl2, 500 mM KC1, PtTX3023 PtTX3027 PtTX3029 PtTX3035 a See Elsik et al. (2000) for methods pH 8.3 (Roche, Indianapolis, Ind buffer 2 of 10×stock – 500 mM Tris-HCl, 200 mM ammonium sulfate, 120% sucrose, l mg /ml gelatine. The final concentration of magnesium chloride varied depending on the primer-pair (Table 2). Complex banding patterns for a microsatellite locus were viewed as either multiple alleles or multiple loci. These markers were tested on ten haploid megagametophytes from a P. elliottii var. elliottii individual (2PEE1-102) and a P. caribaea var. hondurensis individual (1PCH1-063) and a population of 93 controlledpollinated hybrid offspring. Allele segregation patterns also distinguished homozygotes from heterozygotes with null alleles. Successful microsatellite transfer to P. elliottii var. elliottii and P. caribaea var. hondurensis was based on a clear amplification product of the expected size and polymorphism. The four classes for microsatellite marker transfer criteria were as fo1lows: (1) polymorphic, (2) monomorphic, (3) poor amplification resulting in low product yield or non-specific amplification or (4) no amplification. Microsatellite sequencing Six P. taeda (PtTX) microsatellite loci were sequenced in P. elliottii var. elliottii and P. caribaea var. hondurensis to compare repeat structure and sequence using haploid megagametophyte DNA. The sequencing of a PCR-amplified product was carried out using Big Dye Sequencing Kit (Perkin Elmer). Gel separation of the sequencing reactions was carried out at the Australian Genome Research Facility, Brisbane. Sequences were aligned using CLUSTAL-W multiple (accurate) (Thompson et al. 1994) using a gap opening penalty of 10 and manual alignment adjustments. Statistical analysis Heterozygosity (H) was calculated according to Crow (1986) using the following formula. (1) where fi is the frequency of the ith allele and n is the allele number. 822 Fig. 1 Transfer of polymorphic Pinus spp. microsatellite loci categorised by source. Microsatellite sources are identified by the focal taxon and library type. Library types: G genomic, LC low-copy, EST expressed sequence tag Table 3 Transfer of Pinus spp. microsatellite loci to nonfocal population species as categorised by the focal taxa and library type Locus source (taxon and library type) Number of loci tested Number of loci amplified Number of loci polymorphic P. strobus – Ga P. radiata – G P. taeda – G P. taeda – LC P. taeda – EST Totals 5 9 5 23 8 50 3 4 4 12 6 29 0 4 3 10 6 23 a b Library types: G Total genomic library including enriched · LC low-copy library · EST EST library Number of amplified loci refers to number of loci which were polymorphic or monomorphic Linear regression and t-tests were calculated using a spreadsheet. One-way analysis of variance (ANOVA), least significant difference test (LSD) and Fisher’s Exact tests were calculated using Statistica v 4 (Statsoft, Tulsa, Okla.) Results Microsatellite marker transfer to P. elliottii var. elliottii and P. caribaea var. hondurensis was successful for 23 of the 50 loci (Table 1). Twenty-nine microsatellite loci amplified in the two nonfocal taxa, and 23 of these 29 were polymorphic (Tables 2 and 3). If a microsatellite was polymorphic within a hard pine focal species, then it was also polymorphic within the nonfocal taxa. The two exceptions were PtTX 3011 for which P. caribaea var. hondurensis was monomorphic and NZPR2 where P. elliottii var. elliottii was monomorphic. Six of the 50 microsatellite loci were monomorphic in both focal and nonfocal species, and 21 microsatellite loci had either poor or no amplification. Five sources of microsatellites markers were recognised based on the focal taxa and the library type from which they were derived. The three library types were EST, total genomic or low-copy. The P. strubus and P radiata markers were coded as PS – G and PR – G, re- spectively. The P. taeda markers were coded by library type: EST (PT – EST), total genomic (PT – G) or lowcopy (PT – LC). With respect to transfer by amplification, the PT – G and PT – EST classes had high transfer rates, 80% and 75%, respectively. The PS – G and PT – LC classes demonstrated moderate rates of transfer (60% and 52%, respectively), whereas PR – G had the lowest transfer (44%). When transfer was assessed, by polymorphism, PT – EST had the highest proportion (75%) of polymorphic markers, followed by PT – G (60%), PR – G (44%) and PT – LC (43%) were lower and similar (Fig. 1). A Fisher’s exact test was used to compare the transfer rate of polymorphic markers at a taxon level. As many markers from P. radiata transferred and were polymorphic as P. taeda, despite the greater phylogenetic distance for P. radiata (Fisher’s Exact test; two-tailed P value = 0.72). Heterozgyosity was estimated for 19 of the 23 markers where at least four individuals amplified in each taxon (Table 4). Heterozygosity values ranged from zero to 0.86 for individual loci within the two taxa with an average of 0.5 ± 1.9 for P. elliottii var. elliottii and 0.49 ± 0.23 for P. caribaea var. hondurensis. The mean H value and the mean allele number for all loci were the same in both P. elliottii var. elliottii and P. caribaea var. hondu- 823 Table 4 Heterozygosity (H) and number of alleles for 19 Pinus spp. microsatellite markers based on a sample of four P. elliottii var. elliottii and five P. caribaea var. hondurensis individuals. AlLocusa PtTX 3013 PtTX 2008 PtTX 2123 PtTX 3034a PtTX 3025 PtTX 3019 PtTX 3020 PtTX 2128 RPTest 01 RPTest 05 RPTest 08 RPTest 09a RPTest 11 RPTest 13 RPTest 20 NZPR1 NZPR5a NZPR6 NZPR7 Mean±SD leles were sized by capillary electrophoresis (see Methods and materials). Where a primer-pair was multicopy, only alleles at one locus were examined Focal sp. P. caribaea var. hondurensis P. elliottii var. elliottii Nonfocal population Allele sizes (bp) Number of alleles Allele size range (bp) H Number of alleles Allele size range (bp) H Total no. non-focal alleles Percentage of alleles unique to each taxa 129,132 305,305 204,204 199,211 256,266 219,223 171,183 244,270 124,124 199,199 194,194 264,279 214,214 267,276 254,254 133,133 112,114 184,184 113,139 4 2 2 2 3 4 3 2 3 1 2 4 2 3 1 8 3 5 5 3.00±1.57 128–141 305,316 198,201 205,213 254–268 210–221 147–163 231,238 112–130 195 192,195 269–282 209,214 267,276 254 121–145 80–84 172–190 77–101 0.66 0.38 0.50 0.38 0.41 0.72 0.53 0.50 0.53 0.00 0.22 0.69 0.22 0.53 0.00 0.86 0.64 0.74 0.80 0.49±0.23 3 2 2 7 2 2 4 2 2 2 2 2 3 2 2 5 3 5 5 2.86±1.42 138–145 305,316 198,201 191–209 252,268 205,208 151–159 231,238 112,124 192,195 192,195 252,258 209–217 267,276 232,254 121–149 78–82 184–196 93–113 0.59 0.38 0.47 0.84 0.22 0.50 0.75 0.50 0.38 0.22 0.38 0.22 0.41 0.50 0.47 0.76 0.41 0.78 0.83 0.50±0.19 6 2 2 8 4 6 7 2 3 2 2 6 3 3 2 11 4 8 9 4.5±2.7 67 0 0 88 75 100 100 0 33 50 0 33 33 33 50 82 50 75 89 49±33 a Multicopy loci are indicated by an “a” suffix on the locus label. Null alleles are known in NZPR7 but this was scored as a homozygote to be consistent with other loci where it was unknown whether there were null alleles rensis (two-tailed t-tests, P values > 0.5). The number of alleles for each locus ranged from one to eight in P. caribaea var. hondurensis and from two to seven in P. elliottii var. elliottii (Table 4). Across both taxa, the number of alleles ranged from two to eleven per locus, with an average of 4.5 ± 2.7 and with 49% ± 33% of alleles unique to each taxa. The H values for eight PtTX markers tested in this study were compared to H values reported for the same markers in P. taeda (Elsik et al. 2000). The loci tested were PtTX 3025, PtTX2008, PtTX2123, PtTX2128, PtTX3019, PtTX3034, PtTX3013 and PtTX3020. There was no difference in the average H for the markers when they were transferred to the nonfocal population of P. elliottii var. elliottii and P. caribaea var. hondurensis (mean ± SD of H focal = 0.59±0.24, nonfocal = 0.55±0.10; two-tailed paired t-test, t value of 0.68, P = 0.51). To test whether different sources of markers yielded different levels of variability in the nonfocal population, we compared the H values between sources of polymorphic transpecific markers. The PS – G source was excluded because all of these soft pine microsatellites were monomorphic. A one-way ANOVA indicated a significant difference in average H between the sources (oneway ANOVA F-value = 6.9, P value = 0.004; Table 5). The PR – G source had the highest average H (0.73±0.14), although it was not significantly higher than the PT – LC or PT – G sources (0.53±0.14 and 0.49±0.01, respectively). The PT – EST source had the Table 5 A comparison of microsatellite variability (H) in the nonfocal population by focal taxa and library typea. Least significant difference (LSD) test was applied at P<0.05 following a one-way ANOVA test; P value = 0.004 Locus source Number of loci (taxa and library type) H (mean ± SD) LSDb P. radiata – G P. taeda – LC P. taeda – G P. taeda – EST 4 6 2 7 0.73±0.14 0.53±0.14 0.49±0.01 0.34±0.14 a a,b a,b,c c a b See Table 3 Different letters indicate that H was significantly different (See Results) lowest H (0.34±0.14) and was significantly lower than all other sources except PT – G. The variability at microsatellite loci in different repeat type classes was compared using an average of the heterozygosity for each of the two nonfocal taxa. Di-nucleotide repeat microsatellite loci were significantly more variable than the tri-nucleotide repeat loci derived from either genomic or EST libraries (Table 6). Markers NZPR7 and PtTX2037 had null alleles. No other null alleles were confirmed in a sample of 21 polymorphic or monomorphic markers (see Table 1). As previously reported for hard pines, microsatellite loci RPTest 09 and NZPR5 were multilocus in P. elliottii var. elliottii and P. caribaea var. hondurensis (Echt and Burns 1999; 824 Table 6 A comparison of microsatellite variability (H) for microsatellite repeat type and library type in the nonfocal population of P. elliottii var. elliottii and P. caribaea var. hondurensis. Heterozygosity is an average of values for P. elliottii var. elliotttt and P. caribaea var. hondurensis. See Table 3 for library type categories. Only loci which were perfect or compound as defined by Weber (1990) and modified by Elsik (2000) were included. Loci which had a combination of di- and trimers were excluded from the analysis. Least significance difference test applied following one-way ANOVA on the three group: F = 9.5, P-value=0.001 One or two alleles from each of six PtTX loci that amplified from megagametophytes from both a P. elliottii var. elliottii and a P. caribaea var. hondurensis individual were sequenced and aligned with the corresponding loci from P. taeda. The repeat unit number accounted for most of the differences between the focal and the nonfocal species (Table 7). The repeat motif was conserved in all transfer microsatellites except for locus PtTX 2037. At this locus, a dinucleotide motif (GA) was present in both nonfocal species that was absent in P. taeda. The focal and nonfocal species were found to also differ by a total of three single-base pair insertions and 13 single-base pair substitutions within the flanking regions of the six loci. Linear regression was carried out on the PCR product length and repeat length for four loci where there was sufficient data and variability. The relationship was highly significant for PtTX 3013 (R2 = 0.99, P value << 0.001) and PtTX 3034 (R2 = 1, P value = 0), significant for PtTX 2037 (R2 = 0.99, P value = 0.03) but not highly significant for PtTX 2123 (R2 = 0. 89, P value = 0.06). Ascertainment bias was not evident between P. taeda and the two nonfocal species. Within the nonfocal group, only two single base pair substitutions were found between loci in P. elliottii var. elliottii and P. caribaea var. hondurensis. The sequence length of alleles in the nonfocal group were generally shorter for 12 of the 15 alleles, but there was not a significant reduction in sequence or repeat length between the focal and nonfocal population (two-tailed paired t-tests; t = 1.49, P = 0.19 and t = 1.628 P = 0.17 for sequence and repeat length, respectively). Table 7 Repeat region structural and flanking sequence mutations at six microsatellite loci from an individual from each of the taxons P. elliottii var. elliottii, P. caribaea var. hondurensis, and a P.taeda. Mutations are relative to sequence position in Genbank accessions for P. taeda. Substitutions noted as, for example, G-C represents a change from a G in P. taeda to a C in P. elliottii var. elliottii and P. caribaea var. hondurensis. Insertions noted as, for example, -T is an insertion of a T into P. elliottii var. elliottii and P. caribaea var. hondurensis. AC to T mutation at locus PtTX3002 only occurs in P. caribaea var. hondurensis Microsatellite repeat typea and source class n H (mean ± SD) LSDb Di – G Tri – G Tri – EST 8 6 6 0.64±0.13 0.49±0.13 0.32±0.15 a b b a b Di Di-nucleotide repeat motif · Tri tri-nucleotide repeat motif Different letters indicate a significant difference in H (see Results) Fisher et al. 1998). Microsatellites NZPR3, PtTX3029 and PtTX2034 also appeared to be multicopy in P. elliottii var. elliottii and P. caribaea var. hondurensis. Transpecific polymorphism: a sequence-based comparison Locus Taxona GenBank acc. no. Allele Repeat motif Repeat length (bp) Sequence length (bp) Mutuations in flanking sequence PtTX2037 PT PE PC PT AF143959 1 1 1 1 (GTGA)8 (GT)14 (GTGA)1 (GA)2 (GT)12 (GTGA)1 (GA)4 (GT)17 (AGC)8 60 32 46 24 176 150 164 202 A-C (34); 2 ×-T (139, 154) 1 1 2 1 1 1 1 1 2 1 2 1 1 1 1 1 2 (AGC)8 (AGC)8 (AGC)7 (AGG)6 (AGG)6 (AGG)6 (GTT)10 (GTT)9 (GTT)12 (GTT)9 (GTT)12 (CAA)8 (CAA)1 (GT)10 (GA)13 (GT)9 (GA)10 (GT)14 (GA)12 (GT)9 (GA)13 24 24 21 18 18 18 30 27 36 27 36 24 3 46 38 52 44 203 203 200 194 194 194 134 132 141 132 141 211 190 207 199 213 205 PtTX2123 PtTX3002 PtTX3013 PtTX3020 PtTX3034 a PE PC PC PT PE PC PT PE PE PC PC PT PC PT PE PC PC AF143960 AF277846 AF143966 AF143969 AF143974 PT Pinus taeda · PE P. elliottii var. elliottii · PC P. caribaea var. hondurensis -G (301); 3 × T-A (318, 325, 327); 3 × A-G (321, 323, 326) C-T (75); T-C (93) G-T (95); -T (175) G-A (152) G-C (101); T-A (172) 825 Discussion Microsatellite loci transferred to P. elliottii var. elliottii and P. caribaea var. hondurensis equally from P. radiata and P. taeda. The transfer rates to both of the nonfocal taxa was similar, and no ascertainment bias was detected. The length of the repeat sequence in microsatellites transferred to the nonfocal population closely corresponded with PCR product size. Changes in the repeat structure or flanking sequence were minor and consistent with current taxonomic and phylogenetic relationships. Homologous microsatellite loci were as polymorphic in the two nonfocal species as in the focal species. Efficient microsatellite transfer The success of the transfer of polymorphic markers from P. radiata (subsection Oocarpae) was the same as for markers from P. taeda (subsection Australes). Also, despite a higher proportion (36/50) of loci from a focal species within the same subsection than nonfocal species, the overall transfer of polymorphic loci was similar (23/50) to the rate across a subsection. The rate of transfer for polymorphic loci across a subsection (4/9) was similar to that found previously for transfer of loci from P. radiata to P. taeda (9/20) (Devey et al. 1999). These subsections are believed to have evolved separately for at least 10 million years but perhaps as long as 35 million years, (Miller 1993; Krupkin et al. 1996). Microsatellite markers originating from focal species external to a nonfocal species subsection may be as useful as markers transferred from within the same subsection in hard pines, at least where they they have only diverged within the past 10–35 million years. Microsatellite loci from P. strobus (section Strobus), while transpecific, were not polymorphic between the taxa. This was consistent with previously reported low levels of variation for these markers amongst species of hard pines (Echt et al. 1999). The hard and soft pine sections have been separated for around 130 million years (Mirov 1967). Nonetheless, the transfer of microsatellite markers over these evolutionary – and even greater (140–195 million years) – distances in the pines and of those which exhibit variation amongst hard pine taxa has been possible (Karhu et al. 2000; Kutil and Williams 2001). Conserved microsatellite loci in pines were those with perfect triplet-repeat motifs (Kutil and Williams 2001). The type of library from which a microsatellite originated affected its H in the nonfocal group. EST-derived microsatellites were less polymorphic than loci identified from other library types in P. taeda. In rice, microsatellites in EST were found to be less polymorphic than those derived from genomic libraries (Cho et al. 2000). Our data suggests this may also be the case in Pinus. However, caution is required as the effect may be confounded with differences due to repeat types, as all ESTderived loci we tested were perfect trinucleotide or com- pound trinucleotide repeat loci. Loci with trinucleotide repeats as a group were less variable than dinucleotide repeat sequences in our study, and a similar effect has also been found in both Drosophila and humans (Chakraborty et al. 1997; Schug et al. 1998). A comparison of EST-derived microsatellite loci with microsatellite loci derived from genomic libraries unconfounded by differences in repeat types will be required to confirm whether EST-derived loci have lower polymorphism. Although EST-derived microsatellite markers with trinucleotide repeats had a lower average H than dinucleotide repeats, they had similar H to other microsatellite loci containing trinucleotide repeats. EST-derived microsatellite markers also had a higher transfer rate than the average for all markers. This high transfer rate of ESTderived microsatellites compared with markers from other sources probably resulted from a higher sequence conservation in primer binding sites. Furthermore, EST-derived markers were proportionately well represented amongst loci that were mapped in a segregating population (data not shown), indicating that despite a lower polymorphism EST are a valuable source of markers. Repeat structure and flanking mutations are phylogenetically informative Differences in repeat structure and flanking sequence mutations were found to align with the taxonomic assignment of P. elliottii var. elliottii, P. caribaea var. hondurensis and P. taeda and were in agreement with phylogenetic relationships based on other molecular studies (Dvorak et al. 2000; Little and Critchfield 1969; Nelson et al. 1994), Dvorak et al. (2000) proposed evolutionary scenarios for the Australes and Oocarpae subsections based on a random amplified polymorphic DNA (RAPD) phylogeny, other marker data and information from crossing experiments and observations of natural hybridisation and distributions. They proposed a common ancestral origin for the Mesoamerican Oocarpae and Australes subsections. Because P. caribaea var. hondurensis has a genetic constitution of both Oocarpae and Australes, it shared a common ancestor with Mesoamerican Oocarpae and Australes but diverged before this split. A divergence between an ancestral. P. caribaea var. hondurensis and the ancestors that formed two clades within Australes – one consisting of the remaining P. caribaea varieties, caribaea and bahamensis, and the two P. elliottii varieties, elliottii and densa, and the second clade containing P. taeda, P. palustris and P. echinata – was not resolved. On the basis of this genetic distance data, P. caribaea var. hondurensis should be equidistant to P. elliottii var. elliottii, and P. taeda. Nevertheless, morphological and biochemical data supports a close relationship between P. caribaea var. hondurensis and P. elliottii var. elliottii, and they were once classified as a single species (Little and Dorman 1954; Nikles 1966). The affinity of P. elliottii var. elliottii to P. caribaea var. hondurensis compared with P. taeda was also evident in studies 826 of chloroplast DNA variation (Nelson et al. 1994). Pinus elliottii var. elliottii shared its second most common haplotype with Pinus caribaea var. hondurensis, but neither taxa shared haplotypes with P. taeda. Hence, in our study, it was expected that P. elliottii var. elliottii and P. caribaea var. hondurensis would show a greater similarity in DNA sequence to each other than to P. taeda. Most of the variation in allele length was attributed to changes in the number of repeat units in our study. Where the repeat structure differed in the nonfocal samples from focal taxa, the change was consistent between the two nonfocal taxa. Similarly, the majority of flanking sequence single-base pair insertions or deletions and substitutions were constant between P. elliottii var. elliottii and P. caribaea var. hondurensis. Our data demonstrated a strong affinity between P. elliottii var. elliottii and P. caribaea var. hondurensis compared to P. taeda, consistent with previous phylogenic analysis. Microsatellite repeat structure and flanking sequence mutations should be useful for phylogenetic analysis within these groups. Repeat structure in microsatellites and flanking sequence mutations has been found to be phylogenetically informative across a subfamily of wasps (Zhu et al. 2000). No evidence of ascertainment bias for P. taeda microsatellite loci Ascertainment bias is the reduction in allele length or variability in the nonfocal species (Ellegren et al. 1995). It is attributed to an artifact of the microsatellite marker development process such that loci with large numbers of repeats tend to be preferentially isolated and cloned. Evidence for ascertainment bias from reciprocal studies has been found for Drosophila spp. (Hutter et al. 1998) and swallows (Kirchman et al. 2000) but did not fully explain the reduction in allele sizes of human microsatellites transferred to chimpanzees (Cooper et al. 1998) or may only be a minor affect in other animals (Crawford et al. 1998). We found no evidence for ascertainment bias for markers developed from total genomic or low-copy libraries in P. taeda as there was no reduction in the mean H values in the nonfocal population. This lack of significant loss in variability may reflect the relative closeness of the taxa in this study, as there should be a negative relationship between genetic distance and the degree of polymorphism with ascertainment bias (Ellegren et al. 1995). Microsatellite markers from other hard pines transferred equally to P. elliottii var. elliottii and P. caribaea var. hondurensis. Variation in the structure and mutations at these loci are consistent with current phylogenetic relationships, and the lack of ascertainment bias for markers is promising for exchanging markers in population and mapping studies in an important group of tropical pines. Microsatellite marker transfer was an efficient strategy for obtaining polymorphic markers among closely re- lated taxa. Microsatellite markers from P. contorta (section Pinus, subsection Contortae) (Hicks et al. 1998) and P. sylvestris (section Pinus, subsection Sylvestres) (Kostia et al. 1995; Soranzo et al. 1998) may also be useful in P. elliottii var. elliottii and P. caribaea var. hondurensis, although more distant than P. radiata (Krupkin et al. 1996). Acknowledgements The authors thank P. Toon and L. Perkins for assistance in collecting P. elliottii var. elliottii and P. caribaea var. hondurensis samples, A. Cameron and M. Devey for tissue or DNA of P. radiata and P. strobus, respectively, and B. Potts, M. Rossetto, R. Stokoe and K. Shepherd for comments on early versions of the manuscript. This work was supported by the Cooperative Research Centre for Sustainable Production Forestry. Experiments complied with the current laws in Australia. References Chakraborty R, Kimme1 M, Stivers DN, Davison LJ, Deka R (1997) Relative mutation rates at di-, tri-, and tetranucleotide microsatellite loci. Proc Natl Acad Sci USA 94:1041–1046 Cho YG, Ishii T, Temnykh S, Chen X, Lipovich L, McCouch SR, Park WD, Ayres N, Cartinhour S (2000) Diversity of microsatellites derived from genomic libraries and GenBank sequences in rice (Oryza sativa L.). Theor Appl Genet 100:713–722 Cooper G, Rubinsztein DC, Amos W (1998) Ascertainment bias cannot entirely account for human microsatellites being longer than their Chimpanzee homologues. Hum Mol Genet 7:1425– 1429 Crawford AM, Kappes SM, Paterson KA, Degotari MJ, Dodds KG, Freking BA, Stone RT, Beattie CW (1998) Microsatellite evolution – testing the ascertainment bias hypothesis. J Mol Evol 46:256–260 Crow JF (1986) Basic concepts in population, quantitative and evolutionary genetics. Freedman, New York Devey M, Sewell MM, Uren TL, Neale DB (1999) Comparative mapping in loblolly and radiata pine using RFLP and microsatellite markers. Theor Appl Genet 99:656–662 Dvorak W, Jordan A, Hodge G, Romero J (2000) Assessing evolutionary relationships of pines in the Oocarpae and Australes subsections using RAPD markers. New For 20:163–192 Echt C, Burns R (1999) SSR derived from Pinus taeda ESTs. http://dendromeucdavisedu/Data/primerhtml Echt CS, Maymarquardt P, Hseih M, Zahorchak R (1996) Characterisation of microsatellite markers in eastern white pine. Genome 39:1102–1108 Echt CS, Vendramin GG, Nelson CD, Marquardt P (1999) Microsatellite DNA as shared genetic markers among conifer species. Can J For Res 29:365–371 Ellegren H, Primmer C, Sheldon B (1995) Microsatellite ‘evolution’: directionality or bias. Nat Genet 11:360–362 Elsik CG, Minihan VT, Hall SE, Scarpa AM, Williams CG (2000) Low-copy microsatellite markers for Pinus taeda L. Genome 43:550–555 Fisher PJ, Richardson TE, Gardner RC (1998) Characteristics of single- and multi-copy microsatellites from Pinus radiata. Theor Appl Genet 96:969–979 Graham GC, Henry RJ, Godwin ID, Nikles DG (1996) Phylogenetic position of Hoop Pine. Aust Syst Bot 9:893–902 Hicks M, Adams D, O’Keefe S, Macdonald E, Hodgetts R (1998) The development of RAPD and microsatellite markers in lodgepole pine (Pinus contona var. latifolia). Genome 41:797–805 Hutter CM, Schug MD, Aquadro CF (1998) Microsatellite variation in Drosophila melanogaster and Drosophila simulans: a reciprocal test of the ascertainment bias hypothesis. Mol Biol Evol 15:1620–1636 Karhu A, Dieterich JH, Savolainen O (2000) Rapid expansion of microsatellite sequences in pines, Mol Biol Evol 17:259–265 827 Kirchman JJ, Whittingham LA, Sheldon FH (2000) Relationships among cave swallow populations (Petrochelidonfulva) determined by comparisons of microsatellite and cytochrome b data. Mol Phylo Evol 14:107–121 Kostia S, Varvio SL, Vakkari P, Pulkkinen P (1995) Microsatellite sequences in a conifer, Pinus sylvestris. Genome 38:1244–1248 Krupkin AB, Liston A, Strauss SH (1996) Phylogenetic analysis of the hard pines (pinus subgenus Pinus, Pinaceae) from chloroplast DNA restriction site analysis. Am J Bot 83:489–498 Kutil B, Williams CG (2001) Triplet-repeat microsatellites shared among hard and soft pines. J Hered (in press) Lamb A (1973) Pinus caribaea: volume 1: fast-growing timber trees of the lowland tropics. Trop For Pap no. 6, University of Oxford, Oxford Little L, Critchfield W (1969) Subdivisions of the Genus Pinus. Report no. 1144, US For Serv Misc Publ Little E, Dorman K (1954) Slash pine (Pinus elliottii), including south Florida slash pine: nomenclature and description (SE36). USDA For Serv Res Pap Miller C (1977) Mesozoic conifers. Bot Rev 43:217–280 Miller C (1993) Impact of the Ecocene on the evolution of Pinus L. Ann Mo Bot Gard 80:471–498 Mirov NT (1967) The Genus Pinus. Ronald Press, New York Nelson CD, Nance WL, Doudrick RL (1993) A partial genetic linkage map of slash pine (Pinus elliottii Engelm. var. elliottii) based on random amplified polymorphic DNAs. Theor Appl Genet 87:145–151 Nelson CD, Nance WL, Wagner DB (1994) Chloroplast DNA variation among and within taxonomic varieties of Pinus caribaea and Pinus elliottii. Can J For Res 24:424–426 Nikles D (1966) Comparative variability and relationships of Caribbean pine (Pinus caribaea Mor.) and slash pine (Pinus elli- ottii Engelm.) PhD thesis, North Carolina State University, Raleigh, N. C. Peakall R, Gilmore S, Keys W, Morgante M, Rafalski A (1998) Cross-species amplification of soybean (Glycine max) simple sequence repeats (SSRs) within the genus and other legume genera: implications for the transferability of SSRs in plants. Mol Bio1 Evol 15:1275–1287 Pfeiffer A, Olivieri AM, Morgante M (1997) Identification and characterisation of microsatellites in Norway Spruce (Picea abies K). Genome 40:411–419 Schug MD, Hutter CM, Wetterstrand KA, Gaudette MS, Mackay TFC, Aquadro CF (1998) The mutation rates of di-, tri- and tetranucleotide repeats in Drosophila melanogaster. Mol Bio1 Evol 15:1751–1760 Scott LJ, Cross M, Shepherd M, Maguire T, Henry W (1999) Increasing the efficiency of microsatellite discovery from poorly enriched libraries in coniferous forest species. Plant Mol Bio1 Rep l7:351–354 Smith DN, Devey ME (1994) Occurrence and inheritance of microsatellites in Pinus radiata. Genome 37:977–983 Soranzo N, Provan J, Powell W (1998) Characterisation of microsatellite loci in Pinus sylvestris L. Mol Ecol 7:1247–1263 Thompson J, Higgins D, Gibson T (1994) CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res 22:4673–4680 Weber J (1990) Informativeness of human (dC-dA)n - (dG-dT)n polymorphisms. Genomics 7:524–530 Zhu Y, Queller DC, Strassmann JE (2000) A phylogenetic perspective on sequence evolution in microsatellite loci. J Mol Evol 50:324–338