Transcriptome-derived microsatellite markers for Dioon (Zamiaceae) cycad species.

PREMISE OF THE STUDY: Dioon (Zamiaceae) is an endangered North American cycad genus of evolutionary and ornamental value. We designed and validated a set of microsatellite markers from D. edule that can be used for population-level and conservation studies, and that transferred successfully to D. angustifolium, D. spinulosum, and D. holmgrenii. METHODS AND
RESULTS: We tested 50 primers from 80 microsatellite candidate loci in the OneKP D. edule transcriptome. Genotypes from 21 loci in 20 D. edule individuals revealed up to 14 alleles per locus and observed heterozygosity from 0.15 to 0.92; one locus was monomorphic. Seven of those 21 loci were polymorphic in D. angustifolium, D. spinulosum, and D. holmgrenii, with up to seven alleles, and an observed heterozygosity up to 0.89.
CONCLUSIONS: The transcriptome-derived microsatellites generated here will serve as tools to advance population genetic studies and inform conservation strategies of Dioon, including the identification and origin of illegal plants in the cycad trade.

Of Permian origin, cycads (Cycadales) are dioecious gymnosperms distributed in tropical and subtropical regions (Norstog and Nicholls, 1997). Dioon Lindl. is composed of 14 species, 13 of which are distributed in Mexico and one in Honduras (Osborne et al., 2008). Dioon is of economic importance as an ornamental plant, as well as an alternative food source to main crops, and it has cultural value throughout its distribution area (Bonta and Osborne, 2008). Dioon edule Lindl. is a medium-sized cycad with an erect trunk of 1.5 m and a rigid crown of 15–25 long, blue-green leaves that is endemic to eastern Mexico, growing mainly in tropical deciduous thorn forests and oak forests (Octavio-Aguilar et al., 2008). They can live up to 2000 yr and have slow growth rates with long reproductive cycles (Vovides, 1990). Despite legal protection in the Norma Oficial Mexicana of the Secretaría de Medio Ambiente y Recursos Naturales (Nom-059 SEMARNAT-2010) and the Convention on International Trade in Endangered Species of Wild Fauna and Flora (CITES; 2008 Appendix II), poaching of Dioon species is still common and wild populations are disappearing.

Allozyme variation in D. edule has shown that there is unusually high genetic diversity in this genus (González-Astorga et al., 2003; Cabrera-Toledo et al., 2010, 2012), but these markers are limited in their scale of analysis and in their use for conservation, as they lack individual-level resolution and are dominant markers. Independent laboratories (J. C. Bede, A. Cibrián-Jaramillo, D. Cabrera-Toledo, and L. Yañez-Espinosa, personal communication) have been unable to replicate microsatellites previously developed for D. edule (Moynihan et al., 2007). Therefore, there is a need to develop robust genetic markers to understand population genetic history and to inform conservation strategies in Dioon.

METHODS AND RESULTS

Genomic DNA was isolated from 20 randomly chosen samples out of 40 D. edule individuals, representing four populations found in the states of San Luis Potosí and Veracruz, Mexico (Appendix 1). Leaflets were ground in liquid nitrogen and sieved through a fine 0.5-mm mesh to remove cuticle and fiber particles. DNA was extracted using the DNeasy Plant Mini Kit (QIAGEN, Valencia, California, USA) following the manufacturer’s protocols after adjustment for the extraction of 200 mg of tissue. The D. edule transcriptome was accessed through the OneKP project (www.onekp.com), and a total of 121,771 contigs were analyzed for tandem repeats using the algorithm mreps version 2.5 (Kolpakov and Kucherov, 1999) available at the Mobyle Portal (http://mobyle.pasteur.fr/cgi-bin/portal.py?#forms::mreps). We targeted five or more tandem nucleotide repeats prioritizing di-, tri-, and tetranucleotide repeats with adjacent 5′ and 3′ 15–30-nucleotide sequences for primer design. Primers were designed with Primer3Plus (Untergasser et al., 2007), and self-annealing and heterodimer formation was tested with OligoAnalyzer version 3.1 (Integrated DNA Technologies, Coralville, Iowa, USA). Eighty of 150 microsatellites identified in the D. edule transcriptome had candidate primer sites. We designed primers for 80 loci and tested 50 of them in three randomly selected D. edule samples using a PCR protocol as shown for the M13(–21) fluorescent label. We chose 21 loci that produced bands consistently as evidenced by an agarose gel (Table 1). The issues with the remaining loci were double or multiple bands or lack of amplification. Forward primers for D. edule contained a 5′ extension of M13 following the protocol of Schuelke (2000). We used an infrared dye–labeled (LI-COR Biosciences, Lincoln, Nebraska, USA) M13(–29) sequence (CACGACGTTGTAAAACGAC) and a 6-FAM fluorescently labeled M13(–21) sequence (TGTAAAACGACGGCCAGT) (Sigma-Aldrich, St. Louis, Missouri, USA) (Table 1). We used 6-FAM to genotype other congeners: 10 individuals for D. angustifolium Miq. from one population, and 40 each for D. holmgrenii De Luca, Sabato & Vázq. Torres and D. spinulosum Dyer ex Eichler, as based on the availability of these individuals in the field. These three species are representatives of separate phylogenetic clades, with D. spinulosum being sister to the rest of Dioon, and D. angustifolium and D. edule sister to D. holmgrenii, according to González et al. (2008). The PCR mixture contained a minimum of 40 ng/μL of DNA template, 200 μM deoxynucleotides (New England Biolabs, Ipswich, Massachusetts, USA), 0.25 units of Taq DNA polymerase (New England Biolabs), 0.08 μM of forward primer with the 5′-M13 tail, and 0.2 μM reverse primer; we added 0.05 μM infrared dye–labeled primer for M13(–29) and 0.2 μM for the fluorescently labeled M13(–21) primer, plus 1 μL of 10× PCR buffer (Sambrook et al., 1989). PCR conditions for M13(–29) were: 94°C for 3 min; 16 cycles at 94°C for 30 s, 58–60°C for 1 min, and 72°C for 30 s; 10 cycles at 94°C for 30 s, 55°C for 1 min, and 72°C for 30 s; and 72°C for 2 min. PCR conditions for M13(–21) were: 95°C for 2 min; followed by 30 cycles at 95°C for 30 s, 55–58°C for 30 s, and 72°C for 1 min; then eight cycles at 95°C for 30 s, 53°C for 30 s, and 72°C for 1 min; and 70°C for 10 min. All protocols are modifications of Schuelke (2000). The annealing temperature for specific primer pairs is shown in Table 1. We included information on protein sequence matches to our loci according to a TBLASTX search against the nucleotide collection (nr/nt) with default parameters at the National Center for Biotechnology Information (NCBI). One of the 21 loci was monomorphic (locus 2015232), and 20 were polymorphic in 20 randomly chosen individuals from four D. edule localities (Table 2). Seven of those 21 loci, including the monomorphic locus 2015232, were consistently polymorphic in D. angustifolium, D. spinulosum, and D. holmgrenii (Table 3). The remaining 14 loci would require additional optimization to remove stutter bands. Amplicons for D. edule were separated on a 6.5% acrylamide gel on a NEN 4300 DNA Analyzer (LI-COR Biosciences) and compared to the LI-COR size standard 4200-44 (50–350 bp) (LI-COR Biosciences). Bands were scored using SAGAGT version 3.3 software (LI-COR Biosciences). The seven loci tested in Dioon congeners were genotyped in a separate run with an ABI 3730xl sequencer with GeneScan 500 LIZ size standard (Applied Biosystems/Thermo Fisher Scientific, Waltham, Massachusetts, USA) and interpreted with Geneious version 8.1.3 software (Biomatters, http://www.geneious.com/). Scoring errors that may result from stuttering, large allele drop out, or null alleles were identified using MICRO-CHECKER version 2.2.3 (van Oosterhout et al., 2004). Observed (Ho) and expected (He) heterozygosities were calculated using the R package Adegenet (Jombart, 2008). No evidence of scoring errors due to peak stuttering or large allele dropout was observed. Ho and He of the 21 microsatellite markers in D. edule ranged from 0.15 to 0.92 and 0.41 to 0.87, respectively (Table 2). Loci 2002082 and 2002757 had a homozygote excess that was not evenly distributed across all homozygote classes, which could be indicative of null alleles (van Oosterhout et al., 2004). PCR amplification with lower temperatures (52–56°C) did not recover any additional alleles for these loci. The observed heterozygosity confirms previous allozyme studies and is congruent with Dioon’s mating system (González-Astorga et al., 2003; Cabrera-Toledo et al., 2010). The number of alleles for the transferred loci ranged from one to seven, and Ho and He ranged from 0.33 to 0.89 and 0.24 to 0.71, respectively (Table 3), which suggests variability in other species. Vouchers were deposited at the Jardín Botánico Francisco Javier Clavijero in Xalapa, Veracruz, Mexico, and at the McGill University Herbarium (MTMG), Québec, Canada.

Table 1.

Characteristics of 21 microsatellite loci developed for Dioon edule.

Locus	Primer sequences (5′–3′)	Repeat motif	Ta (°C)	GenBank accession no.	Organism/Putative function/Accession no.a	E-value
2001304b	F: CTGGGCCTCGACATAACATT	(ATAA)5	60	KT289910	Picea sitchensis EST BT123050	0
	R: TCAAAATCATTCCGGCTTTC
2001597b	F: CTTACAAGCGGCACCATTG	(TA)12	60	KT289911	None	—
	R: AAGCAGGCCAGACTTCAGAC
2001955b	F: CTGCCGAGGAGGGACA	(TGC)8	58	KT289912	Pinus monticola WRKY TF EU269755	3.00E-41
	R: CGCAGGGTTGGAGAGC
2002082b	F: TGACCTTGCCTTAGGTCAAAA	(GT)14	60	KT289913	None	—
	R: AGATGTGGGTGACACGTCCT
2002349b	F: AGAGCTGCTTCCACGTTCAT	(TCT)6	60	KT289914	None	—
	R: GCGGAAACTTCTTCAACAGC
2002757b	F: TGGGAAATGCACACCTAAAA	(CA)20	58	KT289915	None	—
	R: ACCTGGGCCACTTGAGG
2003643b	F: CGAACTTGAAGACGATGACG	(GCA)7	60	KT289916	None	—
	R: CGGGTAGCACCAAAGATTGT
2014158b	F: CACCGTGCCCGTCATT	(TTC)9	60	KT289917	None	—
	R: GCTGGCCCTGCAAAGA
2016409b	F: AGCGCCGCTGTCATTC	(ATT)9	59	KT289918	None	—
	R: TTCGGCTGTGCATCTCAA
2016799b	F: GAAGGGTGGGTATGGCACT	(GA)14	59	KT289919	None	—
	R: TCCTGCCTGCAAAGCAC
2011473b	F: CATAGGAGGCCACCATGTAGA	(TG)15	60	KT289920	None	—
	R: GCAAGTGCTTTGGATATGCTC
2017825b	F: GGGGCTGCATCCTTCC	(GA)10	60	KT289921	Picea glauca EST BT107925	0
	R: GATGGGGCAATGGGAAT
2015001b	F: TTGTTTGGGCAATTCCTGA	(TG)10	58	KT289923	Picea glauca EST BT116353	0
	R: CAAAGCCAACAAAGACATGA
2015907b	F: TGCCATCTTCCCATCACA	(AT)11	59	KT289924	Oryza brachyantha heat shock protein XM_006663065	0
	R: CCCAGGCCCCAAGATT
2014311c,e	F: TTAGGGGGCCGAGGAG	(CCA)6	58	KT261421	Zea mays Predicted formin-like protein XM_00867779	1.00E-44
	R: CCGACCGCCAAGAGAA
2015232c,d,e	F: TGGAGATCAACAACGACCAA	(CAT)6	58	KT261422	None	—
	R: CGAGACCCAGAGARCCTGAC
2018276c,e	F: CACCCTGCCAAAGGTCAT	(TG)14	58	KT261423	Picea glaucea EST BT106585	0
	R: GCACCCATTGTTGGACA
2019766c,e	F: CGTGCGACCAGCAAGA	(TCG)7	58/55	KT261424	None	—
	R: TCGCCGGTGAGAACAAG
2018116c,e	F: GGCAGATTAGCTCCAGCAG	(GAC)8	58/55	KT261425	Picea glauca EST BT117927	9.00E-57
	R: CAGAGCTTCCAATTCCTTGC
2018893c,e	F: GCGGTAGCTGGAGAGGTTC	(GAA)10	58/55	KT261426	None	—
	R: AGTCTGGGGCCTCATCAAC
2019765c,e	F: CAAATTCCTGTGGGAGATGG	(AGC)8	58/55	KT261427	Picea sitchensis EST WS02775_I20	7.00E-157
	R: GCAGGCAGTTTGGAAAGAAC

Note: Ta = annealing temperature in D. edule/other Dioon species.

Search made with the online version of TBLASTX versus the nr/nt database with default parameters at NCBI (http://blast.ncbi.nlm.nih.gov/Blast.cgi?PROGRAM=tblastx&PAGE_TYPE=BlastSearch&LINK_LOC=blasthome).

Forward primer contains a 19-bp M13(–29) infrared dye extension at the 5′ end.

Forward primer contains an 18-bp M13(–21) fluorescent dye extension at the 5′ end.

Monomorphic in D. edule but polymorphic in congeners.

Loci transferable to D. angustifolium, D. holmgrenii, and D. spinulosum.

Table 2.

Genetic diversity in 21 microsatellite loci developed in Dioon edule.

Locus	Allele size range (bp)	A	Ho	He
2001304	183–203	4	0.619	0.604
2001597	197–211	6	0.6	0.713
2001955	202–214	4	0.35	0.411
2002082	240–256	7	0.15	0.798
2002349	368–386	7	0.75	0.725
2002757	256–312	14	0.6	0.879
2003643	225–246	6	0.65	0.703
2014158	164–176	3	0.667	0.647
2016409	237–270	7	0.6	0.822
2016799	250–272	11	0.688	0.877
2011473	237–269	8	0.8	0.825
2017825	248–272	9	0.867	0.838
2015001	194–200	5	0.923	0.754
2015907	211–223	5	0.733	0.638
2014311b	243–255	4	0.666	0.722
2015232b,c	249	1	—	—
2018276b	252–278	6	1	0.781
2019766b	231–252	3	0.571	0.540
2018116b	189–204	5	0.666	0.694
2018893b	255–261	3	0.666	0.611
2019765b	167–179	5	0.714	0.704

Note: A = number of alleles; He = expected heterozygosity; Ho = observed heterozygosity.

Tested in 20 randomly chosen individuals from four D. edule localities; see Appendix 1.

Transferred to three other species.

Monomophic in D. edule but polymorphic in other species.

Table 3.

Genetic diversity of seven microsatellite loci in related Dioon species.

	D. angustifolium (n = 10)				D. holmgrenii (n = 40)				D. spinulosum (n = 40)
Locus	Allele size range (bp)	A	Ho	He	Allele size range (bp)	A	Ho	He	Allele size range (bp)	A	Ho	He
2014311	246–252	4	0.75	0.656	240–252	4	0.333	0.597	240–243	2	0	0.5
2015232a	249–252	2	0	0.244	240–249	2	0.642	0.497	240–258	5	0.466	0.62
2018276	268–274	3	0.7	0.505	266–280	5	0.461	0.636	260–283	6	0.5	0.57
2019766	246	1	—	—	231–246	2	0.833	0.503	231–246	2	0.772	0.447
2018116	201–210	4	0.5	0.563	192–204	4	0.894	0.68	204–213	4	0.5	0.546
2018893	255–261	3	0.666	0.5	255	1	—	—	258	1	—	—
2019765	170–179	3	0.5	0.402	155–179	7	0.896	0.717	161–179	5	0.476	0.399

Note: A = number of alleles; He = expected heterozygosity; Ho = observed heterozygosity; n = number of individuals sampled.

Locus 2015232 was monomorphic in D. edule.

CONCLUSIONS

We identified and validated 21 new microsatellite loci in D. edule, one being monomorphic for this species. Seven of these 21 markers are polymorphic in the congeners D. angustifolium, D. holmgrenii, and D. spinulosum, which are representatives of divergent phylogenetic clades. This suggests that these markers are likely transferable to additional Dioon species. Our loci are useful for Dioon population genetics and have great potential to be used in in situ and ex situ conservation strategies, including as a means to help authorities identify the origins of illegal plant material.

Appendix 1.

Voucher information for Dioon species used in this study.

Species	Voucher no.a	Collection localityb	N
D. edule	2001-114	Actopan, Veracruz	10
D. edule	2002-013	Cd. Valles, San Luis Potosí	10
D. edule	2011-AGA21 (MTMG)	Agua de Gamotes, San Luis Potosí	10
D. edule	2011-GRAP43 (MTMG)	Los Pocitos, San Luis Potosí	10
D. angustifolium	JGA 10-2003	Aldama, Tamaulipas	10
D. holmgrenii	JGA 15-2005	Rancho El Limon, Oaxaca	20
D. holmgrenii	JGA 16-2005	San Bartolome Loxicha, Oaxaca	20
D. spinulosum	JGA 23-2007	Soyaltepec, Oaxaca	20
D. spinulosum	JGA 24-2007	Cerro Bola, Oaxaca	20

Note: N = number of individuals.

Vouchers are deposited at the Jardín Botánico Francisco Javier Clavijero in Xalapa, Veracruz, Mexico, and at the McGill University Herbarium (MTMG), Québec, Canada; JGA = Jorge González-Astorga.

Locality in Mexico; exact coordinates are not listed to protect endangered populations but are available upon request.