The tribe Euglossini comprises 218 species in 5 genera, including the genus Eulaema with 29 described species (Oliveira 2006; Nemésio 2009). Euglossine bees (Hymenoptera: Apidae), commonly known as orchid bees, are charismatic insects characterized by extremely long tongues and shiny iridescent colors (Roubik and Hanson 2004). Orchid bees are abundant in the Neotropics (López-Uribe et al. 2008) and are considered keystone species in lowland forests because of the ecological role that they play as pollinator of orchids (Dressler 1982) and many other flowering plants (Ramírez et al. 2002).

Orchid bees have recently been targets of conservation concern (Zayed 2004). There is evidence demonstrating that the species diversity of euglossine bees is negatively affected by habitat fragmentation (Brosi 2009). In addition, genetic studies using allozyme markers have shown that some orchid bee populations exhibit high frequencies of diploid males indicating high levels of inbreeding and/or low effective population size (Roubik et al. 1996; Zayed 2004; López-Uribe et al. 2007; but see Takahashi et al. 2001). However, a recent study looking for diploid males using microsatellite markers (Souza et al. 2010) found diploid males to be rare in euglossine bee populations suggesting that the high frequency of diploid males previously reported may be the result of technical flaws in the allozyme-based studies. Therefore, the development of microsatellite markers is essential for the study of population structure and conservation genetics of this group of bees. Here, we describe and characterize eight polymorphic microsatellite loci in Eulaema meriana, and tested these loci across seven other Eulaema species.

A genomic DNA library enriched for 12 microsatellite repeat motifs was created from one individual of E. meriana using a universal linker and ligation procedure (Hamilton et al. 1999; Grant and Bogdanowicz 2006). Transformed bacterial colonies were then screened for microsatellites through hybridization to 33P-radiolabeled oligonucleotides. More than 800 positive clones were obtained from this method and ~200 of them were sequenced with universal M13 primers that flank the cloned insert for microsatellite primer design. PCR primer pairs were designed for 29 microsatellite loci using the software PrimerSelect (DNASTAR). Nine of these loci were tested for PCR amplification quality and variability.

For microsatellite PCR amplifications, a universal tag method with three primers was employed (Schuelke 2000). This approach allows fluorescent labeling of PCR fragments with a single dye-labeled tag used simultaneously with the unlabeled locus-specific (ULS) forward primer containing 20 additional bases at the 5′-end and the ULS reverse primer. The reverse primer was modified by adding a six base pair ‘pigtail’ (GTTTCT) to the 5′-end (Brownstein et al. 1996) to facilitate genotyping by reducing stutter. PCR amplifications contained 5× GoTaq buffer pH 8.5, 2 mM MgCl2, 0.2 mM dNTPs, 0.1 μM ULS forward primer, 0.2 μM dye-labeled tag, 0.2 μM ULS reverse primer, 1U GoTaq DNA polymerase (Promega) and 10–50 ng DNA in 20 μl total volume. PCR cycling conditions consisted of one cycle at 94°C for 30 s, 35 cycles at 94°C for 30 s, 45 s at the locus-specific annealing temperature (Table 1) and 45 s at 72°C, followed by one step of 7 min at 72°C. Cycling was carried out using a Biometra TGradient thermal cycler. Labeled PCR products were analyzed on an Applied BioSystems 3730 l× DNA Analyzer using the allele size standard GeneScan-500 LIZ and called using the software PeakScanner (Applied BioSystems).

Table 1 Primer sequences, repeat motif and annealing temperature (T a) of eight microsatellite loci isolated from Eulaema meriana

Genomic DNA was extracted from males of E. meriana (N = 55), Eulaema cingulata (N = 15), Eulaema bombiformis (N = 10), Eulaema chocoana (N = 3), Eulaema luteola (N = 2), Eulaema mocsaryi (N = 2), Eulaema nigrifacies (N = 1) and Eulaema nigrita (N = 1) (Table 1) using the QIAGEN DNeasy Tissue kit. Characterization of each locus was based on one E. meriana population (N = 40) from La Selva, Costa Rica (Table 1). All loci were checked for amplification variability in four E. meriana populations and across the other seven Eulaema species (Table 2). Due to the haploid nature of the data, tests for Hardy–Weinberg equilibrium and linkage disequilibrium were not performed. Number of alleles per locus (N A) and expected heterozygosity (H E) were calculated using Microsatellite Analyser (MSA) (Dieringer and Schlotterer 2003).

Table 2 Characterization of microsatellite loci isolated from Eulaema meriana from four localities and cross-species amplification for seven other species of the genus Eulaema

The number of alleles per locus for E. meriana ranged from 4 to 9 in the population from La Selva (Costa Rica) (Table 1) and from 4 to 11 when including individuals from the other 3 populations analyzed (Table 2). Null alleles were only detected for the locus EM40 in one E. meriana individual. All microsatellite loci were easily genotyped in all species except for locus EM107 in E. luteola and E. nigrita. Stutter was only evident in locus EM70 for E. luteola. None of the 90 individuals analyzed showed a diploid genotype. Successful cross-species amplification of these loci shows that the microsatellite markers here described will be useful tools for future population and conservation genetic studies in E. meriana and several species of the genus Eulaema.