Trypanosomatidae are obligate parasites of mainly invertebrate hosts, although some species such as Leishmania species and Trypanosoma species can also cause human diseases. A species that occurs in honey bees, Crithidia mellificae Langridge and McGhee, primarily occurs in the rectum of its host, but little is known about the effect on infected individuals (Schwarz et al. 2015). There was little research on C. mellificae since its first description by Langridge & McGhee (1967), until Runckel et al. (2011) revealed the presence of C. mellificae in commercial beekeeping operations transporting bees among Mississippi, California, and South Dakota. The lack of research on C. mellificae is surprising because the related C. bombi Gorbunov is known to have serious effects on bumble bee health. For example, Bombus (Hymenoptera: Apidae) colonies that were founded by infected queens were shown to be 40% less fit, displaying smaller colony sizes and producing fewer reproductive males (Brown et al. 2003). Bumble bee workers infected with C. bombi show increased handling time while foraging for nectar, visiting fewer flowers than their uninfected counterparts in timed studies (Otterstatter et al. 2005).

Ravoet et al. (2013) found that, in addition to Varroa destructor Anderson and Trueman, the presence of C. mellificae and the microsporidian Nosema ceranae (Fries et al.) are predictive markers of winter mortality for honey bees in Belgium, but little else has been documented on C. mellificae's effects on honey bees. Efforts to characterize the effects of this pathogen have also been complicated by taxonomic confusion among the difficult-to-identify trypanosomatids. In 2015, a new species of trypanosome parasite of honey bees was identified, Lotmaria passim Schwarz, from honey bee colonies in Maryland, USA (Schwarz et al. 2015). Since its description, L. passim has also been detected in Belgium, Japan, and Switzerland (Ravoet et al. 2015). However, much of the past literature on trypanosomatid infections has assumed that C. mellificae was the only trypanosomatid species afflicting honey bees (Schwarz et al. 2015).

Schmid-Hempel & Tognazzo (2010) developed a small subunit (SSU) primer set that will amplify, via polymerase chain reaction (PCR), DNA from Crithidia species, and that will work for L. passim as well other trypanosomes. Recently, Ravoet et al. (2015) developed a molecular diagnostic technique to distinguish L. passim from C. mellificae using amplicon size polymorphisms in the ribosomal DNA first internal transcribed spacer region (ITS1). Although this technique allowed for species-level identification of some infections, less than 30% of their samples were successfully amplified (Ravoet et al. 2015), inspiring the development of a more universally robust method. The purpose of this study was to develop a molecular diagnostic technique to distinguish L. passim from C. mellificae using a PCR primer set specific for L. passim and Crithidia species in a multiplex PCR with the highly conserved nuclear SSU gene.

Honey bees were collected from 2013 to 2014 from the Hawaiian Islands of Hawaii (Big Island), Lanai, Kauai, Molokai, Oahu, and Maui, as well as from American Samoa. Samples were preserved in 70 to 100% ethanol. DNA was extracted from individual honey bees by using a salting-out protocol with in-house reagents (Sambrook & Russell 2001). Samples of C. bombi and C. expoeki Schmid-Hempel and Tognazzo were obtained from infected bumble bees, with trypanosomatid identity verified through sequencing (GenBank accession numbers KU937107 and KU937108, respectively). Positive controls for C. mellificae and L. passim were obtained from type strains (30254 and PRA-422, respectively) deposited at the American Type Culture Collection (ATCC, Manassas, Virginia), extracted with the same methods as for the honey bee samples.

Amplification of trypanosome DNA was done using the SSU PCR primers CB-SSUrRNA-F2 and B4 (Schmid-Hempel & Tognazzo 2010), which yield an approximately 716 to 724 bp product. A volume of 2 mL of extracted DNA was used for PCR, and the remainder of the reaction mixture followed Taylor et al. (1997). The PCR temperature profile followed Schmid-Hempel & Tognazzo (2010). Amplicon verification was conducted by gel electrophoresis using 2% agarose gels, and PCR products were visualized using a BioDoc-it™ Imaging System (UVP, Inc., Upland, California). Positive samples were purified and concentrated with VWR centrifugal devices (VWR, Radnor, Pennsylvania) and sent to Eurofins Genomics (Huntsville, Alabama) for direct sequencing in both directions.

Sequences were aligned visually (GenBank accession numbers KU499926, KU499927), and a BLAST search (National Center for Biotechnology Information) was conducted with Geneious 6.0.3 (Auckland, New Zealand). Additional sequences were downloaded from GenBank for molecular phylogenetic analysis to confirm that the positive samples were L. passim (Fig. 1). Bayesian phylogenetic analysis was conducted with the MrBayes (Ronquist & Huelsenbeck 2003) plug-in within Geneious with 100,000 burn-in and 1,000,000 replications using a GTR+G model based on Akaike's Information Criterion results from jModelTest v2.1.3 (Darriba et al. 2012). Leishmania donovani (Laveran and Mesnil) (GenBank GQ332356) was used as the outgroup taxon. Two samples of Hawaiian honey bees, one from Maui and the other from Kauai were L. passim based on the molecular phylogenetic analysis. These samples formed a common clade with L. passim sequences from Belgium (GenBank KM066236) and Maryland, USA (KJ713376).

Fig. 1.

Bayesian molecular phylogenetic tree showing relationship of 2 Hawaiian Lotmaria passim positive samples relative to other trypanosomes from Gen-Bank for a 608 bp region of the rDNA SSU gene.

Using DNA sequences of L. passim from Hawaiian honey bees, along with sequences of Lotmaria and Crithidia species from GenBank (Fig. 1), PCR primers were designed to only amplify L. passim or Crithidia species by aligning the DNA sequences using Geneious software. A region of the alignment was found that had nucleotide polymorphisms between L. passim and Crithidia species. From this region, we selected the PCR primer L.passim18S-F, (5′-AGGGATATTTAAACCCATCGAAAATCT-3′), which has 4 conserved nucleotide differences between L. passim sequences and Crithidia species sequences, with 3 nucleotide differences occurring at the 3′ end of the primer. The primer yields a 499 bp product in conjunction with primer CBSSU rRNA B4. A second PCR primer was designed to only amplify Crithidia species. This primer, designated as C. mel 474-F (5-TTTACGCATGTCATGCATGCCA-3″), has 4 nucleotide differences relative to L. passim, with 2 of them due to a ‘TG’ insertion in L. passim that does not occur in Crithidia. This PCR primer combined with CBSSU rRNA B4 results in a 245 bp amplicon. Both the L.passim18S-F + CBSSU rRNA B4, and C. mel 474-F + CBSSU rRNA B4 amplicons were confirmed by DNA sequencing of 3 positive samples for L. passim and 1 for C. mellificae. Multiplex PCR was done as outlined above with a PCR temperature profile consisting of holding the samples for 2 min at 94 °C, then 40 cycles of 94 °C for 45 s, 55 °C for 1 min, and 72 °C for 1 min, followed by a final extension of 72 °C for 5 min.

The multiplex PCR with primers CBSSU rRNA P2, CBSSU rRNA B4, L.passim18S-F, and C. mel 474-F yielded a 716 to 724 bp product for L. passim and Crithidia species, a 499 bp product for only L. passim, and a 245 bp product for Crithidia species (Fig. 2). The multiplex PCR diagnostic for L. passim was evaluated on 96 worker honey bees collected from 6 Hawaiian Islands, and 6 honey bees collected from American Samoa. In total, 14 samples from Hawaii (Big Island), Kauai, Lanai, Maui, and Molokai, and 1 sample from American Samoa were positive for L. passim, and none were positive for Crithida species based on the molecular diagnostic technique.

Fig. 2.

Multiplex PCR gel showing the 716 to 724 bp amplicon for Lotmaria passim and Crithidia species, the L. passim specific 499 bp amplicon, and the Crithidia specific 245 bp amplicon.

We thank the numerous beekeepers in the state of Hawaii along with Stacey Chun and Craig Kaneshige for assisting with sample collection. We also thank Mark Schmaedick for collecting samples from American Samoa. This research was supported in part by the University of Arkansas, Arkansas Agricultural Experiment Station and the Hawaii Department of Agriculture Apiary Program.