Journal of Invertebrate Pathology 90 (2005) 98–103 www.elsevier.com/locate/yjipa PCR diagnostic methods for Ascosphaera infections in bees R.R. James a,¤, J.S. Skinner b a USDA-ARS Bee Biology and Systematics Laboratory, Logan, UT, USA b Department of Horticulture, Oregon State University, Corvallis, OR, USA Received 19 May 2005; accepted 5 August 2005 Abstract Fungi in the genus Ascosphaera are the causative agents of chalkbrood, a major disease aVecting bee larval viability. IdentiWcation of individual Ascosphaera species based on morphological features has been diYcult due to a lack of distinguishing characteristics. Most identiWcations are based on the size and shape of the ascomata, spore balls and conidia. Unfortunately, much overlap occurs in the size of these structures, and some Ascosphaera species will not produce sexual structures in vitro. We report a quick and reliable diagnostic method for identifying Ascosphaera infections in Megachile bees (leafcutting bees) using PCR markers that employ genus-speciWc primers for Ascosphaera, and species-speciWc primers for species known to be associated with Megachile spp. Using these methods, species identi- Wcations can be performed directly on bees, including asymptomatic individuals. Furthermore, the PCR markers can detect co-infections of multiple Ascosphaera species in a single host. We also identiWed a marker for Ascosphaera apis, the predominant cause of chalkbrood in Apis mellifera, the honey bee. Our diagnostic methods eliminate the need for culturing samples, and could be used to process a large num- ber of Weld collected bee larvae. 2005 Elsevier Inc. All rights reserved. Keywords: Apis mellifera; Ascosphaera; Bees; Chalkbrood; Disease diagnosis; Megachile rotundata; PCR 1. Introduction morphological characteristics, and this placement is sub- stantiated by sequence analysis of the ribosomal gene The alfalfa leafcutting bee, Megachile rotundata (Hyme- region (Berbee and Taylor, 1992). Anderson et al. (1998) noptera: Megachilidae), is managed for the pollination of proposed a phylogeny of 20 Ascosphaera spp. based on the alfalfa seed crops. These bees nest in holes drilled or DNA sequence of this same region. molded into wood or polystyrene blocks. M. rotundata is The Ascosphaera have proved to be a diYcult group to highly susceptible to chalkbrood, a disease caused by fungi identify due to a lack of distinguishing characteristics, and in the genus Ascosphaera (Ascomycota: Plectomycetes: some species are diYcult to culture, requiring specialized Ascosphaearales), and this disease is prevalent enough in growth media or conditions. Most identiWcations are based the U.S. that most alfalfa seed growers rely on purchasing on the appearance of the ascomata in vivo, and the size and new bees each year to meet their pollination needs. shape of the spore balls and conidia. Unfortunately, much To date, 22 species of Ascosphaera have been identiWed overlap occurs between species in the size of these struc- (Bissett et al., 1996; Youssef and McManus, 2001), and all tures. Molecular methods for identifying some species have of these are found only in association with bees, either as a been previously developed, including isozyme analysis pathogen or as a saprophyte on the pollen stores in nests. (Maghrabi and Kish, 1985, 1987, 2001) and RAPD-PCR The Ascosphaera are placed in the Plectomyces based on (Lu et al., 1996). However, both of these techniques require either pure cultures of spores, or nearly pure cultures with little or no bee tissue. The pathogens must be isolated and * Corresponding author. Fax: +1 435 797 0461. grown in culture, or the infected bees must be symptomatic E-mail address: [email protected] (R.R. James). cadavers that consist primarily of fungal tissues. Thus, it 0022-2011/$ - see front matter 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.jip.2005.08.004 R.R. James, J.S. Skinner / Journal of Invertebrate Pathology 90 (2005) 98–103 99 would be diYcult to use this method to identify infections All fungal species were cultured on Saboraud dextrose in bees that are still alive and asymptomatic. In addition, agar and potato dextrose agar, except Ascosphaera the results from such analyses are confounded if co-infec- aggregata, which was cultured on V-8 agar (James and tions occur in a single host, a condition that is likely to arise Buckner, 2004). V-8 Agar consists of 50 ml low sodium (Johnson et al., 2005). V-8 Juice (Campbell Soup, Camden, NJ), 5.3 g maltose, Polymerase chain reaction (PCR) techniques utilize highly 1.6 g yeast extract, 0.11 g MgSO4, 0.04 g thiamine, 1.1 mg speciWc primers and amplify only target DNA sequences in biotin, 7.5 ml Graces Medium (Invitrogen, Grand samples contaminated with other DNA. PCR markers could Island, NY), 0.5 ml canola oil, 0.5 ml 0.1% Triton X-100, provide a quick and reliable diagnostic method for Ascosph- and 1.5 g agar, and enough deionized water to bring the aera infections. Primers developed for the diVerent species of volume to 100ml. The pH was adjusted to 6.0 before Ascosphaera, could also be used to identify co-infections. We autoclaving. report here on PCR diagnostic methods to identify the Ascosphaera species that occur in association with Megachile 2.2. DNA extraction and PCR conditions bees, and for A. apis, the predominant cause of chalkbrood in Apis mellifera, the honey bee. DNA was extracted from pure Ascosphaera cultures, uninfected bee larvae, and larvae infected with A. aggregata 2. Methods using Ultra Clean plant DNA isolation kits (Mo Bio Labo- ratories, Solana Beach, CA). PCR analyses were conducted 2.1. Source of fungal cultures in 25 l reactions consisting of 19.9 l reverse osmosis puri- Wed water, 2.5 l of 10£ buVer, 0.5 l of 10 mM dNTP mix, We obtained pure cultures for eight species of Ascosph- 0.5 l of each primer (at 20 M each), 1 l DNA, and 0.1 l aera (Table 1), but were unable to obtain cultures of A. col- Taq DNA polymerase (5 U/l, Eppendorf HotMaster Taq, umbrina, A. variagata, and A. asterophora, which are also Brinkmann Instruments, Westbury, NY). Primer sequences recorded from Megachile hosts. These 11 species comprise are listed in Table 2. Thermocycle conditions were: an ini- all the recorded species to occur in Megachile spp. or Apis tial denaturing for 10 min at 94 °C; 30 cycles of 45 s dena- mellifera. In addition, we obtained Eremascus albus and turing at 94 °C, 45 s annealing at 62 °C, and 1 min extension Chrysosporium farinicola, due to their close relationship to at 72 °C; followed by a Wnal extension for 5 min at 72 °C. the Ascosphaera. For both of these, the strains obtained PCR products were electrophoretically separated on stan- were the type specimen. C. farinicola is the anamorph of dard agarose gels. Bettsia alvei, a fungus that grows saprophytically on bee pollen provisions. Eurotium, Monascus, and Sordaria cul- 2.3. Primer design tures (Table 1) were also obtained to verify that our speciWc primers would not amplify DNA from other common, The primers used here were all derived from conserved unrelated fungal species. regions of the DNA sequences reported by Anderson et al. (1998) for the 5.8S ribosomal DNA and the internal tran- scribed spacers (ITS 1 and ITS 2) using the program Gene- Table 1 Runner (Fig. 1). Fungal isolates used for testing genus- and species-speciWc primers Species Strain Source 2.4. Testing primers on pure fungal cultures Ascosphaera aggregata Wild1 Isolated from Megachile rotundata, Logan, UT To test the speciWcity of the primer pairs that we wanted a Ascosphaera aggregata 690 ARSEF to use as diagnostic markers, PCR reactions were con- Ascosphaera acerosa 201316 ATCCb Ascosphaera apis 13785 ATCC ducted for each primer using DNA extracted from every Ascosphaera apis 13786 ATCC one of our fungal cultures. Ascosphaera atra 693 ARSEF To test the sensitivity of the PCR reactions, DNA was Ascosphaera atra 5147 ARSEF extracted from A. aggregata (wild1) and A. acerosa X Ascosphaera ava 5144 ARSEF (ATTC 201316), then diluted to 116 ng/ml. The DNA con- Ascosphaera larvis 262708 ATCC X X Ascosphaera proliperda 696 ARSEF centration was determined using a uorometer (Pico uor Ascosphaera proliperda 28358 ATCC Handheld, Turner Designs, Sunnyvale, CA) and Pico- Ascosphaera pollenicola 62712 ATCC Green (Molecular Probes, Eugene, OR), a Xuorescent dye Eurotium cheraleri B. Kropp, Utah State University with speciWcity to double stranded DNA. A series of Wve- Chrysosporium farinicola 18053 ATCC fold dilutions were prepared in 0.1% TE buVer to obtain Eremascus albus 11665 ATCC Monascus perperaseens B. Kropp, Utah State University solutions of 23.2, 4.64, 0.93, 0.185, and 0.037 ng DNA/ml. Sordaria spp. B. Kropp, Utah State University PCR analyses were conducted using 1 l of each dilution a USDA Agricultural Research Service Entomopathogenic Fungi Col- as template DNA to assess the minimum concentration lection, Ithaca, NY. that would amplify. The sensitivity analyses were repeated b American Type Culture Collection, Manassas, VA. three times. 100 R.R. James, J.S. Skinner / Journal of Invertebrate Pathology 90 (2005) 98–103 Fig. 1. Relative location of each primer on the ITS1–5.8S–ITS2 genomic region, and the Ascosphaera species that each primer will amplify. All species are ampliWed by the TW81/AB28 and AscoAll1/AscoAll2 primer pairs. Species in italics were not tested, but ampliWcation is predicted based on the DNA sequences. Table 2 symptomatic by the prepupal stage, and so asymptomatic Primers used as markers for Ascosphaera spp. detection prepupae were considered to be free of infection. IdentiWer Forward Primer sequence (5Ј–3Ј) Fifty A. aggregata-infected and 50 healthy prepupae or reverse were removed from the stored boards, individually surface TW81a F GTT TCC GTA GGT GAA CCT GC cleaned by submerging in a 1% chlorine solution (sodium AB28a R ATA TGC TTA AGT TCA GCG GGT dichloro-s-triazinetrione dihydrate, Quantum Biochemical, AscoAll1 F GCA CTC CCA CCC TTG TCT A Alpharetta, GA) for 3–7 min, followed by three rinses in AscoAll2 R GAW CAC GAC GCC GTC ACT 1a-F2 F GGA AAA GAC CCT CGA CGA G sterile, distilled water.