:1 of Ap: ::Itorol Rereo'd: or:d Bc World 47(3): 194-201 (2008) BRA 2008 DOl: 0.3827/IBRA. I .47.3.07 A l RESEARCH ARTICIll I B R A Detection and characterization of Kodamaea ohmeri associated with small hive beetle Aethina tumida infesting honey bee hives. Nicole D Benda", Drion Boucias°, Baldwyn Tort0 3 and Peter Teal'. LJDA ARS, Center for Med cal, Agricultural, and Veterinary Entomology, 1600-1700 S.W. 23rd Drive, Ca new lie, FL 32608, USA. n titute of Food and Agricultural Sciences, Department of Entomology and Nematology, University of Florida, 110620, Gainesville, F! 3261 1, USA. International Centre of Insect Physiology and Ecology, P0. Box 30772-00)00, Nairobi, Kenya. Received 20 November 2007, revised manuscript received 8 April 2008, accepted for publication 8 April 2008. Corresponding author. Email: Nicole.Benda©ars.usda.gov Summary Honey bee colony infestation by the small hive beetle (SHB) is associated with fermentation of hive materials. Pollen, beetles, and robbing bees (ten of each) were collected from hives infested with SHB in both Florida and Kenya. Plating of homogenized bodies of beetles and bees and comb swabs resulted in smooth cream-colored yeast colonies that formed pseudomycelial cells as they aged. Fatty acid profiles of yeast isolates from Florida and Kenya most closely matched the profiles of Candida krusei and C. sake, respectively. However, the DNA sequence of the 28S and 5.8S-ITS2 of both the Florida and Kenya isolates were, 99-100 % homologous to Kodamaea ohmeri. The ITS I region differed between the two geographic strains. The two strains produced similar volatile profiles which were attractive to SHB and contained compounds also found in honey bee alarm pheromone. Detección y caracterización de Kodarnoea o/irneri asociada a la infestación de colmenas de abejas por el pequeño escarabajo de las colmenas Aethina turn/do. Resu men La infestación de las colmenas de abejas por el pequeno escarabajo de las colmenas (PEC) está asociada a la fermentación de los materiales de las colmenas. Se recolectó polen, escarabajos y abejas pilladoras de colmenas infestadas con PEC en Florida y Kenia. El cultivo en placas de homogeneizados del cuerpo de los escarabajos y de las abejas, y de frotis de los cuadros resultó en colonias de levaduras de color crema que formaron células pseudomiceliales cuando crecieron. El perfll de ácidos grasos de los aislados de levaduras de Florida y Kenia se aproximó mucho al de Candida krusei y C. sake, respectivamente. Sin embargo, las secuencias de ADN de las regiones 28S y 5,8S-ITS2 de los aislados de Florida y Kenia fueron 99- 100% homologas a Kodamaea ohmeri. La region ITS I fue diferente segUn su origen geográfico. Las dos variedades produjeron perflles volátiles similares que son atractivos para el PEC y contenIan compuestos que también se encuentran en la feromona de alarma de las abejas. Keywords: Aethina tumido, small hive beetle, Apis mellifera, FKodamaea ohmerl, yeast, volatile attraction 195 Yeast of SHB / honey bee association Introduction Materials and Methods Yeasts occupy a wide variety of habitats where sugars are available, such as flowers, insect honeydew, and certain plant material. Insects present in these habitats may acquire (and vector) the various yeasts. Yeasts have been found in association with a broad range of insects, including lacewings, bees, and beetles. In many cases, consistent yeast / insect associations are observed over a large geographical region. The insect appears to benefit nutritionally from these yeasts, and it has been suggested by Suh et al. (2005) that their presence allows insects to survive on nutrient-limited substrates. Yeasts have commonly been found in the digestive ti-act of sap beetles (family: Nitidulidae) that feed on flowers, decaying fruit, fermenting plant juices, and mushrooms (Lachance and Bow/es, 2002; Suh et al., 2004; Suh, et al., 2006). The yeasts associated with flower-feeding nitidulids, including species of Metschnikowia, Wickerhamiella, and Kodamaea (Lachance and Bowles, 2002: Lachance et 0/., 1999; Lachance et al., 1998), are distinct from those yeasts (various Candida species) associated with mushroom feeding sap beetles such as Pa//odes spp. (Suh et al., 2006). Significantly, yeast-associated fermentation volatiles attract certain nitidulids (Nout and Bartelt, 1998; Williams et 0/., 1992) and serve as oviposition cues for these beetles (Phelan and [in, /991). Recently, the small hive beetle (SHB) ,Aethina tumida (Coleoptera: Nitidulidae), a nest parasite of honey bees native to Africa, has invaded the United States and Australia (Neumann and Elzen, 2004). The European honey bee used for pollination in the USA is particularly susceptible to SHB. The beetles feed on the pollen, brood, and honey, and infestations destroy the colony. Fermentation of honey in the comb is associated with mass SHB infestation. From initial sampling of infested hives and SHB, we isolated a yeast species and identified it as Kodamaea ohmeri. The yeast grows prolifically on honey bee combs infested with SHB and this growth produces volatiles attractive to SHB (Torto et al., 2007). Interestingly, the volatiles overlap with those identified as alarm pheromone components of honey bees. It is unknown whether SHB or the honey bee is responsible for the introduction of these yeasts into the hive. Because yeasts are commonly found in flowers, bees, and nitidulids, and because either can become contaminated during the common behavioural interactions between honey bees and SHB, either or both routes of entry are feasible. Reliable procedures to detect and identify K. ohmeri in various biological samples are necessary to further define the interactions of this yeast with SHB and honey bees. Herein, we describe how K. ohmeri was detected in samples from hives of both European and African honey bees and associated SHB and identified using a combination of morphological, molecular, and chemical characters. Isolation and cultivation of the yeast in infested hives Yeast samples were collected from SHB from four European honey bee hives maintained at the Centre for Medical, Agricultural and Veterinary Entomology USDA-ARS (Florida), and from four African honey bee hives maintained at the International Centre of Insect Physiology and Ecology (ICIPE) (Kenya). SHB larvae and adults were removed from an infested hive and surface-sterilized by immersion in 70% ethanol for —30 seconds followed by two rinses in sterile water. Homogenates of washed beetles were prepared in sterile water and resulting suspensions streaked for isolation on Sabouraud Dextrose Agar plus 1% yeast extract (SDAY). In addition to insect homogenates, a series of honey and pollen samples were collected from the infested hive and inoculated on plates. Sampling tools (metal spatula, forceps, micropestle) were sterilized with alcohol between individual samples. Inoculated plates were incubated at 28°C for 1-3 days. Individual yeast colonies were selected and sub-cultured on SDAY and incubated at 28°C. Selected isolates were inoculated into Durham tubes containing autoclaved bee pollen broth (1% aqueous pollen) and incubated at 3°C for five days.The gasproducing [-27 isolate produced a colony morphology characteristic of the majority of yeast colonies observed in the initial isolation.To typify growth on different media types, the [-27 isolate was plated on pollen agar (1% pollen plus 1.5% agar), Lee's agar, Czapek-Dox broth, malt extract agar, M40Y agar (a high sucrose, osmotic-stress medium), and moistened autoclaved pollen. Inoculated plates were incubated at 28°C for 7-10 days. Fatty acid profile typing Selected isolates representative of the major colony phenotype detected from both larval and adult SHB homogenates were subcultured on SDAY (two isolates each from Florida and Kenya hives). Cells were harvested (80 mg wet weight) and treated chemically to extract and convert the fatty acids present in the cell wall or cell membrane fractions to fatty acid methyl esters (FAMEs). Total cellular fatty acid methyl esters were subjected to gas chromatographic (GC) analysis. Resulting profiles were analyzed with the Microbial Identification System (MIDI Inc.) that matched the test organism's profiles to the available yeast FAME databases using Sherlock Version 4.5 software (Anonymous, 1993). Similarity indices were calculated: indices of 0 to 0.3, 0.3 to 0.6, and 0.6 to 1.0 were considered as poor, possible, and excellent matches, respectively. DNA extraction and PCR amplification Eight isolates from the Florida hives and two isolates from the Kenya hives were inoculated in SMY (sabour and maltose with yeast extract) broth and incubated overnight at 26°C. Cells were pelleted and DNA was extracted using the Masterpure'° Yeast DNA purification kit (Epicentre, Madison, WI). The quantity and quality of the DNA preparations were evaluated using ethidium bromide-stained agarose gels. Aliquots of DNA were amplified with a mixture of Taq DNA polymerase (Promega) and PFU polymerase (Stratagene), using the primers 181 and AB28 for the ITS-5.8S (Curran, et 0/., 994) and NL- I and NL-4 primers for the 5' divergent domain (the Dl /D2 region) of the LSU 196 Benda, Boucias, Torro, Teal rDNA (Kurtzman and Robnett, 1997). PCR products were extracted using QlAquick PCR extraction kit (Qiagen) and each PCR-amplicon bidirectionally sequenced using an ABI Prism DNA Sequencer at the Interdisciplinary Center for Biotechnology Research Core Facility at the University of Florida, Gainesville, FL. DNA sequences were compared to those deposited in GenBank with BLAST (blastn) using the default settings. Volatile production, collection, and analysis To compare the yeast strains collected in Florida and Kenya, beecollected pollen (20 g) was inoculated with the L-27 clone from a Florida hive and an A-I clone from the Kenya hive. To compare the effect of substrate, three different sterilized media (20 g each) were inoculated with the L-27 strain (collected from SDAY plates, diluted in sterile water): I) moistened pollen, 2) moistened commercial pollen substitute BeePro ® (Mann Lake; Hackensack, MN, USA) and 3) SDAY (Sabouraud Dextrose Agar plus 1% yeast extract). Inoculated and non-inoculated plates were sealed with Parafllm® and incubated at 28°C for 7 days. After incubation, the Parafllm® seals and lids were removed and each plate was placed in a glass aeration chamber (46 cm long x 19 cm wide) (Torto et al., 2007). Volatiles were collected by passing charcoal-filtered and humidified air at 0.5 ]/minute over the plate in the aeration chamber and then through a Super-Q adsorbent trap (30 mg, Alltech; Nicholasville, KY USA), which trapped the volatiles, for I hour at room temperature. Chemicals were eluted from traps with 150 p1 methylene chloride and 174 ng of butyl butyrate were added as an internal standard to 40 p1 of the extract. Samples (I p1) were analyzed by gas chromatography on a HP 6890 equipped with a HP-I column (30 m x 0,25 mm ID x 0.25 pm;j &W Scientific; Folsom, California, USA) linked to a HP 5973 mass spectrometer using electron impact mode (70 eV, Agilent; Palo Alto, CA, USA), with helium as the carrier gas. Samples were injected in splitless mode at an injector temperature or 240°C, and a split valve delay of 0.5 minutes. The oven temperature was held at 35°C for I minute, then rorsc.:l at 10°C / minute to 230°C and held at this temperature for IC minutes. The ion source temperature was 230°C. Volatile compounds were identified by comparison of their chromatographic retention times and mass spectra to those of commercially available standards analyzed on the same instrument. CFU!l0 mg pollen. Interestingly, the robber bees (N= 10) collected from SHB damaged hives contained yeasts at levels that ranged from 150 to more than 10 CFU / bee. Isolates derived from isolation plates grew on various mycological media as well as Lee's agar, pollen agar, and moistened pollen. Scant growth was observed when these cultures were plated on either Czapeks-Dox or M40Y agar plates. Young —4 day old SDAY cultures appeared as uniform creamcolored, smooth colonies (Fig. IA). These colonies contained globose to ellipsoidal multi-polar budding yeast cells, ranging from 2-7 pm in diameter, with the majority of cells under 4 pm in diameter (Fig. 2B). As the cultures aged, pseudomycelial cells formed, changing the colony morphology (Fig. I B). Within five days post-inoculation, crenellated colonies with an undulate surface formed on SDAY plates. The pseudomycelium consisted of chains of cylindrical cells 2-3 pm in diameter (occasionally swollen larger) by 9-20 pm in length (Fig. 2C) Small raised scars Smooth colony phenotype (A) of the L-27 isolate incubated for 3 days at 26°C on SDAY Differential interference contrast (DIC) micrograph of cells (B) harvested from smooth colony. Note presence of numerous ovoid to oblong shaped yeast cells. Bar represents 10 microns. Fig I. Results ' t\ Yeast prevalence and morphology Preparations of larval and adult SHB, comb material (pollen, honey), and associated robber bees (from heavily infested hives) produced a lawn of uniform yeast colonies within 36 h when plated on SDAY Spot dilution plating of these preparations demonstrated that the number of yeast colonies varied among the different samples. All of the SHB larvae (N= 10) sampled from heavily infested hives consistently harboured yeasts (I 03 101 colony forming units or CFU I larvae) whereas 8 of 10 adult SHB harboured fewer yeasts (10- 100 CFU / adult). The presence of yeast in pollen samples was variable and ranged from 0 to 100 , — ut t C 'C- '' e n . L , •— c " Fig 2. The \ :' C- -- ' t', - - L-27 cultures after 7 days on SDAY at 26°C produce a matte-like, wrinkled colony phenotype (A, B). DIC images of the cells (C, D) reveal the presence of elongate hyphal bodies intermixed with round oblong cells. In certain field, the base of these cells appeared to be swollen (D). Bars represent 15 microns. 197 Yeast of SHB / honey bee association Summary of FAME profiles of the Florida ([-28) and Kenya (A- I) yeast isolates. Note the difference in library match between these geographical isolates. RI retention time. Table I. Yeast L-28 RT 1.652 1.753 7.185 8.715 9.995 0.067 10.37 11.733 3.364 13.457 13.846 14.228 6.055 Peak Name SOLVENT PEAK Percent 14:0 15:0 16:1 Cis 7 (w9) 16:1 Cis 7 (co7) 16:0 17:1 Cis 9 ((o8) 18:2 CIS 0.65 0.21 0.39 5.32 13 0.98 9,12118:0a 20.02 Summed Feature 8* 18:0 Summed Feature 10* 18:0 20H 13.815 57.82 0.98 13.906 14.301 14.674 Peak Name SOLVENT PEAK Percent 14:0 16:1 Cis9(w7) 6:0 17:1 Cis9(w8) 18:2 CIS 9, 12/18:Oa Summed Feature8* 18:0 Summed Feature 10* 1.49 10.09 6.3 0.91 25.29 43.15 1.65 1.12 0.35 0.27 Library Results 0704, Candida krusei YST28 krusei krusei* 3.80 YSTCLN 3.80 Yeast A-I RT 2.323 2.583 2.646 7.705 10.546 10.846 12.2 Library Results 0763, Candida sake YST28 3.80 0.828, Candida krusei *Summed features consist of the following fatty acids, which could not be separated. Summed Feature 8: 18:1 CIS9 ((09) and/or I 8: I (w8); Summed Feature 10: I 8: I Cis 9 DMA and/or an unknown (ECL I 8.21 8). on pseudomycelial cells indicated the prior formation of yeast cells and / or the prior points of attachment for branches. Adjacent cells were separated by constrictions, not by formal septal structures. The K. ahmeri strains isolated from the hives reproduced asexually by multilateral budding. Attempts at crossing different isolates on MEA malt extract agar plates did not result in ascospore formation. Fatty acid methyl ester (FAME) analysis MIDI anaysis associated both the Florida (L-27, L-28) and Kenya (A- 1, A-2) isolates with species in the genus Candida (Table I). Fatty acid profiles generated from cultures of the L-27 and [-28 strains were most similar to the profiles of the Candida krusei reference (similarity indices 0.704 and 0.828). The profiles of Kenya isolates A- I and A-2 were most similar to the profiles of C. sake and C. va/ida (similarity index 0.763 and 0.62 I). DNA analysis PCR-amplification of the 28S (D 102) and ITS 1-5.8S-ITS2 of the Florida L-27 isolate produced 509 bp and 377 bp fragments (Fig. 3). The 28S and ITS I -5.8S-ITS2 DNA sequences for the [-27 isolate have been submitted to the GenBank database and assigned accession numbers AY9 11384 and AY9 11385 respectively. The ITS I 5.8S-ITS2 DNA sequence for the A-I isolate has also been submitted to GenBank (accession number of both the EU569326). BLAST searches (Altschul et al., 1997) 1 2 3 4 5 6 7 3. Eth dium bromide stained gel of the Florida [-27 (lanes 2, 3), and the Kenya A-I (lanes 4,5) and A-2 (lanes 6, 7) amplified with primers for the DI /D2 region of the 28S rDNA (lanes 2, 4, 6) and the ITS I-5.8S-ITS2 rDNA (lanes 3, 5, 7). Lane I contains molecular weight markers (100-1000 bp). Fig partial 28S and ITS I-5.8S-ITS2 sequences produced matches with extremely low Expect (E) values. The 509 bp D I /D2 sequence was identical to the culture established from the type strain of Endamycapsis ohmeri ( Kodamaea ohmeri) (GenBank U45702), as well as other strains of K. ahmeri (AF335976, AY26782 I, AY267824), and to an unidentified yeast (AF335975). Since 98 Benda, Boucias, Torto, Teal L- 27 a-AD 28 A-la-AB28 TGTTTCCGTAGGTGCC TGCGGAAG GATCATTCATAATATTC TTACACAC TGTTTTT 60 TGTTTCCGTAGGTGAACCTGCGGAAGGATCAT TAACATTAATTTACTACACACTGTTTTT 60 ************************************** * ** ************** L-27a-AB28 A-la-AB28 TTACAACAAAACAAACATATCTAATCTATAAAT CTACGTTTTAAAATTCTTAAAACTTTC 120 TTAAAACTTTC 118 *************** ******* * ** *** L-27a-AB28 A-la-AB28 CAACGGATCTCTTGGTTCTCGCATCGATGGA ACGCAGCGATGCGATACGTAATA 180 CAACGGATCTCTTGGTTCTCGCATCGATGGA ACGCAGCGATGCGATACGTAATA 178 ************************************************************ L-27a-AB2S A-la-AB28 CGAATCGCAGCTCTCGGAATCATCGAATCTTTG CGCACATTGCACCATTGGGTATTCC 240 CGAATCGCAGCTCTCGGAATCATCGAATCTTTGAACGCACATTGCACCATTGGGTATTCC 238 **************************** ******************************** L-27a-AB28 A-la-AB28 CAATGGTATGCTTGTTTGAGCGAATACTTCCCTA ATCCTCACGGATTGTATTGTGTTTGC 300 CAATGGTATGCTTGTTTGAGCGAATACTTCCCTAA TCCTCACGGATTGTATTGTGTTTGC 298 **** ******************************************************** L-27a-AB28 A-la-AB28 ACGAAAATAATGACGACAGTACTCTACAAAACG GTACCGTCAGTACGCTCATTTTTTTTC 360 ACGAAAATAATGACGACAGTACTCTACAA CGGTACCGTCAGTACACTCATTTTTT 358 ********************************************** *********** * L-27a--AB28 A-la-AB2S CTC 355 TCA 361 Fig 4. Alignment of the ITS I -5.8S-ITS2 sequence of the Florida ([-27) and Kenya (A- 1)yeast isolates. The shaded region in the ITS2 region demonstrates the heterogeneity between the geographical isolates. The underlined sequence represents the small subunit 5.8S rDNA. U45702 is the type sequence, these results suggest that the yeast conserved 5.8S gene had 80% homology to available C. sake we found associated with SHB honey bee hive infestation is K. sequences. Furthermore, the Dl /D2 region of different C. sake ohmeri. Other similar sequences may be from misidentified 28S genes had little homology to the corresponding A I cultures. For example, although the two C. membronifociens sequence. sequences Aj508563 and AB304737 were identical to the 28S DNA sequence of L-27, the identities of these strains are suspect, Analysis of volatiles released from yeast as they do not match the C. membronifoc/ens type sequence Yeast samples from both the Florida and Kenya hives ([-27 and (U45792). Since the 28S sequences from the Kenya A-I and A-2 A- 1)released a complex ofvolatiles when grown on media isolates were identical to the 28S sequence of the Florida L-27 containing pollen. These included: 3-hydroxy-2-butanone, 2- ethyl strain, these isolates were also grouped with K. ohmeri. Because propionate, 3-methyl-butan- I -ol, 2-methyl-butan I -ol, ethyl-2both the 28S and ITS 1-5.8S-ITS2 DNA sequences of the different methyl propionate, 2-methylpropyl acetate, ethyl butyrate, ethylisolates were identical within country of origin, isolates of the 2-hydroxy propionate, isopentyl acetate, 2-methylbutan- I -yl Florida [-27 and Kenya A-I were deposited at the Agricultural acetate, 2-heptanone, ethyl pentanoate, 3-octanone, 2-octanone, Research Service Culture Collection (accession number NRRL 1ethyl hexanoate, ethyl 3-hexenoate, hexyl acetate, and ethyl 48473). heptanoate (Fig. 5). As shown in Fig. 5, the ratios of the Further confirming the identity of [-27 as K. ohmeri, the 5.8S compounds released by [-27 and A-I were essentially identical. sequence was 99-100% homologous to various K. ohmeri SDAY media inoculated with yeast released only 3-methyl-butan(AY 168786, AF2 19004, AF2 I 8977) and several unidentified yeast I -ol and 2-methyl-butan I -ol, and the amounts released were isolates (AF5362 I I, AF536209). The ITS2 and 5.8S regions and significantly lower than those released by yeast growing on media the short sequences of the flanking I 8 and 28S subunits of the containing pollen (Table 2). Interestingly, these same volatiles were Florida and Kenya isolates were highly similar (Fig. 4). However, collected from yeast growing on Bee-Pro plus SDAY media and the ITS region of the [-27 sequence was distinct from both non-inoculated Bee-Pro plus SDAY media. SDAY media alone Kenya isolates, varying by 13 base pairs and 2 deletions. produced none of the volatiles identified as being released by Interestingly, it was the Kenya ITS I sequences that had 100% yeast (Table 2). homology to available K, ohmeri BLAST database sequences. Comparing the results of the FAME and BLAST analyses, the Florida [-27 sequence of the highly conserved 5.8S gene had only 78% homology to available C. krusei sequences. Furthermore, the DI /D2 region of different C. krusei 28S genes (AY305680, AY305674) had little homology to corresponding sequences generated from the SHB yeasts. The Kenya A-I sequence of the Yeast of SHB 199 I honey bee association a, C 0 a a, 0 U a) a) 0 12.00 9.00 6.00 0 Time (min) Fig 5. Representative total ion chromatograms of Super Q volatile extracts of control and yeast-inoculated media. A) Sterilized bee-collected pollen control, B) sterilized bee-collected pollen inoculated with yeast (A- I) from larvae of the SHB from African honey bee colonies in Kenya, and C) sterilized bee-collected pollen inoculated with yeast (L-27) from larvae of the SHB from European honey bee colonies in Florida. I- 3-hydroxy-2-butanone, 2- ethyl propionate, 3- 3-methyl-butan- I -ol, 4- 2-methyl-butan- I ol, 5- ethyl-2-methyl propionate, 6-2-methylpropyl acetate, 7- ethyl butyrate, 8- ethyl-2-hydroxy propionate, 9- isopentyl acetate, 10- 2-methylbutan- I -yl acetate, I I - 2-heptanone, 12- ethyl pentanoate, 13- 3-octanone, 14- 2-octanone, 5- ethyl hexanoate, I 6butyl butyrate, the internal standard. ethyl 3-hexenoate, 17- hexyl acetate, 18- ethyl heptanoate, 19- unknown, 20- unknown. IS Volatiles collected from SDAY media either inoculated with yeast or not and containing various supplements. (+++) indicates high levels of compound; (++) indicates amounts less than 50% of +++, (+) indicates amounts less than 20% of +++ and (-) indicates no detectable amount. Table 2. Compound SDAY+ Yeast+ SDAY+ Yeast Pollen SDAY+ Yeast+ + Bee-Pro + + + + + + Pollen 3-hydroxy-2-butanone +++ + Eth yl propionate 3-methy1-1-butanol +++ ++ + 2-m ethyl -l-butanol EthyI-2-methylpropionate +++ 2-methylpropyl acetate Ethyl butanonate +++ Ipentyl acetate 2-methyl- I -butanol acetate +++ +++ ptanone +++ Ethyl pentanoate ++± 3-octanone +++ 2-octanone Ethyl-2-hydroxy propionate Ethyl hexanoate Ethyl-3-henenoate______ Hexyl acetate Ethyl heptanoate +++ +++ SDAY+ SDAY+ + + Bee-Pro SDAY 200 Benda, Boucias, Torto, Teal Discussion In general, the samples from honey bee hives heavily infested with yeasts (and other microorganisms) are found in healthy hives, SHB yielded a lawn of morphologically consistent yeast colonies. but only K. ohmer/ was present at the height of SHB infestation. Similar samples from healthy hives yielded several species of What prevents other yeasts from growing under these filamentous fungi and yeasts (N. Benda, unpublished). The circumstances is not understood and calls for investigation. predominant colony phenotype detected in the SHB hives (and identified as K. ohmeri) was similar to that of the K. ohmeri isolate described by Suh and Blackwell (2005). Failure to induce ascospore formation by crossing different isolates is consistent with observations by Suh and Blackwell (2005). These researchers found no ascospore formation when crossing various We thank E. R. Dickstein (Bacterial Identification and Fatty Acid K. ohmeri strains. The sequences AJ508563 and AB304737 were Lab, University of Florida) for conducting the MIDI analysis: the identical to the 28S DNA sequence of L-27, and therefore should UF ICBR Sequencing Facility: Dusti Purcell, Alfredo Platinetty, be assigned to K. ohmeri rather than C. membranifaciens. The Steve Willms (U.S. Department of Agriculture Center for problem is almost certainly the result of a nomenclatural Medical, Agricultural, and Veterinary Entomology), and Eluid Mull problem, but because there is no mechanism for third-party (International Center of Insect Physiology and Ecology) for annotation of GenBank entries, the original submitters have been technical assistance: and the advice of Cletus P Kurtzman notified and must be relied on to make the corrections. (Microbial Genomics and Bioprocessing Research, USDA-ARS). Isolates sampled from the SHB infested European honey bee hives and from the SHB infested African honey bee hives produced distinct FAME profiles that were matched with different yeasts. These results highlight strain differences between the isolates from Florida and Kenya. It should be noted that minor changes in nutrient inputs has been reported to modulate fatty ALTSCHUL, S F, MADDEN, T L; SCHAEFER A A; ZHANG,J H; ZHANG, Z; acid profiles (Kellogg et al., 1999). This, and the absence of large MILLER, W; LIPMAN, D (1997) Gapped BLAST and PSI-BLAST; a new generation of protein database search programs. Nucleic Acids Research numbers of insect-associated yeasts in the FAME database, may 25(17); 3389-3402. account for differences in the FAME profiles and in the ANONYMOUS (1993) Microbial Identification System operating manual (Version 4).. differences between the FAME and BLAST search results of the Microbial ID Inc.; Newark DE, USA. strains we isolated. The marked differences at sites in the ITS I CURRAN,j; DRIVER, F; BALLARD,) W 0; MILNER, R (1994) Phylogeny of region of the DNA sequence provide additional support of strain Metorhizium - analysis of ribosomal DNA-sequence data. Mycolog i cal Research differences between the Florida and Kenya isolates. 98; 547-552. Limited sampling has demonstrated an association between KELLOGG,) A; BANKERT D A; CHATURVEDI V (1999) Variation in Microbial Identification System accuracy for yeast identification depending on the yeast, SHB, and honey bee hive in two distant locations commercial source of Sabouraud dextrose agar)ournal of Clinical Microbiology (Florida, USA and Nairobi, Kenya). The yeast growing on pollen 37(6):2080-2083. sources produces bee alarm pheromones that serve to recruit KURTZMAN, C P; ROBNEUi C J (1997) Identification of clinically important additional SHB (Torto et al., 2007). Interestingly, the yeast ascomycetous yeasts based on nucleotide divergence in the 5' end of the produced these attractants only when grown on media containing large-subunit (26S) ribosomal DNA gene.)ournol of Clinical Microbiology 35(5); pollen and not on media containing the commonly used pollen 1216-1223. substitute Bee-Pro, which lacks pollen. Pollen collected by bees LACHANCE M A; BOWLES,) M (2002) Metschnikowio orizorsensis and Metschnikowio dekortorum, two new large-spored yeast species associated with may contain factors required for the yeast to produce the alarm florjcolous beetles, FEMS Yeast Research 2(2); 81-86. pheromones and other volatiles attractive to beetles. LACHANCE, M A; BOWLES, j M; STARMER, W T; BARKER, S F (1999) Kodomoeo j How the yeast initially enters the hive is not well understood. kokoduensis and Coridido tolerons, two new ascomycetous yeast species from Potentially, adult beetles originating from heavily damaged hives or Australian Hibiscus flowers. Conodion )ournol of Microbiology 45(2); 172-177. from some unknown substrate inoculates the healthy hives with LACHANCE, M A; ROSA, CA; STARMER, W T; SCHLAG-EDLER B; BARKER,] S F; yeast residues. Association of K. ohmeri with two other nitidulid BOWLES, J M (1998) Wickerhamiella oustroliensis, Wickerhamiella cocticolo, species has been documented (Suh and Blackwell, 2005). Other Wicker6omiello occidentolis, Condido drosophilae and Condida lipophila, five new yeasts may also be associated with SHB. The quantification and related yeast species from flowers and associated Insects. lnternotiarial)ournal of Systematic Bacteriology 48; 1431-1443. identity of these other yeasts and the quantification of K. ohmeri NEUMANN, P; ELZEN. P (2004) the biology of the small hive beetle (Aethirsa on SHB should be resolved. Although many insect-yeast tumido, Coleoptera; Nitidulidae); Gaps in our knowledge of an invasive species. associations have been identified, the type of interaction Apidologie 35(3): 229-247. (mutualism, commensalism, or even competitive) is rarely NOUTi M J R; BARTELT R (1998) Attraction of a flying nitidulid (Carpophilus understood. Understanding the relationship between SHB and humerolis) to volatiles produced by yeasts grown on sweet corn and a cornK. ohmeri will help in developing control strategies for SHB. The based mediumjournol of Chemicol Ecology 24(7): 1217-1239. attractiveness of yeast-inoculated pollen to SHB (Torto et al., PHELAN, P L; LIN, H C (1991) Chemical characterization of fruit and fungal volatiles attractive to dried-fruit beetle, Corpophilus hemipserus (L) 2007) indicates a relationship beneficial to the beetle. Other (Coleoptera, Nitidulidae).)ournal of Chemical Ecology 17(6); 1253-1272. Acknowledgements References Yeast of SHB I honey bee association SUH, S 0; BLACKWELL, M (2005) Four new yeasts in the Candela mesentenca dade associated with basidiocarp-feeding beetles. Mycologio 97( l): 67-177. SUH, S 0; MCHUGH, j V; BLACKWELL M (2004) Expansion of the Condida tanzowoensis yeast dade: 16 novel Condido species from basidiocarp-feeding beetles. International Journol of Systematic and Evolutionary Microbiology 54: 2409-2429. SUH, S 0; NGUYEN, N H; BLACKWELL, M (2005) Nine new Candida species near C. membronifociens isolated from insects. Mycological Research 109: 1045-1056. SUH, S 0; NGUYEN, N H; BLACKWELL, M (2006) A yeast dade near Cosdido kruisii uncovered: nine novel Condido species associated with basidioma-feeding beetles. Mycological Research 110: 1379-1394, TORTO, B; BOUCIAS, D G; ARBOGAST, RI; TUMLINSON, J H; TEAL, P E A (2007) Multitrophic interaction facilitates parasite-host relationship between an invasive beetle and the honey bee. Proceedings of the National Academy of Sciences of the United States of America 104(20); 8374-8378. WILLIAMS, RN; BLACKMER,J L; RICHMOND, D 5; ELLIS, M 5 (1992) Nitidulidae (Coleoptera) diversity in three natural preserves in Portage County, Ohio. Ohio Journal ef Science 92(4): 82-87. 201