1202 Chiang Mai J. Sci. 2018; 45(3) Chiang Mai J. Sci. 2018; 45(3) : 1202-1213 http://epg.science.cmu.ac.th/ejournal/ Contributed Paper Screening of Antagonistic Bacterial Isolates from Hives of Apis cerana in Vietnam Against the Causal Agent of American Foulbrood of Honey Bees, Paenibacillus larvae Sasiprapa Krongdang [a,b], Jeffery S. Pettis [c], Geoffrey R. Williams [d] and Panuwan Chantawannakul* [a,e,f] [a] Bee Protection Laboratory, Department of Biology, Faculty of Science, Chiang Mai University, Chiang Mai 50200, Thailand. [b] Interdisciplinary Program in Biotechnology, Graduate School, Chiang Mai University, Chiang Mai 50200, Thailand. [c] USDA-ARS, Bee Research Laboratory, Beltsville, MD, 20705, USA. [d] Department of Entomology & Plant Pathology, Auburn University, Auburn, AL, 36849, USA. [e] Center of Excellence in Bioresources for Agriculture, Industry and Medicine, Chiang Mai University, Chiang Mai, 50200, Thailand. [f] International College of Digital Innovation, Chiang Mai University, 50200, Thailand. * Author for correspondence; e-mail: [email protected] Received: 15 February 2017 Accepted: 20 June 2017 ABSTRACT American foulbrood (AFB) is a virulent disease of honey bee brood caused by the Gram-positive, spore-forming bacterium; Paenibacillus larvae. In this study, we determined the potential of bacteria isolated from hives of Asian honey bees (Apis cerana) to act antagonistically against P. larvae. Isolates were sampled from different locations on the fronts of A. cerana hives in Vietnam. A total of 69 isolates were obtained through a culture-dependent method and 16S rRNA gene sequencing showed affiliation to the phyla Firmicutes and Actinobacteria. Out of 69 isolates, 15 showed strong inhibitory activity against P. larvae; Bacillus pocheonensis (VN101) showed the largest zone of inhibition (26 ± 1 mm). In this study, the diversity and richness of antagonistic isolates indicated that Bacillus spp. are the most promising as inhibitors of P. larvae. These finding suggested that certain bacterial isolates can act as antagonist to control P. larvae and may have other biotechnological applications. Keywords: honey bee, 16S rRNA, microbial ecology, Paenibacillus larvae, antagonistic bacteria 1. INTRODUCTION The Asian honey bee, Apis cerana, along A. mellifera colonies has been a major concern with the European honey bee, A. mellifera, because extensive colony losses threaten serve as critical insect pollinators of multiple agriculture’s reliance on pollination services economically important crops in Southeast from honey bees [2]. Various factors are most Asia [1]. In recent years, the health of likely acting in concert to fuel this pollinator Chiang Mai J. Sci. 2018; 45(3) 1203 crisis, including pests and pathogens such as AFB is frequently reported in A. mellifera, viruses [3], microsporidians [4], ectoparasitic and often results in death of an infected mites [5], fungi and bacteria [6]. In response colony. AFB is known in parts of Asia, but to increased A. mellifera colony losses, has not been documented in some regions, A. cerana has received increased interest such as Thailand, India and Vietnam [1]. from researchers as a “healthier” honey bee Despite a lack of information regarding its alternative for beekeeping, with their prevalence in A. cerana and other Asiatic demonstrably faster hygienic response that honey bee species [1], it is important to gain a has the potential to better control the spread deeper understanding of the microbial of pathogens within colonies [7]. Coupled symbionts found in microenvironments of with the drive to identify worker traits the Asian honey bee, and how these bacteria that could improve honey bee health is an may interact with pathogenic bacteria such as increased interest in understanding how P. larvae. Therefore, the objectives of this nest microenvironments [8] and the colony study were to improve our understanding of microbiome (both in the nest and from the microbial community of A. cerana hives and bees themselves) contributes to the health to investigate the potential use of these and productivity of colonies [9-11] microbes as natural antagonistic agents There is a growing understanding of against P. larvae. We achieved this using 16S the ways in which bacterial communities rRNA gene sequence analysis to identify interact with their hosts as well the wider cultivable bacterial species that were present colony environment [11]. Past studies have in samples collected from the fronts of examined the relationship amongst colony A. cerana hives in Vietnam. These microbes microbes, fitness, nutrition and disease were then cultured so that they could be resistance in honey bees [10-12]. Recently, evaluated in vitro for potential antagonist much focus has centered on bacterial honey activity against P. larvae. bee communities and their presence in various in bee body parts [13], their contribution to 2. MATERIALS AND METHODS bee bread [14] and their influence on honey 2.1 Sampling and Isolation of Hive [15]. The majority of research has focused Associated Bacteria on bacterial communities in A. mellifera Microbial samples were collected from colonies, but studies are broadening three apiaries of A. cerana colonies that were to investigations of A. cerana colonies [13]. housed in wooden hives and maintained by a Of great relevance to honey bee health commercial beekeeper in Da Chong, Ba Vi are microbes that might act against the District, Hanoi Province, Vietnam in year 2013. Gram-positive spore-forming bacterium A total of 9 field samples were collected Paenibacillus larvae; a causative agent of using sterile cotton swabs that were rubbed American foulbrood (AFB). This disease is on three different locations of the hives: among the most virulent of honey bee brood (a) at top of the hive front, (b) 5 cm inside [16]. Recent studies have demonstrated the the entrance on the hive bottom, and (c) potential of honey bee bacterial symbionts 10 cm above the entrance. The swabs were that have with clear antagonistic effects against aseptically packed and sent to the Bee Research P. larvae; these bacteria have been isolated Laboratory, USDA-ARS in Beltsville, MD, from A. mellifera larvae [17] and the gut of USA, where they were processed following A. cerana japonica [18]. Good Laboratory Practices (GLP), with all 1204 Chiang Mai J. Sci. 2018; 45(3) materials autoclaved and properly disposed Biosystems 3130 DNA Genetic Analyzer. of following use. Prior to cultivation, swabs were 2.3 Phylogenetic Analysis moistened with 100 μl of sterile distilled DNA sequence data from this study water. All samples were cultivated on nutrient were submitted to GenBank (accession agar (NA) plates at 37 °C for 24 h. Bacterial numbers KU060150-KU060220), and morphology was primarily used to distinguish compared to closely related 16S rRNA isolates; quantity was measured as cfu/plate. sequences retrieved from GenBank All bacterial isolates were re-cultured and database by BLAST-N searches (http:// kept in 50% glycerol solution and stored blast.ncbi.nlm. nih.gov/Blast.cgi). Multiple at -20 °C for further study. sequence alignments were performed with CLUSTALW multiple alignment software 2.2 DNA Extraction and Sequencing embedded in BioEdit [19]. A phylogenetic Bacteria were taken directly from NA relationship was constructed using MEGA6 plates for DNA extraction. One full loop of software that employed maximum likelihood bacteria was collected to be re-suspended algorithms, based on paired alignments in 1 ml of sterile water in a microcentrifuge of nucleotide sequences of the 16S rRNA tube. This tube was then centrifuged at 12,000 genes, to compare our isolates with other rpm for 1 min so that the supernatant could corresponding organisms. The Kimura-2 be removed. Following the manufacturer’s parameter was applied for generating instructions, 200 ul of InstaGene™ Matrix phylogenetic tree [20], which were (Bio-Rad, Hercules, CA) was added to the bootstrapped with 1,000 bootstrap pellet. 16S rRNA genes were amplified by replications; only values greater than PCR using universal eubacterial primers eu27.F 70% were considered above the branches. (5′ GAGAGTTTGATCCTGGCTCAG 3′) and eu1495.R (5′ CTACGGCTACCTTGTT 2.4 Antagonist inhibition assays ACGA 3′)[17]. PCR was performed in a final P. larvae strain ATCC9545 was used as a volume of 30 μl, which consisted of 1 μl type reference strain. It was obtained bacterial extract (approximately 200 ng/μl from the American Type Culture Collection of total genomic DNA), 2 U Taq DNA (ATCC, USA) and cultivated in brain polymerase (Qiagen, Germantown, MD) heart infusion (BHI) agar (Difco, Sparks, MD) with 1× reaction buffer, 1 mM dNTPs mix, at 37 °C for 3 days. Inhibitory activities of and 0.1 μM of each primer. PCR was each isolate against P. larvae were assayed performed using a thermal cycler set at using a modified well diffusion technique, 30 cycles of 93 °C 1 min, 54 °C 1 min, and as described previously [18]. Wells (6 mm in 72 °C 1 min. Bands of an appropriate size diameter) were cut on an agar plate using were confirmed using 1.5 % agarose gel a cork borer. Bacterial suspensions from electrophoresis. PCR products were then P. larvae cultures, grown overnight, were purified directly with QIAquick® PCR suspended in sterile distilled water and purification kit (Qiagen, Germantown, MD). measured comparing McFarland standard Sequencing was performed using Big Dye no. 1. The suspension then spread evenly 2.0 (Applied Biosystems, Foster City, CA) over the surface of each BHI agar plate, end-terminal cycle sequencing, followed as previously described by Evans and by separation and analysis on an Applied Armstrong (2006) [17]. At the same time, Chiang Mai J. Sci. 2018; 45(3) 1205 each cultivable bacterial colony on NA agar Three bacterial isolates showed <96 % plate was suspended as described above. sequence similarity. VN 59 and VN 95 showed Then, 100 μl of each bacterium suspension the best match to Bacillus odysseyi, whereas was applied to one well of each plate. VN109 was closest to Paenibacillus ginsengarvi.