1 Länderinstitut für Bienenkunde, Friedrich-Engels-Str. 32, 16540 Hohen Neuendorf, Germany
2 Oxoid GmbH, Am Lippeglacis 4-8, 46467 Wesel, Germany
Correspondence
Elke Genersch
elke.genersch{at}rz.hu-berlin.de
![]() |
ABSTRACT |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
![]() |
INTRODUCTION |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
The oval-shaped spores represent the infectious stage of P. l. larvae. AFB is transmitted by spore-containing honey being fed to newly hatched larvae. So far, the only known host for P. l. larvae is honey bee larvae. The spores germinate in the midgut lumen. The vegetative forms of P. l. larvae then penetrate the gut epithelium and proliferate within the larval tissue. Within 72 h the bee larvae are reduced to tissue detritus, which forms a glue-like colloid (rope stage). Later still the larval remains dry down to a scale adhering to the side of the cell. This scale is highly infectious since it contains billions of spores (Bailey & Ball, 1991; Gregorc & Bowen, 1998
).
Recent genotyping of German field isolates of P. l. larvae revealed at least four different genotypes, named AB, Ab, ab and B (Genersch & Otten, 2003
). Here we present data on the further characterization of the genotypes AB, Ab and ab, and of the reference strain DSM 7030/ATCC 9545. Biochemical fingerprinting was performed using the Biolog system. This system involves the determination of the metabolism of 95 carbon sources in a microtitre plate format. Since the metabolism of bacteria is adapted to their natural environment or host, each bacterium prefers or uses particular carbon sources. Hence, determination of the metabolic profile of a micro-organism can be used for identifying and characterizing the organism.
Earlier reports show that some isolates of P. l. larvae harbour plasmid DNA (Benada et al., 1988; Bodorova-Urgosikova et al., 1992
). Therefore, we analysed all isolates for the presence of extrachromosomal DNA to see whether or not the occurrence of plasmid DNA might be an additional feature suitable for typing purposes and epidemiology.
![]() |
METHODS |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Paenibacillus larvae subsp. pulvifaciens (P. l. pulvifaciens) reference strains DSM 8442 and DSM 8443 were obtained from DSMZ. P. l. pulvifaciens DSM 8442 and DSM 8443 are synonyms for P. l. pulvifaciens reference strains NRRL NRS 1683 and NRRL NRS 1684, respectively.
Eighty-six German field isolates from P. l. larvae were isolated from honey samples originating from 86 AFB-positive hives diagnosed in the course of foulbrood monitoring programs between 2000 and 2003 (Table 1). Diagnosis was based on clinical symptoms as well as on isolation, cultivation and absolute identification of the causative agent, P. l. larvae. Honey samples had been stored at 4 °C until P. l. larvae was cultivated for scientific purposes from these samples on bacterial plates. Two representatives of the recently identified P. l. larvae genotype AB (Genersch & Otten, 2003
) have been deposited at the DSMZ (accession numbers DSM 16115, DSM 16116).
|
The culture of P. l. larvae from honey samples was performed essentially as previously described (Genersch & Otten, 2003). Briefly, for growth of spore-forming bacteria, honey samples were solubilized overnight at 37 °C. Subsequently, samples were diluted in double-distilled water to obtain a 50 % (w/v) honey solution. To select for spores, samples were incubated at 90 °C for 6 min. Samples were allowed to cool down at room temperature prior to plating them (200 µl per plate) onto Columbia sheep blood agar plates. Three plates were prepared from each sample. Plates were incubated at 37 °C and evaluated for bacterial growth after 3 and 6 days. After 6 days, P. l. larvae-like colonies were identified by catalase and Plagemann tests as well as by PCR detection prior to subsequent detailed analysis.
For further analysis, all isolates were stored as bacterial suspensions in 25 % glycerol in BHI (brain heart infusion) broth at 70 °C.
Biochemical analysis of P. l. larvae reference strain DSM 7030 was performed by subculturing this strain and taking 24 independent subcultures.
Catalase and Plagemann tests.
For absolute identification, colonies with a P. l. larvae-like morphology were further analysed by catalase and Plagemann tests. For the catalase test part of the colony in question was transferred to a microscopic slide using a wooden stick and mixed with a drop of 3 % H2O2. Production of air bubbles is indicative for catalase activity, whereas no air bubbles indicates a lack of catalase activity. For the Plagemann test (Plagemann, 1985), the liquid part of Columbia sheep blood agar slants was inoculated with part of the bacterial colony in question. The tube was air-tight sealed with Parafilm and incubated at 37 °C for 10 days. Subsequently, the liquid part was analysed for the presence of spores and giant whips by phase-contrast microscopy. P. l. larvae is characterized by the lack of catalase activity and giant whips occurring during sporulation (Ritter, 1996
; Hansen & Brodsgaard, 1999
).
P. l. larvae-specific PCR.
For PCR identification of bacterial colonies grown on agar plates, part of the colony in question was resuspended in 50 µl double-distilled water and subsequently incubated at 90 °C for 15 min. Probes were centrifuged at 5000 g for 10 min. The supernatant containing the DNA was transferred to a new tube and directly used for PCR analysis. PCR analysis was based on 16S rDNA sequences of P. l. larvae (accession numbers AY030079 for strain NRRL B-3555 and X60619 for strain ATCC 9545) and on the partial sequence of the gene for a 35 kDa metalloprotease from P. l. larvae (AF111421). Primer sequences were designed using MacVector 6.5 software and compared with published sequences in the GenBank databases using BLAST (Altschul et al., 1990): Pll-16S E1, 5'-GCAAGTCGAGCGGACCTTGTG-3'; Pll-16S E2, 5'-AAACCGGTCAGAGGGATGTCAAG-3'; Pll-16S F6, 5'-GCACTGGAAACTGGGAGACTTG-3'; Pll-16S B11, 5'-CGGCTTTTGAGGATTGGCTC-3'; Pll-MP F3, 5'-CGGGCAGCAAATCGTATTCAG-3'; Pll-MP B1, 5'-CCATAAAGTGTTGGGTCCTCTAAGG-3'.
PCR analyses were carried out in a final volume of 25 µl consisting of 1x Qiagen reaction buffer, 250 µM dNTPs (dATP, dCTP, dGTP, dTTP), 10 µM primer and 0·3 U HotStarTaq polymerase (Qiagen). Concentrations of MgCl2 were adjusted so that all three reactions could be performed at a final annealing temperature of 56 °C: 2·3 mM for primer pair Pll-16S E1/E2, 1·7 mM for primer pair Pll-16S F6/B11 and 1·5 mM for primer pair Pll-MP F3/B1. After the initial activation step (15 min, 95 °C), the reaction conditions for the touch-down PCR were as follows. All denaturation steps were performed at 94 °C for 30 s, all elongation steps were performed at 72 °C for 30 s, and all annealing steps were performed for 1 min. For annealing, temperatures of 66, 62 and 58 °C were used and 5 cycles were run at each temperature. At the final annealing temperature of 56 °C, 30 cycles were run followed by a final elongation step at 72 °C for 8 min. Five microlitres of the PCR samples was analysed on a 0·8 % agarose gel. The DNA bands were stained with ethidium bromide and visualized by UV light.
The expected lengths of specific amplicons generated with Pll-16S E1/E2, Pll-16S F6/B11 and Pll-MP F3/B1 were 965, 665 and 273 bp, respectively. For all P. l. larvae genotypes, specificity of amplicons was verified by sequencing (Medigenomix, Germany). When tested for specificity with P. l. pulvifaciens reference strains DSM 8442 and DSM 8443, none of the primers generated any PCR product (Fig. 1). Specific PCR products were not generated when control PCR analyses were performed with reference strains for Paenibacillus alvei (DSM 29), Paenibacillus apiarius (DSM 5581, DSM 5582, DSM 5612), Paenibacillus polymyxa (DSM 36), Bacillus licheniformis (DSM 13), Bacillus mycoides (DSM 299), Bacillus thuringiensis (DSM 6029) and other unidentified bacilli isolated from honey samples (data not shown).
|
Biochemical fingerprinting (Biolog system).
The Biolog system (obtained through Oxoid) is a carbon source test, where the ability of a bacterial isolate to metabolize 95 different carbon sources is used for identification purposes. Cultivation and preparation of P. l. larvae isolates for metabolic analysis using the Biolog system were performed according to the manufacturer's instructions for spore-forming, Gram-positive rods, with minor modifications to meet the growth requirements of P. l. larvae. Briefly, single pure colonies of P. l. larvae were subcultured on BUG-M-T agar plates (bacterial universal growth agar supplemented with 0·25 % maltose and swapped with thioglycolate) with one colony per plate. Using sterile wooden sticks a special streaking technique was performed resulting in a plus' sign on the centre of the plate. The goal of this technique is to restrain cell growth to two thin lines so that the cells along the edges have a good supply of food. This keeps the cells in an active state and decreases sporulation. For the analysis of one isolate, eight pure colonies of this isolate were subcultured on eight BUG-M-T plates. Growth of bacteria at 37 °C was continued for 48 h since P. l. larvae is a slow-growing bacterium. Subsequently, the inoculum was prepared by taking only those colonies starting at the ends of the plus' sign to half way down the junction of the two lines constituting the plus' sign. Bacteria from the centre or close to the centre must not be taken. Colonies were picked up with a wooden stick and rubbed around the walls of an empty, sterile dry glass tube. The bacterial film was suspended in 5 ml inoculation solution (GN/GP IF; Oxoid) to obtain a homogeneous mixture. After adding the remaining fluid (10 ml), turbidity was adjusted to match the turbidity standard (Oxoid) at 28±2 % turbidity (OD=0·55). Subsequently, a GP2 MicroPlate (Oxoid) was inoculated with 150 µl bacterial suspension per well and incubated at 37 °C for 24 h. Metabolic activity was determined by reading the plates (end point reading method) in a microplate reader (EL800; BIO-TEK Instruments) with a primary wavelength of 590 nm and a reference wavelength of 750 nm. Positive wells were then entered manually into the Biolog MicroLog1 (release 4.20) software after choosing plate type (GP2), strain type (GP-ROD SB) and incubation time (1624 h). Interpretation of the results and identification of the bacterial strain in question was performed automatically by the software. The Biolog database does include a standard profile for P. l. larvae (Table 2) and, therefore, is able to identify P. l. larvae. The software does not include P. l. pulvifaciens.
|
Preparation of plasmid DNA.
Frozen bacterial suspensions were thawed, plated on Columbia sheep blood agar plates and allowed to grow for 3 days at 37 °C. Colonies were scraped off and resuspended in 300 µl BHI broth. Up to six plates were pooled to yield a sufficient amount of bacteria for plasmid preparation. Subsequently, bacteria were pelleted by centrifugation at 5000 g for 10 min. The bacterial pellet was used for plasmid preparation performed with the QIAprep Spin Miniprep kit (Qiagen) by exactly following the manufacturer's protocol. A 12 µl sample of each eluent was analysed on a 0·8 % agarose gel. The DNA bands were stained with ethidium bromide and visualized by UV light.
Restriction analysis of plasmid DNA.
Depending on band intensity, 28 µl of each eluent containing plasmid DNA was used for restriction analysis. Restriction reactions were carried out in a final volume of 10 µl using 10 U of the corresponding restriction enzyme together with the appropriate reaction buffer. Reactions were incubated at 37 °C for 30 min. Subsequently, 2 µl 6x DNA loading buffer was added and the reactions were analysed on a 0·8 % agarose gel. The DNA bands were stained with ethidium bromide and visualized by UV light.
![]() |
RESULTS |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
|
|
|
|
![]() |
DISCUSSION |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Earlier reports on biochemical characterization of P. l. larvae using traditional macro (Jelinski, 1985; Alippi & Aguilar, 1998
) or commercial micro methods (Carpana et al., 1995
; Dobbelaere et al., 2001
) showed the usefulness of such tests for the classification of P. l. larvae, although the results obtained were somewhat contradictory. It was suggested that these differences are due to the different systems used (Dobbelaere et al., 2001
). Here we present evidence that the discrepancies between different studies on the biochemical properties of P. l. larvae are rather due to genotype-specific differences. We analysed the metabolic pattern of four different genotypes of P. l. larvae (AB, Ab, ab and a
) by using the Biolog system. The system involves the determination of the metabolism of 95 different carbon sources. Interpretation of positive reactions is performed via an ELISA reader. Allotting the metabolic fingerprints to the different genotypes revealed characteristic patterns for each genotype (Table 2
). To our knowledge this is the first time that genotype-specific metabolic profiles could be defined for P. l. larvae, indicating that differences in genotype correlate with differences in biochemical phenotype. When compared to other genotypes, P. l. larvae genotype AB exhibits the most striking metabolic pattern, since it is the only strain able to metabolize the carbohydrates D-fructose (100 %) and D-psicose (96 %), and the only strain unable to use glycerol as carbon source. The other strains, of genotypes Ab, ab and a
, also show characteristic metabolic patterns, but are nevertheless more similar to each other. Analysing the metabolic pattern of two reference strains for P. l. pulvifaciens, DSM 8442 and DSM 8443, resulted in no identification using the Biolog software, since their biochemical fingerprint differed in more than 20 carbon sources from the profiles of P. l. larvae (J. Kilwinski, M. Peters, A. Ashiralieva & E. Genersch, unpublished results). Therefore, the Biolog system allows not only the identification of P. l. larvae, but also the definite discrimination between genotype AB and the other genotypes. Since the Biolog system is used in microbiological diagnosis, this result will be of diagnostic relevance if differences in virulence can be assigned to differences in genotype.
Our results show that the characteristic metabolic patterns always contain some variables for each genotype. Hence, groupings based on biochemical properties where these properties are understood as static traits will lead to results which hardly correlate with genotyping.
Comparing our results with the Biolog standard we can conclude that at least genotype AB was not included when establishing the standard metabolic fingerprint for P. l. larvae. D-fructose and D-psicose are both given 0 %, meaning that it was never accepted as a carbon source by any isolate included in the survey. In contrast, D-fructose and D-psicose are metabolized by 100 and 96 %, respectively, by isolates belonging to genotype AB. Furthermore, some strains included in the Biolog standard are not represented in our study, since we never found any isolate able to metabolize 3-methyl-D-glucose, turanose or L-alanine. The exact metabolic relationships between the different genotypes of P. l. larvae in relation to the Biolog standard are given in the dendrogram (Fig. 4). The distance in biochemical phenotype between genotype AB and the other genotypes and, in contrast, the relative closeness of genotypes Ab, ab and a
become obvious.
Jelinski (1985) distinguished seven biochemical types (IVII) according to seven possible combinations of three biochemical properties: reduction of nitrate to nitrite, hydrolysis of mannitol and acid production from salicin. The same study revealed the ability to metabolize glycerol as a consistent feature of P. l. larvae. Based on our analysis using the Biolog system, this holds true for genotypes Ab, ab and the reference strain DSM 7030. In contrast, no isolate belonging to genotype AB was able to use glycerol as carbon source. It has been reported that comparison between the biochemical type (IVII; Jelinski, 1985
) and the genotype of isolates rendered no obvious link between both features (Alippi & Aguilar, 1998
). This result is in disagreement with our results showing a clear link between biochemical and rep-PCR fingerprints. In the study performed by Alippi & Aguilar (1998)
genotyping is based on rep-PCR performed with primers BOX A1R and REP (REP1R.I and REP2-I), a primer combination having less discriminatory power than BOX A1R combined with MBO REP1 (Genersch & Otten, 2003
), as used in our study. Therefore, it is likely, that the genotypes defined by Alippi & Aguilar (1998)
would split up if MBO REP1 primers were used instead of REP primers, possibly allowing a better correlation between biochemical type and genotype. Furthermore, the discriminatory power of only three metabolic properties is quite poor as compared to the Biolog system where the metabolism of a total of 95 carbon sources is analysed. Above all, if the three metabolic properties chosen by Alippi & Aguilar (1998)
are variable features within the genotypes it will be impossible to find any obvious linkage between both features.
Based on the API 50CHB system, Carpana et al. (1995) determined the ability of P. l. larvae to metabolize 49 carbohydrates and their derivatives and presented a detailed comparison of the results with those reported in the literature. In particular for galactose, fructose and mannitol the results were contradictory. We found that among the isolates investigated in our study only genotype AB was able to metabolize D-fructose and mannose, for example. Therefore, we propose that the discrepancies between different studies are in part due to different genotypes analysed in these studies that in turn might be due to the differential geographic origin of the isolates.
Our data show that genotype AB is outstanding in respect to colony morphology and metabolic fingerprint. When screening all isolates used in this study for extrachromosomal DNA, only representatives of genotype AB were found to harbour plasmids. So far, no plasmid DNA has been detected in isolates from genotypes Ab, ab or a. Therefore, another characteristic feature of genotype AB is the presence of plasmids. The occurrence of plasmid DNA in P. l. larvae has been reported (Benada et al., 1988
; Bodorova-Urgosikova et al., 1992
; Drobnikova et al., 1994
). The plasmid denoted pBL423/728 is about 9·4 kb in size and this seems to be in agreement with our data at first. But whereas plasmid pBL423/728 does not contain an XbaI restriction site and digestion with EcoRI results in two fragments of 3·6 and 5·8 kb (Bodorova-Urgosikova et al., 1992
), plasmid pPll9.4 characterized in our study is linearized by XbaI and gives rise to two EcoRI fragments of 850 and 8550 bp. Hence, the two plasmids, although similar in size, are not identical. Since pPll11.0 differs from pPll9.4 only by a 1600 bp insertion, this plasmid also is not related to plasmid pBL423/728.
It has long been recognized that the proteins comprising the parasporal Cry toxins of Bacillus thuringiensis, an insecticidal, Gram-positive, spore-forming bacterium, are generally encoded by large plasmids (Gonzalez et al., 1981; for review see Schnepf et al., 1998
). Nothing is known so far about toxins expressed by P. l. larvae. It will be interesting to screen the newly found plasmids of P. l. larvae for genes possibly involved in pathogenicity.
In some countries, P. l. larvae has been treated in bee colonies by the antibiotic oxytetracyclin for several decades. Recently, widespread resistance to oxytetracyclin has been reported (e.g. Miyagi et al., 2000). A recent study analysing the origin of oxytetracyclin resistance in P. l. larvae did not include any search for plasmids, but rather focused on the correlation between 16S rDNA haplotypes and resistance (Evans, 2003
). No convincing correlation was found, thus leading to the speculation that resistance might be epigenetic in nature, specifically through the presence of plasmids and mobile genetics entities that produce proteins involved in resistance (Adams et al., 1998
). Although no specific resistance to sulphonamides, antibiotics, mercury chloride or cadmium nitrate connected with the presence of plasmid pBL423/728 in P. l. larvae was found (Benada et al., 1988
), this is still an open question for pPll9.4 and pPll11.0. Since both these plasmids are not related to pBL423/728 it will be interesting to further characterize the newly identified plasmids and look for any resistance-connected genes carried by those plasmids.
Overall, our study has identified and characterized the exceptional P. l. larvae genotype AB for the first time. Considering what is known so far about this genotype, it may be the first choice for more detailed analyses with respect to virulence, pathogenicity and antibiotic resistance.
![]() |
ACKNOWLEDGEMENTS |
---|
![]() |
REFERENCES |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Alippi, A. M. & Aguilar, O. M. (1998). Characterization of isolates of Paenibacillus larvae subsp. larvae from diverse geographical origin by the polymerase chain reaction and BOX primers. J Invertebr Pathol 72, 2127.[CrossRef][Medline]
Altschul, S. F., Gish, W., Miller, W., Myers, E. W. & Lipman, D. J. (1990). Basic local alignment search tool. J Mol Biol 215, 403410.[CrossRef][Medline]
Bailey, L. & Ball, B. V. (1991). Honey Bee Pathology, 2nd edn. New York, London: Academic Press.
Benada, O., Drobnikova, V., Kalachova, L. & Ludvik, J. (1988). Plasmid DNA in Bacillus larvae. J Apic Res 27, 3539.
Bodorova-Urgosikova, J., Benada, O. & Tichy, P. (1992). Large-scale isolation and partial characterization of plasmid DNA form B. larvae. Folia Microbiol 37, 8286.
Carpana, E., Marocchi, L. & Gelmini, L. (1995). Evaluation of the API 50CHB system for the identification and biochemical characterization of Bacillus larvae. Apidologie 26, 1116.
Dobbelaere, W., de Graaf, D. C., Peeters, J. E. & Jacobs, F. J. (2001). Comparison of two commercial kits for biochemical characterization of Paenibacillus larvae larvae in the diagnosis of AFB. J Apic Res 40, 3740.
Drobnikova, V., Richter, V., Häusler, J. & Pytelova, I. (1994). Characterization of Bacillus larvae and related bacilli by chromatography of cell fatty acids. J Apic Res 33, 6974.
Evans, J. D. (2003). Diverse origins of tetracycline resistance in the honey bee bacterial pathogen Paenibacillus larvae. J Invertebr Pathol 83, 4650.[CrossRef][Medline]
Genersch, E. & Otten, C. (2003). The use of repetitive element PCR fingerprinting (rep-PCR) for genetic subtyping of German field isolates of Paenibacillus larvae subsp. larvae. Apidologie 34, 195206.[CrossRef]
Gonzalez, J. M., Jr, Dulmage, H. T. & Carlton, B. C. (1981). Correlation between specific plasmids and -endotoxin production in Bacillus thuringiensis. Plasmid 5, 351365.
Gregorc, A. & Bowen, I. D. (1998). Histopathological and histochemical changes in honeybee larvae (Apis mellifera L.) after infection with Bacillus larvae, the causative agent of American foulbrood disease. Cell Biol Int 22, 137144.[CrossRef][Medline]
Hansen, H. & Brodsgaard, C. J. (1999). American foulbrood: a review of its biology, diagnosis, and control. Bee World 80, 523.
Heyndrickx, M., Vandemeulebroecke, K., Hoste, B., Janssen, P., Kersters, K., de Vos, P., Logan, N. A., Ali, N. & Berkeley, R. C. (1996). Reclassification of Paenibacillus (formerly Bacillus) pulvifaciens (Nakamura 1984) Ash et al. 1994, a later subjective synonym of Paenibacillus (formerly Bacillus) larvae (White 1906) Ash et al. 1994, as a subspecies of P. larvae, with emended descriptions of P. larvae as P. larvae subsp. larvae and P. larvae subsp. pulvifaciens. Int J Syst Bacteriol 46, 270279.[Abstract]
Jelinski, M. (1985). Some biochemical properties of Bacillus larvae White. Apidologie 16, 6976.
Miyagi, T., Peng, C. Y. S., Chuang, R. Y., Mussen, E. C., Spivak, M. S. & Doi, R. H. (2000). Verification of oxytetracyclin-resistant American foulbrood pathogen Paenibacillus larvae in the United States. J Invertebr Pathol 75, 9596.[CrossRef][Medline]
Office International des Epizooties (2000). Manual of standards for diagnostic tests and vaccines. OIE Manual 2000, 784788.
Plagemann, O. (1985). Eine einfache Kulturmethode zur bakteriologischen Identifizierung von Bacillus larvae mit Columbia-Blut-Schrägagar. Berl Münch Tierärztl Wschr 98, 6162.
Ritter, W. (1996). Diagnostik und Bekämpfung von Bienenkrankheiten. Jena: Gustav Fischer.
Schnepf, E., Crickmore, N., van Rie, J., Lereclus, D., Baum, J., Feitelson, J., Zeigler, D. R. & Dean, D. H. (1998). Bacillus thuringiensis and its pesticidal crystal proteins. Microbiol Mol Biol Rev 62, 775806.
Versalovic, J., Schneider, M., de Bruijn, F. J. & Lupski, J. R. (1994). Genomic fingerprinting of bacteria using repetitive sequence-based polymerase chain reaction. Methods Mol Cell Biol 5, 2540.
Received 1 March 2004;
revised 30 March 2004;
accepted 7 April 2004.
HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
INT J SYST EVOL MICROBIOL | MICROBIOLOGY | J GEN VIROL |
J MED MICROBIOL | ALL SGM JOURNALS |