Journal of Applied Microbiology 2001, 91, 212216

Disinfection of wooden structures contaminated with Paenibacillus larvae subsp. larvae spores
W. Dobbelaere1,2, D.C. de Graaf1, W. Reybroeck3, E. Desmedt2, J.E. Peeters1 and F.J. Jacobs2
1

Veterinary and Agrochemical Research Centre, Brussels, 2Laboratory for Zoophysiology, University of Ghent, Ghent, 3Department Animal Product Quality, Agricultural Research Centre Ghent, Melle, Belgium

723/01/01: received 12 January 2001, revised 2 March 2001 and accepted 2 March 2001

W. DOBBELAERE, D.C. DE GRAAF, W. REYBROECK, E. DESMEDT, J.E. PEETERS A N D F . J . J A C O B S . 2001.

Aims: The aim of the study is to examine the disinfection of wood contaminated with Paenibacillus larvae subsp. larvae spores, in order to nd a practical decontamination method for hive materials. Methods and Results: The number of viable spores recovered after the treatment, on the surface by swabbing, and in the deeper parts of the wood by scraping, was used to test the efciency of the disinfection. Our results indicate that chemical disinfection is only complete when high concentrations (> 50%) of the disinfectant are used. Heat treatment in general was found to be very effective. The scorching of wood was not satisfactory as it only killed spores at the surface. Conclusions: Complete disinfection is only possible with some heat treatments or by using high concentrations of chemical disinfectants. Signicance and Impact of the Study: This study puts forward some methods that can provide complete decontamination, which is necessary for an effective control of American foulbrood disease.
INTRODUCTION Paenibacillus larvae subsp. larvae is the causative agent of American foulbrood disease (AFB), a fatal disease of the honeybee. The spores are the only transmissive stage of the bacteria. They are highly resistant and can remain infectious for more than 35 years (Haseman 1961). The disinfection of beehive equipment is a crucial step in the control of AFB, especially when the apiary is confronted with clinical disease, which is accompanied by the production of extremely high numbers of spores. Okayama et al. (1997) found that products based on sodium hypochlorite (NaOCl) and glutaraldehyde among others were effective against the spores of P. larvae subsp. larvae. However, those tests were only performed on spores in suspension. It is reasonable to assume that the disinfection of bee equipment, for instance the wooden beehives, is much more difcult. Del Hoyo et al. (1998) described a successful disinfection method, tested under circumstances corresponding more to practical beekeeping. They found the dipping of wooden frames in hot parafn to be very effective. The use of gamma irradiation was also proven to be effective (Hornitzky and Wills 1983). In Belgium, a frequently applied method is scorching with a blowtorch. However, in the past decade, there were several cases of clinical AFB where the disease reappeared in the same apiary after 45 years, putting to question the worth of the disinfection method used. The aim of the work presented here was to compare several methods of decontamination, which could be useful in practical circumstances with the aim of obtaining a complete disinfection of the material. Several chemical disinfectants that are commonly used in the food industry were selected for examination and compared with known heat decontamination methods. Gamma irradiation was not tested, as no such facilities are available to Belgian beekeepers. As our results were conicting with earlier papers on the destruction of P. larvae subsp. larvae spores in suspension, supplementary a European Suspension Test
2001 The Society for Applied Microbiology

Correspondence to: Wim Dobbelaere, Veterinary and Agrochemical Research Centre, Groeselenberg 99, B-1180 Brussels, Belgium (e-mail: widob@var.fgov.be).

DISINFECTION OF WOODEN HIVE MATERIAL

213

(EST) was performed for comparison. The EST is a standardized test in which the efcacy of a chemical disinfectant against a certain organism in suspension is tested using standard contact and neutralization times, at standard temperature (Anon. 1987, 1988). The test was performed in both presence and absence of organic material. MATERIALS AND METHODS Spores The stock solution of spores used in this study was obtained by suspending diseased larvae in sterile water. After a heat shock (80C, 10 min) the spore concentration of a stock solution was determined by plating a serial dilution (10)1 to 10)8) on a semiselective MYPGP medium, which consists of a normal MYPGP medium (Dingman and Stahly 1983) supplemented with 20 lg ml)1 pipemidic acid and 9 lg ml)1 nalidixic acid (Alippi 1995). The number of colony forming units (cfu) was determined after 4 d incubation at 37C. The stock solution was stored at 4C until used. Chemical disinfection of wood Small sticks (7 4 25 cm) of clean, untreated pinewood were used, with a similar structure to the wood normally used for beehives. We spotted 100 ll of the undiluted spore stock solution on a marked area of 7 cm2 on the stick and subsequently let it dry at room temperature for 2 d. The chemical disinfection was performed by submerging the wooden sticks in the chemicals for 30 min at room temperature. The following chemicals were tested: three products based on sodium hypochlorite Hygienius Ultra (Alfa Laval Agri), Hygienius Super (Alfa Laval Agri) and the popular household product bleach (local brand); one product, AV5 (Atlantol), was based on glutaraldehyde and formaldehyde; another, Tego2000 (DiverseyLever), was based on amphoteric compounds. Different concentrations, varying from 05 to 75%, of each product were tested. The disinfected sticks were dried again at room temperature for at least 48 h. No actual neutralization of the disinfectant was performed in this test in order to remain as close as possible to the situation in practice. The sampling for bacteriological examination was twofold. Firstly, the marked area on the wooden stick was thoroughly sampled by rubbing it with a sterile cotton swab, which was subsequently rinsed with 1 ml of sterile water. Then 100 ll of the rinsing uid were plated directly or after dilution on semiselective MYPGP. The number of cfu was determined after 4 d. Secondly, in order to evaluate the disinfecting activity in deeper parts of the wood, wood bres were scraped off in the marked area using a sterile chisel, up to a depth of 23 mm. All the

scrapings were then suspended in 1 ml of sterile water and vortex mixed thoroughly. Again 100 ll of the suspension were plated on semiselective MYPGP. The results are shown as the mean values obtained from three repeated experiments. European Suspension Test The EST was performed as described by Anon. (1987, 1988). However, the organic material to mimic the contamination found in practical circumstances used was yeast instead of albumin. The EST was performed with bleach only. The sodium hypochlorite was inactivated by a 05% sodium thiosulphate solution. The results are shown as the mean values obtained from three repeated experiments. Heat disinfection of wood The dipping in hot parafn was performed as described by Del Hoyo et al. (1998). The dry heat method was performed in an oven at three different temperatures (140, 160 and 180C) for 2 h. For the disinfection with wet heat a standard sterilization at 121C for 20 min in an autoclave was performed. The results are shown as the mean values obtained from three repeated experiments. The scorching method was carried out with an industrial roong burner. The wood was singed for as long as possible ( 30 s) making sure the wooden sticks did not catch re. The distance between the burner and the wood was approx. 30 cm. This test was repeated eight times. RESULTS Articial contamination The stock solution contained 49 108 cfu. Thus each wooden stick was spotted with 49 107 cfu. Recovery rate on positive controls for external samples was 37% and for internal samples 118% of the total inoculum, giving a total recovery rate of cultivable spores of 155% per untreated stick. Chemical disinfection The disinfection with the group of chemical products based on sodium hypochlorite was only 100% effective at very high concentrations. Figure 1 shows the number of cfu per stick recovered after 30 min submersion in different concentrations of Hygienius Ultra. The level of decontamination varied little when dilutions of 525% were used. Compared to the untreated wooden blocks there was very good (> 9999%) but not complete disinfection at the surface. The internal disinfection was slightly less efcient

2001 The Society for Applied Microbiology, Journal of Applied Microbiology, 91, 212216

214 W . D O B B E L A E R E E T A L .

Fig. 1 Logarithmic representation of the average number of cfu recovered from inoculum of 49 107 spores per stick after treatment of wooden sticks with different concentrations of Hygienius Ultra (Alfa Laval Agri). Positive control (PC) is a contaminated and untreated piece of wood, negative control (NC) is a noncontaminated and untreated piece of wood. Differentiation is made between the place of sampling and the concentration of the product used. Error bars represent the standard deviation of the different values obtained; when not shown, the standard deviation was 0. Values above the bars indicate the average number of cfu per stick found

Fig. 2 Logarithmic representation of the average number of cfu recovered from inoculum of 49 107 spores per stick after treatment of wooden sticks with different concentrations of bleach. Positive control (PC) is a contaminated and untreated piece of wood, negative control (NC) is a noncontaminated and untreated piece of wood. Differentiation is made between the place of sampling and the concentration of the product used. Error bars represent the standard deviation of the different values obtained; when not shown, the standard deviation was 0. Values above the bars indicate the average number of cfu per stick found

(between 9993 and 9995%). Only with solutions of 50% or more of Hygenius ultra was the disinfection complete, both supercially and internally. Hygienius Super, another product of the same manufacturer, showed similar properties, with > 9999% disinfection obtained with a 25% solution, and complete supercial decontamination if a solution of 50% or more was used. A 50% solution of Hygenius Super did not achieve complete internal disinfection (> 9999%); this was obtained only if a 75% solution was used. Normal working solution concentration of both Hygienius Super and Hygienius Ultra is 05%. After treatment with the most concentrated bleach (20, French chlorometric degree, indicating the volume Cl2-gas released from 1 l of bleach under conditions of 0C and 1013 kPa, 20 6% NaOCl), only < 001% of the spores were recovered at the surface (Fig. 2). As can be expected, the lower the concentration of bleach, the lower its efcacy. Internal disinfection was slightly less efcient (9990%). Prolongation of the contact period showed no further improvement of the product's effect. These results seem to be somewhat conicting with those published by Okayama et al. (1997). These authors found that low concentrations (005%) of sodium hypochlorite were sufcient to destroy P. larvae subsp. larvae spores. To corroborate our ndings, we did an EST with bleach. We found that spores were effectively destroyed by concentra-

tions corresponding with those found by Okayama et al. (1997). Without organic contamination a 03125 solution ( 01% NaOCl) resulted in a complete decontamination, whereas in the presence of 2% yeast extract a 5 ( 155% NaOCl) solution was required. In these conditions, lower concentrations resulted in incomplete decontamination (03125: 914%; 0625: 926%; 125: 948%; 25: 981% efciency). Contact periods of 5 min and 30 min produced the same results. The results obtained using AV5 were comparable with those given by products based on sodium hypochlorite. Complete disinfection (both supercially and internally) was only possible with concentrations of 50% or higher (gures not shown). Application of Tego2000 did not result in a decrease in cfu compared to the positive control. Heat disinfection After dipping the wooden sticks in parafn at a temperature of 120C, > 9999% of the spores were killed at the surface and 9998% internally. At 145C or more, complete decontamination at the surface was achieved. At 170C there was a complete decontamination at both levels (Fig. 3). Disinfection using dry heat produced similar results as the hot parafn treatment. Complete decontamination, both

2001 The Society for Applied Microbiology, Journal of Applied Microbiology, 91, 212216

DISINFECTION OF WOODEN HIVE MATERIAL

215

(between 57 102 and 31 103 cfu per stick) remained viable internally. DISCUSSION The reappearance of American foulbrood disease in hives that have been disinfected by scorching, which we prove to be > 999% effective, clearly shows the need for complete decontamination. The high virulence and long viability of these pathogenic strains leave no other option than a complete disinfection of the contaminated hives. Several papers have described the successful use of chemical disinfectants against P. larvae subsp. larvae spores in suspension. However, there seems to be a discrepancy between disinfection of spores in suspension and those situated within wooden structures. Chemical disinfectants are effective in the treatment of wooden surfaces but can only provide a complete decontamination at extremely high concentrations, which is economically and environmentally difcult to accept. This reduces the value of those chemicals in practical beekeeping, as most beehives are still constructed of wood. The EST results partially explain why wood is much more difcult to decontaminate: the presence of as little as 2% organic material considerably reduces the efcacy of chemical disinfection, and wooden bres probably will act in a similar way on the disinfectants. Moreover, all disinfectants have a limited penetration capacity, so spores deep down in the wood will remain largely unaffected. Disinfection methods based on heat provide a good alternative. These methods are both cheaper and environmentally more acceptable. Del Hoyo et al. (1998) already pointed out that dipping in hot parafn is a good method for the disinfection of wooden frames, and this was conrmed by our ndings. The use of dry heat can also be considered to be a very good method. The temperatures we worked with were comparable with those applied in the parafn dipping method, and in both methods the wood seemed undamaged after the treatment. An important difference between these two methods was the time of exposure needed. Although 10 min in hot parafn is sufcient, 2 h in the oven are needed to obtain complete disinfection. Another point to consider is the safety of the procedure: the manipulation of hot parafn requires some skill. The use of an autoclave is also very effective, but not available to most beekeepers. Scorching of the wooden beehives cannot be recommended. In conclusion, the decontamination of wooden structures such as hives remains a problem. This seems to be mainly due to the structure of the wood rather than to the nature of the spores. The use of synthetic hives and hive material would thus enable a more effective decontamination and should for that reason be recommended.

Fig. 3 Logarithmic representation of the average number of cfu recovered from inoculum of 49 107 spores per stick after disinfection of wooden sticks by dipping in hot parafn. Positive control (PC) is a contaminated and untreated piece of wood, negative control (NC) is a noncontaminated and untreated piece of wood. Differentiation is made between the place of sampling and the temperature of the parafn. Error bars represent the standard deviation of the different values obtained; when not shown, the standard deviation was 0. Values above the bars indicate the average number of cfu per stick found

supercially and internally, was obtained with temperatures of 160C or more (Fig. 4). After autoclaving, no viable spores were found at the surface of the wooden sticks and only very low numbers (< 10 cfu per stick) internally. Scorching of wooden blocks gave a complete supercial decontamination, while substantial numbers of viable spores

Fig. 4 Logarithmic representation of the average number of cfu recovered from inoculum of 49 107 spores per stick after disinfection of wooden sticks by dry heat. Positive control (PC) is a contaminated and untreated piece of wood, negative control (NC) is a noncontaminated and untreated piece of wood. Differentiation is made between the place of sampling and the temperature used. Error bars represent the standard deviation of the different values obtained; when not shown, the standard deviation was 0. Values above the bars indicate the average number of cfu per stick found

2001 The Society for Applied Microbiology, Journal of Applied Microbiology, 91, 212216

216 W . D O B B E L A E R E E T A L .

ACKNOWLEDGEMENTS This research was supported by funds from the Ministry of Small Enterprises, Traders and Agriculture, DG6 (S-5886) and by a grant of the Flemish beekeepers federation (Koninklijke Vlaamse Imkersbond). We also like to thank Alfa Laval Agri and DiverseyLever for the kind donation of samples of the different chemical decontaminants, and Dr D. Vandekerchove for her critical reading of the manuscript and her comments regarding it. REFERENCES
Alippi, A.M. (1995) Detection of Bacillus larvae spores in Argentinian honeys by using a semi-selective medium. Microbiologia SEM 8, 115118. Anon. (1987) Test Methods for the Antimicrobial Activity of Disinfectants in Food Hygiene. Strasbourg: Council of Europe.

Anon. (1988). British Standard Draft for Development 177 Method of Test for the Antimicrobial Activity of Disinfectants in Food Hygiene. London: British Standards Institution. Del Hoyo, M., Basualdo, M., Torres, J. and Bedascarrasbure, E. (1998) Use of DHT-equipment for disinfection of AFB-contaminated beehive materials in Argentina. American Bee Journal 138, 738740. Dingman, D.W. and Stahly, D.P. (1983) Medium promoting sporulation of Bacillus larvae and metabolism of medium components. Applied and Environmental Microbiology 46 (4), 860869. Haseman, L. (1961) How long can spores of American foulbrood live? American Bee Journal 101, 298299. Hornitzky, M.A.Z. and Wills, P.A. (1983) Gamma radiation inactivation of Bacillus larvae to control American foul brood. Journal of Apicultural Research 22 (3), 196199. Okayama, A., Sakogawa, T., Nakajima, C. and Hayama, T. (1997) Sporicidal activities of disinfectants on Paenibacillus larvae. Journal of Veterinary and Medical Sciences 59 (10), 953954.

2001 The Society for Applied Microbiology, Journal of Applied Microbiology, 91, 212216