ISSN 00036838, Applied Biochemistry and Microbiology, 2013, Vol. 49, No. 1, pp. 79–84. © Pleiades Publishing, Inc., 2013. Original Russian Text © N.V. Rusenova, P. Parvanov, S. Stanilova, 2013, published in Prikladnaya Biokhimiya i Mikrobiologiya, 2013, Vol. 49, No. 1, pp. 88–93.
Development of Multiplex PCR for Fast Detection of Paenibacillus Larvae in Putrid Masses and in Isolated Bacterial Colonies N. V. Rusenovaa, P. Parvanova, and S. Stanilovab a
Faculty of Veterinary Medicine, Department of Veterinary Microbiology, Infectious and Parasitic diseases, Trakia University, Stara Zagora 6000, Bulgaria b
Faculty of Medicine, Department of Molecular Biology, Trakia University, Stara Zagora 6000, Bulgaria email:
[email protected]~~V Received Junuary 26, 2012
Abstract—The present study was performed to develop a fast and sensitive multiplex polymerase chain reac tion protocol for routine diagnostics of American foulbrood. A new approach for detection of Paenibacillus larvae in putrid masses was described. Forty five samples of putrid masses obtained from bee combs suspicious for American foulbrood, a reference strain Paenibacillus larvae (NBIMCC 8478), clinical isolates and 4 strains of closely related bacterial species were included in experiments. Bacterial colonies’ DNA was iso lated by heat and centrifugation method (standard procedure) and with prepGem commercial kit. DNA from putrid masses was isolated by standard and modified procedure. Three pairs of primers specific for 16S rRNA and one pair specific for 35 kDa metalloproteinase genes of Paenibacillus larvae were tested as single pair and in different combinations as multiplex PCR. The sensitivity of the multiplex PCR protocol for putrid masses, developed in study was 100%, versus 45.2% for the standard protocol. The developed multiplex PCR protocol could be successfully used for rapid and specific detection of Paenibacillus larvae in both putrid masses and isolated bacterial colonies. DOI: 10.1134/S0003683813010171
tional methods for microbiology investigation, how ever, are time consuming and not reliable enough when other sporeforming bacteria are present in clin ical samples. Commercial identification kits are also available [9, 10] but they require some days for obtain ing results. Recently, some genotypes have been described among P. larvae strains in European apiaries differing in their colony morphology and this might additionally complicate routine diagnosing [11]. Therefore, the development of a fast and sensitive method for detection of the pathogen is needed to pre vent and control AFB. Recently, PCRbased methods for the identifica tion of various pathogens, including P. larvae were published [11–14]. Nevertheless, it is still little infor mation in specialized literature on detection of P. lar vae in putrid masses by multiplex PCR. The aim of the present study was to develop a rapid, sensitive and specific multiplex PCR protocol for the detection of P. larvae directly in putrid masses and in isolated bacterial colonies for the purposes of routine diagnosing.
American foulbrood (AFB) is one of the most delete rious bacterial honey bee diseases though affecting only the larval stages of honey bees [1]. The causative agent is Paenibacillus larvae (P. larvae), which is a Grampositive and sporeforming bacterium [2]. Spores of P. larvae are extremely tenacious and are the only infectious form of that organism [3]. Ten or less spores are sufficient to cause the disease in very young larvae [4]. AFB has spread worldwide causing considerable economic losses to apiculture [5, 6]. Various factors predispose this event such as transmission of spores by routine beekeeping practice, international trade with bee products, robbery or swarming of infected bee col onies [7, 8]. Therefore, timely detection of P. larvae is of great importance to prevent dissemination of infec tion. Even though clinical symptoms are specific for AFB, in many countries including the Republic of Bulgaria field diagnosis has to be confirmed after lab oratory examination. Laboratory diagnosis of AFB is based on analysis of clinical signs followed by isolation and identification of the etiological agent. Conven 79
80
RUSENOVA et al.
Table 1. Selected primers for the detection of P. larvae. Primer
Sequence
Gene [ref.]
1.
5'AAGTCGAGCGGACCTTGTGTTTC3'
2.
5'TCTATCTCAAACCGGTCAGAGG3'
3.
F3,5'CGGGCAGCAAATCGTATTCAG3'
4.
B1,5'CCATAAAGTGTTGGGTCCTCTAAGG3'
5.
E1,5'GCAAGTCGAGCGGACCTTGTG3'
6.
E2,5'AAACCGGTCAGAGGGATGTCAAG3'
7.
F6,5'GCACTGGAAACTGGGAGACTTG3'
8.
B11,5'CGGCTTTTGAGGATTGGCTC3'
MATERIALS AND METHODS Putrid masses and bacterial strains. Forty five sam ples of putrid masses obtained from bee combs suspi cious for AFB, a reference strain Paenibacillus larvae NBIMCC 8478 (National Bank for Industrial Micro organisms and Cell Cultures, Sofia, Bulgaria), 66 clin ical isolates and 4 strains of closely related bacterial species—Brevibacillus laterosporus (NBIMCC 8036), Bacillus licheniformis (NBIMCC 3346), Paenibacillus alvei and Bacillus subtilis subsp. subtilis (clinical iso lates from bee combs determined by BioLog Gen III system, USA) were included in our experiments. Field strains were isolated on tryptic soy agar (Fluka, Swit zerland) supplemented with 5% sheep blood and incu bated at 37°C for 72 h under aerobic conditions. Isolation of DNA from bacterial strains. DNA from the tested strains was isolated in two different proce dures. The first was described previously by Dob belaere [12] and cited in OIE Terrestrial Manual [15], so it was called standard. Bacterial suspensions were heated for 20 min at 95°C and centrifuged for 10 min at 5000 g. The supernatant was used as DNA template in PCR assays. The second procedure was made according to the manufacturer’s instructions of the DNA bacterial extraction kit prepGem (ZyGem, USA). Isolation of DNA from putrid masses. DNA from putrid masses was also isolated in two ways. First, putrid masses of two brood cells were suspended in 1 mL of sterile distilled water and mixed thoroughly. Hundred µL of this suspension was diluted in 900 µL sterile distilled water and vortexed. Hundred µL of this dilution was used to extract DNA by the standard heating and centrifiigation method [12]. Second, we modified the described procedure with dilution of putrid masses in CASO broth (Fluka, Switzerland) instead of distilled water and incubation of 100 µL of
Length of amplicon, bp
16S rRNA [19]
973
Metalloproteinase gene [11]
273
16S rRNA [11]
965
16S rRNA [11]
665
each dilution for 1 h at 37°C in water bath prior to heating and centrifiigation step. All DNA templates were stored at –20°C till used. Selection of primers. Selected primers, targeted genes and the length of expected amplicons are pre sented in Table 1. The choice of primers was based on the previously published reports for the detection of P. larvae [11, 19]. Primers were tested as single pair in standard PCR and in combination in multiplex PCR. The following combinations were used: 1–2/3–4; 3– 4/5–6 and 3–4/7–8. Primers were supplied by Meta bion, Germany. PCR amplification. Multiplex PCR reaction mix ture (20 µL) contained 1 × PCR buffer (100 mM Tris HCl, pH 8.8, 500 mM KC1), 1.5 mM MgCl2; 200 µM of each dNTP, 0.25 µM of each primer; 1.0 U Taq polymerase and 1 µL of DNA template. The suitable annealing temperature for our experimental condi tions was determined by gradient PCR. PCR amplifi cation was done in a PCR System (Quanta Biotech QB96 thermocycler, England) and consisted of an initial denaturation step at 94°C for 3 min, followed by 30 cycles of: denaturation at 94°C for 1 min; annealing at 54°C for 0.5 min; extension at 72°C for 1 min and a final extension step of 7 min at 72°C completed the reaction. PCR products (10 µL) were separated on 1.0% agarose gel stained with ethidium bromide (0.5 µg/mL), visualized on a UV transilluminator (ImageQuant 150, Taiwan) and photographed using a gel documentation system. Reagents for PCR reaction were supplied by Fermentas, Lithuania. Sensitivity and specificity of the used multiplex PCR protocol. The sensitivity of the multiplex PCR proto col was determined by serial dilutions of P. larvae (NBIMCC 8478) DNA within the range of 120– 15 ng/µL and the specificity of the reaction was tested with closely related bacterial species.
APPLIED BIOCHEMISTRY AND MICROBIOLOGY
Vol. 49
No. 1
2013
DEVELOPMENT OF MULTIPLEX PCR FOR FAST DETECTION M
Ref SZ1 Lch44 1K
81
SZ1 Lch44 2K
Ref
a
b
Fig. 1. Detection of P. larvae by using multiplex PCR mixtures representing combination of primers 1–2/3–4 (a) and 3–4/5–6 (b) in 1% agarose gel stained with ethidium bromide; M—molecular marker (100 bp), Ref—referent strain (NBIMCC 8478); SZ 1 and Lch 44—field strains; 1K and 2K—non template controls of the two reaction mixtures.
M
1
2
3
4
5
6
7
8
9
100 bp
500
100
Fig. 2. Detection of the tested P. larvae strains in the ethidium bromide stained 1.0 % agarose gel demonstrating the PCR products using the combination of primers 1–2 (973 bp amplicon) and 3–4 (273 bp amplicon). M—100 bp marker; lanes 1–9—the tested P. larvae strains.
RESULTS A standard PCR with single pair of primers for the detection of P. larvae targeted genes was preliminary performed. DNA from the reference strain’s bacterial
colonies was used as a positive template control. Using the optimized PCR conditions, expected length of amplification products was generated. Successful amplification of the pointed genes allowed us to carry out experiments with multiplex PCR. One pair of
Table 2. Sensitivity and specificity of the multiplex PCR protocol for the detection of P. larvae in putrid masses Procedure of DNA isolation PCR standard
modified
Sensitivity in one amplified gene
21.4%
90.5%
Sensitivity in two amplified genes
45.2%
Specificity of 16S rRNA gene
29%
Specificity of Mlp gene APPLIED BIOCHEMISTRY AND MICROBIOLOGY
100%
40.5% Vol. 49
No. 1
100%
2013
90.5%
82
RUSENOVA et al. M
1
2
3
4
5
6
7
8
9
a
b
Fig. 3. Detection of P. larvae in putrid masses by standard (a) and modified (b) procedure of DNA isolation. Ethidium bromide stained 1.0% agarose gel showing the PCR trials. M—marker (100 bp); lanes (1–9)—the tested P. larvae strains.
primers was always for the detection of 16S rRNA gene of P. larvae, and the other one was applied for detec tion of metalloproteinase gene precursor (Mlp). Results from gradient PCR indicated 54°C as the most appropriate annealing temperature. Our results showed that the best combination of primers was 1– 2/3–4 though the combination 3–4/5–6 also led to satisfactory results (Fig. 1). Further, all strains of our collection were tested with primers 1–2 and 3–4. Fig ure 2 represents PCR products with the expected length of amplicons, obtained with primers 1–2 and 3–4, respectively. It is obvious that the fragment of 16S rRNA gene of strain No.1 and Mlp gene of strain No.3 were not well amplified. Moreover, it is necessary to take into account that the range of DNA concentra tion in samples varied from 30.8 to 143 ng/µL. No specific PCR products were obtained when control PCR analyses were performed with closely related bacterial species. The sensitivity of this multi plex PCR protocol had a detection limit of DNA less than 15 ng/µL. The developed multiplex PCR protocol was then run directly with putrid masses according to the stan dard procedure for DNA isolation. Surprisingly, we obtained a low sensitivity of the method—only 45.2%. That is why, we modified the standard procedure with an incubation step as described in Materials and Methods. Our modification has led to increase in the sensitivity of the multiplex PCR protocol up to 100%. Visualization of PCR products after electrophoresis of treated putrid masses was interpreted by a four plus system—from barely detected (1+) to strongly visual
ized (4+) amplicons. The specificity of the reaction of 16S rRNA gene in the standard procedure was 29%, versus 100% in the modified one. The specificity of the Mlp gene was 40.5% and 90.5%, respectively (Table 2). Moreover, the results were repeated twice in indepen dent attempts for DNA isolation. Experiments with putrid masses are shown on Fig. 3. DISCUSSION The rapid detection of P. larvae is essential for pre vention of infection spread and efficient control of the disease. Moreover, beekeepers suffering losses due to the disease are compensated only on the basis of a doc ument with laboratory confirmed diagnosis. Our clin ical and laboratory experience confirms the belief of Gilliam [16], that in many instances, the detection of P. larvae with conventional microbiological methods is complicated by sample contamination with other sporeforming microorganisms present in bee hives. That is why the amplification of specific target bacte rial genome fragments is being increasingly used for detection and diagnostic purposes [17]. The incon testable advantages of PCR as utmost specificity, per formance speed and very good reproducibility, make it superior to conventional and phenotypic methods for identification. With the reclassification of P. larvae subsp. larvae and P. larvae subsp. pulvifaciens into a single species, the scope of the more specific PCR assays has become too narrow for diagnostic purposes [18].
APPLIED BIOCHEMISTRY AND MICROBIOLOGY
Vol. 49
No. 1
2013
DEVELOPMENT OF MULTIPLEX PCR FOR FAST DETECTION
In this study, we developed a multiplex PCR proto col for simple and rapid detection of P. larvae in both bacterial colonies and putrid masses using the stan dard PCR method. To obtain a maximum specificity, two different genes were selected for amplification— the 16S rRNA and Mlp genes. The primer combina tion 1–2/3–4 turned out to be the best with lack of nonspecific PCR products, as those observed with other combinations despite the optimized conditions under which the multiplex PCR reaction was run. These are the primers specific for the 16S rRNA gene of P. larvae, selected by Govan [19] with GenBank accession No X60619 and for the Mlp gene with Gen Bank accession No AF111421 [11]. Amplification of fragments from 16S rRNA and Mlp genes of P. larvae in bacterial colonies is reported by others [11, 13], but using touchdown PCR. Our results also showed that the simultaneous amplification of two different genes was a better choice for P. larvae detection as seen from Fig. 2. The insufficient number of copies from a single gene could be poorly visualized in agarose gel and cause a false negative result. The developed protocol showed a high DNA detection threshold—less than 15 ng/µL. The primers combination 1–2/3–4 did not generate any specific amplification products with DNA from closely related bacterial species, as reported by other researchers [20]. Our attempts for detection of P. larvae in putrid masses with the standard protocol described yielded a surprisingly low sensitivity of the method—only 45.2%, having in mind that analyzed specimens were putrid masses from which P. larvae was isolated and identified by multiplex PCR. This finding is contra dictory to the data of Dobbelaere [12], who yielded amplicons with the expected length from all tested brood samples. In both cases samples were collected from brood with clinical AFB. The authors have tested 4 different combinations of two pairs of primers syn thesized on the basis of the of 16S rRNA gene sequence of P. larvae, including primers synthesized by Govan [19]. All this required modification of the DNA isolation procedure with adding a complemen tary incubation step at 37°C for at least 1 h before the 20 min heating at 95°C that increased the sensitivity of the PCR protocol up to 100%. Although the ampli cons in some of samples processed by the modified procedure were barely detected (1+), they were evi dence in subsequent independent DNA isolations. Putrid mass suspensions contain mainly spores, about 2.5 billion [21], as well as vegetative cell debris. During incubation in a rich nutrient medium such as the CASO broth, the permeability of spore envelope is altered and metabolic processes involved in germina tion are activated [22]. Under such optimal condi tions, DNA of cells is probably undergoing conforma tional changes during the initiation of the division cycle, resulting in easier access to the respective genes APPLIED BIOCHEMISTRY AND MICROBIOLOGY
83
subject to amplification. Han [23] reported an ultra rapid PCR protocol for P. larvae detection, but this method used a specific microscale chipbased real time system not available in every diagnostic labora tory. A realtime PCRbased strategy for the diagnosis and screening of AFB was developed by Martinez [24] but this technique is still expensive and not applicable in many developing countries. In conclusion, we believe that the developed multi plex PCR protocol could be successfully used in rou tine diagnostics for rapid detection of P. larvae in both putrid masses and isolated colonies. REFERENCES 1. Genersch, E., J. Verbr. Lebensm., 2008, vol. 3, no. 4, pp. 429–434. 2. Genersch, E., Forsgren, E., Pentikainen, J., Ash iralieva, A., Rauch, S., Kilwinski, J., and Fries, I., Int. J. Syst. Evol. Microbiol., 2006, vol. 5, no. 3, pp. 501– 511. 3. Rauch, S., Ashiralieva, A., Hedtke, K., and Genersch, E., Appl. Environ. Microbiol., 2009, vol. 75, no. 3, pp. 3344–3347. 4. Bailey, L. and Lee, D.C., J. Gen. Microbiol., 1962, vol. 29, no. 4, pp. 711–717. 5. Ellis, J.D. and Munn, P.A., Bee World, 2005, vol. 86, no. 4, pp. 88–101. 6. Fries, I., Lindstrom, A., and Korpela, S., Vet. Micro biol., 2006, vol. 114, nos. 3–4, pp. 269–274. 7. Hornitzky, M.A.Z., J. Apic. Res., 1998, vol. 37, no. 4, pp. 261–265. 8. Lindstroöm, A., Korpela, S., and Fries, I., Apidologie, 2008, vol. 39, no. 5, pp. 1–9. 9. Carpana, E., Marocchi, L., and Gelmini, L., Apidolo gie, 1995, vol. 26, no. 1, pp. 11–16. 10. Dobbelaere, W., De Graaf, D.C., Peeters, J.E., and Jacobs, F.J., J. Apic. Res., 2001, vol. 40, no. 1, pp. 37–40. 11. Neuendorf, S., Hedtke, K., Tangen, G., and Genersch, E., Microbiology, 2004, vol. 150, no. 7, pp. 2381–2390. 12. Dobbelaere, W., De Graaf, D.C., and Peeters, J.E., Apidologie, 2001, vol. 32, no. 4, pp. 363–370. 13. Kilwinski, J., Peters, M., Ashiralieva, A., and Gener sch, E., Vet. Microbiol., 2004, vol. 104, nos. 1–2, pp. 31–42. 14. Lauro, F.M., Favaretto, M., Covolo, L., Rassu, M., and Bertoloni, G., Int. J. Food Microbiol., 2003, vol. 81, no. 3, pp. 195–201. 15. Manual of Diagnostic Tests and Vaccines for Terrestrial Animals. OIE Terrestrial Manual, 6th ed., Paris, 2008, pp. 395–404. 16. Gilliam, M., FEMS Microbiol. Lett., 1997, vol. 155, no. 1 P, pp. 1–10. 17. Vaneechoutte, M., and VanEldere, J, J. Med. Micro biol., 1997, vol. 46, no. 3, pp. 188–194.
Vol. 49
No. 1
2013
84
RUSENOVA et al.
18. De Graaf, D.C., Alippi, A.M., Brown, M., Evans, J.D., Feldlaufer, M., Gregorc, A., and Hornitzky, M., Per naI, S.F., Schuch, D.M.T., Titéra, D., Tomkies, V., and Ritter, W, Lett. Appl. Microbiol., 2006, vol. 43, no. 6, pp. 583–590. 19. Govan, V.A., Allsopp, M.H., and Davison, S., Appl. Environ. Microbiol., 1999, vol. 65, no. 5, pp. 2243– 2245. 20. Piccini, C., D’Alessandro, B., Antunez, K., and Zunino, P., J. Microbiol. Biotechnol., 2002, vol. 18, no. 8, pp. 761–765.
21. Shimanuki, H. and Knox, D.A., Diagnosis of Honey Bee Diseases, U.S. Agriculture Handbook, AH690, Belts ville, MD: US Department of Agriculture, USA, 2000. 22. Moir, A., J. Appl. Microbiol., 2006, vol. 101, no. 3, pp. 526–530. 23. Han, S.H., Lee, D.B., Lee, D.W., Kim, E.H., and Yoon, B.S., J. Invertebr. Pathol., 2008, vol. 99, no. 1, pp. 8–13. 24. Martínez, J., Simon, V., Gonzalez, B., and Conget, P., Lett. Appl. Microbiol., 2010, vol. 50, no. 6, pp. 603–610.
APPLIED BIOCHEMISTRY AND MICROBIOLOGY
Vol. 49
No. 1
2013
