Letters in Applied Microbiology ISSN 0266-8254
ORIGINAL ARTICLE
A simple and fast method for determining colony forming units S. Sieuwerts1,2,3, F.A.M. de Bok1,2, E. Mols4, W.M. de Vos1,3 and J.E.T. van Hylckama Vlieg1,2 1 2 3 4
Top Institute Food and Nutrition, Wageningen, The Netherlands Nizo Food Research, Ede, The Netherlands Laboratory of Microbiology, Wageningen University, Wageningen, The Netherlands Biologic, Dordrecht, The Netherlands
Keywords colony forming units, Escherichia coli, faster enumeration, lactic acid bacteria, plating, Saccharomyces cerevisiae. Correspondence Johan van Hylckama Vlieg, Nizo Food Research, PO Box 20, 6710 BA Ede, The Netherlands. E-mail:
[email protected]
2007 ⁄ 2029: received 18 December 2007, revised and accepted 23 May 2008 doi:10.1111/j.1472-765X.2008.02417.x
Abstract Aims: To develop a flexible and fast colony forming unit quantification method that can be operated in a standard microbiology laboratory. Methods and Results: A miniaturized plating method is reported where droplets of bacterial cultures are spotted on agar plates. Subsequently, minicolony spots are imaged with a digital camera and quantified using a dedicated plugin developed for the freeware program ImageJ. A comparison between conventional and minicolony plating of industrial micro-organisms including lactic acid bacteria, Eschericha coli and Saccharomyces cerevisiae showed that there was no significant difference in the results obtained with the methods. Conclusions: The presented method allows downscaling of plating by 100-fold, is flexible, easy-to-use and is more labour-efficient and cost-efficient than conventional plating methods. Significance and Impact of the Study: The method can be used for rapid assessment of viable counts of micro-organisms similar to conventional plating using standard laboratory equipment. It is faster and cheaper than conventional plating methods.
The growth and maintenance of microbes on agar-containing media in Petri dishes has long been a common practice in microbiology. Traditionally, the preferred method for quantitative population analysis of pure and mixed cultures relies on plating of serial dilutions and subsequent counting of colony forming units (CFUs). In recent years a range of alternative, high-throughput (HT), methods relying on quantitative PCR (Neeley et al. 2005; Haarman and Knol 2006), fluorescent labelling (Blasco et al. 2003; Lay et al. 2005) or genome probing with micro arrays (Bae et al. 2005) have gained popularity (Liu et al. 2004a). However, most of these methods measure different entities, i.e. all cells, including nonviable cells. Moreover, these methods may require the use of special equipment or extensive protocol development. In addition, some of these methods are poorly compatible with complex substrates and environmental samples. This largely explains why enumeration of microbes by colony
counting is still a widely applied methodology. Presently, microbiology is increasingly moving towards HT analyses, which may require the use of large quantities of plates. This results in serious drawbacks when using conventional plating and colony counting techniques. The preparation of media and plates as well as the counting of colonies is time-consuming and labour-intensive. Moreover, large volumes may lead to significant costs, in particular when expensive indicator or reporter substrates are used. Finally, many methods consume large amounts of materials including disposables, compromising their sustainability. Several reports describe alternative plating technologies in which the required volumes are down-scaled (Jett et al. 1997; McNulty and Dunn 1999; Tornero and Dangl 2001; Hamilton et al. 2002) or the colony counting process is automated (Marotz et al. 2001; Dahle et al. 2004; Putman et al. 2005). However, these examples require equipment
ª 2008 The Authors Journal compilation ª 2008 The Society for Applied Microbiology, Letters in Applied Microbiology 47 (2008) 275–278
275
Miniaturized plating of microbes
S. Sieuwerts et al.
that is not present in standard microbiology laboratories (Gilchrist et al. 1973; Liu et al. 2004b), may result only in limited downscaling (Gilchrist et al. 1973; Jett et al. 1997; Hamilton et al. 2002), or are poorly suited for automation (McNulty and Dunn 1999; Hamilton et al. 2002). Building upon this work, we report a simple and flexible method for determining cell counts capitalizing on microtitre plate formats. It integrates rapid miniaturized plating and (semi-) automated counting of minicolonies allowing 100-fold downscaling of the process compared with conventional CFU counting procedures. It can be fully operated with little more than a multichannel pipette, a digital camera and ImageJ, an image processing package that is available as freeware (http://rsb.info.nih. gov/ij/). A number of examples are presented to illustrate the power of the approach for rapid assessment of viable counts of important prokaryotic and eukaryotic industrial micro-organisms. For plating, we used 10-fold serial dilutions that were prepared in 96-well micro plates using a Genex Alpha 12-channel pipette (Genex Labs, Torquay, UK). For each dilution, samples of five ll were pipetted onto a platecontaining agar medium using the same pipette. Typically, 60 to 120 droplets were applied per square 12 cm plate. The plates were air dried and then incubated until colonies were visible with an average size of 200–500 lm. Subsequently, these minicolonies were photographed with a Canon EOS 350D high-resolution digital camera (Canon 0020X665; Canon Inc., Japan) equipped with a Sigma 105 mm macro lens (AF 105MM F2Æ8 EX MACRO F; Sigma Corporation, Japan). Plates were put on a black surface and illuminated from the side in order to achieve a large contrast between colonies and background. Digital images were further processed in ImageJ using a newly developed plug-in that can be downloaded as Supporting Information. The counting process is schematically represented in Fig. S1. Briefly, the colour images were converted into 8-bit greyscale and inverted. By manual intervention using the ‘threshold’ function of ImageJ; colonies were selected from the background. Then all pictures were stacked. After inversion, the median was taken from each pixel with its neighbouring pixels for noise reduction. The ‘watershed’ function was used to separate merged colonies and the ‘analyse particles’ function was used for counting. The output was saved as a text file and subsequently processed in Microsoft Excel. In our study, we specifically aimed at developing fast CFU counting protocols for industrial microbes including lactic acid bacteria, Escherichia coli and Saccharomyces cerevisiae. When these strains are enumerated by conventional CFU counting, typically between 30 and 300 colonies per standard agar Petri dish of 8 cm in diameter results in optimal counting. Larger numbers may easily 276
result in underestimation as individual colonies cannot be discriminated. Low numbers of colonies per plate result in large standard deviations, which is the squared root of the average in Poisson distribution. This implies that in case five colonies are counted the number of cells in the plated samples is between 1 and 9 with 95% confidence whereas for 100 cells these values are 80 and 120. In the latter case, the 95% confidence interval is 20% deviation of the average. We therefore aimed at developing a protocol that would allow the counting of at least 100 colonies. This implies that, in any process of downscaling of platecounting, it is crucial to increase the number of colonies that can be counted per cm2 of agar surface. In our experiments, this was achieved by counting minicolonies with a size of 200–500 lm. Therefore, colonies were counted as soon as minicolonies are visible which is earlier than normally is done with conventional counting. Moreover, growing many colonies on a relatively small surface area results in smaller colonies for instance due to limited substrate availability (Liu et al. 2004b). This effectively increases the number of colonies that can be counted per cm2 of agar surface. We used the procedures described above to perform viable counting of various industrial micro-organisms and bench-marked these data to conventional counting procedures. Therefore, Streptococcus thermophilus CNRZ1066, Lactococcus lactis MG1363, Lact. plantarum WCFS1, E. coli DH5a and S. cerevisiae CBS57957 were cultured in appropriate media (Table 1) for approximately 24 h at 37C. Culture dilutions were prepared in 96-well plates in 12-fold and these were plated using both methods. For the conventional method 50 ll samples were pipetted onto a round agar plate of 8 cm in diameter with a Gilson pipetman P100 and spread using a glass swab. The average CFU counts were comparable and the standard deviations comparable or lower with the minicolony method confirming the suitability of the method for rapid CFU counting. From the standard deviations in Table 1, it can be concluded that the sensitivity of the fast method is comparable to the conventional method, although the variation between counts may differ with the type of pipettes used, as argued previously by Jett et al. (1997). For example, Lact. plantarum spots containing minicolonies were around 9 mm in diameter and therefore it is possible to count up to 200 colonies per cm2 efficiently on MRS-galactose agar, which is typically one to two orders of magnitude higher than with conventional plating. We found that the sizes of spots and minicolonies may vary with the type of agar medium, drying time, sample matrice and species of interest (not shown). It may affect the number of colonies that can be enumerated per 5 ll spot, typically in the range between 10 and 150, but not on the number of colonies that can be
ª 2008 The Authors Journal compilation ª 2008 The Society for Applied Microbiology, Letters in Applied Microbiology 47 (2008) 275–278
S. Sieuwerts et al.
Miniaturized plating of microbes
Table 1 Culturing and plating conditions used in the comparison of the conventional plating method and minicolony plating method and resulting average CFU ml)1 counts with corresponding standard deviations within 12 replicates Average CFU ml)1 (SD)
Temperature
Time (h)
Old method
Minicolony. method
30C
28 (new method) 40 (old method) 20
8Æ85E (3Æ25E 1Æ88E (3Æ17E 2Æ86E (9Æ21E
+ + + + + +
07 07) 08 07) 07 06)
2Æ69E (4Æ25E 1Æ04E (2Æ30E 1Æ37E (1Æ21E
+ + + + + +
08 07) 08 07) 07 06)
2Æ04E (1Æ58E 8Æ80E (7Æ29E
+ + + +
08 08) 07 07)
3Æ12E (8Æ29E 4Æ87E (2Æ19E
+ + + +
08 07) 07 07)
Dilution medium
Plaiting medium
M17 broth + 1% glucose M17 broth + 1% glucose 10% skim milk (Nilac) in co-culture with Strep. thermophilus TY broth
M17 broth + 1% glucose M17 broth + 1% glucose MRS broth
M17 agar + 1% glucose M17 agar + 1% glucose MRS agar + 1% galactose
37C
30C
24
TY broth
TY agar
30C
18
ME broth
ME broth
ME agar
30C
18 (new method) 22 (old method)
Species
Liquid growth medium
Streptococcus thermophilus Lactococcus lactis Lactobacillus plantarum Eschericha coli Saccharomyces cerevisiae
Incubation of plates
counted per cm2 for this certain species. The detection limit of the minicolony method is similar to that of the conventional method. The dynamic range of the new method was found equal to the conventional method (data not shown). In recent years, HT alternatives and variants have been developed for many conventional microbiological techniques. The best examples are probably liquid batch cultivations for which various multi-well alternatives belong to standard laboratory equipment now-a-days. Also several HT alternatives for counting of CFU are reported (Dahle et al. 2004; Liu et al. 2004b; Putman et al. 2005). Throughputs are increased by focusing either on automation of the counting process or on the miniaturization of the plating itself. The novelty of the method reported here relies on the integration of these aspects while it requires only standard laboratory equipment, a digital camera, imaging software that is available as freeware, and a dedicated plug-in that is available as Supplementary Material to this paper. The resulting fast plating and counting protocol is suitable for a quick determination of viable cell counts. The method is highly flexible, because it can easily be implemented for different microbial species and it is easy-to-use. Due to its miniaturization it reduces the amount of necessary materials by approximately 100fold, this makes it a cost and labour-efficient alternative for conventional methods. Because the counting is partially automated, the user can monitor critical steps in data acquisition and processing without the variability encountered from manual counting of CFUs (Lumley et al. 1997). We anticipate that this protocol is a valuable tool for routine enumeration of industrial microbes in research and quality control laboratories.
Acknowledgements The authors thank Colin Ingham and Patrick Janssen for technical suggestions. References Bae, J.W., Rhee, S.K., Park, J.R., Chung, W.H., Nam, Y.D., Lee, I., Kim, H. and Park, Y.H. (2005) Development and evaluation of genome-probing microarrays for monitoring lactic acid bacteria. Appl Environ Microbiol 71, 8825–8835. Blasco, L., Ferrer, S. and Pardo, I. (2003) Development of specific fluorescent oligonucleotide probes for in situ identification of wine lactic acid bacteria. FEMS Microbiol Lett 225, 115–123. Dahle, J., Kakar, M., Steen, H.B. and Kaalhus, O. (2004) Automated counting of mammalian cell colonies by means of a flat bed scanner and image processing. Cytometry A 60, 182–188. Gilchrist, J.E., Campbell, J.E., Donnelly, C.B., Peeler, J.T. and Delaney, J.M. (1973) Spiral plate method for bacterial determination. Appl Microbiol 25, 244–252. Haarman, M. and Knol, J. (2006) Quantitative real-time PCR analysis of fecal Lactobacillus species in infants receiving a prebiotic infant formula. Appl Environ Microbiol 72, 2359– 2365. Hamilton, C.M., Anderson, M., Lape, J., Creech, E. and Woessner, J. (2002) Multichannel plating unit for high-throughput plating of cell cultures. Biotechniques 33, 420–423. Jett, B.D., Hatter, K.L., Huycke, M.M. and Gilmore, M.S. (1997) Simplified agar plate method for quantifying viable bacteria. Biotechniques 23, 648–650. Lay, C., Sutren, M., Rochet, V., Saunier, K., Dore, J. and Rigottier-Gois, L. (2005) Design and validation of 16S rRNA probes to enumerate members of the Clostridium
ª 2008 The Authors Journal compilation ª 2008 The Society for Applied Microbiology, Letters in Applied Microbiology 47 (2008) 275–278
277
Miniaturized plating of microbes
S. Sieuwerts et al.
leptum subgroup in human faecal microbiota. Environ Microbiol 7, 933–946. Liu, B., Li, S. and Hu, J. (2004a) Technological advances in high-throughput screening. Am J Pharmacogenomics 4, 263–276. Liu, X., Wang, S., Sendi, L. and Caulfield, M.J. (2004b) Highthroughput imaging of bacterial colonies grown on filter plates with application to serum bactericidal assays. J Immunol Methods 292, 187–193. Lumley, M.A., Burgess, R., Billingham, L.J., McDonald, D.F. and Milligan, D.W. (1997) Colony counting is a major source of variation in CFU-GM results between centres. Br J Haematol 97, 481–484. Marotz, J., Lubbert, C. and Eisenbeiss, W. (2001) Effective object recognition for automated counting of colonies in Petri dishes (automated colony counting). Comput Methods Programs Biomed 66, 183–198. McNulty, J.J. and Dunn, J.J. (1999) High-throughput transformation and plating using petristrips. Biotechniques 26, 390–392. Neeley, E.T., Phister, T.G. and Mills, D.A. (2005) Differential real-time PCR assay for enumeration of lactic acid bacteria in wine. Appl Environ Microbiol 71, 8954–8957.
278
Putman, M., Burton, R. and Nahm, M.H. (2005) Simplified method to automatically count bacterial colony forming unit. J Immunol Methods 302, 99–102. Tornero, P. and Dangl, J.L. (2001) A high-throughput method for quantifying growth of phytopathogenic bacteria in Arabidopsis thaliana. Plant J 28, 475–481.
Supporting information Additional supporting information may be found in the online version of this article. Figure S1 Flow chart of the process of image analysis using the ImageJ plugin. Dashed boxes are steps that require human intervention. ImageJ_macro.zip Image processing plug-in for the ImageJ software Please note: Blackwell publishing are not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing material) should be directed to the corresponding author for the article.
ª 2008 The Authors Journal compilation ª 2008 The Society for Applied Microbiology, Letters in Applied Microbiology 47 (2008) 275–278
