Simplified agar plate method for quantifying viable bacteria.

We have developed a simple and reproducible technique that we have termed “track dilution” for enumerating viable bacteria, which reduces time and materials required compared to standard agar plate colony counting methods. Many microbiological applications require an accurate determination of bacterial numbers prior to inoculation into, or subsequent to their isolation from, experimental models and systems. For example, our experimental infection models require injection of predetermined numbers of viable colony-forming units (cfu) (1,4). The model further requires careful monitoring of bacterial growth rates at specific time points during infection. In our models of rodent gastrointestinal ecology (3), bacterial concentration must be determined from extraordinarily large numbers of fecal samples. The most commonly used technique for enumerating bacterial cfu involves serial dilution of samples followed by spreading 50–100 μL of each dilution onto agar media. Subsequent to incubation, colonies are counted and bacterial concentrations in the original sample estimated. This technique, though relatively accurate, has some drawbacks, including time required and a considerable expenditure of disposable plastics and media. To meet criteria for statistical accuracy of bacterial numbers in a given specimen samples are usually plated in triplicate, and cfu counted only from plates yielding between 20 and 200 visible colonies (2). Furthermore, since the actual number of bacteria present in a specimen at the completion of an experiment is unknown, samples must frequently be plated from dilutions ranging from 10-1 to 10-7. The result of such exhaustive measures is that most plated samples will yield either too many colonies to count or no colonies at all. For example, an overnight bacterial broth culture containing 109 cfu/mL, serially diluted 10fold over a range of 10-1 to 10-7 and plated (100 μL per plate) in triplicate, will yield only three agar plates with countable and statistically valid numbers of cfu; the remaining 18 plates must be discarded. To simplify bacterial enumeration and reduce time and material expenditures, we have developed a track-dilution technique whereby six 10-fold serial dilutions of a sample containing an unknown quantity of bacteria are plated onto a single agar plate. We compared track dilution with the standard spreadplate method by: (i) recording time required to process samples, (ii) recording materials expended and (iii) statistically analyzing the data for accuracy and reproducibility. Enterococcus faecalis and Staphylococcus aureus were propagated in brain-heart infusion (BHI) broth (Difco Laboratories, Detroit, MI, USA) overnight at 37°C without aeration. Bacillus subtilis and Escherichia coli DH5α (Life Technologies, Gaithersburg, MD, USA) were propagated in LB overnight at 37°C with aeration. Cultures were serially diluted 10-fold by sequential transfer of 100 μL into 900 μL phosphate-buffered saline (PBS). Experiments were performed in triplicate to allow statistical comparisons between the two techniques. For conventional spread-plate counting, cultures were diluted by a factor of 10-7, from which 100-μL samples were spread onto the surface of prewarmed BHI 1.5% agar plates (100× 15-mm round plates; Fisher Scientific, Pittsburgh, PA, USA) using a sterile bent-glass rod and an inoculating turntable (Fisher Scientific). Plates were incubated upright for ap-