In microbiology, colony-forming unit (CFU, cfu or Cfu) is a unit which estimates the number of microbial cells (bacteria, fungi, viruses etc.) in a sample that are viable, able to multiply via binary fission under the controlled conditions. Counting with colony-forming units requires culturing the microbes and counts only viable cells, in contrast with microscopic examination which counts all cells, living or dead. The visual appearance of a colony in a cell culture requires significant growth, and when counting colonies, it is uncertain if the colony arose from one cell or a group of cells. Expressing results as colony-forming units reflects this uncertainty.

Theory

A dilution made with bacteria and peptoned water is placed in an Agar plate (Agar plate count for food samples or Trypticase soy agar for clinic samples) and spread over the plate by tipping in the pattern shown.

The purpose of plate counting is to estimate the number of cells present based on their ability to give rise to colonies under specific conditions of nutrient medium, temperature and time. Theoretically, one viable cell can give rise to a colony through replication. However, solitary cells are the exception in nature, and most likely the progenitor of the colony was a mass of cells deposited together.^[1][2] In addition, many bacteria grow in chains (e.g. Streptococcus) or clumps (e.g., Staphylococcus). Estimation of microbial numbers by CFU will, in most cases, undercount the number of living cells present in a sample for these reasons. This is because the counting of CFU assumes that every colony is separate and founded by a single viable microbial cell.^[3]

The plate count is linear for E. coli over the range of 30 to 300 CFU on a standard sized Petri dish.^[4] Therefore, to ensure that a sample will yield CFU in this range requires dilution of the sample and plating of several dilutions. Typically, ten-fold dilutions are used, and the dilution series is plated in replicates of 2 or 3 over the chosen range of dilutions. Often 100µl are plated but also larger amounts up to 1ml are used. Higher plating volumes increase drying times but often don't result in higher accuracy, since additional dilution steps may be needed.^[5] The CFU/plate is read from a plate in the linear range, and then the CFU/g (or CFU/mL) of the original is deduced mathematically, factoring in the amount plated and its dilution factor.

A solution of bacteria at an unknown concentration is often serially diluted in order to obtain at least one plate with a countable number of bacteria. In this figure, the "x10" plate is suitable for counting.

An advantage to this method is that different microbial species may give rise to colonies that are clearly different from each other, both microscopically and macroscopically. The colony morphology can be of great use in the identification of the microorganism present.^[6]

A prior understanding of the microscopic anatomy of the organism can give a better understanding of how the observed CFU/mL relates to the number of viable cells per milliliter. Alternatively it is possible to decrease the average number of cells per CFU in some cases by vortexing the sample before conducting the dilution. However many microorganisms are delicate and would suffer a decrease in the proportion of cells that are viable when placed in a vortex.^[7]

Uses

Colony-forming units are used to quantify results in many microbiological plating and counting methods, including:

• The Pour Plate method wherein the sample is suspended in a Petri dish using molten agar cooled to approximately 40–45 °C (just above the point of solidification to minimize heat-induced cell death). After the nutrient agar solidifies the plate is incubated.^[11]
• The Spread Plate method wherein the sample (in a small volume) is spread across the surface of a nutrient agar plate and allowed to dry before incubation for counting.^[11]
• The Membrane Filter method wherein the sample is filtered through a membrane filter, then the filter placed on the surface of a nutrient agar plate (bacteria side up). During incubation nutrients leach up through the filter to support the growing cells. As the surface area of most filters is less than that of a standard Petri dish, the linear range of the plate count will be less.^[11]
• The Miles and Misra Methods or drop-plate method wherein a very small aliquot (usually about 10 microliters) of sample from each dilution in series is dropped onto a Petri dish. The drop dish must be read while the colonies are very small to prevent the loss of CFU as they grow together.^[12]

However, with the techniques that require the use of an agar plate, no fluid solution can be used because the purity of the specimen cannot be unidentified and it is not possible to count the cells one by one in the liquid.^[13]

Tools for counting colonies

The traditional way of enumerating CFUs with a "click-counter" and a pen. When the colonies are too numerous, it is common practice to count CFUs only on a fraction of the dish.

Counting colonies is traditionally performed manually using a pen and a click-counter. This is generally a straightforward task, but can become very laborious and time-consuming when many plates have to be enumerated. Alternatively semi-automatic (software) and automatic (hardware + software) solutions can be used.^[14][15][16]

Automated systems

Many of the automated systems are used to counteract human error as many of the research techniques done by humans counting individual cells have a high chance of error involved. Due to the fact that researchers regularly manually count the cells with the assistance of a transmitted light, this error prone technique can have a significant effect on the calculated concentration in the main liquid medium when the cells are in low numbers.^[27]

An automated colony counter using image processing.

Completely automated systems are also available from some biotechnology manufacturers.^[28][29] They are generally expensive and not as flexible as standalone software since the hardware and software are designed to work together for a specific set-up.^[30] Alternatively, some automatic systems use the spiral plating paradigm.^[31]

Some of the automated systems such as the systems from MATLAB allow the cells to be counted without having to stain them. This lets the colonies to be reused for other experiments without the risk of killing the microorganisms with stains. However, a disadvantage to these automated systems is that it is extremely difficult to differentiate between the microorganisms with dust or scratches on blood agar plates because both the dust and scratches can create a highly diverse combination of shapes and appearances.^[32]

See also

• Cell counting
• Growth medium
• Miles and Misra method
• Most probable number
• Replica plating
• Viral plaque

References

Further reading

• Fishman, William H.; Bernfeld, Peter (1955). [31] Glucuronidases. Methods in Enzymology. Vol. 1. pp. 262–9. doi:10.1016/0076-6879(55)01035-5. ISBN 978-0-12-181801-2.