Bacteriology 102:
The Concept of Coliforms
and their Enrichment, Detection,
Isolation and Identification
from natural and clinical samples

John L's Bacteriology Pages >
Bact.102 Website–Fall 2006 >
Page Last

Coliforms are considered indicator organisms in food and water analysis – not because they are "bad" in themselves, but because they are often associated with potential health problems. Two examples:

  • Identification of an isolate from any source as Escherichia coli would then implicate the source as being contaminated by intestinal waste, which is the habitat of E. coli. It is much easier to find whether or not E. coli is present through the enrichment and isolation procedure than it is to test for each specific intestinal pathogen (bacterial, protozoan and viral).

  • Coliforms are abundant in soil, and if they are found in well water, that would be an indication of probable surface soil contamination of the water. Usually such coliforms are identified as Klebsiella and Enterobacter.

The basic definition one encounters about coliforms is that they are gram-negative rods that ferment lactose rapidly with the production of gas – i.e., insoluble gas that is detectable in a Durham tube. A more precise definition would specify that they can grow and show gas production from fermentation in Lactose Lauryl Tryptose Broth and Brilliant Green Bile Broth at 35°C. Those which additionally do so in EC Broth at 44.5°C belong to the subset of coliforms called "fecal coliforms."

Selective Enrichment and Initial Detection

With these characteristics in mind, coliforms happen to be easy to detect with the appropriate selective-differential media in the enrichment process. These media tend to inhibit gram-positive bacteria, and the presence of coliforms (among the other organisms which would also be present) is suggested by gas from lactose fermentation in the Durham tube.

  • As the bacterial cells in the sample to be tested may have been impaired from osmotic incompatability, starvation, chlorination or other harmful conditions, it would be additionally harmful to subject them to an enrichment medium with harsh selective agents. Thus we use a two-stage enrichment where the initial medium (Lactose Lauryl Tryptose Broth) aims primarily to "revive" the damaged coliform cells such that they can increase their numbers and be detected by gas resulting from lactose fermentation.

  • The subsequent enrichment media (Brilliant Green Lactose Bile Broth and EC Broth, below) can then select more effectively against gram-positive organisms without putting undue stress on the coliform population.

The enrichment media in the following table increase in selectivity for the desired organisms from left to right. We never expect a pure culture to develop in any enrichment medium (selective or otherwise), so the subsequent use of an appropriate solid medium (for obtaining isolated colonies) is important as in any enrichment-isolation procedure. For a tube which shows growth and gas, suggesting the presence of coliforms, all it might have taken for a population of coliforms to develop in the tube could have been just one coliform cell in the inoculum. Cells of non-coliforms may or may not be able to grow in any of these media. However, one way or another in the enrichment and subsequent isolation process, they will be eliminated from consideration.

Organisms in the water sample that may grow in LLTB.Note that growth with gas indicates the possible presence of coliforms in LLTB and is confirmatory in the other two media that are inoculated from LLTB.
LLTBBGLBEC Broth at 44.5°C
The true coliforms ("fecal" and others).growth & gasgrowth & gasgrowth & gas only for fecal coliforms
Occasional strains of Bacillus and Clostridium that ferment lactose to acid and gas ("false coliforms").growth & gasno growthno growth
Some Gram-negative bacteria other than coliforms. (Pseudomonas may persist in any of these media.)growth with no gasgrowth of fewer kinds (no gas)growth of still fewer kinds (no gas)
Some rare Gram-positive bacteria other than those in the second row.growth with no gasno growthno growth

Isolation and Identification

With the use of EMB Agar (which contains lactose and gram-positive inhibitors) in the isolation process, coliforms are detected by their acidic colonies from which isolates are chosen for identification by the use of the appropriate differential media.

One must realize that coliforms do not constitute a discrete taxonomic group. Usually they are ultimately identified as various species of enteric bacteria of the genera Escherichia, Enterobacter, Klebsiella and Citrobacter, and a coliform isolate from an EC Broth enrichment is usually identified as E. coli. That does not mean that these genera are to be therefore categorized as coliforms! Out in the real world, one may encounter many "exceptional" strains of these genera that do not conform to the definition of coliform. Many pathogenic E. coli strains do not ferment lactose, and it is sometimes very difficult to differentiate them from Shigella. Also, many strains of Citrobacter very weakly attack lactose (if at all) and may be initially confused with Salmonella when isolations are made from clinical or natural samples. So, considering the absolute definition of this non-taxonomic term, there can be no such thing as a non-lactose-fermenting coliform.

Overall Summary of the Procedure



Remember the definition of coliforms:

  • These are strains that share the characteristics of being gram-negative rods and also fermenting lactose with the production of acid and gas.

  • You may have seen an extended definition which states that they ferment lactose (with visible gas production) in Lactose Lauryl Tryptose Broth (LLTB) and also Brilliant Green Lactose Bile (BGLB) Broth. These media are important in the Presumptive Tests where we are selectively enriching for them and also detecting their presence among the mixed cultures in these media; these media are selective against Gram-positive bacteria, especially BGLB.

  • We could have used LLTB and BGLB in the Completed Tests when we tested the isolates we obtained off EMB Agar, but we simply tested for acid and gas production in Lactose Fermentation Broth. (And instead of doing the gram stain, we assumed that growth on EMB Agar was indicative of being gram-negative.)

  • You may have seen in some of the literature that coliforms are "aerobic or facultatively anaerobic." This is an old-school description of what we call facultative anaerobes. They do grow as strict aerobes unless we give them an opportunity to grow under anaerobic conditions – such as fermentation. As they all ferment lactose, they will all ferment glucose as well. (If any organism ferments a sugar, it is assumed they can ferment at least glucose.)

Also remember that coliforms are not the problem but rather an easily-detectable type of organism that is associated with and therefore indicative of certain problems:

  • If a coliform is identified as Escherichia coli, that means the environment was contaminated with fecal matter. It is a lot easier to look for E. coli than any of the other intestinal pathogens which may be bacterial, protozoan or viral!

  • Remember we mentioned (and it is often forgotten) that well water is tested for coliforms. Many coliforms are found growing naturally in soil, and if they are detected in drinking water, that means that surface soil runoff is getting into the well, bringing problems that affect the water's safety.

  • It is true that certain types of E. coli and Klebsiella happen to be pathogenic, but our emphasis was on the use of coliforms as indicator organisms of problems associated with the environment in which they are found.

The term "coliform" should not imply that there is a taxonomic group consisting of coliforms:

  • There is no genus or species that is classified as a coliform.

  • Recalling the strict definitions of "species" and "strain," any particular coliform isolate is usually identified to the species level (examples: Escherichia coli, Enterobacter aerogenes), and some of these species can consist of lactose-positive and negative strains out in the environment. For example, many strains of E. coli which cause intestinal illness do not ferment lactose at all and can resemble Shigella to a great degree. Remembering the strict definition of coliforms, such lactose-negative organisms would not be called "coliforms."

Selected Groups of Bacteria
Bacteriology 102 Website
Index of the General Bacteriology Pages
Page last modified March, 2020.
John Lindquist, Department of Bacteriology,
University of Wisconsin – Madison