Feature

Deep in the Bowels

DNA fingerprinting paints an interesting and turbulent picture of the bacteria of the human gut.

Anne McCartney and Gerald Tannock

Bacteria may be small creatures, only about 0.5 mm in diameter, but there is no escaping them. All humans whether rich or poor, tall or short have a normal microflora -- an extremely large and diverse collection of microorganisms which inhabit the human body even when it's in full health. In fact, it is estimated that the bacterial cells (10¹⁴) outnumber the human cells (10¹³) by approximately ten to one.

The largest proportion of the normal microflora of humans is found in the large intestine, otherwise known as the colon. The members of this bacterial population consist largely of anaerobes or bacteria that do not require oxygen for life. Found here are species belonging to the genera Bacteroides, Bifidobacterium, Eubacterium, and Clostridium.

Why worry about these bacteria at all? After all, if healthy humans host a large collection of microbes surely it can't be bad -- "Nature knows best", as the old adage goes. It is a long-standing and common belief that the normal microflora is beneficial to their human host. For example, colonization of the intestinal tract by the members of the normal microflora protects the individual from invading pathogens (disease-causing bacteria). This idea is called "microbial interference", and studies have shown that some bacteria do inhibit the growth of other bacteria. Mechanisms of microbial interference range from the physical aspect (this patch is not big enough for the two of us -- one organism out-competes the others) to antagonistic aspects (one kills the other or produces metabolic end-products that inhibit the other's growth).

Additionally, the normal microflora is of interest because some members, such as certain Bacteroides species, are known to produce disease if they become established in other sites in the body. For example, if during surgery a puncture is made in the intestinal wall and gut contents escape into the surrounding tissues, bacteria from the normal microflora may cause an infection, known as post-surgical sepsis. Likewise, some bacterial species commonly found in the normal microflora of the colon are involved in urinary tract infections.

The normal gut microflora is also of interest because of a possible relationship between bacterial enzymatic activity and the development of cancer. Do the members of the normal microflora play a role in causing colon cancer? Some studies suggest that there is a correlation between low concentrations of plant fibre in the diet and an increased likelihood of colon cancer occurring. Do higher fibre diets result in changes in the bacterial composition of the normal microflora and/or the enzyme production of the bacterial species present? Without information on the normal microflora under standard, everyday life conditions it is impossible to accurately determine the effects of any change in diet.

Eating Bacteria

Fermented milk has, for many years, have thought to affect the biochemistry of the colon contents. Lactic acid bacteria (LAB) are used in the dairy industry to make fermented milk products such as cheese. Yoghurt makers include lactobacilli in starter cultures for yoghurt production. In recent years, new types of yoghurts containing additional LAB such as "bifidus" and "acidophilus" (Bifidobacterium and Lactobacillus, respectively), that are considered beneficial members of the normal microflora, have appeared. A big question is what effect the consumption of such products by healthy humans has on microbial interference, bacterial populations and enzyme activities?

Such questions are of interest not just to researchers, but also to organisations which can make use of such information commercially. The New Zealand Dairy Research Institute, for example, has supported our studies.

At present, we lack the basic scientific facts needed to enable satisfactory evaluation of the effects of dietary modifications. Previous dietary studies have involved separate groups of individuals whose dietary intakes differed in one particular area (for example, high versus low fibre diets). However, what is the variation in the composition of the normal microflora between individuals, even when they are on similar diets? Without information on variation between individuals, we cannot ensure that the differences observed between subjects from different diet groups are really due to the dietary changes. There is, therefore, a need to study in detail the composition of the normal microflora.

Our work so far has concentrated on the Bifidobacterium population. Bifidobacteria are branching LAB that are numerous in the human colon. A genetic fingerprinting technique, known as ribotyping, has been used to study the bifidobacterial composition of faecal samples obtained from two healthy individuals over a twelve-month period. Results from these studies have shown that multiple strains of bifidobacteria were present in both individuals. Additionally, variations in the strain composition were demonstrated over a period of time for both individuals, with a marked difference in the complexity of the bifidobacterial population of the samples between the two subjects. Certain strains were demonstrated consistently for both individuals while other strains appeared to be of a more transient nature (appearing only once or appearing, disappearing, and reappearing).

The observation of such variation under standard conditions is important when analysing results from dietary modification studies. Such a variation demonstrates the difficulty in accurately monitoring the effects of dietary modification on the intestinal microflora of humans, and indicates the need for each individual to act as their own "control". That is, a baseline must first be obtained for each subject under standard dietary conditions, over a period of time, prior to making any dietary modifications. This would enable an accurate comparison of the effect of the different dietary conditions to be made.

With the use of DNA fingerprinting techniques we have been able to demonstrate the complexity of the bifidobacterial population of humans, and have raised many more questions concerning the relationship between humans and their bacterial associates.

Anne McCartney is a PhD student in the Department of Microbiology at the University of Otago, Dunedin.
Gerald Tannock works in the Department of Microbiology at the University of Otago, Dunedin.