Water - Microbiological Analysis of Water
- Waterborne pathogens are a leading cause of disease and death worldwide
- Routine microbiological testing of drinking water supplies, recreational waters and environmental waters is essential for the protection of public health
- Microbiological water testing is based on the detection of indicators of faecal contamination rather than specific pathogens
- Most routine laboratory methods are now based on membrane filtration
According to the World Health Organisation (WHO) more than 3.4 million people die each year from waterborne disease, most of whom are young children. In addition, it is estimated that around 50% of the population in developing countries is suffering from a water-related disease at any one time. That makes infections contracted from contaminated water supplies a leading cause of illness and death worldwide and helps to explain why the provision of safe drinking water is such a high priority for governments and aid agencies. The great majority of waterborne infections are the result of faecal contamination, either from humans or animals, and preventing sewage and agricultural effluent from entering water supplies is the key to safe drinking water.
The infrastructure and treatment procedures needed to provide safe water supplies are well understood and have been part of life in most developed countries for so long that safe drinking water is often taken for granted. But maintaining an uncontaminated water supply requires constant attention and regular monitoring by a programme of testing. The most effective way to check water supplies for faecal contamination remains microbiological analysis and a range of test methods designed for that purpose has been developed for the water industry. Testing is important not just for drinking water supplies, but also for recreational and environmental waters where human contact with contaminated water could occur.
Microbiological water testing
A large range of pathogenic microorganisms could be present in contaminated water and some of the more important water-borne pathogens are listed below.
Aeromonas spp, Campylobacter spp, Clostridium spp., Escherichia coli (including VTEC types such as O157), Legionella spp., Leptospira spp., Pseudomonas aeruginosa, Salmonella enterica, Shigella spp., Vibrio spp. and Yersinia spp.
Adenoviruses, Hepatitis A, Noroviruses, Poliovirus, Rotaviruses
Cryptosporidium spp., Giardia spp.
Testing for all these pathogens directly is not practical, partly because of the difficulty and expense of conducting such comprehensive testing on large numbers of water samples, but also because pathogens tend to be present only in low numbers even in heavily contaminated water. Since the majority, though not all, of the species listed above are present in water as a result of contamination with human or animal faeces, it is usual to examine water samples for evidence of such faecal contamination by testing for the presence of so-called ‘indicator’ bacteria.
Indicator organisms are bacterial species that are present in high numbers in human and animal faeces and are also present in sewage effluent. Ideally they should not originate from other sources, should not be able to multiply in water supplies or aquatic environments and should be relatively simple to isolate from water samples. Unfortunately, no one organism completely meets all these requirements, but it is E. coli that most closely matches the criteria for an ideal indicator species.
Other members of the coliform group, such as Klebsiella spp. and Enterobacter spp. are also used as faecal indicators, but are not of exclusively faecal origin, unlike E. coli. Other species that are present in faeces in lower numbers may also be used as indicator organisms, notably enterococci and to a lesser extent Clostridium perfringens.
The basic framework of microbiological testing of drinking water supplies, treated and untreated recreational waters and environmental waters is built on the detection of these indicator organisms and decades of use has demonstrated the effectiveness of this approach in maintaining safe water supplies. Tests for total coliforms and faecal coliforms (coliform species able to grow at 44oC) are used routinely to screen samples for faecal indicator species. However, it should be noted that the absence of indicators does not guarantee the absence of pathogens any more than their presence can be taken to mean that pathogens are necessarily present. Neither pathogens nor indicator organisms can survive indefinitely in water, especially in environments subject to chlorination or high levels of ultra-violet radiation.
There are some situations where it is necessary to test directly for water borne pathogens. For example, public drinking water supplies are routinely monitored for the presence of the protozoan parasite Cryptosporidium, which has caused water borne disease outbreaks in many developed countries. Testing for Cryptosporidium in water samples is a specialised operation and is not considered further here. The same applies to detection of Legionella spp. in water where aerosols may be generated, such as contained in cooling towers and humidifiers. However, testing for Pseudomonas aeruginosa, especially in treated recreational waters, such as swimming pools, is included below.
Water quality parameters
In addition to tests for indicator organisms and certain specific pathogens, non-selective colony counts are also routinely carried out to determine the population of heterotrophic bacteria present. Counts at two temperatures (22oC and 37oC) are typically performed to provide information on the general microbiological population of the water and detect sudden changes in water quality. Counts at 37oC have been used to indicate faecal contamination in the past, but this is not generally considered to be reliable.
Obtaining representative water samples is a critical part of microbiological water analysis. Samples should be collected in sterile containers, which for chlorinated water should contain an appropriate quantity of sodium thiosulphate to neutralise residual chlorine. It is also important to ensure that the sampler does not contaminate the inside of the sample container and rubber gloves should be worn where necessary. Ideally water from piped distribution systems or tanks should be taken from hygienically designed sample taps. Bacterial growth may occur in taps and it is good practice to disinfect the tap with alcohol or another suitable disinfectant before sampling. Water should be allowed to run through the tap for several minutes to flush out any contamination within the tap and ensure that the sample is representative. The samples should be tested as soon as possible after collection. There are many considerations to be borne in mind when sampling water and these vary with the type of sample being taken and the location. Comprehensive official advice on sampling from distribution systems and other water sources is available in published guides, such as the UK Environment Agency booklet The Microbiology of Drinking Water (2010) - Part 2 - Practices and procedures for sampling and the US EPA’s Interactive Sampling Guide for Drinking Water System Operators, which includes procedures for microbiological sampling. Staff required to take water samples for microbiological analysis should be trained according to the principles outlined in such publications.
In recent years, water microbiologists have become increasingly aware of the importance of biofilms for microbiological populations in water systems. Biofilms are now recognised as complex microbial communities, which form on surfaces. Biofilms typically consist of a variety of microbial cells, potentially including pathogens, within a matrix composed of exopolysaccharides (EPS) secreted by certain bacterial species. Biofilms develop over time, becoming more complex and extensive, and can protect individual bacterial cells from chlorine and other antimicrobial compounds in water. Biofilms are also notoriously difficult to remove from surfaces and can act as a sporadic source of microbial contamination as bacterial cells are sloughed off from the matrix into the surrounding water. It is now recognised that most of the bacteria in drinking water distribution systems are present within biofilms rather than free living in the water itself. Pathogens isolated from within biofilms include Salmonella Typhimurium, Campylobacter, Pseudomonas aeruginosa and Aeromonas hydrophila. Biofilms may affect general microbiological water quality, cause objectionable tastes and odours and accelerate corrosion within distribution systems. The presence of significant biofilm growth may make it difficult to obtain representative water samples and may influence the results of microbiological analysis. High heterotrophic plate counts may be indicative of biofilm formation in distribution systems. In some cases it may be necessary to sample biofilms directly using swabs or by allowing a film to develop on the surface of removable metal coupons or within specially designed sections of pipework.
Microbiological test methods
Traditional culture techniques using pour and spread plate count methods are not sufficiently sensitive for the detection of indicator organisms and pathogens in water, although they are still used routinely for enumerating heterotrophic bacteria. Methods capable of testing a larger volume of water (typically 100 ml) are needed. For many years the method of choice was the multiple tube ‘most probable number’ (MPN) technique, in which measured volumes of the water sample are added to a series of tubes containing differential media and incubated. Growth is indicated by a colour change in the medium and the result is calculated from the distribution of positive tubes. Although the method is simple and inexpensive in terms of equipment and materials, it is labour intensive and requires large amounts of incubator space. It is also an indirect method and does not allow the further examination of individual colonies. MPN tests for routine water microbiology have now been largely replaced by membrane filtration (MF) methods, although they may still be useful for occasional tests conducted in small laboratories or in the field, and commercial test kits based on MPN methods are available for coliforms and enterococci.
A typical MF method for water analysis is performed by passing a known volume of water through a sterile membrane filter with a pore size small enough to retain bacterial cells (typically 0.45µm). The filter is then transferred aseptically to the surface of an agar plate, or an absorbent pad saturated with a suitable selective medium, and incubated. Colonies are allowed to develop on the surface of the filter and can be counted and examined directly. MF methods are quick and easy to perform, require little incubator space and can handle very large volumes of water if required. Over the last 30 years they have become the preferred methods for microbiological examination of water for indicator organisms. There are a number of official published methods based on MF, notably a series of ISO methods, such as ISO 9308-1 for coliforms and E. coli and ISO 7899-2 for enterococci. The US Environmental Protection Agency (EPA) has also published official microbiological methods for water testing. Laboratories routinely testing drinking waters, recreational waters and environmental samples should use the appropriate official method recommended by their local enforcement agency. Laboratories testing water supplies for industrial use, such as food processing, are advised to use the same methods when the water supply is required to be of potable quality. In terms of equipment, MF methods require suitable filtration apparatus consisting of a base supporting a porous disc, on which the filter is placed, and a sterile filter funnel, which can be secured to the base, clamping the filter in position. A variety of filtration systems are commercially available, including multiple units in the form of manifolds, so that more than one sample can be filtered at once. Disposable single-use sterile filter funnels are also available for convenience. Suppliers include Millipore, Sartorius and Membrane Solutions. The filter unit is connected to a suitable vacuum source to draw the samples through the filter. This may be a vacuum line or a stand-alone pump unit. Compact pump units specifically designed for use with MF methods are now available. An example is the Millipore EZ-Stream unit, which can run the filtered sample straight to drain, thus saving time spent emptying and cleaning the waste sample containers used with traditional laboratory vacuum pumps. A variety of membrane filters are available for different applications, but microbiological water analysis is typically carried out with 47mm diameter mixed cellulose ester-based filters of 0.45µm pore size. The filters are usually marked with a grid to aid colony counting.
Many selective media have been developed for the detection of indicator organisms in water by MF methods. Recommended media for coliforms and E. coli include membrane lauryl sulphate broth or agar, MI agar and broth and membrane lactose glucuronide agar. Membrane enterococcus agar (mEA) and membrane-Enterococcus Indoxyl-ß-D-Glucoside Agar (mEI) can be used for detection and enumeration of enterococci, while Tryptose sulphite cycloserine agar without egg yolk can be used to culture Clostridium perfringens on membrane filters. Pseudomonas aeruginosa can also be detected by an MF method using Pseudomonas agar. Further culturing, or biochemical testing can then be used to confirm the identity of suspect colonies growing on filters placed on selective media. Increasingly, chromogenic and fluorogenic media are being used in water microbiology. Based on the detection of specific enzymes in the target bacterial species by substrates containing chromogenic or fluorogenic groups, producing highly diagnostic coloured colonies, these media can be less harsh than other selective media, resulting in fewer false negative results, and reduce the time needed to confirm results. Examples include the ChromoCult® media range produced by Merck and the BBL™ range of prepared chromogenic media for water testing.
Although most official methods for microbiological water analysis still rely on traditional culture methods and MF methods, the time taken to obtain results, typically 24-48 hours, has focused attention on alternative rapid methods. Techniques such as flow cytometry and immuno-magnetic separation have been investigated, but molecular biology-based methods, notably those utilising quantitative PCR (QPCR) technology, have shown particular promise, both for the detection of indicator organisms like E. coli, and specific pathogens. Combining MF with QPCR detection and enumeration has been shown to be a particularly rapid and effective means of analysing water samples. The main disadvantage of this method is that it may detect non-viable cells and overestimate the population, but it seems likely that QPCR-based methods will become increasingly important in water microbiology, leading to the development of commercial products similar to those already used for food analysis.
If you would like to get the latest on water monitoring products sent to your email, please subscribe to the rapidmicrobiology eNewsletter.