Controlling bacteria levels in pure water

Dr Paul Whitehead looks at effective systems for controlling the problem of bacterial levels in pure water for laboratories.

Bacterial contamination of purified water can cause major problems in the laboratory. Even if organic and inorganic chemical impurities are removed down to the limits of detection, bacterial growth can still occur, even though very pure water provides an extremely harsh environment with apparently negligible nutrient content. To avoid metallic contamination of the water, laboratory water purifiers are constructed using plastics. The bacteria can use these materials that are in contact with the pure water as a carbon food source to sustain them, and then when they die they release further contaminants into the water. If this bacterial growth is not minimised, it can cause significant difficulties in the day-to-day operation of the laboratory.

The bacteria themselves are not the only problem; they also produce endotoxins and nucleases. Endotoxins are fragments of Gram-negative cell membrane that are released during bacterial cell metabolism, and are also produced at the death of Gram-negative cells. Endotoxins a the most common pyrogens a are powerful immune stimulants, raising temperature if they are injected into the bloodstream. This can even lead to Gram-negative sepsis and death.

Nucleases present in the cells of living organisms play a number of roles in the replication of DNA and the translation of the genetic code into proteins. There are two types: DNase and RNase, which degrade or destroy DNA and RNA respectively. Any laboratory technique where water or made-up reagents will come into contact with DNA or RNA can be affected by nucleases in the water. These include gel electrophoresis, polymerase chain reactions, hybridisation probing, northern and southern transfers and DNA sequencing.

Minimising microbiological activity represents a major challenge for the design engineer, who needs to ensure that the necessary technologies and process operations are in place for making and maintaining a pure, contaminant-free water supply. But what specific operational requirements need to be considered to ensure that laboratory water purifiers are successfully controlling bacterial levels?

Many different purification technologies can be used, both alone and in series, which can remove bacteria and their by-products, such as endotoxins and RNase, that cause difficulties. While some technologies, particularly those providing a large surface area, for example ion exchange and adsorbents like activated carbon, can be very efficient at removing organisms and by-products to start with, once they start to saturate, they can allow large concentrations of these contaminants to be eluted into the system. Ultraviolet light, microfilters and ultra-microfilters are excellent at removing micro-organisms, but less good at removing endotoxins. The only method that has been proved to be highly efficient at removing endotoxins is an ultrafiltration.

ELGA LabWater's PURELAB Option R, fed with mains water, uses reverse osmosis filtration, and exposure to UV light combined with recirculation to minimise bacterial numbers in its purified water. An optional point-of-use filter can also be added. Another system, the PURELAB Ultra Genetic is fed with pre-purified water to produce ultra-pure water. The water is recirculated through UV light and an ultrafilter, giving ultra-pure water with minimal bacterial and endotoxin content. Again, an optional point-of-use filter is available (Fig.1).

When assessing the ability of alternative filters to remove contaminants, challenge tests are the standard method used. Essentially, the microfilter or ultrafilter is exposed to a huge dose of, for example, bacteria ­ far higher than is ever likely to occur in practise and the levels of bacteria, measured in the outlet water. If the concentrations measured are acceptable, then the filter can be concluded to be effective at removing the much lower levels usually seen.

Table 1 shows the results of a typical challenge test on a 0.2 micron membrane filter. While the bacterial removal is excellent, the reduction in endotoxins is limited. This sort of filter is designed to be used at point-of-use to provide a final bacterial removal, and prevent back-contamination into the system, as the outlet (along with the storage reservoir) is by far the most vulnerable part of the system for airborne contaminants to enter. The filters can usually be autoclaved, or replaced frequently.

The results from a similar test carried out on an ultrafilter are shown in Table 2. It clearly demonstrates that ultrafiltration with a low molecular weight cut-off membrane is extremely successful at removing both bacteria and endotoxins. This technology is typically operated dead-headed, with an occasional high flowrate rinse-to-drain to remove filtered material from the system.

Ultraviolet irradiation is also very effective at destroying micro-organisms. Although it is not a barrier process, relatively low energy doses of ultraviolet light greatly reduce overall bacterial levels, minimising the challenge on downstream purification processes. UV light with a wavelength of 254nm inactivates bacterial cells, preventing them from replicating. It also causes some photo-oxidative degradation of endotoxins, but light with a shorter wavelength is needed to maximise this effect. Results of a challenge test using UV at wavelengths of 254nm and 185nm are given in Table 3, showing good reduction in endotoxins and especially bacteria.

RNase and DNase are particularly troublesome to remove from water for molecular biological applications. Autoclaving removes DNase but not RNase and, while chemical treatment with DEPC will remove both, it is toxic, expensive and time-consuming, while generating ionic and organic contamination. However, the combination of an ultrafilter and irradiation with 185nm UV light removes both enzymes, as illustrated in Fig.2 and 3, which give the results of RNase and DNase assays respectively.

Because bacteria breed rapidly in static water, a dynamic water system is infinitely preferable to storing the water statically in a reservoir. Recirculation disrupts the establishment of colonies and enables repeat treatment to ensure that the background level of organisms remains low. All of the purified water ­ including the contents of any reservoir ­ is recirculated through some form of active technology. As the recirculation process creates energy and hence warms the water - giving better growth temperatures for any remaining organisms, periodic recirculation is often used to minimise the warming effect. Contamination levels well below the target specification of 1 CFU/ml are routinely seen over long periods using this approach.

There is a need to periodically chemically sanitise the water purification system. As formaldehyde is becoming less acceptable as a sanitant because of associated health concerns the most common approach is to use an oxidant such as chlorine, hydrogen perioxide or peracetic acid. As well as killing bacteria directly they can also serve to disrupt biofilm within the system. It is essential that as much as possible of the system is sanitised in this way. This will minimise subsequent regrowth and reduce the need for frequent sanitisations.

When used in conjunction with planned maintenance regimes and sanitisation protocols, low bacterial and endotoxin specifications can be achieved throughout the operational life of the water purifying system. Ultrapure, effectively bacteria-free, water is an essential commodity in most laboratories. With correct design and operation and careful application of protocols, low microbial specifications can be assured.

Dr Paul Whitehead is R&D Laboratory Manager with ELGA LabWater. e-mail: info@elgalabwater.com. www.elgalabwater.com

Recent Issues