Summary of a Research project carried out to investigate the use of ATP detection using Bioluminescence as an Alternative Method to Total Viable Counts in Determining Microbial Contamination of Water Systems
Research carried out by David Tovey: B & V Water Treatment
Summary report compiled by Jillian Cooper: B & V Water Treatment
Summary
ATP detection using bioluminescence can be used to determine microbial contamination in water. It is simple to perform and offers results in minutes rather than days taken with traditional plate count methods. The detection limit for a dip swab ATP test was found to be around 105 bacteria/ml. This rather high limit of detection does limit the use of this technology in some areas. The testing carried out also investigated any interference due to the chemical composition of various chemicals used in water treatment with the ATP test method. Significant interference was found with one particular component regularly used in the formulation of cooling water inhibitors.
Aims and Objectives
To determine if the ATP method shows a significant correlation to standard plate count methods and if it is a suitable alternative in testing water systems for microbial contamination.
To ascertain the detection limits of the ATP method and determine the variability of results produced.
To investigate whether the components in standard water treatment chemical inhibitor formulations interfere with ATP results obtained.
ATP Detection using Bioluminescence
As ATP is present in all cells, testing for its presence can determine biological contamination. Luciferase catalysis the following reaction:
ATP+ D-Luciferin + 02 →Oxyluciferin + AMP + Pyrophosphate +C02 + light (560nm)
When ATP is the limiting component in the reaction the proportion of light emitted is proportional to the concentration of ATP. Many commercially available kits are available to test for ATP using the luciferase reaction. A luminometer is used to determine the amount of light that is emitted which is expressed as relative light units (RLU). Luciferase produces light by converting the chemical energy of luciferin oxidation through an electron transition, forming a molecule of oxyluciferin. Firefly luciferase is a 61KDa monomeric protein which requires Mg2+ and the hydrolysis of ATP to ADP + inorganic phosphate to provide energy.
Microbiological monitoring of cooling water systems
Cooling tower systems are open to external contamination and the make up water is often not from a mains supply. Weekly monitoring of microbial contamination is performed to asses the effectiveness of the water management system. The make up water should be tested quarterly and the cooling water weekly. Microbial monitoring is undertaken using an aerobic counting method. A result of less than 104 bacteria/ml indicates that the system is under control and that the water management system is adequately controlling the proliferation of bacteria. A count of greater than 104 bacteria/ml and up to 105 bacteria/ml indicates that the water management system needs reviewing and corrective action should be undertaken to control microbial numbers. A count of greater than 105 bacteria/ml indicates that the system is out of control and the water treatment system needs immediate attention. (Health and Safety Commission 2000). ATP monitoring of these systems will provide immediate results which will give an indication of microbiological levels in the system. The research carried out indicated that it is not possible to determine accurately any levels below 105 bacteria/ml with this monitoring method. The ATP method of monitoring microbial levels in cooling towers is therefore useful purely as an indication of levels greater than105 bacteria/ml which would require attention and with this rapid method the results could be acted on immediately.
The water within cooling towers is also tested for the number of Legionella spp present. (Health and Safety Commission 2000) It is also stated that an alternative method to aerobic plate count may be used to monitor microbial contamination but these methods must be validated to show a clear relation to results obtained from traditional plate count methods.
Water Treatment chemicals tested for possible interference:
Three different corrosion inhibitor formulations commonly used in cooling tower systems were tested to determine whether their chemical composition interfered with the accuracy of the ATP results:
Formulation 1 Silicate based soft water corrosion inhibitor
Formulation 2 Molybdate based soft water corrosion inhibitor
Formulation 3 Zinc based soft water corrosion inhibitor.
When standard use concentrations of formulation 2 and 3 were added to the test cultures there was no significant difference obtained in the relationship already established between cfu/ml and RLU. It was therefore concluded that these formulations did not contain any chemical component which interfered with the ATP reaction.
A significant reduction in RLU when compared with cfu/ml was obtained when standard use concentrations of formulation 1 were added to all test cultures used. Further work was therefore carried out in order to determine which component of this formulation had an inhibitory effect on the ATP reaction.
Initial work indicated that the interference was due to the phosphate component of the formulation so research was carried out on the effect of the following phosphate based chemicals and other commonly used raw materials in order to obtain data on their inhibitory effect on the ATP reaction:
Chemicals tested for their potential inhibitory effect on the ATP reaction
HEDP - 1-hydroxyethylene-1,1-diphosphonic acid B - Benzotrizole,
NTA - Surfact PD38 PP - Potassium pyrophosphate
SHP - Sodium hexametaphosphate PS - Potassium silicate
SHO - Sodium hydrogen orthophosphate
STP - Penta sodium triphosphate
The results above would indicate that Potassium Pyrophosphate, a commonly used constituent in many cooling tower scale and corrosion inhibitors, has a significant inhibitory effect on the ATP reaction even at a concentration of 10ppm. This test was also carried out with cultures of other commonly found bacteria and the same inhibitory effect on the RLU level obtained in relation to cfu was noted.
No inhibitory effect on the ATP reaction was found from the addition of SHO – Sodium hydrogen orthophosphate or STP- Penta sodium triphosphate to the test cultures.
Conclusions:
The chemical structure of potassium pyrophosphate (chain of 2 inorganic phosphates) is similar to ATP which has a chain of 3 inorganic phosphates.
Chemical structure of ATP (Adenosine Tri Phosphate)
Chemical structure of Potassium pyrophosphate
Pyrophosphate is an intermediate in the reaction below and therefore is an integral part of the energy generation reaction within living cells.
ATP → AMP + PPi
The reaction catalyzed by firefly luciferase takes place in two steps, with Pyrophosphate being a product of the initial 1st step of the reaction
It is possible that the interference with the ATP reaction demonstrated by the addition of Potassium Pyrophosphate to the test cultures is due to competition with ATP for the catalytic site on luciferase. The actual method of inhibition of the ATP reaction by Potassium pyrophosphate though is not fully understood at this time.
Limitations of the ATP Method
1). The amount of intracellular ATP varies between different species of micro-organisms and changes depending on the metabolic activity i.e. there may be up to a 70-90% reduction in intracellular ATP between the exponential growth phase and the stationary phase of growth. This means that it is impossible to accurately quantify the relationship between cfu and RLU.
2). Luciferase reacts optimally at pH 7.73 and at a temperature of 23-25oC. Salts and non-ionic chemicals may impair light production by changing the pH of the reaction, the test also has to be conducted at a suitable temperature for optimum luciferase activity. It is difficult, if not impossible, to ensure that the above exact temperature and pH optimals are always adhered when these tests are carried out by operatives working on sites.
3). It is possible to test for non-microbial ATP and then total ATP. From these results it is possible to calculate the levels of microbial ATP. Unless this separation procedure is introduced to the test there is no discrimination between intracellular /extra cellular and prokaryotic /eukaryotic ATP in the sample. The test could therefore pick up the ATP present in pollen, general organic debris etc as well as microbial ATP and produce a high reading which is not related to Microbial levels present.
4). ATP Swabs must be stored between 2-8oC and allowed to reach room temperature before being used. This may be difficult to achieve and may not always be adhered to which may affect results.
5). ATP swabs will detect both aerobic and anaerobic bacteria. Anaerobic bacteria will not grow on dip slides or when standard TVC testing is carried out in a laboratory. This fact makes it even more difficult to correlate RLU to aerobic TVC.
6). ATP bioluminescence monitoring will have very little value in systems which generally contain low levels of bacteria such as potable water and down water services as the limit of detection of this method even under ideal laboratory conditions was found to be 1.5 x 105 bacteria/ml
7). Water treatment chemicals may contain components that inhibit ATP detection resulting in inaccurate reporting. It is essential to ensure that where ATP is used as a monitoring method, particularly in open cooling systems, that Potassium pyrophosphate is not present as part of the inhibitor formulation. The presence of Potassium pyrophosphate will very significantly reduce the RLU results obtained giving a falsely low result. This may, if ATP is the only method of microbial monitoring, give cooling tower operatives inaccurate results and suggest that bacterial levels are within acceptable limits when this is not actually the case.
Benefits of the ATP Method:
Traditional plate counting methods take 3 days for results to become available and this may be further increased by laboratory reporting times. If high microbial contamination is detected from plate counting methods it may be too late for any corrective actions to be implemented. As long as the above limitations are taken into consideration ATP monitoring of some systems, generally in conjunction with standard dip slide monitoring, allows for rapid detection of high levels of contamination so problem areas are highlighted quickly and corrective actions can be implemented straight away.
