Quantification of adenosine triphosphate (ATP) in fuels and fuel-associated waters was first presented at the Technische Akademie Esslingen 6th International Fuels Colloquium in 2007. At the time, two issues limited the overall usefulness of ATP as a test parameter: inability to differentiate between bacteria and fungi and inability to detect dormant microbes. Recent research has addressed both of these issues. This paper presents protocols for detecting dormant microbes - identified as microbes that are not metabolically active in the sampled fluid, but which can become active under appropriate conditions - and for differentiating fungi from bacteria. The newly developed protocols achieve >90% detection of bacterial endospores in fluid samples. They also provide >90% differentiation between bacterial and fungal contaminants in these fluids.
The measurement of adenosine triphosphate (ATP) in fuels and fuel-associated water, using ASTM Method D7687 , has proven to be a fast, accurate and precise tool for quantifying microbial contamination in fuel systems [2, 3]. However, the ASTM test method has two limitations: it does not detect microbes that are dormant in the sample; nor does it differentiate between bacteria and fungi. Dormant microbes include fungal spores, bacterial endospores and persistor cells. This paper presents protocols for detecting dormant cells and for differentiating between bacteria and fungi.
Adenosine Triphosphate Testing
ASTM Method D7687 was used for all ATP tests. Briefly, a 5 mL sample was pressure filtered through a 0.7 µm in-line, glass-fiber, depth filter. The retentate was then washed with a proprietary rinsing solution (all ATP test reagents are from LuminUltra Technologies, Ltd., Fredericton, NB, Canada) and air dried. A proprietary lysing regent was then used to extract ATP into a 17 x150 mm reaction tube. The 1.0 mL extract was then diluted with 9.0 mL of a proprietary buffer. Finally, 100 µL of diluted ATP extract was mixed with 100 µL of Luciferin-Luciferase reagent in a 12 x 55 mm cuvette, and the cuvette was placed into a luminometer to obtain a relative light unit (RLU) reading. All RLU data were converted to pg ATP mL-1 by comparison with the RUL of a 1.0 ng mL-1 ATP reference solution. For the dormant cell detection effort, pg ATP mL-1 data were transformed to Log10 pg ATP mL-1.
In order to investigate the effect of the surfactant selected for the bacterial-fungal differentiation protocol, a 1:1 mixture of fungi and bacteria (see Bacterial and Fungal Contaminant Differentiation; Test Microbes, below) was tested by two protocols: ASTM D7687 and a second protocol in which 1.0 mL aliquants of the mixture are dispensed into each of two reaction tubes. One reaction tube contained 1.0 mL of the aforementioned proprietary lysing agent. The second reaction tube contained 9.0 mL of a proprietary stabilizing reagent. After vigorous shaking and one-minute standing, the contents of the first reaction tube were transferred to a reaction tube containing 8 mL of the proprietary buffer mentioned in the previous paragraph plus resin beads. This tube is inverted several times to ensure complete but gentle mixing. Once the resin beads have settled, a 100 µL portion was transferred to a 12 x 55 mm cuvette and reacted with 100 µL Luciferin-Luciferase reagent to obtain RLU and pg ATP mL-1, as described above. Similarly, a 100 µL portion of the stabilizer-diluted subsample was transferred to a 12 x 55 mm cuvette and reacted with 100 µL Luciferin-Luciferase reagent to obtain RLU and pg ATP mL-1, as described above. The results yielded Total ATP and Dissolved ATP, respectively. In context of this work, Total ATP - tATP - is defined as the ATP detected in a sample from which no effort is made to separate whole cells from cell-fragments or extracellular ATP before the lysis step. Dissolved ATP - dATP - is ATP detected in a sample in which cell lysis has been inhibited. Cellular ATP - cATP - is computed as the difference between tATP and dATP, and biomass stress index (BSI%) is computed as the percentage ratio of dATP to tATP.
Detecting Dormant Cells
A commercially available, Bacillus thermogenesis, spore paste preparation ((BTK; Safer's BTK Biological Insecticide; Woodstream Corp, Brampton ON) was used to model a dormant bacterial population. Based on the protocol described by Min et al. , BTK was suspended in liquid media to give ~1.3 x 107 IU mL-1. During preliminary experiments, ATP was tested immediately after BTK suspensions were prepared and after suspensions had incubated in a 37° C water bath for 15 min. Ultimately, the 37°C immersion period was optimized at 30 min. Four media were compared for their effect on endospore germination: brain-heart infusion broth (BHI; DIFCO, Becton Dickinson, Franklin Lakes, NJ), LiquiCult TM broth (LCB; MCE, Inc., Lake Placid, NY), nutrient broth (DIFCO) and trypticase soy broth (TSB; DIFCO). After determining the broth that produced the greatest post-induction yield, ATP was tested as a function of germination induction time at 37°C in TSB. All testing was performed in duplicate or triplicate.
Bacterial and Fungal Contaminant Differentiation
Preliminary method development experiments were conducted using Saccharomyces cerevisiae; selected for its size (5 to 10 µm dia), single-cell morphology and commercially available as baker's yeast (Fleischmann's - Active Dry Yeast; ACH Food Companies, Inc., Memphis, TN, USA), suspended in Sabouraud-dextrose broth (Oxoid, Ltd, Basingstoke, Hampshire, UK) as the representative fungus. An uncharacterized, Gram negative bacterium, previously isolated from a contaminated metalworking fluid (MWF) sample, was used as the representative bacterium. For subsequent experiments, two filamentous fungal species (tentatively identified as Aspergillus niger and Penicillium chrysogenum) which had previously been isolated from contaminated MWF were used.
The following filter media were evaluated: glass fiber filters mixed cellulose esters (MCE; 5 µm; EMD Millipore Corporation, Billerica, MA, USA), polycarbonate (PC; 12 µm, 10 µm and 8 µm; EMD Millipore Corporation, Billerica, MA, USA), polyvinylidene difluoride (PVDF; 5.0 µm; EMD Millipore Corporation, Billerica, MA, USA) and polytetrafluoroethylene (PTFE; 5.0 µm; EMD Millipore Corporation, Billerica, MA, USA). The PVDF and PTFE filters were supplied as in-line, syringe filters in housings. The other filtration media were 25 mm dia filter disks. Filter disks were placed into Millipore in-line filter housings for use. During testing, each of the described filters were placed in series with the ASTM D7687 standard in-line filter so that the larger pore-size filters were positioned upstream of the standard D7687 filter. In order to test filtration efficiency, sample ATP was tested on both filtered and unfiltered portions. Efficiencies were computed as ratios of duplicate tests before and after filtration. The target filtration efficiencies were <10 % of total ATP being recovered in the yeast filtrate (>90% in the yeast retentate) and >90% of total ATP being recovered in the bacterial filtrate (>90% passing through the larger pore-size filter and being captured by the D7687 filter).
Two surfactants were evaluated for their ability to disaggregate bacterial flocs without lysing either bacteria or fungi. Surfactant A was tested at 1.8%, 0.9%, 0.45% and 0.045% (v/v), and surfactant B) was tested at 0.8%, 0.4%, 0.2% and 0.1% (v/v). In all cases, 10x working stocks were prepared in deionized water. Performance was evaluated by comparing biomass retention of 5.0 mL untreated MWF with biomass retention of 5.0 mL of MWF to which 0.5 mL of surfactant solution had been added and shaken vigorously for 30 sec, before filtration.
Dr. Frederick Passman, PhD is a Certified Metalworking Fluids Specialist with over 35 years experience in Environmental & Industrial Microbiology. His company, Biodeterioration Control Associates, Inc. (BCA) provides clients with unparalleled expertise in Microbial Contamination Control.
©Copyright - All Rights Reserved
DO NOT REPRODUCE WITHOUT WRITTEN PERMISSION BY AUTHOR.