How to construct an ON LINE analytical procedure to control the microbial growth in non-biotechnical processes like those in mining and paper manufacturing?
First of all, it would be a good idea to
map the microbiological problems of the process. They can be slime production by biofilm bacteria, biodeterioration of valuable raw materials or products, biocorrosion by sulfate reducing bacteria - these are the main subjects but, depending on the process in question, there are others, too. As an example, bacterial spores cause hazard for the hygiene of food packaging boards and papers, fermentative/anaerobic bacteria cause bad odours etc. HACCP (Hazard Analysis & Critical Control Points) examination, applied first in food industry over decades ago, helps a lot (more about HACCP in other post).
Next step (and very important one) is the
chose of the analytical method. There are several of them and many alternatives have been reviewed already in 1990 when "RAMI-90",
Sixth International Congress on Rapid Methods and Automation in Microbiology and Immunology was held in Helsinki-Espoo, Finland, 7-10. June 1990. The German institute "
Papiertechnische Stiftung" has also performed evaluations of rapid mibi methods on 80's. A summarizing article called "Microbiological Control of Pigments and Fillers" was presented by me in
PIRA Symposium, Cambridge, England in 1997 and published in the series of "The Fundamentals of Papermaking Materials". It describes the evaluations performed in Research Centre of ENSO Ltd. (currently STORA ENSO Ltd) and Helsinki University / Dept. Appl.Chem. and Microbiology). Some novel methods have appeared after this publication but, as we found in this research, ATP (Adenosine Triphosphate) analysis seems to be a valuable tool even today.
The
microbiological variables to be controlled must also be taken into account. If total growth is the main subject of the control, rapid biochemical reactions like ATP Assay or staining of the cells with fluorochromes like Acridine Orange are recommended. These two measurements are relatively simple and their results - values of light emission or fluorescence - can easily be handled as raw data, derived by optical measurements. RR (Respiratory Rate) Test is slightly more complicated because a certain incubation period of the sample is needed but it can also be automatized. Two drawbacks of this method are its low sensitivity and selectivity (only microbes with aerobic respiration can be detected). Whenever some special species (like coliforms) or groups (like SRO's = Sulfate Reducing Bacteria) are to be controlled, more time and money consuming methods like selective cultivations (in PMEU) or PCR are needed - and they are also very difficult to apply into an ON LINE control system.
Collection and use of the data derived from the processes shall also be planned. Time series with certain transformations are usually most beneficial meters to show any kind of trend in the densities of free-floating (= non-biofilm) microbes. Growth rates of single-cell microbes in water environment will usually be presented after log transformation of microbial densities which gives a straight line in xy plots when the time scale is presented with equal intervals (semi-logarithmic plot). This means that the alarm threshold should be set wisely because the amount of microbial cells increases with a factor of ten in every time unit and the period of time which are needed for growth from 10 to 100 cfu/ml or 1 000 to 10 000 cfu/ml are equal (if anything like lack of nutrients or any kind of inhibition doesn't prevent the growth). Statistical conclusions of biofilm and filamentous microbes will follow other guidelines and semi-logarithmic plot may not be the best framework to collect data. Solutions for the questions of statistical significance of rising or dropping microbial densities can be found in special textbooks like "
Statistical Methods in Biology" by Norman T J Bailey (Edward Arnold, London). A Finnish lesson about this subject, written by me for professional training centers, is also available by request.
Economical considerations will not be discussed deeper here because they are very much depending on the equipment and reagents. A very rough estimate for the price of one analysis is 10 - 30 € (which is, by the way, a relevant estimate for colony count analyses, too). When planning an automatized system, commercial instruments like luminometers may be applied to the system but they have to be modified to work ON LINE (sampling, dosing of reagents, cleaning of the detectors etc.). The costs of control are very much depending on the time schedules and too frequent sampling shall be avoided.
Conclusion: an ON LINE microbiological control system can be constructed to collect time series of microbe density data and give early warnings of hazards whenever the target organisms and their critical growth sites are mapped (HACCP) and the control method, depending of the specificity of organism(s), is chosen. Time series help to evaluate biocidic treatments, effects of the process conditions (temperature, pH, redox potential and so on), overall contamination of the process etc. and, finally, threshold levels for the alarms, based on the control experiences, can be set.
