Introduction to BioTransformation Rule Prioritization

On May 6-7, 2005, at the University of Minnesota, a panel of biodegradation experts was convened, including: Ron Crawford, University of Idaho; Joanna Jaworska, Procter & Gamble, Inc.; Ken Hammel, University of Wisconsin; Sol Resnick, DOW, Inc.; Jim Spain, Georgia Tech; Alfred Spormann, Stanford; Larry Wackett, University of Minnesota. Prior to the meeting, they had been assigned sets of 50 rules for biotransformation from the University of Minnesota Pathway Prediction System. Each rule was prioritized independently by at least two experts. At the workshop, those rules with wider variance in scores were discussed; a consensus score was obtained in all cases. Two panelists agreed to prioritized future rules.

The reaction rules to be evaluated are considered for biodegradation under aerobic conditions, in soil (moderate moisture) or water, at neutral pH, 25°C, with no competing or toxic other compounds.

The experts use a five-point Likkert scoring scale; instructions for its use are:

1. Very likely reaction. This is to be reserved for reactions that will almost certainly occur and occur with the highest priority. For example, if an acid chloride is generated, these compounds almost invariably undergo spontaneous hydrolysis in water very rapidly. So this would likely occur as the next step in any metabolic pathway in any bacterium. EAWAG-BBD btrule bt0026, Acid chloride -> Carboxylate is an example of this type of rule.

2. Likely reaction. This is to be used when almost all bacteria can catalyze a given reaction with a functional group present in a molecule. For example, if the substrate has an ester linkage, it is often hydrolyzed by very common esterases, found both extracellularly and intracellularly. So giving an ester hydrolysis rule a score of 2 would give it a high priority but after an acid chloride hydrolysis reaction. You should also use 2 for a reaction that is significantly likely to occur once a certain intermediate has been generated. For example, aromatic ring cis-dihydrodiols are likely to be dehydrogenated to form catechols. Most organisms that make cis-dihydrodiols will also catalyze their dehydrogenation, thus the latter reaction is likely due to the linkage. EAWAG-BBD btrule bt0255, Dihydrodihydroxyaromatic -> 1,2-Dihydroxyaromatic is an example of this type of rule.

3. Possible reaction (neutral). This applies to reactions that are common but not certain to occur in every system. For example, hydrocarbon oxygenation reactions are quite possible, but may or may not be likely to occur depending on what the substrate is. These must be looked at individually. Some may be likely, some may be possible and some may be unlikely based on current knowledge (an example of the latter may be oxygenases that work on 5-ring polycyclic aromatic hydrocarbons). EAWAG-BBD btrule bt0002, secondary Alcohol -> Ketone is an example of this type of rule.

4. Unlikely reaction. This would be the case for reactions that clearly might occur, but are either very rarely catalyzed in bacterial and fungal populations, or that don't seem likely to occur because of the initial conditions we are using or other chemical/biochemical reason. EAWAG-BBD btrule bt0029, organoHalide -> RH, which is unlikely to occur under aerobic conditions, is an example of this type of rule.

5. Very unlikely reaction. These reactions are ones, for example, that have never been observed under aerobic conditions and the enzymes are oxygen sensitive. Thus, given our initial conditions, we would expect that these reactions are highly unlikely. EAWAG-BBD btrule bt0270, Toluene -> Benzylsuccinate is an example of this type of rule.

6. No decision. This is reserved for cases where you cannot assign a number for whatever reason.