Workshop Report 24

Screening for environmental persistence

The aim of this session was to identify recent advances in the assessment of biodegradation and persistence using screening test methodology and where further improvements are needed. A series of presentations introducing the issues relating to screening studies was followed by a panel discussion and two syndicate sessions.

Gary Bending (University of Warwick, UK) presented research undertaken by his group investigating issues of pragmatism and realism related to the assessment of persistence. The environmental relevance of laboratory screening biodegradation studies is limited because standardised test conditions represent only a small range of environmental complexity and heterogeneity. The challenge is to obtain reproducible test outcomes whilst achieving environmental relevance. There are many factors influencing the catabolic potential of an inoculum including density, diversity, composition of inocula as well as water chemistry and compartment variability.

Gary presented data comparing the biodegradation of p-nitrophenol by different inocula and under different environmental conditions. Microbial biofilms, that are absent from biodegradation screening studies, have several advantages over inocula from water including higher density and active biomass, community interactions, microhabitat heterogeneity, greater genetic variation and exchange, better consistency (longer site history). Seasonal variation was a greater determinant of bacterial community composition than proximity to the outflow of a sewage treatment plant (STP). River water collected from the STP outflow showed more consistent degradation of p-nitrophenol than water collected from upstream or downstream. River biofilms provided similar rates of biodegradation to river water, despite the larger amounts of biomass applied in degradation assays. A lack of degradation of p-nitrophenol in some river water samples was associated with factors controlling bacterial proliferation rather than an absence of catabolic potential (competent organisms).

At environmental concentrations of most xenobiotics (low µg/L) degradation is likely to be via co-metabolism, whereas at higher concentrations typically used in lab studies, biodegradation is likely to be growth-linked. The p-nitrophenol experiments support this with more realistic (low) concentrations undergoing more variable and lower rates of biodegradation. Furthermore, studies also demonstrated that light can also be an important factor, as algal photosynthesis increased pH in non-buffered waters resulting in the reduced growth of degrading organisms. Currently, all biodegradability studies are conducted in the absence of light.

The addition of complexity into test systems may therefore affect the outcome of biodegradation tests in a manner which is hard to predict.

Kees van Ginkel (Akzo Nobel, the Netherlands) discussed modified and enhanced ready biodegradation tests. Ready biodegradability tests only detect growth-linked biodegradation because there is only one substrate and this is presented at a high concentration. This, combined with the low inocula density, leads to a stringent test. However, a pass in a ready test indicates a high probability that the chemical tested will undergo rapid and complete biodegradation in the environment. Modifications and enhancements of the ready biodegradability tests have been designed to reduce the likelihood of false negatives and to overcome several difficulties with chemicals that are poorly water soluble, i.e. are of limited bioavailability, or toxic to the inoculum. For example, the toxicity of quaternary ammonium compounds can be avoided by the addition of silica gel, humic acids or lignosulphonic acids salts. Addition of silicone oil has been shown to reduce the toxicity and volatilisation of a fragrance. Agitation of the test media and addition of surfactants have also been used to successfully increase bioavailability of poorly water soluble substances. The slow release of poorly water soluble chemicals in the environment may limit the rate of biodegradation. Although this may suggest the chemical may be persistent, the risks may be low because the bioavailable fraction can be degraded.

Enhanced (less stringent) biodegradation tests can be achieved by pre-adapting inocula and extending the test duration. This allows competent microorganisms present in low numbers in the environment to multiply, not true for prolonged ready biodegradability test results. At least one competent organism was present at the start of the test, multiplying to numbers which enable detection of biodegradation of test chemicals in screening tests. For example, although N-methylpiperazine is not readily or inherently biodegradable, competent micro-organisms were obtained after acclimatisation to low concentrations of N-methylpiperazine in SCAS (semi-continuous activated sludge) units. Pre-adapted inocula also degraded this chemical in a closed bottle test when grown on glass beads and additional nutrients (glucose) to stimulate growth.

It was concluded that modifications and enhancements have improved the assessment of persistence in the environment. There still is an element of trial and error involved in the assessment of biodegradation with modified tests. Pre-adaptation is useful in assessing potential for inherent biodegradation but it may not indicate how widespread a competent micro-organisms might be in the environment. Running screening tests for a prolonged period (60 days) was considered generally appropriate. Interpretation of the results of modified and enhanced test should be more science based. For instance, in environmental microbiology growth-linked biodegradation is considered superior compared to cometabolic transformations often detected in simulation tests. There was interest in comparing the performance of chemicals in both the new enhanced tests and in standard ready biodegradation tests.

Russell Davenport (Newcastle University, UK) presented the outcome of a research project that has provided a better understanding of the factors that influence screening tests. Ready biodegradability tests (RBTs) have been core to the determination of the biodegradation of chemicals in regulatory frameworks for 2-3 decades. There are seven test methods varying in inoculum source and preparation and consequently there are variable probabilities of inclusion of specific degraders in the test system. This leads to a high failure rate and high variability largely due to the use of low inocula concentrations (microbial concentrations in inocula can vary by 4 orders of magnitude within the standardised biodegradation test guidelines). Together with their short duration, this makes them unreliable for persistence assessments. Variation in density of inocula should be reduced. Enhanced screening tests, for use in persistence assessment rather than ready biodegradability, may include increasing inocula to environmentally relevant concentrations. This increases the likelihood that the microbial diversity will be environmentally realistic.

The CEFIC/LRI ECO 11 (www.cefic-lri.org//ECO11: Towards rationally designed hazard, risk and persistency assessment: Putting the “bio” back into biodegradability tests)project has been investigating how variations in inocula concentration, community composition and diversity relate to the variation and reliability in screening test outcomes. In addition, the bias and pragmatism of different methods to concentrate cells in inocula for enhanced tests has been assessed. These enhancements were validated using a set of reference chemicals (chosen by CEFIC/LRI ECO 12 project) that represent different rates and extents of biodegradation in the environment. It was found that greater variation occurred between sources (different compartments) of inocula than between different inocula from the same compartment. Enhancement of activated sludge inocula concentrations had a greater effect on reliability than test volume in scaled-up biodegradation tests, but not necessarily for marine inocula which showed particularly high variability in the lag phase. Further investigation is needed to achieve a reliable screening test for the marine compartment. Russell recommended a move to high throughput screening tests using multiple sources of inocula and a probabilistic approach to interpreting the results. This would be aided by better understanding of the influence of genetic diversity on biodegradation. In most cases a 100-fold increase in the concentration of inocula resulted in (i) less variability within and between studies, (ii) shorter and more consistent lag periods, (iii) a higher probability of reaching the pass-criterion for ready biodegradability for chemicals considered as non-persistent, and (iv) no false positives for chemicals considered to be persistent.

Dan Salvito (RIFM, USA) introduced the difficulties of testing natural complex substances. Natural Complex Substances (NCS) are materials extracted from plants and used in the preparation of fragrance mixtures for a variety of consumer products. Typically these are classified as UVCBs (Unknown, Variable Composition, or Biologicals), or, minimally, Multi-Component Substances (MCS) for the less chemically complex extracts. Assessment of these materials is required under various regulatory schemes including REACH. While there are methods for considering the ecotoxicity of a mixture using additivity, little has been published on approaches for either environmental fate studies or other assessment methods for NCS. The International Fragrance Association’s Environmental Task Force has provided recommended approaches for NCS biodegradation assessment. Dan presented studies using NCS as examples to assess the ready biodegradability of key constituents of these mixtures in order to provide an overall assessment of the biodegradability of the NCS itself. Two approaches were presented. If the mixture contains mostly substances that are structurally similar, then the mixture could be tested. More work is needed to develop this approach. If the mixture contains structurally dissimilar substances then individual constituents should be assessed. In this case the outcome of the constituent assessments can be used in providing a statement about the biodegradability of the NCS. When individual constituents are unavailable to test as single substances, e.g. in the case of sesquiterpenes which are common constituents of fragrance oils, QSBRs can be used to predict biodegradability. These predictions need to be considered carefully as predictions from different versions of the US EPA EPI Suite BIOWIN software are often contradictory and also may not concur with experimental tests.