Induction, adaptation and recovery of biological responses: implications for environmental monitoring.

A wide range of biological responses have been used to identify exposure to contaminants, monitor spatial and temporal changes in contamination levels, provide early warning of environmental deterioration and indicate occurrences of adverse ecological consequences. To be useful in environmental monitoring, a biological response must reflect the environmental stress over time in a quantitative way. We here argue that the time required for initial induction, maximum induction, adaptation and recovery of these stress responses must first be fully understood and considered before they can be used in environmental monitoring, or else erroneous conclusions (both false-negative and false-positive) may be drawn when interpreting results. In this study, data on initial induction, maximum induction, adaptation and recovery of stress responses at various biological hierarchies (i.e., molecular, biochemical, physiological, behavioral, cytological, population and community responses) upon exposure to environmentally relevant levels of contaminants (i.e., metals, oil, polycyclic aromatic hydrocarbons (PAHs), organochlorines, organophosphates, endocrine disruptors) were extracted from 922 papers in the biomarker literature and analyzed. Statistical analyses showed that: (a) many stress responses may decline with time after induction (i.e., adaptation), even if the level of stress remains constant; (b) times for maximum induction and recovery of biochemical responses are positively related; (c) there is no evidence to support the general belief that time for induction of responses at a lower biological hierarchy (i.e., molecular responses and biochemical responses) is shorter than that at higher hierarchy (i.e., physiological, cytological and behavioral responses), although longer recovery time is found for population and community responses; (d) there are significant differences in times required for induction and adaptation of biological responses caused by different types of contaminants; (e) times required for initial and maximum induction of physiological responses in fish are significantly longer than those in crustaceans; and (f) there is a paucity of data on adaptation and recovery of responses, especially those at population and community levels. The above analyses highlight: (1) the limitations and possible erroneous conclusions in the present use of biomarkers in biomonitoring programs, (2) the importance of understanding the details of temporal changes of biological responses before employing them in environmental management, and (3) the suitability of using specific animal groups as bioindicator species.