Establishing Causality between Pollution and Effects at Different Levels of Biological Organization: The VALIMAR Project

The study summarizes the objectives of the VALIMAR project and gives selected examples of biomarker responses that allow causal relationships to be established between exposure and biological effects at different levels of biological organization. In this project, active and passive biomonitoring experiments with brown trout (Salmo trutta f. fario) and stone loach (Barbatula barbatula) were performed in two small streams in southern Germany between 1995 and 1999 in parallel with investigations on contaminant mixtures in the laboratory in order to evaluate the suitability of biomarkers representing different levels of biological organization for the assessment of pollution in small streams. In addition to biomarker studies, the morphology of the test streams was characterized and limnological and chemical parameters were monitored. Early life stage tests and ecological studies of brown trout and stone loach population demography, of the fish assemblages, and the macro- and meiozoobenthos communities in the two test streams were included in the project. Several causality criteria were addressed by means of combined (1) laboratory and field studies, (2) chemical, biological, and statistical investigations, and (3) in vivo and in vitro studies that allowed establishment of cause-effect relationships at different biological levels. The comparison of results obtained at these levels allowed identification of mechanisms responsible for the respective effects (coherence of association, biological plausibility). Finally, individual responses (biomarkers, bioindicators) could be extrapolated to higher biological levels (population, community) thus addressing the criteria of ‘time order’ and ‘coherence of association’.     

Hepatic biotransformation enzymes of fish exposed to non-point source pollution in small streams

The response of hepatic phase I and phase IIbiotransformation enzymes in brown trout(Salmo trutta f. fario) and stone loach(Barbatula barbatula) to differentpollution exposure scenarios was investigatedduring a 5-year-study. Phase I enzymes wereassessed by means of7-ethoxyresorufin-O-deethylase (EROD) activityand cytochrome P4501A (CYP1A) protein level,and phase II enzymes were measured by thecatalytic activity of glutathione-S-transferase(GST). Fish were exposed in the laboratory totap water (controls) or under semi-fieldconditions in so-called bypass systems to waterfrom two streams, with one of the streams(Krähenbach) receiving contaminants mainlyfrom non-point sources (for instance, surfacerun-off), and the second stream (Körsch)receiving contaminants from both non-pointsources and point sources (sewage plants). Inaddition to the fish in the bypass systems,feral fish from the streams were sampled. Thequestions addressed by the study were: (1) isit possible to discriminate the study sites onthe basis of hepatic biotransformation indices?(2) are the metabolic enzyme responses of thetwo species, brown trout and stone loach,comparable? (3) are the metabolic enzymeresponses of fish from the bypass systemscomparable to those of feral fishes? HepaticEROD activities were elevated in both browntrout and stone loach exposed in the bypasssystems compared to the laboratory control. Adiscrimination of the two field sites relativeto pollution intensity was possible by hepaticEROD activities of brown trout only, whereasbiotransformation enzymes of stone loach showedno consistent differences between the stream.Hepatic GST levels of both speciesdid not discriminate between the streams orbetween streams and laboratory. The enzymeresponses measured in fishes from the bypasssystems were comparable to those of feralfishes living in the streams. The findings ofthis study point to critical factors for thebiological monitoring of low-level pollutionwith complex chemical mixtures, including theselection of appropriate test species,monitoring endpoints, and sampling strategies.     

The use of histopathological indicators to evaluate contaminant-related stress in fish

As a component of a large research program toevaluate the effects of contaminants on fish healthin the field, histopathological studies have beenconducted to help establish causal relationshipsbetween contaminant exposure and various biologicalresponses. Brown trout (Salmo trutta f. fario)and loach (Barbatula barbatula) were exposedto water diverted from polluted streams undersemi-field conditions at various times during theyear. The histopathological studies revealedseasonal differences in the types and severity oforgan lesions between fish of the two streams. Bothtoxicant-induced alterations and organ lesionsresulting from natural stressors (physicochemicaland limnological water parameters) and secondarystress effects of pollution (diseases) could bedetected. In evaluating the general health ofexperimental and control fish, the use ofhistopathological studies are recommended for makingmore reliable assessments of biochemical responsesin fish exposed to a variety of environmentalstressors. Stereological analysis providesquantitative data on pathological lesions whichhelps to establish correlation with other biomarkersthereby increasing the probability of identifyingcause (stressor) and effect (biomarker) relationships.     

The biology of stone loach Barbatula barbatulus in the Rivers Goloustnaya and Olkha, East Siberia

Five hundred and four stone loach 63–202 mm in total length, of I + to 8 + age groups, were sampled in two rivers of East Siberia from April to October 1989–90. This paper presents data on the seasonal dynamics of fecundity, ripening, fatness and feeding intensity. A second spawning from the end of August to October has been identified for the first time for this species. The absolute fecundity ranged from 1710 to 27600, and the relative fecundity varied from 216 to 833 eggs g⁻¹: (fish weight). Stone loach in the River Olkha grew faster than those in the River Goloustnaya and, according to this index, are closer to stone loach from streams of northern England.     

Using indicators of fish health at multiple levels of biological organization to assess effects of stream pollution in southwest Germany

The health of fish populations and assemblagesin two degraded streams in south-westernGermany were investigated in order to validatethe use of biomarkers for the assessment ofsmall streams as part of the Valimar project.In the more severely polluted stream,recruitment was impaired and populationabundances were low. Extinction ofsite-specific species and appearance of nonsite-specific species has resulted in a shiftin the species composition from thatcharacteristic of unpolluted reference streams.In the less severely degraded stream, abundantrecruitment and relatively high densities ofadults indicated healthy, self sustaining fishpopulations. Missing species, however,indicated some effects of human influenceswithin the reference stream. The reproductivestatus of adult stone loach, Barbatulabarbatula, was not affected in any stream asreflected by normal production of eggs andreproductive behavior. Indices of energystorage differed between the investigated sitesbut could not be related to pollution. Theevaluation of biological integrity based onfish assemblages revealed that both sites inthe more severely impacted stream were in poorcondition. The two reference sites wereclassified in moderate and good condition,respectively. By excluding effects of migrationbarriers on the fish assemblage there was astronger correlation between chemical waterquality and biomarker responses. The resultsindicate the importance of assessing theeffects of pollution within the content oflongitudinal gradients in streams and theusefulness of measuring health effects overseveral levels of biological organization.     

Temporally variable macroinvertebrate–stone relationships in streams

Stones were used to sample macroinvertebrates and characterise microhabitats at monthly or bimonthly intervals in six Ecuadorian streams covering a gradient in four different stability measures and other stream characteristics. The physical variables current velocity, water depth, horizontal position, embeddedness and size were measured to characterise stone microhabitats and presumed to be affected by or related to physical impact during hydrological disturbances. My first objective was to analyse how density, the number of families and a richness measure (residuals from a power regression of families vs. individuals) were related to the physical characteristics of individual stone habitats. My second objective was to quantify temporal variability in fauna–stone relationships and to analyse if such variability was related to overall stability of stream reaches. Partial Least Squares (PLS) multiple regression analyses showed high temporal variability between sampling dates in factor loadings of specific stone micro habitat variables. In spite of this, there was a clear negative effect of depth and a positive effect of current on density and number of families. Stone size was consistently negatively related to density and positively related to number of families. Patterns were less clear for richness residuals. Simple linear regressions of fauna vs. stone parameters generally confirmed the results reached by the PLS analysis, although few of the regressions were significant. For all fauna–stone regressions the variability in slopes was much higher among sampling dates within streams (temporal variability) than among streams (spatial variability), and significant slopes were even inverted on different sampling dates. Although the coefficients of variation (CV) of slopes of a given combination of fauna parameter and stone variable from different sampling dates (n=9–11) were rarely correlated to any of the measures of stream stability, this study has demonstrated high temporal variability in fauna–stone relationships (CV’s of regression slopes). Consequently, temporally un-replicated studies of such relationships do not necessarily reveal general patterns.     

The hepatic stress protein (hsp70) response to interacting abiotic parameters in fish exposed to various levels of pollution

Two indigenous fish species, brown trout (Salmo trutta f. fario) and stone loach(Barbatula barbatula) were exposed tocomplex stressors (mixtures of environmentalpollutants) in laboratory and semi-fieldexperiments (aquaria connected to stream water)and in field studies. As a biomarker of effect,the level of the 70 kD heat shock protein(hsp70) was quantified in the liver of troutand loach. Laboratory experiments withdifferent pollutant mixtures did not mimic thehsp70-inducing or inhibiting potential of fieldconditions, whereas effects of long-termexposure in the bypass systems showed asignificant correlation with effects recordedin feral fish. Laboratory as well as semi-fieldstudies revealed the stress response to followan optimum curve, resulting in a maximum hsp70level under stress but rather low hsp70 levelswhen stressors (chemicals, high temperature)become too severe. Consequently, the hsp70level in the liver of both species was highlyseason-dependent with two peaks in late springand fall, and rather low hsp70 levels insummer, particularly in fish exposed to waterand sediment of the complexly polluted stream.In winter, the low hsp70 level of lab controlswas elevated by exposure to natural streamwater only, but elevation did not occur undercontrol conditions independent of apre-exposure to polluted streamwater two months earlier. Despite the highvariability of the hsp70 level within one yearand among five subsequent years, patternanalysis indicated the prevailing importance ofwater temperature on stress protein response.Temperature alone, however, could not explainthe regularly observed low summer levels inhsp70. Non-linear regression analysis on labcontrols revealed an optimum temperature(T_opt) for the highest hsp70 level forboth fish species. In both investigated streams, thechemical influence led to a decrease in thehsp70 level only when T_opt was surpassedby the ambient temperature at the same time.Otherwise, the chemical impact resulted in anelevated hsp70 level relative to the control.The study demonstrated the suitability of hsp70stress protein levels to integrate the responsedynamics of several different stressors and,therefore, to effectively function as abiomarker for the integrated effect of allenvironmental stressors acting on an organism(not only of chemical pollution). Rathercomplex kinetics of hsp70 elevation anddecrease should be taken into consideration.     

Brown trout Salmo trutta f. Fario liver ultrastructure as a biomarker for assessment of small stream pollution

T he effects of environmental pollutants from two small streams in south-west Germany on the liver of brown trout (S almo trutta f. Fario ) were studied as biomarkers by means of quantitative and semi-quantitative electron microscopy, and quantitatively by morphometrical measurements. Cellular damage was assessed semi-quantitatively based on a classification of ultrastructural responses. Both methods revealed more severe cellular effects in the liver of trout which had been exposed to the highly polluted stream than in those exposed to the lightly polluted river. Morphometrical studies showed a significant reduction of glycogen storage and a significant increase in number of mitochondria, peroxisomes and cisternae of the endoplasmic reticulum. The biomarker responses of this study were correlated with the results obtained by limnological and analytical investigations, and reflect the levels of pollution in each stream.     

Alterations of selected metabolic enzymes in fish following long-term exposure to contaminated streams

In order to assess the suitability ofalterations in activities of selected metabolicenzymes as biomarkers of chemicalcontamination, juvenile brown trout (Salmotrutta f. fario) and adult loach (Barbatula barbatula) were exposed to nativesurface waters from Krähenbach andKörsch, two differently polluted smallstreams in Southern Germany. As biomarkers ofexposure, a set of metabolic enzymes comprisinghexokinase, phosphofructokinase,glucose-6-P-dehydrogenase, malic enzyme,cytochrome c oxidase, succinate dehydrogenase,citrate synthase, alanine aminotransferase,esterase, superoxide dismutase, catalase, acidphosphatase as well as acetylcholinesterase(brain) were measured by means of establishedenzyme assays. In parallel studies, loach andbrown trout sampled from corresponding fieldsites were investigated to elucidate therepresentativeness of a stream bypass system.Although the episodic exposure scenarios ineither natural stream resulted in considerablevariability of enzyme activities, biochemicalchanges in both species allowed a cleardiscrimination between the two differentlypolluted streams. Similarities in changes ofenzyme activities between fish exposed in thebypass system and the field reached levels of63% and 73.1% in brown trout and loach,respectively. In conjunction with biomarkersfrom other studies, alterations in metabolicenzyme activities were able to serve as a toolfor the sensitive identification ofenvironmental pollutants, which in turn form fromthe basis for an improved understanding ofunderlying toxic processes and an interpretationof toxicant-related effects.     

Ultrastructural changes in fish gills as biomarker to assess small stream pollution

In order to verify the principal suitability of gill ultrastructure as a biomarker, semi-field studies with two endigoneous fish, trout (Salmo trutta f. fario) and loach (Barbatula barbatula), were performed. The fish were exposed in flow-through systems to one heavily polluted (K?rsch) and one lightly polluted small stream (Kr?henbach) in South-West Germany. Ultrastructural responses in gills were correlated with limnological and chemical data recorded over a 2 year period in each stream. After 8 weeks of exposure to the heavily polluted stream, fish showed ultrastructural changes in the gills, such as cell proliferation, dilation of the endoplasmic reticulum, hyperplasia, hypersecretion, and epithelial lifting in chloride, epithelial, and mucus cells. The results of the study demonstrate that ultrastructural reactions in the gills of fish kept under semi-field conditions are potentially useful biomarkers indicating small stream pollution.     

Spatial variation in density of stream benthic fishes in northern Hokkaido, Japan: does riparian vegetation affect fish density via food availability?

The densities of two benthic fishes, the Siberian stone loach (Noemacheilus barbatulus) and the wrinklehead sculpin (Cottus nozawae), and the biomass of their food resources (i.e., periphyton and benthic invertebrates) were compared between forest and grassland streams in northern Hokkaido, Japan, to examine whether riparian deforestation had positive effects on the benthic fishes via enhancement of food availability. The comparisons indicated that riparian vegetation had little influence on periphyton, invertebrates, or fishes. Regression analysis indicated that spatial variations in loach and sculpin densities were explained more by substrate heterogeneity, competitor abundance, or both, rather than by food abundance. However, when the two species were combined as benthic insectivores, a strong correlation was found between total benthic fish density and invertebrate biomass. Our results suggest that, although total benthic fish abundance was food limited, riparian vegetation had no positive effects via food availability on the benthic fishes in our streams.     

The critical thermal limits for the stone loach, Noemacheilus barbatulus, from three populations in north-west England

1. The chief objective was to determine the upper and lower thermal limits for feeding and survival in the stone loach, Noemacheilus barbatulus, using juveniles (total length 30–45 mm, live weight 0.25–0.80 g) from one population and adults (total length 77–100 mm, live weight 3.6–7.9 g) from three populations. 2. Fish were acclimatized to constant temperatures of 3, 7, 10, 15, 20, 25 and 27°C; then the temperature was changed at a rate of 1°C/30min to determine the critical limits for feeding, survival over 7 days (incipient lethal temperature), or survival for 10 min or less (ultimate lethal temperature). The rate of 1°C/30min was the optimum value from preliminary experiments, using nine rates from 0.5°C/48h to 18°Ch^?1. As values for adults were not significantly different between populations, they were pooled to provide arithmetic means (with 95% CL) for the thermal limits at each acclimation temperature. 3. Feeding limits increased with acclimation temperature to upper and lower mean values of 28.0 ± 0.15°C and 5.1 ± 0.55°C for adults, 25.0 ± 0.54°C and 6.1 ± 0.92°C for juveniles. Incipient lethal levels defined a tolerance zone within which stone loach survive for a considerable time; upper limits increased with acclimation temperature to reach a maximum plateau of 29.1 ± 0.18°C for adults and 29.0 ± 0.40°C for juveniles; lower limits also increased from near 0°C to 3.0 ± 0.40°C for adults and juveniles. Upper limits for the ultimate lethal level increased with acclimation temperature to a maximum plateau of 33.5°C for adults (95% CL ± 0.19) and juveniles (95% CL ± 0.40), whilst the lower limits increased from near 0°C to 2.5 ± 0.30°C. At acclimation temperatures below 20°C, upper incipient and ultimate lethal values were significantly lower for juveniles than those for adults. 4. The thermal tolerance of stone loach was higher than that of juvenile Atlantic salmon or brown trout, one or both of these species often being dominant in streams with stone loach.     

Response of stream fish to improved water quality: a comparison between the 1960s and 1990s

1. The effect of improved water quality on fish assemblages in streams in southern Sweden was assessed by comparing species composition at 161 sites and water quality at twenty-nine sites in the 1960s and the 1990s. 2. Water quality had improved and there was an increase in the number of sites or catchments with brown trout (Salmo trutta), stone loach (Barbatula barbatula) and eel (Anguilla anguilla). The response was greatest for brown trout and was best explained by increased oxygen concentrations. 3. The number of sites with nine-spined stickleback (Pungitius pungitius), ide (Leuciscus idus) and brook lamprey (Lampetra planeri) decreased between the 1960s and 1990s. The decrease was greatest for nine-spined stickleback and was related to the increase in sites with trout, suggesting that nine-spined stickleback may be sensitive to predation or competition. 4. Improved water quality has led to recolonization by brown trout, probably enabling biotic interactions to play a larger role in structuring fish assemblages. Improving water quality was an effective method for rehabilitating fish populations in streams where natural colonization was possible.     

Fish abundance and habitat relationships in forest and grassland streams, northern Hokkaido, Japan

The relationships between habitat variables and population densities of masu salmon (Oncorhynchus masou), rosyface dace (Leuciscus ezoe), Siberian stone loach (Noemacheilus barbatulus) and wrinklehead sculpin (Cottus nozawae) were examined by data collected at 55 reaches in forest and grassland streams in northern Hokkaido, Japan. Regression analysis suggested that salmon and dace densities were affected by water temperature (negative for salmon, positive for dace) and structural habitat factors (woody debris for salmon, pools for dace). Salmon density was higher in forest reaches than in grassland reaches, whereas dace density was higher in grassland reaches, suggesting that the removal of riparian forest had raised water temperature and allowed upstream invasions by dace. In contrast to salmon and dace, neither the density of loach nor sculpin differed between the forest and grassland reaches. For their densities, a negative effect of each on the other was most important, suggesting a strong effect of interspecific competition between loach and sculpin on their distributions. However, regression models also suggested that substrate heterogeneity mediated the outcome of their interspecific competition. On the basis of the results, a scenario is predicted for a fish-assemblage change with a typical land-development process in Hokkaido, and the importance of leaving or restoring riparian buffer for conservation and restoration of stream habitat is emphasized.     

Studies on the toxicity of zinc sulphate and of cadmium sulphate to stone loach Noemacheilus barbatulus (L.) in hard water

The toxicity of zinc sulphate and cadmium sulphate to stone loach Noemacheilus barbatulus was examined in an attempt to explain the distribution of the species in two polluted streams. Stone loach in hard water were more sensitive than rainbow trout to zinc, but much more resistant to cadmium. Their disappearance from the Willow Brook, Northants, in 1971–72 could be accounted for by zinc toxicity. A behavioural change, the loss of the instinct to hide during daylight, occurred in subsequently lethal concentrations of cadmium. The accumulation of cadmium in several tissues was also determined.     

Spatial and ontogenetic variability in the microhabitat use of stream-dwelling spined loach (Cobitis taenia) and stone loach (Barbatula barbatula)

Ontogenetic and spatial variability in microhabitat use of spined loach Cobitis taenia (Linnaeus), considered as one species for the purposes of this study, and stone loach Barbatula barbatula (Linnaeus) were examined in the River Great Ouse basin, England, using multivariate and habitat suitability methods, including a technique for handling spatial variation in collections of preference curves. Distinct ordinations of spined age classes and stone loach developmental stages, respectively, in canonical correspondence analysis of species × variables × samples relationships suggest that the two species occupy completely different microhabitats; however, young‐of‐the‐year spined loach occurred more often than expected with all developmental stages of stone loach except young larvae. Water velocity and filamentous algae were the most influential microhabitat variables, the latter decreasing in importance with increasing age of both fish species. Preferred water velocities generally decreased with age in spined loach and increased in stone loach, with substratum size generally increasing with fish age in both species. Spatial variation in microhabitat preferences was great in both species but less so in the spined loach, suggesting that limited plasticity in habitat use could account, at least in part, for the latter species’ limited distribution and abundance in the catchment. Preference curves for a species, if generated and verified for all life intervals and all seasons, could be used as a management tool for a given stream or sector of river basin. But preference curves should be generated for each location to ensure that river management decisions with regard habitat and species conservation consider local‐level species requirements. Thus, a multi‐(eco)species and multi‐scale approach is required in habitat suitability assessments.     

Histopathological alterations and parasite infection in fish: indicators of multiple stress factors

Within the scope of the multidisciplinaryresearch project Valimar (1995–1999), thepresent study emphasizes the use ofhistopathological investigations to evaluatethe effects of the different chemical impactof two small streams on the general health andcondition of the two sentinel fish species,brown trout (Salmo trutta f.fario) and loach (Barbatula barbatula).Parasitological investigations were alsoincluded to assess a possible relationshipbetween toxicant exposure, parasitic load, andthe occurence of histopathological organchanges. According to the multitiered approachof the Valimar project, fish from the field,individuals exposed to stream water undersemi-field conditions, and fish fromlaboratory experiments were investigated.Under semi-field conditions, thehistopathological responses in brown troutproved to reflect the different levels ofcontamination of the two small streams moreclearly than those detected in loach. Whencaptured from the field, both fish speciesrevealed, in most cases, a higher degree ofhistopathological alterations than thoseindividuals exposed at the respective siteunder semi-field conditions. Simulation of thetoxic load of the more polluted stream in alaboratory experiment did not result in thesame kind of tissue lesions in brown trout asin the field and semi-field investigations.Whereas results from the laboratoryexperiments could be specified astoxicant-induced, the histopathologicalalterations in fish from both semi-field andfield studies represent stress responses whichintegrate the effects of various abiotic andbiotic stress factors including both toxicantsand parasitic diseases.     

A functional model for maximum growth of immature stone loach, Barbatula barbatula, from three populations in north-west England

1. The chief objective was to develop a functional model for growth of stone loach, Barbatula barbatula, using immature fish from three populations. The growth model had been developed previously for brown trout, Salmo trutta, but new estimates of the five parameters for the stone loach had to be obtained from laboratory experiments. 2. Fish from four size groups (initial arithmetic mean live weights 0.053 g, 0.231 g, 0.840 g, 1.612 g, with five fish per group) from Great Oaks Wood Beck were acclimatized to constant temperatures of either 3, 5, 10, 15, 20 or 25 °C. Each fish was kept in a separate tank and fed to satiation on freshwater shrimps. Weights and lengths of each fish were recorded at the start and finish of a growth period of 35 days. For each of the other populations (Black and Ford Wood Becks), there were only three temperatures (5, 10, 20 °C) with ten fish per temperature. 3. The growth model was an excellent fit (P < 0.001, R² > 0.99) for the 120 fish from Great Oaks Wood Beck. Growth rates were negative at 3 °C, close to zero at 5 and 25 °C, and positive at 10, 15 and 20 °C, with an optimum value of 19 °C. When growth rates were positive, they decreased markedly with increasing fish weight for small fish but decreased more slowly for larger fish. At the start of the experiments, weight–length relationships were similar for fish from all three populations and were well described by a power function. There was excellent agreement between growth rates estimated from the fitted growth model for fish from Great Oaks Wood Beck and values obtained for fish from Black and Ford Wood Becks. Data from all three populations were therefore pooled (n = 180) to obtain new estimates of the five parameters in the model. 4. Comparisons between parameter estimates for trout and stone loach showed that the latter grew better in warmer waters (e.g. optimum value for growth was 19.0 °C for stone loach and 13.1 °C for trout, with ranges for growth of 5.0–25.0 °C and 3.6–19.5 °C, respectively).     

Habitat factors affecting the distribution and abundance of the spinous loach <Emphasis Type="Italic">Cobitis shikokuensis</Emphasis> in southwestern Japan

We examined habitat factors related to the distribution and abundance of the spinous loach Cobitis shikokuensis, an endangered benthic fish, in the Shigenobu River system, southwestern Japan. In the study river, the spinous loach was distributed widely along the main stem, from headwater to near the mouth, whereas it was rarely found in tributary streams. Classification tree analysis showed that the presence/absence of spinous loach was explained by a combination of percent pebble and length of river fragment between artificial barriers. Spinous loach incidence was high in sites with abundant pebble (>27.7%), but low in sites with short river fragment (≤0.97 km) even if pebbles were abundant. A regression tree model for loach density retained only percent pebble as a single best predictor, with sites with higher percent pebble (>40.4%) having higher density. These results suggest that substrate condition is an important factor determining the distribution and abundance of spinous loach and also that habitat fragmentation by artificial barriers has great potential to threaten the spinous loach population in this river. Considering the highly fragmented situation of the study river and prevention of upstream migration by barriers, we conclude that maintenance of suitable habitats in upper reaches has high priority for conservation of the spinous loach.