doctor rerum naturalium

(Dr. rer. nat.)

im Fach Biologie

eingereicht an der

der Humboldt-Universität zu Berlin

von

Präsident der Humboldt-Universität zu Berlin

Prof. Dr. J. Mlynek

Naturwissenschaftlichen Fakultät I

Prof. T. Buckhout, PhD

Tag der mündlichen Prüfung: 15. Dezember 2005

Einreichungsdatum: 06. Juli 2005

1. Summary

1.1 Zusammenfassung

2. Publications

2.1 References to this dissertation

2.2 Further references

2.3 Contributions to Congresses

3. Introduction

3.1 Behaviour as toxicological endpoint

3.2 Behaviour of fish influenced by chemical stressors

3.3 Chronobiological aspects of behaviour

3.4 Test substances

3.4.1 Microcystin-LR (MC-LR)

3.4.1.1 Cyanotoxins

3.4.1.2 Concentration and persistence of MC-LR in aquatic systems

3.4.1.3 Toxicity of MC-LR

3.4.2 Trichlorobiphenyl (PCB 28)

3.4.2.1 Technical use of PCBs

3.4.2.2 Chemical and physical properties of PCBs

3.4.2.3 Concentrations in aquatic systems and bioaccumulation

3.4.2.4 Toxicity of PCBs

3.5 Aim of this study

4. Material and Methods

4.1 Fish species

4.2 Test conditions

4.3 Test substances

4.3.1 Microcystin-LR

4.3.2 Trichlorobiphenyl

4.4 Recording Procedures

4.5 Statistics and calculations

4.5.1 Analysis of mean motility and mean number of turns

4.5.2 Regression between motility and turns

4.5.3 Zeitgeber analysis

4.5.4 Cosinor analysis

4.5.5 Periodic frequency analysis

5. Results

5.1 Effects of Microcystin-LR

5.1.1 Mean motility over the whole exposure period

5.1.2 Mean motility during light and dark phases

5.1.3 Temporal development of mean motilities

5.1.4 Number of turns over the whole exposure period

5.1.5 Regression between motilities and turns

5.1.6 Effects of zeitgeber

5.1.7 Cosinor analysis and Polar Plots

5.1.8 Periodic frequency analysis

5.2 Effects of Trichlorobiphenyl

5.2.1 Mean motility over the whole exposure period

5.2.2 Mean motility during light and dark phases

5.2.3 Temporal development of mean motilities

5.2.4 Number of turns over the whole exposure period

5.2.5 Regression between motilities and turns

5.2.6 Effects of zeitgeber

5.2.7 Cosinor analysis and Polar Plots

5.2.8 Periodic frequency analysis

6. Discussion

6.1 Toxicity of Microcystin-LR and Trichlorobiphenyl

6.1.1 Microcystin-LR

6.1.2 Trichlorobiphenyl

6.2 Fish-species specific differences to chemical stressors

6.2.1 Effects on locomotor activity of

*Danio rerio*and*Leucaspius d**e**lineatus*6.2.1.1 Microcystin-LR induced behavioural effects

6.2.1.2 Trichlorobiphenyl induced behavioural effects

6.2.2 Effects on chronobiology of

*Danio rerio*and*Leucaspius d**e**lineatus*6.2.2.1 Cyclic aspects of behaviour under unexposed conditions

6.2.2.2 Microcystin-LR induced chronobiological changes

6.2.2.3 Trichlorobiphenyl induced chronobiological changes

6.2.3 Stress potential of Microcystin-LR and Trichlorobiphenyl

6.3 Applications of behavioural approaches in ecotoxicology

6.3.1 Hormesis theory

6.3.2 Behavioural strategies

6.3.3 Chronobiological analysis

6.4 Digression: Biomonitoring

References

Appendix

Abbreviations

Erklärung

Acknowledgements

Tab. 1. Classification scheme for behavioural toxicants compiled by Barron (2002).

Tab. 2. Physico-chemical properties of PCB 28 (from Paasivirta et al. 1999).

Tab. 3. Mean total length (TL) and mean body mass (BM) of

*Danio rerio*and*Le**u**caspius delineatus*individuals for the tests with MC-LR and PCB 28.Tab. 4. Physico-chemical parameter of the aquaria water for the experiments with

*Danio rerio*and*Leucaspius delineatus*under the influence of MC-LR and PCB 28.Tab. 5. Number of turns versus motility with fitted linear function: y = a*x + b for

*Danio rerio*and*Leucaspius delineatus*exposed by MC-LR; Variables of the function: a (slope), b (intercept), r^{2 }(determination coefficient).Tab. 6. Rhythmical parameters of cosinor analysis: MESOR (M), amplitude (A), acrophase (K) and period length (P) for

*Danio rerio*and*Leucaspius delineatus*exposed by microcystin-LR (* > means that the value of the exposure group is significantly higher and * < means that the value of the exposure group is significantly lower compared to the control at the 0.05% level; n.s. means that there are no significant differences between the exposure group and the control group). The F- and p-values indicate the overall difference between the exposure group and the control.Tab. 7. Degree of Functional Coupling (DFC) and the harmonic portion (HP) of

*Danio rerio*and*Leucaspius delineatus*exposed by MC-LR.Tab. 8. Number of turns over motility with fitted linear function: y = a*x + b for Danio rerio and Leucaspius delineatus exposed by PCB 28.

Tab. 9. Rhythmical parameters of cosinor analysis: MESOR (M), amplitude (A), acrophase (K) and period length (P) for

*Danio rerio*and*Leucaspius delineatus*exposed by PCB 28. (* > means that the value of the exposure group is significantly higher and * < means that the value of the exposure group is significantly lower compared to the control at the 0.05 % level; n.s. means that there are no significant differences between the exposure group and the control group). The F- and p-values indicate the overall difference between the exposure group and the control.Tab. 10. Degree of Functional Coupling (DFC) and the harmonic portion (HP) of

*Danio rerio*and*Leucaspius delineatus*exposed by PCB 28.

Fig.1 The integrative role of behaviour. Behaviour is influenced by a variety of abiotic, biotic and internal factors.

Fig.2 Chemical structure of microcystin-LR.

Fig.3 Chemical structure of 2.4.4`-trichlorobiphenyl (C

_{12}H_{7}Cl_{3}).Fig.4 Scheme of the equipment.

Fig.5 Screen shot of the movement tracks of one fish group during one measuring interval of 2 minutes. The different lines represent the single individuals.

Fig.6 Scheme of a circadian rhythm.

Fig.7 Average motility of

*Danio rerio*and*Leucaspius delineatus*exposed to MC-LR over the whole measuring time per day (23 h per day) and the whole exposure period. Means and 95% confidence intervals are shown.Fig.8 Average motility of

*Danio rerio*and*Leucaspius delineatus*divided in light (white bars) and dark phases (grey bars) over the whole period of exposure to MC-LR. Means and 95% confidence intervals are shown.Fig.9 Smoothed curve of average motility of

*Danio rerio*and*Leucaspius delineatus*over the whole period of exposure to MC-LR.Fig.10 Average motility of

*Danio rerio*and*Leucaspius delineatus*during the time interval of 3 hours after the onset of light over the whole period of exposure to MC-LR. Means and 95% confidence intervals are shown.Fig.11 Differences in motility between the different exposure groups and the controls for

*Danio rerio*and*Leucaspius delineatus*, divided into light (white bars) and dark phases (grey bars). The whole period of exposure to MC-LR was divided into five time intervals of 3 days each and one interval of two days. Means and 95% confidence intervals are shown.Fig.12 Average number of turns of

*Danio rerio*and*Leucaspius delineatus*divided in light (white bars) and dark phases (grey bars) over the whole period of exposure to MC-LR. Means and 95% confidence intervals are shown.Fig.13 Number of turns over motility with fitted linear function: y = a*x + b for

*Danio rerio*and*Leucaspius delineatus*exposed by MC-LR.Fig.14 Effects of zeitgeber with 95% confidence intervals for

*Danio rerio*and*Le**u**caspius delineatus*exposed by MC-LR. Asterisks indicate significant differences to the value of 0.5. At values between 0 and 0.5, the animals are nocturnally active and between 0.5 and 1 they are diurnally active. Means and 95% confidence intervals are shown.Fig.15 Circadian motility rhythms of

*Danio rerio*determined by the cosinor method. Comparison of control and MC-LR exposed groups. Average motilities per hour (grey line) and cosinor fitted curves (black line).Fig.16 Circadian motility rhythms of

*Leucaspius delineatus*determined by the cosinor method. Comparison of control and MC-LR exposed groups. Average motilities per hour (grey line) and cosinor fitted curves (black line).Fig.17 Polarogram of cosinor analysis for

*Danio rerio*exposed by MC-LR. Angular axis represents the amplitude [pix s^{-1}]. Radial axis represents the time of the day [h]. Confidence ellipse covers the 95% interval.Fig.18 Polarogram of cosinor analysis for

*Leucaspius delineatus*exposed by MC-LR. Angular axis represents the amplitude [pix s^{-1}]. Radial axis represents the time of the day [h]. Confidence ellipse covers the 95% interval.Fig.19 Power spectrum of motility of

*Danio rerio*and*Leucaspius delineatus*exposed to MC-LR.Fig.20 Average motility of

*Danio rerio*and*Leucaspius delineatus*exposed by PCB 28 over the whole measuring time per day (23 h per day) and the whole exposure period. Means and 95% confidence are shown.Fig.21 Average motility of

*Danio rerio*and*Leucaspius delineatus*divided in light (white bars) and dark phases (grey bars) over the whole period of exposure to PCB 28. Means and 95% confidence intervals are shown.Fig.22 Average motility of

*Danio rerio*and*Leucaspius delineatus*during the time interval of 3 hours after the onset of light over the whole period of exposure to PCB 28. Means and 95% confidence intervals are shown.Fig.23 Smoothed curve of average motility of

*Danio rerio*and*Leucaspius delineatus*over the whole period of exposure to PCB 28.Fig.24 Differences in motility between the different exposure groups and the controls for

*Danio rerio*and*Leucaspius delineatus*, divided into light (white bars) and dark phases (grey bars). The whole period of exposure to PCB 28 was divided into four time intervals of 2 days each. Means and 95% confidence intervals are shown.Fig.25 Average number of turns of

*Danio rerio*and*Leucaspius delineatus*divided in light (white bars) and dark phases (grey bars) over the whole period of exposure to PCB 28. Means and 95% confidence intervals are shown.Fig.26 Number of turns over motility with fitted linear function: y = a*x + b for

*Danio rerio*and*Leucaspius delineatus*exposed by PCB 28.Fig.27 Effects of zeitgeber with 95% confidence intervals for

*Danio rerio*and*Le**u**caspius delineatus*exposed by PCB 28. Asterisks indicate significant differences to the value of 0.5. At values between 0 and 0.5, the animals are nocturnal active and between 0.5 and 1 they are diurnal active. Means and 95% confidence intervals are shown.Fig.28 Circadian motility rhythms of

*Danio rerio*determined by the cosinor method. Comparison of control and PCB 28 exposed groups. Average motilities per hour (grey line) and cosinor fitted curves (black line).Fig.29 Circadian motility rhythms of

*Leucaspius delineatus*determined by the cosinor method. Comparison of control and PCB 28 exposed groups. Average motilities per hour (grey line) and cosinor fitted curves (black line).Fig.30 Polarogram of cosinor analysis for

*Danio rerio*and*Leucaspius delineatus*exposed by PCB 28. Angular axis represents the amplitude [pix s^{-1}]. Radial axis represents the time of the day [h]. Confidence ellipse covers the 95% interval.Fig.31 Power spectrum of motility

*Danio rerio*and*Leucaspius delineatus*exposed to PCB 28.