With love, to Leon and Gaia.

Abstract in English

The rise of mammals from premammalian cynodonts during the Late Triassic was an important transition in vertrebrate evolution. The similarities in body size, orbit size, and tooth shape of early mammalian fossils, as e.g. Morganucodon and Megazostrodon, to modern shrews, tenrecs, and hedgehogs led paleontologists to the assumption that the first mammals were nocturnal, living in the shadow of the dinosaurs. For over 30 years, this view has been generally accepted and published in textbooks. Moreover, a nocturnal lifestyle would have gone hand in hand with the evolution of fur and of endothermy, which, among other features, contributed to the origin of this highly diverse and successful animal group.

One of the limitations of paleontology is the lack of soft tissue preservation; because eye tissue is not preserved in early mammalian fossils, nocturnality as the ancestral state in these taxa will always remain an assumption. Fortunately, in recent years there have been major improvements in molecular techniques; e.g. ancestral sequence reconstructions and in vitro expression systems, as well as in selective constraint analyses, allowing certain types of evolutionary questions regarding the evolution of visual systems to be addressed in novel ways. 

This thesis investigates whether early mammals had indeed been nocturnal by combining paleontology and molecular techniques, focusing on the only visual pigment in the vertebrate eye that is responsible for vision at night and/or dim-light; the rhodopsin.

First, for a more reliable taxon sampling, the rhodopsin gene of the echidna, one of the two living families of the most basal mammalian lineage, the monotremes, was sequenced and was successfully expressed in vitro, together with two self-designed mutants with unique substitutions at sites 158 and 169. Biochemical and functional analyses revealed that the echidna rhodpsin displays some cone-like characteristics, likely due to rhodopsin being expressed in cones as well. Furthermore, site 169 was found to affect the strength of photon absorption in the echidna. With the echidna being a nocturnal animal, this thesis comprises the first characterisation of a rhodopsin of a nocturnal animal.

Second, based on a comprehensive alignment of 27 tetrapod rhodopsin sequences, ancestral rhodopsin sequences for the nodes Amniota, Mammalia, and Theria (i.e. marsupials and placentals) were inferred using Maximum likelihood estimates. The most likely of these were successfully expressed in vitro. All expressed pigments were functional and rod-like. Most importantly, meta II half lifes, which specify the time in which rhodopsin is in its active state activating the visual transduction cascade, were found to differ; Amniota shows the same rate as bovine, whereas Mammalia and Theria display a much higher t_1/2. A high t_1/2 has been said to facilitate better vision at low-light levels. Due to inconsistency in the available data, the result also suggests that, with the visual signaling cascade being such a complex and interconnected system, erecting ecological interpretations based on single biochemical and functional reactions is problematic.

Third, selective constraint analyses that investigate positive selection were completed. Positive selection is characterised by a high number of non-synonymous substitutions that change the subsequent amino acid and, thus, lead to changes in and the adaptation of a protein. These analyses revealed that the branches leading to Theria and marsupials were the only ones that experienced positive selection acting on the rhodopsin. The positive selection found at the therian branch likely reflects the rapid diversification into modern ecological habitats during the Triassic and Jurassic, as indicated by recent additions to the fossil record. Furthermore, it has been found that the branch leading to Mammalia experienced positive selection in synonymous substitutions, which do not change the subsequent amino acid; instead, these silent sites have an effect on mRNA stability and tRNA translation efficiency, increasing the number of rhodopsin molecules. This results in a scenario where the mammalian rhodopsin might have experienced positive selection on synonymous substitutions in order to increase its molecule number as an adaptation to vision at night, followed by later adaptive changes due to ecological diversification.

Though molecular techniques permit valuble insights regarding the nocturnality of the earliest mammals, additional data as well as novel investigative approaches are needed in order to address this fascinating aspect of evolutionary history. Nonetheless, this thesis emphasises the inherent value of paleontology and molecular methods working in tandem.

Abstract in German

Die Evolution der Säugetiere in der späten Trias zählt zu den bedeutendsten Ereignissen in der Wirbeltiergeschichte.

Fossilien belegen, dass die ersten Säugetiere, z. B. Morganucodon oder Megazostrodon, klein, sehr agil und aktiv waren. Sie besaßen große Augen und hatten Zähne, die auf eine insektivore Ernährung hindeuten. Die Ähnlichkeit mit heute lebenden Igeln, Spitzmäusen und Tenreks hat Paläontologen seit über 30 Jahren zu der Annahme verleitet, diese ersten Säugetiere wären nachtaktiv gewesen. Eine nachtaktive Lebensweise hätte bei der Entstehung eines endothermen Metabolismus, einer für die Säugetierevolution entscheidenden Anpassung, unterstützend gewirkt.

Auch wenn der Fossilbericht der ersten Säugetiere in den letzten Jahren massiv an Quantität und auch Qualität zugenommen hat, kann dieser aufgrund fehlender Weichteilerhaltung keine neuen Erkenntnisse bezüglich einer nachtaktiven Lebensweise dieser Tiere liefern. Dank bedeutender Fortschritte in Wissen und Techniken der molekularen Evolutionsbiologie ist es heutzutage jedoch möglich, anzestrale Gensequenzen zu rekonstruieren und im Labor das darausfolgende Protein zu synthetisieren, sowie Selektionsdrücke, die auf Proteine gewirkt haben, genau zu analysieren.

Hier setzt die vorliegende Arbeit an. Sie untersucht das einzige Sehpigment in der Netzhaut von Wirbeltieren, welches für das Sehen bei Nacht und/oder Dämmerung verantwortlich ist: das Rhodopsin.

Zuerst wurde das Rhodopsin der nachtaktiven Echidna, die zu einer der zwei letzten lebenden Familien von Monotrematen, der basalsten lebenden Säugetiere, gehört, sequenziert. Zusammen mit zwei selbstkreierten Mutanten wurde dieses erfolgreich in vitro exprimiert, die biochemischen und funktionellen Eigenschaften analysiert und verglichen mit dem Rhodopsin der tagaktiven Kuh, welches bereits bestens in diversen Studien charakterisiert wurde. Die Untersuchungen ergaben, dass das Rhodopsin der Echidna auch Charakteristika von Farb-Sehpigmenten aufweist, was auf eine Expression von Rhodopsin in Zapfen hindeutet. Tests an Mutante 169 ergaben, dass diese Aminosäure an der Regulierung der Absorptionsstärke des Rhodopsins der Echidna beteiligt war.

Des Weiteren, basierend auf einem umfassenden Alignment von 27 Tetrapoden-Rhodopsinen, wurden anzestrale Proteinsequenzen für die Knotenpunkte Amniota, Mammalia und Theria (d.h. Marsupialia und Plazentalia) mithilfe der Maximum-Likelihood-Methode berechnet und wiederum erfolgreich in vitro synthetisiert: alle Pigmente erwiesen sich funktional und zeigten typische Rhodopsin-Charakteristika.

Ausserdem ergab die Messung der Halbwertszeit von Meta II, einem entscheidenden Aktivatorzustand des Rhodopsins in der visuellen Signalkaskade, einen im Vergleich zum Kuh-Rhodopsin erhöhten Wert, sowohl im hypothetischen Säugetier- als auch im hypothetischen Theria-Rhodopsin. Dies deutet auf eine Anpassung an besseres Sehen bei schwachen Lichtverhältnissen oder bei Dunkelheit hin. Es erwies sich aber als schwierig, aus einzelnen Funktionstests Schlussfolgerungen auf ökologisch-bedingte Anpassungen zu ziehen, da die visuelle Signalkaskade ein sehr komplexes und durch viele Proteine vernetztes System darstellt.

Zuletzt wurden mithilfe der Maximum-Likelihood-Methode Selektionsdrücke, die auf nicht-synonyme Substitutionen des Rhodopsins gewirkt haben, untersucht. Positive Selektion führt dazu, dass ein Protein sich Veränderungen in der Umwelt anpasst, wohingegen negative Selektion die ursprüngliche Funktion des Proteins manifestiert. Starke positive Selektion wurde allein entlang der Linie, die zu den Theria und auch derjenigen, die zu den Marsupialia führt, ermittelt. Entlang der Theria-Linie, im Mesozoikum, sind mehrere Einnischungsevents von Säugetiertaxa in neue Lebensräume im Fossilbericht belegt. Sehr wahrscheinlich spiegeln sich Anpassungen an neue Lebensräume in einem so adaptiven System wie dem der Sehpigmente wider. Des Weiteren wurde gezeigt, dass positive Selektion auf synonyme Substitutionen im Rhodopsin nur entlang der Mammalia-Linie gewirkt hat, was Auswirkungen auf die Stabilität der mRNA sowie die Translation der tRNA hat und weiter zu einer Zunahme der Rhodopsin-Moleküle führt. Diese Ergebnisse beschreiben ein mögliches Szenario, in dem die Säugetiere im Vergleich zu anderen Amnioten zunächst die Anzahl ihrer Rhodopsin-Moleküle gesteigert haben, möglicherweise als Anpassung an das Nachtsehen. Später erfuhr das Rhodopsin adaptive Veränderungen als Antwort auf die starke ökologische Diversifikation.

Die vorliegende Arbeit zeigt mithilfe bioinformatischer und molekularbiologischer Techniken, dass das Säugetier-Rhodopsin einige Veränderungen erfahren hat. Des Weiteren bringt sie zum Ausdruck, dass Paläontologie und Molekularbiologie sich gegenseitig unterstützen können und müssen, um interessante makroevolutionsbiologische Fragen zu lösen.

Inhaltsverzeichnis

• 1. Introduction

  • 1.1. The origin and evolution of mammals

    • 1.1.1. The origin of mammals

    • 1.1.2. Nocturnality – a prerequisite of endothermy

    • 1.1.3. Evolution of therapsids and the acquisition of endothermy

  • 1.2. Enigmatic monotremes, the most basal mammals

    • 1.2.1. Monotremes

    • 1.2.2.  Tachyglossus aculeatus, the short-beaked echidna

  • 1.3. Rhodopsin, a vertebrate visual pigment

    • 1.3.1. The visual signaling cascade

    • 1.3.2. Rhodopsin, a G protein-coupled receptor

  • 1.4. Ancestral sequence reconstruction and selective constraint analyses

    • 1.4.1. Resurrecting ancient genes

    • 1.4.2.  In vitro expression systems in vision research

    • 1.4.3. Selective constraint analyses

  • 1.5. Objectives of this thesis

• 2. Material and methods

  • 2.1. In the molecular lab

    • 2.1.1. Genomic DNA isolation

    • 2.1.2. Genome-walking PCR 

    • 2.1.3. Gene synthesis and site-directed mutagenesis 

    • 2.1.4. An adequate expression vector

    • 2.1.5. Protein expression 

    • 2.1.6. Western blot 

    • 2.1.7. Spectrophotometry 

    • 2.1.8. Functional assays: acid bleach, hydroxylamine sensitivity, and meta II decay rate 

  • 2.2. Maximum likelihood analyses

    • 2.2.1. PAML

    • 2.2.2. The dataset

    • 2.2.3. Selective constraint analyses

      • 2.2.3.1. Introduction

      • 2.2.3.2. Likelihood ratio test

      • 2.2.3.3. Branch models

      • 2.2.3.4. Branch-site models

    • 2.2.4. Ancestral sequence reconstruction

• 3. Results

  • 3.1. In the molecular lab

    • 3.1.1. The echidna rhodopsin sequence

    • 3.1.2. Three ancestral sequences 

    • 3.1.3. Western blot

    • 3.1.4. Dark and light spectra

    • 3.1.5. Acid bleach

    • 3.1.6. Hydroxylamine sensitivity

    • 3.1.7. Meta II decay by fluorescence spectroscopy

  • 3.2. The ancestral sequences and their structure

    • 3.2.1. Interesting sites

    • 3.2.2. Rhodopsin 3D structure

  • 3.3. Comparing protein-coding rhodopsin sequences from living taxa

    • 3.3.1. Substitutions unique to a taxon

    • 3.3.2. Substitutions unique to monophyletic groups

    • 3.3.3. Similar substitutions in different clades

  • 3.4. Selective constraint acting on the rhodopsin visual pigment 

    • 3.4.1. Introduction

    • 3.4.2. Branch models

    • 3.4.3. Branch-site models

    • 3.4.4. Summary

• 4. Discussion

  • 4.1. Nocturnal vs. diurnal 

    • 4.1.1. Characterisation of the echidna rhodopsin

    • 4.1.2. Characterisation of the two echidna mutants

    • 4.1.3. Inferring life habits from absorption maxima of living taxa

    • 4.1.4. Conclusions

  • 4.2. The ancestral rhodopsins

    • 4.2.1. Characterisation of the three ancestral rhodopsins

    • 4.2.2. The meta II decay rate

    • 4.2.3. Weak points of Maximum likelihood Inferences 

    • 4.2.4. Conclusions

  • 4.3. Positive selection on non-synonymous substitutions along the Therian branch

    • 4.3.1. Therian diversity during the Late Jurassic 

    • 4.3.2. The tetrapod opsin complement

    • 4.3.3. Selective constraint on synonymous substitutions in the mammalian rhodopsin

    • 4.3.4. Conclusions

  • 4.4. Summary and future prospects

• Acknowledgements

• Literature cited

• Publications

• Conference presentations 

• Erklärung

• Table 1. Self-designed degenerate primers used in first round hot-start PCR. 

• Table 2. Primers used in site-directed mutagenesis PCR in order to create echidna mutants T158A and F169A. 

• Table 3. Accession numbers of all sequences which were downloaded from NCBI and used in this study. 

• Table 4. Alignment of rhodopsin amino acid sequences used in this study.

• Table 5. Parameters of branch models used in this study.

• Table 6. Parameters of branch-site models used in this study.

• Table 7. Genomic DNA (gDNA) sequence, and complementary DNA (cDNA) sequence of the rhodopsin of the short-beaked echidna, Tachyglossus aculeatus.

• Table 8. Most likely hypothetical ancestral nucleotide sequences for the nodes Amniota, Mammalia, and Theria, inferred by maximum likelihood estimates.

• Table 9. Molecular weight estimates based on protein sequences

• Table 10. Absorption peaks of all rhodopsins expressed in this study.

• Table 11. Molar extinction coefficients determined for all proteins expressed in this study.

• Table 12. Meta II decay results and their coefficient of determination (R²) of ancestral pigments and bovine rhodopsin as positive control.

• Table 13. Meta II decay results and coefficients of determination (R²) of echidna rhodopsin and bovine as positive control.

• Table 14. Branch model estimates for the branch Amniota. np is number of parameters, LnL is log likelihood of the model.

• Table 15. Branch model estimates for the branch Reptilia. np is number of parameters, LnL is log likelihood of the model.

• Table 16. Branch model estimates for the branch Mammalia. * indicates statistical significance. np is number of parameters, LnL is log likelihood of the model.

• Table 17. Branch model estimates for the branch Monotremata. * indicates statistical significance. np is number of parameters, LnL is log likelihood of the model.

• Table 18. Branch model estimates for the branch Theria.

• Table 19. Branch model estimates for the branch Marsupialia.

• Table 20. Branch model estimates for the branch Placentalia.

• Table 21. Branch-site model estimates for the branch Amniota. np is number of parameters, df is degrees of freedom in Likelihood Ratio Test, LnL is log likelihood of the model.

• Table 22. Positively selected sites estimated by BEB analysis in branch-site model MA (Yang et al. 2005), with posterior probabilities, for branches Amniota, Reptilia, Monotremata, Theria, Marsupialia, and Placentalia.

• Table 23. Branch-site model estimates for the branch Reptilia. np is number of parameters, df is degrees of freedom in Likelihood Ratio Test, LnL is log likelihood of the model.

• Table 24. Branch-site model estimates for the branch Mammalia. np is number of parameters, df is degrees of freedom in Likelihood Ratio Test, LnL is log likelihood of the model.

• Table 25. Branch-site model estimates for the branch Monotremata. np is number of parameters, df is degrees of freedom in Likelihood Ratio Test, LnL is log likelihood of the model.

• Table 26. Branch-site model estimates for the branch Theria.

• Table 27. Branch-site model estimates for the branch Marsupialia.

• Table 28. Branch-site model estimates for the branch Placentalia.

• Table 29. 42 tetrapod taxa used in a Kruskal-Wallis test.

• Figure 1. Reconstruction of Morganucodon, an early mammal from the Late Triassic of China, South Africa, India, and Europe (Kemp 2005).

• Figure 2. Simplified synapsid phylogeny based on accepted literature

• Figure 3. A short-beaked echidna, Tachyglossus aculeatus, in Australia

• Figure 4. Wavelength diagram. SWS 1 absorbs light at 355-440 nm, SWS2 at 410-490 nm, MWS at 480-535 nm, LWS at 490-570 nm, and Rhodopsin at about 500 nm.

• Figure 5. Structural formula of 11-cis retinal.

• Figure 6. The phototransduction cascade in the vertebrate eye

• Figure 7. Secondary structure of bovine rhodopsin

• Figure 8. Three-dimensional structure of bovine rhodopsin.

• Figure 9. The ancestral gene resurrection strategy

• Figure 10. 1% agarose gel showing all three elutions and two DNA ladders.

• Figure 11. Establishing a genome walker library

• Figure 12. Structural formula of hydroxylamine.

• Figure 13. Reaction scheme of rhodopsin photoproducts

• Figure 14. Tetrapod phylogeny used in this study, with coelacanth and lungfish as outgroups. 

• Figure 15. Secondary structure of the echidna rhodopsin (modified after Sakmar et al. 2002). 

• Figure 16. Western blot analysis of expressed rhodopsin pigments.

• Figure 17. Dark (in red) and light (in black) absorption spectra of expressed and purified rhodopsins, i.e.

• Figure 18. Acid bleaches of (A) bovine, (B) echidna, (C) mutant T158A, (D) mutant F169A, (E) Amniota, (F) Mammalia, and (G) Theria rhodopsin. 

• Figure 19. Hydroxylamine assays performed on (A) bovine, (B) echidna, (C) mutant T158A, (D) mutant F169A, (E) Amniota, (F) Mammalia, and (G) Theria rhodopsins. 

• Figure 20. Amino acid alignment of the three inferred ancestral rhodopsins.

• Figure 21. Rhodopsin 3D structure of all pigments from this study.

• Figure 22. Summary figure showing selective constraints acting on rhodopsin along branches and on sites.

• Figure 23. Phylogeny showing meta II decay rates derived from this study. 

• Figure 24. Phylogeny of Mesozoic and extant mammalian groups (after Luo 2007). 

• Figure 25. Visual pigment loss in tetrapods.

• Figure 26. Distribution of G/C-ending codons in mammalian rhodospin gene.