Abstract

Microtubules (MTs) are hollow polymeric tubes formed of α/β-tubulin heterodimers that are in a constant dynamic equilibrium between growth and shrinkage. MTs are an essential constituent of the cytoskeleton and so are critical in many cellular processes, such as mitosis, cargo trafficking and cell migration. Consequentially, modulation of MT dynamics is a promising objective in the development of biological research tools, as well as therapeutic drugs. However, since MTs fulfil many different physiological roles at the same time, global manipulation of the MT network often causes serious off-site effects. This poses a major challenge in MT biology research. A major goal is thus to develop MT-targeting research tools whose activity can be precisely controlled: in terms of location (spatial) to selectively address the area under study, and at a given time point (temporal). Owing to its unrivalled precision, non-invasiveness, and reversibility, light is an almost ideal stimulus for such spatiotemporally precise modulation of bioactivity. Photopharmacology harnesses these unique features of light to create high-precision photopharmaceuticals. Reversible photopharmaceuticals are bioactive small molecules derivatized with a molecular photoswitch that has two photoisomers that differ in binding affinity towards the biological target. Driven by light, these two isomers are interconverted between each other, thus enabling optical control over protein activity. Previous approaches towards optical control over MT dynamics have mainly been based on azobenzene-derived photopharmaceuticals. However, azobenzenes are by no means ideal photoswitches, in that they suffer from the need of potentially phototoxic UV light, incomplete photoisomerization, incompatibility with certain functional groups and metabolic lability, among others. Therefore, this thesis aims at the development of MT photopharmaceuticals that overcome some of the limitations and weaknesses of azobenzene-based approaches. These photopharmaceuticals are based on hemithioindigo (HTI), an emerging photoswitch with various beneficial properties, whose use in photopharmacology has however been scarce and far from systematical. In the following, I describe the development of hemithioindigo-based photopharmaceuticals for MT biology. Paper One demonstrates the first use of HTIs in cellular photopharmacology (HOTubs). We showcase their potential for long-term photocontrol over the MT network, and their suitability for photopharmacology in general. We show in live cells that these HTIs light-dependently induce cytotoxicity, disruption of the MT network and mitotic cell cycle arrest. They also bring unique possibilities: for most photoswitches including azobenzenes, isomerization is accompanied by large changes in sterics and consequentially it is often the sterics of the protein binding site that decides which photoisomer is the stronger binding one; but since both HTI isomers are nearly planar, we could use rational design rather than protein sterics to control which of the two HOTub isomers was the more bioactive. Taken together with the absence of unspecific toxicity and the avoidance of potentially phototoxic UV light, our findings strongly recommend HTIs as general and cell-compatible scaffolds for photopharmacology. Paper Two aims at increasing the cellular potency of our HTI-based tubulin photopharmaceuticals. I design a SAR study and outline how the structural similarity of a highly potent indanone and hemithioindigo can be harnessed to create research tools (HITubs) exceeding the potency of all MT photopharmaceuticals known at the time. Another key aspect of this paper is to show that, unlike azobenzenes, the metastable isomer of HTI photopharmaceuticals has slow enough relaxation even with a strongly electron-donating, tautomerizable -OH group in the para-position, to allow its use as a bistable photoswitch. This substantially widens the chemical scope of photopharmaceuticals and thus expands the range of biological targets addressable by photopharmacology. Paper Three describes how structural finetuning of the photoswitch scaffold yields MT photopharmaceuticals with near-quantitative photoisomerization (PHTubs). This enables studies void of undesired background of bioactivity which would stem from incomplete photoisomerization. For this purpose, I design a SAR study and synthesize tools based on pyrrole hemithioindigo (PHT), a photoswitch closely related to HTI. We further show that our PHTubs are sturdy photopharmaceuticals, that unlike azobenzenes, resist glutathione-mediated degradation in both isomeric states. This is an important prerequisite for more advanced studies, and we illustrate how PHTs can be implemented to light-dependently modulate the MT network in living cells and with single-cell precision. Beside these topics, I briefly present my unpublished work on two other projects. The first aims for optical control over γ-tubulin, an isoform involved in cellular MT nucleation. Taking advantage of the similarity between an isoflavone-based γ-tubulin specific inhibitor and HTI, I design an HTI-based photopharmaceutical thought to target MT nucleation (phatastatin). We are currently optimizing a key step en route to phatastatin. The second project aims towards optical control over proteasomal degradation of tubulin. In the context of a research stay in the Trauner lab at New York University (NYU), I designed and synthesized a series of photoswitchable tubulin degraders (TubPHOTACs), and we are currently evaluating two promising compounds for their cellular mechanism. In summary, this work contributes to light-dependent modulation of MT dynamics and introduces indigoid photoswitches (HTI and PHT) as sturdy photoswitches allowing for visible-light powered, reversible and highly precise studies in photopharmacology.