Expanding and exploring the bioconjugation toolbox

A large variety of biomolecules such as lipids, DNA, carbohydrates and proteins exist inside cells. Every biomolecule fulfills a different type of function and it is highly relevant to study these functions to eventually understand the development of certain diseases. In order to study the role of a single protein, it is necessary to selectively mark this protein inside cells. The interest in methods that can be used for the site-selective modification of biomolcules has increased tremendously over the past few years. For example, by selectively modifying a protein of interest with a fluorescent label inside cells or organisms, it is possible to learn more about its function, localization and interactions with other molecules. A frequently used method for the visualization of proteins is to fuse a second, fluorescent protein to the protein of interest. However, this fluorescent protein is large and might interfere with the function of the protein of interest. Therefore, scientists started searching for chemical reactions that can be used to couple a small dye to a protein in a very controlled manner. In addition to fundamental research, the siteselective modification of proteins is also important for therapeutic applications, e.g. in the development of biopharamceuticals. The biostability of a pharmaceutical protein can be improved by the attachment of certain molecules. A high level of control of the attachment is important in order to avoid interference with the binding of the protein to its target and hence influencing the therapeutic effect of the protein. Classical methods for protein modification involve attachment via natural functionalities such as the amine of the lysine residue or the thiol of the cysteine residue. These methods are slowly being replaced by chemical reactions that are highly selective, functional under physiological conditions, and are non-toxic. These reactions are called biorthogonal reactions. One of the most popular bioorthogonal reactions is the reaction of an azide with a cyclooctyne, called strain-promoted azide–alkyne cycloaddition (known as SPAAC or copper-free click chemistry). This reaction inspired many researchers to search for more reactions that can be used for the labeling of proteins and other biomolecules. Chapter one gives an overview of the application of bioorthogonal reactions in fundamental as well as applied research. Chapter two describes the genetic encoding of the new unnatural amino acid bicyclononyne-lysine (BCN-lysine). The pyrrolysyl-tRNA and a mutant of the pyrrolysyl-tRNA synthetase from Methanosarcina mazei were used to site-specifically incorporate BCN-lysine into GFP in response to the amber stop codon. A Forster resonance energy transfer assay was used to study the kinetics of the reaction of BCNmodified GFP with azido- or tetrazine-containing fluorogenic dyes and was compared to an amino acid harboring a plain cyclooctyne moeity. BCN was shown to react approximately 10 times faster with azidotetramethylrhodamine than plain cylcooctyne. The reaction with tetrazine was 80 times faster and is one of the fastest bioconjugation reactions reported so far. BCN-lysine could also be incorporated into proteins inside mammalian cells and could be used for fast protein labeling inside cells. We have developed a new bioconjugation reaction based on the reaction of quinones with cyclootynes, which is called strain-promoted oxidation-controlled cyclooctyne–1,2-quinone cycloaddition (SPOCQ). In Chapter three we characterize this reaction and show its application for the labeling of peptides and proteins. 1,2-quinones can be generated by oxidation from the corresponding catechols, which can be used as an activatable bioconjugation reaction. The reaction between 1,2-quinones and BCN was shown to be much faster (k = 496 ± 70 M-1s-1) than the reaction of BCN with azides (strain-promoted azide-alyne cycloaddition, SPAAC). In chapter four we studied hydrogel formation using the SPOCQ reaction. 4- armed star-PEG-BCN and 4-armed star-PEG containing 3,4-dihydroxyphenylacetic acid (star-PEGDHPA) were mixed and upon the addition of the oxidizing agent hydrogels formed almost instantaneously. When the enzyme, mushroom tyrosinase, was used as the oxidizing agent gelation times could be tuned from seconds to several minutes by using different concentrations of the enzyme. By using an excess of the star-PEG-BCN polymer for hydrogel formation, we showed that hydrogels could be formed and simultaneously functionalized with azido-compounds because of the large difference in rate constants between the SPAAC and SPOCQ reactions. In chapter five, we aimed to prepare a new vaccine formulation against bacterial infections by decorating surfaces of bacteria with Fc fragments, which is the tail of an antibody that activates the immune system, using bioorthogonal chemistry. The surface of Staphylococcus aureus were functionalized with azidogroups by culturing the bacteria in the presence of D-azidoalanine. D-Azidoalanine incorporation was validated using a fluorescent cyclooctyne-dye. Fluorescence microscopy and flow cytometry analysis showed facile incorporation of the azido-groups. An enzymatic as well as chemical approaches was used to modify the Fc protein with a cyclooctyne moiety. The latter method resulted in a higher labeling efficiency showing approximately 80% functionalization of Fc protein based on gel-shift assay analysis. Cyclooctyne-functionalized Fc fragments could not yet be coupled to the bacteria yet, because the reaction is most likely sterically hindered by cell-wall glycopolymers that are present in the cell wall of Staphylococcus aureus. Chapter six introduces a tagless workflow for the enrichment of proteins for mass spectrometry-based proteomics to study protein-protein interactions (PPIs). GFP, either fused to the N- or C-terminus of a protein of interest, is a frequent tag for the enrichment of a single protein and its interactors. We aimed to develop an alternative strategy for fusion tag-based enrichment protocols by combining amber suppression technology and bioorthogonal chemistry. Azido-phenylalanine (azF) was first genetically encoded into the model protein mCherry-GFP. We then tested the efficiencies of copper-catalyzed and copper-free click chemistry for the labeling and enrichment of this model protein from cell lysate. Protein precipitation was observed with the copper-catalyzed click chemistry, whereas incubation with azadibenzocyclooctyne-functionalized agarose beads resulted in selective enrichment of azido-tagged mCherry-GFP from cell lysate. Testing the feasibility of our novel enrichment strategy for studying interactomes, we incorporated azF at position 211 into methyl-CpG-binding domain protein 3 (MBD3), a member of the Nucleosome Remodeling and Deacetylase (NuRD) complex. Mass spectrometry analysis showed that MBD3 could be selectively enriched with two of its known interactors. This finding shows that direct enrichment of a protein of interest and its interactors is feasible. Based on this result, we hypothesize that MS-data from experiments with different mutants, each containing one azF at another position, might be combined for the complete identification of all PPIs.