Forschung / Research

Gene regulation

Regulatory signal transduction systems link extra- and intracellular stimuli with cellular responses. The success of an organism depends decisively on the optimal adaptation to the prevailing conditions. On the one hand, the existing regulatory mechanisms play a role when it comes to natural fluctuations within the ecological niche. On the other hand, the flexibility of the regulatory networks is also important, especially when it comes to the conquest of new habitats with a changed range of fluctuation. Predominately, bacteria rely on two major types of regulation systems when it comes to transcriptional control. The first, are so called one-component system, which combine input and output domains within a single protein. Second, in two-component signal-transduction input and output are separated on at least two distinct proteins - a histidine kinase and a response regulator - in turn enabling more complex regulatory circuits. Whereas the number of outputs is limited and most often results in transcriptional control a sheer unlimited amount of signal can be recognized by an arsenal of input domains.
In my group we are interested to identify new transcriptional control regimes towards non-canonical amino acids such as fructose lysine and to understand these systems at the molecular level.

Translational stress response

Synthesis of polyproline proteins leads to translation arrest. To overcome this ribosome stalling effect, bacteria depend on a specialized translation elongation factor P (EF-P), being orthologous and functionally identical to eukaryotic/archaeal elongation factor e/aIF-5A (recently renamed ‘EF5’). EF-P binds to the stalled ribosome between the peptidyl-tRNA binding and tRNA-exiting sites, and stimulates peptidyl transferase activity, thus allowing translation to resume. Most interestingly EF-P also assists in peptide bond formation with non-canonical amino acids, to enable production of specific synthetic proteins.
In their active form, both EF-P and e/aIF-5A are post-translationally modified at a positively charged residue, which protrudes toward the peptidyl transferase center when bound to the ribosome. While archaeal and eukaryotic IF-5A strictly depend on (deoxy-) hypusination of a conserved lysine, bacteria have evolved diverse analogous modification strategies to activate EF-P. In Escherichia coli and Salmonella enterica a lysine is extended by lysylation and subsequently hydroxylated, whereas in Pseudomonas aeruginosa an arginine in the equivalent position is rhamnosylated.
In my group we are interested to understand why nature has evolved such specialized translation factor and investigate its impact on condition-dependent proteome shaping. We are further interested in the chemical diversity of post-translational EF-P modification systems and how we utilize different EF-P subtypes to repurpose the translational machinery for improved yields of heterologously expressed genes. We have therefore developed a fluorescent reporter that allows us to specifically enrich cells, with improved translational properties.

Post-translational modifications

Post-translational modifications (PTM) are the evolutionary solution to challenge and extend the boundaries of genetically predetermined proteomic diversity. Since PTMs are highly dynamic, they also hold an enormous regulatory potential. It is therefore not surprising that out of the 20 proteinogenic amino acids, 15 can be post-translationally modified. In higher eukaryotes, up to 5 % of the total genome can be dedicated to such modifiers. Although bacteria are often considered to be simple organisms with very basal cellular regulation, their proteome is also subject to substantial directed post-translational changes of largely unknown molecular function.

A) The sweet sites of bacteria: Unraveling the regulatory and biotechnological potential of prokaryotic protein sugar modifications

The attachment of a sugar moiety onto a protein is a potent physiological strategy to alter its structural and functional properties. Although these post-translational modifications were believed to be restricted to eukaryotes, it is now broadly accepted that they also occur in bacteria (Nothaft & Szymanski, 2010). However, there is virtually no study about the physiological role of the non-enzymatic glycation in prokaryotes (Mironova et al., 2001) and most publications on enzymatic protein glycosylation have been focused on one single pathway previously identified in Campylobacter jejuni (Nothaft & Szymanski, 2010). Thus, it has remained enigmatic whether bacteria have evolved alternative strategies of protein glycosylation and its impact on physiology. This is especially surprising as glycosylations are most likely the most widespread post-translational modifications in nature (Khoury et al., 2011) and also play a role in organismic interaction such as pathogenicity development as demonstrated by our discovery of the glycosyltransferase EarP (Lassak et al., 2015, Li et al., 2016, Krafczyk et al., 2017). In the past, proteome studies have presumably been hampered by insufficient detection and enrichment strategies for glycoproteins. Now we have a set of rhamnosyl specific antibodies in hand, and with this already succeeded in the detection of novel mono-glycosylated proteins both in bacteria as well as in human, proofing this approach highly promising. In close collaboration with Prof. Anja Hoffmann-Röder heading the LMU Munich organic chemistry group, we will continue to develop new glycoconjugates and antibodies as a prerequisite for further glycoprotein enrichment that precedes an MS-based proteome analysis. Having determined a bacterial glycoproteome we will investigate the impact of glycosylation on protein function and will identify and analyze the corresponding glycosyltransferases. Currently, we have a particular interest to understand a novel group of effector glycosyltransferases, which we recently identified in pseudomonads and have a presumed role in host-microbe interaction. We are also intrigued by EarP mediated rhamnosylation and how it contributes to Pseudomonas infection. In this regard we are currently investigating cyclized small peptides as inhibitors of the glycosyltransferase reaction. We also have the vision to engineer in a way that allows for targeted transfer of a variety of sugars including the most important N-acetylglucosamine. The latter is of particular biotechnological impact as it will allow to reproduce eukaryotic glycan pattern in bacteria.
 

B) Genetic code expansion

Nature employs a limited and conservative set of amino acids to synthesize proteins. The ability to genetically encode an extended set of building blocks can be used in diverse applications, including approaches to study and control protein function as well as to design novel therapeutics.
Non-natural amino acids (NAA) are co-translationally incorporated into proteins by orthogonal pairs consisting of aminoacyl-tRNA synthetase and cognate tRNA. However, the current repertoire can neither display the full natural diversity of NAAs and is especially limited for backbone modifications. Excitingly, I now succeeded in reverse engineering a protein ligase into a new-to-nature tRNA synthetase (aaRS-β) that can load tRNA with β-amino acids (βaa): Ribosomal protein biosynthesis is naturally restricted to L-α-amino acids. By contrast, diverse non-ribosomal peptides contain various β-amino acids (βaa) showing a selective advantage (Kudo et al., 2014). Indeed, these βaa are essential for bioactivity, membrane permeability, specificity and stability (Cabrele et al., 2014). Moreover, peptides bearing stretches of consecutive βaa can form unique secondary structures, such as the 10-, 12- and 14-helices, which generally have higher stabilities than their α-helical counterparts (Katoh & Suga, 2018). Some β-peptides even self-assemble into higher-order structures, including nanofibers, nanosheets, helical bundles and liquid crystals (Gopalan et al., 2015, Kwon et al., 2015, Seoudi & Mechler, 2017). Due to these characteristics, many foldamers contain or are exclusively composed of βaa residues.
We have now made the first big step towards synthesis of such proteins and we will apply our knowledge for the production of antimicrobial peptides with high protease stability and increased activity against multidrug-resistant pathogens.
Additionally, we generated a variant aarRS-α accepting α-amino acids that cannot be integrated into the genetic code by any other means. Among them are advanced glycation end products (AGEs): The accumulation of AGEs on nucleotides, lipids, and especially peptides/proteins is an inevitable part of the aging process in all eukaryotic organisms, including humans (Delgado-Andrade, 2016). There is ample evidence that AGEs and their functionally impaired adducts are related to, and possibly responsible for, changes during aging and the development of many age-related diseases (Di Sanzo et al., 2021). Among AGEs Nε-carboxymethylysine (CML) is one of the most abundant. Its formation is triggered, among other routes, by glyoxal. CML is known to be chemically stable, to accumulate in human tissues in diabetes, atherosclerosis, neurodegeneration, and thus is an important biomarker here. More and more evidence are provided for a causal relationship between the buildup of AGEs and aging and individual diseases. In fact, protein glycation is increasingly seen as driver of metabolic disease and aging, and may elicit specific effects by targeting signaling proteins. However, due to the spontaneous nature of this PTM and a lack of tools to expand the genetic code with CML, a deeper understanding is currently missing. With my aarRS-α I am currently expanding the genetic code with CML and other AGEs to understand how these PTMs impact on aging disease development at the molecular level.

C) Non-canonical food: Prokaryotic metabolism of protein glycations

AGEs are not only generated by our body but also by thermal food processing (Lassak et al. 2019 & 2022). Notably, humans are not able to degrade them on their own but rely on the gut microbiota (Hellwig et al. 2019). However, our current understanding of the bacterial metabolism towards NAAs in general and AGEs in particular is scarce. We are currently focusing on CML and its higher ortholog Nε-Carboxymethyllysine (CEL) both of which are especially enriched. e. g. in milk or protein powder. Removal of this ingredients is of high biotechnological significance as CML/CEL are not only important as PTM in human protein but the free compound was also found to have negative implications on human health. Thus, it is also our mission to identify bacteria with such novel metabolic capabilities or to utilize enzymes in model organisms with underground activity against CML and CEL. In this regard we have now identified the first enzyme catalyzing CML degradation by E. coli. At the same time, we have isolated pseudomonads, encoding a CML/CEL-Lyase (an enzyme which has not known thus far) enabling the utilization as sole carbon source (Mehler et al., 2022).
When understanding especially the uptake of NAAs, we have also the tools in hand to reduce cost of protein production that relies on their external supply and inefficient uptake by promiscuous transporters.