Refine
Year of publication
Document Type
- Article (120)
- Doctoral Thesis (93)
Language
- English (213) (remove)
Has Fulltext
- yes (213)
Is part of the Bibliography
- no (213)
Keywords
- - (72)
- Biokatalyse (19)
- biocatalysis (19)
- Proteindesign (11)
- Biocatalysis (10)
- Enzym (8)
- protein engineering (7)
- Protein-Engineering (6)
- Protein Engineering (5)
- Bacteroidetes (4)
Institute
- Institut für Chemie und Biochemie (213) (remove)
Publisher
- Wiley (39)
- MDPI (29)
- Frontiers Media S.A. (13)
- Springer Nature (10)
- John Wiley & Sons, Ltd (7)
- John Wiley & Sons, Inc. (3)
- Nature Publishing Group (2)
- Oxford University Press (2)
- BioMed Central (BMC) (1)
- Blackwell Publishing Ltd (1)
In their idealized forms, enzymes can facilitate complex reactions with extreme specificity and selectivity. Additionally, in this imaginative form, they only require mild reaction conditions, resulting in low energy consumption, and they are biodegradable, efficient, reusable, and sustainable. Unfortunately, this idealized form often deviates significantly from reality, where enzymes are more likely to be associated with marginal stability and low reaction rates, leaving them less than desirable for many industrial applications. As such, if we could master the process of engineering the configuration of a protein towards a given task, the implications could be staggering.
This thesis aims to contribute to the process of protein engineering, mainly how computational tools can be used to make the protein engineering process more efficient and accessible.
Article I explores the current state of the art in machine learning-guided directed evolution and serves as a foundation for Article II, which is a concrete application of these techniques to an engineering campaign. Despite successfully improving overall activity and selectivity, we also observe limitations and constraints within the methodology. Article III then delves into these drawbacks and attempts to lay the foundation for a more generalizable and, more importantly, efficient engineering workflow, balancing the strengths and weaknesses of computational techniques with advances in gene synthesis. We then validated this novel pipeline in Article IV, where we show the potential of this methodology. Article V describes a more standard protein engineering campaign on squalene-hopene cyclases for potentially interesting products in the flavor and fragrance industry. Lastly, Article VI outlines a PyMol plugin for molecular docking.
Marine algae are essential for fixation of carbon dioxide, which they transform into complex polysaccharides. These carbohydrates are degraded e.g., by marine Bacteroidetes and the understanding of their decomposition mechanism can expand our knowledge how marine biomasses can be accessed. This understanding then gains insights into the marine carbon
cycle. This thesis summarizes the current knowledge of marine enzymatic polysaccharide degradation in review Article I and extents a previously discovered ulvan degradation pathway in Article II with the description of a novel dehydratase involved in the ulvan degradation pathway. This enlarged ulvan-degradation pathway can be used to generate fermentable sugars from the algal derived polysaccharide ulvan. A potential biorefinery process is proposed in Article III, where B. licheniformis was engineered to degrade ulvan, thus establishing the initial steps for a microbial cell factory development. In addition to ulvan, also plenty of other complex carbohydrate sources are present in the ocean. The enzymatic elucidation principles previously developed were thus adapted towards a new marine carbohydrate. In Article IV a xylan utilization pathway was elucidated, using enzymes present in Flavimarina Hel_I_48 as model bacterium. The Flavimarina genome contains two separated genome clusters which potentially targets xylose containing polymers reflecting the diversity and adaptions towards different marine xylan-like substrates. Besides, marine Bacteroidetes are adapted towards decomposition of methylated polysaccharide, e.g., porphyran, via demethylation catalyzed by cytochrome P450 monooxygenases. This reaction results in the formation of toxic formaldehyde and thus the marine Bacteroidetes require formaldehyde detoxification principles. The analysis of potential formaldehyde detoxification mechanisms revealed a marine RuMP pathway (Article V) and a novel auxiliary activity of an alcohol dehydrogenase of which the encoding gene is adjacent to the demethylase cluster (Article VI).
Impact of proteostasis and the ubiquitin proteasome system on myeloid cell function in the CNS
(2023)
Cellular protein homeostasis (proteostasis) maintains a functional proteome and thus proper cell function. Proteostasis is facilitated by the ubiquitin-proteasome system (UPS), an intracellular protein turnover machinery ensuring clearance of damaged, misfolded, old and/or unneeded regulatory proteins. This is particularly important in the central nervous system (CNS), where it is linked to neurodegeneration. Disruptions of the proteostasis systems cause the accumulation of misfolded proteins which are commonly seen in progressive neurodegenerative diseases also linked to neuroinflammation. Proper UPS function can protect cells from the accumulation of defective proteins, neurodegeneration and neuroinflammation. Furthermore, it has been found that loss of function mutations in the genes encoding UPS components are linked to systemic inflammation including neuroinflammation and/or neurodevelopmental disorders. Proteasome defects in patients suffering from these disorders cause decreased proteasome activity, accumulation of proteins, activation of proteotoxic stress responses and systemic inflammation. However, the molecular link between proteotoxic stress and the initiation of inflammatory signalling remained unclear. In Article 2, we summarized the importance of the UPS in immune cell proteostasis and function including activation of innate and adaptive immune responses. Although UPS function is notably important in innate immune signalling, the current understanding of the role of UPS in myeloid cell function in the CNS is limited. We also indicated the involvement of impaired UPS function in sterile systemic inflammation including neuroinflammation as well as tumour diseases and pathogen manipulation of immune cells.
To investigate the molecular link behind proteasome impairment and systemic inflammation in the brain, we focused on microglia cells as the only immune residents of the CNS. In Article 1, we used a pharmacological inhibitor called bortezomib which targets β5 and β5i/LMP7 subunit activities in standard proteasome (SP) and immunoproteasome (IP), respectively. We showed for the first time on the molecular level that inhibition of proteasome activity by bortezomib triggers the accumulation of ubiquitylated proteins, proteotoxic stress responses and innate immune signalling activation depending on the induced proteotoxic stress response called unfolded protein response (UPR) in murine microglia. In particular, activation of the inositol-requiring protein 1α arm of UPR upon bortezomib treatment leads to systemic inflammation as indicated by type I interferon (IFN) response.
IP enhance the proteolytic capacity of UPS by rapid clearance of proteins upon immune signalling activation. Microglia, like other immune cells, exhibit constitutive expression of IP as well as SP to maintain their cellular proteostasis. In Manuscript 3, we studied the particular impact of IP impairment on microglial cellular function. We showed accumulation of ubiquitin-modified proteins and activation of proteotoxic stress responses in IP-impaired mouse and human microglia models. Moreover, we identified possible IP substrates in microglia using β5i/LMP7 knockout mice as an IP deficiency model and, examined how IP deficiency affects microglia function. IP deficient microglia affected the ubiquitylation levels of proteins involved in multiple pathways such as immune responses, energy metabolism, cytoskeleton organisation, cell cycle and ribosome function. Based on the molecular analysis, we confirmed sterile activation of innate immune signalling mechanisms in IP impaired microglia. This is driven by the proteotoxic stress sensor protein kinase R (PKR). In addition, we were able to show that IP impairment altered levels of the microglial activation markers, which are also involved in motility, adhesion and phagocytosis of microglia.
In this thesis, we highlight that UPS function is necessary to maintain microglial proteostasis and, that impairment of proteasome activities triggers sterile inflammation in microglia via activation of proteotoxic stress responses. The described activation of innate immune signalling mechanisms in microglia upon proteasome impairment may be considered as new therapeutic targets for patients suffering from rare protesomapathies or other disorders linked to dysregulated immune signalling.
In this thesis, new catalysts as well as unprecedented approaches for the
valorization of sustainable carbon sources were investigated. The first part deals with the design of catalysts for photocatalytic CO2 reduction (Articles I&II). The promiscuous activity of phenolic acid decarboxylase from Bacillus subtilis (BsPAD) was found to catalyze CO2 reduction (Article I). This cofactor-free enzyme could facilitate the replacement of (noble) metal catalysts regularly employed in CO2 reduction. Based on these findings, additional enzyme catalysts were identified for photocatalytic CO2 reduction. The second part (Articles III-VII) focuses on the valorization of resources obtained from biomass, such as olive mill waste water or lignin, by the promising acyltransferases/hydrolase PestE from Pyrobaculum calidifontis VA1 (Articles IV-VII). The potential of PestE for the valorization of sustainable sources has been demonstrated by enzyme engineering and use in (chemo)enzymatic cascade reactions leading to value-added products.
Emerging zoonotic viruses are a constant threat to human and animal health. Therefore, knowledge about the host factors influencing viral pathogenicity is highly welcome as a basis for developing treatment or vaccine strategies. In order to identify host factors that potentially determine the
pathogenicity of three highly pathogenic (’high consequence’) zoonotic viruses, the interactomes of
selected viral proteins were analysed in parallel with the interactomes of the homologous proteins from closely related viruses which lack high pathogenicity. For this purpose, affinity purification mass spectrometry (AP-MS) was performed with the virus proteins as baits and lists of candidate proteins were generated that may determine the pathotype and warrant follow-up studies to characterise their function concerning the viral life cycles. In detail, the interactomes of virus pairs from the arenaviruses, filoviruses and henipaviruses were studied. The following protein homologues were selected: for filoviruses, the transcription factor VP30, the co-transcription factor VP35 and matrix protein VP40 of the non-pathogenic Reston virus
(RESTV, species Reston ebolavirus), the pathogenic Ebola virus (EBOV, species Zaire ebolavirus),
and, in addition, the Lloviu virus (LLOV, species Lloviu cuevavirus); in case of the arenaviruses
the nucleoprotein (NP), matrix protein (Z) and glycoprotein (GP) of the pathogenic Junín virus (JUNV, species Argentine mammarenavirus) and the non-pathogenic Tacaribe virus (TCRV, species Tacaribe mammarenavirus); and for the henipaviruses, the fusion protein F of the apathogenic Cedar virus (CedV, species Cedar henipavirus) and the pathogenic Nipah virus (NiV, species Nipah henipavirus). The experimental approach was to express the tagged bait proteins in human cells by transfection with appropriate constructs, purify the interactomes by affinity enrichment and analyse their protein content by MS. Quantitation was performed by labelling with stable isotopes or by label-free quantification (LFQ). High-confidence interactions for the LFQ approach were identified using the Mass Spectrometry interaction STatistics (MiST) scoring tool. Qualitative and quantitative data were used to identify a limited number of candidates for follow-up research. Additionally,
the interactomes were analysed with bioinformatical tools like term enrichment analysis and network analysis to identify cellular pathways which are possibly impacted by the expression of viral proteins. A novel specific interactor of EBOV VP30 was identified, ubiquitin carboxyl-terminal hydrolase7
(USP7, also known as HAUSP), and the interaction was partially characterised. The interaction was confirmed by reverse-pull-down experiments, and the Kd value (determined by Microscale Thermophoresis, MST) was found to be lower than for the interaction of USP7 with the RESTV VP30.
This work adds insight into virus protein interactomes, especially for the often neglected low pathogenic virus species. Furthermore, the pathogenicity of the viruses was refl ected to some degree
in the interactomes of their proteins. The generated interactome data for the different virus species
create a basis in the search for interactions that determine pathogenicity.
This dissertation explores and tries to unravel the fundamental basis of G-quadruplex end-folding as well as G-quadruplex interactions with small molecules by thermodynamic and structural approaches. Selective targeting of G-quadruplexes with ligands remains elusive, either because the ligand has
considerable binding affinity for other DNA structures or because it fails to discriminate between different G-quadruplex topologies. Unique structural motifs on the G-quadruplex may enhance or inhibit ligand binding to the G-quadruplex. For such aspects, it is necessary to understand the effect of G-quadruplex motifs or elements on the end-folding in order to better tune certain G-quadruplex topologies as model systems. Importantly for targeting G-quadruplex with ligands, motifs called Quadruplex-duplex (QD) junctions and interfaces are shown to be a binding hotspot
for various G-quadruplex ligands containing an intercalator motif. Binding affinity and selectivity of the ligands are discussed with the support of the NMR structures.
This dissertation focuses on the characterization of novel enzymes and metabolic pathways that fulfill crucial functions during marine carbohydrate degradation by Bacteroidetes and thus contributes to an advanced understanding of the global carbon cycle. Depolymerization and utilization of marine polysaccharides by Bacteroidetes requires a tremendous repertoire of enzymes with a wide range of functions. For instance, during the breakdown of the marine red algal polysaccharide porphyran, an oxidative demethylation of the methoxy sugar 6-O-methyl-D-galactose (G6Me) by cytochrome P450 monooxygenases occurs. This reaction produces huge amounts of cytotoxic formaldehyde, marine bacteria capable of degrading porphyran must therefore possess suitable formaldehyde detoxification pathways. Consequently, Article I focus on the identification of possible formaldehyde detoxification pathways in marine
Flavobacteriia, which led to the discovery of the ribulose monophosphate pathway as specific pathway for the detoxification of formaldehyde in certain Bacteroidetes like Zobellia galactanivorans. Furthermore, it was demonstrated in Article II that alcohol dehydrogenases play an essential role in the microbial utilization of G6Me and therefore possess a function in porphyran degradation. Discovering novel enzymes, entire enzymatic cascades or biotechnologically important microorganisms that can metabolize these marine carbohydrates also contributes to the utilization of marine polysaccharides as feedstock for potential biotechnological applications. A prospective biorefinery process was proposed in Article III by the identification of Bacillus licheniformis as promising utilizer of marine carbohydrate-derived monosaccharides and the creation of a microbial cell factory capable of growing on ulvan, a marine carbohydrate obtainable from algal bloom-dominating green algae, enabling an industrial use of the renewable and abundant algal biomass in future.
The transition to Ni‐based battery cathodes enhances the energy density and reduces the cost of batteries. However, this comes at the expense of losing energy efficiency which could be a consequence of charge–discharge hysteresis. Here, a thermodynamic model is developed to understand the extent and origin of charge–discharge hysteresis in battery cathodes based on their cyclic voltammograms (CVs). This was possible by defining a Gibbs energy function that weights random ion insertion/expulsion, i. e., a solid solution pathway, against selective ion insertion/expulsion, i. e., a phase separation route. The model was verified experimentally by the CVs of CoOOH and Ni(OH)2 as solid‐solution and phase‐separating cathodes, respectively. Finally, a microscopic view reveals that phase separation and hysteresis are a consequence of large ionic radii difference of the reduced and oxidized central metal atoms.
Amine transaminases (ATAs) are powerful biocatalysts for the stereoselective synthesis of chiral amines. Machine learning provides a promising approach for protein engineering, but activity prediction models for ATAs remain elusive due to the difficulty of obtaining high-quality training data. Thus, we first created variants of the ATA from Ruegeria sp. (3FCR) with improved catalytic activity (up to 2000-fold) as well as reversed stereoselectivity by a structure-dependent rational design and collected a high-quality dataset in this process. Subsequently, we designed a modified one-hot code to describe steric and electronic effects of substrates and residues within ATAs. Finally, we built a gradient boosting regression tree predictor for catalytic activity and stereoselectivity, and applied this for the data-driven design of optimized variants which then showed improved activity (up to 3-fold compared to the best variants previously identified). We also demonstrated that the model can predict the catalytic activity for ATA variants of another origin by retraining with a small set of additional data.
Boronate esters formed by reaction of an oligonucleotide carrying a 5′-boronic acid moiety with the 3′-terminal cis-diol of another have been shown previously to assist assembly of fragmented DNAzymes. Here we demonstrate that boronate esters replacing the natural phosphodiester linkage at selected sites of two functional RNAs, the hairpin ribozyme and the Mango aptamer, allow assembly of functional structures. The hairpin ribozyme, a small naturally occurring RNA that supports the reversible cleavage of appropriate RNA substrates, is very sensitive to fragmentation. Splitting the ribozyme at four different sites led to a significant decrease or even loss of cleavage and ligation activity. Ribozymes assembled from fragments capable of boronate ester formation showed restoration of cleavage activity in some but not all cases, dependent on the split site. Ligation proved to be more challenging, no supportive effect of the boronate ester was observed. Split variants of the Mango aptamer also showed a dramatic loss of functionality, which however, was restored when 5′-boronic acid modified fragments were used for assembly. These studies show for the first time that boronate esters as internucleoside linkages can act as surrogates of natural phosphodiesters in functional RNA molecules.
Baeyer-Villiger monooxygenases (BVMOs) are important flavin-dependent enzymes which perform oxygen insertion reactions leading to valuable products. As reported in many studies, BVMOs are usually unstable during application, preventing a wider usage in biocatalysis. Here, we discovered a novel NADPH-dependent BVMO which originates from Halopolyspora algeriensis using sequence similarity networks (SSNs). The enzyme is stable at temperatures between 10 °C to 30 °C up to five days after the purification, and yields the normal ester product. In this study, the substrate scope was investigated for a broad range of aliphatic ketones and the enzyme was biochemically characterized to identify optimum reaction conditions. The best substrate (86 % conversion) was 2-dodecanone using purified enzyme. This novel BVMO could potentially be applied as part of an enzymatic cascade or in bioprocesses which utilize aliphatic alkanes as feedstock.
Combining solid acid catalysts with enzyme reactions in aqueous environments is challenging because either very acidic conditions inactivate the enzymes, or the solid acid catalyst is neutralized. In this study, Amberlyst-15 encapsulated in polydimethylsiloxane (Amb-15@PDMS) is used to deprotect the lignin depolymerization product G−C2 dioxolane phenol in a buffered system at pH 6.0. This reaction is directly coupled with the biocatalytic reduction of the released homovanillin to homovanillyl alcohol by recombinant horse liver alcohol dehydrogenase, which is subsequently acylated by the promiscuous acyltransferase/hydrolase PestE_I208A_L209F_N288A in a one-pot system. The deprotection catalyzed with Amb-15@PDMS attains up to 97 % conversion. Overall, this cascade enables conversions of up to 57 %.
o-Hydroxyarylphosphanes are fascinating compounds by their multiple-reactivity features, attributed to the ambident hard and soft Lewis- and also Brønstedt acid-base properties, wide tuning opportunities via backbone substituents with ±mesomeric and inductive, at P and in o-position to P and O also steric effects, and in addition, the configurational stability at three-valent phosphorus. Air sensitivity may be overcome by reversible protection with BH3, but the easy oxidation to P(V)-compounds may also be used. Since the first reports on the title compounds ca. 50 years ago the multiple reactivity has led to versatile applications. This includes various P−E−O and P=C−O heterocycles, a multitude of O-substituted derivatives including acyl derivatives for traceless Staudinger couplings of biomolecules with labels or functional substituents, phosphane-phosphite ligands, which like the o-phosphanylphenols itself form a range of transition metal complexes and catalysts. Also main group metal complexes and (bi)arylphosphonium-organocatalysts are derived. Within this review the various strategies for the access of the starting materials are illuminated, including few hints to selected applications.
In this article, we address the transition of the Kolbe electrolysis of valeric acid (VA) to n-octane as an exemplary electrosynthesis process from a batch reaction to a continuous, self-regulated process. Based on a systematic assessment of chemical boundary conditions and sustainability aspects, we propose a continuous electrosynthesis including a simple product separation and electrolyte recirculation, as well as an online-pH-controlled VA feeding. We demonstrate how essential performance parameters such as product selectivity (S) and coulombic efficiency (CE) are significantly improved by the transition from batch to a continuous process. Thus, the continuous and pH-controlled electrolysis of a 1 M valeric acid, starting pH 6.0, allowed a constantly high selectivity of around 47 % and an average Coulomb efficiency about 52 % throughout the entire experimental duration. Under otherwise identical conditions, the conventional batch operation suffered from lower and strongly decreasing performance values (Sn-octane, 60min=10.4 %, Sn-octane, 240min=1.3 %; CEn-octane, 60min=7.1 %, CEn-octane, 240min=0.5 %). At the same time, electrolyte recirculation significantly reduces wastes and limits the use of electrolyte components.
Enzymatic degradation and recycling can reduce the environmental impact of plastics. Despite decades of research, no enzymes for the efficient hydrolysis of polyurethanes have been reported. Whereas the hydrolysis of the ester bonds in polyester‐polyurethanes by cutinases is known, the urethane bonds in polyether‐polyurethanes have remained inaccessible to biocatalytic hydrolysis. Here we report the discovery of urethanases from a metagenome library constructed from soil that had been exposed to polyurethane waste for many years. We then demonstrate the use of a urethanase in a chemoenzymatic process for polyurethane foam recycling. The urethanase hydrolyses low molecular weight dicarbamates resulting from chemical glycolysis of polyether‐polyurethane foam, making this strategy broadly applicable to diverse polyether‐polyurethane wastes.
Poly(vinyl alcohol) (PVA) is a water‐soluble synthetic vinyl polymer with remarkable physical properties including thermostability and viscosity. Its biodegradability, however, is low even though a large amount of PVA is released into the environment. Established physical‐chemical degradation methods for PVA have several disadvantages such as high price, low efficiency, and secondary pollution. Biodegradation of PVA by microorganisms is slow and frequently involves pyrroloquinoline quinone (PQQ)‐dependent enzymes, making it expensive due to the costly cofactor and hence unattractive for industrial applications. In this study, we present a modified PVA film with improved properties as well as a PQQ‐independent novel enzymatic cascade for the degradation of modified and unmodified PVA. The cascade consists of four steps catalyzed by three enzymes with in situ cofactor recycling technology making this cascade suitable for industrial applications.
In general, concepts such as chemical equilibria and classical analytical chemistry, as well as instrumental methods such as potentiometry and flow injection analysis (FIA), are taught separately in the curricula of chemistry-related degree programs. However, FIA provides an opportunity to connect all of these important topics. This article shows how simple experiments can be used to link the fundamentals of FIA and potentiometry to the fundamentals of chemical equilibria and teach them in a sustainable way. The experiments were designed for undergraduate environmental science students with basic knowledge of chemical equilibria and classical analytical chemistry and minimal knowledge of electrochemistry. However, they can also be integrated into various courses on instrumental analysis in chemistry-related degree programs.
The vast majority of RNA splicing in today‘s organisms is achieved by the highly regulated and precise removal of introns from pre-mRNAs via the spliceosome. Here we present a model of how RNA splicing may have occurred in earlier life forms. We have designed a hairpin ribozyme derived spliceozyme that mediates two RNA cleavages and one ligation event at specific positions and thus cuts a segment (intron) out of a parent RNA and ligates the remaining fragments (exons). The cut-out intron then performs a downstream function, acting as a positive regulator of the activity of a bipartite DNAzyme. This simple scenario shows how small RNAs can perform complex RNA processing dynamics, involving the generation of new phenotypes by restructuring segments of given RNA species, as well as delivering small RNAs that may play a functional role in downstream processes.
Marine Bacteroidetes that degrade polysaccharides contribute to carbon cycling in the ocean. Organic matter, including glycans from terrestrial plants, might enter the oceans through rivers. Whether marine bacteria degrade structurally related glycans from diverse sources including terrestrial plants and marine algae was previously unknown. We show that the marine bacterium Flavimarina sp. Hel_I_48 encodes two polysaccharide utilization loci (PULs) which degrade xylans from terrestrial plants and marine algae. Biochemical experiments revealed activity and specificity of the encoded xylanases and associated enzymes of these PULs. Proteomics indicated that these genomic regions respond to glucuronoxylans and arabinoxylans. Substrate specificities of key enzymes suggest dedicated metabolic pathways for xylan utilization. Some of the xylanases were active on different xylans with the conserved β-1,4-linked xylose main chain. Enzyme activity was consistent with growth curves showing Flavimarina sp. Hel_I_48 uses structurally different xylans. The observed abundance of related xylan-degrading enzyme repertoires in genomes of other marine Bacteroidetes indicates similar activities are common in the ocean. The here presented data show that certain marine bacteria are genetically and biochemically variable enough to access parts of structurally diverse xylans from terrestrial plants as well as from marine algal sources.
This study investigated, if a mixed electroactive bacterial (EAB) culture cultivated heterotrophically at a positive applied potential could be adapted from oxidative to reductive or bidirectional extracellular electron transfer (EET). To this end, a periodic potential reversal regime between − 0.5 and 0.2 V vs. Ag/AgCl was applied. This yielded biofilm detachment and mediated electroautotrophic EET in combination with carbonate, i.e., dissolved CO2, as the sole carbon source, whereby the emerged mixed culture (S1) contained previously unknown EAB. Using acetate (S2) as well as a mixture of acetate and carbonate (S3) as the main carbon sources yielded primarily alternating electrogenic organoheterotropic metabolism with the higher maximum oxidation current densities recorded for mixed carbon media, exceeding on average 1 mA cm−2. More frequent periodic polarization reversal resulted in the increase of maximum oxidative current densities by about 50% for S2-BES and 80% for S3-BES, in comparison to half-batch polarization. The EAB mixed cultures developed accordingly, with S1 represented by mostly aerobes (84.8%) and being very different in composition to S2 and S3, dominated by anaerobes (96.9 and 96.5%, respectively). S2 and S3 biofilms remained attached to the electrodes. There was only minor evidence of fully reversible bidirectional EET. In conclusion the three triplicates fed with organic and/or inorganic carbon sources demonstrated two forms of diauxie: Firstly, S1-BES showed a preference for the electrode as the electron donor via mediated EET. Secondly, S2-BES and S3-BES showed a preference for acetate as electron donor and c-source, as long as this was available, switching to CO2 reduction, when acetate was depleted.