Pascal Falter-Braun

Plant-colonizing mutualistic symbionts confer beneficial effects to their hosts, which often includes increased growth as well as biotic and abiotic stress resilience. How these benefits are activated on a molecular level is mostly unknown. Here, we describe effector candidates of the fungal symbiont Serendipita indica (Si), which modulate plant stress signaling pathways. By analyzing the Si effector interactome, we reveal frequent targeting of stress-related host proteins, which are linked to the identified effector signaling functions. Moreover, functional data indicate that Si effectors modulate abiotic stress response of Arabidopsis, as well as resistance to pathogen infection. Analysis of symbiont effectors might not only uncover previously unreported molecular mechanisms that increase plant fitness but might also be used to identify potential genetic traits for crop improvement under changing climates.

The transport protein particle II (TRAPPII) complex is a conserved regulator of post-Golgi membrane trafficking. In Arabidopsis, phosphorylation of the TRAPPII-specific subunit TRS120 by SHAGGY-like kinases modulates adaptive growth responses, but the phosphatases that reverse this phosphorylation remain unknown. Here, proteomic analyses identified subunits of Protein Phosphatase 2A (PP2A) in the TRAPPII interactome. PP2A subunits physically associated with TRAPPII, and double-mutant analyses revealed genetic interactions between PP2A and TRAPPII. Loss of TRAPPII function reduced the relative membrane association of PP2A scaffolding subunits. We established an in vitro assay for Arabidopsis PP2A holoenzyme activity using complexes transiently co-expressed and affinity-purified from Nicotiana benthamiana . Structural modelling and interface analysis predicted binding of a phosphorylated peptide encompassing a TRS120 phosphosite cluster at the PP2A catalytic interface, while biochemical assays showed that a PP2A holoenzyme containing the B2 regulatory subunit dephosphorylated this peptide. Together, these findings identify a B2-containing PP2A holoenzyme as a candidate phosphatase for TRS120 and support a model in which antagonistic SHAGGY-like kinase and PP2A activities couple signalling to membrane trafficking during plant development and environmental adaptation.

Plant colonizing mutualistic symbionts confer beneficial effects to their hosts, which often includes increased growth, and biotic and abiotic stress resilience. How these benefits are activated on a molecular level is mostly unknown. Here, we describe effector candidates of the fungal symbiont Serendipita indica ( Si ), which modulate plant stress signalling pathways. By analyzing the Si effector interactome, we reveal frequent targeting of stress related host proteins, which are linked to the identified effector signalling functions. Moreover, functional data indicate that Si effectors modulate abiotic stress response of Arabidopsis, as well as resistance to pathogen infection. Analysis of symbiont effectors might not only uncover previously unreported molecular mechanisms that increase plant fitness but might also be used to identify potential genetic traits for crop improvement under changing climates.

Genotype-phenotype relationships are mediated through intricate networks of physical and functional interactions among macromolecules. Knowledge of the interactome is vital to understand and model genetics and cellular biology. Recent advances in accurately predicting tertiary protein structures using artificial intelligence (AI) approaches such as AlphaFold ¹ have revived the vision that the protein-protein interactome might be fully predictable through computational modeling of quaternary structures. Here we present a comprehensive experimental framework to systematically assess the impact of AI-driven interactome predictions for yeast ² and human ³ . We find that the quality of high-confidence predictions is on par with established experimental approaches. However, in proteome-wide screening, the tested AI approaches underperform in the discovery of strictly novel protein-protein interactions (PPIs) compared to experimental reference interactome maps. In particular, the yeast interactome map described here identifies >40-fold more novel PPIs than its AI counterpart. Strikingly, AlphaFold provides structural models for a substantial number of experimentally identified PPIs missed by the virtual screens. Our results suggest that, at this stage, the main contribution of AI predictions is to provide quaternary structure models for experimentally identified PPIs.

Environmental factors are often detrimental; however, certain environments enhance immune resilience. Notably, children raised on traditional farms show reduced allergies and asthma prevalence. Here, we investigated how a beneficial environment, using farm dust (FD) extract, influenced lung immune function in ovalbumin-induced allergic inflammation. FD exposure reduced allergic lung inflammation and increased monocyte-derived macrophage (MDM) recruitment. Single-cell RNA sequencing revealed that FD-exposed MDMs had altered gene expression, including dampened Ccl8 and major histocompatibility complex class II expression, impairing eosinophil recruitment and antigen presentation. RNA sequencing and assay for transposase-accessible chromatin using sequencing confirmed FD-induced epigenetic reprogramming ex vivo, on bone marrow–derived macrophages. This modulation, seen in both human and murine cells, relied on histone deacetylase activity sustained by peroxisome proliferator–activated receptor γ signaling. These findings suggest that beneficial environmental exposures can reprogram immune cells and may offer a previously unidentified strategy for asthma prevention.

Pseudomonadota (formerly Proteobacteria) are prevalent in the commensal human gut microbiota, but also include many pathogens that rely on secretion systems to support pathogenicity by injecting proteins into host cells. Here we show that 80% of Pseudomonadota from healthy gut microbiomes also have intact type III secretion systems (T3SS). Candidate effectors predicted by machine learning display sequence and structural features that are distinct from those of pathogen effectors. Towards a systems-level functional understanding, we experimentally constructed a protein–protein meta-interactome map between human proteins and commensal effectors. Network analyses uncovered that effector-targeted neighbourhoods are enriched for genetic variation linked to microbiome-associated conditions, including autoimmune and metabolic diseases. Metagenomic analysis revealed effector enrichment in Crohn’s disease but depletion in ulcerative colitis. Functionally, commensal effectors can translocate into human cells and modulate NF-κB signalling and cytokine secretion in vitro. Our findings indicate that T3SS contribute to microorganism–host cohabitation and that effector–host protein interactions may represent an underappreciated route by which commensal gut microbiota influences health.

Accurate reconstruction of gene regulatory networks (GRNs) from single-cell transcriptomic data remains a major methodological challenge. Recent machine learning approaches, particularly graph neural networks and graph autoencoders, have reported improved performance, yet these gains do not consistently translate to realistic biological settings. Here, we show that a key reason for that is the way negative regulatory interactions are sampled for supervised training and evaluation. We find that widely used sampling strategies introduce node-degree biases that allow models to exploit trivial graph-structural cues rather than biological signals. Across multiple benchmarks, simple degree-based heuristics match or exceed state-of-the-art graph neural network models under these biased evaluation protocols. We further introduce a degree-aware sampling approach that eliminates these artifacts and provides more reliable assessments of GRN inference methods. Our results call for standardized, bias-aware benchmarking practices to ensure meaningful progress in supervised GRN inference from single-cell RNA-seq data.

SUMMARY: Pathogens deploy effector proteins to exploit host cell biology, and most pathogen open reading frames (ORFs) are rapidly evolving and lack functional annotation. We developed the eORFeome, a scalable functional genomics platform encompassing 3,835 effector ORFs from diverse viruses, bacteria, and parasites. High-throughput barcoded screens across NFκB, apoptosis, p53, cGAS–STING and MHC-I pathways revealed functions for hundreds of uncharacterized eORFs, unexpected new activities for known effectors, and distinct pathway-specific functions encoded by single ORFs. Illustrating the power of the approach, we identify HHV6A U14 as a p53 antagonist, HHV7 U21 as a dual-function STING antagonist and MHC-I antigen display inhibitor, and adenoviral 13.6K/i-leader protein as a de novo evolved TAP inhibitor that suppresses MHC-I display. These results establish a general framework for systematic effector annotation, uncover new mechanisms of host–pathogen interaction across kingdoms, and highlight pathogen effectors as a versatile toolkit for rewiring and probing human cellular pathways.

Climate change is projected to intensify weather extremes like droughts that threaten global crop yields. While many crops are sensitive to water deprivation, they have drought-tolerant wild relatives; however, the molecular basis for this differential tolerance among closely related species remains unclear. We previously investigated this using transcriptomics in the Brassicaceae family, comparing the drought-sensitive reference plant Arabidopsis thaliana (Ath) with its drought-tolerant relatives Arabidopsis lyrata (Aly) and Eutrema salsugineum (Esa) , uncovering key differences that may underlie their respective phenotypes. Here, to elucidate the regulatory architecture driving these differences, we mapped drought-responsive gene regulatory networks (GRNs) for all three species using the yeast one-hybrid (Y1H) approach. We cloned promoters of orthologs with divergent drought-induced expression patterns and screened them against the Ath transcription factor (TF) collection, generating a quality-controlled cross-species drought-responsive GRN map encompassing 1,137 high-confidence TF-promoter interactions. Comparative analysis revealed that the tolerant species share greater regulatory similarity and exhibit higher network connectivity than Ath . An Esa -specific expansion of bZIP TF interactions was observed, consistent with an enrichment of G-box motifs in its genome and in drought-upregulated genes. Several observed changes affect the ABA signalling pathway: Ath gained a susceptibility linked GBF3- CYP707A1 edge, whereas Esa acquired tolerance associated ABF- mTERF10 interactions. Furthermore, network-rewiring analysis uncovered a novel role for ASIL2 in stress response. Finally, our network highlights TFs (CSDP1, ERF4, HB6, and MYB73) that likely contribute to the “stress-primed” state of Aly and Esa . Our work provides a broadly applicable framework for comparative GRN mapping and a valuable resource for improving drought tolerance in crops.