Peggy Stolt-Bergner

Mutations in the Kirsten rat sarcoma viral oncogene homolog (KRAS) protein are highly prevalent in cancer. However, small-molecule concepts that address oncogenic KRAS alleles remain elusive beyond replacing glycine at position 12 with cysteine (G12C), which is clinically drugged through covalent inhibitors. Guided by biophysical and structural studies of ternary complexes, we designed a heterobifunctional small molecule that potently degrades 13 out of 17 of the most prevalent oncogenic KRAS alleles. Compared with inhibition, KRAS degradation results in more profound and sustained pathway modulation across a broad range of KRAS mutant cell lines, killing cancer cells while sparing models without genetic KRAS aberrations. Pharmacological degradation of oncogenic KRAS was tolerated and led to tumor regression in vivo. Together, these findings unveil a new path toward addressing KRAS-driven cancers with small-molecule degraders.

Background: Recombinant protein production and purification of large protein complexes in eukaryotes requires efficient methods to generate multi-gene expression constructs, where each individual gene is under the control of its own promoter and terminator. Current methods are based either on serial rounds of combination of several vectors containing loxP sites via the Cre-lox technology, or on multiple rounds of gene combination via PCR or other methods. These methods are multi-step, have lower efficiencies than single gene cloning, and may require laborious processes to verify that all genes of interest are present in the final product. Here, we describe a rapid and simple Golden Gate-based system for the generation of multi-gene expression constructs compatible with baculovirus expression vector systems (BEVS) using either Tn7 transposition or KO1629-based homologous recombination, which we refer to as “GoldenBac”. Results: This method is based on the construction of a series of vectors containing a promoter-gene of interest-terminator cassette flanked by cleavage sites of the BsaI type IIS restriction enzyme. This series of vectors can be cut by BsaI to excise cassettes with unique overhangs. In the same reaction, the cassettes are then ligated in the correct sequence in a final destination vector to generate multi-gene expression constructs containing 2–15 genes. Individual expression constructs can therefore be combined into a single vector in a single reaction, with over 90% efficiency when combining up to 14 expression cassettes. We demonstrate successful construction and expression of three different co-expression systems, the proteosomal lid complex, the anaphase promoting complex/cyclosome (APC/C), and a series of constructs used to test the effect of chaperone co-expression on the solubility of the HOIP protein. Conclusions: This robust, single-step cloning system provides an easy-to-use method for generation of multi-gene expression constructs for both transposition and homologous recombination-based baculovirus systems, making this technology available across all laboratories using baculovirus expression systems. This highly efficient and simple method allows for rapid incorporation of multi-gene expression cloning into the standardized service portfolio of protein production facilities and can also easily be adopted by any laboratory for routine generation of multi-gene baculovirus constructs.

Plants use surface receptors to perceive information about many aspects of their local environment. These receptors physically interact to form both steady state and signalling competent complexes. The signalling events downstream of receptor activation impact both plant developmental and immune responses. Here, we present a comprehensive study of the physical interactions between the extracellular domains of leucine-rich repeat receptor kinases (LRR-RKs) in Arabidopsis. Using a sensitized assay, we tested reciprocal interactions among 200 of the 225 Arabidopsis LRR-RKs for a total search space of 40,000 interactions. Applying a stringent statistical cut-off and requiring that interactions performed well in both bait-prey and prey-bait orientations resulted in a high-confidence set of 567 bidirectional interactions. Additionally, we identified a total of 2,586 unidirectional interactions, which passed our stringent statistical cut-off in only one orientation. These datasets will guide further investigation into the regulatory roles of LRR-RKs in plant developmental and immune signalling decisions.

Resolving functions of closely linked genes is challenging or nearly impossible with classical genetic tools. Four members of the Catharanthus roseus receptor-like kinase 1-like (CrRLK1L) family are clustered on Arabidopsis chromosome five. To resolve the potentially redundant functions of this subclass of CrRLK1Ls named MEDOS1 to 4 (MDS1 to 4), we generated a single CRISPR/Cas9 transformation vector using a Golden Gate based cloning system to target all four genes simultaneously. We introduce single mutations within and deletions between MDS genes as well as knock-outs of the whole 11 kb gene cluster. The large MDS cluster deletion was inherited in up to 25% of plants lacking the CRISPR/Cas9 construct in the T2 generation. In contrast to described phenotypes of already characterized CrRLK1L mutants, quadruple mds knock-outs were fully fertile, developed normal root hairs and trichomes and responded to pharmacological inhibition of cellulose biosynthesis similar to wildtype. Recently, we demonstrated the role of four CrRLK1L in growth adaptation to metal ion stress. Here we show the involvement of MDS genes in response to Ni²⁺ during hypocotyl elongation and to Cd²⁺ and Zn²⁺ during root growth. Our finding supports the model of an organ specific network of positively and negatively acting CrRLK1Ls.

In this Letter, an incorrect version of the Supplementary Information file was inadvertently used, which contained several errors. The details of references 59–65 were missing from the end of the Supplementary Discussion section on page 4. In addition, the section ‘Text 3. Y2H on ICD interactions’ incorrectly referred to ‘Extended Data Fig. 4d’ instead of ‘Extended Data Fig. 3d’ on page 3. Finally, the section ‘Text 4. Interaction network analysis’ incorrectly referred to ‘Fig. 1b and Extended Data Fig. 6’ instead of ‘Fig. 2b and Extended Data Fig. 7’ on page 3. These errors have all been corrected in the Supplementary Information.

A high-throughput assay is used to analyse 40,000 potential extracellular domain interactions of a large family of plant cell surface receptors (LRR-RKs) and provide a cell surface interaction network for these receptors. Cell surface receptors mediate communication between the interior of a cell and its external environment. Specifically, the extracellular domains (ECDs) of such receptors interact with external molecules. It is less clear how interactions between ECDs of different receptors help to form receptor complexes for signal transduction. Youssef Belkhadir and colleagues investigate systems-level organization of leucine-rich repeat receptor kinases (LRR-RKs)—a large family of plant cell surface receptors with roles in processes including plant defence and development. The authors use a high-throughput assay to study 40,000 potential ECD interactions. They develop a cell surface interaction network for these receptors and study its dynamics. The team demonstrate the power of this network for detecting biologically relevant interactions by predicting and validating the function of previously uncharacterized LRR-RKs in plant growth and immunity. The cells of multicellular organisms receive extracellular signals using surface receptors. The extracellular domains (ECDs) of cell surface receptors function as interaction platforms, and as regulatory modules of receptor activation1,2. Understanding how interactions between ECDs produce signal-competent receptor complexes is challenging because of their low biochemical tractability3,4. In plants, the discovery of ECD interactions is complicated by the massive expansion of receptor families, which creates tremendous potential for changeover in receptor interactions5. The largest of these families in Arabidopsis thaliana consists of 225 evolutionarily related leucine-rich repeat receptor kinases (LRR-RKs)5, which function in the sensing of microorganisms, cell expansion, stomata development and stem-cell maintenance6,7,8,9. Although the principles that govern LRR-RK signalling activation are emerging1,10, the systems-level organization of this family of proteins is unknown. Here, to address this, we investigated 40,000 potential ECD interactions using a sensitized high-throughput interaction assay3, and produced an LRR-based cell surface interaction network (CSILRR) that consists of 567 interactions. To demonstrate the power of CSILRR for detecting biologically relevant interactions, we predicted and validated the functions of uncharacterized LRR-RKs in plant growth and immunity. In addition, we show that CSILRR operates as a unified regulatory network in which the LRR-RKs most crucial for its overall structure are required to prevent the aberrant signalling of receptors that are several network-steps away. Thus, plants have evolved LRR-RK networks to process extracellular signals into carefully balanced responses.

Chromosome ends are sequestered from aberrant DNA repair activities within the DNA binding channel of the Ku heterodimer.

A method for producing multiprotein complexes engineered with site–specifically introduced noncanonical amino acids is described, enabling applications in biochemical and biophysical analysis, as well as in biotechnology. We present a baculovirus-based protein engineering method that enables site-specific introduction of unique functionalities in a eukaryotic protein complex recombinantly produced in insect cells. We demonstrate the versatility of this efficient and robust protein production platform, 'MultiBacTAG', (i) for the fluorescent labeling of target proteins and biologics using click chemistries, (ii) for glycoengineering of antibodies, and (iii) for structure–function studies of novel eukaryotic complexes using single-molecule Förster resonance energy transfer as well as site-specific crosslinking strategies.

Expression in Escherichia coli represents the simplest and most cost effective means for the production of recombinant proteins. This is a routine task in structural biology and biochemistry where milligrams of the target protein are required in high purity and monodispersity. To achieve these criteria, the user often needs to screen several constructs in different expression and purification conditions in parallel. We describe a pipeline, implemented in the Center for Optimized Structural Studies, that enables the systematic screening of expression and purification conditions for recombinant proteins and relies on a series of logical decisions. We first use bioinformatics tools to design a series of protein fragments, which we clone in parallel, and subsequently screen in small scale for optimal expression and purification conditions. Based on a scoring system that assesses soluble expression, we then select the top ranking targets for large-scale purification. In the establishment of our pipeline, emphasis was put on streamlining the processes such that it can be easily but not necessarily automatized. In a typical run of about 2 weeks, we are able to prepare and perform small-scale expression screens for 20–100 different constructs followed by large-scale purification of at least 4–6 proteins. The major advantage of our approach is its flexibility, which allows for easy adoption, either partially or entirely, by any average hypothesis driven laboratory in a manual or robot-assisted manner.