The genetic architecture of protein stability.

There are more ways to synthesize a 100-amino acid (aa) protein (20100) than there are atoms in the universe. Only a very small fraction of such a vast sequence space can ever be experimentally or computationally surveyed. Deep neural networks are increasingly being used to navigate high-dimensional sequence spaces1. However, these models are extremely complicated. Here, by experimentally sampling from sequence spaces larger than 1010, we show that the genetic architecture of at least some proteins is remarkably simple, allowing accurate genetic prediction in high-dimensional sequence spaces with fully interpretable energy models. These models capture the nonlinear relationships between free energies and phenotypes but otherwise consist of additive free energy changes with a small contribution from pairwise energetic couplings. These energetic couplings are sparse and associated with structural contacts and backbone proximity. Our results indicate that protein genetics is actually both rather simple and intelligible.

Mapping the energetic and allosteric landscapes of protein binding domains.

Allosteric communication between distant sites in proteins is central to biological regulation but still poorly characterized, limiting understanding, engineering and drug development1-6. An important reason for this is the lack of methods to comprehensively quantify allostery in diverse proteins. Here we address this shortcoming and present a method that uses deep mutational scanning to globally map allostery. The approach uses an efficient experimental design to infer en masse the causal biophysical effects of mutations by quantifying multiple molecular phenotypes-here we examine binding and protein abundance-in multiple genetic backgrounds and fitting thermodynamic models using neural networks. We apply the approach to two of the most common protein interaction domains found in humans, an SH3 domain and a PDZ domain, to produce comprehensive atlases of allosteric communication. Allosteric mutations are abundant, with a large mutational target space of network-altering 'edgetic' variants. Mutations are more likely to be allosteric closer to binding interfaces, at glycine residues and at specific residues connecting to an opposite surface within the PDZ domain. This general approach of quantifying mutational effects for multiple molecular phenotypes and in multiple genetic backgrounds should enable the energetic and allosteric landscapes of many proteins to be rapidly and comprehensively mapped.

Maternal age generates phenotypic variation in Caenorhabditis elegans.

Genetically identical individuals that grow in the same environment often show substantial phenotypic variation within populations of organisms as diverse as bacteria, nematodes, rodents and humans. With some exceptions, the causes are poorly understood. Here we show that isogenic Caenorhabditis elegans nematodes vary in their size at hatching, speed of development, growth rate, starvation resistance, fecundity, and also in the rate of development of their germline relative to that of somatic tissues. We show that the primary cause of this variation is the age of an individual's mother, with the progeny of young mothers exhibiting several phenotypic impairments. We identify age-dependent changes in the maternal provisioning of the lipoprotein complex vitellogenin to embryos as the molecular mechanism that underlies the variation in multiple traits throughout the life of an animal. The production of sub-optimal progeny by young mothers may reflect a trade-off between the competing fitness traits of a short generation time and the survival and fecundity of the progeny.

Fibroblast-led collective invasion of carcinoma cells with differing roles for RhoGTPases in leading and following cells.

Imaging of collectively invading cocultures of carcinoma cells and stromal fibroblasts reveals that the leading cell is always a fibroblast and that carcinoma cells move within tracks in the extracellular matrix behind the fibroblast. The generation of these tracks by fibroblasts is sufficient to enable the collective invasion of the squamous cell carcinoma (SCC) cells and requires both protease- and force-mediated matrix remodelling. Force-mediated matrix remodelling depends on integrins alpha3 and alpha5, and Rho-mediated regulation of myosin light chain (MLC) activity in fibroblasts, but these factors are not required in carcinoma cells. Instead, carcinoma cells use Cdc42 and MRCK (myotonic dystrophy kinase-related CDC42-binding protein kinases) mediated regulation of MLC to follow the tracks generated by fibroblasts.

Transgenerational transmission of environmental information in C. elegans.

The environment experienced by an animal can sometimes influence gene expression for one or a few subsequent generations. Here, we report the observation that a temperature-induced change in expression from a Caenorhabditis elegans heterochromatic gene array can endure for at least 14 generations. Inheritance is primarily in cis with the locus, occurs through both oocytes and sperm, and is associated with altered trimethylation of histone H3 lysine 9 (H3K9me3) before the onset of zygotic transcription. Expression profiling reveals that temperature-induced expression from endogenous repressed repeats can also be inherited for multiple generations. Long-lasting epigenetic memory of environmental change is therefore possible in this animal.

Impaired DNA replication derepresses chromatin and generates a transgenerationally inherited epigenetic memory.

Impaired DNA replication is a hallmark of cancer and a cause of genomic instability. We report that, in addition to causing genetic change, impaired DNA replication during embryonic development can have major epigenetic consequences for a genome. In a genome-wide screen, we identified impaired DNA replication as a cause of increased expression from a repressed transgene in Caenorhabditis elegans. The acquired expression state behaved as an "epiallele," being inherited for multiple generations before fully resetting. Derepression was not restricted to the transgene but was caused by a global reduction in heterochromatin-associated histone modifications due to the impaired retention of modified histones on DNA during replication in the early embryo. Impaired DNA replication during development can therefore globally derepress chromatin, creating new intergenerationally inherited epigenetic expression states.

Oncogenic BRAF induces melanoma cell invasion by downregulating the cGMP-specific phosphodiesterase PDE5A.

We show that in melanoma cells oncogenic BRAF, acting through MEK and the transcription factor BRN2, downregulates the cGMP-specific phosphodiesterase PDE5A. Although PDE5A downregulation causes a small decrease in proliferation, its major impact is to stimulate a dramatic increase in melanoma cell invasion. This is because PDE5A downregulation leads to an increase in cGMP, which induces an increase in cytosolic Ca(2+), stimulating increased contractility and inducing invasion. PDE5A downregulation also this leads to an increase in short-term and long-term colonization of the lungs by melanoma cells. We do not observe this pathway in NRAS mutant melanoma or BRAF mutant colorectal cells. Thus, we show that in melanoma cells oncogenic BRAF induces invasion through downregulation of PDE5A.

Collective cell migration requires suppression of actomyosin at cell-cell contacts mediated by DDR1 and the cell polarity regulators Par3 and Par6.

Collective cell migration occurs in a range of contexts: cancer cells frequently invade in cohorts while retaining cell-cell junctions. Here we show that collective invasion by cancer cells depends on decreasing actomyosin contractility at sites of cell-cell contact. When actomyosin is not downregulated at cell-cell contacts, migrating cells lose cohesion. We provide a molecular mechanism for this downregulation. Depletion of discoidin domain receptor 1 (DDR1) blocks collective cancer-cell invasion in a range of two-dimensional, three-dimensional and 'organotypic' models. DDR1 coordinates the Par3/Par6 cell-polarity complex through its carboxy terminus, binding PDZ domains in Par3 and Par6. The DDR1-Par3/Par6 complex controls the localization of RhoE to cell-cell contacts, where it antagonizes ROCK-driven actomyosin contractility. Depletion of DDR1, Par3, Par6 or RhoE leads to increased actomyosin contactility at cell-cell contacts, a loss of cell-cell cohesion and defective collective cell invasion.

The Golgi mitotic checkpoint is controlled by BARS-dependent fission of the Golgi ribbon into separate stacks in G2.

The Golgi ribbon is a complex structure of many stacks interconnected by tubules that undergo fragmentation during mitosis through a multistage process that allows correct Golgi inheritance. The fissioning protein CtBP1-S/BARS (BARS) is essential for this, and is itself required for mitotic entry: a block in Golgi fragmentation results in cell-cycle arrest in G2, defining the 'Golgi mitotic checkpoint'. Here, we clarify the precise stage of Golgi fragmentation required for mitotic entry and the role of BARS in this process. Thus, during G2, the Golgi ribbon is converted into isolated stacks by fission of interstack connecting tubules. This requires BARS and is sufficient for G2/M transition. Cells without a Golgi ribbon are independent of BARS for Golgi fragmentation and mitotic entrance. Remarkably, fibroblasts from BARS-knockout embryos have their Golgi complex divided into isolated stacks at all cell-cycle stages, bypassing the need for BARS for Golgi fragmentation. This identifies the precise stage of Golgi fragmentation and the role of BARS in the Golgi mitotic checkpoint, setting the stage for molecular analysis of this process.

Mitotic Golgi partitioning is driven by the membrane-fissioning protein CtBP3/BARS.

Organelle inheritance is an essential feature of all eukaryotic cells. As with other organelles, the Golgi complex partitions between daughter cells through the fission of its membranes into numerous tubulovesicular fragments. We found that the protein CtBP3/BARS (BARS) was responsible for driving the fission of Golgi membranes during mitosis in vivo. Moreover, by in vitro analysis, we identified two stages of this Golgi fragmentation process: disassembly of the Golgi stacks into a tubular network, and BARS-dependent fission of these tubules. Finally, this BARS-induced fission of Golgi membranes controlled the G2-to-prophase transition of the cell cycle, and hence cell division.