William Shih, Harvard University, USA

DNA origami, in which a long scaffold strand is assembled with a large number of short staple strands into parallel arrays of double helices, has proven a powerful method for custom nanofabrication of shapes up to 100 nm in size. The scaffold represents about half the mass of an origami, therefore the origami size is restricted by the length of the scaffold. However, it is impractical and prohibitively expensive to scale the length of the scaffold. Here I will discuss a strategy, that we call crisscross polymerization, that combines all-or-nothing scaffold-dependent initiation of folding with scaffold-independent growth, therefore allowing for sizes unbounded by the length of the scaffold. Assembly using single-stranded DNA slats enables digital counting of molecular analytes, where each molecular detection event triggers growth of a single filament resolvable by low-cost microscopy. Assembly using DNA-origami slats, conversely, enables fabrication of fully addressable structures that are twice the mass of the E. coli genome, and that span an area of two microns by two microns.

Claudia Höbartner, University of Würzburg, DE

Synthetic functional nucleic acids such as fluorogen-activating RNA aptamers and RNA-alkylating ribozymes have emerged as enabling tools for tagging and visualizing RNA in vitro and in cells. Novel activities of ribozymes and aptamers are generated by iterative cycles of selection and amplification from an initially random library. This presentation will focus on ribozymes that catalyze site-specific RNA modification using small molecule cofactors.

Natural RNA modifications such as site-specifically methylated nucleotides are conserved throughout evolution and expand the structural and functional diversity of RNA. Synthetic RNA modifications enable RNA labelling and visualization to study RNA localization, folding and structural dynamics. We discovered a methyltransferase ribozyme (MTR1) that catalyzes the installation of 1-methyl-adenosine (m¹A) in a target RNA. The ribozyme shows a broad RNA sequence scope, as exemplified by site-specific adenosine methylation in native tRNAs and synthetic mRNAs. The crystal structure of MTR1 revealed an active site reminiscent of natural guanine riboswitches and suggested the mechanistic involvement of a protonated cytidine. Further research revealed alkyltransferase ribozymes using established benzylguanine substrates for site-specific fluorescent labelling of RNA. Recently we found an alkyltransferase ribozyme that uses a synthetic, stabilized S-adenosyl-methionine (SAM) analogue and catalyses the transfer of a propargyl group to a specific adenosine in the target RNA. A genetically encoded version of the SAM analogue-utilizing ribozyme (SAMURI) was expressed in eukaroytic cells, and intracellular propargylation of the target adenosine was confirmed. The transferred alkyne group was efficiently conjugated to different azides, including fluorophores and biotin, to enable studies of RNA localization, folding and structural dynamics. Such engineered ribozymes provide attractive means for tracking RNA localization, folding and structural dynamics, and may be used as protein-free tools to modulate the cellular RNA modification landscape.

Luisa de Cola, University of Milan, IT

Advancements in the use of nanoparticles for biomedical applications have clearly shown their potential for the preparation of improved imaging and drug-delivery systems. However, only a few successfully materials translate into clinical practice, because, of their incomplete elimination, difficulties to cross barriers and lack of slectivity. We have recently reported disulfide-bridged organosilica nanoparticles with cage-like morphology, and assessed in detail their bioaccumulation in vivo. The fate of intravenously injected 20 nm nanocages was investigated in both healthy and tumor bearing mice. Interestingly, the nanoparticles exclusively co-localize with hepatic sinusoidal endothelial cells (LSECs), while avoiding Kupffer-cell uptake in both physiological and pathological condition. Our findings suggest that organosilica nanocages hold the potential to be used as nanotools for drug delivery and for crossing important body barriers.

But how to target specific organs or cancer tissues? To improve selective uptake we have recently developed a technology based on the use of extracellular vesicles, EVs. Through a strategy we are able to separate the membrane of these vesicles and reconstruct it on top of our nanomaterials. The membrane contains all the original targeting proteins and receptors and depending from which tumor is isolated able to target specific organs or metastasis. We show that not only in vitro we have an excellent selectivity, but also in vivo we are able to demonstrate an excellent selectivity towards cancer cells vs normal cells of the same tissue. The use of melanoma EVs showed a tropism, of the hybrid materials, towards lungs and quantitative analysis on mice models suggested that the targeting behavior of the EVs can be indeed used as a strategy for the targeting of lungs and reduces dramatically the accumulation in liver.

Finally silica nanoparticles containing single-stranded nucleic acids, that are covalently embedded in the silica network, have been reported. The system can be programmed to be more dynamic and responsive by designing supramolecular organo-silica systems based on PNA- derivatives that can self-assemble through direct base paring or can be joined through a bridging functional nucleic acid, such as the ATP-binding aptamer. These systems can be followed by confocal microscopy in different cell lines and their biological effect was measured in cells to assess the biological effect of the aptamer.

Giacomo Renno, University of Geneva, CH

Pnictogens represent the last frontier of the σ-hole bonds, non-covalent interactions highly attractive for the development of powerful tools in supramolecular chemistry. In this study, artificial enzymes capitalizing on pnictogen bonding are reported as a new tool, absent in biocatalysis. To tackle this challenge, stibine catalysts were decorated with a biotin moiety and combined with streptavidin mutants. The transfer hydrogenation of fluorogenic quinolines mediated by a hydrophilic Hantzsch ester was used as benchmark reaction. σ-Hole deepening resulted into an improvement of the catalytic performances, best for mutants which position negative charges (D, E) or hydrophobic residues (I) in the active pocket. Michaelis-Menten analysis showed transition-state recognition in the low micromolar range. Lastly, an emerging stereoselectivity further corroborated the promising potentiality of pnictogen-bonding catalysts in such ordered systems.

Mathieu Baltussen, Radboud University Nijmegen, NL

Chemical reaction networks, such as those found in metabolism and signalling pathways, enable cells to process information from their environment. Current approaches to molecular information processing and computation typically pursue digital computation models and require extensive molecular-level engineering. These approaches demonstrate how molecular systems may perform computation, but do not achieve the information processing capabilities of living systems. Unlocking the full potential of molecular systems requires (1) moving beyond a strict adherence to reproducing digital computation principles and (2) finding an approach that overcomes the laborious nature of bottom-up ‘molecule-by-molecule’ design patterns.

We recently reported on the discovery of emergent information processing capabilities and implementation of chemical reservoir computation in the formose reaction. This prebiotic, complex, self-organizing reaction network produces a rich diversity of possible chemical compositions that are non-linearly dependent on a small number of input chemicals. Under flow conditions, the distribution of chemical compositions can be modulated using changes in reactor input concentrations, allowing a range of complex reaction responses to be controlled with a relatively simple set of input parameters. These properties, and the experimental tractability of the formose reaction, make it an excellent candidate system for exploring chemical information processing using the model of physical reservoir computation. We demonstrate how the formose reaction can be used as a reservoir computer and investigate its capabilities for several computational tasks. We first show it can emulate all Boolean logic gates, and several, more advanced nonlinear classification tasks in parallel, achieving performance similar to machine learning algorithms. We next demonstrate its suitability for more complex computational tasks, by using the formose reservoir to predict the dynamics of an E. coli carbon-metabolism pathway in a fluctuating environment. Finally, we achieve time-series forecasting of chaotic changing environments, and evaluate the memory properties of the reservoir by quantifying the propagation of information through the formose network over time.

Our work shows how chemical reaction networks process information on the basis of self-organization, and, much like biological systems, can achieve a variety of powerful computational tasks using information from their environment, obviating the need for complex bottom-up design and creating new opportunities for scalable molecular computing. This in-chemico information processing system provides proof of principle for the emergent computational capabilities of complex chemical reaction networks, paving the way for a new class of biomimetic information processing systems.

Sijbren Otto, University of Groningen, NL

How the immense complexity of living organisms has arisen is one of the most intriguing questions in contemporary science. We have started to explore experimentally how organization and function can emerge from complex molecular networks in aqueous solution. We focus on networks of molecules that can interconvert, to give mixtures that can change their composition in response to external or internal stimuli. Noncovalent interactions within molecules in such mixtures can lead to the spontaneous formation of foldamers of remarkable structural complexity. In contrast, molecular recognition between molecules in such mixtures leads to their mutual stabilization, which drives the synthesis of more of the privileged structures. As the assembly process drives the synthesis of the very molecules that assemble, the resulting materials can be considered to be self-synthesizing. In this process the assembling molecules are replicating themselves, where replication is driven by self-recognition of these molecules in the dynamic network. We have witnessed spontaneous differentiation (a process akin to speciation as it occurs in biology) in a system made from a mixture of two building blocks. When such systems are operated under out-of-equilibrium conditions, replicators can complexify.

Replicators that are able to catalyse reactions other than their own formation have also been obtained, representing a first step towards metabolism. Rudimentary Darwinian evolution of purely synthetic molecules has also been achieved and the prospect of synthesizing life de-novo is becoming increasingly realistic.

Joachim Spatz, MPI Medical Research, DE

The evolution of cellular compartments for spatially and temporally controlled assembly of biological processes was an essential step in developing life by evolution. Synthetic approaches to cellular-like compartments are still lacking well-controlled functionalities, as would be needed for more complex synthetic cells. With the ultimate aim to construct life-like materials such as a living cell, matter-to-life strives to reconstitute cellular phenomena in vitro – disentangled from the complex environment of a cell. In recent years, working towards this ambitious goal gave new insights into the mechanisms governing life. With the fast-growing library of functional modules assembled for synthetic cells, their classification and integration become increasingly important. We will discuss strategies to reverse-engineer and recombine functional parts for synthetic eukaryotes, mimicking the characteristics of nature’s own prototype. Particularly, we will focus on large outer compartments, complex endomembrane systems with organelles and versatile cytoskeletons as hallmarks of eukaryotic life. Moreover, we identify microfluidics and DNA nanotechnology as two highly promising technologies which can achieve the integration of these functional modules into sophisticated multifunctional synthetic cells.

Edoardo Sisti, University of Pisa, IT

NanoAmp is an innovative PCR-based technology designed to detect antigens and antibodies through an easy and quick single-step method. Developed by UlisseBiomed S.p.A., NanoAmp leverages the target-induced increase in local concentration, mediated by interactions between nucleic acid-based elements. It combines both isothermal polymerase reactions and hot-start PCR amplification, which can be performed using a single enzyme in its most advanced form. This approach enables rapid, sensitive, and quantitative biomolecule detection in a single well within 30 minutes. Compared to traditional biomolecule detection methods such as ELISA and CLIA, which are used for high-throughput formats and involve complex, time-consuming, multi-step processes, NanoAmp offers an easier and more flexible solution. Its unique chemistry allows for testing on a few samples or even single-sample analyses without sacrificing extreme sensitivity. Additionally, compared to PCR-based protein detection methods, which are sensitive but lengthy and prone to contamination due to their multiple steps, NanoAmp addresses these challenges by providing a straightforward, single-step method equivalent to a standard PCR analysis for antigen detection. This study demonstrates NanoAmp's efficacy as a modular platform accommodating various binding moieties for diverse biological targets. The platform successfully operated with antibodies, proteins, aptamers, and peptides, showcasing its versatility and high sensitivity. Employing the DIG binding motif, the system detected anti-DIG antibodies at concentrations as low as 66 picomolar (pM) and exhibited effectiveness in crude biological samples. NanoAmp aims to revolutionize the world of immunoassays by offering a platform capable of producing highly sensitive and easy-to-use assays for one-pot detection of a plethora of antigens. Additionally, its modular nature allows for the development of customizable research use only (RUO) assays, as different analyte binding motifs can be easily paired with the core system to tailor tools for specific needs.

Robert Strutt, ETH Zurich, CH

In vitro pharmacology is constrained by the biological relevance of the experimental system, experimental throughput and analytical method flexibility. To address these bottlenecks, fundamentally new approaches are required. Across the course of the NCCR MSE, the Bioanalytics lab has developed a series of novel techniques which blend advances from the fields of artificial cell engineering and analytical chemistry. This contribution will focus on recent developments with artificial cell networks assembled from droplets. With our experimental approach, we exert spatial-temporal control over nanoliter droplets containing chemical and biological stimuli. With this, we have unlocked pharmacokinetic / pharmacodynamic (PK/PD) modelling in single droplets and droplet networks.

In these systems, the PK component is defined by transport between droplet compartments connected by biomimetic membranes. Through assessment of an FDA approved drug library, we categorized drugs with a propensity to undergo passive diffusion. Similar physicochemical features which predict in vivo bioavailability correlate with the drug kinetics in our system. Physiologically relevant variables such as temperature, buffer, pH and membrane composition provide tools for modulating the rate of transport and improving system biomimicry. Using our approach, we can reconstitute drug concentration profiles, which capture the peak and trough flux of oral and intravenously administered drugs. In response, the PD component is measured in situ through the inclusion of living cells. By arranging the number of membrane barriers between a source of antibiotic and the target site of action, our system can mimic intracellular infections. Control over the number and arrangement of connected droplets passively distributes antibiotic throughout the system. Through modelling and simulation, we couple the experimentally measured PK and PD responses, facilitating detailed interrogation of antibiotic efficacy under subtle variations in the drug concentration profile. With this simple to use and automatable methodology, we therefore introduce a novel framework for designing and conducting pharmacological assays.

Lee Cronin, University of Glasgow, UK

Recent advancements in automation and digitization of chemistry have opened new avenues for exploring chemical complexity. In this talk I will explain how Assembly Theory and Chemputation can be used to develop a new paradigm to understand and harness the principles of Assembly Theory in chemical synthesis. Assembly Theory provides a framework for quantifying molecular complexity and understanding the emergence of complex chemical systems. Chemputation, on the other hand, offers a standardized method for digitizing and automating chemical synthesis through modular robotic platforms and a chemical programming language (χDL). By combining these approaches, researchers can systematically explore vast chemical spaces, optimize reaction conditions, and potentially discover novel molecules and materials. The integration of these two methodologies enables a new approach to explore chemical space with autonomous experimentation and discovery. As these technologies continue to evolve, they promise to accelerate chemical research, improve reproducibility, provide new insights into the fundamental nature of chemical complexity, as well as an entirely new language.

Martin D. Burke, University of Illinois, USA

Toolmaking made us human, and over the last 2 million years we’ve gotten pretty good at it. But we’ve only been intentionally making tools on the molecular scale for about 200 years. And currently only a tiny fraction of a fraction of the world’s population can meaningfully participate in the molecular innovation process. Consider that many of the most important challenges facing society today likely have molecular solutions that await discovery. Then imagine the impact we could achieve together if everybody could make molecules.

Small molecules in particular possess tremendous functional potential that remains largely untapped due to “the synthesis bottleneck”. This limits both the efficiency with which small molecules can be made by chemists and the participation of non-specialists in the molecular innovation process. An automatable modular platform based on iterative coupling of iminodiacetic acid boronate building blocks is accelerating and expanding access to small molecule synthesis and functional discovery. This platform has specifically been leveraged to develop molecular prosthetics – small molecules that autonomously perform protein-like functions and thus have the potential to treat a wide range of currently incurable diseases, including cystic fibrosis and anemias. Recent advances in Csp3 cross-coupling are substantially expanding the scope of complex small molecules that are accessible. Interfacing this approach with frontier AI and automated functional testing methods has enabled closed-loop discovery of new molecular functions, including organic lasers and photovoltaics. A first of its kind “Molecule Maker Lab” has now been created at UIUC that has opened the door for non-specialists to enter the molecular innovation process. And we are dreaming about globally accessible innovation competitions that leverage such resources to capture so many brilliant imaginations that the field of chemistry has been missing out on. Continued advances in these directions have the potential to democratize molecular innovation and thereby broadly empower tomorrow’s molecular innovators worldwide.

Gregory Stephanopoulos, MIT, USA

The importance of liquid fuels in transportation is well established, yet, there are presently no viable options for their cost-effective production from renewable feedstocks. During the past 15 years we have been engineering the molecular and bioprocess system for the conversion of sugar substrates and gasses to oils and alkanes. Despite achieving near theoretical yields, production of liquid fuels from sugars in not economical, due mainly to the high substrate cost. Mixtures of gasses, on the other hand, like CO₂ and hydrogen (or CO) is more promising using a two-stage system comprising anaerobic fixation of CO₂ and conversion of the CO₂ fixation product (for example, acetate) to lipids, from which biodiesel can be produced. In another application, the CO₂ fixation product is converted to alkanes. Our work includes both the molecular engineering of the microbes and development of a process to achieve gas to liquid conversion in prototype systems. These systems are scalable, make no use of land (beyond what is needed for generating renewable electricity for hydrogen production), do not compete with food and are cost competitive based on high level cost analysis and TEA. I will present the essential features of this process in my talk.

Brigitte Städler, Aarhus University, DK

Bottom-up synthetic biology aims to design life-like units (aka artificial cells) that can substitute for missing/lost cellular activity or to add non-native function to mammalian cells and tissue. Artificial cells are minimal, simplistic structures that imitate selected structural or functional aspects of living cells.

We focus our efforts on hydrogel-based artificial cells equipped with a specific liver-like function and their integration and communication with mammalian cells. Specifically, the artificial cells support their living counterpart in fighting reactive oxygen species either by direct conversion or by deploying supportive nano-units. Further, we showed that hepatic cell aggregates could be 3D bioprinted together with artificial cells to boost catalytic activity for at least 2 weeks. In addition, we illustrated that artificial cells can eavesdrop on a typical activity of a liver cell due to co-existence in a semi-synthetic tissue.

Our efforts illustrate the potential of nano-engineered artificial cells for tissue engineering purposes.

Marcos Manero Carranza, ETH Zurich, CH

The discovery and repurposing of CRISPR effectors has allowed for the development of a genome editing toolbox capable of performing programmable and efficient genetic interventions for a wide range of therapeutic and research purposes. However, most available tools directly modify the DNA sequence or alter gene expression level, leaving RNA modification comparatively unexplored. RNA editing, however, offers the opportunity to introduce either stable or transient modifications to nucleic acid sequences, without the risk of permanent off-target effects. This could be applied for sensing, labelling, or repairing of RNA transcripts, but installation of arbitrary edits into the transcriptome is currently infeasible. Trans-splicing-based RNA editing technologies can install diverse edits, insertions, and deletions by introducing exogenous templates via competition with the endogenous exon in the pre-spliced mRNA. However, currently they suffer from low efficiency, even after extensive screening, which hinders their applicability. Here, we describe Programmable RNA Editing & Cleavage for Insertion, Substitution, and Erasure (PRECISE), a versatile RNA editing method for writing RNA of arbitrary length and sequence into existing pre-mRNAs via 5′ or 3′ trans-splicing. We demonstrate PRECISE editing across 11 distinct endogenous transcripts of widely varying expression levels, showcasing more than 50 types of edits, including all possible transversions and transitions and a wide range of insertions and deletions. We show high efficiency replacement of MECP2 last exon, addressing most mutations that drive the Rett Syndrome; editing of SHANK3 transcripts, a gene involved in Autism; and replacement of exon 1 of HTT, removing the hallmark repeat expansions of Huntington′s disease. Furthermore, we combine payload engineering and ribozymes for protein-free, high-efficiency trans-splicing, with demonstrated efficiency in editing HTT exon 1 via AAV delivery. We show that the PRECISE achieves editing in non-dividing neurons and patient-derived Huntington’s disease fibroblasts. Our results provide the scientific community with a novel tool that markedly broadens the scope of RNA editing, is straightforward to deliver, lacks permanent off-targets, and can enable any type of edit, including those not otherwise possible with current RNA base editors. Ultimately, these findings expand the current RNA editing toolbox for therapeutic or research purposes and widen the spectrum of addressable diseases.

Denise Di Lena, University of Parma, IT

Monoclonal antibodies (mAbs) represent a key therapeutic option for managing chronic conditions. Therapeutic drug monitoring (TDM) presents a promising method for personalized treatment plans, especially in chronic diseases where excessive treatment can result in significant adverse effects. To advance TDM strategies, the NanoHybrid (NH) platform developed by Ulisse BioMed has been redesigned into an innovative protein-DNA nanoswitch sensor. This new configuration utilizes entire protein binding units to detect mAbs by taking advantage of multiple identifiable epitopes. Nanoswitch probes, composed of specifically designed DNA strands conjugated with whole tumor necrosis factor α (TNFα) proteins, facilitated the one-step quantification of various mAbs such as Infliximab and Adalimumab in concentrations ranging from 2.4 µg/mL to 19 µg/mL directly in blood serum. Given that mAbs can induce patient-specific immune responses, anti-drug antibodies (ADAs) are frequently generated against biological medications, significantly affecting the actual bioavailability of therapeutic mAbs. Inspired by this, the nanoswitch platform was utilized to detect Infliximab in the presence of an anti-Infliximab antibody, revealing a notable decrease in the detected concentrations of Infliximab when such ADA was present. These findings demonstrated that the developed nanoswitch platform can specifically identify bioavailable antibodies, offering valuable insights for pharmacokinetic research. With its adaptable design, the current nanoswitch platform has potential as a multi-epitope nanosensor for the measurement of bioavailable mAbs and biosimilars.

Barbara Treutlein, ETH Zurich, CH

Pluripotent stem cell derived organoids are exciting, complex in vitro models to study human organ development. Integrative, multi-modal single-cell technologies are needed to understand the mechanisms underlying fate specification during human organoid development.

In my talk, I will present our efforts to develop and use single-cell technologies combined with genetic and environmental perturbations to dissect the mechanisms underlying patterning and fate specification during human organoid development with a focus on brain and vasculature. Further, I will highlight our attempts to further improve organoid development and maturation by introducing missing lineages. Together, our work highlights the power of single-cell and organoid technologies to understand cell fate and state specification during human organ development.

Seraphine Valeska Wegner, University of Münster, DE

Bottom-up synthetic biology aims to construct cell-like systems starting from molecular building blocks and give insight into principles that give rise to cell function. Many functions in cells arise directly from the spatial and temporal control of processes such as protein localization, cell migration, tissue assembly and cell-to-cell communication. In this talk, I will present strategies of how such spatiotemporal control over adhesions in synthetic cells can be achieved with visible light using photoswitchable proteins and functions that arise from these. The photoswitchable adhesions allow us to recapitulate cell migration, to self-assemble and self-sort synthetic cells into multicellular functional architectures with high precision. Moreover, the organization in these multicellular communities is of significance for their communication and the overall arising behaviors. These synthetic cell-mimetic systems, which reduce complexity and yet capture key features of natural cells, allow us to quantify and correlate cell behavior with molecular information. Further, complementary approaches pursued with synthetic minimal cells as well as bacterial and mammalian cells allow translating concepts between different systems and integration into hybrid structures. Overall, our work on one hand provides insight into underlying design principles of life and on the other hand engineer new synthetic cell biology.

Christopher Reinkemeier, ETH Zurich, CH

Engineering new functionalities into living systems has a tremendous potential for biotechnology and medicine. However, using canonical routes of evolution can be limited by the necessity to avoid interference with endogenous processes. Cells themselves do not only rely on molecular evolution to prevent undesired crosstalk, but they often utilize compartmentalization, in the form of membrane-enclosed or membraneless organelles to spatially isolate distinct molecular processes and execute complex operations that would otherwise interfere with each other.

Inspired by this, we developed synthetic organelles to equip mammalian cells with new functions. Specifically, we combined phase separating proteins such as FUS and EWSR1 with anchor proteins that tether the organelles to specific subcellular structures. With this we can generate organelles spanning vastly different shapes from kinesin-motor based micrometer sized organelles over fiber-like organelles along the microtubule cytoskeleton to film-like organelles on membrane surfaces that have a thickness of less than 100 nm.

We then applied these organelles to improve genetic code expansion (GCE) in mammalian cells. GCE is a powerful tool to control and expand protein function with single-residue precision that is widely used to perform labeling for microscopy or to photocontrol proteins. This technology relies on an orthogonal tRNA/synthetase pair that is introduced into the host, which recodes a stop codon to incorporate a noncanonical amino acid (ncAA) into the nascent chain. While this technique is codon-specific, it cannot select specific mRNAs, so naturally occurring stop codons can be suppressed, causing significant side effects. To overcome this, we integrated a GCE system into our synthetic organelles to develop orthogonally translating (OT) organelles. These OT organelles can perform mRNA specific GCE, installing ncAAs exclusively into desired proteins. We further demonstrate that it is possible to combine different OT organelles in one cell, effectively yielding cells that simultaneously execute multiple distinct genetic codes. Our results demonstrate a simple yet effective approach to generate artificial organelles that enable customized orthogonal translation and protein engineering in semi-synthetic eukaryotic cells.

Andres Garcia-Garcia, University of Basel, CH

Human induced pluripotent stem cells (hiPSCs) have emerged as a powerful tool to engineer robust and reproducible organoids to model human biology and pathology, paving the way for mechanistic and pharmacological studies in human settings. Recently, the two first bone marrow (BM) organoid models based on hiPSCs were presented, and showed to successfully recapitulate key features of human BM while modeling both healthy and malignant hematopoiesis. Although these BM organoids represent a breakthrough in modeling the functional complexity of human BM perivascular niches, they lack the bone compartment, and thus they are not suitable to model human endosteal BM niches (close to the bone surface). Moreover, these organoids remain in the micrometer scale and therefore the vascular structures are non-physiological in shape and size. Finally, they rely on the use of Matrigel as embedding material, which introduces mouse-derived proteins in the system.

Here we present a novel developmentally-guided approach combining hiPSC-derived organoids with macro-scale hydroxyapatite scaffolds to generate a standardized and physiological-like model of the human endosteal BM perivascular niche (engineered vascularized osteoblastic niche, eVON). We developed and validated a protocol to differentiate hiPSCs into vascular cells (endothelial cells and pericytes) and osteoblasts that later self-assemble to create complex and long-lasting vascular networks integrated in a dense osteogenic matrix, resembling the 3D architecture of the native endosteal BM. The eVON was first characterized through flow cytometry, high-resolution imaging and scRNAseq. Second, we demonstrated that it can persist for at least 6 weeks upon in vivo ectopic implantation and integrate within the murine tissue. Then, we assessed its potential to support human hematopoiesis and explored its customization potential by engineering different eVON using different hiPSC lines. Finally, we validated the model for pharmacological studies by targeting VEGF signaling and analyzing the effects on the vasculature. Therefore, this work provides the first standardized and physiological-like model of the human endosteal BM, offering an unprecedented possibility to dissect the contribution of BM endosteal niches to human pathophysiological hematopoiesis.

Randall Platt, ETH Zurich, CH

The ever-growing compendium of genetic variants associated with human pathologies demands new methods to study genotype-phenotype relationships in complex tissues in a high-throughput manner. Here we introduce adeno-associated virus (AAV)-mediated direct in vivo single-cell CRISPR screening, termed AAV-Perturb-seq, a tuneable and broadly applicable method for transcriptional linkage analysis as well as high-throughput and high-resolution phenotyping of genetic perturbations in vivo. We applied AAV-Perturb-seq using gene editing and transcriptional inhibition to systematically dissect the phenotypic landscape underlying 22q11.2 deletion syndrome genes in the adult mouse brain prefrontal cortex. We identified three 22q11.2-linked genes involved in known and previously undescribed pathways orchestrating neuronal functions in vivo that explain approximately 40% of the transcriptional changes observed in a 22q11.2-deletion mouse model. Our findings suggest that the 22q11.2-deletion syndrome transcriptional phenotype found in mature neurons may in part be due to the broad dysregulation of a class of genes associated with disease susceptibility that are important for dysfunctional RNA processing and synaptic function. Our study establishes a flexible and scalable direct in vivo method to facilitate causal understanding of biological and disease mechanisms with potential applications to identify genetic interventions and therapeutic targets for treating disease.