Large Assemblies Research Articles

A Generally Applicable Method for Disentangling the Effect of Individual Noncovalent Interactions on the Binding Energy.

We introduce the fragment-pairwise Local Energy Decomposition (fp-LED) scheme for precise quantification of individual interactions contributing to the binding energy of arbitrary chemical entities, such as protein-ligand binding energies, lattice energies of molecular crystals, or association energies of large biomolecular assemblies. Using fp-LED, we can assess whether the contribution to the binding energy arising from noncovalent interactions between pairs of molecular fragments in any chemical system is attractive or repulsive, and accurately quantify its magnitude at the coupled cluster level - commonly considered as the "gold standard" of computational chemistry. Such insights are crucial for advancing molecular and material design strategies in fields like catalysis and therapeutic development. Illustrative applications across diverse fields demonstrate the versatility and accuracy of this theoretical framework, promising profound implications for fundamental understanding and practical applications.

Precision compensation method for large-scale measurements based on full-space refractive index field reconstruction with vibratory mirror background oriented schlieren (VMBOS)

With the development of the large-scale equipment manufacturing industry, such as aircraft component assembly, wind turbine rotor blade measurement and large ship assembly the demand for optical measurement accuracy in large-dimensional spaces within industrial environments has become increasingly stringent. However, due to the presence of internal interference sources, light will no longer propagate in a straight line, thereby introducing measurement errors that cannot be ignored. To compensate for the measurement error, the refractive index field distribution within the large-dimensional measurement space must be reconstructed. Therefore, this paper proposed a vibratory mirror background oriented schlieren (VMBOS) based on vibratory scanning. This method scans the measurement space through the rotation of vibratory mirror to obtain light deflection angles at different directions. Subsequently, the refractive index field of the large-dimensional space is reconstructed by using the tomographic reconstruction algorithm. Then the optical path of the light in the optical measurement can be corrected through the Hamiltonian ray tracing algorithm, thereby achieving compensation for the measurement error. Finally, the VMBOS method has been verified through both simulation and experimental methods. The experimental results demonstrated that the method proposed in this paper can be applied to optical path correction and error compensation in large scale spaces.

STRPsearch: fast detection of structured tandem repeat proteins.

Structured Tandem Repeats Proteins (STRPs) constitute a subclass of tandem repeats characterized by repetitive structural motifs. These proteins exhibit distinct secondary structures that form repetitive tertiary arrangements, often resulting in large molecular assemblies. Despite highly variable sequences, STRPs can perform important and diverse biological functions, maintaining a consistent structure with a variable number of repeat units. With the advent of protein structure prediction methods, millions of 3D-models of proteins are now publicly available. However, automatic detection of STRPs remains challenging with current state-of-the-art tools due to their lack of accuracy and long execution times, hindering their application on large datasets. In most cases, manual curation remains the most accurate method for detecting and classifying STRPs, making it impracticable to annotate millions of structures. We introduce STRPsearch, a novel tool for the rapid identification, classification, and mapping of STRPs. Leveraging manually curated entries from RepeatsDB as the known conformational space of STRPs, STRPsearch employs the latest advances in structural alignment for a fast and accurate detection of repeated structural motifs in proteins, followed by an innovative approach to map units and insertions through the generation of TM-score profiles. STRPsearch is highly scalable, efficiently processing large datasets, and can be applied to both experimental structures and predicted models. Additionally, it demonstrates superior performance compared to existing tools, offering researchers a reliable and comprehensive solution for STRP analysis across diverse proteomes. STRPsearch is coded in Python. All scripts and associated documentation are available from: https://github.com/BioComputingUP/STRPsearch. Supplementary data are available at Bioinformatics online.

Integrated Profiling Identifies Long-Term Molecular Consequences of Prenatal Dexamethasone Treatment in the Rat Brain-Potential Triggers of Depressive Phenotype and Cognitive Impairment.

Prenatal excess of glucocorticoids (GCs) is considered to be one of the highly impacting factors contributing to depression development. Although GCs are crucial for normal fetal development and their administration (mainly dexamethasone, DEX) is a life-saving procedure for those at risk of preterm delivery, exposure to excess levels of GCs during pregnancy can yield detrimental consequences. Therefore, we aimed to systematically investigate the brain molecular alterations triggered by prenatal DEX administration. We used a rat model of depression based on prenatal exposure to DEX and performed integrative multi-level methylomic, transcriptomic, and proteomic analyses of adult rats' brains (i.e., frontal cortex (FCx) and hippocampus (Hp)) to identify the outcomes of DEX action. Each of the investigated levels was significantly affected by DEX in the long-term manner. Particularly, we found 200 CpG islands to be differentially methylated in the FCx and 200 in the Hp of prenatally DEX-treated rats. Global transcriptomic analysis uncovered differential expression of transcripts mostly in FCx (271) and 1 in Hp, while proteomic study identified 146 differentially expressed proteins in FCx and 123 in Hp. Among the identified enriched molecular networks, we found altered pathways involved in synaptic plasticity (i.e., cAMP, calcium, and Wnt signaling pathways or tight junctions and adhesion molecules), which may contribute to cognitive impairment, observed in DEX-treated animals. Moreover, in the FCx, DEX administration in the prenatal period downregulates the expression of ribosome protein genes associated both with large and small ribosomal subunit assembly which can lead to a global decrease in translation and protein synthesis processes and, indirectly, alterations in the neurotransmission process.

STEM-09. A GROUP 3 MEDULLOBLASTOMA STEM CELL PROGRAM IS MAINTAINED BY OTX2-MEDIATED ALTERNATIVE SPLICING

Abstract Group 3 medulloblastomas (MBs) are highly metastatic pediatric brain tumors with limited therapeutic options. The transcription factor orthodenticle homeobox 2 (OTX2) is a known driver and molecular hallmark of Group 3 MB. We previously demonstrated that OTX2 regulates Group 3 MB cell fate decisions that favor self-renewal or stemness and the retention of a primitive cerebellar rhombic lip signature. Here, we show that OTX2-regulated alternative splicing (AS) is a major driver of Group 3 MB progression. Surprisingly, the functional role of post-transcriptional events, such as AS, in Group 3 MB development has never been explored. This gene regulatory layer represents an untapped source of therapeutic targets. Using a combination of TurboID, RNA-sequencing and single cell RNA-seq, we characterized the Group 3 primary MB AS landscape with a focus on genes that control MB stem cells. OTX2 associates with the multimeric Large Assembly of Splicing Regulators (LASR) complex through protein-protein interactions and can directly or indirectly bind RNA. Together, OTX2 and LASR complex proteins oversee a pro-tumorigenic splicing program that is mirrored in the human cerebellar rhombic lip origins and is associated with Group 3γ MBs that exhibit the worst prognosis. Among our core set of OTX2-regulated spliced genes, periphilin 1 (PPHLN1) is expressed in the most primitive cerebellar rhombic lip stem cells, and mimicking PPHLN1 AS by pre-treating cells with a splice blocking antisense morpholino reduces tumor growth while increasing survival in vivo. We propose that OTX2-mediated AS is a key determinant of cell fate decisions favoring the maintenance of the stem cell state. Targeting transcription factors like OTX2 is notoriously difficult; thus, our work offers fresh opportunities to exploit MB-specific AS events therapeutically to abrogate tumor progression.

A new discrete-geometry approach for integrative docking of proteins using chemical crosslinks.

The structures of protein complexes allow us to understand and modulate the biological functions of the proteins. Integrative docking is a computational method to obtain the structures of a protein complex, given the atomic structures of the constituent proteins along with other experimental data on the complex, such as chemical crosslinks or SAXS profiles. Here, we develop a new discrete geometry-based method, wall-EASAL, for integrative rigid docking of protein pairs given the structures of the constituent proteins and chemical crosslinks. The method is an adaptation of EASAL (Efficient Atlasing and Search of Assembly Landscapes), a state-of-the-art discrete geometry method for efficient and exhaustive sampling of macromolecular configurations under pairwise inter-molecular distance constraints. We provide a mathematical proof that the method finds a structure satisfying the crosslink constraints under a natural condition satisfied by energy landscapes. We compare wall-EASAL with IMP (Integrative Modeling Platform), a commonly used integrative modeling method, on a benchmark, varying the numbers, types, and sources of input crosslinks, and sources of monomer structures. The wall-EASAL method performs better than IMP in terms of the average satisfaction of the configurations to the input crosslinks and the average similarity of the configurations to their corresponding native structures. The ensembles from IMP exhibit greater variability in these two measures. Further, wall-EASAL is more efficient than IMP. Although the current study uses crosslinks, the method is general and any source of distance constraints can be used for integrative docking with wall-EASAL. However, the current implementation only supports binary rigid protein docking, i.e., assumes that the monomer structures are known and remain rigid. Additionally, the current implementation is deterministic, i.e., it does not account for uncertainties in the crosslinking data beyond using distance bounds. Neither of these appears to be a theoretical or algorithmic limitation of the EASAL methodology. Structures from wall-EASAL can be incorporated in methods for modeling large macromolecular assemblies, for example by suggesting rigid bodies or restraints for use in these methods. This will facilitate the characterization of assemblies and cellular neighborhoods at increased efficiency, accuracy, and precision. The wall-EASAL method is available at https://bitbucket.org/geoplexity/easal-dev/src/Crosslink and the benchmark is available at https://github.com/isblab/Integrative_docking_benchmark.

Topological rearrangements activate the HerA-DUF anti-phage defense system.

Leveraging the rich structural information provided by AlphaFold, we used integrated experimental approaches to characterize the HerA-DUF4297 (DUF) anti-phage defense system, in which DUF is of unknown function. To infer the function of DUF, we performed structure-guided genomic analysis and found that DUF homologs are universally present in bacterial immune defense systems. One notable homolog of DUF is Cap4, a universal effector with nuclease activity in CBASS, the most prevalent anti-phage system in bacteria. To test the inferred nuclease function of DUF, we performed biochemical experiments and discovered that the DUF only exhibits activity against DNA substrates when it is bound by HerA. To understand how HerA activates DUF, we determined the structures of DUF and the HerA-DUF complex. DUF forms large oligomeric assemblies with or without HerA, suggesting that oligomerization per se is not sufficient for DUF activation. Instead, DUF activation requires dramatic topological rearrangements that propagate from HerA to the entire HerA-DUF complex, leading to reorganization of DUF for effective DNA cleavage. We further validated these structural insights by structure- guided mutagenesis. Together, these findings reveal dramatic topological rearrangements throughout the HerA-DUF complex, challenge the long-standing dogma that protein oligomerization alone activates immune signaling, and may inform the activation mechanism of CBASS.

The conformational equilibria of a human GPCR compared between lipid vesicles and aqueous solutions by integrative 19F-NMR.

Endogenous phospholipids influence the conformational equilibria of G protein-coupled receptors, regulating their ability to bind drugs and form signaling complexes. However, most studies of GPCR-lipid interactions have been carried out in mixed micelles or lipid nanodiscs. Though useful, these membrane mimetics do not fully replicate the physical properties of native cellular membranes associated with large assemblies of lipids. We investigated the conformational equilibria of the human A2A adenosine receptor (A2AAR) in phospholipid vesicles using 19F solid-state magic angle spinning NMR (SSNMR). By applying an optimized sample preparation workflow and experimental conditions, we were able to obtain 19F-SSNMR spectra for both antagonist- and agonist-bound complexes with sensitivity and linewidths closely comparable to those achieved using solution NMR. This facilitated a direct comparison of the A2AAR conformational equilibria across detergent micelle, lipid nanodisc, and lipid vesicle preparations. While antagonist-bound A2AAR showed a similar conformational equilibria across all membrane and membrane mimetic systems, the conformational equilibria of agonist-bound A2AAR exhibited differences among different environments. This suggests that the conformational equilibria of GPCRs may be influenced not only by specific receptor-lipid interactions but also by the membrane properties found in larger lipid assemblies.

FMO-LC-TDDFTB method for excited states of large molecular assemblies in the strong light-matter coupling regime.

We present a new methodology to calculate the strong light-matter coupling between photonic modes in microcavities and large molecular aggregates that consist of hundreds of molecular fragments. To this end, we combine our fragment molecular orbital long-range corrected time-dependent density functional tight-binding methodology with a generalized Tavis-Cummings Hamiltonian. We employ an excitonic Hamiltonian, which is built from a quasi-diabatic basis that is constructed from locally excited and charge-transfer states of all molecular fragments. To calculate polaritonic states, we extend our quasi-diabatic basis to include photonic states of a microcavity and derive and implement the couplings between the locally excited states and the cavity states and built a Tavis-Cummings Hamiltonian that incorporates the intermolecular excitonic couplings. Subsequently, we demonstrate the capability of our methodology by simulating the influence of the electric field polarization on the polaritonic spectra for a tetracene aggregate of 125 monomers. Furthermore, we investigate the dependence of the splitting of the upper and lower polaritonic branches on the system size by comparing the spectra of five different tetracene clusters. In addition, we investigate the polariton dispersion of a large tetracene aggregate for electric field polarizations in the x, y, and z directions. Our new methodology can facilitate the future study of exciton dynamics in complex molecular systems, which consist of up to hundreds of molecules that are influenced by strong light-matter coupling to microcavities.

Coordinate-based simulation of pair distance distribution functions for small and large molecular assemblies: implementation and applications.

X-ray scattering has become a major tool in the structural characterization of nanoscale materials. Thanks to the widely available experimental and computational atomic models, coordinate-based X-ray scattering simulation has played a crucial role in data interpretation in the past two decades. However, simulation of real-space pair distance distribution functions (PDDFs) from small- and wide-angle X-ray scattering, SAXS/WAXS, has been relatively less exploited. This study presents a comparison of PDDF simulation methods, which are applied to molecular structures that range in size from β-cyclo-dextrin [1 kDa molecular weight (MW), 66 non-hydrogen atoms] to the satellite tobacco mosaic virus capsid (1.1 MDa MW, 81 960 non-hydrogen atoms). The results demonstrate the power of interpretation of experimental SAXS/WAXS from the real-space view, particularly by providing a more intuitive method for understanding of partial structure contributions. Furthermore, the computational efficiency of PDDF simulation algorithms makes them attractive as approaches for the analysis of large nanoscale materials and biological assemblies. The simulation methods demonstrated in this article have been implemented in stand-alone software, SolX 3.0, which is available to download from https://12idb.xray.aps.anl.gov/solx.html.

White Light Emission from Zn(II) and DMSO-Induced Copper Nanocluster Assembly.

An assembly of metal nanoclusters driven by appropriate surface ligands and solvent environment may engender entirely new photoluminescence (PL). Herein, we first synthesize histidine (His) stabilized copper nanoparticles (CuNPs) and, subsequently, copper nanoclusters (CuNCs) from it using 3-mercaptopropionic acid (MPA) as an etchant. The CuNCs originally emit bluish-green (λem=470 nm) PL with a low quantum yield (QY∼1.8 %). However, it transformed into a dual-emissive nanocluster assembly (Zn-CuNCs) in the presence of Zn(II) salt, having a distinct blue emission band (λem=420 nm) and a red emission band (λem=615 nm) with eight times QY (∼9.1 %) enhancement. The temperature-dependent emission spectra of Zn-CuNCs depicted that the blue emission band persists for all the temperature ranges (0-80 °C) while the red emission band vanishes at high temperatures (70-80 °C). Thus, the blue emission may originate from the locally excited state (LES) emission of the nanoclusters, while the red emission originates from through-space interaction (TSI) and Cu(I)…Cu(I) interaction within the assembly. Adding dimethyl sulfoxide (DMSO) further modifies the emission intensities; the red band was amplified four times, while the blue band was diminished by 2.5 times. The transmission electron microscopy (TEM) images unveiled that the Zn-CuNCs are a large assembly of tiny nanoclusters, which become more compact in DMSO. The blue emission possesses steady-state fluorescence anisotropy, while the red emission shows no anisotropy. Further, near-perfect white light emission(WLE) was rendered with CIE coordinates of (0.33, 0.32) by combining the dual emission of the Zn-CuNCs with the original green emission of the CuNCs.

Rational design of metal–organic cages to increase the number of components via dihedral angle control

The general principles of discrete, large self-assemblies composed of numerous components are not unveiled and the artificial formation of such entities is a challenging topic. In metal–organic cages, design strategies for tuning the coordination directions in multitopic ligands by the bend and twist angles were previously developed to solve this problem. In this study, the importance of remote geometric communications between components is emphasized, realizing several types of metal–organic assemblies based on dihedral angle control in multitopic ligands although they have the same coordination directions. Self-assembly of a tritopic ligand with dihedral angles θ = 36.5° and a cis-protected Pd(II) ion affords M9L6 and M12L8 cages as kinetic and thermodynamic products, respectively, whereas an M12L8 sheet is formed when θ = 90°. Geometric analyses of strains in the subcomponent rings reveals that remote geometric communications among neighboring multitopic ligands through coordination bonds are key for large assemblies.

Stacking molecular chips in multiple dimensions

Our quest is to control how molecules come together to form large assemblies. In our earlier studies we showed that we can stack saddle-shaped molecular chips into neat columns that then stick together laterally to form 2D sheets, but can we limit this process to form different morphologies? Here we show how blocking the sides of the columns equalizes the secondary interaction, yielding 1D fibers.

SuFEx Chemistry Enables Covalent Assembly of a 280-kDa 18-Subunit Pore-Forming Complex.

Proximity-enhanced chemical cross-linking is an invaluable tool for probing protein-protein interactions and enhancing the potency of potential peptide and protein drugs. Here, we extend this approach to covalently stabilize large macromolecular assemblies. We used SuFEx chemistry to covalently stabilize an 18-subunit pore-forming complex, CsgG:CsgF, consisting of nine CsgG membrane protein subunits that noncovalently associate with nine CsgF peptides. Derivatives of the CsgG:CsgF pore have been used for DNA sequencing, which places high demands on the structural stability and homogeneity of the complex. To increase the robustness of the pore, we designed and synthesized derivatives of CsgF-bearing sulfonyl fluorides, which react with CsgG in very high yield to form a covalently stabilized CsgG:CsgF complex. The resulting pores formed highly homogeneous channels when added to artificial membranes. The high yield and rapid reaction rate of the SuFEx reaction prompted molecular dynamics simulations, which revealed that the SO2F groups in the initially formed complex are poised for nucleophilic reaction with a targeted Tyr. These results demonstrate the utility of SuFEx chemistry to structurally stabilize very large (here, 280 kDa) assemblies.

A 3-DOF Multi-Mode spherical actuator driven by cooperative piezoelectric units

Stick-slip piezoelectric actuators are widely used in precision engineering due to their high accuracy and small size. However, most of these actuators achieve multi-DOF motion by serially connecting multiple units, resulting in a non-compact structure and large assembly errors.To resolve this issue, this paper proposes a spherical actuator consists of a rotor and four drive units distributed around the rotor. Each drive unit comprises a piezoelectric shear stack and a piezoelectric plate. By utilizing the cooperation of piezoelectric elements, various driving modes with different performance and control complexity can be switched. To understand the influence of position and orientation of the drive units on the actuator’s performance, a dynamic model describing the behavior of the proposed actuator is established. This model takes into account the behavior of the drive units, friction and motion transmission between the drive units and the rotor, as well as transformation of coordinates. Finally, a prototype is developed based on the design, and the three-DOF motion performance of the prototype is tested under different drive modes. This work contributes to the mechanical design of multi-DOF precision positioning devices and the study of motion modes.

Controlled Assembly of Nanowires into Welded Nanowire Networks for Memristor Device Fabrication

Next-generation electronic devices are essential for neuromorphic computing (or brain-inspired computing), which is essential for the high-speed processing of big data using artificial intelligence tools.(1) Memristors, an important constituent of these next-generation electronic devices, have been fabricated using many forms of metal oxides, including nanowires. Nanowires forms of oxides, although reduce the complexity of the memristive device fabrication process, need many technological advances for their widespread use in memristor fabrication.(1) For instance, the cyclability of devices needs to be improved. Furthermore, realizing the high density of nanowire-based memristor devices requires methods for addressing individual nanowires in large assemblies (i.e., arrays) of nanowires, strategies for which also need to be developed.(1) A possible avenue to address these problems is through the use of nanowire networks,(1) specifically those that have nanowires welded together via bridges that have the same chemical composition as those of the nanowires themselves.(2) In this configuration, the resistive switching of both the nanowires and the randomly-formed junctions between the nanowires can be employed to make mesoscale devices, circumventing the need to individually address each nanowire in the nanowire assemblies.(1) Realizing memristive devices based on welded nanowire networks, therefore, requires a strategy for assembly via welding of pre-synthesized nanowires into networks that offer the ability to control the chemical compositions of the interfaces between the nanowires and the densities of nanowires within the assemblies.(2, 3) In the presentation, strategies developed by our group for the assembly of semiconductor nanowires into welded nanowire networks of controlled densities will be discussed in detail. Specific devices that will be discussed include assembly via welding of Mg2Si, TiO2 and TiC nanowires.(2, 3) The assembly via welding of nanowires into networks is expected to accelerate the testing of mesoscale memristive devices, irrespective of the method employed for the synthesis of the nanowires used in the assemblies. This is expected to aid in accelerating both the science and technology needed for the deployment of memristive devices.

Putting It All Together: The Roles of Ribosomal Proteins in Nucleolar Stages of 60S Ribosomal Assembly in the Yeast Saccharomyces cerevisiae.

Here we review the functions of ribosomal proteins (RPs) in the nucleolar stages of large ribosomal subunit assembly in the yeast Saccharomyces cerevisiae. We summarize the effects of depleting RPs on pre-rRNA processing and turnover, on the assembly of other RPs, and on the entry and exit of assembly factors (AFs). These results are interpreted in light of recent near-atomic-resolution cryo-EM structures of multiple assembly intermediates. Results are discussed with respect to each neighborhood of RPs and rRNA. We identify several key mechanisms related to RP behavior. Neighborhoods of RPs can assemble in one or more than one step. Entry of RPs can be triggered by molecular switches, in which an AF is replaced by an RP binding to the same site. To drive assembly forward, rRNA structure can be stabilized by RPs, including clamping rRNA structures or forming bridges between rRNA domains.

Complete list of canonical post-transcriptional modifications in the Bacillus subtilis ribosome and their link to RbgA driven large subunit assembly.

Ribosomal RNA modifications in prokaryotes have been sporadically studied, but there is a lack of a comprehensive picture of modification sites across bacterial phylogeny. Bacillus subtilis is a preeminent model organism for gram-positive bacteria, with a well-annotated and editable genome, convenient for fundamental studies and industrial use. Yet remarkably, there has been no complete characterization of its rRNA modification inventory. By expanding modern MS tools for the discovery of RNA modifications, we found a total of 25 modification sites in 16S and 23S rRNA of B. subtilis, including the chemical identity of the modified nucleosides and their precise sequence location. Furthermore, by perturbing large subunit biogenesis using depletion of an essential factor RbgA and measuring the completion of 23S modifications in the accumulated intermediate, we provide a first look at the order of modification steps during the late stages of assembly in B. subtilis. While our work expands the knowledge of bacterial rRNA modification patterns, adding B. subtilis to the list of fully annotated species after Escherichia coli and Thermus thermophilus, in a broader context, it provides the experimental framework for discovery and functional profiling of rRNA modifications to ultimately elucidate their role in ribosome biogenesis and translation.

Comparative Pore Structure and Dynamics for Bacterial Microcompartment Shell Protein Assemblies in Sheets or Shells.

Bacterial microcompartments (BMCs) are protein-bound organelles found in some bacteria that encapsulate enzymes for enhanced catalytic activity. These compartments spatially sequester enzymes within semipermeable shell proteins, analogous to many membrane-bound organelles. The shell proteins assemble into multimeric tiles; hexamers, trimers, and pentamers, and these tiles self-assemble into larger assemblies with icosahedral symmetry. While icosahedral shells are the predominant form in vivo, the tiles can also form nanoscale cylinders or sheets. The individual multimeric tiles feature central pores that are key to regulating transport across the protein shell. Our primary interest is to quantify pore shape changes in response to alternative component morphologies at the nanoscale. We used molecular modeling tools to develop atomically detailed models for both planar sheets of tiles and curved structures representative of the complete shells found in vivo. Subsequently, these models were animated using classical molecular dynamics simulations. From the resulting trajectories, we analyzed the overall structural stability, water accessibility to individual residues, water residence time, and pore geometry for the hexameric and trimeric protein tiles from the Haliangium ochraceum model BMC shell. These exhaustive analyses suggest no substantial variation in pore structure or solvent accessibility between the flat and curved shell geometries. We additionally compare our analysis to hydroxyl radical footprinting data to serve as a check against our simulation results, highlighting specific residues where water molecules are bound for a long time. Although with little variation in morphology or water interaction, we propose that the planar and capsular morphology can be used interchangeably when studying permeability through BMC pores.

A group 3 medulloblastoma stem cell program is maintained by OTX2-mediated alternative splicing

OTX2 is a transcription factor and known driver in medulloblastoma (MB), where it is amplified in a subset of tumours and overexpressed in most cases of group 3 and group 4 MB. Here we demonstrate a noncanonical role for OTX2 in group 3 MB alternative splicing. OTX2 associates with the large assembly of splicing regulators complex through protein–protein interactions and regulates a stem cell splicing program. OTX2 can directly or indirectly bind RNA and this may be partially independent of its DNA regulatory functions. OTX2 controls a pro-tumorigenic splicing program that is mirrored in human cerebellar rhombic lip origins. Among the OTX2-regulated differentially spliced genes, PPHLN1 is expressed in the most primitive rhombic lip stem cells, and targeting PPHLN1 splicing reduces tumour growth and enhances survival in vivo. These findings identify OTX2-mediated alternative splicing as a major determinant of cell fate decisions that drive group 3 MB progression.