Whitehead Institute for Biomedical Research

Plasmids are the workhorses of molecular biology: fast, flexible, and often taken for granted. We clone, overexpress, tag, and mutate freely, assuming they will faithfully produce RNA transcripts that match the intended DNA sequence. This assumption is rarely tested and often invalidated. Sequences in plasmid backbones, epitope tags, and codon-optimized regions may inadvertently harbor cryptic promoters or splice sites. The resulting unexpected transcripts and proteins, while often undetected, can distort results and propagate false conclusions through papers, grants, and even clinical trials. In this perspective, we highlight published cases where plasmids have distorted results and misled interpretation. We examine the mechanisms and consequences of plasmid-associated expression artifacts and offer practical strategies to minimize them. Finally, we call for a revision of community standards for experiments using transgenes: deposit complete plasmid sequences and verify the resulting transcripts using RNA-seq.

The clinical success of cancer drug candidates depends on efficacy across many different individuals. Because xenografts are challenging to scale, we currently rely on a limited set of in vivo preclinical models. Here, to address this limitation, we introduce GENEVA, a scalable single-cell-resolution platform for measuring responses to drug perturbations. GENEVA models cancer genetic diversity by combining multiple patient-derived cell lines and cancer cell lines into pooled three-dimensional cultures and xenograft models, allowing us to study drug responses across a wide range of genetic backgrounds within a single experiment. We apply GENEVA to investigate KRAS-G12C inhibitors and demonstrate that mitochondrial activation is a key driver of cell death following KRAS inhibition, while epithelial-to-mesenchymal transition is a prominent resistance mechanism. These findings highlight the utility of GENEVA to identify therapeutic targets and optimize combination therapies with the potential to bridge the gap between preclinical cancer models and patient outcomes.

In Saccharomyces cerevisiae , prolonged cellular stress induces some introns to accumulate post-splicing as stable, linear, spliceosome-protected RNAs 1 . These stable introns are defined by having short distances from their branchpoint (BP) sequences to their 3′-splice sites (3′SSs). Stable introns sequester splicing components, thereby reducing splicing activity and affecting cell growth in the stressed conditions. The mechanism by which these normally ephemeral products of pre-mRNA splicing persist cannot be explained by the current understanding of the splicing pathway, which derives primarily from studies of unstressed cells and their extracts 2,3 . Here, we determined the cryo-electron microscopy (cryo-EM) structure of a stable-intron complex purified from saturated-culture conditions. This structure and experimental follow-up show that a B act -like spliceosome protects stable introns from degradation, and that the short BP–3′SS distances of stable introns render this conformation of the spliceosome resistant to remodelling by helicases. Spliceosomes can also assemble onto artificial introns that have the same sequences as authentic stable introns but do not rely on splicing for their biogenesis, which demonstrates that spliceosomes arrive at this B act -like conformation by reassembling onto linear introns after their excision from pre-mRNAs. This reassembly activity is maintained in both stressed and unstressed cells. Thus, most yeast introns compete with pre-mRNAs for access to the splicing machinery, and budding yeast has co-opted this activity to adapt to environmental insults.

Metabolic dysfunction-associated steatotic liver disease (MASLD) is a chronic disease with multiple etiologies, stemming from the interplay between local and systemic genetic, diet, and gene-environment interactions. To understand the progression of MASLD in a controlled setting, we utilized a human liver microphysiological system (MPS) to establish a physiologically relevant metabolic baseline and probe how primary human hepatocytes respond to perturbations in insulin, glucose, and free fatty acids (FFAs). Replicate liver MPS were maintained in media with either 200 pM or 800 pM insulin for up to 3 weeks alone and in combination with standard glucose (5.5 mM), hyperglycemia (11 mM glucose), normal (20 µM) and elevated FFA (100 µM). Together, hyperinsulinemia along with elevated glucose and FFAs, induces the release of pro-inflammatory chemokines, accumulation of triglycerides, and predisposes hepatocytes to insulin resistance. Treatment with the thyroid receptor β agonist resmetirom normalizes hepatic fat content and partially rescues insulin sensitivity, but paradoxically induces higher CXCL1 and IL8 expression in male and female donors. In aggregate, our enhanced in vitro MPS model establishes a metabolic baseline and perturbed condition that recapitulates a spectrum of phenotypes observed in MASLD, offering improved quantification and insight into disease progression with relevance to human physiology.

Understanding how tissues remodel in response to perturbations requires computational tools that can untangle condition-specific changes from the conserved tissue architecture. We present Haruka, a spatially aware contrastive learning framework that identifies salient (condition-specific) and background (shared) spatial domains across tissue slices and experimental conditions. Haruka integrates contrastive variational inference with an auxiliary microenvironment reconstruction task, enabling the model to learn spatial-context-informed embeddings that capture both perturbation effects and local neighborhood context. Through benchmarking on simulated and real datasets, Haruka outperforms state-of-the-art methods in detecting spatially heterogeneous responses. Applied to diverse spatial omics platforms, Haruka distinguished immunotherapy responders in melanoma, traced fibrosis progression in human lung tissue, and mapped treatment-resistant microenvironments in KRASG12D-mutated lung cancer. Thus, Haruka provides a generalizable framework for spatial contrastive analysis, enabling systematic dissection of tissue organization, cellular plasticity, and microenvironmental remodeling across disease, development, and therapeutic response.

Cancer cell fate has been widely ascribed to mutational changes within protein-coding genes associated with tumor suppressors and oncogenes. In contrast, the mechanisms through which the biophysical properties of membrane lipids influence cancer cell survival, dedifferentiation and metastasis have received little scrutiny. Here, we report that cancer cells endowed with high metastatic ability and cancer stem cell-like traits employ ether lipids to maintain low membrane tension and high membrane fluidity. Using genetic approaches and lipid reconstitution assays, we show that these ether lipid-regulated biophysical properties permit non-clathrin-mediated iron endocytosis via CD44, resulting in significant increases in intracellular redox-active iron and enhanced ferroptosis susceptibility. Using a combination of in vitro three-dimensional microvascular network systems and in vivo animal models, we show that loss of ether lipids from plasma membranes also strongly attenuates extravasation, metastatic burden and cancer stemness. These findings illuminate a mechanism whereby ether lipids in carcinoma cells serve as key regulators of malignant progression while conferring a unique vulnerability that can be exploited for therapeutic intervention.

Alzheimer's (AD) and other neurological diseases (NDs) cause substantial health-related and economic burdens, but progress toward preventative or ameliorative treatments has been limited. Genome-wide association studies (GWAS) have identified hundreds of risk loci containing single nucleotide polymorphisms (SNPs) that alter ND risk, but >90% of these SNPs are in noncoding regions, which are cell type-specific and difficult to study. To address this limitation, we have combined iPSC-based cell models and functional genomics to pinpoint the causal genes at 10 loci associated with risk for AD and multiple sclerosis (MS). We generated neural progenitors, excitatory neurons, astrocytes and microglia from iPSCs and performed RNA-seq, H3K27ac ChIP-seq and Promoter Capture Hi-C. We compared these data to those from uncultured primary (ex vivo) human brain cells to assess their physiological relevance, and determined global patterns of ND risk SNP enrichments in cell type-specific enhancers. We transduced iPSC-derived microglia-like cells (iMGLs) with a lentiviral sgRNA library targeting AD and MS risk SNPs and performed scRNA-seq to identify altered genes and cellular pathways. iPSC-derived cell types displayed gene signatures and enhancer patterns consistent with their cell identities, with iMGLs displaying the strongest similarity to primary cells. Neuronal and glial cells from both iPSC and primary brain sources show consistent enrichment of ND risk SNPs in cell type-specific enhancers, with microglia showing enrichment of AD and MS SNPs. Enhancer/promoter interactions identified in iMGLs and primary microglia nominated potential target genes of these SNPs, and these predictions were tested by CRISPRi single cell screening. We identified multiple novel AD and MS risk genes, including the transcription factor MAF, which is regulated by both AD and MS risk SNPs, and ELMO1, which we characterize as a novel MS risk gene involved in microglial homeostasis. The combination of iPSC-based cell models and functional genomics allowed for systematic perturbation of AD and MS risk SNPs, revealing previously uncharacterized risk genes linking neuroimmune pathways to disease pathophysiology. Ongoing studies are exploring the role of altered expression levels of MAF, ELMO1 and other screen hits in microglial phenotypes, including effects on neurons and astrocytes in co-culture and organoid systems.

SUMMARYRare genetic diseases collectively affect millions of individuals. A common target of many rare diseases is the mitochondria, intracellular organelles that originated through endosymbiosis. Eukaryotic cells require related proteins to function both within the mitochondria and in the host cell. By analyzing N-terminal protein isoforms generated through alternative start codon selection, we identify hundreds of differentially localized isoform pairs, including dual-localized isoforms that are essential for both mitochondrial and host cell function. Subsets of dual mitochondria-localized isoforms emerged during early eukaryotic evolution, coinciding with mitochondrial endosymbiosis. Importantly, we identify dozens of rare disease alleles that affect these alternative protein variants with unique molecular and clinical consequences. Alternative start codon selection can bypass pathogenic nonsense and frameshift mutations, thereby selectively eliminating specific isoforms, which we term isoform-selective alleles (ISAs). Together, our findings illuminate the evolutionary and pathological relevance of alternative translation, offering insights into the molecular basis of rare human diseases.