BioByte 083: neurolipid atlas in neurodegenerative disease, the role of iron in large-scale cell death, Benchling and Flagship CEOs on AI
Welcome to Decoding Bio’s BioByte: each week our writing collective highlight notable news—from the latest scientific papers to the latest funding rounds—and everything in between. All in one place.
What we read
Blogs
AI-Driven Behavior Change Could Transform Health Care [Sam Altman, Ariana Huffington, Time, July 2024]
90% of USA’s $4.1 trillion annual healthcare spending goes to treating chronic physical and mental-health conditions. While medical care and genetics play a role, behavior change is crucial for both prevention and optimized treatment. Recognizing this, OpenAI and Thrive Global have partnered to fund a new AI-powered health coach app. This app leverages OpenAI’s LLM expertise alongside personalized health data. The app encourages small, manageable daily changes to improve sleep, stress management, social connections, nutrition, and exercise habits, leading to better overall health outcomes.
Flagship's Noubar Afeyan on uncertainty, AI and biotech’s future [Endpoints, July 2024]
Noubar Afeyan discusses key points for Flagship Pioneering’s current and future focus after announcing their $3.6B raise:
Partnerships are a ‘bigger and bigger deal’ - Flagship is focusing on increasing and building on its current industry partnerships (such as with Pfizer, Samsung, Thermo Fisher and Novo Nordisk). Afeyan recognises the increasing importance of such partners to help Flagship efficiently develop and grow its companies. Flagship states the clear benefit it provides to such players such as pharma given its heavy R&D and suite of innovative assets.
Programmable medicines - treatments that can be rapidly personalized for a given target, patient or scenario are becoming a huge focus for the incubator.
What can’t AI do? Flagship is a firm believer in the potential of AI to disrupt biotech and is doubling down on this focus with the launch of initiatives such as Pioneering Intelligence.
Efficiency and uncertainty - Flagship is famous for a unique approach to company creation, where the incubator takes bold ideas with high uncertainty and quickly sets out to resolve this uncertainty via killer experiments. Afeyan states that Flagship will continue to work on such high risk ideas with a focus on becoming more efficient at this process.
Could billions of dollars in AI funding lead to the same number of new drugs? [Ashu Singhal and Sajith Wickramasekara]
Benchling’s founders take to STAT News to raise questions about the value the billions of dollars being poured into the AI-driven biotech or “techbio” companies may or may not bring. They argue that investing heavily in one part of the drug development process is a risk to the rest, that the AI focus cannot just be on discovery, but also needs to be on how the pharma R&D system evolves to meet this increased output. They speak a lot about pharma R&D in particular—defined here as how drugs are developed, tested, approved, and manufactured—and say that it is largely becoming less efficient over time. Large pharma companies spend over $6B on R&D per approved drug, which is a 150x increase from what was required in the 1950s. 85% of that funding is used after discovery. The duo claim that fixing this issue with AI requires rethinking how R&D organizations are designed—it’s not as simple as applying AI and throughput to current processes. They envision labs in the future to incorporate more automation that allows AI-driven experimentation to be run at scale and AI-driven systems to analyze data that is output. They also consider other layers where AI can be useful like clinical trial design and scientific communication. It is very likely that pharma looks entirely different from an organizational standpoint with the incorporation of AI at every layer.
The strategy behind one of the most successful labs in the world [Gebel et al., Nature Comment, June 2024]
Since its inception in the late 1940s, the Medical Research Council’s Laboratory of Molecular Biology (LMB) in Cambridge, UK has procured a dozen Nobel laureates, including James Watson, Francis Crick and Fred Sanger. In the past 15 years, four LMB scientists have received the award: “Venkatraman Ramakrishnan for determining the structure of ribosomes, Michael Levitt for computer models of chemical reactions, Richard Henderson for cryo-electron microscopy (cryo-EM) and Gregory Winter for work on the evolution of antibodies''. Between 2015 and 2019, 36% of the LMB’s publications were in the top 10% of the world’s most-cited papers.
To find out what is the LMB’s secret sauce, the authors interviewed 12 senior scientists and analyzed 60 years’ worth of archival documents from the lab: from publications to meeting minutes to internal management reports.
The main difference is that the lab is organized in a way that increases the likelihood of discoveries. The lab’s approach is to “identify new and important scientific questions in uncrowded fields that require pioneering technologies to answer them. The lab develops that technology to open up the field; continual improvements bring more breakthroughs, which can be scaled up to enter markets”.
The LMB’s management strategy prioritizes three elements:
Culture: encourage scientific diversity, recruitment of groups with diverse but aligned interests enabling economies of scale while retaining flexibility to innovate in emerging fields (e.g. the development of cryo-EM benefited both structural studies division and neurobiology division)
Incentives: Promotes junior members rather than bringing internal talent. Many of its Nobel prizewinners like Greg Winter began their careers at the lab and were promoted internally. Although the LMB is structured in divisions, almost all career scientists have independent but aligned scientific programmes, leading to creative combinations of ideas and sharing of failure and resilience to it. External grant bidding is limited as these have short-term, result-oriented requirements which do not align with LMB’s longer-term ambitions.
Management Oversight: focuses on managing tensions between technology developers and scientists. Management, like experienced PIs, will translate scientific terms into technical engineering requirements and vice versa. Funding for technology development is allocated centrally through the lab’s executive committee.
The LMB focuses on long-term goals rather than short-term incremental gains, and therefore the internal evaluation system for researchers is more concerned with “the potential of the overall scientific programme than with standard individual performance metrics, such as the number of journal publications and citations, personal impact factors, grant funding, awards and collaborations.”
The authors call for the LMB’s strategy to be used as a blueprint for other institutes focusing on fundamental research.
Papers
The Neurolipid Atlas: a lipidomics resource for neurodegenerative diseases uncovers cholesterol as a regulator of astrocyte reactivity impaired by ApoE4 [Feringa et al., BioRxiv, July 2024]
Alzheimer's disease (AD) and other neurodegenerative disorders have been increasingly linked to alterations in brain lipid metabolism. However, detailed understanding of the specific lipid changes associated with different diseases and cell types in the brain has been lacking. To address this, a group of researchers from the Netherlands have developed a new resource called the Neurolid Atlas - a data commons and analysis tool for exploring lipidomics data across various neurodegenerative conditions.
The researchers used a standardized pipeline combining isogenic induced pluripotent stem cell (iPSC) technology and comprehensive lipidomics analysis capable of quantifying over 1000 different lipid species. They generated iPSC-derived neurons, astrocytes and microglia, confirming these cells have distinct lipid profiles resembling in vivo lipid compositions. Focusing on AD, they found cholesterol esters accumulate significantly in post-mortem AD brain tissue. Intriguingly, iPSC-derived astrocytes carrying the AD risk gene variant ApoE4 also showed cholesterol ester accumulation. Through multi-omic analysis, they discovered ApoE4 suppresses immune activation of astrocytes by enhancing cholesterol esterification.
This study provides several key advances for the field. The Neurolipid Atlas offers a powerful new tool for exploring brain lipidomics data across diseases and model systems. The finding that ApoE4 drives cholesterol ester accumulation and suppresses astrocyte immune function provides new mechanistic insights into how this major genetic risk factor may contribute to AD pathogenesis. More broadly, this work establishes altered cholesterol metabolism as a key feature of AD and demonstrates how lipid changes can directly impact cellular functions relevant to disease.
Emergence of large-scale cell death through ferroptotic trigger waves [Co et al., Nature 2024]
Large scale cell death is a common feature across development and disease, which raises questions on how it is possible to coordinate cell death across space and time. Although well established that cell death can trigger the release of various cytotoxic molecules, these are often diffusion limited and reach about 2-3 cells adjacent to the initiating one. Large scale cell death necessitates trigger waves, which are self propagating chemical waves that can travel long distances without compromising on speed or intensity. These have been found across nature, from neuronal firing, mitotic waves in developing frogs, and more.
Ferroptosis is a form of cell death mediated by iron-catalyzed lipid peroxidation, and has been shown to enable coordinated cell death across long distances. To study mechanics of ferroptosis propagation, the authors developed an assay wherein retinal pigment epithelial cells (sensitive to blue light irradiation) were allowed to propagate, and were treated with erastin (a ferroptosis enabler). By using blue light to selectively activate ferroptosis in a defined spot, they found that this propagated many millimeters across cells, with the wave of cell death transmitting at a defined speed that resulted from the kinetics of reactions between iron, oxygen, and glutathione, with diffusion of reactive oxygen species (ROS) as a possible coupling mechanism for ferroptosis trigger waves.
The authors subsequently studied development of chicken limbs; notably, they found high levels of 4-hydroxynonenal (4-HNE), a lipid peroxidation marker and indicator of ferroptosis localized at areas of cell death. Treating developing limbs with a ferroptosis inhibitor notably deranged muscle fiber development, resulting in excessive fibers along with poor anterior-posterior segregation. This data adds to the growing body of evidence that ferroptosis has critical roles in development – at least in chicken limbs. However, the mechanisms of how exactly pro-ferroptotic signals remains unclear. That said, this paper provides significant experimental evidence on how ferroptosis is critical for large scale cell death, and raises the question of how this mechanism can be modulated in the context of disease.
In Case You Missed It
Ginkgo Earnings: Why Biotech Platforms are Hard $DNA
Bioptimus launches H-optimus-0, the world's largest open-source AI foundation model for pathology
Notable Deals
Flagship Pioneering raises $3.6B to support the creation and development of approximately 25 bioplatform companies across human health, sustainability and AI. Since Moderna’s COVID breakthrough, the Boston-based incubator has announced partnerships with players such as Pfizer, Samsung, Thermo Fisher and Novo Nordisk to innovate on their company creation and innovation pipeline.
Eli Lilly has made one of the largest acquisitions of 2024, purchasing Boston-based biotech Morphic Therapeutics for $3.2 billion. While Morphic has a range of experimental therapies in development across autoimmune, cancer, and fibrotic diseases, the key asset of interest for Lilly is MORF-057, an α4β7 integrin inhibitor currently in Phase II clinical trials for ulcerative colitis. Due to the lengthy timelines and high costs of autoimmune clinical trials, smaller biotechs often turn to pharmaceutical giants like Lilly to expedite development and bring promising therapies to market sooner.
Myricx Bio raises $114M to develop NMT inhibitors for ADC payloads with potential to work in tumors resistant to the toxins contained in typical ADCs (e.g. Enhertu, Trodelvy). Investors include Abingworth, Brandon Capital, British Patient Capital, Cancer Research Horizons, Eli Lilly and Company, Novo Holdings, and Sofinnova Partners. The start-up is based in London.
Komo BioSciences launches to develop integrase-like tools for large-scale genomic change. Komo plans to license its technology to other biotechs, rather than develop genetic medicines itself.
Illumina acquires FluentBio for access to single-cell analysis technology that eliminates the need for complex, expensive instrumentation and microfluidic consumables.
Biovance debuts with €51M to lift up biotech scene in Portugal and southern Europe
Courier Health raises $16.5M Series A to build a patient data CRM
What we liked on socials channels
Field Trip
Did we miss anything? Would you like to contribute to Decoding Bio by writing a guest post? Drop us a note here or chat with us on Twitter: @ameekapadia @ketanyerneni @morgancheatham @pablolubroth @patricksmalone