Translation – Details, episodes & analysis

Episode Summary: 

Millions of people die every year from chronic diseases. Traditional drug discovery has failed in identifying solutions to many of these persistent health challenges. Functional genomics is offering a way forward by identifying gene networks and enabling the development of drugs with very specific targets. But, rather than just relying on gene targets within humans, Linda and her company, Fauna Bio, are casting a wider net across the animal kingdom. Extreme adaptation is common across many mammals, giving us an incredible pool of potential targets to go after. Whereas a single heart attack can kill a person, certain animals not only survive 25 heart attacks a year but also go on to thrive, living 2x longer than other mammals their size. By identifying and understanding the gene networks underlying these extreme adaptations, Fauna can identify novel targets across 415 different species, map them to human genes, and develop drugs that exploit our natural protective physiological mechanisms.

About the Guest 

• Linda is the Co-Founder and CTO at Fauna Bio, a biotechnology company leveraging the science of hibernation to improve healthcare for humans. She earned an MPhil in Computational Biology from the University of Cambridge and got her Ph.D. in Genetics and Genomics from Harvard University. She previously held positions at the Broad Institute and Stanford University studying comparative mammalian genomics and human disease genetics.

Key Takeaways 

• Many mammals have evolved complex adaptations that enable them to survive in extreme environments or withstand physiological events that humans cannot.
• At Fauna Bio, Linda Goodman and her team are working to better understand the biological networks that underlie these adaptations, in hopes of developing therapeutics inspired by the adaptations of the animal kingdom.

Impact 

• Drawing on a completely new source of knowledge about the defense mechanisms of living organisms, Fauna Bio goes beyond the limitations of traditional drug development and looks for better, more effective drugs based on natural defense mechanisms.

Company: Fauna Bio

Episode Summary

The expression of genes in our genome to produce proteins and non-coding RNAs, the building blocks of life, is critical to enable life and human biology. So, the ability to predict how much of a gene is expressed based on that gene’s regulatory DNA, or promoter sequence, would help us both understand gene expression, regulation, and evolution, and would also help us design new, synthetic genes for better cell therapies, gene therapies, and other genomic medicines in bioengineering.

However, the process by which gene transcription is regulated is incredibly complex; thus, prediction transcriptional regulation has been an open problem in the field for over half a century. In his work, Eeshit used neural networks to predict the levels of gene expression based on promoter sequences. Then, he reverse engineered the model to design specific sequences that can elicit desired expression levels. Eeshit’s work developing a sequence-to-expression oracle also provided a framework to model and test theories of gene evolution.

About the Guest

• Eeshit earned his double major in CS & Engineering and Biological Sciences & Engineering from the Indian Institute of Technology in Kanpur. 
• During his PhD at MIT, working on Dr. Aviv Regev’s team, he published 4 papers in Nature-family journals, including 2 on the cover and 1 on the cover as first and corresponding author. Eeshit’s work is in Cell, Nature Biotechnology, Nature Medicine, Nature Communications, and beyond.

Key Takeaways

• cis-regulatory elements like promoters interact with transcription factors in the cell to regulate gene expression.
• Variation in cis-regulatory elements drives phenotypic variation and influences organismal fitness.
• Modeling the relationship between promoter sequences and their function – in this case, the expression levels they induce – is important to better understand regulatory evolution and also enable the engineering of regulatory sequences with specific functions with applications across therapeutics and cell-based biomanufacturing.
• By cloning 50 million sequences into a yellow fluorescent protein (YFP) expression vector in S. cerevisiae and measuring the YFP levels they induced, Eeshit generated a rich dataset to map yeast promoter sequence to expression levels.
• Next, Eeshit trained neural network models, including convolutional neural networks and Transformers, to predict expression from sequence with high accuracy.
• Eeshit then “reverse-engineered” these convolutional models to create genetic algorithms that designed sequences which could induce desired expression levels.
• Finally, Eeshit’s sequence-to-expression oracle allowed for the computational evaluation of regulatory evolution across different evolutionary scenarios, including genetic drift, stabilizing selection, and directional selection.

Impact

• Eeshit’s work developing a sequence-to-expression oracle provided a framework to model and test theories of gene evolution.
• This framework can help us both understand gene expression, regulation, and evolution, and design new, synthetic genes for better cell therapies, gene therapies, and other genomic medicines in bioengineering.

Paper: The evolution, evolvability and engineering of gene regulatory DNA 

Episode Summary:

COVID-19 tests have become synonymous with jamming a swab up our nose to find out whether we have an active infection. But as we progress through this pandemic, a test that tells us whether people have antibodies against the virus will be massively important to creating public health initiatives and deciding who to vaccinate next. Unfortunately, these serology tests are exceedingly tedious to perform, inhibiting their widespread use. Realizing this problem, Susanna talks us through how she utilized protein engineering to create a novel serology test that is massively easier and quicker than traditional methods. Importantly, this test can be used in resource low settings to help end the pandemic worldwide.

Episode Notes:

About the Author

• Susanna’s scientist parents and love for the natural world drove her to research biology and chemistry.
• Susanna is most excited about adding new dimensions to biomolecules through bioconjugation to enhance their function.

Key Takeaways

• A serology test is used to see whether a person has antibodies against a specific pathogen.
• Positive serology tests can tell us whether getting the disease led to immunity, whether a vaccine worked, or whether a person is protected from new variants.
• This could be massively useful to help understand who is protected and who to vaccinate next to finally beat the SARS-CoV-2 pandemic.
• Traditional serology tests use hard to scale and overly laborious methods that hinder their adoption, especially in a low resource setting.
• Susanna used protein engineering and leveraged the shape of antibodies to develop an entirely new serology test.
• She engineered protein fusions that when simply mixed with a human sample such as serum or saliva, will generate light if antibodies against COVID-19 are present.
• This much easier test as well as the variety of human samples it can use as inputs make it a much more approachable option and enables its use in low-resource settings.

Translation

• Susanna and her colleagues are working to make this test available for field studies by making the protein easier to ship and making a handheld device that can measure the readout.
• Productizing this test will require more research in how to stabilize the components, incorporate controls, and most importantly, make it high-throughput.
• Susanna hopes to leverage this technology to help us beat the variants of SARS-CoV-2 and eventually rapidly test for other infectious diseases and autoimmunity.

First Author: Susanna Elledge

Paper: Engineering luminescent biosensors for point-of-care SARS-CoV-2 antibody detection

Episode Summary:

Brain machine interfaces untangle the complex web of neurons firing in our brains and relay the underlying meaning to a computer. These devices are being adapted to help patients regain motor control, monitor our mental well being, and may one day even make us more empathetic. State of the art methods to do this have massive trade-offs, either being high resolution yet requiring devices to be embedded in our heads or low resolution but non-invasive. Finding a key middle ground, Sumner uses advances in ultrasound to monitor the brain activity of monkeys performing specific tasks. With this data, he can not only record the brain activity associated with performing the task itself but also the intention of doing it before the subject even has a chance to move.

Episode Notes:

About the Author

• Sumner started his career in mechanical and aerospace engineering, performing research on haptics and mechatronics.
• This developed a love for how humans and computers interact, leading him to earn a PhD developing exoskeleton robots for motor learning and control.
• Through this, he realized that to translate these technologies, we need better methods to get information out of the brain.

Key Takeaways

• Ultrasound technologies are leveraged to monitor brain activity.
• The signal that is generated when these methods “listen” to the brain is extremely complex and entangled, akin to trying to make out a sentence from across a loud stadium.
• Sumner taught monkeys how to perform a task, reading the brain with ultrasound and using machine learning to decode the message.
• With it, they were able to read which way the monkey intended to move, when the movement would occur, which way the monkey actually moved, and whether it would move its hands or eyes.

Translation

• This technology has massive potential to help those suffering from motor impairment and could one day connect us all on a deeper level.
• To get there, the device will need to be optimized to find the best way to maximize signal-to-noise but minimize invasiveness.
• Additionally, advances in miniaturization, wireless connections, lowered cost of goods, and finding the right balance between AI and BMI control are needed to get this extremely new technology into the hands of everyone.

First Author: Sumner Norman

Paper: 

Single-trial decoding of movement intentions using functional ultrasound neuroimaging

Episode Summary: Enzymes that break down other proteins, or proteases, could be used as a powerful therapeutic if they could specifically chew-up disease causing entities. However many proteases are non-specific, breaking any protein in their path, while the specific ones target proteins that would provide no therapeutic benefit. Travis and his colleagues developed a riff on the method known as PANCE that utilizes bacteria and bacterial viruses known as phages to evolve proteins toward a specific goal. With it, he retrains the sequence-specific protease, botulinum neurotoxin, toward new targets and away from its original ones. The novel enzymes Travis generates have the potential to not only stimulate nerve regeneration but also deliver itself to the correct cell types for a whole new type of therapy. 

Episode Notes:

About the Author

• Travis is a postdoc who performed this work in the lab of Professor David Liu at Harvard University. The Liu lab is famous for engineering and evolving proteins that can be utilized as massively impactful tools for overcoming diverse diseases.  
• Travis’s teachers fostered a curiosity that created a passion for chemistry and ultimately led him to engineer new biochemistries. 

Key Takeaways

• Proteases are enzymes that cut up other proteins.
• Proteases can either be non-specific, a nuke obliterating any protein in their path,  or sequence-specific, a heat seeking missile only cutting very specific protein motifs.
• Sequence-specific proteases that target disease causing proteins would make great drugs but therapeutically useful proteases rarely exist in nature.
• Travis focuses on re-engineering the sequence-specific protease known as botulinum neurotoxin so that it cuts an entirely new, therapeutically relevant protein sequence.
• Using a method called PANCE that utilizes bacteria and bacterial viruses (phages), Travis trains botulinum neurotoxin toward cutting a new target and leaving its original target alone.

Translation

• Botulinum neurotoxin has a cutting domain that Travis engineered toward a therapeutically relevant target, and a targeting domain that delivers the protein toward neurons.
• The enzymes generated could be used to cure neural pathologies but the PANCE could also be applied to change which cell type the protease targets, creating a highly programmable therapeutic protease platform.
• The platform has a ton of interest from industry and Travis is continuing to work on it outside of academia so that these proteases make it to the clinic and impact patient lives.

First Author: Travis Blum

Paper: 

Phage-assisted evolution of botulinum neurotoxin proteases with reprogrammed specificity

Episode Summary: 

Novel drugs that boost the immune system to fight cancer have become pharma darlings in the few short years since their approval. These drugs, known as immunotherapies, have so far focused on improving T cell responses and can be used to cure a multitude of different cancer types. Yet more often than not, immunotherapies have no effect on a patient, leaving doctors guessing on whether to prescribe the drug. To find the reason why some people respond while others don’t, Kevin and his team create a huge database of sequences derived from immunotherapy-treated patients. With it, he discovers biomarkers, mutational signatures, and immune profiles that correlate to response with the hopes that one day, these measurements form a diagnostic to ensure we treat the right patients.

Episode Notes:

About the Author

• Kevin is a group leader at University College London and performed this work in the lab of Charles Swanton at the Francis Crick Institute. Dr. Swanton and his group are experts in studying the genome instability and evolution of cancer.
• Kevin started his career as a mathematician but was always driven to apply his skills to improving medicine.

Key Takeaways

• Immunotherapies aim to cure cancer by “taking the breaks off” your immune system, supercharging it to attack tumors.
• Two immunotherapies known as checkpoint inhibitors (CPI), anti-CTLA-4 and anti-PD-1, work by enhancing T cells and have recently become blockbuster drugs for the treatment of multiple different cancer types.
• These immunotherapies don’t work in many patients and medicine has yet to understand why.
• Kevin aggregated DNA and RNA sequencing data across multiple studies to generate a dataset that contained over 1,000 CPI treated patients who did and did not benefit from treatment.
• With this data, Kevin discovers mutational signatures, biomarkers, and immune profiles that correlate to whether a patient will respond to treatment.

Translation

• Kevin finds measurable signatures of a patient’s cancer that could be used to determine whether a patient should receive CPIs.
• This retrospective analysis will need to be validated as a prospective study to determine whether Kevin’s findings actually predict response.
• More tumor data as well as information about the patient’s genetics is being brought in to improve the accuracy of this prediction.
• Collaborations between academics, medical centers, non-profits, and industry partners will enable the findings to make an impact on patient outcomes.

First Author: Kevin Litchfield

Paper: Meta-analysis of tumor- and T cell-intrinsic mechanisms of sensitization to checkpoint inhibition

Episode Summary:

In a single decade, CRISPR has made a dramatic impact on literally every facet of biotechnology. This game-changing system is traditionally programmed to make cuts at very specific parts of the genome, altering the code to cure disease. But a new class of CRISPRs discovered by Leo’s colleagues don’t simply cut DNA -- they integrate entirely new genetic material at targeted locations. With it, Leo generates a new method to perform very specific and highly efficient genome engineering on bacteria and describes the multitude of ways it can generate strains that revolutize commodity molecule synthesis and medicine.

Episode Notes:

About the Author

• Leo is a PhD candidate who performed this work under Professor Sam Sternberg at Columbia University in New York City. Dr. Sternburg and his team are world experts in CRISPR biology having discovered multiple new CRISPR systems, including the function of Cas9 during his time in Professor Jennifer Doudna’s lab.
• Leo was driven to become a synthetic biologist after being exposed to all the ways nature has engineered biology to overcome problems. 

Key Takeaways

• A new class of CRISPRs have been discovered that don't cut DNA but instead integrate new DNA on the genome.
• Leo hijacks this CRISPR’s novel functionality to integrate whatever new DNA he wants into whatever location on a genome he desires. 
• Through the tools of synthetic biology, the system generates extremely targeted integrations at high efficiency in bacterial cells.
• This CRISPR tool allows for integration of huge genetic payloads, iterative integrations, and integration of payloads at multiple locations in a single step, all of which create entirely new options for strain engineering.
• The tool can be applied to multiple bacterial species and has proven utility in engineering the microbiome in situ as well as modifying industrially sought after strains. 

Translation

• Leo demonstrates that the system is highly effective in laboratory settings and can be optimized to overcome new challenges in new bacterial hosts.
• The tool is undergoing further development and optimization to do population scale engineering -- making targeted and useful modifications to bacteria in communities like those seen in our gut or in nature. 
• Further research is needed to move this powerful integration tool into human cells as a novel method to overcome genetic disease and engineer future cell therapies.

First Author: Leo Vo

Paper: 

CRISPR RNA-guided integrases for high-efficiency, multiplexed bacterial genome engineering, Nature Biotechnology, 2020.

Whether it's Multiple Sclerosis, Type 1 Diabetes, Lupus, or Crohn's Disease, autoimmunity is a rapidly growing problem that traditional pharmaceuticals have failed to completely cure. While these diseases have very different symptoms, they all have the same root cause -- the body’s immune system is attacking its own healthy organs. Lurking within ourselves are a group of T cells called regulatory T cells that have the power to suppress immune function. These cells have huge potential to be engineered and utilized as a platform to cure any autoimmune disease. Unfortunately, they easily lose their suppressive abilities and can even exacerbate autoimmunity if handled incorrectly. Looking to stabilize regulatory T cells, Jessica and her colleagues perform a CRISPR screen to map which genes are responsible for maintaining their suppressive function. Using this data, Jessica takes the first step to bring this incredibly powerful cell type to the clinic to help millions of patients suffering from a myriad of diseases.

About the Author

• Jessica performed this work in the lab of Professor Alex Marson at the University of California, San Francisco. The Marson lab is renowned for their work in building and applying synthetic biology tools to understand and improve the therapeutic value of immune cells.
• Jessica is driven to understand and cure autoimmune diseases because her mother, her sister, and her have all been diagnosed with autoimmune diseases.

Key Takeaways

• Regulatory T cells can suppress immune reactions, making them an attractive therapeutic to be used to cure any autoimmune disease.
• These regulatory T cells do not easily maintain their suppressive function, necessitating some engineering to make sure they maintain their therapeutic value.
• With CRISPR, Jessica turned every gene off one-by-one in regulatory T cells to find which genes were involved in maintaining its suppressive function.
• Jessica found a gene, USP22, that when expressed, inhibited regulatory T cell function making it a useful target for both autoimmunity and cancer.

Translation

• While Jessica focused on one of the hits from the screen, there were many more that have massive potential as drug targets or as engineering steps for T cell therapies against autoimmunity.
• Maintaining a stable regulatory T cell is the vital first step to creating a world where all autoimmune diseases are cured using cells.

First Author: Jessica Cortez

Paper: 

CRISPR screen in regulatory T cells reveals modulators of Foxp3

Engineered T cells that hunt and kill blood cancers have recently obtained three landmark FDA approvals, forever changing the way we treat this disease. Even with its massive clinical success, these cells come with life-threatening neurotoxicities. But is neurotoxicity a set feature of using T cell therapies or is our engineering accidentally targeting the brain? Utilizing advances in bioinformatics and the huge sequencing datasets available to science, Kevin uncovers similarities between a cell type in our brain and the cancer we target with engineered cells. Finding this needle in a haystack, Kevin creates a link between how we engineer these cells and the neurotoxicities we see, discovering a potential root cause of the problem and generating a rule for how to engineer around it.

About the Author

• Kevin recently received his PhD from Stanford University in the labs of Professor Howard Chang and Professor Ansuman Satpathy. These labs specialize in uncovering the molecular mechanisms of disease using advanced sequencing modalities.
• Bridging both biology and computer science, Kevin’s background and expertise made him uniquely suited to hunt down the culprit of CAR T cell neurotoxicity.

Key Takeaways

• CAR T cells are excellent at killing blood cancers but are not without side-effects -- they can cause severe neurotoxicities.
• The receptor engineered into CAR T cells was thought to be specific to these blood cancers, ensuring the therapies don't attack healthy tissue.
• Kevin looked at publically available single cell sequencing data to find a small subset of brain cells hiding in plain sight that the CAR T cells could attack. 
• In mice, engineered “blood cancer specific” T cells attack the brain, demonstrating that neurotoxicity is an off-target effect of the therapy, not a byproduct.

Translation

• The finding points to the potential need for different engineered receptors to be used to target these blood cancers.
• As CAR T cells expand to other cancers and malignancies, this process can be run to ensure we engineer cells that minimize the opportunity for damage to healthy tissue.

First Author: Kevin Parker

Paper: 

Single-Cell Analyses Identify Brain Mural Cells Expressing CD19 as Potential Off-Tumor Targets for CAR-T Immunotherapies

Small molecules are a pillar of human health, making up a majority of the drugs we have in our healthcare arsenal. Many of these drugs are obtained by utilizing synthetic chemistry to modify the composition of some small molecule found in nature. Derivatives of tropane alkaloids, for example, alleviate neuromuscular disorders and are derived from a chemical found in nightshade plants. However, sourcing these plants have become exceedingly difficult as climate change, the pandemic, and geopolitics ravage the supply chain. Looking to overcome these challenges, Prashanth recapitualed the biochemical pathway that makes these tropane alkaloids in yeast. In the most complex feat of metabolic engineering to date, Prashanth can make these life-saving drugs in a bioreactor, insulated from the issues that make them expensive and in short-supply.

About the Author

• Prashanth is a graduate student at Stanford University and published this work in the lab of Professor Christina Smolke. Christina and her team are world experts in metabolic engineering and broke multiple records in generating yeast that perform complex biosynthesis.
• Prashanth’s love of science was fostered by his teacher who encouraged him to combine his fascination with biology and his unique perspective on chemistry.

Key Takeaways

• Drugs are often sourced from natural sources like plants that have extremely precarious supply chains.
• The same biosynthetic pathways that makes the drug in plants can be recapitulated in yeast so that the small molecule can be brewed anywhere.
• Moving this biosynthetic pathway from one organism to another is not easy and still requires a ton of novel biology to be discovered in order to succeed.
• Here, Prashant had to hunt for new enzymes, cut-out wasted chemical reactions, and engineer ways to move the molecule and proteins to the specific parts of the cell.

Translation

• Scaling these microbes to make them economically viable first requires maximizing the amount of drug that each yeast can make.
• Directed evolution of useful enzymes, importing new molecular transporters, and optimizing growth conditions will be used to spin-out this microbe.
• The strain will be licensed through Stanford to pharmaceutical companies.

First Author: Prashanth Srinivasan

Paper: 

Biosynthesis of medicinal tropane alkaloids in yeast