2030’s Biotech, According to Future Nobel Laureates
A reflection on biotech after attending Canada’s Gairdner Gala (trends, core questions & insights)
In October 2022, I had the privilege of attending the Canada Gairdner Award lectures and exclusive gala.
The Gairdner foundation honours biomedical research from around the world. This year’s winners included the inventors of the COVID-19 mRNA vaccine. One-quarter of the winners eventually earn a Nobel Prize. And regardless, all of their work has an expansive impact on medicine and science.
My experience learning from ingenious scientific conversations was like, what I imagine, children feel when they see their cartoon heroes at Disney world. Magic. Unbelievable. Fireworks.
I am thankful to the Morehead-Cain Foundation for funding this trip and to The Knowledge Society for getting me an invite to attend!
The experience was a turning point for me. It clarified the direction of my education: to the cutting edge of biochemical inventions. I’m thankful for the opportunity to get a sneak peek into the real world of science.
I’d like to take my reflections on this experience forward — to the questions and concepts at the brink of scientific disruption.
In this post I’ll highlight some of the big-picture areas of biotech I’m excited about and the molecules and tactics I head through Gairdner’s programming that profoundly piqued my interest! ❤️❤️
Part 1: 3 Lessons from A Big Picture Framework for Biotech
How will we…
Cure cancer? Make an HIV vaccine? Reverse fatal genetic conditions like cystic fibrosis, sickle-cell and Huntington’s? Distribute medical care in low-resource communities? Build new treatments for heart failure, the #1 cause of death in Canada?
We’ll build tools.
Particularly, tools for understanding and tools for adjustment.
- We will want to understand the liver functions of HIV-positive patients in Uganda to adjust their drugs to avoid liver cirrhosis while suppressing the HIV virus as best as possible.
- We will want to understand the microbiome’s interface with tumour microenvironments so we could adjust the microbiome via medication to reduce cancer proliferation or migration.
- We will want to understand the genes and proteins involved in autoimmune diseases so we can adjust their autoimmune qualities.
Tools can be vaccines, treatments, imaging techniques, experts, systems, small molecules, etc… ultimately, each tool will be something that accomplishes some goal of either understanding or adjustment.
Biological Sciences + Technology = BioTech 🧬🦾
We’ll combine biological understanding with our tools to create biotechnology; these biotechnologies will spread to hospitals, health centres, and ultimately into people’s hands (or livers or arms or hearts… wherever they need to go).
Important definition: biotech is the application of science (technology) in the realm of biology.
These tools of biotech rely on four components:
- Someone’s hypothesis
- Biological knowledge (which is governed by chemistry, physics and math)
- Turning the knowledge into a reproducible tool which we call technology
- Safely scaling the technology
And what this will all look like… well, let’s turn to the people posing the hypotheses. Particularly, the frontier scientists that have and will receive Nobel Prizes.
What Prized Scientists Have to Say
Nobel prize-winning work has changed the world:
- The discovery of penicillin (the first antibiotic used on humans)
- The foundational surgical work on blood vessels and organ transplantation
- The existence of insulin (which opened up treatments for diabetics)
- Method for genome editing (i.e., CRISPR-Cas 9)
- Proof that x-rays cause genetic mutations (which is why pregnant people don’t / rarely get x-rays)
The Nobel Prize brings light to important scientific and societal contributions, but they do so retroactively. Often, prizes come years or decades after discovery and they’ve already made a dent in the world.
How then can we predict/think about biotech in the next 10–20 years?
Probable Nobel Prize winners!
The Canada Gairdner Awards recognize the world’s most creative and accomplished biomedical scientists from around the world. 25% of the Canada Gairdner Laureates go on to win Nobel Prizes.
Having the opportunity to learn from the 2022 Gairdner laureates, I got a little sneak into their ways of thinking, ideating and building. Their research is the foundation for biotech
7 Amazing Scientists; a million new ideas:
It genuinely felt like a million new synapses clicked in my brain! The raw excitement and scholarship in the Gairdner room was unlike anything I’d experienced before. It was over 15 hours of programming dedicated to the bleeding edge of biotechnology. Wow, it was just a dream.
In short, here are the 2022 Canada Gairdner Winners:
Dr.Katalin Karikó, Dr.Drew Weissman and Dr.Pieter Cullis: for creating the base technologies that enabled the COVID-19 Vaccine.
Dr. John Dick: for discovering leukemic stem cells and the idea of hierarchies of cells within cancer.
Dr. Stuart Orkin: for researching the molecular mechanisms behind blood disorders and applying that research to develop cures for diseases like sickle-cell anemia
Dr. Zulfiqar Bhutta: for developing and evaluating evidence-based interventions for maternal and child health in marginalized communities.
Dr. Deborah Cook: for pioneering and informing global critical care practices.
And it doesn’t stop there:
Even more stellar scientists have inspired me recently, in particular:
Dr.Mina Bissel: for characterizing the “dynamic reciprocity” between the inside and outside of cells, and how the outside of a cell could change gene expression, and how gene expression can change the outside of the cell.
Dr.Anna Blakney: for working on next-generation gene-delivery technologies and studying their interface with the immune system!
Dr. Carolyn Bertozzi: 2022 Chemistry Nobel Prize winner for discovering bioorthogonal chemistry; types of reactions that could be used to study biological systems, without chemically altering them. (ps. Her Nobel Prize lecture is live)
Not only are these 10 people awesome starting points for a deep dive, but their work is (and will be) incredibly relevant for the future of biotech.
Lesson #1: Specificity is the Start of Something Broad
How do you build a house? You start by putting down a brick.
How do you build a brick? Not by slicing out a piece of a house. You start by mixing ground clay with water.
Great and vast systems, at some point, started as a building block. To build something up, we have to have the building blocks… and better yet, we should understand them and their chemistry.
This is why scientists start with sharp, specific and measurable questions. They help them understand the building blocks, the legos, that can build into something impressive.
If you’ve built any lego castles you’d know the process is more dynamic than “knowing the pieces” and putting them together.
We may start by clumping some pieces together, and then pieces those clumps into even bigger clusters. Along the way, we may make a terrible miscalculation and have to redo an entire section. Then we try to imagine the bigger picture, perhaps even peek at the instructions, and we could hypothesize that the cat ate one of our pieces and that’s why it’s all falling apart.
Start with the lego pieces.
You’re about to see a ‘big picture’ summary of my biotech learnings.
I wrote this section well after I combed through pages in my notebook, memories in my hippocampus, neocortex and amygdala and research pertaining to the prize topics. I started with specificity and derived the ‘big picture.’
I think this is how science ought to be. As far as I know, there are no practical big-picture blueprints, other than dreams, that detail how to get to scientific breakthroughs. We can only start by hunting for the legos.
Lesson #2: Trends in Biotech are Interconnected and Complementary
Before the event, I went through a stack of scientific research from the 1990s — Present day. It was helpful to contextualize the production of biotechnology (e.g. the mRNA Pfizer vaccine to treat SARs-CoV-2) along a timeline of three stages:
- Identification. This is the scientific inquiry that reveals the truth. We get to know our ‘lego pieces.’
- Leverage. How can we combine pieces? What pieces of information can they connect? We create ‘lego structures.’
- Optimize. How can we create treatments for humans, increase the yield or use our new ‘lego structures’ in the context of the world?
It’s a simple way of classifying the evolution of science.
Since science is the pursuit of truth, the macro look at this framework is much less simple. One lego piece can go on to optimize another set, it can contribute to another’s leverage or identification. The truth, within science, works in a collaborative web:
Order is hard to define.
Even just reading papers from the seven laureates, I could feel their web of collaboration. Inspiration from the same source. Citations. Similar implications. It was almost claustrophobically intertangled.
Lesson #3: Biotech is Philosophical
There were four major themes for biomedical research.
Toxicity: too much of a good protein can be toxic. Too little of a good protein can be toxic.
There are two toxicity classes: tolerable and intolerable substances.
Tolerable substances have a ‘happy medium’ within a system, but too much or too little can be a bad, toxic, thing. Intolerable substances have no happy medium within a system.
Toxicity is crucial to biotechnology for two reasons.
- We don’t want to put toxic treatments into our bodies
- The development of therapeutics is dependent on science which is dependent on technology. Technology helps us sense what goes on in a system. Our technology cannot be toxic, whether it be a molecule we indirectly measure or a state-of-the-art microscope. If we poison the body we are studying we cannot see the glorious positive outcomes of an intervention.
Greyness in Definition: like what we saw in toxicity, a protein may not be ‘good’ or ‘bad’ inherently; it can depend on its dosage. Or its environment.
There is a greyness to the bounds of labels within science. It’s quite a philosophical undertaking to examine the inherent definitions of cells and their morality. But this is important. We must understand the dynamic states of complex systems and what, at a moment’s notice, can cause them to become “bad.”
For example: when does a cell go from being ‘good’ to ‘cancerous?’ — what is the line? It’s grey. It’s multivariate.
In-vivo: we can minimize the equipment needed to carry out biological therapeutics by using the most advanced biological machinery: mother nature’s human bodies.
Many industries, like animal mining and raw materials extraction, will move towards in-vitro (outside the living system) production. This is more of a comment on human independence and our ability to make very specific, replicable products at scale.
Making a hamburger and treating an individual’s cancer are two totally different problems. We can make a million of the same hamburger. But we cannot treat a million cancers the same way.
For the sake of personalized effectiveness and eventually cost, in-vivo, in medical contexts, is a word to pay attention to.
Targeting: in this vein, the next era of medicine and biotech is personalized. Inherent to personalization is targeting. To target something, we must know what we are looking for. Otherwise, what will we target?
Part 2: Piqued Interests + Biotech Case Studies 🤓
1961’s Unstable Intermediate
Dr. Katilin Karikó started her Gairdner talk by introducing the “unstable intermediate” discovered in 1961. mRNA. The basis for the COVID-19 Pfizer-BioNTech and Moderna vaccines.
Dr. Karikó’s experiments, decades ago, made it possible to build an effective COVID-19 vaccination in a perceivably short period of time. And while mRNA vaccines hold incredible promise for other hard-to-vaccinate diseases (like HIV), the significance of mRNA holds implications beyond disease prevention.
The title of the 1961 paper by Brenner, Jacob and Meselson was: “An Unstable Intermediate Carrying Information from Genes to Ribosomes for Protein Synthesis.”
mRNA’s debut adjectives were “unstable” and “intermediate” !
Let’s leave its instability for a moment and focus on mRNA’s role as an intermediate. And intermediate in what?
A few years prior, in 1957, Francis Crick proposed an idea we’d go on to call the central dogma of biology. This paper explores the history more, but I’ll summarize below:
Information flows in only one direction: nucleic acid (like DNA or RNA) to protein. Not the other way around. DNA and RNA can replicate themselves, and they can interchange (i.e. DNA makes RNA and RNA makes DNA), but proteins cannot make nucleic acids.
See below for Crick’s original 1956 note:
Here is the crucial bit: DNA’s genes are turned into RNA which is then used as a script for protein synthesis.
In particular, we call the RNA that interacts between DNA and the final protein messenger RNA, or mRNA. It’s the intermediate.
This unlocks a medical gem: the body has the machinery to produce proteins, what if we fed it mRNA or DNA to direct what it produces?
Now to the second adjective: unstable. RNA molecules are very short-lived… one study found they lived an average of two minutes. Understandably, this makes them hard to study!
But why might nature produce such a short-lived molecule? Well, let’s look at why DNA produces RNA.
Each cell in the body has the same DNA. What then differentiates a heart cell from a liver cell? The proteins they express!
Proteins in the cell do several things: they may be messengers (signalling other for other cells like immune cells to come over), they may be receptors (resting on the outside of cells waiting for some signal), they may be structural proteins (say, helping the heart contract)…
A cell doesn’t produce its proteins all at once. Protein synthesis is a dynamic process that responds to the needs of the cell! So, having an unstable intermediate like mRNA makes it easier to regulate what proteins are produced. Since mRNA doesn’t stick around for very long, the cell doesn’t have to worry about excess protein being produced. Instead, it can match its mRNA to the amount and types of proteins it needs!
Also, DNA is very long-lived! The body already has a long-term way of storing information… mRNA fills a different niche.
Recap: Mother Nature is awesome. She built a clever design for protein synthesis where information is stored long-term, but the usage of information can be regulated in the short term. 🔑 We can leverage Mother Nature’s design for modern medicine: use the body’s protein synthesis machinery; adjust the DNA and RNA ‘instruction manuals.’
Vaccines Specifically: vaccines are substances used to train our body’s immunity.
We start by exposing the body to something that resembles the pathogen, without it being the active pathogen itself.
Our body then develops immune cells that recognize parts of the pathogen. These immune cells then make memory cells. These memory cells stick around to support your body for the “real deal” pathogen if it does come along.
✨ An analogy:
Imagine you are from Hawai’i and never left the island. At age 18, you get into a university in Boston, and you decide to visit for a week during your March break.
You get to Boston, and you realize it is extremely cold!! You go to the store to buy some warm clothes. You spend the week in Boston and decide to accept the school's offer.
You return to Hawai’i and place your warm clothes at the very back of your closet. In September, you return to Boston, with the clothes from the back of your closet and are much more prepared for the weather.
You use your warm clothes as a starting point for your wardrobe, but you eventually expand it so you have more options.
Vaccines are the week-long trip where you learn something (i.e., that Boston is cold) and that helps you prepare yourself for the “real deal” (i.e., moving to cold cold Boston).
Take this analogy and replace “clothes for warmth” with “antibodies.” Getting a vaccine looks something like the graph below:
Where does mRNA come in? It could be used to design elements of the pathogen so your body knows to recognize these elements! For instance, the COVID-19 vaccine used the Corona Virus’s spike protein. Traditionally, we have used deactivated viruses for vaccines, but mRNA allows us to focus on specific elements of the virus. This makes producing vaccines much faster and easier at scale!
mRNA, the “unstable intermediate”, gets degraded in the body after 8 hours!
The Third Generation of Therapeutics
First, there were small molecule drugs (think advil) then biologics (think insulin) and now, the next generation of therapeutics are nucleic acids.
Nucleic acids are really a platform that can give rise to therapies for diseases, rare or common. They’re like the internet: enabling businesses from all sorts of sectors.
But how do we go from the 1961 unstable intermediate to an emergency response pandemic vaccine? I want to bring it back to Dr. Karikó’s talk.
Skipping along to the 1990s, our scientific understanding of RNA really matured. We discovered that mRNA had certain modifications in-vivo (inside the body) that helped it from being degraded and scientists performed experiments that provided mRNA could have anti-viral effects.
But the research was limited by one thing: mRNA is an unstable intermediate. The implication? mRNA was unstable, easily degraded, and immunogenic with low yields of the desired protein. None of these adjectives seemed promising for therapeutics… especially since there were other things to study at the time. There are always other things to study.
“People are usually impatient, so many left the field… the mRNA studies were not followed up upon” — Dr. Karikó
Reducing the immune response of synthesized mRNA
Dr.Karikó worked in her lab to create RNA. After doing so, she compared her RNA to mammalian RNA, specifically looking at an inflammation marker.
Her RNA caused the most inflammatory response but tRNA (a type of mammalian RNA called transfer RNA) stimulated almost none!
There was already knowledge that tRNA had many modifications. The logic followed that if tRNA was modified, mRNA has some modifications too. In 2005 she published a study discussing the impact of certain modifications and how they affected immune response. The implication: we must modify synthetic RNA so that our body does not see it as a foreign pathogen it wants to attack. In the case of vaccines, we want our body to attack the protein that the mRNA produces, not the mRNA itself.
There are nearly 100 modifications possible for RNA (read this paper for the details, note: it focuses on tRNA although many RNAs share modifications). What is important is this: it took years to optimize mRNA to improve its stability and yield and to reduce its immunogenic effects. And that’s what Dr.Karikó did. The graphs that she shared in her presentation are proof.
The modification of RNA opens up a lot of interesting discussions. But I’ll focus on one thing: the blue greek letter psi (Ψ) above the right graph. Ψ represents pseudouridine “the most abundant modified nucleoside in RNA.” In Karikó’s 2005 paper, she states that pseudouridine and other uridine modifications uniquely reduce immune response.
Pseudouridine is a slightly altered arrangement of uridine (play spot the difference below):
This is just one of 100+ possible modifications that we know of for RNA. Hopefully, you can start to appreciate the engineering it takes to optimize these molecules. It’s a process we still haven’t perfected!
Now, a lesser-known fact on mRNA vaccines is that the first one to hit trials was for the Zika virus in 2017! That’s a pretty amazing feat if you ask me. With mRNA vaccines, we create a platform to design many many many more vaccines because the fundamental “format” of the vaccine is similar, we just need to change the sequence of the mRNA. With conventional vaccines, we need to change the deactivated virus.
Hopefully, this helps illustrate how the mRNA COVID vaccines are not a new drug, but an application of 30 years of mRNA research.
This story of the creation of stable mRNA brings up a crucial idea. Engineering from attributes. For instance, viral mRNA can be used to study modifications that allow foreign RNA to survive the body’s immune system. Viruses, although often the thing we are trying to fight, have attributes we can learn from… for instance to create mRNA to fight the coronavirus!
It almost introduces this paradox in our philosophy. We learn the tactics to evade the immune system (e.g., producing synthetic mRNA that can survive it) in order to enhance the immune system! ‘Evil’ can teach us skills to bring onto ‘good.’
mRNA Vaccines
In the first minute of Dr. Drew Weissman’s lecture, he makes it clear what mRNA vaccines are and why they’re important:
🔑It’s an infrastructural technology: “If you want to change the vaccine target, you just change the sequence design and everything else stays identical”
🔑It’s quick to produce: [Referring to the in-vitro transcription]: “that’s a two-hour reaction”
Cancer and Stem Cells and How They’re Not So Far Apart
In the news, you might hear “stem cell therapy” or “stem cell cure.” While stem cell research has gathered some ethical publicity, for the most part, it’s a hopeful groundbreaking medicine for many diseases.
Cancer on the other hand is a 6-letter word that no one wants to hear.
Simply: cancer = bad, stem cells = good.
However… they’re not exactly on opposite ends of the field. They’re actually quite connected.
For starters, cancer cells and stem cells share a lot of similar properties. Both cancer cells and stem cells express an enzyme called telomerase, whereas ‘regular’ body cells, called somatic cells, do not. Telomerase keeps the ends of DNA alive by adding to the “telomeres” (which are like the protective cap at the end of shoelaces)… by doing so, cancer cells and stem cells have large reproductive potential.
Large reproductive potential is good for stem cells because they can regenerate damaged cells or you know… grow babies. But this attribute that is magical in many ways is also detrimental in the case of cells with large reproductive potential growing tumours.
Here’s where it gets extra fuzzy: there are cancer stem cells too.
It makes sense that cancer stem cells exist! Stem cells have hallmark properties that explain cancer’s behaviour: long self-renewal and the ability to differentiate into other cells.
This means that furthering our understanding of stem cells can further our understanding of cancer’s survival tactics.
For instance, WNT signalling is a pathway common to embryonic development and tissue regeneration (i.e., the territory of normal stem cells) and cancer.
Also, ATP-binding cassettes (ABC) transporters are a class of integral membrane proteins. They are like the toll gate at a border that lets cars through for a fee and after inspection… but for molecules and substances into the cell. Common to normal and cancer stem cells, studies have found them to contribute to drug resistance in cancer treatment.
Stem cells can give rise to different cell types (that is how you go from an embryo to a full human), and while great for babies, it’s not great for tumours.
🔑 “Although much focus has been placed on the genetic and biochemical mechanisms that cause drug resistance, there is increasing awareness that tumour heterogeneity contributes to therapy failure and eventually disease relapse” (source)
What’s the source of these cancer stem cells? Mutations of normal stem cells or more specialized descendants of stem cells, progenitors (source).
One of the characteristics of stem cells is that they have strict control of their population size, whereas cancer stem cells do not. It’s a paradox: stem cells have to loosen their strict control to become cancer stem cells.
Cancer’s like a bee colony
Dr. John Dick’s research in acute myeloid leukemia links cancer stem cells as the origin of relapse in patients who at some point achieved complete remission (👉paper).
It’s important to first break down the r-words (I paraphrased from this source):
*Relapse (a.k.a. recurrence) = when cancer comes back after a period of being gone.
*Remission can be partial or ‘complete.’
- Partial remission = reduced tumour size or less cancer (for non-solid tumour cancers, like cancers of the blood) for at least one month
- Complete remission = indetectable cancer given the tests, exams and scans available. This is distinct from cancer-free. Laboratory tests are limited by their limit of detection; i.e., they can only zoom in up to a certain point. In complete remission, given our tools and technologies, there is no cancer detectable for at least one month.
So relapse occurs when small amounts of cancer, indetectable from our tests, grow to a point where we can detect them. That’s how a patient can go from ‘complete remission’ (what can be mistaken as cancer-free) to relapse (detectable cancer again).
There are two key points here:
🔑We’re limited by the eyesight of our technology. Technologies allow us to see beyond our eyes (like internal imagining), but they do not see infinitely small. They have a limit of detection which determines how small they can detect. These devices also have ranges of values for which they’re quantitatively accurate (for instance, a range of tumour sizes where the imaging could accurately predict tumour size). Sharpening these values sharpens our understanding as scientists/physicians.
For qualitative decisions (i.e., is something present, yes or no) only the limit of detection matters. A lower limit of detection = we can detect the presence of our thing of interest sooner.
But it isn’t perfect…
🔑Relapse can happen because undetectable cancer cells proliferate again. But not all cells are created equal! Dr. John Dick was the first to discover leukemia stem cells (LSCs) which established the idea that cancer cells are organized in a cellular hierarchy. His work showed that LSCs can survive typical treatments and thus cause relapse up to a decade later.
🐝How’s it like a bee colony? Bee colonies have many worker bees (cancer cells) but only one queen bee (cancer stem cell) that controls the kingdom.
One of Dr.Dick’s papers (source) states a few remarkable things about the origins of cancer and why it matters:
- “There is overwhelming evidence that virtually all cancers are clonal and [originate] from a single cell” | insight: it only takes one cell, a LSC, for relapse
- “Multiple genetically distinct subclones co-exist with the dominant clone” | insight: while the dominant clone is the origin of the ‘first’ cancer, any of the clones could cause relapse. (Their study found that 50% of relapses were not from the dominant clone)
- “A clear understanding of the genomic landscape of tumours is required in order to devise targeting strategies that eliminate not only the dominant clone but also the subclonal reservoirs from which recurrence can arise” | insight: we must know the landscape of the disease before we build more targeted treatments/diagnostic tools; our tools are limited by our status quo understanding of cancer biomarkers.
- “The very first steps in cancer development remain poorly defined” | insight: there’s an opportunity to uncover the possibly igniting mechanisms behind cancer!
From Bench to Biomarker
The research behind LSCs led to a 17-gene ‘stemness score’, which assesses the relapse risk in acute myeloid leukaemia patients (one type of leukaemia).
This work resembles an essential pipeline in biotechnology: insight > biomarker > predictive or diagnostic tool. It all relies on that fundamental insight, but from that, we can measure what biological markers are predictive of that insight’s outcome (in this case, relapse of leukemia) and finally, that can become a tool in a clinical setting.
Stem Cells as Teachers
One of my favourite parts of Dr.Dick’s lecture was hearing about all the unknowns of stem cells… because that’s where future discoveries will happen 😋.
He brought up several interesting angles from which we ought to understand stem cells: epigenetics, exit regulation, metabolism, how increased autophagy is linked to stemness and stem cells’ increased ability to survive stress.
While these are all fertile research grounds, I want to highlight two key points:
🔑Latency is a hallmark of stemness. Stem cells need to be thoughtful with when, where and into what they replicate into; they have great potential to build things in the body, but with that power comes responsibility.
“It is critical for stem cells to tightly control this balance between the two different modes of division, both during development and adulthood, because failure in maintaining cellular homeostasis may lead to incomplete tissue or organ development, whereas uncontrolled proliferation can lead to tumorigenesis” (source)
🔑Sensors, cancer and high-growth environments: stem cells' latency is partly influenced by their ability to be great sensors of their environment. They have to know what is happening in order to utilize their powers responsibly. Maybe they could teach us about how to be sensors or what to sense?
I also brought up the line between stem cells and cancer stem cells; many of the qualities of stem cells could teach us about the qualities of tumours.
Finally, the ‘fetal environment’ and its fertility for stem cells and starting up life is something we can apply to tissue engineering and derivative technologies like cellular agriculture. Stem cells can be our teachers ❣️.
Finding the needles in the haystacks, rocks in the vacuum of space, and shells in beachy sands.
Dr. Stewart Orkin lectured on his four decades of work on genetic disorders. His talk emphasized that genetic insights (e.g., what genetic mutations lead to a fatal condition) are like finding a needle in a haystack.
Human DNA, if unravelled, could make 150,000 trips around the moon (i.e., 108,000,000,000 km). That is a lot of genetic material to understand!
Additionally, only 2% of our DNA actually codes for proteins. These proteins go on to serve our bodies, whether they’re our eye colour, a cell membrane receptor or proteins in your sweat to provide immune protection, etc.
The rest of the DNA is considered “non-coding,” meaning that it doesn’t directly translate to amino acids, the building blocks of proteins. However, this 98% has some regulatory roles, and potentially other roles yet to be discovered.
We can think about coding and non-coding DNA like a celebrity and their supporting people. Sure, a celebrity is an individual, but they were raised by someone, had teachers nurture their talent, an agent or agency, a camera crew, a makeup artist, a plane crew to support their flight to their concert, a stage crew at the concert, fans that purchase tickets, a life partner or kids to go home to, and so forth. Yes, they are a celebrity (coding region), but many more non-celebrities are enabling them (non-coding region) behind the scenes.
Traditionally, genetic knowledge discovery was focused on the coding region. We even called the non-coding region “junk DNA” because its function was not clear, and therefore, we thought it was useless. We’ve since realized our error and found that non-coding DNA serves the body… but there is much more to be uncovered.
The other thing we must remember for our journey of finding genetic insights is that the DNA of any two people on earth is 99.6% similar. That does leave 12 million “base pairs” (letters within our genetic material, A, G, C or T) of difference.
These 3 points: the vast length of DNA, its majority of non-coding regions and strong similarity between humans highlight some of the major reasons hunting for genetic insights is hard.
There is so much data to go through, and so much of that data is similar; computers, machine learning and automation will inevitably need to be part of our quest for genetic insights.
As for the non-coding regions, they’ve taught us two important things:
- DNA has sophisticated methods for regulating and protecting protein expression; we cannot just study the existence or types of proteins expressed from DNA. These non-coding regions are not “junk”. They might hold some of the “golden” insights of genetics.
- The conventional wisdom of DNA as a template for proteins is too simplistic. DNA holds much more functionality than an instruction manual for proteins; we must invite more complexity into our definition of DNA. It’s time to move forward.
It’s Fertile Ground for Genetic “Needles”
Dr.Orkin’s research was integral to discovering that the protein BCL11A is responsible for switching from fetal to adult hemoglobin. (The failure for this switch to occur is what happens in patients with sickle-cell anemia).
Knowing the gene that codes the protein essential for proper function is game-changing for therapeutics. If we don’t know, what information are we going to use to build the therapeutic?
🔑 genetic needles in the haystack are the basis for innovative therapies.
Closer Look Into Biomarkers
TL;DR: biomarkers are essential for targeting and targeting is split into 3 main categories:
Biomarkers are “signals’’ we can use to identify the state of a biological system.
Some examples:
- The gene sequence of the coronavirus spike protein
- A genetic mutation indicative of Huntington’s disease
- The hCG hormone in urine determines if a person is pregnant.
Biomarkers can indicate if someone is “normal,” “diseased,” or “responding to an intervention.” They can be used to predict, diagnose or monitor the normal or diseased states of an individual.
They speak a language using nucleotides, proteins, hormones, radiation, etc as their alphabet.
The FDA defines four different classes of biomarkers:
- Molecular (usually measured through biological samples)
- Histological (tissues and cells under a microscope)
- Radiographic
- Physiological (measures of bodily processes like heart rate, blood pressure, menstruation, and frequency of migraines)
We usually use biomarkers in the context of diagnosis or risk analysis. For example, what’s the risk of someone getting breast cancer given their family history, hormone levels and genetics?
The Gairdner lectures opened my eyes to a less-obvious application of biomarkers. It starts with a single word: targeting.
As Dr. Drew Weissman put it: “if you’ve worked in this field, you know the critical difficulty is targeting.”
Targeting is an important verb for therapeutics. We aspire to target a disease or problem area.
However, the quality of our targeting may have unintended effects. E.g., chemotherapy: it targets cancer cells but also kills healthy cells.
🔑 Precision medicine is no secret. It’s essential to the future and current generation of biotechnologies. But how do we get precision?
We need to understand our targets (and biomarkers help us do this) 🎯.
There are three buckets for biomarkers and targeting:
- Contextualizing the playing field
- Molecular labelling for more precise therapeutics
- Measuring progress
1: Contextualizing the Playing Field (Targeting as an Adjective)
The “playing field” is the thing we are working on. Maybe it’s a diseased organ. Maybe it is the entire immune system. Whatever it is, we know it is not made up of the same material (unless we are dealing with a singular element in the periodic table… but that’s not biotech). Each system has its own diverse landscape.
Now imagine we are playing darts. On the board, there are areas worth more points than others. The ‘playing field’ is the board. Imagine we have two options:
- Use a dart board painted in the same colour. I.e., we don’t know what areas are worth more points, we know what the board is. We can throw aimlessly.
- A dart board where different points are labelled in different colours.
Uhhh, probably #2! If our objective is to be effective, we want to know where on a board (or in a ‘playing field’) we should divert our resources to.
📍 Targetting as an adjective: use biomarkers to help label a more targeted picture of what we are working with.
^ For example, use florescent pigments to label different parts of cells (e.g., nucleus, cell membrane and mitochondria). This type of targeting is descriptive.
2: Molecular Labelling For More Precise Therapeutics (Targeting as a Verb)
Ok, now that we have a board to play with, we want to get our darts to the parts of the board worth the most points!
Now targeting becomes a verb! We want to make something happen to a targeted subsection of our sample. We want the bull’s eye!
To appreciate this concept, I want to loop in Dr.Pieter Cullis’s work on lipid nanoparticles (LNPs). LNPs are the delivery vessel that enabled the COVID-19 vaccines, and his research, which he calls ’50 years of lipids,’ made LNPs a therapeutic reality!
These LNPs are the delivery system for the mRNA element of the vaccine, like how the mailman’s car (LNPs) is used to deliver you a package (mRNA).
First, let’s understand the base of the LNPs that enabled the COVID-19 vaccine, then I’ll go into how we can build on top of these principles.
🧪 Case Study: Lipid Nanoparticles (LNPs)
Lipid nanoparticles are a type of very small molecule (1 to 100 nanometres in size; 1 nm = 0.000000001 m). They’re delivery mechanisms for things like the COVID-19 vaccine.
LNPs can be made from a variety of different kinds of lipids. The precise definition of a lipid is less important than the big point: many varieties exist. With so many options, we’ve created the breeding grounds for a huge optimization question.
But why do we need delivery mechanisms for things like vaccines in the first place? Our immune system is designed to attack foreign invaders. Despite our rosy intentions, the therapeutics we design are ‘foreign invaders’ (at least according to your body).
Placing a nucleic acid (like mRNA) directly into the body is a great way to activate the immune system. Nucleic acids are negatively charged, and our bodies see too much charge as a toxic/bad thing.
Delivery vesicles (e.g., LNPs) help us evade immune system triggering. It’s like wearing a disguise.
We have to understand our enemy to evade them (i.e., the immune system that would prevent the entrance of our helpful therapeutic).
Lipid Nanoparticles have four core components that do this job^:
- They’re ionizable. At a range of non-physiological pHs, they are positively charged. In the lab, we use positive lipids to attract negative nucleic acids. When they enter the pH of our bodies (7.4), they’re neutral. This allows them to unite with the therapeutic but make it through the body without sounding off alarms. Permanent positively charged LNPs are toxic!
- We use Polyethylene glycol (PEG)-functionalized lipids for stability.
- Cholesterol also provides stability by monitoring the membrane of the LNP.
- Structural lipids, like phospholipids (which make up the membranes in your body!).
Simplistically, these elements form a shell around the nucleic acid (and they could be used to deliver other things, like cancer therapies).
🏃♀️Running the Race
The journey of an LNP from the lab bench to the targeted area is not like a marathon.
It’s more of an ultra-complex triathlon.
It will have to go through different terrains that require different skills and equipment. At a high level, these terrains are:
- Engineered: the environment where it is created
- Body: its journey through the body to the target cell (it needs to survive the immune system!)
- Cell: the LNP has to enter the highly protected membrane of the cell.
- Released: once in the cell, the LNP has to release the nucleic acid.
One of the breakthroughs in LNP research was building ionizability into the system. The LNP system could start positively charged in the engineering phase to associate with the negatively charged nucleic acid, neutral through the body and entering the cell, and then charged inside the cell to release its contents.
🔑we have to optimize the LNP for the obstacle course from lab to the cytoplasm.
“However, enhancing the ability of nanocarriers to target specific cells in the body remains the most difficult challenge” (source)
The three most common techniques for targeting:
- Altering the LNP’s composition of lipids (recall: cholesterol, PEG, ionizable lipids and structural lipids)
- Antibodies (Y-shaped protein that can recognize ‘something’ and that ‘something’ is what we call an antigen)
- Proteins
Right now the most-studied targets for LNPs are immune cells, liver cells and lung cells. However, the spleen, tumours, eyes and kidneys are emerging as areas of interest too!
The Path of Accumulation
With targeting, we are interested in how many of our LNPs end up at our designated site. Following intramuscular injection, one study found this order of accumulation:
It makes sense that most accumulation happens near the beginning of the injection.
🔑This illustrates an important idea: improving accumulation at one site also means impairing accumulation at preceding sites. For example, liver uptake is a major barrier to spleen uptake (also stated in this paper).
This liver (hepatic) blockage is a barrier in targetted delivery: “Nucleic acid drugs have shown promise in oncology, however, their use is limited by hepatic clearance… a large part of the intravenously-administered dose does not reach the therapeutic target.”
Some LNP Tweeking Strategies:
- Charge: “The LNP surface charge has been shown to influence the liver accumulation of intramuscularly administered LNP, with negatively charged LNP having greater liver uptake” (source)
- Oxidizing cholesterol
- Changing lipid composition: e.g., including constrained lipids was found to target a different type of liver cell than usual (source)
- Nanoprimers: they reduce the accumulation of LNPs in the liver by pre-priming the body and temporarily clearing the liver cells (source)
- mRNA engineering: adding suppressive microRNA as part of the mRNA inside the LNP to reduce expression (source)
- Size: larger LNPs, for instance, accumulate less at liver sites. Smaller and negatively charged LNPs have better distribution at lymph nodes, as another example (source)
- Combination of PEG, Helper Lipids and Charge (source)
The list above is incomplete, but it does illustrate the multivariable nature of designing LNPs. There are billions of combinations and designs of LNPs for nucleic acids possible… an opportunity area screaming for technological disruption.
Back to the Verb
Recap: LNPs are disguises that can deliver therapeutics. They’re highly variable and complicated, which leaves room for engineering!
Creating more targeted LNPs relies on biomarkers. Specific biomarkers can be like keys: they only work in certain holes; they won’t unlock everything.
What is uniquely on our cell target of interest? What is uniquely on a cell we want to avoid? We start with those questions, and then we can design around them.
But once we build our targeted therapy, how do we measure its effectiveness?
3: Measuring Progress (Targeting as an Adverb)
Once we have our dart board mapped and a mechanism/strategy to target our areas of interest and avoid the ones not of interest, we need to measure our precision.
This measurement is essential to improving the targeting mechanism (the verb). Hence, it’s an adverb since it’ll be used to modify the verb of action :) (Alas, I have totally exposed my grammar nerd spirit)
For example, here’s an excerpt from this paper: “[We used] DNA barcode-labelled oligonucleotides [to] quantify targeted delivery of nucleic acids in multiple tissues.”
There’s a technology that helps us quantify something of interest. We use biomarkers to decipher between our measurements (to filter for the data we care about most).
This brings me to one of my favourite quotes of the day from Dr. Cullis:
🔑 “When you need an instrument that isn’t available, build it” — Dr. Pieter Cullis.
Science is an interface between subjects and technologies; our instruments limit us.
From Biomarker to Biotech
Once we have a theoretical hunch behind pathogenesis, we want to build some sort of benefit. I observed two big themes for the future of biotech ‘benefits’:
- In-vivo: utilizing our bodies’ existing machinery since they can make their own medicines/treatments. Doing more things in-vivo (as opposed to in-vitro) reduces the transition times between technology and the body. There is less transferring, and this has the potential to reduce costs.
- Platform technologies: infrastructure that can rapidly adapt to different goals at low costs/innovation. mRNA vaccines enable this. The mRNA vaccine part is the infrastructure which can be consistent across vaccines with only the mRNA sequence changing (to target different viruses, for instance). 🔑in essence, achieving a breakthrough can enable rapid iteration and adoption.
Places to go from mRNA vaccines
From the lectures and my conversations, here are some of the opportunities to improve this technology (special thanks to Dr.Anna Blakney for her talk and for sharing some information via email afterwards):
Reducing the dose of RNA (why: dose is directly correlated with adverse effects)
- Improving the stability of mRNA in-vivo so that it lasts longer
- Identifying other RNA molecules that could be used (Dr.Blakney’s lab looks at self-amplifying RNA, saRNA)
- Improving the efficiency of delivery. I.e., the number of mRNA molecules that get into the cells.
Engineering the immune response (why: vaccines build up our immune system’s memory, training our body to fight the real pathogen if it comes in)
- There’s plenty of room to understand immune sensing. That is, the pathways and mechanisms that trigger the immune system and how we can a) avoid this while our mRNA is delivered to the cells but b) trigger this after the mRNA is translated into the virus fragment.
- Delivery mechanisms… because at the end of the day, we are trying to bring a negatively charged molecule into the body. Delivery mechanisms, like LNPs, are crucial.
CAR T Cell Therapy
What is CAR T cell therapy? It’s a type of cancer therapy where a patient’s specific immune cells (called T-cells) are removed from the body and engineered to specifically attack cancer cells.
How do they attack cancer cells? This brings us to the “CAR” part of the process. CAR stands for chimeric antigen receptor, which is what we engineer onto the T-cells. These receptors specifically recognize cancer cells.
Adding them to the T-cells helps the T-cells go and ignite (or amplify) an immune response to the cancer cells.
Chimeric means ‘formed from many parts.’ The engineered receptor is chimeric in that it helps the T-cells find the cancer cells and activate an immune response.
What’s the catch? It’s really expensive: hundreds of thousands of dollars! We have to capture cells, engineer them in the lab (it takes more than one step… it’s not just a simple arrow) and put them back into the body.
But what if we could deliver the genes to T-cells inside the body for CAR receptors? We’d have a cheaper, in-vivo CAR-T cell therapy.
And then, if we crack the code for in-vivo CAR-T cell therapies… we can use it as a platform technology to target other cancers (right now, blood cancer is the primary use for CAR T-cell therapies).
Well… Dr.Weissman’s lab has already started and proven the concept of in-vivo CAR T cell therapy.
This sentence from their abstract summarizes the thesis of this sub-section, ‘from biomarker to biotech, ’ well: “In vivo generation of CAR T cells may hold promise as a therapeutic platform to treat various diseases.”
Science is a Small Group: The End
The Gairdner experience emphasized that science is interconnected. The community is growing but still small. A room of 200 people felt like a family dinner. Instead of political mismatches, conversations build on each other because scientific ideas/discoveries are so fertile.
It’s back to one of my lessons: specificity is the start of something broad. You take an RNA expert, a lipid expert and a vaccine expert and suddenly, you engineer mRNA lipid-nanoparticle vaccines. It’s the product of overlap… applicable to billions of people.
Everything’s linked because the body, animals and nature are systems.
It was a beautiful experience to shadow and immerse myself in overlapping scientific ideas. I am totally stoked to see some of these brilliant researchers on the Nobel prize stage in the coming years! 😉
Thank you again to the Morehead-Cain Foundation, The Knowledge Society and the Gairdner Foundation for this experience ❤️
Ps. go read something for fun on nature.com! I dare you!
