The main thing:
→ Church’s opinion on whether scientists can reproduce human intelligence in a machine
→ Discussion about civic science and the idea of open consent in genomic confidentiality
→ A brief diatribe on CRISPR (and a review of other genetic engineering tools)
→ Why gene therapy is the future of precision medicine
→ Ways to the real appeal of aging
You probably know the scientist George Church from unusual headlines in the media that one way or another referred to his work - the return of a mammoth
, the use of CRISPR in encoding animated GIF in living DNA
and the treatment (or treatment) of aging
. In his personal opinion, a professor at the Harvard Medical School and the Massachusetts Institute of Technology says things that seem implausible.
“Many of the things I said when I was young were ignored or ridiculed by people in my own laboratory,” said Church. “Now everything is different, and even if I say absurd things, they do not ridicule me. And even the most ridiculous things have a serious version. "
What makes Church worthy of the title is a rare combination of characteristics: he is a pioneer in science, a defender of "civil science", and his appearance is tall, wearing glasses, with wavy gray hair and a big gray beard - he looks like Albert Einstein
, Doc Brown
and Charles Darwin
It was these first two characteristics that made us visit Church in his laboratory at Harvard Medical School at the end of 2017. The combination of scientific contribution and the ability to attract the public with great ideas is rarely found among scientists. Church’s contribution includes, among other things, the development of first direct genome sequencing methods and concepts that lay the foundation for sequencing the “next generation” (that is, allowing sequential and fast sequencing of whole genomes) and the first use of CRISPR / Cas9 for editing genes in human stem cells.
In addition to Professor of Genetics at Harvard Medical School and Professor of Medical Science and Technology at Harvard and the Massachusetts Institute of Technology, Church leads or advises in dozens of companies
(sequencing, diagnostics, therapy, non-profit organizations, etc.); He is the founder of Wyss Institute for Biologically Inspired Engineering
, where he heads the direction of synthetic biology; He received several awards, including the Bauer Award and the Award for Scientific Achievement and was elected to the National Academy of Sciences.
When we arrived, Church and his team of nearly 100 laboratory members, interns, and visiting scientists worked on projects such as genetically modified pigs whose organs could be transplanted to people, changing wild species to eradicate malaria and Lyme disease, and creating cells resistant to all viruses. We talked about these projects and their impact on human health. While Church showed his usual openness and friendliness, allowing us to be at home in the laboratory - he advised us to avoid the fifth floor of a building not indicated on the elevator buttons panel.
“We’re joking that the fifth floor is beyond the scope of our world,” he said. “People say that mammoths and Neanderthals are on it. I can neither confirm nor deny this rumor. ”George Church at Harvard Medical School in Boston, Massachusetts
Before reading: glossary
Civil science : Broadly defined as public participation in research.
Open consent : an idea relating to genomic confidentiality . With open consent, participants who share their data recognize risks by disclosing information about themselves.
Synthetic biology : genetic engineering, which takes advantage of other areas, such as electrical, mechanical engineering and civil engineering, and at the same time uses a wide range of tools and resources from the wild (for example, CRISPR).
Genomic engineering : a subset of synthetic biology, including the assembly of a whole chromosome from natural genomic sequences .
Homologous recombination : exchange of DNA chains of a similar or identical nucleotide sequence .
Next Generation Sequencing : A generic term for current DNA sequencing technology that allows for the rapid sequencing of entire genomes.
Transcription factors : proteins that regulate gene transcription.
Oligonucleotides : short DNA or RNA molecules.
Nucleases : enzymes that cut nucleic acids (DNA and RNA). They play a role in the natural mechanism of cell DNA repair and are important tools in genome editing.
Civil science, gene therapy and the treatment of agingEH
: I want to start with a question that caught the attention of the public. In the podcast,
with your colleague, a professor at the Massachusetts Institute of Technology and the Future of Life Institute
, Max Tegmark, he discussed the nature of consciousness, superhuman AI, and whether intelligence depends on the carrier. Tell us everything you know, we will have super smart cars?Church
: I'm inclined to be skeptical. I think there is a longtime meme or idea that we can reproduce our intelligence in the machine. In particular, individual intelligence, like yours or mine, in an individual machine, and not some kind of common machine that can add numbers or play movies.
The words "impossible" are rarely used in my laboratory. If this is a short-term, high priority, you can use it. If you ask what is the smartest computer in the world, with the most amazing computing power, it is the human brain. He can do such things as the five works on physics performed by Einstein in 1905, each of
which could receive the Nobel Prize
, and one of them received it
. It would be difficult to imitate this with a machine.
This is no less difficult for other people, but the fact is that people are capable of it - and with a power of 20 watts. Just to beat a person in a simple task, for example, chess or Jeopardy or Go, you need a 100 kilowatt machine. And this is a very limited thing in which people are not very good. It would seem that a person is competing with a jet engine. This is not what people evolved to, so it is not surprising that they cannot do this. Things people are good at are some of the most valuable things: thinking outside of constraints, thinking about the future, diplomatic speaking, teaching ethics. We cope very well with these cognitive tasks, and we teach them machines very poorly.EH
: The reason why Endpoints
exist, and why we are doing these interviews, is to encourage participation in science. How do you feel about the role of a non-scientist in science?Church
: Historically there have been various “civil science”: weather, astronomy, gardening, and genealogy. And I think it is extremely important that we have a civil science. I was involved in DIY biology
from the very beginning of this movement in San Francisco, New York and Boston.
We need to inspire the next generation, like NASA and Disney animation, inspired me when I was young. It is equally important to make decisions in an increasingly technological world. We need our citizens and leaders to be technically savvy. They do not have to have academic degrees - for example, as one congressman, Bill Foster
, but if we do not take care in advance, it will put us at a disadvantage.EH
: One of your main projects, encouraging civic participation, is the Personal Genome Project
. What is it?Church
: The original goal was to provide the genomes and the other sombs and medical data so that scientists can find the cause and effect, or at least the correlation between the media and the genes. We also develop cellular therapies, gene therapies, synthetic biology; they can now be checked on PGP cells, many of them are mine, but they can be from anyone. This can be compared with various data sets that are openly available.
PGP is mainly about measurement, not modification, but I think measurement is an important part of any research. PGP is designed to provide the way in which various studies can exchange information. In principle, we can be representatives of the world: we agree, and the world can use data and materials in various ways. I think that we are the only biomedical research project in the world that gives people full disclosure, full publication of any aspect.
Now this is an international project in five countries, and I hope that this autumn we will add China and Mongolia. We do not need all 7.5 billion people, although it would be nice. The bigger, the better.EH
: Can you explain the concepts of full disclosure and open consent?Church
: When I developed another sequencing technology, I realized that we can quickly move the bacterial genomes to humans. And we did it.
I figured we would need to get approval from the institutional review board
(IRB) to do something in public, even in sequencing, so I hired a law student, Dan Warhus, and ethics, and we published several articles on this issue
and understood: you could not promise de-identification
in a world where both DNA and your traits are identifiable; you could not promise that your information would not be stolen in the world of Wikileaks, and where almost all medical records can be hacked because they are valuable.
Therefore, we switched to a mechanism in which, instead of making false promises, we mitigated the consequences as much as we could, and at least made sure that people know what they are going into. Thus, open consent is only frankly with people about the reality of medicine and life in general, and this is what your data will escape and be re-identified.
The problem is that if you agree to something less than full disclosure, then something happened - the data went away or they were identified - you violated your trust and contract. This is unacceptable. It's not about "We will try to do everything possible so that this does not happen." It gives the feeling that it can be prevented. If this could have been prevented, the Ministry of Defense and the Democratic Party
could have prevented this, because the stakes there are much higher than that of the individual patient.EH
: So a person can share his genome with PGP. What is the advantage of this?Church
: First, if you have a family disease and you force your family to discover, it is likely that the disease will receive a higher priority. Instead of lobbying the congress for additional funding, use the project. More personal knowledge. Some projects not only do not share your data with other scientists, they do not share them with the person who donated it, which struck me. The idea of protecting you from your own data is ridiculous.
This is an opportunity to be part of the ecosystem of scientists and patients who can contribute.EH
: Did PGP do anything?Church
: We developed CRISPR on induced pluripotent stem cells from PGP (they were my cells), which was the first use of CRISPR for human stem cells and focused on various alternative ways of achieving accurate editing. We also have the first complete set of human transcription factors. We can essentially turn skin cells into any cell in the body. We also used this to reverse aging, where we used transcription factors to epigenetic changes in the cell state. All this was in cell culture.
Gene editing, four ways
CRISPR : A genome editing tool created from a system that protects bacteria against viruses and plasmids. Here is a detailed explanation .
Meganucleases : a family of nucleases, useful in genetic engineering, because of their longer recognition sequence, which means that they have less chance of error.
TALEN : Transcriptor Activator-Like Effector Nucleases . Another editing tool derived from bacteria. Scientists can change the order of amino acids in TALEN to recognize a specific DNA sequence.
Lambda Red : also derived from bacteria. Unlike other editing tools, lambda red does not double the chain break in DNA. Instead, it uses donor DNA during the natural process of cell replication.
EH: What opportunities does CRISPR give us?
Church: I am calm about CRISPR. I try to downplay its meaning in order to resist deception. Our largest, most efficient, and most accurate genome development projects do not include CRISPR. Our record for CRISPR - and the world record - 62 changes immediately in pig cages
. Our record for other genome editing tools is 62,000 in the project, and this was done using lambda red.
CRISPR was not the first genome editing tool. We and other groups used meganucleases, TALEN, lambda red
and others. CRISPR has a lot of problems, so one of our projects is developing alternatives. We are looking for things that do not cause double breaks - because this is due to toxicity and the inability to get accurate editing.
CRISPR is the fourth major way to make nucleases. CRISPR is a bit easier than TALEN. In each case, when you want to change the position in the genome you are aiming at, you order it. In one case [CRISPR], you order the indicated RNA, and otherwise you order TALEN. And, in fact, TALEN comes bundled, so as soon as you have the kit, you no longer need to order anything. You just drip. In a sense, you can argue that it is simpler than CRISPR, since you do not need to order extra. But this means that they are much more complicated and you cannot make them. Thus, the difference is ease of use.
Another difference is how effective they are when they are in the cage? In other words, once you are in a cage, what are the chances that you will get what you want? If you want to turn off something, CRISPR is quite effective. In some cases, it can be two to four times better.
Most "book" people are called editing off the gene. This is not very accurate editing. Exact editing — for example, you have a sickle-anemia gene, and you want to change A to T. That's right. If you want to get rid of the hemoglobin gene, CRISPR is your tool for this.
But if you want to do something exact, you get a race between random and accurate double-chain repair. Why and need such things as lambda red. For CRISPR, you need two nucleic acids: a guide RNA and a donor DNA. In lambda red, you just need to put in donor DNA, and that’s for sure, because it doesn’t make a double break. It interferes with normal replication when you have few single-strand circuits, and replaces a single piece. (When you copy DNA, you have short pieces that are naturally single-stranded, it simply binds to them.)EH
: Why are people excited about CRISPR?Church
: I think there is a tendency to create icons. The Apple logo has nothing to do with apples. Behind him is electronics, software, design, testing, strength, human factors, all of this. But everyone says it is Apple. The same thing about CRISPR. This is not CRISPR. This is editing.
This is my little diatribe.EH
: Before we dive into editing genes, let's take a big step back and talk about what you are doing in your lab, in general.Church
: What we do as a whole is the development of transformative technologies, ideally, something with an amazing philosophical component, a technological component with a factor of a million times, not two, and social advantages. You are trying to take all this into account.
Examples of things we were involved in include next-generation
sequencing, capable of sequencing a human genome for less than $ 1,000; in situ sequencing
, where you can see the component sequences in the cell; writing new genomes, not just editing them, and then using them to precisely and inexpensively create organisms, organs, ecosystems. Much of what we do is related to security, price, and the egalitarian distribution of technology.EH
: These transformative technologies are sometimes known as “synthetic biology”, and in the case of Wyss, “biologically inspired engineering”. What does it mean?Church
: I will describe it like this. CRISPR is a tiny subset of editing. Editing is a tiny subset of genomic technology. Gene therapy is a subset of genetic engineering. Engineering is a subset of synthetic biology. Synthetic biology is a subset of biologically inspired engineering that includes biological engineering, as well as things that are not actually biological or biomedical, for example, small robotic bees and swarm technology.
You can also say that synthetic biology is a real genetic engineering with all the advantages of other areas such as electrical, mechanical engineering and civil engineering, but also with its own advantages: a wide range of tools and resources that come from wildlife, for example CRISPR . It was a very complicated thing, and its adaptation from bacteria to mammals was in our power. It was difficult, but we did it in a year. If we tried to invent CRISPR from scratch from atoms, it would be much more difficult.
We have very complex nanomachines from nature, plus the ability to independently engage in evolution. With a mobile phone or car, you can have one prototype or a pair. With the help of biotechnology, you can do trillions of tests. You can make billions of prototypes in a test tube and test them simultaneously.EH
: So let's talk about gene editing. What are applications?Church
: We use it to “humanize” pigs so we can take their organs
. We use it in wild species to be resistant to Lyme disease and malaria
. It takes a lot of engineering effort to make them sustainable, because you do not want it to fade. You want mice or mosquitoes to remain stable. The most extreme form of editing makes the host resistant to all viruses, even those that you have never seen before. It turns out that viruses do not have their own genetic code. They depend on the owner. You can change the host so as not to harm him, but the virus cannot even begin to develop in it.
We showed this by changing the genetic code of a single industrial microorganism, and now we are expanding intake to several organisms and more stringent tests for resistance to all viruses.EH
: This year, the FDA approved the first gene therapy, as you said, a subset of genetic engineering. Why are gene therapies valuable?Church
: Gene therapy is especially attractive to us, because you can go from theory to hypothesis and to therapy almost instantly. You do not need to open a small molecule. Small molecule - indirect approach. If you know your molecular goal, you do not necessarily know the composition of the molecule that affects it. If you know the phenomenon, you do not necessarily know your purpose or the composition of the molecule. If you have a theory in which you know the mechanism, you can go directly to gene therapy without any problems. We will move from a theory that can work in some simple organism to testing either in human organoids or in mice, and then on dogs, and on humans.
Another argument in favor of gene therapy is the fact - in principle, you can permanently change the cell. It's like Newton's law: the movement is unchanged, until you do something. It means that you do not need to take a pill one or more times a day. Potential advantage, but it is not realized, because there are not so many gene therapies on the market. Will be soon
. I’m pretty sure that gene therapy will be one of the most accurate and safe mechanisms we have, because they are easily programmable. They are easy to make smart: where they are delivered, what they do, the logic that they double check that they are doing the right things in the right place at the right time. [Here is a brief description
of how they work.]
The advantage of a clever drug that does not depend on a person who needs to remember a lot. You drink a pill blindly if you do not take blood samples and do not analyze them on a computer that tells you when and what to do. Most pills are taken uniformly. I get the same dose as you, regardless of our weight, time of day, regardless of my stress, and so on. We are committed to precision medicine, but we have not yet come to it. Gene therapy is our best chance for precision medicine. It is automated at the nano level.EH
: How can exact gene therapy be done?Church
: They can be delivered intravenously in systemic delivery. We have some evidence that we can get at least small levels in several different tissues. They can be intramuscular, or in the retina, the brain, through the intestines. It may even be a pill. She will have to go through the gastric juice, then through the intestinal barrier, and then, if you want to enter the brain, she must cross another barrier to the brain.EH
: I heard that you are working on gene therapy that reverse aging, so let's talk about aging. Is there an acceptable causal theory of aging?Church
: There are hypotheses and different schools of thought. It is not so mature that there is a consensus. There are relatively few interesting areas of biology where there is a consensus. There is a school saying that this is all damage, and you must repair the damage done. There is another school, about regulation and epigenetics, and if you get a cell in the correct epigenetic state, then it can repair the damage itself; the young cell is much more efficient in reparation.Hallmarks of aging
– – , , . , . , .
, : «» , , , , . , . - , . , , , , , , . , 150 , . , , , .
, , , - , , - , , .EH
: , . ?
: , , .
. – , , . , , . , , , .
, , , , . , , , , , . 20-, 60 .EH
: c, João Pedro de Magalhães, , GenAge
. 304 , , , . , , , , . , , , , , , - . , , CRISPR.
, . , 100% . . , , .EH
: . , : , , , . - , , .EH
: , . , - . , . – .EH
: , , , . Rejuvenate Bio
- , , . . , . , . FDA , – . , , .EH
: FDA , ?
: , , . -, , . , . , -, , . FDA , .EH
: . , , . , , , , . – , . .
. . , , , , – , , 100% , , - . , , .EH
: , - CRISPR, « A Crack in Creation: Gene Editing and the Unthinkable Power to Control Evolution
» , . ?
: , , . , . : , 100 . , . - , . , , . , , .
, 7,5 , , , , , LCA, Editas
, . – , - . . , , . . , .EH
: , , ?
: , , , , – , , . . , . , , , , , .
, SNP, , 23andMe
. , , Veritas
, – . , $999, , .
, , , , . , , , lostridium difficile, .
, . .EH
: , , . , , . : « , – ?» , . , . , , , – , .
, – , , . , . . .