Skip to main content
Loading Explore guides…
DNA Replication Discovery · Acedra
Home
Explore
Biology
DNA Replication Discovery
Sign in to save
Chat
If YouTube asks you to sign in, open the video on YouTube.
Watch on YouTube
Reload player
Transcript
60 segments
Copy
Hide video
0:06
Hello, I'm Bruce Alberts and I'm going to talk about something I spent 30 years on, trying to understand how DNA is replicated, a very fundamental process in all living systems. I'm going to give you a little history because I think that outlining a pathway to discovery where we were lost in the woods, not knowing where we were going and misunderstanding things can help to make sense of what we now know. It also, trying to convey a sense of how science works, in this short talk.
0:44
So, this is a picture taken when I was in high school, 1953, a major breakthrough in our understanding of life happened when Watson and Crick correctly predicted the structure of DNA double helix and as soon as they got that structure, our understanding of living things, and specifically of heredity, changed dramatically. I'd like to point out that at the time, Watson was only 25 years old. I'm a big supporter of young scientists,
1:18
most creativity on the part of science happens when they're much younger than I am. This is the famous paper, first paper published on the double helix and it has this famous sentence, outlined in yellow here: "It has not escaped our notice that," basically our model suggests how genetic information is transferred between one generation and the next. They were correct. Now, the point is, they subsequently published a short paper in the same year which basically outlined this scheme,
2:05
called semi-conservative DNA replication, where each old strand of the DNA double helix templates the formation of a new one using the four base pairs that the observed in the DNA double helix. But they had, it's important to note, that they had no idea how this actually would happen. This, they understood in principle how it could happen and the next major breakthrough came only a few years later, a biochemist, Arthur Kornberg, who also earned the Nobel Prize,
2:40
as did Watson and Crick, discovered the enzyme DNA polymerase. This enzyme was thought to be very remarkable at the time. It will add a new nucleotide to the end of one DNA strand, called the primer strand, only if that strand is paired to a complementary strand, that's to serve as its template and there are two problems, basically I'll describe with their finding as far as really understanding how it worked. Because they thought the template strand must be single stranded
3:15
with its bases exposed, and second of all, that the DNA polymerase enzyme moves in only one chemical direction, that is growing the 3 prime (3') end of a new DNA strand. So, here's a little schematic view, primer gets added onto one nucleotide at a time, which nucleotide gets added depends on what base is on the template strand at that point and the 3' end of the primer extends one nucleotide at a time. So, this is a more detailed view, showing some of the chemistry,
3:54
we actually, as discovered by Kornberg, the source for the energy of polymerization is an incoming deoxyribonucleoside triphosphate and the two phosphates that are marked in red get split off every time this single unit is added to the 3' end. And again, as you can better see here, which base is added depends on specific base pairing, here C with a G, the G is on the template, C is coming in as a new monomer.
4:30
So, I started working on DNA replication, way back then, I was a beginning graduate student and I was fascinated by trying to solve the problem of how DNA might replicate, what was the machinery that would carry out the process that wasn't predicted and one of my professors, a famous professor, talked about, I still remember him giving this lecture about how it might work and here's one the models that was suggested. Since the DNA polymerase only goes in one direction,
5:05
maybe replication only starts at one end of the chromosome and opposite ends get synthesized going along this mechanism diagrammed here and I just want to show you, this would create a lot of single stranded DNA. Because this strand opposite the template strand would be released into a DNA single strand and so, if you looked at a replicating chromosome, you should see a lot of single-stranded DNA. Well, just one year later, a man by the name of John Cairns did a famous experiment
5:47
where he actually did electron microscopy of what you see when DNA is replicated. He discovered the replication fork. That is there isn't any of this large amounts of single-stranded DNA. Instead, DNA is replicated on both sides, both strands of DNA double helix roughly at the same time. And so at a replication fork, the two daughter DNA strands appear to grow at opposite ends, and yet we know that DNA polymerase only goes one way. How is this possible?
6:22
This was the mystery that faced me when I started work in the lab as a graduate student. And this was the way people drew it, actually, same professor is famous for drawing a fig leaf a little later over the replication fork. Saying we had no idea what happened at that point. So, 1961, when I started my work as a graduate student and when I did this, I was given the freedom to choose anything I wanted to work on, that was wonderful, and I wanted to solve the mystery of how DNA replicated.
7:01
We assumed, everybody assumed that a single enzyme, DNA polymerase, alone replicated DNA. But how was this possible? Two problems of course that DNA polymerase works only on single stranded DNA and yet DNA is a double helix inside the cell, as Watson and Crick showed, and secondly, the two DNA strands, as Watson and Crick showed went in opposite directions, they were antiparallel. And yet the polymerase only goes one way. Well, this led to very unsuccessful career as a graduate student
7:38
because I had been inspired by Watson and Crick to believe that theoretical biology could predict things and I think not only me, but many graduate students were predicting things and trying to test their ideas, we're more sophisticated now, so I predicted how DNA polymerase might be able to start DNA synthesis on a chromosome, I actually used it as a term paper in one of my graduate courses and I got an A+, so, oh my goodness, this must be right.
8:09
So, I then spent years doing experiments to see if my theory was right. Of course, my theory wasn't right, and as you'll see, the whole hypothesis was wrong, that one enzyme could replicate DNA. And I had a very unsuccessful, difficult time as a graduate student, from which I learned a lot, you could hear about that if you go through the short talk, I've given on iBio as well, it's also here in a section about lessons learned from failure
8:39
and I've had many failures, that's one of them I explained how failure's important for a career. You have to learn from your failures, that's what makes you better the next time. But then I finally got my PhD and I moved to Geneva, Switzerland for a year as a post-doctoral fellow. I immediately discovered, to my great surprise and someone consternation that if I had been reading the scientific literature I would have known better
9:06
because work done in bacteriophage T4, a wonderful model system for early stages of molecular biology, it's a large, double-stranded chromosome virus, infects E. coli and it makes all its own replication machinery and it had about a hundred genes discovered by genetics and published in 1963, in the middle of my thesis and the fact is that there were at least 7 genes in this virus that were absolutely required for DNA replication of this virus,
9:40
but the geneticists who did this work had no idea what these genes did. They gave them numbers, 32, 41, 43, 44, 45, 61 and 62. And a little later, one of these genes, the gene 43 protein, had been shown to produce T4 bacteriophage DNA polymerase, so that we understood, that made sense. But the question was, why are these six additional proteins needed for any replication of the T4 chromosome when the virus infects E. coli. And this meant that I and others had been completely naive
10:17
in thinking that only a single protein was needed. These sub-proteins must be needed and therefore replication must be much more complicated than anyone had imagined. So then I was getting ready to start my own lab and I needed a strategy and how was I going to, every young, starting scientist really should have the opportunity to start a new problem from scratch, that's the way we get innovation, and again, I was still devoted to solving this problem that I had failed to solve
10:53
and I had a new strategy, a completely different one. I decided I would make a column out of DNA and looked at all the proteins that bind to DNA in that column, as a biochemist, break open the cell, pass the crude extract over the column, and some of the proteins, a small fraction of them, will bind to the DNA column, they could be then removed from the DNA column by raising the salt concentration and I found that many proteins in a T4 infected cell,
11:22
made by the T4 bacteriophage genes, bound to this column. Then the question was, would one of these proteins be missing if a mutant virus is used instead of the normal virus from the infection. So, then I could find a protein that the genetics said was need for the very mysterious machinery that replicated DNA. Here's a slide taken from an experiment we published in Cold Spring Harbor Symposium, 1968, the first real discovery from my independent lab,
11:55
at that time at Princeton. We had done that experiment I just described and found that when you infected a cell with a gene 32 mutant, a major band that was present in all the other infections was missing. You know, pretty obvious, and because it was missing only when you had a gene 32 mutant, we concluded that that must be the product of gene 32, called the 32 protein and that therefore, it must also be required for DNA synthesis.
12:35
But of course, we had no idea what it did, so we had to do biochemistry and basically, what we discovered is drawn in this little schematic. The 32 protein binds tightly and cooperatively to single-stranded DNA, forming long clusters on the DNA. In so doing, it straightens out the DNA molecule because DNA, single-stranded DNA, single stranded RNA, would otherwise be folded into little knots because these little hairpin helices that can form with imperfect base pairing
13:15
and this open form of DNA is the kind of DNA that DNA polymerase can more effectively use as a template. So, here's one of our great experiments, you know, seeing is believing, I collaborate with electron microscopist Hajo Delius at Cold Spring Harbor and here's what we found. When we took circular DNA single strands, those are actually the viral DNA from a small virus, but it's completely circular and single-stranded, and without 32 protein, it's this little ball,
13:57
you see an example with that arrow, without 32 protein, it's this little ball, all tangled up in a little knot, when you add 32 protein, the whole molecule's extended in a beautiful way, in the form that turns out leaves the DNA bases exposed so that DNA polymerase can still see them. So, if we now go to the next slide, this is what we conclude after doing a number of experiments, that the form in which the 32 protein holds the DNA, it's a much better template for DNA polymerase,
14:34
allowing it to go much more rapidly than it would go otherwise, over this structured single-stranded DNA which would, the form of DNA inside the cell without the 32 protein. Now, there were 5 mysterious proteins left. Of course, we looked for some of the other weak bands on our columns, we didn't see them, so we tried a different approach called an in vitro complementation assay. It was a very simple idea, if you break open the infected cell
15:08
and add precursors to DNA, we see that some of that precursor nucleotides are incorporated into DNA. A very little amount, it's a very bad reaction because everything's being degraded as it's being made and everything's mixed together inappropriately, but with that assay, we asked, could we find the stimulation of DNA synthesis in this crude mixture that depended on adding one of the five missing proteins and fortunately for us, it worked for all the missing proteins.
15:42
And this is the kind of data we got, basically, if you added any extract of an infected cell to a crude lysate, made with a mutant, say in this case, it was 62 mutant, you could get a little stimulation very rapidly, stopped this lousy DNA synthesis, two-fold stimulation in the amount of DNA made, if you added the original extract of course, or uninfected cells, nothing happened, so we had an assay and soon as a biochemist has an assay,
16:22
you can purify anything, by following it through a series of chromatography steps. This work was mostly done by my graduate students at Princeton and finally, after years of collecting each of these proteins, every time we got a new one, we added it to DNA polymerase and see if it would replicate. The answer was no, but when we had all 7 and only when we had mixed all 7, suddenly, we had this magic day, which remember, I'd been working on this since 1960,
16:53
very unsuccessfully, now we could do what I could never do before, replicate a double-helical DNA template. If we left out any one of those seven proteins, it didn't work. Now, important point here is that this kind of in vitro biochemistry with purified proteins is essential if we're going to understand how life works because only then can you manipulate the system to actually see what actually is going on.
17:24
And it took us another 6 years then to figure out why each of the proteins was needed in this reaction. I'm not going to go through all the details by any means, I'll just give you one example that's simple. The gene 45 protein turned out to form a sliding clamp for the DNA polymerase which allowed the polymerase to move for many thousands of nucleotides, hundreds of thousands, without dissociating. This is how we viewed DNA replication in the old days.
17:51
This is how we viewed life in the old days; basically everything was a random collision. And so here's DNA polymerase, coming in, adding one nucleotide, falling off, a second DNA polymerase, because collisions are very rapid in the cytosol, in chemistry, we knew that from chemistry, so the second DNA polymerase comes in, it adds another one, then it falls off, another one comes in. Well, it turns out that kind of mechanism for life in general
18:17
is so inefficient relative to what actually happens that it's only rarely used. Usually things are much more organized, as I'll explain. Turned out that the 45 protein produced a sliding clamp for the DNA polymerase that we can show biochemically enabled the DNA polymerase to go for hundreds of thousands of nucleotides, we tried to figure out why that was so and I had a graduate student who worked very hard on this,
18:44
but we missed the fact that, as first discovered in the E. coli system by Mike O'Donnell, the reason why it worked like this is it was a ring, it formed a ring around the DNA that created a tether for the DNA polymerase and the DNA polymerase is kept associated with the same DNA molecule because it's pulling this ring along as it moves. And this is a structure that was solved originally for the E. coli protein.
19:16
I've pointed out on several slides that there are human versions of all these proteins. In this case, the human protein is called PCNA, it looks the same. A ring, pulled along by DNA polymerase. So, I skipped an important point here because I was going about, talking about what I was doing, but early in our work, by 1970 or so, we knew from the work of a lab in Japan, two Okazaki, husband and wife, and their co-workers, that when you looked at the newly synthesized DNA inside of a cell,
20:02
actually they used radioactive post-labeling to see what the new DNA was like, the DNA, half the DNA, that made on the so-called lagging strand, the strand that was so mysterious how you could move at a replication fork, that DNA was made backwards, again, it's made by the same DNA polymerase, which can now go backwards, because that's the direction it goes, but it's made as a short series of fragments, which are subsequently run into each other and sealed together to create a continuous strand
20:36
on the lagging strand of the DNA replication fork, just like the DNA on the leading strand is continuous because it's made continuous. Now this is a mechanism that was completely unpredicted. I think if anybody had predicted this mechanism in 1960, they would have said you're crazy, because this is too sloppy, it would create a mess, all these fragments would get mixed up and you would have all kinds of errors. But in fact, this is the way life works.
21:13
More specifically, each Okazaki fragment was made by the same DNA polymerase on the leading strand, that DNA polymerase needs a special primer to get started and that primer must be made very frequently on the lagging strand, to start each of the Okazaki fragments every hundred or thousand nucleotides depending on the system and this requires a separate enzyme. It turned out to be one of the seven proteins that was so mysterious in our system. This enzyme is called DNA primase,
21:50
it makes a short fragment that DNA polymerase then takes over. Very importantly, because you don't want RNA in your final DNA product, that RNA is recognized and specifically erased before all these fragments are sealed together in the end to make a continuous leading strand. Now, in retrospect, we could understand why DNA replication is so complicated. We never predicted it. It's because of the need for enormous accuracy. DNA polymerase turns out to have two active sites,
22:27
one for polymerization and one for proofreading. Why is proofreading needed? Proofreading is needed because although we might imagine that the base-pairing that Watson and Crick discovered would be perfect, A could only pair with T and C could only pair with G, this violates all the rules of chemistry and life must exist with chemistry as it is in the real world and it turns out that you can't have perfection in chemistry and bases have rare tautomeric forms which are of higher energy,
23:05
but nevertheless exist in one part in ten to fourth or so, one part in 10,000, and in their incorrect, minor, tautomeric form, they can pair incorrectly and for example, a C could pair with an A, be added by DNA polymerase as it's moving. DNA polymerase evolved so it could recognize that mistake because, when it, before it will add onto the previous place where it just added, it makes sure that that 3' end is base paired and so, as indicated in this diagram,
23:42
it can recognize that something's wrong, back up, chop off the mistake it made and go forward again. And this, it turns out, to increase the accuracy by something like a factor of a hundred over what would happen without self-correcting polymerase. Anyway, to make a long story short, I'm not explaining this really, only the enzyme that goes in this direction can do that effectively. We understand that and that's why we don't have DNA polymerase going continuously on the lagging strand.
24:14
It's also why DNA polymerase must have a primer to start, it can't start a new chain by itself because it looks for a perfect pairing before it can start and thus, also explains why an erasable RNA primer makes so much sense and why you need a DNA primase enzyme. Well, the net result of all this, plus things I can't talk about, is DNA is replicated with only about one mistake per billion base pairs that are copied. One part in 10 to the 9th.
24:50
This in incredible. Humans can't do anything like this. And we know that without such accuracy, a complex organism like a human could not exist. We'd have too many mutations, one of the things that would happen is that every embryo would get cancer and be destroyed. Many other things would go wrong. And so what we could imagine is that as cells evolve to more and more complex forms, all these mechanisms got added and somehow the original mechanism
25:25
by which DNA replicated obviously couldn't have been this complicated, but after all these many, many rounds of evolution, billions of years of natural selection, we ended up with the mechanism that I was fortunate to be involved in helping to discover, which was of course a real privilege and a great pleasure to solve this mystery. So, here's the DNA replication fork with its proteins on it. I didn't even talk about some of these. Let me, I talked about most of them.
25:59
The two green molecules are the DNA polymerase, one on the leading strand, and one on the lagging strand, same, in our system it's exactly the same DNA polymerase molecule, the red ring behind it is the sliding clamp that keeps the polymerase from dissociating from the DNA, the DNA polymerase is tethered by this clamp to the DNA. On the leading strand, this is all very simple, it just keeps going. On the lagging strand, it's very complex.
26:28
It stops and starts every few hundred nucleotides in human cells, and every time you want to start a new one, you have to make a new RNA primer and that involves an enzyme, which is one of our 7 proteins, called a DNA primase. It's a special protein that makes a short RNA primer, a few nucleotides long, just long enough to start the synthesis by the DNA polymerase each time. You can see the single-strand DNA binding protein coating all the single-stranded DNA.
27:00
It's probably also ahead of the DNA polymerase molecule on the lagging strand, not indicated there. The new molecule I haven't described is again one of our seven proteins, was an enzyme called a DNA helicase. Very important enzyme which uses ATP energy, hydrolyzing ATP, to propel itself along the DNA and opening up the helix ahead of the leading strand DNA polymerase to make the DNA single-stranded in the form that the DNA polymerase needs in order to copy that template.
27:35
Now, the fork is actually folded, this was our, probably, my most exciting discovery, the realization, basically, I was giving a seminar somewhere and I was, like I'm doing now, and I suddenly realized that it made no sense for the DNA polymerase on the lagging strand to have to come on and off every time it made an Okazaki fragment, that I could actually test whether, as a biochemist, whether it was being retained, and used over and over again and we actually found that was true.
28:11
So, the fork is actually folded to allow the DNA polymerase on the lagging strand to be used over and over again, therefore allowing very rapid DNA synthesis. So, the, really, the whole point of my talk really is to introduce you to this movie, my favorite movie. It was not made by me, it was made at Cold Spring Harbor laboratory, by Bruce Stillman and coworkers who work on human and yeast DNA replication. It was part of the 50 year DNA celebration there, 2003.
28:44
There was a big party, I was there, and this movie was produced and distributed for everybody to use, originally through the Cold Spring Harbor website. And I'll let this movie play, it's narrated, as it's narrated in our textbook, and you can find this movie on YouTube if you want, as well as on my talk. So, this shows the replication fork in action and what I want you to get from this movie is the general idea of how all of life works.
29:20
It's, life, I can talk to you only because life, the biochemistry of life, the chemistry of life, is this complicated and elegant. (Female voice) "Using computer animation, based on molecular research, we're able to picture how DNA is replicated in living cells. You're looking at an assembly line of miniature biochemical machines that are pulling apart the DNA double helix and cranking out a copy of each strand. The DNA to be copied enters the production line from bottom left.
29:56
The whirling blue molecular machine is called a helicase. It spins the DNA as fast as a jet engine as it unwinds the double helix into two strands. One strand is copied continuously and can be seen spooling off to the right. Things are not so simple for the other strand because it must be copied backwards. It is drawn out repeatedly in loops and copied one section at a time. The end result is two new DNA molecules." Well, that movie I find inspiring,
30:38
it really shows this mechanism that we have been working on for so many years, me and many other laboratories. And it's so satisfying to work on mysteries for decades and finally solve mysteries, so that's why it's my favorite movie. So, now, it turns out that almost every process in the cell is recognized to be driven in a similar way, different mechanisms, by a complex of ten or more proteins that works like what we call a protein machine.
31:15
These protein machines function very much like the machines in everyday life that are driven by electric energy, except that of course, they're not driven that way, they also need ordered movements, in this case ordered movements of protein and RNA molecules and those ordered movements are driven by proteins in the set that specifically hydrolyze a nucleotide, ATP or GTP, to give it unidirectional motion and to harness the energy of that hydrolysis
31:46
to make the machine be effective, because you need ordered movements to make machines. That's why we use electricity to drive machines in everyday life. So, here are some, now at the end of my talk, this is basically some personal lessons that I've learned as a scientist. I guess the most important one is not listed here, which is that being a scientist is exciting, you get to get paid for solving problems that both are fascinating to try to solve,
32:28
puzzles, but also, in the end, turn out to be useful, because although we weren't intending this, the kind of information we discovered is very important for our understanding of what happens in human diseases where these things go wrong, including cancer. But, the major mystery, the major point is, to work on something that's important that you don't understand. I call them mysteries because if you then solve that problem, it will help the whole world understand
33:00
whatever process your process is connected with much better. I mean, obviously, we needed to know how DNA replicated, that, we didn't know what it would be useful for, but it obviously has helped us understand many other things. Second thing which is, I've been stressing, and it often gets lost in this push to, you know, to go take government grant money and solve some disease problem, is that the shortest path to actually solving those kinds of human problems often lies in backing up a bit
33:40
and coming to understand how basic processes work. And this is often best done not in human cells, but in a model organism, like the one I used, or a bacteria, or a yeast cell, which are, all of which are much more readily manipulatable and easier to understand than our human cells. And finally, of course, the major lesson from my PhD thesis is that in many ways the Watson and Crick discovery, which was so critical,
34:12
was unique. Most theoretical predictions of biologists turn out not to be correct and to be oversimplified because nearly all cell processes will be based on elegant mechanisms which are too hard to predict, we just have to go find out how it works by experiment. So, thank you for listening, I hope this helped you to better understand a basic process that's going on as, every minute of the day, in billions of cells in your body,
34:49
if it wasn’t going on, of course, you'd be in very bad trouble, you couldn't be alive. Thank you.
Here to help you learn
Sign in to keep a private conversation with your course tutor.
Ask me anything about the material…
Sign in to chat