Constructing and Screening a Recombinant DNA Library | MIT 7.01SC Fundamentals of Biology
Articles Blog

Constructing and Screening a Recombinant DNA Library | MIT 7.01SC Fundamentals of Biology


PROFESSOR: Good morning,
good morning. So last time, we talked about
the most remarkable biochemical purification
procedure ever invented– cloning. You remember the issue
with biochemistry. You’re going to grind up a cell,
you’re going to take the contents and run it over
different kinds of separation columns or centrifuge it or
things, in order to separate some things from other things. The problem with purifying a
gene that way, away from all the rest of DNA in a cell, is
that the gene has exactly the same biochemical properties
as all the other genes. How in the world are
you going to do it? The solution was devilishly
simple and devilishly complex at the same time. All you have to do to purify
something is dilute it. If I take any substance
an I add enough water, it’s very dilute. And in any little drop in any
little test tube there will only be one DNA molecule. It’s now purified. I could do that for every
molecule there is. The problem is, it’s not very
useful unless I have a way to take that molecule and make
extra copies of it. But if I could do that, dilution
is purification. The trick we use initially for
making more copies is we ask E. coli to do it for us. That, as usual, is the solution
to most issues. Find something in life
that already does it. And so, just to remember what
we did, we took total DNA. Maybe it was yeast DNA, maybe
it was human DNA, maybe it was zebra DNA. And we cut it up with some
restriction enzyme. Our restriction enzyme we used
was EcoR1 And it therefore cut at the EcoR1 sites, G, A, A, T,
T, C. We could have used a different restriction enzyme. We then added it to a vector. The vector was cut open
at an EcoR1 site. The vector had an origin of
replication, ORI, and the vector had a resistance marker,
some resistance gene that made a protein that could
break down some antibiotic found in nature. We combine these two pieces. Vector now gets its insert. We call this vector, we
call this insert. We attach them together
using what? AUDIENCE: Ligase. PROFESSOR: Ligase, the enzyme
that ligates DNA. We then take this, we transform
it into a bacterium. The scale is obviously
off, right? This DNA plasmid here is tiny
compared to this bacteria. But if I draw it to scale,
it won’t be very helpful. So we then transform
it in here. We do that in a test tube. We treat the bacteria a little
roughly so it likes to slurp up the DNA. We then plate it on a plate. And those bacteria that have
acquired our plasmid containing the antibiotic
resistance gene are able to grow on this plate that
has antibiotic on it. And we call this thing
a library. So that’s it. We can make a library. And we have in effect,
then, what I said. We’ve deluded the individual
molecules out, and each one went into its own bacteria, and
each one got replicated by that bacteria. Those plasmids are replicated
by the bacteria. In fact, we choose plasmids
that are called multicopy plasmids, where there’s not
just one copy but the cell might make 50 copies of it. We grow up a whole colony
of it and there you go. We talked about some of
the issues with it. Where do we get the restriction
enzymes, the ligases, the vectors,
et cetera? It’s in the catalog, right? We used to purify them
ourselves, but they’re all in the catalog. So any questions about this? We had some questions. I have a question about this. How come when we add ligase the
vector itself just doesn’t close up into a circle without
an insert in it? It might. What would happen then? AUDIENCE: You’d just have
your original vector. PROFESSOR: You’d just have
your original vector. And what happens when I
transform into the bacteria? AUDIENCE: It’ll still survive. PROFESSOR: It’ll
still survive. So will the unimolecular closure
of that circle be more common than the bimolecular
interaction between a vector and an insert? AUDIENCE: Probably. PROFESSOR: Probably, because the
two ends of the circle are pretty close to each other. So what do I do? Yeah? AUDIENCE: Because [INAUDIBLE] are the same. [INAUDIBLE] same restriction enzymes
because if they’re two different enzymes so
they don’t match. PROFESSOR: Ooh. That’s cute trick. Two different restriction
enzymes so that they couldn’t reclose. But then my fragments better
have those two different sites, too, and only be able to
clone those fragments that have the two different ones. But that could work. So you want a trick for making
sure it doesn’t reclose. Any other tricks? But the problem is, I won’t get
all the fragments, only the ones that have, say,
an Eco at a Bam site. But that’s– So I bring this up not because
it’s particularly important, but to tell you the kind of
engineering that really does have to go on in molecular
biology. What happens is when you have
your DNA put in here, we have a sugar-phosphate backbone
in both cases. And if we look up close, one
of these sides has the phosphate, the other
has a hydroxyl. Phosphate, hydroxyl, phosphate,
hydroxyl, right? And ligase comes in
and joins that. So this guy has a phosphate on
this strand, and this guy has a phosphate on that
strand, hydroxyl there, hydroxyl there. What would happen if
I got rid of the phosphates on the vector? Could it reclose? AUDIENCE: No. PROFESSOR: No. So if there were an enzyme that
removed phosphates, I could treat my vector
first with the phosphate-removing enzyme. And now it couldn’t possibly
reclose on itself. And is there such an enzyme? And what’s it called? Phosphatase, because it takes
phosphates so easy. Phosphatase, it takes
off the phosphates. And then it can’t
reclose anymore. And where do we get
phosphatase? It’s in the catalog. Exactly. So now what happens
is that the vector only has an OH here. What happens to ligase? When I put an insert in here,
ligase can make a covalent join on this strand. But it can’t actually
make a covalent join on the other strand. But does it matter? No, because I’ve closed up my
circle on that strand, and I close up the circle there on the
other strand, and I just throw it into E. coli. And you know what happens when
it goes into E. coli? It’s got a nick, obviously,
on that strand. It hasn’t closed up. But what does E. coli do
when it sees that DNA? Must be damaged DNA. I’ll fix it. So E. coli actually does the
last little trick of closing that up for you with
its own enzymes for repairing its own DNA. I bring this up not because it’s
crucial that you should worry about it, but because I
want to know that there’s a whole layer of these interesting
engineering tricks that get developed. Every one of them exploits
enzymes that we know. Every one of them deals with
questions like, will my vector reclose on itself, how
do I avoid that? And there’s a vast cooking
book of protocols in molecular biology. And we constantly are just
cribbing from things life does to make our protocols more
and more efficient. So I bring it up more because
it’s kind of a cool thing that all that goes on, and because
it helps you remember that these phosphates are
very important to joining things up. That’s a digression. Now, let’s go to the topic. How do we actually
read the library? How do we read our library? How do we use the library,
read from the library? Well let’s say we’re going to
try to find the arginine gene. We talked about the gene
for arginine in yeast. So I’d like to clone
the gene for ARG1. We found mutants before that
were unable to grow without supplemental arginine. They somehow had a defect in
producing their own origin. It’s a mutant. I want to find the
gene, please. How do I do it? We’ve got to think about
what’s our insert DNA. What are our vectors? What insert DNA should
we start with, zebra? No. Human? No. How about yeast? We’re trying to clone a gene
from yeast, right? So let’s start with yeast. OK. So we’re going to start
with yeast DNA. We’re going to cut
up yeast DNA. We’re going to attach it to
our vector, we’re going to transform it into E.
coli, E. coli will grow up on our plates. And one of these guys, I happen
to know it’s that one there, contains the ARG1 gene. The problem is I happen to
know it, but you don’t. How are you going to find out
where the ARG1 gene is? Any takers? Yeah? AUDIENCE: It could be like when
you put the gene in make it flourescent. PROFESSOR: A fluorescent tag? So I should just attach
the fluorescent tag to the ARG1 gene? AUDIENCE: Yes. PROFESSOR: How do I do that? All the DNA looks the
same in the tube. How do I know where to attach
the fluorescent tag? AUDIENCE: Maybe you could
size it [INAUDIBLE]. PROFESSOR: There’s a lot
of pieces of DNA there. And my eyes are not that good. AUDIENCE: Separate it? PROFESSOR: Separate it. But will I know which
one is ARG1? See, I don’t actually know
anything about ARG1. All I know is I made a mutant. The mutant is unable to
grow without arginine. I haven’t got a clue
what that gene is. I don’t know what it encodes, I
don’t know how big it is, I don’t know nothing. All I know is that whatever it
is, it’s a gene which when mutated prevents you from
growing without arginine. AUDIENCE: Could you plate all
of your colonies onto a– could you put [INAUDIBLE]? PROFESSOR: Minimal medium. What if I plate on
minimal medium? Now what? What are you hoping for? AUDIENCE: The one that has the
ARG1 gene will not grow. PROFESSOR: The one that
has the ARG1 gene won’t be able to grow– oh wait, yeah. But something like that. Let’s work it through. We’ve got my idea here. What are we going
to do with it? AUDIENCE: [INAUDIBLE] PROFESSOR: I’ve got a working
ARG1 Mutate ARG1 afterwards? AUDIENCE: [INAUDIBLE] PROFESSOR: OK. How will that work? I’m open for– got an idea here? AUDIENCE: If you have your
different colonies [INAUDIBLE], you could have a
secondary plate them all over to one of middle medium. The ones that die are the ones
that already [INAUDIBLE]– PROFESSOR: So guys,
I have a concern. I’m just transferring
this into E. coli. E.coli grows just fine
without arginine. I mean, I’m going to take
this yeast DNA. I’m going to put
it in E. coli. E. coli was kind enough
to grow it for me. But frankly, E. coli doesn’t
need this ARG1 gene. E. coli can grow without
arginine. I can plate this with and
without arginine, E. coli grows just fine. But you’re on to something. You’re onto the idea that
somehow, the only thing we know about ARG1 is that the
functional, wild-type copy of that gene confers an ability
to grow without arginine. And who does it confer it on? And what kind of yeast? Haploid mutant yeast. Ah. So suppose I put a working copy
of ARG1, a good copy, a wild-type copy, into
a mutant yeast. Now what would happen to
that mutant yeast? What would happen? The mutant yeast before, could
it grow without arginine? No. If I put in a working
copy of the ARG1 gene, what will happen? It grows. Now let’s design a scheme. Do I want to use
E. coli at all? No. What do I want to use? I want to use a yeast. So let’s get rid of E. coli and
let’s instead use yeast. And which yeast should we use,
wild-type or mutant? Mutant yeast. What mutation? ARG1 mutant yeast. ARG1 minus yeast. Now, if I plate ARG1 minus
yeast on minimal medium, what happens? It doesn’t grow. It dies. What DNA should I
be putting in? Yeast DNA. Mutant or wild type? Why wild type? Because it’ll have a working
copy of ARG1. So I want yeast, wild type. Now what happens? One of these guys, and only one
of these guys here, this one, has a ARG1 gene. That’s ARG1. When it goes in, that plasmid
has the ARG1 plus gene, whereas other plasmids don’t. That cell that inherits that
gene there, that gets that gene, is not green. I just drew it green for you. But it has the ARG1 gene. And when I plate this on
minimal medium, what’s distinctive about it? It grows. That’s how you can clone
the ARG1 gene. You clone it by the only thing
you know about it. Namely, it confers a function. This is called cloning
by function. Or, what did we do when we
crossed two mutants together to see if things were
in different genes? It was a test of
complementation. Really what we’re asking is,
is there a plasmid that complements the defect? In effect what’s happening is
in this cell, right here, we have a yeast cell. And the yeast cell has
a defect in its ARG1. But the plasmid has
a working ARG1. So for that one gene, this cell
could be thought of maybe a little bit like a diploid,
just at that one gene. And we’ve done a
complementation, just a teeny little complementation
for one gene. And we could call this cloning
complementation. It’s essentially cloning
by function. Any questions? Yes? AUDIENCE: [INAUDIBLE] –they all have functioning
[INAUDIBLE]? PROFESSOR: Yep. AUDIENCE: So why does that
then have a function? PROFESSOR: Oh, oh. You see, the yeast genome has
about 4,000 different genes. I chop it up with my EcoR1. Some plasmids get ARG1 but most
of them get leucine 2 or [INAUDIBLE] or other things. And each yeast cell in my
library only picks up a plasmid with one chunk
of DNA, one gene. So it turns out that the yeast
cells in my library, each one has one of thousands of
alternative possibilities. And it’s just the guy who
gets ARG1 who grows. AUDIENCE: But you’re saying
that the yeast [INAUDIBLE] plasmid. All code for– PROFESSOR: That’s right. AUDIENCE: But I don’t get why
you would end up with [INAUDIBLE]. PROFESSOR: What do I–? AUDIENCE: Why do you end
up with a strain that has ARG1 in it? PROFESSOR: Oh. ARG1 is the ARG1 working copy. In the yeast, I’m talking about
this yeast here has the working copy of ARG1,
ARG1 plus. So this guy has an ARG1 plus. It also has lots
of other genes. Each of these plasmids
gets one gene. Some of them get an ARG plus. Some of them happen to get a
leucine gene or some other gene that’s irrelevant. And the plasmids that contain
the working copy of the gene, they, when they go into the
cell, give the cell the ability to grow. So that’s why. All right. So that’s how we get
a gene by function.

36 thoughts on “Constructing and Screening a Recombinant DNA Library | MIT 7.01SC Fundamentals of Biology

  1. You could also screen by putting each gene fragment into a vector such that it was inside a partial lacZ gene (at least in E.coli) -that is, you put the polylinker in the plasmid lacZ. You can then do a simple blue/white screening.

  2. It does remove both phosphates. The phosphate used in the ligation comes from the insert which has a phosphate on one strand and an OH on the other

  3. Thank you for this sharing! I cannot forget my last visited to MIT when I participated in iGEM. Really hope I can have the chance to study there

  4. Dr. Lander, I love your lectures, learned from you a lot!!! it's and a great review for molecular biology!!

  5. he is a brilliant lecturer! everything is put into context and explained so well, I wish I had him for my molecular biology course. lucky students.

  6. you could point a great professor just listening to his sound, the pitch on his voice. Like a music of science.. as if he becomes knowledge and the knowledge becomes him goes so comfortably…

  7. What does he mean by the "catalog"? Do molecular biologists look at a catalog and order DNA ligase and different enzymes that are mass produced in a lab somewhere so they don't have to make everything they need themselves? That sounds cool.

  8. LOL I love his lecture and how he ask interesting questions that students can engage and think like a scientist by just starting with questions ! LOVE IT

  9. What if mutant yeast cell that has the plasmid with arg+ gene does attack the plasmid with its own restriction enzymes ?

Leave a Reply

Your email address will not be published. Required fields are marked *

Back To Top