This is the second lecture on DNA.

In the last lecture, we saw...

we talked about
how many bases there are.

We saw that the cell and DNA
contains many more bases

than just the four standard bases.

In this lecture,
I want to talk about pairing.

How is pairing controlled?

And also, how does DNA replication

achieve the amazingly low mutation rate
of ten to the minus four.

So, usually when we talk about base pairs
they're presented in this way,

where T really fits only A
and G only fits C.

It's simply the shapes...
don't fit any other way .

And, that then would be what matters,
which would be then translated to

T and A having simply
two hydrogen bonds

and G and C having three

And, that's what allows the DNA
to replicate with such high fidelity.

DNA replication is really
an amazing mechanism.

Here, we see...

Of course, no one managed to videotape
how DNA replicates

but it would look about like this -
not with colors.

And, I will not talk about
this whole machinery,

I will just talk about...
slightly about this DNA polymerase,

which allows DNA replication
to achieve a very low rate.

Let's look at the reaction.

So, usually in any reaction
we have reactants and products.

And, we're interested in the rate of going
from reactants to products...

or both rates going forward
and backward.

Rate going forward usually is calculated
by the Arrhenius equation,

which gives us e to the minus EA -
the activation energy - divided by kT.

So, the higher the activation energy,
the slower the reaction will be.

And, activation energy is actually often
what enzymes affect

and also, in this case,
what DNA polymerase affects.

The reverse reaction will include
the same activation energy often -

so you will have to go
over the same peak.

But sometimes, if reactants and products
have different free energy,

then there's also Delta G.

So, the activation energy
going backwards

includes another term, Delta G,

and the rate going back
could be slower or faster.

The ratio between the two gives us then
e to the minus Delta G over kT.

And, when we have a case
where we are talking about

the right base versus the wrong base
we're looking at two different reactions.

And then, what we're interested in
is the Delta Delta G,

the difference between the two Delta Gs,

which then gives us the ratio
of the products

of the right or the wrong type.

There's also... one can also talk about
the difference in the activation energy,

which is important in DNA polymerase
and we'll talk about later.

But, at first, we'll start with
this Delta Delta G.

So, does Delta Delta G,
the difference in binding strength -

difference in the hydrogen bonds...

Does this explain how well DNA binds?

When we have... we start with a certain
ratio of right to wrong bases,

and then we go through a reaction
where both of them drop

by a difference in Delta G,
Delta Delta G.

Then, the initial ratio is amplified by
e to the power of Delta Delta G over kT.

When we look then at the bases,
and this is...

the base pairing energy actually depends
on the neighbor bases.

And, here I just showed you the Delta Gs
for when the previous space was CG.

So, when the previous base is CG,

and here you see in blue
the Watson-Crick base pairs -

and in red, the other ones -
you see that the difference between...

the biggest difference that you can see
between right and wrong

would be about 2.5 kilocalories per mole.

And, if we want to get the ratio that...

the amplified ratio that we get out of
such a difference in Delta G,

it's very easy to calculate.

We simply take the 2.5,

multiply it by about 4000
to get it into joule,

then divide by the Avogadro's number,
k and T.

We can put k and T...
k and the Avogadro's number together

to get the gas constant,
which is 8.31.

And, in the end,
we get about 60.

So, the biggest difference
that we would get

just out of the binding energies
of base pairs

would be a sixty-fold difference,
which is very far from ten to the ten.

How can we get a better...

effective rate.

One proposal that was brought forth

is something called
"kinetic proofreading,"

whereby having non-reversible reactions
we can actually enhance this ratio.

Let's see how it works.

So, we said that if we start with
a certain ratio of right to wrong bases

and do a reaction
where there is a Delta Delta G,

the ratio is amplified by
e to the power of Delta G over kT.

If we add several more reactions
into this

where still we have the same
Delta Delta G or smaller,

then at equilibrium,
the amplification is still the same -

nothing changes.

As long as everything is reversible,

the ratios will simply be
the ratios of the Delta Gs.

However, if we add the...
irreversible reaction here,

then something different happens.

Now, the first reaction will have
an amplification

of e to the Delta Delta G
over kT.

The second reaction by itself
will also have such an amplification,

so that overall we get
an amplification of about

a square of e to the power of
Delta Delta G over kT.

So, we...

took the ratio that we have
and put it to the power of two.

So, from sixty you would get
like 3600,

still far from ten to the ten
but a bit closer.

In the case of DNA polymerase,
this is probably not what happens

because we don't know of
really an irreversible reaction

that uses energy

that goes on...
when we attach another base

Instead, it seems that,
in the case of DNA polymerase,

what's important is not really
the difference in the free energies,

but instead, the shape of the bases.

So, let's look at
what DNA polymerase does.

This is a schematic representation
of DNA polymerase.

Usually, it's thought of as something
like a hand with fingers and... thumb,

and at the bottom you see
some DNA polymerases have also

the gray exonuclease activity
that we'll talk about in a second.

So, dNTP arrives - dNTP is a single base
of the DNA

that still has the three
phosphates attached.

These three phosphates carry the energy
that will be used to bind this base.

Going forward,
it seems like the rate of attaching

the right base pair to the wrong one
is about ten to the four.

Again, we don't exactly understand
why that is.

Now, once... that base attached,
there's two possibilities

in the case where there is
an exonuclease activity.

We can either go forward
and attach the next base

or we can a go a bit backward,

move the double-stranded DNA
to a single-stranded DNA,

and move that to the exonuclease part
where the last base pair is removed.

And here, the wrong base pair
actually has an advantage

although this happens...
a hundred times more

when the wrong base pair is then
than the right one.

This together seems to give us a rate
that is close to ten to the ten.

We still don't exactly understand
how all these steps work.

But, it seems what we see is that,

when the base is in the pocket
of the DNA polymerase,

it seems that the shape of the base
and where its every atom is exactly,

determines whether the reaction
will go forward

or go towards the exonuclease activity.

So, the exact shape
of the DNA polymerase -

here I showed <i>Taq</i> polymerase.

The exact shape is important.

What kind of shapes
will the DNA polymerase encounter?

Of course, at first there's the four
regular Watson-Crick pairs.

So, there's C-A-T-G,
and they're pairs,

This gives us a total of four pairs
that DNA polymerase needs

to be able to recognize.

As we said, just the binding energy
gives us about a fifty-fold enrichment

for the right base versus the wrong one.

But, in addition, the right one
has three-fold disadvantage

versus the wrong types
because there's three times more

of the wrong base than the right base.

In addition to the four regular bases,
we said that at least two more bases,

but probably many others,
are recognized by DNA polymerase,

so that DNA can replicate
even if they're there.

So, that means that DNA polymerase

has at least six pairs
that it can recognize,

probably many more.

In addition, in the cell,
we also have many NTPs - not dNTPs.

So, NTPs don't have the oxygen removed.

These are the RNA bases.

They might go into the pocket
but have to not attach to DNA.

So, the pocket has to..
and it has like a small little,

supposedly a little part
that doesn't allow RNA to go in.

In addition, we talked about
all the non-standard bases

that can be around in the cell.

There are specific mechanisms
that lower the concentration

to a very low level so that
they don't interfere with DNA replication.

They could either stop it or they could
then be introduced into the DNA.

If we disable these mechanisms,
we do see that they then enter the DNA.

So, this is what a DNA polymerase
encounters when it replicates.

It needs to be able to recognize
several shapes and reject others.

In addition to the activity
of DNA polymerase,

there is activity of correcting errors
on the DNA.

These errors might both be introduced
by the DNA polymerase itself

during replication and through UV

and other chemical actions
that change the DNA itself.

The first of these mutation correction
mechanisms is DNA mismatch repair,

also called "MMR,"

where this amazing molecule
called "MutS" is able to recognize

areas where the DNA doesn't...
seem to have the right shape,

maybe it has a nick
or something like this.

Because the wrong base was inserted,
it recognized these places.

And then, another molecule,
MutL,

recognizes - at least in prokaryotes -

a location where
there's a CTAG with 6mA,

so a methylated amine.

When DNA replicates,

this methylation is not replicated,

and therefore it's possible to recognize
where the old strand is the new strand -

the new strand will not have
the A methylated.

And then, at that place...
the new strand is nicked,

this whole new strand is removed
up to the mismatch,

and then repolymerized
by DNA polymerase -

hopefully without an error.

So, this is one thing that removes errors
mainly during DNA replication.

Two other mechanisms are
direct reversal and base excision.

So, you see here 24 bases that are
pretty much specifically recognized

for these mechanisms.

Direct reversal - you reverse exactly
the high frequency errors

that can be introduced
by chemistry or UV

and you revert them
to the original base.

Bases excision repair -

you remove again
a recognized error,

and then that - or a region around it -

and then DNA polymerase
replicates again.

What we see in all these mechanisms

is that DNA actually does
error corrections through the fact that

every bit of information appears twice
on one strand and on the other strand.

So, if you know -
in the case of MutS -

if you know which strand is old
and which is new,

you can copy the more correct information
that doesn't have replication error again.

And, in these cases of reversal
or base excision,

again, you recognize the error
and through that are able to correct.

So, the error correction here

uses the fact that
DNA is double-stranded.

Let's summarize.

The point of these two lectures
was to show that

what is important in DNA
isn't really the bases.

It isn't the bases that carry
the information of how amazing DNA is.

We saw that, in additional to...
the four bases,

there's many more bases
that appear in the DNA

carrying information
or not carrying information,

and many more bases
appear in the cell.

We saw that dNTP,
the concentration of the bases -

these triple phosphate groups -

are tightly controlled by the cell

so that they will not interfere
with DNA replication.

If more bases were around
in the TP state,

the DNA would not replicate as well.

We saw that it is mainly DNA polymerase
that controls pairing

and recognizes what is the right pair
versus what is the wrong pair,

and it isn't really the energy
of the binding

between the pairs that does it.

And finally, we saw that
there is a large cloud,

a large number of specific
or not so specific enzymes,

that recognize wrong pairs
or wrong bases

and excise them or correct them.

And, only this whole machinery
of all these correction mechanisms

allow DNA to work as we see it,
with four bases that pair to each other

and replicate with a high fidelity.

Thank you.