1
00:00:01 --> 00:00:08
Good morning.
Good morning.

2
00:00:08 --> 00:00:14
So, I'd like to pick up where we
left off last time and just finish

3
00:00:14 --> 00:00:20
off translation and then step back
and look at how this central dogma

4
00:00:20 --> 00:00:26
of DNA is replicated into DNA, is
read into RNA, and is translated

5
00:00:26 --> 00:00:30
into protein. Or,
actually, as Francis Crick

6
00:00:30 --> 00:00:33
really put it, all
information flow from nucleic

7
00:00:33 --> 00:00:37
acid to protein. How that
varies amongst organisms.

8
00:00:37 --> 00:00:40
Because first we're going through
it and looking at the absolutely

9
00:00:40 --> 00:00:43
common features, DNA
replication, so it's five prime

10
00:00:43 --> 00:00:46
to three prime, et
cetera, et cetera,

11
00:00:46 --> 00:00:50
transcription, translation. But
in a moment I'd like to turn to

12
00:00:50 --> 00:00:53
the variations between
different kinds of organisms.

13
00:00:53 --> 00:00:56
But let me briefly finish up, if
I may, the bit about translation

14
00:00:56 --> 00:01:00
in general so we can
look at its variation.

15
00:01:00 --> 00:01:06
As we talked about last time,
we have a messenger RNA that has

16
00:01:06 --> 00:01:13
been transcribed from a specific
region of the chromosome starting at

17
00:01:13 --> 00:01:20
a promoter and going to
some stop of transcription.

18
00:01:20 --> 00:01:27
And that messenger RNA will
include some particular sequence,

19
00:01:27 --> 00:01:34
and I'll copy one here, A-U-A-C-G-A-U-G-A-A-G-A-G-G-C-C-C,

20
00:01:34 --> 00:01:41
et cetera, et cetera,
et cetera, out to a UAG.

21
00:01:41 --> 00:01:45
And this is the direction
five prime to three prime.

22
00:01:45 --> 00:01:49
We'll remember that all nucleic
acid polymerization goes five prime

23
00:01:49 --> 00:01:53
to three prime. So, what
happens is the cell begins

24
00:01:53 --> 00:01:57
scanning this message. And
it does that by this message

25
00:01:57 --> 00:02:01
being exported into the cytoplasm of
the cell. The ribosome coming along

26
00:02:01 --> 00:02:05
and glomming onto this
message and scanning on for the

27
00:02:05 --> 00:02:09
place to start.
It looks, it looks,

28
00:02:09 --> 00:02:13
it looks, it looks, and it
finds the first AUG. Footnote,

29
00:02:13 --> 00:02:18
this isn't 100% true. There are
occasional messages that start their

30
00:02:18 --> 00:02:22
translation not at an AUG,
and there are even occasional,

31
00:02:22 --> 00:02:27
there are even more messages that
don't quite start at the first AUG

32
00:02:27 --> 00:02:31
because the ribosome is really is
looking for something a little bit

33
00:02:31 --> 00:02:36
special, but to a first
order approximation.

34
00:02:36 --> 00:02:40
Good enough for the textbooks.
It goes along to the first AUG.

35
00:02:40 --> 00:02:44
In reality it's a little
more subtle than that.

36
00:02:44 --> 00:02:48
But it starts at the first AUG.
And what it does is it builds a

37
00:02:48 --> 00:02:52
protein that corresponds to it
according to a three letter genetic

38
00:02:52 --> 00:02:56
code. And you all know the
lookup table. It's in your book.

39
00:02:56 --> 00:03:00
AUG, always the first amino
acid put in. A methionine.

40
00:03:00 --> 00:03:04
Then AAG. Lysine, I
think. Then arginine.

41
00:03:04 --> 00:03:08
Then a proline. Now, I mean
this is this particular sequence.

42
00:03:08 --> 00:03:12
Any other sequence would
be different. Et cetera.

43
00:03:12 --> 00:03:17
How does it accomplish this
matching between three letters of

44
00:03:17 --> 00:03:21
the genetic code? Oh,
and when it gets to AUG,

45
00:03:21 --> 00:03:25
that is one of the three singles
for stop, don't put in any more amino

46
00:03:25 --> 00:03:30
acids. There are three
such stop signals.

47
00:03:30 --> 00:03:37
AUG, sorry, UAG, UGG and
U, oops, what did I just do

48
00:03:37 --> 00:03:44
here? Let's get that right.
UAG, UGG and UGA. Those are the

49
00:03:44 --> 00:03:51
three stop codons. So, how
many total codons are there?

50
00:03:51 --> 00:03:59
64 codons. Three of them
spell stop. 61 of them spell

51
00:03:59 --> 00:04:05
specific amino acids. And how
many amino acids are there?

52
00:04:05 --> 00:04:11
20. So, the average redundancy
is three. Some are specified by

53
00:04:11 --> 00:04:16
multiple codons. The
most extreme is some amino

54
00:04:16 --> 00:04:21
acids are specified by as
many as six codons. Did I,

55
00:04:21 --> 00:04:27
oh, thank you. Come back down. Of
course. U-A, so it's UAG, right?

56
00:04:27 --> 00:04:32
Sorry, UAA and
UGA and UAG.

57
00:04:32 --> 00:04:37
Thank you. Very good. All right.
So, now, how does it accomplish

58
00:04:37 --> 00:04:42
this feat of taking amino acids,
of taking nucleotide sequence, RNA

59
00:04:42 --> 00:04:47
sequence and converting it into
the sequence of amino acids?

60
00:04:47 --> 00:04:52
As I mentioned last time, there
was lots of original somewhat

61
00:04:52 --> 00:04:57
nutty thinking about some looping
codes that would make the RNA fold

62
00:04:57 --> 00:05:03
up in such a way to bind
the amino acids and all that.

63
00:05:03 --> 00:05:12
But, as Francis Crick thought up,
there had to be some kind of an

64
00:05:12 --> 00:05:21
adapter molecule that would take
the RNA sequence and would somehow

65
00:05:21 --> 00:05:30
connect it up to the correct
amino acid, and that was UAC.

66
00:05:30 --> 00:05:35
A particular transfer RNA molecule.
And the tRNA molecule is an adapter

67
00:05:35 --> 00:05:40
sequence that has three nucleotides
here that match up to the three

68
00:05:40 --> 00:05:45
nucleotides of the codon that
we're trying to translate,

69
00:05:45 --> 00:05:50
and it has the appropriate amino
acid that's been stuck on the end of

70
00:05:50 --> 00:05:55
it. And how does it get there?
How does the right tRNA, the tRNA

71
00:05:55 --> 00:06:00
to match this codon have the
right amino acid put on it?

72
00:06:00 --> 00:06:03
There's a dedicated enzyme that
recognizes that tRNA and puts on

73
00:06:03 --> 00:06:07
that amino acid. It's
aminoacyl-tRNA synthetase.

74
00:06:07 --> 00:06:11
It sticks the right amino
on the right transfer RNA.

75
00:06:11 --> 00:06:15
So, that's how it accomplishes the
physical recognition of these three

76
00:06:15 --> 00:06:18
bases and has the right
amino acid attached to it.

77
00:06:18 --> 00:06:22
There's an enzymatic machinery
that has all of these tRNAs floating

78
00:06:22 --> 00:06:26
around in the cell which can be
used for this translation here.

79
00:06:26 --> 00:06:30
How does this actually
happen physically?

80
00:06:30 --> 00:06:35
It happens in this vast
machine called the ribosome.

81
00:06:35 --> 00:06:41
In the ribosome, if we have,
say, our codon here and we have a

82
00:06:41 --> 00:06:47
tRNA that, well, we'll
put that actually in the

83
00:06:47 --> 00:06:53
ribosome that, say, has
the first amino acid here,

84
00:06:53 --> 00:06:59
methionine, there's a cavity
for this guy and there's a cavity

85
00:06:59 --> 00:07:05
for the next guy. And other
tRNAs come into the cell

86
00:07:05 --> 00:07:11
carrying their next amino acid.
Maybe it will be here a lysine that

87
00:07:11 --> 00:07:17
matches up with the codon and the
anti-codon. And when the right tRNA

88
00:07:17 --> 00:07:23
fits in the next cavity over,
the ribosome itself catalyzes a

89
00:07:23 --> 00:07:30
peptide bond between
these amino acids.

90
00:07:30 --> 00:07:35
Then it chugs over by one, it
translocates by one moving this

91
00:07:35 --> 00:07:40
bit of the complex to the left,
and the peptide chain continues to

92
00:07:40 --> 00:07:45
grow out this end as each new
codon is moved into position,

93
00:07:45 --> 00:07:50
a tRNA comes in bringing the right
amino acid until finally a stop

94
00:07:50 --> 00:07:55
codon is hit. And what happens
when you hit a stop codon?

95
00:07:55 --> 00:08:00
It stops. And is there
a tRNA for a stop?

96
00:08:00 --> 00:08:02
It turns out there's not. There
actually isn't. There's some

97
00:08:02 --> 00:08:05
other factor. There's a protein
factor that helps recognize the

98
00:08:05 --> 00:08:08
stops. So, that just continues
to chug on. Those of you who are

99
00:08:08 --> 00:08:11
computer scientists or
mathematicians will recognize this

100
00:08:11 --> 00:08:14
is a two-tape Turing machine.
It is the small two-tape Turing

101
00:08:14 --> 00:08:17
machine that I know to exist. If
you don't know what that means,

102
00:08:17 --> 00:08:20
you can forget about that comment.
In any case, but some of you know

103
00:08:20 --> 00:08:23
what that is. So,
that's how it proceeds.

104
00:08:23 --> 00:08:26
That is your basic
protein translation.

105
00:08:26 --> 00:08:28
And, I must say, what I
really love about this was

106
00:08:28 --> 00:08:31
that Francis Crick kind of figured
out what had to happen just on first

107
00:08:31 --> 00:08:34
principles and was able to think
through it much more clearly and

108
00:08:34 --> 00:08:37
direct people to know what
to look for in the laboratory.

109
00:08:37 --> 00:08:40
And if people had not had the
clarity of thinking that Crick

110
00:08:40 --> 00:08:43
provided by saying, look,
there's got to be this kind of

111
00:08:43 --> 00:08:46
adapter, I don't think they
would have found it as quickly.

112
00:08:46 --> 00:08:49
But once he said this is
what you've got to look for,

113
00:08:49 --> 00:08:52
golly, it was there. You
can't do that very often,

114
00:08:52 --> 00:08:55
but Francis Crick seemed to have
a very good track record of doing

115
00:08:55 --> 00:08:58
those things. OK. So,
that was just finishing off

116
00:08:58 --> 00:09:02
translation. Now what
I'd like to do is turn to

117
00:09:02 --> 00:09:08
variations on the theme as
the major issue for today.

118
00:09:08 --> 00:09:14
How does this central dogma, DNA
replicates, is transcribed into

119
00:09:14 --> 00:09:20
RNA and is translated into protein,
vary amongst the different kinds of

120
00:09:20 --> 00:09:26
organisms that we
might be interested in?

121
00:09:26 --> 00:09:32
The kinds of organisms
we might be interested in,

122
00:09:32 --> 00:09:39
eukaryotes, prokaryotes,
viruses. Sample eukaryote,

123
00:09:39 --> 00:09:46
MIT undergraduate. Prokaryote,
E. coli. And virus, many possible

124
00:09:46 --> 00:09:53
viruses. The eukaryotes'
big nucleated cells.

125
00:09:53 --> 00:10:00
So, in here we're going to
have our nucleated cells.

126
00:10:00 --> 00:10:04
DNA living in there. In
our prokaryotes we have no

127
00:10:04 --> 00:10:09
distinct nucleus. The
DNA is not in a distinct

128
00:10:09 --> 00:10:14
nucleus, although it's not
entirely freely floating around.

129
00:10:14 --> 00:10:19
It tends to be clustered together.
In the virus the nucleic acid

130
00:10:19 --> 00:10:24
resides in some kind of a capsid,
some kind of a, it could be a

131
00:10:24 --> 00:10:29
protein capsid. There
are some of them that have

132
00:10:29 --> 00:10:34
lipid capsids with lipid particles
around them, but some kind of a coat

133
00:10:34 --> 00:10:39
around nucleic acid there. Do
they all do exactly the same

134
00:10:39 --> 00:10:44
things with regard to DNA
replication, RNA transcription and

135
00:10:44 --> 00:10:49
protein translation?
Well, not entirely. So,

136
00:10:49 --> 00:10:54
as a way, in a way to reinforce
what we know about these,

137
00:10:54 --> 00:11:00
let's look at how they differ.
DNA replication. Eukaryotes.

138
00:11:00 --> 00:11:06
What's the structure of one of
your chromosomes? Is it a long line,

139
00:11:06 --> 00:11:12
a long linear molecule, or
is it a circular molecule?

140
00:11:12 --> 00:11:18
How many of you have linear
chromosomes? How many of you have

141
00:11:18 --> 00:11:24
circular chromosomes? I heard
there were some people with

142
00:11:24 --> 00:11:30
circular. And how many of you
are unsure about your chromosomes?

143
00:11:30 --> 00:11:38
OK. That's good. Well,
then I'm pleased to inform

144
00:11:38 --> 00:11:47
you that you have long linear
chromosomes. Every human chromosome

145
00:11:47 --> 00:11:55
is a long double-stranded molecule
of DNA. Linear double-stranded DNA.

146
00:11:55 --> 00:12:02
They can be extremely long.
You have 23 chromosomes,

147
00:12:02 --> 00:12:08
and together they make up three
billion nucleotides of DNA.

148
00:12:08 --> 00:12:13
A typical chromosome could be 150
million bases long as an average

149
00:12:13 --> 00:12:19
size for a chromosome.
And it's a single connected

150
00:12:19 --> 00:12:24
molecule. 150 million bases long in
the human is a typical chromosome.

151
00:12:24 --> 00:12:30
One tricky little bit
about replicating DNA.

152
00:12:30 --> 00:12:34
Let's just think back to our
little model of replicating DNA.

153
00:12:34 --> 00:12:38
Let's come to the chromosome end
here. It's five prime to three

154
00:12:38 --> 00:12:42
prime. Five prime to three prime.
We're going to start replicating.

155
00:12:42 --> 00:12:47
We're getting to the end
of chromosome number one.

156
00:12:47 --> 00:12:51
We've got a primer here, and
the primer is going to be used

157
00:12:51 --> 00:12:55
to extend, extend, extend.
We get right to the end.

158
00:12:55 --> 00:13:00
That's good. Tell me how
we're going to replicate back.

159
00:13:00 --> 00:13:04
We need a little primer to start
it, right? And where's that primer

160
00:13:04 --> 00:13:09
going to land? Maybe
over here it will start

161
00:13:09 --> 00:13:14
replicating back. Oh,
boy, we haven't done this

162
00:13:14 --> 00:13:19
figure. So, what do we have to
do there? So, we need to primer a

163
00:13:19 --> 00:13:24
little further back.
OK. But, you know what,

164
00:13:24 --> 00:13:29
the chance that we're going
to get that right at the end,

165
00:13:29 --> 00:13:34
that we're going to get a primer
exactly at the end is pretty low.

166
00:13:34 --> 00:13:37
And if we don't have a
primer exactly at the end,

167
00:13:37 --> 00:13:41
what's going to be wrong with
that copy of the chromosome?

168
00:13:41 --> 00:13:45
Too short. Now, big deal. So,
it's short by maybe 20 bases.

169
00:13:45 --> 00:13:49
But that's just this cell division.
What about next cell division? It

170
00:13:49 --> 00:13:53
will be short on average by a little
bit, and then the next cell division

171
00:13:53 --> 00:13:57
and the next cell division.
It's actually pretty tricky to

172
00:13:57 --> 00:14:01
replicate a linear chromosome
on the lagging strand,

173
00:14:01 --> 00:14:05
unless you can land the primer
in exactly the right place,

174
00:14:05 --> 00:14:09
which doesn't happen. So,
a special little solution is

175
00:14:09 --> 00:14:15
used. The ends of chromosomes
here are called telomeres,

176
00:14:15 --> 00:14:21
telo meaning end. These telomeres
have very specific structures.

177
00:14:21 --> 00:14:27
In the human they repeat,
T-T-A-G-G-G, again and

178
00:14:27 --> 00:14:32
again and again. At the end
of the chromosome there's

179
00:14:32 --> 00:14:36
a special enzyme that will come
along and add some extra telomere to

180
00:14:36 --> 00:14:41
the chromosome. That,
sorry? Did I say leading

181
00:14:41 --> 00:14:46
strand? It's the, oh, yeah,
sorry. It's the lagging,

182
00:14:46 --> 00:14:50
sorry. It's the leading strand. No,
no, no, this is the lagging strand.

183
00:14:50 --> 00:14:55
This is the leading strand because
it's running along happily not

184
00:14:55 --> 00:15:00
having to make a primer. The
okazaki fragment should be here.

185
00:15:00 --> 00:15:04
I'll stick by that.
We'll debate it later.

186
00:15:04 --> 00:15:08
Anyway, they, we get the point.
But it's lagging because you've got

187
00:15:08 --> 00:15:12
the ogzocy fragments there.
So, anyway, we have a problem of

188
00:15:12 --> 00:15:16
replication. And the way the cell
solves it is the actual replication

189
00:15:16 --> 00:15:20
is shorter, but since it manages to
stick some repeat at the end of the

190
00:15:20 --> 00:15:24
chromosome it adds back some
more T-T-A-G-G-G, T-T-A-G-G-G,

191
00:15:24 --> 00:15:29
T-T-A-G-G-G, and it keeps
dynamically adding more.

192
00:15:29 --> 00:15:33
What do you think would happen if
you didn't, or what's the enzyme

193
00:15:33 --> 00:15:37
that adds telomeres?
Telomerase. Telomerase adds that.

194
00:15:37 --> 00:15:41
What cells do you think need
to have active telomerase?

195
00:15:41 --> 00:15:45
Rapidly dividing cells would
need to have telomerase.

196
00:15:45 --> 00:15:49
Cells that are not rapidly dividing,
cells that have stopped dividing can

197
00:15:49 --> 00:15:53
shut off their telomerase.
But if a cell is going to go

198
00:15:53 --> 00:15:57
through lots and lots of
cell divisions it's got to,

199
00:15:57 --> 00:16:01
it's got to tidy up its telomeres
each time because they're

200
00:16:01 --> 00:16:06
getting too short. You've
got to have an enzyme that's

201
00:16:06 --> 00:16:10
adding back ends of chromosomes.
What cells do you think

202
00:16:10 --> 00:16:14
particularly care about
having telomerase on them?

203
00:16:14 --> 00:16:19
Cancers. It turns out that
this is not a trivial point.

204
00:16:19 --> 00:16:23
More than 90% of cancers turn
on actively the telomerase gene,

205
00:16:23 --> 00:16:27
which would be a shut off in
normal cells because the cell is

206
00:16:27 --> 00:16:32
not dividing anymore. Part
of becoming a cancer is having

207
00:16:32 --> 00:16:36
to turn on this repair mechanism for
the ends, this extension mechanism

208
00:16:36 --> 00:16:40
for the ends of your chromosomes.
And so, various people are trying

209
00:16:40 --> 00:16:44
to make drugs to inhibit cancers by
inhibiting this telomerase enzyme.

210
00:16:44 --> 00:16:49
So, understanding just your linear
replication of chromosomes is a kind

211
00:16:49 --> 00:16:53
of useful thing even in
dealing with things like cancer.

212
00:16:53 --> 00:16:57
Genome sizes. I mentioned,
how big was the human genome?

213
00:16:57 --> 00:17:02
Three times ten to the ninth
bases. The mouse genome?

214
00:17:02 --> 00:17:06
It's almost as big, about
2.7 times ten to the ninth

215
00:17:06 --> 00:17:11
bases, 2.7 million bases. The
elephant genome? I actually

216
00:17:11 --> 00:17:15
just found this out last week
because we just finished sequencing

217
00:17:15 --> 00:17:20
elephant DNA, and I can
now tell you I think it's 3.

218
00:17:20 --> 00:17:25
. The dog is 2. times
ten to the ninth.

219
00:17:25 --> 00:17:29
Anyway, it's about, for most
mammals it's pretty close

220
00:17:29 --> 00:17:33
to three billion bases. And
there is some fluctuation.

221
00:17:33 --> 00:17:37
Some are a little bigger.
Some are a little smaller.

222
00:17:37 --> 00:17:41
It doesn't scale with
sizing the animal, though,

223
00:17:41 --> 00:17:45
because the dog has a smaller genome,
for example, than the mouse does,

224
00:17:45 --> 00:17:48
but the elephant is a big bigger
than us. And check in later in the

225
00:17:48 --> 00:17:52
term, I'll tell you about the
aardvark. We should know in a

226
00:17:52 --> 00:17:56
little while. But here are, for
example, fruit flies. The fruit

227
00:17:56 --> 00:18:00
fly, it has a genome of
two times ten to the eighth.

228
00:18:00 --> 00:18:04
I'm giving, I'm being
quite approximate. In fact,

229
00:18:04 --> 00:18:08
I'll make it, I'll give you
1. times ten to the eighth. 150

230
00:18:08 --> 00:18:12
million bases.
Yeast, by contrast,

231
00:18:12 --> 00:18:17
has a genome of 1.2 times ten to
the seventh. So, that's 12 million,

232
00:18:17 --> 00:18:21
150 million give or take,
and about three billion,

233
00:18:21 --> 00:18:25
so 3,000 million. So,
genome sizes can vary quite

234
00:18:25 --> 00:18:30
dramatically amongst
different eukaryotes.

235
00:18:30 --> 00:18:36
Now, what about prokaryotes?
How do the prokaryotes differ?

236
00:18:36 --> 00:18:43
Prokaryotes differ because their
genomes are typically not linear

237
00:18:43 --> 00:18:49
chromosomes. The typical
prokaryotic chromosome is a

238
00:18:49 --> 00:18:56
double-stranded circle. It's
a double-stranded circular DNA.

239
00:18:56 --> 00:19:02
Now, the
double-stranded circular

240
00:19:02 --> 00:19:07
DNA doesn't have this problem
of telomeres. You just keep

241
00:19:07 --> 00:19:12
replicating around and you get to
the end. So, there you have a much

242
00:19:12 --> 00:19:17
simpler replication system than
having to worry about your ends of

243
00:19:17 --> 00:19:22
chromosomes. You also
have much smaller genomes.

244
00:19:22 --> 00:19:27
The typical prokaryotic genome size,
it's on the order of a few million

245
00:19:27 --> 00:19:31
bases. E. coli, 4.6 million
bases. There are, for example,

246
00:19:31 --> 00:19:35
mycobacteria, such as the
mycobacteria that caused

247
00:19:35 --> 00:19:39
tuberculosis or leprosy,
have on the order of, well,

248
00:19:39 --> 00:19:43
actually, not quite them, but other
mycobacteria have on the order of

249
00:19:43 --> 00:19:46
about a million bases or so.
Mycobacteria, M. genitalia has

250
00:19:46 --> 00:19:50
actually slightly less
than a million basis.

251
00:19:50 --> 00:19:54
So, these are basically
several million bases.

252
00:19:54 --> 00:19:58
So, there's a huge
variation in genome size.

253
00:19:58 --> 00:20:02
Your genome is about a
thousand times bigger than E.

254
00:20:02 --> 00:20:07
coli's genome. Now, you do actually
have one circular chromosome.

255
00:20:07 --> 00:20:11
Do you know what it is? I speak
about the 23 pairs of human

256
00:20:11 --> 00:20:16
chromosomes. There's actually
one more human chromosome.

257
00:20:16 --> 00:20:20
The mitochondria have their
own chromosome. It's a circle.

258
00:20:20 --> 00:20:25
That's very odd that you would have
one chromosome that's a circle that

259
00:20:25 --> 00:20:30
looks like a bacterial chromosome.
Do you know why that is?

260
00:20:30 --> 00:20:34
The mitochondria arose as a
symbiotic bacterium that became a

261
00:20:34 --> 00:20:38
symbiont of eukaryotic cells
about 1. billion years ago.

262
00:20:38 --> 00:20:42
It was a bacterium taken up
into another cell, and that's how

263
00:20:42 --> 00:20:46
eukaryotes evolved. And
we can even see that little

264
00:20:46 --> 00:20:50
signature of it having been a
prokaryote from the fact that it's

265
00:20:50 --> 00:20:54
got one of these circular
prokaryotic looking chromosomes.

266
00:20:54 --> 00:20:58
Now, it, because it's living in
your cells, has thrown out all sorts

267
00:20:58 --> 00:21:02
of genes that it doesn't
need anymore because the main,

268
00:21:02 --> 00:21:06
the nucleus supplies
most of the proteins.

269
00:21:06 --> 00:21:10
So, your mitochondrial genome
is a circle that's a mere 16,

270
00:21:10 --> 00:21:14
00 bases long. It's a very small
circle encoding a very limited

271
00:21:14 --> 00:21:18
number of genes,
but it's, in fact,

272
00:21:18 --> 00:21:22
the residue of the bacterial
symbiont that lead to the formation

273
00:21:22 --> 00:21:26
of euks. Now, viruses,
what do viruses have?

274
00:21:26 --> 00:21:30
Do they have double-strained
linear chromosomes? Which is it?

275
00:21:30 --> 00:21:38
Is it double-stranded linear DNA or
is it double-stranded circular DNA?

276
00:21:38 --> 00:21:46
Circular DNA. So, who votes for
linear? Who votes for circular?

277
00:21:46 --> 00:21:54
Who's undecided? Ah, the
undecided are very larger here. So,

278
00:21:54 --> 00:22:01
the answer is both. Some
viruses have double-stranded

279
00:22:01 --> 00:22:07
linear DNA. Some viruses have
double-stranded circular DNA.

280
00:22:07 --> 00:22:14
It's worse than that, though.
Some viruses have single-stranded

281
00:22:14 --> 00:22:20
linear, circular DNA.
Ha? How does that work?

282
00:22:20 --> 00:22:26
Some viruses actually infect
the cell injecting DNA,

283
00:22:26 --> 00:22:32
and it's just single-stranded.
As soon as it gets into the cell,

284
00:22:32 --> 00:22:36
however, it's replicated to make a
double-stranded DNA which can then

285
00:22:36 --> 00:22:41
be transcribed, et
cetera, et cetera,

286
00:22:41 --> 00:22:46
et cetera. But it travels around
as a single-stranded piece of DNA.

287
00:22:46 --> 00:22:50
And it's actually weirder
than that. Some viruses,

288
00:22:50 --> 00:22:55
viruses being very small
can experiment with all

289
00:22:55 --> 00:23:02
sorts of things. Some viruses
actually consist not of

290
00:23:02 --> 00:23:10
DNA at all but of RNA,
single-stranded RNA. How does it do

291
00:23:10 --> 00:23:18
that? So, in other words, in
the capsid there's a single

292
00:23:18 --> 00:23:26
strand of RNA. When
it gets into the cell,

293
00:23:26 --> 00:23:32
what does it do?
Sorry? It creates DNA.

294
00:23:32 --> 00:23:36
How does it create DNA? From
the RNA. How's it going to do

295
00:23:36 --> 00:23:41
that? Well, how is it going
to turn itself into DNA?

296
00:23:41 --> 00:23:46
It needs an enzyme to do that?
Reverse transcriptase. You would

297
00:23:46 --> 00:23:50
like to reverse the transcription
process, and you would like to name

298
00:23:50 --> 00:23:55
that reverse transcriptase. And
where are you going to get this

299
00:23:55 --> 00:24:00
reverse transcriptase from?
Laying around. Laying around where?

300
00:24:00 --> 00:24:04
I mean the cell is just sitting
there with reverse transcriptase

301
00:24:04 --> 00:24:08
waiting to obligingly
reverse transcribe this virus?

302
00:24:08 --> 00:24:12
Your RNA. So make it how? With
ribosomes. So, in other words,

303
00:24:12 --> 00:24:17
if I'm an RNA, why don't
I encode the sequence for

304
00:24:17 --> 00:24:21
reverse transcriptase and
actually translate myself.

305
00:24:21 --> 00:24:25
So, if you were really cleaver,
you might decide to put in the

306
00:24:25 --> 00:24:30
genetic code for
reverse transcriptase.

307
00:24:30 --> 00:24:36
And when that message gets into the
cell, it will first act as an mRNA,

308
00:24:36 --> 00:24:42
a messenger RNA, translate, make,
here's the reverse transcriptase

309
00:24:42 --> 00:24:48
enzyme, which is then going to go,
and it's going to reverse transcribe

310
00:24:48 --> 00:24:54
this thing into,
say, DNA. So, wow.

311
00:24:54 --> 00:25:00
Now, that's a good one.
This is a plus strand virus.

312
00:25:00 --> 00:25:05
It encodes its own reverse
transcriptase in its instructions.

313
00:25:05 --> 00:25:10
There actually are
minus-strand viruses that don't,

314
00:25:10 --> 00:25:15
but what they do is instead in their
own code, in their own package bring

315
00:25:15 --> 00:25:20
a longer reverse transcriptase.
So, either you can encode your own

316
00:25:20 --> 00:25:25
reverse transcriptase or in the
package you can include your own

317
00:25:25 --> 00:25:30
reverse transcriptase.
Do you know any viruses?

318
00:25:30 --> 00:25:36
And then the reverse transcriptase
is then used to transcribe the DNA,

319
00:25:36 --> 00:25:43
the RNA into DNA, and eventually
into a double-stranded DNA which,

320
00:25:43 --> 00:25:49
in some of the viruses, can then be
slammed into and inserted into your

321
00:25:49 --> 00:25:56
own chromosomes. So, a DNA
copy of the virus can be

322
00:25:56 --> 00:26:03
installed into your own chromosomes,
which is somewhat insidious.

323
00:26:03 --> 00:26:08
What viruses do you know that
do this? HIV. More generally

324
00:26:08 --> 00:26:13
retroviruses are the class
of these viruses that can,

325
00:26:13 --> 00:26:19
in fact, run this replication
process from RNA to DNA and install

326
00:26:19 --> 00:26:24
DNA copies of them in your genome.
And how do you then get the DNA

327
00:26:24 --> 00:26:30
copy out of your
genome? You don't.

328
00:26:30 --> 00:26:33
It doesn't come out.
Retroviral insertions don't come

329
00:26:33 --> 00:26:37
out. That's one of the issues in
dealing with AIDS is once this DNA

330
00:26:37 --> 00:26:40
copy is in a cell it's not coming
out. We have no way to remove it.

331
00:26:40 --> 00:26:44
We have to make sure that the virus
is shut down by other mechanisms

332
00:26:44 --> 00:26:47
that might inhibit its products,
et cetera, but once its stuck a DNA

333
00:26:47 --> 00:26:51
copy into your chromosomes, you
know, there's no way of getting

334
00:26:51 --> 00:26:55
it out. So, if we had
to try to inhibit the

335
00:26:55 --> 00:27:00
action of the AIDS virus, we
might wish to make inhibitors of

336
00:27:00 --> 00:27:05
this aspect of replication,
inhibitors or reverse transcription.

337
00:27:05 --> 00:27:10
And, of course, as probably many of
you know, some of the important AIDS

338
00:27:10 --> 00:27:15
drugs are reverse transcriptase
inhibitors, very important to

339
00:27:15 --> 00:27:20
limiting the replication of the
AIDS virus. And there are many other

340
00:27:20 --> 00:27:25
kinds of weirdnesses.
Viruses pretty much explore,

341
00:27:25 --> 00:27:30
everything you possibly can do,
viruses come up with ways to do.

342
00:27:30 --> 00:27:35
Let's take now the
process of transcription.

343
00:27:35 --> 00:27:40
We have replication up there.
Let's look at transcription. And

344
00:27:40 --> 00:27:45
this time let's start with
prokaryotes. For the simple aspect

345
00:27:45 --> 00:27:50
of transcribing genes, the
prokaryotic genome looks just

346
00:27:50 --> 00:27:55
like the simple model I gave you.
There is some kind of a promoter

347
00:27:55 --> 00:28:00
that tells RNA polymerase
to come sit down here.

348
00:28:00 --> 00:28:07
RNA polymerase hops on, RNA
polymerase begins to copy in RNA,

349
00:28:07 --> 00:28:15
and eventually it hits the signal
that says to terminate transcription.

350
00:28:15 --> 00:28:22
OK. This is not a stop codon
which is about translation.

351
00:28:22 --> 00:28:30
This is a termination
of transcription.

352
00:28:30 --> 00:28:35
And this RNA then goes off.
A perfectly happy thing, a

353
00:28:35 --> 00:28:40
messenger RNA, mRNA.
So, there's nothing weird

354
00:28:40 --> 00:28:46
about proks compared to the simple
description that we gave before.

355
00:28:46 --> 00:28:51
But eukaryotes are different.
There are some funny things that

356
00:28:51 --> 00:28:57
happen in the eukaryote. Well,
first off it starts the same.

357
00:28:57 --> 00:29:03
There's a promoter. RNA
polymerase sits down there,

358
00:29:03 --> 00:29:09
it starts transcribing, it makes
an mRNA, it hits the transcriptional

359
00:29:09 --> 00:29:16
termination signal, it
stops, and then this RNA gets

360
00:29:16 --> 00:29:22
processed in interesting ways.
The first thing that happens is

361
00:29:22 --> 00:29:29
three modifications happen. The
first one is at the five prime

362
00:29:29 --> 00:29:35
end, remember five prime to three
prime, a funny modification is put

363
00:29:35 --> 00:29:41
on. It's a, if the message, say,
were A-U-C-U-G-G-C et cetera,

364
00:29:41 --> 00:29:47
a G triphosphate is put on
backwards. It's actually a methyl G

365
00:29:47 --> 00:29:53
triphosphate is put on backwards,
so going in the other direction.

366
00:29:53 --> 00:30:00
You have the triphosphate
bond there, a methyl G.

367
00:30:00 --> 00:30:04
And the only thing that
you share care about that,

368
00:30:04 --> 00:30:09
I don't care if you know the
structure, is that there's a funny

369
00:30:09 --> 00:30:13
cap. This thing is called a
cap that is put on this message.

370
00:30:13 --> 00:30:18
And that cap is very important
to signaling to the cell this is a

371
00:30:18 --> 00:30:23
messenger RNA to be dealt with in a
certain way, to get the ribosome to

372
00:30:23 --> 00:30:27
hop on, to get this thing
processed properly, et cetera.

373
00:30:27 --> 00:30:32
At the other end of the message
a long string of As is added

374
00:30:32 --> 00:30:37
to messenger RNAs. This
long string of As is called,

375
00:30:37 --> 00:30:41
very sensibly, a poly A tail.
The poly A tail is added to the

376
00:30:41 --> 00:30:46
messenger RNA,
and very often,

377
00:30:46 --> 00:30:51
I mean it's, if you wanted to purify
messenger RNAs from your own human

378
00:30:51 --> 00:30:55
cells, you can actually use poly T
as a reagent because it turns out,

379
00:30:55 --> 00:31:00
because messenger RNAs have
a poly A tail, they'll bind to

380
00:31:00 --> 00:31:04
and stick to poly T. So,
people actually purify messenger

381
00:31:04 --> 00:31:08
RNAs by binding them to poly
T and they get the poly A tail.

382
00:31:08 --> 00:31:11
But it is broadly believed that the
reason for this poly A tail is not

383
00:31:11 --> 00:31:15
to make things convenient for
molecular biologists to purify

384
00:31:15 --> 00:31:18
messages. To the contrary, it
is an important function for the

385
00:31:18 --> 00:31:22
cell. And it turns out that this
is important in regulating the

386
00:31:22 --> 00:31:25
stability of messages. If,
in fact, you don't have a poly

387
00:31:25 --> 00:31:29
A tail, if you contrive to make the
same message without the poly A tail,

388
00:31:29 --> 00:31:33
the message will be
degraded rather rapidly.

389
00:31:33 --> 00:31:36
And the lengths of the poly A tails
control aspects of the degradation,

390
00:31:36 --> 00:31:39
et cetera. So, in a
complex eukaryotic cell,

391
00:31:39 --> 00:31:43
already it's how to attach
a little signal at the front,

392
00:31:43 --> 00:31:46
some signals at the back that
says process me in a certain way,

393
00:31:46 --> 00:31:49
et cetera, don't degrade me yet.
You could even imagine that this

394
00:31:49 --> 00:31:53
poly A tail could serve as a little
bit of a clock for how long that

395
00:31:53 --> 00:31:56
message sticks around. It's
not quite that simple but

396
00:31:56 --> 00:31:59
there are ways to do it. But
all of these pale in comparison

397
00:31:59 --> 00:32:03
to the third way in which eukaryotic
messages differ from prokaryotic

398
00:32:03 --> 00:32:10
messages. These small
modifications are,

399
00:32:10 --> 00:32:21
as I say, small. The most striking
way in which they differ is that

400
00:32:21 --> 00:32:33
only a small portion often
of the gene, here's my gene,

401
00:32:33 --> 00:32:44
matters for the protein that
is made. So, my mRNA gets made.

402
00:32:44 --> 00:32:54
It includes the whole long sequence.
And then the cell comes along and

403
00:32:54 --> 00:33:05
splices this message together.
So, this is the immature RNA.

404
00:33:05 --> 00:33:13
It is processed by clipping
out this, clipping out this,

405
00:33:13 --> 00:33:22
clipping out this. And what you get
is a splice where the mature message

406
00:33:22 --> 00:33:30
throws this stuff out,
splices between here and here,

407
00:33:30 --> 00:33:39
splices here, splices here,
splices here, and you get a

408
00:33:39 --> 00:33:47
much shorter mRNA. And
this is a mature mRNA.

409
00:33:47 --> 00:33:53
This splicing is a remarkable
phenomenon. In fact,

410
00:33:53 --> 00:34:00
it was discovered by Phil Sharp
here, for which he won a Nobel prize.

411
00:34:00 --> 00:34:04
This splicing is a very
complex operation. First off,

412
00:34:04 --> 00:34:08
how does, well, actually,
what accomplishes splicing?

413
00:34:08 --> 00:34:12
It should be splicase, right?
But it turns out it's not a single

414
00:34:12 --> 00:34:17
enzyme. It's a big body of stuff.
So, instead it's the splicosome, OK.

415
00:34:17 --> 00:34:21
Everything is either ase or
some or something like that.

416
00:34:21 --> 00:34:25
So, it turns out it's the
splicosome that does that.

417
00:34:25 --> 00:34:30
It's just wonderful how all those
names work out. The splicosome.

418
00:34:30 --> 00:34:36
The splicosome comes along and
splices it. How does the splicosome

419
00:34:36 --> 00:34:42
know how to do this? Well,
there are kind of codes.

420
00:34:42 --> 00:34:48
It turns out that there are some
information encoded along in these

421
00:34:48 --> 00:34:54
messages. It turns out that
there is, you know, slight biases.

422
00:34:54 --> 00:35:00
Typically the sequence just after
where the slice starts here is a GU

423
00:35:00 --> 00:35:06
and the sequence here is an AG,
but that's obviously not enough

424
00:35:06 --> 00:35:10
information, right? It's not
enough bases of information

425
00:35:10 --> 00:35:14
to get this right. And
so there's a little more

426
00:35:14 --> 00:35:18
preferences for what bases use, but
the truth is we don't fully know.

427
00:35:18 --> 00:35:21
Our best picture right now
involves some cellular factors help

428
00:35:21 --> 00:35:25
recognizing the parts that are
supposed to stay in some sequences

429
00:35:25 --> 00:35:29
here. But the truth is we
don't have the simple codes.

430
00:35:29 --> 00:35:33
Because if we had the simple codes,
we'd be able to take a long stretch

431
00:35:33 --> 00:35:37
of DNA and figure out exactly
where the splices go based on just

432
00:35:37 --> 00:35:42
computer analysis. And
we can't do that so well.

433
00:35:42 --> 00:35:46
These bits that stay in are called
exons. The bits that go out are

434
00:35:46 --> 00:35:51
called introns. This is
the source of extraordinary

435
00:35:51 --> 00:35:55
confusion for students because you
might think that the bits that are

436
00:35:55 --> 00:36:00
excised are the
exons, but they're not.

437
00:36:00 --> 00:36:04
The bits that stay in are the exons.
Why are they called exons if they

438
00:36:04 --> 00:36:08
stay in and ex is a prefix meaning
out? Well, because the introns are

439
00:36:08 --> 00:36:12
named because they're intervening
sequences. Once the introns,

440
00:36:12 --> 00:36:17
the intervening sequences were named
as intervening sequences or introns,

441
00:36:17 --> 00:36:21
you were stuck then having to name
the things that stay in as exons.

442
00:36:21 --> 00:36:25
This was all done by a Harvard
professor, don't blame me.

443
00:36:25 --> 00:36:30
In any case, a good
friend Harvard professor.

444
00:36:30 --> 00:36:37
But, nonetheless, I'm not
sure that this was the best

445
00:36:37 --> 00:36:44
way to name them. But
you're stuck with it.

446
00:36:44 --> 00:36:52
So, for a typical human gene,
typical human gene, the length of

447
00:36:52 --> 00:36:59
the gene itself might be 30,
00 bases. But the mature RNA,

448
00:36:59 --> 00:37:07
the mature mRNA might be
one and a half, 1,500 bases.

449
00:37:07 --> 00:37:11
That's remarkable. Out
of 30,000 letters in the

450
00:37:11 --> 00:37:15
initial transcript that is made,
the genes start, the promoter, and

451
00:37:15 --> 00:37:19
the transcription will stop
30, 00 bases away. The cell goes

452
00:37:19 --> 00:37:24
through the trouble of making
an RNA of 30,000 bases long,

453
00:37:24 --> 00:37:28
and then it trims it
down by throwing out 28,

454
00:37:28 --> 00:37:33
00 of the bases, keeping
only 1, 00 bases at the end.

455
00:37:33 --> 00:37:37
Now, this may seem profligate but
it ain't nothing compared to some

456
00:37:37 --> 00:37:42
extreme cases. The
clotting factor gene,

457
00:37:42 --> 00:37:47
the factor 8 gene, the gene that
has mutated in individuals with

458
00:37:47 --> 00:37:52
hemophilia, that gene is 200, 00
bases long, and it gets spliced

459
00:37:52 --> 00:37:57
down to a mere 10, 00 bases.
190,000 bases are thrown

460
00:37:57 --> 00:38:02
away. But that's
nothing compared to

461
00:38:02 --> 00:38:08
Duchene muscular dystrophy. The
Duchene muscular dystrophy is

462
00:38:08 --> 00:38:13
the all time winner. That
gene makes an immature initial

463
00:38:13 --> 00:38:19
RNA of 2 million bases. RNA
polymerase hops on at the

464
00:38:19 --> 00:38:24
promoter and it gets off at the
end of the Boston Marathon on here 2

465
00:38:24 --> 00:38:30
million bases later having made
an RNA of 2 million bases long.

466
00:38:30 --> 00:38:36
Calculate the speed of RNA
polymerase and you'll find out that

467
00:38:36 --> 00:38:42
it's at it for hours. It
hops on and it stays on for

468
00:38:42 --> 00:38:48
hours until it gets to the other end.
And then for all its troubles this

469
00:38:48 --> 00:38:54
gene is spliced down to 16,
00 bases in the mature message.

470
00:38:54 --> 00:39:00
Yup? How would it increase
the chance of mutations? Yup.

471
00:39:00 --> 00:39:05
So, splicing mutations could be a
problem. Some diseases could arise

472
00:39:05 --> 00:39:10
from errors in splicing.
Do you think that happens?

473
00:39:10 --> 00:39:15
Sure does. There could
be mutations that create,

474
00:39:15 --> 00:39:20
that change a splicing, or
mutations that create a new

475
00:39:20 --> 00:39:25
splicing, and all of that
could screw up the gene.

476
00:39:25 --> 00:39:30
Why do this? What in
the world is going on?

477
00:39:30 --> 00:39:35
Just think about the energetic cost.
I mean count up the ATPs involved

478
00:39:35 --> 00:39:40
in synthesizing a nucleotide, and
then the ATPs involved in adding

479
00:39:40 --> 00:39:45
nucleotides up. You know,
think about this totally

480
00:39:45 --> 00:39:50
wasted energy.
What is the point?

481
00:39:50 --> 00:39:55
I might be able to encode multiple
proteins with the same gene.

482
00:39:55 --> 00:40:00
How would I do that? Ooh,
wouldn't that be cleaver?

483
00:40:00 --> 00:40:05
I might be able to take a single
gene and make a mix and match

484
00:40:05 --> 00:40:10
product. It might be, do you
mean like one type of cell

485
00:40:10 --> 00:40:15
might splice up that message one
way to produce a certain protein,

486
00:40:15 --> 00:40:20
but a different cell type might
splice the same gene another way to

487
00:40:20 --> 00:40:25
produce a different protein?
Ooh. So, you're proposing, if I

488
00:40:25 --> 00:40:30
understand you correctly,
alternative splicing.

489
00:40:30 --> 00:40:34
Alternative splicing could
create multiple proteins,

490
00:40:34 --> 00:40:38
multiple distinct proteins. It
might be, for example, that you

491
00:40:38 --> 00:40:42
might make one protein that has a
cytoplasmic tail and another protein

492
00:40:42 --> 00:40:46
that doesn't have cytosplasmic
tail or a different tail or,

493
00:40:46 --> 00:40:50
or, this is true. This actually
happens. It's very cleaver.

494
00:40:50 --> 00:40:54
Anything that can happen
does happen somewhere,

495
00:40:54 --> 00:40:58
and it's fairly regularly used.
A typical gene in the human being

496
00:40:58 --> 00:41:02
has at least two alternative
splice forms, on average.

497
00:41:02 --> 00:41:05
Most, many don't, but there
are some that have large

498
00:41:05 --> 00:41:08
numbers. The most extreme
is there's a gene known,

499
00:41:08 --> 00:41:11
drosophila, that has more than a
thousand alternative splice forms.

500
00:41:11 --> 00:41:15
How does it know, how does the cell
know whether to splice it one way in

501
00:41:15 --> 00:41:18
the liver and one way in a heart or
something? We don't fully know but

502
00:41:18 --> 00:41:21
there's machinery and signals people
are trying to work out for that.

503
00:41:21 --> 00:41:25
Now, I don't want to confuse
you too much about it.

504
00:41:25 --> 00:41:28
You know, mostly, when we give you
a gene, you should think about it

505
00:41:28 --> 00:41:31
spliced out introns, exons.
But the truth is it is more

506
00:41:31 --> 00:41:35
complicated than that. There
can be alternative splicing

507
00:41:35 --> 00:41:39
that allows genes to be
used in multiple ways.

508
00:41:39 --> 00:41:43
Sometimes they don't make multiple
proteins. They may splice into

509
00:41:43 --> 00:41:46
portions of the mRNA that
are not translated, but,

510
00:41:46 --> 00:41:50
there is that, but, boy,
it's a huge amount of overhead

511
00:41:50 --> 00:41:54
here just to do that.
Is it justified? Yes?

512
00:41:54 --> 00:41:58
That is by computer if I just
gave you the sequence? Not quite.

513
00:41:58 --> 00:42:01
Almost. Maybe. Sort of.
It turns out that the

514
00:42:01 --> 00:42:04
computer programs for automatically
recognizing the matter of the human

515
00:42:04 --> 00:42:07
genome are sort of, they're
mediocre, not very good.

516
00:42:07 --> 00:42:09
We have some idea of the signals,
and various people have trying to

517
00:42:09 --> 00:42:12
write better and better
algorithms for doing that,

518
00:42:12 --> 00:42:15
but the cell knows what it's
doing and we don't fully know,

519
00:42:15 --> 00:42:18
as evidenced by the fact that we
can't write a clean computer program

520
00:42:18 --> 00:42:21
to do it yet. We need to get
information from the cell or from

521
00:42:21 --> 00:42:24
evolution or various other things
like that, and that's the ultimate

522
00:42:24 --> 00:42:27
test. If we knew what we were
talking about we'd just be able to

523
00:42:27 --> 00:42:30
write a computer program
and splice it out.

524
00:42:30 --> 00:42:34
And we don't. There's another
reason why people think these big

525
00:42:34 --> 00:42:38
introns and exons, these
big introns are helpful,

526
00:42:38 --> 00:42:42
and that is an evolutionary reason.
The evolutionary reason is a little

527
00:42:42 --> 00:42:47
bit harder to follow,
but let me try it on you.

528
00:42:47 --> 00:42:51
Suppose a random event happens
and a chromosome breaks,

529
00:42:51 --> 00:42:55
that happens, and suppose a random
breakage sticks one part of a

530
00:42:55 --> 00:43:00
chromosome to some other
part of the chromosome.

531
00:43:00 --> 00:43:04
If it lands smack dab in the middle
of the coding sequence of a gene

532
00:43:04 --> 00:43:09
that's bad new. But it
turns out that if it lands

533
00:43:09 --> 00:43:13
in the introns of two different
genes and sticks them together it

534
00:43:13 --> 00:43:18
could make a new gene that would
still work. By having a random

535
00:43:18 --> 00:43:23
break between two genes in their
introns and slamming them together,

536
00:43:23 --> 00:43:27
you could make a gene that had a
bunch of exons from one gene and a

537
00:43:27 --> 00:43:32
bunch of exons from another gene.
And this intervening sequence in the

538
00:43:32 --> 00:43:38
middle and it would get spliced up.
Evolution might like that because

539
00:43:38 --> 00:43:44
it would be a very easy way for
evolution to build new genes that

540
00:43:44 --> 00:43:49
had a portion of one protein
and a portion of another protein.

541
00:43:49 --> 00:43:55
This kind of mix and match domain
swapping could be very useful.

542
00:43:55 --> 00:44:01
And when we look across genomes,
we see lots and lots of examples of

543
00:44:01 --> 00:44:07
genes that have a similar first
half but different second halves.

544
00:44:07 --> 00:44:10
Or have some portion in the middle,
a domain that we recognize, that we

545
00:44:10 --> 00:44:13
see in multiple proteins. And
so, in fact, an argument for

546
00:44:13 --> 00:44:17
why we have all of this intronic DNA,
one that's impossible to prove but

547
00:44:17 --> 00:44:20
is an argument is that from
an evolution point of view,

548
00:44:20 --> 00:44:24
this allows a great deal
of evolutionary innovation.

549
00:44:24 --> 00:44:27
You have to be careful that you say
those organisms that have this extra

550
00:44:27 --> 00:44:31
space are able to mix and match and
create more new kinds of combination

551
00:44:31 --> 00:44:34
proteins, et cetera, et
cetera, and therefore survived

552
00:44:34 --> 00:44:38
better, et cetera,
et cetera, et cetera.

553
00:44:38 --> 00:44:41
Why don't bacteria have this?
Sorry? They're not as complicated.

554
00:44:41 --> 00:44:45
That's one though is we can take
a sort of condescending attitude to

555
00:44:45 --> 00:44:48
these bacteria.
They're not very,

556
00:44:48 --> 00:44:52
they're just not so complicated.
There's another point of view which

557
00:44:52 --> 00:44:56
is bacteria are far more
sophisticated than we are because

558
00:44:56 --> 00:45:00
they're under incredibly
rigorous evolutionary selection.

559
00:45:00 --> 00:45:04
You might argue that if I'm a
bacteria, can I really afford all

560
00:45:04 --> 00:45:08
this extra DNA? Now, the
metabolic cost of all that

561
00:45:08 --> 00:45:12
extra DNA is huge to a bacteria
which competes on replication.

562
00:45:12 --> 00:45:16
It's got to divide every 20 minutes,
and trying to put in all these extra

563
00:45:16 --> 00:45:20
bases would be very news. So,
you might imagine, just to be,

564
00:45:20 --> 00:45:24
you know, stand things on its head,
that early life all had introns and

565
00:45:24 --> 00:45:28
bacteria, in the process of
competing to be more and more

566
00:45:28 --> 00:45:31
efficient go rid of their introns.
There's actually a large camp of

567
00:45:31 --> 00:45:34
people who think it went that way,
that early life evolved with introns,

568
00:45:34 --> 00:45:37
and then bacteria, in
the pressure to compete,

569
00:45:37 --> 00:45:41
got rid of them. And there's
some evidence to support that.

570
00:45:41 --> 00:45:44
Bacteria don't have introns.
Small eukaryotes like yeast that

571
00:45:44 --> 00:45:47
sort of do compete on
replication have some introns,

572
00:45:47 --> 00:45:50
but a small number. There
are only about 250 introns in

573
00:45:50 --> 00:45:53
yeast. Only about 5% of the genes
have an intron and they're small.

574
00:45:53 --> 00:45:57
Bigger eukaryotes have bigger
introns. And the bigger you get,

575
00:45:57 --> 00:46:00
often on average the bigger
the genome sizes are the more

576
00:46:00 --> 00:46:03
you can tolerate it.
And so I actually think,

577
00:46:03 --> 00:46:07
I actually probably favor this
notion that introns were the

578
00:46:07 --> 00:46:10
original state and
they've been gotten rid of.

579
00:46:10 --> 00:46:14
And the more pressure you're under
to replicate rapidly the less you

580
00:46:14 --> 00:46:17
can tolerate this interesting
and complicated innovation.

581
00:46:17 --> 00:46:21
Anyway, that's another
way that things differ.

582
00:46:21 --> 00:46:24
And then, finally, viruses
can do it either way.

583
00:46:24 --> 00:46:28
Viruses, depending on whether
they are prokaryotic viruses or

584
00:46:28 --> 00:46:31
eukaryotic viruses,
are able to replicate,

585
00:46:31 --> 00:46:35
are able to either do
or don't have splicing.

586
00:46:35 --> 00:46:41
Last topic. Translation.
Here eukaryotes are relatively

587
00:46:41 --> 00:46:48
simple. You get a message, you
get a gene, you get an mRNA.

588
00:46:48 --> 00:46:54
The mRNA goes to the
ribosome. Here's a ribosome.

589
00:46:54 --> 00:47:01
The ribosome goes to the mRNA,
actually, and it starts turning out

590
00:47:01 --> 00:47:09
one protein as it chugs along.
Prokaryotes differ in an interesting

591
00:47:09 --> 00:47:18
way. I get a promoter here that
is transcribed into my mRNA,

592
00:47:18 --> 00:47:27
but it turns out that the mRNA can
encode multiple independent proteins,

593
00:47:27 --> 00:47:36
protein one, protein two,
protein three on the same mRNA.

594
00:47:36 --> 00:47:40
And a ribosome will hop on
here and synthesize this one.

595
00:47:40 --> 00:47:44
A ribosome will hop on here
and synthesize this one,

596
00:47:44 --> 00:47:48
and a ribosome will hop on
here and synthesize that one.

597
00:47:48 --> 00:47:52
And you have what is called
a polycistronic message.

598
00:47:52 --> 00:47:56
Poly, many. Cystronic, cystrons
were an old name for coding

599
00:47:56 --> 00:48:00
regions of genes here.
Polycystronic messages.

600
00:48:00 --> 00:48:03
Why would you want to do that,
have a single mRNA that encodes

601
00:48:03 --> 00:48:06
multiple distinct proteins, each
starting with its own ribosome

602
00:48:06 --> 00:48:09
start site there?
Efficiency. Maybe,

603
00:48:09 --> 00:48:12
in fact, these would be, how
about, oh, this would be cleaver,

604
00:48:12 --> 00:48:15
make them multiple steps
in a biochemical pathway?

605
00:48:15 --> 00:48:18
Have them coded on a single
messenger so then you'd only have to

606
00:48:18 --> 00:48:21
worry about regulating that
once. If you have the regulatory

607
00:48:21 --> 00:48:24
machinery to turn on, you'll
make all the enzymes for the

608
00:48:24 --> 00:48:27
pathway. And that's exactly what
bacteria do. They tend to put all

609
00:48:27 --> 00:48:30
the enzymes for a pathway on a
single message so when they want to

610
00:48:30 --> 00:48:33
call up, let's digest hexose this
morning, they have a whole thing

611
00:48:33 --> 00:48:37
that will let them be able
to do that, poly-cystronic.

612
00:48:37 --> 00:48:41
That's because they're small
genomes. They're pressed for space.

613
00:48:41 --> 00:48:46
And, because of that, they have
to slam a lot into a single unit.

614
00:48:46 --> 00:48:50
And this single unit that has
multiple genes encoded in a single

615
00:48:50 --> 00:48:55
message is called an operon,
and we'll talk more about that.

616
00:48:55 --> 00:49:00
Last of all viruses. Viruses.
Viruses have very little room.

617
00:49:00 --> 00:49:05
Their genomes can be tiny. A
typical virus might have a genome

618
00:49:05 --> 00:49:10
of 5,000 bases to 10, 00
bases to, in some cases,

619
00:49:10 --> 00:49:15
200,000 bases, but it hasn't got a
lot of room. It wants to pack a lot

620
00:49:15 --> 00:49:21
of protein coating information in.
And some viruses have come up with

621
00:49:21 --> 00:49:26
the most extraordinary way of doing
that. Some viruses have gone to the

622
00:49:26 --> 00:49:32
extreme of having RNAs that get made
from them that have a sequence --

623
00:49:32 --> 00:49:39
I'm just going to pick up in
the middle of the sequence here.

624
00:49:39 --> 00:49:46
A-C-U-A-C-U-A-C-U-A-C-U. You might
decide to read the sequence like

625
00:49:46 --> 00:49:53
this, that those are the codons,
and you'd get a certain protein.

626
00:49:53 --> 00:50:00
But I might also decide to read
that sequence C-U-A-C-U-A-C-U-A.

627
00:50:00 --> 00:50:04
And, of course, I'm giving
this in a repeating form

628
00:50:04 --> 00:50:08
because it's easy to note. I
could give you any sequence and I

629
00:50:08 --> 00:50:12
could read it in this reading frame,
I could read it in this reading

630
00:50:12 --> 00:50:17
frame, or I could read
it as U-A-C-U-A-C-U-A-C.

631
00:50:17 --> 00:50:21
In other words, there are
three reading frames that,

632
00:50:21 --> 00:50:25
in principle, you could translate
a protein from. In a typical

633
00:50:25 --> 00:50:30
prokaryotic gene or eukaryotic
gene only one of those is used.

634
00:50:30 --> 00:50:34
You start at the first AUG and
that sets the reading frame.

635
00:50:34 --> 00:50:39
But some viruses are so pressed for
space and are so cleaver and are so

636
00:50:39 --> 00:50:43
efficient that they make messages
that have tricks that they actually

637
00:50:43 --> 00:50:48
use two or, in some cases, all
three reading frames, which is

638
00:50:48 --> 00:50:53
an extraordinary packing of
information density into a simple

639
00:50:53 --> 00:50:57
message. So, the basic point.
We have a simple model. DNA is

640
00:50:57 --> 00:51:02
replicated.
Transcribed into RNA.

641
00:51:02 --> 00:51:06
Translated into protein. But
there are a lot of important

642
00:51:06 --> 00:51:11
variations between eukaryotes,
prokaryotes and viruses. And

643
00:51:11 --> 00:51:15
understanding them can be
useful for treating cancer,

644
00:51:15 --> 00:51:20
for treating AIDS, and for
treating viral and bacterial

645
00:51:20 --> 51:25
infections.
Next time.