1
00:00:00,320 --> 00:00:04,640
In today's episode, we're going to talk 
about the wonderful world of proteins.
 

2
00:00:04,640 --> 00:00:08,000
Proteins are all around our body. 
We use them in our daily lives,  

3
00:00:08,000 --> 00:00:10,800
and they do amazing things to keep us going.
 

4
00:00:10,800 --> 00:00:17,120
Protein design just won a Nobel Prize 
and we are going to do a mini-series of  

5
00:00:17,120 --> 00:00:23,920
episodes here to talk about AlphaFold and 
other AI systems used to design proteins,  

6
00:00:23,920 --> 00:00:29,520
whether people can increasingly design dangerous 
proteins, not just medicines, and whether protein  

7
00:00:29,520 --> 00:00:36,720
design can help us get cures for some of the 
toughest diseases that still plague humanity.
 

8
00:00:39,600 --> 00:00:45,280
But first, let's start with the basics. You might 
remember being in high school biology and seeing  

9
00:00:45,280 --> 00:00:51,440
a simple diagram of a cell. It probably looked a 
bit like a fried egg or a sunny side up. There was  

10
00:00:51,440 --> 00:00:56,080
the nucleus, which was a bit like the egg yolk. 
And then there were a few other things scattered  

11
00:00:56,080 --> 00:01:01,600
around, like mitochondria and ribosomes, 
but that was a massive simplification.
 

12
00:01:01,600 --> 00:01:07,840
In reality, cells are incredibly busy. There 
are billions of molecules in every cell,  

13
00:01:07,840 --> 00:01:13,680
including loads of proteins, which have different 
functions. So let me just think about what are  

14
00:01:13,680 --> 00:01:18,400
the different things that the proteins are 
doing? Well, there are structural proteins;  

15
00:01:18,400 --> 00:01:23,920
they provide shape and strength to cells. 
There are storage proteins; they store  

16
00:01:23,920 --> 00:01:28,640
little molecules. There are signalling proteins 
that help cells communicate with each other.  

17
00:01:28,640 --> 00:01:34,240
So insulin, for example, is a hormone, and 
it's made in the pancreas and it tells cells  

18
00:01:34,240 --> 00:01:39,920
to take up glucose from the bloodstream, 
and that lowers blood sugar after eating.
 

19
00:01:39,920 --> 00:01:45,600
There are also transport proteins that 
move molecules between cells. Haemoglobin,  

20
00:01:45,600 --> 00:01:50,640
for example, is a protein in red blood cells 
that binds to oxygen and carries it around in  

21
00:01:50,640 --> 00:01:56,320
the blood. There are also enzymes — enzymes speed 
up chemical reactions in our body, by lowering  

22
00:01:56,320 --> 00:02:02,720
the activation energy needed for them. There are 
regulatory proteins that control other proteins  

23
00:02:02,720 --> 00:02:08,800
and pathways. And there are defence proteins 
that protect us from attack; so antibodies are  

24
00:02:08,800 --> 00:02:13,760
a type of protein. Snakes and spiders have 
venoms, which are proteins that help them  

25
00:02:13,760 --> 00:02:20,640
disable their threats. There are so many different 
types of jobs that a protein might have, and many  

26
00:02:20,640 --> 00:02:26,320
proteins have multiple jobs at the same time.
And this means that this basic diagram view,  

27
00:02:26,320 --> 00:02:32,640
that you might've had of a cell, was quite 
simple. In reality, the cell is extremely  

28
00:02:32,640 --> 00:02:39,840
busy. It's more like a bustling city, and there 
are literally billions of molecules, proteins,  

29
00:02:39,840 --> 00:02:46,560
DNA, RNA, fats, sugars, and ions — all moving 
around, reacting and interacting with each other.
 

30
00:02:46,560 --> 00:02:53,520
Every part of the cell has its own job and it's 
a bit like different districts in the city.  

31
00:02:53,520 --> 00:02:59,840
There's a great blog post by Niko McCarty where he 
describes this, and I thought it would be helpful  

32
00:02:59,840 --> 00:03:06,480
just to have a sense of what's going on. He says, 
"A microbe's guts are a veritable Times Square,  

33
00:03:06,480 --> 00:03:12,080
crowded with sugars, proteins, and water molecules 
that ricochet and smash into each other billions  

34
00:03:12,080 --> 00:03:19,040
of times each second. Space is limited. A 
bacterium's insides are 70% water by mass;  

35
00:03:19,040 --> 00:03:25,760
the other 30% is dominated by proteins first, 
followed by RNA and lipids. DNA accounts for  

36
00:03:25,760 --> 00:03:34,080
just 1%. And all of this stuff fits inside a 
volume that is one quadrillionth of a litre.
 

37
00:03:34,080 --> 00:03:37,920
That's a lot of proteins and I 
can't even see one of them.
 

38
00:03:37,920 --> 00:03:46,160
Right? They're so small. And so if you think 
of this city — of each cell — the nucleus is  

39
00:03:46,160 --> 00:03:51,600
something like the city hall, it's managing the 
information; it has instructions for what should  

40
00:03:51,600 --> 00:03:57,840
happen. There are mitochondria; the power stations 
of the cell. There are ribosomes that construct  

41
00:03:57,840 --> 00:04:02,240
new proteins. And then there are proteins, that 
are the workers and the machines of the city,  

42
00:04:02,240 --> 00:04:07,120
but they're also the structural components and the 
signalling molecules and all of these things.
 

43
00:04:07,120 --> 00:04:11,600
Our body is doing so much 
with all of those proteins.  

44
00:04:11,600 --> 00:04:19,520
Are proteins used outside of the body too?
They are! In fact, if you've done any cooking,  

45
00:04:19,520 --> 00:04:25,520
you would know, for example, that chemical 
reactions change the proteins that you're cooking  

46
00:04:25,520 --> 00:04:32,720
with. So, for example, if you cook an egg white, 
it becomes firm when it's cooked. That's because  

47
00:04:32,720 --> 00:04:38,880
the heat denatures the proteins — it makes them 
unfold — and then it makes them coagulate into a  

48
00:04:38,880 --> 00:04:44,400
different kind of mesh, and that makes it opaque.
There's also gluten, which is a protein that gives  

49
00:04:44,400 --> 00:04:50,160
bread its stretchy texture — that's made of two 
proteins. There are also lots of proteins that  

50
00:04:50,160 --> 00:04:57,360
are used in industry and biotechnology. If you've 
done your laundry recently, you might have used  

51
00:04:57,360 --> 00:05:04,880
a detergent that was made of enzymes, and the 
enzymes break down stains, like fat or blood.  

52
00:05:04,880 --> 00:05:10,240
Then there are a bunch of proteins that are used 
in baking and brewing and textile manufacturing.  

53
00:05:10,240 --> 00:05:14,720
Of course there are lots of proteins that are 
used in medicine as well. So I mentioned that  

54
00:05:14,720 --> 00:05:21,600
antibodies are a type of protein, and lots of 
medicines are types of antibodies. There's also  

55
00:05:21,600 --> 00:05:29,680
insulin, which people use in diabetes; it's 
a protein that is also a therapeutic drug.
 

56
00:05:29,680 --> 00:05:37,120
What actually are proteins? What do 
they look like and how do they form?
 

57
00:05:37,120 --> 00:05:45,680
Proteins are long chains of amino acids. You 
can sort of think of that as like beads on a  

58
00:05:45,680 --> 00:05:53,200
string. And then that string, or that chain, is 
folded into some kind of 3D shape. The string  

59
00:05:53,200 --> 00:06:00,720
is the protein's backbone, and each bead is an 
amino acid. Each amino acid has unique features.  

60
00:06:00,720 --> 00:06:06,240
So as this string falls into a structure, you 
can kind of imagine that maybe happening at a  

61
00:06:06,240 --> 00:06:12,640
small scale — maybe there's like a little helix 
of the string in some place, or maybe there are  

62
00:06:12,640 --> 00:06:18,320
two parallel strings next to each other. But 
imagine that... we have to kind of zoom out  

63
00:06:18,320 --> 00:06:25,200
and this whole 3D shape of the protein could also 
be connected to another protein; it could be two  

64
00:06:25,200 --> 00:06:33,200
proteins together, making a protein complex.
How is that made? I know I eat some protein,  

65
00:06:33,200 --> 00:06:38,880
but I think we make some too.
That's right. So you have lots of  

66
00:06:38,880 --> 00:06:46,320
DNA in your cells, and the DNA, which is the 
code of life, is the instructions for which  

67
00:06:46,320 --> 00:06:52,320
proteins to make and how they should look. 
The DNA is transcribed into RNA, which is  

68
00:06:52,960 --> 00:07:00,800
typically this temporary molecule, and then the 
RNA is then translated into protein by ribosomes.  

69
00:07:01,840 --> 00:07:10,240
They sort of form one-by-one into this chain, and 
then rapidly fold into a much bigger structure.
 

70
00:07:10,800 --> 00:07:16,640
This was kind of interesting to me because when 
I was reading this, I was thinking, okay, how did  

71
00:07:16,640 --> 00:07:23,520
the first protein that was ever discovered look? 
What did people think when they first saw it?  

72
00:07:24,080 --> 00:07:32,960
And that was fascinating because the first protein 
whose structure was determined was in 1958,  

73
00:07:32,960 --> 00:07:40,560
and that was myoglobin. This was determined 
by John Kendrew, a British scientist. When he  

74
00:07:40,560 --> 00:07:47,360
discovered this, it was only four years after 
the discovery of DNA's structure — DNA is of  

75
00:07:47,360 --> 00:07:55,840
course very beautiful; it has this symmetrical 
structure, of this helix. And he was really  

76
00:07:55,840 --> 00:08:01,920
disappointed when he figured out what myoglobin 
looked like. He wrote in this paper: "Perhaps the  

77
00:08:01,920 --> 00:08:07,040
most remarkable features of the molecule are 
its complexity and its lack of symmetry."
 

78
00:08:07,040 --> 00:08:13,520
Oh no, it's ugly.
But in hindsight, the irregularity is exactly what  

79
00:08:13,520 --> 00:08:23,360
makes proteins so powerful. It's not really like 
DNA, which has this kind of linear messaging — it  

80
00:08:23,360 --> 00:08:30,480
has the code, and then the code just linearly 
turns into RNA. But a protein is actually doing  

81
00:08:30,480 --> 00:08:37,600
multiple things. It's in the cell being bombarded 
sometimes with lots of different molecules,  

82
00:08:37,600 --> 00:08:43,600
and it needs to be able to recognise these 
different shapes and structures, and sometimes,  

83
00:08:43,600 --> 00:08:50,720
it has multiple functions — and this function 
of every protein depends on that 3D structure.
 

84
00:08:51,760 --> 00:08:57,360
The folded shape means that there are like 
little pockets, grooves and surfaces that  

85
00:08:57,360 --> 00:09:04,240
the protein uses to bind to other molecules, 
or carry out specific chemical reactions,  

86
00:09:04,240 --> 00:09:10,960
or even receive signals and then change shape in 
response. That means the same protein molecule  

87
00:09:10,960 --> 00:09:16,160
might be doing multiple things at once. It could 
be doing a chemical reaction, but also binding  

88
00:09:16,160 --> 00:09:21,360
to something else, and then when it gets some 
regulatory signal, it could be changing shape and  

89
00:09:21,360 --> 00:09:28,240
stopping that chemical reaction from happening.
So there's benefits to being a weird blob. There's  

90
00:09:28,240 --> 00:09:39,440
nothing wrong with being a weird blob.
I thought it would be fun if we both share  

91
00:09:39,440 --> 00:09:48,240
some fun facts about proteins. I found these from 
the book Biology by the Numbers, which is a great  

92
00:09:48,240 --> 00:09:56,400
textbook, and it's also free online. The authors 
create these rough estimates and pull together key  

93
00:09:56,400 --> 00:10:03,920
numbers on lots of different things related to 
cell biology. Some of them are rough estimates,  

94
00:10:03,920 --> 00:10:09,120
but they're kind of our best guess right now.
Hit me.
 

95
00:10:09,120 --> 00:10:15,520
Alright, first one, how many 
proteins are in a human cell?
 

96
00:10:15,520 --> 00:10:21,600
They're busy, so I'm going to guess a lot. 
And I'm going to guess it depends on the cell,  

97
00:10:21,600 --> 00:10:30,320
but I will go with a hundred million.
That is a lot, and it does depend on  

98
00:10:30,320 --> 00:10:38,480
the cell. But the estimate for the average 
number is ten billion proteins per cell.
 

99
00:10:38,480 --> 00:10:44,320
Oh no. Two orders of magnitude wrong, not 
a good start. Okay, well, I've got one.  

100
00:10:47,040 --> 00:10:54,320
Which is bigger: the protein or the 
mRNA that codes for the protein?
 

101
00:10:54,320 --> 00:11:04,873
Um... surely the protein is bigger, no? Why would 
the instructions be bigger than the protein?
 

102
00:11:04,873 --> 00:11:10,480
That's what I always think, and it's the other 
way around. So the mRNA is bigger — you look at  

103
00:11:10,480 --> 00:11:18,720
them side by side - well, images of 'em - and 
the mRNA is like 10 times bigger. Because each  

104
00:11:18,720 --> 00:11:23,840
amino acid is coded for by three nucleotides, 
and the nucleotides themselves are bigger and  

105
00:11:23,840 --> 00:11:31,520
heavier. So it's counterintuitive to me, 
but you know, it makes sense, I guess,  

106
00:11:31,520 --> 00:11:36,480
when you think about it physically.
That does make sense... well,  

107
00:11:36,480 --> 00:11:39,600
I don't know if that makes sense. I feel 
like I need to think about this more.
 

108
00:11:39,600 --> 00:11:44,160
Yeah, it doesn't make sense from a computer 
science point of view, but from a physical point  

109
00:11:44,160 --> 00:11:48,240
of view it feels like, yeah.
Right.
 

110
00:11:48,240 --> 00:11:54,080
I have one. So, you know, as a small person, 
I wanted to find out which protein was the  

111
00:11:54,080 --> 00:11:59,040
smallest. Do you have any guesses?
The protein that's the smallest? Well,  

112
00:11:59,040 --> 00:12:05,840
the definition of a protein... I wonder if I'm 
allowed to have- it's got to have at least two  

113
00:12:05,840 --> 00:12:12,000
amino acids, so I know it's not going to be 
less than two, but that probably wouldn't  

114
00:12:12,000 --> 00:12:16,320
count as a protein because it wouldn't fold 
into anything, wouldn't have much function.  

115
00:12:16,320 --> 00:12:23,920
So I'm going to guess philosophically, 
two, and then, literally, more than two.
 

116
00:12:23,920 --> 00:12:31,920
Well, you're right. I think the typical definition 
of a protein is something that floats on its own  

117
00:12:31,920 --> 00:12:37,920
in water and can fold into a stable shape. If 
you use that definition, then the smallest ones  

118
00:12:37,920 --> 00:12:44,080
are some 20 to 30 amino acids long. There are 
actually lots of really tiny proteins, and these  

119
00:12:44,080 --> 00:12:51,680
tiny proteins are called "micro proteins", and 
they're less than a hundred amino acids or so. One  

120
00:12:51,680 --> 00:13:01,600
example that's actually even smaller than 20 or 30 
is somatostatin, which is a hormone that controls  

121
00:13:01,600 --> 00:13:07,680
other hormones — so it controls growth hormone and 
insulin. — and that's only 14 amino acids long.
 

122
00:13:07,680 --> 00:13:12,640
Oh wow, it's that small. Oh okay.
Right. It still has a stable shape,  

123
00:13:12,640 --> 00:13:18,320
because parts of the chain are connected to each 
other. So it's not considered a typical protein,  

124
00:13:18,320 --> 00:13:23,680
but it's a Itpeptide and it's very small.
Got it, okay. What's the biggest? I think you  

125
00:13:23,680 --> 00:13:28,160
know the answer to this one.
I think I do. Is it titin?
 

126
00:13:28,160 --> 00:13:32,640
It's titin. That's the biggest human protein at 
least, I don't know outside of humans. But that  

127
00:13:32,640 --> 00:13:37,840
one is 33,000 amino acids long.
I got one. What's the most  

128
00:13:37,840 --> 00:13:43,120
abundant protein on earth?
I am going to guess it has something  

129
00:13:43,120 --> 00:13:49,040
to do with photosynthesis, because that seems 
like one of the biggest functions on earth.
 

130
00:13:49,040 --> 00:13:56,480
Very good guess. So it's kind of a tie, and 
we're not really sure which one is more abundant,  

131
00:13:56,480 --> 00:13:59,200
so that was a bit of a trick question.
Oh wow.
 

132
00:13:59,200 --> 00:14:08,080
But one of them is RuBisCO, and that is used in 
photosynthesis; it's used to grab carbon from the  

133
00:14:08,080 --> 00:14:17,040
air and turn it into useful organic material. And 
that's used by all photosynthetic organisms. And  

134
00:14:17,040 --> 00:14:22,720
scientists estimate that there are about five 
kilogrammes of RuBisCO per person on earth.
 

135
00:14:22,720 --> 00:14:28,880
Oh my god. What?! Wow.
I guess there are a lot of plants.
 

136
00:14:28,880 --> 00:14:32,640
Yeah, fair enough. They're winning. 
They're winning... for now...
 

137
00:14:32,640 --> 00:14:37,040
There's actually the second, which 
might be ahead. We're not sure-
 

138
00:14:37,040 --> 00:14:42,480
Oh right.
-and that is collagen. That is  

139
00:14:42,480 --> 00:14:51,040
used as a kind of structural protein, and it makes 
up about 30% of the protein mass in your body — so  

140
00:14:51,040 --> 00:14:57,280
about three kilogrammes of collagen per person. 
But it's not just humans that have collagen,  

141
00:14:57,280 --> 00:15:05,920
it's also the livestock and all animals. That 
means there's- well, the total number- the total  

142
00:15:05,920 --> 00:15:12,320
mass of livestock is also enormous, right? And so 
this means there's roughly four to six kilogrammes  

143
00:15:12,320 --> 00:15:17,520
of collagen per person on earth.
Ready for another fun fact?
 

144
00:15:17,520 --> 00:15:20,240
Yes.
Well, enzymes are a  

145
00:15:20,240 --> 00:15:28,480
type of protein that speed up reactions... so how 
much do you think enzymes speed up reactions?
 

146
00:15:28,480 --> 00:15:37,520
Mmm... a thousand times, maybe? Two thousand? 
I feel like... a lot. But I don't know.
 

147
00:15:37,520 --> 00:15:45,920
A lot. A lot. And I bet some do a thousand, but 
if you're really looking at the best of the best,  

148
00:15:45,920 --> 00:15:54,400
we're talking billions of times, and possibly 
trillions of times, so we're talking millions  

149
00:15:54,400 --> 00:16:01,040
of reactions per second per enzyme in some 
cases, and just totally changing what is  

150
00:16:01,040 --> 00:16:06,640
happening at the molecular level.
That's crazy. That means, I guess,  

151
00:16:06,640 --> 00:16:10,640
some reactions just wouldn't happen 
if the enzymes weren't there.
 

152
00:16:10,640 --> 00:16:14,000
Oh, absolutely. Yeah. I mean, 
statistically speaking, yeah.
 

153
00:16:14,000 --> 00:16:18,720
So we were talking about protein folding 
the other day, and I was thinking: well,  

154
00:16:18,720 --> 00:16:22,480
how fast do proteins fold into 
shape? Do you have any guesses?
 

155
00:16:22,480 --> 00:16:25,758
Oh... that is a tough one because, well, we just 
had a very long protein that took forever, but  

156
00:16:25,758 --> 00:16:29,680
I bet most proteins don't take long at all. The 
folding has to happen quickly, otherwise they'll  

157
00:16:29,680 --> 00:16:40,560
get distracted by other forces. So I will go with 
tenths of seconds, no, hundredths of seconds.
 

158
00:16:40,560 --> 00:16:44,480
Pretty close. So, on average, 
proteins fold in milliseconds,  

159
00:16:44,480 --> 00:16:52,800
but some proteins fold really quickly, in micro 
seconds, which are a millionth of a second. And  

160
00:16:52,800 --> 00:16:57,440
I guess you're right that it really does have 
to happen fast, because there's so much other  

161
00:16:57,440 --> 00:17:03,280
stuff going on in the cell. It could just be 
bombarded with something else before it folds.
 

162
00:17:03,280 --> 00:17:04,868
Yeah, well, no fun. One final one 
from me. How quick do they move? Let's  

163
00:17:04,868 --> 00:17:04,941
say you're in a cell. How quick does 
the protein move across the cell?
 

164
00:17:04,941 --> 00:17:05,035
I love the idea that I've shrunk myself to the 
size that I can fit inside a cell. And now I'm  

165
00:17:05,035 --> 00:17:05,119
trying to race with these little proteins. 
To get across a cell... uh... I dunno. A  

166
00:17:05,119 --> 00:17:05,200
second? Maybe half a second? I dunno.
A small protein could be 10 milliseconds  

167
00:17:05,200 --> 00:17:05,300
to get across a cell. The thing, though, is that 
cells are small. So if you haven't shrunk yourself  

168
00:17:05,300 --> 00:17:05,385
all the way down, and are just visualising 
the human scale, how long would it take a  

169
00:17:05,385 --> 00:17:05,479
protein to move a whole centimetre? Well, then 
you'd need 20 days for some of the proteins.
 

170
00:17:05,479 --> 00:17:05,559
Well, so at first I thought you said - 
okay, that's quite fast - they're taking  

171
00:17:05,559 --> 00:17:05,646
10 milliseconds to cross the cell. But 20 
days to travel one centimetre is quite slow,  

172
00:17:05,646 --> 00:17:05,714
I could do that much faster.
Yeah, I think you're going to win.
 

173
00:17:05,714 --> 00:17:10,560
... but maybe not if I'm shrink to that size. 
Okay, I got another one. How fast are enzymes  

174
00:17:10,560 --> 00:17:15,600
colliding with other molecules in the cell? Or 
how many collisions are there per second?
 

175
00:17:15,600 --> 00:17:20,640
Okay. I have the sense that things are 
just crazy up in there and everyone's  

176
00:17:20,640 --> 00:17:26,720
sort of bumping around. So I'm going to 
say a thousand collisions a second.
 

177
00:17:26,720 --> 00:17:33,600
Well, you were right with the idea.
Oh no, I should have just said "A lot."
 

178
00:17:33,600 --> 00:17:42,640
But I think the estimate is 500,000 molecules 
are colliding with an enzyme per second.
 

179
00:17:42,640 --> 00:17:47,920
Wow.
And that's a lot! And that makes me think that  

180
00:17:47,920 --> 00:17:54,000
proteins have to be really specific in how they 
bind to their targets. It's like, you know, if  

181
00:17:54,000 --> 00:18:00,720
you're at a really crowded party and you're trying 
to find a friend, you would just bump into so many  

182
00:18:00,720 --> 00:18:05,600
people before you actually find your friend. So 
you have to actually be able to recognise them  

183
00:18:05,600 --> 00:18:12,160
among the 500,000 random strangers around you.
Yep. That's tricky. Okay, Saloni,  

184
00:18:12,160 --> 00:18:18,000
what's your favourite protein?
My favourite protein is tubulin. It's part  

185
00:18:18,000 --> 00:18:25,680
of microtubules. The microtubules are kinda the 
skeletons of your cells... That sounds a bit grim,  

186
00:18:25,680 --> 00:18:33,840
actually. But they are basically formed of these 
hollow tubes that are made of this protein,  

187
00:18:33,840 --> 00:18:41,040
and each of the little structures is kind of 
like a tiny corn kernel. That tube can sort of  

188
00:18:41,040 --> 00:18:47,920
assemble and disassemble in response to signals, 
and that means that the entire skeleton can kind  

189
00:18:47,920 --> 00:18:54,480
of assemble and disassemble... which means the 
whole cell can change its shape or its size and  

190
00:18:54,480 --> 00:19:02,880
move around, because of these microtubules. 
The microtubules also act as tracks to move  

191
00:19:02,880 --> 00:19:09,360
things around, so they're a bit like a cellular 
railway or something, which I think is just super  

192
00:19:09,360 --> 00:19:14,720
cool. And I remember learning about this in 
my undergrad and just seeing some diagrams  

193
00:19:14,720 --> 00:19:18,320
and thinking, wow, that's amazing.
That's a good one. I haven't even  

194
00:19:18,320 --> 00:19:25,680
better one though, which is gluten 
in bread! Woo! I'm a bread guy.
 

195
00:19:25,680 --> 00:19:29,760
That's a good one.
We each have our favourites.
 

196
00:19:29,760 --> 00:19:36,960
This was the first of a series of mini episodes 
we're doing on proteins. Stay tuned for our next  

197
00:19:36,960 --> 00:19:45,280
episode on the history of Insulin. And if you like 
this, share it with your friends and subscribe.