TED2008
Craig Venter: On the verge of creating synthetic life
Filmed:
Readability: 4.7
1,196,566 views
"Can we create new life out of our digital universe?" Craig Venter asks. His answer is "yes" -- and pretty soon. He walks through his latest research and promises that we'll soon be able to build and boot up a synthetic chromosome. NOTE: This talk was given in 2008, and this field of science has developed quickly since then. Read "Criticisms & updates" below for more details.
Craig Venter - Biologist, genetics pioneer
In 2001, Craig Venter made headlines for sequencing the human genome. In 2003, he started mapping the ocean's biodiversity. And now he's created the first synthetic lifeforms -- microorganisms that can produce alternative fuels. Full bio
In 2001, Craig Venter made headlines for sequencing the human genome. In 2003, he started mapping the ocean's biodiversity. And now he's created the first synthetic lifeforms -- microorganisms that can produce alternative fuels. Full bio
Double-click the English transcript below to play the video.
00:19
You know, I've talked about some of these projects before --
0
1000
2000
00:21
about the human genome and what that might mean,
1
3000
4000
00:25
and discovering new sets of genes.
2
7000
3000
00:28
We're actually starting at a new point:
3
10000
3000
00:31
we've been digitizing biology,
4
13000
4000
00:35
and now we're trying to go from that digital code
5
17000
3000
00:38
into a new phase of biology
6
20000
2000
00:40
with designing and synthesizing life.
7
22000
3000
00:43
So, we've always been trying to ask big questions.
8
25000
3000
00:48
"What is life?" is something that I think many biologists
9
30000
2000
00:50
have been trying to understand
10
32000
2000
00:52
at various levels.
11
34000
2000
00:54
We've tried various approaches,
12
36000
3000
00:57
paring it down to minimal components.
13
39000
3000
01:01
We've been digitizing it now for almost 20 years;
14
43000
2000
01:03
when we sequenced the human genome,
15
45000
2000
01:05
it was going from the analog world of biology
16
47000
3000
01:08
into the digital world of the computer.
17
50000
4000
01:12
Now we're trying to ask, "Can we regenerate life
18
54000
4000
01:16
or can we create new life
19
58000
2000
01:18
out of this digital universe?"
20
60000
3000
01:21
This is the map of a small organism,
21
63000
3000
01:24
Mycoplasma genitalium,
22
66000
2000
01:26
that has the smallest genome for a species
23
68000
3000
01:29
that can self-replicate in the laboratory,
24
71000
3000
01:32
and we've been trying to just see if
25
74000
2000
01:34
we can come up with an even smaller genome.
26
76000
3000
01:38
We're able to knock out on the order of 100 genes
27
80000
2000
01:40
out of the 500 or so that are here.
28
82000
3000
01:43
When we look at its metabolic map,
29
85000
2000
01:45
it's relatively simple
30
87000
2000
01:47
compared to ours --
31
89000
2000
01:49
trust me, this is simple --
32
91000
2000
01:51
but when we look at all the genes
33
93000
2000
01:53
that we can knock out one at a time,
34
95000
3000
01:56
it's very unlikely that this would yield
35
98000
2000
01:58
a living cell.
36
100000
2000
02:01
So we decided the only way forward
37
103000
2000
02:03
was to actually synthesize this chromosome
38
105000
3000
02:06
so we could vary the components
39
108000
3000
02:09
to ask some of these most fundamental questions.
40
111000
4000
02:13
And so we started down the road of:
41
115000
2000
02:15
can we synthesize a chromosome?
42
117000
3000
02:19
Can chemistry permit making
43
121000
2000
02:21
these really large molecules
44
123000
2000
02:23
where we've never been before?
45
125000
2000
02:25
And if we do, can we boot up a chromosome?
46
127000
3000
02:28
A chromosome, by the way, is just a piece of inert chemical material.
47
130000
3000
02:32
So, our pace of digitizing life has been increasing
48
134000
3000
02:35
at an exponential pace.
49
137000
3000
02:38
Our ability to write the genetic code
50
140000
3000
02:41
has been moving pretty slowly
51
143000
2000
02:43
but has been increasing,
52
145000
3000
02:46
and our latest point would put it on, now, an exponential curve.
53
148000
4000
02:51
We started this over 15 years ago.
54
153000
2000
02:53
It took several stages, in fact,
55
155000
3000
02:56
starting with a bioethical review before we did the first experiments.
56
158000
3000
03:00
But it turns out synthesizing DNA
57
162000
2000
03:02
is very difficult.
58
164000
2000
03:04
There are tens of thousands of machines around the world
59
166000
3000
03:07
that make small pieces of DNA --
60
169000
2000
03:09
30 to 50 letters in length --
61
171000
3000
03:12
and it's a degenerate process, so the longer you make the piece,
62
174000
3000
03:15
the more errors there are.
63
177000
2000
03:17
So we had to create a new method
64
179000
2000
03:19
for putting these little pieces together and correct all the errors.
65
181000
3000
03:23
And this was our first attempt, starting with the digital information
66
185000
3000
03:26
of the genome of phi X174.
67
188000
2000
03:28
It's a small virus that kills bacteria.
68
190000
3000
03:32
We designed the pieces, went through our error correction
69
194000
3000
03:35
and had a DNA molecule
70
197000
2000
03:37
of about 5,000 letters.
71
199000
3000
03:40
The exciting phase came when we took this piece of inert chemical
72
202000
4000
03:44
and put it in the bacteria,
73
206000
2000
03:46
and the bacteria started to read this genetic code,
74
208000
4000
03:50
made the viral particles.
75
212000
2000
03:52
The viral particles then were released from the cells
76
214000
2000
03:54
and came back and killed the E. coli.
77
216000
3000
03:57
I was talking to the oil industry recently
78
219000
3000
04:00
and I said they clearly understood that model.
79
222000
3000
04:03
(Laughter)
80
225000
3000
04:06
They laughed more than you guys are. (Laughter)
81
228000
3000
04:10
And so, we think this is a situation
82
232000
2000
04:12
where the software can actually build its own hardware
83
234000
3000
04:15
in a biological system.
84
237000
2000
04:17
But we wanted to go much larger:
85
239000
2000
04:19
we wanted to build the entire bacterial chromosome --
86
241000
3000
04:22
it's over 580,000 letters of genetic code --
87
244000
4000
04:26
so we thought we'd build them in cassettes the size of the viruses
88
248000
3000
04:29
so we could actually vary the cassettes
89
251000
2000
04:31
to understand
90
253000
2000
04:33
what the actual components of a living cell are.
91
255000
3000
04:36
Design is critical,
92
258000
2000
04:38
and if you're starting with digital information in the computer,
93
260000
3000
04:41
that digital information has to be really accurate.
94
263000
4000
04:45
When we first sequenced this genome in 1995,
95
267000
3000
04:48
the standard of accuracy was one error per 10,000 base pairs.
96
270000
4000
04:52
We actually found, on resequencing it,
97
274000
2000
04:54
30 errors; had we used that original sequence,
98
276000
3000
04:57
it never would have been able to be booted up.
99
279000
3000
05:00
Part of the design is designing pieces
100
282000
2000
05:02
that are 50 letters long
101
284000
3000
05:05
that have to overlap with all the other 50-letter pieces
102
287000
3000
05:08
to build smaller subunits
103
290000
2000
05:10
we have to design so they can go together.
104
292000
3000
05:13
We design unique elements into this.
105
295000
3000
05:16
You may have read that we put watermarks in.
106
298000
2000
05:18
Think of this:
107
300000
2000
05:20
we have a four-letter genetic code -- A, C, G and T.
108
302000
3000
05:23
Triplets of those letters
109
305000
3000
05:26
code for roughly 20 amino acids,
110
308000
2000
05:28
such that there's a single letter designation
111
310000
3000
05:31
for each of the amino acids.
112
313000
2000
05:33
So we can use the genetic code to write out words,
113
315000
3000
05:36
sentences, thoughts.
114
318000
2000
05:39
Initially, all we did was autograph it.
115
321000
2000
05:41
Some people were disappointed there was not poetry.
116
323000
3000
05:44
We designed these pieces so
117
326000
2000
05:46
we can just chew back with enzymes;
118
328000
3000
05:50
there are enzymes that repair them and put them together.
119
332000
3000
05:53
And we started making pieces,
120
335000
2000
05:55
starting with pieces that were 5,000 to 7,000 letters,
121
337000
4000
05:59
put those together to make 24,000-letter pieces,
122
341000
4000
06:03
then put sets of those going up to 72,000.
123
345000
4000
06:07
At each stage, we grew up these pieces in abundance
124
349000
2000
06:09
so we could sequence them
125
351000
2000
06:11
because we're trying to create a process that's extremely robust
126
353000
3000
06:14
that you can see in a minute.
127
356000
3000
06:17
We're trying to get to the point of automation.
128
359000
3000
06:20
So, this looks like a basketball playoff.
129
362000
2000
06:22
When we get into these really large pieces
130
364000
2000
06:24
over 100,000 base pairs,
131
366000
4000
06:28
they won't any longer grow readily in E. coli --
132
370000
2000
06:30
it exhausts all the modern tools of molecular biology --
133
372000
4000
06:34
and so we turned to other mechanisms.
134
376000
4000
06:38
We knew there's a mechanism called homologous recombination
135
380000
3000
06:41
that biology uses to repair DNA
136
383000
3000
06:44
that can put pieces together.
137
386000
3000
06:47
Here's an example of it:
138
389000
1000
06:48
there's an organism called
139
390000
1000
06:49
Deinococcus radiodurans
140
391000
2000
06:51
that can take three millions rads of radiation.
141
393000
3000
06:54
You can see in the top panel, its chromosome just gets blown apart.
142
396000
4000
06:58
Twelve to 24 hours later, it put it
143
400000
3000
07:01
back together exactly as it was before.
144
403000
2000
07:03
We have thousands of organisms that can do this.
145
405000
3000
07:06
These organisms can be totally desiccated;
146
408000
2000
07:08
they can live in a vacuum.
147
410000
2000
07:11
I am absolutely certain that life can exist in outer space,
148
413000
3000
07:14
move around, find a new aqueous environment.
149
416000
3000
07:17
In fact, NASA has shown a lot of this is out there.
150
419000
4000
07:21
Here's an actual micrograph of the molecule we built
151
423000
4000
07:25
using these processes, actually just using yeast mechanisms
152
427000
4000
07:29
with the right design of the pieces we put them in;
153
431000
3000
07:32
yeast puts them together automatically.
154
434000
3000
07:35
This is not an electron micrograph;
155
437000
2000
07:37
this is just a regular photomicrograph.
156
439000
2000
07:39
It's such a large molecule
157
441000
2000
07:41
we can see it with a light microscope.
158
443000
3000
07:44
These are pictures over about a six-second period.
159
446000
3000
07:47
So, this is the publication we had just a short while ago.
160
449000
4000
07:51
This is over 580,000 letters of genetic code;
161
453000
3000
07:54
it's the largest molecule ever made by humans of a defined structure.
162
456000
5000
07:59
It's over 300 million molecular weight.
163
461000
3000
08:02
If we printed it out at a 10 font with no spacing,
164
464000
3000
08:05
it takes 142 pages
165
467000
2000
08:07
just to print this genetic code.
166
469000
4000
08:11
Well, how do we boot up a chromosome? How do we activate this?
167
473000
3000
08:14
Obviously, with a virus it's pretty simple;
168
476000
3000
08:17
it's much more complicated dealing with bacteria.
169
479000
3000
08:20
It's also simpler when you go
170
482000
2000
08:22
into eukaryotes like ourselves:
171
484000
2000
08:24
you can just pop out the nucleus
172
486000
2000
08:26
and pop in another one,
173
488000
2000
08:28
and that's what you've all heard about with cloning.
174
490000
3000
08:31
With bacteria and Archaea, the chromosome is integrated into the cell,
175
493000
4000
08:35
but we recently showed that we can do a complete transplant
176
497000
4000
08:39
of a chromosome from one cell to another
177
501000
2000
08:41
and activate it.
178
503000
3000
08:44
We purified a chromosome from one microbial species --
179
506000
4000
08:48
roughly, these two are as distant as human and mice --
180
510000
3000
08:51
we added a few extra genes
181
513000
2000
08:53
so we could select for this chromosome,
182
515000
2000
08:55
we digested it with enzymes
183
517000
2000
08:57
to kill all the proteins,
184
519000
2000
08:59
and it was pretty stunning when we put this in the cell --
185
521000
3000
09:02
and you'll appreciate
186
524000
2000
09:04
our very sophisticated graphics here.
187
526000
3000
09:07
The new chromosome went into the cell.
188
529000
3000
09:10
In fact, we thought this might be as far as it went,
189
532000
2000
09:12
but we tried to design the process a little bit further.
190
534000
3000
09:15
This is a major mechanism of evolution right here.
191
537000
3000
09:18
We find all kinds of species
192
540000
2000
09:20
that have taken up a second chromosome
193
542000
2000
09:22
or a third one from somewhere,
194
544000
2000
09:24
adding thousands of new traits
195
546000
2000
09:26
in a second to that species.
196
548000
2000
09:28
So, people who think of evolution
197
550000
2000
09:30
as just one gene changing at a time
198
552000
2000
09:32
have missed much of biology.
199
554000
3000
09:35
There are enzymes called restriction enzymes
200
557000
2000
09:37
that actually digest DNA.
201
559000
2000
09:39
The chromosome that was in the cell
202
561000
2000
09:41
doesn't have one;
203
563000
2000
09:43
the chromosome we put in does.
204
565000
2000
09:45
It got expressed and it recognized
205
567000
2000
09:47
the other chromosome as foreign material,
206
569000
3000
09:50
chewed it up, and so we ended up
207
572000
2000
09:52
just with a cell with the new chromosome.
208
574000
4000
09:56
It turned blue because of the genes we put in it.
209
578000
3000
09:59
And with a very short period of time,
210
581000
2000
10:01
all the characteristics of one species were lost
211
583000
3000
10:04
and it converted totally into the new species
212
586000
3000
10:07
based on the new software that we put in the cell.
213
589000
3000
10:10
All the proteins changed,
214
592000
2000
10:12
the membranes changed;
215
594000
2000
10:14
when we read the genetic code, it's exactly what we had transferred in.
216
596000
4000
10:18
So, this may sound like genomic alchemy,
217
600000
3000
10:21
but we can, by moving the software of DNA around,
218
603000
4000
10:25
change things quite dramatically.
219
607000
4000
10:29
Now I've argued, this is not genesis;
220
611000
2000
10:31
this is building on three and a half billion years of evolution.
221
613000
4000
10:36
And I've argued that we're about to perhaps
222
618000
2000
10:38
create a new version of the Cambrian explosion,
223
620000
3000
10:41
where there's massive new speciation
224
623000
3000
10:45
based on this digital design.
225
627000
2000
10:47
Why do this?
226
629000
2000
10:49
I think this is pretty obvious in terms of some of the needs.
227
631000
2000
10:51
We're about to go from six and a half
228
633000
2000
10:53
to nine billion people over the next 40 years.
229
635000
3000
10:56
To put it in context for myself:
230
638000
2000
10:58
I was born in 1946.
231
640000
2000
11:00
There are now three people on the planet
232
642000
2000
11:02
for every one of us that existed in 1946;
233
644000
4000
11:06
within 40 years, there'll be four.
234
648000
3000
11:09
We have trouble feeding, providing fresh, clean water,
235
651000
3000
11:12
medicines, fuel
236
654000
2000
11:14
for the six and a half billion.
237
656000
3000
11:17
It's going to be a stretch to do it for nine.
238
659000
2000
11:19
We use over five billion tons of coal,
239
661000
3000
11:22
30 billion-plus barrels of oil --
240
664000
3000
11:25
that's a hundred million barrels a day.
241
667000
4000
11:29
When we try to think of biological processes
242
671000
2000
11:31
or any process to replace that,
243
673000
3000
11:34
it's going to be a huge challenge.
244
676000
2000
11:36
Then of course, there's all that
245
678000
2000
11:38
CO2 from this material
246
680000
2000
11:40
that ends up in the atmosphere.
247
682000
3000
11:43
We now, from our discovery around the world,
248
685000
2000
11:45
have a database with about 20 million genes,
249
687000
4000
11:49
and I like to think of these as the design components of the future.
250
691000
4000
11:53
The electronics industry only had a dozen or so components,
251
695000
3000
11:56
and look at the diversity that came out of that.
252
698000
4000
12:00
We're limited here primarily
253
702000
2000
12:02
by a biological reality
254
704000
2000
12:04
and our imagination.
255
706000
2000
12:07
We now have techniques,
256
709000
2000
12:09
because of these rapid methods of synthesis,
257
711000
3000
12:12
to do what we're calling combinatorial genomics.
258
714000
4000
12:16
We have the ability now to build a large robot
259
718000
3000
12:19
that can make a million chromosomes a day.
260
721000
3000
12:23
When you think of processing these 20 million different genes
261
725000
3000
12:26
or trying to optimize processes
262
728000
2000
12:28
to produce octane or to produce pharmaceuticals,
263
730000
3000
12:31
new vaccines,
264
733000
3000
12:34
we can just with a small team,
265
736000
3000
12:37
do more molecular biology
266
739000
2000
12:39
than the last 20 years of all science.
267
741000
3000
12:42
And it's just standard selection:
268
744000
2000
12:44
we can select for viability,
269
746000
2000
12:46
chemical or fuel production,
270
748000
2000
12:48
vaccine production, etc.
271
750000
2000
12:50
This is a screen snapshot
272
752000
3000
12:53
of some true design software
273
755000
3000
12:56
that we're working on to actually be able to sit down
274
758000
3000
12:59
and design species in the computer.
275
761000
3000
13:03
You know, we don't know necessarily what it'll look like:
276
765000
3000
13:06
we know exactly what their genetic code looks like.
277
768000
3000
13:09
We're focusing on now fourth-generation fuels.
278
771000
5000
13:15
You've seen recently, corn to ethanol
279
777000
2000
13:17
is just a bad experiment.
280
779000
2000
13:19
We have second- and third-generation fuels
281
781000
2000
13:21
that will be coming out relatively soon
282
783000
3000
13:24
that are sugar, to much higher-value fuels
283
786000
3000
13:27
like octane or different types of butanol.
284
789000
3000
13:30
But the only way we think that biology
285
792000
3000
13:33
can have a major impact without
286
795000
2000
13:36
further increasing the cost of food and limiting its availability
287
798000
3000
13:39
is if we start with CO2 as its feedstock,
288
801000
3000
13:42
and so we're working with designing cells to go down this road.
289
804000
4000
13:47
And we think we'll have the first fourth-generation fuels
290
809000
3000
13:50
in about 18 months.
291
812000
2000
13:52
Sunlight and CO2 is one method ...
292
814000
2000
13:54
(Applause)
293
816000
5000
13:59
but in our discovery around the world,
294
821000
2000
14:01
we have all kinds of other methods.
295
823000
2000
14:03
This is an organism we described in 1996.
296
825000
4000
14:07
It lives in the deep ocean,
297
829000
2000
14:09
about a mile and a half deep,
298
831000
2000
14:11
almost at boiling-water temperatures.
299
833000
2000
14:13
It takes CO2 to methane
300
835000
3000
14:16
using molecular hydrogen as its energy source.
301
838000
3000
14:19
We're looking to see if we can take
302
841000
2000
14:21
captured CO2,
303
843000
2000
14:23
which can easily be piped to sites,
304
845000
2000
14:25
convert that CO2 back into fuel
305
847000
3000
14:28
to drive this process.
306
850000
3000
14:31
So, in a short period of time,
307
853000
2000
14:33
we think that we might be able to increase
308
855000
4000
14:37
what the basic question is of "What is life?"
309
859000
3000
14:40
We truly, you know,
310
862000
2000
14:42
have modest goals
311
864000
2000
14:44
of replacing the whole petrol-chemical industry --
312
866000
3000
14:47
(Laughter) (Applause)
313
869000
3000
14:50
Yeah. If you can't do that at TED, where can you? --
314
872000
3000
14:53
(Laughter)
315
875000
2000
14:55
become a major source of energy ...
316
877000
2000
14:57
But also, we're now working on using these same tools
317
879000
3000
15:00
to come up with instant sets of vaccines.
318
882000
3000
15:03
You've seen this year with flu;
319
885000
2000
15:05
we're always a year behind and a dollar short
320
887000
3000
15:08
when it comes to the right vaccine.
321
890000
2000
15:10
I think that can be changed
322
892000
2000
15:12
by building combinatorial vaccines in advance.
323
894000
3000
15:16
Here's what the future may begin to look like
324
898000
3000
15:19
with changing, now, the evolutionary tree,
325
901000
4000
15:23
speeding up evolution
326
905000
2000
15:25
with synthetic bacteria, Archaea
327
907000
3000
15:28
and, eventually, eukaryotes.
328
910000
3000
15:32
We're a ways away from improving people:
329
914000
2000
15:34
our goal is just to make sure that we have a chance
330
916000
3000
15:37
to survive long enough to maybe do that. Thank you very much.
331
919000
3000
15:40
(Applause)
332
922000
7000
ABOUT THE SPEAKER
Craig Venter - Biologist, genetics pioneerIn 2001, Craig Venter made headlines for sequencing the human genome. In 2003, he started mapping the ocean's biodiversity. And now he's created the first synthetic lifeforms -- microorganisms that can produce alternative fuels.
Why you should listen
Craig Venter, the man who led the private effort to sequence the human genome, is hard at work now on even more potentially world-changing projects.
First, there's his mission aboard the Sorcerer II, a 92-foot yacht, which, in 2006, finished its voyage around the globe to sample, catalouge and decode the genes of the ocean's unknown microorganisms. Quite a task, when you consider that there are tens of millions of microbes in a single drop of sea water. Then there's the J. Craig Venter Institute, a nonprofit dedicated to researching genomics and exploring its societal implications.
In 2005, Venter founded Synthetic Genomics, a private company with a provocative mission: to engineer new life forms. Its goal is to design, synthesize and assemble synthetic microorganisms that will produce alternative fuels, such as ethanol or hydrogen. He was on Time magzine's 2007 list of the 100 Most Influential People in the World.
In early 2008, scientists at the J. Craig Venter Institute announced that they had manufactured the entire genome of a bacterium by painstakingly stitching together its chemical components. By sequencing a genome, scientists can begin to custom-design bootable organisms, creating biological robots that can produce from scratch chemicals humans can use, such as biofuel. And in 2010, they announced, they had created "synthetic life" -- DNA created digitally, inserted into a living bacterium, and remaining alive.
Craig Venter | Speaker | TED.com