1
00:00:08,914 --> 00:00:10,080
UYANGA TSEDEV: So hi, girls.

2
00:00:10,080 --> 00:00:11,420
My name is Uyanga.

3
00:00:11,420 --> 00:00:13,820
And I am currently
at the MIT Koch

4
00:00:13,820 --> 00:00:17,154
Institute, which is a building
just for cancer research.

5
00:00:17,154 --> 00:00:18,570
And today, I'm
going to talk about

6
00:00:18,570 --> 00:00:22,010
the biological engineering
that we do in order to find

7
00:00:22,010 --> 00:00:26,910
and image deeply embedded and
difficult-to-reach tumors.

8
00:00:26,910 --> 00:00:29,990
So currently in the clinic,
the number one therapy method

9
00:00:29,990 --> 00:00:36,840
for cancer is-- thank
you-- for cancer

10
00:00:36,840 --> 00:00:41,390
is surgical debulking, which is
basically removing of the tumor

11
00:00:41,390 --> 00:00:43,480
masses, and then chemotherapy.

12
00:00:43,480 --> 00:00:46,240
And so it's very
important for surgeons

13
00:00:46,240 --> 00:00:49,390
to be able to locate
precisely and early

14
00:00:49,390 --> 00:00:51,920
on in the stage of the
tumor the masses that they

15
00:00:51,920 --> 00:00:57,030
want to remove in order to do
the most effective therapy.

16
00:00:57,030 --> 00:01:00,310
And so cancer
cells are basically

17
00:01:00,310 --> 00:01:03,040
cells that have unusual
dividing abilities.

18
00:01:03,040 --> 00:01:05,430
So they proliferate
without stop.

19
00:01:05,430 --> 00:01:10,090
And they are differentiated from
the regular cells around them

20
00:01:10,090 --> 00:01:13,770
by certain markers on
the surface of the cell.

21
00:01:13,770 --> 00:01:16,610
And so what we want to do is
use these markers on the surface

22
00:01:16,610 --> 00:01:22,340
in the cell in order to
find and then locate where

23
00:01:22,340 --> 00:01:23,964
the tumors are, and
tell the surgeons,

24
00:01:23,964 --> 00:01:25,380
this is the place
where you should

25
00:01:25,380 --> 00:01:27,310
remove the biological tissue.

26
00:01:27,310 --> 00:01:29,730
And you don't want to damage
the healthy tissue around it.

27
00:01:29,730 --> 00:01:31,021
So you want to be very precise.

28
00:01:38,280 --> 00:01:41,600
So there are two tools
that we use in our lab

29
00:01:41,600 --> 00:01:45,200
in order to find the
tumors in the body.

30
00:01:45,200 --> 00:01:48,120
The first is the
custom-built imager

31
00:01:48,120 --> 00:01:51,710
that we're working on with
collaboration at the Lincoln

32
00:01:51,710 --> 00:01:54,850
Labs, and the second is
the biological probe.

33
00:01:54,850 --> 00:01:56,750
It's a little shuttle
that we use in order

34
00:01:56,750 --> 00:02:00,714
to go ahead and find
the cells in the body.

35
00:02:00,714 --> 00:02:02,380
So we can talk about
both of them today.

36
00:02:05,210 --> 00:02:06,950
So the biological
marker that we use

37
00:02:06,950 --> 00:02:09,090
is called the M13 bacteriophage.

38
00:02:09,090 --> 00:02:12,300
And bacteriophage really
just means bacteria eater.

39
00:02:12,300 --> 00:02:13,620
So it's a virus.

40
00:02:13,620 --> 00:02:20,790
So as with typical viruses, it
infects bacteria, [INAUDIBLE]

41
00:02:20,790 --> 00:02:24,240
takes over the local
mechanism of the bacteria,

42
00:02:24,240 --> 00:02:27,200
and begins to replicate itself.

43
00:02:27,200 --> 00:02:29,680
So what we can do
is hack this system

44
00:02:29,680 --> 00:02:32,410
in order to use this molecule
for our own benefits.

45
00:02:32,410 --> 00:02:36,050
It's a very beautiful, very
simple sort of molecule.

46
00:02:36,050 --> 00:02:38,250
It's just a piece
of DNA-- so a DNA

47
00:02:38,250 --> 00:02:40,300
that codes for its
genetic information--

48
00:02:40,300 --> 00:02:43,870
and a bunch of proteins that's
wrapped around the body.

49
00:02:43,870 --> 00:02:45,930
And so because we can
manipulate the DNA,

50
00:02:45,930 --> 00:02:49,770
we can change the peptides that
are on the body of the phage.

51
00:02:49,770 --> 00:02:51,890
And so we can change
the protein so that they

52
00:02:51,890 --> 00:02:55,780
have specific binding
abilities to certain materials

53
00:02:55,780 --> 00:02:58,040
or certain ligands,
or certain markers.

54
00:02:58,040 --> 00:03:01,750
So this is the tumor cell
markers that I'm talking about.

55
00:03:01,750 --> 00:03:05,940
We can change the protein
at the head of the virus

56
00:03:05,940 --> 00:03:10,290
in order to bind specifically
to the surface of a cancer cell.

57
00:03:10,290 --> 00:03:12,580
And we can change the
proteins on the body

58
00:03:12,580 --> 00:03:15,680
of the virus to bind to
a specific imaging agent

59
00:03:15,680 --> 00:03:17,900
or therapy agent that
we can then shuttle

60
00:03:17,900 --> 00:03:19,611
to the surface of our cells.

61
00:03:24,330 --> 00:03:28,500
So for our system here, the
imaging agent that we chose

62
00:03:28,500 --> 00:03:30,570
is called a carbon nanotube.

63
00:03:30,570 --> 00:03:34,900
And so a carbon nanotube
is just a piece of graphene

64
00:03:34,900 --> 00:03:36,940
that's been rolled up.

65
00:03:36,940 --> 00:03:39,960
And what graphene
is is it's just

66
00:03:39,960 --> 00:03:41,140
a single layer of graphite.

67
00:03:41,140 --> 00:03:44,370
And graphite is what you find
in your pencils, in lead.

68
00:03:44,370 --> 00:03:47,090
And so it has very interesting
optical and electrical

69
00:03:47,090 --> 00:03:48,860
properties, and it's
very well-studied.

70
00:03:48,860 --> 00:03:50,370
But for us, what's
interesting about

71
00:03:50,370 --> 00:03:53,460
it is the wavelength at
which it emits light.

72
00:03:53,460 --> 00:03:57,440
This is in the second window
near infrared wavelength.

73
00:03:57,440 --> 00:04:00,320
And so it's slightly longer
than the visible wavelength.

74
00:04:00,320 --> 00:04:05,670
And the cool thing about this
is it's able to penetrate deeper

75
00:04:05,670 --> 00:04:06,927
into the biological tissue.

76
00:04:06,927 --> 00:04:09,010
And you have less scattering
and less interference

77
00:04:09,010 --> 00:04:10,301
from the surrounding materials.

78
00:04:15,510 --> 00:04:20,220
So finally, this is the
design of our shuttles

79
00:04:20,220 --> 00:04:22,920
that we send in to
find tumor cells.

80
00:04:22,920 --> 00:04:25,710
We have the M13
bacteriophage, which

81
00:04:25,710 --> 00:04:28,640
has a particular
protein at the head that

82
00:04:28,640 --> 00:04:30,160
detects the tumor cells.

83
00:04:30,160 --> 00:04:34,870
And we have on the body the
CNT, the carbon nanotube,

84
00:04:34,870 --> 00:04:36,570
that is going to
light up and tell us

85
00:04:36,570 --> 00:04:38,778
this is where we should go
ahead and remove the mass.

86
00:04:42,637 --> 00:04:44,970
So the first model that we
have actually done in our lab

87
00:04:44,970 --> 00:04:46,080
is on ovarian tumor.

88
00:04:46,080 --> 00:04:48,930
And ovarian tumor is one
of the tumor [INAUDIBLE]

89
00:04:48,930 --> 00:04:52,900
that is very difficult
to detect early on.

90
00:04:52,900 --> 00:04:55,040
Unlike things like breast
tumor, you can't feel it.

91
00:04:55,040 --> 00:04:57,930
And usually when the surgeon
or the doctor diagnoses,

92
00:04:57,930 --> 00:05:00,110
it's quite late in the
stage of the tumor.

93
00:05:00,110 --> 00:05:01,920
So it could have
turned metastatic.

94
00:05:01,920 --> 00:05:04,614
And metastatic means
that the tumor cells

95
00:05:04,614 --> 00:05:06,280
have detached from
the body and traveled

96
00:05:06,280 --> 00:05:09,807
to other parts of the body, and
so proliferated other tumors

97
00:05:09,807 --> 00:05:10,640
throughout the body.

98
00:05:10,640 --> 00:05:13,950
So this is a stage you
don't want to reach.

99
00:05:13,950 --> 00:05:18,900
So it'd be very cool if you
can put our probes into mice

100
00:05:18,900 --> 00:05:21,850
with a ovarian tumor,
light up the tumors,

101
00:05:21,850 --> 00:05:24,000
give the surgeon the
image, and the surgeon can

102
00:05:24,000 --> 00:05:26,870
go ahead and remove the tumors,
and the mice will then survive.

103
00:05:26,870 --> 00:05:29,714
So this is a first step
in putting our system

104
00:05:29,714 --> 00:05:30,880
to its clinical application.

105
00:05:33,750 --> 00:05:36,180
So this is what
happens in real life.

106
00:05:36,180 --> 00:05:39,990
You see here the stomach
cavity of the mice.

107
00:05:39,990 --> 00:05:43,470
And in the bright orange,
you see all the tumor masses

108
00:05:43,470 --> 00:05:44,550
that have lit up.

109
00:05:44,550 --> 00:05:47,130
So we have injected
the mice with trillions

110
00:05:47,130 --> 00:05:49,010
of our tiny little
molecular shuttles

111
00:05:49,010 --> 00:05:52,480
that carry our CNT to the
location of the tumor.

112
00:05:52,480 --> 00:05:57,980
And using our imager, which
we have built with help

113
00:05:57,980 --> 00:06:00,270
from the Lincoln
Labs, the surgeon

114
00:06:00,270 --> 00:06:02,440
is able to go in and
remove all these nodules.

115
00:06:02,440 --> 00:06:04,980
And the cool thing
about the nodules

116
00:06:04,980 --> 00:06:07,030
here is we're getting
the ones that are even

117
00:06:07,030 --> 00:06:08,680
below a millimeter in diameter.

118
00:06:08,680 --> 00:06:11,720
So usually, when a surgeon
goes in without any guidance,

119
00:06:11,720 --> 00:06:14,000
he might miss these because
he's just going by eye.

120
00:06:14,000 --> 00:06:15,700
But because he
has the imager, he

121
00:06:15,700 --> 00:06:17,741
can go ahead and remove
most of the tumors.

122
00:06:21,850 --> 00:06:24,440
So here's an example of what
we're doing in real time.

123
00:06:24,440 --> 00:06:28,680
So during surgery, if
the surgeon can directly

124
00:06:28,680 --> 00:06:31,600
refer to the images
that he's getting,

125
00:06:31,600 --> 00:06:34,400
then he can go ahead and
implement our technology

126
00:06:34,400 --> 00:06:36,040
in a very real way.

127
00:06:36,040 --> 00:06:39,425
So here, we see Dr.
[INAUDIBLE] from MGH.

128
00:06:39,425 --> 00:06:41,280
He's doing surgery on mice.

129
00:06:41,280 --> 00:06:45,107
And you can see on the screen,
he can see brightly lit up--

130
00:06:45,107 --> 00:06:46,690
this is where the
tumors are, and this

131
00:06:46,690 --> 00:06:47,830
is where I should take it out.

132
00:06:47,830 --> 00:06:49,980
And so here's a serial
image of what's happening.

133
00:06:49,980 --> 00:06:54,710
So here's the probes lighting
up where the tumors are.

134
00:06:54,710 --> 00:06:57,550
This is what happens when
the surgeon goes in by eye

135
00:06:57,550 --> 00:06:59,550
and is like, OK,
from my experience,

136
00:06:59,550 --> 00:07:01,410
this is where I should
take out the tumors.

137
00:07:01,410 --> 00:07:04,050
And you see that there's
quite a bit of mass left.

138
00:07:04,050 --> 00:07:06,520
But if we have the
imager, then we

139
00:07:06,520 --> 00:07:08,730
have a much cleaner image
at the end of the surgery.

140
00:07:08,730 --> 00:07:10,626
And so there's a
lot less of a chance

141
00:07:10,626 --> 00:07:12,000
that the tumor is
going to recur.

142
00:07:16,539 --> 00:07:19,080
So for me personally, there are
two things I'm doing in order

143
00:07:19,080 --> 00:07:22,620
to expand on our system
that we have already built.

144
00:07:22,620 --> 00:07:26,650
The first is manipulating
the M13 phage

145
00:07:26,650 --> 00:07:28,390
to different geometries.

146
00:07:28,390 --> 00:07:29,590
So I call this inhophage.

147
00:07:29,590 --> 00:07:33,860
Inho just means
small in Portuguese.

148
00:07:33,860 --> 00:07:36,760
And so based on the aspect
ratio, which basically

149
00:07:36,760 --> 00:07:39,630
is the diameter versus
the length of our phage,

150
00:07:39,630 --> 00:07:42,410
they might have very
different traveling abilities

151
00:07:42,410 --> 00:07:43,450
in the bloodstream.

152
00:07:43,450 --> 00:07:44,980
So because of
their oblong shape,

153
00:07:44,980 --> 00:07:48,760
they have a tendency to stick
to the walls and tumble around.

154
00:07:48,760 --> 00:07:51,781
But if we can change this
shape to an optimal size,

155
00:07:51,781 --> 00:07:53,280
then there's a
chance that they will

156
00:07:53,280 --> 00:07:56,070
travel for a longer period
of time in the bloodstream,

157
00:07:56,070 --> 00:07:58,060
and so find the tumors
that are farther

158
00:07:58,060 --> 00:08:00,950
from the site of injection.

159
00:08:00,950 --> 00:08:03,690
The second thing is we are
expanding on the library

160
00:08:03,690 --> 00:08:07,080
of cancers that we can attack.

161
00:08:07,080 --> 00:08:09,420
So so far, we have
looked at ovarian tumor,

162
00:08:09,420 --> 00:08:11,664
and now we want to
look at brain tumor.

163
00:08:11,664 --> 00:08:13,580
And the most difficult
thing about brain tumor

164
00:08:13,580 --> 00:08:15,290
is the blood-brain barrier.

165
00:08:15,290 --> 00:08:17,700
So it's just a layer of cells
that separate your blood

166
00:08:17,700 --> 00:08:20,020
from the matter in your brain.

167
00:08:20,020 --> 00:08:21,610
And the molecules
that pass through

168
00:08:21,610 --> 00:08:23,480
are very specific
and very small.

169
00:08:23,480 --> 00:08:26,980
So it's very hard to deliver
from your bloodstream

170
00:08:26,980 --> 00:08:32,250
various drugs or imaging agents
to the actual brain mass.

171
00:08:32,250 --> 00:08:35,299
So what we have done
here is actually

172
00:08:35,299 --> 00:08:37,700
engineer the head
peptides-- so the proteins

173
00:08:37,700 --> 00:08:42,440
at the head of our phage--
so that it allows for travel

174
00:08:42,440 --> 00:08:46,630
across the blood-brain barrier
so it can deliver these imaging

175
00:08:46,630 --> 00:08:48,810
molecules to tumors
in the brain.

176
00:08:52,760 --> 00:08:56,700
So one thing to remember here
is that the scale at which we're

177
00:08:56,700 --> 00:08:58,680
working is very, very small.

178
00:08:58,680 --> 00:09:01,650
So if you look at
CNT or our phage,

179
00:09:01,650 --> 00:09:05,986
really, it's about
100,000 times smaller

180
00:09:05,986 --> 00:09:07,110
than a strand of your hair.

181
00:09:07,110 --> 00:09:10,110
So these are really,
really tiny things.

182
00:09:10,110 --> 00:09:13,040
So when I make
these small phage,

183
00:09:13,040 --> 00:09:15,010
they're even smaller
than our regular phage.

184
00:09:15,010 --> 00:09:18,500
Our regular phage is
about 800 nanometers.

185
00:09:18,500 --> 00:09:22,530
So here's an example of 280
nanometer phage, 100 nanometer

186
00:09:22,530 --> 00:09:24,782
phage, and 50 nanometer phage.

187
00:09:24,782 --> 00:09:26,740
So when I make them and
I want to look at them,

188
00:09:26,740 --> 00:09:28,580
it's really hard because
they're so small.

189
00:09:28,580 --> 00:09:31,440
And you can't just use
regular light microscopy

190
00:09:31,440 --> 00:09:33,090
to detect them.

191
00:09:33,090 --> 00:09:34,740
So another interesting
thing that I do

192
00:09:34,740 --> 00:09:37,152
is use atomic force microscopy.

193
00:09:37,152 --> 00:09:38,610
I don't know if
you've heard of it.

194
00:09:38,610 --> 00:09:42,330
But what it is is basically
you take a sample,

195
00:09:42,330 --> 00:09:44,360
and you have this
very, very tiny needle

196
00:09:44,360 --> 00:09:46,780
that scans across the surface.

197
00:09:46,780 --> 00:09:49,450
And so you're getting a
topology of the surface.

198
00:09:49,450 --> 00:09:52,460
And from that scanning, kind
of like reading Braille,

199
00:09:52,460 --> 00:09:55,640
you get a good idea of what
the profile of your image is.

200
00:09:55,640 --> 00:09:58,101
So these are the images that
I get from this sort of needle

201
00:09:58,101 --> 00:09:58,600
probing.

202
00:10:04,280 --> 00:10:06,740
For the brain tumor,
another interesting factor--

203
00:10:06,740 --> 00:10:10,620
not only can we bring
the phage to the surface

204
00:10:10,620 --> 00:10:13,592
of the tumor-- so this is
tumor here shown in green.

205
00:10:13,592 --> 00:10:16,050
This is tumor that actually
has green fluorescent proteins,

206
00:10:16,050 --> 00:10:18,600
so that means they glow.

207
00:10:18,600 --> 00:10:21,170
And if we send in our
phage, we see that, oh,

208
00:10:21,170 --> 00:10:26,430
yeah, our phage goes ahead
and coats in red the tumor.

209
00:10:26,430 --> 00:10:27,980
But then the other
thing we notice

210
00:10:27,980 --> 00:10:31,381
is that the phage is actually
internalized into the tumor

211
00:10:31,381 --> 00:10:31,880
cells.

212
00:10:31,880 --> 00:10:35,050
So the phage comes and
attaches to the surface,

213
00:10:35,050 --> 00:10:37,400
and then the tumor
cells eats the phage.

214
00:10:37,400 --> 00:10:41,840
So in the blue here is
the nucleus of a phage.

215
00:10:41,840 --> 00:10:45,700
And in the red spread around
the nucleus in the Golgi region

216
00:10:45,700 --> 00:10:46,970
is our phage.

217
00:10:46,970 --> 00:10:50,480
So it just tells us we can
possibly deliver things to

218
00:10:50,480 --> 00:10:53,150
inside the cells of the cancer.

219
00:10:53,150 --> 00:10:55,260
So that means we could
send in therapies

220
00:10:55,260 --> 00:10:58,360
that disrupt the function,
and so in this way,

221
00:10:58,360 --> 00:10:59,830
kill the tumor cells.

222
00:10:59,830 --> 00:11:02,220
And so can we not only
detect where they are

223
00:11:02,220 --> 00:11:04,060
and see if it can
remove them effectively,

224
00:11:04,060 --> 00:11:07,100
but we can also
locally cure them.

225
00:11:11,110 --> 00:11:17,340
So overall, today, we've looked
at the very real medical impact

226
00:11:17,340 --> 00:11:20,050
that we can do with
imaging and the probe

227
00:11:20,050 --> 00:11:22,080
that we have biologically
engineered in our lab.

228
00:11:22,080 --> 00:11:24,440
And the two things
that I've done

229
00:11:24,440 --> 00:11:27,844
is look at what we can do
to change the geometry,

230
00:11:27,844 --> 00:11:30,260
and what we can do to change
the different cancers that we

231
00:11:30,260 --> 00:11:30,810
can attack.

232
00:11:34,742 --> 00:11:36,200
And all of this
work, of course, is

233
00:11:36,200 --> 00:11:38,930
impossible without
the collaboration that

234
00:11:38,930 --> 00:11:43,190
happens between the colleagues I
have at the MIT Koch Institute,

235
00:11:43,190 --> 00:11:46,142
as well as the scientists
we have at the Lincoln Labs.

236
00:11:46,142 --> 00:11:48,600
And so that's another really
cool thing about this project.

237
00:11:48,600 --> 00:11:50,840
It's so very broad,
and so it brings

238
00:11:50,840 --> 00:11:52,590
in so many different
techniques and brings

239
00:11:52,590 --> 00:11:53,770
in so many different people.

240
00:11:53,770 --> 00:11:55,750
And so I really
want to thank them.

241
00:11:55,750 --> 00:12:00,320
And another aspect of this work
that I really find meaningful

242
00:12:00,320 --> 00:12:04,000
is that I am personally
funded by families

243
00:12:04,000 --> 00:12:06,180
that are affected by
these types of tumors,

244
00:12:06,180 --> 00:12:07,250
these types of diseases.

245
00:12:07,250 --> 00:12:12,140
And so I really want to thank
the Goodwin, [INAUDIBLE],

246
00:12:12,140 --> 00:12:15,830
[INAUDIBLE] and [INAUDIBLE]
families who have been really

247
00:12:15,830 --> 00:12:17,940
generous with their
funding for this project.

248
00:12:21,370 --> 00:12:24,959
And finally, I want to say
that for me, the M13 is

249
00:12:24,959 --> 00:12:26,000
kind of a building block.

250
00:12:26,000 --> 00:12:28,330
We're here all as
Girls Who Build.

251
00:12:28,330 --> 00:12:32,190
And so for me, for my building
project, it's the M13.

252
00:12:32,190 --> 00:12:36,970
And I hope that you guys are
able to understand from this

253
00:12:36,970 --> 00:12:39,110
that as an engineer,
for me, it's really just

254
00:12:39,110 --> 00:12:43,670
about finding creative ways to
use my building blocks to make

255
00:12:43,670 --> 00:12:46,270
a difference, to address
the problems that I think

256
00:12:46,270 --> 00:12:49,150
are most important to me.

257
00:12:49,150 --> 00:12:52,580
And so thank you very
much for listening.

258
00:12:52,580 --> 00:12:55,780
Please feel free to
ask me any questions.

259
00:12:55,780 --> 00:12:59,480
I can talk more about myself, or
I can talk more about research.

260
00:12:59,480 --> 00:13:01,760
But yes, thank you.

261
00:13:01,760 --> 00:13:05,260
[APPLAUSE]

262
00:13:08,260 --> 00:13:10,443
PROFESSOR: Any
questions for her?

263
00:13:10,443 --> 00:13:11,234
UYANGA TSEDEV: Yes?

264
00:13:11,234 --> 00:13:13,604
AUDIENCE: You were
using specific examples

265
00:13:13,604 --> 00:13:18,520
for the cancer, but can these
detect any type of cancer,

266
00:13:18,520 --> 00:13:21,852
or is it just specific,
or a couple of types?

267
00:13:21,852 --> 00:13:23,810
UYANGA TSEDEV: It can
detect any type of cancer

268
00:13:23,810 --> 00:13:28,830
as long as you have a specific
differentiating marker

269
00:13:28,830 --> 00:13:31,460
on the cancer that you know of.

270
00:13:31,460 --> 00:13:34,320
So not only is this
useful in cancer,

271
00:13:34,320 --> 00:13:37,320
it's also useful in
detecting any other material.

272
00:13:37,320 --> 00:13:39,460
As long as you're able
to modify the head so

273
00:13:39,460 --> 00:13:44,700
that it is specific to that
material or to that cell,

274
00:13:44,700 --> 00:13:48,160
then you can do that, which
makes it a very elegant, very

275
00:13:48,160 --> 00:13:50,650
cool molecule, which is
why I think it's something

276
00:13:50,650 --> 00:13:51,630
I should be using.

277
00:13:51,630 --> 00:13:52,830
[CHUCKLING]

278
00:13:56,130 --> 00:13:59,130
PROFESSOR: Any other questions?

279
00:13:59,130 --> 00:13:59,730
Great.

280
00:13:59,730 --> 00:14:00,480
Thank you, Uyanga.

281
00:14:00,480 --> 00:14:01,230
[APPLAUSE]

282
00:14:01,230 --> 00:14:02,430
UYANGA TSEDEV: Thank you.

283
00:14:02,430 --> 00:14:05,780
[APPLAUSE]