1
00:00:00,500 --> 00:00:03,270
The following content is
provided under a Creative

2
00:00:03,270 --> 00:00:04,630
Commons license.

3
00:00:04,630 --> 00:00:07,140
Your support will help
MIT OpenCourseWare

4
00:00:07,140 --> 00:00:11,470
continue to offer high quality
educational resources for free.

5
00:00:11,470 --> 00:00:14,100
To make a donation or
view additional materials

6
00:00:14,100 --> 00:00:18,050
from hundreds of MIT courses,
visit MIT OpenCourseWare

7
00:00:18,050 --> 00:00:19,000
at ocw.mit.edu.

8
00:00:24,121 --> 00:00:25,110
REUBEN: I'm Reuben.

9
00:00:25,110 --> 00:00:28,070
I am currently a
senior in Chemistry.

10
00:00:28,070 --> 00:00:31,820
And I am a UROP
in Bob Sauer's lab

11
00:00:31,820 --> 00:00:34,520
over in the
Department of Biology.

12
00:00:34,520 --> 00:00:36,116
As you guys know
from this class,

13
00:00:36,116 --> 00:00:37,490
one of the modules
that you cover

14
00:00:37,490 --> 00:00:42,920
is ClpXP, the bacterial
analog of the proteasome.

15
00:00:42,920 --> 00:00:46,970
And today I hope to tell you
about some of the cool work

16
00:00:46,970 --> 00:00:50,060
using modern biophysical
techniques that people

17
00:00:50,060 --> 00:00:54,260
in Bob's lab have done to gain
some really, in my opinion,

18
00:00:54,260 --> 00:00:59,060
interesting insight into the
actual mechanical mechanism

19
00:00:59,060 --> 00:01:03,230
of this motor protein ClpXP
as it unfolds and degrades

20
00:01:03,230 --> 00:01:05,060
protein substrates.

21
00:01:05,060 --> 00:01:07,500
Before I get started,
I want to say,

22
00:01:07,500 --> 00:01:10,400
please stop me at any
time to ask questions.

23
00:01:10,400 --> 00:01:14,420
If I say something that doesn't
make sense, please let me

24
00:01:14,420 --> 00:01:17,090
know, and I will try to clarify.

25
00:01:17,090 --> 00:01:19,550
So today I'm mostly
going to be talking

26
00:01:19,550 --> 00:01:25,670
about modern single molecule
methods, which are the cutting

27
00:01:25,670 --> 00:01:27,010
edge in biophysics.

28
00:01:27,010 --> 00:01:29,150
And so the first question
I wanted to ask you

29
00:01:29,150 --> 00:01:32,060
guys is, what are the
possible advantages

30
00:01:32,060 --> 00:01:34,730
of looking at
molecules one at a time

31
00:01:34,730 --> 00:01:38,800
rather than performing
some sort of bulk assay?

32
00:01:38,800 --> 00:01:42,740
Do any of you have any ideas?

33
00:01:42,740 --> 00:01:45,500
AUDIENCE: Maybe the average of
a bunch of different molecules

34
00:01:45,500 --> 00:01:47,990
isn't truly
representative of what

35
00:01:47,990 --> 00:01:50,399
a single molecule looks like.

36
00:01:50,399 --> 00:01:50,940
REUBEN: Yeah.

37
00:01:50,940 --> 00:01:53,790
So the average is the average.

38
00:01:53,790 --> 00:01:56,750
But the average as a statistic
has some sort of weakness.

39
00:01:56,750 --> 00:02:00,950
It obscures a lot of
the vagaries of behavior

40
00:02:00,950 --> 00:02:04,780
that may be lost
in that average.

41
00:02:04,780 --> 00:02:08,889
So it's worth saying that
in classical biochemistry

42
00:02:08,889 --> 00:02:12,110
you are looking at a lot of
molecules simultaneously.

43
00:02:12,110 --> 00:02:15,680
In one microliter of
one micromolar solution,

44
00:02:15,680 --> 00:02:18,730
that's 6 times 10 to
the 11th molecules.

45
00:02:18,730 --> 00:02:24,620
So taking some sort of
bulk measurement averages

46
00:02:24,620 --> 00:02:26,750
across all of these molecules.

47
00:02:26,750 --> 00:02:29,630
And as you said, you can lose
a great deal of information

48
00:02:29,630 --> 00:02:34,350
about the variation and the
dynamics of these molecules.

49
00:02:34,350 --> 00:02:37,970
So another thing that
you can lose, in addition

50
00:02:37,970 --> 00:02:40,640
to information
about variation, is

51
00:02:40,640 --> 00:02:42,920
you can lose information
about the procession

52
00:02:42,920 --> 00:02:46,440
through some sort of
complicated biological process.

53
00:02:46,440 --> 00:02:50,240
So let's say that you
are studying ClpXP

54
00:02:50,240 --> 00:02:54,620
as it recognizes, unfolds, and
degrades protein substrates.

55
00:02:54,620 --> 00:02:56,510
If you were looking
at, for example,

56
00:02:56,510 --> 00:02:59,090
6 times 10 to the 11th
molecules of ClpXP

57
00:02:59,090 --> 00:03:01,190
sitting in a test
tube, then you're

58
00:03:01,190 --> 00:03:05,000
smearing across all of these
different time points of ClpXP,

59
00:03:05,000 --> 00:03:07,730
where you have some ClpXPs
that are unfolding substrates,

60
00:03:07,730 --> 00:03:09,590
some ClpXPs that are
not bound to substrate

61
00:03:09,590 --> 00:03:12,740
at all, some ClpXPs that
are translocating substrate.

62
00:03:12,740 --> 00:03:15,570
So any sort of
readout that you have

63
00:03:15,570 --> 00:03:19,550
loses this information about the
different states that ClpXP can

64
00:03:19,550 --> 00:03:20,360
inhabit.

65
00:03:20,360 --> 00:03:22,760
And also, it's very
difficult to gain

66
00:03:22,760 --> 00:03:26,070
detailed kinetic information
about how ClpXP would,

67
00:03:26,070 --> 00:03:29,360
for example, transit between
these different states

68
00:03:29,360 --> 00:03:33,810
if all you can see
is a bulk average.

69
00:03:33,810 --> 00:03:36,870
So another way of
putting this is

70
00:03:36,870 --> 00:03:39,540
that a bulk measurement
like fluorescence, you're

71
00:03:39,540 --> 00:03:43,050
looking at the unsynchronized
activity of these molecules.

72
00:03:43,050 --> 00:03:45,580
But by studying single
molecules at a time,

73
00:03:45,580 --> 00:03:48,240
you can do some sort of
post hoc synchronization

74
00:03:48,240 --> 00:03:53,250
to actually gain insight into
the kinetics, or the order

75
00:03:53,250 --> 00:03:55,620
of the different
states during some sort

76
00:03:55,620 --> 00:03:57,903
of biophysical process.

77
00:03:57,903 --> 00:04:00,236
JOANNE STUBBE: You can do
better than that, even in bulk

78
00:04:00,236 --> 00:04:01,930
you can synchronize, right?

79
00:04:01,930 --> 00:04:02,610
REUBEN: Yeah.

80
00:04:02,610 --> 00:04:05,259
So there are certainly
tools like stopped-flow.

81
00:04:05,259 --> 00:04:06,842
JOANNE STUBBE: Which
is what we talked

82
00:04:06,842 --> 00:04:08,646
about in the first
couple of recitations.

83
00:04:08,646 --> 00:04:09,548
That's why I brought it up.

84
00:04:09,548 --> 00:04:10,089
REUBEN: Yeah.

85
00:04:10,089 --> 00:04:12,110
So stopped-flow is
a fantastic tool,

86
00:04:12,110 --> 00:04:14,210
which has been used
really successfully

87
00:04:14,210 --> 00:04:17,050
to study a lot of systems,
particularly, I think,

88
00:04:17,050 --> 00:04:19,479
the Rodnina lab, which I
think you guys covered,

89
00:04:19,479 --> 00:04:21,200
used stopped-flow
very effectively

90
00:04:21,200 --> 00:04:24,630
to study the early
stages of translation.

91
00:04:24,630 --> 00:04:28,220
But I think that stopped-flow
is good for looking

92
00:04:28,220 --> 00:04:30,310
at some early
stages of processes,

93
00:04:30,310 --> 00:04:33,440
but for trying to track a
long, complicated process

94
00:04:33,440 --> 00:04:35,870
across a significant
period of time,

95
00:04:35,870 --> 00:04:38,510
stopped-flow is not
really an ideal technique.

96
00:04:38,510 --> 00:04:42,110
Because what happens is that
the rates of transitions

97
00:04:42,110 --> 00:04:44,500
between different processes--

98
00:04:44,500 --> 00:04:47,270
let's say that there is
a single rate constant.

99
00:04:47,270 --> 00:04:49,250
The actual time that
an individual molecule

100
00:04:49,250 --> 00:04:51,910
spends in a state before
switching to the next state

101
00:04:51,910 --> 00:04:53,000
is stochastic.

102
00:04:53,000 --> 00:04:54,470
It's an exponential
decay process

103
00:04:54,470 --> 00:04:55,920
based on the rate constant.

104
00:04:55,920 --> 00:05:00,140
So if you are looking at a
bunch of rates all put together,

105
00:05:00,140 --> 00:05:02,360
and you start a
stopped-flow experiment,

106
00:05:02,360 --> 00:05:04,970
it's true that for the
first couple of seconds

107
00:05:04,970 --> 00:05:06,590
all of the molecules
are synchronized.

108
00:05:06,590 --> 00:05:08,089
But over the rest
of the experiment,

109
00:05:08,089 --> 00:05:10,737
they blur out all the way
across the time space.

110
00:05:10,737 --> 00:05:13,070
And so you can really lose a
lot of detailed information

111
00:05:13,070 --> 00:05:14,187
about the kinetics.

112
00:05:14,187 --> 00:05:15,770
Whereas if you're
looking at molecules

113
00:05:15,770 --> 00:05:19,255
one molecule at a time, you
can track an entire process.

114
00:05:19,255 --> 00:05:21,296
JOANNE STUBBE: But again,
the issue where you can

115
00:05:21,296 --> 00:05:22,740
only look at them in one way--

116
00:05:22,740 --> 00:05:23,435
REUBEN: Oh, absolutely.

117
00:05:23,435 --> 00:05:23,720
JOANNE STUBBE: --ways.

118
00:05:23,720 --> 00:05:25,670
And so, if you
don't know anything

119
00:05:25,670 --> 00:05:27,146
about the bulk system--

120
00:05:27,146 --> 00:05:30,590
I hate it when people are
touting single-molecule,

121
00:05:30,590 --> 00:05:33,145
which I think is very
powerful, but the fact

122
00:05:33,145 --> 00:05:35,144
is that you have
many ways to look

123
00:05:35,144 --> 00:05:38,000
at reactions that
can't be carried over

124
00:05:38,000 --> 00:05:39,428
into the single-molecule venue.

125
00:05:39,428 --> 00:05:41,190
So that's the caveat.

126
00:05:41,190 --> 00:05:42,550
REUBEN: I completely agree.

127
00:05:42,550 --> 00:05:44,966
I was actually just about to,
after I finished this slide,

128
00:05:44,966 --> 00:05:47,270
ask what were the disadvantages
of single-molecule.

129
00:05:47,270 --> 00:05:48,561
JOANNE STUBBE: All right, sure.

130
00:05:48,561 --> 00:05:50,302
REUBEN: And you raise
a very good point.

131
00:05:50,302 --> 00:05:53,420
No, I always appreciate
comments from the audience--

132
00:05:53,420 --> 00:05:54,180
[LAUGHTER]

133
00:05:54,180 --> 00:05:56,040
--and from the faculty
in the audience.

134
00:06:00,360 --> 00:06:01,204
I don't know.

135
00:06:01,204 --> 00:06:03,120
It's true that there are
some measurements you

136
00:06:03,120 --> 00:06:04,630
can't make in single-molecule.

137
00:06:04,630 --> 00:06:06,796
But it's also true that
there are some measurements,

138
00:06:06,796 --> 00:06:08,630
such as the measurements
in this paper,

139
00:06:08,630 --> 00:06:10,420
that you can't make in bulk.

140
00:06:10,420 --> 00:06:13,810
So you can measure some unique
properties of molecules--

141
00:06:13,810 --> 00:06:17,140
particularly molecular machines
such as force generation

142
00:06:17,140 --> 00:06:18,810
and processive motion--

143
00:06:18,810 --> 00:06:21,360
using measurement techniques
such as the optical tweezers,

144
00:06:21,360 --> 00:06:24,510
which I'm going to talk
about a lot in this lecture,

145
00:06:24,510 --> 00:06:29,370
that there's no bulk analog
of a measurement like this.

146
00:06:29,370 --> 00:06:34,546
So JoAnne mentioned a couple
of the possible disadvantages

147
00:06:34,546 --> 00:06:35,670
of single-molecule studies.

148
00:06:35,670 --> 00:06:37,086
Do any of you have
any other ideas

149
00:06:37,086 --> 00:06:40,650
of potential disadvantages that
might make sure that you want

150
00:06:40,650 --> 00:06:42,940
to do some bulk experiments?

151
00:06:45,880 --> 00:06:49,650
So the really big one is that
single molecules, there aren't

152
00:06:49,650 --> 00:06:51,469
great ways to look at them.

153
00:06:51,469 --> 00:06:53,010
There's single-molecule
fluorescence,

154
00:06:53,010 --> 00:06:54,420
there's optical tweezers.

155
00:06:54,420 --> 00:06:57,330
After that, you know, I'm not
really sure what else there is.

156
00:06:57,330 --> 00:07:01,380
And both of these techniques
study from a significant noise

157
00:07:01,380 --> 00:07:02,010
issue.

158
00:07:02,010 --> 00:07:04,620
You can really only look
at one thing at a time.

159
00:07:04,620 --> 00:07:07,320
And a single
fluorophore, for example,

160
00:07:07,320 --> 00:07:09,080
in a single-molecule
fluorescence study,

161
00:07:09,080 --> 00:07:10,560
is not very bright.

162
00:07:10,560 --> 00:07:13,660
You have enormous noise coming
from your instrumentation.

163
00:07:13,660 --> 00:07:17,190
And it can be very difficult
to tangle out the signal

164
00:07:17,190 --> 00:07:20,850
that you're trying to go
after from your noise.

165
00:07:20,850 --> 00:07:23,880
Another major issue is that
many of these experiments

166
00:07:23,880 --> 00:07:25,574
are actually very
difficult to run,

167
00:07:25,574 --> 00:07:27,990
and it can take, for example,
months and months and months

168
00:07:27,990 --> 00:07:29,940
to do single-molecule
experiments, which,

169
00:07:29,940 --> 00:07:32,550
you know, a bulk protein
degradation experiment, you

170
00:07:32,550 --> 00:07:34,110
can mix some proteins
and run a gel,

171
00:07:34,110 --> 00:07:36,193
and actually learn a great
deal about the behavior

172
00:07:36,193 --> 00:07:38,060
of the proteins.

173
00:07:38,060 --> 00:07:40,590
But all the same, there
are some measurements,

174
00:07:40,590 --> 00:07:42,540
such as the measurements
in this paper,

175
00:07:42,540 --> 00:07:45,690
that you cannot take without
single-molecule measurements.

176
00:07:45,690 --> 00:07:49,350
So I hope that today I
will be able to tell you

177
00:07:49,350 --> 00:07:52,200
about some modern biophysical
techniques, particularly

178
00:07:52,200 --> 00:07:55,050
optical tweezers, so that
you can all learn about some

179
00:07:55,050 --> 00:07:58,990
of the latest tools in the
biophysicist's toolkit.

180
00:07:58,990 --> 00:08:01,311
So I am going to talk about
optical tweezers today,

181
00:08:01,311 --> 00:08:02,810
but I did want to
mention that there

182
00:08:02,810 --> 00:08:05,190
is another
single-molecule method,

183
00:08:05,190 --> 00:08:06,930
single-molecule fluorescence.

184
00:08:06,930 --> 00:08:09,330
And this is actually what
I do in my experiments.

185
00:08:09,330 --> 00:08:11,160
That's also very
useful for looking

186
00:08:11,160 --> 00:08:14,700
at the dynamical properties
of complicated biological

187
00:08:14,700 --> 00:08:16,330
processes.

188
00:08:16,330 --> 00:08:19,980
So single-molecule
fluorescence basically

189
00:08:19,980 --> 00:08:21,450
involves tracking
the fluorescence

190
00:08:21,450 --> 00:08:23,790
of single fluorophores
over time,

191
00:08:23,790 --> 00:08:25,634
and quantitating it,
and then studying

192
00:08:25,634 --> 00:08:28,050
the dynamics of the switching
between the different states

193
00:08:28,050 --> 00:08:29,550
of the fluorophore.

194
00:08:29,550 --> 00:08:31,860
Almost all of these
experiments use

195
00:08:31,860 --> 00:08:35,030
a physical property
called FRET, which

196
00:08:35,030 --> 00:08:38,789
is energy transfer between
two nearby fluorophores,

197
00:08:38,789 --> 00:08:41,133
the efficiency of which
is distance dependent.

198
00:08:41,133 --> 00:08:43,049
I believe that you guys
have learned about it.

199
00:08:43,049 --> 00:08:44,910
It's not that
difficult to understand

200
00:08:44,910 --> 00:08:47,790
how you could go from
doing FRET in bulk

201
00:08:47,790 --> 00:08:49,432
to FRET in single-molecule.

202
00:08:49,432 --> 00:08:51,390
If you haven't heard
about FRET, I'm sure you--

203
00:08:51,390 --> 00:08:52,890
JOANNE STUBBE: We'll hear
it in the last recitation.

204
00:08:52,890 --> 00:08:54,420
REUBEN: I'm sure you will
hear about it very soon.

205
00:08:54,420 --> 00:08:56,070
JOANNE STUBBE: [INAUDIBLE].

206
00:08:56,070 --> 00:08:57,240
REUBEN: Yeah.

207
00:08:57,240 --> 00:09:00,350
So this is a very
useful technique

208
00:09:00,350 --> 00:09:04,650
for answering some
questions, but in some ways

209
00:09:04,650 --> 00:09:06,484
it's a little bit
easier to understand

210
00:09:06,484 --> 00:09:07,400
than optical tweezers.

211
00:09:07,400 --> 00:09:09,441
So I figured I should talk
about optical tweezers

212
00:09:09,441 --> 00:09:11,810
today, because I think it's
also kind of cool that you

213
00:09:11,810 --> 00:09:15,790
can do the experiments that
are in the paper today.

214
00:09:15,790 --> 00:09:21,360
So optical tweezers--
the main technique

215
00:09:21,360 --> 00:09:23,880
used in the Olivares
and Sauer paper

216
00:09:23,880 --> 00:09:25,530
that I'm talking about today.

217
00:09:25,530 --> 00:09:27,660
And the general idea
of optical tweezers

218
00:09:27,660 --> 00:09:30,000
is that you can use
the momentum of light.

219
00:09:30,000 --> 00:09:32,790
Because as you guys know
from quantum mechanics,

220
00:09:32,790 --> 00:09:38,370
light has momentum to trap
certain types of particles

221
00:09:38,370 --> 00:09:41,820
within a beam of light and
directly apply forces to them

222
00:09:41,820 --> 00:09:43,920
and directly make
unique measurements

223
00:09:43,920 --> 00:09:48,930
about distance using this tool.

224
00:09:48,930 --> 00:09:53,220
So with optical tweezers, you
can measure nanometer motions

225
00:09:53,220 --> 00:09:54,992
at sub-millisecond
time resolution.

226
00:09:54,992 --> 00:09:56,450
And in a couple of
slides I'm going

227
00:09:56,450 --> 00:09:58,030
to tell you how that works.

228
00:09:58,030 --> 00:10:00,620
And you can also
directly apply force

229
00:10:00,620 --> 00:10:03,690
to probe mechanics
and biochemistry.

230
00:10:03,690 --> 00:10:06,440
So you guys have probably all
heard about optical tweezers,

231
00:10:06,440 --> 00:10:08,940
even if you haven't heard about
them in the case of studying

232
00:10:08,940 --> 00:10:10,470
single molecules of protein.

233
00:10:10,470 --> 00:10:12,990
So this is a really
cool video that I found.

234
00:10:12,990 --> 00:10:16,890
These are 12 beads, all trapped
in their own optical trap.

235
00:10:16,890 --> 00:10:19,230
And some graduate
student wrote a program

236
00:10:19,230 --> 00:10:21,450
to steer one of the beads
around the other beads.

237
00:10:21,450 --> 00:10:24,150
So the system that we used
in this experiment, that I'm

238
00:10:24,150 --> 00:10:26,100
going to tell you about
today, is basically

239
00:10:26,100 --> 00:10:28,140
the same as this system,
except you're not

240
00:10:28,140 --> 00:10:29,740
looking at 12 beads at a time.

241
00:10:29,740 --> 00:10:32,250
And each of the beads is
conjugated to some sort

242
00:10:32,250 --> 00:10:34,690
of molecule of protein.

243
00:10:34,690 --> 00:10:39,900
So the general setup in
single-molecule optical

244
00:10:39,900 --> 00:10:44,550
tweezers experiment in
biophysics is to take a bead--

245
00:10:44,550 --> 00:10:47,070
usually the bead is
made of polystyrene,

246
00:10:47,070 --> 00:10:49,540
because it happens to have
some nice properties--

247
00:10:49,540 --> 00:10:53,160
and to conjugate onto it a
single molecule of protein.

248
00:10:53,160 --> 00:10:55,110
In this case, they
conjugated onto it

249
00:10:55,110 --> 00:10:58,003
a single molecule of kinesin,
which is a motor protein.

250
00:10:58,003 --> 00:11:00,128
JOANNE STUBBE: [INAUDIBLE]
like polystyrene shrinks

251
00:11:00,128 --> 00:11:03,020
and it traps that use
polystyrene columns.

252
00:11:03,020 --> 00:11:04,960
Is that--

253
00:11:04,960 --> 00:11:09,920
REUBEN: These beads are
functionalized on the surface,

254
00:11:09,920 --> 00:11:12,280
and then covered with
streptavidin, which I'll

255
00:11:12,280 --> 00:11:13,854
talk about in just a second.

256
00:11:13,854 --> 00:11:15,520
The beads don't really
shrink very much,

257
00:11:15,520 --> 00:11:17,728
because you're not really
putting them under pressure

258
00:11:17,728 --> 00:11:21,110
in any of these experiments.

259
00:11:21,110 --> 00:11:24,490
And so what they did
in these experiments

260
00:11:24,490 --> 00:11:26,920
is that they took
kinesin, and they put it

261
00:11:26,920 --> 00:11:28,390
on the surface of a bead.

262
00:11:28,390 --> 00:11:30,280
And then they took
some microtubules,

263
00:11:30,280 --> 00:11:34,360
and they put their microtubules
on the surface of a cover slip.

264
00:11:34,360 --> 00:11:36,850
And they brought the
kinesin on the bead close

265
00:11:36,850 --> 00:11:40,300
to the microtubules on the
cover slip, and they added ATP.

266
00:11:40,300 --> 00:11:45,610
And kinesin is a molecular motor
which walks along microtubules.

267
00:11:45,610 --> 00:11:49,390
So what happened is that kinesin
grabbed onto the microtubules,

268
00:11:49,390 --> 00:11:52,140
and it began to bind
and hydrolyze ATP.

269
00:11:52,140 --> 00:11:54,040
And it began walking.

270
00:11:54,040 --> 00:11:56,620
And it dragged the
bead with it as it

271
00:11:56,620 --> 00:11:58,390
walked along the microtubule.

272
00:11:58,390 --> 00:12:00,940
And with an optical
trap, you can actually

273
00:12:00,940 --> 00:12:03,390
detect how far from
the center of the trap

274
00:12:03,390 --> 00:12:06,700
a bead has been dragged with
extremely high, less than one

275
00:12:06,700 --> 00:12:08,410
nanometer, precision.

276
00:12:08,410 --> 00:12:11,830
So they were able to tell that
this kinesin was taking very

277
00:12:11,830 --> 00:12:15,160
discrete eight-nanometer
steps as it walked

278
00:12:15,160 --> 00:12:17,620
along the microtubules,
which could be related very

279
00:12:17,620 --> 00:12:20,950
effectively to the size
of these tubulin domains

280
00:12:20,950 --> 00:12:24,620
that the kinesin was
actually making contact with.

281
00:12:24,620 --> 00:12:27,550
So using this they could, for
example, study the average time

282
00:12:27,550 --> 00:12:29,890
in between kinesin
steps as a function

283
00:12:29,890 --> 00:12:31,780
of the concentration of ATP.

284
00:12:31,780 --> 00:12:34,420
And they could begin to ask
some questions about how

285
00:12:34,420 --> 00:12:37,270
the binding and
hydrolysis of ATP

286
00:12:37,270 --> 00:12:40,750
actually led to kinesin's
processive mechanical motion.

287
00:12:40,750 --> 00:12:42,790
So before I tell you more
about the biophysics,

288
00:12:42,790 --> 00:12:44,290
I just want to tell
you a little bit

289
00:12:44,290 --> 00:12:45,790
about how optical traps work.

290
00:12:45,790 --> 00:12:48,700
Because I think that it's not
particularly intuitive before

291
00:12:48,700 --> 00:12:49,630
you hear about it.

292
00:12:49,630 --> 00:12:51,440
Although, in the
scheme of things,

293
00:12:51,440 --> 00:12:54,490
they're actually not
that complicated.

294
00:12:54,490 --> 00:12:57,350
So I'm going to tell you
three ways of thinking

295
00:12:57,350 --> 00:13:01,640
about optical tweezers, all of
which are basically the same,

296
00:13:01,640 --> 00:13:03,860
but use different
approach to look

297
00:13:03,860 --> 00:13:06,270
at the behavior of the system.

298
00:13:06,270 --> 00:13:10,250
So I'm just going to start by
pretending that light is just

299
00:13:10,250 --> 00:13:12,050
a ray, not a wave,
because that's

300
00:13:12,050 --> 00:13:13,880
a pretty good
approximation in a lot

301
00:13:13,880 --> 00:13:16,270
of circumstances,
including this one.

302
00:13:16,270 --> 00:13:18,590
So the way to think
about optical tweezers

303
00:13:18,590 --> 00:13:22,010
is to imagine two rays
of light, one coming

304
00:13:22,010 --> 00:13:25,230
from each side of a lens.

305
00:13:25,230 --> 00:13:27,350
And both of these rays,
they have momentum,

306
00:13:27,350 --> 00:13:28,880
because they're light.

307
00:13:28,880 --> 00:13:31,220
And to think about these
rays of light interacting

308
00:13:31,220 --> 00:13:35,330
with a dielectric particle such
as a polystyrene bead, which

309
00:13:35,330 --> 00:13:38,030
has a higher index of
refraction than the surrounding

310
00:13:38,030 --> 00:13:40,670
media, which is usually water.

311
00:13:40,670 --> 00:13:45,500
So what that means is that
when the light hits the bead,

312
00:13:45,500 --> 00:13:47,150
the angle of the
light will change.

313
00:13:47,150 --> 00:13:49,760
Basically the bead
will deflect the light.

314
00:13:49,760 --> 00:13:54,470
And so what that means is that
the bead is actually changing

315
00:13:54,470 --> 00:13:56,330
the momentum of the light.

316
00:13:56,330 --> 00:13:58,310
So what we know
from Newton is that,

317
00:13:58,310 --> 00:14:00,850
if the bead is changing
the momentum of the light,

318
00:14:00,850 --> 00:14:06,440
then the light must be exerting
an equal and opposite force

319
00:14:06,440 --> 00:14:11,570
on the bead itself, opposing
the momentum change of the light

320
00:14:11,570 --> 00:14:12,560
itself.

321
00:14:12,560 --> 00:14:16,640
So we can see here that this
ray A coming from the left

322
00:14:16,640 --> 00:14:21,040
is being deflected
up and to the right.

323
00:14:21,040 --> 00:14:24,080
And this ray B
coming from the right

324
00:14:24,080 --> 00:14:27,870
is being deflected
down and to the right.

325
00:14:27,870 --> 00:14:30,620
So we can think about
the force on the bead

326
00:14:30,620 --> 00:14:32,870
from each of these rays
being equal and opposite

327
00:14:32,870 --> 00:14:35,810
to the direction of the
momentum change of these rays.

328
00:14:35,810 --> 00:14:39,170
And say that ray A
is imparting a force

329
00:14:39,170 --> 00:14:42,310
on the bead which is pointing
down and to the left,

330
00:14:42,310 --> 00:14:46,910
and ray B is imparting a
force on the bead which

331
00:14:46,910 --> 00:14:49,430
is pointing up and to the left.

332
00:14:49,430 --> 00:14:52,550
So we can basically sum
both of these forces

333
00:14:52,550 --> 00:14:55,310
and find that, in
this system here,

334
00:14:55,310 --> 00:14:57,350
where the center of
the bead is moved

335
00:14:57,350 --> 00:15:01,130
to the right of the true focus
of these two rays of light,

336
00:15:01,130 --> 00:15:05,960
then the light is imposing a
force on the bead which pushes

337
00:15:05,960 --> 00:15:07,620
it back toward the center.

338
00:15:07,620 --> 00:15:10,700
And it turns out that,
whenever the center of the bead

339
00:15:10,700 --> 00:15:15,780
is not in the same spot as the
center of the focused light,

340
00:15:15,780 --> 00:15:18,350
the light will impart
a force on the bead,

341
00:15:18,350 --> 00:15:20,900
pushing it toward the
direction of that center,

342
00:15:20,900 --> 00:15:24,412
no matter how you move it
around within the trap.

343
00:15:24,412 --> 00:15:27,550
So do you all understand
the ray approach

344
00:15:27,550 --> 00:15:30,006
to understanding how
optical traps work?

345
00:15:30,006 --> 00:15:31,920
Cool.

346
00:15:31,920 --> 00:15:35,290
So I'm going to give a slightly
more realistic picture.

347
00:15:35,290 --> 00:15:38,970
So as you guys know,
light is not a ray.

348
00:15:38,970 --> 00:15:40,240
Light is a wave.

349
00:15:40,240 --> 00:15:43,860
And what that means is
that you can't focus light

350
00:15:43,860 --> 00:15:46,380
into just a point.

351
00:15:46,380 --> 00:15:48,960
Instead, you can focus light
into a small volume which

352
00:15:48,960 --> 00:15:52,440
is diffraction-limited in size
to approximately a diameter

353
00:15:52,440 --> 00:15:55,140
of one-half the wavelength
of the light, which is often

354
00:15:55,140 --> 00:15:57,430
called the confocal volume.

355
00:15:57,430 --> 00:16:01,440
So to think about the shape
of the confocal volume,

356
00:16:01,440 --> 00:16:03,360
it's not quite a
Gaussian, but it's

357
00:16:03,360 --> 00:16:04,574
very similar to a Gaussian.

358
00:16:04,574 --> 00:16:06,240
So it's useful to
think about Gaussians,

359
00:16:06,240 --> 00:16:08,130
to say that, you know,
it's easy to imagine

360
00:16:08,130 --> 00:16:11,130
a one-dimensional Gaussian or
a two-dimensional Gaussian.

361
00:16:11,130 --> 00:16:15,000
At a confocal volume of
light, what you actually

362
00:16:15,000 --> 00:16:17,760
have is a three-dimensional
Gaussian in which

363
00:16:17,760 --> 00:16:20,980
each point in space
has a scalar property,

364
00:16:20,980 --> 00:16:24,210
which is the intensity of the
light at that point in space.

365
00:16:24,210 --> 00:16:27,770
And if you move from the
center out in any direction,

366
00:16:27,770 --> 00:16:31,360
it basically decays according
to this Gaussian curvature.

367
00:16:31,360 --> 00:16:33,990
So there's a gradient, a
Gaussian-shaped gradient

368
00:16:33,990 --> 00:16:37,140
of light intensity going
out in every direction

369
00:16:37,140 --> 00:16:39,539
from the center, if this
is like the optical axis,

370
00:16:39,539 --> 00:16:41,080
and say this is like
the x direction,

371
00:16:41,080 --> 00:16:42,659
and this is the y direction.

372
00:16:42,659 --> 00:16:44,700
I don't know if my laser
printing is particularly

373
00:16:44,700 --> 00:16:47,760
helpful in this situation.

374
00:16:47,760 --> 00:16:49,800
So what you can
think about is, say

375
00:16:49,800 --> 00:16:56,808
that from your beam of
light in this Gaussian shape

376
00:16:56,808 --> 00:17:00,570
you have rays of
different intensity.

377
00:17:00,570 --> 00:17:02,460
I've only shown
two here, but try

378
00:17:02,460 --> 00:17:04,800
to imagine these rays
coming from every part

379
00:17:04,800 --> 00:17:08,790
of this light beam, all being
focused here so that rays

380
00:17:08,790 --> 00:17:10,470
from the center have
a lot of intensity,

381
00:17:10,470 --> 00:17:11,800
therefore they have
a lot of momentum.

382
00:17:11,800 --> 00:17:13,091
There's a lot of photons there.

383
00:17:13,091 --> 00:17:16,170
So they impart more
force per deflection

384
00:17:16,170 --> 00:17:17,950
than rays coming
over from the side.

385
00:17:17,950 --> 00:17:22,800
So if you imagine rays coming
from this whole Gaussian volume

386
00:17:22,800 --> 00:17:25,230
here, then it's not
too difficult to see

387
00:17:25,230 --> 00:17:27,972
that whenever you move
the bead from the center,

388
00:17:27,972 --> 00:17:30,180
there's going to be a
restoring force pushing it back

389
00:17:30,180 --> 00:17:32,480
to the center.

390
00:17:32,480 --> 00:17:35,070
And the last explanation
for how optical traps work

391
00:17:35,070 --> 00:17:36,900
is my favorite, but
it's a little bit more

392
00:17:36,900 --> 00:17:38,983
difficult to understand,
even though it's actually

393
00:17:38,983 --> 00:17:40,840
classical rather than quantum.

394
00:17:40,840 --> 00:17:44,235
So what we know--

395
00:17:44,235 --> 00:17:46,490
a dielectric particle
such as a bead,

396
00:17:46,490 --> 00:17:48,900
what a dielectric
actually is, a dielectric

397
00:17:48,900 --> 00:17:52,870
is a material which is
polarized by an electric field.

398
00:17:52,870 --> 00:17:56,970
So polystyrene is
polarized quite a lot

399
00:17:56,970 --> 00:17:58,350
by an electric field.

400
00:17:58,350 --> 00:18:01,410
And what that does is that,
if you have an electric field,

401
00:18:01,410 --> 00:18:04,020
say, that has a
negative charge here

402
00:18:04,020 --> 00:18:06,000
and a positive charge
here, then the dielectric

403
00:18:06,000 --> 00:18:08,040
will be polarized
in such a direction

404
00:18:08,040 --> 00:18:11,080
that it is opposing
that electric field.

405
00:18:11,080 --> 00:18:13,140
And the electric field
within the dielectric

406
00:18:13,140 --> 00:18:16,440
is lower than the electric
field outside of the dielectric.

407
00:18:16,440 --> 00:18:19,770
And it turns out that, for very
small particles, in this case

408
00:18:19,770 --> 00:18:21,630
particles that are
actually smaller

409
00:18:21,630 --> 00:18:23,940
than the wavelength of
light that we are actually

410
00:18:23,940 --> 00:18:26,580
using to trap it,
you can approximate

411
00:18:26,580 --> 00:18:31,020
a dielectric material very well
as just a simple dipole whose

412
00:18:31,020 --> 00:18:33,900
dipole moment is facing
in the opposite direction

413
00:18:33,900 --> 00:18:37,260
of the electric field that
it's actually sitting in.

414
00:18:37,260 --> 00:18:40,380
So think about what
light actually is.

415
00:18:40,380 --> 00:18:42,840
Light is a rapidly
oscillating electric field

416
00:18:42,840 --> 00:18:45,860
that also happens to have
a magnetic component.

417
00:18:45,860 --> 00:18:51,870
So the particle that
you're studying basically

418
00:18:51,870 --> 00:18:54,007
has a rapidly
oscillating dipole,

419
00:18:54,007 --> 00:18:55,590
which is always
opposing the direction

420
00:18:55,590 --> 00:18:57,030
of the magnetic field.

421
00:18:57,030 --> 00:19:03,210
And so we know that there is
a cost in potential energy

422
00:19:03,210 --> 00:19:05,100
to separating two charges.

423
00:19:05,100 --> 00:19:07,530
And putting a
dielectric material

424
00:19:07,530 --> 00:19:09,510
in between that
charge separation

425
00:19:09,510 --> 00:19:15,190
decreases the potential cost
of this charge separation.

426
00:19:15,190 --> 00:19:19,020
So if you have a gradient
of electric fields,

427
00:19:19,020 --> 00:19:21,880
as you do at the center
of a confocal volume,

428
00:19:21,880 --> 00:19:24,480
the dielectric will
be most favorable

429
00:19:24,480 --> 00:19:26,850
if it is in the place where
it is opposing the strongest

430
00:19:26,850 --> 00:19:27,810
electric field.

431
00:19:27,810 --> 00:19:29,550
And if you move it
from the center,

432
00:19:29,550 --> 00:19:32,980
then it will oppose a
weaker electric field.

433
00:19:32,980 --> 00:19:36,790
And we know that force is
the gradient of potential.

434
00:19:36,790 --> 00:19:42,000
So the dielectric particle
will feel a force driving it

435
00:19:42,000 --> 00:19:45,510
toward the strongest area
of electric field, which

436
00:19:45,510 --> 00:19:49,110
takes place at the very
center of the confocal volume.

437
00:19:49,110 --> 00:19:52,980
And so it will always be
restored toward that spot.

438
00:19:52,980 --> 00:19:56,610
So this is kind of an
intuitive explanation,

439
00:19:56,610 --> 00:19:58,481
but it takes a little
bit of thinking about.

440
00:19:58,481 --> 00:20:00,230
If you have any more
questions about this,

441
00:20:00,230 --> 00:20:05,580
I'm totally happy to answer them
either now or after the talk.

442
00:20:05,580 --> 00:20:07,560
So do you all sort of
feel like you understand

443
00:20:07,560 --> 00:20:11,020
how you can use light to
trap a dielectric particle

444
00:20:11,020 --> 00:20:12,883
such as a polystyrene bead?

445
00:20:12,883 --> 00:20:15,258
AUDIENCE: What wavelength of
light are you talking about?

446
00:20:15,258 --> 00:20:17,079
Like, is this monochromatic?

447
00:20:17,079 --> 00:20:17,620
REUBEN: Yeah.

448
00:20:17,620 --> 00:20:20,200
So in this case we're using
monochromatic infrared light.

449
00:20:29,070 --> 00:20:31,880
The wavelength which is
chosen for these studies,

450
00:20:31,880 --> 00:20:34,280
it turns out to not
damage proteins very much

451
00:20:34,280 --> 00:20:35,909
compared to other
wavelengths of light,

452
00:20:35,909 --> 00:20:37,700
and you're using it at
very high intensity.

453
00:20:37,700 --> 00:20:40,074
And it's pretty easy to make
very high intensity infrared

454
00:20:40,074 --> 00:20:42,240
lasers.

455
00:20:42,240 --> 00:20:45,350
But you could use, you know,
the thing that really matters

456
00:20:45,350 --> 00:20:47,570
is what the index of
refraction is at the wavelength

457
00:20:47,570 --> 00:20:48,940
that you're studying.

458
00:20:48,940 --> 00:20:51,312
So as long as you
have a big difference,

459
00:20:51,312 --> 00:20:53,270
as long as you have a
dielectric particle which

460
00:20:53,270 --> 00:20:56,060
has a higher index of
refraction than the surrounding

461
00:20:56,060 --> 00:20:59,432
media at the wavelength
that you're studying, then

462
00:20:59,432 --> 00:21:00,890
you can use any
wavelength of light

463
00:21:00,890 --> 00:21:03,510
for this sort of experiment.

464
00:21:03,510 --> 00:21:05,420
So I'm not going to
go too much in depth

465
00:21:05,420 --> 00:21:08,897
into the actual apparatus
for an optical trap,

466
00:21:08,897 --> 00:21:10,730
but I do want to just
give you the schematic

467
00:21:10,730 --> 00:21:13,790
and talk about the very basics.

468
00:21:13,790 --> 00:21:17,060
So the general idea is that you
have a trapping laser, which

469
00:21:17,060 --> 00:21:19,310
is being focused right
here on the specimen stage,

470
00:21:19,310 --> 00:21:21,590
basically just a microscope
where you actually

471
00:21:21,590 --> 00:21:23,520
have the bead trapped.

472
00:21:23,520 --> 00:21:26,280
And then you have a
second detection laser,

473
00:21:26,280 --> 00:21:29,295
which is much weaker, 100-fold
weaker than the trapping laser,

474
00:21:29,295 --> 00:21:31,290
and is at a
different wavelength.

475
00:21:31,290 --> 00:21:35,090
And this is being focused
onto the same volume.

476
00:21:35,090 --> 00:21:38,390
And you have a
quadrant photodiode,

477
00:21:38,390 --> 00:21:42,530
which basically gives off a
different voltage in response

478
00:21:42,530 --> 00:21:47,391
to a different location of
the trapping laser on it.

479
00:21:47,391 --> 00:21:48,890
And this is how you
actually measure

480
00:21:48,890 --> 00:21:52,730
the where the bead actually is
within the center of the trap.

481
00:21:52,730 --> 00:21:57,610
So if the bead is moved out
of the center of the trap,

482
00:21:57,610 --> 00:22:01,040
then this detection laser
is deflected slightly,

483
00:22:01,040 --> 00:22:04,050
because it's also
refracts through the bead.

484
00:22:04,050 --> 00:22:09,110
And so that shifts the
centroid of the detection laser

485
00:22:09,110 --> 00:22:12,170
as it's read out on this
position-sensitive detector.

486
00:22:12,170 --> 00:22:16,460
And so you can take this
reading 100,000 times a second.

487
00:22:16,460 --> 00:22:21,260
And that's how you can get this
very detailed, one-nanometer

488
00:22:21,260 --> 00:22:25,640
level of resolution, which
describes where the bead has

489
00:22:25,640 --> 00:22:28,910
been pulled by some
biological molecule out

490
00:22:28,910 --> 00:22:30,050
of the center of the trap.

491
00:22:32,750 --> 00:22:34,375
So one other thing
I should talk about,

492
00:22:34,375 --> 00:22:36,250
I want to give a couple
of examples of things

493
00:22:36,250 --> 00:22:38,020
that people have used
optical traps for.

494
00:22:38,020 --> 00:22:40,480
But it's worth mentioning
that most experiments

495
00:22:40,480 --> 00:22:44,180
with optical traps are no
longer done with just one trap.

496
00:22:44,180 --> 00:22:46,570
They're now done with
two traps in which you

497
00:22:46,570 --> 00:22:48,940
have two different
beads, each trapped

498
00:22:48,940 --> 00:22:50,710
within their own optical trap.

499
00:22:50,710 --> 00:22:53,200
And you do this because it
basically mechanically isolates

500
00:22:53,200 --> 00:22:54,010
the system.

501
00:22:54,010 --> 00:22:55,870
So normally you
can have vibrations

502
00:22:55,870 --> 00:22:59,465
just in the room that would
shake your specimen stage.

503
00:22:59,465 --> 00:23:01,840
And it turns out that the
distances that you're measuring

504
00:23:01,840 --> 00:23:04,390
are so small that these
vibrations dramatically

505
00:23:04,390 --> 00:23:06,590
increase the noise
of the system.

506
00:23:06,590 --> 00:23:08,110
But if you have
both beads trapped

507
00:23:08,110 --> 00:23:09,568
within their own
optical trap, that

508
00:23:09,568 --> 00:23:11,350
provides significant
damping from any sort

509
00:23:11,350 --> 00:23:13,520
of mechanical vibration.

510
00:23:13,520 --> 00:23:16,990
And you can measure the
bead-to-bead distance

511
00:23:16,990 --> 00:23:19,110
using this apparatus.

512
00:23:19,110 --> 00:23:21,970
And so here, this is from
Stephen Block's group.

513
00:23:21,970 --> 00:23:26,050
They actually were able to
measure single base pair

514
00:23:26,050 --> 00:23:29,700
incorporation into
a growing mRNA chain

515
00:23:29,700 --> 00:23:33,340
as RNA polymerase
walked along a template.

516
00:23:33,340 --> 00:23:34,990
So it's not perfect.

517
00:23:34,990 --> 00:23:36,430
It doesn't look totally linear.

518
00:23:36,430 --> 00:23:41,239
But it's pretty incredible to
see two-angstroms resolution.

519
00:23:41,239 --> 00:23:43,030
And from this they
could learn a great deal

520
00:23:43,030 --> 00:23:46,090
about the kinetics of
RNA polymerase stepping

521
00:23:46,090 --> 00:23:46,974
along a template.

522
00:23:46,974 --> 00:23:48,640
And they could even
keep it in registers

523
00:23:48,640 --> 00:23:51,280
so that they could know which
base is being incorporated,

524
00:23:51,280 --> 00:23:53,650
which change in distance
here represents which

525
00:23:53,650 --> 00:23:55,300
base incorporation,
so they could

526
00:23:55,300 --> 00:23:57,920
study the kinetics
of how sequence

527
00:23:57,920 --> 00:24:02,720
effects the rate of insertion.

528
00:24:02,720 --> 00:24:04,600
Another use of
optical traps which

529
00:24:04,600 --> 00:24:07,970
has proved very fruitful
for some people is to,

530
00:24:07,970 --> 00:24:10,920
you know, you can use
them to measure forces

531
00:24:10,920 --> 00:24:14,260
that molecules apply on
a bead, but you can also

532
00:24:14,260 --> 00:24:16,120
directly apply
forces on molecules

533
00:24:16,120 --> 00:24:17,850
by turning the system around.

534
00:24:17,850 --> 00:24:21,280
If you move the center
of the trap a little bit,

535
00:24:21,280 --> 00:24:23,860
then you can apply a
force on a molecule which

536
00:24:23,860 --> 00:24:26,110
is, for example,
bound in the trap,

537
00:24:26,110 --> 00:24:28,810
and see how a change in
the force on the molecule

538
00:24:28,810 --> 00:24:30,400
will change the
bead-to-bead distance,

539
00:24:30,400 --> 00:24:34,450
basically get a force
extension curve.

540
00:24:34,450 --> 00:24:41,800
So here they trapped a protein
domain in between two beads

541
00:24:41,800 --> 00:24:43,840
with a long DNA
handle, which basically

542
00:24:43,840 --> 00:24:46,930
is designed to be a relatively
rigid linker that just keeps

543
00:24:46,930 --> 00:24:49,190
the two beads
relatively separated

544
00:24:49,190 --> 00:24:52,540
so that the optical traps
don't overlay at all.

545
00:24:52,540 --> 00:24:53,980
And they applied force.

546
00:24:53,980 --> 00:24:58,660
And what you can see is that
as you increase the force by--

547
00:24:58,660 --> 00:25:01,972
in this case, they moved one
of these beads just a little

548
00:25:01,972 --> 00:25:02,680
bit to the right.

549
00:25:05,860 --> 00:25:08,120
You get a little
bit of stretching,

550
00:25:08,120 --> 00:25:11,710
which is partially in the DNA
and partially in the protein

551
00:25:11,710 --> 00:25:13,190
domain itself.

552
00:25:13,190 --> 00:25:18,040
But then once you reach a
certain threshold force, which

553
00:25:18,040 --> 00:25:21,550
in this case is 17
piconewtons, they actually

554
00:25:21,550 --> 00:25:24,130
unfold this protein domain.

555
00:25:24,130 --> 00:25:27,340
And you get a sudden jump
in bead-to-bead distance

556
00:25:27,340 --> 00:25:31,060
as this now unfolded
polypeptide no longer provides

557
00:25:31,060 --> 00:25:34,110
any sort of force
pulling it back together.

558
00:25:34,110 --> 00:25:37,000
And then if they
release the force

559
00:25:37,000 --> 00:25:40,900
and allow the molecule
to relax, they actually

560
00:25:40,900 --> 00:25:43,840
get a different relaxation curve
than their original extension

561
00:25:43,840 --> 00:25:46,210
curve, because this
represents the relaxation

562
00:25:46,210 --> 00:25:48,940
of this unfolded
polypeptide rather than

563
00:25:48,940 --> 00:25:51,376
the refolding of this domain.

564
00:25:51,376 --> 00:25:54,270
And another cool thing that
they did in this experiment

565
00:25:54,270 --> 00:25:59,990
is they brought it right
up to this unfolding force,

566
00:25:59,990 --> 00:26:02,200
and then they left it there.

567
00:26:02,200 --> 00:26:05,040
And it turns out that
at this unfolding force,

568
00:26:05,040 --> 00:26:08,790
this protein rapidly
unfolds and refolds,

569
00:26:08,790 --> 00:26:11,160
because it's a very small
protein domain, so the protein

570
00:26:11,160 --> 00:26:12,690
folding is extremely quick.

571
00:26:12,690 --> 00:26:18,100
So this is the rapid transition
between folding and unfolding

572
00:26:18,100 --> 00:26:21,650
which is taking place at
this transition force.

573
00:26:21,650 --> 00:26:24,630
So they could use that to study
the kinetics of the protein

574
00:26:24,630 --> 00:26:29,140
folding under force
for this system.

575
00:26:29,140 --> 00:26:32,580
So today, the paper that you
guys read for this recitation

576
00:26:32,580 --> 00:26:36,090
is about ClpXP,
which is this protein

577
00:26:36,090 --> 00:26:38,510
that, as you guys
studied in class,

578
00:26:38,510 --> 00:26:40,530
it recognizes protein
substrates that have

579
00:26:40,530 --> 00:26:43,530
been tagged for degradation.

580
00:26:43,530 --> 00:26:49,590
And it uses mechanical
energy released

581
00:26:49,590 --> 00:26:55,320
by ATP binding and hydrolysis
to unfold folded protein domains

582
00:26:55,320 --> 00:26:58,560
and translocate the
unfolded protein

583
00:26:58,560 --> 00:27:01,500
substrate through an axial
pore into an associated

584
00:27:01,500 --> 00:27:04,890
sequestered peptidase, where
the protein is rapidly chewed up

585
00:27:04,890 --> 00:27:06,360
into small peptides
that are then

586
00:27:06,360 --> 00:27:10,370
recycled into their
constituent amino acids.

587
00:27:10,370 --> 00:27:13,630
So I have another figure,
which someone else in my lab

588
00:27:13,630 --> 00:27:17,890
made more recently, that
illustrates a little bit

589
00:27:17,890 --> 00:27:22,600
better the actual unfolding
and translocation steps, which

590
00:27:22,600 --> 00:27:25,690
were the steps that we
studied in this paper.

591
00:27:25,690 --> 00:27:36,130
So the general idea is
that ClpX will translocate

592
00:27:36,130 --> 00:27:39,040
a folded protein domain until
it's reached a point where

593
00:27:39,040 --> 00:27:43,120
the folded domain is too big
to fit through the axial pore.

594
00:27:43,120 --> 00:27:46,900
And then it will apply what
we call power strokes, which

595
00:27:46,900 --> 00:27:51,840
are somehow related to ATP
hydrolysis and binding, which

596
00:27:51,840 --> 00:27:57,850
yank on the ssrA tag and pull
the protein down in a repeated

597
00:27:57,850 --> 00:27:59,590
attempt to unfold it.

598
00:27:59,590 --> 00:28:02,170
And you can say that there's
some sort of strained protein

599
00:28:02,170 --> 00:28:02,934
structure.

600
00:28:02,934 --> 00:28:04,600
It turns out that
most of these attempts

601
00:28:04,600 --> 00:28:07,720
do not actually successfully
unfold the protein.

602
00:28:07,720 --> 00:28:10,330
But some proportion
of these attempts--

603
00:28:10,330 --> 00:28:13,387
maybe ClpX gets a
particularly good grip.

604
00:28:13,387 --> 00:28:15,220
Maybe there's some sort
of transient thermal

605
00:28:15,220 --> 00:28:17,890
destabilization in
the protein itself.

606
00:28:17,890 --> 00:28:20,420
The protein
successfully unfolded,

607
00:28:20,420 --> 00:28:22,530
and then it continues to
take these small steps,

608
00:28:22,530 --> 00:28:26,630
these small power strokes
along the unfolded polypeptide,

609
00:28:26,630 --> 00:28:31,648
translocating it into the
peptidase for degradation.

610
00:28:31,648 --> 00:28:36,040
So in the experiments
that Adrian actually

611
00:28:36,040 --> 00:28:43,720
did for this system, what
he did is he attached ClpXP

612
00:28:43,720 --> 00:28:47,470
to a bead using streptavidin
and biotin, which

613
00:28:47,470 --> 00:28:50,980
is, I believe, an interaction
you guys are familiar with.

614
00:28:50,980 --> 00:28:55,780
And he attached a
multi-domain protein substrate

615
00:28:55,780 --> 00:28:59,980
to a DNA linker so
that the ssrA tag was

616
00:28:59,980 --> 00:29:02,470
at the end of this substrate.

617
00:29:02,470 --> 00:29:05,920
And he was able to move
the two beads together

618
00:29:05,920 --> 00:29:10,410
so that, at some point, the
ClpXP bound to the ssrA tag.

619
00:29:10,410 --> 00:29:12,910
And then he could record
the bead-to-bead distance

620
00:29:12,910 --> 00:29:15,890
during the unfolding
and translocation

621
00:29:15,890 --> 00:29:17,620
of this substrate.

622
00:29:17,620 --> 00:29:21,640
So what he saw is that,
when ClpXP was successful

623
00:29:21,640 --> 00:29:24,880
in unfolding a protein domain,
there was a sudden jump

624
00:29:24,880 --> 00:29:29,380
in bead-to-bead
distance as the folded--

625
00:29:29,380 --> 00:29:30,880
the whole experiment
is taking place

626
00:29:30,880 --> 00:29:32,760
under a small amount of force.

627
00:29:32,760 --> 00:29:36,200
So as the polypeptide
has now unfolded

628
00:29:36,200 --> 00:29:37,930
for that small
amount of force, it's

629
00:29:37,930 --> 00:29:39,930
quickly pulled pretty taut.

630
00:29:39,930 --> 00:29:43,840
And then ClpXP translocates
that substrate, now

631
00:29:43,840 --> 00:29:45,910
unfolded, through
its axial pore,

632
00:29:45,910 --> 00:29:47,980
decreasing the
bead-to-bead distance.

633
00:29:47,980 --> 00:29:49,480
And it continues
this translocation

634
00:29:49,480 --> 00:29:52,195
until it reaches the next
folded protein substrate,

635
00:29:52,195 --> 00:29:53,520
at which point it stops.

636
00:29:53,520 --> 00:29:55,620
It can't translocate
any further.

637
00:29:55,620 --> 00:29:59,710
And it begins to attempt to
unfold the substrate again.

638
00:29:59,710 --> 00:30:02,650
And eventually, after
some sort of dwell

639
00:30:02,650 --> 00:30:04,870
that we call a
pre-unfolding dwell,

640
00:30:04,870 --> 00:30:07,420
it will be successful at
unfolding the substrate.

641
00:30:07,420 --> 00:30:09,940
There will be another jump
in bead-to-bead distance,

642
00:30:09,940 --> 00:30:11,610
and the process will repeat.

643
00:30:11,610 --> 00:30:12,110
Go ahead.

644
00:30:12,110 --> 00:30:14,018
AUDIENCE: So you said
that it was held taut.

645
00:30:14,018 --> 00:30:16,403
Is there an additional
force that it

646
00:30:16,403 --> 00:30:18,788
experiences because
it's being held taut,

647
00:30:18,788 --> 00:30:20,710
or is that accounted for?

648
00:30:20,710 --> 00:30:22,792
REUBEN: It's being
held a little bit taut.

649
00:30:22,792 --> 00:30:23,750
I shouldn't say "taut."

650
00:30:23,750 --> 00:30:26,770
There is a small amount
of force on the system.

651
00:30:26,770 --> 00:30:28,500
You know, often we
record in the range

652
00:30:28,500 --> 00:30:33,300
of about five piconewtons,
which is just required

653
00:30:33,300 --> 00:30:36,030
for the optical trapping
apparatus to actually work,

654
00:30:36,030 --> 00:30:38,310
but should not have
a significant effect

655
00:30:38,310 --> 00:30:40,450
on the behavior of the system.

656
00:30:40,450 --> 00:30:43,950
So I should have mentioned
it back at the kinesin.

657
00:30:43,950 --> 00:30:46,770
You can understand
the stall force

658
00:30:46,770 --> 00:30:49,320
for a molecular motor,
which is the force that it

659
00:30:49,320 --> 00:30:52,650
takes to restrain it from
taking additional steps.

660
00:30:52,650 --> 00:30:55,090
And it turns out that the
stall force for kinesin

661
00:30:55,090 --> 00:30:56,800
is about seven piconewtons.

662
00:30:56,800 --> 00:31:00,300
The stall force for ClpXP
is about 25 piconewtons.

663
00:31:00,300 --> 00:31:03,450
So it's dramatically higher
than the small amount of force

664
00:31:03,450 --> 00:31:05,160
exerted on it by the trap.

665
00:31:05,160 --> 00:31:07,200
So it shouldn't have
a significant impact

666
00:31:07,200 --> 00:31:08,130
on the results.

667
00:31:08,130 --> 00:31:11,820
What it just means is that, once
you go from folded to unfolded,

668
00:31:11,820 --> 00:31:14,640
since there is nothing
holding the unfolded structure

669
00:31:14,640 --> 00:31:17,250
in a coil, even a
small amount of force

670
00:31:17,250 --> 00:31:20,199
will pull it out a little
bit and cause this increase

671
00:31:20,199 --> 00:31:21,240
in bead-to-bead distance.

672
00:31:21,240 --> 00:31:22,740
JOANNE STUBBE: So
I have a question.

673
00:31:22,740 --> 00:31:24,487
When you're developing
these methods,

674
00:31:24,487 --> 00:31:26,070
like when we used
to teach [INAUDIBLE]

675
00:31:26,070 --> 00:31:30,594
DNA polymerases, how you put
the complex onto anything.

676
00:31:30,594 --> 00:31:32,260
REUBEN: Oh, I have a
slide on that next.

677
00:31:32,260 --> 00:31:32,605
JOANNE STUBBE: OK.

678
00:31:32,605 --> 00:31:34,465
Because to me,
that's the key thing.

679
00:31:34,465 --> 00:31:37,840
And the question is, in
this particular experiment,

680
00:31:37,840 --> 00:31:40,475
how many iterations did
they have to go through

681
00:31:40,475 --> 00:31:45,295
before they figured out length,
attachment, all of that stuff?

682
00:31:45,295 --> 00:31:47,410
REUBEN: So fortunately,
other groups

683
00:31:47,410 --> 00:31:49,420
have worked out
many of these issues

684
00:31:49,420 --> 00:31:52,570
for other protein systems, so
we were able to adapt those--

685
00:31:52,570 --> 00:31:53,110
JOANNE STUBBE: And you
would be able extrapolate--

686
00:31:53,110 --> 00:31:53,530
REUBEN: --relatively easily.

687
00:31:53,530 --> 00:31:54,530
JOANNE STUBBE: from one?

688
00:31:54,530 --> 00:31:57,050
Because with the polymerases,
you couldn't do that.

689
00:31:57,050 --> 00:32:00,730
REUBEN: Other people had studied
these AAA's, particularly

690
00:32:00,730 --> 00:32:05,710
helicases, as well as various
nucleotide translocases that

691
00:32:05,710 --> 00:32:08,380
are actually, in
mechanical activity,

692
00:32:08,380 --> 00:32:10,060
somewhat similar to ClpXP.

693
00:32:10,060 --> 00:32:11,950
So we could basically
adapt their systems

694
00:32:11,950 --> 00:32:16,030
without significant
trial and error, which

695
00:32:16,030 --> 00:32:18,475
meant that we could get
closer to the biology,

696
00:32:18,475 --> 00:32:21,300
or closer to the biophysics
quickly, which was very nice.

697
00:32:21,300 --> 00:32:23,990
But other groups certainly spent
20 years getting these methods

698
00:32:23,990 --> 00:32:26,039
to actually work well.

699
00:32:26,039 --> 00:32:28,080
JOANNE STUBBE: Are there
any generalizations that

700
00:32:28,080 --> 00:32:31,820
came out of that optimization?

701
00:32:31,820 --> 00:32:33,134
REUBEN: Yes.

702
00:32:33,134 --> 00:32:35,050
So there are a couple
of generalizations which

703
00:32:35,050 --> 00:32:37,090
came out of that optimization.

704
00:32:37,090 --> 00:32:39,180
The first is that
biotin-streptavidin

705
00:32:39,180 --> 00:32:41,140
is a really useful interaction.

706
00:32:41,140 --> 00:32:43,720
Basically all modern
biophysical techniques

707
00:32:43,720 --> 00:32:48,130
use biotin-streptavidin
or other basically binary,

708
00:32:48,130 --> 00:32:53,920
basically permanent
interactions for these sorts

709
00:32:53,920 --> 00:32:55,000
of biophysical studies.

710
00:32:55,000 --> 00:32:56,958
JOANNE STUBBE: But
biotin-enhanced streptavidin

711
00:32:56,958 --> 00:32:58,283
has four binding sites.

712
00:32:58,283 --> 00:33:02,010
Are they using a streptavidin
with one binding site?

713
00:33:02,010 --> 00:33:02,510
REUBEN: No.

714
00:33:02,510 --> 00:33:06,590
You give the-- so when
you're actually taking a bead

715
00:33:06,590 --> 00:33:08,720
and adding ClpXP
to it, what you do

716
00:33:08,720 --> 00:33:11,204
is, you add a very
small amount of ClpXP.

717
00:33:11,204 --> 00:33:12,620
And then once
you've added it, you

718
00:33:12,620 --> 00:33:15,952
add just straight biotin to fill
the rest of the binding sites.

719
00:33:15,952 --> 00:33:17,660
Because one thing that
you never want is,

720
00:33:17,660 --> 00:33:20,330
you never want two
ClpXPs on one bead

721
00:33:20,330 --> 00:33:22,460
to engage two substrates
on another bead,

722
00:33:22,460 --> 00:33:23,660
because that would
just totally screw up

723
00:33:23,660 --> 00:33:24,951
the data that you're recording.

724
00:33:24,951 --> 00:33:26,540
So you make sure
that it's very, very

725
00:33:26,540 --> 00:33:29,610
sparse labeling of molecules
of ClpXP on the surface.

726
00:33:29,610 --> 00:33:32,210
You know, I think we
probably use femtomolar ClpXP

727
00:33:32,210 --> 00:33:36,140
during the actual labeling
of the bead process.

728
00:33:36,140 --> 00:33:38,390
Another major takeaway
from these experiments

729
00:33:38,390 --> 00:33:41,450
is that, for these
dual bead experiments,

730
00:33:41,450 --> 00:33:43,580
it's very important
to have a DNA linker.

731
00:33:43,580 --> 00:33:46,040
So the distance here is
not really representative,

732
00:33:46,040 --> 00:33:49,070
but this DNA linker
actually goes over here.

733
00:33:49,070 --> 00:33:50,570
Because you really
want to maintain

734
00:33:50,570 --> 00:33:54,110
bead-to-bead separation on
the order of more than a

735
00:33:54,110 --> 00:33:57,824
micron, so that your two
traps don't overlay at all.

736
00:33:57,824 --> 00:33:59,240
Because this leads
to much clearer

737
00:33:59,240 --> 00:34:01,760
readouts on your
position-sensitive display,

738
00:34:01,760 --> 00:34:04,216
allowing much better
data acquisition.

739
00:34:04,216 --> 00:34:06,148
JOANNE STUBBE: So
again, just technically,

740
00:34:06,148 --> 00:34:07,600
I'm interested in this--

741
00:34:07,600 --> 00:34:08,670
REUBEN: Oh, sure.

742
00:34:08,670 --> 00:34:10,920
JOANNE STUBBE: So when you
have very low concentration

743
00:34:10,920 --> 00:34:13,590
of anything, usually you
have a lot of problems,

744
00:34:13,590 --> 00:34:16,050
because the stuff
sticks to everything.

745
00:34:16,050 --> 00:34:19,345
So how do you avoid, you
know, just inherently,

746
00:34:19,345 --> 00:34:20,689
especially proteins.

747
00:34:20,689 --> 00:34:22,489
I mean, I've dealt
a lot with proteins.

748
00:34:22,489 --> 00:34:24,526
The more glue you
get, the worse-behaved

749
00:34:24,526 --> 00:34:26,533
they are, because they
stick to everything,

750
00:34:26,533 --> 00:34:30,157
even when you modify
the containers

751
00:34:30,157 --> 00:34:32,813
with different kinds of
polymers and stuff like that.

752
00:34:32,813 --> 00:34:33,780
So--

753
00:34:33,780 --> 00:34:35,730
REUBEN: Yeah, so that
is a major issue.

754
00:34:35,730 --> 00:34:38,190
In this case, it's not a
particularly big issue,

755
00:34:38,190 --> 00:34:41,219
because you're really-- all you
need is one ClpXP on the bead

756
00:34:41,219 --> 00:34:44,010
and one molecule of
substrate on the other bead.

757
00:34:44,010 --> 00:34:46,110
So it doesn't really
matter what happened

758
00:34:46,110 --> 00:34:48,420
to most of the molecules
that you add to your mixture

759
00:34:48,420 --> 00:34:52,090
as long as you have one
active molecule here and one

760
00:34:52,090 --> 00:34:53,699
accessible molecule there.

761
00:34:53,699 --> 00:34:55,710
So it's probably
true that there are

762
00:34:55,710 --> 00:34:58,290
ClpXPs which, at this
very low concentration,

763
00:34:58,290 --> 00:35:00,930
are bound non-specifically
to the surface of the bead,

764
00:35:00,930 --> 00:35:02,790
and probably bound
non-specifically

765
00:35:02,790 --> 00:35:04,590
into the actual cover slip.

766
00:35:04,590 --> 00:35:07,244
But it turns out
that those molecules

767
00:35:07,244 --> 00:35:08,535
have no signal in this process.

768
00:35:11,610 --> 00:35:13,640
So it's not as big of an issue.

769
00:35:13,640 --> 00:35:15,570
And there's no
concentration dependence

770
00:35:15,570 --> 00:35:17,310
that you're trying
to measure using

771
00:35:17,310 --> 00:35:18,990
these very low concentrations.

772
00:35:18,990 --> 00:35:23,070
So even if 99% of the stuff
you add binds non-specifically,

773
00:35:23,070 --> 00:35:25,140
it's just not that
big of a deal.

774
00:35:25,140 --> 00:35:28,590
For single-molecule fluorescence
it can be a really major deal,

775
00:35:28,590 --> 00:35:31,285
and so you have to work really
hard to basically [INAUDIBLE]

776
00:35:31,285 --> 00:35:32,160
all of your surfaces.

777
00:35:34,695 --> 00:35:35,662
So-- no, go ahead.

778
00:35:35,662 --> 00:35:38,120
AUDIENCE: I guess it doesn't
matter as much for this system

779
00:35:38,120 --> 00:35:40,610
since you have two optical
traps you're only measuring

780
00:35:40,610 --> 00:35:41,890
a one-dimensional distance.

781
00:35:41,890 --> 00:35:44,505
But for optical traps, do you
just measure displacement,

782
00:35:44,505 --> 00:35:47,800
or can you know two-dimensional,
like which direction

783
00:35:47,800 --> 00:35:48,790
it was displaced?

784
00:35:48,790 --> 00:35:50,290
REUBEN: You can
tell which direction

785
00:35:50,290 --> 00:35:54,100
it's displaced, because
actually that little picture--

786
00:35:54,100 --> 00:35:56,039
basically it will
deflect the beam of light

787
00:35:56,039 --> 00:35:57,580
onto this
position-sensitive display.

788
00:35:57,580 --> 00:36:00,970
AUDIENCE: But the voltage is
dependent upon both x and y

789
00:36:00,970 --> 00:36:02,710
directions when
working with that?

790
00:36:02,710 --> 00:36:04,210
REUBEN: I can tell you more
about the apparatus later.

791
00:36:04,210 --> 00:36:05,820
It's dependent on
both x and y, yeah.

792
00:36:05,820 --> 00:36:07,840
It's called a
quadrupole detector.

793
00:36:07,840 --> 00:36:09,730
So it basically has
four different diodes,

794
00:36:09,730 --> 00:36:11,860
and you look at the relative
ratio from these four

795
00:36:11,860 --> 00:36:12,580
different detectors.

796
00:36:12,580 --> 00:36:13,746
That tells you where it is--

797
00:36:16,200 --> 00:36:18,900
basically a four-pixel
camera, I guess.

798
00:36:18,900 --> 00:36:25,280
So the actual substrate,
the attachment

799
00:36:25,280 --> 00:36:28,520
onto the DNA, it's via a
sort of chemical biology

800
00:36:28,520 --> 00:36:30,560
which is worth knowing
about, which is something

801
00:36:30,560 --> 00:36:32,330
called the HaloTag.

802
00:36:32,330 --> 00:36:36,560
So it turns out that there
is this enzyme which,

803
00:36:36,560 --> 00:36:40,010
if you make a couple of
mutations to ruin it,

804
00:36:40,010 --> 00:36:44,661
it will make a covalent
bond with haloalkanes.

805
00:36:44,661 --> 00:36:46,910
Basically any haloalkane
that you add to your reaction

806
00:36:46,910 --> 00:36:48,620
mixture, this
enzyme will actually

807
00:36:48,620 --> 00:36:52,022
form a covalent bond with the
haloalkane in the active site.

808
00:36:52,022 --> 00:36:54,230
I forget what the original
function of the enzyme is.

809
00:36:54,230 --> 00:36:57,010
It's some sort of
halogenase or something.

810
00:36:57,010 --> 00:37:01,070
So in this case, we synthesized
this long DNA linker such

811
00:37:01,070 --> 00:37:03,380
that it had a haloalkane
at one of its ends.

812
00:37:03,380 --> 00:37:05,360
And then we added
this halo domain

813
00:37:05,360 --> 00:37:08,900
at the N-terminus of
our long substrate

814
00:37:08,900 --> 00:37:13,250
such that it would
form a covalent bond

815
00:37:13,250 --> 00:37:16,280
with the DNA which had a
biotin at the other end.

816
00:37:16,280 --> 00:37:17,810
And then we added
that to a bead.

817
00:37:17,810 --> 00:37:20,060
And it turns out that these
beads are slightly smaller

818
00:37:20,060 --> 00:37:21,107
than these beads.

819
00:37:21,107 --> 00:37:22,940
So when we actually
start these experiments,

820
00:37:22,940 --> 00:37:25,145
we trap a big bead in
one beam and a small bead

821
00:37:25,145 --> 00:37:27,140
in another beam,
and we basically

822
00:37:27,140 --> 00:37:30,680
move them very slowly
with respect to each other

823
00:37:30,680 --> 00:37:33,680
until you eventually get an
attachment between a ClpXP

824
00:37:33,680 --> 00:37:35,030
and a halo domain.

825
00:37:35,030 --> 00:37:38,210
And then you begin the actual
recording of the experiment.

826
00:37:38,210 --> 00:37:43,500
So I have a nice animation of
this actual unfolding process.

827
00:37:43,500 --> 00:37:48,080
So this illustrates the effects
of an unfolding under force.

828
00:37:48,080 --> 00:37:52,980
So once ClpXP successfully
unfolds a domain,

829
00:37:52,980 --> 00:37:56,420
you get this big jump in
bead-to-bead distance, which

830
00:37:56,420 --> 00:38:02,450
decreases during translocation
and then increases again

831
00:38:02,450 --> 00:38:04,140
during the next unfolding step.

832
00:38:04,140 --> 00:38:04,640
Alex?

833
00:38:04,640 --> 00:38:06,400
ALEX: Why do you
use a DNA linker?

834
00:38:06,400 --> 00:38:09,800
REUBEN: We use a DNA linker
to keep the Y-DNA instead

835
00:38:09,800 --> 00:38:10,970
of something else.

836
00:38:10,970 --> 00:38:13,840
It's really easy to make
a long piece of DNA.

837
00:38:13,840 --> 00:38:15,499
You know, this is
like a micron long.

838
00:38:15,499 --> 00:38:17,540
There are very few proteins
that are well-behaved

839
00:38:17,540 --> 00:38:18,970
at that sort of length.

840
00:38:18,970 --> 00:38:19,512
DNA is very--

841
00:38:19,512 --> 00:38:21,386
ALEX: But why not work
with PEG or something?

842
00:38:21,386 --> 00:38:22,752
REUBEN: --well behaved.

843
00:38:22,752 --> 00:38:24,132
Stiffness.

844
00:38:24,132 --> 00:38:25,580
PEG is very floppy.

845
00:38:25,580 --> 00:38:27,387
DNA is relatively stiff.

846
00:38:27,387 --> 00:38:28,970
There's still some
sort of floppiness,

847
00:38:28,970 --> 00:38:31,470
which you account for using
what they call a worm-like chain

848
00:38:31,470 --> 00:38:33,680
model, which if you have
any questions about that,

849
00:38:33,680 --> 00:38:35,630
ask me afterward.

850
00:38:35,630 --> 00:38:37,430
But DNA is much
stiffer than just

851
00:38:37,430 --> 00:38:39,764
a single PEG, and also
very strong compared

852
00:38:39,764 --> 00:38:40,430
to a single PEG.

853
00:38:44,020 --> 00:38:48,340
So again, what you see is
unfolding translocation,

854
00:38:48,340 --> 00:38:50,739
and then unfolding again
at the next substrate.

855
00:38:50,739 --> 00:38:52,780
So do any of you have
questions about the readout

856
00:38:52,780 --> 00:38:55,681
that we get when we're
actually looking at acquiring

857
00:38:55,681 --> 00:38:56,680
the data for this paper?

858
00:38:56,680 --> 00:38:59,030
Does this make
sense to you guys?

859
00:38:59,030 --> 00:39:00,716
Cool.

860
00:39:00,716 --> 00:39:02,090
So I'm going to
get into a couple

861
00:39:02,090 --> 00:39:03,630
of the figures of the paper.

862
00:39:03,630 --> 00:39:06,482
This paper is very difficult,
and has a lot of data in it.

863
00:39:06,482 --> 00:39:07,940
But some of the
figures I think are

864
00:39:07,940 --> 00:39:10,320
relatively easy to understand.

865
00:39:10,320 --> 00:39:14,450
So I believe you covered these
different mutations of titin,

866
00:39:14,450 --> 00:39:16,820
which change the
mechanical stability

867
00:39:16,820 --> 00:39:19,830
and change the rate of ClpXP
unfolding in degradation.

868
00:39:19,830 --> 00:39:22,430
So using these mutations,
we were actually

869
00:39:22,430 --> 00:39:28,310
able to directly investigate
the unfolding strength of ClpXP.

870
00:39:28,310 --> 00:39:31,880
So this titin I27 domain,
there are a couple

871
00:39:31,880 --> 00:39:35,210
of mutations you can make
right by the C-terminus

872
00:39:35,210 --> 00:39:39,380
of the domain, where you
basically switch a valine

873
00:39:39,380 --> 00:39:42,500
or something for a proline,
which significantly decreases

874
00:39:42,500 --> 00:39:45,030
the mechanical stability.

875
00:39:45,030 --> 00:39:47,780
And we made these
multi-domain substrates

876
00:39:47,780 --> 00:39:51,350
with these various
mutants of titin inserted.

877
00:39:51,350 --> 00:39:57,560
And then we investigated the
time of the pre-unfolding dwell

878
00:39:57,560 --> 00:39:59,600
before this big
bead-to-bead distance

879
00:39:59,600 --> 00:40:02,510
for each of these unfoldings
events before a domain.

880
00:40:02,510 --> 00:40:05,540
And what we saw is that,
for a wild-type titin,

881
00:40:05,540 --> 00:40:08,840
the most stable, we
saw relatively long

882
00:40:08,840 --> 00:40:11,690
pre-unfolding dwells.

883
00:40:11,690 --> 00:40:14,090
Whereas for the
V15P titin, which

884
00:40:14,090 --> 00:40:16,550
has the intermediate
mechanical strength,

885
00:40:16,550 --> 00:40:19,750
we saw shorter
pre-unfolding dwells.

886
00:40:19,750 --> 00:40:23,750
And for the V13P titin,
we saw the shortest

887
00:40:23,750 --> 00:40:28,440
pre-unfolding dwells, which
basically are the flat periods

888
00:40:28,440 --> 00:40:30,470
in between these jumps.

889
00:40:30,470 --> 00:40:32,530
So it turns out that--

890
00:40:32,530 --> 00:40:36,260
so you can quantitate the
length of these dwells

891
00:40:36,260 --> 00:40:39,700
by basically plotting
the number of dwells

892
00:40:39,700 --> 00:40:43,050
of a certain length,
the frequency of dwells

893
00:40:43,050 --> 00:40:46,600
of a certain length or
shorter, versus that length

894
00:40:46,600 --> 00:40:48,300
on the x-axis.

895
00:40:48,300 --> 00:40:50,690
So this is basically making
a cumulative distribution

896
00:40:50,690 --> 00:40:53,330
plot of the dwells.

897
00:40:53,330 --> 00:40:56,160
And it's pretty
easy to show that,

898
00:40:56,160 --> 00:40:58,700
if you're looking at a
single exponential decay

899
00:40:58,700 --> 00:41:01,260
process such as a
single kinetic step,

900
00:41:01,260 --> 00:41:04,281
then this plot should
have an exponential shape

901
00:41:04,281 --> 00:41:06,530
where it's fit very well by
a single exponential decay

902
00:41:06,530 --> 00:41:07,870
process.

903
00:41:07,870 --> 00:41:11,120
And so you can look at the
rate of that exponential

904
00:41:11,120 --> 00:41:12,800
to actually gain
a lot of insight

905
00:41:12,800 --> 00:41:16,880
into the rate of that
particular kinetic step.

906
00:41:16,880 --> 00:41:20,380
So for the wild
type titin, we found

907
00:41:20,380 --> 00:41:22,930
that the half-life for
this unfolding event,

908
00:41:22,930 --> 00:41:26,170
which you can extract from
this exponential decay rate,

909
00:41:26,170 --> 00:41:28,420
was about 55 seconds.

910
00:41:28,420 --> 00:41:32,140
Whereas for this V15P
titin, which we knew

911
00:41:32,140 --> 00:41:36,110
is less mechanically stable,
it was about 17 seconds.

912
00:41:36,110 --> 00:41:40,430
And for this V13P titin
it was about six seconds.

913
00:41:40,430 --> 00:41:42,280
And we saw that
these plots are not

914
00:41:42,280 --> 00:41:44,430
perfectly fit by a
single exponential,

915
00:41:44,430 --> 00:41:46,390
but they are fit very well.

916
00:41:46,390 --> 00:41:48,160
And this sort of
indicates that there

917
00:41:48,160 --> 00:41:52,870
is one rate limiting step,
one major kinetic step

918
00:41:52,870 --> 00:41:55,450
in the actual process
of unfolding, which

919
00:41:55,450 --> 00:41:57,160
is what we assumed,
because we believed

920
00:41:57,160 --> 00:41:59,740
that these domains unfolded
very cooperatively.

921
00:41:59,740 --> 00:42:03,490
And so basically this suggests
that a single successful power

922
00:42:03,490 --> 00:42:06,580
stroke by ClpXP is
responsible for the unfolding

923
00:42:06,580 --> 00:42:08,445
of these domains.

924
00:42:08,445 --> 00:42:10,820
So do any of you have any
questions about how we measured

925
00:42:10,820 --> 00:42:13,492
unfolding in this circumstance?

926
00:42:13,492 --> 00:42:15,200
AUDIENCE: Well, just
looking at the fits,

927
00:42:15,200 --> 00:42:16,960
it almost looks like
it decays slower

928
00:42:16,960 --> 00:42:19,161
than a monoexponential decay.

929
00:42:19,161 --> 00:42:19,660
Right?

930
00:42:19,660 --> 00:42:24,560
Do you have an explanation
for why that is like that?

931
00:42:24,560 --> 00:42:26,660
REUBEN: No.

932
00:42:26,660 --> 00:42:28,870
There aren't that
many data points here.

933
00:42:28,870 --> 00:42:30,820
And some of these
data points out

934
00:42:30,820 --> 00:42:34,700
at these high distances
are a little bit hairy.

935
00:42:34,700 --> 00:42:39,650
I would say there's quite a
bit of error in biochemistry,

936
00:42:39,650 --> 00:42:44,820
and my guess is that the quality
of these fits is like 99.9.

937
00:42:44,820 --> 00:42:46,820
And without
dramatically more data,

938
00:42:46,820 --> 00:42:50,330
it's not very
useful to speculate

939
00:42:50,330 --> 00:42:54,140
about what this other
exponential might be.

940
00:42:54,140 --> 00:42:58,100
So I'm not going to say.

941
00:42:58,100 --> 00:42:59,950
I honestly have no idea.

942
00:42:59,950 --> 00:43:01,450
JOANNE STUBBE: But
I think that when

943
00:43:01,450 --> 00:43:02,824
you're looking at
data like that,

944
00:43:02,824 --> 00:43:06,650
sometimes it's better to look
way down here so you can see

945
00:43:06,650 --> 00:43:09,630
the quality of the
fit, so you can

946
00:43:09,630 --> 00:43:11,000
look at the quality of the fit.

947
00:43:11,000 --> 00:43:13,210
It's hard to see
it on that graph.

948
00:43:13,210 --> 00:43:14,070
REUBEN: Oh, sorry.

949
00:43:14,070 --> 00:43:14,190
Yeah.

950
00:43:14,190 --> 00:43:16,815
JOANNE STUBBE: [INAUDIBLE] would
mean a lot of the data points.

951
00:43:16,815 --> 00:43:18,024
And then you look at the fit.

952
00:43:18,024 --> 00:43:18,564
REUBEN: Yeah.

953
00:43:18,564 --> 00:43:20,910
JOANNE STUBBE: I think the
data looks [INAUDIBLE]..

954
00:43:20,910 --> 00:43:21,451
REUBEN: Yeah.

955
00:43:21,451 --> 00:43:24,280
Given-- I mean, this data took
Adrian like a year to acquire,

956
00:43:24,280 --> 00:43:27,470
because you get so few
data points per experiment.

957
00:43:27,470 --> 00:43:29,870
With the error, this
is pretty darn good

958
00:43:29,870 --> 00:43:32,750
for this type of setup.

959
00:43:32,750 --> 00:43:36,200
It may be that sometimes it
takes two very rapid power

960
00:43:36,200 --> 00:43:39,461
strokes, and that might add
some longer exponential decay

961
00:43:39,461 --> 00:43:39,960
process.

962
00:43:39,960 --> 00:43:43,910
But it's very difficult to say.

963
00:43:43,910 --> 00:43:46,940
So we know from the
crystal structures,

964
00:43:46,940 --> 00:43:48,740
as well as from
biochemical studies,

965
00:43:48,740 --> 00:43:52,340
that ClpXP has these pore loops
along its axial pore, which

966
00:43:52,340 --> 00:43:55,010
are actually directly involved
in the mechanical activity

967
00:43:55,010 --> 00:43:56,100
of the protein.

968
00:43:56,100 --> 00:43:59,780
So we believe that there
are these RKH loops, which

969
00:43:59,780 --> 00:44:02,360
are on the top face of ClpXP.

970
00:44:02,360 --> 00:44:04,520
And these loops make
this initial interaction

971
00:44:04,520 --> 00:44:06,200
with the ssrA tag.

972
00:44:06,200 --> 00:44:09,380
And then there are pore-1
loops within the pore,

973
00:44:09,380 --> 00:44:12,380
and we believe that the pore-1
loops are the loops that

974
00:44:12,380 --> 00:44:15,320
are basically the levers of
mechanical force application

975
00:44:15,320 --> 00:44:19,260
to a substrate during
unfolding and translocation.

976
00:44:19,260 --> 00:44:22,260
So these pore loops,
you can imagine

977
00:44:22,260 --> 00:44:24,540
them undergoing a
conformational change, where

978
00:44:24,540 --> 00:44:27,060
they grab onto an
unfolded polypeptide

979
00:44:27,060 --> 00:44:29,500
and basically are
translocated downward

980
00:44:29,500 --> 00:44:32,100
in some sort of
nucleotide-dependent manner,

981
00:44:32,100 --> 00:44:34,370
dragging the substrate with it.

982
00:44:34,370 --> 00:44:37,470
And if the substrate is
unfolded, then it comes along.

983
00:44:37,470 --> 00:44:40,050
If it's folded, then this
can be either a successful

984
00:44:40,050 --> 00:44:42,820
or an unsuccessful
unfolding event.

985
00:44:42,820 --> 00:44:44,950
And then there are also
these pore-2 loops,

986
00:44:44,950 --> 00:44:46,650
which we think are
involved in holding

987
00:44:46,650 --> 00:44:49,380
the substrate at the
bottom of the pore

988
00:44:49,380 --> 00:44:53,640
as the pore-1 loops are reset
for the next power stroking

989
00:44:53,640 --> 00:44:55,530
event.

990
00:44:55,530 --> 00:44:57,780
And we've done a
lot of experiments

991
00:44:57,780 --> 00:45:03,930
which try to explain how these
power stroking events are

992
00:45:03,930 --> 00:45:08,190
related to the actual
process of ATP hydrolysis

993
00:45:08,190 --> 00:45:12,270
to ADP and phosphate, and then
phosphate release, and then ADP

994
00:45:12,270 --> 00:45:13,920
release.

995
00:45:13,920 --> 00:45:16,110
And we have some
evidence which suggests

996
00:45:16,110 --> 00:45:19,590
that phosphate release
after ATP hydrolysis

997
00:45:19,590 --> 00:45:22,080
is the step which is
most intimately coupled

998
00:45:22,080 --> 00:45:25,320
with this particular
conformational change that

999
00:45:25,320 --> 00:45:27,270
applies to mechanical force.

1000
00:45:27,270 --> 00:45:28,290
But it's--

1001
00:45:28,290 --> 00:45:29,665
JOANNE STUBBE: So
how do you look

1002
00:45:29,665 --> 00:45:31,790
at that kind of measurement?

1003
00:45:31,790 --> 00:45:34,370
REUBEN: The way that we've
measured this most effectively

1004
00:45:34,370 --> 00:45:37,760
is to do single-molecule
experiments of the sort I'm

1005
00:45:37,760 --> 00:45:41,540
going to show on the next plot
with either excess phosphate

1006
00:45:41,540 --> 00:45:45,170
or excess vanadate.

1007
00:45:45,170 --> 00:45:49,400
So the idea is that
after phosphate leaves,

1008
00:45:49,400 --> 00:45:52,760
there is some sort of
a rate constant, which

1009
00:45:52,760 --> 00:45:57,440
describes the time it takes
for this conformational change

1010
00:45:57,440 --> 00:45:58,470
to occur.

1011
00:45:58,470 --> 00:46:00,890
So if you have phosphate
or a phosphate analog that

1012
00:46:00,890 --> 00:46:04,940
rebinds before that
conformational change can

1013
00:46:04,940 --> 00:46:08,585
occur, then that should
increase the dwell time.

1014
00:46:08,585 --> 00:46:10,835
JOANNE STUBBE: That's only
one phosphate binding site?

1015
00:46:10,835 --> 00:46:12,791
I mean, you have
six-- how many--

1016
00:46:12,791 --> 00:46:14,747
you have multiple subunits.

1017
00:46:14,747 --> 00:46:18,659
And depending on whether the
other nucleotide is around,

1018
00:46:18,659 --> 00:46:20,615
you only have one
binding site if you have

1019
00:46:20,615 --> 00:46:24,530
a huge excess of Pi, or is it--

1020
00:46:24,530 --> 00:46:27,830
REUBEN: We think that the
important phosphate-leaving

1021
00:46:27,830 --> 00:46:31,590
step is taking place
when there is ADP bound.

1022
00:46:31,590 --> 00:46:35,390
So in this subunit where
you have ADP bound and also

1023
00:46:35,390 --> 00:46:37,539
phosphate bound, and then
that phosphate leaves,

1024
00:46:37,539 --> 00:46:39,080
we think that that
is the step that's

1025
00:46:39,080 --> 00:46:41,366
related to the actual
mechanical motion.

1026
00:46:41,366 --> 00:46:43,490
But we're not going to
claim that that is the step,

1027
00:46:43,490 --> 00:46:45,500
we're going to say we
think that's the step.

1028
00:46:45,500 --> 00:46:47,570
And our evidence
supports that versus any

1029
00:46:47,570 --> 00:46:53,510
of the other steps such as
ATP binding or ADP release,

1030
00:46:53,510 --> 00:46:59,000
because you can show that
adding a lot of competitor ADP

1031
00:46:59,000 --> 00:47:01,310
or changing the
rate of ATP does not

1032
00:47:01,310 --> 00:47:04,610
have the effect on this
dwell time in between steps

1033
00:47:04,610 --> 00:47:07,280
that you would expect if
those were the steps that had

1034
00:47:07,280 --> 00:47:09,680
this central mechanical role.

1035
00:47:09,680 --> 00:47:14,570
But the evidence is
not totally compelling.

1036
00:47:14,570 --> 00:47:16,450
But just imagine that
something related

1037
00:47:16,450 --> 00:47:18,580
to this process
of ATP hydrolysis

1038
00:47:18,580 --> 00:47:21,400
is basically causing
this pore loop

1039
00:47:21,400 --> 00:47:24,220
to be translocated downward.

1040
00:47:24,220 --> 00:47:27,220
So what we could do is--

1041
00:47:27,220 --> 00:47:30,340
I showed these plots where you
see this unfolding and then

1042
00:47:30,340 --> 00:47:31,980
the slow translocation.

1043
00:47:31,980 --> 00:47:34,900
You can actually, because
these optical traps

1044
00:47:34,900 --> 00:47:38,170
have such high resolution,
you can actually

1045
00:47:38,170 --> 00:47:40,510
monitor the individual
steps that ClpXP

1046
00:47:40,510 --> 00:47:43,150
takes as it walks along a
substrate, which I actually

1047
00:47:43,150 --> 00:47:45,260
think is pretty awesome.

1048
00:47:45,260 --> 00:47:50,950
So you can fit these
steps to a model

1049
00:47:50,950 --> 00:47:53,380
to get cleaner quantitation.

1050
00:47:53,380 --> 00:47:57,670
And then you can quantitate
the length of these steps.

1051
00:47:57,670 --> 00:48:00,070
And you can see that
ClpXP takes steps

1052
00:48:00,070 --> 00:48:03,460
that are a range between
about one nanometer and about

1053
00:48:03,460 --> 00:48:05,060
four nanometers.

1054
00:48:05,060 --> 00:48:07,960
And we quantize this.

1055
00:48:07,960 --> 00:48:12,640
We say that ClpXP can take
steps between one and four

1056
00:48:12,640 --> 00:48:15,850
in multiples of one, because
we know from the crystal

1057
00:48:15,850 --> 00:48:18,430
structures that this
particular pore loop

1058
00:48:18,430 --> 00:48:22,966
translocation downward, it
moves about one nanometer.

1059
00:48:22,966 --> 00:48:28,990
And so what we've
found is that you

1060
00:48:28,990 --> 00:48:31,480
can look at the order
of these step sizes.

1061
00:48:31,480 --> 00:48:34,920
And you find that
ClpXP, it rarely takes--

1062
00:48:34,920 --> 00:48:38,050
so the order of steps,
which you can see A,

1063
00:48:38,050 --> 00:48:40,095
it's taking a one-nanometer
step, and then

1064
00:48:40,095 --> 00:48:42,982
a two-nanometer step, and
then a one-nanometer step,

1065
00:48:42,982 --> 00:48:45,190
and then a couple other
one-nanometer steps, and then

1066
00:48:45,190 --> 00:48:47,350
some twos and then some threes.

1067
00:48:47,350 --> 00:48:50,400
And you can see that for
B and for C, as well.

1068
00:48:50,400 --> 00:48:57,090
We found that the order of these
steps is relatively random.

1069
00:48:57,090 --> 00:49:00,960
It's very difficult to use
one step to predict what

1070
00:49:00,960 --> 00:49:02,640
the next step is going to be.

1071
00:49:02,640 --> 00:49:06,150
So from that, we think that
there is a significant degree

1072
00:49:06,150 --> 00:49:12,750
of stochasticity in some
aspect of the mechanism

1073
00:49:12,750 --> 00:49:15,600
of this enzyme,
determining possibly which

1074
00:49:15,600 --> 00:49:18,600
subunits are hydrolyzing
ATP to power a power stroke.

1075
00:49:18,600 --> 00:49:21,570
But we say that it's not
completely stochastic.

1076
00:49:21,570 --> 00:49:24,300
The events are not all
equally likely to occur

1077
00:49:24,300 --> 00:49:25,900
after each other event.

1078
00:49:25,900 --> 00:49:29,400
So for example, after
a four-nanometer step,

1079
00:49:29,400 --> 00:49:32,250
we very rarely see a
second four-nanometer step

1080
00:49:32,250 --> 00:49:33,750
or a three-nanometer step.

1081
00:49:33,750 --> 00:49:36,060
We are much more likely to
see a one-nanometer step

1082
00:49:36,060 --> 00:49:38,430
or a two-nanometer step.

1083
00:49:38,430 --> 00:49:42,510
Whereas it's much more likely
that after a one-nanometer

1084
00:49:42,510 --> 00:49:45,810
step, we might
see a longer step.

1085
00:49:45,810 --> 00:49:51,300
So we have come up with a couple
of complicated kinetic models

1086
00:49:51,300 --> 00:49:53,890
that can explain
some of this data.

1087
00:49:53,890 --> 00:49:56,280
And we're not saying
that it's correct.

1088
00:49:56,280 --> 00:49:58,860
What we're more saying is
that this is the sort of thing

1089
00:49:58,860 --> 00:50:02,280
that you have to think
about when really asking

1090
00:50:02,280 --> 00:50:04,800
deep questions about
the mechanical activity

1091
00:50:04,800 --> 00:50:05,790
of this machine.

1092
00:50:05,790 --> 00:50:08,460
So a model which explains
this behavior where

1093
00:50:08,460 --> 00:50:11,550
you don't see several
long steps in a row

1094
00:50:11,550 --> 00:50:15,240
is to say that these steps,
which are quantized according

1095
00:50:15,240 --> 00:50:17,970
to this one nanometer, what
they actually represent

1096
00:50:17,970 --> 00:50:22,410
is cascades of steps, possibly
a two-nanometer step being two

1097
00:50:22,410 --> 00:50:25,170
very quick steps in a row, or
a four-nanometer step being

1098
00:50:25,170 --> 00:50:27,240
four very quick
steps in a row, which

1099
00:50:27,240 --> 00:50:28,920
occurred too quickly
for our instrument

1100
00:50:28,920 --> 00:50:31,380
to actually catch them.

1101
00:50:31,380 --> 00:50:34,350
And you can say that
each step is basically

1102
00:50:34,350 --> 00:50:36,590
controlled by
nucleotide hydrolysis

1103
00:50:36,590 --> 00:50:39,680
or by phosphate release
in a single subunit.

1104
00:50:39,680 --> 00:50:44,940
And say that possibly a
four-nanometer step involves

1105
00:50:44,940 --> 00:50:49,505
phosphate release in four
subunits, sort of one

1106
00:50:49,505 --> 00:50:50,130
after the next.

1107
00:50:50,130 --> 00:50:52,800
Maybe they're all contiguous,
or maybe they're not.

1108
00:50:52,800 --> 00:50:55,920
We don't really have any
evidence going either way.

1109
00:50:55,920 --> 00:51:00,240
And so you have boom,
boom, boom, boom,

1110
00:51:00,240 --> 00:51:02,940
in a cascade of four
very quick steps.

1111
00:51:02,940 --> 00:51:05,190
Whereas a two-nanometer
step involves

1112
00:51:05,190 --> 00:51:07,860
hydrolysis in two subunits,
and a one-nanometer step

1113
00:51:07,860 --> 00:51:10,014
involves hydrolysis in
just a single subunit.

1114
00:51:10,014 --> 00:51:11,430
So the reason that
we're attracted

1115
00:51:11,430 --> 00:51:14,190
to models such as
this is that after you

1116
00:51:14,190 --> 00:51:16,430
have one of these
four-nanometer steps,

1117
00:51:16,430 --> 00:51:19,410
you have basically
lost your phosphate,

1118
00:51:19,410 --> 00:51:23,380
or lost your ATP, or whatever,
in four different subunits.

1119
00:51:23,380 --> 00:51:26,580
And so it takes
time for ATP to bind

1120
00:51:26,580 --> 00:51:29,190
and possibly to be hydrolyzed
again in these four

1121
00:51:29,190 --> 00:51:30,490
different subunits.

1122
00:51:30,490 --> 00:51:33,720
So it's unlikely
that you're going

1123
00:51:33,720 --> 00:51:36,570
to have the time for
four ATPs to bind

1124
00:51:36,570 --> 00:51:39,810
and be hydrolyzed before
your next power stroke.

1125
00:51:39,810 --> 00:51:41,550
Whereas it only
requires binding in one

1126
00:51:41,550 --> 00:51:47,610
subunit to power a
single-nanometer step.

1127
00:51:47,610 --> 00:51:52,200
So this model, we don't
actually have much evidence

1128
00:51:52,200 --> 00:51:55,620
that suggests that it has to be
sequential orders of subunits

1129
00:51:55,620 --> 00:51:58,020
going around the ring like that.

1130
00:51:58,020 --> 00:52:00,570
We've chosen this
counterclockwise direction

1131
00:52:00,570 --> 00:52:02,124
completely arbitrarily.

1132
00:52:02,124 --> 00:52:03,540
The reason that
we think that it's

1133
00:52:03,540 --> 00:52:05,580
limited to four-nanometer
steps rather

1134
00:52:05,580 --> 00:52:08,100
than five-nanometer
or six-nanometer steps

1135
00:52:08,100 --> 00:52:12,180
is that we know that ClpXP, even
at saturating concentrations

1136
00:52:12,180 --> 00:52:17,370
of ATP, basically never
binds more than about four

1137
00:52:17,370 --> 00:52:20,170
equivalents of ATP per hexamer.

1138
00:52:20,170 --> 00:52:22,470
And we also know that
there are these U subunits,

1139
00:52:22,470 --> 00:52:25,050
these unloadable subunits,
which are actually

1140
00:52:25,050 --> 00:52:28,270
not competent for ATP binding.

1141
00:52:28,270 --> 00:52:31,380
So possibly a
four-nanometer step

1142
00:52:31,380 --> 00:52:34,230
could involve phosphate
release from all four

1143
00:52:34,230 --> 00:52:37,230
of the subunits
that have ATP bound.

1144
00:52:37,230 --> 00:52:40,250
But then the step stops
when this cascade event,

1145
00:52:40,250 --> 00:52:42,270
this cascade of
conformational changes,

1146
00:52:42,270 --> 00:52:46,080
reaches an unloadable subunit,
which has nothing bound.

1147
00:52:46,080 --> 00:52:49,870
So we are trying to add
more to this mechanism,

1148
00:52:49,870 --> 00:52:54,270
but the behavior of this
protein is very complicated.

1149
00:52:54,270 --> 00:52:56,839
So I should just say,
I didn't actually

1150
00:52:56,839 --> 00:52:58,380
record any of the
data in this paper.

1151
00:52:58,380 --> 00:53:02,240
I don't work on
optical traps at all.

1152
00:53:02,240 --> 00:53:05,920
If you have any questions about
the experiments that I do,

1153
00:53:05,920 --> 00:53:08,640
which more directly
investigate the model I just

1154
00:53:08,640 --> 00:53:11,490
showed on the last page using
single-molecule fluorescence,

1155
00:53:11,490 --> 00:53:13,200
feel free to ask me anytime.

1156
00:53:13,200 --> 00:53:15,750
These experiments
were done by Adrian,

1157
00:53:15,750 --> 00:53:17,730
who's a postdoc in the
lab, who's probably

1158
00:53:17,730 --> 00:53:21,600
going to start as a professor
at UC San Francisco next year.

1159
00:53:21,600 --> 00:53:23,010
This is the Sauer lab.

1160
00:53:23,010 --> 00:53:24,320
It's a great place to Year Up.

1161
00:53:24,320 --> 00:53:27,960
If any of you are looking for
a new Year Up, come say hi.

1162
00:53:27,960 --> 00:53:30,210
So if you have any other
questions about optical traps

1163
00:53:30,210 --> 00:53:33,470
or ClpXP, feel free to ask me.