Control of Embryonic Axis Formation in Drosophila involving Gurken

Part 1: Axes Formation in the Drosophila Egg

17:02.2 where gurken indicates the dorsal position of the oocyte nucleus
06:44.3 and then whenever we see something of interest to our question,
15:35.1 and that will lead to the expression of certain target genes.
06:30.0 that are homozygous for a particular mutagenized chromosome,
09:57.2 and when this wildtype recipient embryo grows up into a fly,
06:48.3 we can retain the lines and then study them in more detail.
21:54.3 both major body axes of the egg and embryo are established.
13:09.3 to receive this information that comes from the germline,
13:56.1 is required multiple times during Drosophila development,
14:42.1 dorsal follicle cells as well as ventral follicle cells.
16:13.1 why is just found in this particular part of the oocyte?
17:08.1 but even earlier in oogenesis, gurken is already active.
18:17.2 but this time the signal indicates to the follicle cells
01:57.1 which is accumulating at the posterior pole of the egg.
04:15.0 can also already be predicted when we look at this egg.
07:29.1 you see an egg that doesn't have any dorsal appendages.
12:34.3 It's also important for the surrounding follicle cells.
17:51.1 results in a reorganization of the microtubule network
01:17.1 have nuclei with now express different sets of genes.
02:36.1 and you can see that there are many egg chambers here
04:21.0 in the egg chambers during later stages of oogenesis.
05:39.0 molecular processes that establish these asymmetries,
11:57.1 we only see the phenotype, the ventralized phenotype,
13:03.2 And torpedo, which is required in the follicle cells,
14:02.2 It can be activated by a number of different ligands.
21:36.2 and the follicle cells then signal back to the oocyte
02:32.2 an ovary of a fruit fly dissected out of the female,
07:43.1 ventralized and it will only form ventral structures
08:01.2 on the dorsal side, that form the dorsal appendages,
13:00.2 is somehow signaled to the overlying follicle cells.
18:56.3 just accumulating at the anterior end of the oocyte,
19:01.1 but mutations in which gurken/EGF receptor signaling
19:19.2 to the posterior end of the egg chamber and the egg,
05:23.2 because they have to form this asymmetric egg shell
08:44.1 and by the embryo that forms inside the egg itself.
08:51.0 we have mutations that affect two different tissues
17:42.1 and this signal is still not quite well understood,
18:29.2 early to establish the anterior-posterior asymmetry
20:45.0 where gurken again signals through the EGF receptor
00:41.2 The model organism in which we study this question
02:48.1 so that we have the oldest egg chambers down here.
09:54.2 and implant them into a wildtype recipient embryo,
11:37.1 when we did the converse, or the same experiments,
13:42.3 So, this is a receptor molecule that's very large.
14:28.0 accumulates very asymmetrically within the oocyte.
21:28.2 through the gurken/EGF receptor signaling pathway.
02:54.0 at a single one of these strings of egg chambers,
12:43.1 look at their own genotype for pattern formation,
19:36.1 we need the gurken/EGF receptor signaling pathway
20:14.1 has not be totally determined in molecular terms,
03:08.3 which will differentiate into the eventual eggs.
05:12.0 by the follicle cells which surround the oocyte.
05:42.1 and how can we understand how the follicle cells
06:42.2 how the embryos look that are inside these eggs,
15:57.0 and is used to create dorso-ventral asymmetries.
20:30.1 in the reorganization of the microtubule network
01:10.2 how do we go from the egg with a single nucleus
05:29.0 that are present in the oocyte and nurse cells.
07:16.0 by these two structures, the dorsal appendages,
08:35.3 because we see a change in patterning, in fact,
17:46.2 but we know that the signal back to the oocyte,
18:14.2 through the EGF receptor to the follicle cells,
19:29.2 and instead accumulates somewhere in the middle
20:51.3 the dorsoventral pattern of the egg and embryo.
02:41.2 At the top, we have the youngest egg chambers,
03:01.2 At the top of the ovariole we have stem cells.
03:43.1 and the whole entity is called an egg chamber,
03:48.2 that the pattern and polarity are established.
15:02.2 Remember, gurken encodes a signaling molecule,
15:43.1 where the follicle cells don't see high levels
16:19.1 is at the anterior dorsal corner at this stage
04:41.3 occupies a dorsal position within the oocyte,
05:09.2 And, the egg shell is formed during oogenesis
06:05.2 what genes are important for these processes.
11:05.2 wildtype germ cells into mutant female hosts,
14:49.3 in a very restricted part of the egg chamber.
15:30.2 that see high levels of this gurken molecule,
16:08.1 well, if we see the gurken protein localized,
16:48.1 and also of the protein and RNA distribution,
18:00.3 and over to just one side of the egg chamber.
21:01.3 how in Drosophila both the anterior-posterior
02:14.1 localized to the vegetal pole of the oocyte.
09:11.1 And there are methods to ask these questions
10:17.2 where we have a mutant oocyte and nurse cell
11:11.1 we found that in this particular case, then,
12:02.2 We don't see it when the germline is mutant.
12:15.2 between the germline and the follicle cells,
14:56.1 in the dorsal anterior corner of the oocyte,
15:22.1 to signal the position of the oocyte nucleus
18:47.1 So, one very famous anteriorly-localized RNA
20:20.2 are trying to find exactly how this happens.
00:15.1 in the question of how pattern and polarity
01:37.3 and in particular there are some important,
05:04.2 This is what we are really looking at here,
09:31.3 where one is mutant and the other wildtype.
12:19.2 because gurken is obviously doing something
14:13.2 by the oocyte to signal to the EGF receptor
14:33.2 you see the expression of the EGF receptor,
15:07.1 and this can then bind to the EGF receptor.
15:11.1 So, now, we can understand that the oocyte,
17:17.1 the gurken protein already being expressed.
18:40.2 where we look at the particular anteriorly-
19:11.0 both at the anterior and the posterior end.
20:10.0 This signal, as I said, is at the moment...
21:22.1 but in both cases these initial asymmetries
21:48.1 between the germline and the follicle cells
00:31.2 that contains a single fertilized nucleus,
00:53.2 just as it's been deposited by the female,
03:18.2 and the actual differentiation of the egg:
03:34.2 and 1 of the cells will become the oocyte.
04:30.0 and you can predict the anterior-posterior
05:46.1 know where the anterior and posterior ends
06:23.2 breed such lines through some generations,
09:49.1 so we can for instance take the germ cells
10:53.2 but surrounded by wildtype follicle cells,
10:58.2 the eggs and the embryos were ventralized.
12:29.2 but it's not only important for the embryo
13:06.2 might then be acting in the follicle cells
14:21.2 when we looked at the protein distribution
15:41.0 will not be activated on the ventral side,
16:35.2 and then it's accumulating in this corner,
19:41.1 the right anterior and posterior addresses
20:34.2 and also causes the oocyte nucleus to move
21:18.3 and moving the oocyte nucleus to one side,
00:27.2 is the question of how we go from an egg,
03:25.1 and they will form a cluster of 16 cells.
03:29.2 15 of these cells will become nurse cells
06:19.2 We can then cross these males to females,
06:58.2 different mutant phenotypes that we found
08:13.0 and have tried to make dorsal appendages.
12:58.2 that whatever gurken does in the germline
14:00.0 and it acts in several different tissues.
14:36.1 which is expressed in the follicle cells.
16:33.0 it's rapidly transported into the oocyte,
16:54.1 the gurken/EGF receptor signaling pathway
19:59.2 and induces the posterior follicles cells
04:37.0 at the posterior end of the egg chamber,
08:24.2 and no mesoderm and no ventral ectoderm.
08:56.1 where is the wildtype gene really active
09:16.2 and they usually involve making mosaics.
12:26.3 that is important for pattern formation,
13:40.1 of the epidermal growth factor receptor.
17:27.1 at the posterior end of the egg chamber.
19:57.2 where gurken signals to the EGF receptor
20:07.0 then send a signal back to the germline.
21:40.0 and cause asymmetries within the oocyte,
01:31.1 but that there is a lot of organization
03:40.1 is then enveloped by the follicle cells
04:57.1 Now, these asymmetries that are visible
05:36.1 Now, how can we find out the underlying
07:11.2 and the dorsal side of the wildtype egg
08:08.0 from all around the egg's circumference
08:19.2 the embryos inside are also dorsalized,
09:06.2 Where is the pattern established first?
09:38.3 we're using transplantation of cells...
10:20.2 surrounded by wildtype follicles cells,
10:35.1 that ventralize the egg and the embryo.
11:29.3 Whenever the follicle cells are mutant,
13:20.2 These results were further corroborated
13:31.2 that are encoded by gurken and torpedo.
15:54.1 is then signaled to the follicle cells,
19:24.2 in the mutants, it doesn't seem to know
20:58.2 I've tried to explain to you in my talk
21:26.3 are then signaled to the follicle cells
01:26.2 that eggs are actually not just simply
01:35.2 as they are being produced by females,
05:06.1 the outside of the egg, the egg shell.
05:26.2 in correspondence with the asymmetries
06:01.1 that we can do mutagenesis experiments
07:27.0 So, here, in the middle, for instance,
07:34.2 It has lost all its dorsal structures,
07:45.2 such as mesoderm and ventral ectoderm.
08:10.2 have become dorsal, and have migrated,
08:42.1 which is formed by the follicle cells,
09:22.1 where we're interested in the question
11:42.1 with the other gene that also produced
11:46.2 torpedo, we found the opposite result.
13:16.0 to act and implement this information.
14:06.2 Gurken, which is one of these ligands,
14:39.1 It's expressed in all follicle cells -
15:15.3 which has localized its oocyte nucleus
17:19.3 The oocyte at this stage is very small
18:54.2 which you can see here in the wildtype
21:33.2 are induced to take on specific fates,
01:42.3 that are often localized within eggs.
04:12.3 and then the dorsal and ventral sides
06:11.3 is to perform mutagenesis experiments
07:01.0 when we did such mutagenesis screens.
07:58.1 very defined groups of follicle cells
10:22.3 and we can then ask what will happen.
18:04.2 Once it's in its asymmetric position,
18:07.2 the gurken RNA and the gurken protein
18:20.1 to become dorsal rather than ventral.
21:17.0 by moving the oocyte to the posterior
00:09.1 and a professor of molecular biology
03:04.1 These stem cells divide continuously
04:06.2 for instance, the head of the embryo
05:18.1 the follicle cells have to have also
05:56.1 we have the advantage that we can...
06:36.1 to see whether they lay normal eggs,
07:24.1 for dorso-ventral pattern formation.
08:16.2 And, when these eggs are fertilized,
09:08.2 What comes first, what comes second?
09:24.2 of patterning between the egg itself
10:04.1 in most of its body parts, wildtype,
10:14.2 and so then we end up with a mosaic,
16:06.1 Of course we then have the question,
19:45.3 so that these RNAs know where to go.
20:40.1 and then once the nucleus has moved,
00:13.1 In my laboratory, we are interested
00:56.0 and it has a single nucleus inside.
02:59.0 we can look at this in more detail.
04:01.2 after it's been laid by the female.
04:18.2 The asymmetries are already visible
04:52.2 that will be the future dorsal side
05:01.1 are then also seen in the eggshell.
07:18.2 that protrude from the egg surface.
10:40.1 because of their mutant phenotypes.
10:50.2 that when we had a mutant germline,
11:15.1 wildtype eggs and embryos resulted.
12:00.1 when the follicle cells are mutant.
12:12.1 some sort of communication going on
12:45.3 but whenever the germline is mutant
12:48.1 they will for an abnormal egg shell
13:28.0 and found out what the proteins are
17:57.2 to move off of its axis of symmetry
18:44.1 or posteriorly-localized molecules.
20:42.2 we have the next stage of signaling
00:18.3 is established during development.
02:29.0 we have to look back in oogenesis.
04:09.0 will form in this part of the egg,
07:03.2 On the top here is a wildtype egg,
07:09.2 You remember it has a dorsal side,
07:32.1 It's a completely ventralized egg.
08:05.2 here on the bottom, follicle cells
10:30.2 by such pole cell transplantations
11:52.0 is required in the follicle cells,
12:08.3 This suggested that there would be
17:49.0 from the posterior follicle cells,
20:17.2 and we, in our lab, and other labs
20:26.3 this posterior signal then results
21:44.2 and so by signaling back and forth
01:24.2 over the past twenty years or so,
03:14.1 there is a little step in between
04:49.2 the oocyte nucleus is closest to,
06:33.1 and then we can test such females
09:13.2 that we use a lot in development,
09:42.2 we can transplant, in Drosophila,
10:12.0 will give rise to mutant oocytes,
10:28.3 at the results that were obtained
10:37.3 They're called torpedo and gurken
10:45.0 Performing such transplantations,
13:48.3 and it can bind ligand molecules.
21:04.3 and the dorso ventral asymmetries
21:09.1 are established during oogenesis.
00:07.1 I'm a Howard Hughes investigator
01:47.2 here on the upper panel is again
01:52.1 and you can see a key molecular,
03:06.2 and they give off daughter cells
03:46.2 and it's within this egg chamber
04:46.2 and whichever side of the oocyte
10:48.1 we found, in the case of gurken,
11:01.2 Conversely, when we transplanted
11:25.2 Whenever the germline is mutant,
12:40.1 the follicle cells don't seem to
13:37.1 encodes the Drosophila homologue
13:46.1 It sits in the membrane of cells
14:25.2 we found that the gurken protein
15:24.2 to the overlying follicle cells.
15:49.2 how this dorso-ventral asymmetry
17:22.1 and gurken signals at this stage
17:30.1 It induces those follicle cells,
19:17.1 which is normally just localized
03:21.0 the daughters of the stem cells
03:58.3 can already be seen in the egg,
06:39.1 whether they lay abnormal eggs,
06:55.2 I'm just showing you a panel of
07:37.2 and when this egg is fertilized
11:27.2 mutant eggs and embryos result.
12:32.2 that forms inside the germline.
15:18.0 to one side of the egg chamber,
15:33.0 will activate the EGF receptor,
16:43.2 After we did some more analysis
17:32.2 at this stage, to be posterior.
17:39.1 then signal back to the oocyte,
18:10.1 then follow the oocyte nucleus,
18:38.0 in situ hybridization pictures,
18:51.2 in oogenesis is the bicoid RNA,
20:36.2 to one side of the egg chamber,
02:06.1 and in this oocyte you can see
03:23.0 will undergo four more mitoses
04:26.1 So, here in this picture here,
07:40.0 the embryo inside will also be
07:49.0 Conversely, here at the bottom
08:22.2 and they form only dorsal skin
15:47.1 So, we can start to understand
17:54.2 that causes the oocyte nucleus
18:12.2 and now gurken signals, again,
00:38.2 of many different cell types.
00:44.3 is Drosophila, the fruit fly.
01:01.1 as we go through development,
02:25.1 where molecules are localized
02:43.2 and as they develop they move
07:13.2 is particularly characterized
07:22.0 They serve as a good landmark
07:51.2 is an egg that is dorsalized.
10:33.0 for two of the mutants, here,
11:21.1 that gurken, the gurken gene,
11:44.1 ventralized eggs and embryos,
11:49.2 Here, it appears that torpedo
14:10.2 is specifically used, though,
14:54.0 It's just found at this stage
00:36.3 that is complex and consists
01:03.0 we end up with an adult fly,
02:03.2 you see an amphibian oocyte,
02:23.1 in these structured manners,
04:34.2 because the oocyte is always
05:59.2 when we have such questions,
06:03.2 and let the organism tell us
06:09.3 What we can do in Drosophila
07:54.3 Here, instead of just having
08:49.0 And, whenever in development
09:01.0 Is it in the follicle cells?
09:45.0 germ cells at an early stage
09:51.2 out of a mutant donor embryo
10:08.2 but the implanted germ cells
10:56.0 we saw the mutant phenotype:
11:18.3 This leads to the conclusion
11:23.2 is required in the germline.
14:31.1 So here, on the upper panel,
14:47.2 you can see accumulates here
14:59.3 close to the oocyte nucleus.
16:25.1 is localized to this corner.
16:38.1 close to the oocyte nucleus.
17:00.1 not only at this later stage
17:36.2 The posterior follicle cells
18:26.0 The fact that gurken signals
18:34.2 is illustrated here in these
19:51.3 we can look at this cartoon.
20:04.3 The posterior follicle cells
20:47.2 to the dorsal follicle cells
02:30.3 This picture here shows you
03:55.0 both the anterior-posterior
05:49.0 of the egg are going to be?
06:15.2 by feeding males a mutagen.
08:39.2 here, both in the eggshell,
09:47.1 from one embryo to another,
11:32.2 that doesn't really matter.
12:50.3 even though they themselves
13:14.0 and help the follicle cells
15:52.1 that arises in the germline
16:57.3 is used twice in oogenesis,
17:13.0 So, in this uppermost panel
17:15.1 you see in the small oocyte
19:08.2 now the bicoid RNA is found
19:27.1 where the posterior end is,
19:34.1 So then, that tells us that
21:50.2 over the time of oogenesis,
01:08.3 and the questions then is,
01:15.1 where different cell types
03:56.2 and the dorso-ventral axis
04:54.2 of the egg and the embryo.
05:54.0 In using a model organism,
07:06.1 again, for your reference.
08:53.2 we have to ask, well then,
11:54.2 because in the transplants
14:45.1 Gurken, on the other hand,
16:31.0 transcribe the gurken RNA,
19:06.2 you can see down here that
19:14.3 Conversely, the oskar RNA,
00:48.1 Here, in the upper panel,
01:05.0 which has a head, a tail,
02:10.1 a key RNA molecule, Vg-1,
08:30.1 However, these phenotypes
09:05.0 Or maybe in both tissues?
10:26.0 We're going to look, now,
13:24.1 after we cloned the genes
13:53.2 In fact, the EGF receptor
00:11.0 at Princeton University.
00:25.2 in developmental biology
00:51.2 you see a fertilized egg
01:13.0 to this complex organism
01:50.0 an oocyte of Drosophila,
02:19.1 If we want to understand
02:21.1 how these eggs are built
02:27.1 to one end or the other,
03:38.1 This cluster of 16 cells
09:03.1 Is it in the egg itself?
09:20.1 In this particular case,
11:35.0 Now, very interestingly,
12:24.0 in the wildtype germline
16:17.1 Well, the gurken protein
16:46.0 of the mutant phenotypes
20:24.1 We do know, though, that
00:23.1 One of the big problems
05:16.1 What this means is that
09:27.1 and the follicle cells,
12:38.1 With respect to gurken,
12:53.1 have a normal genotype.
16:51.1 we learned that in fact
21:14.2 makes these asymmetries
01:07.1 it has wings and legs,
03:16.2 the stem cell division
04:28.0 you see an egg chamber
04:32.3 and dorso-ventral axis
04:39.2 and the oocyte nucleus
09:36.2 Here, in this example,
15:45.1 of the gurken protein.
16:22.0 because the gurken RNA
17:06.0 to the follicle cells,
18:32.1 within the egg chamber
19:54.0 So, early in oogenesis
19:56.0 we have a first signal
21:11.3 The germline initially
00:34.3 to an adult organism,
00:59.1 Later in development,
01:40.3 informative molecules
02:52.0 If we just look, now,
15:28.3 These follicle cells,
17:24.1 to the follicle cells
19:38.3 early on to establish
00:05.2 I'm Trudi Schupbach.
01:23.0 It has become clear,
02:39.3 that are developing.
06:27.1 until we get females
06:53.0 In this slide, here,
01:28.2 unstructured cells,
02:01.2 On the lower panel,
09:29.1 we can make mosaics
16:28.1 So, the nurse cells
03:52.2 So, in Drosophila,
04:11.0 and the tail here,
04:59.2 in the egg chamber
05:21.2 prior information,
10:01.1 it will be itself,
12:55.2 That then suggests
13:34.3 Torpedo, we found,
14:23.2 of these two genes
15:39.0 These target genes
19:49.1 To summarize this,
20:32.3 within the oocyte,
20:49.2 and induces, thus,
21:31.2 The follicle cells
01:45.3 So, for instance,
02:46.1 to the posterior,
04:04.0 So, we know that,
14:16.1 during oogenesis.
15:20.0 uses gurken, now,
20:56.1 In summary, then,
21:07.1 of egg and embryo
08:33.1 posed a problem,
14:19.2 Very excitingly,
20:02.1 to be posterior.
03:12.1 Now, in insects
08:59.2 in development?
10:06.1 meaning normal,
01:54.1 the oskar RNA,
19:31.2 of the oocyte.
02:56.2 an ovariole,
16:10.2 how is it...
01:33.2 within eggs
19:04.3 is blocked,

Part 2: grk RNA Localization

00:00:04.20 I'm Trudi Schupbach.
00:00:06.18 I'm a Howard Hughes Investigator
00:00:09.13 and a professor of Molecular Biology
00:00:11.05 at Princeton University.
00:00:13.14 In the first part of my talk,
00:00:15.15 I have told you about
00:00:19.18 establishment of pattern and polarity
00:00:21.25 during Drosophila oogenesis.
00:00:24.17 I've also told you that there is a key molecule,
00:00:26.29 gurken, that gets localized during oogenesis,
00:00:30.02 and in the second part of my talk
00:00:32.13 I would like to introduce you to some of the experiments
00:00:36.01 that we are doing in order to understand
00:00:38.18 this gurken localization process in more detail.
00:00:43.14 Just to remind you, however,
00:00:45.13 I'll show you again the picture of a Drosophila ovary.
00:00:49.26 At the top, you have the youngest egg chambers.
00:00:52.11 They develop continuously
00:00:54.28 down to the older stages.
00:00:59.12 Pattern and polarity can already be seen
00:01:01.17 at later stages of oogenesis in the egg chambers,
00:01:05.15 and in particular we will concentrate here
00:01:07.28 on the dorsoventral asymmetry,
00:01:10.03 and that is predictable at this stage
00:01:12.09 because the oocyte nucleus is localized
00:01:15.01 to one particular corner of the developing egg chamber,
00:01:19.02 and whichever side of the egg chamber it's closest too,
00:01:22.17 that's going to be the future dorsal side
00:01:24.24 of both the egg and the embryo.
00:01:29.11 The gurken molecule,
00:01:31.29 which encodes a signaling molecule,
00:01:35.06 is produced by the germline
00:01:38.19 and accumulates in close vicinity
00:01:41.29 to the oocyte nucleus.
00:01:44.06 You can see here the gurken RNA
00:01:46.09 surrounding the oocyte nucleus,
00:01:48.08 and this RNA then gives rise to the protein,
00:01:51.18 the gurken protein,
00:01:53.18 which is a transmembrane protein
00:01:55.12 and accumulates to a higher level in the oocyte membrane,
00:01:57.20 but, again, to a very restricted part
00:02:00.06 of the oocyte membrane,
00:02:02.18 just in the dorsal anterior corner.
00:02:06.04 The gurken molecule
00:02:08.18 encodes a signaling molecule of the TGFɑ-like
00:02:12.07 family of growth factors.
00:02:15.16 It can be released from the oocyte
00:02:19.12 and then binds and activates
00:02:22.09 the Drosophila EGF receptor,
00:02:24.22 which is a large transmembrane receptor
00:02:27.16 that sits in the membrane of the follicle cells,
00:02:31.02 and upon receiving (binding) the gurken signal,
00:02:33.29 the EGF receptor will activate
00:02:36.15 its internal tyrosine kinase domain
00:02:39.06 and that leads, eventually,
00:02:42.18 to the activation and repression
00:02:45.00 of subsets of genes in the follicle genes,
00:02:48.12 and induces the follicle cells
00:02:50.20 to become dorsal follicle cells.
00:02:55.18 In order to understand how this gurken RNA molecule
00:02:59.13 gets localized to this particular place
00:03:01.15 within the oocyte,
00:03:04.09 we used again the mutagenesis approach
00:03:07.06 because that allows us to see, without bias,
00:03:12.10 what type of genes might be involved
00:03:15.11 in getting the RNA to be accumulating
00:03:18.07 just in this one particular place during oogenesis.
00:03:22.11 So, in mutagenesis experiments
00:03:24.12 that either we or other people did,
00:03:28.05 we found a number of genes in the Drosophila genome
00:03:31.06 that, when mutant, resulted in a mislocalization
00:03:34.24 of the gurken RNA.
00:03:36.29 In particular, there is a group of genes
00:03:40.10 that has the following phenotype:
00:03:42.23 in these genes, the gurken RNA,
00:03:44.29 rather than restricted to just one side of the egg chamber,
00:03:48.28 is found all around the anterior circumference of the oocyte.
00:03:53.17 You might appreciate this better
00:03:56.02 in a cross-section shown in this picture here.
00:03:59.12 You can see, in wildtype,
00:04:01.19 the gurken RNA is just localized
00:04:04.19 sort of like a cap above the oocyte nucleus
00:04:07.04 on the dorsal side of the egg chamber.
00:04:10.10 In contrast, in the mutants here,
00:04:12.22 you find the RNA all around the anterior circumference.
00:04:18.13 Therefore, the genes that are mutated
00:04:22.29 in these various mutant lines
00:04:26.00 must be involved, somehow, in getting the RNA
00:04:29.14 to be just localized to one side of the egg chamber,
00:04:32.29 and so of course we're interested in
00:04:36.06 what these genes encode.
00:04:38.24 The ultimate result of this mislocalization
00:04:42.28 is a dorsalized egg and embryo.
00:04:46.19 This is because when the RNA
00:04:49.11 is not localized to just one side of the egg chamber,
00:04:52.13 but is found all around the anterior circumference,
00:04:55.28 then we also see the gurken protein
00:04:58.24 all around the anterior circumference,
00:05:01.16 and now follicle cells from all sides,
00:05:04.08 both the dorsal and the ventral sides,
00:05:06.29 will see high levels of this gurken protein,
00:05:09.21 think that they are dorsal,
00:05:12.11 and start migrating and producing dorsal appendage material.
00:05:20.05 The first gene in this class
00:05:22.14 that we looked at in more detail
00:05:24.26 is the gene squid.
00:05:27.13 Squid was originally cloned
00:05:30.04 by two different laboratories simultaneously.
00:05:33.14 It encodes a protein that can bind RNAs,
00:05:38.03 and it can also shuttle in and out of nuclei.
00:05:41.19 Amanda Norvell, a former postdoc in the lab,
00:05:45.15 looked at squid in more detail and she found that
00:05:49.09 squid can indeed bind the gurken RNA directly.
00:05:53.17 She also discovered that
00:05:56.23 squid is not only required for the localization of the RNA,
00:05:59.24 but it's also required for the translation repression
00:06:04.01 of the RNA before it's localized.
00:06:07.14 Many localized RNAs are initially translationally repressed.
00:06:13.20 The cell will do this because
00:06:16.26 the whole goal of RNA localization is to get the RNA
00:06:19.27 to the place where the protein should be localized,
00:06:23.19 and so the RNA should not already be translated
00:06:27.23 during its journey to that location,
00:06:31.15 and for that reason, very often,
00:06:33.29 these localized RNAs are associated with
00:06:37.24 repressive complexes that keep them from being translated
00:06:41.02 while they're still being transported.
00:06:43.16 This seems also to be true for the gurken RNA,
00:06:46.26 and squid is an important factor
00:06:50.12 in organizing such a translational repression.
00:06:54.12 Now, squid is of course not doing everything by itself,
00:06:58.10 localizing and translational repression,
00:07:01.15 and so we wanted to see what other proteins
00:07:04.21 would associate with squid and help
00:07:07.11 in the function of gurken localization
00:07:10.04 and translational regulation.
00:07:12.17 Two graduate students in the lab,
00:07:14.26 Jennifer Goodrich and Nicole Clouse,
00:07:17.07 undertook some screens to find proteins
00:07:20.05 that would interact with squid.
00:07:22.09 They used different methods.
00:07:24.06 One method is to go to yeast
00:07:26.28 and perform what we call a yeast two-hybrid screen,
00:07:32.00 where we transform Drosophila proteins into yeast,
00:07:36.12 and the system is set up in a way that
00:07:38.23 if two of these Drosophila proteins bind to each other
00:07:42.14 the yeast will be able to produce a blue color.
00:07:46.02 So, in this way,
00:07:48.14 Jennifer and Nicole found this gene, Hrb27C,
00:07:53.06 it's also sometimes called Hrp48,
00:07:57.02 which can associate with squid.
00:07:59.19 They also verified this by doing more directed
00:08:03.10 physical interaction tests,
00:08:06.08 for instance using IP methods
00:08:09.20 to immunoprecipitate with the help of squid,
00:08:14.06 the complex,
00:08:16.03 and then testing whether Hrb27C could also be seen
00:08:18.25 in this complex.
00:08:22.03 Now, showing that two proteins interact physically
00:08:25.26 doesn't necessarily tell us
00:08:28.24 whether they really also functionally
00:08:31.23 act in the same process,
00:08:34.11 and so Jennifer Goodrich tested mutations in Hrb27C
00:08:39.05 to see whether they really also affected gurken localization.
00:08:43.17 This wasn't so straight forward in this case,
00:08:47.16 because mutations in this gene are actually lethal.
00:08:52.05 That means we can't just simply take homozygous females
00:08:54.12 and check their eggs,
00:08:57.11 but we have to use a more complicated way
00:09:00.14 of making mosaics
00:09:03.21 where the females in general
00:09:06.22 are heterozygous for the mutation,
00:09:09.16 and therefore survive,
00:09:11.22 but then using mitotic recombination
00:09:13.28 we can create homozygous germ cells
00:09:16.15 and then test what the effect is
00:09:18.29 of losing this gene now in the germline.
00:09:22.15 Performing such a test, Jennifer found that,
00:09:25.22 indeed, the eggs produced by such germline clones
00:09:30.02 were also dorsalized,
00:09:32.19 and importantly the gurken RNA
00:09:35.04 was not localized correctly in these mutant egg chambers.
00:09:41.12 Hrb27C has a number of other phenotypes.
00:09:45.21 You can see these eggs are smaller...
00:09:48.06 we also see abnormalities in the nurse cells,
00:09:51.02 so this gene has multiple functions during oogenesis,
00:09:55.05 but, importantly,
00:09:58.02 it does appear to create the same mutant phenotype
00:10:01.06 as mutations in squid,
00:10:03.20 and therefore seems indeed to interact with squid
00:10:06.15 in this particular process of gurken RNA localization
00:10:10.03 and translational control.
00:10:13.07 Performing more such screens and tests,
00:10:16.16 we came up with the following
00:10:19.21 cartoon interpretation of our findings.
00:10:24.02 The gurken RNA is transcribed in the nucleus
00:10:27.23 and, presumably, in the nucleus
00:10:30.28 it will already interact with genes like squid and Hrb27C,
00:10:36.26 which accumulate in the nucleus
00:10:39.24 and can bind to RNAs
00:10:42.08 as they are being transcribed and spliced.
00:10:46.20 These factors then seem to leave the nucleus
00:10:50.07 with the RNA,
00:10:53.01 and in the cytoplasm it appears that
00:10:55.24 squid and Hrb27C
00:10:59.08 can organize a complex on the gurken RNA
00:11:03.04 that allows the RNA to move,
00:11:06.19 to associate with the localization machinery,
00:11:10.08 but also to keep the RNA translationally repressed.
00:11:15.07 Once the RNA has reached its destination,
00:11:21.01 there seems to be some sort of anchoring mechanism
00:11:24.24 that anchors the localized RNA in this place,
00:11:28.29 and this also results, somehow,
00:11:31.24 in a reorganization of the protein complexes
00:11:35.02 that are associated with gurken,
00:11:37.20 so that the translational repressors leave the RNA
00:11:42.09 and open it up for translation,
00:11:45.00 so that now, in this localized state,
00:11:48.18 the gurken protein can be produced.
00:11:53.25 Now,
00:11:58.07 we have of course many questions here.
00:12:00.22 For instance, what is this anchor that anchors
00:12:04.26 the gurken protein in this dorsal anterior corner?
00:12:08.02 We have a candidate protein
00:12:10.04 that we think might be involved in this anchoring
00:12:12.29 and that is the protein encoded by a gene called Encore.
00:12:17.28 Encore is a gene
00:12:20.19 that also has multiple functions during oogenesis,
00:12:23.14 but at this particular stage in oogenesis
00:12:26.20 it seems to be required
00:12:29.17 for high levels of gurken protein accumulation.
00:12:36.16 In the Encore mutants,
00:12:38.27 you can see the gurken RNA is somewhat localized,
00:12:43.11 there's more RNA on the dorsal side
00:12:46.10 than on the ventral side,
00:12:48.11 but it's not very tightly localized
00:12:50.24 to the dorsal anterior corner.
00:12:53.05 And then, very dramatically,
00:12:55.07 the amount of gurken protein that we see
00:12:57.20 at this stage of oogenesis
00:12:59.28 is very much reduced.
00:13:02.03 Because there is not much gurken protein in these eggs,
00:13:05.16 this results then in a ventralized egg
00:13:08.06 and embryo phenotype,
00:13:10.19 so that instead of having two populations
00:13:13.05 of dorsal follicle cells
00:13:15.21 that produce dorsal appendages,
00:13:18.02 we lose the dorsal midline
00:13:20.24 and we have only a reduced population,
00:13:23.19 a single population of follicle cells that will migrate
00:13:26.06 and form dorsal appendages,
00:13:28.07 and the embryos that develop inside these eggs
00:13:30.15 are also ventralized.
00:13:33.15 When Cheryl Van Buskirk,
00:13:35.20 another graduate student in the lab,
00:13:37.27 made an antibody to the Encore protein,
00:13:41.23 we found that this protein also accumulates
00:13:45.13 in the dorsal anterior corner to higher levels.
00:13:49.01 You can see here in the cross-section,
00:13:51.13 Encore protein around the oocyte nucleus
00:13:54.05 at higher levels than you find it on the ventral side.
00:13:57.27 The Encore protein is not quite as tightly localized
00:14:01.17 as, say, the gurken RNA
00:14:03.23 which is shown here for your reference in this panel,
00:14:06.25 but nevertheless it appears that Encore
00:14:09.13 shows an asymmetry that makes it a candidate
00:14:12.03 for being such an anchor of the gurken RNA.
00:14:16.05 Another finding about Encore
00:14:18.24 is that it interacts with a polyA-binding protein.
00:14:22.21 PolyA-binding proteins
00:14:25.07 are usually required to upregulate translation in cells,
00:14:31.19 and we've found that Encore
00:14:34.19 seems to directly interact
00:14:37.09 with such a polyA-binding protein.
00:14:40.24 Putting all these results together, then,
00:14:43.06 into a scheme,
00:14:45.19 we now think that once the gurken RNA
00:14:48.09 reaches its destination in the dorsal anterior corner
00:14:51.08 Encore might help to anchor the gurken RNA,
00:14:55.19 and it may also be involved in upregulating
00:14:58.13 translation of gurken at this stage,
00:15:02.01 through its interaction with a polyA-binding protein.
00:15:06.26 Of course, there are still many more questions
00:15:09.22 that we would like to answer.
00:15:12.24 We would like to understand how the gurken RNA complex
00:15:16.01 is actually moved through the cytoplasm.
00:15:19.28 This cartoon, here,
00:15:22.05 was obtained by either doing genetic experiments
00:15:25.10 or by checking for direct physical interactions,
00:15:30.16 but a lot of it is based on interpretation of these results,
00:15:34.23 and what we'd really like to do in the long run
00:15:38.11 is to visualize the RNA and the proteins
00:15:41.25 in real-life oocytes and ask,
00:15:44.23 are these proteins really all together,
00:15:47.26 sitting on the RNA as it's moving?
00:15:50.17 Who comes on when?
00:15:52.09 Who leaves the complex where?
00:15:54.13 And, what happens in a more dynamic
00:15:56.22 and biophysically described manner?
00:16:00.22 As a first step in this direction,
00:16:03.23 we have started a collaboration
00:16:06.18 with my colleague Elizabeth Gavis
00:16:09.01 in our department at Princeton University,
00:16:11.25 to visualize the gurken RNA localization process live.
00:16:18.08 The methods that we're using here
00:16:21.11 is to try to label the gurken RNA
00:16:23.26 so that we can see it live in the microscope.
00:16:28.14 The methodology for such
00:16:31.14 messenger RNA labelings
00:16:34.00 have initially been worked out by Rob Singer and his colleagues
00:16:37.04 in yeast,
00:16:39.01 and then have been adapted for Drosophila by Liz Gavis
00:16:41.23 and her students.
00:16:44.07 The system consists of two parts.
00:16:47.27 For one, one takes the RNA one wants to study,
00:16:52.12 gurken in our case,
00:16:55.02 and inserts some stem loops into this RNA
00:16:58.01 in a place where it doesn't interfere with the function of the RNA.
00:17:02.13 These stem loops are derived from an RNA virus,
00:17:07.22 and there is a viral coat protein
00:17:10.20 that has a very high binding affinity
00:17:13.07 to these stem loops.
00:17:15.13 So then, the second part of the labeling system
00:17:18.14 involves the viral coat protein,
00:17:23.19 now fused to GFP,
00:17:25.29 which is coexpressed with the marked RNA,
00:17:30.24 and then when both of these parts are together
00:17:33.16 in the same oocyte
00:17:36.19 the GFP-marked coat protein
00:17:39.12 will bind to the stem loops
00:17:42.21 and thus mark us the live RNA,
00:17:45.27 so that we can see on the microscope where it is.
00:17:48.23 Angela Jaramillo, a postdoc in the lab,
00:17:52.16 used this system and made the gurken constructs
00:17:56.19 and was able to see, indeed,
00:18:00.21 that we could visualize the gurken mRNA
00:18:04.18 accumulating either at the posterior end of the egg chamber,
00:18:08.00 or at the dorsal anterior corner later in oogenesis,
00:18:11.29 in live imaging.
00:18:16.18 Angela then used this system
00:18:19.27 to ask a question about the dynamics
00:18:23.04 of the gurken localization process,
00:18:25.28 and for this she used the technique called
00:18:28.24 fluorescence recovery after photobleaching.
00:18:31.27 What this involves is to take a laser
00:18:36.04 and make a small...
00:18:40.04 illuminate a small portion of the localized RNA
00:18:45.10 and make those molecules, there,
00:18:47.15 that are in the laser beam, dark.
00:18:50.26 Now, if the molecules are stably anchored,
00:18:55.12 they will just stay there,
00:18:57.23 and therefore there will be a dark hole in this spot,
00:19:00.23 persisting.
00:19:02.19 On the other hand,
00:19:04.29 if the RNA molecules are not stably anchored in this position,
00:19:08.12 but are constantly exchanged
00:19:10.23 with new molecules coming from they cytoplasm,
00:19:14.00 then this initial dark spot will over time
00:19:17.10 recover fluorescence.
00:19:19.17 And so, that's what Angela did,
00:19:22.02 and then she measured what would happen
00:19:24.28 after she used the laser to bleach the molecules
00:19:28.04 in particular spots.
00:19:30.09 So, I'm going to show you two movies.
00:19:32.20 The first movie, here,
00:19:35.22 was taken at an early stage in oogenesis
00:19:39.02 when the gurken RNA is localized to the posterior pole,
00:19:43.05 and you will then see,
00:19:45.13 for a short amount of time,
00:19:48.00 as you could just see now,
00:19:50.23 a dark spot where the RNA was bleached,
00:19:54.07 but you also probably notice that very quickly
00:19:58.16 this spot was filled in
00:20:01.13 and the fluorescence was recovered.
00:20:07.19 We will now compare this to the second movie,
00:20:12.22 which was taken at a later stage,
00:20:14.29 stage 9 of oogenesis,
00:20:17.21 and if we run the movie now
00:20:20.00 you'll see, again, the dark spot appearing here,
00:20:22.25 but this time it's not being filled in very rapidly.
00:20:26.16 It remains there
00:20:29.02 during the duration of the time that we took this movie,
00:20:32.19 about half an hour,
00:20:34.22 and was not filled it,
00:20:37.01 telling us, therefore, that at this stage
00:20:39.20 the molecules are much less dynamic than in the early stage.
00:20:44.01 Angela of course measured the fluorescence
00:20:49.10 very carefully,
00:20:52.27 so that she could show that,
00:20:56.07 after the initial drop in fluorescence, fluorescence gets recovered,
00:20:59.28 and that this recovery changes over time.
00:21:03.02 So, in summary... when she summarized all her results
00:21:07.06 she found that over time there is a decrease
00:21:10.20 in the recovery of fluorescence
00:21:14.01 from young to old stages,
00:21:16.17 and in particular there is also a big jump in recovery time
00:21:19.24 from the younger to the older stages.
00:21:23.12 All of this tells us that
00:21:26.15 the gurken RNA seems to be localized
00:21:29.04 at the two different stages by two different mechanisms.
00:21:32.19 Early, at the posterior end,
00:21:35.13 the gurken RNA seems to be very dynamic.
00:21:39.08 This recovery that we saw,
00:21:41.26 this rapid recovery of fluorescence
00:21:44.17 is consistent with a continuous exchange
00:21:48.20 of new molecules coming in from the cytoplasm,
00:21:52.18 and it might therefore reflect constant transport.
00:21:56.22 In contrast, at the later stage
00:21:59.12 the gurken RNA seemed much more static.
00:22:02.01 It was not able to recover very quickly
00:22:04.22 or very well at all.
00:22:07.12 Therefore, we would conclude that
00:22:11.03 at this later stage there is an additional mechanism
00:22:13.18 that really anchors the RNA.
00:22:15.22 And, remember now,
00:22:17.26 in the beginning of my talk I said that we require these molecules...
00:22:21.02 squid, Hrb27C, Encore...
00:22:24.17 mostly at this stage of oogenesis.
00:22:27.10 So, they seem then to be particularly important
00:22:30.19 to get the RNA to this anchored stage,
00:22:34.08 which after all lasts for a long time.
00:22:37.11 The RNA needs to be present
00:22:41.13 over quite a long time in oogenesis.

Part 3: Gurken Gradient and Follicle Cell Response

00:00:03.23 I'm Trudi Schupbach.
00:00:05.14 I'm a Howard Hughes Investigator,
00:00:07.05 and a professor of Molecular Biology
00:00:09.02 at Princeton University.
00:00:11.00 In my previous segments of my talk,
00:00:13.29 I have introduced you to
00:00:16.13 the question of pattern and polarity
00:00:18.13 during Drosophila oogenesis.
00:00:20.19 I've also told you that
00:00:23.26 there is a key signaling process
00:00:26.01 that takes place during oogenesis
00:00:28.16 whereby a signaling molecule
00:00:30.18 encoded by the gene gurken
00:00:32.29 signals to the Drosophila EGF receptor,
00:00:35.18 which is expressed in the follicle cells,
00:00:38.03 and induces dorsoventral pattern formation
00:00:42.18 in the follicle cells.
00:00:46.04 The localization of the gurken protein
00:00:49.17 depends, as I've told you in the second part of my talk,
00:00:53.00 on the localization of the gurken RNA.
00:00:56.10 Once the RNA is localized
00:00:58.05 and the protein is made,
00:01:00.14 it is then released from the oocyte
00:01:03.09 and diffuses through the space
00:01:05.26 between the oocyte and the overlying follicle cells,
00:01:09.10 binds to the EGF receptor,
00:01:11.19 and starts this pattern formation process.
00:01:17.12 In the third part of my talk,
00:01:19.25 I would actually like to start telling you about
00:01:23.29 some of the experiments we are doing
00:01:26.09 to try to understand what the response process is
00:01:29.16 that goes on in the follicle cells,
00:01:32.03 and also how the gurken gradient
00:01:34.19 actually shapes this response in the follicle cells.
00:01:39.27 Just to recap what I've already told you...
00:01:45.19 so, the gurken RNA is transcribed during oogenesis
00:01:49.05 and then there's a set of proteins
00:01:52.04 that are involved in the localization of the RNA
00:01:55.05 to the dorsal anterior corner.
00:01:57.04 There is also translational regulation of the gurken RNA,
00:02:00.06 and once it's localized
00:02:03.15 the protein is made and then secreted by the oocyte,
00:02:06.19 binds to the EGF receptor in the follicle cells,
00:02:09.24 and this binding activates the receptor
00:02:13.12 and activates the downstream signaling process
00:02:17.04 that then results in the establishment of dorsal follicle cell fate.
00:02:22.27 One of the outcomes of this activity of the EGF receptor
00:02:27.00 in the dorsal follicle cells
00:02:29.19 is to repress a new signal
00:02:33.08 that is created by the follicle cells
00:02:37.10 and that is transmitted back to the egg
00:02:40.28 and then serves to establish the dorsoventral pattern
00:02:44.11 of the embryo.
00:02:47.06 So, ultimately,
00:02:49.20 what starts in oogenesis with the movement of the oocyte nucleus,
00:02:52.27 ends during embryogenesis
00:02:55.25 with the formation of a gradient of a protein in the embryo,
00:03:02.03 a protein that is called dorsal.
00:03:06.02 The dorsal protein accumulates
00:03:08.14 to very high levels
00:03:11.07 in the nuclei on the ventral side of the embryo,
00:03:14.14 and it is mostly cytoplasmic
00:03:17.07 and not in the nucleus on the dorsal side of the embryo.
00:03:20.25 Dorsal encodes a transcriptional regulator
00:03:23.26 and, once it's in the nuclei
00:03:27.04 of the ventral embryonic cells,
00:03:29.27 it regulates transcription in these cells
00:03:32.16 and thus establishes ventral cell fates.
00:03:38.00 Now, during my talk before,
00:03:40.20 I've told you that during oogenesis
00:03:43.14 we have a signaling process, gurken/EGF receptor,
00:03:45.28 going on on the dorsal side of the egg chamber.
00:03:51.00 So, how does this all hang together?
00:03:53.15 So, during oogenesis we have gurken
00:03:56.04 localized to the dorsal side.
00:03:58.08 It activates the EGF receptor on the dorsal side,
00:04:01.23 and that then restricts
00:04:05.08 a new signaling process to the ventral side
00:04:07.25 of the follicle cell epithelium.
00:04:10.22 This new signaling process
00:04:13.18 requires the activity of a gene
00:04:15.27 called pipe,
00:04:18.10 and the pipe gene starts this new ventral signal
00:04:22.19 by activating a protein called spaetzle,
00:04:26.13 which can bind to a receptor that's present
00:04:29.21 on the egg plasma membrane,
00:04:33.28 and once this receptor is now activated
00:04:36.09 on the ventral side
00:04:38.11 it leads to the import of this dorsal transcription factor
00:04:43.01 into the nuclei on the ventral side,
00:04:46.11 and the formation of this nuclear gradient
00:04:49.27 of the dorsal protein.
00:04:52.14 So, the effect of gurken/EGF receptor signaling
00:04:56.00 on pipe expression
00:04:59.02 is shown here in this slide,
00:05:02.04 taken by the laboratory of David Stein,
00:05:04.28 where he showed here in the wildtype
00:05:09.12 that pipe expression is seen on the ventral side
00:05:12.12 of the egg chamber,
00:05:15.03 whereas in a gurken mutant...
00:05:16.26 so when gurken doesn't signaling the EGF receptor is not activated,
00:05:20.00 and now pipe is expressed
00:05:22.28 all throughout the egg chamber.
00:05:25.06 So, pipe is then a target gene of the EGF receptor,
00:05:28.29 and is repressed by EGF receptor activity.
00:05:38.21 So, what we are then interested in...
00:05:42.28 in understanding is,
00:05:45.16 how does the EGF receptor
00:05:47.26 affect the expression of pipe?
00:05:50.13 Is this is a direct repression
00:05:53.08 in response to EGF receptor activity,
00:05:55.21 or are there many steps in between?
00:05:58.16 Are the other signaling processes
00:06:01.06 that intervene and that eventually
00:06:04.08 regulate the expression of pipe
00:06:07.03 and restrict it to the ventral side of the egg chamber?
00:06:11.08 In order to get at these questions,
00:06:14.11 we have again undertaken some mutagenesis experiments
00:06:17.04 to find genes
00:06:20.04 that would alter EGF receptor signaling
00:06:24.21 or its outcome in the embryo
00:06:27.11 to try to understand how this process works
00:06:30.26 at the molecular level.
00:06:33.28 In these mutagenesis screens,
00:06:37.20 we used a technique
00:06:40.12 that allows us to make mosaic egg chambers.
00:06:45.17 We suspect that many of the genes
00:06:48.05 that might intervene here
00:06:49.24 downstream of EGF receptor signaling
00:06:52.06 might be lethal for the homozygous carrier,
00:06:55.15 and for that reason we can't just simply
00:06:57.15 do mutagenesis and look at homozygous mutant flies,
00:07:02.05 but we have to work with heterozygous flies
00:07:04.17 that can survive,
00:07:07.00 and then induce homozygous mutant follicle cells
00:07:10.00 inside these heterozygous flies.
00:07:14.12 The way we can do this in flies
00:07:17.13 is by using a system
00:07:20.07 that allows us to generate mitotic recombination.
00:07:23.21 So, normally, during mitosis, of course,
00:07:26.16 there's no recombination,
00:07:29.00 but by introducing homologous recombination enzymes
00:07:33.29 and target sites from yeast,
00:07:36.11 fly biologists have engineered chromosomes
00:07:39.18 that can now be used to make such
00:07:43.08 homozygous mutant cells
00:07:45.21 within a heterozygous carrier fly.
00:07:49.22 Initially, this happens at mitosis.
00:07:52.08 A single cell will be homozygous mutant,
00:07:55.28 but then as the cell divides
00:07:58.21 it will form a patch or a clone of cells
00:08:01.17 that are all homozygous mutant.
00:08:04.08 What this looks like is illustrated here in this slide,
00:08:07.10 where you can see we made
00:08:10.00 a patch of mutant follicle cells.
00:08:14.11 These follicle cells were mutant
00:08:17.01 for a mutation that makes a very thin egg shell,
00:08:20.08 so you can see here, when we use the technique,
00:08:23.29 we create a mosaic egg shell
00:08:26.04 where we just have a large path of mutant cells
00:08:29.20 surrounded and embedded in
00:08:33.11 normal wildtype follicle cells.
00:08:38.00 A postdoc in the lay, Li-Mei Pai,
00:08:41.15 performed such a screen and she found a number of mutations
00:08:44.08 that showed interesting phenotypes.
00:08:46.27 These phenotypes, here,
00:08:50.29 can be classified as dorsalized egg shells,
00:08:54.10 and the embryos that developed inside these eggs
00:08:56.25 were also partially dorsalized.
00:09:00.16 Li-Mei then analyzed these mutations
00:09:04.10 at the molecular level
00:09:06.08 and she found that, for instance,
00:09:08.13 the one here shown on the top,
00:09:10.23 is a mutation in a Drosophila gene called Cbl,
00:09:14.18 which is a conserved gene
00:09:17.08 that's also found in vertebrates.
00:09:19.22 It has a C. elegans homologue
00:09:22.15 and then two different protein isoforms
00:09:25.10 are made from this gene in flies
00:09:28.11 by differential splicing.
00:09:31.07 What does the Cbl gene do?
00:09:33.28 Well, about ten years ago
00:09:37.01 there were a series of biologists
00:09:40.10 that investigated this Cbl gene,
00:09:42.24 because it is actually an important oncogene
00:09:45.15 that can cause...
00:09:48.14 be implicated in cancer formation in mutants in humans,
00:09:51.29 and they found that the Cbl genes
00:09:55.26 can act to downregulate the activity
00:09:59.17 of major receptor tyrosine kinases,
00:10:03.01 such as EGF receptor.
00:10:05.22 So, normally,
00:10:07.13 when these receptor tyrosine kinases are activated,
00:10:09.27 because they bind their ligands,
00:10:13.04 they cross-phosphorylate
00:10:16.10 and once they are phosphorylated
00:10:19.06 they are active
00:10:21.05 and now can also phosphorylate target proteins
00:10:24.15 inside the cell.
00:10:27.16 What Cbl does is it has a
00:10:30.13 protein phosphotyrosine-binding domain
00:10:33.05 and it binds to these activated tyrosines,
00:10:36.18 phosphorylated tyrosines in the receptor
00:10:39.16 and then acts as a ubiquitin ligase
00:10:42.05 to get the receptor marked with ubiquitin.
00:10:46.13 This serves both as a targeting signal
00:10:49.09 to downregulate the receptor,
00:10:51.01 and also a degradation signal
00:10:53.18 to get the receptor degraded.
00:10:56.10 So, the Cbl proteins have therefore
00:10:59.11 an important function
00:11:02.05 in limiting the activity of the receptor.
00:11:04.27 So, that makes, now,
00:11:05.10 sense with respect to the phenotypes
00:11:08.14 that Li-Mei saw in these mutant egg chambers,
00:11:12.28 in fact we now can understand that,
00:11:16.02 normally, the EGF receptor has to be active
00:11:18.27 to make us a dorsal side,
00:11:21.02 but then if it's overly active
00:11:23.19 or stays active too long,
00:11:26.14 this will result in dorsalized eggs and embryos.
00:11:32.25 Now, we wanted to really understand better, however,
00:11:37.04 whether Cbl is really only required on the dorsal side
00:11:39.23 to downregulate the receptor
00:11:42.01 or whether it has any effects on the ventral side,
00:11:44.25 and how, in general,
00:11:47.12 it shapes the activity of the receptor.
00:11:51.09 For this we used some target genes
00:11:54.15 of the EGF receptor
00:11:56.14 to see how they could be misregulated
00:11:58.28 by the loss of Cbl function.
00:12:01.17 Kekkon is one of such EGF receptor response genes.
00:12:06.04 It usually is activated by receptor activity,
00:12:10.20 and therefore at this stage of oogenesis
00:12:13.01 you can see that there is a high level of
00:12:16.00 Kekkon expression on the dorsal side of the egg chamber
00:12:19.28 in response to EGF receptor activity.
00:12:23.24 Now, when Li-Mei made clones,
00:12:28.09 mutant clones that had lost Cbl activity
00:12:33.00 on the ventral side of the egg chamber,
00:12:35.29 and they're shown here in black,
00:12:38.10 so the wildtype cells in this egg chamber
00:12:41.07 are labeled in red, but as they became mutant
00:12:44.09 they lost the red marker and that tells us that
00:12:47.14 they are homozygous mutant for Cbl.
00:12:51.01 But interestingly, in these ventral cells
00:12:53.24 that are mutant for Cbl,
00:12:56.06 you can now see high Kekkon expression
00:12:58.08 on the ventral side, where it should normally
00:13:00.17 not be expressed.
00:13:03.12 That tells us that Cbl must be acting,
00:13:06.08 in general, to keep the EGF receptor activity
00:13:10.25 low on the ventral side,
00:13:13.04 and that if Cbl doesn't act,
00:13:15.23 EGF receptor activity
00:13:18.16 gets high on the ventral side
00:13:20.13 and activates a dorsal response gene.
00:13:23.20 Interestingly, also, if you look very carefully,
00:13:26.17 you can see that the expression of Kekkon,
00:13:30.11 the misexpression of Kekkon,
00:13:32.19 is very cell autonomous:
00:13:34.19 it just happens in the mutant cells,
00:13:37.03 not in the neighboring cells.
00:13:39.27 That tells us, then, that
00:13:43.01 the activation of Kekkon by EGF receptor
00:13:47.20 is very direct.
00:13:50.02 It happens...
00:13:51.25 each cell, in a way,
00:13:54.03 looks at its own EGF receptor activity
00:13:57.25 and decides whether or not to turn the response gene,
00:14:02.13 Kekkon, on or not.
00:14:04.03 It doesn't look at its neighboring cells,
00:14:06.03 it just measures, somehow,
00:14:08.07 the level of EGF receptor activity
00:14:10.09 inside its own cytoplasm,
00:14:12.06 and then turns on the response gene
00:14:15.21 cell autonomously.
00:14:17.28 We can look at another response gene, pipe,
00:14:20.09 and we find the same situation there as well.
00:14:23.18 Now, pipe is repressed by EGF receptor activity,
00:14:28.21 and you can see in wildtype,
00:14:30.20 and this egg is a little bit tilted
00:14:32.15 so we're looking mostly at the ventral side,
00:14:34.08 but you can see it has a
00:14:35.29 sharp on/off expression pattern.
00:14:37.29 It's off on the dorsal side and it's on on the ventral side.
00:14:41.09 And, if you make a clone of mutant cells for Cbl,
00:14:44.26 here on the ventral side,
00:14:47.06 pipe gets repressed,
00:14:49.05 again telling us that Cbl
00:14:53.01 normally has to keep EGF receptor activity
00:14:55.17 under control on the ventral side
00:14:58.03 in order to allow pipe to be expressed.
00:15:03.01 Not shown in this slide,
00:15:04.28 but also done in experiments by Li-Mei,
00:15:08.05 is that she was able to show that
00:15:10.09 all of this Cbl activity
00:15:12.24 depends entirely on the presence of gurken.
00:15:16.03 So, in a gurken mutant,
00:15:17.26 pipe is just simply expressed everywhere,
00:15:20.20 whether or not Cbl is there or not,
00:15:23.19 telling us, therefore,
00:15:25.24 that what we're looking at here
00:15:27.16 really is the result of gurken/EGF receptor signaling
00:15:30.14 and not some other signaling pathway
00:15:32.28 that's affected by the Cbl mutant.
00:15:40.18 These results, though,
00:15:44.00 then led us to revise our view
00:15:48.16 of the gurken gradient to some extent.
00:15:51.19 The fact that, in the absence of Cbl,
00:15:54.29 gurken can activate the EGF receptor
00:15:58.00 even in ventral follicle cells
00:16:00.21 tells us that some gurken protein
00:16:03.07 must actually reach the ventral side.
00:16:06.19 So, shown here in this cartoon form,
00:16:09.18 gurken RNA is localized
00:16:12.23 to the dorsal side of the egg chamber
00:16:14.26 in a very sharp domain,
00:16:17.22 but then the protein that's released from the egg
00:16:21.07 must be able to diffuse to a certain extent
00:16:23.26 to the ventral side.
00:16:26.26 Under normal circumstances,
00:16:29.08 Cbl is there to help keep
00:16:32.28 EGF receptor activity low on the ventral side
00:16:36.12 so that these occasional gurken molecules
00:16:38.19 that do make it to the ventral side
00:16:41.00 don't have the effect of
00:16:44.16 repressing pipe or activating Kekkon
00:16:47.02 under normal circumstances.
00:16:48.26 So, in the wildtype we see Kekkon
00:16:50.24 only expressed on the dorsal side,
00:16:52.26 where we have high EGF receptor activity,
00:16:55.11 and pipe expressed on the ventral side,
00:16:58.07 where there's low EGF receptor activity.
00:17:01.02 But then, in the Cbl mutant,
00:17:03.18 where EGF receptor activity overall
00:17:06.23 gets increased,
00:17:09.06 now Kekkon can be expressed from dorsal to ventral,
00:17:12.29 and pipe gets repressed even on the ventral side.
00:17:21.16 We concluded that gurken must reach the ventral side
00:17:25.19 by doing these mosaic experiments
00:17:28.28 and looking at the expression of the response genes.
00:17:32.13 Meanwhile, Li-Mei Pai, in her own laboratory,
00:17:35.07 has gone on and she has used
00:17:38.17 a system where she can visualize gurken molecules
00:17:42.19 during oogenesis
00:17:46.02 much better than we could with the regular antibody,
00:17:48.06 and she has in fact found that, indeed,
00:17:50.25 she can see some gurken molecules
00:17:53.25 in the follicle cells on the ventral side of the egg chamber.
00:18:01.01 However,
00:18:04.26 we then...
00:18:06.29 we now have a new question.
00:18:08.28 So, the gurken gradient is established
00:18:10.29 and it reaches from the dorsal side
00:18:13.05 all the way to the ventral side.
00:18:15.11 So, what are the biophysical parameters
00:18:17.17 of this gradient?
00:18:19.12 How can it activate response genes?
00:18:21.28 What are thresholds
00:18:24.01 for these response genes?
00:18:26.19 How can we think about pattern formation
00:18:29.14 from the dorsal all the way to the ventral side?
00:18:33.18 We can use antibodies to look at
00:18:36.11 the gurken protein...
00:18:39.03 Li-Mei has been able to
00:18:41.13 visualize the gurken molecule, here,
00:18:43.18 with a fusion protein...
00:18:45.20 but the protein that is vexing the field
00:18:48.19 is that, while we can see the gurken protein
00:18:52.23 in various stages,
00:18:55.11 it's hard to know which of these molecules
00:18:58.10 that we're visualizing here
00:19:00.05 are actually the active protein.
00:19:02.29 The protein is made as a precursor.
00:19:05.12 It's cleaved and then can diffuse,
00:19:07.23 and then it's taken up by the follicle cells
00:19:10.08 and eventually degraded inside the follicle cells.
00:19:14.01 So, where are the molecules
00:19:16.04 that are really actively forming the gradient?
00:19:18.26 How we can think about those,
00:19:21.20 or how can we get some impression of
00:19:26.01 what the shape of the gurken gradient is?
00:19:29.11 In order to address this question,
00:19:33.00 I have been for a while
00:19:35.19 collaborating with a colleague of mine at Princeton,
00:19:38.20 Stas Shvartsman,
00:19:41.04 and he and his students
00:19:43.26 have been modeling
00:19:46.17 dorsoventral pattern formation in oogenesis
00:19:49.16 and together we had this project
00:19:53.27 where we were combining
00:19:57.11 some experimental approaches
00:19:59.28 with the modeling approaches
00:20:02.03 to gain some description
00:20:04.23 of the shape of the gurken gradient.
00:20:07.11 So, how is that done?
00:20:10.24 Well, Lea Goentoro,
00:20:13.04 a graduate student of Stas Shvartsman's,
00:20:18.16 employed both
00:20:23.06 experimental manipulation of the gurken gradient
00:20:28.02 and she also did modeling of the gurken gradient,
00:20:32.06 and then she did parameter fitting
00:20:35.12 and estimations of parameters
00:20:38.03 that allowed us to get some sense
00:20:40.11 of the gurken gradient.
00:20:43.23 So, initially, Lea had to write
00:20:47.10 all the equations...
00:20:49.08 she had to model the gurken gradient.
00:20:51.08 This would involve the production of the molecule,
00:20:53.17 its diffusion,
00:20:55.24 its binding to the receptor...
00:20:58.21 the receptor, of course,
00:21:00.23 has also to be made in the follicle cells
00:21:02.19 and deployed on the plasma membrane,
00:21:05.11 and all of this can then be summarized
00:21:09.05 in a relatively complicated equation
00:21:12.29 that I won't go into here,
00:21:15.21 but the important thing to note here
00:21:19.10 is that really none of the parameters
00:21:24.15 have been determined in absolute terms,
00:21:26.24 which means that this equation cannot just simply be solved
00:21:30.05 by putting actual numbers into this equation.
00:21:33.17 They all can only be described
00:21:36.22 in a series of differential equations.
00:21:40.00 So, then, how can we still about
00:21:43.04 this relatively complicated gradient formation.
00:21:46.21 Well, there are some things
00:21:49.22 that are intuitively understandable.
00:21:52.29 We can think of two different extremes.
00:21:56.27 We can have something that's a long-range gradient
00:21:59.28 or we could have, conversely,
00:22:02.09 something that's a short-range gradient.
00:22:05.08 A long-range gradient could be formed,
00:22:07.13 for instance,
00:22:09.07 if the molecule we're looking at
00:22:11.06 has a very high diffusion rate,
00:22:13.06 if it has a low binding constant to the receptor,
00:22:16.02 so it diffuses long before it binds to the receptor,
00:22:19.29 or it it has a very long half-life
00:22:22.08 relative to the space it has to travel through.
00:22:27.05 And conversely, of course,
00:22:28.26 a short-range gradient would be one
00:22:32.02 that could be formed by molecules that diffuse very slowly,
00:22:35.01 that bind with affinity to the receptor
00:22:37.16 so that they immediately bind
00:22:39.15 as soon as they're released from the oocyte,
00:22:41.19 or if they have a very short half-life.
00:22:45.26 We can now test whether gurken
00:22:48.07 is a long- or a short-range gradient
00:22:50.03 by manipulating some of the parameters,
00:22:53.17 and we decided to do this
00:22:55.17 by changing the level of the EGF receptor in oogenesis.
00:23:01.18 In flies, we have convenient systems
00:23:03.24 that allow us to express target genes at different levels,
00:23:11.24 it's called a Gal4-UAS system
00:23:14.22 that's been imported from yeast,
00:23:17.03 and by having different Gal4 transcription factors
00:23:20.27 that drive expression of the UAS target gene
00:23:24.12 at different level in oogenesis,
00:23:26.25 we could vary the amount of receptor
00:23:29.18 that was expressed in the follicle cells.
00:23:32.15 And then, Lea used the pipe gene
00:23:35.24 as a target readout for the gurken gradient.
00:23:41.02 The pipe gene was very convenient
00:23:43.06 because it has a very sharp on/off response
00:23:46.29 and so it was easy to quantify this response,
00:23:51.01 and you can see that with increasing receptor level
00:23:54.28 the pipe boundary was shifted.
00:24:03.23 When Lea quantified everything,
00:24:06.07 we saw that in fact
00:24:09.14 the response to these manipulations of the system
00:24:15.17 was quite noticeable.
00:24:18.18 So, the gradient therefore
00:24:22.02 seems to be easily manipulated,
00:24:26.26 which already implies
00:24:29.15 that it can't be just that the gurken molecule
00:24:35.00 immediately binds to the receptor.
00:24:37.17 It seems that by just changing receptor concentrations,
00:24:43.22 even at small levels,
00:24:46.12 we can make quite a noticeable effect
00:24:49.07 in the pipe response.
00:24:52.24 Putting all these results together
00:24:55.13 and then fitting the results to the model,
00:25:01.06 Lea was able to conclude
00:25:03.09 that gurken, indeed, is a long-range signal,
00:25:07.12 and in particular she was able to calculate that
00:25:12.17 approximately the difference
00:25:16.17 between the concentration of gurken
00:25:19.01 on the dorsal side versus the ventral side
00:25:21.08 is only ten-fold.
00:25:24.11 She also could give approximate
00:25:27.11 threshold levels to the gurken gradient
00:25:30.16 for the pipe and the Kekkon response.
00:25:35.19 We think this is an interesting approach
00:25:39.00 that could potentially also be used
00:25:41.13 to estimate other gradients in development,
00:25:43.27 where it might also be hard
00:25:46.13 to really visualize the active molecules,
00:25:49.01 but by performing a simple manipulation
00:25:52.10 such as giving extra receptor copies to the tissue
00:25:56.18 and seeing whether the tissue responds
00:25:58.25 and by how much,
00:26:00.28 it should be possible to get similar estimates
00:26:03.09 for other gradients as well.
00:26:07.08 At the end, now,
00:26:09.13 I would like to thank
00:26:11.29 everybody who has made it so far
00:26:14.03 and listened to my talk on dorsoventral pattern formation
00:26:18.18 in Drosophila oogenesis,
00:26:21.08 and, last but not least,
00:26:23.13 I need of course to thank all the many people
00:26:25.25 in my laboratory
00:26:27.21 who have contributed to this work over the years,
00:26:31.08 and I also would like to thank
00:26:33.19 my collaborators at Princeton,
00:26:35.23 Stas Shvartsman and Liz Gavis,
00:26:39.05 who have collaborated with us in very exciting new projects,
00:26:44.07 and I hope we'll do more of those in the future.
00:26:49.15 Thank you.

Videos in this Talk

Part 1: Axes Formation in the Drosophila Egg
Audience:
- Researcher
- Educators of Adv. Undergrad / Grad
Part 2: grk RNA Localization
Audience:
- Researcher
- Educators of Adv. Undergrad / Grad
Part 3: Gurken Gradient and Follicle Cell Response
Audience:
- Researcher
- Educators of Adv. Undergrad / Grad

Speaker: Trudi Schupbach

Total Duration: 1:11:52

Recorded: February 2009

All Talks in Development and Stem Cells

Talk Overview

How do complex multicellular organisms develop from single celled eggs with a single nucleus? We study this question in the fruitfly, Drosophila. In these insects, as in many other organisms, the major body plan is predetermined during oogenesis, or egg development. In the first part of the lecture, I will give an introduction to oogenesis in Drosophila, and the techniques we use to find genes that are responsible for determining the major axes of the egg and embryo. Interestingly, our analysis revealed that this process requires cell to cell communication between the oocyte and the surrounding follicle cells. It involves a signaling molecule, Gurken, which provides a localized signal from the oocyte to the follicle cells and ultimately sets up both the anterior-posterior as well as the dorso-ventral axis of the egg.

In the second part of the lecture I will introduce our ongoing efforts to understand axis formation in Drosophila oogenesis at a molecular level. In the first part of the lecture, I introduced the localized signaling molecule, Gurken. The RNA that encodes Gurken accumulates in a very restricted area of the oocyte. This localization signals spatial information to the surrounding follicle cells. I will explain how we use a combination of biochemical techniques and live imaging to unravel the mechanisms that localize this RNA during oocyte development. This work is in progress, and the lecture will provide a snapshot of what we know and what the open questions are.

Part 3: The third part of the lecture focuses on the spatial information that is conveyed by the oocyte to the surrounding follicle cells. I discuss how the spatially restricted activation of the EGF receptor by the signaling molecule, Gurken, is relayed into a cascade of information that ultimately sets up the dorso-venral axis of the embryo. I explain how we use genetic mosaicism in the follicle epithelium to ask questions about this signaling process. I also summarize results that were obtained in collaboration with my colleague, Stas Shvartsman, where we used a combination of experiments and modelling to determine the shape of the Gurken gradient.

Speaker Bio

Trudi Schupbach

Trudi Schupbach is Professor of Molecular Biology at Princeton University and a Howard Hughes Investigator. She grew up in Switzerland, and did her undergraduate and graduate work at the University of Zurich studying the development of the genital disc and sex determination in the germline of Drosophila. She moved to Princeton as a research associate… Continue Reading