Upload
nguyenminh
View
219
Download
0
Embed Size (px)
Citation preview
1
Drosophila oocyte polarity and cytoskeleton organization requires regulation of 1
Ik2 activity by Spn-F and Javelin-like 2
Simha Amsalem1, Anna Bakrhat
1, Tetsuhisa Otani
2, Shigeo Hayashi
2,3, Bareket 3
Goldstein1, Abdu Uri
1,# 4
1Department of Life Sciences, Ben-Gurion University, Beer-Sheva 84105, Israel. 5
2Laboratory for Morphogenetic Signaling, RIKEN Center for Developmental Biology, 6
2-2-3, Minatojima-Minamimachi, Chuo-ku, Kobe, Hyogo 650-0047, Japan. 7
3Department of Biology, Kobe University Graduate School of Science, Kobe, Hyogo 8
657-8501, Japan 9
10
11
12
Keywords: Cytoskeleton, Drosophila, Gurken, Ik2, Javelin-like, Oskar, Spn-F, 13
Running title: Regulation of Ik2 activity in oocyte polarity. 14
#Correspondence should be addressed to Uri Abdu: [email protected] 15
16
MCB Accepts, published online ahead of print on 9 September 2013Mol. Cell. Biol. doi:10.1128/MCB.00713-13Copyright © 2013, American Society for Microbiology. All Rights Reserved.
on March 18, 2018 by guest
http://mcb.asm
.org/D
ownloaded from
2
17
Abstract 18
The Drosophila Spn-F, Ik2 and Javelin-like (Jvl) proteins interact to regulate oocyte 19
mRNA localization and cytoskeleton organization. However, the mechanism by which 20
these proteins interact remains unclear. Using antibodies to activated Ik2, we showed 21
that this protein is found at the region of oocyte and follicle cell where microtubule 22
minus-ends are enriched. We demonstrate that germline Ik2 activation is diminished 23
both in jvl and in spn-F mutant ovaries. Structure-function analysis of Spn-F revealed 24
that the C-terminal end is critical for protein function, since it alone was able to rescue 25
spn-F sterility. On the other hand, germline expression of Spn-F lacking its conserved 26
C-terminal region (Spn-F〉C) phenocopied ik2, leading to production of ventralized 27
eggshell and bicaudal embryos. In Spn-F〉C-expressing oocytes, Gurken protein is 28
mislocalized and oskar mRNA and protein localization is disrupted. Expression of Ik2 29
rescued Spn-F〉C ovarian phenotypes. We found that whereas Spn-F physically 30
interacts with Ik2 and Jvl, Spn-F〉C physically interacts with Ik2 but not with Jvl. 31
Thus, expression of Spn-F〉C, which lacks the Jvl-interacting domain, probably 32
interferes with Ik2 and Jvl interaction. In summary, our results demonstrate that Spn-F 33
mediates the interaction between Ik2 and Jvl to control Ik2 activity. 34
35
36
37
38
on March 18, 2018 by guest
http://mcb.asm
.org/D
ownloaded from
3
39
Introduction 40
During development and cell differentiation, mRNA localization is a crucial step 41
in the regulation of gene expression of many transcripts. Accurate mRNA localization 42
permits precise temporal and spatial regulation of protein production during 43
development in a variety of organisms and cell types. RNA localization has been 44
described in organisms as diverse as yeast and humans and has been observed in many 45
polarized cells, such as oocytes, fibroblasts or neurons. In Drosophila, mRNA 46
localization provides a particularly important mechanism for polar localization of axis-47
determining factors during oogenesis. The asymmetric localization of mRNA within the 48
developing egg chamber relies on both microtubules (MTs) and actin networks, as well 49
as on motor proteins. Although the organization of MTs and actin during mid-oogenesis 50
has been revealed, the mechanism that leads to this complex cytoskeleton organization 51
is still not fully understood. 52
The Drosophila spindle-F (spn-F), IKKi homologue (Ik2), and the novel 53
microtubule (MT)-associated protein, Javelin-like (Jvl), together produce a complex of 54
proteins that affect both oogenesis and bristle development (1-4). We and others have 55
shown that females carrying mutations in these genes produce eggs and embryos with 56
polarity defects that arise due to disruptions in cytoskeleton organization and mRNA 57
localization during oocyte development (1-2, 4). We, moreover, have demonstrated that 58
these three proteins physically interact and that their proper cell localization and 59
function are inter-dependent (3-4). In their physical interaction, Ik2 phosphorylates Spn-60
F, although such phosphorylation does not affect the stability of the protein (4). In 61
addition, ik2 has also been found to be involved in other processes, including spindle 62
organization (5-6), dendrite pruning (7), bristle MT function (8-9), F-actin assembly 63
regulation (10-11) and in the shuttling of recycling endosomes during bristle cell 64
elongation (12). 65
Closer examination of spn-F and ik2 ovarian defects reveals that whereas both 66
mutants share the same defects in terms of cytoskeleton organization, they differ in their 67
effects on mRNA localization. In the mutants, both transport towards the minus-end of 68
the MT and organization of the MTs that surround the oocyte nucleus are strongly 69
affected (1-2). The spn-F and ik2 mutants also present the same defects in terms of grk 70
mRNA and protein localization. However, while over 90% of the embryos produced by 71
on March 18, 2018 by guest
http://mcb.asm
.org/D
ownloaded from
4
ik2 mutant females are bicaudal (2), this phenotype is only rarely found in spn-F mutant 72
embryos (1). This difference could be attributed to the fact that in ovaries and embryos 73
produced by ik2 mutant females, oskar (osk) mRNA and protein are localized 74
posteriorly and anteriorly, while in spn-F mutant ovaries, osk mRNA and protein 75
localization are not affected. The difference seen in osk mRNA but not in grk mRNA 76
localization defects between spn-F and ik2 mutants raises the question as to which 77
molecular mechanisms control the actions of these proteins. 78
To better understand the function of these genes in mRNA localization and 79
cytoskeleton organization during development, structure-function analysis of Spn-F 80
protein was conducted. We show that the Spn-F protein may act as mediator between 81
Ik2 and Jvl to regulate Ik2 activity. Thus, our results provide a new perspective on the 82
function of these proteins in pattern formation of the Drosophila egg and embryo 83
demonstrating that Spn-F and Jvl act on the core Ik2 function to augment the activity of 84
this complex. 85
86
Materials and Methods 87
Drosophila stocks 88
Oregon-R served as a wild-type control. The following mutants and transgenic flies 89
were used: spn-F1, Df(3R)tll-e, (1), jv1
l, Df(3R)Exel6275 (3), UASp:GFP-Spn-F, 90
UASp:GFP-Ik2 (4); UASp:GFP-Rad9 (13) kinesin く-GAL insertion line KZ503 and the 91
Nod く-GAL insertion line NZ143.2 (14, 15); P{w+, osk-LacZ} (16). Germline 92
expression was performed with P{matg4- GAL-VP16}V37 (subsequently referred to as 93
g-tub Gal4-3), obtained from the Bloomington Stock Center. 94
95
Cloning and transgenic flies 96
The Spn-F-N and Spn-F-C constructs were described in Dubin-Bar et al. (2008). To 97
create the truncated Spn-F〉C, CC2 and CT Spn-F protein, the sequence coding for that 98
region spanning nucleotides 1–840, 573-819 and 810-1092 respectively, of the gene 99
were amplified by PCR and cloned into the pUASp plasmid in fusion with DNA 100
encoding GFP using the XbaI restriction site. To create a non-GFP-tagged Spn-F〉C 101
construct, Spn-F〉C was cloned into plasmind pUASp using the KpnI site. P-element-102
mediated germline transformation of these constructs was carried out according to 103
standard protocols (17). 104
on March 18, 2018 by guest
http://mcb.asm
.org/D
ownloaded from
5
105
106
Eggshell and cuticle preparations 107
Eggs and embryos were allowed to age for 24 h on plates containing a molasses agar 108
medium at 25ºC. For chorion visualization, eggs were collected and prepared as 109
previously described (18). For cuticle preparation, eggs were collected and washed with 110
0.05% Triton X-100 and dechorionated in 50% commercial bleach (3% sodium 111
hypochlorite) for 2-3 min. Following methanol devitellinization, embryos were mounted 112
in Hoyer’s medium (19), and incubated at 65ºC overnight. Eggs and embryos were 113
viewed by phase contrast microscope. 114
115
In situ hybridization 116
RNA in situ hybridization on ovaries and embryos was carried out as described 117
previously (1, 20). 118
119
ß-galactosidase and antibody staining 120
ß-galactosidase staining of ovaries was performed according to Peretz et al. (2007), with 121
the exception that the ovaries were incubated in X-gal stock solution at room 122
temperature. Antibody staining of ovaries was performed as described previously (4). 123
The following primary antibodies were used: mouse anti-Grk (1:10, clone 1D12) (21), 124
rabbit anti-Oskar (1:3000) (22), mouse anti-g-tubulin (1:100, Sigma), and rabbit anti-125
pIKK (10). Goat anti-mouse Cy3 or Cy2, and goat anti-rabbit Cy3 or Cy2 secondary 126
antibodies (Jackson ImmunoResearch) were used at a dilution of 1:100. For g-tubulin 127
staining, ovaries were kept at room temperature after fixation to prevent MT 128
depolymerization. The following dyes were used: Oregon green 488 and Alexa Fluor 129
568 phalloidin (1:250, Molecular Probes). All pictures were imaged on an Olympus 130
FV1000 laser-scanning confocal microscope. 131
132
Western blot analysis 133
Dissected ovaries were ground in Laemmli sample buffer (10 たl per ovary). The protein 134
extracts were boiled for 5 min and loaded onto a 10% polyacrylamide gel. Following 135
electrophoresis, proteins were transferred to nitrocellulose membranes for 1 h at 300 136
mA. The nitrocellulose membranes were blocked by incubation in TTBS (0.2 M Tris-137
HCl, pH 7.5, 1.5 M NaCl, 9 mM Tween 20) containing 2.5% non-fat dry milk for 30 138
on March 18, 2018 by guest
http://mcb.asm
.org/D
ownloaded from
6
min at room temperature followed by either a 1 h incubation with anti-acetylated tubulin 139
(1:1000, Sigma) or overnight with anti-Ikk (1:20, 10) primary antibodies. The 140
membranes were washed in TTBS and incubated for 30 min with horseradish 141
peroxidase (HRP)-labeled anti-mouse antibodies (1:2000, Amersham). Antibody 142
labeling was visualized in a Fujifilm LAS300 imager using an ECL detection kit 143
(Biological Industries). 144
145
Co-immunoprecipitation and co-localization assay 146
S2 cells expressing constructs as described in the text were treated with lysis buffer and 147
immuno-complexes were recovered using GFP-Trap_A (Chromotek) according to the 148
manufacturer's instruction. To detect interactions between proteins, Western blot with 149
g-mCherry or g-Ikk antibodies was performed. To detect protein localization patterns, 150
mCherry-Jvl, Myc-tagged Ik2, GFP-Spn-F and GFP-Spn-F〉C constructs were used. 151
For Ik2 detection, primary mouse anti-c-Myc antibodies (1:150, Santa Cruz 152
Biotechnology) were used. Goat anti-mouse Alexa Fluor 633 secondary antibodies 153
(Molecular Probe) were used at a dilution of 1:100. 154
155
Yeast two hybrid assay 156
Yeast two hybrid analysis was performed using the Yeast Two Hybrid Phagemid vector 157
kit (Stratagene), following the manufacturer's instructions. The pAD-Spn-F and 158
truncated version plasmids were used as bait while a plasmid encoding full length Jvl 159
protein was used as prey. 160
161
Results 162
Activation of Ik2 phosphorylation is dependent on spn-F and javelin-like genes 163
To study the activation pattern of Ik2 during oogenesis, we used antibodies raised 164
against Ik2 phosphorylated on serine 175 (pIKK�) (12). We found that in egg chambers 165
from stage 5-6, pIKK� is found throughout the oocyte, with higher accumulation seen 166
at the posterior end (Fig. 1A). Later on, after the oocyte nucleus migrates to the dorsal-167
anterior corner of the oocyte, pIKK� is found at the anterior ring, with higher 168
accumulation seen in the vicinity of the nucleus (Fig. 1B) and in a punctate pattern in 169
nurse cells (Fig. 1B). We also noticed that pIKK� is present on the apical side of the 170
on March 18, 2018 by guest
http://mcb.asm
.org/D
ownloaded from
7
follicle cells (Fig. 1B'-B’’’). Thus, we found that pIKK� accumulates at MT minus 171
end-rich regions both in the oocyte and in follicle cells. Next, we studied the 172
localization pattern of pIKK� in spn-F and in jvl mutant ovaries and found that whereas 173
in spn-F mutants, anti-pIKK� antibody staining of the oocyte and nurse cells was 174
abolished (Fig. 1C-D), in jvl mutants no such pIKK� staining was detected in the 175
oocyte but was still seen in the nurse cells (Fig. 1D). Also, in both the spn-F (Fig. 1 D') 176
and jvl mutants (Fig. 1F'), anti-pIKK� antibody staining was still evident on the apical 177
side of the follicle cells. Thus, our results suggest that both spn-F and jvl are required 178
for Ik2 function in the germline but not in somatic follicle cells. 179
Next, we studied whether the absence of anti-pIKK� antibody staining in the 180
germline of spn-F and jvl mutants could arise due to defects in Ik2 protein stabilization. 181
Using antibodies directed against the Ik2 protein (10), we thus compared the levels of 182
Ik2 protein in ovarian extracts from wild type flies and in spn-F (Fig. 1G) and jvl 183
mutants (Fig. 1H). We found that in the ovarian extracts of both mutants, the level of 184
Ik2 protein was similar to that of the wild type ovarian extract. Thus, our results show 185
that spn-F and jvl are required for Ik2 activation but not for its stabilization in the 186
germline. 187
188
Expression of C-terminally truncated Spn-F affects the antero-posterior and 189
dorso-ventral axes 190
In this study, we sought to understand how Spn-F affects Ik2 activity by conducting 191
structure-function analysis of the Spn-F protein. Our search for conserved regions of 192
Spn-F revealed the presence of two coiled-coil domains, the first extending from amino 193
acid residue 32 to 114 and the second from residue 210 to 243. We had previously 194
shown that the C-terminal end (but not the N-terminal end) of Spn-F is crucial for 195
interaction with Ik2 (4). To investigate the functional importance of these Spn-F 196
domains, we tested the functions of several mutant Spn-F transgenes, deleted of 197
sequences encoding different domains of the protein. Examination of multi-species spn-198
F protein sequences aligned by ClustalX showed that the two coiled-coil domains are 199
conserved in all species considered. Using this alignment, an additional C-terminal 200
conserved region spanning from amino acid 285 to the end of the protein sequence was 201
found (data not shown). Three deletion constructs were thus generated in plasmid 202
pUASp to yield truncated versions of Spn-F N-terminally tagged with GFP. The first 203
variant encodes the N-terminal region of Spn-F (residues 1-162, hereafter termed Spn-204
on March 18, 2018 by guest
http://mcb.asm
.org/D
ownloaded from
8
F-N), encompassing the first coiled-coil domain. The second construct encodes the C-205
terminal domain of Spn-F (residues 165-364, hereafter termed Spn-F-C) that includes 206
the second coiled-coil domain of the protein. Finally, the third construct encodes a 207
version of Spn-F that includes the two coiled-coil domains but lacks the 84 C-terminal 208
residues (i.e. residues 1-280, hereafter termed Spn-F〉C) (Fig. 2A). Transgenic flies 209
expressing each of the truncated proteins under the control of the UAS/Gal4 system 210
were created. Three independent transgenic lines were tested for the expression of each 211
of the constructs. All lines showed the same pattern, as described below. To drive the 212
expression of the different Spn-F-encoding constructs, we used mat gTub-GAL4-VP16, 213
a GAL4-VP16 fusion protein expressed under the control of the alphaTub67C promoter 214
(Bloomington stock # 7062 or 7063). These Gal4 drivers lead to higher protein 215
expression when starting from stage 5-6 egg chambers. 216
First, we tested the ability of each of the constructs to rescue spn-F mutant 217
female sterility. We found that neither spn-F-N nor spn-F〉C germline expression 218
rescued spn-F female sterility. On the other hand, 95% of the eggs (n=236) laid by 219
females expressing spn-F-C in spn-F mutants background hatched, demonstrating that 220
the C-terminal half of Spn-F is sufficient for Spn-F function. Interestingly, we noticed 221
that whereas germline expression of spn-F-N and spn-F-C in a wild type background 222
had no effect on female fertility, expression of spn-F〉C led to complete female sterility. 223
Closer examination revealed that 92% (n=1036) of eggs laid by females expressing spn-224
F〉C in the germline produced ventralized eggshells (Figure 2 B-E). Furthermore, we 225
found that expression of spn-F〉C severely affected the anterior-posterior axis of the 226
embryos. 98% (n=126) of the embryos produced by females expressing spn-F〉C 227
showed a strong bicaudal phenotype, with several abdominal segments appearing in 228
mirror image symmetry, and Filzkoerper and telson at both ends (Fig. 2G). The same 229
results were obtained with non-tagged constructs, demonstrating that the defects seen 230
with spn-F〉C flies are not due to fusion of the GFP tag to the protein. 231
232
Expression of C-terminally truncated Spn-F affects Gurken (Grk) protein 233
localization 234
Dorsal-ventral polarity defects can be attributed to disruptions in the grk-Egfr signaling 235
pathway. Since specific expression of C-terminally truncated Spn-F protein in the 236
germline led to ventralized egg production, we examined the localization and expression 237
of grk RNA and Grk protein in these ovaries. In situ hybridization analysis with a grk 238
on March 18, 2018 by guest
http://mcb.asm
.org/D
ownloaded from
9
probe was performed in female flies expressing Spn-F〉C and wild type ovaries. We 239
found that similar to wild type ovaries (arrow in Fig. 3A, 23-25), ovaries expressing 240
spn-F〉C showed no effect on grk mRNA localization, with grk mRNA being detected 241
as a cap around the oocyte nucleus in the anterior-dorsal corner of the oocyte (arrow in 242
Fig. 3B). Next, we tested the localization pattern of Grk protein and found that 243
expression of spn-F〉C in the germline led to a profound effect on Grk localization 244
throughout oocyte development. In wild type ovaries from stage 7 onward, Grk is 245
restricted to the anterior-dorsal corner of the oocyte (arrow in Fig. 3C), much as seen in 246
terms of grk mRNA localization. In stage 7 (data not shown) to stage 9 (arrow in Fig. 247
3D) egg chambers from Spn-F〉C-expressing females, Grk protein was localized to 248
abnormally large puncta close to the oocyte. These results indicate that the dorsal-249
ventral defects in Spn-F〉C-expressing eggs are due to mislocalization of Grk protein. 250
251
Secreted and microfilament-related proteins are localized to ectopic actin cages in 252
oocytes from flies expressing spn-F〉C. 253
The integrity of both the microtubule and actin networks is essential for correct mRNA 254
localization in the oocyte. To examine the organization of the actin cytoskeleton, 255
ovaries were stained with rhodamine-conjugated phalloidin (Fig. 3). We found that 256
expression of Spn-F〉C affected oocyte actin organization. Whereas in a stage 8-9 wild 257
type egg chamber actin is evenly distributed at the cortical surface (Fig. 3C), the ectopic 258
F-actin network is juxtaposed to the oocyte nucleus (arrow in Fig. 3E). 259
Ectopic actin clumps were also observed in trailer hitch (tral) and BicC mutants, 260
where it was shown that also other microfilament-related and secreted proteins associate 261
with the ectopic actin cages (26-27). We saw that in flies expressing Spn-F〉C, the 262
secreted protein Grk is co-localized with the abnormal actin clumps in oocytes (arrow in 263
Fig. 3F). Moreover, we found that Spectrin, a microfilament-related protein, is also 264
associated with the actin clumps (Fig. 3G-J). Thus, our results show that both secreted 265
proteins, such as Grk, and the microfilament-related protein, Spectrin, co-localized to 266
the ectopic actin network in oocytes from flies expressing Spn-F〉C. 267
268
Expression of C-terminally truncated Spn-F affects osk mRNA and protein 269
localization 270
Establishment of the anterior-posterior axis of the embryo depends on the localization 271
of bcd and osk mRNA to the anterior and posterior poles of the oocyte, respectively (28-272
on March 18, 2018 by guest
http://mcb.asm
.org/D
ownloaded from
10
30). Indeed, most of the bicaudal phenotypes reported arise from either mis-localization 273
of osk mRNA to the anterior of the oocyte (29, 31) or from premature translation of osk 274
mRNA (32-34)0 As described above, 98% of the eggs from females expressing Spn-275
F〉C produced bicaudal embryos. We initially examined bcd mRNA localization and 276
found that in Spn-F〉C-expressing egg chambers, bcd mRNA was localized to the 277
anterior ring of the oocyte, similarly to wild type flies (data not shown). Next, we 278
examined the localization pattern of osk mRNA and protein in ovaries and eggs from 279
Spn-F〉C-expressing females. In wild type ovaries, osk mRNA was tightly localized to 280
the posterior pole of the oocyte (Fig. 4A). However, in 47% of stage 9 or 10 egg 281
chambers (n=36) from Spn-F〉C-expressing females, osk mRNA was found both at the 282
posterior and anterior poles and/or abnormally accumulated in the middle of the oocyte 283
(arrows in Fig. 4B). Analysis of Osk protein localization revealed that in comparison to 284
wild type oocytes, where Osk protein is found as a tight crescent at the oocyte posterior 285
end (Fig. 4C), in Spn-F〉C-expressing stage 10 oocytes, Osk protein was distributed 286
more diffusely (Fig. 4D). 287
To characterize the localization pattern of Osk protein in oocytes at later stages 288
of oogenesis, we used an osk-LacZ construct as a reporter of osk mRNA translation 289
since most antibodies fail to penetrate at these stages (16). We found that in mature eggs 290
produced by wild type females, Osk-LacZ protein localized exclusively to the posterior 291
pole (Fig. 4E). However, in eggs from Spn-F〉C-expressing females, Osk-LacZ protein 292
accumulated at the anterior end (arrow in Fig. 4F) and was seen at high levels 293
throughout the egg, in addition to its normal localization to the posterior pole, 294
suggesting that anteriorly-localized osk mRNA in Spn-F〉C-expressing egg chambers is 295
only translated in the mature egg and does not undergo premature translation during 296
oogenesis. Next, we analyzed osk mRNA localization in early embryos and found that 297
in wild type embryos, osk transcripts are concentrated at the posterior of the embryo, as 298
well as being expressed at low levels throughout the embryo (Fig. 4H). In 90% of the 299
embryos (n=65) produced by Spn-F〉C-expressing females, osk mRNA levels were 300
found at the posterior pole and all over the embryo, as compared to wild type, and also 301
slightly accumulated at the anterior end of the embryo (arrow in Fig. 4I). 302
303
Microtubule polarity defects in oocytes from flies expressing Spn-F〉C 304
on March 18, 2018 by guest
http://mcb.asm
.org/D
ownloaded from
11
We next examined the integrity of the microtubule network, as revealed by g–tubulin 305
antibody staining, and found that expression of Spn-F〉C affects MT organization in 306
mid-stage egg chambers. In stage 9 egg chambers, oocyte MTs are organized in a 307
gradient, with higher accumulation observed at the anterior cortex of the oocyte (arrow 308
in Fig. 5A). In oocytes from females expressing Spn-F〉C, a significant reduction in g-309
tubulin levels along the anterior cortex was observed (arrow in Fig. 5B). 310
To investigate the polarity and functionality of the MT network, we tested 311
transport to MT plus and minus ends, using the kin:く-gal and Nod:く-gal markers, 312
respectively (14-15). In wild type stage 8 egg chambers, Nod:く-gal is localized to the 313
anterior cortex of the oocyte and is enriched in the dorsoanterior corner, near the oocyte 314
nucleus (arrows in Fig. 5C). However, in egg chambers from females expressing Spn-315
〉C, Nod:く-gal could not be detected (Fig. 5D). In analyzing transport to the MT plus 316
end, we found that whereas in wild type stage 9-10 egg cambers, kin:く-gal is tightly 317
concentrated at the oocyte posterior (arrow in Fig. 5E), in Spn-F〉C-expressing egg 318
chambers, kin:く-gal is abnormally localized. In 58% of the egg chambers examined 319
(n=64), the protein was not detected (Fig. 5F), while in 12% of the egg chambers, the 320
protein was found at the posterior end and in the middle of the oocyte (arrows in Fig. 321
5G). In 27% of cases, the protein accumulated only at the center of the oocyte (arrow 322
Fig. 5H). These results suggest that transport towards the MT minus and plus ends is 323
severely disrupted in Spn-F〉C-expressing ovaries. 324
325
Expression of Ik2 rescues the sterility of females expressing Spn-F〉C 326
Since expression of Spn-F〉C phenocopied ik2 but not spn-F loss-of-function, we tested 327
whether Spn-F〉C ovarian phenotypes are due to inactivation of Ik2 activity. We saw 328
that expression of Ik2 significantly rescued Spn-F〉C ovarian phenotypes. Whereas 329
expression of Spn-F〉C (co-expressed with UASp-Rad9, which serves as a negative 330
control) led to complete female sterility (n=1036), in females co-expressing Spn-F〉C 331
and Ik2, 61% (n= 228) of the eggs hatched. Moreover, in Spn-F〉C-expressing females, 332
94% of the eggs (n=321) laid showed a ventralized pattern (a detailed description of 333
these eggshell patterns is found in Fig. 2 C-F), 17% of them were weakly ventralized, 334
74% were ventralized, and 3% were strongly ventralized eggshell (Fig. 6A). Following 335
the addition of Ik2 protein, these ventralized eggshell pattern was reduced to 26% 336
(n=293), where 19% were weakly ventralized, 4% were ventralized, and 3% were 337
strongly ventralized eggshell (Fig. 6A). Also, expression of Ik2 rescued the effect of 338
on March 18, 2018 by guest
http://mcb.asm
.org/D
ownloaded from
12
Spn-F〉C expression on the anterior-posterior pattern of the embryo, reducing the 339
bicaudal phenotypes from 98% to 39% (n=144). 340
341
Expression of C-terminally truncated Spn-F protein affects Ik2 activation 342
Next, we studied the localization pattern of pIKK� in ovaries expressing C-terminally 343
truncated Spn-F. We found that expression of Spn-F protein lacking its C-terminal 344
completely abolished pIKK� in the germline (Fig. 6 A-C) but did not affect Ik2 protein 345
stability (Fig. 6E). The use of the mat-GAL4 promoter, which drives high expression 346
from stage 6-7, allows for demonstration of the dramatic effects of Spn-F〉C expression 347
on pIKK�. In stage 5 egg chambers, where GFP:Spn-F〉C expression is low 348
(arrowhead in Fig. 6A), anti-pIKK� immune-staining in the oocyte is similar to that 349
seen in the wild type (arrow head in Fig. 6B). However in stage 8 egg chambers, where 350
GFP-tagged Spn-F〉C is expressed at high levels (arrow in Fig. 6A), anti-pIKK� 351
antibody staining was abolished (arrow head in Fig. 6B). Anti-pIKK� immuno-352
staining was also evident in follicle cells (Fig. 6B), since Spn-F〉C was expressed 353
exclusively in the germline. 354
Since we found that expression of Ik2 rescues the sterility of females expressing 355
Spn-F〉C, we tested whether Ik2 expression also restores Ik2 activity. We saw that 356
expression of Ik2 is sufficient to also restore Ik2 activity in flies expressing the C-357
terminally truncated Spn-F protein, as revealed by the restoration of anti-pIKK� 358
immunostaining in the ovaries of these flies (arrows in Fig. 6D). 359
360
Spn-F〉C protein can bind Ik2 but not Jvl 361
To better understand the mechanism by which Spn-F〉C over-expression affects 362
Ik2 activity, we studied the physical interaction between Spn-F〉C and Ik2 or Jvl. Our 363
previous results showed that the C-terminal but not the N-terminal end of the Spn-F 364
protein binds the Ik2 protein (4). The truncated version of Spn-F employed here 365
includes the second coiled-coil domain and the conserved region at the C-terminal 366
region of the protein (spanning from residue 285 to the end of the protein). Since the 367
expression of Spn-F〉C, which includes the second coiled-coil domain but lacks the C-368
terminal conserved region, affects Ik2 activity, we asked whether Spn-F〉C binds Ik2. 369
Accordingly, we co-expressed myc-Ik2 with either GFP-Spn-F or GFP-Spn-F〉C in S2 370
cells and then performed immunoprecipitation with anti-GFP antibodies. We found that 371
myc-Ik2 was co-precipitated with both GFP-Spn-F constructs, demonstrating that Spn-F 372
on March 18, 2018 by guest
http://mcb.asm
.org/D
ownloaded from
13
〉C is able bind the Ik2 protein (Fig 7B). These results suggest that the second coiled-373
coil domain within Spn-F is required for Ik2 binding. Indeed, the second coiled-coil 374
domain alone is able to co-precipitate with Ik2 (Shigeo Hayashi, personal 375
communication). 376
Next, we studied the ability of Jvl to interact with Spn-F〉C in a yeast two 377
hybrid assay. We found that whereas Spn-F physically interacts with Jvl (Fig. 7C and 378
Dubin-Bar et al., 2011), no interaction between Spn-F〉C and Jvl was detected (Fig. 379
7C). Most importantly, we were able to show that the C-terminal region of Spn-F is able 380
to bind Jvl (Fig. 7C). To verify these results, we considered the ability of Jvl to 381
physically interact with Spn-F protein and its truncated forms in immunoprecipitation 382
assays in S2 cells. Similar to the results of the yeast two hybrid assay, we found that full 383
length Spn-F protein and the C-terminal region of Spn-F but not Spn-F〉C were able to 384
bind Jvl (Fig 7D). 385
386
Spn-F but not Spn-F〉C protein co-localized with Ik2 and Jvl 387
The results showing that different domains within Spn-F bind Jvl and Ik2 suggest that 388
Spn-F, Jvl and Ik2 act as complex of proteins, where Spn-F acts as mediator between 389
Ik2 and Jvl. Thus, deleting Jvl-binding domain in Spn-F will affect the formation of the 390
complex. To test this prediction, we studied the localization pattern of all three proteins 391
in S2 cells. Expressions of each of this protein alone revealed that mCherry-Jvl is 392
localized to a filamentous structure (Fig. 8A), shown to be MT network (3). Myc-tagged 393
Ik2 (Fig. 8B) and GFP-Spn-F (Fig. 8C) were found in a punctate pattern, while GFP-394
Spn-F〉C (Fig 8D) was found mainly throughout the cytoplasm. We found that when 395
mCherry-Jvl, Myc-tagged Ik2 and GFP-Spn-F were co-expressed, all three protein co-396
localized in a filamentous structure which resembles the localization of mCherry-Jvl 397
alone (Fig. 8E-F). One the other hand, when GFP-Spn-F〉C was co-expressed with 398
mCherry-Jvl and Myc-tagged Ik2, both Myc-tagged Ik2 and GFP-Spn-F〉C were no 399
longer associated with mCherry–Jvl (Fig. 8 H-J). 400
401
Discussion 402
Previously, it was shown that Ik2 is activated locally at the tip of bristles (12). 403
In this study, we have demonstrated that during oogenesis, Ik2 is also locally activated 404
at MT minus end regions in the oocyte and follicle cells, as well as in the nurse cells, 405
on March 18, 2018 by guest
http://mcb.asm
.org/D
ownloaded from
14
where Ik2 presents a punctate pattern. We found that in the germline, jvl and spn-F are 406
required for activation of Ik2. To better understand the mechanism by which Spn-F 407
affects Ik2 activation, we performed structure-function analysis of the Spn-F protein. 408
Using this approach, we were able to demonstrate that the C-terminal end of Spn-F is 409
sufficient for protein function. Following the expression of several truncated Spn-F-410
encoding constructs in the germline, we noticed that expression of Spn-F lacking 84 411
amino acids from the C-terminal end produced similar defects in both the dorsal-ventral 412
and anterior-posterior axes as found in ik2 loss of function ovaries (2). A high 413
percentage of ventralized eggs and bicaudal embryos are produced by both the ik2 414
mutant (2) and Spn-F〉C-expressing females. Most importantly, the fact that expression 415
of Ik2 was able to significantly rescue defects in eggshell and embryo development, as 416
detected by Spn-F〉C expression, suggests that the C-terminal end of Spn-F regulates 417
Ik2 protein function. 418
The results of the current study demonstrated that in terms of mRNA 419
localization, expression of Spn-F〉C protein produced, similarly to ik2 mutants, high 420
percentages of bicaudal embryos due to defects in osk mRNA localization. Previously, 421
we showed that females mutant for spn-F produce low percentage of bicaudal embryos 422
ranging from embryos with the typical reduced head skeleton to rare symmetrical 423
bicaudals (1). Thus, we believe that Ik2 function in oocyte anterior-posterior patterning 424
has two components, one that depends on Spn-F and the other that does not. 425
One can thus ask how Ik2 activity affects osk mRNA localization. The role of 426
the cytoskeleton in transporting osk mRNA to its final destination required cooperation 427
between MTs and between MTs and actin motor proteins. Initially, osk mRNAs are 428
transcribed in nurse cells and transported into the MT minus end at the anterior of the 429
oocyte by dynein, along with the accessory factors BicD and Egalitarian (35-36). 430
Within the oocyte, it was shown that the localization of osk to the posterior end required 431
MTs, Khc and myosin V (37-39). Several models explaining how osk transcripts are 432
transported toward the posterior of the oocyte have been proposed, including active 433
transport to the posterior (38), diffusion and trapping (40), or exclusion from the 434
anterior and lateral cortex (41). One recent model suggested that osk mRNA is actively 435
transported along microtubules in all directions, with a slight bias toward the posterior 436
(42). As to the role of Khc in osk mRNA transport, it was suggested that Khc is required 437
either directly (38, 41) or indirectly (43-45). In the present study, we have demonstrated 438
that over-expression of Spn-F〉C, which eliminates Ik2 activity, affects Khc-LacZ 439
on March 18, 2018 by guest
http://mcb.asm
.org/D
ownloaded from
15
function and the posterior localization of osk mRNA. Based on our results, it is possible 440
that regulation of Ik2 by the Spn-F C-terminal region affects MT-biased polarity 441
towards the oocyte posterior, thus indirectly affecting Khc and osk mRNA localization. 442
Alternatively, the effect on osk mRNA localization could be due to a direct effect on the 443
regulation of Khc motor protein activity. 444
We have demonstrated that the defects in dorso-ventral axis in flies expressing 445
Spn-F〉C are due to the Grk protein but not mRNA mislocalization. Moreover, we 446
showed that Grk protein, a secreted protein, as well as Spectrin, a microfilament-related 447
protein, is localized to ectopic actin clumps in the oocyte. The localization of Grk to 448
ectopic actin clumps was reported for several mutants, including Bic-C (26-27), trailer 449
hitch (tral) (26-27), spn-F (1) and ik2 (2). In all of these mutants, grk mRNA was 450
mislocalized but in a different pattern than was Grk protein, suggesting that defects in 451
grk mRNA localization cannot account for defects in Grk protein localization. It was 452
suggested that Bic-C and tral are part of the same pathway that regulate efficient Grk 453
secretion (26-27). Accumulation of ectopic Grk protein in the oocyte was also found in 454
Khc and Dhc mutants, and it was suggested that both genes are also required for Grk 455
protein exocytosis (44). Thus, we suggest that the aberrant localization of Grk protein in 456
ik2 mutants and in flies expressing Spn-F〉C revealed a role of these proteins in 457
regulating Grk protein secretion. 458
To better understand the mechanism by which over-expression of Spn-F〉C 459
affects Ik2 activity, we studied interactions between Spn-F〉C and the Ik2 and Jvl 460
proteins. Previously, we had shown that Spn-F is able to directly bind Ik2 and Jvl. In 461
this study, we demonstrated that whereas Ik2 physically interacts with Spn-F〉C, Jvl 462
was not able to bind Spn-F〉C. We also found that Ik2 binds to the second coiled-coil 463
domain of Spn-F (Shigeo Hayashi, personal communication), while Jvl interacts with 464
the conserved C-terminal region of this protein. Moreover, we were able to show that 465
Spn-F but not Spn-F〉C protein forms a complex with Ik2 and Jvl. We believe that 466
expression of Spn-F〉C, which able to bind Ik2 but not Jvl, interfered with the 467
interaction between Ik2 and Jvl. These results suggest that specific interference with the 468
interaction between Ik2 and Jvl, as revealed upon Spn-F〉C expression, is critical for 469
Ik2 core functions during oogenesis. The fact that Spn-F mediates Ik2 interaction with 470
Jvl, a MT-associated protein, and the finding that Ik2 is activated at the MT-minus-end, 471
together with the specific effects of ik2 on oocyte MT organization (2), suggest that Ik2 472
plays a crucial role in MT organization and/or function during oogenesis. 473
on March 18, 2018 by guest
http://mcb.asm
.org/D
ownloaded from
16
474
475
476
Acknowledgements 477
We thank Anna Ephrussi and the Bloomington stock center for generously providing fly 478
strains and reagents. This research was supported by Israel Science Foundation grant 479
968/10 (to U.A.). 480
481
482
483
484
on March 18, 2018 by guest
http://mcb.asm
.org/D
ownloaded from
17
485
486
Figure legends: 487
Figure 1: spn-F and javelin-like (jvl) are required for Ik2 activation. (A-F) 488
Confocal images of egg chambers stained with antibodies against phosphorylated Ik2, 489
pIKK (red), and actin (green). In all figures, the egg chamber posterior is to the right. 490
(A-B) Wild type, (C-D) spn-F mutant, (E-F) jvl hemizygous. (A, C, and E) Stage 6 egg 491
chambers, (B, D and F) stage 7-8 egg chambers. (A) pIKK shows posterior 492
accumulation in the oocyte. (B) pIKK is localized to the anterior end of the oocyte and 493
close to the oocyte nucleus. In nurse cells, pIKK is found in a punctuate pattern 494
(arrows). (B'-B'') pIKK is also found at the apical side of follicle cells. (C-D) pIKK is 495
not detected in the spn-F germline but is found in follicle cells (arrows in D'). (E-F) 496
pIKK is not detected in jvl mutants oocytes but is found in nurse cells (arrows in F) and 497
follicle cells (arrows in F'). (G-H) Western blot analysis of Ik2 levels in ovaries. Ik2 498
levels in ovarian extracts from wild type and mutant flies were detected using antibodies 499
against this protein. The level of acetylated tubulin served as a loading control. 500
Mutations in spn-F and jvl do not affect the level of Ik2 protein. Df- Deficiency. 501
502
Figure 2: Expression of Spn-F〉C leads to the appearance of ventralized eggshell 503
and bicaudal embryos. (A) Schematic presentation of Spn-F protein domains and 504
deletion constructs used to make transgenic flies. CC1, coiled-coil 1, CC2, coiled- coil 505
2, CT, conserved C-terminal. (B) Western blot analysis of ovarian extract from flies 506
expressing either GFP- Spn-F〉C, GFP-Spn-F-C or GFP- Spn-F-N. The level of actin 507
served as a loading control. (C-F) Eggshells from flies expressing Spn-F〉C. (B) 8% of 508
the eggshells had a wild type appearance, (C) 13% had a weakly ventralized eggshell 509
with fused appendage, (D) 76% had a ventralized eggshell with one appendage, and (E) 510
3% had a strongly ventralized eggshell with no dorsal appendages. (G) Wild type 511
embryo, (H) 98% of embryos from flies expressing Spn-F〉C had a bicaudal phenotype. 512
In all figures, the Gal4 that was used was P{matg4- GAL-VP16}V37. 513
514
Figure 3: Secreted and microfilament-related proteins are localized to ectopic actin 515
cages in oocytes from flies expressing spn-F〉C. (A-B) In situ grk mRNA localization 516
in stage 9 egg chambers. (A) Wild type, (B) Spn-F〉C, grk mRNA is found at anterior-517
on March 18, 2018 by guest
http://mcb.asm
.org/D
ownloaded from
18
dorsal corner of the oocyte (arrows in A and B). (C-F) Confocal images of egg 518
chambers stained with antibodies against Grk (white) and actin (red). (C) Stage 9 wild 519
type egg chamber, (D-F) stage 9 egg chamber from flies expressing Spn-F〉C. In the 520
wild type, Grk protein is localized to the dorsal-anterior corner of the oocyte (arrow in 521
C), however, in egg chambers from flies expressing Spn-F〉C, Grk protein is localized 522
to ectopic actin clumps (arrow in D, E and in F). (G-J) Confocal images of egg 523
chambers stained with antibodies against Spectrin (white) and actin (red). (G) Stage 9 524
wild type egg chamber, (H-J) stage 9 egg chamber from flies expressing Spn-F〉C. 525
Spectrin protein is also localized to ectopic actin clumps (arrows in H-J). 526
527
Figure 4: Defects in osk mRNA and protein localization in flies expressing Spn-528
F〉C. (A-B) In situ osk mRNA localization in stage 10 egg chambers. A) Wild type, (B) 529
Spn-F〉C. (C-D) Confocal images of egg chambers stained with antibodies against Osk 530
(red). (C) Wild type, (D) Spn-F〉C. (E-F) く-galactosidase staining of Osk-く-GAL in egg 531
chambers from (E) wild type, (F) Spn-F〉C flies. (H-I) osk mRNA in embryos from (H) 532
wild type, (I) Spn-F〉C flies. Both osk mRNA (arrows in B and I) and Osk protein 533
(arrow in F) are mislocalized upon Spn-F〉C expression. 534
535
Figure 5: Microtubule organization and function is affected upon Spn-F〉C 536
expression. (A-B) Confocal images of egg chambers stained with antibodies against 537
tubulin (red). Stage 9 egg chambers from (A) wild type, (B) flies expressing Spn-F〉C. 538
The anterior to posterior tubulin gradient detected in the wild type (arrow in A) is 539
abolished upon Spn-F〉C expression (arrow in B). (C-H) く-galactosidase staining of 540
Nod く-GAL (C-D) and kinesin く-GAL (E-H). Wild type egg chamber (C and E) and 541
Spn-F〉C egg chambers (D, F-H). In flies expressing Spn-F〉C, wild type anterior Nod 542
く-GAL staining (arrows in C) is abolished (D). In wild type flies and in 3% of egg 543
chambers from flies expressing Spn-F〉C, kin:く-gal is accumulated at the oocyte 544
posterior (arrow in E). Kin:く-gal in flies expressing Spn-F〉C was either not detected 545
(F) or was found at the posterior end and in the middle of the oocyte (arrows in G) or 546
accumulated only at the center of the oocyte (arrow in H). 547
548
Figure 6: Expression of Spn-F〉C affects Ik2 activation. (A) Expression of Ik2 549
rescues the dorsal-ventral defects of Spn-F〉C expression. Whereas 94% (n=321) of 550
eggs laid by females expressing spn-F〉C produced a range of ventralized eggshells 551
on March 18, 2018 by guest
http://mcb.asm
.org/D
ownloaded from
19
(detailed description of these eggshells pattern is found in the legends to Fig. 2 C-F), 552
expression of Ik2 significantly suppressed these dorsal-ventral defects. (B-E) Confocal 553
images of egg chambers stained with antibodies against pIKK (red). (B-D) Egg 554
chambers from flies co-expressing GFP::Spn-F〉C and Flag-tagged Rad9. (E) Egg 555
chamber from flies co-expressing GFP::Spn-F〉C and GFP::Ik2. Whereas high 556
expression of Spn-F〉C during mid-oogenesis (arrows in B) eliminates Ik2 activation 557
(arrow in C), the lower expression levels seen in early oogenesis (arrows head in B) had 558
no effect on Ik2 activation (arrow head in C). (F) Western blot analysis of Ik2 levels in 559
ovaries. Ik2 levels in ovarian extracts from wild type and Spn-F〉C-expressing flies 560
were detected using antibodies against Ik2. The level of acetylated tubulin served as a 561
loading control. Spn-F〉C expression did not affect the level of Ik2 protein. 562
563
Figure 7: Spn-F〉C physically interacts with Ik2 but not with Jvl. (A) Schematic 564
presentation of Spn-F protein domains and deletion constructs used in this study. (B) 565
Co-immunoprecipitation of GFP-Spn-F constructs with Myc-Ik2 from S2 cells. GFP-566
Spn-F was precipitated using a Protein Trap kit and Myc-Ik2 was detected by western 567
blotting using anti-Ikk antibodies. Ik2 binds to the Spn-F〉C. (C) A yeast two hybrid 568
assay was used to detect the interaction between Spn-F and Jvl. Jvl binds to the 569
conserved C-terminal domain of Spn-F. (D) Co-immunoprecipitation of GFP-Jvl with 570
mCherry-Spn-F constructs from S2 cells. GFP-Jvl was precipitated using a Protein Trap 571
kit and mCherry-Spn-F was detected by western blotting using g–mCherry antibodies. 572
Jvl binds to the conserved C-terminal domain of Spn-F. 573
574
Figure 8: Spn-F〉C do not co-localized with Ik2 and Jvl. (A-D) Confocal images of 575
Schneider cell expressing mCherry-Jvl (A), Myc-tagged Ik2 (B, white pseudo color), 576
GFP-Spn-F (C) and GFP-Spn-F〉C (D). Confocal images of Schneider cells co-577
expressing either mCherry-Jvl, Myc-tagged Ik2 and GFP-Spn-F (E-G) or mCherry-Jvl, 578
Myc-tagged Ik2 and GFP-Spn-F〉C (H-J). Co-expression of mCherry-Jvl, Myc-tagged 579
Ik2 and GFP-Spn-F led to co-localization of all three proteins in a filamentous structure. 580
However, when GFP-Spn-F〉C was co-expressed with mCherry-Jvl and Myc-tagged 581
Ik2, the three proteins no longer co-localized. 582
583
584
on March 18, 2018 by guest
http://mcb.asm
.org/D
ownloaded from
20
585
586
587
References 588
1) Abdu U, Bar D, Schüpbach, T. 2006. spn-F encodes a novel protein that 589
affects oocyte patterning and bristle morphology in Drosophila. Development. 133: 590
1477-1484. 591
592
2) Shapiro RS, Anderson, KV. 2006. Drosophila Ik2, a member of the IせB 593
kinase family, is required for mRNA localization during oogenesis. Development. 594
133: 1467-1475. 595
596
3) Dubin-Bar D, Bitan A, Bakhrat A, Amsalem S, Abdu U. 2011. Drosophila 597
javelin-like encodes a novel microtubule-associated protein and is required for mRNA 598
localization during oogenesis. Development 138: 4661-4671. 599
600
4) Dubin-Bar D, Bitan A, Bakhrat A, Kaiden-Hasson R, Etzion S, Shaanan 601
S, Abdu U. 2008. The Drosophila IKK-related kinase (Ik2) and Spindle-F 602
proteins are part of a complex that regulates cytoskeleton organization during 603
oogenesis. BMC Cell Biol. 9: 51. 604
605
5) Bettencourt-Dias M, Giet R, Sinka R, Mazumdar A, Lock WG, Balloux F, 606
Zafiropoulos PJ, Yamaguchi S, Winter S, Carthew RW, Cooper M, Jones D, Frenz 607
L, Glover DM. 2004. Genome-Wide Survey of Protein Kinases Required for Cell 608
Cycle Progression. Nature. 432: 980-987. 609
610
6) Somma MP, Ceprani F, Bucciarelli E, Naim V, De Arcangelis V, Piergentili R, 611
Palena A, Ciapponi L, Giansanti MG, Pellacani C, Petrucci R, Cenci G, Vernì F, 612
Fasulo B, Goldberg ML, Di Cunto F, Gatti M. 2008. Identification of Drosophila 613
Mitotic Genes by Combining Co-Expression Analysis and RNA Interference. PLoS 614
Genet. 4, e1000126. 615
616
on March 18, 2018 by guest
http://mcb.asm
.org/D
ownloaded from
21
7) Lee HH, Jan LY, Jan YN. 2009. Drosophila IKK-related kinase Ik2 and Katanin 617
p60-like 1 regulate dendrite pruning of sensory neuron during metamorphosis. Proc. 618
Natl Acad Sci. USA. 106(15): 6363-8 619
620
8) Bitan A Guild GM, Bar-Dubin D, Abdu U. 2010 Asymmetric microtubule 621
function is an essential requirement for polarized organization of the Drosophila 622
bristle. Mol Cell Biol. 30(2): 496-507 623
624
9) Bitan A, Rosenbaum, I, Abdu U. 2012. Stable and Dynamic Microtubule 625
Networks Coordinate to Determine and Maintain Drosophila Bristle Shape. 626
Development. 139(11):1987-96. 627
628
10) Oshima K, Takeda M, Kuranaga E, Ueda R, Aigaki T, Miura M, Hayashi S, 629
2006. IKK epsilon regulates F-actin assembly and interacts with 630
Drosophila IAP1 in cellular morphogenesis. Curr Biol. 16(15): 1531-7. 631
632
11) Kuranaga E, Kanuka H, Tonoki A, Takemoto, K, Tomioka T, Kobayashi 633
M, Hayashi S, Miura M. 2006. Drosophila IKK-related kinase regulates 634
nonapoptotic function of caspases via degradation of IAPs. Cell. 126(3): 583-96. 635
636
12) Otani T, Oshima K, Onishi S, Takeda M, Shinmyozu K, Yonemura S, Hayashi 637
S. 2011. IKKセ Regulates Cell Elongation through Recycling Endosome 638
Shuttling. Dev. Cell. 20(2): 219-32. 639
640
13) Kadir R, Bakhrat A, Tokarsky R, Abdu, U. 2012. Localization of the Drosophila 641
rad9 protein to the nuclear membrane is regulated by the C-terminal region and is 642
affected in the meiotic checkpoint. PLoS One 7(5): e38010. 643
644
14) Clark I, Giniger, E, Ruohola-Baker, H, Jan LY, Jan YN. 1994. Transient 645
posterior localization of a Kinesin fusion protein reflects anteroposterior polarity of the 646
Drosophila oocyte. Curr Biol. 4: 289-300. 647
648
on March 18, 2018 by guest
http://mcb.asm
.org/D
ownloaded from
22
15) Clark IE, Jan LY, Jan YN. 1997. Reciprocal localization of Nod and Kinesin 649
fusion proteins indicates microtubule polarity in the Drosophila oocyte, epithelium, 650
neuron and muscle. Development. 124: 461–470. 651
652
16) Gunkel N, Yano T, Markussen FH, Olsen LC, Ephrussi A. 1998. 653
Localization-dependent translation requires a functional interaction between the 5ガ and 654
3ガ ends of oskar mRNA. Genes Dev. 12: 1652–1664. 655
656
17) Spradling AC, Rubin GM. 1982. Transposition of cloned P elements into 657
Drosophila germ line chromosomes. Science. 218: 341-7. 658
659
18) Peretz G, Bakhrat A, Abdu, U. 2007. Expression of the Drosophila 660
melanogaster GADD45 homologue (CG11086) affects egg asymmetric 661
development which is mediated by the p38/JNK pathway. Genetics 177: 1691-702. 662
663
19) Nusslein-Volhard C, Wieschaus E. 1980. Mutations affecting segment number 664
and polarity in Drosophila. Nature 287: 795-801 665
666
20) Roth S, Schüpbach, T. (1994). The relationship between ovarian and embryonic 667
dorsoventral patterning in Drosophila. Development. 120(8): 2245-57. 668
669
21) Queenan AM, Barcelo G, Van Buskirk C, Schupbach, T. 1999. The 670
transmembrane region of Gurken is not required for biological activity, but is necessary 671
for transport to the oocyte membrane in Drosophila. Mech Dev. 89: 35–42. 672
673
22) Vanzo NF, Ephrussi A. 2002. Oskar anchoring restricts pole plasm formation to 674
the posterior of the Drosophila oocyte. Development. 129: 3705-3714. 675
676
23) Neuman-Silberberg FS, Schüpbach T. 1993. The Drosophila dorsoventral 677
patterning gene gurken produces a dorsally localized RNA and encodes a TGF-g-like 678
protein. Cell 75: 165-174. 679
680
24) Neuman-Silberberg FS, Schüpbach T. 1996. The Drosophila TGF 681
alphalike protein Gurken: expression and cellular localization during Drosophila 682
on March 18, 2018 by guest
http://mcb.asm
.org/D
ownloaded from
23
oogenesis. Mech. Dev. 59: 105-113. 683
684
25) Gonzalez-Reyes A, Elliott H, St Johnston D. 1995. Polarization of both 685
major body axes in Drosophila by gurken-torpedo signalling. Nature. 375: 654-658. 686
687
26) Kugler JM, Chicoine J, Lasko P. 2009. Bicaudal-C associates with a Trailer 688
Hitch/Me31B complex and is required for efficient Gurken secretion. Dev Biol. 689
328(1):160-72. 690
691
27) Snee MJ, Macdonald PM. 2009. Bicaudal C and trailer hitch have similar roles in 692
gurken mRNA localization and cytoskeletal organization. Dev Biol. 328(2):434-44. 693
694
28) Berleth T, Burri M, Thoma G, Bopp D, Richstein, S, Frigerio G, Noll M, 695
Nusslein-Volhard C. 1988. The role of localization of bicoid RNA in organizing the 696
anterior pattern of the Drosophila embryo. EMBO J. 7: 1749-1756. 697
698
29) Ephrussi A, Dickinson LK, Lehmann R. 1991. oskar organizes the germ 699
plasm and directs localization of the posterior determinant nanos. Cell. 66: 37–50. 700
701
30) Kim-Ha J, Smith JL, Macdonald PM. 1991. Oskar mRNA is localized to the 702
posterior pole of the Drosophila oocyte. Cell. 66: 23–35. 703
704
31) Wharton RP, Struhl G. 1989. Structure of the Drosophila Bicaudal-D protein and 705
its role in localizing the posterior determinant nanos. Cell, 59: 881–892. 706
707
32) Cinnamon E, Gur-Wahnon D, Helman A, St Johnston D, Jiménez G, Paroush 708
Z. 2004. Capicua integrates input from two maternal systems in Drosophila terminal 709
patterning. EMBO J. 23(23): 4571-82. 710
711
33) Mahone M, Saffman EE, Lasko PF. 1995. Localized bicaudal-C RNA 712
encodes a protein containing a KH domain, the RNA binding motif of FMR1. EMBO J. 713
14: 2043–2055. 714
on March 18, 2018 by guest
http://mcb.asm
.org/D
ownloaded from
24
715
34) Smith JL, Wilson JE Macdonald PM. 1992. Overexpression of oskar directs 716
ectopic activation of nanos and presumptive pole cell formation in Drosophila embryos. 717
Cell. 70(5): 849-59. 718
719
35) Bullock S, Ish-Horowicz D. 2001. Conserved signals and machinery for RNA 720
transport in Drosophila oogenesis and embryogenesis. Nature. 414: 611–616. 721
722
36) Clark A, Meignin C, Davis I. 2007. A Dynein-dependent shortcut rapidly 723
delivers axis determination transcripts into the Drosophila oocyte. Development. 134: 724
1955–1965. 725
726
37) Theurkauf WE, Alberts BM, Jan YN, Jongens TA. 1993. A central role for 727
microtubules in the differentiation of Drosophila oocytes. Development. 118(4):1169-728
80. 729
730
38) Brendza RP, Serbus LR, Duffy JB, Saxton, WM. 2000. A Function for Kinesin I 731
in the Posterior Transport of oskar mRNA and Staufen Protein. Science. 289: 2120-732
2122. 733
734
39) Krauss J, López de Quinto S. Nüsslein-Volhard C, Ephrussi A. 2009. Myosin V 735
regulates oskar mRNA localization in the Drosophila oocyte. Curr Biol. 19(12): 1058-736
1063. 737
738
40) Glotzer JB, Saffrich R, Glotzer M, Ephrussi A. 1997. Cytoplasmic flows localize 739
injected oskar RNA in Drosophila oocytes. Curr Biol. 7(5): 326-37. 740
741
41) Cha BJ, Serbus LR, Koppetsch BS, Theurkauf WE. 2002. Kinesin Idependent 742
cortical exclusion restricts pole plasm to the oocyte posterior. Nat Cell Biol. 4: 592-598. 743
744
42) Zimyanin VL, Belaya K, Pecreaux J, Gilchrist MJ, Clark A, Davis I, St 745
Johnston D. 2008. In vivo imaging of oskar mRNA transport reveals the mechanism of 746
posterior localization. Cell 134: 843-853 747
748
on March 18, 2018 by guest
http://mcb.asm
.org/D
ownloaded from
25
43) Brendza R, Serbus L, Saxton W, Duffy J. 2002. Posterior localization of Dynein 749
and dorsal-ventral axis formation depend on Kinesin in Drosophila oocytes. Curr. Biol. 750
12: 1541–1545. 751
752
44) Januschke J, Gervais L, Dass S, Kaltschmidt JA, Lopez-Schier H, St Johnston 753
D, Brand AH, Roth S, Guichet A. 2002. Polar transport in the Drosophila oocyte 754
requires Dynein and Kinesin I cooperation. Curr. Biol.12: 1971-1981. 755
756
45) Palacios IM, St Johnston D. 2002. Kinesin light chain-independent function of the 757
kinesin heavy chain in cytoplasmic streaming and posterior localisation in the 758
Drosophila oocyte. Development 129: 5473–5485. 759
760
761
762
763
764
765
766
767
768
on March 18, 2018 by guest
http://mcb.asm
.org/D
ownloaded from