First published online August 4, 2003
doi: 10.1242/10.1242/dev.00617
Mechanism of inhibition of the Drosophila and mammalian EGF receptors by the transmembrane protein Kekkon 1
Christian Ghiglione1,2,*,
Laufey Amundadottir1,3,*,
Margret Andresdottir3,
David Bilder1,4,
John A. Diamonti5,6,
Stéphane Noselli2,
Norbert Perrimon1,7,
and
Kermit L. Carraway III5,6,
1 Department of Genetics, Harvard Medical School, 200 Longwood Avenue, Boston,
MA 02115, USA
2 Centre de Biochimie, UMR6543/CNRS, Faculté des Sciences, 06108 Nice,
France
3 Division of Cancer Genetics, deCODE Genetics, Sturlugata 8, 108 Reykjavik,
Iceland
4 Department of Molecular and Cell Biology, University of California, Berkeley,
CA 94720-3200, USA
5 Division of Signal Transduction, Beth Israel Deaconess Medical Center, 330
Brookline Avenue, Boston, MA 02215, USA
6 UC Davis Cancer Center, 4645 2nd Avenue, Sacramento, CA 95817, USA
7 Howard Hughes Medical Institute, Harvard Medical School, 200 Longwood Avenue,
Boston, MA 02115, USA
View larger version (79K):
[in a new window]
|
Fig. 1. kek1 expression in wing and eye discs. Expression of the
kek1-lacZ enhancer trap line in the third instar wing and eye discs
(A and D, respectively). Ectopic expression of UAS-DERDN in the wing
pouch using MS1096-Gal4 and behind the morphogenetic furrow of the eye disc
using GMR-Gal4 strongly reduce kek1-lacZ expression (C and F,
respectively). Conversely, expression of an activated DER
(UAS- top) using the same drivers leads to an expansion of
kek1-lacZ expression in the corresponding domains of the wing (B) and
eye discs (E).
|
|
View larger version (184K):
[in a new window]
|
Fig. 2. Kek1 antagonizes DER activity in the eye and wing. (A-H) Eye phenotypes.
(A) A wild-type adult eye posses around 750 ommatidia arranged in a highly
ordered pattern. (B) Eyes from kek1 mutants look wild type. (C)
Elp/+ eyes are rough and this phenotype is strongly enhanced when
homozygous for kek1 (D). Elp/Elp eye (E).
UAS-kek1/GMR-Gal4 (F) and UAS-(kek1)2/GMR-Gal4
(G) are rough and reduced in size, similar to UAS-DERDN/GMR-Gal4 eyes
(H). (I-N) Wing phenotypes. (I) A wild-type adult wing with its five
longitudinal veins and two crossveins. (J) Wings from kek1 mutants
look wild type. (K) Elp/+ wings have a weak extra wing vein phenotype
(arrows), and this phenotype is enhanced when homozygous for kek1
(L). Overexpression of UAS-kek1 using MS1096-Gal4 results in severe
reduction in the vein material (M), similar to UAS-DERDN
overexpression (N).
|
|
View larger version (23K):
[in a new window]
|
Fig. 3. Structure-function analysis of Kek1. (A) Schematic representation of the
different Kek1 constructs: SP, signal peptide; NT, N-terminal region; LRR,
Leucine-Rich Repeat domains; Ig, Immunoglobulin-like domain; TM, transmembrane
domain. The relative efficiencies of these truncated proteins to inhibit DER
signalling after overexpression in the follicle cells and the wing imaginal
discs are indicated. (B) In vitro association between Kek1 and DER. Sf9 cells
were infected with baculovirus encoding DER and co-infected with nothing (lane
1) or viruses encoding Myc-tagged versions of wild-type Kek1 (lane 4),
Kek1 LRR (lane 2), Kek1Ig (lane 3). Lane 5 is a control with Sf9
infected with baculovirus encoding Kek1-Myc alone. Anti-DER and anti-Myc
immunoprecipitates, lane 1 and lane 2-5, respectively, were blotted with
anti-DER (upper panel) and then reprobed with anti-Myc (lower panel).
|
|
View larger version (59K):
[in a new window]
|
Fig. 4. Activity of Kek1-DER chimeras. (A) Schematic representation of the
different Kek1-DER chimeras. (B)Wild-type egg with its two dorsal appendages.
The eggs laid by females UAS-kek1-DER/T155-Gal4 (C) and UAS-kek1
DIg-DER/T155-Gal4 (E) are strongly dorsalized. (D) Eggs laid by females
UAS-kek1 DLRR-DER/T155-Gal4 are wild type. Eggs obtained after overexpressing
UAS-kek1-DER in the follicle cells of top homozygous females are strongly
ventralized (F), similar to eggs laid by top/top females (G).
|
|
View larger version (68K):
[in a new window]
|
Fig. 5. The cytoplasmic domain of Kek1 is required for subcellular localization.
(A-D) Confocal microscope sections showing anti-Myc immunostaining after
overexpression of Myc-tagged UAS-kek1 and UAS-kek1TM in
embryos (A and B, respectively) and wing imaginal discs (C and D,
respectively) by using en-Gal4 as a driver. Apicobasal polarity is shown
(a b) with apical orientation upwards. Apical surface of wing imaginal
discs is revealed with rhodamine-phalloidin. Anti-Myc staining is in green and
rhodamine-phalloidin is in red. (E,F) Resulting UAS-kek1/en-Gal4 and
UAS-kek1TM/en-Gal4 adult wings (compare with a wild-type wing in
Fig. 2I).
|
|
View larger version (32K):
[in a new window]
|
Fig. 6. Kek1 association and inhibition of human EGFR. (A) Association of Kek1-HA
with human EGFR in two stably transfected HEK293-Ecr cell lines. Cells were
treated without or with Ponasterone A (PonA) to induce Kek1-HA expression, and
lysates were immunoprecipitated (IP) with antibodies to EGFR. Precipitates
were then immunoblotted with anti-HA. Cell lysate (right lane) was included as
a positive control for blotting. (B,C) Inhibition of human EGFR signalling in
HEK293 cells by Kek1. 293-Ecr stably transfected cells (clone 4) were treated
without and with PonA for 24 hours, and then treated without and with EGF as
indicated. (B) Inhibition of EGFR autophosphorylation: upper panel, lysates
from treated cells were immunoprecipitated with anti-phosphotyrosine
antibodies; precipitates were blotted with anti-EGFR. Lower panel, lysates
were blotted with anti-HA to detect Kek1-HA expression. (C) Inhibition of
Erk1/2 activation: lysates from treated cells were blotted with antibodies
specific for phosphorylated Erk1 and Erk2 (upper panel) and re-probed with
antibodies that recognize the total Erk2 population (middle panel). Bands were
quantified and relative Erk activity plotted (lower panel).
|
|
View larger version (60K):
[in a new window]
|
Fig. 7. Mechanism of EGFR inhibition by Kek1. (A) Inhibition of EGF binding and
EGF-stimulated receptor tyrosine phosphorylation by Kek1. Sf9 insect cells
were infected with baculoviruses encoding either Kek1-Myc or EGFR, or
co-infected with both viruses. Cells were treated without or with 30 nM EGF as
indicated. For the [125I]-labeled EGF crosslinking experiment
(lower panel), trace levels (0.1 nM) of iodinated growth factor and 1 mM
BS3 crosslinker were added to all samples at the time of EGF
addition. Lysates from cells were immunoprecipitated with antibodies to either
Myc epitope or to EGFR. Precipitates were exposed to autoradiography (lower
panel), or were blotted with antibodies to phosphotyrosine (upper panel) or
EGFR (middle panel). (B) Co-localization of Kek1 and EGFR at the cell surface.
Sf9 insect cells were infected at a low multiplicity of infection with
baculoviruses encoding Kek1-Myc and human EGFR. Cells were fixed and stained
with both rabbit anti-EGFR (left panel) and mouse anti-Myc epitope (middle
panel). Images were merged to show co-localization (right panel). (C,D) Kek1
can inhibit DER in trans. (C) 6% of the eggs (n=112) laid by females
UASp-kek1/nos-Gal4 are weakly ventralized (partial or total fusion of
the dorsal appendages). (D) Among the 64% of the ventralized eggs
(n=96) laid by top/+; UASp-kek1/nos-Gal4 females,
8% are strongly ventralized.
|
|
View larger version (26K):
[in a new window]
|
Fig. 8. Inhibition of mammalian cell growth by Kek1. (A) Inhibition of
anchorage-dependent cell growth. The growth rate of cells stably transfected
with vector alone (v.o.) or cells transfected with epitope-tagged Kek1 were
compared for MDAMB-468 human mammary tumor cells, and NF-639, IJ9921 and
AC-816 mouse mammary tumor cells. Growth rates of HEK293 cells treated without
and with Kek1 induction by PonA were also compared. Experiments were carried
out in triplicate and repeated at least three times. (B) Inhibition of
tumorigenic growth properties by Kek1. The growth of cells in soft agar or as
tumors in nude mice was compared. Plotted is the percent inhibition by Kek1
transfectants relative to vector alone transfectants. Error bars represent the
standard error of the mean of three to six determinations. Experiments were
repeated at least three times.
|
|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati 
Twitter What's this?
© The Company of Biologists Ltd 2003
