First published online 2 February 2005
doi: 10.1242/dev.01645
Development 132, 977-986 (2005)
Published by The Company of Biologists 2005
aPKC, Crumbs3 and Lgl2 control apicobasal polarity in early vertebrate development
Andrew D. Chalmers1,2,
Michael Pambos1,2,
Julia Mason1,2,
Stephanie Lang3,
Chris Wylie3 and
Nancy Papalopulu1,2,*
1 Wellcome Trust/Cancer Research UK Gurdon Institute, Tennis Court Road,
Cambridge CB2 1QR, UK
2 Department of Anatomy, University of Cambridge, Downing Site, Cambridge CB2
3DY, UK
3 Division of Developmental Biology, Children's Hospital Medical Center, 3333
Burnet Avenue, Cincinnati, OH 45229-3039, USA
View larger version (80K):
[in a new window]
|
Fig. 1. aPKC, Crumbs3 and Lgl2 show specific localisation in early epithelial
cells. (A) GFP-Lgl2 localised exclusively to the basolateral membrane at stage
8. (B) Crumbs3-GFP localised to the apical (arrowhead) and basolateral
membrane at stage 8, and to unknown internal structures (arrow). (C) GFP
control. The examples shown are after injecting 1 ng of RNA. (D) RLDX control.
GFP localised nonspecifically in the cytoplasm, nucleus and points of cell
contact, as did the lineage label RLDX. Because of the high yolk content of
early Xenopus cells, cytoplasmic fluorescence of the controls has a
latticed appearance. This is very distinct from the localisation of the fusion
proteins shown. (E) Antibody staining showing that aPKC localises to the
apical membrane.
|
|
View larger version (61K):
[in a new window]
|
Fig. 2. aPKC overexpression produces rounded, protruding, hyper-pigmented cells.
(A,B) Mouse and Xenopus aPKC overexpression produced embryos with
protruding superficial cells and extended pigmented (apical) surface when
compared with controls (C,D). (E) Overexpression of a truncated version of the
Xenopus tropicalis protein, PKC NT, which lacks the entire kinase
domain, failed to produce this phenotype. (F) Crumbs3 overexpression
caused cell protrusion and over-apicalisation, similar to that of aPKC, but
was less effective in that the percentage of affected embryos was lower.
Quantification was carried out blind, by counting the number of embryos with
protruding cells. Right panels show the percentage of affected embryos at each
concentration of injected RNA. Each experiment was carried out at least three
times and the average is shown.
|
|
View larger version (76K):
[in a new window]
|
Fig. 7. aPKC and Lgl act by a process of mutual inhibition. (A,B) GFP-Lgl was
injected on its own (A) or with aPKC (B). Addition of aPKC inhibited the
basolateral localisation of GFP-Lgl2. GFP was visualised by using an anti-GFP
antibody. (C,D) Overexpression of Lgl2 inhibited the apical localisation of
aPKC but overexpression of GFP did not. (E-G) Lgl2, but not
GFP injections, can rescue the apicalisation caused by injecting
aPKC. There are more rounded cells in aPKC plus
GFP-injected embryos than in aPKC plus
Lgl2-injected embryos. The graph shows the average percent of embryos
with apicalised cells from three experiments. The experiment was scored blind
as for Fig. 2.
|
|
View larger version (96K):
[in a new window]
|
Fig. 3. aPKC is sufficient to promote apical and inhibit basal lateral membrane
identity without disrupting tight junctions. (A,B) aPKC
overexpression (B) caused expansion of the apical marker keratin compared with
GFP control (A). (C-F) aPKC caused reduction in the basolateral
markers, occludin (D) and ß1-integrin (F) compared with controls (C,E).
(G,H) aPKC caused tight junctions (as marked by cingulin) to be
maintained but relocated to the new apicobasolateral border. The borders of
the markers used in each panel are delineated with arrows. (I,J) aPKC staining
in GFP-injected controls (I) and aPKC staining in
aPKC-injected (J) embryos. The apicalised cells have inherited
overexpressed aPKC. (K) Diagrammatic representation of the result; aPKC causes
protruding hyper-apical cells, which still have tight junction markers.
Apical, red; basolateral, black; tight junctions, green. Albino embryos were
injected with aPKC RNA and stained for antibody markers of cell
polarity. Each experiment was carried out three times.
|
|
View larger version (79K):
[in a new window]
|
Fig. 4. Loss of aPKC function expands basolateral membrane domain into the apical
side and disrupts the apical domain. (A) The aPKC NT construct has the
Par6-binding site but no kinase domain and so acts as a dominant-negative
fragment. (B) The effect of this construct can be rescued by overexpressing
full-length aPKC. Injections of 4.5 ng aPKC NT + 0.5 ng GFP, 4.5 ng aPKC NT +
0.5 ng aPKC, or 5 ng GFP were carried out. The average of four experiments
scored blind is shown. (C,D) aPKC NT dominant-negative fragment caused pigment
defects (D) compared with control (C) (5 ng of each). (E,H) aPKC NT
was co-injected with GFP showing that the pigment defects occurred in
the injected region. The arrows in C,E and D,H highlight the same cells. (F-J)
aPKC NT (I,J) caused ectopic localisation of the basolateral markers
ß1-integrin and occludin to the apical side (arrow) and tight junctions
were also lost (J, arrowhead) when compared with GFP control (F,G). (K)
Diagram of the observed phenotype. Colours are as above. Pigmented embryos
were injected as this allowed the affected area to be easily identified, they
were then fixed and stained for markers of cell polarity.
|
|
View larger version (90K):
[in a new window]
|
Fig. 5. Lgl2 promotes basolateral and inhibits apical identity. (A) Injection of 5
ng GFP did not affect the cells. (B,C) Injection of Xenopus
Lgl2 caused loss of pigment and also a block in cytokinesis at high doses
(B, 5 ng; C, 0.5 ng). (D-K) Injection of GFP (D-G) or Lgl2
(H-K) and immunostaining with the markers shown in each panel.
GFP-injected embryos were entirely normal. (I) Injection of
Lgl2 caused a reduction in keratin to the levels normally seen in the
basolateral region (arrow) and loss of tight junctions (cingulin, arrowhead).
(J,K) Injection of Lgl2 caused ectopic localisation of
ß1-integrin (J) and occludin (K) to the apical side (arrow) and loss of
tight junctions (arrowhead). (L) diagrammatic representation of phenotype,
colours as above. Experiments were carried out three times in both albino and
pigmented embryos (except for the keratin where the staining is obscured by
the pigment and therefore was carried out only in albinos), and the same
result was obtained in both.
|
|
View larger version (72K):
[in a new window]
|
Fig. 6. Time-lapse images showing the gradual but direct depigmentation of the
apical side by Lgl2. A pigmented embryo was injected animally with
Lgl2 RNA at the two-cell stage and filmed. A small site of
cytoplasmic leakage helps to verify the site of injection. Evidence of apical
membrane disruption starts as a concentration of pigment spots (arrow) that
appear quite suddenly and spread quickly. The even distribution of
pigmentation is lost and the pigment is gradually cleared from the apical
side. Interestingly, pigment becomes concentrated to the periphery of the
apical domain. There is no evidence of inner cells coming to the surface of
the embryo or outer cells falling in.
|
|
View larger version (12K):
[in a new window]
|
Fig. 8. A model showing the antagonistic action of aPKC and Lgl2 in maintaining the
apical and basolateral domain. Increased aPKC causes expansion of the apical
domain (red), while reduced aPKC or increased Lgl2 causes expansion of the
basolateral domain (black). Tight junctions are shown in green.
|
|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati 
Twitter What's this?
© The Company of Biologists Ltd 2005
