First published online January 10, 2007
doi: 10.1242/10.1242/dev.02754
Development 134, 591-600 (2007)
Published by The Company of Biologists 2007
The glycosyltransferase Fringe promotes Delta-Notch signaling between neurons and glia, and is required for subtype-specific glial gene expression
Graham B. Thomas1 and
Donald J. van Meyel2,3,4,*
1 Graduate Program in Neurological Sciences, 1650 Cedar Avenue, Montreal, QC,
H3G 1A4, Canada.
2 Centre for Research in Neuroscience, 1650 Cedar Avenue, Montreal, QC, H3G 1A4,
Canada.
3 Department of Neurology and Neurosurgery, McGill University, 1650 Cedar
Avenue, Montreal, QC, H3G 1A4, Canada.
4 McGill University Health Centre Research Institute, 1650 Cedar Avenue,
Montreal, QC, H3G 1A4, Canada.
View larger version (113K):
[in this window]
[in a new window]
|
Fig. 1. Fng expression in a subset of longitudinal glia (LG). (A-C)
Dorsal view of the nervous system of a filleted wild-type embryo at stage
16-17. (A) Anti-HRP (green) reveals the brain lobes (asterisks) and the major
axon tracts of the VNC. (B) Repo-positive lateral glia (magenta). (C) Overlay
of A and B. (D) Three segments of the VNC of a htl-Gal4,
UAS-nGFP embryo. Anti-HRP (green) shows the anterior and posterior
commissures in each segment, and the two longitudinal connectives over which
the LG (anti-GFP, magenta) are positioned. (E) The six most anterior LG
in each hemisegment express Pros (magenta), as do a small number of neurons
(arrow). (F) Overlay of D and E without anti-HRP. Note that the Pros
channel in D has been converted to green in F, clarifying the distinction
between the anterior LG, which express Pros (white), and posterior LG, which
do not (magenta). (G) Whole-mount fng in situ hybridization of
a stage 15 embryo. Transcripts for fng are observed in a two rows of
cells at the dorsal surface of the VNC. (H) Fluorescence in situ
hybridization for fng (red), showing a single hemisegment with axons
(anti-HRP, blue) and the commissures labeled. (I) The six anterior LG
coexpress fng mRNA (red) and Repo (green). Anterior is at the top in
all panels of all figures. Scale bars: 50 µm in A for A-C; 10 µm in D
for D-F. AC, anterior
commissure; PC, posterior commissure.
|
|
View larger version (80K):
[in this window]
[in a new window]
|
Fig. 2. Mutations in fng cause defects of Pros expression in LG. All panels
except C and F show three segments of stage 15-16 VNC stained with anti-HRP
(green) and co-stained for either Pros or ß-galactosidase, as indicated
(magenta). (A) fngL73 heterozygotes, like wild
type, have six Pros-positive LG per hemisegment. (B) The reporter
loco-lacZ reveals the number and position of LG in fng
heterozygotes. (C) Overlay of D and E without anti-HRP, and with Pros
converted to green. (D) fng mutants have fewer Prospositive
LG, and the level of Pros expression is often reduced. (E) The number
and position of LG in fng mutants appears normal. (F) Overlay
of D and E. (G) fngDf/+ hemizygotes also have six
Pros-positive LG per hemisegment as normal. (H) Reduced Pros expression
in an fng13/Df embryo (D). (I) Pros expression is
rescued in fng13/Df mutants using repo-Gal4 to
express UAS-fng. (J) Quantification of Pros-positive LG per
hemisegment. Every Pros-expressing cell in the vicinity of the neuropil was
counted as positive, regardless of staining intensity. The number of abdominal
hemisegments counted for each genotype is indicated (N). Because one of the
six Pros-expressing LG per hemisegment is often obscured in the compiled
images, the results were binned as five or six Prosexpressing LG, versus four
or fewer.
|
|
View larger version (81K):
[in this window]
[in a new window]
|
Fig. 3. Expression of N and its ligands in the VNC. (A-F) Stage 15
embryo (genotype repo-Gal4;UAS-mCD8GFP) co-labeled for N (red), GFP
(blue) and HRP (green). (A) N is found on most neurons and is highly expressed
in the neuropil. (B,C) Single optical section dorsal to the neuropil reveals N
expression (red) along the perimeter of many LG marked by the
membrane-targeted GFP. (B',C') In a cross-section through LG and
the longitudinal connectives, N co-localizes with the membrane of LG
(arrowheads) and channel glia (arrow). (D,E) N expression overlaps with
anti-HRP in the dorsal neuropil (arrowheads in E), while HRP is excluded from
glia (F). The dotted line in B indicates the anteroposterior (a-p) level of
the cross-section in B',C', and D-F. The dotted line in B'
indicates the dorsoventral (d-v) level of section for B and C. (G) Dl
expression in a cluster of cells (arrow) near the lateral edge of the VNC of a
stage 15-16 embryo. (H) Dl is expressed on proximal axons (arrowheads
in G,H) although it is largely excluded from the distal axons in the region of
the longitudinal connectives. (I) Ser is expressed on the cell bodies
(arrowheads) and axons of two or three neurons in each hemisegment. Scale bar:
10 µm. (J) Cartoon depicting a summary of the expression of N, Dl
and Ser relative to expression of Pros in the anterior LG.
|
|
View larger version (74K):
[in this window]
[in a new window]
|
Fig. 4. N activity specifically influences Pros expression in LG. All panels
show three segments of a stage 15-16 VNC with axons (anti-HRP) shown in green
in A,C,G,H. All other antibodies are as labeled. htl-Gal4 was used to
drive expression of UAS transgenes. (A) Misexpression of Hairless (H)
abolishes nearly all Pros expression in LG, while Pros in neurons is
unaffected (arrowhead). (B) Overlay of Pros in A with anti-GFP reveals
that H does not alter the number or positioning of LG. (C)
Misexpression of Numb also abolishes Pros expression. (D,E) In
wild type, the enzyme Gs2 is restricted to the six anterior LG. (F) In
contrast to Pros, Gs2 expression in the anterior LG is unaffected by
misexpression of Numb. (G) Misexpression of NICD causes Pros
expression in most or all LG, but does not alter the number or position of LG
(H). (I) Overlay of G and H without anti-HRP and with Pros
converted to green. (J) Quantification of Pros-positive LG per
hemisegment: misexpression studies. Quantification performed as in
Fig. 2J, with results binned as
shown in the legend.
|
|
View larger version (123K):
[in this window]
[in a new window]
|
Fig. 5. Pros expression in LG depends on Dl-N signaling through neuron-glial
interactions. All panels (except C and H) show three segments of a stage
15-16 VNC with the axon tracts shown in green (anti-HRP) and Pros in magenta.
(A) Misexpression of UAS-fng with htl-Gal4
leads to ectopic expression of Pros in posterior LG, labeled 7 and 8.
(B,C) Ectopic Pros is also observed upon misexpression of either
Dl (B) or Ser (not shown) in postmitotic neurons using
elavC155-Gal4. (C) Co-labeling for Pros (green) and the LG
marker Distal-less (magenta) shows that the ectopic Prospositive cells are
indeed posterior LG. (D) By contrast, Pros is inhibited upon
misexpression of Dl in LG with htl-Gal4. (E) Pros expression
in fng mutants is further reduced in embryos also heterozygous for
Dlrev10 (compare with
Fig. 2D). (F) Pros
expression is unaffected in Serrev2-11/SerRx82
mutants. (G) Fewer cells express Pros in embryos in which misexpression
of UAS-Tom is driven by scrt11-6-Gal4 in
neuronal lineages. (H) Three segments of a stage 16-17 embryo with
scrt11-6-Gal4 driving UAS-Tom. Anti-Eve
immunochemistry revealed no transformations of cell fate of RP2 (arrows) or
other Eve-positive neurons. (I) Pros expression is unaffected by Tom
misexpression in LG.
|
|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati 
Twitter What's this?
© The Company of Biologists Ltd 2007
