First published online 3 August 2005
doi: 10.1242/dev.01968
Development 132, 3947-3961 (2005)
Published by The Company of Biologists 2005
Genomic characterisation of a Fgf-regulated gradient-based neocortical protomap
Stephen N. Sansom1,2,
Jean M. Hébert3,4,
Uruporn Thammongkol1,2,
James Smith1,
Grace Nisbet1,
M. Azim Surani1,5,
Susan K. McConnell3 and
Frederick J. Livesey1,2,*
1 Wellcome Trust/CR UK Gurdon Institute, University of Cambridge, Tennis Court
Road, Cambridge CB2 1QN, UK
2 Department of Biochemistry, University of Cambridge, Tennis Court Road,
Cambridge CB2 1QN, UK
3 Department of Biological Sciences, Gilbert Hall, Stanford University,
Stanford, CA 94305, USA
4 Department of Neuroscience, Albert Einstein College of Medicine of Yeshiva
University, Rose F. Kennedy Center, 1410 Pelham Parkway South, Room 903,
Bronx, NY 10461, USA
5 The Physiological Laboratory, University of Cambridge, Downing Site, Cambridge
CB2 3EG, UK
View larger version (32K):
[in a new window]
|
Fig. 1. A genomics-based investigation into the nature and composition of a
neocortical protomap successfully identified genes with described spatial
differences in expression. (A) Strategy for a genomics-based identification of
protomap components. Rostral and caudal thirds of the neocortex were dissected
from mouse embryos at two ages (E11.5 and E13.5) in both inbred and outbred
strains. Comparison of gene expression between rostral and caudal tissue was
carried out within single litters of embryos to reduce variability in the
developmental stages of embryos. In addition to spatial gene expression, the
same samples were also used to analyse temporal changes in gene expression.
(B) Screen efficacy was demonstrated by the successful identification of
almost all of the genes that have been previously described as having
differential rostrocaudal expression, including COUP-TFI, Lhx2, Sfrp1
and Pax6. Plots of gene expression ratios (expressed in base 2)
between rostral and caudal neocortex are shown. The graphs show changes in
gene expression for single genes for multiple dye-swapped samples. The stage
and strain of origin of the samples is as indicated, with each datapoint
representing a measurement from a single microarray. Measurements are
presented as dye-swapped pairs, producing the alternating positive and
negative expression patterns. The two caudally expressed genes
COUP-TFI and Lhx2 show an opposite expression pattern to the
two rostral genes Sfrp1 and Pax6.
|
|
View larger version (40K):
[in a new window]
|
Fig. 2. Comprehensive identification of genes demonstrating differences in
expression along the rostrocaudal axis of the developing neocortex. (A)
Rostrocaudal gene expression differences in the neocortex at E11.5. Scatter
plot of observed gene expression ratios against expected ratios, as calculated
using the significance analysis of microarrays algorithm (SAM, see text for
details). (B) Clustering of genes from A with rostrocaudal expression
differences at E11.5 passing a 5% false discovery rate cutoff in SAM. Each
column represents a single microarray and each row expression data for a
single gene. Alternate microarrays are dye-swaps (technical replicate) of the
preceeding microarray, the ratios of which have been reversed. The 10
microarrays shown here represent five dye-swapped pairs, and the analysis of
five separate samples of rostral and caudal neocortical tissue. The numbers of
rostral and caudal genes identified are indicated in brackets. By convention,
positive differences in expression (upregulation) are represented in red, and
negative differences (downregulation) in green, with the colour intensity
reflecting the magnitude of the underlying expression ratio. (C) Rostrocaudal
gene expression differences at E13.5. Scatter plot of observed gene expression
ratios against expected ratios, as calculated using the significance analysis
of microarrays algorithm (SAM, see text for details). (D) Clustering of genes
from C with rostrocaudal expression differences at E13.5 passing a 5% false
discovery rate cutoff in SAM. (E) Combined analysis of E11.5 and E13.5 data.
Scatter plot of observed gene expression ratios against expected ratios, as
calculated using the significance analysis of microarrays algorithm (SAM, see
text for details). (F) Clustering of genes passing 5% false discovery rate
cutoff in SAM from E. (G) An array screen for genes demonstrating periodic
gene expression along the rostrocaudal axis of the developing neocortex. E13
neocortices were divided into thirds along the rostrocaudal axis, and tissue
pooled from single cortical hemispheres from four different embryos to
generate three separate pools of tissue for each third (three pools each of
rostral, middle and caudal tissue). Global gene expression was compared
between every possible pair of types of pool in a set of 18 independent array
hybridisations. The resulting cluster of genes identified as reproducibly
rostrally or caudally enriched, compared with middle and rostral or caudal
tissue, is shown. (H) Genes identified as more highly expressed in the caudal
neocortex and expressed in gradients, or as more highly expressed in the
rostral neocortex and expressed in gradients, show the appropriate rank order
of expression by microarray analysis. For example, the magnitude of the
expression ratio is highest comparing caudal and rostral samples, and these
genes are expressed at higher levels in caudal samples than middle, and in
middle samples than rostral.
|
|
View larger version (77K):
[in a new window]
|
Fig. 3. Spatial gene expression across the field of neocortical progenitor cells is
in gradients, rather than discrete domains. (A) Expression of rostrally
enriched transcripts in neocortical progenitor cells at E11.5 and E13.5. In
situ hybridisation on 14-µm parasagittal sections at each age. Rostrocaudal
orientation and gene identifiers are as indicated. (B) Expression of caudally
enriched transcripts at E11.5 and E13.5. (C) Expression of genes predicted
from array analyses as temporally regulated and correlating with the
neurogenetic gradient. Both genes are expressed rostrally initially (E11.5)
and subsequently (E13.5) expression spreads caudally. (D) Whole-mount in situ
hybridisation for caudally enriched transcripts. With the exception of the
EphA3 whole-mount, which was carried out at E12, the embryos are all E11.5.
Genes are as labelled in each panel. Arrow indicates rostral forebrain
expression. Scale bars: 0.5 mm.
|
|
View larger version (34K):
[in a new window]
|
Fig. 4. Candidate protomap genes are up- and downregulated in Fgfr1 mutant
cortices in a manner consistent with their predicted regulation. (A) Cluster
of genes up- and downregulated in the conditional Fgfr1 mutant dorsal
telencephalon at E12.5. Nine independent comparisons of gene expression
between single homozygous and heterozygous null mutant embryos were carried
out and the data analysed by t-test to identify reproducible
differences in expression between those two genotypes. (B) Examples of the
expression in the Fgfr1 mutant cortex of genes normally expressed
caudally (green) and rostrally (red) in the E13 neocortex. Caudal genes are
upregulated, whereas rostral genes are downregulated in the Fgfr1
E12.5 mutant cortex. Average log base 2 (mean from nine independent
hybridisations) gene expression values are shown. Gene names are as indicated:
Osr1, odd-skipped related 1; Dct, dopachrome tautomerase.
(C) Neurogenic genes and genes expressed in differentiating neurons are
upregulated in the Fgfr1 mutant E12.5 neocortex. Genes shown include
a neurogenic gene expressed in progenitor cells (neurogenin 2), transcription
factor genes expressed in differentiating neurons (Neurod, Myt1l, Tbr1,
Tbr2, Lmo1) and genes expressed in differentiated neurons
(Epha3, alpha-synuclein, chromogranin B, Snap25).
|
|
View larger version (22K):
[in a new window]
|
Fig. 5. Neocortical progenitor cell expression of Mest is directly
regulated by Fgf8 signalling. (A) Explants of the middle third of the E11.5
neocortex were cultured on polycarbonate membranes in defined, serum-free
medium, as shown. Two explants were used for each single treatment. (B) Four
hours of Fgf8 exposure strongly induced the expression of Mest, Etv5
and Pax6 and strongly repressed expression of COUP-TF1. No
significant changes in the expression of Hey1 and Emx2 were
seen. Histograms of the expression levels of all six genes relative to the
average expression level of each gene from four control, non-Fgf8-treated
experiments are shown. Results from three independent experiments are shown,
with each experiment containing two neocortical explants. Expression levels
within each sample were normalised to that of the abundant transcript encoding
the ribosomal protein rpS17, whose expression does not change in this
system.
|
|
View larger version (39K):
[in a new window]
|
Fig. 6. Mest encodes a highly conserved, vertebrate-specific hydrolytic
enzyme. (A) Phylogram of the multiple sequence alignment of the predicted
protein sequences for Mest from several vertebrate and invertebrate genomes.
Organisms are as labelled. Note that the most similar proteins from C.
elegans, D. melanogaster and Ciona intestinalis group together
and are relatively dissimilar from the other proteins. In all three organisms,
the most similar protein in the mouse or human genome is not Mest but is
instead a hydrolase lacking the hydrophobic N terminus. (B) Multiple sequence
alignment of the vertebrate Mest protein sequences demonstrate a very high
level of conservation among the different organisms.
|
|
View larger version (38K):
[in a new window]
|
Fig. 7. Functional analysis of a candidate protomap component, Mest/Peg1, reveals a
role in Fgf8-mediated patterning of rostral cortex. (A) Gene expression
changes in the E12.5 Mest mutant neocortex. A hierarchical cluster
analysis of the set of genes showing statistically significant differences in
gene expression between E12.5 Mest mutant and wild-type cortices. (B)
Changes in the expression of genes showing higher rostral and caudal
expression in the E13 microarray screen in the Mest mutant cortex.
Note that approximately 20% of such rostral genes are upregulated in the
Mest mutant and a similar percentage of caudal genes are
downregulated. (C) Caudal patterning genes (red) are downregulated and rostral
patterning genes are upregulated (green) in the Mest mutant E12.5
cortex. The average gene expression change (n=6 hybridisations) for
each gene is shown. (D,E) A comparison of Klf3 expression in E12.5 wild-type
and Mest mutant neocortex shows marked changes in both the level of
expression and the extent of the rostrocaudal gradient, with a caudal shift of
the gradient in the mutant cortex. Fluorescent in situ hybridisation on
parasagittal sections from mutant (E) and wild-type (D) E12.5 embryos is
shown, with rostral to the left. Arrowheads indicate the rostral region of
Klf3 expression in each section.
|
|
View larger version (45K):
[in a new window]
|
Fig. 8. Mest is a potential negative regulator of an Fgf-regulated neocortical
patterning pathway. (A) Genes upregulated in the Fgfr1 mutant E12.5
cortex are downregulated in the E12.5 Mest mutant cortex.
Hierarchical cluster of the set of genes showing significantly different
changes in gene expression between the Fgfr1 and Mest mutant
cortices (P<0.05). (B) Candidate caudal neocortical patterning
genes (as labelled) are downregulated in the Mest mutant cortex but
upregulated in the Fgfr1 mutant cortex. Average fold changes (log
base 2) from multiple independent hybridisations are shown (Mest,
n=6; Fgfr1, n=9). (C) Candidate rostral neocortical patterning
genes (as labelled) are upregulated in the Mest mutant cortex but
downregulated in the Fgfr1 mutant cortex.
|
|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati 
Twitter What's this?
© The Company of Biologists Ltd 2005
