First published online August 4, 2003
doi: 10.1242/10.1242/dev.00640
The chemokine SDF1/CXCL12 and its receptor CXCR4 regulate mouse germ cell migration and survival
Kathleen A. Molyneaux1,*,
Hélène Zinszner2,*,
Prabhat S. Kunwar2,
Kyle Schaible1,
Jürg Stebler3,
Mary Jean Sunshine4,
William O'Brien4,
Erez Raz3,
Dan Littman4,
Chris Wylie1,
and
Ruth Lehmann2,
1 Division of Developmental Biology, Children's Hospital Research Foundation,
Cincinnati, OH 45229, USA
2 Developmental Genetics Program, Skirball Institute of Biomolecular Medicine,
New York University Medical Center and Howard Hughes Medical Institute, New
York, NY 10016, USA
3 Max Planck Institute for Biophysical Chemistry, Am Fassberg 11, 37077
Gottingen, Germany
4 Molecular Pathogenesis Program, Skirball Institute of Biomolecular Medicine,
New York University Medical Center and Howard Hughes Medical Institute, New
York, NY 10016, USA
View larger version (60K):
[in a new window]
|
Fig. 1. CXCR4 is expressed in PGCs during and after colonization of the gonad. (A)
Probes prepared from E10.5 (two biological replicates) and E12.5 (two
biological replicates) PGCs were applied to Affymetrix chips. Chip data was
analyzed using MicroArray Suite v5.0 software (Affymetrix) to generate signal
intensity data, and to statistically determine presence and absence calls. The
average chip signal was normalized to an arbitrary value of 1000. The bar
graph shows the signal intensity of the CXCR4 probe set for each sample. CXCR4
was called present in all samples. The PGC marker gene Kit was also
called present in all samples. The somatic marker gene Steel (Kitl -
Mouse Genome Informatics) was called absent. (B) CXCR4 expression was
confirmed by RT-PCR. Lane 1, E10.5 PGC cDNA; Lane 2, RT-; Lane 3,
H2O blank; Lane 4, E12.5 PGC cDNA; Lane 5, E10.5 whole-embryo cDNA.
(C) Cxcr4 message is enriched in PGCs relative to the somatic tissue.
The level of Cxcr4 transcripts in E11.5 PGCs (GFP+) or somatic tissue
(GFP-) were quantified by SYBR-green based RT-PCR. The meiotic marker, STAG3
(Pezzi et al., 2000 ) was used
as a positive control for PGCs, and the gonadal markers SPARC and cystatin C
(CST3) (Wertz and Herrmann,
2000) were used as positive controls for the somatic component of
the gonad. For CXCR4 and STAG3, the PGC cDNA was used to generate a standard
curve and expression in this tissue was set to an arbitrary value of 100. For
SPARC and CST3, the somatic tissue cDNA was used to generate standard curves.
Expression was normalized to ODC levels in the somatic and PGC samples. (D)
SDF1 is expressed in the genital ridge area. Anti-SDF1 staining is shown in
red. This slice was taken from an E9.5 +/Oct4 PE:GFP+ embryo.
A BSA-coated bead was placed in the aorta and the slice was cultured for 12
hours, then fixed and stained with 2.5 µg/ml anti-SDF1 antibody. Arrows
indicate the region of more intense SDF1 staining in the floor plate of the
neural tube, and arrowheads indicate the mesonephros and adjacent mesenchyme.
The germ cells are marked by bright green GFP fluorescence. (E) Control for
SDF1 staining. This control slice was treated in the same way as the
experimental slice in D, but the anti-SDF1 antibody was omitted. Scale bar in
D: 82 µm for D,E.
|
|
View larger version (79K):
[in a new window]
|
Fig. 2. SDF1 alters PGC behavior in an organ culture assay. (A) A slice taken from
an E9.5 +/Oct4PE:GFP+ embryo after 20 hours in culture. PGCs
(green) have formed clusters at the genital ridges. A few cells (arrowhead)
remain in the hindgut. (B) An E9.5 slice incubated for 20 hours in the
presence of 1.6 µg/ml SDF1. Some PGCs have colonized the genital ridges,
but many cells remain scattered across the midline. (C) Summary of migration
data from four culture experiments. Slices were incubated in the presence of a
range of SDF1 concentrations (5 ng/ml to 1 µg/ml). PGC migration was
affected at all concentrations (data not shown). Beads coated with 50 µg/ml
SDF1 caused a similar effect to treatment with soluble SDF1. (D) An example of
a control slice that was scored as dead. The majority of cells remain trapped
in the hindgut. (E) Example of a control slice scored as confused. The circled
cells have failed to clear the midline. (F) Example of a control slice scored
as normal. The midline of the body wall is clear and only a few PGCs
(arrowheads) remain in the hindgut. Scale bars: in A, 46 µm for A,B; in D,
68 µm for D-F.
|
|
View larger version (56K):
[in a new window]
|
Fig. 3. Soluble SDF1 decreases the average and maximum velocity of PGCs. (A) An
E9.5 control slice. (B) An E9.5 slice in the presence of 1.6 µg/ml SDF1.
(C,D) Slices shown in A,B after 7.5 hours of culture, respectively. (E,F)
Trajectories of cells in control (E) and SDF1-treated (F) slices. Green
outlines represent starting positions of cells and red outlines indicate
ending positions of cells. Arrow indicates the position of the gut. Black
lines follow trajectories of five selected cells in the experimental and
control slices. (G) Summary of PGC velocity data from five control and four
SDF1-treated slices. Error bars show standard deviation. Scale bar: 55 µm
(A-F).
|
|
View larger version (20K):
[in a new window]
|
Fig. 4. SDF1-coated beads capture migrating PGCs. (A) Trajectories of cells in a
control slice. A BSA-coated bead was placed into the aorta of an E9.5 slice
and the slice was filmed for eleven hours. Green outlines represent starting
positions of cells and red outlines represent ending positions. A dotted
outline indicates the position of the bead. (B-D) Trajectories of cells near
beads coated with 10 (B), 25 (C) and 50 (D) µg/ml SDF1. (E) Summary of
migration data from control and SDF1-coated bead experiments. Movies of three
slices per treatment were analyzed, except for the 50 µg/ml treatment where
movies of two slices were analyzed (see Movies at
http://dev.biologists.org/supplemental/).
Error bars show standard deviation. Scale bar: 55 µm (A-D).
|
|
View larger version (54K):
[in a new window]
|
Fig. 5. Reduction in the number of PGCs reaching the genital ridge in CXCR4-mutant
embryos. Cryosections of E10.5 (A,B), E11.5 (C,D) and E12.5 (E,F) control
(+/+, A,C; +/-, E) and mutant (-/-, B,D,F) embryos. PGCs are stained with AP:
black arrows indicate PGCs in the genital ridge and white arrows point to lost
PGCs in mutant embryos. DA indicates position of dorsal aorta in each section.
Scale bars: 0.5 mm. (G) Number of PGCs in control and mutant embryos. For each
bar, PGCs were counted in eight transverse 20 µm sections distributed along
the genital ridge. Error bars show standard deviation. There is a
statistically significant difference in the number of PGCs between control and
mutant embryos (Student's t-test, P<0.0001).
|
|
View larger version (67K):
[in a new window]
|
Fig. 6. Soluble SDF1 enhances PGC survival in E10.5 slices. (A) A control slice
before culture. This slice was optically sectioned and the picture shown is a
projection (z=156.4 µm). Red marks indicate counted cells. (B) The slice
shown in A after 20 hours in culture. The picture shown is a projection
(z=83.7 µm). Slices compress while in culture. (C) An SDF1-treated slice
prior to culture (z=150.0 µm). (D) The slice shown in C after 20 hours in
the presence of 500 ng/ml SDF1 (z=76.8 µm). (E) Summary of survival data
from two culture experiments. The slice shown in A is Control 3 in the graph,
and the slice shown in C is SDF1 5. Most control slices have poor survival,
whereas SDF1 treatment enhances survival and/or proliferation. Scale bar: 55
µm (A-D).
|
|
© The Company of Biologists Ltd 2003
