Supplemental Figure 1
-
Fig. S1. Aberrant ductal pattern in Foxa1-deficient prostates.
AR staining for 4-week-old (A,B,E,F) and 15-week-old (C,D,G,H) rescued Foxa1–/– and wild-type DLPs. (E-H) High
magnifications illustrating the solid epithelial cords in null prostates (E,G),
and normal luminal epithelium in wild type controls (F,H). These lobe
identities were tentatively determined.
Supplemental Figure 2
-
Fig. S2. Intact Foxa1+/–
mouse dorsal prostates show a mild phenotype indicating haploid insufficiency.
(A-D) Sexually mature dorsal prostate (Hematoxylin and Eosin staining in A,B;
Ck14 staining in C,D) from intact 10-week-old wild-type and heterozygous (Foxa1+/–) mice. The dorsal prostates from
heterozygous animal show epithelial cell hyperproliferation and abnormal ductal
development (B,D), indicating haploid insufficiency can result in a phenotype.
Arrowheads indicate basal cells that show hyperproliferation in the dorsal
heterozygous prostates (D). Arrows indicate basal cells within the lumen.
Supplemental Figure 3I-L
-
Fig. S3. Foxa1-deficiency leads to altered epithelial cell
population and modified stromal pattern. (A,B) Ck5 staining in 4-week-old
tissue recombinants derived from rUGM+Foxa1–/– BLE (A) and rUGM+Foxa1+/+ BLE (B). Ck5-expressing basal cells are
enriched in the null recombinants. (C-H) Four-week-old rescued Foxa1–/– and Foxa1+/+ prostates were double-stained for Ck14
(red) and AR (green). Tentatively identified DLPs are shown in C,D, VPs in E,F,
and APs in G,H. (I,J) Four-week-old tissue recombinants stained for SMA. Smooth
muscle cell layers are dramatically thickened in rUGM+Foxa1–/– recombinants. (K,L) Four-week-old
tissue recombinants stained for smooth muscle g-actin.
Supplemental Figure 4
-
Fig. S4. Identification of adjacently positioned forkhead and
androgen-response elements in Sbp promoter. (A) Two forkhead binding sites were
identified on Sbp promoter, each immediately flanked by an androgen response
elements (ARE). (B) EMSA confirmed these binding sites for Foxa1 and AR.
Radiolabeled probes are indicated at bottom of each panel. Lanes 1,4,6,9, probe
only; lanes 2,7, controls (no Foxa1 templates added to the in vitro
transcription/translation reaction); lanes 3,8, probe plus in vitro sythesized
Foxa1; lanes 5,10, probe plus recombinant AR. Arrowheads and arrows indicate
Foxa1/DNA and AR/DNA complexes, respectively. (C) Organized ARE (red) and
Foxa1-binding sites (green) with fixed orientation (boxed arrows) in Sbp
promoter and in multiple prostate-specific gene enhancers (Gao et al., 2003).
Supplemental Figure 5
-
Fig. S5. Continued elevation of Shh and Foxa2 in
Foxa1-deficient epithelium. (A-C) No focal Shh staining was detected in
4-week-old rescued wild-type prostate epithelium. (A) Shh (red) alone; (B) DAPI
(blue); (C) merge of A and B. (D) Strong and focused Shh expression was seen in
4-week-old rescued Foxa1–/–
prostate epithelium. (E) Serial section was stained for SMA. Compare the
Shh-expressing null epithelial buds (arrows in D) and the surround thick smooth
muscle cell layers (arrows in E). (F) Nuclear Foxa2 expression was detected in
15-week-old rescued Foxa1–/–
epithelium. Foxa2-expressing cells form numerous microlumen (arrows),
reminiscent a pattern of hypercellularity. (G) Foxa2 was stained positive in
4-week-old tissue recombinants derived from wild-type rUGM plus Foxa1–/– epithelium.
