Table 1.
Antibodies used in this study.
Figure 1.
Sus cells differentiate normally in developing OE of Ascl1 KO mice.
Sustentacular (Sus) cells are marked in adult wild-type mice (A–C) and in wild-type embryos (WT; D–F) and knockout (Ascl1 KO; G–I) littermates at E19.5 by staining with SUS4 (A, D, G), anti-ezrin (B, E, H) and anti-REEP6 (C, F, I), as shown here on three consecutive sections from each animal. Prominent cytoplasmic labeling of both SUS4 and anti-REEP6 at the apical processes of Sus cells (large white arrows) is comparable between wild-type and knockout mice. Anti-ezrin heavily labels the microvilli at the apical surface of the Sus cells (black triangles) as well as Bowman’s duct/gland cells in the adult (B) and the staining pattern is consistent in wild-type (E) and knockout (H) mouse embryos. White arrowheads indicate the basal lamina. Scale bar in I is 50 µm, and also applies to A–H.
Figure 2.
Sus cells differentiate on schedule, but lack robust HES1 expression in the Ascl1 KO OE.
Sections of nasal epithelium harvested at various embryonic time points were stained with K18 (to label differentiating Sus cells), HES1, and Hoechst 33258 (to label nuclei). Boxed areas in A–D, I–L are shown at higher magnification beneath each image as the green channel (left image for HES1 staining) and three channel merged images (right image). White lines overlaying the basal lamina in B–D, J–L highlight the transitional area from respiratory epithelium situated ventrally to olfactory epithelium situated dorsally. Arrows in B–D, K, L show developing duct/gland units. Arrowheads in high magnification panels (E–H and M–P) indicate basal lamina. In heterozygote (HET), HES1-expressing differentiating sus cells coalesce to form a tight apical layer by E14.5 and maintain this pattern throughout subsequent development, with the exception of rare scattered cells deep in the OE. In contrast, in Ascl1 knockout (Ascl1 KO) some apical nuclei lack detectable labeling with HES1 antibody. This layer of differentiated sus cells is less compact and the relative overall reduction in HES1 expression is indicated by comparison with respiratory epithelium; there is a sharper drop-off of RE to OE in the Ascl1 KO. Arrowheads in insets indicate basal lamina. Scale bar in L (50 µm) applies to A–D, I–L; scale bar in P (20 µm) applies to E–H, M–P.
Table 2.
The labeling intensity of Hes1 is reduced in Ascl1 knockout compared to heterozygote.
Figure 3.
βIV TUBULIN strongly labels respiratory epithelium.
Sections of nasal epithelium were harvested from E14.5 HET and Ascl1 KO animals and stained for PGP9.5, a marker of neuronal differentiation, and βIV TUBULIN, which strongly labels respiratory epithelium. Dotted lines identify the olfactory/respiratory border. Boxed region in A is shown at higher magnification in B. Note that the βIV TUBULIN labeling is strictly limited to ventral epithelium and has minimal overlap with the PGP9.5(+) olfactory epithelium, substantiating its usefulness as a respiratory epithelial marker. C. An image of the olfactory/respiratory border in age-matched Ascl1 KO littermates, showing the expected lack of neurons (lack of PGP9.5 labeling) in the olfactory area and a region of βIV TUBULIN(+) epithelium, which marks the respiratory epithelium in these animals as well. Scale bar in A is 100 µm. Scale bar in C panel is 25 µm and applies to B and C.
Figure 4.
Horizontal Basal Cells are largely absent from Ascl1 KO olfactory epithelium at E19.5.
Sections of nasal epithelium at E19.5 were stained for βIV TUBULIN and K14. K14 marks horizontal basal cells (HBCs) and βIV TUBULIN marks respiratory epithelium (A–D). Open white arrows designate boundary between respiratory epithelium (RE) and olfactory (OE). Open red arrows designate the limits of the dorsal extent of K14 staining in OE. In HET, the extent differs between rostral (A) and caudal (B). (A), (C) At rostral levels, K14(+) HBCs extend a number of cell diameters beyond the respiratory/olfactory boundary into the olfactory epithelium. At rostral levels of both HET and Ascl1 KO K14(+) HBCs extend into the OE to a roughly comparable extent (red arrows in A and C). (B), (D) In contrast, K14 immunoreactivity at caudal levels of the heterozygotes (HET) is extends much further into the dorsal septum, as compared to Ascl1 KO, where K14 immunoreactivity (red arrows) is rarely found more than a few cell diameters beyond the RE/OE boundary (white arrows). Sections of Ascl1 KO OE at PND0 stained for these same two proteins indicate that K14(+) HBCs are found both at the border with respiratory epithelium (E), as well as far dorsally in the OE (F), indicating that the emergence of HBCs is not prevented in areas away from the OE-RE transition zone. Scale bar in D (50 µm) applies to A–D. Scale bar in F (25 µm) applies to E and F.
Figure 5.
HBC Emergence is aberrant but not aborted completely as a consequence of Ascl1 KO.
Matched sections from E19.5 HET and Ascl1 KO nasal epithelium were stained for K14 to mark HBCs and βIV TUBULIN to mark respiratory epithelium. K14 staining was highlighted and cartooned in black and βIV TUBULIN was highlighted and cartooned in red. Representative sections were arranged from Rostral to Caudal levels (top to bottom, matched levels numbered 1 to 5). Coincident black and red indicate regions of respiratory epithelium that contain a dense population of K14 (+) basal cells, while areas of black only represent the emergence of K14(+) HBCs in olfactory epithelium. In the HET (left side) HBCs are relatively abundant in the caudal olfactory areas, while few if any are seen in comparable regions of the OE in the Ascl1 KO (right side).
Figure 6.
Bowman’s duct and gland development appear normal in the Ascl1 KO OE.
OE sections harvested at various ages were stained for K18 to mark sustentacular cells and developing Bowman’s duct and gland units. In sections harvested from E16.5 and later, examples of developing duct/gland units and gland acini are indicated by arrows and shown at higher magnification in insets. At E12.5 and E14.5, ducts are not yet distinct. However, occasional intensely stained cells that are indicated by arrowheads extend toward the basal lamina, and are presumably developing ducts (B). By E16.5, epithelial spanning ducts in the lamina propria are seen in both HET and Ascl1 KO (arrows in C, D, H, I, J). Insets show the duct/gland structures identified by arrows. Scale bar in J (50 µm) applies to A–J. Scale bar in the inset of panel J (25 µm) applies to all insets.
Figure 7.
SOX2 is expressed by the overwhelming majority of the cells in the Ascl1 KO epithelium.
OE sections harvested from E14.5 animals were stained for SOX2, K18 and Hoechst 33258. (A), (D) Single channel green images of SOX2 staining. (B), (E) Merged images of SOX2 (green), K18 (red), and Hoechst 33258 (blue, to label nuclei). (C), (F) Blue-only images showing labeled nuclei. (A–C) In HET epithelium anti-SOX2 labels two major populations, apical Sus cells (identified by co-labeling with K18), and GBCs. (D–F) In Ascl1 KO epithelium anti-SOX2 stains cell nuclei throughout the entire apical to basal extent of the epithelium with only a very few exceptions (e.g., white arrow). Arrowheads indicate basal lamina. Scale bar in H is 25 µm and applies to all panels.
Figure 8.
NOTCH1-labeled cells are proliferating and upstream of differentiated neurons.
OE from E16.5 wild type embryos was stained for NOTCH1 and co-stained with Ki67, a marker of proliferating cells, and NCAM, a marker of neuronal differentiation. (A) Many NOTCH1(+) cells are Ki67(+) (vertical arrows). (B) The vast majority of NOTCH1(+) cells are NCAM (–). Horizontal arrows in B illustrate the border between NOTCH1(+) basal cells and the NCAM (+) neuronal layer. Arrowheads indicate basal lamina. Scale bar in B (20 µm) applies to both panels.
Figure 9.
NOTCH1-labeled cells are absent from the Ascl1 KO epithelium.
OE harvested at various embryonic ages were stained for NOTCH1. Boxed regions are shown at higher magnification in the insets. Arrowheads in insets indicate basal lamina. (A–D) NOTCH1 is expressed by basal cells throughout the developing OE of heterozygous mice; (E–H) Notch1 staining is largely absent from the epithelium of Ascl1 KO mice. Higher magnification examination of the tissue on both sides of the basal lamina (insets B–D, F–H) demonstrate that NOTCH1 staining in the Ascl1 KO is found within the lamina propria, and not the epithelium, indicating that the absence of stained cell in the OE of knockout mice is not an artifact. Scale bar in lower magnification image of H (10 µm) applies to all panels. Scale bar for the high magnification inset in H is 25 µm and applies to all insets.
Figure 10.
cKIT(+) basal cells are markedly reduced in Ascl1 KO epithelium.
Sections from E14.5 and PND0 Ascl1 HET and KO animals were stained for TuJ1 (which marks immature neurons heavily and also lightly stains some GBCs see Fig. 11 and Packard et al., 2011) and cKIT. (A–D) Low power images provide an overview of the distribution of cKIT relative to the neuronal population. Boxes are shown at higher magnification in panels to the right of each low magnification image (upper – merged red and green channels, lower – red channel). (A, C) In the HET, cKIT(+) basal cells are found deep to the neurotubulin (TuJ1) positive cells at both E14.5 and PND0. Intensity of cKIT staining appears to increase from E14.5 to PND0. (B, D) In contrast, much of the tangential extent of the Ascl1 KO olfactory epithelium lacks detectable cKIT expression. The exception to this is found in domains of epithelium that are adjacent to or co-extensive with the areas that show neuronal sparing (arrows in B1, D1). Arrowheads mark the basal lamina. Scale bar in B is 100 µm and applies only to B. Scale bar in D is 100 µm and applies to A, B, and D. Scale bar in D2 (lower) is 50 µm and applies to all high magnification panels.
Figure 11.
A subset of TuJ1(+) cells are proliferating.
Sections of OE harvested from E16.5 wild type animals were stained with TuJ1 and co-stained with phospho-histone 3 (pH3, A–C) or proliferating cell nuclear antigen (PCNA, D–F) and Hoechst 33258 to label nuclei. (A, D) Red channel only. (B, E) Merged images including Hoechst nuclear stain (blue). (C, F) Green channel only. Arrows indicate double positive cells. Scale bar in F is 25 µm and applies to all panels.
Figure 12.
A model of how cell differentiation of non-neuronal cells is affected by the neurogenic failure in Ascl1 knockout mice.
The model depicts proposed cell lineage relationships during normal development in the embryo (left side) as compared to Ascl1 knockout (Ascl1 KO) animals (right side). Italics designate genes expressed by the various cell types. Arrow thickness on the left and right sides of the model represent differences in differentiation pathways utilized in normal development as compared to the Ascl1 knockout – thicker arrows indicate a pathway that is used more, while thinner arrows indicate less. Faded cell types on the knockout panel (right) symbolize the loss of those cell stages because of Ascl1 knockout (Cau et al., 1997). Ascl1 knockout pushes the epithelium toward the expansion of upstream OPP/GBCs and away from HBCs (which are normally generated by OPP/GBCs; Packard et al., 2011) in the face of aborted production of neurons. However, to a limited degree, neurons are made by, or in apparent association with, cKIT-expressing progenitors. Abbreviations: OPP/GBC = olfactory placodal progenitor/globose basal cell, GBCTA-OSN = transit amplifying globose basal cell producing olfactory sensory neurons, GBCINP = immediate neuronal precursor, HBC = horizontal basal cell.