Lhx2 Determines Odorant Receptor Expression Frequency in Mature Olfactory Sensory Neurons

Conditional deletion of Lhx2 in immature OSNs decreases the abundance of most OR mRNAs. A, In 123Cre:Z/EG mice, recombination-dependent GFP expression marks OSNs in the immature and mature OSN layers of the olfactory epithelium. Dashed line, location of the basal lamina separating the olfactory epithelium and lamina propria. Scale bar, 20 µm. B–E, Gap43 immunoreactivity (B, D) and in situ hybridization for Gap43 mRNA (C, E) identify immature OSN cell bodies in 123Cre:Lhx2^fl/fl mice (D, E) and littermate controls (B, C). F–I, Omp immunoreactivity (F, H) and in situ hybridization for Omp mRNA (G, I) identify mature OSN cell bodies in 123Cre:Lhx2^fl/fl mice (H, I) and littermate controls (F, G). Scale bars, 20 µm. J, In 123Cre:Lhx2^fl/fl mice, the abundance of 676 OR mRNAs decreases (p < 0.01; n = 4; FDR <1%). •, Significant decrease; +, no difference from littermate controls. K, The percentage of class I ORs affected by Lhx2 deletion in immature OSNs is very similar to the percentage of Class II ORs affected, and the magnitude of the decreases are similar for both classes of ORs. Error bars represent SDs.

Conditional deletion of Lhx2 in immature OSNs reduces OR gene expression frequencies. A, In situ hybridization for Olfr615 labels scattered OSNs in the dorsal olfactory epithelium. B, In 123Cre:Lhx2^fl/fl mice, the frequency of OSNs expressing Olfr615 appears to be reduced. C, Normalized cell counts show that ORs are consistently less frequently expressed in 123Cre:Lhx2^fl/fl mice compared with littermate controls. Numbers in control bars are the numbers of littermate pairs tested. Diff., Normalized fold differences from the microarray experiment; *Significant difference (p < 0.01; n = 4; FDR <1%). NC, Air space of the nasal cavity; S, nasal septum; arrows, blood vessels in the lamina propria. Scale bars, 100 µm.

Deletion of Lhx2 in mature OSNs decreases the abundance of most OR mRNAs. A, In OmpCre:Z/EG mice, recombination-dependent GFP expression marks OSNs in the mature OSN layer. Dashed line, Location of the basal lamina separating the olfactory epithelium and lamina propria. Scale bar, 20 µm. B, In OmpCre:Lhx2^fl/fl mice, the abundance of 765 OR mRNAs is decreased relative to littermate controls (p < 0.01; n = 4; FDR <1%). •, Significant decrease; +, no difference from littermate controls; ▴, increased abundance. C, Deletion of Lhx2 in immature OSNs (123Cre:Lhx2) and mature OSNs (OmpCre:Lhx2) produced similar effects on the abundance of most OR mRNAs relative to littermate controls. Line, Pearson linear correlation fit; •, significant decrease in at least one of the two datasets; +, no decrease compared with littermate controls.

Conditional deletion of Lhx2 in mature OSNs reduces OR expression frequencies. A, In situ hybridization shows that Olfr156 is expressed in scattered OSNs in the ventral olfactory epithelium. B, In OmpCre:Lhx2^fl/fl mice, the frequency of OSNs expressing Olfr156 is reduced. C, Normalized cell counts from in situ hybridization images show that the ORs tested are less frequently expressed in OmpCre:Lhx2^fl/fl mice compared with littermate controls, which is in agreement with normalized fold differences (Diff.) measured in the microarray experiment. *Significant difference (p < 0.01; n = 4; FDR < 1%). T, Turbinate; S, nasal septum. Scale bars, 100 µm.

Conditional deletion of both Lhx2 and Emx2 in immature OSNs decreases the abundance of 755 OR mRNAs and increases the abundance of 10 OR mRNAs. •, Significant decrease; +, no difference from littermate controls; ▴, increased abundance.

Conditional deletion of both Lhx2 and Emx2 in immature OSNs reduces OR expression frequencies. A, In situ hybridization shows that Olfr308 is expressed in scattered OSNs in the ventral olfactory epithelium. B, In 123Cre:Lhx2^fl/fl:Emx2^fl/fl mice, the frequency of OSNs expressing Olfr308 appears to be reduced. NC, Air space of the nasal cavity; S, nasal septum; T, turbinate. Scale bars, 100 µm. C, Normalized cell counts from in situ hybridization images show that ORs are less frequently expressed in 123Cre:Lhx2^fl/fl:Emx2^fl/fl mice compared with littermate controls. Diff., Normalized fold differences from the microarray experiment. *Significant difference (p 0.01; n 4; FDR 1%).

The absence of Lhx2 and the absence of Emx2 have distinctly different effects on Olfr15 expression frequency. A, In situ hybridization image showing normal Olfr15 expression in the ventral region of the olfactory epithelium in a littermate control. B, When Lhx2 is deleted in OSNs, Olfr15 expression frequency decreases. C, An image of the left half of a more caudal section from a littermate control shows the normal expression of Olfr15 in the ventral olfactory epithelium. This image shows one of the rare instances of the misexpression of Olfr15 in the dorsal olfactory epithelium in control genotypes (arrow). D, When both Lhx2 and Emx2 are deleted, Olfr15 expression frequency decreases in the ventral region, where it is normally expressed, but misexpression increases in the dorsal region. The approximate boundaries between dorsal and ventral olfactory epithelium are marked in C and D. NC, Air space of the nasal cavity; S, nasal septum; T, turbinate. Scale bars, 100 µm.