Fig. 4.
Another sine oculis/SIX-type homeobox gene diversifies another radially symmetric neuron class. (A) RMD neck motor neuron overview. Other connectivity difference between RMD subtypes exist (e.g., dopamine neurons; or lateral neuron-specific feedback to RIA) (10) but are not illustrated here for clarity. (B) A GFP-tagged ceh-32 reporter allele (ot1040) is expressed across all larval stages in the ventral and dorsal RMD subtypes (circled in red), located, respectively, dorsally and ventrally. ceh-32 expression is never observed in the lateral RMD. The ceh-32 expressing neuron classes that are closest to the RMD (RIA/RIB/RID/RME) are labeled. (C) Loss-of-function experiments. RMD subtype marker expression in ceh-32(ok343) null mutant animals. slrf-7prom::TagRFP is present on all transgenes to label overall RMD fate. Top: expression of a mgl-1 reporter, expressed in dorsoventral RMD (yellow triangles), is lost in two different lines (otEx7843 and otEx7844). Bottom: a lgc-37 reporter, expressed in lateral RMD (red triangles) is ectopically expressed in more than just two RMDs, as seen in two different lines (otEx7688 and otEx7689). The number of slrf-7(+) neurons is not affected (expression in the dorsoventral RMDs appears brighter in ceh-32 mutants). Quantifications are found in SI Appendix, Fig. S6C. (D) Gain-of-function experiments. ceh-32 is misexpressed in all six RMD neurons using the slrf-7 driver. The dorsoventral subtype markers nlp-45 (reporter allele ot1032) and unc-46 (fosmid otIs568) can be found in more than the usual four expected RMDs (using ceh-32 misexpressing arrays otEx7935 and otEx7936 for nlp-45 reporter allele, and otEx7909 for unc-46 fosmid line; another line, otEx7910, had no effect). In the case of nlp-45, the cells expressing both nlp-45 and a slrf-7p::tagRFP marker as part of the otEx7935/7936 misexpression arrays are shown with red triangles. The lateral RMD subtype markers nlp-11 (reporter allele syb4759) and flp-19 (reporter allele syb3278) are repressed by pan-RMD expression of ceh-32 (expressed with arrays otEx7937 and otEx7938 in the nlp-11 reporter allele and otEx7912 and otEx7913 in the flp-19 reporter allele). In the nlp-11 reporter allele background, we also used slrf-7p::tagRFP as part of the misexpression array to help identify the RMD neurons. The lateral RMDs are circled in white. For flp-19, we show ventral images, the lateral RMDs are circled in red and would be anterior to the two bright cells in the middle (AIAs). Quantifications are found in SI Appendix, Fig. S6D. (E) In an unc-42(ot1164) null allele, the ceh-32 reporter allele ot1040 becomes dimmer in both dorsal and ventral RMDs. We quantified the RMDD/V (circled red) expression as bright vs. dim, using the brightness of RIA and RIB (circled blue) as reference. Those neurons do not express unc-42 and stay unchanged in unc-42 null mutant animals. P values are done via Fisher two-sided test. (F) Summary. Terminal selectors operate in a nested regulatory configuration in which they regulate the expression of subtype terminal selectors that specify subtype identity. Loss of a subtype selector results in homeotic identity transformation to another subtype identity. Subtype selectors promote the ability of a terminal selector to activate specific subtype identity features (“subtype 1 battery”) and antagonize the ability of a terminal selector to activate specific subtype identity features (“subtype 2 battery”). Open questions are how the subtype specificity of subtype regulators are controlled and whether subtype selectors act directly on target genes (as indicated here) or through intermediary factors.