Redefining the serotonergic system by genetic lineage

Abstract

Central serotonin-producing neurons are heterogeneous-differing in location, morphology, neurotoxin sensitivity and associated clinical disorders-but the underpinnings of this heterogeneity are largely unknown, as are the markers that distinguish physiological subtypes of serotonergic neurons. Here we redefined serotonergic subtypes on the basis of genetic programs that are differentially enacted in progenitor cells. We uncovered a molecular framework for the serotonergic system that, having genetic lineages as its basis, is likely to have physiological relevance and will permit access to genetically defined subtypes for manipulation.

Figure 1. Intersectional and subtractive cell marking distinguishes r1 (En1-cre)- from non-r1-derived 5-HT precursors

(a) RC::Fela dual recombinase responsive indicator allele. CAG promoter/enhancer elements followed by, 5′ to 3′, an FRT-flanked PGK-neo^R cassette with two consecutive SV40 polyadenylation (pA) sequences; a loxP-flanked eGFP cassette with five consecutive bovine growth hormone pA sequences; and a nlacZ sequence followed by a single pA sequence was targeted to the Rosa26 locus. (b) Strategy for single (Flp-FRT) and dual (Flp-FRT and cre-loxP) recombinase-based fate mapping. Tubes represent the neural tube at early (left) to late (right) developmental stages. Top row illustrates early transient expression of GeneA. Middle row illustrates expression of GeneA::Flpe and GeneB::cre transgenes. Bottom row illustrates coupling of GeneA::Flpe and GeneB::cre transgenes with dual recombinase-responsive indicator allele. Cells expressing Flpe recombinase activate expression of eGFP (the subtractive population, green); cells expressing Flpe and Cre recombinase activate expression of nβgal (the intersectional population, red). Reporter molecule activation is permanent, and expression is retained in all descendent cells (bottom right). (c) Cartoon schematic illustrates sagittal section of embryonic day (E) 12.5 brain. (d, e) E12.5 ePet::Flpe transgenic embryo. A 40kb ePet1 enhancer region coupled with β-globin minimal promoter followed by an intron-Flpe cassette was used to generate the ePet::Flpe transgene. Adjacent sagittal sections, and horizontal sections inset, show expression of Pet1 (d) and Flpe (e) mRNA by in situ hybridization detectable in the mantle zone (mz) immediately ventral to the ventricular zone (vz). (f) E11.5 En1-cre; indicator double transgenic embryo, and horizontal section inset, showing En1 descendents as βgal⁺ by X-gal detection. (g-l) En1-cre; ePet::Flpe; RC::Fela triply transgenic animals. E12.5 sagittal sections (g-i) and P21 coronal sections (j-l) stained to detect nβgal (the r1 (En1-cre)-derived intersectional 5-HT population) by X-gal detection (g) and immunofluorescence (h,I,k,l), and eGFP (the non-r1-derived 5-HT subtractive population) by immunofluorscence (h,I,k,l).

Figure 2. The organization of 5-HT neurons as defined by embryonic origin and developmental gene expression profile is different from that based on anatomical architecture

(a) Shown are cartoon schematics of the embryonic hindbrain (left) and the mature brainstem (right): dark blue represents the 5-HT progenitor cells in r1 (left) or r1-derived mature 5-HT neurons (right); yellow, r2 5-HT progenitors (left) or descendants (right); light blue, r3; pink, presumed r5; gray, r6-r7. The mature sagittal diagram is compressed along the mediolateral axis. The B1-B9 abbreviations refer to the names given to nuclei defined anatomically, some of which are also referred to as the dorsal raphe nucleus (DR) or median raphe nucleus (MR). Dashed lines represent coronal sections (b-f) presented rostral to caudal, top to bottom. r1-derived 5-HT neurons (dark blue) populate the B7, B6 and B4 groups in their entirety, as well as aspects of the B9, B8 and B5 nuclei. r2-derived 5-HT neurons (yellow) populate the B9, B8 and B5 nuclei intermingled with both the r1-derived (dark blue) and r3-derived (light blue) 5-HT neurons. Presumed r5-derived 5-HT neurons (pink) populate the B3 nucleus intermingled with more caudally (r6-r7-) derived 5-HT neurons (gray). Boxes highlight images of coronal sections from various triple transgenic P21 brainstem stained to detect nβgal (red) and eGFP (green) by immunofluorescence: (g-k) En1::cre (r1); ePet::Flpe; RC::Fela; (l-p) Rse2::cre (r2); ePet::Flpe; RC::Fela; (q-u) Egr2::cre (r3/r5); ePet::Flpe; RC::Fela,. 5-HT neurons with a combined history of rhombomere specific::cre and ePet::Flpe activity are identified as nβgal⁺ (the intersectional population, in other words that population derived from a particular rhombomere), while cells with a history of only ePet::Flpe activity are eGFP⁺ only (the subtractive population, in other words those neurons derived from the remaining pool of 5-HT progenitor cells).

Figure 3. r1 (En1-cre)- and non-r1-derived 5-HT neurons populate the postnatal Nkx2.2−/− mutant brainstem

(a-f) Coronal brainstem sections from P0.5 Nkx2.2^−/− animals (d-f) and wild-type littermates (Nkx2.2^+/+, a-c) stained to detect 5-HT. Adjacent images are taken from approximately the same axial levels. At the level of the B4, B6 and B7 nuclei (shown is B7), there is a comparable distribution of 5-HT positive neurons in Nkx2.2^+/+ and Nkx2.2^−/− sections. In more ventral and caudal territories there is a clear diminution of the B9, B8, B5, B3, B2 and B2 nuclei (shown is B8 and B3). (g-l) Coronal sections from P0.5 En1-cre; ePet::Flpe;RC::Fela;Nkx2.2−/− animals (g-i) and wild-type littermates (En1-cre; ePet::Flpe; RC::Fela;Nkx2.2^+/+, j-l) stained to detect nβgal (the r1 (En1-cre)-derived intersectional 5-HT population) and eGFP (the non-r1-derived 5-HT subtractive population) by immunofluorescence. (j-l) In Nkx2.2−/− brains the rostral 5-HT nuclei (B4-B9) receives contributions exclusively from r1-derivied 5-HT progenitors (j,k) and the caudal 5-HT nuclei (B1–3) is populated by non-r1-derived 5-HT progenitors (l).