Molecular characteristics and laminar distribution of prefrontal neurons projecting to the mesolimbic system

Abstract

Prefrontal cortical influence over the mesolimbic system - including the nucleus accumbens (NAc) and the ventral tegmental area (VTA) - is implicated in various cognitive processes and behavioral malfunctions. The functional versatility of this system could be explained by an underlying anatomical complexity; however, the detailed characterization of the medial prefrontal cortical (mPFC) innervation of the NAc and VTA is still lacking. Therefore, combining classical retrograde and conditional viral tracing techniques with multiple fluorescent immunohistochemistry, we sought to deliver a precise, cell- and layer-specific anatomical description of the cortico-mesolimbic pathways in mice. We demonstrated that NAc- (mPFC_NAc) and VTA-projecting mPFC (mPFC_VTA) populations show different laminar distribution (layers 2/3-5a and 5b-6, respectively) and express different molecular markers. Specifically, calbindin and Ntsr1 are specific to mPFC_NAc neurons, while mPFC_VTA neurons express high levels of Ctip2 and FoxP2, indicating that these populations are mostly separated at the cellular level. We directly tested this with double retrograde tracing and Canine adenovirus type 2-mediated viral labeling and found that there is indeed minimal overlap between the two populations. Furthermore, whole-brain analysis revealed that the projection pattern of these populations is also different throughout the brain. Taken together, we demonstrated that the NAc and the VTA are innervated by two, mostly nonoverlapping mPFC populations with different laminar distribution and molecular profile. These results can contribute to the advancement in our understanding of mesocorticolimbic functions and its disorders in future studies.

Keywords: cortical organization; laminar topography; mouse; neuroscience; nucleus accumbens; prefrontal cortex; pyramidal cell types; ventral tegmental area.

Figure 1.. Nucleus accumbens (NAc) is innervated by L2/3 and L5 medial prefrontal cortical (mPFC) cells.

(A) Experimental design. (B) A representative retrograde tracer (Fluoro-Gold [FG], green) injection site in the NAc. (C) Extent of injection sites in the NAc of three animals. Each case is represented with different color. (D) Plotted distribution of retrogradely labeled cells throughout the mPFC of the same animals as in C (same colors represent same animals). Each dot represents one labeled mPFC_NAc cell. (E) Distribution of labeled mPFC_NAc neurons in relation to parvalbumin (PV) (orange) immunofluorescent labeling outlining the PrL cortex (Figure 1—figure supplement 1). (F) Pooled anteroposterior distribution of mPFC_NAc neurons for three animals. (G) Distribution of mPFC_NAc cells in individual mPFC subregions. (H) Laminar distribution of mPFC_NAc neurons in the mPFC. (I–K) Confocal images showing the distribution of FG-labeled cells (green) in the PrL at Br. + 1.98 mm (I1–K1) with the counterstaining of Calb1 (purple, I2), Ctip2 (gray scale, J2), and FoxP2 (red, K2) (Figure 1—figure supplement 1). Note that most labeled cells are localized in the L2/3 (Calb1) and L5a (Ctip2). (I3–K3) High-magnification confocal images showing the coexpression of FG and Calb1 (I3), Ctip2 (J3), or FoxP2 (K3). White arrowheads indicate colabeling, empty arrowheads indicate the lack of marker expression. (L–N) Bar graphs showing the proportion of Calb1- (L), Ctip2- (N), and FoxP2-expressing (M) mPFC_NAc cells. All data are shown as mean ± standard deviation (SD), n = 3 mice. Numbers in the bars represent cell counts and circles represent individual animal data. For detailed quantitative data see Tables 1 and 2. Scale bars: (B, E, I1–K1, I2–K2) 200 µm; (I3–K3) 20 µm. aca, anterior commissure, anterior part; BNST, bed nucleus of the stria terminalis; CPu, caudate putamen; fmi, forceps minor of the corpus callosum; LS, lateral septum; LV, lateral ventricle; VDB, nucleus of the vertical limb of the diagonal band; VP, ventral pallidum.

Figure 1—figure supplement 1.. Parvalbumin (PV), Calb1, Ctip2, and FoxP2 staining define medial prefrontal cortical (mPFC) subregion borders and layers.

(A) The lack of PV + axonal and cellular immunolabeling (gray scale) in the L2/3 defines the territory of PrL (asterisk). (B) Calb1-expressing L2/3 (green) is clearly visible in most parts of the mPFC. A thickened L1 and a more compact L2/3 and L5 identify the deep peduncular cortex (DP) (number sign). (C) Ctip2 (magenta) is expressed in the L5b (bigger neurons) and L6 (smaller neurons). (D) FoxP2 (cyan) is expressed throughout the L6 of the mPFC. Narrowed L6 and the lack of FoxP2-expressing cells (cyan) in the L5 reveal DP (cross). (E) Schematic summary of PV, Calb1, FoxP2, and Ctip2 distribution in the mPFC. Scale bar: 200 µm.

Figure 2.. Ventral tegmental area (VTA) is innervated by two medial prefrontal cortical (mPFC) cell clusters.

(A) Experimental design. (B) A representative retrograde tracer (Fluoro-Gold [FG], green) injection site in the VTA. (C) Full extent of the injection sites in the VTA in three animals. Each case is represented with different color. (D) Plotted distribution of retrogradely labeled neurons throughout the mPFC of the same animals as in (C) (same colors represent same animals). Each dot represents one labeled mPFC_VTA cell. (E) Distribution of labeled neurons in the mPFC in relation to parvalbumin (PV) (orange) immunofluorescent labeling outlining the PrL cortex. (F) Pooled anteroposterior distribution of mPFC_VTA neurons for three animals. (G) Distribution of mPFC_VTA cells in individual mPFC subregions. (H) Laminar distribution of mPFC_VTA neurons in the mPFC. (I–K) Confocal images showing the layer-specific distribution of FG-labeled cells (green) in the PrL (I1–K1) with counterstaining of Calb1 (purple, I2), Ctip2 (gray scale, J2), and FoxP2 (red, K2). Note that the labeled cells are almost exclusively localized in the L5b (Ctip2) and L6 (Ctip2 + FoxP2) layers. (I3–K3) High-magnification confocal images showing the coexpression of FG and Calb1 (I3), Ctip2 (J3), or FoxP2 (K3). White arrowheads indicate colabeling, empty arrowheads indicate the lack of marker expression. Bar graphs showing the proportion of Calb1- (L), Ctip2- (M), and FoxP2-expressing (N) mPFC_VTA cells. All data are shown as mean ± standard deviation (SD), n = 3 mice. Numbers in the bars represent cell counts and circles represent individual animal data. For detailed quantitative data see Tables 1 and 3. Scale bars: (B, E, I1–K1, I2–K2) 200 µm; (I3–K3) 20 µm. aca, anterior commissure, anterior part; fmi, forceps minor of the corpus callosum; IPR, interpeduncular nucleus, rostral subnucleus; ml, medial lemniscus; RN, red nucleus; SN, substantia nigra; VO, ventral orbital cortex.

Figure 3.. Distinct nucleus accumbens (NAc) and ventral tegmental area (VTA) innervation by genetically identified medial prefrontal cortical (mPFC) cell populations.

Delineation of AAV-DIO-mCherry/eYFP injection sites in the mPFC of the Calb1- (A), Rbp4- (B) Ntsr1- (C) FoxP2- (D), and Thy1-Cre (E) strains (n = 3 mice in each strain). Viral labeling was always analyzed after immunohistochemical enhancement (Figure 3—figure supplement 1). For higher magnification distribution of labeled cells in the mPFC and M1 see Figure 3—figure supplement 2. Confocal images showing virally labeled prefrontal axons (red) in the NAc of Calb1- (F), Rbp4- (G), Ntsr1- (H), FoxP2- (I) and Thy1-Cre (J) mouse strains. Calb1 (cyan) immunofluorescent staining was used to identify the NAc. (K–O) Distribution of labeled axons (red) from the same animals, respectively, in the VTA defined with TH staining (gray scale). Scale bars: 200 µm. aca, anterior commissure, anterior part; fmi, forceps minor of the corpus callosum; IPR, interpeduncular nucleus, rostral subnucleus; LS, lateral septum; LV, lateral ventricle; ml, medial lemniscus; Pir, piriform cortex; R, raphe; RN, red nucleus; SN, substantia nigra; VDB, nucleus of the vertical limb of the diagonal band; VP, ventral pallidum.

Figure 3—figure supplement 1.. IHC enhancement is necessary for reliable detection of viral fluorescent signal.

(A–B) Confocal images showing the native mCherry signals in the somatodendritic (A1) and axonal compartments (B1) and after IHC_Fluo signal enhancement (A2, B2) of the same slice taken from the medial prefrontal cortex (mPFC) (A) (injection site) and from the nucleus accumbens (NAc) (B). Arrows indicate dendrites, arrowheads indicate neurons visible in both the native and IHC_Fluo-enhanced sample. White circles indicate neurons not visible in the native sample. (C, D) Brightfield images showing IHC_DAB-Ni-enhanced samples taken from a neighboring slice. Scale bars: 100 µm. aca, anterior commissure, anterior part.

Figure 3—figure supplement 2.. Layer-specific Cre mouse lines reveal different laminal distribution of medial prefrontal cortical (mPFC) and primary motor cortical cells.

(A–E) Confocal images (left) and plots (right) showing the adeno-associated viral vector (AAV)-labeled cells in the mPFC in the same animals as in Figure 3, respectively. (F–J) Localization of labeled neurons in the M1 cortex in the same mouse strains. Note the different distribution of Rbp4- (B, G), Ntsr1- (C, H), and FoxP2-expressing (D, I) cells in the two cortical regions. Dashed lines indicate layer borders defined with IHC_Fluo against Calb1, Ctip2, or FoxP2 (Figure 1—figure supplement 1). Open circles in A, F represent putative GABAergic interneurons. Scale bars: 200 µm.

Figure 4.. Ventral tegmental area (VTA) and nucleus accumbens (NAc) are mostly innervated by nonoverlapping medial prefrontal cortical (mPFC) cell populations.

(A) Experimental design of double retrograde tracing experiments. (B) Representative CTB (magenta) injection site in the NAc (top) and Fluoro-Gold (FG) (green) in the VTA (bottom). (C) High-magnification confocal image showing the distribution of mPFC_NAc (magenta) and mPFC_VTA cells (green) in the mPFC. Inset shows higher magnification of the same slice with arrowheads representing double-labeled cells. (D) Only a small proportion of labeled mPFC cells innervated both VTA and NAc. All data are shown as mean ± standard deviation (SD), n = 3 mice. Exact cell counts are written in the bars. For detailed quantitative data see Table 4. Scale bars: 200 µm, (C) inset: 100 µm. aca, anterior commissure, anterior part; IPR, interpeduncular nucleus, rostral subnucleus; LV, lateral ventricle; ml, medial lemniscus; RN, red nucleus.

Figure 5.. Nucleus accumbens (NAc) or ventral tegmental area (VTA) preference of medial prefrontal cortical (mPFC) cells.

(A) Experimental design of CAV2-Cre-mediated viral tracing experiments. BDA was used to visualize the exact location of injection sites. (B) Representative CAV2-Cre + 5% BDA (cyan) injection site (B1) in the NAc and AAV-DIO-mCherry (red) injection site (B2) in the mPFC of the same animal. (C) Representative CAV2-Cre + BDA (cyan) injection site (C1) in the VTA counterstained with TH (grayscale) and AAV-DIO-mCherry (red) injection site (C2) in the mPFC of the same animal. (D–G) High-magnification confocal images showing the distribution of mCherry (red)-labeled axons in the NAc (left) and the VTA (right) in a mPFC_NAc (D), mPFC_VTA (E), mPFC_Rbp4 (F); same animal as in Figure 3C, H, M and Figure 3—figure supplement 1B and mPFC_Thy1 (G); same animal as in Figure 3F, K, P and Figure 3—figure supplement 1E animal. (H) Quantification of relative axon density (RAD) in the mPFC_NAc, mPFC_VTA, mPFC_Rbp4, and mPFC_Thy1 animals F_{(3, 8)} = 55.56; p = 0.000011; mPFC_NAc vs. mPFC_VTA, p = 0.0000026; mPFC_NAc vs. mPFC_Rbp4, p = 0.028; mPFC_NAc vs. mPFC_Thy1, p = 0.18; mPFC_VTA vs. mPFC_Rbp4, p = 0.000018; mPFC_VTA vs. mPFC_Thy1, p = 0.0000072; one-way analysis of variance (ANOVA), least significant difference (LSD) post hoc test; *p < 0.05; ***p < 0.001; n.s., not significant. All data are shown as mean ± standard deviation (SD), n = 3 mice in each group. For detailed quantitative data see Table 5 and Figure 5—source data 1. Scale bars: (B–C) 200 µm, (D–G) 20 µm. aca, anterior commissure, anterior part; BDA, biotinylated dextran amine; IPR, interpeduncular nucleus, rostral subnucleus; LV, lateral ventricle; ml, medial lemniscus; RN, red nucleus.

Figure 6.. mPFC_NAc and mPFC_VTA neurons possess different efferent connections.

(A–I) Brightfield images showing the distribution of CAV2-Cre-mediated AAV-DIO-mCherry-labeled axons visualized with IHC_DAB-Ni at different AP levels. Arrows indicate adeno-associated viral vector (AAV) injection sites in the medial prefrontal cortex (mPFC) (B). Note the clear difference between the mPFC_NAc (left column) and mPFC_VTA (right column) populations, most prominently in the striatum (C), different cortical areas (A–H), the hypothalamus (D–F), and the brainstem (F–I) – including the ventral tegmental area (VTA) (G). Note the almost complete lack of contralateral cortical projection in mPFC_VTA animals as opposed to mPFC_NAc animals (A, B, asterisks). For experimental design see Figure 6A. (J) Summary table showing the innervation intensities of mPFC_NAc (top row) and mPFC_VTA (bottom row) populations (n = 3–3 mice) in different brain regions. Darker color indicates stronger innervation. For details see Table 6. Scale bars: 500 µm. 3V, 3rd ventricle; 4V, 4th ventricle; aca, anterior commissure, anterior part; AO, anterior olfactory nucleus; Astr, amygdalostriatal transition area; Aq, aqueduct; BA, basolateral amygdaloid nucleus; BNST, bed nucleus of the stria terminalis; BS, brainstem; cc, corpus callosum; CG, central gray; CPu, caudate putamen; CTX, cortex; D3V, dorsal 3rd ventricle; DM, dorsomedial hypothalamic nucleus; DR, dorsal raphe; Ect, ectorhinal cortex; Ent, entorhinal cortex; Hb, habenula; I, intercalated amygdalar nuclei; ic, internal capsule; Ins, insular cortex; LH, lateral hypothalamus; LC, locus coeruleus; LDTg, laterodorsal tegmental nucleus; LP, lateral posterior thalamic nucleus; LS, lateral septum; LV, lateral ventricle; MR, medial raphe; MS, medial septum; nRT, reticular thalamic nucleus; PVT, paraventricular thalamic nucleus; VDB, nucleus of the vertical limb of the diagonal band; VP, ventral pallidum; opt, optic tract; PAG, periaqueductal gray; PB, parabrachial nucleus; Pir, piriform cortex; PnO, pontine reticular nucleus, oral part; PO, preoptic area; PPTg, pedunculopontine tegmental nucleus; PRh, perirhinal cortex; Re, reuniens thalamic nucleus; rf, rhinal fissure; RS, retrosplenial cortex; RtTg, reticulotegmental nucleus of the pons; SC, superior colliculus; scp, superior cerebellar peduncle; STR, striatum; Sub, subiculum; TeA, termporal association cortex; TH, thalamus; Tu, olfactory tubercule; ZI, zona incerta.

Figure 7.. Summary: molecular characteristics and laminar distribution of the two identified projection groups in the medial prefrontal cortex (mPFC).

Neurons that innervate the nucleus accumbens (NAc) (‘IT-like’) are mostly localized in the upper layers of the mPFC (L2/3–5a) and express Calb1 (green), Ntsr1 (purple), Rbp4 (orange), and to a lesser extent, Ctip2 (magenta). mPFC cells that innervate the ventral tegmental area (VTA) (‘PT-like’) are mostly localized in the deeper layers (L5b–6) and express Ctip2, FoxP2 (cyan), and Rbp4. Connections between NAc and VTA, and ascending VTA pathways (gray arrows) are based on literature data (see Introduction, Discussion).