Thalamic dopaminergic neurons projects to the paraventricular nucleus-rostral ventrolateral medulla/C1 neural circuit

Abstract

Paraventricular nuclei (PVN) projections to the rostral ventrolateral medulla (RVLM)/C1 catecholaminergic neuron group constitute the pre-autonomic sympathetic center involved in the neural control of systemic cardiovascular function. However, the role of extra-hypothalamic and thalamic dopaminergic inputs in this circuit remains underexplored. Using retrograde neuroanatomical tracing and high contrast confocal imaging methods, we investigated the projections and morphology of the discrete thalamic dopaminergic neuron groups in the dorsal hypothalamic area (DHA) and their contribution to the PVN-RVLM neural circuit. We found that dopaminergic (DA) neuron sub-groups in the Zona Incerta (Zi; 60%) and Reuniens thalamic nuclei (Re; 40%) were labeled comparably to the PVN (85%) after a retrograde tracer was injected into the RVLM/C1 (p<0.01 mean±SEM). The Re/Zi DA neuron sub-groups were characterized by angulated cell bodies, superiomedial and inferiomedial projections reaching the contralateral Re/Zi and ipsilateral PVN DA neurons respectively. Ultimately, we deduced that the dopaminergic projections of the Re/Zi to the PVN contribute to the PVN-RVLM/C1 neural circuit. As a result of these connections, the Re/Zi dopaminergic neuron groups may regulate preautonomic sympathetic events associated with the PVN-RVLM pathway.

Introduction

The paraventricular nucleus (PVN) is an integral part of the hypothalamic neuroendocrine system involved in the control of appetite, fluid balance and metabolism (Yang et al., 2000). Specifically, the neural circuit involved in the control of systemic cardiovascular function is composed of projections between the PVN, subfornical organ, rostral ventrolateral medulla and the intermediolateral grey column of the spinal cord (Badoer, 2011; Salman, 2016; Shell et al., 2016). Major projections to the PVN originate from local sites within the hypothalamus and are composed of excitatory cathecolaminergic, glutamatergic, and inhibitory GABAergic inputs (Pandit et al., 2015). The summed activity of these excitatory and inhibitory inputs regulates putative pulsatile release of effector molecules from the hypothalamic area in response to stress and fluid-electrolyte balance (Herman et al., 2002).

Projections from the PVN have been identified in the brain stem; the rostral ventrolateral medulla (RVLM) and adjacent C1 catecholaminergic region. The RVLM/C1 neurons are mostly excitatory and are involved in the sympathetic control of blood pressure through the intermediolateral cell column of the thoracolumbar spinal cord (Shell et al., 2016; Bourassa et al., 2015; Xu et al., 2012).

Previous studies have highlighted the structure of the PVN and ventral hypothalamic areas (nuclei systems). Neuronal projections from the PVN converge below the third ventricle while contributing to the arcuate paraventricular formation and arcuate fibers projecting to the contralateral side of the third ventricle (Yang et al., 2000; Pandit et al., 2005). Interestingly, the distribution of dopaminergic neurons in the arcuate paraventricular fibers has been shown to constitute part of the descending pathway that converges in the brain stem (Papp and Palkovits, 2014). However, the contribution of dopaminergic neuron groups in the median thalamus to the PVN-brain stem circuitry has not been elucidated (Bell et al., 2000). In this study, we described the comparative morphology of thalamic (Re/Zi) and paraventricular hypothalamic dopaminergic neuron groups. In addition, we demonstrate the contributions of these neurons to the PVN-RVLM circuit, which constitute a possible negative feedback loop involved in the control of hypothalamic activity.

Methods

Animal Preparation

Male Adult WT C57BL/6 (N=10) 18–20 g were used for this study. All animal handling procedures were approved by the Institutional Animal Use and Research Committee of the Louisiana State University. Surgical injections of retrograde tracer were performed using stereotaxic coordinates under anesthesia (Ketamine; 100mg/Kg) + Xylazine; 10mg/Kg).

Anatomical Tract Tracing

The projections from the PVN, Re and Zi were identified using fluorescent retrograde anatomical tract tracing and immunohistochemistry. Flame pulled glass pipettes were back-filled with mineral oil and mounted on a Nanoject (Drummond Instruments). The tip of the pipette was trimmed with forceps to increase the width to 200µm and reduce resistance to the injected bolus. Rhodamine B + Amine [10,000MW] (Life Technologies; Molecular Probe) was diluted (10%) in saline and drawn into tip of the glass pipette. After the animal has been deeply sedated, the toe was pinched to confirm the absence of pain. The scalp was carefully dissected to expose the cranium, and identify the bregma and lambdoid sutures. Subsequently, a cotton wool swab dabbed in hydrogen peroxide solution was used to clean the incised area and to remove the periosteal covering of the calvaria. The coordinate of the RVLM was determined using a stereotaxic apparatus following which a marker was used to mark the AP and ML positions relative to bregma (RVLM: Anteroposterior -6.68 mm, Mediolateral +1.35 mm) (Franklin and Paxinos, 2007). A dental drill was used to remove the bone on the marked AP/ML intersection until the dura became visible. The exposed brain area was kept moist with oxygenated artificial cerebrospinal fluid [aCSF; in mM 125 NaCl, 25 NaHCO₃, 3 KCl, 1.25 NaH₂PO₄, 1 MgCl₂, 2 CaCl₂ and 25 Glucose] throughout the duration of the experiment. The pipette (Nanoject) was lowered gradually into the RVLM (Dorsoventral: +5.7 mm) following which two boli (5nL each) were injected at 120 secs interval. After a resting period of 3 minutes, the pipette was lowered further (DV: +5.8 mm) to reach the C1 cathecolaminergic area following which the tracer was injected as described previously. After a 7-day survival period, animals were decapitated following which the whole brain was removed and kept in ice-cold oxygenated aCSF. Vibrotome sliced coronal sections (400µm thick) exposing the PVN (Bregma - 0.58mm to -1.22mm) and RVLM (Bregma -6.60mm to -6.70mm) were prepared in cold oxygenated artificial cerebrospinal fluid (ACSF) and fixed in 4% 10 mM phosphate buffered paraformaldehyde (4%PB PFA).

Immunohistochemistry

Brain slices (PVN and RVLM) were fixed in 4% PB PFA for 24 hours following which the tissue was moved to 4% PB PFA containing 30% Sucrose (cryopreservation). Subsequently, free floating cryostat sections (40 µm thick) were obtained in 10 mM PBS (pH 7.4) and washed (3 times) in PBS. In order to block non-specific binding, sections were incubated in normal goat serum (Vector Labs) prepared in 10mM PBS (with 0.03% Triton-X 100) at room temperature for 1h. Sections were washed three times in 10mM PBS following which they were incubated in primary antibody solution [Rabbit Anti-Tyrosine Hydroxylase (1:500; Abcam-ab6211), normal goat serum, 0.03% Triton-X 100, 10mM PBS-pH 7.4] overnight at 4°C. The sections were washed three times in 10mM PBS and incubated in secondary antibody (Goat anti Rabbit Alexa-568) solution at room temperature for 1h. Subsequently, the sections were washed three times and incubated in the conjugated second primary antibody (Anti-NeuN Alexa-488 conjugate; Abcam-ab190195) at room temperature for 2 hours at a dilution of 1:300. Labeled sections were washed and mounted on gelatin coated slides with plain anti-fade mounting medium (Vector Labs).

Confocal Microscopy and Statistical Analysis

Retrogradely labeled serial coronal brainstem and hypothalamic sections were imaged using an Olympus BX 51 to characterize the RVLM/C1 injection site, labeled neurons of the PVN, Re and Zi. Morphology/count of retrogradely labeled neurons and tyrosine hydroxylase immunostaining was acquired using an Olympus FluoView 10i confocal laser scanner (Olympus America Inc). The magnitude of connection between two sites in traced circuits was determined as a function of percentage of total retrogradely labeled neurons and tyrosine hydroxylase positive cell count in n=10 fields for n=6 sections. The was calculated as percentage labelled cells in total number cells/unit area (field). Cell counting was done in Image J (NIH, USA) using the method by Sonni and co-workers (2016). Comparison was done between the paraventricular area and dorsal hypothalamic area (Re/Zi) using Student’s t-test (GraphPad Prism Version 5) with Mann-Whitney post-hoc test.

Diagrammatic illustration of morphology of neurons

High contrast confocal images of the PVN, Re and Zi were selected based on the outcome of cell counting, and screening through identification. Among counted cells, the predominant retrogradely labelled cell appearance were highlighted (drawn) using a Graphic Tablet connected to PC (Huion Animation). The illustrations were made using the Draw platform on Microsoft PowerPoint (Microsoft Inc. USA). Subsequently, colors were selected to denote neurons with similar morphology following which the lines were converted into an image, and scaled approximately.

Results

Projections to RVLM/C1 from the PVN and Dorsal Hypothalamic area (Re/Zi)

Retrograde tracer (Rhodamine B + Amine) injected into the RVLM/C1 region of the brain stem (N=10; Fig. 1A) resulted in significant labeling in the PVN, and some labeling in the dorsal hypothalamic Re/Zi (p<0.01; mean±SEM) region above the third ventricle. Specifically, in the DHA the Re contained 40% of the retrogradely labeled cells while the Zi had 60% of labeled cells (Fig. 1B).

Figure 1. A-B| Retrograde neuroanatomical tracing of the PVN and DHA neuronal projections to the RVLM.

(A) Wet mount fluorescence image of retrograde tracer (Rhodamine B+) injection site in the RVLM/C1 region of the medulla [mlf: medial longitudinal fasciculus, C1: C1 adrenaline cells, Gi: Gigantocellular reticular nucleus, RVLM: Rostral ventrolateral medulla] (n=10; scale bar=200µm). (B) Low magnification wet mount image of the PVN showing the localization of Rhodamine after a survival period of 5 days. Right pane: distribution of the vesicles was also visible in the upper margins of the PVN, reuniens thalamic nuclei (Re) and Zona Incerta (Zi). Right pane: Quantification of Rhodamine fluorescence showed a higher yield of vesicles in the PVN when compared with the DHA (n=6; p<0.01) (n=10; scale bar=30µm).

In subsequent analyses, sections were immunolabeled for tyrosine hydroxylase and neuronal marker (NeuN). In the brainstem, tyrosine hydroxylase positive neurons were abundant at the injection sites - and the adjacent Gigantocellular nuclei (GiV) - when compared with the regions around the medial longitudinal fasciculus (Fig. 2A). In the hypothalamic area, the PVN, Re and Zi contained dopaminergic projections from the zona incerta to the ipsilateral Re and PVN (Fig. 2B). Taken further, we examined triple-labelling of PVN, Re and Zi neurons (Rhodamine, NeuN, Tyrosine hydroxylase) in order to quantify the percentage of retrogradely labeled dopaminergic neurons (Fig. 2C),which was significantly higher in the PVN (80%) when compared with the Re (40%) and Zi (60%) (DHA) (Fig. 2D-E; p<0.01, mean±SEM).

Figure 2. A-E| Neuronal projections to the RVLM/C1 from PVN and dorsal hypothalamic dopaminergic neuron groups (Re and Zi).

(A) Rhodamine B + Amine injection site in the RVLM/C1 region of the brainstem. Triple labeling of neurons, tyrosine hydroxylase and retrograde vesicle showed the localization of dopaminergic (catecholaminergic) neurons in the RVLM/C1 population from the anterior (Bregma -6.64mm) to the posterior (Bregma -7.08mm) RVLM. Confocal (montaged) image demonstrating the distribution of dopaminergic neurons and the retrograde tracer relative to the medial longitudinal fasciculus. [mlf: medial longitudinal fasciculus, RoB: Raphe Obscurus Nucleus, GiV: Gigantocellular nuclei] (n=12; scale bar=20µm). (B-C) Confocal image showing the localization of retrograde vesicles and tyrosine hydroxylase positive neurons of the PVN and DHA (n=12; scale bar=20µm). (D) DHA dopaminergic neurons are less clustered. Cell bodies are characterized by superiomedial and inferior projections. PVN neurons project inferiomedially with prominent elongated cell bodies (n=10) (n=12; scale bar=10µm). (E) Quantification of tyrosine hydroxylase and rhodamine retrograde vesicles in the PVN versus DHA (n=10; p<0.05; scale bar=5µm).

Morphology of PVN and Re/Zi Dopaminergic Neurons

The tyrosine hydroxylase positive cell bodies and axons also presented an interesting pattern around the third ventricle. Our results showed that retrogradely labeled DA neurons of the Zi are composed of inferiomedial and superiomedial fibers projecting to the ipsilateral PVN and contralateral Re respectively. Furthermore, horizontal groups of tyrosine hydroxylase positive fibers crisscrossed the median plane in the area anatomically occupied by the Re (Fig. 2B-C). Based on these findings, we highlighted the morphology of discrete DA neuron groups using enhanced contrast confocal imaging techniques. Similar to the observations from retrograde tracing, bilateral and median sub-groups of tyrosine hydroxylase positive neurons were identified in the Re, PVN and Zi (Fig. 3A). The morphology of the Re/Zi neurons can be described as mostly bipolar (80%), with angulated cell bodies (directed laterally). Superiomedial cellular projections from the Re contributes to the horizontal crossing fibers (median group) originating from the ipsilateral side and extends to the contralateral group of Re/Zi neurons (Fig. 3A). Inferiomedial (descending) fibers of the Re were observed to be continuous with the Zi tyrosine hydroxylase positive fibers, which converges on the ipsilateral group of PVN DA neurons (Fig. 3A). By contrast, the PVN DA neurons have slender cell bodies and projections directed inferiomedially on the lateral side of the third ventricle towards the ventral hypothalamus and arcuate paraventricular formation (Fig. 3B). Taken together, an ipsilateral Re/Zi/PVN DA convergence, together with contralateral Re-Zi DA neuronal projections suggest the presence of a superior connection for the PVN-RVLM circuit above the third ventricle: through thalamic dopaminergic inputs (Fig. 3B-VI).

Figure 3. Confocal Imaging of the DHA/PVN showing the interconnection between dopaminergic neurons.

(A) Superiomedial projections from ipsilateral Re dopaminergic neurons extends across the median plane to the contralateral Re dopaminergic neuron population. Inferiomedial projections descend into the PVN, while converging with ipsilateral Zi projections beside third ventricle (n=12) (n=12; scale bar=20µm, 10µm). (B) Diagrammatic illustration of morphology of Re/Z1, convergence zone (CZ) and PVN tyrosine hydroxylase positive neurons [(I-II)Superiomedial and inferior projections from these neurons (Blue neurons: smaller neuron sub-types) communicate with the crossing fibers and CZ neurons respectively, (III) Two sub-types of Re/Zi dopaminergic neurons projecting to the CZ are characterized by angulated cell bodies and extensive neuronal projections, (IV) Descending projections of Convergence neurons and PVN dopaminergic neurons. Three sub-types are identifiable along the lateral border of the third ventricle, (V) Horizontal crossing fibers with prominent lateral and inferiorly directed projections extends between the contralateral Re and Zi; above the third ventricle (VI) illustration of contralateral and ipsilateral projections from DA neuronal population in the Zi and Re dorsal hypothalamic area (scale bar=30µm).

Discussion

Discrete neurotransmitters systems in the PVN play important roles in the homeostatic regulation of neuroendocrine and systemic cardiovascular functions (Flak et al., 2009). In pathophysiological states, dysregulation of the excitatory (Dopamine/Glutamate) or inhibitory (GABA) neurotransmitter system is often associated with the progression of sympathoexcitation and putative release of hypothalamic neurotransmitter hormones (Angiotensin II/Corticotrophin releasing hormone - CRH); involved in stress modulation and inflammation (Masson et al., 2015; Du et al., 2015). Taken together, we have shown that the dopaminergic neuron groups of the PVN, Re and Zi are of varying morphology and are connected to the RVLM/C1. Furthermore, the Re/Zi DA neuron groups project to the ipsilateral PVN, and contralateral Re/Zi; thus forming a superior connection for the right and left PVN above the third ventricle. Together, our results create a premise for which the function of Re/Zi DA neurons, activated by PVN/RVLM inputs, may be highlighted in the complex thalamic-hypothalamic function.

Stereotaxic injection of retrograde label (Rhodamine B+ Amine, 10,000 MW) into the RVLM/C1 showed that a large percentage of PVN neurons projects to these region (Fig. 1A-B; p<0.001 versus Re/Zi). However, mapping of retrogradely labeled cells suggests that neuron sub-populations in the Zona incerta and reuniens thalamic nuclei may contribute minimally to cathecolaminergic PVN projections to the RVLM/C1. Neurochemical characterization of the injection site (RVLM/C1), and retrogradely labeled (PVN, Zi and Re) neurons showed the relative abundance of tyrosine hydroxylase positive neuron groups in the brain stem, PVN and dorsal and ventral hypothalamic area (Fig. 2A and Fig. 2B-C). In previous studies, the hypothalamic dopaminergic neuronal circuit has been identified in the physiological control of motivation, behavior, feeding, metabolism, cardiovascular function and reward (Stuber and Wise, 2016; Baskerville and Douglas, 2010). Together, the PVN and dorsal hypothalamic Re/Zi dopaminergic neurons constitute part of the cell clusters already identified in primates (A13) (Francis and Paxinos, 2007, Peyron et al., 1998). Furthermore, these cells are known projections to brainstem neuron groups involved in the control of homeostasis and metabolism (Paxinos and Franklin, 2007; Geerling et al., 2010; Stocker et al., 2006). Other studies have identified neuronal efferent from the Reuniens thalamic nuclei through the Zona incerta to reach the medial forebrain bundle and the internal capsule (Berendse and Groenewegen, 1991). In agreement with our findings, Bentivoglio at al. (1991) has described contralateral connections between the Reuniens thalamic nuclei and its role in amnesia and other behavioural alterations. Similarly, Studies by Sánchez-González et al. (2005) showed that thalamic dopaminergic projections in the monkey projects to more diverse DA sub-groups than earlier identified.

We have highlighted the structural variations in the morphology of DA neurons groups of Re, Zi and PVN in mice. Our data showed that the Re/Zi DA neurons are bipolar and angulated; oriented into superiomedial and inferior (medial) axonal projections (Fig. 2D and 3A-B). The superiomedial neuronal projections from the Re and Zi extends into the median group of horizontal neurons (thalamic; possibly Re), and further in to the contralateral Re/Zi DA neuron group. By contrast, the inferiomedial Re projections descend into the ipsilateral paraventricular area and are continuous with the Zi DA fibers in the upper part of the PVN rich in retrograde vesicles and tyrosine hydroxylase positive neurons (Fig. 2C). In the PVN, retrogradely labelled dopaminergic neurons differ significantly in morphology when compared with the Re/Zi DA neurons (Fig. 3A-C). Tyrosine hydroxylase positive/retrogradely labelled neurons in the PVN consist of vertical projections and slender cell body descending along the tapered lateral border of the ipsilateral third ventricle, into the periventricular and arcuate paraventricular regions of the ventral hypothalamus (Fig. 4).

Figure 4. Schematic Illustration of the PVN-RVLM circuit showing the area of overlap lateral to the PVN.

Retrograde tracing of PVN and Re/Zi projections to the RVLM elucidates neural connections that control pre-autonomic systemic cardiovascular function, metabolism and behavior. Furthermore, previous studies have described the role of PVN dopaminergic neurons in the control of CRH secretion and fluid balance through the hypothalamic-hypophysial system (Liposits and Paull, 1989; Stadler et al., 1992; Tchekalarova et al., 2015). However, the role of thalamic excitatory projections to the PVN in the regulation RVLM/C1 circuit has not been explored. Thus, understanding the contributions of Re/Zi dopaminergic neurons to the PVN catecholaminergic projections to the RVLM is crucial to evaluating external excitatory PVN inputs in sympathoexcitation and hypothalamic dopaminergic alteration of behavior.

Conclusion

Thalamic (Re/Zi) and PVN DA neuron groups are of varying morphology, and are connected above the third ventricle as part of the PVN-RVLM circuit.