Article Figures & Data
Figures
- Figure 1.
Cell filling of individual GnRH neurons reveals a unique projection to the median eminence (ME). A, Montage of confocal z-projections of GnRH neuron (yellow/red) in the anterior hypothalamic area projecting to the ME, which is labeled green by the dense fiber plexus of the other GFP-expressing GnRH neuron fibers. The top inset right shows the location of the neuron within the horizontal brain slice preparation. The optic tract was removed during dissection. B, Camera lucida-like reconstruction of two GnRH neurons projecting to the median eminence; one shown in A above (red) and another with its cell body located close to the median eminence (blue). a–d, Positions of the respective high-magnification pseudocolored insets. Red, Neurobiotin; green, transgene GFP; blue, B. simplicifolia Lectin 1 blood vessel label. Arrowhead in b indicates a spine. d1, Filled GnRH neuron terminal (white/pink) among terminals of other GnRH neurons (green) in the external zone of the median eminence juxtaposed to a blood vessel (blue). d2, Isosurface rendering of d1 demonstrating the close proximity between filled terminal and blood vessel. “X” in B indicates fibers that continue but could not be photographed.
- Figure 2.
GnRH neuron processes do not label for classical axonal or dendritic markers. All markers (red label) labeled appropriate neuronal compartments in various other brain areas (top) but did not localize to GnRH neurons (green, middle). The bottom part of the same field, without GnRH GFP, is given for each marker to show that it is not expressed by GnRH neurons. Dendritic markers used were microtubule-associated protein 2 (MAP2) and dephosphorylated neurofilament and kinesin-like protein 17 (KIF17). Axonal markers were Tau1 and phosphorylated neurofilaments (NF). The general neuronal marker β-3-tubulin did not label GnRH neurons. 3V, third ventricle; rPOA, rostral preoptic area; SN, substantia nigra; VTA, ventral tegmental area. Scale bars: 10 μm.
- Figure 3.
The GnRH neuron projections to the median eminence receive synaptic input throughout their length. A, Camera lucida-like reconstruction of a filled GnRH neuron (red) projecting to the median eminence (ME), double-labeled with synaptobrevin 2 (VAMP2; pseudocolored cyan) antibodies to show putative synaptic inputs. a–d, Positions of single-plane confocal high-magnification. Multiple VAMP2 puncta (arrowheads) can be observed along the projection and into the ME. B, Isosurface renderings of the cell body (top) and the distal projection (bottom) showing putative synaptic inputs positive for VAMP2 labeling (cyan). C, VAMP2 pre-adsorption control.
- Figure 4.
The GnRH neuron projections to the median eminence express the spike initiation site marker Ankyrin G. Camera lucida-like reconstruction of a filled GnRH neuron in which both projections innervate the median eminence and an Ankyrin G-positive initial segment exists in the direct projection to the median eminence (ME). Black box indicates approximate area shown to the left. a, GFP-expressing GnRH neuron with white box showing area highlighted to right. a1, Ankyrin G immunoreactivity pseudocolored magenta. a2, Endogenous GFP label with arrowheads indicating the location of overlap with the Ankyrin G signal above. Note that spines exist proximal, within, and distal to the initial segment. b, The short “stripes” of Ankyrin G label, seen here in the neocortex, are absent when the primary antibody is omitted as seen in c. Note that the second dendrite of this neuron enters the median eminence as well. “X” in reconstruction indicates fibers that continue but could not be photographed.
- Figure 5.
Neuroendocrine dopaminergic neurons observed in TH-GFP mice in the arcuate nucleus exhibit typical dendrites and an axonal projection to the median eminence (ME). Camera lucida-like reconstruction of a Neurobiotin-filled neuroendocrine dopaminergic neuron (red) from a sagittal slice with three dendrites and a separate axon that enters the ME. The relatively short, thick dendrites bear multiple long spines (a, b), are confined to the arcuate nucleus, and do not colocalize with Ankyrin G label (pseudocolored cyan; bottom). The thin axon has an Ankyrin G-positive initial segment shortly after leaving the soma (c, arrowheads), does not exhibit spines (d), and projects to the ME.
- Figure 6.
Median eminence projecting processes integrate synaptic signals to regulate excitability. A, Image of a GnRH neuron filled with Alexa Fluor 555 (red) via the recording electrode among other GnRH neuron elements (green) including a GnRH neuron cell body on the edge of the median eminence adjacent to the filled fiber in a GnRH-GFP brain slice. Scale bar represents 20 μm. B, C, Peak depolarization evoked by puff application of glutamate (B) or GABA (C) at different distances along the GnRH neuron process. Black dots, puff sites outside of the median eminence; red dots, puff sites within the lateral aspect of median eminence. Insets are representative responses. D, Flashes of 473 nm light (2 ms, blue bars) in the region around the median eminence evoke fast synaptic currents in GnRH neurons from Vgat-ChR2;GnRH-GFP mice that are blocked with the GABAA receptor antagonist picrotoxin (100 μm). E, Left, Experimental setup. Middle, Representative on-cell recording from the soma (blue traces) and 181 μm along the projection process (green traces). Traces are 10 trials overlaid. Puff application of glutamate to the soma or projection process elicited a burst of action potentials at both recording sites. Right, Example of a spike recorded at the soma and process after process glutamate puff showing the small latency difference. F, The average number of action potentials evoked with a glutamate puff onto the projection process decreased at more distal sites.
- Figure 7.
Modulation of conducted spikes by glutamate receptor activation. A, Spikes were recorded from a GnRH neuron projection with an on-cell recording electrode. Action potentials were evoked by somatic current injection either alone (black) or simultaneously with glutamate uncaging (red). B, Glutamate uncaging produced a significant broadening of the spike recorded in the projection process when compared with a control spike (p < 0.05, n = 8 neurons), with no change in spike amplitude (p > 0.05, n = 8 neurons).
