Pan: Drag the map with the mouse. Zoom: Press the ALT-key and turn the mouse-wheel.
Subsections
The review papers by
Nakamura (2013);
Michelsen et al. (2007) provide a good overview
over the serotonin releasing nucleus.
The DRN seems to generate 5HT to “wait to obtain a reward” behaviour
(Nakamura, 2013).
A detailed tracing / optogenetic study can be found in
Pollak Dorocic et al. (2014). LHb, mPFC and LH appear to be the main
afferents to the DRN (
Vertes et al., 2010;
Sparta and Stuber, 2014)
(
Lee et al., 2003).
The OFC has strong reciprocal connections to with DRN (
Zhou et al., 2015)
where the OFC is probably the main nucleus of the DRN's ability to
track the long term anticipated reward and reversal learning
(
Roberts, 2011).
The mPFC (in particular its ventral part) has GLU projections to the
DRN (
Gonçalves et al., 2009;
Lee et al., 2003).
The conventional view is that the mPFC’s glutamatergic projections to
the DRN connect to locally inhibitory neurons that then target 5HT
neurons (Celada et al., 2001). Stimulation of mPFC neurons usually
inhibit DRN neurons which makes a strong case for these scenario.
However, (Pollak Dorocic et al., 2014) found that the mPFC has direct
excitatory control of 5HT which they consider to be potentially
critical for the correct function of the serotonergic system.
Similarly with regards to LHb projections to the DRN the conventional
view is that the LHb neurons target local GABA neurons that then
inhibit 5HT. However, (
Pollak Dorocic et al., 2014) found a direct
connection to the DRN. Contrary to (
Pollak Dorocic et al., 2014),
(
Ogawa et al., 2014) found few monosynaptic connections from the LHb to
the DRN and instead posits that the LHb inhibits DRN 5HT via the
rostral medial tegmental nucleus (RMTg). This has also been confirmed
by (
Sego et al., 2014).
Overall the picture emerges that the LHb exerts its influence via the
RMTg and that direct connections from the LHb to the DRN are rare.
Strong monosynaptic glutamatergic projections have been shown by
(
Lee et al., 2003) and (
Aghajanian et al., 1990). It's interesting that of
these most prominent projections mentioned above it appears that this
is the only excitatory one.
Tracing studies have shown robust projections from the Amygdala (CEA)
to the DRN (
Pollak Dorocic et al., 2014) which are monosynaptic and are
most likely glutamatergic (
Swanson and Petrovich, 1998). In contrast to the
GABAergic inputs above this seems to be one of the few excitatory
inputs.
Several nuclei from the basal ganglia project to the DRN including SNr
and globus pallidus which is inhibitory in nature
(
Pollak Dorocic et al., 2014).
The paper by
Pollak Dorocic et al. (2014) has also shown an excitatory
pathway from the ACA to the DR.
Most recently the projections to the limbic cortices have been
identified as the strongest (
Linley et al., 2013) (
Roberts, 2011)
whereas in this older publication (
Reisine et al., 1984) states that the
DRN projects to the striatum and caudate nucleus.
The paper by (Vertes et al., 2010) claims the DRN efferents include the
VTA, SNc, LH and NAcc core. (Nakamura, 2013) stresses the
projections from the DRN to the SNr, the VTA (inhibitory), amygdala,
cortex (with inhibition of the mPFC) and to the NAcc where 5HT has at
least partially a disinhibitory effect by targeting interneurons.
The actual effect of 5HT depends on the prominent receptor type in the
target area. Some 5HT receptors are excitatory, some inhibitory and
some ramp up plasticity (Frazer and Hensler, 1999).
The review paper by Michelsen et al. (2007) makes a distinction between
dorsal, medial and ventral pathways:
All parts of the striatum ranging from the Nacc shell over core to the dorsal striatum
and to a lesser extent the globus pallidus (GP).
The media part projects main to the SNr.
This pathway targets a large number of different limbic nuclei. In order of density:
- Septum (dense)
- Amygdala (dense)
- Habenula (dense)
- Piriform, insular and frontal cortices (dense)
- Occipital, entorhinal, perirhinal, frontal orbital, anterior
cingulate and infralimbic cortices (moderate)
- Thalamic and hypothalamic nuclei (dense to moderate)
- Olfactory bulb. (Lottem et al., 2016) has shown that the
spontaneous activity of the olfactory cortex is suppressed by 5HT
release but that odor evoked activity is unaffected by 5HT.
- Hippocampus
- Interpeduncular nucleus
- Geniculate body
In contrast to dopamine 5HT has a vast array of different
receptors. While the DRN and MRN generate a global 5HT signal the
effects on different brains areas can be vastly different because of
every brain area has their own 5HT receptor distribution
(
Palacios et al., 1990) (
Carhart-Harris and Nutt, 2017). Before we go into the
efferents we present the different receptors and where they are
located. If not otherwise cited it's based on the review by
(
Mengod et al., 2010) and the classic (
Palacios et al., 1990).
The 5HTR1 has numerous subclasses:
- 5HTR1A receptors are found on excitatory/pyramidal neurons and
inhibit those. This receptor has been called the "limbic" receptor
because it is prominent in the limbic areas of the brain:
hippocampus, lateral septum, cortex (cingulate/entorhinal) and
Raphe nucleus. These receptors are often co-expressed with the
excitatory 5HT2A receptor. They are located on the soma or
dendrite and thus can inhibit the firing of neurons
(Riad et al., 2000).
- 5HTR1B are found mainly on inhibitory neurons and inhibit
those but occasionally also on pyramidal neurons. They are very
prominent in the basal ganglia, in particular in the GP, SNR, VP
and EP (which are the output nuclei of the BG). They are both auto
and heterosynaptic receptors and are located at the axon terminals
(Riad et al., 2000) and control rather the release of transmitter in
contrast to 5HTR1A which control spiking.
- 5HTR1D are located in the caudate putamen, Nacc, olfactory
cortex, dorsal Raphe nucleus and locus ceruleus. It's
predominantly located on axon terminals of both 5HT and and non
5HT neurons and inhibit release of neurotransmitters.
- 5HTR1E is prominent in the (entorhinal) cortex, caudate
putamen and claustrum.
- 5HTR1F has its highest levels in the cortical regions,
olfactory bulb, Nacc, parascicular nucleus, thalamus, medial
mamillary nucleus, hippocampus, subiculum and amygdala.
Having both inhibitory 5HT receptors on both excitatory and inhibitory
neurons means that this can cancel out in average but will possibly
change the dynamics of the network.
- 5HTR2A: These receptors are excitatory and enhancing
inputs when activated meaning they are located on dendrites and on
the soma. These receptors are very prominent in the cortex and
have been localised on GABAergic interneurons but also
glutamatergic projection neurons.
- 5HTR2B: it's function and localisation is still
poorly understood
- 5HTR2C: has only been found in the CNS and there in
the choroid plexus, cortex, NAcc, hippocampus, amygdala, caudate
and SNr. They are also postsynaptic but might be also presynaptic.
It's highest concentration is in the dorsal vagal complex of the brainstem.
Is a fast excitatory receptor which is mainly located in the
hippocampus (and possibly amygdala) (Palacios et al., 1990) and is
co-expressed with GLU receptors in the hippocampus.
Some evidence points to it controlling DA release (Mengod et al., 2010).
This receptor seems to control primarily plasticity, for both LTP and
LTD (
Peñas-Cazorla and Vilaró, 2015). In an experiment by (
Mlinar et al., 2006)
stimulation of this receptor causes LTP in hippocampal slices which
were very long lasting for over 2hrs.
Even more impressive are the findings by (Hagena and Manahan-Vaughan, 2017) who show
that 5HTR4 activation shifts the frequency threshold between LTD and
LTP: it is generally accepted that under LFS LTD is induced whereas
under HFS LTP is induced. The frequency threshold where LTD turns into
LTP can be shifted by the 5HTR4 receptor. If this receptor is
stimulated even lower frequencies can cause LTP which otherwise would
have caused LTD!
In the rat it is located mainly in the limbic system: hippocampus,
striatum, inferior colliculus, SNr, ventral pallidum, fundus striatae,
olfactory tubercle, septum and amygdala. It has also high
concentrations in the parietal cortex.
As shown by (
Li et al., 2016): the activity increases when "when a mouse
voluntarily seeks and acquires sucrose, food, sex and social
interaction". 5HT neurons are activated by surprising reward events
(such as VTA neurons do) and reward predicting cues is presented. The
activity only drops off after the reward has been experienced. In
particular the DRN activity stays active while the animal is waiting
for a reward.
Numerous studies have reported that 5HT is required for delayed reward
scenarios, for example where a rat has to wait in front of a dispenser
to retrieve a reward (
Khani and Rainer, 2016) as already mentioned above
where the activity was measured during a delayed reward scenario
(
Li et al., 2016).
This has been combined into the proposal by (Miyazaki et al., 2012) that
5HT controls patience and reward.
Premature responding is increased after (Fletcher et al., 2007)
application of a 5HT(2A) receptor antagonist and decreased after
5-HT(2C) application. In earlier studies impulsivity could also be
increased by 5HT depletion (Harrison et al., 1997).
