Long-term depression

In neurophysiology, long-term depression (LTD) is an activity-dependent reduction in the efficacy of neuronal synapses lasting hours or longer following a long patterned stimulus. LTD occurs in many areas of the CNS with varying mechanisms depending upon brain region and developmental progress.^[1]

As the opposing process to long-term potentiation (LTP), LTD is one of several processes that serves to selectively weaken specific synapses in order to make constructive use of synaptic strengthening caused by LTP. This is necessary because, if allowed to continue increasing in strength, synapses would ultimately reach a ceiling level of efficiency, which would inhibit the encoding of new information.^[2] Both LTD and LTP are forms of synaptic plasticity.

LTD in the hippocampus and cerebellum have been the best characterized, but there are other brain areas in which mechanisms of LTD are understood.^[1] LTD has also been found to occur in different types of neurons that release various neurotransmitters, however, the most common neurotransmitter involved in LTD is L-glutamate. L-glutamate acts on the N-methyl-D- aspartate receptors (NMDARs), α-amino-3-hydroxy-5-methylisoxazole-4-propionic acid receptors (AMPARs), kainate receptors (KARs), and metabotropic glutamate receptors (mGluRs) during LTD. It can result from strong synaptic stimulation (as occurs in the cerebellar Purkinje cells) or from persistent weak synaptic stimulation (as in the hippocampus). Long-term potentiation (LTP) is the opposing process to LTD; it is the long-lasting increase of synaptic strength. In conjunction, LTD and LTP are factors affecting neuronal synaptic plasticity. LTD is thought to result mainly from a decrease in postsynaptic receptor density, although a decrease in presynaptic neurotransmitter release may also play a role. Cerebellar LTD has been hypothesized to be important for motor learning. However, it is likely that other plasticity mechanisms play a role as well. Hippocampal LTD may be important for the clearing of old memory traces.^[3][4] Hippocampal/cortical LTD can be dependent on NMDA receptors, metabotropic glutamate receptors (mGluR), or endocannabinoids.^[5] The result of the underlying-LTD molecular mechanism in cerebellum is the phosphorylation of AMPA glutamate receptors and their elimination from the surface of the parallel fiber-Purkinje cell (PF-PC) synapse.^[6]

It is highly important for neurons to maintain a variable range of neuronal output. If synapses were only reinforced by positive feedback, they would eventually come to the point of complete inactivity or too much activity. To prevent neurons from becoming static, there are two regulatory forms of plasticity that provide negative feedback: metaplasticity and scaling.^[7] Metaplasticity is expressed as a change in the capacity to provoke subsequent synaptic plasticity, including LTD and LTP.^[8] The Bienenstock, Cooper and Munro model (BCM model) proposes that a certain threshold exists such that a level of postsynaptic response below the threshold leads to LTD and above it leads to LTP. BCM theory further proposes that the level of this threshold depends upon the average amount of postsynaptic activity.^[9] Scaling has been found to occur when the strength of all of a neuron’s excitatory inputs are scaled up or down.^[10] LTD and LTP coincide with metaplasticity and synaptic scaling to maintain proper neuronal network function.

Long-term depression can be described as either homosynaptic plasticity or heterosynaptic plasticity. Homosynaptic LTD is restricted to the individual synapse that is activated by a low frequency stimulus.^[11] In other words, this form of LTD is activity-dependent, because the events causing the synaptic weakening occur at the same synapse that is being activated. Homosynaptic LTD is also associative in that it correlates the activation of the postsynaptic neuron with the firing of the presynaptic neuron.^[2] Heterosynaptic LTD, in contrast, occurs at synapses that are not potentiated or are inactive. The weakening of a synapse is independent of the activity of the presynaptic or postsynaptic neurons as a result of the firing of a distinct modulatory interneuron. Thus, this form of LTD impacts synapses nearby those receiving action potentials.^[11]

Endocannabinoids affect long-lasting plasticity processes in various parts of the brain, serving both as regulators of pathways and necessary retrograde messengers in specific forms of LTD. In regard to retrograde signaling, cannabinoid receptors function widely throughout the brain in presynaptic inhibition. Endocannabinoid retrograde signaling has been shown to effect LTD at corticostriatal synapses and glutamatergic synapses in the prelimbic cortex of the nucleus accumbens (NAc), and it is also involved in spike-timing-dependent LTD in the visual cortex. Endocannabinoids are implicated in LTD of inhibitory inputs (LTDi) within the basolateral nucleus of the amygdala (BLA) as well as in the stratum radiatum of the hippocampus. Additionally, endocannabinoids play an important role in regulating various forms of synaptic plasticity. They are involved in inhibition of LTD at parallel fiber Purkinje neuron synapses in the cerebellum and NMDA receptor-dependent LTD in the hippocampus.^[23]

Spike timing-dependent plasticity (STDP) refers to the timing of presynaptic and postsynaptic action potentials. STDP is a form of neuroplasticity in which a millisecond-scale change in the timing of presynaptic and postsynaptic spikes will cause differences in postsynaptic Ca²⁺ signals, inducing either LTP or LTD. LTD occurs when postsynaptic spikes precede presynaptic spikes by up to 20-50 ms.^[24] Whole-cell patch clamp experiments "in vivo" indicate that post-leading-pre spike delays elicit synaptic depression.^[24] LTP is induced when neurotransmitter release occurs 5-15 ms before a back-propagating action potential, whereas LTD is induced when the stimulus occurs 5-15 ms after the back-propagating action potential.^[25] There is a plasticity window: if the presynaptic and postsynaptic spikes are too far apart (i.e., more than 15 ms apart), there is little chance of plasticity.^[26] The possible window for LTD is wider than that for LTP^[27] – although it is important to note that this threshold depends on synaptic history.

When postsynaptic action potential firing occurs prior to presynaptic afferent firing, both presynaptic endocannabinoid (CB1) receptors and NMDA receptors are stimulated at the same time. Postsynaptic spiking alleviates the Mg²⁺ block on NMDA receptors. The postsynaptic depolarization will subside by the time an EPSP occurs, enabling Mg²⁺ to return to its inhibitory binding site. Thus, the influx of Ca²⁺ in the postsynaptic cell is reduced. CB1 receptors detect postsynaptic activity levels via retrograde endocannabinoid release.^[28]

STDP selectively enhances and consolidates specific synaptic modifications (signals), while depressing global ones (noise). This results in a sharpened signal-to-noise ratio in human cortical networks that facilitates the detection of relevant signals during information processing in humans.^[29]

Long-term depression has long been hypothesized to be an important mechanism behind motor learning and memory. Cerebellar LTD is thought to lead to motor learning, and hippocampal LTD is thought to contribute to the decay of memory. However, recent studies have found that hippocampal LTD may not act as the reverse of LTP, but may instead contribute to spatial memory formation.^[30] Although LTD is now well characterized, these hypotheses about its contribution to motor learning and memory remain controversial.^[31]

Studies have connected deficient cerebellar LTD with impaired motor learning. In one study, metabotropic glutamate receptor 1 mutant mice maintained a normal cerebellar anatomy but had weak LTD and consequently impaired motor learning.^[32] However the relationship between cerebellar LTD and motor learning has been seriously challenged. A study on rats and mice proved that normal motor learning occurs while LTD of Purkinje cells is prevented by (1R-1-benzo thiophen-5-yl-2[2-diethylamino)-ethoxy] ethanol hydrochloride (T-588).^[33] Likewise, LTD in mice was disrupted using several experimental techniques with no observable deficits in motor learning or performance.^[34] These taken together suggest that the correlation between cerebellar LTD and motor learning may have been illusory.

Studies on rats have made a connection between LTD in the hippocampus and memory. In one study, rats were exposed to a novel environment, and homosynaptic plasticity (LTD) in CA1 was observed.^[30] After the rats were brought back to their initial environment, LTD activity was lost. It was found that if the rats were exposed to novelty, the electrical stimulation required to depress synaptic transmission was of lower frequency than without novelty.^[30] When the rat was put in a novel environment, acetylcholine was released in the hippocampus from the medial septum fiber, resulting in LTD in CA1.^[30] Therefore, it has been concluded that acetylcholine facilitates LTD in CA1.^[30]

LTD has been correlated with spatial learning in rats, and it is crucial in forming a complete spatial map.^[35] It suggested that LTD and LTP work together to encode different aspects of spatial memory.^[35][36]

New evidence suggests that LTP works to encode space, whereas LTD works to encode the features of space.^[36] Specifically, it is accepted that encoding of experience takes place on a hierarchy. Encoding of new space is the priority of LTP, while information about orientation in space could be encoded by LTD in the dentate gyrus, and the finer details of space could be encoded by LTD in the CA1.^[35]

Research on the role of LTD in neurological disorders such as Alzheimer's disease (AD) is ongoing. It has been suggested that a reduction in NMDAR-dependent LTD may be due to changes not only in postsynaptic AMPARs but also in NMDARs, and these changes are perhaps present in early and mild forms of Alzheimer-type dementia.^[38]

Additionally, researchers have recently discovered a new mechanism (which involves LTD) linking soluble amyloid beta protein (Aβ) with the synaptic injury and memory loss related to AD. While Aβ's role in LTD regulation has not been clearly understood, it has been found that soluble Aβ facilitates hippocampal LTD and is mediated by a decrease in glutamate recycling at hippocampal synapses. Excess glutamate is a proposed contributor to the progressive neuronal loss involved in AD. Evidence that soluble Aβ enhances LTD through a mechanism involving altered glutamate uptake at hippocampal synapses has important implications for the initiation of synaptic failure in AD and in types of age-related Aβ accumulation. This research provides a novel understanding of the development of AD and proposes potential therapeutic targets for the disease. Further research is needed to understand how soluble amyloid beta protein specifically interferes with glutamate transporters.^[39]

In the field of research of cerebellum disorders, auto-antigens are involved in molecular cascades for induction of LTD of synaptic transmissions between parallel fibers (PFs) and Purkinje cells (PCs), a mechanism of synaptic plasticity in the cerebellum. Anti-VGCC, anti-mGluR1, and anti-GluR delta Abs-associated cerebellar ataxias share one common pathophysiological mechanism: a deregulation in PF-PC LTD. This causes an impairment of restoration or maintenance of the internal model hold by the cerebellum and triggers cerebellar ataxias. These diseases are LTDpathies.^[40]

The mechanism of long-term depression has been well characterized in limited parts of the brain. However, the way in which LTD affects motor learning and memory is still not well understood. Determining this relationship is presently one of the major focuses of LTD research.

• Brodmann area 25
• Hebbian theory
• BCM theory
• Electrical synapse
• Excitatory postsynaptic potential
• Homeostatic plasticity
• Inhibitory postsynaptic potential
• Long-term potentiation (LTP)
• Spike timing dependent plasticity (STDP)
• Neural Facilitation (Short-term plasticity)
• Neuroplasticity
• Postsynaptic potential
• Actin remodeling of neurons