PHYSCI 147 Lecture 6 Audio Notes

Introduction to Learning Memory

• Mention of inherently controversial ideas about the fundamental mechanisms of learning and memory within neuroscience. These controversies often stem from the immense complexity of synaptic plasticity, the diverse brain regions involved, and the inherent difficulty in precisely isolating specific cellular and molecular contributions to such intricate processes.

• Announcement of personal involvement in cutting-edge AI research, specifically with a published patent on neuron-centric AI developed in collaboration with UCLA. This detail highlights the speaker's engagement in interdisciplinary science and suggests a practical application or deep theoretical understanding that extends beyond traditional neurobiology.

• Expressed uncertainty regarding the future success or impact of this AI venture, reflecting the speculative and challenging nature of innovative research and development in rapidly evolving fields like AI and neuroscience. This personal reflection adds a human element to the scientific discourse.

Focus on Long-Term Potentiation (LTP)

Background on LTP

• Overview of critical experiments conducted in the early 1990s, which were specifically designed to precisely identify the cellular and molecular mechanisms underlying long-term potentiation (LTP), a persistent strengthening of synapses based on recent patterns of activity. A central and often debated question during this period was whether the expression of LTP (i.e., the sustained increase in synaptic strength observable after induction) is primarily presynaptic (due to changes in the probability or amount of neurotransmitter release from the presynaptic terminal) or postsynaptic (due to changes in the sensitivity, number, or properties of neurotransmitter receptors on the postsynaptic membrane). Both mechanisms could contribute, and discerning their relative roles was a significant challenge.

• The significance of these findings is profound for understanding plasticity in the central nervous system, particularly in the hippocampus, a brain region universally recognized as crucial for the formation of new declarative and spatial memories. Elucidating LTP mechanisms in the hippocampus therefore provides fundamental insights into how memories are encoded and stored.

Methodology of Experiments

• Recording from Synapses:
  The experiments specifically focused on recording from pyramidal cells located within the CA1 region of the hippocampus. These CA1 neurons receive excitatory presynaptic inputs primarily from Schaffer collaterals, which are axons originating from CA3 pyramidal neurons. This specific synaptic pathway (CA3 to CA1) is a well-established and extensively studied model system for investigating the cellular and molecular basis of LTP due to its prominent role in hippocampal function and memory.

• Recording Techniques:
  Researchers employed the whole-cell patch clamp recording method, a highly sensitive electrophysiological technique that allows for real-time measurement of ionic currents flowing into or out of a single neuron. During these recordings, neurons were typically voltage-clamped, meaning their membrane potential was actively held constant at a desired level (e.g., $-70 \text{ mV}$). This voltage clamp allows researchers to precisely measure synaptic currents elicited by presynaptic stimulation without confounding changes in the membrane potential, which would otherwise influence current flow.
  The setup involves meticulously forming a giga-ohm seal ( \text{G\Omega} seal) between the ultra-fine tip of a glass patch pipette and the cell membrane. This seal electrically isolates the small patch of membrane under the pipette tip from the external medium, dramatically reducing noise and enabling the accurate measurement of minute synaptic currents.
  A major procedural difference from the standard cell-attached patch clamp (where the membrane remains intact under the tip) involves applying stronger suction after forming the giga-ohm seal. This suction is used to break through the small patch of membrane, gaining complete electrical and chemical access to the entire intracellular compartment of the neuron. This whole-cell configuration facilitates comprehensive recordings of currents from all active channels across the neuron's membrane.

Experimental Setup

• Stimulation Method:
  Researchers utilized minimal stimulation, a sophisticated technique where the intensity of the presynaptic stimulus (typically electrical current applied to presynaptic axons) is carefully lowered to a point where it is just sufficient to reliably activate a very small number of, ideally, a single presynaptic axon. The explicit goal is to isolate the synaptic response originating from one presynaptic input; however, there is an inherent experimental difficulty and no absolute guarantee that only a single axon is activated due to the variability in neuronal excitability, the anatomical complexity of axon bundles, and the practical limitations of stimulating electrodes. This technique aims to study the properties of individual synaptic connections.

• Recording Synaptic Responses:
  Successful synaptic currents, characterized by measurable postsynaptic potentials or currents (e.g., excitatory postsynaptic currents, EPSCs), were identified and meticulously distinguished from synaptic failures. Synaptic failures are defined as instances where a presynaptic action potential is unambiguously shown to be elicited (e.g., by monitoring the presynaptic fiber volley) but does not produce any detectable postsynaptic response in the recorded neuron.
  The significant meaning of synaptic failure lies in its direct implication about the probability of neurotransmitter release ($P$) from the presynaptic terminal or, in some cases, the very existence of functional synaptic connections. A high proportion of failures during minimal stimulation profoundly suggests a low probability of release, offering critical insights into presynaptic mechanisms of plasticity.

Synaptic Failures and Analysis

• Proportion of Failures:
  Synaptic responses, including both the amplitude distribution of successes and the proportion of failures, were meticulously classified and analyzed by their coefficients of variation (CV). The CV is a statistical measure of the relative variability of a set of data, calculated as the ratio of the standard deviation to the mean ($CV = \sigma / \mu$). In this neurophysiological context, changes in CV following LTP induction provided crucial clues to infer whether presynaptic or postsynaptic mechanisms were at play in altering synaptic strength.
  The importance of calculating the mean synaptic response ($M$) is underscored by the fundamental quantal hypothesis of synaptic transmission, represented by the formula:

  $M = N \times P \times Q$

  Where:

  • $N$ = The number of functional synaptic contacts or active release sites contributing to the measured response.

  • $P$ = The probability of neurotransmitter release from a single active zone when an action potential arrives.

  • $Q$ = The quantal size, representing the postsynaptic response amplitude elicited by the release of a single vesicle of neurotransmitter (a 'quantum').

• Insight on LTP Changes:
  A key insight derived from quantal analysis, particularly involving the CV, is that if the CV of synaptic responses is modified following LTP induction, it strongly suggests that the underlying change must originate from presynaptic alterations, primarily changes in $P$ (the probability of release). If LTP were solely postsynaptic (e.g., an increase in $Q$), the CV would theoretically decrease because the mean would increase while the standard deviation of quantal events might remain constant or increase proportionally differently. Conversely, an increase in $P$ would lead to a decrease in the number of failures and potentially a change in CV that is consistent with presynaptic changes. This analytical framework allowed researchers to distinguish between presynaptic and postsynaptic loci of plasticity.

Key Research Findings

• Experiment by Bolshakoff and Siegelbaum (1995):
  A seminal experiment by Bolshakoff and Siegelbaum (1995) provided compelling evidence for presynaptic LTP. They involved examining single CA1 neurons that were minimally stimulated by individual CA3 neurons. Their detailed analysis revealed distinct Gaussian curves for the amplitude distributions of synaptic failures (which clustered at zero amplitude) and successes (which showed a distribution of various amplitudes, reflecting quantal events).
  Following the induction of LTP, their results strikingly showcased shifted distributions of synaptic responses towards higher amplitudes (fewer failures, larger average success amplitude) without a significant change in the mean quantal size ($Q$), as inferred from the amplitude of the smallest successful responses. This finding strongly supported the idea that the probability of release ($P$) increases during LTP induction, rather than a primary change in the sensitivity or number of postsynaptic receptors (which would be reflected in an altered $Q$). This work significantly contributed to the understanding of presynaptic contributions to LTP.

• Rebuttal from Alternative Findings

• Despite the strong evidence from studies like Bolshakoff and Siegelbaum, the field of LTP research is characterized by robust debate. Presentation of conflicting results from another research laboratory (e.g., from Malone and colleagues, among others) highlighted instances where the presynaptic axon activation under minimal stimulation conditions involved more synapses or multiple release sites than initially expected (i.e., technical challenges in ensuring true "minimal" stimulation, or the involvement of silent synapses). These alternative findings suggested that perceived changes attributed exclusively to presynaptic activity might be misinterpreted or that postsynaptic modifications could still play a significant, perhaps even primary, role during LTP expression. This ongoing scientific discussion emphasizes the complexity of distinguishing between presynaptic and postsynaptic loci of plasticity and highlights the need for rigorous experimental controls and diverse methodologies.

Mechanisms of Postsynaptic and Presynaptic Changes

AMPA vs. NMDA Receptor Currents

• Characteristics of AMPA Receptors:
  AMPA ( $\alpha$ -amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid) receptors are ligand-gated ion channels that mediate the majority of fast excitatory synaptic transmission in the central nervous system. Upon binding of the excitatory neurotransmitter glutamate, these channels rapidly open, allowing a swift sodium influx ($\text{Na}^+$) into the postsynaptic neuron. This rapid influx of positive charge causes a rapid depolarization (an EPSC), which is typically very brief, with a response duration of around ten milliseconds. This fast kinetic property makes AMPA receptors critical for rapid signal propagation and basic information processing.

• Characteristics of NMDA Receptors:
  NMDA (N-methyl-D-aspartate) receptors are another class of ligand-gated ion channels, but they play a distinct and crucial role in synaptic plasticity, learning, and memory. Their function is slower and more complex compared to AMPA receptors. They exhibit a unique voltage-dependent magnesium block ($\text{Mg}^{2+}$) at resting membrane potentials (around $-70 \text{ mV}$). This magnesium block must first be physically relieved by significant postsynaptic depolarization (e.g., from strong or repetitive AMPA receptor activation) before glutamate binding alone can lead to the opening of the channel. Once simultaneously depolarized and bound by glutamate, NMDA receptors open, allowing the influx of both calcium ($\text{Ca}^{2+}$) and sodium ($\text{Na}^+$) ions. The influx of calcium, in particular, acts as a crucial intracellular second messenger that triggers a cascade of downstream signaling pathways, including activation of protein kinases (like CaMKII and PKC), leading directly to the enduring changes that characterize LTP expression.
  The clear distinctions drawn between AMPA and NMDA receptor activity during synaptic currents highlight their complementary roles: AMPA receptors mediate fast, baseline transmission, while NMDA receptors act as critical "coincidence detectors," signaling when presynaptic glutamate release coincides with strong postsynaptic depolarization, thus initiating plastic changes.

Silent Synapses and Their Role

• Identification of Silent Synapses:
  Silent synapses are a fascinating class of excitatory synapses characteristically identified as having functional NMDA receptors in their postsynaptic membrane but lacking or having a very low density of functional AMPA receptors. Consequently, at resting membrane potentials where NMDA receptors are blocked by magnesium, these synapses appear functionally "silent" or "dormant" when only weak presynaptic stimulation is applied, as they cannot mediate fast AMPA-receptor-mediated currents. Their "silence" suggests minimal functional activity under normal conditions.
  The presence of these silent synapses is highly significant in memory processes. They are thought to represent a potentially large, latent storage capacity within neural circuits. The idea is that these silent synapses can become "unsilenced" or functionally active during learning through the insertion of new AMPA receptors into their postsynaptic membranes, thereby transforming a previously non-functional connection into an active one. This mechanism provides a powerful way for neural circuits to rapidly form new excitatory connections and adapt their computational properties, contributing significantly to learning and memory formation without requiring de novo synapse formation. It allows for the rapid potentiation of previously weak or non-existent connections.

Mechanisms for LTP Expression and the Insertion of AMPA Receptors

Insertion Mechanisms

• AMPA Receptors Become Active:
  A major postsynaptic mechanism for LTP expression involves the dynamic regulation of AMPA receptors. This typically occurs through several processes by which existing AMPA receptors, previously residing in intracellular pools (e.g., recycling endosomes) or in perisynaptic regions (outside the active synaptic zone), rapidly migrate into the postsynaptic density (PSD) of the synapse. Alternatively, newly synthesized receptors can be transported and integrated into the postsynaptic membrane. This increase in the number of functional AMPA receptors at the synaptic cleft leads to a corresponding increase in the postsynaptic cell's sensitivity to glutamate and thus a strengthening of the synaptic response (an increase in $Q$). This entire process of AMPA receptor trafficking is highly regulated and is a cornerstone of postsynaptic LTP.

Regulatory Proteins in AMPA Insertion

• Role of TARP (Transmembrane AMPA Receptor Regulatory Proteins):
  The insertion and stabilization of AMPA receptors in the postsynaptic membrane are tightly controlled by various scaffolding and regulatory proteins. Among these, TARP (Transmembrane AMPA Receptor Regulatory Proteins) family members play a crucial role. TARP proteins (e.g., stargazin, $\gamma$ -2) act as auxiliary subunits that bind directly to AMPA receptors and are essential for their transport from the endoplasmic reticulum to the cell surface, their surface expression, and their anchoring within the postsynaptic density. Specifically, the phosphorylation of TARP proteins, such as stargazin, by kinases like CaM kinase II (Calcium/calmodulin-dependent protein kinase II) following NMDA receptor activation and $\text{Ca}^{2+}$ influx, is a critical step. This phosphorylation facilitates the enhanced transport and synaptic integration of AMPA receptors, thus contributing directly to the expression phase of LTP by mechanically linking AMPA receptors to the scaffolding proteins in the PSD.

Summary of Outcomes

• Implications for LTP:
  The discussion highlights that Long-Term Potentiation (LTP) is not a singular phenomenon mediated by a single mechanism but rather a complex outcome potentially driven by a confluence of cellular and molecular changes. These various mechanisms include: 1) The rapid and sustained insertion of AMPA receptors into the postsynaptic membrane, increasing postsynaptic sensitivity. 2) The growth of new presynaptic and postsynaptic terminals (e.g., synaptic boutons, dendritic spines), which can increase the number of functional contacts or release sites ($N$). 3) An increase in the probability of neurotransmitter release ($P$) from presynaptic terminals. 4) The mobilization and conversion of silent synapses to active synapses, effectively forming new functional connections.
  This body of research leads to a recognition of multiple, sometimes parallel or sequential, perspectives on presynaptic and postsynaptic contributions in LTP expression. The general consensus now often considers LTP as involving a dynamic interplay between both sides of the synapse, with the relative contributions potentially varying depending on the specific brain region, type of synapse, induction protocol, and developmental stage.

Conclusion and Further Study

• The final remarks emphasize the profound complexity of synaptic interactions and their multifaceted functional implications for memory formation and the modulation of synaptic efficiencies. The ability of synapses to undergo long-lasting changes in strength and efficacy is fundamental to how the brain learns and adapts. Therefore, understanding these mechanisms is paramount to comprehending higher cognitive functions.

• The lecture encourages further exploration, particularly highlighting the significant and intriguing role of silent synapses and their dynamic involvement in the intricate processes of learning and memory. Investigating how these dormant connections are recruited and modified continues to be a fertile area of research in neuroscience, promising further insights into the fundamental principles that govern brain function.