Copyright by Gregory James Ordemann 2021

Home , James Rothman, Vagusstoff

The Dissertation Committee for Gregory James Ordemann Certifies that this is the approved version of the following dissertation:

Voltage Gated Ion Channel Control of CA1 Pyramidal Neuron Function in Wild Type and fmr1 KO mice

Committee:

Darrin Brager, Supervisor

Nace Golding, Co-supervisor

Laura Colgin

Daniel Johnston

Jonathan Pierce Voltage Gated Ion Channel Control of CA1 Pyramidal Neuron Function in Wild Type and fmr1 KO mice

Gregory James Ordemann

Dissertation

Presented to the Faculty of the Graduate School of The University of Texas at Austin in Partial Fulfillment of the Requirements for the Degree of

Doctor of Philosophy

The University of Texas at Austin May 2021

To my parents For supporting me as I follow my dreams and Being there whenever I need you.

To my wife Who has made every day better than the last.

Acknowledgements

The accomplishments that have brought me to this point are far from mine to claim on my own. I have been surrounded by people, both in and out of the lab, who have guided me and kept me motivated to accomplish the behemoth task of achieving a Ph.D. I came to neuroscience with great interest, and very little experience. I will be forever grateful to Dr. Jonathan Pierce for providing me the opportunity to work as a post-baccalaureate research assistant in his lab. Dr. Brager has been an incredibly supportive, engaged, and enthusiastic mentor from answering my every inane question to listening to me ramble on about far-fetched ideas. Dr. Brager provided me with the training to succeed, the space to allow me to develop as a scientist, and the drive to design and perform experiments with precision and integrity. I have learned and achieved more than I thought was possible, and that is thanks to Dr. Brager’s excellent mentorship. I would like to thank the members of my committee, Drs. Laura Colgin, Nace Golding, Daniel Johnston, and Jonathan Pierce, for their enthusiasm and advice on topics both scientific and otherwise. The lab environment has changed drastically throughout my time at UT, however, the people have always been a positive source of support, feedback, and distraction (when necessary). I would like to thank Drs. Elizabeth Arnold, Federico Brandalise, Brian Kalmbach, Ruchi Malik,

Raymond Chitwood, Jennifer Siegel, Chung Sub Kim, and Niraj Desai, and Megan Volquardsen, Christopher Apgar, and Brandy Routh for being incredible lab mates. Dr. Richard Gray has been incredibly helpful in teaching me the importance of understanding the equipment I work with, and to be unafraid of diving into a problem head-first to fix it myself.

v Dr. Gray and Lauren Hewitt have been incredible friends, creating an environment of camaraderie that has been invaluable in developing a welcoming and collaborative environment in which ideas (and graduate students) can thrive. For that I cannot thank them enough. Thank you to the members of my cohort, Dr. Morgan Hernandez, Matthew Whitmire, and Philip Lambeth, who have been amazing friends and a fantastic support system. I would like to thank and acknowledge the tremendous amount of work done by Krystal Phu and the office staff of the Center for Learning and memory, who work tirelessly to allow graduate students to focus on their lab work. Finally, I would like to thank my family. My parents instilled in me the drive and confidence to pursue my goals, while providing love and support at every step along the way. I could not ask for better siblings, who are also some of my best friends. Thank you for keeping me from taking myself too seriously and for bringing me back to reality when I need it. To my wife, who has been willing to listen to and discuss every success and failure along the way. Thank you for being there for me every time, whether I ask or not.

Voltage Gated Ion Channel Control of CA1 Pyramidal Neuron Function in Wild Type and fmr1 KO mice

Gregory James Ordemann, Ph.D. The University of Texas at Austin, 2021

Supervisor: Darrin Brager Co-Supervisor: Nace Golding

Changes in the complement or function of ion channels can drastically affect individual neurons and their constituent circuits. Neuron function is a highly tunable system. The difference between function and dysfunction can exist on a razor’s edge. Studying neurons in disease can provide insight into the operation of brain structures. This dissertation focuses on ion channel control over individual neuron and neuronal circuit function of CA1 pyramidal neurons in wild type mice and a model of Fragile X syndrome.

Using somatic recordings, we investigated differences in CA1 pyramidal neurons across the dorsoventral axis of mouse hippocampus. Ventral neurons show depolarized resting Vm, have greater RN, and have reduced dendritic branching compared with dorsal neurons. Action potential firing was not different across the dorsoventral axis of mouse hippocampus. However, ventral neurons have a more depolarized action potential threshold compared to dorsal neurons. Action potential threshold in ventral neurons was more sensitive to block of KV1 channels compared to dorsal neurons. Outside-out voltage clamp recordings showed larger slowly inactivating K+

vii currents in ventral neurons. Despite differences in subthreshold properties between dorsal and ventral CA1 neurons, action potential output is normalized by the differential functional expression of D-type K+ channels. In investigating the effects of fmr1 KO on CA1 pyramidal neurons no difference was identified in intrinsic function across the dorsoventral axis of mouse hippocampus between wild type and fmr1 KO neurons.

We further investigated differences in wild type and fmr1 KO CA1 neurons using somatic and dendritic recordings to investigate synaptic transmission at distal inputs from entorhinal cortex. We found that TA-LTP was impaired in male fmr1 KO mice. Synaptically evoked dendritic

Ca2+ signals were smaller in fmr1 KO neurons. Threshold for Na+ dependent dspikes was depolarized in fmr1 KO mice. Dspike threshold and TA-LTP were restored by block of A-type K+ channels. TA-LTP impairment, coupled with previously described enhanced Schaffer collateral

LTP, may contribute to spatial memory alterations in FXS. Furthermore, as both of these LTP phenotypes are attributed to changes in A-type K+ channels in FXS, our findings provide a potential therapeutic target to treat cognitive impairments in FXS.

viii

Table of Contents

List of Tables ...... x

List of Figures ...... xi

Chapter 1: Introduction ...... 1

Chapter 2: D-type K+ channels normalize action potential firing between dorsal and ventral CA1 pyramidal neurons of the mouse hippocampus ...... 62

Chapter 3: Intrinsic properties of dorsal and ventral fmr1 KO CA1 pyramidal

neurons ...... 94

Chapter 4: Impaired long-term potentiation and synaptically evoked Ca2+ signaling in the TA pathway in a mouse model of Fragile X syndrome ...... 109

Chapter 5: Altered A-type K+ channel function impairs dendritic spike initiation

and TA LTP in Fragile X syndrome ...... 127

Chapter 6: Discussion ...... 164

Appendix: Methods ...... 169

References ...... 179

ix List of Tables

Table 3.1: Comparison of wild type and fmr1 KO dorsal and ventral CA1 pyramidal

neurons ...... 106

x List of Figures

Figure 1.1: The action potential model of Hodgkin and Huxley ...... 12

Figure 1.2: Single channel recordings using the patch clamp method ...... 26

Figure 1.3: Axon projections in the major excitatory pathways projecting onto CA1 pyramidal neurons within the hippocampus ...... 29

Figure 1.4: Hypothesized physiology of active dendrites in CA1 pyramidal neurons ... 35

Figure 2.1: Mapping of slice location along the dorsal ventral axis of mouse hippocampus ...... 66

Figure 2.2: Subthreshold intrinsic membrane properties in dorsal and ventral CA1 neurons ...... 70

Figure 2.3: h-channel and KIR activity does not differ between dorsal and ventral CA1 neurons ...... 72

Figure 2.4: Greater dendritic branching in dorsal compared with ventral CA1 pyramidal neurons ...... 75

Figure 2.5: Action potential threshold is more depolarized in ventral compared with dorsal CA1 neurons ...... 78

Figure 2.6: D-type K+ channel conductance density is higher in ventral compared with dorsal CA1 neurons ...... 81

Figure 2.7: Application of 4-AP abolishes the difference in action potential t

hreshold between ventral and dorsal CA1 neurons ...... 85

Figure 3.1: Mapping slice location along the dorsal ventral axis of hippocampus ...... 96

Figure 3.2: Action potential firing is not different between dorsal and ventral fmr1 KO CA1 pyramidal neurons ...... 98

Figure 3.3: Subthreshold intrinsic properties of dorsal and ventral fmr1 KO CA1 pyramidal neurons ...... 102

xi Figure 3.4: Measurements of voltage threshold for action potential generation in dorsal and ventral fmr1 KO CA1 pyramidal neurons ...... 104

Figure 4.1: TA-LTP is impaired in fmr1 KO CA1 pyramidal neurons ...... 112

Figure 4.2: CA1 pyramidal neuron morphology is not different between wild t ype and fmr1 KO mice ...... 114

Figure 4.3: Greater distance dependent decay of bAP amplitude in the dendrites of wild type compared with fmr1 KO CA1 pyramidal neurons ...... 116

Figure 4.4: NMDARs account for approximately 50% of synaptically evoked change in intracellular [Ca2+] ...... 118

Figure 4.5: Synaptic Ca2+ signaling reduced in fmr1 KO dendrites ...... 120

Figure 4.6: Changes in intracellular [Ca2+] during TBS of the TA pathway in wild type and fmr1 KO CA1 pyramidal neurons ...... 122

Figure 4.7: Dendritic recordings show a lack of TA-LTP in fmr1 KO CA1 pyramidal neurons ...... 125

Figure 5.1: Block of h-channels by 20 µM ZD7288 does not rescue TA-LTP in fmr1 KO CA1 pyramdial neurons ...... 130

Figure 5.2: NMDAR synaptically evoked activity is not different in the TA pathway between wild type and fmr1 KO CA1 pyramidal neurons ...... 132

Figure 5.3: Dendritic complex spikes are not different between wild type and fmr1 KO CA1 pyramidal neurons ...... 134

Figure 5.4: Dendritic Ca2+ spikes are not different between wild type and fmr1 KO CA1 pyramidal neurons ...... 136

Figure 5.5: Ba2+ application rescues TA-LTP in fmr1 KO CA1 pyramidal neurons ... 138

Figure 5.6: Functional measures of KIR activity show no differences between wild type and fmr1 KO CA1 pyramidal dendrites ...... 140

xii Figure 5.7: Dendritic depolarization rescues TA-LTP in fmr1 KO CA1 neurons ...... 142

Figure 5.8: Associative paring of SC and TA inputs results in normal TA-LTP in fmr1 KO CA1 pyramidal neurons ...... 144

Figure 5.9: Method for the determination of dspike threshold ...... 146

Figure 5.10: Dspike threshold is more depolarized in in fmr1 KO compared with wild type CA1 pyramidal neurons ...... 147

Figure 5.11: Somatic Na+ channels are not different between wild type and fmr1 KO CA1 pyramidal neurons ...... 150

Figure 5.12: Dendritic Na+ channels are not different between wild type and fmr1 KO CA1 pyramidal neurons ...... 151

Figure 5.13: Block of A-type K+ channels with Ba2+ hyperpolarizes dspike threshold in fmr1 KO but not wild type CA1 pyramidal neurons ...... 154

Figure 5.14: Block of A-type K+ channels hyperpolarizes dspike threshold in fmr1 KO but not wild type CA1 pyramidal dendrites ...... 155

Figure 5.15: Block of A-type K+ channels restores TA-LTP in fmr1 KO neurons ...... 157

xiii Chapter 1: Introduction

There are more things in heaven and earth, Horatio,

Than are dreamt of in your philosophy.

-William Shakespeare

Written in the 16th century, this quote from William Shakespeare’s play, Hamlet, captures a fundamental truth behind scientific pursuit. The importance of humility in what we do not know, and flexibility in what we think we understand. We do not dream of what we may find, only examine the evidence displayed before us. In this way, the open-minded exploration suggested in the quote embodies the attitude of unbiased observation that is the ideal in the pursuit of scientific advancement.

In the Summer of 1816, a group of poets found themselves isolated together due to long stretches of dismal weather in the countryside of Switzerland. For entertainment, a series of ghost stories were read amongst them. As may well be inevitable in the presence of poets of the romantic era, a challenge was issued. Each member of the group was to write a ghost story. Discussions of the recently developed theory of galvanism, stating that animals had intrinsic electricity that powered the body, had sparked Mary Shelley’s imagination. Furthermore, the recent publication of work by Luigi Galvani’s nephew, Giovanni Aldini, in which he applied electrical stimulation to recently deceased criminals to trigger muscle contraction, further stimulated the young writer’s imagination. These circumstances led Shelley to the invention of science fiction when she conceived of Frankenstein.

The novel logically advanced a recent scientific discovery to its horrifying conclusion, the reanimation of dead flesh through the use of electricity. While raising life from dead tissue has eluded medical science, it is undeniable that our bodies are powered by electrical impulses. The generation of nerve impulses causes the heart to beat, providing blood flow and oxygenation to tissue throughout the body. Our ability to move our limbs, expand and contract our lungs, and even learn are all driven by electrical activity within our bodies. These nerve impulses, or action potentials, produce the function necessary to live and interact with the world around us.

This dissertation centers around our understanding of biological electricity and its significance in the transmission of nerve impulses as the primary form of communication between neurons of the brain. The advances and discoveries that have led to the techniques used in this dissertation are presented in this introduction, providing a historical perspective leading to the understanding of neuronal function attained in the field of neuroscience.

The Action Potential

The Galvani-Volta Controversy

The experimental technique instrumental in the discovery of animal electricity was developed in the mid-17th century by Dutch natural scientist Jan Swammerdam (Cole, 1937). The preparation consisted of an excised muscle, along with the attached nerve, from a frog’s leg that was cleaned and pierced at either end by needles. By stimulating the nerve, Swammerdam observed muscle contraction in his preparation.

Luigi Galvani was an Italian scientist in the 18th century. Galvani, along with his wife,

Lucia Galeazzi, and nephew, Giovanni Aldini, utilized Swammerdam’s frog leg preparation. He

2 further developed the technique by exposing the muscles and nerves of a frog’s lower extremities and attaching a metal hook through the vertebral canal, allowing for the direct application of electricity to the nerves innervating the leg muscles. When the preparation was shocked, Galvani observed contraction of the muscle. He experimented widely with differing sources of electricity, even attaching a wire to the muscle to directly transmit atmospheric charge to stimulate the muscle.

The finding that connecting nerves from two separate frog preparations caused muscle contractions in both preparations that led him to believe that the stimulation originated from the animal tissue itself. Galvani went on to elucidate his findings by proposing fluid filled canals that transmit electrical signals in animal physiology, even suggesting that a differential build-up of positive and negative charges could underlie the generation of intrinsic electricity. The concept was termed animal electricity.

Alessandro Volta, a contemporary of Galvani, was skeptical of animal electricity. Volta claimed that Galvani’s animal electricity came from wetting two different metals. Galvani used brass hooks and an iron railing to hold the preparation in place. Volta even invented an early chemical battery by alternating zinc and copper discs separated by cloth soaked in salt-water to show how electricity could be generated in Galvani’s experimental setup. Volta’s unwavering conviction that Galvani’s animal electricity was an experimental artefact caused the concept of animal electricity to remain relatively unexplored for nearly 50 years (Schuetze, 1983; Barbara and Clarac, 2011).

The Membrane Theory

In 1842 the Italian physicist, Carlo Mateucci, published his work in which he measured injury current. Mateucci’s experiments were made possible by the invention of the galvanometer

3 by C.L. Nobili, a device that uses electromagnetism to indicate the direction and magnitude of current. For the first time physiologists were able to measure the electrical nature of muscle contractions. Using a galvanometer, Mateucci connected the intact portion of a muscle to a cut end of the same muscle and found a current was generated between the two points, indicating the presence of a voltage differential.

Mateucci’s work became known to Johannes Peter Mueller, a German physiologist.

Mueller had two students who made important contributions toward the understanding of the action potential. In 1848, Emil du Bois-Reymond repeated Mateucci’s experiments and performed the same measurement on nerves. Du Bois-Reymond identified a decrease in current across the nerve and termed it the negative variation, he had made the first recording of an action potential.

In 1849, another of Mueller’s students, Hermann von Helmholtz, measured the propagation rate of nerve impulses. Helmholtz designed a device that started a timer with the stimulation of a segment of nerve, and another trigger that stopped the timer at the other end of the nerve segment.

Helmholtz found the rate of propagation along the nerve to be 30 meters per second (Schuetze,

1983; Barbara and Clarac, 2011; Carmeliet, 2019).

Julius Bernstein, a student of du Bois-Reymond, took the next stride forward in 1868 with the development of the differential rheotome. Du Bois-Reymond had measured the action potential using a galvanometer, however, the galvanometer was best suited to measure steady currents because of its poor temporal precision. Bernstein was tasked with measuring the time course of the negative variation in addition to measuring its propagation rate, which would confirm that phenomenon was, in fact, a nerve impulse.

In order to overcome the limitations of the galvanometer’s slow initiation time, Bernstein developed a method to sample the current generated by nerve stimulation at a much faster rate.

The differential rheotome, or current slicer, made it possible to record the current from a nerve at a set delay from the stimulation of the nerve for a period of only 0.3 milliseconds. By repeating his experiment many times and varying the delay between stimulation and recording onset time,

Bernstein was able to reconstruct the current of the negative variation, or the action potential.

Furthermore, this method allowed Bernstein to calculate the velocity of electrical propagation of the negative variation. He calculated the rate at 28.7 meters per second, a value very similar to

Helmholtz’s calculation of 30 meters per second. The technical advance made by Bernstein with his differential rheotome allowed for the first measurement of the time course of the action potential.

Bernstein published two works in 1902 and 1912 that laid out his explanation of nerve activity in what he called the membrane theory. Bernstein postulated that nerve cells were surrounded by a membrane that, under normal conditions, was permeable to potassium ions. Since the concentration of potassium was greater inside the cell membrane compared with the outside, the intracellular potential would be negative. Bernstein applied the equation for diffusion potentials, developed in 1889 by Nernst, to nerve cells to express the negative resting potential caused by the concentration differential of potassium on either side of the membrane:

! �� [� ]" � = �� ! �� [� ]#

This equation, applied to a membrane permeable only to potassium, explained the baseline current observed in muscle nerves. Bernstein went on to describe the negative variation as a loss of selective permeability of the membrane. Upon stimulation the membrane would allow the 5 intracellular and extracellular spaces to equilibrate. Furthermore, this theory depends on the current of the negative variation not reaching beyond zero. At the time, Bernstein’s membrane theory was a comprehensive explanation of the current understanding of neurophysiology; however, there were two key findings that were not explained within his hypothesis. First, was one of his own. In his experiments in which he coupled multiple nerves that were stimulated simultaneously to obtain a more salient measurement he clearly observed the current of the negative variation overshoot zero, a finding that would not be possible under his membrane theory.

Secondly, a British scientist, Ernest Overton, found in 1902 that no muscle contractions could occur in the absence of either sodium or lithium ions (Schuetze, 1983; Barbara and Clarac, 2011;

Carmeliet, 2019).

Around the time of the development of Bernstein’s membrane theory, several advances were made in the understanding of the nature of the membrane surrounding nerve cells. By experimenting with anesthetics mixed with varying ratios of oil and water, Meyers and Overton independently performed a series of experiments that showed that the greater the oil content in the anesthetic, the greater its efficacy. This led to the conclusion that the membrane must be composed of lipids. The presence of a lipid membrane indicated the existence of a membrane capacitance.

Fricke and Morse were later able calculate the capacitance in erythrocytes of 0.81 microfarads per square centimeter (Carmeliet, 2019; Fricke and Morse, 1925). Finally, by laying erythrocytes out in a monolayer and finding an area twice that of intact erythrocytes, Gorter and Grendel proposed a bilayer structure to cell membranes (Gorter and Grendel, 1925; Carmeliet, 2019).

The membrane theory, along with greater understanding of membrane composition and physics, was the beginning of a molecular understanding of electrical excitation within animal

6 nerves. Bernstein’s insights into the ionic mechanism of cellular function provided a working model upon which future physiologists could build an understanding of neuron function.

Ionic Theory

Early action potential recordings relied on galvanometers to record changes in current. In

1897 the physicist Karl Ferdinand Braun invented the cathode ray tube to illustrate the emission of a negatively charged subatomic particle. Advancements in cathode ray tubes (Johnson, 1922) led to a significant reduction in the voltage level necessary to trigger the emission of electrons.

Thus, Gasser and Erlanger coupled a series of amplifiers to a cathode ray tube to create the cathode ray oscillograph (Gasser and Erlanger, 1922). Instead of relying on the relatively slow response of mechanical measurement techniques, the cathode ray oscillograph projected a visible image of the time course and amplitude of action potentials. This was the first observation of the action potential in real time and opened the door for more rigorous investigation of nerve impulses.

John Z. Young, in describing the anatomy of cephalopods, identified an extraordinarily large nerve that ranged from 0.5 to 1mm in diameter. Recognizing the value in the unusual scale of the axon, Young traveled to Woods Hole, Massachusetts in 1936 to share what he saw as an unmatched opportunity for the study of the action potential in a new preparation. Young’s preparation found a willing experimentalist in Kenneth S. Cole. Alan Hodgkin, while spending a year at the New York Rockefeller University working with Herbert Gasser, met Cole and learned how to prepare the squid giant axon for experimentation. In 1939 Cole and Curtis measured the impedance of the squid giant axon during the stimulation of an action potential. Their experiment showed that a large increase in conductance occurred during action potentials in the squid giant axon (Cole and Curtis, 1939). In the summer of that same year Alan Hodgkin and Andrew Huxley

7 measured voltage during action potential generation in the squid giant axon by running an electrode down the length of an axon segment. Using this technique, they were able to measure a resting potential in the axon of approximately -50 millivolts. Furthermore, when stimulated, the voltage of the action potential surpassed 0 millivolts and reversed polarity, something that was not thought to occur based on Bernstein’s membrane theory (Hodgkin and Huxley, 1939). Further experimentation was delayed due to the start of world war II shortly after their initial finding.

After the war, experimentation continued in earnest. It had become clear that the membrane theory failed to account for the clear reversal of membrane potential observed during the action potential (Hodgkin and Huxley, 1939; 1945). Hodgkin and Katz wrote: “Experiments with internal electrodes suggest that the active reaction of nerve is not a simple depolarization of the kind postulated by Bernstein” (Hodgkin and Katz, 1949). A new conceptualization of nerve impulse transmission was necessary. Hodgkin and Katz tested the hypothesis that nerve cell membranes become much more permeable to sodium ions during an action potential. The estimated 10-fold difference in concentration between the extracellular space (sea water for the squid) and the intracellular space would account for a sodium reversal potential of approximately 60 mV, explaining the action potential overshoot. By exchanging the normal sea water external solution with solutions made up of 33%, 50%, and 71% sea water, Hodgkin and Katz demonstrated that the delay, amplitude, and rise time of the action potential was altered with changes in sodium concentration (Hodgkin and Katz, 1949).

In 1952, Hodgkin and Huxley described their voltage clamp system, the experimental preparation of the squid giant axon, and their findings on sodium and potassium currents in nerve cells (Hodgkin and Huxley, 1939; 1952a; 1952b; 1952c; 1952d; Hodgkin et al., 1952). A diagram

8 of the voltage clamp system is presented in Figure 1.1A. Two wires were inserted into the squid giant axon, one a current wire used to apply a DC current to hold the voltage, the other a voltage sensor to measure the membrane potential of the axon relative to the extracellular space. Electrodes to measure current were placed in the extracellular space in close proximity to the axon in order to measure the ionic current flow through the membrane of the squid giant axon. In this way, the current-voltage relationship of the ion movement underlying the action potential could be studied

(Hodgkin et al., 1952).

Hodgkin and Huxley observed two distinct portions of the current response to axonal stimulation. The first they termed the early inward current, the second the late outward current.

While it was theoretically and experimentally well established that potassium permeability was a key component to the resting potential of the cell membrane, further confirmation of the sodium selectivity was necessary to confirm its contribution to the generation of the action potential.

Comparisons of current responses in the nerve cell were recorded in the presence of extracellular solutions containing seawater, 10% seawater, and a prepared solution in which choline replaced sodium. The experiment showed that the early inward component of the action potential current was highly sensitive to the concentration of sodium present in the extracellular solution (Hodgkin and Huxley, 1952b). Additionally, preventing the movement of sodium ions by replacing them with choline allowed for the separation of the sodium current and the potassium current. Thus, it was shown that the primary currents underlying the action potential are driven by sodium and potassium permeability of the axonal membrane.

Hodgkin and Huxley developed the ionic theory based on their observations of currents in the squid giant axon. They were able to develop a circuit-based model describing the function of

9 the axon membrane (Figure 1.1B) that accounts for a membrane resistance as well as capacitance.

This allowed for the application of mathematical descriptions of electrical circuits to sub-threshold graded potentials that did not elicit an action potential:

t = RC

Where tau is the time constant of changes in membrane voltage, R is the membrane resistance, and

C is the membrane capacitance. Furthermore, the amplitude of these sub-threshold events could be understood using ohm’s law:

V = IR

In which V is the voltage, I is the current, and R is the resistance of a system. By modifying

Ohm’s law, Hodgkin and Huxley represented the inverse of resistance, or conductance, to calculate the current of a given ion across the membrane at a given membrane potential. This modification of Ohm’s law is as follows:

�$ = �#"%(�$ − �#"%)

Where Im is the current across the membrane of a single ion, Gion is the conductance across the membrane of that same ion, and the phrase Vm − Eion is membrane potential minus the reversal potential of an ion, or the driving force. An ion that is being conducted across the membrane will flow in the direction that moves the membrane potential toward its equilibrium potential. So, as the membrane potential moves farther from the equilibrium potential, the electrical force applied to that ion will increase, a concept known as driving force.

Hodgkin and Huxley observed that the capacity of the axon membrane causes a capacitive current that occurs with changes in membrane voltage. This consists of either a charging or a discharging of membranes in response to changes in voltage. In order to account

10 for the membrane capacitance when calculating the total current across the membrane, Hodgkin and Huxley expressed the total membrane current with the equation:

�� = � + � $ �� #

In which I is the total current, Cm is the membrane capacitance, dV/dt is the rate of change of the membrane voltage, and Ii is the ionic current across the membrane. The proposed membrane circuit is composed of membrane capacitance, as well as a passive leak channel creating the basic RC circuit of the membrane. Hodgkin and Huxley identified the sodium and potassium currents as voltage dependent and represented them as batteries in the circuit because in combination they generate a non-linear signal that propagates down the axon membrane

(Hodgkin et al., 1952; Hodgkin and Huxley, 1952c).

On the basis of their observations in studying the squid giant axon using their voltage clamp system, Hodgkin and Huxley developed a model to describe the generation of action potentials. Their model accounts for the necessity of several components to move within the membrane before ions can be passed across the membrane. Additionally, the model includes the parameter of sodium inactivation. To describe the function of sodium channels, the expression, m3h fit the activation (m3) and inactivation (h) of sodium currents. For the slower potassium channel activation, n4 provided the best fit. When the model was used to calculate the axonal response to stimulation (point by point!) the result was remarkably similar to experimental data

(Figure 1.1C; (Hodgkin and Huxley, 1952a).

The findings of Hodgkin and Huxley’s five papers of 1952 has provided a basis for the understanding of neuron function that is still widely taught in neuroscience classrooms to this day.

A B

Figure 1.1: The action potential model of Hodgkin and Huxley. A Diagram of Hodgkin and Huxley’s voltage clamp system. B Circuit representation of neuronal membrane. C Action potential traces from Hodgkin and Huxley’s model (top) and experimental data (bottom).

Synaptic Transmission

The Neuron Doctrine versus Reticular Theory

The advent of the microscope was a key landmark in the advancement of biological studies; however, it was several hundred years between its conception and when it was directed toward the study of neurons. Invented in 1595 by Hans and Zacharias Jensen, the microscope was originally more of a curiosity that a true investigatory tool. In the 17th century, Robert Hooke and Anton van

Leeuwenhoek improved upon the tool, observing for the first time the structures that are the building blocks of all living things, what Hooke named cells (Bennett, 1999). The ability to directly observe the fundamental building block of the brain, the neuron, brought about a revolution in the study and understanding of structural components of the brain.

Advancements in brain sectioning and histological techniques by Jan Purkinje allowed for vast improvements in neuron visualization, and in 1837 Purkinje described the cell, that is now his namesake, in detail (Bennett, 1999). Despite the improvement, neurons were difficult to observe beyond the cell body. Developments in histological processing later allowed Karl Deiters to more clearly examine neurons, identifying two separate categories of process extending from the cell body. One he named protoplasmic extensions (dendrites) and the other he named the axis cylinder

(axon). In 1871 another anatomist, Joseph von Gerlach, hypothesized that action currents traveled along the processes of neurons to provide communication pathways throughout the brain.

Furthermore, he suggested that neurons were interconnected, forming a vast reticula of interconnected cells throughout the brain. A hypothesis that became known as reticular theory.

In 1873 Camillo Golgi developed his process, that he called the black reaction (now called the Golgi stain), providing for the first time the ability to clearly visualize and identify the

13 extremities of neuronal processes. The Golgi stain was particularly effective because it only stained around 1% of cells in a treated region. Those cells that were stained, however, were stained completely. This allowed for a detailed examination of individual neurons unobscured by surrounding cells. Using his technique, Golgi characterized many neurons of the brain. Golgi became a champion for reticular theory, unable to identify any separation between neuronal processes.

Converging evidence as to the inaccuracy of reticular theory was mounting. However, a large portion of the initial evidence was found in muscle fibers and it took several decades for the ideas to disseminate to the study of neurons. During his experiments measuring the action potential, Bernstein noted in 1871 that direct stimulation yielded no delay; yet, indirect stimulation produced a consistent delay of about 3 milliseconds. Observations by two anatomists, Krauss and

Kuhne, in the 1860’s led to their assertion that the muscle contracts in response to the action potential in the nerve fiber. Furthermore, the nerve acts upon the muscle through an electrical field created as the action potential invades the end plate adjacent to the muscle fibers. The discharge hypothesis elucidated a potential explanation for the transmission of electrical potential between anatomically discrete cells. du Bois-Reymond, along with his students, Carl Sachs and Sigmund

Meyer, tested the discharge hypothesis. While his findings did not confirm the discharge theory of signal transmission, neither could he discount the theory entirely. He postulated that only two methods of transmission were possible, the release of an excitant substance from the nerve ending, or electrical excitation (Grundfest, 1957). Further evidence for the separation between neurons was found by Auguste-Henri Forel who observed in 1886 that when a part of a neuron is damaged or severed, only that neuron degenerates. His findings led him to conclude that there is no evidence

14 of anastomosis, or direct connection, between the terminals of neurons. Finally, in 1849 while experimenting with the poison curare, Claude Bernard found that application prevented the contraction of muscles. While it is known now that curare is an antagonist to acetylcholine receptors, at the time the results of Bernard’s experiments would have been very difficult to interpret before the widespread knowledge of the action current, then only recently identified, had spread widely.

Santiago Ramon y Cajal began working with the Golgi method in 1887. By 1889 Ramon y Cajal had improved upon the method, presenting illustrations of his observations that vehemently contradicted reticular theory. Ramon y Cajal realized that the myelin of axons interfered with the

Golgi staining process. To overcome this interference, he used tissue from younger animals as well as birds, both measures that reduce the amount of myelin in his tissue samples. Ramon y Cajal used his more finely detailed staining technique to show that there was, in fact, separation between the terminals of neurons. Additionally, he hypothesized correctly not only the direction of information travel in neurons, but also that dendritic spines were not an artefact of histological processing, but the primary site of contact between neurons.

Although it was Ramon y Cajal’s work that refuted the assertions of Golgi and the followers of reticular theory, it was a review written by Wilhelm von Waldemeyer that outlined the neuron doctrine. The neuron doctrine stated that the cells of the brain (that he termed neurons) were individual cells that make up the fundamental structural and functional units of the nervous system. The experiments of Ramon y Cajal as well as others in support of the neuron doctrine were enough to virtually end the argument and establish the neuron doctrine as the prevailing theory of neuronal structure within the scientific community.

A key synthesis of data from the realms of electrophysiology and histology was performed by Charles Sherrington. Sherrington’s work on the reflex arc within the spinal cord led him to the idea that there is some sort of valve system which restricted flow of impulse signals to a single direction between neurons. He hypothesized a mechanism, discrete from the action potential, by which information could be translated between neurons. Sherrington also termed the junction between cells the synapse, providing nomenclature that is still used today (Bennett, 1999;

Glickstein, 2006).

The Soup versus Spark Debate

Although the biological concepts espoused by the neuron doctrine had proved to be accurate, another controversy emerged. If neurons are not connected and thus cannot directly transmit signals through cytoplasmic connections, how then are signals transmitted across the gaps between neurons?

In the early 20th century two primary theories for the transmission of signals across the synaptic cleft were proposed. One group, led primarily by John Eccles, held that transmission across the synapse is electrical in nature (spark). The other, championed by Henry Dale and Otto

Loewi, claimed that synaptic transmission is chemical (soup).

Henry Dale’s work was primarily in the peripheral nervous system, studying neuromodulators and even suggesting that noradrenaline was a more likely chemical transmitter than adrenaline. In 1913 Dale observed that intravenous injections of acetylcholine resulted in a suppression of heart rate in cats, mimicking parasympathetic nerve activity (Todman, 2008a). In

1921, Otto Loewi, a collaborator of Henry Dale performed a now famous experiment (that came to him in a dream). Loewi stimulated the vagus nerve of the heart, known to decrease heart rate.

Saline from the chamber of the stimulated heart was transferred to a chamber containing a different, unstimulated frog heart. The unstimulated heart, when exposed to the saline from the stimulated heart, slowed its rate, as if its vagus nerve had been stimulated. Loewi termed the chemical solution Vagusstoff, which was later shown by Henry Dale to be the neurotransmitter acetylcholine (Todman, 2008a). The experiments performed by Henry Dale and Otto Loewi were sufficient to allay arguments that electrical transmission occurred at peripheral synapses; however, the significantly shorter delay in nerve transmission at central nervous system synapses led many to believe that such fast transmission could only be electrical.

Chief among the proponents of electrical transmission within the central nervous system was John Eccles. Eccles was a student of Charles Sherrington, studying the reflex arc of the spinal cord in cats. Eccles persisted in his adamant stance in favor of electrical transmission until 1951, even developing a theory of electrical inhibition through intermediary cells. The advent of the intracellular micropipette as well as the design of an amplifier and recording circuit connected to an oscilloscope by Jack Coombs provided Eccles and his team with the precision of measurement necessary to provide definitive evidence on the nature of transmission between neurons of the spinal cord. By recording from neurons of the spinal cord and stimulating neurons known to have an inhibitory effect, they were able to confine their experiment to two possible outcomes. In an electrical model, stimulation of the inhibitory neuron would result in the microelectrode itself being in a positively moving electrical field (causing a measured depolarization). In a chemical model, stimulation of the inhibitory neuron would cause a hyperpolarization of the motor neuron.

As is the case in well-designed experiments, the results were clear. Eccles reliably observed hyperpolarizing potentials elicited from the excitation of inhibitory neurons of the spinal cord,

17 indicating that synaptic transmission between central neurons is chemical in nature (Todman,

2008b). With the conversion of its most vehement opponent, chemical transmission quickly became widely accepted as the primary method of communication between neurons both central and peripheral.

The Quantal Hypothesis of Neurotransmitter Release

Bernard Katz, a student of Eccles, worked with Paul Fatt to characterize the end plate potential at the neuromuscular junction using intracellular microelectrodes (Fatt and Katz, 1951).

During their experimentation they noticed spontaneous subthreshold potentials that were highly localized to the muscle end plate. Fatt and Katz hypothesized that the events they called miniature end plate potentials (mEPPs) were caused by small discharges of acetylcholine (Fatt and Katz,

1952). They reduced the extracellular concentration of calcium and measured the effects on the mEPPs. We now know that by reducing the extracellular calcium content, they reduced the probability of release at the presynaptic terminal, allowing for the observation of the spontaneous release of vesicles containing neurotransmitter. Fatt and Katz observed that, under conditions of reduced calcium, the mEPP amplitudes were very stereotyped. Furthermore, they found that mEPP amplitudes were always multiples of a baseline value of approximately 0.5 to 1 mV. This observation led to the concept of a quantum of neurotransmitter release, or a minimum amount of transmitter that can be released (Fatt and Katz, 1952; Del Castillo and Katz, 1954).

A quantitative analysis of spontaneous mEPPs led to the development of a mathematical model describing quantal release and the calculation of the probability of synaptic release. By reducing calcium and increasing magnesium in their extracellular solution, del Castillo and Katz were able to reduce the likelihood of the release of a quantum of action. They identified that the

18 release of quanta occurred stochastically and could be described by a Poisson distribution. An equation was developed to describe the release of quanta at the neuromuscular junction:

� = ��

In which m is the mean quantal content, or the ratio of the evoked end plate potential and the mEPP. The variable n is the total number of units available to act upon the end plate. And p is the average probability of response. Del Castillo and Katz speculated on the importance of calcium in adjusting m, suggesting its involvement in some aspect of quantal release (Del Castillo and Katz,

1954). The work by Katz and his colleagues provided a key piece of information to the understanding of synaptic transmission and the nature of fast chemical signaling between cells.

The invention of the electron microscope in 1931 by Ernst Ruska and Max Knoll (Knoll and Ruska, 1932) provided an incredibly powerful tool for the observation of biological material.

The electron microscope is 10 times more powerful and can resolve structures 100 times smaller than a standard light microscope. It was not until 1954 that this tool was turned to the synapse to investigate with precision the observations of Santiago Ramon y Cajal in the previous century.

Palay and Palade turned the powerful electron microscope to the synapse, providing the first images of synaptic vesicles. Palade suggested that the vesicles were related to the mEPPs observed by Katz, providing a structural basis for the theory of quantal synaptic transmission (Palay, 1956).

The Calcium Hypothesis

The importance of calcium in the release of neurotransmitter was established by Katz and his student Ricardo Miledi by removing calcium from a preparation of the squid giant synapse and finding a total lack of synaptic transmission (Katz and Miledi, 1965); however, it was in 1973 in an experiment by Ricardo Miledi that the necessity of calcium ions in the presynaptic terminal for

19 vesicle release was identified (Miledi, 1973). Miledi used intracellular micropipettes to record from the post synaptic neuron while simultaneously injecting calcium directly into the presynaptic terminal. He found that application of calcium to the presynaptic terminal triggered post-synaptic potentials where previously there had been none (Miledi, 1973). This relationship was further explored by Rodolfo Llinas in a study that identified a linear relationship between presynaptic calcium current and postsynaptic membrane potential (Llinás et al., 1976). Despite the understanding of the importance of calcium in synaptic transmission, the understanding of the molecular mechanism underlying calcium’s role in vesicle release did not come until nearly 20 years later.

The identification of the molecular link between synaptic vesicle exocytosis and presynaptic terminal calcium concentration came when groups led by Thomas Sudhof and Richard

Scheller showed that the protein synaptotagmin binds calcium. Studies using spider venom illustrated the involvement of synaptotagmin in synaptic vesicle exocytosis (Petrenko et al., 1991).

Furthermore, studies had shown that synaptotagmin was highly conserved across species, making it a likely candidate for vesicle release (Wendland et al., 1991; Perin et al., 1991). A study in which the synaptotagmin was isolated and exposed to physiological concentrations of calcium revealed that the protein cooperatively binds calcium to trigger the release of synaptic vesicles (Brose et al.,

1992; Söllner, 1993a). James Rothman, in collaboration with Scheller identified protein complex responsible for vesicle docking and fusion in the presynaptic terminal. They identified what is now called the SNARE complex, which consists of binding proteins on the presynaptic membrane, synaptobrevin and VAMP-2, and on the vesicle membrane, SNAP-25 and syntaxin, that bind to dock the vesicle on the presynaptic membrane. This complex also binds to synaptotagmin which,

20 once bound, calcium influx will cause the fusion of the vesicle membrane to the terminal membrane and thus the release of the neurotransmitter contents into the synaptic cleft. After the release of transmitter, SNAP and NSF disrupt binding of synaptotagmin and trigger the clathrin mediated process of budding and reuptake of leftover neurotransmitter for the reforming of a synaptic vesicle (Elferink et al., 1993; Söllner et al., 1993b; Hussain and Davanger, 2011).

Technical Advancements

Through the early history of the study of neuroscience it can be clearly seen that leaps in understanding of the brain are preceded by the development of technical tools. Often these tools are developed in the direct pursuit of an experimental question, as was the case with Bernstein and his current slicer. Alternatively, some technical developments sit on a shelf as a curiosity for years before the practical application for experimentation is realized, as was the case with the microscope. Perhaps the most influential experimental tool in the development of modern electrophysiology was the application of the cathode ray tube to create the oscilloscope for the direct measurement of neuron potentials. Without the ability to record neural activity at a rate that reflects the speed of biological electricity, the understanding of neuronal function would be virtually impossible. In this section I will discuss several technical advances critical to the completion of the work presented in this dissertation.

The Patch-clamp method

The development by Kenneth Cole of the intracellular micro-electrode was a key step in the understanding of electrical signals within neurons. However, Cole’s electrodes required the large scale of the squid giant axon. Experimentation was limited to a single preparation and did

21 not allow for the study of intracellular signals in neurons of the brain or muscle nerve fibers, both areas in which physiologists were historically interested.

It was with this limitation in mind that Ralph Gerard, an investigator at the University of

Chicago, set the task of developing an intracellular microelectrode for the measurement of transmembrane potentials to two of his students in succession. In 1946, Judith Graham created the first intracellular microelectrodes (Graham and Gerard, 1946). There were some limitations to these first microelectrodes; the tips were quite large, meaning that when inserted into tissue there was considerable damage. Gerard’s next student, Gilbert Ning Ling, improved upon Graham’s design by increasing the taper length and decreasing the diameter of the pipette opening (Ling and

Gerard, 1949). Although early intracellular micropipettes were pulled by hand, and thus difficult to produce reliably, this development allowed a more detailed study of excitable membranes than had previously been possible.

A major drawback of the sharp microelectrode method of cell impalement was the decreased resistance in the cell due to the leaky junction between microelectrode and membrane.

This caused microelectrode recordings to be very noisy, often making the interpretation of experimental results difficult. Additionally, it was a goal of those studying individual channel currents to obtain a clear measurement of channel activity. Both of these problems were solved with the introduction of the patch pipette. Erwin Neher and Bert Sakmann sought to isolate a patch of membrane in order to limit the amount of membrane observed in a single recording and thus identify the electrical activity of single channels. To achieve this, they used micropipettes with a large diameter, and polished the tips of the pipettes in order to seal off a section of membrane for measurement. Their technique was a success, and they saw a reduction in noise and for the first

22 time, the current of single channel openings in the muscle of a frog leg (Figure 1.2; (Neher and

Sakmann, 1976).

Over the next five years the technique was refined to improve the signal to noise ratio and to include a wide variety of recording configurations that provided researchers with the ability to interrogate single cell electrical activity with unprecedented accuracy and flexibility (Hamill et al.,

1981). The ability to perform the work that makes up this dissertation can be traced directly to the advancement of the patch clamp technique.

Visualization of living tissue

Although microscopes had advanced significantly since their conception, techniques for visualizing living tissue were lacking. Electron microscopy was a significant development in the anatomical study of neurons; however, the requirement of tissue processing eliminates the possibility of its use in physiological recordings. The thickness of tissue slices and the diffraction of light through that tissue means that light microscope images of the slice surface were ambiguous guides for recording pipettes at best. Several techniques for neuron visualization made it possible to visualize and target neuron populations for recording.

Georges Nomarski developed a method to create the illusion of three-dimensional contours in an observed sample (Nomarski, 1957). The technique, called differential interference contrast imaging (DIC), works by splitting light through a prism and polarizing it before it reaches the sample. The light then passes through the sample before being recombined by another prism. By limiting the orientation of the polarized light that is observed, differences in the optical path of the polarized light create the illusion of contours in the surface of the sample. In brain slices, this causes neurons to appear as rounded objects that cast an artificial shadow based on interference of

23 the polarized orientations of light passing through the sample. This allows for the visualization of neurons within a slice in what would otherwise appear as a flat expanse of grey tissue. Application of DIC optics with an infrared light source results in crisp images that can even be used to determine the health of individual neurons before electrophysiological recording.

Dodt contrast imaging was developed by Hans-Ulrich Dodt (Dodt and Zieglgänsberger,

1990). Dodt imaging works by masking a portion of the light being projected onto a sample. This results in oblique illumination of the sample and observable contours in the surface of the sample and thus observation of cells within a brain slice. Both techniques provide the ability for visualization of neurons within a brain slice without the necessity of staining. The observed images even look similar, both providing images clear enough even for the targeting and recording of dendrites.

Fluorescent indicators and calcium imaging

Further advances in imaging techniques as well as calcium chelators that release a fluorescent signal when bound to calcium ions allowed for the direct imaging of changes in intracellular calcium concentration. Roger Tsien, by combining calcium chelators (such as EGTA and BAPTA) with fluorescent molecules, created a dye that was highly selective for calcium and altered its fluorescent intensity with the binding of calcium ions (Tsien, 1980). Thus, experimenters were able to observe changes in calcium concentration within neurons and measure the impact of calcium signaling on neuronal function. This technique was difficult to implement based on insufficiencies in the speed and resolution of existing cameras. This limitation was addressed by the development, in 1969, of the digital charge coupled device (CCD) camera

(Janesick, 2001). CCD cameras operate by absorbing photons of light into cells and pooling them

24 to transmit an image where the light intensity of each cell is determined by the number of absorbed photons. CCD cameras are the basis for modern digital cameras and allow for the capture of images at frame rates high enough to reliably observe biological phenomena, such as changes in intracellular calcium concentrations caused by action potential mediated membrane depolarizations.

Two-photon microscopy

In 1990 the development of two-photon imaging, an imaging technique that allowed for precise imaging of a sample regardless of the depth of the imaging target within brain tissue, provided a significant advance in the ability to monitor cellular processes in living tissue. Two- photon calcium imaging operates by releasing photons of light at a frequency that, individually, would not be enough to activate a fluorophore; however, when two photons of light excite a fluorophore simultaneously the fluorescent signal will be emitted. This prevents the excitation of any fluorescent label at any point in the tissue except for the point of convergence of the emitted photons, precluding any obfuscation of signal from out of focus, excited fluorophores (Denk et al.,

1990). This method allows for precise measurements of cellular structure as well as intracellular signaling processes like changes in calcium concentration. Modern two-photon microscopes can scan a region of interest at rates exceeding 1,000 Hz, providing detailed, real time feedback of physiological processes within neurons and neuron circuits.

Figure 1.2: Single channel recordings using the patch clamp method. Current recordings of acetylcholine receptors in denvervated frog muscle.

The Hippocampus

Based on its unique structure and clear borders, it is no wonder that the hippocampus has long been identified as a subregion of the brain. The hippocampus was identified by an Italian surgeon, Giulio Cesare Aranzio, in 1587 (Engelhardt, 2016). The hippocampus is named for its resemblance to the seahorse, which in turn is named for the mythological half-horse half-fish steeds of Poseidon. The development of the Golgi stain by Camillo Golgi and its subsequent refinement by Santiago Ramon y Cajal brought about the first detailed descriptions of the organization and the proposed flow of electrical information within the hippocampus. A student of Ramon y Cajal, Lorente de No, further characterized the hippocampal formation. No made insights into propagation of nerve impulses and provided a basis for the understanding of hippocampal circuitry that would inform the understanding of neuron function at large

The hippocampal formation is a part of the limbic system located in the temporal lobe. The hippocampus itself is a curved structure located ventrocaudally to the temporal cortex. Transverse slices of the hippocampus reveal the highly organized laminar structure displaying the subregions of the hippocampus. The regions of the hippocampal formation include entorhinal cortex, subiculum, pre-subiculum, parasubiculum, the cornu ammonis regions (CA1-3), and the dentate gyrus (Amaral and Witter, 1989).

Figure 1.3 shows a transverse brain slice displaying the hippocampal circuit. The illustration exhibits a simplified view of the connections of the major neuron types within the hippocampus, with arrows indicating axonal projections. Within the hippocampal formation there are two primary pathways by which information travels through the circuit. The more widely studied is named the tri-synaptic pathway. This path begins with information being transmitted

27 from layer II of entorhinal cortex through the medial perforant path to the dendrites of dentate gyrus granule cells. Granule cells project to CA3 pyramidal neurons which in turn project to the proximal apical and basal dendrites of CA1 pyramidal neurons via the Shaffer collaterals. The second primary excitatory pathway onto CA1 neurons is a direct connection between entorhinal cortex layer III and the distal apical dendrites of CA1 pyramidal neurons. This pathway is often named the lateral perforant path, although a more anatomically descriptive name is the temporoammonic (TA) pathway. Information leaving CA1 neurons is transmitted either to the subiculum or the entorhinal cortex before projecting out of the hippocampal formation.

Descriptions of the major pathways of information transfer within the hippocampus make it clear that CA1 pyramidal neurons are a site of convergence in the hippocampus. As such, the proposed function for CA1 neurons as a comparator to identify novel stimuli. In practice, this novelty detection would occur by comparing previous experience retrieved through the tri-synaptic circuit with sensory information transmitted directly from entorhinal cortex to CA1 neurons. The timing and strength of incoming inputs, along with factors such as inhibition and neuromodulation, could result in signal convergence in CA1 pyramidal neurons (Lisman and Grace, 2005).

The relationship between hippocampal function and learning and memory was originally identified when a patient, referred to as H.M., with intractable epilepsy had a bilateral temporal lobectomy. Post-surgery, H.M. was unable to form new long-term memories, establishing a link between forming new memories and the hippocampal formation (Scoville and Milner, 1957). The link between the hippocampus and learning and memory has been studied thoroughly. The development of tools to assess learning in rodents has contributed to the understanding of the integration of sensory information into long-lasting knowledge about the environment.

CA1

CA3

Tri-synaptic loop ECIII ECII Temporoammonic pathway

Figure 1.3: Axon projections in the major excitatory pathways projecting onto CA1 pyramidal neurons within the hippocampus. ECIII: entorhinal cortex layer III. ECII: entorhinal cortex layer II. DG: dentate gyrus. CA3: cornu ammonis area 3. CA1: cornu ammonis area 1.

A significant stride in understanding the workings of the hippocampus came when the association between a sub-population of CA1 pyramidal neurons and spatial location was identified (O'Keefe and Dostrovsky, 1971; O'Keefe, 1976). Termed place cells, these neurons provide a concrete example of neurons representing the external world. The spatial representations were later found to be directly maintained by cortical inputs to the hippocampus from entorhinal cortex (Brun et al., 2002; Fyhn et al., 2004). These findings provided concrete, circuit based evidence for the function of memory through experience of the outside world, and further established the hippocampus as a center for learning and memory processes.

The development of a method of chemical lesioning using ibotenic acid injection provided researches with the ability to selectively lesion the hippocampus. Previous methods, such as aspiration or electrolytic lesion, were unable to limit the affected area to the hippocampus alone causing difficulty in the interpretation of experimental results. This, more spatially restricted, lesion method allowed researchers to test more specifically the learning and memory tasks controlled by the hippocampus while maintaining the integrity of surrounding structures within the hippocampal formation. Lesion of the hippocampus resulted in impairment of spatial memory function in locating rewards based on external spatial cues in a radial arm maze task and in locating a hidden platform in the Morris water maze. Interestingly, performance on a cue driven task using texture to indicate the correct arm for reward in a radial arm maze was unaffected after the lesioning of the hippocampus. This work supports the spatial mapping theory of the hippocampus developed alongside the identification of hippocampal place cells (Jarrard, 1993).

Further teasing apart the function of the hippocampus, a study in which the entorhinal cortex fibers projecting directly to CA1 pyramidal neurons (TA pathway) were lesioned. In testing

30 hippocampal dependent memory function, researchers found that lesioned animals showed normal short-term memory, but impaired long-term memory for platform location in the Morris water maze. These results indicate that the direct entorhinal cortex input to CA1 pyramidal neurons is critical for the consolidation of short-term memory into long-term memory (Remondes and

Schuman, 2004).

The hippocampus is a key structure for the understanding of the nature of learning and memory. The highly organized structure provides a system in which it is easy to study specific neuron types and input pathways in a targeted manner that can be significantly more complicated in other structures of the brain. The important role the hippocampus has played in our understanding of neural structure and function cannot be overstated. Although understanding of one structure by no means yields understanding of another structure, the ability to examine neurons in an easily controlled experimental environment has allowed for insights that have translated to many cell types throughout the brain.

Neuronal Dendrites

Until the middle of the 20th century the prevailing theory of dendritic function characterized dendrites as passive integrators of incoming synaptic information (Yuste and Tank, 1996).

Hodgkin and Rushton established that subthreshold potentials in invertebrate axons propagate and decay as expected based on the cable properties of neuronal processes, exemplifying the concept of decremental conduction in living tissue (Hodgkin and Rushton, 1946). Applying decremental conduction to dendrites, Lorente de No and Condouris postulated that the summation of inputs combining at branch points could be a primary point of neuronal computation before the summated

31 signals travel to the soma and axon initial segment. Under these circumstances, properly timed inputs throughout the dendritic arbor would summate at branch points to provide sufficient depolarization to elicit an action potential that would then propagate down the axon (Nó and

Condouris, 1959). The concept of dendritic decremental conduction was thought to preclude any all or nothing event that would obscure information contained within the amplitude of synaptic events. Wilfrid Rall, building upon the action potential model developed by Hodgkin and Huxley, expanded the theory to include passive dendritic properties of signal integration (Rall, 1962).

Rall’s model allowed for complex dendritic arbors to be collapsed into single compartments for physiologically relevant computational analysis. Although rooted in the idea that dendrites operate as passive integrators adhering to the fundamental properties of cable theory, Rall’s model stands as an accurate theoretical depiction of subthreshold membrane properties.

Intracellular micropipette recordings allowed for a much more detailed examination of neuronal potentials; which led to a more nuanced understanding of the computations made within neurons. The discovery that active events occur in neuronal dendrites indicated a theoretical break from the concept of neurons as only a spike or no-spike calculation in response to synaptic input.

Eccles observed in motoneurons in which the axon had been severed, that there were patches of dendritic membrane that showed abnormal excitability (Eccles et al., 1958). A true conceptualization of regenerative local activity in the dendrites was developed based on the observation of a phenomena called the fast pre-potential (FPP). It was observed in CA1 pyramidal neurons that preceding many action potentials there was a slower event that reliably triggered action potentials in the axon initial segment (Spencer and Kandel, 1961). It was concluded that

FPPs were all or nothing potentials generated in the dendrites based on their consistent amplitude

32 and time constant of decay that was faster than the membrane time constant. They proposed that there was a trigger zone in the dendrites of CA1 pyramidal neurons, likely at the nexus of dendritic arborization, where regenerative events could be initiated that would reliably propagate to the soma to elicit an action potential (Figure 1.3). The work of Spencer and Kandel pushed against the dendritic theory of passive integration. Their conclusions, drawn from only somatic recordings, were highly contentious. Interestingly, although the concept of regenerative events local to the dendrites has become well established, the identity and origination of FPPs remains unclear.

The concept of active dendrites was shown in Purkinje cell dendrites of alligator cerebellum through field potential recordings. Experimenters showed that extracellular potentials deviated from expectations based on passive propagation (Nicholson and Llinás, 1971).

Intracellular dendritic recordings later confirmed the existence of local dendritic spikes in Purkinje cells in response to depolarization. Furthermore, it was identified that the primary underlying conductance of the dendritic spikes was voltage gated calcium channels (Llinás and Sugimori,

1980). The technical advance of using brain slices to isolate brain regions and provide improved access and precision for neuron recordings made much more detailed analyses of neurophysiological phenomena. Additionally, experiments using a brain slice in a saline bath made concentration specific pharmacological manipulations possible (Yuste and Tank, 1996).

Experiments in CA1 pyramidal neurons, recording from isolated dendrites in which the soma had been severed, revealed sodium and calcium mediated regenerative events of dendritic origin (Benardo et al., 1982). Although intradendritic recordings using sharp microelectrodes vastly improved the ability to monitor neuronal potentials, the application of patch clamp recordings to the dendritic membrane provided a systematic method for the accurate recording of

33 membrane potential changes and channel activity within the dendritic membrane. The advancements in recording techniques and visualization of brain slices provided the tools necessary for a detailed dissection of both subthreshold and regenerative events in neuronal dendrites.

Figure 1.4: Hypothesized physiology of active dendrites in CA1 pyramidal neurons.

Subthreshold Dendritic Properties

The effects of membrane properties such as resistance and capacitance on dendritic potentials was well described in Wilfrid Rall’s dendritic model (Rall, 1959; 1962). Although the electrical properties of membrane are useful tools in the understanding of signal integration and propagation that occurs in neurons dendrites, the expression of ion channels and their modulation through second messenger signaling is a vast and complex system. This section focuses on ion channels that affect subthreshold membrane properties examined within this dissertation.

Experimental evidence from intracellular dendritic recordings revealed that subthreshold events did not always follow the expected patterns predicted by cable theory and decremental conduction. Most notably, measurements of pyramidal neuron dendrites uncovered a membrane

“anomalous rectification” that occurred when subthreshold current injections were applied to the membrane (Benardo et al., 1982). Instead of rising in an exponential curve to a steady state potential, the membrane voltage would rise to a peak and then rectify to a reduced amplitude. It was later identified that the channel underlying this effect is the h-channel.

h-channels are hyperpolarization activated, non-inactivating cyclic nucleotide gated

(HCN) mixed conductance cation channels (Halliwell and Adams, 1982; Maccaferri et al., 1993;

Magee, 1998). A significant number of h-channels are open near the resting membrane potential.

As such Ih contributes to both the resting membrane potential and input resistance (Halliwell and

Adams, 1982; Maccaferri et al., 1993; Magee, 1998). Ih also acts to oppose changes in membrane potential, as the membrane potential depolarizes (or hyperpolarizes) h-channels will deactivate (or activate), thus reducing an inward current and opposing the depolarization (or adding an inward current and opposing the hyperpolarization). This opposition to membrane depolarization allows

36 h-channels to control the temporal summation of synaptic inputs (Magee, 1998; 1999). The combination of biophysical properties and increasing dendritic density in CA1 pyramidal neurons allow h-channels to normalize the contribution of synaptic inputs to somatic depolarization independent of distance along the apical dendrite (Magee, 1998; Lörincz et al., 2002). The effects that Ih has on dendritic input resistance and resting membrane potential modulate active events involved in the induction of long-term potentiation (LTP). In CA1 pyramidal neurons, the enriched expression of h-channels in the distal dendrites constrains LTP and LTD (Brager and Johnston,

2007; Narayanan and Johnston, 2007; Tsay et al., 2007; Brager et al., 2013). Ih expression, paired with membrane filtering properties, causes the dendritic membrane to act as a band pass filter tuned to events in the theta frequency (Narayanan and Johnston, 2007; Vaidya and Johnston,

2013).

Another potent regulator of dendritic input resistance and membrane potential are inward rectifying potassium channels. Inward rectifying potassium channels are channels that pass current into the cell membrane more easily than out, making them important for the regulation of intracellular potassium (Hibino et al., 2010). Classes of these channels can be activated by either voltage or G-protein coupled receptor activity, meaning that a subset of the channels are consistently activated. The effect inward rectifying potassium channels can have on the properties of neuronal dendrites is best displayed by the differential expression across the dorsal-ventral extent of the hippocampus. The dorsal hippocampus has a greater expression of inward rectifying potassium channels compared with ventral hippocampus. This leads to differences in action potential firing, resting membrane potential, input resistance, and constrains the induction of LTP

37 in CA1 pyramidal neurons (Dougherty et al., 2012; Kim and Johnston, 2015; Malik et al., 2015;

Malik and Johnston, 2017).

Ion channels that modulate the subthreshold function within neuronal dendrites can have profound effects on the activity of a neuron, whether in the formation of regenerative events like the action potential or in the modulation of syaptically generated events.

Active Events in Neuronal Dendrites

Investigations into suprathreshold events that occur in neuronal dendrites have revealed the complexity and diversity of active events and the impact they have on cellular function. These events are greatly influence by the pattern and vast diversity of ion channel expression within the dendritic membrane. This section focuses on the active events localized within the dendrites and the ion channels that influence their shape and generation as well as those channels most relevant to this dissertation.

The understanding within the neuroscience community of dendritic function has expanded tremendously since the advent of intradendritic recording. However, our understanding is far from complete. The integration of findings from all levels of study, from molecular to behavioral, are necessary to understand how individually characterized events interact within a whole organism to result in animal behavior. Despite the need for a cohesive approach to the study of how dendritic properties impact whole systems, many of the events that occur under physiological conditions consist of a mixture of subtypes of active events that have been individually characterized. It is important to understand the individual phenomena and their impact on neuron function to understand how a system functions as a whole.

Back-Propagating Action Potentials

When an action potential is triggered at the axon initial segment the impulse travels orthodromically down the axon toward the axon terminal and the synaptic cleft. However, the impulse also travels antidromically back through the soma and into the dendritic compartment.

Patch clamping dendrites in a mammalian brain slice preparation (Stuart et al., 1993) provided unprecedented access to the physiological details of dendrites. The technique of patch clamping, through the outside out patch configuration, allowed for direct demonstration that voltage gated sodium channels were present and functional within the dendritic membrane of cortical pyramidal neurons (Stuart and Sakmann, 1994). Through patch clamping both the soma and dendrite of a single neuron, it was observed that the dendritic spike always followed the somatic spike initiated in the axon initial segment. It was concluded that dendritic sodium spikes were generated in the axon initial segment before back-propagating into the dendrites where voltage gated sodium channels would actively boost the signal (Stuart and Sakmann, 1994).

Back-propagating action potentials (bAPs) rapidly decrease in amplitude as they invade the dendritic arbor, nearly disappearing in the distal tuft. A portion of this decay can be explained by decreased dendritic input resistance, largely due to the greater expression of Ih in pyramidal neuron dendrites (Magee, 1998), and passive decay of signal due to the cable properties of dendrites. However, these factors do not fully account for the observed decrement in bAP signal, especially given the presence of dendritic voltage gated sodium channels that should act to boost the amplitude of bAPs. It was identified that a transient outward current expressed in pyramidal neuron dendrites regulates the amplitude of supra threshold signals (Hoffman et al., 1997).

A-type potassium channels are rapidly activating and inactivating voltage-gated channels

(Connor and Stevens, 1971; Rudy et al., 1988). Evidence suggests that Kv4.2 is the putative pore- forming subunit of A-type K+ channels in CA1 pyramidal neurons (Chen et al., 2006). Research shows that A-type K+ channels are expressed at higher densities in CA1 neuron dendrites compared with WT neurons (Hoffman et al., 1997). The increasing density of IKA with increasing distance from the soma in CA1 neuron dendrites acts to suppress the amplitude of action potentials backpropagating from the soma and thus normalizes the resulting influx of calcium between the larger more proximal dendritic arbor and the smaller distal dendrites (Frick et al., 2003). Because calcium is the molecular trigger for synaptic plasticity, changes in the regulation of calcium within neuronal dendrites has far-reaching implications for cellular learning and memory. The rapid inactivation of A-type K+ channels causes them to act as dendritic coincidence detectors (Hoffman et al., 1997;

Migliore et al., 1999; Stuart and Häusser, 2001). This effect occurs through the tightly regulated timing of synaptic potentials and backpropagating action potentials, both of which are modulated by IKA. The amplitude of dendritic events has been shown to be directly associated with the timing between excitatory synaptic potentials and antidromically propagating action potentials triggered from the axon hillock, with the maximal response occurring when EPSPs precede bAPs (Hoffman et al., 1997; Migliore et al., 1999).

Unsurprisingly, the amplification of dendritic signal due to the timing of synaptic input and action potential back-propagation increases the likelihood of LTP induction at activated synapses

(Magee and Johnston, 1997; Johnston et al., 2003). Furthermore, PKA and PKC activity can increase the activation voltage of IKA, limiting its modulatory effect on bAPs and allowing for

40 greater amplitude of bAPs and increased propagation distance into the dendritic arbor (Hoffman and Johnston, 1998).

Dendritic Sodium Spikes

A detailed, biophysical characterization of voltage gated sodium and calcium channels was performed, finding an abundance of both throughout the dendritic membrane of CA1 pyramidal neurons of the hippocampus (Magee and Johnston, 1995a; 1995b). Dendritic sodium channels in

CA1 pyramidal neurons have a hyperpolarized activation voltage relative to somatic sodium channels. Additionally, protein kinase activity depolarizes the activation curve for sodium channels, indicating that phosphorylation of dendritic sodium channels could significantly modulate the generation of active dendritic events (Gasparini and Magee, 2002). Although sharp microelectrode studies had shown that sodium and calcium spikes could occur in neuronal dendrites (Llinás and Sugimori, 1980; Benardo et al., 1982), the biophysical properties of the channels underlying those spikes were thus far uncharacterized. Furthermore, it was unclear whether dendrites could generate local, sodium dependent spikes or if the bAP signal was the primary trigger for active dendritic signaling. To directly test whether local dendritic sodium spike generation could occur dual recordings were made during synaptic stimulation of distal synaptic inputs. It was definitively shown that local dendritic spikes could, in fact, be elicited through synaptic stimulation (Golding and Spruston, 1998). The initiation of dendritic spikes is dependent on highly constrained spatial and temporal coordination. Activation of multiple inputs in relatively tight groupings are required for the local generation of dendritic sodium spikes (Gasparini et al.,

2004). Dendritic sodium spikes in the distal dendritic inputs onto CA1 pyramidal neurons are critical triggers of LTP (Golding et al., 2002; Kim et al., 2015), acting in a similar way as bAPs at

41 proximal CA1 neuron inputs as a coincident signal for the activation of NMDA receptor dependent processes.

Dendritic Calcium Spikes

The pharmacological blockade of sodium channels in slice preparations results in the generation of calcium dependent spikes in response to depolarizing current or synaptic stimulation

(Benardo et al., 1982; Andreasen and Lambert, 1995; Schiller et al., 1997; Golding et al., 1999;

Larkum et al., 1999). Dendritic calcium spikes are behaviorally relevant signals that coordinate activity within cortical layers during sensory-motor tasks (Xu et al., 2012). A study using cell attached patch clamp recordings of dendritic voltage gated calcium channels in CA1 pyramidal neurons identified low and high voltage activated calcium channels in dendrites of CA1 pyramidal neurons (Magee and Johnston, 1995a).

Voltage gated calcium channels can be separated into classes based on their voltage of activation and functional properties. High voltage activated calcium channels include L, N, P/Q, and R-type voltage gated calcium channels (VGCC). L-type VGCCs produce long lasting, large scale currents. L-type channels are of particular functional importance in post-synaptic membrane in the brain based on their proposed role as the synaptic tag in LTP as well as their role in the structural formation and modification of synapses (Dolphin, 2006; Wheeler et al., 2008; 2012;

Dolphin, 2016; Li et al., 2016; Cohen et al., 2018). N-type and P/Q-type calcium channels produce a non-inactivating current and are primarily involved in the release of synaptic vesicles (Dolphin,

2006; 2016). N-type channels are primarily expressed in neurons and are implicated in a type of presynaptic LTP in the distal dendritic inputs of CA1 pyramidal neurons (Ahmed and Siegelbaum,

2009). R-type calcium channels are slowly inactivating channels that are of particular importance

42 in synaptic signaling during synaptic stimulation in CA1 pyramidal neurons (Dolphin, 2006;

Takahashi and Magee, 2009; Dolphin, 2016). High voltage activated calcium channels include the transiently activated T-type channels. These channels often require an initial hyperpolarization to become available for activation and based on their transient current profile, support bursting activity in many neuron subtypes (Dolphin, 2006; Cain and Snutch, 2013; Dolphin, 2016).

The relevance of dendritic calcium spikes, however, is largely interdependent with the activation of voltage gated sodium channels. The under physiological circumstances these events occur together, particularly in the hippocampus, where dendritic sodium spikes often trigger the generation of dendritic calcium spikes (Golding et al., 1999).

Dendritic Complex Spikes and Plateau Potentials

Complex spikes are generated in a subset of CA1 pyramidal neuron dendrites (Andreasen and Lambert, 1995; Golding et al., 1999). It is hypothesized that these large-scale calcium signals activate signaling pathways involved in LTP and signal integration in neuronal dendrites, however, a direct experimental link has yet to be drawn between these events. Complex spikes are triggered by the generation of a dendritic sodium spike that activates voltage gated calcium channels resulting in 1-3 calcium dependent spikes (Golding et al., 1999). The duration and amplitude of the calcium dependent portion of these events is largely regulated by various voltage activated potassium channels (Andreasen and Lambert, 1995; Golding et al., 1999).

Exerting the greatest influence on the generation of complex spikes are D-type potassium channels (Golding et al., 1999). D-type potassium channels are delayed rectifiers of the Kv1 family that activate at relatively hyperpolarized potentials and are slow to inactivate. D-type potassium channels are expressed primarily in the soma, with lower expression levels in the dendrites of CA1

43 pyramidal neurons (Golding et al., 1999; Routh et al., 2013). D-type potassium channels influence action potential threshold, particularly at the soma (Storm, 1988; 1993; Bekkers and Delaney,

2001; Guan et al., 2007; Yu et al., 2008; Higgs and Spain, 2011). In addition to affecting somatic action potential threshold, these channels suppress the generation of calcium spikes in the soma and proximal dendrites of CA1 pyramidal neurons (Golding et al., 1999).

Dendritic plateau potentials are associative signals that occur with the activation of multiple input pathways in the dendritic arbor of CA1 pyramidal neurons (Takahashi and Magee,

2009). During high frequency synaptic stimulation of both the proximal and distal input pathways onto CA1 pyramidal neurons it was observed that large scale depolarization persisted even after the stimulation train ended. These plateau potentials were found to be dependent upon R-type calcium channel and NMDA receptor activity and are powerful triggers of LTP (Takahashi and

Magee, 2009). Plateau potentials in the distal dendrites of CA1 neurons result in prolonged depolarization that is reliably transmitted to the soma, eliciting burst firing at the axon initial segment. Artificial induction of plateau potentials in rodents exploring a track induced the formation of place cells, cells sensitive to a particular region in space. These findings indicate that distal dendritic plateau potentials are critical to the function of spatial memory within the hippocampus (Bittner et al., 2017).

The evidence makes a clear argument for the importance of active dendritic events and integrative properties in cell signaling and animal behavior. The complex computations made within the dendritic arbor are likely a key element to understanding the overall function of neuronal circuits in the generation of behavior.

Long-Term Synaptic Plasticity

Donald Hebb, a neurophysiologist in the mid-20th century, theorized on the nature of learning and memory in the brain in his text, The Organization of Behavior: A Neurophysiologcial

Theory. While his ideas were influenced by other researchers of the time, his postulate has become widely known as an almost prescient view on the nature of synaptic interactions. He postulates that, “When an axon of cell A is near enough to excite a cell B and repeatedly or persistently takes part in firing it, some growth process or metabolic change takes place in one or both cells such that

A’s efficiency, as one of the cells firing B, is increased” (Hebb, 1949). Indeed, what has become known as Hebbian learning was shown to be correct over 20 years after Hebb’s presentation of his theory.

Long-term plasticity, the mechanisms by which neurons can adjust the strength of synaptic connections in the brain, was first identified in rabbits at the axonal connection between entorhinal cortex layer II and dentate gyrus granule cells of the hippocampus (Bliss and Lomo, 1973). By delivering high frequency, presynaptic stimulation the synaptic response to single stimuli was increased. This increase in the efficacy of single stimuli to elicit a response in the postsynaptic membrane was termed long-term potentiation (LTP). The discovery of LTP generated excitement in the study of synaptic transmission, in particular the speculation presented by Bliss and Lomo on the possibility of LTP being involved in the process of learning and memory.

In the decades after the discovery of long-term plasticity a great deal has been learned concerning the structural and functional mechanisms involved. In conjunction with a mechanism to increase synaptic efficacy, there must be an opposite to downregulate synaptic strength in order to maintain balance within the nervous system. This mechanism does exist and is called long-term

45 depression (LTD). It was first identified in the purkinje cells of the cerebellum by Ito and Kano in

1982 (Ito and Kano, 1982). The process of LTD creates a use dependent model where the greater the frequency of synaptic stimulation the more effective that synaptic connection becomes.

Conversely, if the frequency of presynaptic activation decreases, then depression can occur.

Although there are numerous mechanisms underlying the processes of long-term plasticity, this dissertation focuses on LTP. Thus, the background provided here will include relevant mechanisms of LTP within the hippocampal circuit.

NMDA receptor mediated LTP: Induction

The laminar structure of the hippocampus and the well characterized circuitry made the region ideal for the study of LTP. The ability to elicit LTP in hippocampal slices (Schwartzkroin and Wester, 1975) added to the experimental viability of the hippocampus as a region of interest.

The most common form of LTP resembles that identified at the Schaffer collateral synapse between CA3 and CA1.

For LTP to occur the delivery of high frequency stimuli is required (usually around 100

Hz). This phase is often referred to as the induction phase. The next stage of LTP in which synaptic strength is increased is termed the expression phase. A final phase, that often lasts beyond experimental observation, is called the maintenance phase; called so because the increase in synaptic strength is maintained long after the induction stimulus is delivered. Early experiments explored the necessary conditions for the induction stimulus to effectively result in LTP. It was identified that a weak high frequency stimulation (or tetanus) could not induce LTP; whereas, a strong stimulus reliably resulted in LTP. This concept was termed cooperativity. Additionally, if a single synapse is weakly stimulated, such that no LTP would be induced on its own, and another

46 neighboring synapse is strongly stimulated then both synapses would undergo LTP. This phenomenon was called associativity (McNaughton et al., 1978; Levy and Steward, 1979). Also identified during these experiments was the fact that a synapse that was not activated by tetanic stimulation could not undergo LTP, indicating that LTP induction shows input specificity.

The work of Jeff Watkins was instrumental in the identification of glutamate as the primary amino acid and AMPA, NMDA, and Kainate receptors as the major post-synaptic effectors of chemical transduction in the central nervous system (Olverman et al., 1984; Watkins and Jane,

2006). Additionally, he provided an invaluable tool for the study of synaptic transmission and long-term plasticity, the NMDA receptor blocker D-AP5 (Evans et al., 1982). In fact, the year after its identification D-AP5 was applied during tetanic stimulation to show that the induction of LTP requires NMDA receptor activation (Collingridge et al., 1983). Further evidence of the importance of glutamate in synaptic transmission and LTP was provided by identifying a large increase in glutamate levels during tetanic stimulation in the hippocampus (Dolphin et al., 1982).

Much of the mechanistic underpinnings of the induction of LTP was made clear by the understanding of NMDA receptor function. The key attribute of NMDA receptors is their ability to function as a coincidence detector. In 1984 it was discovered that magnesium ions block NMDA receptors at resting membrane potentials (Mayer et al., 1984; Nowak et al., 1984). The negatively charged intracellular space exerts an attractive electrical force on the positively charged magnesium ion that binds to the pore of NMDA receptors. When the postsynaptic cell is depolarized the electrical attraction exerted on the magnesium ion is lessened or even reversed causing a release of the magnesium ion blocking an NMDA receptor. Thus, if the postsynaptic cell is sufficiently depolarized to cause a release of magnesium and there is glutamate in the synaptic

47 cleft available to bind to the NMDA receptor to cause activation, then there can be current flow through the channel pore. The relation between these two stipulations for NMDA receptor activation and LTP became clear when it was discovered that postsynaptic depolarization is required for LTP. Furthermore, LTP could be blocked by hyperpolarization of the postsynaptic membrane, and a weak stimulus could induce LTP with an associated non-synaptic depolarizing stimulus (Kelso et al., 1986; Malinow and Miller, 1986; Gustafsson et al., 1987). These findings also assigned a mechanism to the LTP trait of associativity identified nearly a decade previously.

Once again calcium rose to the forefront of the investigation of synaptic transmission when it was identified that the calcium chelator, EGTA, prevents the induction of LTP (Lynch et al.,

1983). It was later found that NMDA receptors are calcium permeable, indicating that activation of and calcium entry through NMDA receptors are critical for the induction of LTP (MacDermott et al., 1986; Ascher and Nowak, 1988). The induction period of the phenomenon of LTP had clearly been linked to NMDA receptor function, however, the question remained as to whether the site of expression of LTP was pre or postsynaptic.

To address the question of the location of synaptic alteration during LTP, quantal analysis, developed by del Castillo and Katz (Del Castillo and Katz, 1954) at the neuromuscular junction, was applied to synapses of the central nervous system. If there was a change in the number of failures to elicit vesicle release, then the locus of LTP was presynaptic based on an increase in the probability of transmitter release. If the probability remained the same but the spontaneous EPSP amplitude increased, then the locus was postsynaptic. Quantal analysis was first applied to LTP at the crayfish neuromuscular junction. It was shown that the amplitude of mEPPs remained unchanged after tetanic stimulation, however, the quantal content (m) of EPPs was increased to

48 the same degree as the potentiated synaptic voltage signals indicating an increase in neurotransmitter release and a presynaptic locus of LTP (Baxter et al., 1985). The same methodology was later applied to central synapses of the mammalian hippocampus with similar results indicating presynaptic expression of LTP (Bekkers and Stevens, 1990; Malinow and Tsien,

1990). These findings were supported by previous results showing an increase in neurotransmitter during LTP induction (Dolphin et al., 1982). So, it was widely accepted that LTP induction occurred through the activation of NMDA receptors, and LTP expression occurred in presynaptic terminals.

The identification of silent synapses simultaneously reopened the argument for a postsynaptic locus of the expression of LTP and shed light on the greater complexity of the function of central synapses versus those of the muscle end plate. In the hippocampus of young animals, a large number of synapses express NMDA receptors but not AMPA receptors. Thus, when a single stimulus is applied, little to no response is observed. However, when a high frequency tetanus of synaptic stimulation is applied to induce LTP, glutamate activation of NMDA receptors coupled with post-synaptic depolarization results in NMDA receptor activation. The activation of NMDA receptors results in the insertion of readily available AMPA receptors into the postsynaptic membrane, in turn resulting in an increase in EPSP amplitude through a mechanism of postsynaptic LTP expression (Isaac et al., 1995; Liao et al., 1995). The insertion of

AMPA receptors after high frequency synaptic stimulation mimics an increase quantal content.

The identification of silent synapses and AMPA receptor insertion during LTP indicated a postsynaptic locus for LTP expression and explained the previous results that argued for a presynaptic expression.

NMDA receptor mediated LTP: Expression

While the induction of LTP is largely mediated by NMDA receptor activation, the expression of LTP arises from the insertion and phosphorylation of AMPA receptors. The contribution of protein kinases was identified by interrupting protein kinase activity at various key time points during the course of LTP. By interrupting activated protein kinase activity with the drug H7, researchers suppressed LTP that had already been induced, showing that ongoing protein kinase activity was required for the expression of LTP. Furthermore, when H7 was washed out of the bath saline, LTP was restored suggesting an ongoing signaling effect maintaining LTP

(Malinow et al., 1988). These results revealed a mechanism by which synaptic strength could be increased and maintained through the expression phase of LTP. Later research showed more specifically the effectors acting to phosphorylate proteins as well as their targets. It was identified that protein kinase C (PKC) and Ca2+/calmodulin kinase type II (CaMKII) were key to the expression of postsynaptic LTP (Malinow et al., 1989). While the induction of LTP relies on Ca2+ influx through NMDA receptors, activated protein kinases act on AMPA receptors to affect use dependent changes in synaptic strength. Activation of protein kinase A (PKA), PKC, or CaMKII phosphorylates AMPA receptors, increasing the unitary conductance and thus increasing synaptic strength (Lisman et al., 2002; Ren et al., 2013; Park et al., 2021).

Another method by which kinase activity increases synaptic strength involves the activation of phosphatidylinositol 3 kinase (PI3K), a kinase that phosphorylates phosphoinositides.

Activation of PI3K triggers the insertion of AMPA receptors into the postsynaptic membrane thus increasing synaptic strength (Lu et al., 2001; Man et al., 2003).

Morphological changes also contribute to the expression of LTP. During LTP, dendritic spines undergo physical expansion. The post synaptic density, directly adjacent to the presynaptic active zone and the region with the highest density of receptor expression, expands to accommodate the greater AMPA receptor expression characteristic of LTP (Harris et al., 2003).

NMDA receptor mediated LTP: Maintenance

After the expression phase of LTP follows LTP maintenance. Without continuing action, the changes in AMPA receptor function and expression would not last indefinitely. LTP maintenance is provided through cellular mechanisms that renew and maintain the changes that result in LTP. Although the mechanisms for the ongoing maintenance of LTP are not well understood, there are several hypothesized mechanisms for the maintenance of LTP.

The involvement of calcium/calmodulin kinase II (CaMKII) in LTP was originally identified because disruption of its activity, or its absence (in a CaMKII genetic knockout) disrupted LTP (Silva et al., 1992; Pettit et al., 1994; Lledo et al., 1995). CaMKII binds to NMDA receptors in the postsynaptic density and acts to increase AMPA receptor function, AMPA receptor trafficking, and contributes to structural changes involved in LTP (Lisman et al., 2002). Once activated, CaMKII auto-phosphorylates. This attribute provides the potential for perpetual activation and action at a potentiated synapse (Miller and Kennedy, 1986; Lisman et al., 2012;

Kim et al., 2016). These functional attributes of CaMKII make it a clear candidate for the long- term maintenance of LTP, however, definitive evidence as to the role of CaMKII in the maintenance of LTP beyond several hours post high frequency stimulation has yet to be provided.

Another candidate molecule involved in the maintenance of LTP is protein kinase M zeta

(PKMz), an atypical PKC isoform. This protein kinase is persistently active following LTP;

51 however, unlike the CaMKII knockout, the PKMz knockout mouse does not show impaired LTP

(Lee et al., 2013; Volk et al., 2013).

The synaptic tag hypothesis for the ongoing maintenance of LTP puts forth the idea that strong stimulation of a synapse results in a molecular tag that signals for greater allocation of cellular resources involved in LTP maintenance (Frey and Morris, 1998). While this hypothesis has been quite popular as an explanation for the persistence of LTP, very little data has been generated on the potential identity of the synaptic tag. Recent evidence has suggested that L-type calcium channels may signal soma to synapse production and transport of proteins involved in

LTP (Wheeler et al., 2008; 2012; Li et al., 2016; Cohen et al., 2018).

Temporoammonic pathway LTP

The lateral perforant pathway, or temporoammonic (TA) pathway, was first identified by

Ramon y Cajal in his detailed investigation of hippocampal neuron morphology. This observation was further elucidated by Ramon y Cajal’s student, Lorente de No. Despite the observation of direct projections between entorhinal cortex and the distal dendrites of CA1 pyramidal neurons, the 1960’s and 1970’s saw conflicting observations that sparked a controversy concerning the existence of a functionally relevant direct input onto CA1 pyramidal neurons. Early electrophysiological studies of the hippocampal circuit in a slice preparation failed to identify a direct entorhinal cortex to CA1 projection, leading to the question of whether the connection existed at all (Cragg and Hamlyn, 1957).

The Nauta silver degeneration technique provided robust evidence for the presence of the

TA pathway. The technique involved impregnation of a sample with silver before ablating a chosen region. The degeneration of the affected neurons coupled with the silver stain allowed for the

52 tracing of axon projections (Switzer, 2000). When the entorhinal cortex was lesioned under this preparation axon projections to the dendrites of CA1 neurons were clearly identified (Blackstad,

1958). Interestingly, the use of electron microscopy in conjunction with the Nauta silver degeneration technique found few degenerating terminals in area CA1 (Simonsen and Jeune,

1972).

As the use of slice preparations became widespread in the field of electrophysiology the evidence for the existence of the TA pathway mounted (Andersen et al., 1966). However, it was through the measurement of synaptic delay after entorhinal cortex stimulation in the dentate gyrus

(medial perforant path) and CA1 (lateral perforant path) that the presence of the TA pathway became widely accepted. By showing that the delay in postsynaptic response between dentate gyrus and CA1 was the same after entorhinal cortex stimulation, it became clear that there was a direct connection as the response in CA1 was too fast to be caused by the tri-synaptic circuit

(Doller and Weight, 1982). It was later shown that not only was there a direct entorhinal cortex to

CA1 projection, but high frequency stimulation of the pathway results in robust LTP (Doller and

Weight, 1985; Colbert and Levy, 1993; Dvorak-Carbone and Schuman, 1999).

This dissertation focuses on LTP of the direct inputs from entorhinal cortex to the distal dendritic arbor of CA1 pyramidal neurons, the TA pathway. While LTP at this synapse shares some similarity with SC-LTP, TA-LTP has several key differences and as a whole is not as well understood as SC-LTP.

LTP induction at TA inputs requires strong, coordinated activation of entorhinal cortex fibers (Callaway and Ross, 1995; Golding et al., 2002). Unlike SC-LTP, TA-LTP does not rely on bAP signals from the soma and can be induced even if all action potential activity is suppressed at

53 the soma (Golding et al., 2002). The generation of dendritically originating sodium mediated spikes trigger the activation of both NMDA and voltage gated calcium channels, both of which are important for the full expression of TA-LTP (Golding et al., 2002; Remondes and Schuman, 2003;

Kim et al., 2015).

The investigation of the induction of TA-LTP through electrical stimulation has focused exclusively upon the induction phase. What is known about the expression and maintenance phases of TA-LTP is gleaned from second messenger signaling cascades. The activation of estrogen receptor alpha (ERa) results in robust TA-LTP expression through the activation of PI3K and the subsequent insertion of AMPA receptors in the post-synaptic membrane (Clements and Harvey,

2020). Interestingly, TA-LTP induced through electrical stimulation or ERa activation are mutually exclusive, indicating that there is a shared mechanism of expression between the two forms of induction.

Another form of TA-LTP involves the activation of the serotonin receptor 5HT-1b (Cai et al., 2013). This mechanism involves the phosphorylation of AMPA receptors through activation of CAMKII. Interestingly, TA-LTP induction through 5HT-1b activation is occluded by high frequency stimulation induced TA-LTP, indicating a common mechanism through this pathway as well.

There is clearly a gap in the understanding of the mechanisms underlying TA-LTP induced by electrical stimulation and the involvement of second messenger signaling systems. It is likely that, under physiological circumstances, these mechanisms do not occur in isolation. In fact, it is possible that the activation of second messengers coincides with coordinated TA pathway activation and acts to reduce the threshold for TA-LTP induction.

Fragile X Syndrome

Fragile X syndrome, initially called Martin Bell syndrome, was originally identified in a case study concerning cognitive disability expressed within a single family (Martin and Bell,

1943). A mother came to a clinic worried about mental deficiency in her child that she had observed in other members of her family. J. Purdon Martin and Julia Bell identified the disorder as a novel form of cognitive disability. Within the family, the mothers of the affected sons all displayed normal intelligence, leading to the conclusion that what they observed is a sex-linked inheritable disorder affecting cognitive function (Martin and Bell, 1943). The identification of the link between the X-chromosome and Martin Bell syndrome occurred when a researcher was imaging chromosomes of typically developing and affected individuals. It was observed that the long arm of the X-chromosome of those with Martin Bell Syndrome appeared to be partially broken, thus the name Fragile X syndrome was adopted for the disease (Lubs, 1969). While the understanding of Fragile X syndrome has grown significantly, knowledge of the functional impact the disease has on neurons and neuron circuits is still lacking. Furthermore, the underlying reason for memory dysfunction and cognitive disability remains elusive. This dissertation focuses on the relevant physiological understanding of CA1 neurons and circuit function within Fragile X syndrome.

Fragile X syndrome (FXS) is the most common monogenetic cause of intellectual disability and autism, affecting approximately 1 in 4000 males and 1 in 6000 females (Brown et al., 1982;

Kemper et al., 1988; Turner et al., 1996; Rogers et al., 2001). FXS is associated with increased instances of autism, seizures, problems with attention and hyperactivity, increased sensitivity to

55 external stimuli, and aggressive behavior (Fryns, 1984; Fryns et al., 1984; Musumeci et al., 1988;

1999; Rogers et al., 2001; Haessler et al., 2016). The symptomatology of FXS results in a complex presentation of cognitive, psychological, and neurological dysfunction that is difficult to effectively treat and adversely affects the quality of life experienced by people with FXS. Although the genetic basis for FXS is well understood, the neurophysiological basis for the presentation of the disorder in patients remains unclear. This lack of physiological understanding, along with comorbidities, increases the difficulty in designing and providing effective treatments for FXS.

FXS is caused by the expansion of CGG tri-nucleotide repeats of the fragile x mental retardation 1 (fmr1) gene on the long arm of the X chromosome. The expansion of CGG repeats results in hypermethylation and transcriptional silencing of the fmr1 gene. The loss of fmr1 mRNA leads to a complete absence of the fmr1 protein product, Fragile X mental retardation protein

(FMRP) (Lubs, 1969; Bell et al., 1991; Verkerk et al., 1991; Tassone et al., 2000). The presentation of intellectual disability has led researchers to identify specific deficits that could be targeted for therapy or used as a diagnostic tool (Kemper et al., 1988; Reiss and Freund, 1992; Santos et al.,

2014). Although genetic screening for FXS has made behavioral testing a secondary diagnostic tool, researchers have characterized deficits in working memory and cognition associated with the disorder. Studies have identified cognitive deficits that suggest hippocampal dysfunction in FXS

(Greicius et al., 2004; Ornstein et al., 2008). Given the high expression levels of FMRP and well- described cellular architecture, the hippocampus is a crucial structure in the investigation of FXS

(Devys et al., 1993; Steward and Schuman, 2003). Because FMRP is expressed in nearly every brain region (Devys et al., 1993; Steward and Schuman, 2003), the effects of FXS can be seen in a wide variety of cell types in many brain regions. Alterations in cellular signaling, morphology,

56 and physiology have been identified throughout the brain (Rudelli et al., 1985; Hinton et al., 1991;

Comery et al., 1997; Weiler et al., 1997; Brown et al., 2001; Li et al., 2002; Miyashiro et al., 2003;

Zalfa et al., 2003; Galvez and Greenough, 2005; Larson et al., 2005; McKinney et al., 2005; Zhao et al., 2005; D'Hulst et al., 2006; Gantois et al., 2006; Meredith et al., 2007; Bilousova et al., 2009;

Ferron et al., 2014; Kalmbach et al., 2015; Wahlstrom-Helgren and Klyachko, 2015; Khalfallah et al., 2016; Nagaoka et al., 2016; Routh et al., 2017), creating an incredibly complex cellular portrait of the disease that has proven difficult to reconcile with the symptoms of FXS observed in human subjects. FMRP regulates protein expression by either promoting or repressing translation of target mRNAs, many of which are localized in neuronal dendrites (Eberhart et al., 1996; Feng et al.,

1997; Bassell and Warren, 2008). In addition to functioning as a translational regulator, FMRP also regulates protein activity through direct protein-protein interactions (Brown et al., 2010; Deng et al., 2013). Through regulating the expression and function of target proteins, FMRP exerts strong influence over synaptic structure, integration, and plasticity (Zalfa et al., 2003; Dictenberg et al., 2008). This led to the common characterization of FXS as a disorder of synaptic structure and function. Recently, dendritic channelopathies associated with FXS have been identified in the hippocampus and surrounding structures suggesting a broader role for FMRP in neuronal function in not only synaptic transmission but in signal processing and integration.

FXS researchers have identified cellular disease phenotypes such as, morphological abnormalities in dendritic spines, changes in synaptic plasticity and function, and altered signaling pathways associated with synaptic transmission (Rudelli et al., 1985; Hinton et al., 1991; Comery et al., 1997; Huber et al., 2002; Galvez and Greenough, 2005; McKinney et al., 2005; Bilousova et al., 2009; Deng et al., 2013; Routh et al., 2013; Wang et al., 2016). However, the understanding

57 of how changes in cellular properties affect the physiological function of a neuron as a whole is lacking.

Hippocampal Channelopathies in FXS:

Several hippocampal channelopathies have recently been identified that affect circuit interactions between neurons of the hippocampus. A calcium activated potassium (BK) channelopathy modulating the release of neurotransmitter by reducing BK channel activity has been discovered in afferents from CA3-CA1 as well as Layer II/III of entorhinal cortex to CA1

(Deng et al., 2013). An increase in persistent sodium channel current has been described in pyramidal neurons of the entorhinal cortex that project into the hippocampus. This increase in persistent sodium current results in a decrease in action potential threshold and an increase in cellular excitability (Deng and Klyachko, 2016a). Research from our lab has identified two hippocampal channelopathies: an up-regulation of hyperpolarization activated non-selective cation channels (HCN/h-channel) in CA1 pyramidal neuron dendrites and a decreased current and hyperpolarized activation voltage of A-type potassium channels in CA1 neurons of fmr1 knockout

(fmr1 KO) mice (Brager et al., 2012; Routh et al., 2013). While all of these channelopathies influence information processing in the hippocampal circuit, h- and A-type potassium channelopathies are intrinsic to CA1 pyramidal neurons and have direct effects on signal processing and integration of synaptic signals in CA1 neuron dendrites.

Studies in CA1 pyramidal neurons using an HCN1 knockout mouse have shown that, in neurons lacking h-channel expression, TA-LTP is increased (Nolan et al., 2004). Knockout of

HCN1 results in the removal of an inward, positive resting conductance, thus the resting membrane potential is hyperpolarized in HCN1 knockouts compared with WT animals. Hyperpolarizing the

58 resting membrane potential causes an increase in the availability of T-type voltage gated calcium channels for activation, increasing the calcium response to synaptic stimuli (Tsay et al., 2007).

Based on these findings, it is reasonable to postulate that increased dendritic h-channel expression will influence calcium signals and alter TA LTP in fmr1 KO hippocampal pyramidal neurons.

The reduction in dendritic IKA recorded in fmr1 KO CA1 pyramidal neurons leads to an increase of bAP amplitude and the distance travelled into the dendritic arbor. This FXS associated reduction in IKA also results in a decreased threshold for LTP induction in SC synapses in fmr1

KO CA1 pyramidal neurons (Routh et al., 2013). These findings suggest that FXS associated channelopathies have direct effects on normal neuronal function. While the impact of the loss of

A-type K+ channel function on back-propagating action potentials (bAPs) in fmr1 KO has been examined, it remains unclear how the reduction of IKA might impact TA inputs.

The thoroughly investigated SC synapse has made an ideal setting in which to investigate how a disease state interrupts normal synaptic function. As a result, FXS researchers have utilized knowledge of this pathway to study synaptic phenotypes at the SC inputs onto CA1 pyramidal neurons in fmr1 KO animals. The loss of FMRP in FXS results in physiological alterations of synaptic plasticity at SC inputs (Huber et al., 2002; Lauterborn et al., 2007; Hu et al., 2008; Deng et al., 2013; Routh et al., 2013; Toft et al., 2016; Wang et al., 2016). Changes in ion channel expression due to FXS have been shown to affect both synaptic signal integration and action potential propagation (Brager et al., 2012; Deng et al., 2013; Routh et al., 2013). Decreases in A- type potassium channel current in particular have been shown to reduce the threshold for the induction of theta burst pairing SC LTP due to increased amplitude of back-propagating action potentials (Routh et al., 2013). Investigations into synaptic plasticity at Schaffer collateral synapses

59 have been important in understanding LTP as a phenomenon and, more recently, synaptic pathophysiology in FXS. As such, it is important to characterize dendritic changes in membrane potential in response to different patterns of stimulation in order to thoroughly study the disease state.

The comparatively few studies on TA inputs in FXS have produced somewhat conflicting results. Whole cell recordings from CA1 pyramidal neurons in fmr1 KO mice revealed no difference in excitatory synaptic transmission from TA inputs (Wahlstrom-Helgren and Klyachko,

2015). In contrast, extracellular field potential recording in stratum lacunosum-moleculare, the region where TA fibers contact CA1 dendrites, showed a reduction in excitatory synaptic strength in fmr1 KO TA synapses (Booker et al., 2020). While both these studies provide insights into how

TA synapses are altered in FXS, neither investigated LTP.

TA-LTP is dependent on calcium transmission through NMDA receptor activation and voltage gated calcium channels (Golding et al., 2002; Remondes and Schuman, 2003), impairments in one or both of these channels may affect TA-LTP in fmr1 KO animals. The possibility of NMDA receptor dysfunction is supported by the finding that medial entorhinal cortex projections onto granule cells of the dentate gyrus that parallel TA projections to CA1 express decreased NMDA dependent signaling in a model of FXS (Yun and Trommer, 2011; Bostrom et al., 2015). It has also been shown that layer 2/3 neurons of the prefrontal cortex exhibit a decrease in functional L-type calcium signal in the synaptic spines of neuronal dendrites (Meredith et al.,

2007). Although this effect was observed in a different brain region it is possible that FMRP modulates L-type calcium channel function and/or expression. Since TA-LTP is, in part, dependent

60 upon L-type calcium channels, dysfunction in these channels may affect the induction of TA-LTP in fmr1 KO mice.

Conclusion

The work presented in this dissertation seeks to increase our understanding of neuronal integration, and the function of the hippocampal circuit under normal conditions as well as in disease. Understanding how small changes within a single neuron can alter circuit functions can lead to greater overall understanding of how a system functions under baseline conditions. The brain is a plastic, adaptable organ. This dissertation reveals that, despite variability in expression, function, and physiology, the brain displays an incredible ability to adapt and produce an intended function in the face of the dysfunction of individual components.

There is a dual nature to approaching a scientific problem. One must understand what has come before, yet also have the insight to build upon those previous findings to make new discoveries. With feet firmly in the past, we look toward the horizon and the understanding that awaits us there.

Chapter 2: D-type potassium channels normalize action potential firing between dorsal and ventral CA1 neurons of the mouse hippocampus*

The hippocampus has long been associated with learning and memory (Scoville and

Milner, 1957; Jarrard, 1983; Squire, 1992; Savage et al., 2004; Kim and Frank, 2009) and as such the circuitry of the hippocampus has been particularly well characterized (Blackstad, 1956; Hjorth-

Simonsen, 1971; Amaral, 1978; Ishizuka et al., 1990; Chevaleyre and Siegelbaum, 2010; Roy et al., 2017). This thorough understanding of hippocampal organization and function makes the hippocampus an attractive model system for studying the cellular nature of learning and memory.

Within the hippocampus the intrinsic and synaptic properties are not uniform along the dorsal- ventral axis of the hippocampus (Papatheodoropoulos and Kostopoulos, 2000; Sotiriou et al., 2005;

Maggio and Segal, 2007; Dougherty et al., 2012; 2013; Kim and Johnston, 2015; Malik et al.,

2015; Malik and Johnston, 2017). These differences may be critical to understanding how the functionally distinct regions of the hippocampus are pivotal to the mechanisms of learning and memory specific to the dorsal and ventral regions of the hippocampus.

The dorsal and ventral poles of hippocampus are functionally distinct (Swanson and

Cowan, 1977; Dong et al., 2009; Fanselow and Dong, 2010). The dorsal hippocampus shows greater involvement in spatial memory tasks (Moser et al., 1993; 1995), with a greater density of place fields compared to ventral hippocampus (Jung et al., 1994). The ventral hippocampus is

* Data included in this chapter is published in The Journal of Neurophysiology: Ordemann GJ, Apgar CJ, Brager DH (2019) D-type potassium channels normalize action potential firing between dorsal and ventral CA1 neurons of the mouse hippocampus. J Neurophysiol 121:983–995. Contributions: GJO, CJA, and DHB performed and analyzed experiments; GJO and DHB conceived experiments and wrote the manuscript.

62 associated with memory tasks involving an emotional component that relates to greater connectivity with the amygdala (Jay et al., 1989; Henke, 1990). These regional differences are accompanied by differences in neuron morphology, excitability, and ion channel expression along the dorsal-ventral axis of the hippocampus (Dougherty et al., 2012; Marcelin et al., 2012;

Dougherty et al., 2013; Hönigsperger et al., 2015; Kim and Johnston, 2015; Malik et al., 2015;

Milior et al., 2016).

The majority of the physiological investigations comparing dorsal and ventral hippocampus were performed using rats as a model organism. However, many models of neurological disease rely on mice, and limitations in our knowledge of the dorsal-ventral differences in the cellular properties across the mouse hippocampus make comparative interpretation of previous studies difficult to parse. In the rat hippocampus, ventral hippocampal

CA1 neurons have a more depolarized resting membrane potential (Vm) and higher input resistance

(RN) compared with dorsal neurons due to morphological differences and the differential contributions of h-channels and GIRK channels (Dougherty et al., 2012; Kim and Johnston, 2015;

Malik et al., 2015). As a result, ventral CA1 neurons in rat fire more action potentials on average than dorsal neurons. While the comparison of hippocampal properties across the dorsoventral axis in rat has greatly increased our understanding of hippocampal structure and function, it remains unclear whether the differences identified in rat hippocampus translate to the mouse. A study of dorsal and ventral properties of CA1 neurons in mouse hippocampus found that subthreshold intrinsic properties showed differences that align with data from rat hippocampus. However, action potential firing, while different at lower amplitude current injections, were much more similar between dorsal and ventral neurons of mouse compared with rat hippocampus (Milior et al., 2016).

In this study, we use a combination of histology, current clamp, and outside-out patch clamp to systematically investigate the mechanisms resulting in normalized action potential firing between dorsal and ventral CA1 neurons of the mouse hippocampus. In agreement with studies from rat, we found that dorsal CA1 neurons had a more hyperpolarized VM and lower RN compared to ventral CA1 neurons. Also consistent with rat, we found that dorsal CA1 neurons have greater dendritic branching stratum radiatum compared to ventral neurons. However, our findings, contrary to those in rat hippocampus, show no difference in the functional expression of h-channels and GIRK/IRK channels. We found that a more depolarized action potential threshold in ventral neurons was consistent between rat and mouse. Our results show that higher functional expression

+ of D-type (KV1-like) K channels is responsible for this difference in excitability between dorsal and ventral CA1 neurons of mouse hippocampus

Results

Active properties across the dorsal ventral axis

In order to determine the location of our acute slices within the mouse hippocampus, we used a mapping system based on anatomical markers within the hippocampal formation to estimate the position of acute slices within either the dorsal or ventral poles of the hippocampus (Malik et al., 2015). At the end of electrophysiological recording, measurements were made of the dentate gyrus, CA3, and CA1 regions of dorsal and ventral hippocampal slices (Figure 2.1A top; see methods for details). When slices were mapped back into the mouse hippocampal formation, the ventral pole is represented by negative numbers and the dorsal pole more positive numbers, we

64 found that the acute slices used for this study were from either the dorsal or ventral poles of the mouse hippocampus (Figure 2.1A bottom).

Previous electrophysiological data from rat hippocampus shows higher spike firing in ventral compared with dorsal CA1 hippocampal neurons (Dougherty et al., 2012; Malik et al.,

2015). To test whether this was consistent in mouse, we recorded the number of action potentials fired in response to a 500 ms current injection of varying amplitudes (50 to 450 pA at 50 pA intervals) (Figure 2.1B). In contrast to rat, we found that the average number of spikes fired in dorsal and ventral CA1 neurons were not different when cells were held at resting Vm (Figure

2.1C; ventral: n=18, dorsal n=22, animals=33; RM 2-way ANOVA: F(1, 38)=0.37, p=0.54) or at the common potential of -65 mV (Figure 2.1D; ventral: n=21, dorsal n=22, animals=33; RM 2- way ANOVA: F(1, 41)=0.93, p=0.34). Refer to table 1 for detailed information on statistical tests.

In rat, a combination of differences in anatomy as well as subthreshold and suprathreshold properties contribute to the higher action potential output in ventral CA1 neurons. The lack of difference between dorsal and ventral mouse CA1 neurons suggests that the distribution of ion channels and/or anatomical differences between the dorsal and ventral regions of the mouse hippocampus may not align with observations in rat.

Figure 2.1: Mapping of slice location along the dorsal ventral axis of hippocampus.

Figure 2.1: Mapping of slice location along the dorsal ventral axis of hippocampus. Action potential firing is not different between dorsal and ventral neurons of mouse hippocampus. A Slices from ventral (top left) and dorsal (top right) hippocampus. Histogram of slice location along the dorsal-ventral axis of hippocampus (bottom). Ventral is represented by -2 to 0 mm and dorsal by 2 to 4 mm. (n=27 ventral cells and 25 dorsal cells from 37 mice). B Action potential firing in a ventral (top) and dorsal (middle) CA1 neuron during a 150 pA current injection (bottom). C-D The mean number of action potentials fired in dorsal and ventral CA1 neurons at rest (C) or -65 mV (D). Analyzed using 2-way ANOVA (n=18 ventral cells and 22 dorsal cells from 33 mice).

Subthreshold properties of dorsal-ventral mouse CA1 pyramidal neurons

We performed whole cell current clamp recordings from the soma of either dorsal or ventral

CA1 pyramidal neurons. Figure 2.2A shows representative traces of subthreshold voltage responses to current injections. We found that the resting Vm was ~2 mV more depolarized in ventral neurons compared to dorsal neurons (Figure 2.2B; ventral: -62.31 ± 0.74 mV, n=29; dorsal:

-64.05 ± 0.45 mV, n=38, animals: n=43; Mann-Whitney: U=393.5, p=0.045 h2=0.06). Although small, this difference is in agreement with previous findings in rat hippocampus (Dougherty et al.,

2012; Malik et al., 2015) as well as in mouse hippocampus (Milior et al., 2016). RN was measured at both the resting Vm (ventral: 140.6 ± 11.33 MW, n=10; dorsal: 99.0 ± 10.94 MW, n=8; n=10 animals) and at a common Vm (-65 mV; ventral: 129.9 ± 8.4 MW, n=13; dorsal: 92.01 ± 7.41 MW, n=8; n=10 animals) (Figure 2.2C-D). In both cases, and consistent with findings in rat hippocampus, RN was significantly higher in ventral CA1 neurons compared to dorsal CA1

2 neurons (Figure 2.2C-D; RN at rest – unpaired t-test: t(16)=2.6, p=0.02, h =0.3; RN at -65 mV – unpaired t-test: t(19)=3.09, p=0.006, h2=0.33).

In rat, the difference in Vm and RN is due to the difference in both h-channels (Ih) and G- protein coupled inwardly rectifying K+ (GIRK) channels (Dougherty et al., 2012; 2013; Kim and

Johnston, 2015). Subthreshold differences in intrinsic properties contribute to the greater firing of action potentials in ventral compared with dorsal neurons in rat hippocampus. Therefore, we investigated the possibility that differences in the functional expression in h- and GIRK/IRK- channels could underlie the normalized action potential firing between dorsal and ventral CA1 neurons. To estimate difference in Ih between dorsal and ventral CA1 neurons, we measured voltage sag and rebound, two properties that are reflective of h-channel function owing to the slow

68 kinetics of Ih (Magee, 1998; Brager and Johnston, 2007; Brager et al., 2012; Dougherty et al., 2013;

Malik et al., 2015). There was no significant difference in either sag (sag at rest – ventral: 1.08 ±

0.034 mV, n=10; dorsal: 1.02 ± 0.011 mV, n=8; n=10 animals; Mann-Whitney: U=21, p=0.094. sag at -65 mV – ventral: 1.07 ± 0.016 mV, n=10; dorsal: 1.03 ± 0.012, n=8; n=10 animals; Mann-

Whitney: U=26.5, p=0.065) or rebound (rebound at rest – ventral: -0.02 ± 0.015 mV/mV, n=10; dorsal: -0.04 ± 0.021 mV, n=8; n=10 animals; unpaired t-test: t(16)=0.57, p=0.58. rebound at -65 mV – ventral: -0.1 ± 0.019 mV, n=13; dorsal: 0.058 ± 0.014, n=8; n=10 animals; unpaired t-test: t(19)=1.39, p=0.18) between dorsal and ventral CA1 neurons at either the resting Vm or -65 mV

(Figure 2.2E-H). As these results disagreed with previous results in rat, we measured three additional h-channel sensitive properties: resonance frequency (fR), summation, and membrane time constant (τM). In rat, ventral CA1 neurons have a higher fR, indicative of the higher expression of functional h-channels compared with dorsal CA1 neurons (Dougherty et al., 2013). In contrast, we found that there was no significant difference in fR between dorsal and ventral CA1 neurons in mouse (Figure 2.3A-B; ventral: n=19; dorsal: n=18; n=23 animals; RM 2-way ANOVA:

F(1,35)=0.001, p=0.97). Additionally, summation (Figure 2.3C; ventral: n=19; dorsal: n=15; n=23 animals; RM 2-way ANOVA: F(1,32)=0.33, p=0.57) and τM (Figure 2.3D; ventral: n=19; dorsal: n=18; n=23 animals; RM 2-way ANOVA: F(1, 35)=0.66, p=0.42) were also not different. These results suggest that, unlike in rat, differences in Vm and RN cannot be explained by a difference in

Ih between dorsal and ventral CA1 neurons of the mouse hippocampus.

Figure 2.2: Subthreshold intrinsic membrane properties in dorsal and ventral CA1 neurons. A Voltage responses from ventral (top) and dorsal (middle) CA1 neurons in response to a series of current injections (bottom). B Ventral neurons have a more depolarized resting Vm compared to dorsal CA1 neurons. (n=29 ventral cells and 38 dorsal cells from 43 mice). C-D RN of ventral and dorsal CA1 neurons measured from either rest (C: n=10 ventral cells and 8 dorsal cells from 10 mice) or -65 mV (D: n=13 ventral cells and 8 dorsal cells from 10 mice). E-F Voltage sag in ventral and dorsal CA1 neurons measured from either rest (E: n=10 ventral cells and 8 dorsal cells from 10 mice) or -65 mV (F; n=13 ventral cells and 8 dorsal cells from 10 mice). G-H Rebound slope in ventral and dorsal CA1 neurons measured from either rest (G; n=10 ventral cells and 8 dorsal cells from 10 mice) or -65 mV (H; n=13 ventral cells and 8 dorsal cells from 10 mice). A, E, and F analyzed using Mann-Whitney test. C, D, G, and H analyzed using t-test. See table 1 for details.

In rat, there is a greater functional expression of GIRK channels in dorsal compared with ventral rat hippocampus (Kim and Johnston, 2015). We hypothesized that the depolarized Vm and higher RN in mouse ventral CA1 neurons could be due to lower functional expression of GIRK channels. Low concentrations of extracellular Ba2+ can be used to block inwardly rectifying K+ channels (IRK) including GIRK channels (Kim and Johnston, 2015; Malik and Johnston, 2017).

2+ In rat hippocampus, Ba has greater effect on Vm and RN in dorsal compared with ventral neurons

2+ (Kim and Johnston, 2015). Bath application of 25 µM Ba depolarized Vm and increased RN in both dorsal and ventral mouse hippocampal neurons (Figure 2.3E, F). In contrast to rat however,

2+ we found that the effect of Ba on Vm (Figure 2.3G; ventral: 7.38 ± 0.53 mV, n=10; dorsal: 6.3 ±

1.38 mV, n=16; n=19 animals; Mann-Whitney: U=61, p=0.33) and RN (Figure 2.3H; ventral: 54.18

± 11.92 MW, n=10; dorsal: 27.71 ± 10.44 MW, n=16; n=19 animals; unpaired t-test: t(24)=1.63, p=0.12) was not significantly different between dorsal and ventral CA1 pyramidal neurons. These results suggest that, unlike in rat, difference in resting Vm and RN between dorsal and ventral mouse

CA1 neurons does not result from differences in h- or GIRK/IRK channel expression.

Figure 2.3: h-channel and inwardly rectifying K+ channel activity does not differ between dorsal and ventral CA1 neurons.

Figure 2.3: h-channel and inwardly rectifying K+ channel activity does not differ between dorsal and ventral CA1 neurons. A Impedance amplitude profiles that show the resonant frequency, fR, of representative ventral (3.4 Hz) and dorsal (3.2 Hz) CA1 neurons. Inset: voltage traces showing response to a chirp current injection. B fR measured while varying Vm in dorsal and ventral CA1 neurons. C summation measured while varying Vm indorsal and ventral CA1 neurons. Inset: representative voltage responses to injection of 5 alpha waveforms at 20Hz. D τM measured while varying Vm in dorsal and ventral CA1 neurons. Inset: representative voltage responses to -100 pA current injection used to determine τM. B-D n=19 ventral cells and 18 dorsal cells from 23 mice. 2+ E Time dependent effect of extracelluar 25 µM Ba on resting Vm and RN dorsal and ventral CA1 neurons. Black bar represents the application of Ba2+. F Voltage response to -50 pA current 2+ injection before and after Ba wash-on. Note the depolarization of Vm and increase in RN. G-H 2+ The effect of extracellular Ba application on resting Vm (G) and RN (H) in dorsal and ventral CA1 neurons (n=10 ventral cells and 16 dorsal cells from 19 mice). Data presented as mean ± s.e.m. B, C, and D analyzed using 2-way ANOVA. G was analyzed using a two-tailed Mann- Whitney test and H was analyzed using a two-way unpaired t-test. See table 1 for details.

Neuronal Reconstructions

In the rat hippocampus, dorsal CA1 neurons have a shorter somato-dendritic axis, greater number of branches, and greater surface area compared with ventral CA1 neurons (Dougherty et al., 2012; Malik et al., 2015). We used Neurolucida software to reconstruct dorsal and ventral mouse CA1 neurons and asked if these same differences were present (Figure 2.4A). Sholl analysis revealed that dendritic branching was significantly greater in dorsal compared with ventral neurons of the mouse hippocampus, specifically in the stratum radiatum region of CA1 (Figure 2.4B; ventral: n=7; dorsal: n=7; n=10 animals; RM 2-way ANOVA: F(1,432)=11.9, p=0.0006, h2=0.01).

In contrast with rat data however, somato-dendritic length (Figure 2.4C; ventral: 484.7 ± 13.8 µm, n=7; dorsal: 462.1 ± 34.13 µm; n-10 animals; unpaired t-test: t(11)=0.65, p=0.53) and surface area

(Figure 2.4D; ventral: 13.52 ± 1.12 mm2, n=7; dorsal: 17.2 ± 1.36 mm2, n=7; n=10 animals; unpaired t-test: t(12)=2.1, p=0.06) were not different between dorsal and ventral CA1 neurons.

Figure 2.4: Greater dendritic branching in dorsal compared with ventral CA1 pyramidal neurons. A Representative cellular reconstructions of ventral (left) and dorsal (right) CA1 neurons of the hippocampus, scale bar represents 50 µm. B Dendritic branching as measured by Sholl analysis (see methods) in dorsal and ventral CA1 neurons. C Somato-dendritic distance in dorsal and ventral CA1 neurons. D Total dendritic surface area in dorsal and ventral CA1 neurons. B-D: n=7 ventral cells and 7 dorsal cells from 10 mice. B analyzed using two-way ANOVA. C and D analyzed using t-test. See table 1 for details.

Action potential threshold

One possible explanation for the normalized action potential firing between dorsal and ventral neurons in mouse hippocampus is a difference in action potential threshold. It was previously reported that ventral CA1 neurons in rat have a significantly more depolarized action potential threshold compared with dorsal CA1 neurons (Dougherty et al., 2012; Malik et al.,

2015).However, the underlying cause of the difference in threshold in rat hippocampus remains uninvestigated. In order to compare to studies in rat hippocampus, we measured the threshold of the first action potential in a train where the first action potential occurred with a latency of ~50 ms (Dougherty et al., 2012) (see Methods for details). Consistent with previous reports in rat, we found that action potential threshold in ventral neurons was significantly more depolarized than in dorsal CA1 neurons in mouse hippocampus (Figure 2.5A; ventral: -45.11 ± 1.49 mV, n=9; dorsal:

-49.41 ± 0.74 mV, n=15; n=23 animals; unpaired t-test: t(22)=2.8, p=0.009, h2=0.27).

To further investigate the difference in action potential threshold between dorsal and ventral CA1 neurons, we utilized a more systematic approach for measuring action potential threshold (Higgs and Spain, 2011; Kalmbach et al., 2015). Current injections of varying durations

(1.5 - 100 ms) were delivered to dorsal and ventral CA1 neurons from a common Vm of -65 mV.

The current amplitude was adjusted to elicit a single action potential at the end of the current step

(Figure 2.5B). Action potential threshold was significantly more depolarized in ventral neurons at all measured current durations (Figure 2.5C; ventral: n=16; dorsal: n=15; n=19 animals; RM 2- way ANOVA: F(1, 29)=16.44, p=0.0003, h2=0.33). Action potential amplitude (ventral: n=14; dorsal: n=13; n=16 animals; RM 2-way ANOVA: F(1, 25)=0.063, p=0.81), duration (ventral: n=14; dorsal: n=13; n=16 animals; RM 2-way ANOVA: F(1, 25)=0.52, p=0.48), and maximum

76 rate of rise (ventral: n=14; dorsal: n=13; n=16 animals; RM 2-way ANOVA: F(1, 25)=0.21, p=0.65) were not significantly different between dorsal and ventral CA1 pyramidal neurons

(Figure 2.5D-F).

Figure 2.5: Action potential threshold is more depolarized in ventral compared with dorsal CA1 neurons. A Action potential threshold in dorsal and ventral CA1 neurons (n=9 ventral cells and 15 dorsal cells from 23 mice). B Voltage responses showing action potentials elicited near the end of a 24 ms current injection. Inset: Expanded area indicated by the black box showing action potential threshold (black dot). C Action potential threshold in dorsal and ventral CA1 neurons. (*: p <0.05; n=16 ventral cells and 13 dorsal cells from 18 mice). D-F Action potential amplitude (D), duration (E), and maximum dV/dt (F) in dorsal and ventral CA1 neurons (n=16 ventral cells and 15 dorsal cells from 16 mice). Analyzed using 2-way ANOVA with Bonferroni’s multiple comparison test. See table 1 for details.

K+ channel voltage clamp measurements

Slowly-inactivating D-type K+ channels affect action potential threshold during longer current steps while having few effects on other physiological properties (Bekkers and Delaney,

2001; Guan et al., 2007; Yu et al., 2008; Higgs and Spain, 2011; Sosanya et al., 2015; Kalmbach et al., 2015). The change in threshold, but not in other action potential properties, led us to hypothesize that differences in the functional expression of D-type K+ channels contributed to the difference in threshold between dorsal and ventral mouse CA1 neurons. In order to test for differences in voltage-gated K+ channel function we used outside-out voltage clamp to measure

K+ current from somatic patches from dorsal and ventral CA1 neurons. We used a series of voltage commands to separate IK-TOTAL into three components: a rapidly inactivating IK-FAST, a slowly inactivating IK-SLOW, and non-inactivating IK-SUS (Figure 2.6A) (Kalmbach et al., 2015; Routh et al., 2017). There was no significant difference in IK-TOTAL between dorsal and ventral CA1 neurons

(Figure 2.6B; ventral: 216.8 ± 45.76 pS/µm2, n=7; dorsal: 281.6 ± 59.27 pS/µm2, n=6; n=9 animals; unpaired t-test: t(11)=0.88, p=0.4). We found that IK-SLOW conductance density was greater in patches from ventral CA1 neurons compared to dorsal CA1 neurons (Figure 2.6C-D; ventral: 155.6 ± 21.0 pS/µm2, n=7; dorsal: 94.17 ± 16.61 pS/µm2, n=6; n=9 animals; unpaired t- test: t(11)=2.24, p=0.047, h2=0.31). There was no significant difference in either the rapidly inactivating (ventral: 110.5 ± 19.3 pS/µm2, n=7; dorsal: 181.3 ± 26.45 pS/µm2, n=6; n=9 animals; unpaired t-test: t(10)=2.16, p=0.056) or sustained current (ventral: 15.26 ± 5.08 pS/µm2, n=5; dorsal: 16.88 ± 4.09 pS/µm2, n=5; n=9 animals; unpaired t-test: t(8)=0.25, p=0.8) (Figure 2.6E-F).

Accordingly, the slowly inactivating current makes up a greater percentage of IK in ventral than in dorsal neurons (Figure 2.6G; ventral: 50.94 ± 3.9%, n=7; dorsal: 36.71 ± 3.71%, n=6; n=9 animals;

79 unpaired t-test: t(11)=2.61, p=0.035, h2=0.38). There was no significant difference in the activation of either IK-FAST or IK-SLOW between dorsal and ventral CA1 neurons (Figure 2.6H-I).

These findings support the hypothesis that the more depolarized action potential threshold in ventral CA1 neurons is due to higher functional expression of slowly inactivating D-type K+ current.

Figure 2.6: D-type K+ channel conductance density is higher in ventral compared with dorsal CA1 neurons.

Figure 2.6: D-type K+ channel conductance density is higher in ventral compared with dorsal CA1 + neurons. A IK recorded from an outside-out patch (left) was separated in three K current components based on voltage-dependence (IK-FAST, top-right; IK-SLOW, bottom-right; and IK-SUS, bottom-left). Total IK was measured by using 800 ms voltage steps ranging from -70 to 50 mV at 20 mV intervals from a holding potential of -90 mV (top-left). A 100-ms pre-step to -20 mV was used to remove the rapidly inactivating component (middle-left). Sustained IK was isolated by using 800 ms voltage steps ranging from -70 to 50 mV at 20 mV intervals from a holding potential of -20 mV (bottom-left). Offline subtraction of isolated IK components provided individual fast, slow and sustained K+ currents (Kalmbach et al., 2015; Routh et al., 2017). B The K+ conductance density dorsal and ventral CA1 neurons (n=7 ventral cells and 6 dorsal cells from 9 mice). C Traces showing the maximum IK-SLOW from ventral (black) and dorsal (gray) CA1 pyramidal neurons. D the conductance density for IK-SLOW in dorsal and ventral CA1 neurons (n=7 ventral cells and 6 dorsal cells from 9 mice). E Conductance density for IK-FAST in dorsal and ventral CA1 neurons (n=7 ventral cells and 6 dorsal cells from 9 mice). F Conductance density for IK-SUS in dorsal and ventral CA1 neurons (n=5 ventral cells and 5 dorsal cells from 9 mice). G The percentage of IK made up by IK-SLOW in dorsal and ventral CA1 neurons (n=7 ventral cells and 6 dorsal cells from 9 mice). H-I Activation curves for IK-FAST (H) or IK-SLOW (I) in dorsal and ventral CA1 neurons (n=7 ventral cells and 6 dorsal cells from 9 mice). Analyzed using t-test. See table 1 for details.

4-AP Experiments

D-type K+ channels are sensitive to low concentrations of the K+ channel blocker 4- aminopyridine (4-AP). Low concentrations of 4-AP can be used to test for the role of D-type K+ channels on action potential threshold (Castle et al., 1994; Russell et al., 1994; Higgs and Spain,

2011; Sosanya et al., 2015; Kalmbach et al., 2015). It was previously shown that the slowly inactivating component of IK (IK-SLOW) is blocked by low concentrations of 4-AP or α-dendrotoxin

+ (Kalmbach et al., 2015), suggesting that IK-SLOW is mediated by KV1-containing D-type K channels. Based on the greater IK-SLOW in ventral neurons, we hypothesized that action potential threshold in ventral CA1 neurons will be more sensitive to low 4-AP compared to dorsal CA1 neurons. We measured action potential threshold before and after bath application of 50 µM 4-AP to both dorsal and ventral CA1 neurons. 4-AP hyperpolarized action potential threshold in dorsal neurons at varying current step durations by between 0 and -2.3 mV (dorsal: n=7 from 9 animals;

RM 2-way ANOVA: F(1,42)=16.89, p=0.0002, h2=0.02) and in ventral neurons by between -1 and -4.5 mV (ventral: n=9 from 9 animals; RM 2-way ANOVA: F(1,63)=31.26, p=0.0001, h2=0.16. The effect of 4-AP was greater in ventral compared with dorsal CA1 neurons of mouse hippocampus with a mean change in threshold due to 4-AP of 3.11 mV in ventral and 0.94 mV in dorsal hippocampus (Figure 2.7A-D). Application of 4-AP significantly increased the number of action potentials fired in both dorsal and ventral CA1 neurons, however, in ventral neurons the difference in firing was significant at all current amplitudes with an average increase in firing of

3.05 Hz, while in dorsal neurons only the 100 pA current step differed significantly due to the application of 4-AP with an average increase in firing of 2.02 Hz (Figure 2.7F; ventral: n=9 from

9 animals; RM 2-way ANOVA: F(1,64)=124.7, p=0.0001, h2=0.02). These data support the

83 hypothesis that higher functional expression of D-type K+ channels in ventral CA1 pyramidal neurons contributes to the more depolarized threshold relative to dorsal CA1 pyramidal neurons, and thus a normalization of action potential firing between dorsal and ventral CA1 neurons.

Figure 2.7: Application of 4-AP abolishes the difference in action potential threshold between ventral and dorsal CA1 neurons. 85

Figure 2.7: Application of 4-AP abolishes the difference in action potential threshold between ventral and dorsal CA1 neurons. A Voltage responses showing action potentials elicited from a ventral CA1 neuron before and after application of 50 µM 4-AP. Inset: Expanded area indicated by the black box showing action potential threshold (black dot). B Threshold measurements during application of 4-AP in ventral CA1 neurons (*: p<0.05) (n=10 ventral cells from 9 mice). C Voltage responses showing action potentials elicited from a dorsal CA1 neuron before and after application of 50µM 4-AP. Inset: Expanded area indicated by the black box showing action potential threshold (black dot). D Action potential threshold in dorsal CA1 neurons during application of 4-AP (n=7 dorsal cells from 9 mice). E Action potential threshold during 4-AP application in dorsal and ventral CA1 neurons (n=10 ventral cells and 7 dorsal cells from 9 mice). F The mean number of action potentials fired during 1 second current injections of varying amplitudes differed at all current amplitudes in the presence of 4-AP in ventral CA1 neurons. Ventral: n=9; dorsal: n=8; n=9 animals. Analyzed using 2-way ANOVA with Sidak’s multiple comparison test. See table 1 for details.

Discussion

We investigated the mechanism behind the normalized action potential firing between dorsal and ventral CA1 pyramidal neurons of mouse hippocampus. We found that Vm was more depolarized and RN higher in ventral CA1 neurons compared to dorsal CA1 neurons consistent with rat (Dougherty et al., 2012; Kim and Johnston, 2015; Malik et al., 2015). However, unlike rat, these differences do not depend on differential expression of h- and/or GIRK/IRK channels.

Furthermore, although RN was higher in ventral CA1 neurons there was no significant difference in action potential firing between dorsal and ventral CA1 neurons. We found that ventral CA1 neurons have a higher expression of D-type K+ channels, which contributes to a more depolarized action potential threshold in ventral compared with dorsal CA1 neurons. These findings suggest that differences in subthreshold properties between dorsal and ventral CA1 neurons in mouse are largely ameliorated by differential expression of D-type K+ channels.

Dorsal-ventral subthreshold differences are smaller in mouse compared to rat

Research examining dorsal-ventral differences in subthreshold properties has largely been performed in the rat hippocampus. Ventral CA1 neurons consistently have a more depolarized resting Vm (by 4-8 mV) and higher RN (~200%) compared with dorsal CA1 neurons (Dougherty et al., 2012; Kim and Johnston, 2015; Malik et al., 2015). We found that CA1 neurons in the mouse hippocampus express similar differences but at a lesser magnitude, with ventral neurons resting 2 mV depolarized and with a 60% greater RN compared with dorsal CA1 neurons (Figure 2.2). In rat, the difference in subthreshold properties between dorsal and ventral CA1 neurons is due to a combination of: h-channel activity, greater in ventral CA1 neurons; GIRK channel activity, greater

87 in dorsal neurons; and dendritic morphology, more branching in dorsal CA1 neurons (Dougherty et al., 2012; 2013; Kim and Johnston, 2015; Malik et al., 2015; Malik and Johnston, 2017).

We used a variety of h-channel sensitive electrophysiological measurements to test if differences in h-channel function contribute to the differences in Vm and RN in mouse CA1 neurons. Unlike in rat, we found no significant differences in sag, rebound slope, fR, or membrane time constant. Our results suggest that the functional expression of h-channels does not contribute to differences in somatic function between dorsal and ventral CA1 neurons in the mouse hippocampus. A comparison of middle hippocampus between two mouse strains (C57BL6 and

129/SvEv) and one rat strain (Sprague-Dawley) found that there is a lower functional expression of h-channels in mouse compared with rat hippocampus (Routh et al., 2009). These findings raise the possibility that, despite using multiple measures of h-channel function, decreased h-channel functional expression in mouse hippocampus compared with rat hippocampus made it difficult to resolve differences in h-channel function in the soma of dorsal and ventral CA1 neurons in this study. A recent comparison of place cell function between rats and mice found that mice have similar theta oscillations but smaller, less spatially specific place fields (Mou et al., 2018), suggesting relatively conserved function between mouse and rat hippocampal CA1 neurons. These studies suggest that, despite subtle differences in physiology and function between rats and mice, the functional output of the hippocampus remains largely the same.

We used a low concentration of extracellular Ba2+ to test if higher functional expression of

GIRK/IRK channels in dorsal CA1 neurons contributed to the difference in Vm and RN between dorsal and ventral CA1 neurons in the mouse hippocampus (Kim and Johnston, 2015). Ba2+ depolarized and increased the RN of both dorsal and ventral CA1 neurons. However, in contrast to

2+ rat, we found that the effect of Ba on both Vm and RN was not significantly different between dorsal and ventral CA1 neurons in mouse. Although both dorsal and ventral CA1 neurons express

GIRK/IRK channels, as described in rat, there does not appear to be a differential contribution of

GIRK/IRK channels to the subthreshold properties of dorsal and ventral CA1 neurons in mouse.

A non-significant trend was observed in the change in RN indicating slightly greater increase in

2+ ventral hippocampal RN compared with dorsal. However, this Ba induced change is contrary to findings in rat hippocampus (Kim et al., 2015) and fails to explain the lower resting RN in dorsal compared with ventral neurons of the mouse hippocampus.

Prior to this study, a morphological comparison of dorsal and ventral CA1 neurons in the mouse hippocampus was lacking. In agreement with data from rat hippocampus, our results show that dorsal CA1 neurons show greater branching in the stratum radiatum region of the hippocampus compared with ventral CA1 neurons. It is likely that the increased branching underlies the lower RN in dorsal CA1 neurons compared with ventral CA1 neurons of mouse hippocampus. While the difference in dendritic morphology suggests a role for dendritic branching in the difference in RN between dorsal and ventral neurons, our findings do not explain the observed depolarized ventral resting Vm compared with dorsal CA1 neurons. The recordings in this chapter were performed from the soma of CA1 pyramidal neurons while the phenotypes of differential h-channel and GIRK/IRK expression across the dorsoventral axis was explored largely in the dendrites of rat CA1 pyramidal neurons. It is possible that measurements of these channels from the soma do not reflect the reality of dendritic expression.

Somatic excitability is normalized due to dorsal-ventral differences in D-type K+ channel function

Dorsal-ventral comparisons of CA1 rat neurons show that ventral hippocampus is significantly more excitable than dorsal hippocampus. The higher excitability of ventral CA1 neurons arises from the depolarized resting Vm and higher RN (Dougherty et al., 2012; 2013; Kim and Johnston, 2015; Malik et al., 2015). In rat, the consequence of these subthreshold differences is that ventral CA1 neurons fire more action potentials compared to dorsal CA1 neurons. Our findings mostly agree with a previous mouse study which found no significant difference in action potential firing except with small current injections (Milior et al., 2016). This difference at lower amplitude current steps but not with increasing amplitude current injections differs from data in the rat hippocampus in which the divergence in action potential firing increases with increasing current step amplitudes (Dougherty et al., 2012; Malik et al., 2015). Previous studies in rat found that the action potential threshold for ventral CA1 neurons is more depolarized compared to dorsal

CA1 neurons (Dougherty et al., 2012; Malik et al., 2015). A more depolarized threshold should make ventral CA1 neurons less excitable than dorsal CA1 neurons; however, the significantly higher RN and more depolarized Vm of ventral neurons in rat that overcomes this difference in action potential threshold. In contrast, while our findings, concurrent with others, found a depolarized resting Vm and higher RN in ventral CA1 neurons (Milior et al., 2016), the action potential output is not significantly different between dorsal and ventral CA1 neurons. These results suggest a fundamental difference in input-output between CA1 neurons of mouse and rat hippocampus.

Action potential threshold is dictated by the complement of voltage-gated ion channels expressed by a neuron. It is commonly thought that threshold is dictated by the interplay between

Na+ channels and K+ channels (Hodgkin and Huxley, 1952a; 1952b; Stafstrom et al., 1984). We found no difference in maximum dV/dt, often an indicator of Na+ channel function, suggesting that a difference in Na+ channel functional expression did not contribute to the difference in threshold. While we cannot directly rule out a difference in the voltage dependence of Na+ channel activation contributing to the difference in action potential threshold, based on the normalization of threshold between dorsal and ventral neurons by the application of 4-AP strongly suggests that the difference in threshold is primarily mediated by D-type K+ channels. Pharmacological evidence shows that action potential threshold is influenced by dendrotoxin and low 4-AP sensitive (D-type)

+ K channels from the KV1 family (Storm, 1988; 1993; Bekkers and Delaney, 2001; Guan et al.,

2007; Yu et al., 2008; Higgs and Spain, 2011). Our measurements of rapidly-inactivating K+ channels show a small, yet not significant, difference in current with a trend toward dorsal hippocampus having greater IK-FAST compared with ventral. This finding is in support of previous research that shows greater Kv4.2, a rapidly inactivating K+ channel in dorsal compared with ventral CA1 neuron dendrites (Marcelin et al., 2012). Because our measurements were taken from neuron somata instead of dendrites, it is possible that the functional expression of rapidly- inactivating K+ channels is greater in the dendrites of dorsal compared with ventral neurons of mouse hippocampus. However, this difference cannot explain the difference in threshold between dorsal and ventral neurons of mouse hippocampus. Cells analyzed in this study had intact apical and basal dendritic arbors based upon Neurobiotin fills. However, the axon was not visible in the majority of cells, preventing the determination of whether the axon was cut during the preparation of acute slices. While we cannot definitively say that the axon was severed near the soma, which would influence action potential threshold, the majority of our reconstructions had intact basal

91 dendritic arbors. It is thus likely that any severing of the axon that may occur was beyond the extent of the basal dendrites. Together with the observation that the application of 4-AP normalizes action potential threshold it is unlikely that artifacts due to the preparation of dorsal versus ventral hippocampal slices influenced our conclusions.

In agreement with studies in rat hippocampus, we found that ventral CA1 neurons have a depolarized action potential threshold in the mouse hippocampus. While the underlying cause of the dorsal-ventral difference in threshold in rat hippocampus has yet to be identified, we

+ demonstrate here that the functional expression of D-type K channels of the KV1 family was higher in ventral CA1 neurons resulting in the dorsoventral difference in action potential threshold in mouse hippocampus. Our findings indicate that the primary current underlying the dorsoventral difference in action potential threshold by showing that when D-type K+ channels were blocked using 4-AP, action potential threshold was no longer significantly different between ventral and dorsal CA1 neurons (Figure 2.7).

Functional consequences

The ventral hippocampus is more susceptible to insults that result in the development of epilepsy and temporal lobe seizures are more likely to originate within the ventral pole of the hippocampus (Racine et al., 1977; Lothman and Collins, 1981; Papatheodoropoulos et al., 2005;

Ekstrand et al., 2011; Toyoda et al., 2013). It was previously shown that a downregulation of KV1.1 and concomitant hyperpolarization of action potential threshold occurs in a rodent model of TLE

(Sosanya et al., 2015). This change was also accompanied by a reduction in the sensitivity of action potential threshold to low concentrations of 4-AP. Our results suggest that the vulnerability of the ventral hippocampus to epileptogenesis may be due in part to the higher functional expression of

+ 4-AP sensitive (putative KV1, D-type) K channels in ventral CA1 neurons. Given that ventral

CA1 neurons have a more depolarized resting Vm and higher RN, any loss in functional expression of D-type K+ channels would result in an increase in action potential firing and may contribute to temporal lobe seizure activity.

Our results identify the mechanism underlying the more depolarized action potential threshold in ventral as compared to dorsal neurons and the resulting normalization of action potential firing between dorsal and ventral CA1 hippocampal neurons. Application of low concentrations of 4-AP hyperpolarized the action potential threshold in ventral neurons to values comparable to that of dorsal neurons. Furthermore, we identified greater functional expression of a slowly-inactivating potassium current in ventral compared with dorsal CA1 neurons. Taken together, these results support the conclusion that there is an increase in a KV1-type potassium channel in ventral neurons that contributes to a more depolarized action potential threshold.

Interestingly, the more depolarized threshold, in combination with higher RN and depolarized resting Vm in ventral CA1 neurons, is a firing rate that is equivalent between dorsal and ventral

CA1 neurons of the mouse hippocampus. The incongruent findings between our study in mice and research performed in the CA1 region of rat hippocampus across the dorsal-ventral axis highlights the importance of characterizing and understanding neural variability. Despite reported differences in signal processing and output, mouse and rat hippocampi are successful in their intended function. Understanding how the same neural computation can occur through distinct mechanistic pathways is pivotal to understanding how basic physiological functions result in complex animal behavior.

Chapter 3: Intrinsic properties of dorsal and ventral fmr1 KO CA1 pyramidal neurons

The hippocampus has long been a particular structure of interest in Fragile X syndrome

(FXS) research. There are several points in favor of the study of FXS in the hippocampus. The hippocampus exhibits high levels of FMRP expression, the protein product of the fmr1 gene that is lacking in FXS (Ludwig et al., 2014). FXS is a disease of learning and memory, and the hippocampus has long been associated with learning and memory function (Scoville and Milner,

1957). And finally, the hippocampus is among the most well studied regions of the brain, making it ideal for investigations of how a disease may interrupt normal function (Amaral and Witter,

1989; Squire, 1992). While it is true that the hippocampus has been extensively studied, there is still much that is not known about its structure and function. Understanding of the normal function of neurons is important in understanding the underlying mechanisms behind disease phenotypes.

In this chapter we investigate effects of fmr1 KO on the intrinsic properties of dorsal and ventral

CA1 pyramidal neurons.

We identified a novel mechanism for the normalization of action potential firing across the dorsal ventral extent of mouse hippocampus (Ordemann et al., 2019). In this chapter, we measure similar intrinsic properties as presented in Chapter 2 to compare the properties of fmr1 KO dorsal and ventral CA1 pyramidal neurons. Our findings reveal similar differences across the fmr1 KO dorsal ventral axis of hippocampus as were observed in wild type animals. When fmr1 KO and wild type dorsal and ventral CA1 pyramidal neuron properties were compared, we found no

94 differences in any measurements recorded. This suggests that the D-type potassium channel mediated action potential firing normalization is present in fmr1 KO hippocampus.

Results

Mapping slices across the dorsal-ventral axis of hippocampus

We made somatic, whole cell current clamp recordings from dorsal and ventral CA1 pyramidal neurons of fmr1 KO male mice between 2 and 4 months old. Using an algorithm to for the measurement of slice location along the dorsal-ventral axis of the mouse hippocampus (Malik et al., 2015), we calculated the location of each slice used for electrophysiological recording. Slices clearly fell within measurement parameters of either ventral or dorsal (Figure 3.1A). Ventral was defined as falling between -2 and 0 mm and dorsal as between 2 and 4 mm as measured by the length of mouse hippocampus (Figure 3.1B).

Figure 3.1: Mapping slice location along the dorsal ventral axis of hippocampus. A Slices from ventral (left) and dorsal (right) hippocampus. B Histogram of slice location along the dorsal- ventral axis of hippocampus (bottom). Ventral is represented by -2 to 0 mm and dorsal by 2 to 4 mm. (n=10 ventral cells and 7 dorsal cells from 13 mice).

Action potential firing in dorsal and ventral fmr1 KO CA1 pyramidal neurons

In chapter 2, we showed that, unlike rat hippocampus (Dougherty et al., 2012), action potential firing in response to square pulses of varying current amplitude was not different between wild type dorsal and ventral CA1 pyramidal neurons. We performed the same experiment in fmr1

KO CA1 pyramidal neurons. During recording from either dorsal or ventral CA1 neurons, 500 ms current injections of amplitudes varying from 50 to 400 pA in 50 pA intervals were delivered

(Figure 3.2A). As in our wild type data we observed no difference in the rate of action potential firing between dorsal and ventral fmr1 KO CA1 pyramidal neurons at either resting membrane potential (ventral: n=7; dorsal: n=6; n=13 animals; RM 2-way ANOVA: F(1, 11)=0.052, p=0.82) or when held at the common membrane potential of -65 mV (Figure 3.2B-C; ventral: n=7; dorsal: n=6; n=13 animals; RM 2-way ANOVA: F(1, 12)=0.96, p=0.35).

Figure 3.2: Action potential firing is not different between dorsal and ventral fmr1 KO CA1 pyramidal neurons. A Action potential firing in a ventral (top) and dorsal (middle) CA1 neuron during a 150 pA current injection (bottom). C-D The mean number of action potentials fired in dorsal and ventral CA1 neurons at rest (C) or -65 mV (D). Analyzed using 2-way ANOVA (n=7 ventral cells and 6 dorsal cells from 13 mice).

Subthreshold intrinsic properties of dorsal and ventral fmr1 KO CA1 pyramidal neurons

In the hippocampus of both rat and mouse dorsal CA1 pyramidal neurons have a lower input resistance and more hyperpolarized resting membrane potential, although the differences are more pronounced in rat compared with mouse data (Dougherty et al., 2012; Kim and Johnston,

2015; Malik et al., 2015; Milior et al., 2016; Ordemann et al., 2019). We compared input resistance in dorsal and ventral fmr1 KO CA1 pyramidal neurons at varying membrane potential to compare input resistance (Figure 3.3A). Interestingly, we found no difference in input resistance between dorsal and ventral fmr1 KO CA1 pyramidal neurons (Figure 3.3B; ventral: n=12; dorsal: n=12; n=13 animals; RM 2-way ANOVA: F(1, 22)=3.18, p=0.089), however, when compared to wild type dorsal and ventral CA1 neurons input resistances measure using the same methods we found no difference between the genotypes (Table 3.1). The resting membrane potential of fmr1 KO CA1 pyramidal neurons was found to be different between dorsal and ventral cells, matching findings in wild type mouse and rat data (Figure 3.3C; ventral: -63.25 ± 0.57 mV, n=12; dorsal: -65.62 ±

0.58 mV, n=26; n=30 animals; unpaired t-test: t(36)=2.52, p=0.016, h2=0.15).

Dorsal and ventral CA1 pyramidal neurons of the rat hippocampus show differing expression of h-channels, with greater expression in dorsal compared with ventral cells (Dougherty et al., 2013; Arnold et al., 2019). Our findings shown in Chapter 2 showed no physiological signs of an expression gradient between dorsal and ventral CA1 pyramidal neurons in wild type hippocampus. However, it is established that the expression of h-channels is impacted by the absence of FMRP which interacts with h-channels through protein-protein interaction (Brager et al., 2012; Brandalise et al., 2020). We compared physiological markers of h-channel expression in dorsal and ventral fmr1 KO CA1 pyramidal neurons. h-channels contribute to the voltage sag and

99 rebound slope observed in the voltage response to subthreshold square current injections (Magee,

1998). We observed no difference in either voltage sag (ventral: n=12; dorsal: n=12; n=13 animals;

RM 2-way ANOVA: F(1, 22)=1.42, p=0.25) or rebound slope (ventral: n=12; dorsal: n=12; n=13 animals; RM 2-way ANOVA: F(1, 22)=1.73, p=0.2) between dorsal and ventral fmr1 KO CA1 pyramidal neurons (Figure 3.3D-E). Resonant frequency is the input frequency of current injection that elicits greatest voltage response from neuronal membrane. Resonant frequency increases with increased expression of h-channels, which, in conjunction with the properties of the membrane, act as a band pass filter (Ulrich, 2002; Narayanan and Johnston, 2007). Using a current stimulus that varied from 1 to 20 Hz over 20 seconds we showed no difference in the resonant frequency in dorsal and ventral fmr1 KO CA1 pyramidal neurons (Figure 3.3F; ventral: n=12; dorsal: n=12; n=13 animals; RM 2-way ANOVA: F(1, 22)=2.59, p=0.12). A population of expressed h-channels are open at resting membrane potential in CA1 pyramidal neurons (Magee, 1998). Thus, greater h-channel expression results in reduced membrane resistance. Based on the equation:

� = ��

In which � is the membrane time constant, R is the membrane resistance, and C is the capacitance of the membrane. If the membrane resistance is reduced through the expression of a channel open at rest, then the membrane time constant will be shorter. We measured membrane time constant between dorsal and ventral fmr1 KO CA1 pyramidal neurons and found no difference in time constant at multiple membrane potentials (Figure 3.3G; ventral: n=12; dorsal: n=12; n=13 animals;

RM 2-way ANOVA: F(1, 21)=0.7, p=0.41). Similarly, changes in membrane resistance affect the amplitude of membrane voltage responses in accordance with Ohm’s law:

� = ��

100

In which voltage (V) is shown to change with current (I) and resistance (R). A hypothesized increase in h-channel expression would then result in a reduction in membrane resistance and reduced amplitude of the voltage response to current input. We tested input summation, a measure which takes into account both the amplitude and time course of voltage responses, by delivering alpha injections of current designed to mimic EPSP waveforms at a frequency of 50 Hz at varying membrane potentials and measuring the ratio between the amplitude of the fifth EPSP and the first

EPSP. We found no difference in the summation ratio between dorsal and ventral fmr1 KO CA1 pyramidal neurons (Figure 3.3H; ventral: n=12; dorsal: n=12; n=13 animals; RM 2-way ANOVA:

F(1, 20)=0.29, p=0.6). Our measurements on subthreshold properties associated with h-channel expression, as in wild type mouse hippocampus, show no difference in h-channel expression associated physiological markers between fmr1 KO dorsal and ventral CA1 pyramidal neurons.

101

Figure 3.3: Subthreshold intrinsic properties of dorsal and ventral fmr1 KO CA1 pyramidal neurons. A Voltage responses from ventral (top) and dorsal (middle) CA1 neurons in response to a series of current injections (bottom). B RN of ventral and dorsal CA1 neurons measured while varying Vm (n=12 ventral cells and 12 dorsal cells from 13 mice). C Resting Vm in ventral and dorsal neurons (n=13 ventral cells and 26 dorsal cells from 30 mice). D Voltage sag in ventral and dorsal CA1 neurons measured while varying Vm (n=12 ventral cells and 12 dorsal cells from 13 mice). E Rebound slope in ventral and dorsal CA1 neurons measured while varying Vm (n=12 ventral cells and 12 dorsal cells from 13 mice). F fR measured while varying Vm in dorsal and ventral CA1 neurons. Inset: voltage traces showing response to a chirp current injection. G τM measured while varying Vm in dorsal and ventral CA1 neurons. Inset: representative voltage responses to -100 pA current injection. H summation measured while varying Vm in dorsal and ventral CA1 neurons. Inset: representative voltage responses to injection of 5 alpha waveforms at 20Hz.

102

Action potential threshold in dorsal and ventral fmr1 KO CA1 pyramidal neurons

We measured the voltage threshold for the generation of action potentials by selecting trains of action potentials elicited by square pulse current injections in which the first action potential in the train occurred approximately 50 ms after the onset of the current injection (Figure

3.4A; (Dougherty et al., 2012; Ordemann et al., 2019). Action potential threshold was defined as the point at which the rate of rise exceeded 20 mV/ms. Action potential threshold was more depolarized in ventral compared with dorsal neurons in fmr1 KO CA1 pyramidal neurons (Figure

3.4B; ventral: -46.65 ± 0.93 mV, n=9; dorsal: -51.44 ± 0.66 mV, n=9; n=13 animals; unpaired t- test: t(14.45)=4.2, p=0.0008, h2=0.55). In a small number of neurons, a more finely tuned measure of threshold was performed in which the duration of the current stimulus was systematically changed and current delivered was adjusted so that an action potential was elicited at the end of the current step (Figure 3.4C; (Higgs and Spain, 2011; Kalmbach et al., 2015). Our findings revealed no difference in voltage threshold using this method of calculation (Figure 3.4D; ventral: n=4; dorsal: n=4; n=6 animals; RM 2-way ANOVA: F(1, 5)=5.24, p=0.084); however, this is likely due to the small sample size of the experiment as the data are clearly trending to match the findings of Figure 3.4A-B.

103

Figure 3.4: Measurements of voltage threshold for action potential generation in dorsal and ventral fmr1 KO CA1 pyramidal neurons. A Voltage responses to 500 ms square current injections in which action potentials occur approximately 50 ms after current onset. Inset: expanded view of black box, voltage threshold indicated by black dots. B Action potential threshold in dorsal and ventral CA1 neurons (n=9 ventral cells and 9 dorsal cells from 13 mice). C Voltage responses showing action potentials elicited near the end of a 24 ms current injection. Inset: Expanded area indicated by the black box showing action potential threshold (black dot). D Action potential threshold in dorsal and ventral CA1 neurons. (n=4 ventral cells and 4 dorsal cells from 6 mice).

104

Dorsal ventral properties of wild type and fmr1 KO CA1 pyramidal neurons

We measured intrinsic properties in the dorsal and ventral poles of the hippocampus in wild type and fmr1 KO CA1 pyramidal neurons. Chapters 2 and 3 show the results of comparisons between dorsal and ventral measurements in either wild type (Chapter 2) or fmr1 KO (Chapter 3)

CA1 pyramidal neurons. Table 3.1 shows statistical tests comparing properties in either dorsal or ventral regions of hippocampus between wild type and fmr1 KO CA1 pyramidal neurons.

Statistical analysis comparing intrinsic properties measured in dorsal wild type and dorsal fmr1

KO CA1 pyramidal neurons reveal no differences (Table 3.1). The same comparison performed between intrinsic properties of wild ventral and fmr1 KO ventral CA1 pyramidal neurons also revealed no differences (Table 3.1).

105

wild type dorsal versus fmr1 KO dorsal Measurement statistical test test statistic p-value significant? F/I at rest RM 2-way ANOVA F(1, 20)=0.003 0.95 no F/I at -65 mV RM 2-way ANOVA F(1, 20)=1.12 0.3 no

RN RM 2-way ANOVA F(1, 29)=1.87 0.085 no

resting Vm unpaired t-test t(54)=1.55 0.13 no membrane sag RM 2-way ANOVA F(1, 29)=0.18 0.68 no rebound slope RM 2-way ANOVA F(1, 29)=0.7 0.41 no

Rf RM 2-way ANOVA F(1,28)=2.48 0.13 no

�m RM 2-way ANOVA F(1, 27)=0.61 0.44 no summation RM 2-way ANOVA F(1, 23)=0.35 0.55 no threshold (from F/I) unpaired t-test t(21.43)=2.04 0.054 no RM 2-way ANOVA F(1, 9)=0.11 0.75 no threshold (△t)

wild type ventral versus fmr1 KO ventral Measurement statistical test test statistic p-value significant? F/I at rest RM 2-way ANOVA F(1, 16)=0.32 0.58 no F/I at -65 mV RM 2-way ANOVA F(1, 17)=1.49 0.24 no

RN RM 2-way ANOVA F(1, 29)=0.55 0.46 no

resting Vm unpaired t-test t(29)=0.8 0.43 no membrane sag RM 2-way ANOVA F(1, 29)=0.47 0.5 no rebound slope RM 2-way ANOVA F(1, 29)=0.4 0.53 no

Rf RM 2-way ANOVA F(1, 29)=0.04 0.85 no

�m RM 2-way ANOVA F(1, 29)=1.45 0.23 no summation RM 2-way ANOVA F(1, 29)=0.7 0.41 no threshold (from F/I) unpaired t-test t(13.41)=0.88 0.4 no threshold (△t) RM 2-way ANOVA F(1, 8)=0.37 0.56 no

Table 3.1: Statistical tests comparing wild type and fmr1 KO dorsal and ventral CA1 pyramidal neurons.

106

Discussion

We investigated the subthreshold and suprathreshold intrinsic properties of dorsal and ventral fmr1 KO CA1 pyramidal neurons. Our findings reveal no differences between properties measured in wild type and fmr1 KO dorsal and ventral CA1 pyramidal neurons. Consistent with our wild type data from mouse hippocampus, we show a depolarized action potential threshold in ventral compared with dorsal CA1 pyramidal neurons, but no difference in action potential firing.

Our findings suggest that the same difference in D-type potassium channel expression is present in fmr1 KO CA1 pyramidal neurons of mouse hippocampus.

The findings in this Chapter are in conflict with previous research examining action potential firing in wild type and fmr1 KO CA1 pyramidal neurons measured from the soma. It has been observed previously that action potential firing is increased in fmr1 KO neurons based on increased somatic input resistance (Luque et al., 2017). Another study also found increased action potential firing caused by increased length of the axon initial segment (Booker et al., 2020). While the findings presented here do not agree with the findings in these studies, there are several potential explanations. The finding that increased input resistance results in greater spike firing in fmr1 KO CA1 neurons (Luque et al., 2017) was made in an fmr1 KO mouse of a different genetic background than the present study. While both of the knockout strains have been used widely in the study of Fragile X syndrome, the findings made in each have often been in conflict (Spencer et al., 2011). Investigation into the effects axon initial segment length have on somatic action potential firing (Booker et al., 2020) took place in animals of the same genetic background as the present study, however, the study was performed in animals under 1 month of age while the data presented here was taken from animals at least 2 months old.

107

Significant evidence on the synaptic dysfunction of CA1 pyramidal neurons in Fragile X syndrome has been presented. However, until recently, evidence of intrinsic dysfunction in physiological function has been sparse. Dendritic channelopathies affecting the integration of synaptic signals have been identified in CA1 pyramidal neurons as well as in cortical pyramidal neurons (Brager et al., 2012; Deng et al., 2013; Routh et al., 2013; Zhang et al., 2014; Kalmbach et al., 2015; Deng and Klyachko, 2016b; Deng et al., 2019; Brandalise et al., 2020). Our data suggests that the properties of fmr1 KO CA1 pyramidal neuron soma are intact in Fragile X syndrome. Despite normal somatic function, the impact that changes in dendritic function has on overall neuronal and circuit function remains to be fully investigated.

108

Chapter 4: Impaired long-term potentiation and synaptically evoked calcium signaling in the temporoammonic pathway in a mouse model of Fragile X syndrome*

FMRP controls many neuronal proteins, including those involved in synaptic structure, function and plasticity, through translational regulation of target mRNAs (Bassell and Warren,

2008; Darnell et al., 2011) and direct protein-protein interactions (Ramos et al., 2006; Brown et al., 2010; Deng et al., 2013; 2019; Brandalise et al., 2020). The Schaffer collateral CA3 to CA1 synapse has been extensively studied in FXS (Huber et al., 2002; Shang et al., 2009; Brager et al.,

2012; Routh et al., 2013; Bostrom et al., 2015; Toft et al., 2016; Wang et al., 2016). By contrast, few studies have investigated the temporoammonic (TA) entorhinal cortex to CA1 synapses.

Wahlstrom-Helgren and Kylachko found no difference in TA synaptic transmission using somatic whole cell recording (Wahlstrom-Helgren and Klyachko, 2015), while Booker and colleagues found that TA synaptic transmission was reduced in fmr1 KO mice using extracellular field potential recording (Booker et al., 2020). Neither of these studies, however, investigated TA long- term potentiation (LTP). Given the lack of LTP studies and the critical involvement of the TA pathway in the consolidation of long-term memory (Remondes and Schuman, 2004) and its necessity for the generation of hippocampal CA1 place fields (Bittner et al., 2017), we asked whether TA-LTP is altered in FXS.

* Data included in this section is accepted at The Journal of Neuroscience: Ordemann GJ, Apgar CJ, Chitwood RA, Brager DH (2021) Altered A-type K+ channel function impairs dendritic spike initiation and TA LTP in Fragile X syndrome. J Neurosci in press. Contributions: GJO, CJA, RAC, and DHB performed and analyzed experiments; GJO and DHB conceived experiments and wrote the manuscript. 109

Using somatic and dendritic recording, we found that TA-LTP following theta-burst stimulation was impaired in fmr1 KO mice. Our results reveal no difference in baseline synaptic transmission between wild type and fmr1 KO CA1 pyramidal neurons. In order to observe intracellular calcium signaling during patch clamp recordings, we filled individual neurons with calcium sensitive dye and imaged changes in intracellular calcium concentrations in the dendrites of wild type and fmr1 KO CA1 pyramidal neurons. Two photon imaging of back propagating action potentials (bAPs) showed decreased decay of bAP amplitude in fmr1 KO dendrites, in agreement with previous findings (Routh et al., 2013). NMDA receptors contribute approximately half of the intracellular calcium concentration increase during high frequency synaptic stimulation

(see Figure 4.4). Two photon imaging during bursts of TA stimulation as well as during the induction of LTP revealed that dendritic Ca2+ signals were smaller in fmr1 KO neurons. Dendritic recordings from wild type and fmr1 KO CA1 pyramidal neurons showed a lack of TA-LTP as well as no differences in baseline synaptic transmission.

Results

Long-term potentiation of TA synapses is impaired in fmr1 KO mice

We made somatic whole cell current clamp recordings from CA1 pyramidal neurons in middle hippocampus in wild type and fmr1 KO male mice and recorded TA EPSPs before and after theta burst stimulation (TBS) to induce TA-LTP (Tsay et al., 2007) (Figure 4.1A). In wild type neurons TA EPSP slope increased 30 minutes post TBS (Figure 4.1B-C; wild type: n=7 from

6 mice. pre-TBS: 0.24 ± 0.054 mV/ms, post-TBS: 0.84 ± 0.19 mV/ms. Wilcoxon: W=28, p=0.02, h2=0.79.). By contrast, in fmr1 KO CA1 neurons TA EPSP slope was not increased after TBS (Fig.

110

1B-C; fmr1 KO: n=9 from 7 mice. pre-TBS: 0.23 ± 0.04 mV/ms, post-TBS: 0.24 ± 0.05 mV/ms.

Wilcoxon: W= -3, p=0.92). The somatic depolarization during the TBS (area under the curve) was not different between wild type and fmr1 KO CA1 pyramidal neurons (Figure 4.1D; wild type:

12.97 ± 2.38 mV*ms; fmr1 KO: 5.93 ± 1.48 mV*ms; Mann-Whitney: U=13, p=0.06). In agreement with previous work (Wahlstrom-Helgren and Klyachko, 2015), we found no difference in postsynaptic responsiveness to single TA stimuli (Figure 4.1E; wild type: n=5 from 4 mice; fmr1 KO: n=6 from 4 mice; 2-way RM ANOVA: F(1,7)=0.65, p=0.45), baseline paired-pulse ratio

(Figure 4.1F; wild type: n=5 from 4 mice; fmr1 KO: n=6 from 4 mice; 2-way RM ANOVA:

F(1,9)=4.23, p=0.07) or temporal summation (Figure 4.1G; wild type: n=5 from 4 mice; fmr1 KO: n=6 from 4 mice; 2-way RM ANOVA: F(1,9)=0.55, p=0.48) between wild type and fmr1 KO neurons. Paired pulse ratio was also not changed after TBS consistent with a postsynaptic locus of

LTP (wild type: 2-way RM ANOVA: F(1,4)=3.53, p=0.13; fmr1 KO: 2-way RM ANOVA:

F(1,5)=3.98, p=0.1).

111

Figure 4.1: TA-LTP is impaired in fmr1 KO CA1 pyramidal neurons. A Diagram of recording configuration for somatic EPSP measurements. B EPSP slope before and after TBS in wild type but not fmr1 KO CA1 neurons; baseline: (mean of 5 minutes before TBS); post-TBS: (mean of 5 minutes at the end of recording). C Normalized change in EPSP slope for wild type and fmr1 KO CA1 neurons. Inset: Representative EPSP traces from baseline (a) and post-TBS (b). D Graph of area under the curve during TBS. E Input output of TA inputs measuring EPSP slope. Inset: representative experiments from wild type and fmr1 KO neurons. F Paired pulse ratio as a function of interstimulus interval (ISI). Inset: representative 50 ms ISI traces. G Temporal summation of TA EPSPs as a function of frequency. Inset: representative 50 Hz traces.

112

Dendritic morphology is not different in wild type and fmr1 KO CA1 pyramidal neurons

We filled CA1 pyramidal neurons with neurobiotin during whole cell recording for post- hoc morphological reconstruction and analysis (Figure 4.2A). There was no significant difference in neuronal surface area (Figure 4.2B; wild type: n=6 from 4 mice, 13.3 ± 1.1 mm2; fmr1 KO: n=6 from 4 mice, 12.9 ± 1.2 mm2; Mann-Whitney: U=16, p=0.82) or somato-dendritic length, measured as straight line distance from the soma to the tip of the most distal dendrite, (Figure

4.2C; wild type: 426.6 ± 25.01 µm; fmr1 KO: 441.3 ± 24.56 µm; Mann-Whitney: U=15, p=0.7) between wild type and fmr1 KO CA1 pyramidal neurons. We used Sholl analysis to compare dendritic branching between wild type and fmr1 KO neurons. There was no difference in dendritic morphology between wild type and fmr1 KO CA1 pyramidal neurons (Figure 4.2D; 2-way RM

ANOVA: F(1,10)=0.19, p=0.67). Taken together, these data show that although TA-LTP is impaired in fmr1 KO CA1 neurons, basal TA synaptic transmission and CA1 neuron morphology are normal.

113

Figure 4.2: CA1 pyramidal neuron morphology is not different between wild type and fmr1 KO mice. A Representative neuronal reconstructions of wild type (black) and fmr1 KO (red) CA1 pyramidal neurons. B Total dendritic length in wild type and fmr1 KO CA1 neurons. C Surface area in wild type and fmr1 KO CA1 neurons. D Sholl analysis of dendritic branching in wild type and fmr1 KO CA1 pyramidal neurons. 114

Smaller distance dependent decay of bAPs in fmr1 KO dendrites

Using two photon calcium imaging we first set out to replicate previous findings on the decay of bAPs as they propagate into the dendritic arbor (Routh et al., 2013). A-type K+ channels expressed in CA1 pyramidal neuron dendrites constrain the amplitude of bAPs (Hoffman et al.,

+ 1997). In fmr1 KO CA1 dendrites, reduced A-type K channel current (IKA) results in greater amplitude bAPs that decay to a smaller extent as they propagate into the dendritic arbor. This propagation can be observed in the degree to which voltage gated calcium channels are activated by imaging changes in intracellular calcium concentration. By using this measure, it can be deduced that greater change in intracellular calcium concentration equates greater amplitude of bAP (Routh et al., 2013). Using a somatic patch pipette we filled individual wild type or fmr1 KO

CA1 pyramidal neurons with Alexa 594 for the visualization of neuronal structure and cell membrane impermeant OGB-1 for the imaging of changes in intracellular calcium concentrations

(Figure 4.3A). CA1 neuron dendrites were imaged at four locations along the apical dendrite

(indicated by yellow bars in Figure 4.3A). In agreement with previous findings we observed that the amplitude of the change in intracellular calcium concentration in response to somatically generated trains of action potentials was greater in the proximal apical dendrite of fmr1 KO compared with wild type CA1 pyramidal neurons (Figure 4.3B-C). These results validate the calcium imaging techniques utilized in this chapter.

115

Figure 4.3: Greater distance dependent decay of bAP amplitude in the dendrites of wild type compared with fmr1 KO CA1 pyramidal neurons. A CA1 pyramidal neuron filled with OGB-1 (100 µM) and Alexa 594 (40 µM). Yellow bars represent imaging locations along the apical dendrite. B Experiment measuring the amplitude of Ca2+ response to a train of 5 Aps delivered at 20 Hz. Top: voltage traces measured with patch pipette at the soma. Bottom: Ca2+ traces showing responses at 50, 100, 150, and 200 µm from the soma in response to bAPs. C Graph of the change in [Ca2+] in response to bAPs as a function of distance of measurement from the soma.

116

NMDAR contribution to calcium response to bursts of high frequency synaptic stimulation

The induction of TA-LTP requires an increase in intracellular calcium concentration mediated by voltage gated calcium channels and NMDA receptor activation (Golding et al., 2002;

Remondes and Schuman, 2003). We examined the contribution of NMDA receptor activation to changes in intracellular calcium concentration in response to bursts of high frequency synaptic stimulation of the TA pathway at varying baseline EPSP amplitudes (Figure 4.4A-B). At a baseline

EPSP amplitude of 4 mV we found that NMDA receptor activation accounted for approximately

50% of the total increase in intracellular calcium concentration (Figure 4.4C; wild type: n=7 from

6 animals, baseline: 65.62 ± 19.52 %DF/F0, post-AP5: 30.41 ± 15.79 %DF/F0; Wilcoxon: W= -28, p=0.016, h2=0.34).

117

Figure 4.4: NMDARs account for approximately 50% of synaptically evoked change in intracellular [Ca2+]. A Recording paradigm during high frequency bursts of synaptic stimulation. Yellow bar represents region of linescan measurements of changes in intracellular [Ca2+]. B Representative voltage and Ca2+ traces in response to burst of 100 Hz synaptic stimulation with either 2 or 4 mV baseline EPSP amplitudes. C Group data showing the amplitude of the change in intracellular [Ca2+] in response to bursts of synaptic stimulation before and after the application of the NMDAR blocker AP5.

118

Ca2+ entry into fmr1 KO neurons is reduced at TA synapses

A rise in intracellular Ca2+ during TBS is necessary for the induction of TA-LTP (Golding et al., 2002; Remondes and Schuman, 2003; Takahashi and Magee, 2009). We used two photon

Ca2+ imaging to directly measure changes in dendritic Ca2+ during TA TBS in wild type and fmr1

KO CA1 neurons (Figure 4.5A). We varied the initial EPSP amplitude and triggered bursts TA

EPSPs (10 at 100 Hz) (Tsay et al., 2007). Small but detectable Ca2+ signals were observed at EPSP amplitudes greater than 1 mV for both wild type and fmr1 KO neurons. In both wild type and fmr1

KO dendrites, the Ca2+ signal increased with increasing EPSP amplitude; however, the Ca2+ signal was smaller in fmr1 KO compared to wild type dendrites (Figure 4.5B-C; wild type: n=11 from 8 mice; fmr1 KO: n=10 from 7 mice; 2-way RM ANOVA: F(1,19)=13.14, p=0.002, h2=0.13.

Interaction: F(2, 38): 4.86, p=0.013, h2=0.1. Sidak’s test-4 mV: p=0.04, h2=0.29; wild type: 52.73

± 13.38 DF/F, fmr1 KO: 12.26 ± 2.77 DF/F).

119

Figure 4.5: Synaptic Ca2+ signaling reduced in fmr1 KO dendrites. A CA1 neuron filled with OGB-1 (100 µM) and Alexa 594 (40 µM). Yellow bar represents location of linescan imaging in the s.l.m region. B Representative Ca2+ and voltage signals during 100 Hz bursts of TA EPSPs using 1 and 4 mV initial EPSP amplitudes. C Peak intracellular [Ca2+] during bursts of synaptic activity as a function of initial EPSP amplitude. 120

Impaired Ca2+ entry during TA-LTP induction in fmr1 KO CA1 pyramidal neurons

Changes in intracellular Ca2+ concentration were measured during the induction TA-LTP

(Figure 4.6A). As shown previously, wild type neuron EPSP slope increased consistently in response to TBS while fmr1 KO neurons did not (Figure 4.6B-C; wild type: n=2 from 2 mice, pre-

TBS: 0.12 ± 0.07 mV/ms, post-TBS: 0.96 ± 0.58 mV/ms; fmr1 KO: n=2 from 2 animals, pre-TBS:

0.05 ± 0.02 mV/ms, post-TBS: 0.05 ± 0.01 mV/ms). Our findings suggest that Ca2+ entry (wild type: 33.99 ± 26.09 %DF/F0; fmr1 KO: 1.05 ± 0.15 %DF/F0) and depolarization (wild type: 33.84

± 6.23 mV*ms; fmr1 KO: 4.99 ± 0.66 mV*ms) during TBS of TA inputs is greatly reduced in fmr1

KO compared with wild type CA1 pyramidal neuron dendrites (Figure 4.6D-E). Using two photon

Ca2+ imaging we show a clear reduction in intracellular calcium signaling in response to synaptic stimulation in fmr1 KO compared with wild type CA1 pyramidal neuron dendrites.

121

Figure 4.6: Changes in intracellular [Ca2+] during TBS of the TA pathway in wild type and fmr1 KO CA1 pyramidal neurons. A Recording paradigm of voltage recordings and Ca2+ imaging during TA-LTP induction protocol. B EPSP slope before and after TBS of TA inputs in wild type and fmr1 KO CA1 pyramidal neurons. C Normalized EPSP slope 5 minutes before and 30 minutes after TBS of TA inputs. D Top: Voltage traces during TBS of TA inputs. Bottom: Change in intracellular [Ca2+] during TBS of TA inputs. E Area under the curve of the change in intracellular [Ca2+] during TBS. F Area under the curve of voltage signal during TBS.

122

Dendritic recordings of TA synaptic transmission and LTP

Using somatic recordings, we showed a clear lack of TA-LTP in fmr1 KO CA1 pyramidal neurons and dendritic imaging revealed reduced dendritic Ca2+ influx during bursts of TA stimulation. This suggests that the dendrites are the locus of the changes that impair TA-LTP in fmr1 KO CA1 pyramidal neurons. Many of the dendritic events required for TA-LTP are distorted or undetectable using somatic recording due to the filtering properties of CA1 dendrites. Thus, we performed current clamp recordings from the apical dendrites of wild type and fmr1 KO CA1 pyramidal neurons (Figure 4.7A; 40 µM Alexa 594; distance from soma–wild type: 211.8 ± 3.52

µm; fmr1 KO 212.1 ± 4.69 µm; Mann-Whitney: U=99.5; p=0.84). Consistent with our somatic recordings, TA EPSP slope was significantly increased after TBS in wild type but not fmr1 KO neurons (Figure 4.7B-C; wild type: n=8 from 4 mice; pre-TBS: 0.65 ± 0.15 mV/ms, post-TBS:

2.74 ± 0.59 mV/ms. Wilcoxon: W=36, p=0.008, h2=0.4; fmr1 KO: n=7 from 4 mice; pre-TBS:

0.65 ± 0.11 mV/ms, post-TBS: 1.01 ± 0.18 mV/ms. Wilcoxon: W=18, p=0.16). There was no difference in the area under the curve during TBS of TA synapses between wild type and fmr1 KO

CA1 pyramidal neurons (Figure 4.7D; wild type: 13.33 ± 3.14 mV*ms, fmr1 KO: 8.78 ± 3.76 mV*ms. Mann-whitney: U=16, p=0.19). There was no significant difference in the response to single stimuli (Figure 4.7E; wild type: n=6 from 2 mice; fmr1 KO: n=7 from 4 mice; 2-way RM

ANOVA: F(1,17)=0.63, p=0.44), paired-pulse ratio (Figure 4.7F; wild type: n=7 from 4 mice; fmr1 KO: n=7 from 4 mice; 2-way RM ANOVA: F(1,12)=0.022, p=0.88), or summation (Figure

4.7G; wild type: n=6 from 2 mice; fmr1 KO: n=7 from 4 mice; 2-way RM ANOVA:

F(1,12)=0.00001, p=0.99) between wild type and fmr1 KO CA1 neurons. Paired-pulse ratios were not different after TBS (wild type: 2-way RM ANOVA: F(1,6)=1.89, p=0.22; fmr1 KO: 2-way

123

RM ANOVA: F(1,6)=0.02, p=0.89). Dendritic recordings, much like in somatic recordings, show a lack of fmr1 KO TA-LTP.

124

Figure 4.7: Dendritic recordings show a lack of TA-LTP in fmr1 KO CA1 pyramidal neurons. A Representative dendritic recording from a CA1 neuron filled with Alexa 594 (40 µM). B EPSP slope before and TBS of wild type and fmr1 KO CA1 pyramidal neuons. C Normalized EPSP slope for wild type and fmr1 KO CA1 neurons. Inset: Representative EPSP traces from baseline (a) and post-TBS (b). D Graph of area under the curve during TBS. E Input output of EPSP slope at TA inputs in dendritic recordings. Inset: representative dendritic experiments. F Paired pulse ratio in wild type and fmr1 KO CA1 pyramidal neurons. Inset: 50 ms ISI paired pulse. G Temporal summation in wild type and fmr1 KO CA1 pyramidal neurons. Inset: 50 Hz temporal summation experiment.

125

Discussion

The data in this chapter reveals a clear lack of TA-LTP in fmr1 KO CA1 pyramidal neurons. It is clear based on previous investigations into the mechanisms underlying the induction of TA-LTP that understanding the mechanism underlying the impairments in intracellular calcium signaling in fmr1 KO CA1 pyramidal neuron dendrites is the key to understanding why fmr1 KO neurons do not express TA-LTP. Chapter 5 deals with the investigation of the mechanisms underlying the lack of TA-LTP in fmr1 KO CA1 pyramidal neurons. Further discussion of the implications of a lack of TA-LTP is presented in chapter 5.

126

Chapter 5: Altered A-type potassium channel function impairs dendritic spike initiation and temporoammonic long-term potentiation in Fragile X syndrome*

In chapter 4 we identified a lack of TA-LTP accompanied by reduced synaptically evoked

Ca2+ signaling in fmr1 KO CA1 pyramidal neurons. However, the mechanistic underpinnings of the lack of LTP remains to be elucidated. Identifying the differences between wild type and fmr1

KO CA1 pyramidal neurons that results in a lack of TA-LTP requires understanding of the normal progression of TA-LTP. Coordinated activation of TA synapses results in the generation of dendritic Na+ mediated spikes that trigger the activation of NMDA receptors and voltage gated calcium channels that result in a large rise in intracellular Ca2+ concentration (Golding et al., 2002;

Remondes and Schuman, 2003; Kim et al., 2015). Expression of TA-LTP occurs through the activation of PI3K and subsequent insertion of AMPA receptors, causing in increase in synaptic strength (Clements and Harvey, 2020). While little is known about the maintenance of LTP at large, a convergent mechanism for LTP induction involving activation of serotonin receptors in the TA pathway results in CaMKII dependent LTP of TA inputs (Cai et al., 2013). This makes it likely that TA-LTP maintenance is at least partially dependent upon CaMKII activation. It remains unclear at what stage fmr1 KO TA-LTP is disrupted; however, our findings indicating a reduction in intracellular Ca2+ signaling during TA stimulation in fmr1 KO CA1 dendrites suggest that the impairment affects the induction of TA-LTP.

* Data included in this section is accepted to The Journal of Neuroscience: Ordemann GJ, Apgar CJ, Chitwood RA, Brager DH (2021) Altered A-type K+ channel function impairs dendritic spike initiation and TA LTP in Fragile X syndrome. J Neurosci in press. Contributions: GJO, CJA, RAC, and DHB performed and analyzed experiments; GJO and DHB conceived experiments and wrote the manuscript. 127

CA1 pyramidal neuron dendrites express Na+, K+, Ca2+, and h-channels which play crucial roles in dendritic integration and the induction of LTP. We previously showed that the functional expression of dendritic h-channels (Ih) is higher in fmr1 KO CA1 neurons (Brager et al., 2012;

Brandalise et al., 2020). We also demonstrated that the current carried by dendritic A-type K+ channels (IKA) is reduced in fmr1 KO CA1 neurons (Routh et al., 2013). These changes alter the local integrative properties and increase the backpropagation of action potentials respectively.

While the effect of these changes in dendritic h-channels and IKA on Schaffer collateral LTP were previously described (Brager et al., 2012; Routh et al., 2013), the impact on TA-LTP remains unknown.

Using somatic and dendritic recording, we found that the lack of LTP was not due to the higher expression of h-channels as block of Ih with ZD7288 did not rescue LTP. Although complex and pharmacologically isolated Ca2+ spikes recorded in the dendrites were not different, the threshold for fast dendritic spikes (dspikes) was more depolarized in fmr1 KO CA1 pyramidal

2+ neurons. Application of extracellular Ba or AmmTx3 to block IKA rescued dspike threshold and

TA-LTP in fmr1 KO CA1 pyramidal neurons, implicating A-type K+ channels.

Results

Block of Ih does not rescue TA-LTP in fmr1 KO neurons

The expression of h-channels in the distal dendrites of CA1 pyramidal neurons constrains

TA inputs and LTP (Tsay et al., 2007). We previously showed that dendritic Ih is elevated in fmr1

KO CA1 pyramidal neurons compared to wild type (Brager et al., 2012; Brandalise et al., 2020).

Higher Ih in fmr1 KO CA1 neurons may reduce the effectiveness of TA synapses in the distal

128 dendrites and impair LTP (Magee, 1998; 1999). To test this hypothesis, we repeated the TBS TA-

LTP experiments with Ih blocked by 20 µM ZD7288 (Figure 5.1A). With Ih blocked, TBS significantly potentiated TA EPSPs in wild type (n=6 from 4 mice; pre-TBS: 0.076 ± 0.018 mV/ms, post-TBS: 0.23 ± 0.073 mV/ms. Wilcoxon: W=21, p=0.03, h2=0.81), but not fmr1 KO

CA1 pyramidal neurons (Figure 5.1B-C; n=5 from 4 mice; pre-TBS: 0.13 ± 0.029 mV/ms, post-

TBS: 0.12 ± 0.041 mV/ms. Wilcoxon: W= -3, p=0.81). Our results showed no difference in the area under the curve during TBS of TA inputs in the presence of ZD (Figure 5.1D; wild type: 12.3

± 3.8 mV*ms, fmr1 KO: 10.5 ± 2.4 mV*ms. Mann-Whitney: U=11, p=0.84). These results suggest that higher dendritic Ih alone does not account for the lack of TA-LTP in fmr1 KO CA1 pyramidal neurons.

129

Figure 5.1: Block of h-channels by 20 µM ZD7288 does not rescue TA-LTP in fmr1 KO CA1 pyramdial neurons. A Recording paradigm during TA-LTP experiments. B EPSP slope before and after TBS in wild type and fmr1 KO CA1 pyramidal neurons. C Normalized EPSP slope 5 minutes before and 30 minutes after TBS of TA inputs. Inset: Representative baseline and post-TBS EPSPs from wild type and fmr1 KO CA1 neurons. D Left: Wild type and fmr1 KO traces during TBS. Right: Graph of the area under the curve during TBS in the presence of ZD7288.

130

NMDAR-mediated EPSPs are not different between wild type and fmr1 KO TA synapses

N-methyl-D-aspartate receptor (NMDAR) activation is a key source of Ca2+ influx during the induction of TA-LTP (Golding et al., 2002; Remondes and Schuman, 2003; Takahashi and

Magee, 2009). Although we previously showed that there was no significant difference in TA

EPSPs between wild type and fmr1 KO CA1 neurons (Figure 4.1), those experiments did not separate the AMPA and NMDAR contributions to the EPSP. To test if NMDAR-mediated EPSPs are different between wildtype and fmr1 KO neurons, we stimulated TA inputs in the presence of

AMPA receptor blocker DNQX (20 µM) and with 0 mM Mg2+ in the extracellular saline (Figure

5.2A, left). Application of the NMDAR antagonist D-AP5 (50 µM) confirmed isolation of

NMDAR-mediated EPSPs in a subset of experiments (Figure 5.2A, right). We found no difference in isolated NMDAR-mediated TA EPSPs between wild type and fmr1 KO CA1 pyramidal neurons

(Figure 5.2B; wild type: n=7 from 3 mice; fmr1 KO: n=7 from 3 mice; 2-way RM ANOVA:

F(1,12)=0.19, p=0.87).

131

Figure 5.2: NMDAR synaptically evoked activity is not different in the TA pathway between wild type and fmr1 KO CA1 pyramidal neurons. A Representative input output traces of NMDAR dependent EPSPs in 0 mM Mg2+ and the AMPA blocker DNQX (20 µM) (left). Addition of the NMDAR antagonist AP5 (25 µM) confirmed the synaptic response was NMDAR dependent (right). E, Slope of NMDAR EPSPs as a function of stimulation intensity was not different between wild type and fmr1 KO neurons.

132

Dendritic complex spikes are not different between wild type and fmr1 KO neurons

NMDARs, activated by presynaptic glutamate release, are one source of Ca2+ signaling.

We found that subthreshold NMDAR-mediated EPSPs activated by stimulation of TA synapses were not different between wild type and fmr1 KO neurons. Dendritic voltage-gated channels can also contribute to rises in dendritic Ca2+ (Magee and Johnston, 1995a; Golding et al., 2002;

Remondes and Schuman, 2003; Takahashi and Magee, 2009). The activation of distal synapses in

CA1 pyramidal neurons gives rise to complex spikes in CA1 dendrites mediated by voltage-gated

Na+ and Ca2+ channels (Andreasen and Lambert, 1995; Golding and Spruston, 1998; Golding et al., 1999; Takahashi and Magee, 2009; Kim et al., 2015). We therefore tested the hypothesis that complex spikes and Ca2+-dependent action potentials are impaired in fmr1 KO neurons.

In CA1 pyramidal neurons, dendritic complex spikes consist of a fast initial Na+ spike which triggers 1–3 slower, Ca2+ mediated spikes (Golding et al., 1999). We used dendritic current injection (1 second) to compare complex spikes between wild type and fmr1 KO CA1 neurons

(Figure 5.3A). Previous studies using rat hippocampus showed that only a sub-population of CA1 dendrites fire complex spikes (Andreasen and Lambert, 1995; Golding et al., 1999). We found that approximately half of mouse CA1 neurons fired complex spikes and that the proportion was not different between wild type and fmr1 KO CA1 pyramidal neurons (wild type: 51.6%, fmr1 KO:

47.6%). There was no difference in the width of the complex spikes between wild type and fmr1

KO CA1 dendrites (Figure 5.3B-D; wild type: n=16 from 14 mice; fmr1 KO: n=11 from 10 mice;

2-way RM ANOVA: F(1,24)=0.00002, p=0.67) (Takahashi and Magee, 2009).

133

Figure 5.3: Dendritic complex spikes are not different between wild type and fmr1 KO CA1 pyramidal neurons. A Dendritic recording schematic. B Method for measuring complex spike width (see methods for description). C Complex spikes elicited by dendritic current injection in wild type and fmr1 KO neurons (left) and example complex spikes (boxes) shown on expanded time scale (right). D Complex spike width as a function of number of events (1-3).

134

Dendritic Ca2+ spikes are not different between wild type and fmr1 KO CA1 pyramidal neurons

To isolate the dendritic Ca2+ spike component of the complex spike, we applied 0.5 µM

TTX to block voltage-gated Na+ channels and 50 µM 4-AP to block voltage-gated K+ channels and bias the dendrites toward firing dendritic Ca2+ spikes (Benardo et al., 1982; Andreasen and

Lambert, 1995; Golding et al., 1999). Dendritic Ca2+ spikes were evoked by depolarizing current injections and confirmed to be mediated by voltage-gated Ca2+ channels by the addition of 200

µM Cd2+ (Figure 5.4A). The number of elicited Ca2+ spikes was not different between wild type and fmr1 KO CA1 pyramidal neuron dendrites (Figure 5.4B). The first Ca2+ spike elicited was selected for further analysis (Figure 5.4C-G). Dendritic Ca2+ spike amplitude (Figure 5.4D; wild type: n=6 from 5 mice, 35.32 ± 1.08 mV; fmr1 KO: n=7 from 6 mice, 38.08 ± 1.84 mV; Mann-

Whitney: U=9, p=0.18), maximum rate of rise (Figure 5.4E; wild type: 7.66 ± 1.06 mV/ms, fmr1

KO: 7.79 ± 0.51 mV/ms. Mann-Whitney: U=20.5, p=0.98), maximum rate of decay (Figure 5.4F; wild type: -4.52 ± 0.70 mV/ms, fmr1 KO: -4.22 ± -2.80 mV/ms. Mann-Whitney: U=19, p=0.84) and estimated threshold (Figure 5.4G; wild type: -16.47 ± 2.25 mV, fmr1 KO: -14.98 ± 3.76 mV.

Mann-Whitney: U=19, p=0.84) were not different between wild type and fmr1 KO neurons. These results suggest that impairment of dendritic complex spikes or Ca2+ spikes does not contribute to the reduced dendritic Ca2+ signal in fmr1 KO CA1 pyramidal neuron bursts of TA stimulation.

135

Figure 5.4: Dendritic Ca2+ spikes are not different between wild type and fmr1 KO CA1 pyramidal neurons. A Traces showing isolated dendritic Ca2+ spikes in the presence of 500 nM TTX and 50 µM 4-AP. Application of 200 µM Cd2+ confirmed that spikes were generated by voltage gated Ca2+ channels. B The number of Ca2+ spikes as a function of current amplitude. C Representative dendritic Ca2+ spike traces (left). Single Ca2+ spike in the box at left on expanded time scale used for analyses in D-G (right). Black arrows show the estimated threshold. D Dendritic Ca2+ spike amplitude in wild type and fmr1 KO neurons. E Maximum rate of rise in wild type and fmr1 KO neurons. F Minimum rate of decay in wild type and fmr1 KO neurons. G Estimated Ca2+ spike threshold in wild type and fmr1 KO neurons.

136

Block of inwardly rectifying K+ channels rescues TA-LTP in fmr1 KO neurons

Our results thus far suggest that there are no differences in NMDARs at TA synapses, dendritic complex spikes, or dendritic Ca2+ spikes between wild type and fmr1 KO CA1 pyramidal neurons.

Furthermore, despite higher dendritic expression, block of Ih in fmr1 KO neurons does not rescue

+ TA-LTP. Inwardly rectifying K (KIR) channels are open at or near the resting membrane potential, expressed in CA1 pyramidal neuron dendrites and contribute to the induction of hippocampal LTP

(Hoffman et al., 1997; Chen and Johnston, 2004; Malik and Johnston, 2017). In rats, KIR channels in the dendrites constrain dendritic nonlinear events and control LTP (Malik and Johnston, 2017).

To test if KIR channels contribute to the lack of TA-LTP in fmr1 KO neurons, we performed TBS

2+ TA-LTP experiments in the presence of 25 µM Ba to block KIRs (Malik and Johnston, 2017)

(Figure 5.5A). Block of KIR rescued TA-LTP in fmr1 KO CA1 pyramidal neurons (Figure 5.5B-

C; wild type: n=5 from 4 mice; pre-TBS: 0.16 ± 0.029 mV/ms, post-TBS: 0.37 ± 0.98 mV/ms.

Wilcoxon: W=15, p=0.03, h2=0.81; fmr1 KO: n=6 from 3 mice; pre-TBS: 0.11 ± 0.015 mV/ms, post-TBS: 0.31 ± 0.05 mV/ms. Wilcoxon: W=21, p=0.03, h2=0.81). The area under the curve during the induction protocol was not different between wild type and fmr1 KO CA1 pyramidal neurons in the presence of 25 µM Ba2+ (Figure 5.5D-E; wild type: 21.89 ± 8.17 mV*ms, fmr1 KO:

9.78 ± 1.49 mV*ms. Mann-Whitney: U=11, p=0.54).

137

Figure 5.5: Low Ba2+ application rescues TA-LTP in fmr1 KO CA1 pyramidal neurons. A Recording configuration during low Ba2+ TA-LTP experiments. B EPSP slope is significantly increased 30 minutes after TBS in both wild type and fmr1 KO CA1 neurons. C Normalized change in EPSP slope for wild type and fmr1 KO CA1 neurons in the presence of 25 µM extracellular Ba2+. Inset: baseline and post-TBS EPSP traces from wild type and fmr1 KO neurons. D Traces showing TBS of wild type and fmr1 KO neurons during Ba2+ wash-on. E Graph of area under the curve during TBS with Ba2+ included in the bath saline.

138

The functional expression of KIR is not different in wild type and fmr1 KO CA1 neurons

One potential explanation for the rescue of TA-LTP by Ba2+ is that there is a higher dendritic expression of KIRs in fmr1 KO CA1 pyramidal neurons. To test if dendritic KIRs were different between wild type and fmr1 KO CA1 neurons, we measured the dendritic resting

2+ membrane potential (Vm) and input resistance (RN) before and after application of 25 µM Ba

2+ (Figure 5.6A). Extracellular Ba depolarized VM in both wild type and fmr1 KO dendrites (Figure

5.6B; wild type: n=7 from 3 mice; pre-Ba2+: -59.29 ± 2.53 mV, post-Ba2+: -55.29 ± 2.47 mV.

Wilcoxon: W=28, p=0.016, h2=0.79; fmr1 KO: n=7 from 2 mice; pre-Ba2+: -59.86 ± 1.68 mV,

2+ 2 2+ post-Ba : -54.00 ± 1.6 mV. Wilcoxon: W=28, p=0.016, h =0.79). The effect of Ba on VM was however, not significantly different between wild type and fmr1 KO neurons (Figure 5.6C; wild type: 4.00 ± 1.02 mV, fmr1 KO: 5.86 ± 0.94 mV. Mann-Whitney: U=13.5, p=0.16). Extracellular

2+ Ba also increased RN in both wild type and fmr1 KO dendrites (Figure 5.6D-E; wild type: n=7 from 3 mice; 2-way RM ANOVA: F(1,5)=11.81, p=0.019, h2=0.12; fmr1 KO: n=7 from 2 mice;

2 2-way RM ANOVA: F(1,6)=10.13, p=0.019, h =0.1). As for VM, the change in RN was not significantly different between wild type and fmr1 KO dendrites (Figure 5.6F; 2-way RM

ANOVA: F(1,11)=0.005, p=0.95). These results demonstrate that, although block of KIRs by extracellular Ba2+ rescued TA-LTP in fmr1 KO neurons, the functional expression of dendritic

KIRs is not different between wild type and fmr1 KO CA1 pyramidal neurons.

139

Figure 5.6: Functional measures of KIR activity show no differences between wild type and fmr1 KO CA1 pyramidal dendrites. A Recording configuration (left) and representative voltage traces 2+ 2+ before and after application of 25 µM Ba . B Ba significantly depolarizes VM in both wild type 2+ 2+ and fmr1 KO neurons. C The effect of Ba on VM in wild type and fmr1 KO neurons. D Ba 2+ increases RN in wild type dendrites. E Ba increases RN in fmr1 KO dendrites. F The effect of 2+ Ba on RN in wild type and fmr1 KO neurons.

140

Dendritic depolarization rescues TA-LTP in fmr1 KO CA1 pyramidal neurons

Block of KIRs depolarized dendritic Vm, increased dendritic RN and rescued TA-LTP.

Although the effect on KIRs was not different between wild type and fmr1 KO dendrites, it is unclear if Ba2+ rescued TA-LTP in fmr1 KO neurons by affecting other classes of K+ channels. In rat CA1 neurons, direct manipulation of dendritic Vm was able to reproduce the effects of KIRs on dendritic function (Malik and Johnston, 2017). To test whether the depolarization would mimic

2+ the effects of Ba on TA-LTP, we depolarized the dendritic Vm by 10 mV during the delivery of

TBS using steady state current injection to mimic the depolarization of oblique dendrites (Golding et al., 2005) during Ba2+ wash-on (Figure 5.7A). Similar to extracellular Ba2+, dendritic depolarization rescued TA-LTP in fmr1 KO neurons (Figure 5.7B-C; wild type: n=6 from 2 mice; pre-TBS: 0.36 ± 0.048 mV/ms, post-TBS: 1.21 ± 0.16 mV/ms. Wilcoxon: W=21, p=0.03, h2=0.4; fmr1 KO: n=7 from 4 mice; pre-TBS: 0.55 ± 0.21 mV/ms, post-TBS: 2.22 ± 0.61 mV/ms.

Wilcoxon: W=24, p=0.047, h2=0.52). The area under the curve during the induction protocol was not different between wild type and fmr1 KO CA1 pyramidal neurons (Figure 5.7D-E; wild type:

8.65 ± 1.08 mV*ms, fmr1 KO: 15.22 ± 3.6 mV/ms. Mann-Whitney: U=15, p=0.45). Unlike the

2+ Ba experiments above, however, dendritic Vm was only depolarized during delivery of the TBS.

Taken together with Ba2+ wash-on experiments, these results suggest that fmr1 KO CA1 dendrites are unable to reach the threshold for TA-LTP induction under normal conditions.

141

Figure 5.7: Dendritic depolarization rescues TA-LTP in fmr1 KO CA1 neurons. A Recording configuration. Current injection produced a 10 mV depolarization during TBS. B, EPSP slope is increased 30 minutes after TBS in both wild type and fmr1 KO CA1 neurons. C Normalized EPSP slope 5 minutes before and 30 minutes after TBS of TA inputs. Inset: Representative baseline and post-TBS EPSPs from wild type and fmr1 KO CA1 neurons. D Graph of area under the curve during TBS while a steady depolarizing current of 10 mV was applied to the dendritic patch.

142

Associative pairing of TA and SC inputs results in normal TA-LTP in fmr1 KO CA1 neurons

TBS during associative pairing of TA and SC inputs has been shown to elicit dendritic plateau potentials, a dendritic signal that results in prolonged depolarization of the dendritic arbor of CA1 pyramidal neurons and reliably induces LTP at TA synapses (Takahashi and Magee, 2009).

Because dendritic depolarization during TBS was an effective method of TA LTP rescue in fmr1

KO, we hypothesized that increased depolarization during LTP induction provided by associative

SC stimulation would likely have a similar effect. We simultaneously stimulated TA and SC pathways using a TBS LTP induction protocol (Figure 5.8A; (Takahashi and Magee, 2009). TA-

LTP was reliably observed in both wild type and fmr1 KO CA1 pyramidal neurons (Figure 5.8B-

C; wild type: n=5 from 3 animals, pre-TBS: 0.28 ± 0.063 mV/ms, post-TBS: 0.58 ± 0.14 mV/ms;

Wilcoxon: W=15, p=0.03, h2=0.81; fmr1 KO: n=6 from 4 animals, pre-TBS: 0.27 ± 0.051 mV/ms, post-TBS: 0.51 ± 0.12 mV/ms; Wilcoxon: W=19, p=0.03, h2=0.81). No difference was observed in the area of depolarization during TBS of TA and SC inputs (Figure 5.8D; wild type: 23.94 ±

3.49 mV*ms; fmr1 KO: 30.82 ± 3.85 mV*ms; Mann-Whitney: U=6, p=0.22). These results further suggest that fmr1 KO CA1 pyramidal neurons do not potentiate by failing to reach a threshold for the induction of TA-LTP.

143

Figure 5.8: Associative paring of SC and TA inputs results in normal TA-LTP in fmr1 KO CA1 pyramidal neurons. A Recording configuration of associative pairing stimulation paradigm. B EPSP slope before and after TBS in wild type and fmr1 KO neurons. C Normalized EPSP slope 5 minutes before and 30 minutes after associative TBS. Inset: Representative traces from baseline (a) and the last 5 minutes of recording (b). D Traces showing the voltage response to simultaneous TBS of SC and TA inputs. E Graph of area of depolarization during TBS in wild type and fmr1 KO CA1 pyramidal neurons.

144 dspike generation is impaired in fmr1 KO CA1 pyramidal neurons

The large scale, rapid influx of Ca2+ necessary for the generation of LTP at distal synapses in CA1 neurons is dependent on dendritic Na+ spikes (dspike) (Kim et al., 2015). Consequently, one explanation for our results thus far is that dspike generation is impaired in fmr1 KO neurons.

Figure 5.9 displays the method used for the determination of dspike threshold (Gasparini et al.,

2004). We used a double exponential current injection (τrise = 0.1 ms; τdecay = 2 ms) to mimic the time course of dendritic EPSPs and trigger dendritic Na+ spikes in wild type and fmr1 KO dendrites

(Figure 5.10A). Figure 5.10B shows dspike threshold as function of the dendritic recording distance from the soma in wild type and fmr1 KO CA1 neurons. Dspikes in fmr1 KO CA1 neurons had a more depolarized threshold (Figure 5.10C; wild type: n=7 from 5 mice; fmr1 KO: n=7 from

5 mice; wild type: -48.08 ± 1.35 mV, fmr1 KO: -41.48 ± 1.87 mV. Mann-Whitney: U=6, p=0.02, h2=0.34) and slower maximum dV/dt (Figure 5.10D; wild type: 60.07 ± 5.14 mV/ms, fmr1 KO:

44.89 ± 2.40 mV/ms. Mann-Whitney: U=3, p=0.004, h2=0.54) compared to wild type dendrites.

145

Figure 5.9: Method for the determination of dspike threshold. A Representative dspike recordings showing recorded voltage (left), 1st derivative (middle), and 2nd derivative (right). Threshold was determined as 20% of the second peak of the second derivative (arrows).

146

Figure 5.10: Dspike threshold is more depolarized in in fmr1 KO compared with wild type CA1 pyramidal neurons. A Dendritic voltage response to double exponential current injections (inset) of increasing amplitude (500 pA intervals). Thick lines show voltage traces with dspike and the corresponding current injection. B Graph of distance of dendritic recordings as a function of dspike threshold. C voltage threshold for dspikes is more depolarized in fmr1 KO CA1 neurons. D, the maximum rate of rise is significantly slower in fmr1 KO neurons.

147

Voltage-gated Na+ channels do not differ between wild type and fmr1 KO CA1 neurons

CA1 pyramidal neuron dendrites express voltage-gated Na+ channels which contribute to dspikes (Magee and Johnston, 1995a; Golding and Spruston, 1998). The difference in threshold and maximum dV/dt could be accounted for by differences in dendritic voltage-gated Na+ channels between wild type and fmr1 KO CA1 neurons. To test whether there are differences in dendritic

Na+ channel function, we measured Na+ current using somatic and dendritic (200-250 µm from the soma) cell attached voltage clamp recordings from wild type and fmr1 KO CA1 pyramidal neurons. The maximum current and voltage-dependence of Na+ channels we measured in wild type CA1 pyramidal neurons were in good agreement with those obtained from rat CA1 dendrites

(Magee and Johnston, 1995a; Gasparini and Magee, 2002).

Somatic patches showed no difference in Na+ channel I-V curve in wild type and fmr1 KO

CA1 pyramidal neurons (Figure 5.11A). We found no difference in the V1/2 of activation (wild type: n=6 from 2 animals, -22.22 ± 1.54 mV; fmr1 KO: n=7 from 2 animals, -21.47 ± 6.18 mV;

Mann-Whitney: U=20, p=0.95) or the activation slope (wild type: n=6 from 2 animals, 8.86 ± 2.02; fmr1 KO: n=7 from 2 animals, 12.1 ± 2.59; Mann-Whitney: U=12, p=0.23) of the voltage dependence of activation between wild type and fmr1 KO CA1 pyramidal neurons. V1/2 of inactivation (wild type: n=4 from 2 animals, -68.7 ± 1.5 mV; fmr1 KO: n=4 from 2 animals, -69.39

± 12.99 mV; Mann-Whitney: U=8, p=0.99) and inactivation slope (wild type: n=4 from 2 animals,

-5.88 ± 1.66 mV; fmr1 KO: n=4 from 2 animals, -9.23 ± 5.64 mV; Mann-Whitney: U=7, p=0.89), measurements of steady state inactivation of voltage gated sodium channels showed no differences between wild type and fmr1 KO CA1 neurons.

148

We found no difference in the Na+ I-V curve between wild type and fmr1 KO neurons

(Figure 5.12A). Additionally, there was no significant difference in the voltage-dependence of activation or steady-state inactivation between wild type and fmr1 KO Na+ channels (Figure

5.12B). There was no difference in either the V1/2 of activation (Figure 5.12C; wild type: n=10 from 3 mice, -19.11 ± 2.17 mV; fmr1 KO: n=8 from 3 mice, -16.2 ± 4.23 mV. Mann-Whitney:

U=33, p=0.57) or the activation slope factor (Figure 5.12D; wild type: 7.15 ± 1.08. fmr1 KO: 8.57

± 1.13, Mann-Whitney: U=29, p=0.36) between wild type and fmr1 KO CA1 pyramidal dendrites.

Similarly, steady state inactivation properties of V1/2 of inactivation (Figure 5.12E; wild type: n=6 from 2 animals, -63.47 ± 5.59. fmr1 KO: n=6 from 2 animals, -55.51 ± 1.45. Mann-Whitney:

U=11, p=0.31) and inactivation slope (Figure 5.12F; wild type: n=6 from 2 mice, -8.93 ± 1.98; fmr1 KO: n=6 from 2 mice, -4.41 ± 1.03. Mann-Whitney: U=7, p=0.09) were not different between wild type and fmr1 KO neurons. Our results suggest that the differences in Na+ channel function do not contribute to the threshold and dV/dt differences in dspikes between wild type and fmr1

KO CA1 pyramidal neurons.

149

Figure 5.11: Somatic Na+ channels are not different between wild type and fmr1 KO CA1 pyramidal neurons. A Current-voltage plot showing no significant difference in Na+ current between wild type and fmr1 KO dendrites. Inset: representative Na+ current traces in response to voltage steps to -60, -40, -20, and 0 mV. B, Na+ channel activation and inactivation are not different + between wild type and fmr1 KO CA1 dendrites. C Na channel activation V1/2 in the soma of wild type and fmr1 KO neurons. D Na+ channel slope factor in the soma of wild type and fmr1 KO + neurons. E Na channel steady state inactivation V1/2 in the soma of wild type and fmr1 KO + neurons. F Na channel slope factor in the soma of wild type and fmr1 KO neurons.

150

Figure 5.12: Dendritic Na+ channels are not different between wild type and fmr1 KO CA1 pyramidal neurons. A, Current-voltage plot showing no significant difference in Na+ current between wild type and fmr1 KO dendrites. Inset: representative Na+ current traces in response to voltage steps to -60, -40, -20, and 0 mV. B, Na+ channel activation and inactivation are not different between wild type and fmr1 KO CA1 dendrites. C Na+ channel activation V1/2 in wild type and fmr1 KO dendrites. D Na+ channel slope factor in wild type and fmr1 KO dendrites. E Na+ channel steady state inactivation V1/2 in wild type and fmr1 KO dendrites. F Na+ channel slope factor in wild type and fmr1 KO dendrites.

151

A-type K+ channel block restores dspike threshold in fmr1 KO neurons

+ A-type K currents (IKA) are smaller in fmr1 KO CA1 neurons compared to wild type

(Routh et al., 2013). We did not initially consider differences in IKA as a potential cause for the lack of TA-LTP in fmr1 KO neurons as we would expect the reduction in IKA to either promote or have no effect on dspikes. There is, however, a hyperpolarized shift in the activation of A-type K+ channels in fmr1 KO dendrites compared to wild type (Routh et al., 2013). A-type K+ channels that activate at more negative potentials could overlap with Na+ currents and influence dspikes in fmr1 KO neurons. Furthermore, low concentrations of Ba2+ have been shown to affect transient K+ currents expressed in the heart (Shi et al., 2000). We thus hypothesized that A-type K+ channels influence dspikes in fmr1 KO but not wild type CA1 neurons. As an initial test of this hypothesis, we recorded dspikes under control conditions, 25 µM Ba2+ (used in Figure 5.5), and 150 µM Ba2+, a concentration known to block A-type K+ channels (Gasparini et al., 2007; Routh et al., 2013)

(Figure 5.13A-B). Neither concentration of Ba2+ had an effect on dspike threshold in wild type neurons; however, dspike threshold was significantly hyperpolarized by 25 or 150 µM Ba2+ in fmr1 KO neurons (Figure 5.13A-B; wild type: n=7 from 3 mice; fmr1 KO: n=8 from 2 mice; 2- way RM ANOVA: F(1,13)=4.9, p=0.045, h2=0.15. Interaction: F(2, 26)=11.51, p=0.0003, h2=0.13. Sidak’s test-baseline: p=0.001, h2=0.64, wild type: -36.14 ± 1.09 mV, fmr1 KO: -29.3 ±

0.79 mV). These results support the hypothesis that A-type K+ channels contribute to the depolarized dspike threshold in fmr1 KO CA1 neurons. These data also suggest that Ba2+ rescued

TA-LTP in fmr1 KO neurons by hyperpolarizing the threshold for dspike generation. As a further test of the hypothesis that A-type K+ channels depolarize dspike threshold in fmr1 KO CA1 neurons, we measured the effect of the specific A-type K+ channel blocker AmmTx3 (500 nM)

152

(Chittajallu et al., 2020; Hu et al., 2020) on dspike threshold in wild type and fmr1 KO CA1 neuron dendrites. In agreement with our Ba2+ experiments, AmmTx3 significantly reduced dspike threshold in fmr1 KO CA1 dendrites but had no effect on wild type dspikes (Figure 5.14A-B; wild type: n=6 from 3 mice; pre-AmmTx3: -32.56 ± 0.65 mV, post-AmmTx3: -31.91 ± 1.11 mV.

Wilcoxon: W=5, p=0.69; fmr1 KO: n=6 from 3 mice; pre-AmmTx3: -25.52 ± 2.29 mV, post-

AmmTx3: -33.84 ± 1.8 mV. Wilcoxon: W= -21, p=0.03, h2=0.65).

153

Figure 5.13: Block of A-type K+ channels with Ba2+ hyperpolarizes dspike threshold in fmr1 KO but not wild type CA1 pyramidal neurons. A dspikes during baseline and application of 150 µM Ba2+ in wild type and fmr1 KO CA1 pyramidal neurons. Black arrows represent threshold measurments. B Extracellular Ba2+ effects on wild type and fmr1 KO dendrites during successive application of 25 and 150 µM Ba2+.

154

Figure 5.14: Block of A-type K+ channels with 500 nM AmmTx3 hyperpolarizes dspike threshold in fmr1 KO but not wild type CA1 pyramidal dendrites. A Representative dspike recordings from wild type and fmr1 KO neurons before and after application of 500 nM AmmTx3. Arrows indicate threshold for dspike generation. B Effect on threshold of AmmTx3 application on wild type and fmr1 KO CA1 pyramidal neuron dendrites.

155

A-type K+ channel block restores TA-LTP in fmr1 KO neurons

To test if block of A-type K+ channels rescued TA-LTP in fmr1 KO neurons, we repeated the TBS TA-LTP experiments in the presence of the A-type K+ channel blocker AmmTx3 (500 nM). Blockade of A-type K+ channels with 500 nM AmmTx3 rescued TA-LTP in fmr1 KO CA1 pyramidal neurons (Figure 5.15A-C; wild type: n=6 from 3 mice; pre-TBS: 0.13 ± 0.017 mV/ms, post-TBS: 0.23 ± 0.032 mV/ms. Wilcoxon: W=21, p=0.03, h2=0.79. fmr1 KO: n=6 from 3 mice; pre-TBS: 0.11 ± 0.027 mV/ms, post-TBS: 0.21 ± 0.048 mV/ms. Wilcoxon: W=21, p=0.03, h2=0.79). The area under the curve during TBS was not different while AmmTx3 was present in the bath (Figure 5.15D-E; wild type: 22.62 ± 3.51 mV*ms; fmr1 KO: 22.58 ± 4.65 mV*ms. Mann-

Whitney: U=16, p=0.82). Taken together, these results suggest that the hyperpolarized shift in activation of A-type K+ channels in fmr1 KO CA1 pyramidal neuron dendrites results in a depolarized threshold for dspikes and a lack of TA-LTP.

156

Figure 5.15: Block of A-type K+ channels restores TA-LTP in fmr1 KO neurons. A Recording configuration for AmmTx3 TA-LTP experiments. B EPSP slope before and after TBS in wild type and fmr1 KO CA1 neurons. C Normalized EPSP slope 5 minutes before and 30 minutes after TBS of TA inputs. Inset: Representative baseline and post-TBS EPSPs from wild type and fmr1 KO CA1 neurons. D Traces showing the somatic voltage response during TBS. E Group data of area under the curve during TBS.

157

Discussion

TBS induces behaviorally relevant LTP of TA inputs in wild type CA1 neurons that is critical to hippocampal-dependent learning and memory (Remondes and Schuman, 2004; Bittner et al., 2017). We found that TBS fails to induce TA-LTP in fmr1 KO CA1 pyramidal neurons. We previously identified an increase in the expression of dendritic h-channels in fmr1 KO CA1 dendrites (Brager et al., 2012; Brandalise et al., 2020). We thus hypothesized that decreased summation of TA inputs would account for impairments in TA-LTP. However, blocking h- channels with ZD7288 did not rescue TA-LTP in fmr1 KO neurons. We cannot rule out the possibility that h-channelopathy contributes to difficulty in synaptic stimuli summating to reach dspike threshold and cause TA-LTP induction. TA-LTP requires Ca2+ influx through NMDARs and L-type voltage-gated Ca2+ channels (Golding et al., 2002; Remondes and Schuman, 2003). 2- photon Ca2+ imaging demonstrated that synaptically evoked dendritic Ca2+ signals were smaller in fmr1 KO neurons. We found that basal TA synaptic transmission was not different between wild type and fmr1 KO mice as was described previously (Wahlstrom-Helgren and Klyachko, 2015) but see (Booker et al., 2020). Interestingly, we found no difference in NMDAR EPSPs, dendritic complex spikes, or isolated dendritic Ca2+ spikes between wild type and fmr1 KO CA1 pyramidal neurons. Dendritic Na+ mediated spikes are necessary and trigger the influx of Ca2+ necessary for

TA-LTP (Golding et al., 2002; Kim et al., 2015). We found that dspike threshold is depolarized in fmr1 KO dendrites and suggest that an inability of TA TBS to trigger dspikes likely contributes to

TA-LTP dysfunction. Indeed, depolarization of the dendritic membrane potential in fmr1 KO neurons through block of KIR channels or direct dendritic current injection restored TA-LTP. Our lab previously demonstrated that A-type K+ channel expression is reduced in the dendrites of fmr1

158

KO CA1 pyramidal neurons and the activation is shifted to more hyperpolarized potentials (Routh et al., 2013). Despite reduction in A-type K+ channel expression, the shift in activation resulted in a depolarized dspike threshold in fmr1 KO CA1 dendrites, illustrated by the rescue of TA-LTP and dspike threshold in fmr1 KO neurons by the addition of the A-type K+ channel blocker

AmmTx3.

Dendritic nonlinear events in fmr1 KO CA1 neurons

We provide the first direct comparison of dendritic nonlinear events (complex spikes, Ca2+ spikes, and dendritic Na+ spikes) between wild type and fmr1 KO CA1 neurons. Fast, Na+- dependent dspikes are essential for the induction TA-LTP (Kim et al., 2015). We found that the threshold was more depolarized and the maximum dV/dt slower for dspikes in fmr1 KO dendrites compared to wild type. Dendritic complex spikes, which are the combined effect of dendritic Na+ and Ca2+ channels, were not different in frequency or width between wild type and fmr1 KO CA1 pyramidal neurons. Recordings of isolated, dendritic Ca2+ spikes suggest that CA1 dendritic voltage-gated Ca2+ channels are not different between wild type and fmr1 KO neurons. Previous studies have shown that changes in voltage-gated Ca2+ channels in fmr1 KO mice occur in a brain region and cell type specific manner (Meredith et al., 2007; Danesi et al., 2018; Gray et al., 2019).

Our results further illustrate that changes in ion channel function and expression in FXS are not conserved across brain regions.

A-type K+ channels contribute to dspikes in fmr1 KO but not wild type CA1 neurons

The threshold for regenerative events is determined by the complement of ion channels and their relative kinetics (Hodgkin and Huxley, 1952a; 1952b; Stafstrom et al., 1984). In particular, there is an interplay between depolarizing sodium conductances and hyperpolarizing potassium

159 conductances. We provide the first investigation of dendritic voltage-gated Na+ channels in CA1 dendrites of fmr1 KO mice. The properties of voltage-gated Na+ currents in our wild type mouse dendritic recordings were comparable to previous findings in rat (Magee and Johnston, 1995a;

Gasparini and Magee, 2002), and were not different between wild type and fmr1 KO CA1 pyramidal neurons.

It was previously found that the activation of A-type K+ channels is hyperpolarized in fmr1

KO CA1 neuron dendrites (Routh et al., 2013). We hypothesize that the hyperpolarized shift in

+ + IKA activation in fmr1 KO dendrites allows A-type K channels to provide more opposition to Na - dependent depolarization and depolarizes dspike threshold. Similar modulation of action potential threshold is known to occur via somatic D-type K+ channels (Higgs and Spain, 2011; Kalmbach et al., 2015; Ordemann et al., 2019).

This hypothesis is supported by our results showing that direct depolarization of dendrites during TBS was sufficient to rescue TA-LTP in fmr1 KO CA1 neurons. Furthermore, both 25 and

150 µM extracellular Ba2+ hyperpolarized dspike threshold in fmr1 KO but had no effect in wild type CA1 neurons. We confirmed that block of IKA hyperpolarized dspike threshold and rescued

TA-LTP with the specific A-type K+ channel blocker AmmTx3.

Potential mechanism for the hyperpolarized activation of A-type K+ channels

The voltage-dependence of A-type K+ channels in CA1 pyramidal dendrites is modulated by multiple signaling cascades. Stimulation of protein kinase C (PKC), by activation of metabotropic glutamate and muscarinic acetylcholine receptors, and cAMP-dependent protein kinase A (PKA), by dopaminergic and β-adrenergic receptors, produce a depolarizing shift in the of activation of A-type K+ channels (Hoffman and Johnston, 1998; 1999).

160

Reduction in PKC and/or PKA activity could account for the hyperpolarization of A-type K+ channel activation in fmr1 KO CA1 neurons. There is evidence from human and animal models of

FXS exhibiting lower cAMP levels (Berry-Kravis and Huttenlocher, 1992; Choi et al., 2015).

Lower cAMP levels could reduce basal PKA activity and lead to a hyperpolarized shift of A-type

K+ channel activation. A study on cortical synaptoneurosomes found that basal PKC activity was not different between wild type and fmr1 KO mice (Weiler et al., 2004). A more recent study, however, found that PKCε expression was lower in the hippocampus of fmr1 KO mice (Marsillo et al., 2020). PKCε is abundant in the nervous system and activated by G-protein coupled receptors

(Akita, 2002). Furthermore, PKCε is activated by phorbol esters, which were used to investigate

PKC-dependent modulation of A-type K+ channels (Hoffman and Johnston, 1998). Thus, PKCε is a potential candidate for regulating the voltage-dependence of A-type K+ channel activation.

Additionally, FMRP is positive regulator of both PKA (Sears et al., 2019) and PKC (Zhao et al.,

2015); therefore, loss of FMRP in FXS could result in reduced PKA and PKC activity. Thus, changes in the basal activity of PKA and/or PKC could contribute to the hyperpolarized shift in

A-type K+ channel activation that alters dspike threshold and impairs TA-LTP in fmr1 KO CA1 neurons.

Consequences for hippocampal learning and memory

There are two major changes in A-type K+ channel function in fmr1 KO CA1 dendrites: lower maximum current and hyperpolarized activation. LTP induction of excitatory inputs to CA1 pyramidal neurons requires active dendritic events. Schaffer collateral LTP requires backpropagating action potentials to provide the depolarization necessary for NMDA receptor activation (Magee and Johnston, 1997). In the SLM region of CA1 however, where the TA inputs

161 synapse onto the distal dendrites of CA1 pyramidal neurons, backpropagating action potentials are unreliable and often fail to propagate into the distal apical dendrites. Rather, it is locally generated dspikes that provide the necessary depolarizing signal for the induction of LTP. Alterations in dendritic A-type K+ channel function in fmr1 KO CA1 pyramidal neurons causes changes to the active events affecting LTP of both excitatory pathways. Reduced maximal A-type K+ current allows the amplitude of back propagating action potentials to be larger which results in a reduced threshold for Schaffer collateral LTP (Routh et al., 2013). By contrast, the hyperpolarized shift in the activation of A-type K+ channels provides a small, but effective, hyperpolarizing influence on the dendritic membrane potential that depolarizes dspike threshold and impairs TA-LTP. The coordinated plasticity of TA inputs, which convey current information about the external environment, and Schaffer collateral inputs, which convey stored information from prior experiences, is critical for hippocampal dependent memory tasks. The combined changes in A- type K+ channel function alter the dendritic processing of these two critical excitatory pathways.

Our findings show an inability to induce LTP in TA synapses, in a pathway critical for the induction of long-term memory (Remondes and Schuman, 2004). This finding provides critical new information for the understanding of the FXS disease phenotype.

FXS is marked by deficits in learning and memory, making the discovery of specific therapeutic targets necessary for the understanding and future treatment of the disorder. Our previous work, along with other studies, showed how changes in the expression of A-type K+ channels affect

Schaffer collateral LTP in FXS (Gross et al., 2011; Lee et al., 2011; Routh et al., 2013). Here, we show that changes in the properties of A-type K+ channels impact TA-LTP. Therefore, while A-

162 type K+ channels represent a potential therapeutic target for FXS interventions, manipulations must take into account both the change in expression and the shift in gating.

163

Chapter 6: Discussion

All or nothing, regenerative events shape every aspect of neuronal communication. The ability for neurons to produce active events that propagate over large distances is tightly controlled by the complement of ion channels expressed in a neuron. As such, small change in ion channel expression can have a substantial impact on the function of neurons and neuron circuits. The primary determinants of action potential threshold are the expression and biophysical properties of voltage gated Na+ and K+ channels. It is the balance between these two opposing forces that dictates the membrane voltage at which a regenerative event will occur. This dissertation describes in detail examples of ion channel expression or function leading to differences voltage threshold, highlighting the delicate balance required in neuronal systems to produce the appropriate output for a given neuron type.

The investigation of CA1 pyramidal neurons in the dorsal and ventral regions of mouse hippocampus presented here revealed similar subthreshold differences in membrane potential and input resistance consistent with previous findings in mouse and rat hippocampus (Dougherty et al., 2012; Kim and Johnston, 2015; Malik et al., 2015; Milior et al., 2016). However, unlike in rat hippocampus, the firing rate was not different between dorsal and ventral CA1 pyramidal neurons.

In rat hippocampus dorsal neurons fire significantly fewer action potentials compared with ventral neurons due to reduced input resistance caused by greater expression of h-channels and GIRK channels (Dougherty et al., 2013; Kim and Johnston, 2015; Arnold et al., 2019). The lack of difference in firing rate is the result of a differential expression of D-type K+ channels, causing the action potential threshold to be more depolarized in ventral neurons resulting in decreased

164 excitability and a normalization of action potential firing between dorsal and ventral mouse hippocampus.

This dissertation shows that the functional physiological properties of dorsal and ventral hippocampus are not the same between mouse and rat hippocampus. Interestingly, research in the hippocampal field shows that both mouse and rat hippocampus largely yield the same behavioral outcomes despite differences in action potential output. This phenomenon is particularly interesting in light of the study of disease phenotypes that rely on our understanding of normal properties of neurons and neuron circuits to understand how disruptions in function result in disease phenotype. The tremendous flexibility in the mechanisms used by neurons to accomplish the same end goal is incredible.

The difference in individual neuron function that results in the same overall circuit function illustrates the concept of convergent mechanisms. This idea is centered around the ability of structures of the brain to accomplish the same function through entirely different mechanisms.

This raises the possibility that, even within a single neuron, there are multiple backup mechanisms to ensure proper function in the event that a primary mechanism is disrupted.

The findings of this dissertation lead to the conclusion that there are no differences in the intrinsic physiological function of dorsal and ventral CA1 pyramidal neurons between wild type and fmr1 KO mice. Although previous research has shown increased action potential firing in CA1 pyramidal neurons in fmr1 KO compared with wild type neurons, this is the first evidence comparing dorsal and ventral neurons of adult mice with a C57/B6 genetic background.

LTP of the TA pathway is impaired in fmr1 KO compared with wild type CA1 pyramidal neurons. A hyperpolarizing shift in the activation of A-type K+ channels causes their activation to

165 overlap with the activation of dendritic voltage gated Na+ channels. The result of this interaction is a depolarized threshold for the generation of dspikes. During TBS the generation of dspikes is necessary for the induction of TA-LTP. In fmr1 KO CA1 pyramidal neurons high frequency synaptic stimulation is insufficient for the generation of dspikes and thus does not trigger the postsynaptic depolarization necessary for the influx of calcium through NMDA receptors and voltage gated Ca2+ channels that then trigger the expression of TA-LTP. A particularly interesting aspect of these findings lies in the dual role of A-type K+ channel dysfunction in fmr1 KO CA1 pyramidal neurons. Altered A-type K+ channel function is responsible for changes in LTP in both the SC and

TA pathways, the two major excitatory synaptic connections onto CA1 pyramidal neurons.

Furthermore, changes in A-type K+ channel function have differential effects at each of these synapses, hyperpolarizing the threshold for LTP induction at the SC synapse and depolarizing the threshold for LTP induction at the TA synapse.

CA1 pyramidal neurons are hypothesized to act as a comparator between current sensory experience (TA pathway) and previous experience (SC pathway), suggesting that circuit coordination is critical to the development of place fields. Indeed, in vivo patch clamp recordings of mice running on a track showed that plateau potentials, events that require coordinated TA and

SC input, reliably induce place fields that are stable over multiple trials (Takahashi and Magee,

2009; Bittner et al., 2017). It is likely that the increased threshold for dspike generation will impair the transmission of synaptic impulses between the distal dendrites of CA1 pyramidal neurons to the proximal synaptic inputs and the soma, disrupting the coordinated input necessary for proper circuit function.

166

Although the cellular data present here might suggest robust deficits in overall hippocampal function, relatively few studies have identified reliable hippocampal dysfunction in mouse models of FXS. Surprisingly, recent studies on CA1 place fields in fmr1 KO mice showed normal CA1 place field development. In fact, it is argued that the establishment of place fields occurs with less training and is associated with increased SC pathway plasticity (Talbot et al.,

2018). This finding is in agreement with investigations into LTP in the excitatory input pathways onto CA1 pyramidal neurons. First, the SC pathway displays a reduced threshold for the induction of LTP in fmr1 KO animals. Second, associative pairing of TA and SC inputs resulted in normal

TA-LTP. While these findings would initially suggest normal hippocampal function during spatial tasks, impairments in fmr1 KO hippocampus dependent tasks arise when mice are confronted with a change in established rules. A task measuring cognitive flexibility in relation to spatial learning involves placing mice on a rotating disc in which a certain region is associated with an electric shock. Both wild type and fmr1 KO mice quickly learn to avoid the shock zone in the task, forming normal place fields in the process; however, when the shock zone is moved to a new region of the disc, wild type animals are able to consistently adapt to the new rules of the task while fmr1 KO mice have significant difficulty with the adjustment (Radwan et al., 2016; Talbot et al., 2018). The difficulty in adjusting to new rules surrounding a task is referred to as cognitive inflexibility.

During this task it was observed that, despite normal place cell formation, fmr1 KO mice showed decreased coordinated CA1 discharge compared with wild type neurons. It is possible that the depolarized threshold for TA-LTP may well explain the reduced CA1 circuit discharge coordination observed during mouse behavior (Talbot et al., 2018).

167

In summary, the decreased threshold for SC-LTP may result in the initial establishment of normal place fields during experiential learning. However, when the rules shift, the inability to integrate new sensory information being transmitted through the TA pathway with more stable, learned information from the SC pathway could result in the cognitive inflexibility in fmr1 KO animals and underlie deficits in memory observed in patients with FXS.

While this hypothesis requires validation through large scale circuit level investigation and behavioral testing, it is a promising pathway for investigation. Under this hypothesis, the correction of cognitive inflexibility would require the treatment of a single dysfunctional channel to restore normal function. Furthermore, correction would be possible in patients with FXS not only before and during development, but also in adults with FXS.

This dissertation is rooted in the importance of understanding the function of single neurons and neuron circuits in elucidating how brain structures gather and transmit the information that results in downstream behavior. The understanding of both normal system function and dysfunction brought about by disease states can bring clarity to the important aspects that bring about normal behavioral function in our interactions with our environments.

168

Appendix: Methods

Animals

The University of Texas at Austin Institutional Animal Care and Use Committee approved all animal procedures. Male wild type and fmr1 KO C57/B6 mice from 8-16 weeks were used from a colony for all experiments. fmr1 KO male and homozygous female fmr1 KO mice were paired to produce litters of fmr1 KO animals. Mice were weaned at P20. Animals were housed in single sex groups at room temperature with ad libitum access to food and water and set on a reverse 12- hour light cycle in the University of Texas at Austin vivarium located in the Norman Hackerman

Building.

Preparation of acute brain slices

Mice were anesthetized with acute isofluorane exposure followed by injection of ketamine/xylazine cocktail (100/10 mg/kg i.p.). Mice were then perfused through the heart with ice-cold saline consisting of (in mM): 2.5 KCl, 1.25 NaH2PO4, 25 NaHCO3, 0.5 CaCl2, 7 MgCl2,

+ 7 dextrose, 205 sucrose, 1.3 ascorbate and 3 Na pyruvate (bubbled with 95% O2/5% CO2 to maintain pH at ~7.4). The brain was removed, trimmed, and sectioned into 300 µm thick transverse slices of dorsal, ventral, or middle hippocampus using a vibrating tissue slicer (Vibratome 3000,

Vibratome Inc.) (Brager et al., 2012; Brandalise et al., 2020). Slices were held for 30 minutes at

35°C in a chamber filled with artificial cerebral spinal fluid (aCSF) consisting of (in mM): 125

+ NaCl, 2.5 KCl, 1.25 NaH2PO4, 25 NaHCO3, 2 CaCl2, 2 MgCl2, 10 dextrose and 3 Na pyruvate

(bubbled with 95% O2/5% CO2) and then at room temperature until the time of recording.

169

Electrophysiology

Slices were submerged in a heated (32–34°C) recording chamber and continually perfused

(1−2 mL/minute) with bubbled aCSF containing (in mM): 125 NaCl, 3.0 KCl, 1.25 NaH2PO4, 25

+ NaHCO3, 2 CaCl2, 1 MgCl2, 10 dextrose, 3 Na pyruvate, 0.005 CGP and 0.002 gabazine. In experiments comparing dorsal and ventral CA1 pyramidal neurons 0.025 mM AP5 and 0.02 mM

DNQX were bath applied.

Whole-cell current clamp recordings

Patch pipettes (4−8 MΩ somatic, 7–11 MΩ dendritic) were pulled from borosilicate glass and filled with (in mM): 120 K-Gluconate, 16 KCl, 10 HEPES, 8 NaCl, 7 K2 phosphocreatine, 0.3

Na−GTP, 4 Mg−ATP (pH 7.3 with KOH). Neurobiotin (Vector Laboratories; 2%) was included in the internal recording solution to determine the recording location during post-hoc morphological reconstruction. Neurons that had a significant portion of the oblique or apical dendrites cut were excluded from analysis. In some cases, Alexa 594 was used to provide real time feedback of dendritic morphology.

Data were acquired using a Dagan BVC−700 amplifier and Axograph (Axograph X) or custom data acquisition software written using Igor Pro (Wavemetrics). Data were sampled at

10−50 kHz, filtered at 3-5 kHz and digitized using an ITC-18 interface (InstruTech). Pipette capacitance and series resistance were monitored and adjusted throughout each recording. Series resistance was monitored throughout each experiment and the experiment discarded if series resistance exceeded 30 MΩ (50 MΩ for dendritic recordings). Experiments in which the resting

170 membrane potential was more depolarized than -50 mV were discarded. The liquid junction potential was estimated to be 14.3 mV (Patcher’s Power tools IGOR Pro) and was not corrected for.

Extracellular stimulation was performed using bipolar sharp tungsten electrodes (5 MΩ, 8° taper, A-M Systems) connected to a Neurolog NL800A current stimulus isolator (Digitimer). To reduce recurrent excitation, a cut was made between area CA3 and area CA1. CA1 pyramidal neuron dendrites or somata were visually identified using DIC or Dodt contrast optics. The temporoammonic inputs were targeted by visually locating the axon fibers in the SLM region

(³250 µm from CA1 stratum pyramidale) and lowering the tungsten electrode until the tip was approximately 10 µm below the surface of the tissue. Stimulation intensity was increased until reliable EPSP (1-2 mV for somatic recordings and 2-4 mV for dendritic recordings) was elicited.

To isolate NMDAR dependent EPSPs MgCl2 was removed from the extracellular aCSF and 20

µM DNQX added to block AMPA receptors.

Induction of long-term potentiation

Long-term potentiation was performed using theta burst stimulation as previously described (Tsay et al., 2007). Baseline EPSPs were stimulated at 0.067 Hz for 5 minutes. LTP was induced using TBS with bursts of 10 stimuli at 100 Hz, performed in 5 trains at 5 Hz, and each set of TBS was repeated 4 times at 20 second intervals. EPSPs were then recorded for approximately

30 minutes post TBS. The change in EPSP slope was plotted normalized to the baseline period.

171

Analysis of current clamp data

Data were analyzed using Axograph (Axograph X) or custom analysis software written in

Igor Pro (Wavemetrics). EPSP summation was determined from synaptic stimulation or from α-

-αt current injections where I=Imax(t/α)e and quantified as the ratio of the amplitude of the fifth EPSP to the amplitude of the first EPSP. Paired pulse ratio was calculated as the slope of the second

EPSP divided by the slope of the first EPSP. Input-output was measured by increasing the magnitude of stimulus and measuring the slope of the resulting EPSP. For a given input-output experiment the same stimulation electrode was used throughout to reduce variability within the experiment (but note that stimulation amplitudes differed across the experiments).

RN was calculated from the linear portion of the current−voltage relationship generated in response to a family of current injections (−50 to +50 pA, 10 pA steps). Voltage sag was measured by calculating the ratio of the maximum RN to the steady state RN. Rebound slope was measured by plotting the amplitude of the voltage rebound at the offset of a current step as a function of steady state voltage during the current step and calculated by fitting a linear regression to the data points. Resonant frequency (fR) was determined as the frequency where the peak of the impedance amplitude profile occurred in response to a constant amplitude sinusoidal current injection that linearly increased in frequency from 1−15 Hz over 15 seconds (Narayanan and Johnston, 2007).

The membrane time constant was measured by delivering 1 ms current injections (+/- 100 pA) and fitting a double exponential to the decay of the voltage, the slower tau value was used to estimate membrane time constant (Routh et al., 2009). Action potential firing was measured by delivering

500 or 1000 ms current steps of increasing amplitude from 50 to 450 pA in 50 pA intervals.

Somatic action potential threshold was calculated as the membrane voltage at which dV/dt reached

172

20 mV/ms. Threshold was compared between dorsal and ventral CA1 neurons using two methods.

The first was determined as in (Dougherty et al., 2012), in which threshold of the first spike in a train that occurred approximately 50 ms after the onset of the current step was compared between dorsal and ventral neurons. The second method compared threshold using current steps of varying duration (1.5, 3, 6, 12, 24, 50, and 100 ms) adjusting the current amplitude to elicit a single action potential at the end of the current step (Higgs and Spain, 2011). Action potential amplitude was measured from baseline Vm to the peak of the action potential. Action potential duration was measured as the width at the half maximal amplitude. 50 µM 4-AP was bath applied to block D- type potassium channels in experiments measuring voltage threshold.

Complex spikes were elicited by delivering square current injections (50-450 pA for 1 second). The width of a complex spike was calculated differentially depending on the number of

Ca2+ dependent events following the initial fast spike. For one slow spike the width was calculated as the half-width from the initiation point of the spike to the peak of the event. When two or three

Ca2+ dependent spikes occurred, the width was calculated using methods for the measurement of dendritic Ca2+ plateau width modified from (Takahashi and Magee, 2009). For these events, width was taken as the halfway point between the initiation point of the first slow event and the lowest trough between events.

Dendritic Ca2+ spikes were isolated with the addition of 0.5 µM TTX and 50 µM 4-AP.

Spikes were generated by injecting square current pulses 1000 ms long and varying between 50 and 700 pA with 50 pA intervals. Because Ca2+ dependent spikes have a much slower time course compared with Na+ dependent spikes, measurements that utilized the second derivative of the voltage response are not reliable. The threshold of somatically recorded action potentials occurs at

173 approximately 8% of the maximum dV/dt. We therefore estimated the Ca2+ spike threshold using the time when dV/dt was 8% of maximum. Amplitude was measured as the difference from baseline membrane potential to the peak of the first Ca2+ spike.

Dendritic Na+ spike experiments were performed on dendritic recordings between 200 and

300 µm from the soma. Spikes were generated by injecting a series of double exponential currents

(t1 = 0.1 ms t2 = 2 ms) ranging in amplitude between 500 pA and 5000 pA at 100 or 500 pA intervals. The threshold for dspikes was calculated as 20% of the second peak of the second derivative of the voltage response (Gasparini et al., 2004).

2-photon Ca2+ imaging

Ca2+ imaging experiments are performed on a Prairie Ultima 2-photon imaging system

(now Bruker) arranged for in vitro patch clamp recording. An ultra fast, pulsed laser beam (Spectra

Physics: MaiTai) was used at 920 nm for imaging. Recording pipettes were filled with OGB-1

(Invitrogen, 100 µM) and Alexa 594 (Invitrogen, 40 µM). Line scans across the distal dendrites were performed at 500 Hz with a dwell time of 4 µs for between 400 and 1200 ms. Imaging location was chosen approximately 50 μm more proximal to the soma in reference to the extracellular stimulating electrode.

Analysis of line scans were performed using ImageJ (NIH) by placing a line through the fluorescent signal and plotting the profile before converting to a number array(Schneider et al.,

2012). Changes in Ca2+ were quantified using DF/F, where F is the baseline fluorescence prior to stimulation and DF is the change in fluorescence during neuronal stimulation. Ca2+ signal traces were smoothed using a Savitsky-Golay function (IGOR pro, Wavemetrics).

174

Voltage Clamp Recordings of K+ current

Outside out voltage clamp recordings were performed using an Axopatch 200B amplifier

(Molecular Devices). Data were acquired at 10 kHz and filtered at 2 kHz and then digitized using an ITC-18 interface (InstruTech) and recorded using Axograph X software (Axograph). Separation of fast-inactivating, slow-inactivating, and sustained K+ current was performed as in (Kalmbach et al., 2015). IK was measured by applying 800 ms voltage steps ranging to -70 to 50 mV at 20 mV intervals from a holding potential of -90 mV. A 100-ms pre-step to -20 mV was used to remove the rapidly inactivating component (Kalmbach et al., 2015). Isolation of the sustained component was achieved by using the same range of voltage commands but from a holding potential of -20 mV. Offline subtraction of isolated components from IK allowed for the separation of the fast, slow and sustained K+ currents (Kalmbach et al., 2015; Routh et al., 2017). Activation curves were fit with a single Boltzmann function using a least squares program (Kalmbach et al., 2015). Linear leakage and capacitive currents were digitally subtracted by scaling traces at smaller voltages in which not voltage dependent current was activated.

Voltage Clamp Recordings of Na+ current

Cell attached voltage clamp recordings were performed using an Axopatch 200B amplifier

(Molecular Devices). Data were acquired at 50 kHz and filtered at 2 kHz and then digitized using an ITC-18 interface (InstruTech) and recorded using custom Igor Pro software (Igor Pro 7,

Wavemetrics). Pipette solution for cell attached Na+ channel recordings consisted of (in mM): 140

+ NaCl, 2.5 KCl, 2 CaCl2, 1 MgCl2, 10 HEPES, 10 TEA, 1 3,4 D-AP, 1 4-AP. Na currents were

175 elicited using depolarizing voltage commands from -70 to 20 mV in 10 mV steps from a holding potential of -90 mV. Steady state inactivation was measured using depolarizing test pulses to a fixed potential (0 mV) preceded by a series of prepulse conditioning potentials ranging from -100 to -20 mV in 10 mV increments. Linear leakage and capacitive currents were digitally subtracted by scaling traces at smaller voltages in which not voltage dependent current was activated.

Activation data were plotted as normalized conductance and steady-state inactivation as normalized current. Activation and inactivation data were fit to a single Boltzmann function using a least-squares program.

DAB reaction and cellular reconstruction

Cells were filled with Neurobiotin (Vector Laboratories, Burlingame, CA) during current clamp experiments and at the end of the recording, slices were fixed in 3% glutaraldehyde for a minimum of 24 hours. Slices were washed in 0.1 M phosphate buffer (PB) and incubated in 0.5%

H2O2 for 30 minutes. Slices were then washed in PB and incubated in ABC reagent (Vector

Laboratories, Burlingame, CA) containing Avidin DH and biotinylated horseradish peroxidase H for 24-48 hours at 4°C. Slices were then incubated in DAB solution (Vector Laboratories,

Burlingame, CA) in the presence of H2O2 causing a visible color change to the slices and

Neurobiotin filled cell. Slices were dehydrated in glycerol and mounted on glass slides for imaging. Identifiable neurons from both dorsal and ventral slices were reconstructed at 40X magnification using a Leitz Diaplan microscope with Neurolucida 6.0 software (MicroBrightField,

Inc., Williston, VT). Distance from soma was scaled so that there were 10 and 25 concentric circles measuring the basal dendrites and apical dendrites respectively.

176

Dorsal-ventral mapping

We used a predictive algorithm, developed in (Malik et al., 2015), to predict the location of slices used for electrophysiological recordings along the dorsal-ventral axis of the hippocampus.

Anatomical markers of the hippocampal formation in an individual slice are compared to normal hippocampal measurements in order to ensure that recorded cells were from slices in either the dorsal or ventral pole of the hippocampus. Anatomical indicators were: the ratio of the transverse length of area CA1 to the radial length of area SLM, the ratio of the transverse length of area CA3 to the radial length of area CA3, and the transverse length of dentate gyrus to the distance between the horns of dentate gyrus. By using coefficients determined from the measurement of 100 hippocampal slices across the dorsal-ventral axis and anatomical measurement ratios taken from each slice, we estimated the location of slices that were recorded from within the mouse hippocampus.

Experimental Design and Statistical Analysis

The use of male wild type and fmr1 KO mice (Chapters 4 and 5) or male wild type dorsal and ventral mice (Chapters 2 and 3) was interleaved during each set of experiments. Where possible, experimenters were blind to the condition of the animal during experimentation and analysis. Experiments were designed as a comparison between wild type and fmr1 KO or dorsal and ventral mice. The Shapiro-Wilk test was used to assess whether data were normally distributed. Parametric data were analyzed using Student’s t-test and non-parametric data were analyzed using the Mann-Whitney rank sum test or Wilcoxon rank sum test. Two-way repeated

177 measures (RM) ANOVA was applied to experiments with multiple test variables for each genotype. Sidak’s multiple comparison test was used to compare row means between groups. Data are presented as mean ± standard error. Alpha was set to 0.05 for all experiments. Effect size, a measure of the amount of variance accounted for by differences between groups, is reported as h2 only when p<0.05. h2 effect sizes are defined as small – 0.01, medium – 0.06, and large – 0.14

(Cohen, 1988). Statistics were calculated using Prism software (Graphpad).

178

References

Ahmed MS, Siegelbaum SA (2009) Recruitment of N-Type Ca(2+) channels during LTP enhances

low release efficacy of hippocampal CA1 perforant path synapses. Neuron 63:372–385.

Akita Y (2002) Protein kinase C-epsilon (PKC-epsilon): its unique structure and function. J

Biochem 132:847–852.

Amaral DG (1978) A Golgi study of cell types in the hilar region of the hippocampus in the rat. J

Comp Neurol 182:851–914.

Amaral DG, Witter MP (1989) The three-dimensional organization of the hippocampal formation:

a review of anatomical data. Neuroscience 31:571–591.

Andersen P, Holmqvist B, Voorhoeve PE (1966) Excitatory Synapses on Hippocampal Apical

Dendrites Activated by Entorhinal Stimulation. Acta Physiol Scand 66:461–472.

Andrásfalvy BK, Makara JK, Johnston D, Magee JC (2008) Altered synaptic and non-synaptic

properties of CA1 pyramidal neurons in Kv4.2 knockout mice. J Physiol (Lond) 586:3881–

3892.

Andreasen M, Lambert JD (1995) Regenerative properties of pyramidal cell dendrites in area CA1

of the rat hippocampus. J Physiol (Lond) 483 ( Pt 2):421–441.

Arnold EC, McMurray C, Gray R, Johnston D (2019) Epilepsy-Induced Reduction in HCN

Channel Expression Contributes to an Increased Excitability in Dorsal, But Not Ventral,

Hippocampal CA1 Neurons. eNeuro 6.

Ascher P, Nowak L (1988) The role of divalent cations in the N‐methyl‐D‐aspartate responses of

mouse central neurones in culture. J Physiol (Lond) 399:247–266.

179

Barbara JG, Clarac F (2011) Historical concepts on the relations between nerves and muscles.

Brain Res 1409:3–22.

Bassell GJ, Warren ST (2008) Fragile X syndrome: loss of local mRNA regulation alters synaptic

development and function. Neuron 60:201–214.

Baxter DA, Bittner GD, Brown TH (1985) Quantal mechanism of long-term synaptic potentiation.

Proc Natl Acad Sci USA 82:5978–5982.

Bekkers JM, Delaney AJ (2001) Modulation of excitability by alpha-dendrotoxin-sensitive

potassium channels in neocortical pyramidal neurons. J Neurosci 21:6553–6560.

Bekkers JM, Stevens CF (1990) Presynaptic mechanism for long-term potentiation in the

hippocampus. Nature 346:724–729.

Bell MV, Hirst MC, Nakahori Y, MacKinnon RN, Roche A, Flint TJ, Jacobs PA, Tommerup N,

Tranebjaerg L, Froster-Iskenius U (1991) Physical mapping across the fragile X:

hypermethylation and clinical expression of the fragile X syndrome. Cell 64:861–866.

Benardo LS, Masukawa LM, Prince DA (1982) Electrophysiology of isolated hippocampal

pyramidal dendrites. J Neurosci 2:1614–1622.

Bennett MR (1999) The early history of the synapse: from Plato to Sherrington. Brain Res Bull

50:95–118.

Berry-Kravis E, Huttenlocher PR (1992) Cyclic AMP metabolism in fragile X syndrome. Ann

Neurol 31:22–26.

Bilousova TV, Dansie L, Ngo M, Aye J, Charles JR, Ethell DW, Ethell IM (2009) Minocycline

promotes dendritic spine maturation and improves behavioural performance in the fragile X

mouse model. J Med Genet 46:94–102.

180

Bittner KC, Milstein AD, Grienberger C, Romani S, Magee JC (2017) Behavioral time scale

synaptic plasticity underlies CA1 place fields. Science 357:1033–1036.

Blackstad TW (1956) Commissural connections of the hippocampal region in the rat, with special

reference to their mode of termination. J Comp Neurol 105:417–537.

Blackstad TW (1958) On the termination of some afferents to the hippocampus and fascia dentata;

an experimental study in the rat. Acta Anat (Basel) 35:202–214.

Bliss TV, Lomo T (1973) Long-lasting potentiation of synaptic transmission in the dentate area of

the anaesthetized rabbit following stimulation of the perforant path. J Physiol (Lond) 232:331–

356.

Booker SA, Simões de Oliveira L, Anstey NJ, Kozic Z, Dando OR, Jackson AD, Baxter PS, Isom

LL, Sherman DL, Hardingham GE, Brophy PJ, Wyllie DJA, Kind PC (2020) Input-Output

Relationship of CA1 Pyramidal Neurons Reveals Intact Homeostatic Mechanisms in a Mouse

Model of Fragile X Syndrome. Cell Rep 32:107988.

Bostrom CA, Majaess N-M, Morch K, White E, Eadie BD, Christie BR (2015) Rescue of

NMDAR-dependent synaptic plasticity in Fmr1 knock-out mice. Cereb Cortex 25:271–279.

Brager DH, Akhavan AR, Johnston D (2012) Impaired dendritic expression and plasticity of h-

channels in the fmr1(-/y) mouse model of fragile X syndrome. Cell Rep 1:225–233.

Brager DH, Johnston D (2007) Plasticity of intrinsic excitability during long-term depression is

mediated through mGluR-dependent changes in I(h) in hippocampal CA1 pyramidal neurons.

J Neurosci 27:13926–13937.

181

Brager DH, Lewis AS, Chetkovich DM, Johnston D (2013) Short- and long-term plasticity in CA1

neurons from mice lacking h-channel auxiliary subunit TRIP8b. J Neurophysiol 110:2350–

2357.

Brandalise F, Kalmbach BE, Mehta P, Thornton O, Johnston D, Zemelman BV, Brager DH (2020)

Fragile X Mental Retardation Protein Bidirectionally Controls Dendritic Ih in a Cell Type-

Specific Manner between Mouse Hippocampus and Prefrontal Cortex. J Neurosci 40:5327–

5340.

Brose N, Petrenko AG, Sudhof TC, Jahn R (1992) Synaptotagmin: a calcium sensor on the

synaptic vesicle surface. Science 256:1021–1025.

Brown MR, Kronengold J, Gazula V-R, Chen Y, Strumbos JG, Sigworth FJ, Navaratnam D,

Kaczmarek LK (2010) Fragile X mental retardation protein controls gating of the sodium-

activated potassium channel Slack. Nat Neurosci 13:819–821.

Brown V, Jin P, Ceman S, Darnell JC, O'Donnell WT, Tenenbaum SA, Jin X, Feng Y, Wilkinson

KD, Keene JD, Darnell RB, Warren ST (2001) Microarray identification of FMRP-associated

brain mRNAs and altered mRNA translational profiles in fragile X syndrome. Cell 107:477–

487.

Brown WT, Jenkins EC, Friedman E, Brooks J, Wisniewski K, Raguthu S, French J (1982) Autism

is associated with the fragile-X syndrome. Journal of Autism and Developmental Disorders

12:303–308.

Brun VH, Otnass MK, Molden S, Steffenach H-A, Witter MP, Moser M-B, Moser EI (2002) Place

cells and place recognition maintained by direct entorhinal-hippocampal circuitry. Science

296:2243–2246.

182

Cai X, Kallarackal AJ, Kvarta MD, Goluskin S, Gaylor K, Bailey AM, Lee H-K, Huganir RL,

Thompson SM (2013) Local potentiation of excitatory synapses by serotonin and its alteration

in rodent models of depression. Nat Neurosci 16:464–472.

Cain SM, Snutch TP (2013) T-type calcium channels in burst-firing, network synchrony, and

epilepsy. Biochim Biophys Acta 1828:1572–1578.

Callaway JC, Ross WN (1995) Frequency-dependent propagation of sodium action potentials in

dendrites of hippocampal CA1 pyramidal neurons. J Neurophysiol 74:1395–1403.

Carmeliet E (2019) From Bernstein“s rheotome to Neher-Sakmann”s patch electrode. The action

potential. Physiol Rep 7:e13861.

Castle NA, Fadous S, Logothetis DE, Wang GK (1994) Aminopyridine block of Kv1.1 potassium

channels expressed in mammalian cells and Xenopus oocytes. Mol Pharmacol 45:1242–1252.

Chen X, Johnston D (2004) Properties of single voltage-dependent K+ channels in dendrites of

CA1 pyramidal neurones of rat hippocampus. J Physiol (Lond) 559:187–203.

Chen X, Yuan L-L, Zhao C, Birnbaum SG, Frick A, Jung WE, Schwarz TL, Sweatt JD, Johnston

D (2006) Deletion of Kv4.2 gene eliminates dendritic A-type K+ current and enhances

induction of long-term potentiation in hippocampal CA1 pyramidal neurons. J Neurosci

26:12143–12151.

Chevaleyre V, Siegelbaum SA (2010) Strong CA2 pyramidal neuron synapses define a powerful

disynaptic cortico-hippocampal loop. Neuron 66:560–572.

Chittajallu R, Auville K, Mahadevan V, Lai M, Hunt S, Calvigioni D, Pelkey KA, Zaghloul KA,

McBain CJ (2020) Activity-dependent tuning of intrinsic excitability in mouse and human

neurogliaform cells. Elife 9:684.

183

Choi CH, Schoenfeld BP, Weisz ED, Bell AJ, Chambers DB, Hinchey J, Choi RJ, Hinchey P,

Kollaros M, Gertner MJ, Ferrick NJ, Terlizzi AM, Yohn N, Koenigsberg E, Liebelt DA, Zukin

RS, Woo NH, Tranfaglia MR, Louneva N, Arnold SE, Siegel SJ, Bolduc FV, McDonald TV,

Jongens TA, McBride SMJ (2015) PDE-4 inhibition rescues aberrant synaptic plasticity in

Drosophila and mouse models of fragile X syndrome. J Neurosci 35:396–408.

Clements L, Harvey J (2020) Activation of oestrogen receptor α induces a novel form of LTP at

hippocampal temporoammonic-CA1 synapses. British Journal of Pharmacology 177:642–

655.

Cohen, J. (1988) Statistical power analysis for the behavioral sciences. Hillsdale, NJ:

Erlbaum.

Cohen SM, Suutari B, He X, Wang Y, Sanchez S, Tirko NN, Mandelberg NJ, Mullins C, Zhou G,

Wang S, Kats I, Salah A, Tsien RW, Ma H (2018) Calmodulin shuttling mediates cytonuclear

signaling to trigger experience-dependent transcription and memory. Nat Commun 9:2451–

12.

Colbert CM, Levy WB (1993) Long-term potentiation of perforant path synapses in hippocampal

CA1 in vitro. Brain Res 606:87–91.

Cole FJ (1937) Jan Swammerdam, 1637–80. Nature 139:218–220.

Cole KS, Curtis HJ (1939) Electric impedance of the squid giant axon during activity. J Gen

Physiol 22:649–670.

Collingridge GL, Kehl SJ, McLennan H (1983) Excitatory amino acids in synaptic transmission

in the Schaffer collateral-commissural pathway of the rat hippocampus. J Physiol (Lond)

334:33–46.

184

Comery TA, Harris JB, Willems PJ, Oostra BA, Irwin SA, Weiler IJ, Greenough WT (1997)

Abnormal dendritic spines in fragile X knockout mice: maturation and pruning deficits. Proc

Natl Acad Sci USA 94:5401–5404.

Connor JA, Stevens CF (1971) Voltage clamp studies of a transient outward membrane current in

gastropod neural somata. J Physiol (Lond) 213:21–30.

Cragg BG, Hamlyn LH (1957) Some commissural and septal connexions of the hippocampus in

the rabbit; a combined histological and electrical study. J Physiol (Lond) 135:460–485.

D'Hulst C, De Geest N, Reeve SP, Van Dam D, De Deyn PP, Hassan BA, Kooy RF (2006)

Decreased expression of the GABAA receptor in fragile X syndrome. Brain Res 1121:238–

245.

Danesi C, Achuta VS, Corcoran P, Peteri U-K, Turconi G, Matsui N, Albayrak I, Rezov V,

Isaksson A, Castrén ML (2018) Increased Calcium Influx through L-type Calcium Channels

in Human and Mouse Neural Progenitors Lacking Fragile X Mental Retardation Protein. Stem

Cell Reports 11:1449–1461.

Darnell JC, Van Driesche SJ, Zhang C, Hung KYS, Mele A, Fraser CE, Stone EF, Chen C, Fak

JJ, Chi SW, Licatalosi DD, Richter JD, Darnell RB (2011) FMRP stalls ribosomal

translocation on mRNAs linked to synaptic function and autism. Cell 146:247–261.

Del Castillo J, Katz B (1954) Quantal components of the end‐plate potential. J Physiol (Lond)

124:560–573.

Deng P-Y, Carlin D, Oh YM, Myrick LK, Warren ST, Cavalli V, Klyachko VA (2019) Voltage-

Independent SK-Channel Dysfunction Causes Neuronal Hyperexcitability in the

Hippocampus of Fmr1 Knock-Out Mice. J Neurosci 39:28–43.

185

Deng P-Y, Klyachko VA (2016a) Genetic upregulation of BK channel activity normalizes multiple

synaptic and circuit defects in a mouse model of fragile X syndrome. J Physiol (Lond) 594:83–

97.

Deng P-Y, Klyachko VA (2016b) Increased Persistent Sodium Current Causes Neuronal

Hyperexcitability in the Entorhinal Cortex of Fmr1 Knockout Mice. Cell Rep 16:3157–3166.

Deng P-Y, Rotman Z, Blundon JA, Cho Y, Cui J, Cavalli V, Zakharenko SS, Klyachko VA (2013)

FMRP regulates neurotransmitter release and synaptic information transmission by

modulating action potential duration via BK channels. Neuron 77:696–711.

Denk W, Strickler JH, Webb WW (1990) Two-photon laser scanning fluorescence microscopy.

Science 248:73–76.

Devys D, Lutz Y, Rouyer N, Bellocq JP, Mandel JL (1993) The FMR-1 protein is cytoplasmic,

most abundant in neurons and appears normal in carriers of a fragile X premutation. Nat Genet

4:335–340.

Dictenberg JB, Swanger SA, Antar LN, Singer RH, Bassell GJ (2008) A direct role for FMRP in

activity-dependent dendritic mRNA transport links filopodial-spine morphogenesis to fragile

X syndrome. Dev Cell 14:926–939.

Dodt HU, Zieglgänsberger W (1990) Visualizing unstained neurons in living brain slices by

infrared DIC-videomicroscopy. Brain Res 537:333–336.

Doller HJ, Weight FF (1982) Perforant pathway activation of hippocampal CA1 stratum

pyramidale neurons: Electrophysiological evidence for a direct pathway. Brain Res 237:1–13.

Doller HJ, Weight FF (1985) Perforant pathway-evoked long-term potentiation of CA1 neurons in

the hippocampal slice preparation. Brain Res 333:305–310.

186

Dolphin AC (2006) A short history of voltage-gated calcium channels. British Journal of

Pharmacology 147 Suppl 1:S56–S62.

Dolphin AC (2016) Voltage-gated calcium channels and their auxiliary subunits: physiology and

pathophysiology and pharmacology. J Physiol (Lond) 594:5369–5390.

Dolphin AC, Errington ML, Bliss TV (1982) Long-term potentiation of the perforant path in vivo

is associated with increased glutamate release. Nature 297:496–498.

Dong H-W, Swanson LW, Chen L, Fanselow MS, Toga AW (2009) Genomic-anatomic evidence

for distinct functional domains in hippocampal field CA1. Proc Natl Acad Sci USA

106:11794–11799.

Dougherty KA, Islam T, Johnston D (2012) Intrinsic excitability of CA1 pyramidal neurones from

the rat dorsal and ventral hippocampus. J Physiol (Lond) 590:5707–5722.

Dougherty KA, Nicholson DA, Diaz L, Buss EW, Neuman KM, Chetkovich DM, Johnston D

(2013) Differential expression of HCN subunits alters voltage-dependent gating of h-channels

in CA1 pyramidal neurons from dorsal and ventral hippocampus. J Neurophysiol 109:1940–

1953.

Dvorak-Carbone H, Schuman EM (1999) Long-term depression of temporoammonic-CA1

hippocampal synaptic transmission. J Neurophysiol 81:1036–1044.

Eberhart DE, Malter HE, Feng Y, Warren ST (1996) The fragile X mental retardation protein is a

ribonucleoprotein containing both nuclear localization and nuclear export signals. Hum Mol

Genet 5:1083–1091.

Eccles JC, Libet B, Young RR (1958) The behaviour of chromatolysed motoneurones studied by

intracellular recording. J Physiol (Lond) 143:11–40.

187

Ekstrand JJ, Pouliot W, Scheerlinck P, Dudek FE (2011) Lithium pilocarpine-induced status

epilepticus in postnatal day 20 rats results in greater neuronal injury in ventral versus dorsal

hippocampus. Neuroscience 192:699–707.

Elferink LA, Peterson MR, Scheller RH (1993) A role for synaptotagmin (p65) in regulated

exocytosis. Cell 72:153–159.

Engelhardt E (2016) Hippocampus discovery First steps. Dement Neuropsychol 10:58–62.

Evans RH, Francis AA, Jones AW, Smith DAS, Watkins JC (1982) The effects of a series of ω‐

phosphonic α‐carboxylic amino acids on electrically evoked and excitant amino acid‐induced

responses in isolated spinal cord preparations. British Journal of Pharmacology 75:65–75.

Fanselow MS, Dong H-W (2010) Are the dorsal and ventral hippocampus functionally distinct

structures? Neuron 65:7–19.

Fatt P, Katz B (1951) An analysis of the end‐plate potential recorded with an intra‐cellular

electrode. J Physiol (Lond) 115:320–370.

Fatt P, Katz B (1952) Spontaneous subthreshold activity at motor nerve endings. J Physiol (Lond)

117:109–128.

Feng Y, Gutekunst CA, Eberhart DE, Yi H, Warren ST, Hersch SM (1997) Fragile X mental

retardation protein: nucleocytoplasmic shuttling and association with somatodendritic

ribosomes. J Neurosci 17:1539–1547.

Ferron L, Nieto-Rostro M, Cassidy JS, Dolphin AC (2014) Fragile X mental retardation protein

controls synaptic vesicle exocytosis by modulating N-type calcium channel density. Nat

Commun 5:3628.

188

Frey U, Morris RG (1998) Synaptic tagging: implications for late maintenance of hippocampal

long-term potentiation. Trends in Neurosciences 21:181–188.

Frick A, Magee J, Koester HJ, Migliore M, Johnston D (2003) Normalization of Ca2+ signals by

small oblique dendrites of CA1 pyramidal neurons. J Neurosci 23:3243–3250.

Fricke H, Morse S (1925) The electric resistance and capacity of blood for frequencies between

800 and 4½ million cycles. J Gen Physiol 9:151–167

Fryns JP (1984) The fragile X syndrome. A study of 83 families. Clin Genet 26:497–528.

Fryns JP, Jacobs J, Kleczkowska A, van den Berghe H (1984) The psychological profile of the

fragile X syndrome. Clin Genet 25:131–134.

Fyhn M, Molden S, Witter MP, Moser EI, Moser M-B (2004) Spatial representation in the

entorhinal cortex. Science 305:1258–1264.

Galvez R, Greenough WT (2005) Sequence of abnormal dendritic spine development in primary

somatosensory cortex of a mouse model of the fragile X mental retardation syndrome. Am J

Med Genet A 135:155–160.

Gantois I, Vandesompele J, Speleman F, Reyniers E, D'Hooge R, Severijnen L-A, Willemsen R,

Tassone F, Kooy RF (2006) Expression profiling suggests underexpression of the GABA(A)

receptor subunit delta in the fragile X knockout mouse model. Neurobiol Dis 21:346–357.

Gasparini S, Losonczy A, Chen X, Johnston D, Magee JC (2007) Associative pairing enhances

action potential back‐propagation in radial oblique branches of CA1 pyramidal neurons. J

Physiol (Lond) 580:787–800.

189

Gasparini S, Magee JC (2002) Phosphorylation-dependent differences in the activation properties

of distal and proximal dendritic Na+ channels in rat CA1 hippocampal neurons. J Physiol

(Lond) 541:665–672.

Gasparini S, Migliore M, Magee JC (2004) On the initiation and propagation of dendritic spikes

in CA1 pyramidal neurons. J Neurosci 24:11046–11056.

Gasser HS, Erlanger J, (1922) A study of the action currents of nerve with the cathode ray

oscillograph. J Physiol 62:496–524.

Glickstein M (2006) Golgi and Cajal: The neuron doctrine and the 100th anniversary of the 1906

Nobel Prize. Curr Biol 16:R147–151

Golding NL, Jung HY, Mickus T, Spruston N (1999) Dendritic calcium spike initiation and

repolarization are controlled by distinct potassium channel subtypes in CA1 pyramidal

neurons. J Neurosci 19:8789–8798.

Golding NL, Mickus TJ, Katz Y, Kath WL, Spruston N (2005) Factors mediating powerful voltage

attenuation along CA1 pyramidal neuron dendrites. J Physiol (Lond) 568:69–82.

Golding NL, Spruston N (1998) Dendritic Sodium Spikes Are Variable Triggers of Axonal Action

Potentials in Hippocampal CA1 Pyramidal Neurons. Neuron 21:1189–1200.

Golding NL, Staff NP, Spruston N (2002) Dendritic spikes as a mechanism for cooperative long-

term potentiation. Nature 418:326–331.

Gorter E, Grendel F (1925) On bimolecular layers of lipoids on the chromocytes of the blood. The

Journal of Experimental Medicine 41:439–443.

Graham J, Gerard RW (1946) Membrane potentials and excitation of impaled single muscle fibers.

J Cell Comp Physiol 28:99–117.

190

Gray EE, Murphy JG, Liu Y, Trang I, Tabor GT, Lin L, Hoffman DA (2019) Disruption of GpI

mGluR-Dependent Cav2.3 Translation in a Mouse Model of Fragile X Syndrome. J Neurosci

39:7453–7464.

Greicius MD, Boyett-Anderson JM, Menon V, Reiss AL (2004) Reduced basal forebrain and

hippocampal activation during memory encoding in girls with fragile X syndrome.

NeuroReport 15:1579–1583.

Gross C, Yao X, Pong DL, Jeromin A, Bassell GJ (2011) Fragile X mental retardation protein

regulates protein expression and mRNA translation of the potassium channel Kv4.2. J

Neurosci 31:5693–5698.

Grundfest H (1957) Excitation at synapses. J Neurophysiol 20:316–324.

Guan D, Lee JCF, Higgs MH, Spain WJ, Foehring RC (2007) Functional roles of Kv1 channels in

neocortical pyramidal neurons. J Neurophysiol 97:1931–1940.

Gustafsson B, Wigström H, Abraham WC, Huang YY (1987) Long-term potentiation in the

hippocampus using depolarizing current pulses as the conditioning stimulus to single volley

synaptic potentials. J Neurosci 7:774–780.

Haessler F, Gaese F, Huss M, Kretschmar C, Brinkman M, Peters H, Elstner S, Colla M, Pittrow

D (2016) Characterization, treatment patterns, and patient-related outcomes of patients with

Fragile X syndrome in Germany: final results of the observational EXPLAIN-FXS study.

BMC Psychiatry 16:318.

Halliwell JV, Adams PR (1982) Voltage-clamp analysis of muscarinic excitation in hippocampal

neurons. Brain Res 250:71–92.

191

Hamill OP, Marty A, Neher E, Sakmann B, Sigworth FJ (1981) Improved patch-clamp techniques

for high-resolution current recording from cells and cell-free membrane patches. Pflugers

Arch 391:85–100.

Harris KM, Fiala JC, Ostroff L (2003) Structural changes at dendritic spine synapses during long-

term potentiation. Philos Trans R Soc Lond, B, Biol Sci 358:745–748.

Hebb DO (1949) The organization of behavior; a neuropsycholocigal theory. New York, NY.

Wiley.

Henke PG (1990) Hippocampal pathway to the amygdala and stress ulcer development. Brain Res

Bull 25:691–695.

Hibino H, Inanobe A, Furutani K, Murakami S, Findlay I, Kurachi Y (2010) Inwardly rectifying

potassium channels: their structure, function, and physiological roles. Physiol Rev 90:291–

366.

Higgs MH, Spain WJ (2011) Kv1 channels control spike threshold dynamics and spike timing in

cortical pyramidal neurones. J Physiol (Lond) 589:5125–5142.

Hinton VJ, Brown WT, Wisniewski K, Rudelli RD (1991) Analysis of neocortex in three males

with the fragile X syndrome. Am J Med Genet 41:289–294.

Hjorth-Simonsen A (1971) Hippocampal efferents to the ipsilateral entorhinal area: An

experimental study in the rat. J Comp Neurol 142:417–437.

Hodgkin AL, Huxley AF (1939) Action Potentials Recorded from Inside a Nerve Fibre. Nature

144:710–711.

Hodgkin AL, Huxley AF (1945) Resting and action potentials in single nerve fibres. J Physiol

(Lond) 104:176–195.

192

Hodgkin AL, Huxley AF (1952a) A quantitative description of membrane current and its

application to conduction and excitation in nerve. J Physiol (Lond) 117:500–544.

Hodgkin AL, Huxley AF (1952b) Currents carried by sodium and potassium ions through the

membrane of the giant axon of Loligo. J Physiol (Lond) 116:449–472.

Hodgkin AL, Huxley AF (1952c) The components of membrane conductance in the giant axon of

Loligo. J Physiol (Lond) 116:473–496.

Hodgkin AL, Huxley AF (1952d) The dual effect of membrane potential on sodium conductance

in the giant axon of Loligo. J Physiol (Lond) 116:497–506.

Hodgkin AL, Huxley AF, Katz B (1952) Measurement of current-voltage relations in the

membrane of the giant axon of Loligo. J Physiol (Lond) 116:424–448.

Hodgkin AL, Katz B (1949) The effect of sodium ions on the electrical activity of giant axon of

the squid. J Physiol (Lond) 108:37–77.

Hodgkin AL, Rushton WAH (1946) The electrical constants of a crustacean nerve fibre. Proc R

Soc Med 134:444–479.

Hoffman DA, Johnston D (1998) Downregulation of transient K+ channels in dendrites of

hippocampal CA1 pyramidal neurons by activation of PKA and PKC. J Neurosci 18:3521–

3528.

Hoffman DA, Johnston D (1999) Neuromodulation of dendritic action potentials. J Neurophysiol

81:408–411.

Hoffman DA, Magee JC, Colbert CM, Johnston D (1997) K+ channel regulation of signal

propagation in dendrites of hippocampal pyramidal neurons. Nature 387:869–875.

193

Hönigsperger C, Marosi M, Murphy R, Storm JF (2015) Dorsoventral differences in Kv7/M-

current and its impact on resonance, temporal summation and excitability in rat hippocampal

pyramidal cells. J Physiol (Lond) 593:1551–1580.

Hu H, Qin Y, Bochorishvili G, Zhu Y, van Aelst L, Zhu JJ (2008) Ras signaling mechanisms

underlying impaired GluR1-dependent plasticity associated with fragile X syndrome. J

Neurosci 28:7847–7862.

Hu J-H, Malloy C, Tabor GT, Gutzmann JJ, Liu Y, Abebe D, Karlsson R-M, Durell S, Cameron

HA, Hoffman DA (2020) Activity-dependent isomerization of Kv4.2 by Pin1 regulates

cognitive flexibility. Nat Commun 11:1567–18.

Huber KM, Gallagher SM, Warren ST, Bear MF (2002) Altered synaptic plasticity in a mouse

model of fragile X mental retardation. Proc Natl Acad Sci USA 99:7746–7750.

Hussain S, Davanger S (2011) The discovery of the soluble N-ethylmaleimide-sensitive factor

attachment protein receptor complex and the molecular regulation of synaptic vesicle

transmitter release: the 2010 Kavli Prize in neuroscience. Neuroscience 190:12–20.

Isaac JTR, Nicoll RA, Malenka RC (1995) Evidence for silent synapses: Implications for the

expression of LTP. Neuron 15:427–434.

Ishizuka N, Weber J, Amaral DG (1990) Organization of intrahippocampal projections originating

from CA3 pyramidal cells in the rat. J Comp Neurol 295:580–623.

Ito HT, Schuman EM (2009) Distance-dependent homeostatic synaptic scaling mediated by a-type

potassium channels. Front Cell Neurosci 3:15.

194

Ito M, Kano M (1982) Long-lasting depression of parallel fiber-Purkinje cell transmission induced

by conjunctive stimulation of parallel fibers and climbing fibers in the cerebellar cortex.

Neurosci Lett 33:253–258.

Janesick JR (2001) Scientific Charge-coupled Devices. SPIE Press.

Jarrard LE (1983) Selective hippocampal lesions and behavior: effects of kainic acid lesions on

performance of place and cue tasks. Behav Neurosci 97:873–889.

Jarrard LE (1993) On the role of the hippocampus in learning and memory in the rat. Behav Neural

Biol 60:9–26.

Jay TM, Glowinski J, Thierry AM (1989) Selectivity of the hippocampal projection to the

prelimbic area of the prefrontal cortex in the rat. Brain Res 505:337–340.

Johnston D, Christie BR, Frick A, Gray R, Hoffman DA, Schexnayder LK, Watanabe S, Yuan L-

L (2003) Active dendrites, potassium channels and synaptic plasticity. Philos Trans R Soc

Lond, B, Biol Sci 358:667–674.

Johnson BJ (1922) A low voltage cathode ray oscillograph. J Opt Soc Am 6:701–712

Jung MW, Wiener SI, McNaughton BL (1994) Comparison of spatial firing characteristics of units

in dorsal and ventral hippocampus of the rat. J Neurosci 14:7347–7356.

Kalmbach BE, Johnston D, Brager DH (2015) Cell-Type Specific Channelopathies in the

Prefrontal Cortex of the fmr1-/y Mouse Model of Fragile X Syndrome. eNeuro

2:ENEURO.0114–15.2015.

Katz B, Miledi R (1965) The effect of calcium on acetylcholine release from motor nerve

terminals. Proc R Soc Lond B Biol Sci. 161:496–503

195

Kelso SR, Ganong AH, Brown TH (1986) Hebbian synapses in hippocampus. Proc Natl Acad Sci

USA 83:5326–5330.

Kemper MB, Hagerman RJ, Altshul-Stark D (1988) Cognitive profiles of boys with the fragile X

syndrome. Am J Med Genet 30:191–200.

Khalfallah O, Jarjat M, Davidovic L, Nottet N, Cestèle S, Mantegazza M, Bardoni B (2016)

Depletion of the fragile X mental retardation protein in embryonic stem cells alters the kinetics

of neurogenesis. Stem Cells.

Kim CS, Johnston D (2015) A1 adenosine receptor-mediated GIRK channels contribute to the

resting conductance of CA1 neurons in the dorsal hippocampus. J Neurophysiol 113:2511–

2523.

Kim K, Saneyoshi T, Hosokawa T, Okamoto K, Hayashi Y (2016) Interplay of enzymatic and

structural functions of CaMKII in long-term potentiation. J Neurochem 139:959–972.

Kim SM, Frank LM (2009) Hippocampal lesions impair rapid learning of a continuous spatial

alternation task. Tanimoto H, ed. PLoS ONE 4:e5494.

Kim Y, Hsu C-L, Cembrowski MS, Mensh BD, Spruston N (2015) Dendritic sodium spikes are

required for long-term potentiation at distal synapses on hippocampal pyramidal neurons. Elife

4:372.

Knoll M, Ruska E (1932) Das Elektronenmikroskop. Z Physik 78:318–339.

Larkum ME, Zhu JJ, Sakmann B (1999) A new cellular mechanism for coupling inputs arriving at

different cortical layers. Nature 398:338–341.

196

Larson J, Jessen RE, Kim D, Fine A-KS, Hoffmann du J (2005) Age-dependent and selective

impairment of long-term potentiation in the anterior piriform cortex of mice lacking the fragile

X mental retardation protein. J Neurosci 25:9460–9469.

Lauterborn JC, Rex CS, Kramár E, Chen LY, Pandyarajan V, Lynch G, Gall CM (2007) Brain-

derived neurotrophic factor rescues synaptic plasticity in a mouse model of fragile X

syndrome. J Neurosci 27:10685–10694.

Lee AM, Kanter BR, Wang D, Lim JP, Zou ME, Qiu C, McMahon T, Dadgar J, Fischbach-Weiss

SC, Messing RO (2013) Prkcz null mice show normal learning and memory. Nature 493:416–

419.

Lee HY, Ge W-P, Huang W, He Y, Wang GX, Rowson-Baldwin A, Smith SJ, Jan YN, Jan LY

(2011) Bidirectional regulation of dendritic voltage-gated potassium channels by the fragile X

mental retardation protein. Neuron 72:630–642.

Levy WB, Steward O (1979) Synapses as associative memory elements in the hippocampal

formation. Brain Res 175:233–245.

Li B, Tadross MR, Tsien RW (2016) Sequential ionic and conformational signaling by calcium

channels drives neuronal gene expression. Science 351:863–867.

Li J, Pelletier MR, Perez Velazquez J-L, Carlen PL (2002) Reduced cortical synaptic plasticity

and GluR1 expression associated with fragile X mental retardation protein deficiency. Mol

Cell Neurosci 19:138–151.

Liao D, Hessler NA, Malinow R (1995) Activation of postsynaptically silent synapses during

pairing-induced LTP in CA1 region of hippocampal slice. Nature 375:400–404.

197

Ling G, Gerard RW (1949) The membrane potential and metabolism of muscle fibers. J Cell Comp

Physiol 34:413–438.

Lisman J, Schulman H, Cline H (2002) The molecular basis of CaMKII function in synaptic and

behavioural memory. Nat Rev Neurosci 3:175–190.

Lisman J, Yasuda R, Raghavachari S (2012) Mechanisms of CaMKII action in long-term

potentiation. Nat Rev Neurosci 13:169–182.

Lisman JE, Grace AA (2005) The hippocampal-VTA loop: controlling the entry of information

into long-term memory. Neuron 46:703–713.

Lledo PM, Hjelmstad GO, Mukherji S, Soderling TR, Malenka RC, Nicoll RA (1995)

Calcium/calmodulin-dependent kinase II and long-term potentiation enhance synaptic

transmission by the same mechanism. Proc Natl Acad Sci USA 92:11175–11179.

Llinás R, Steinberg IZ, Walton K (1976) Presynaptic calcium currents and their relation to synaptic

transmission: voltage clamp study in squid giant synapse and theoretical model for the calcium

gate. Proc Natl Acad Sci USA 73:2918–2922.

Llinás R, Sugimori M (1980) Electrophysiological properties of in vitro Purkinje cell dendrites in

mammalian cerebellar slices. J Physiol (Lond) 305:197–213.

Lothman EW, Collins RC (1981) Kainic acid induced limbic seizures: metabolic, behavioral,

electroencephalographic and neuropathological correlates. Brain Res 218:299–318.

Lörincz A, Notomi T, Tamás G, Shigemoto R, Nusser Z (2002) Polarized and compartment-

dependent distribution of HCN1 in pyramidal cell dendrites. Nat Neurosci 5:1185–1193.

198

Lu WY, Man HY, Ju W, Trimble WS, MacDonald JF, Wang YT (2001) Activation of Synaptic

NMDA Receptors Induces Membrane Insertion of New AMPA Receptors and LTP in

Cultured Hippocampal Neurons. Neuron 29:243–254.

Lubs HA (1969) A marker X chromosome. Am J Hum Genet 21:231–244.

Ludwig AL, Espinal GM, Pretto DI, Jamal AL, Arque G, Tassone F, Berman RF, Hagerman PJ

(2014) CNS expression of murine fragile X protein (FMRP) as a function of CGG-repeat size.

Hum Mol Genet 23:3228–3238.

Luque MA, Beltran-Matas P, Marin MC, Torres B, Herrero L (2017) Excitability is increased in

hippocampal CA1 pyramidal cells of Fmr1 knockout mice. Charpier S, ed. PLoS ONE

12:e0185067.

Lynch G, Larson J, Kelso S, Barrionuevo G, Schottler F (1983) Intracellular injections of EGTA

block induction of hippocampal long-term potentiation. Nature 305:719–721.

Maccaferri G, Mangoni M, Lazzari A, DiFrancesco D (1993) Properties of the hyperpolarization-

activated current in rat hippocampal CA1 pyramidal cells. J Neurophysiol 69:2129–2136.

MacDermott AB, Mayer ML, Westbrook GL, Smith SJ, Barker JL (1986) NMDA-receptor

activation increases cytoplasmic calcium concentration in cultured spinal cord neurones.

Nature 321:519–522.

Magee J (1999) Dendritic Ih normalizes temporal summation in hippocampal CA1 neurons. Nat

Neurosci 2:848–848.

Magee JC (1998) Dendritic hyperpolarization-activated currents modify the integrative properties

of hippocampal CA1 pyramidal neurons. J Neurosci 18:7613–7624.

199

Magee JC, Cook EP (2000) Somatic EPSP amplitude is independent of synapse location in

hippocampal pyramidal neurons. Nat Neurosci 3:895–903.

Magee JC, Johnston D (1995a) Characterization of single voltage-gated Na+ and Ca2+ channels

in apical dendrites of rat CA1 pyramidal neurons. J Physiol (Lond) 487:67–90.

Magee JC, Johnston D (1995b) Synaptic activation of voltage-gated channels in the dendrites of

hippocampal pyramidal neurons. Science 268:301–304.

Magee JC, Johnston D (1997) A synaptically controlled, associative signal for Hebbian plasticity

in hippocampal neurons. Science 275:209–213.

Maggio N, Segal M (2007) Unique regulation of long term potentiation in the rat ventral

hippocampus. Hippocampus 17:10–25.

Malik R, Dougherty KA, Parikh K, Byrne C, Johnston D (2015) Mapping the electrophysiological

and morphological properties of CA1 pyramidal neurons along the longitudinal hippocampal

axis. Hippocampus 26:341–361.

Malik R, Johnston D (2017) Dendritic GIRK Channels Gate the Integration Window, Plateau

Potentials, and Induction of Synaptic Plasticity in Dorsal But Not Ventral CA1 Neurons. J

Neurosci 37:3940–3955.

Malinow R, Madison DV, Tsien RW (1988) Persistent protein kinase activity underlying long-

term potentiation. Nature 335:820–824.

Malinow R, Miller JP (1986) Postsynaptic hyperpolarization during conditioning reversibly blocks

induction of long-term potentiation. Nature 320:529–530.

Malinow R, Schulman H, Tsien RW (1989) Inhibition of postsynaptic PKC or CaMKII blocks

induction but not expression of LTP. Science 245:862–866.

200

Malinow R, Tsien RW (1990) Presynaptic enhancement shown by whole-cell recordings of long-

term potentiation in hippocampal slices. Nature 346:177–180.

Man HY, Wang Q, Lu WY, Ju W, Ahmadian G, Liu L, D’Souza S, Wong TP, Taghibiglou C, Lu

J, Becker LE, Pei L, Liu F, Wymann MP, MacDonald JF, Wang YT (2003) Activation of PI3-

Kinase Is Required for AMPA Receptor Insertion during LTP of mEPSCs in Cultured

Hippocampal Neurons. Neuron 38:611–624.

Marcelin B, Lugo JN, Brewster AL, Liu Z, Lewis AS, McClelland S, Chetkovich DM, Baram TZ,

Anderson AE, Becker A, Esclapez M, Bernard C (2012) Differential dorso-ventral

distributions of Kv4.2 and HCN proteins confer distinct integrative properties to hippocampal

CA1 pyramidal cell distal dendrites. J Biol Chem 287:17656–17661.

Marsillo A, David L, Gerges B, Kerr D, Lasiychuk V, Sadek R, Salame D, Soliman Y, Menkes S,

Chatterjee A, Mancuso A, Banerjee P (2020) PKC epsilon as A Neonatal Target to Correct

FXS-linked AMPA Receptor Translocation in the Hippocampus, Boost PVN Oxytocin

Expression, and Normalize Adult Behavior in Fmr1 Knockout Mice. Biochim Biophys Acta

Mol Basis Dis:166048.

Martin JP, Bell J (1943) A pedigree of mental defect showing sex-linkage. J Neurol Psychiatry

6:154–157.

Mayer ML, Westbrook GL, Guthrie PB (1984) Voltage-dependent block by Mg 2+ of NMDA

responses in spinal cord neurones. Nature 309:261–263.

McKinney BC, Grossman AW, Elisseou NM, Greenough WT (2005) Dendritic spine

abnormalities in the occipital cortex of C57BL/6 Fmr1 knockout mice. Am J Med Genet B

Neuropsychiatr Genet 136B:98–102.

201

McNaughton BL, Douglas RM, Goddard GV (1978) Synaptic enhancement in fascia dentata:

cooperativity among coactive afferents. Brain Res 157:277–293.

Meredith RM, Holmgren CD, Weidum M, Burnashev N, Mansvelder HD (2007) Increased

threshold for spike-timing-dependent plasticity is caused by unreliable calcium signaling in

mice lacking fragile X gene FMR1. Neuron 54:627–638.

Migliore M, Hoffman DA, Magee JC, Johnston D (1999) Role of an A-type K+ conductance in

the back-propagation of action potentials in the dendrites of hippocampal pyramidal neurons.

J Comput Neurosci 7:5–15.

Miledi R (1973) Transmitter release induced by injection of calcium ions into nerve terminals.

Proc R Soc Lond B Biol Sci. 183:421–425

Milior G, Di Castro MA, Sciarria LP, Garofalo S, Branchi I, Ragozzino D, Limatola C, Maggi L

(2016) Electrophysiological Properties of CA1 Pyramidal Neurons along the Longitudinal

Axis of the Mouse Hippocampus. Scientific Reports 6:38242.

Miller SG, Kennedy MB (1986) Regulation of brain type II Ca2+/calmodulin-dependent protein

kinase by autophosphorylation: a Ca2+-triggered molecular switch. Cell 44:861–870.

Miyashiro KY, Beckel-Mitchener A, Purk TP, Becker KG, Barret T, Liu L, Carbonetto S, Weiler

IJ, Greenough WT, Eberwine J (2003) RNA cargoes associating with FMRP reveal deficits in

cellular functioning in Fmr1 null mice. Neuron 37:417–431.

Moosmang S, Biel M, Hofmann F, Ludwig A (1999) Differential distribution of four

hyperpolarization-activated cation channels in mouse brain. Biol Chem 380:975–980.

202

Moser E, Moser MB, Andersen P (1993) Spatial learning impairment parallels the magnitude of

dorsal hippocampal lesions, but is hardly present following ventral lesions. J Neurosci

13:3916–3925.

Moser MB, Moser EI, Forrest E, Andersen P, Morris RG (1995) Spatial learning with a minislab

in the dorsal hippocampus. Proc Natl Acad Sci USA 92:9697–9701.

Mou X, Cheng J, Yu YSW, Kee SE, Ji D (2018) Comparing Mouse and Rat Hippocampal Place

Cell Activities and Firing Sequences in the Same Environments. Front Cell Neurosci 12:332.

Musumeci SA, Colognola RM, Ferri R, Gigli GL, Petrella MA, Sanfilippo S, Bergonzi P, Tassinari

CA (1988) Fragile-X syndrome: a particular epileptogenic EEG pattern. Epilepsia 29:41–47.

Musumeci SA, Hagerman RJ, Ferri R, Bosco P, Dalla Bernardina B, Tassinari CA, De Sarro GB,

Elia M (1999) Epilepsy and EEG findings in males with fragile X syndrome. Epilepsia

40:1092–1099.

Nagaoka A, Takehara H, Hayashi-Takagi A, Noguchi J, Ishii K, Shirai F, Yagishita S, Akagi T,

Ichiki T, Kasai H (2016) Abnormal intrinsic dynamics of dendritic spines in a fragile X

syndrome mouse model in vivo. Scientific Reports 6:26651.

Narayanan R, Johnston D (2007) Long-term potentiation in rat hippocampal neurons is

accompanied by spatially widespread changes in intrinsic oscillatory dynamics and

excitability. Neuron 56:1061–1075.

Neher E, Sakmann B (1976) Single-channel currents recorded from membrane of denervated frog

muscle fibres. Nature 260:799–802.

Nicholson C, Llinás R (1971) Field potentials in the alligator cerebellum and theory of their

relationship to Purkinje cell dendritic spikes. J Neurophysiol 34:509–531.

203

Nolan MF, Malleret G, Dudman JT, Buhl DL, Santoro B, Gibbs E, Vronskaya S, Buzsáki G,

Siegelbaum SA, Kandel ER, Morozov A (2004) A behavioral role for dendritic integration:

HCN1 channels constrain spatial memory and plasticity at inputs to distal dendrites of CA1

pyramidal neurons. Cell 119:719–732.

Nomarski G (1957) From phase contrast to contrast by interference. Rev Hematol 12:439–442.

Nowak L, Bregestovski P, Ascher P, Herbet A, Prochiantz A (1984) Magnesium gates glutamate-

activated channels in mouse central neurones. Nature 307:462–465.

Nó RL, Condouris GA (1959) Decremental conduction in peripheral nerve. Integration of stimuli

in the neuron. Proc Natl Acad Sci USA 45:592–617.

O'Keefe J (1976) Place units in the hippocampus of the freely moving rat. Exp Neurol 51:78–109.

O'Keefe J, Dostrovsky J (1971) The hippocampus as a spatial map. Preliminary evidence from unit

activity in the freely-moving rat. Brain Res 34:171–175.

Olverman HJ, Jones AW, Watkins JC (1984) L -Glutamate has higher affinity than other amino

acids for [ 3 H]- D -AP5 binding sites in rat brain membranes. Nature 307:460–462.

Ordemann GJ, Apgar CJ, Brager DH (2019) D-type potassium channels normalize action potential

firing between dorsal and ventral CA1 neurons of the mouse hippocampus. J Neurophysiol

121:983–995.

Ornstein PA, Schaaf JM, Hooper SR, Hatton DD, Mirrett P, Bailey DB (2008) Memory skills of

boys with fragile X syndrome. Frances Conners LAA, ed. Am J Ment Retard 113:453–465.

Palay SL (1956) Synapses in the central nervous system. J Biophysic and Biochem Cytol 2:193-

202.

204

Papatheodoropoulos C, Kostopoulos G (2000) Dorsal-ventral differentiation of short-term

synaptic plasticity in rat CA1 hippocampal region. Neurosci Lett 286:57–60.

Papatheodoropoulos C, Moschovos C, Kostopoulos G (2005) Greater contribution of N-methyl-

D-aspartic acid receptors in ventral compared to dorsal hippocampal slices in the expression

and long-term maintenance of epileptiform activity. Neuroscience 135:765–779.

Park P, Georgiou J, Sanderson TM, Ko K-H, Kang H, Kim J-I, Bradley CA, Bortolotto ZA, Zhuo

M, Kaang B-K, Collingridge GL (2021) PKA drives an increase in AMPA receptor unitary

conductance during LTP in the hippocampus. Nat Commun 12:1–15.

Perin MS, Johnston PA, Ozcelik T, Francke U, Sudhof TC (1991) Structural and functional

conservation of synaptotagmin (p65) in Drosophila and humans. Journal of Biological

Chemistry 266:615–622.

Petrenko AG, Perin MS, Davletov BA, Ushkaryov YA, Geppert M, Südhof TC (1991) Binding of

synaptotagmin to the α-latrotoxin receptor implicates both in synaptic vesicle exocytosis.

Nature 353:65–68.

Pettit DL, Perlman S, Malinow R (1994) Potentiated transmission and prevention of further LTP

by increased CaMKII activity in postsynaptic hippocampal slice neurons. Science 266:1881–

1885.

Racine R, Rose PA, Burnham WM (1977) Afterdischarge thresholds and kindling rates in dorsal

and ventral hippocampus and dentate gyrus. Can J Neurol Sci 4:273–278.

Radwan B, Dvorak D, Fenton AA (2016) Impaired cognitive discrimination and discoordination

of coupled theta-gamma oscillations in Fmr1 knockout mice. Neurobiol Dis 88:125–138.

205

Rall W (1959) Branching dendritic trees and motoneuron membrane resistivity. Exp Neurol 1:491–

527.

Rall W (1962) Electrophysiology of a dendritic neuron model. Biophys J 2:145–167.

Ramos A, Hollingworth D, Adinolfi S, Castets M, Kelly G, Frenkiel TA, Bardoni B, Pastore A

(2006) The structure of the N-terminal domain of the fragile X mental retardation protein: a

platform for protein-protein interaction. Structure 14:21–31.

Reiss AL, Freund L (1992) Behavioral phenotype of fragile X syndrome: DSM-III-R autistic

behavior in male children. Am J Med Genet 43:35–46.

Remondes M, Schuman EM (2003) Molecular Mechanisms Contributing to Long-Lasting

Synaptic Plasticity at the Temporoammonic–CA1 Synapse. Learn Mem 10:247–252.

Remondes M, Schuman EM (2004) Role for a cortical input to hippocampal area CA1 in the

consolidation of a long-term memory. Nature 431:699–703.

Ren SQ, Yan JZ, Zhang XY, Bu YF, Pan WW, Yao W, Tian T, Lu W (2013) PKCλ is critical in

AMPA receptor phosphorylation and synaptic incorporation during LTP. The EMBO Journal

32:1365–1380.

Rogers SJ, Wehner DE, Hagerman R (2001) The behavioral phenotype in fragile X: symptoms of

autism in very young children with fragile X syndrome, idiopathic autism, and other

developmental disorders. J Dev Behav Pediatr 22:409–417.

Routh BN, Johnston D, Brager DH (2013) Loss of functional A-type potassium channels in the

dendrites of CA1 pyramidal neurons from a mouse model of fragile X syndrome. J Neurosci

33:19442–19450.

206

Routh BN, Johnston D, Harris K, Chitwood RA (2009) Anatomical and electrophysiological

comparison of CA1 pyramidal neurons of the rat and mouse. J Neurophysiol 102:2288–2302.

Routh BN, Rathour RK, Baumgardner ME, Kalmbach BE, Johnston D, Brager DH (2017)

Increased transient Na(+) conductance and action potential output in layer 2/3 prefrontal

cortex neurons of the fmr1-/y mouse. J Physiol (Lond) 13:4431–4448.

Roy A, Svensson FP, Mazeh A, Kocsis B (2017) Prefrontal-hippocampal coupling by theta rhythm

and by 2-5 Hz oscillation in the delta band: The role of the nucleus reuniens of the thalamus.

Brain Struct Funct 222:2819–2830.

Rudelli RD, Brown WT, Wisniewski K, Jenkins EC, Laure-Kamionowska M, Connell F,

Wisniewski HM (1985) Adult fragile X syndrome. Clinico-neuropathologic findings. Acta

Neuropathol 67:289–295.

Rudy B, Hoger JH, Lester HA, Davidson N (1988) At least two mRNA species contribute to the

properties of rat brain A-type potassium channels expressed in Xenopus oocytes. Neuron

1:649–658.

Russell SN, Publicover NG, Hart PJ, Carl A, Hume JR, Sanders KM, Horowitz B (1994) Block

by 4-aminopyridine of a Kv1.2 delayed rectifier K+ current expressed in Xenopus oocytes. J

Physiol (Lond) 481 ( Pt 3):571–584.

Santoro MR, Bray SM, Warren ST (2012) Molecular mechanisms of fragile X syndrome: a twenty-

year perspective. Annu Rev Pathol 7:219–245.

Santos AR, Kanellopoulos AK, Bagni C (2014) Learning and behavioral deficits associated with

the absence of the fragile X mental retardation protein: what a fly and mouse model can teach

us. Learn Mem 21:543–555.

207

Savage LM, Buzzetti RA, Ramirez DR (2004) The effects of hippocampal lesions on learning,

memory, and reward expectancies. Neurobiol Learn Mem 82:109–119.

Schiller J, Schiller Y, Stuart G, Sakmann B (1997) Calcium action potentials restricted to distal

apical dendrites of rat neocortical pyramidal neurons. J Physiol (Lond) 505 ( Pt 3):605–616.

Schneider CA, Rasband WS, Eliceiri KW (2012) NIH Image to ImageJ: 25 years of image analysis.

Nat Methods 9:671–675.

Schuetze SM (1983) The discovery of the action potential. Trends in Neurosciences 6:164–168.

Schwartzkroin PA, Wester K (1975) Long-lasting facilitation of a synaptic potential following

tetanization in the in vitro hippocampal slice. Brain Res 89:107-119.

Scoville WB, Milner B (1957) Loss of recent memory after bilateral hippocampal lesions. J Neurol

Neurosurg Psychiatry 20:11–21.

Sears JC, Choi WJ, Broadie K (2019) Fragile X Mental Retardation Protein positively regulates

PKA anchor Rugose and PKA activity to control actin assembly in learning/memory circuitry.

Neurobiol Dis 127:53–64.

Shang Y, Wang H, Mercaldo V, Li X, Chen T, Zhuo M (2009) Fragile X mental retardation protein

is required for chemically-induced long-term potentiation of the hippocampus in adult mice. J

Neurochem 111:635–646.

Shi H, Wang H-Z, Wang Z (2000) Extracellular Ba2+ blocks the cardiac transient outward K+

current. American Journal of Physiology-Heart and Circulatory Physiology 278:H295–H299.

Silva AJ, Stevens CF, Tonegawa S, Wang Y (1992) Deficient hippocampal long-term potentiation

in alpha-calcium-calmodulin kinase II mutant mice. Science 257:201–206.

208

Simonsen AH, Jeune B (1972) Origin and termination of the hippocampal perforant path in the rat

studied by silver impregnation. Journal of Comparative Neurology 144:215–231.

Söllner T, Bennett MK, Whiteheart SW, Scheller RH, Rothman JE (1993a) A protein assembly-

disassembly pathway in vitro that may correspond to sequential steps of synaptic vesicle

docking, activation, and fusion. Cell 75:409–418.

Söllner T, Whiteheart SW, Brunner M, Erdjument-Bromage H, Geromanos S, Tempst P, Rothman

JE (1993b) SNAP receptors implicated in vesicle targeting and fusion. Nature 362:318–324.

Sosanya NM, Brager DH, Wolfe S, Niere F, Raab-Graham KF (2015) Rapamycin reveals an

mTOR-independent repression of Kv1.1 expression during epileptogenesis. Neurobiol Dis

73:96–105.

Sotiriou E, Papatheodoropoulos C, Angelatou F (2005) Differential expression of gamma-

aminobutyric acid--a receptor subunits in rat dorsal and ventral hippocampus. J Neurosci Res

82:690–700.

Spencer CM, Alekseyenko O, Hamilton SM, Thomas AM, Serysheva E, Yuva-Paylor LA, Paylor

R (2011) Modifying behavioral phenotypes in Fmr1KO mice: genetic background differences

reveal autistic-like responses. Autism Res 4:40–56.

Spencer WA, Kandel ER (1961) Electrophysiology of hippocampal neurons: IV. Fast

prepotentials. J Neurophysiol 24:272–285.

Squire LR (1992) Memory and the hippocampus: a synthesis from findings with rats, monkeys,

and humans. Psychol Rev 99:195–231.

Stafstrom CE, Schwindt PC, Crill WE (1984) Repetitive firing in layer V neurons from cat

neocortex in vitro. J Neurophysiol 52:264–277.

209

Steward O, Schuman EM (2003) Compartmentalized synthesis and degradation of proteins in

neurons. Neuron 40:347–359.

Storm JF (1988) Temporal integration by a slowly inactivating K+ current in hippocampal

neurons. Nature 336:379–381.

Storm JF (1993) Functional diversity of K+ currents in hippocampal pyramidal neurons. Seminars

in Neuroscience 5:79–92.

Stuart GJ, Dodt HU, Sakmann B (1993) Patch-clamp recordings from the soma and dendrites of

neurons in brain slices using infrared video microscopy. Pflugers Arch 423:511–518.

Stuart GJ, Häusser M (2001) Dendritic coincidence detection of EPSPs and action potentials. Nat

Neurosci 4:63–71.

Stuart GJ, Sakmann B (1994) Active propagation of somatic action potentials into neocortical

pyramidal cell dendrites. Nature 367:69–72.

Swanson LW, Cowan WM (1977) An autoradiographic study of the organization of the efferent

connections of the hippocampal formation in the rat. J Comp Neurol 172:49–84.

Switzer RC (2000) Application of silver degeneration stains for neurotoxicity testing. Toxicol

Pathol 28:70–83.

Takahashi H, Magee JC (2009) Pathway interactions and synaptic plasticity in the dendritic tuft

regions of CA1 pyramidal neurons. Neuron 62:102–111.

Talbot ZN, Sparks FT, Dvorak D, Curran BM, Alarcon JM, Fenton AA (2018) Normal CA1 Place

Fields but Discoordinated Network Discharge in a Fmr1-Null Mouse Model of Fragile X

Syndrome. Neuron 97:684–697.e684.

210

Tassone F, Hagerman RJ, Loesch DZ, Lachiewicz A, Taylor AK, Hagerman PJ (2000) Fragile X

males with unmethylated, full mutation trinucleotide repeat expansions have elevated levels

of FMR1 messenger RNA. Am J Med Genet 94:232–236.

Todman D (2008a) Henry Dale and the discovery of chemical synaptic transmission. Eur Neurol.

60:162–164.

Todman D (2008b) John Eccles (1903–97) and the experiment that proved chemical synaptic

transmission in the central nervous system. Journal of Clinical Neuroscience 15:972–977.

Toft AKH, Lundbye CJ, Banke TG (2016) Dysregulated NMDA-Receptor Signaling Inhibits

Long-Term Depression in a Mouse Model of Fragile X Syndrome. J Neurosci 36:9817–9827.

Toyoda I, Bower MR, Leyva F, Buckmaster PS (2013) Early activation of ventral hippocampus

and subiculum during spontaneous seizures in a rat model of temporal lobe epilepsy. J

Neurosci 33:11100–11115.

Tsay D, Dudman JT, Siegelbaum SA (2007) HCN1 channels constrain synaptically evoked Ca2+

spikes in distal dendrites of CA1 pyramidal neurons. Neuron 56:1076–1089.

Tsien RY (1980) New calcium indicators and buffers with high selectivity against magnesium and

protons: design, synthesis, and properties of prototype structures. Biochemistry 19:2396–

2404.

Turner G, Webb T, Wake S, Robinson H (1996) Prevalence of fragile X syndrome. Am J Med

Genet 64:196–197.

Ulrich D (2002) Dendritic resonance in rat neocortical pyramidal cells. J Neurophysiol 87:2753–

2759.

211

Vaidya SP, Johnston D (2013) Temporal synchrony and gamma-to-theta power conversion in the

dendrites of CA1 pyramidal neurons. Nat Neurosci 16:1812–1820.

Verkerk AJ, Pieretti M, Sutcliffe JS, Fu YH, Kuhl DP, Pizzuti A, Reiner O, Richards S, Victoria

MF, Zhang FP (1991) Identification of a gene (FMR-1) containing a CGG repeat coincident

with a breakpoint cluster region exhibiting length variation in fragile X syndrome. Cell

65:905–914.

Volk LJ, Bachman JL, Johnson R, Yu Y, Huganir RL (2013) PKM-ζ is not required for

hippocampal synaptic plasticity, learning and memory. Nature 493:420–423.

Wahlstrom-Helgren S, Klyachko VA (2015) GABAB receptor-mediated feed-forward circuit

dysfunction in the mouse model of fragile X syndrome. J Physiol (Lond) 593:5009–5024.

Wang H, Ardiles AO, Yang S, Tran T, Posada-Duque R, Valdivia G, Baek M, Chuang Y-A,

Palacios AG, Gallagher M, Worley P, Kirkwood A (2016) Metabotropic Glutamate Receptors

Induce a Form of LTP Controlled by Translation and Arc Signaling in the Hippocampus. J

Neurosci 36:1723–1729.

Watkins JC, Jane DE (2006) The glutamate story. Nature 147:S100-S108.

Weiler IJ, Irwin SA, Klintsova AY, Spencer CM, Brazelton AD, Miyashiro K, Comery TA, Patel

B, Eberwine J, Greenough WT (1997) Fragile X mental retardation protein is translated near

synapses in response to neurotransmitter activation. Proc Natl Acad Sci USA 94:5395–5400.

Weiler IJ, Spangler CC, Klintsova AY, Grossman AW, Kim SH, Bertaina-Anglade V, Khaliq H,

de Vries FE, Lambers FAE, Hatia F, Base CK, Greenough WT (2004) Fragile X mental

retardation protein is necessary for neurotransmitter-activated protein translation at synapses.

Proc Natl Acad Sci USA 101:17504–17509.

212

Wendland B, Miller KG, Schilling J, Scheller RH (1991) Differential expression of the p65 gene

family. Neuron 6:993–1007.

Wheeler DG, Barrett CF, Groth RD, Safa P, Tsien RW (2008) CaMKII locally encodes L-type

channel activity to signal to nuclear CREB in excitation-transcription coupling. J Cell Biol

183:849–863.

Wheeler DG, Groth RD, Ma H, Barrett CF, Owen SF, Safa P, Tsien RW (2012) Ca(V)1 and

Ca(V)2 channels engage distinct modes of Ca(2+) signaling to control CREB-dependent gene

expression. Cell 149:1112–1124.

Xu N-L, Harnett MT, Williams SR, Huber D, O'Connor DH, Svoboda K, Magee JC (2012)

Nonlinear dendritic integration of sensory and motor input during an active sensing task.

Nature 492:247–251.

Yu Y, Shu Y, McCormick DA (2008) Cortical action potential backpropagation explains spike

threshold variability and rapid-onset kinetics. J Neurosci 28:7260–7272.

Yun SH, Trommer BL (2011) Fragile X mice: reduced long-term potentiation and N-Methyl-D-

Aspartate receptor-mediated neurotransmission in dentate gyrus. J Neurosci Res 89:176–182.

Yuste R, Tank DW (1996) Dendritic integration in mammalian neurons, a century after Cajal.

Neuron 16:701–716.

Zalfa F, Giorgi M, Primerano B, Moro A, Di Penta A, Reis S, Oostra B, Bagni C (2003) The fragile

X syndrome protein FMRP associates with BC1 RNA and regulates the translation of specific

mRNAs at synapses. Cell 112:317–327.

213

Zhang Y, Bonnan A, Bony G, Ferezou I, Pietropaolo S, Ginger M, Sans N, Rossier J, Oostra B,

LeMasson G, Frick A (2014) Dendritic channelopathies contribute to neocortical and sensory

hyperexcitability in Fmr1(-/y) mice. Nat Neurosci 17:1701–1709.

Zhao M-G, Toyoda H, Ko SW, Ding H-K, Wu L-J, Zhuo M (2005) Deficits in trace fear memory

and long-term potentiation in a mouse model for fragile X syndrome. J Neurosci 25:7385–

7392.

Zhao W, Wang J, Song S, Li F, Yuan F (2015) Reduction of α1GABAA receptor mediated by

tyrosine kinase C (PKC) phosphorylation in a mouse model of fragile X syndrome. Int J Clin

Exp Med 8:13219–13226.

214