* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Download Ionotropic Glutamate Receptors in Aplysia californica and Molecular
Axon guidance wikipedia , lookup
Caridoid escape reaction wikipedia , lookup
Nonsynaptic plasticity wikipedia , lookup
Biology and consumer behaviour wikipedia , lookup
Premovement neuronal activity wikipedia , lookup
Nervous system network models wikipedia , lookup
Central pattern generator wikipedia , lookup
Optogenetics wikipedia , lookup
Endocannabinoid system wikipedia , lookup
Synaptic gating wikipedia , lookup
Signal transduction wikipedia , lookup
Feature detection (nervous system) wikipedia , lookup
Activity-dependent plasticity wikipedia , lookup
Neurogenomics wikipedia , lookup
Gene expression programming wikipedia , lookup
Stimulus (physiology) wikipedia , lookup
Synaptogenesis wikipedia , lookup
Pre-Bötzinger complex wikipedia , lookup
NMDA receptor wikipedia , lookup
Molecular neuroscience wikipedia , lookup
Aging brain wikipedia , lookup
Neuroanatomy wikipedia , lookup
Channelrhodopsin wikipedia , lookup
University of Miami Scholarly Repository Open Access Dissertations Electronic Theses and Dissertations 2017-04-25 Ionotropic Glutamate Receptors in Aplysia californica and Molecular Changes in the Aging Nervous System Justin B. Greer University of Miami, [email protected] Follow this and additional works at: http://scholarlyrepository.miami.edu/oa_dissertations Recommended Citation Greer, Justin B., "Ionotropic Glutamate Receptors in Aplysia californica and Molecular Changes in the Aging Nervous System" (2017). Open Access Dissertations. 1819. http://scholarlyrepository.miami.edu/oa_dissertations/1819 This Open access is brought to you for free and open access by the Electronic Theses and Dissertations at Scholarly Repository. It has been accepted for inclusion in Open Access Dissertations by an authorized administrator of Scholarly Repository. For more information, please contact [email protected]. UNIVERSITY OF MIAMI IONOTROPIC GLUTAMATE RECEPTORS IN APLYSIA CALIFORNICA AND MOLECULAR CHANGES IN THE AGING NERVOUS SYSTEM By Justin B. Greer A DISSERTATION Submitted to the Faculty of the University of Miami in partial fulfillment of the requirements for the degree of Doctor of Philosophy Coral Gables, Florida May 2017 ©2017 Justin B. Greer All Rights Reserved UNIVERSITY OF MIAMI A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy IONOTROPIC GLUTAMATE RECEPTORS IN APLYSIA CALIFORNICA AND MOLECULAR CHANGES IN THE AGING NERVOUS SYSTEM Justin B. Greer Approved: ________________ Lynne Fieber, Ph.D. Associate Professor of Marine Biology and Ecology _________________ Michael Schmale, Ph.D. Professor of Marine Biology and Ecology ________________ Marjorie Oleksiak, Ph.D. Associate Professor of Marine Biology and Ecology _________________ Sathya Puthanveettil, Ph.D. Professor of Neuroscience Scripps Research Institute ________________ Sawsan Khuri, Ph.D. Research Assistant Professor Department of Computer Science _________________ Guillermo Prado, Ph.D. Dean of the Graduate School GREER, JUSTIN B. (Ph.D., Marine Biology and Ecology) Ionotropic Glutamate Receptors in Aplysia californica (May 2017) and Molecular Changes in the Aging Nervous System Abstract of a dissertation at the University of Miami. Dissertation supervised by Professor Lynne Fieber. No. of pages in text. (157) Aplysia californica is a marine snail with well-defined neural circuits and a relatively short life span of one year, making it a useful model for studies of neural function and aging. Many of these neural circuits likely use L-Glutamate as the excitatory neurotransmitter. The goal of this dissertation was to molecularly characterize ionotropic L-Glutamate receptors in A. californica, and describe molecular changes in these receptors during aging in well-defined neural circuits. The evolutionary relationship between ionotropic L-Glutamate receptors of A. californica and vertebrates was first studied to evaluate the relevance of A. californica as a model of ionotropic L-Glutamate receptor function. Genes belonging to the N-methylD-aspartate receptor (NMDAR) and α-Amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptor (AMPAR) subtypes were found to be particularly well-conserved between A. californica and vertebrates. Both subtypes are critical for synaptic plasticity underlying learning and memory. These results suggest conserved function of NMDAR and AMPAR over large evolutionary distances, thus making A. californica a suitable model for studying their function. Expression of all identified ionotropic L-Glutamate receptors in all A. californica nervous system tissues emphasized their important role in the nervous system. Next, age-related changes in L-Glutamate receptor expression were examined in two sensory neuron clusters. Several NMDAR subtype genes exhibited reduced expression in aged animals compared to sexually mature animals. This suggested that reduced NMDAR in sensory neurons may contribute to previously described declines in L-Glutamate current amplitude and reduced reflexes of aged animals. Finally, whole transcriptome expression changes in aged sensory neurons were studied using RNASeq. Transcription of ion channel genes and genes related to nervous system function were significantly affected in aged compared to mature animals. These alterations may negatively affect synaptic transmission and contribute to previously observed deficits in nervous system function of aged A. californica. TABLE OF CONTENTS Page LIST OF FIGURES ..................................................................................................... v LIST OF TABLES ...................................................................................................... viii Chapter 1 INTRODUCTION ........................................................................................... 1.1 Ionotropic glutamate receptors in the nervous system ............................... 1.2 Aplysia as a model for aging and learning and memory ........................... 1 1 6 2 PHYLOGENETIC ANALYSIS OF IONOTROPIC GLUTAMATE RECEPTOR GENES IN THE BILATERIA, WITH SPECIAL NOTES ON APLYSIA CALIFORNICA ............................................................................ 2.1 Summary .................................................................................................... 2.2 Background ................................................................................................ 2.3 Materials and Methods ............................................................................... 2.4 Results ........................................................................................................ 2.5 Discussion .................................................................................................. 15 15 16 19 22 31 ALTERED EXPRESSION OF IONOTROPIC L-GLUTAMATE RECEPTORS IN AGED SENSORY NEURONS OF APLYSIA .................. 3.1 Summary .................................................................................................... 3.2 Background ................................................................................................ 3.3 Materials and Methods ............................................................................... 3.4 Results ........................................................................................................ 3.5 Discussion .................................................................................................. 47 47 47 52 57 63 3 4 TRANSCRIPTIONAL CHANGES DURING AGING IN SENSORY NEURONS INNERVATING TWR IN APLYSIA ........................................ 81 4.1 Summary .................................................................................................... 81 4.2 Background ................................................................................................ 81 4.3 Materials and Methods ............................................................................... 88 4.4 Results ........................................................................................................ 95 4.5 Discussion .................................................................................................. 100 5 CONCLUSIONS ............................................................................................. 123 REFERENCES ............................................................................................................ 132 iii Appendix A CHAPTER 2 SUPPLEMENTARY FIGURES ............................................... 151 B CHAPTER 4 SUPPLEMENTARY DATA FILES ......................................... 157 iv LIST OF FIGURES Page Chapter 2 2.1 Phylogeny of species used for iGluR analysis ........................................... 42 2.2 Bootstrap consensus phylogenetic tree ...................................................... 43 2.3 Conserved motifs in Aplysia and H. sapiens ............................................. 44 2.4 Hydrophobicity plot of Grin1 in H. sapiens and Aplysia........................... 45 2.5 iGluR expression levels in Aplysia ganglia ............................................... 46 Chapter 3 3.1 Behavioral and weight correlates of aging in two cohorts of Aplysia ....... 71 3.2 Differential expression of iGluR subunits within abdominal ganglia and sensory neuronal clusters ................................................................................. 72 3.3 Calculated number of iGluR transcripts in different ganglia ..................... 73 3.4 Relative expression of iGluR subunits in PVC sensory neurons of Mature and Aged II animals from two cohorts ................................................................... 74 3.5 Absolute transcript quantification of iGluR subunits in PVC sensory neurons of Mature and Aged II animals from cohort 1 ................................................. 75 3.6 Relative expression of iGluR subunits in BSC sensory neurons of Mature and Aged II animals from two cohorts ................................................................... 76 3.7 Absolute transcript quantification of iGluR subunits in BSC sensory neurons of Mature and Aged II animals from cohort 1 ................................................. 77 3.8 Calculated number of iGluR transcripts in Mature and Aged II whole abdominal ganglion of cohort 1 ............................................................ v 78 3.9 Percent composition of NMDAR subtype subunits ................................... 79 3.10 Percent composition of AMPAR subtype subunits ................................. 80 Chapter 4 4.1 Behavioral assessments of aging in cohorts used for RNASeq ................. 112 4.2 Principal component analysis (PCA) of the first two principal components in M and AII PVC neurons .................................................................................. 113 4.3 Overlap of DE genes identified with STAR and Tophat2 ......................... 114 4.4 Scatterplot of mean expression of genes analyzed by STAR vs. their log-fold change ................................................................................................ 115 4.5 Heatmap of the top 50 most significantly DE genes from PVC neuron clusters of M and AII animals ....................................................................................... 116 4.6 Expression of iGluR subunits in M and AII PVC neurons ........................ 117 4.7 Composition of NMDAR and AMPAR in M and AII PVC neurons ........ 118 4.8 The pathways and activities most affected by aging ................................. 119 4.9 Mean expression vs log2 fold change for ion channel associated genes ... 120 4.10 Mean expression vs log2 fold change for stress response genes .............. 121 4.11 Comparison of DE genes in RNASeq and qPCR analyses ...................... 122 Appendix A S2.1 NMDAR subtype only tree ...................................................................... 151 S2.2 AMPAR subtype only tree....................................................................... 152 S2.3 Kainate receptor subtype only tree .......................................................... 153 S2.4 Aplysia iGluR protein tree ....................................................................... 154 vi S2.5 Representative AMPA subunit TMD Hydrophobicity ............................ 155 S2.6 Representative kainate subunit TMD Hydrophobicity ............................ 156 vii LIST OF TABLES Page Chapter 2 2.1 Primers used for qPCR............................................................................... 38 2.2 Placement of Protostomia iGluR into subtypes based on phylogenies ...... 39 2.3 Number of iGluR genes in the protostomes and vertebrates ..................... 40 2.4 Sequence similarities in transmembrane domains of Grin1 between different protostomes and H. sapiens ............................................................................. 41 Chapter 4 4.1 Primer sequences for qPCR ....................................................................... 110 4.2 Read depth and mapping statistics using STAR ........................................ 111 viii Chapter 1: Introduction 1.1 Ionotropic glutamate receptors in the nervous system The mechanisms underlying learning and memory are vastly complex and involve a multitude of molecular changes that may result in learned behaviors. Mammalian systems may be difficult to work use to understand learned behaviors due to highly complex nervous systems; the human brain consists of one trillion neurons and highly complicated neuronal networks, making it difficult to elucidate learning and memory processes. The mammalian hippocampus is believed to be heavily involved in learning, but contains ~40 million neurons (Gundersen, et al. 2001). Tracing how sensory information is received by the hippocampus and how this information is transmitted into behavior through this large neural network is immensely difficult, even without the added complications of learning and memory events. These confounding factors facilitated the need for a simpler model system to understand learning and memory basics. This led Eric Kandel in the early 1960’s to study Aplysia californica (Aplysia), culminating in the 2000 Nobel Prize in Physiology and Medicine for his work on the physiological basis of memory formation and storage. Aplysia is a model system for learning and memory with several distinct advantages compared to mammals. The Aplysia nervous system contains ~20,000 neurons, and simple forms of behavior can be followed from stimulus to behavior with less complexity and involvement of interneurons. Aplysia neurons are some of the largest in the animal kingdom at up to of 1 mm in diameter, can be seen with the naked eye, and are easily reidentifiable in consecutive individual animals, allowing for direct 1 2 electrophysiological recordings from individual cells. Large neurons also allow for nucleic acid isolation from few cells, allowing for targeted studies on subsets of neurons. Glutamate (L-Glu) is the principal fast excitatory neurotransmitter in the central nervous system (CNS) of nearly all animals. In the Aplysia CNS, there is abundant evidence that these sensory neurons use L-Glu as their primary excitatory neurotransmitter. The nerve fibers that connect sensory neurons to other neurons are strongly immunoreactive for L-Glu (Levenson, Sherry, et al. 2000). These sensory neurons have high expression of L-Glu-like receptors (Dale and Kandel 1993; Ha, et al. 2006), and their [L-Glu] is high at 29 mM, compared to 3 mM in unidentified neurons of the abdominal ganglion (Drake, et al. 2005), suggesting that L-Glu mediated responses are prevalent in the Aplysia nervous system. L-Glu receptors are classified into two types, ionotropic and metabotropic receptors. Metabotropic L-Glu receptors (mGluR) are G-protein-coupled receptors in which binding of L-Glu activates intracellular cascades and modification of intracellular proteins; they are not ion channels. In contrast, ionotropic L-Glu receptors (iGluR) are ion channels that convey the majority of fast excitatory signal transmission modulated by L-Glu and have been implicated in most aspects of CNS development and function (Traynelis, et al. 2010a). The binding of L-Glu to iGluR opens transmembrane ion channels that allow ion flux across the plasma membrane and depolarization of the postsynaptic cell. Sufficient agonist-receptor binding results in superthreshold depolarization that triggers an action potential, thereby transmitting information. iGluR also play a significant role in synaptic plasticity associated with learning (Roche, et al. 1994; Kandel 2001; Paoletti, et al. 2013). 3 iGluR are divided into three subtypes; N-methyl-D-aspartate (NMDA), kainate, and α-Amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA), based on selective agonists. The latter two subtypes are known as non-NMDA receptors (nonNMDAR). Overall amino acid identity of iGluR subunits across the three families in mammals is 20-30%, but each contain common structural features that place them together into a single large superfamily (Kew and Kemp 2005). Function and expression varies between subtypes, but there are several common features of all iGluR. In eukaryotes iGluR are tetrameric in structure, consisting of four subunits, and are formed as a dimer of dimers (Tichelaar, et al. 2004). Each complete receptor contains an agonistbinding domain, three transmembrane domains, an extracellular N terminus, and an intracellular C terminus (Traynelis, et al. 2010a). Studies on receptor structure have been limited primarily to vertebrates, and it is unknown if Aplysia iGluR are comprised of this same structure. Insights into composition of these receptors are an important step in furthering the Aplysia model. Although they are related in structure, there are several distinct electrophysiological and activation requirements that are unique to individual iGluR subtypes. Activation kinetics of non-NMDAR show that activation, deactivation, and desensitization occur on a millisecond timescale, while NMDAR display these characteristics much slower, on a timescale of tens to hundreds of milliseconds (Hansen, et al. 2007; Paoletti and Neyton 2007). AMPAR are permeable to Na+, K+, as well as infrequently permeable to Ca2+, depending on the subunit structure, with most AMPAR containing the Ca2+ impermeable Gria2 subunits, and are thus impermeable to Ca2+ 4 (Greger, et al. 2007). Conversely, Ca2+ is the principal ion transported by NMDAR in vertebrate models (Mayer and Westbrook 1987). In mammals, NMDAR are obligatory heterotetramers composed of various arrangements of three subunits, Grin1, Grin2, and Grin3 (Glutamate Receptor Ionotropic/NMDA). This may be due to the fact that, unlike AMPAR, NMDAR require binding of the co-agonist glycine for activation. In mammals, complete NMDAR are comprised of two Grin1 subunits responsible for glycine binding (or d-serine), and either two Grin2 or Grin3 subunits binding L-Glu, although NMDAR containing all three subunits have also been reported (Traynelis, et al. 2010a). AMPAR preferentially form heterotetramers, but are also functional as a homotetramer, containing four of the same subunit (Sobolevsky, et al. 2009). Vertebrate NMDAR are also subject to allosteric modulation by extracellular Mg2+, which blocks the channel constitutively in a voltage-dependent manner (Mayer, et al. 1984). The cation must be excluded by depolarization for Ca2+ to enter the channel. They are also modulated by other small molecules or ions, such as H+ and Zn2+, which are endogenously present in the CNS and act as allosteric modulators (Paoletti 2011). Similar to the work on receptor structure, the majority of these findings were demonstrated in vertebrates, and modifications, if they occur, are unclear in Aplysia. High Mg2+ concentrations in seawater make Mg2+ unlikely to be an effective mechanism for constitutive channel block without very high energy cost to remove it. Seawater contains large amounts of glycine that, similarly, make it unlikely to be an effective coagonist. Aplysia are osmoconformers not believed to selectively exclude ions, resulting in high endogenous Mg2+ and glycine levels in their CNS. 5 The Grin1 subunit is encoded by a single gene in mammals, and contains a varying number of splices depending on species, with as many as eight splice variants in humans and two in mice (Paoletti 2011). Grin2 subunits are represented by four members, each encoded by its own gene, and Grin3 has two members each encoded by a separate gene. Currently, two Grin-like receptors and one Grin2-like receptor have been cloned in Aplysia (Ha, et al. 2006). Analysis in Prosite (Letunic, et al. 2012) predicts binding sites for both L-Glu and glycine in the Aplysia Grin1-like genes. Grin3-like subunits have yet to be found in Aplysia. Incorporation of various subunit compositions plays a major role in receptor functional properties, and changes in expression of subunit (and isoform) mRNA is prevalent during development (Akazawa, et al. 1994; Laurie and Seeburg 1994). Functionally, permeability of NMDAR to Ca2+ as well as the extent of Mg2+ block is directly related to the subunits that comprise the receptor (Schneggenburger 1996). The intracellular N-terminal domain, ~400 amino acids, is thought to be the major determinant of subtype-specific assembly (Paoletti 2011). For several reasons, classification of iGluR subunits into subtypes is less clear in invertebrates. First, there is high divergence of these sequences from their vertebrate counterparts, making molecular phylogenetics difficult because invertebrate iGluR cluster together as a separate node from vertebrate iGluR using full-length sequences. Further confounding classification of Aplysia iGluR, subtype specific agonists are less effective in Aplysia neurons than in vertebrates. The structure of iGluR in Aplysia are sufficiently different that neurons do not respond to AMPA as an agonist (Trudeau and Castellucci 1993; Carlson and Fieber 2012). Kainate analogs do elicit depolarization and 6 non-desensitizing inward current in glutamatergic BSC neurons and abdominal ganglion neurons (Trudeau and Castellucci 1993; Wang, et al. 2013), and NMDA elicits small amplitude currents in BSC neurons (Carlson and Fieber 2012). Of the three iGluR subtypes, AMPAR and NMDAR have been implicated in underlying physiological changes that accompany learning and memory. Aplysia and other invertebrate model organisms have been integral in furthering research on the underlying mechanisms responsible for learning and memory. However, there is a current lack of evidence for the subunits that may play the roles of AMPAR and NMDAR in invertebrates. Identification of these subunits could allow for could allow for targeted studies of these iGluR subunits to gain further insights into iGluR physiology in invertebrate nervous systems. 1.2 Aplysia as a model for aging and learning and memory Aging is associated with sensory and motoneuron function impairments that result in deterioration of cognitive function and reflex systems in aged animals. Significant neuron loss is not proposed to be a major component of reduced neuronal function with age (Hof and Morrison 2004). Rather, a reduced number of synaptic contacts and reduced dendritic spine density may alter the strength of synaptic connections (Wallace, et al. 2007; Dumitriu, et al. 2010). Reduced postsynaptic density in axiospinous synapses in hippocampal neurons has been implicated in learning impairments in aged rats (Nicholson, et al. 2004). Reduced excitatory post synaptic potential amplitude in the aged hippocampus suggests that neuronal communication is reduced during aging (Landfield, et al. 1986; Boxer, et al. 1988; Deupree, et al. 1993). Significant decreases in conduction 7 velocity in several areas of the brain may also be caused by changes in the myelin sheath (Kanda, et al. 1986; Boxer, et al. 1988; Peters 2002). Taken together, these changes in structure and function of neuronal pathways may lead to age-related impairments in the aging brain. Aplysia provides a unique model system with several advantages that make it an excellent model for studies of aging at the behavioral and physiological levels. At the National Resource for Aplysia animals have a lifespan of approximately one year, allowing for study of both mature and aged animals in a relatively short period of time (Gerdes and Fieber 2006). Sexual maturity occurs at ~7-8 months at 13-15°C at a density of 5 animals per cage, but this can be manipulated with changes in either temperature or cage density (Stommes, et al. 2005). Body length and weight increase linearly with age through first sexual maturity, then plateau as food mass in and spawn mass out become approximately equal. Egg mass production reaches its peak ~1-2 months after first sexual maturity (Capo, et al. 2003; Gerdes and Fieber 2006). A decrease in body mass and reduced egg production in the months preceding death signify the onset of senescence (Hirsch and Peretz 1984; Gerdes and Fieber 2006). Arousal that results in increased heart rate as a response to food stimuli was reduced in aged Aplysia (Bailey, et al. 1983). Given the simple nervous system of Aplysia, the underlying neuronal circuits of several behaviors have been studied in detail (Pinsker, et al. 1970; Byrne, Castellucci and Kandel 1978; Peretz, et al. 1982; Walters, et al. 1983a, b; Peretz, et al. 1984). Many of these behaviors are altered in aged animals, and physiological changes in the corresponding neural circuits have been investigated. In the gill-withdrawal reflex (GWR), a weak stimulus applied to the siphon causes the siphon and gill to withdraw into 8 the mantle for protection. GWR responsiveness to a stimulus is reduced in aged animals and habituation is impaired compared to young animals (Rattan and Peretz 1981). Physiologically, retraction of the siphon into the mantle is innervated by 24 sensory neurons that connect with 6 motoneurons in the abdominal ganglion that withdrawal the siphon (Carew, et al. 1971; Byrne, Castellucci, Carew, et al. 1978). Impairment of GWR in aged animals has been correlated with decreased input resistance and decreased size of post synaptic potentials evoked by gill stimulation (Rattan and Peretz 1981). This resulted in reduced excitability and synaptic communication in the GWR circuit and provided evidence that reduced physiology in aged neurons directly involved in behavioral deficits. Aging of the tail withdrawal reflex (TWR) has also been investigated both behaviorally and physiologically. TWR is a monosynaptic reflex in which an electrical or mechanical stimulus to the tail results in retraction of the tail into the body as a defense mechanism (Walters, et al. 1983a). The time to relax the tail after tail touch, a measure of TWR, increased significantly in aged animals and amplitude of tail retraction decreased (Kempsell and Fieber 2014). Tail mechanosensory neurons located in the ventral caudal (PVC) region of the pleural ganglion are activated in response to tail touch. PVC neurons innervate monosynaptic connections to tail motoneurons in the ipsilateral pedal ganglion, withdrawing the tail and completing the reflex (Walters, et al. 1983a). The underlying physiology of decreased TWR with age indicated that aging of PVC sensory neurons may be responsible, and may involve reduced responsiveness to glutamatergic agonists (Kempsell and Fieber 2014). PVC neurons had robust, excitatory iGluR responses to 9 applied L-Glu and D-Asp that declined significantly during aging, whether studied in isolated (cultured) neurons or in in situ preparations (Fieber, et al. 2010; Kempsell and Fieber 2015b). There are also alterations in behaviors and physiology of the biting reflex with age. Biting amplitude was significantly reduced and latency of the response increased in aged animals compared to animals at maturity. Buccal S cluster (BSC) sensory neurons of the biting reflex in aged animals had reduced amplitude and current densities to L-Glu. Thus, reduced current density and amplitude occured sensory neurons of TWR and the biting reflex in aged animals in response to L-Glu. This suggests molecular changes in iGluR that mediate glutamatergic responses. Decreased iGluR to bind L-Glu, a change in sensitivity to agonist, or a change in receptor structure could lead to reduced glutamate physiology in aged PVC and BSC sensory neurons. In Aplysia, both the GWR and the TWR exhibit sensitization and habituation, two forms of non-associative learning in which repetition of a stimulus results in changes in the strength of the response. Sensitization results in amplification of the response, while progressive diminution of a response constitutes habituation. These modifications, known as synaptic plasticity, are the primary foundation of learning and memory. Sensitization and habituation of the aforementioned neuronal circuits can be observed both behaviorally and electrophysiologically in Aplysia. Behaviorally, repeated noxious electrical stimulation of the tail results in an incrementally increased amplitude of TWR that is characteristic of sensitization (Kempsell and Fieber 2014). Electrophysiologically, sensitization leads to an increase in the number of spikes in the motor neuron, and an increase in the amplitude of the monosynaptic excitatory postsynaptic potential (EPSP) in 10 the motoneuron (Carew, et al. 1981; Walters, et al. 1983b; Kempsell and Fieber 2015b). The duration of the memory is a function of the number of repetitions of the stimulus and occurs in two distinct forms. One or a few repetitions induces short-term facilitation (STF) lasting minutes or hours, and many repetitions induce long-term facilitation (LTF) that persists for 24 hours or several days. Each form of memory requires different pathways for activation. STF behavioral and electrophysiological changes are expressed in the presence of protein synthesis inhibitors on both the pre- and postsynaptic neurons, suggesting fastacting mechanisms that are not dependent upon protein synthesis characterize STF (Schwartz, et al. 1971). Electrical stimulation of presynaptic neurons resulted in a doubling of cAMP in the neurons, and presynaptic STF was induced by injection of cAMP without prior stimulation (Cedar and Schwartz 1972; Brunelli, et al. 1976). Increased cAMP resulted in activation of the catalytic subunit of cAMP-dependent protein kinase A (PKA). PKA phosphorylates presynaptic K+ channels, reducing the number of active channels, resulting in greater Ca2+ influx per action potential, and subsequent increases in amount of transmitter release by terminals of the sensory cell (Castellucci, et al. 1980; Klein and Kandel 1980). STF memory appears to be primarily due to these presynaptic mechanisms, resulting in enhanced synaptic strength through increased neurotransmitter release at the synapse. Application of the monoamine neurotransmitter serotonin (5-HT) mimics behavioral training and resulted in increased cAMP and STF or LTF, depending on the number of spaced applications (Cedar and Schwartz 1972). Later, identification of serotonergic neurons in the Aplysia cerebral ganglia that participate in presynaptic 11 facilitation of sensory cells in the abdominal ganglia of the gill-withdrawal reflex confirmed its role in vivo as the neurotransmitter modulating facilitation (Mackey, et al. 1989). In contrast to STF, where enhancement of transmission is dependent only on second messenger-induced phosphorylation, LTF is dependent on both phosphorylation and protein synthesis. Inhibition of protein synthesis in the presynaptic or postsynaptic neuron blocks induction of LTF, suggesting a different mechanism of action and requirement for protein synthesis in both neurons (Castellucci, et al. 1989; Villareal, et al. 2007). 5-HT release from interneurons results in intracellular increases in both cAMP and PKA, as in STF. With repeated pulses of 5-HT invoking LTF, however, PKA and mitogen-activated protein kinase (MAPK) translocate to the nucleus and also phosphorylate cAMP response element binding protein (CREB1) there (Bacskai, et al. 1993). CREB-1 is a transcription factor and binds to a cAMP response element in the promoter of target genes (Kandel 2001). CREB2, a repressor of CREB1-mediated gene expression, acts jointly with CREB1 in coordinated regulation for initiation of LTF (Bailey, et al. 1996). Two other positive transcriptional regulators, the CAAT box enhancer-binding protein (C/EBP) and activation factor (Ap/AF), known as the immediate response genes, are also activated with multiple applications of 5-HT (Alberini, et al. 1994). Activation of immediate response genes stimulates transcription of downstream genes that contribute to the formation of new synaptic connections between sensory and motor neurons, a process not seen in STF (Bailey and Chen 1988b, a). Sensorin, a sensory neuron-specific neuropeptide, is rapidly released from presynaptic 12 terminals, and expression is enhanced with repeated 5-HT applications (Hu, et al. 2004). Sensorin, CREB1, and CREB2 show dynamic regulation during the first 24 hour after LTF induction, and timing of expression appears to vary depending on the gene (Hart, et al. 2011; Liu, et al. 2011). Presynaptic structural changes observed in LTF in culture do not occur without the presence of the motoneuron, suggesting signals from the postsynaptic neuron are a requirement for expression of LTF (Glanzman, et al. 1990). Injection of 1,2-Bis-(2aminophenoxy)ethane-N,N,N’,N’-tetraacetic acid (BAPTA), a Ca2+ chelator, prevented increased sensorin expression in the presynaptic cell body. This maneuver blocked LTF, indicating elevated postsynaptic intracellular Ca2+ also is required for LTF (Cai, et al. 2008). Regulation of postsynaptic NMDAR and AMPAR have been proposed to significantly contribute to LTF. In Drosophila knockout of the Grin1 gene disrupted olfactory learning, and the phenotype was rescued by transgene induction of Grin1 (Xia, et al. 2005). This behavioral experiment was not verified electrophysiologically due to the small size of Drosophila neurons. In the mammalian hippocampus, LTF in the Schaffer collateral pathway is NMDA receptor dependent (Collingridge, et al. 1988). An intracellular rise in Ca2+ is primarily mediated by NMDAR, and this rise in postsynaptic intracellular Ca2+ is a necessary trigger for LTF (Malenka and Bear 2004). Calcium/calmodulin-dependent protein kinase II (CaMKII) is required as a mediator for NMDAR-dependent long term potentiation (Nicoll and Malenka 1999), an electrophysiological proxy for the synaptic strengthening that is a component of learning. Many studies have also found an increase in the number of functional AMPAR in the 13 synapse accompanying LTF (Trudeau and Castellucci 1995; Zhao, et al. 2003). Previous work in Aplysia has shown that application of the AMPAR antagonist DNQX reversed LTF that had been induced by 5-HT, indicating increased functional expression of AMPAR in the motoneuron were necessary to maintain LTF (Chitwood, et al. 2001). Induction of both STF and LTF are impaired in aged Aplysia. In the gill withdrawal reflex long-term retention of habituation is impaired, and acquisition of LTF of this reflex is absent in aged animals (Rattan and Peretz 1981). In TWR, sensitizing tail shocks increased excitability of PVC neurons and enhanced neurotransmitter release at the sensory to motoneuron synapse, strengthening the neuronal response to future stimuli. In aged animals tail shocks failed to enhance TWR amplitude or duration (Kempsell and Fieber 2015b). Additionally, enhanced excitation of PVC neurons and increased EPSP amplitude in motoneurons is not observed. These data suggest regulation of genes vital for induction of facilitation may be disrupted in the aged Aplysia nervous system. Over the past 40 years the Aplysia model has been a pivotal tool in advancing our understanding of the behavioral changes in the aging nervous system and their corresponding electrophysiological alterations. This dissertation attempts to address several outstanding questions that still remained. In Chapter 2, the evolutionary relationship between vertebrate and invertebrate iGluR subunits is addressed. These data give insights into the potential composition of iGluR in Aplysia, an important step in understanding their similarity to well-studied vertebrate iGluR. Identification of subunits most closely related to vertebrate iGluR subtypes allowed for hypotheses of iGluR subunits that may be involved in learning and memory in Aplysia and other invertebrate model organisms. In Chapter 3, transcriptional changes of iGluR subunits was examined 14 to identify subunits that may be involved in altered L-Glu physiology of aged sensory neurons. In Chapter 4, whole-transcriptome profiling of PVC sensory neurons from mature and aged Aplysia was assessed via RNASeq to characterize large-scale transcriptional correlates of aging. Together, these investigations provided a greater understanding iGluR in Aplysia and the molecular effects of aging in the nervous system. This dissertation revealed that reduced physiological function with age in sensory neurons may be the result of significant declines in expression of components critical for neuronal homeostasis. Previously described declines in L-Glu physiology of aged Aplysia were correlated with reduced expression of iGluR that mediate the response. Significant alterations in ion channel associated genes in aged animals may be crucial to observed sensory neuron declines in excitability. NMDAR and AMPAR subtypes were found to be highly conserved between Aplysia and vertebrates, providing evidence for conserved iGluR function, thus making Aplysia a relevant model for iGluR function in the nervous system. Chapter 2: Phylogenetic analysis of ionotropic L-glutamate receptor genes in the Bilateria, with special notes on Aplysia californica 2.1 Summary Invertebrate organisms such as Aplysia californica (Aplysia) are well-studied models for ionotropic L-Glutamate receptor (iGluR)-mediated function, yet no studies to date have analyzed the evolutionary relationships between iGluR genes in Aplysia and those in vertebrates, to identify genes that may mediate plasticity. We conducted a thorough phylogenetic analysis spanning Bilateria to elucidate these relationships. The expression status of iGluR genes in the Aplysia nervous system was also examined. Our analysis shows that N-methyl-D-aspartate receptor (NMDAR) and α-Amino-3-hydroxy5-methyl-4-isoxazolepropionic acid receptor (AMPAR) subtypes were present in the common bilaterian ancestor. NMDAR genes show very high conservation in motifs responsible for forming the conductance pore of the ion channel. The number of NMDAR subunits is greater in vertebrates due to an increased number of splice variants and an increased number of genes, likely due to gene duplication events. AMPAR subunits form an orthologous group, and there is high variability in the number of AMPAR genes in each species due to extensive taxon specific gene gain and loss. qPCR results show that all 12 Aplysia.iGluR subunits are expressed in all nervous system ganglia. Orthologous NMDAR subunits in all species studied suggests conserved function across Bilateria, and potentially a conserved mechanism of neuroplasticity and learning. Our results suggest a significant role for L-Glu mediated responses throughout 15 16 the Aplysia nervous system, consistent with L-Glu’s role as the primary excitatory neurotransmitter. 2.2 Background L-Glu is the most abundant neurotransmitter in the vertebrate brain (Meldrum 2000), and exerts most of its effects by binding to different postsynaptic ligand-gated receptors. L-Glu receptors are classified into two types, ionotropic and metabotropic receptors (Petralia, et al. 1996). Metabotropic L-Glu receptors (mGluR) are G-proteincoupled receptors in which binding of L-Glu activates intracellular cascades and modification of intracellular proteins. Ionotropic L-Glu receptors (iGluR) convey the majority of fast excitatory signal transmission and have been implicated in most aspects of central nervous system (CNS) development and function (Traynelis, et al. 2010b). The binding of L-Glu to iGluR opens transmembrane ion channels that allow ions to cross the plasma membrane leading to depolarization of the postsynaptic cell and triggering of action potentials, thereby transmitting synaptic information. iGluR play important roles in synaptic plasticity, which is the ability of a synapse to strengthen or weaken its interactions with others over time in response to changes in activity. This feature of iGluR is believed to be a key mechanism underlying learning and memory (Danysz, et al. 1995). Several features of iGluR in vertebrates have been revealed using model systems such as rats and mice. Vertebrate iGluR are divided into three subtypes according to selective agonists: N-methyl-D-aspartate receptors (NMDAR), kainate, and α-Amino-3hydroxy-5-methyl-4-isoxazolepropionic acid receptors (AMPAR). Each subtype of iGluR 17 is composed of 4 subunits that form a dimer of dimers to create the functional protein (Tichelaar, et al. 2004). In a phylogenetic analysis of human, rat, and mouse iGluR subunits, each of the 3 different subtypes formed a monophyletic clade, with functional iGluR proteins made only with subunits within each individual clade (Lipsky and Goldman 2003). Overall amino acid identity of iGluR subunits across the 3 subtypes in mammals is only 20-30%, but all subtypes contain common structural features that place them together into a single large superfamily (Kew and Kemp 2005). A fourth iGluR subtype, called delta receptors, show low sequence similarity with other iGluR and do not open ion channels, but are believed to bind D-serine and glycine (Naur, et al. 2007). Invertebrate model species such as Aplysia californica (Aplysia), Drosophila melanogaster (Drosophila), and Caenorhabditis elegans (C. elegans) have been used extensively in studies of iGluR mediated transmission in the nervous system since the 1960’s due to several distinct advantages over vertebrate models (Kandel 2001). Their nervous systems contain 302-135,000 neurons, compared to 1x1011 in the human brain (Kosinski and Zaremba 2007; Herculano-Houzel 2009; Alivisatos, et al. 2012), and simple neuronal circuits underlying various behaviors have been described (Walters, et al. 1983a; Chalfie, et al. 1985; Simpson 2009; Piggott, et al. 2011). There is ample evidence suggesting that L-Glu is the major neurotransmitter in many neural circuits in these species (Dale and Kandel 1993; Levenson, Endo, et al. 2000; Brockie, et al. 2001; Marrus, et al. 2004). These advantages have resulted in widespread use of Aplysia, Drosophila, C. elegans, and other models to study iGluR mediated synaptic plasticity, because learned behaviors can be correlated with both molecular and physiological synaptic changes between sensory- and motoneurons. 18 Physiologically, invertebrate model organisms have greatly enhanced our understanding of L-Glu mediated synaptic plasticity. Many plasticity related changes characterized in invertebrate models subsequently have been demonstrated to occur in the more complex vertebrate hippocampus (Malenka and Bear 2004). For example, AMPAR and NMDAR have been implicated in learning and memory in vertebrates, Aplysia, and D. malanogaster (Lisman, et al. 1998; Lee, et al. 2003; Xia, et al. 2005). Despite the physiological use of invertebrate model species for iGluR-mediated responses, an outstanding question remains as to which iGluR genes are likely to play the functional role of vertebrate NMDAR and AMPAR during plasticity. To address this question we have conducted a phylogenetic analysis spanning all three Bilateria superclades in order to identify iGluR genes in invertebrate model species that are orthologous with vertebrate iGluR, and thus more likely to be functionally similar. Bilateria are organized into three superclades based on embryology, morphology and molecular data: Ecdysozoa (including arthropods like D. melanogaster, nematodes like C. elegans), Lophotrochozoa (including molluscs such as Aplysia), and Deuterostomia (including chordates like rats, mice, humans) (Philippe, et al. 2005). Lophotrochozoa and Ecdysozoa are sister-clades that together form the Protostomia (Halanych 2004; Dunn, et al. 2008). Most studies support the monophyly of these three superclades, and thus a common ancestor for all Bilateria. Several recent studies support a monophyletic grouping of Deuterostomia, with Protostomia as a sister clade (Paps, et al. 2009). Identification of Protostomia iGluR genes homologous to vertebrate NMDAR and AMPAR subunits can allow for predictions of subunits that may be involved in observed 19 synaptic plasticity. This analysis can also add useful information to our poor understanding of subtype-specific agonists in protostomes (Trudeau and Castellucci 1993; Fox and Lloyd 1999; Kimura, et al. 2001). Aplysia is a model organism with a long history of studies of iGluR-mediated nervous system function, in particular for learning and memory paradigms (Bartsch, et al. 1995; Kandel 2001; Kempsell and Fieber 2015b, a). Aplysia NMDAR subunits have been shown to be expressed throughout the nervous system (Ha, et al. 2006), however most other Aplysia iGluR have been identified through similarity to sequences of other species, and their in vivo expression patterns are unknown. Subunits within each iGluR subtype form a monophyletic clade in vertebrates, and complete receptors can only be formed with subunits within each of these clades. An Aplysia-only phylogeny was built to identify subunits that form monophyletic clades, and thus may form functional receptors. In this study we addressed the evolution of bilaterian iGluR, including Aplysia, using phylogenetic analysis. The expression of iGluR throughout the Aplysia nervous system and the number of iGluR genes in Aplysia were also addressed to place this model into an appropriate context with other iGluR model species. 2.3 Materials and Methods Phylogenetic analysis iGluR sequences for phylogenetic analysis were obtained from NCBI for all species except Branchiostoma belcheri (lancelet), which were obtained from the Chinese Lancelet Genome Project (<genome.bucm.edu.cn/lancelet>). Sequences of all identified canonical iGluR proteins were aligned in MEGA7 (Kumar, et al. 2016) using the 20 MUSCLE (multiple sequence comparison by log-expectation) multiple aligner algorithm with default parameters (Edgar 2004). Poorly aligned regions of the alignment were removed using trimAl (Capella-Gutiérrez, et al. 2009) with the –automated1 option to heuristically determine optimal trimming of the alignment, resulting in 473 positions used for phylogenetic analysis. Unrooted phylogenetic trees were then constructed in MEGA7 using maximum likelihood and 1000 bootstrap replicates with the LG amino acid substitution model (Le and Gascuel 2008) with gamma distributed rates and 5 rate categories. The initial tree was obtained by applying the Neighbor-Joining algorithm to a matrix of pairwise distances estimated using a JTT model. Subtype specific trees were built to further clarify the relationships within orthologous groups using the same parameters as the full phylogeny, except 500 bootstrap replicates were performed. These same parameters were also used to build the Aplysia-only iGluR tree. Trees were visualized using FigTree (v1.4.2) (Rambaut 2012). Identification of iGluR genes in Aplysia Previously described iGluR in both Aplysia and chordates were obtained from the NCBI database, and analyzed using tools at the SMART (Letunic, et al. 2012) and Interpro (Hunter, et al. 2011) protein databases for potential binding sites and transmembrane domains to identify all iGluR subunits in Aplysia. These sites are likely to be highly conserved to maintain L-Glu activation, thus their sequences were extensively searched in both the Aplysia published genome (NCBI) and the freely accessible Aplysia transcriptome database (<http://www.aplysiagenetools.org/>). Candidate genes were then translated to protein sequences and run through a BLAST search as well as scanned in Interpro for verification. Genes were confirmed to be in the Aplysia transcriptome by 21 PCR amplification from cDNA of the abdominal ganglia and cloned in a TA vector (Invitrogen) and subsequently sequenced. Hydrophobicity analysis The Kite/Doolittle hydrophobicity scale was used to determine the hydrophobicity of each amino acid for the last 450 amino acids in both human and Aplysia Grin1 subunits, which contains the ligand binding, transmembrane, and intracellular C-terminal domains. For representative AMPAR and kainate receptor subunits only transmembrane domains were analyzed due to lack of agonist-binding site identification in Aplysia. mRNA extraction and quantification of iGluR gene expression in the nervous system To describe quantitative expression of identified iGluR in Aplysia nervous system ganglia, six sexually mature Aplysia californica from a single egg mass of wild-caught animals were obtained from the National Resource for Aplysia at the University of Miami. Animals were anesthetized in a solution of 50:50 isotonic MgCl2:Artificial sea water (ASW). All ganglia were removed, immediately rinsed in ASW, and placed in Trizol (Invitrogen). Tissues were ground in a bead homogenizer to break cells out of the sheath prior to RNA extraction. Both hemiganglia for each tissue from each animal were pooled into a single sample. Total RNA was extracted following the manufacturer’s protocol and samples were treated with DNAse to remove any contaminating DNA. RNA quantities were determined using a Nanodrop (Model ND-1000), and samples were stored at -80°C until further processing. 100 ng of RNA was reverse transcribed into cDNA using the SuperScript III First-Strand Synthesis System (Invitrogen). After dilution of the cDNA (1:5 with H2O) messenger RNA (mRNA) copy number was determined using qPCR on a 22 Stratagene Mx3005P with SYBR Green master mix and the equivalent of 2 ng of starting RNA per well. Primer pairs were designed for each of the thirteen iGluR identified in Aplysia to detect expression levels with quantitative real-time PCR (Table 2.1). Efficiencies of each primer pair was tested by generating standard curves based on regression analyses of the Ct and the log value of 10-time dilution of each target gene for each primer pair. All primers used had efficiencies between 0.9-1.1. mRNA copies were then calculated using standard curves and the average of duplicate cycle threshold (Ct) values. Southern blot Southern blotting from agarose gels was done onto Hybond-N+membrane (Amersham) in sodium saline citrate buffer (10×SSC; 1.5 M NaCl; 0.15 M Na citrate pH 7.0) by capillary action. Hybridization for high stringency blots was conducted in 30% formamide, 5× SSC, 1× Denhardt’s, 0.2 % sodium dodecyl sulphate (SDS), 10% Dextran sulphate, 20 mM sodium phosphate, pH 6.8 and at 42°C. Final washes for high stringency were in 0.1 ×SSC, 0.1 % SDS at 65°C. Each lane contained 4µg of DNA and was cut with one of the following enzymes: EcoR1, HindIII, BamH1, and PstI. The probe was labeled using a random primed DNA labeling kit with [g-32P] dCTP. 2.4 Results Phylogenetic analysis of bilaterian iGluR subunits To investigate the evolutionary relationships of iGluR across the Bilateria, phylogenetic analysis was conducted on full-length protein sequences of all NMDAR, AMPAR, kainate receptor, and delta receptor subunits. Sequences were included from the 23 three major bilaterian lineages: Deuterostomia (Homo sapiens, Mus musculus, Rattus norvegicus, Danio rerio, Branchiostoma lanceolatum, Ciona intestinalis), Ecdysozoa (Limulus polyphemus, Priapulus caudatus, Drosophila melanogaster, Caenorhabditis elegans, Daphnia magna, Tribolium castaneum), and Lophotrochozoa (Aplysia californica, Octopus bimaculoides, Lingula anatina). The species tree showing the evolutionary relationships among the species is shown in Fig. 2.1. The phylogenetic tree of bilaterian iGluR is presented in Fig. 2.2. Identification of iGluR genes as kainate receptor or AMPAR subtypes is unclear for many protostomes, thus these genes are currently named as “GluR” genes without an AMPA or kainate designation. Deuterostome iGluR genes are named Grin (Glutamate Receptor Ionotropic NMDA), GRIA (Glutamate Receptor Ionotropic AMPA), or GRIK (Glutamate Receptor Ionotropic Kainate) based on selective agonists. Canonical sequences, as determined by Uniprot, were used for genes with more than 1 splice variant. Some deep branches of the tree were poorly supported by bootstrapping due to high divergence of sequences between subtypes and large evolutionary distance between species. However, most orthologous groups were well resolved with strong bootstrap support. Based on the phylogeny of the iGluR proteins each protostome subunit was classified as NMDAR, AMPAR, kainate receptor or as orphan genes that do not show a clear relationship with vertebrate subtypes (Table 2.2). NMDAR genes form three orthologous groups corresponding to Grin1, Grin2, and Grin3 subunits, providing evidence that all three ancestral Grin subunits were present in the most recent bilaterian ancestor (Fig. 2.2). This is a unique feature of NMDAR: they are the only iGluR subtype with more than one orthologous copy present before the 24 divergence of protostomes and deuterostomes. In the orthologous group Grin2 each protostome and basal deuterostome has only one ortholog of Grin2, but vertebrates have four Grin2 genes. Vertebrate Grin2 genes form a highly discrete clade, with four paralogous copies that arose early in the vertebrate lineage. The Grin2 paralogs in vertebrates are likely best explained by the 2R hypothesis, which postulates that two rounds of whole genome duplication occurred early in the vertebrate lineage after their split from tunicates (Dehal and Boore 2005; Putnam, et al. 2008). Thus, the 2R hypothesis predicts that vertebrates are expected to have four copies of each gene in comparison to one copy in invertebrates. In the case of Grin2 all four paralogs have been retained in all vertebrates used in this study. In contrast, both in Grin1 and Grin3 there is only a single orthologous group of vertebrates present in the tree. Therefore, in both cases only one of the four that originated during the 2R genome duplications remained active, whereas three of them have been lost early in vertebrate evolution. Despite large evolutionary distances among the studied species, all three Grin orthologs are highly conserved suggesting that they are under high functional constraints slowing their divergence. For AMPAR genes there is a monophyletic clade of all vertebrate AMPAR sequences and several orthologous protostome genes, pointing to an AMPAR gene copy present in the common bilaterian ancestor (Fig. 2.2). The number of AMPAR genes in protostome species is highly variable and appears to be species or taxon specific. For example, in the Lophotrochozoa, Aplysia has 6 paralogous AMPAR genes, Octopus has 1, and Lingula has 2. Ecdysozoan species have 2-3 genes in this orthogroup, except Daphnia and Tribolium, which do not have any genes in the AMPAR orthogroup. 25 Most AMPAR genes of protostome species are most closely related to each other (inparalogs) rather than to AMPAR genes, with Lingula as the exception. This implies that gene duplications occurred after the divergence of each lineage in this study, suggesting that there has been extensive gene gain and loss that has acted independently in each protostome taxon. This appears particularly true in Aplysia. Six AMPAR genes in Aplysia and 1-2 in other lophotrocozoans, including Octopus, suggests several gene duplication events occurred within the Gastropoda lineage. The sampled vertebrates have consistent with the 2R genome duplication scenario all four AMPAR genes, in comparison to one AMPAR ortholog in the common bilaterian ancestor. Kainate receptor genes are the only subtype in the tree in which Protostomia and Deuterostomia genes do not form a strongly supported monophyletic clade. Only three protostome genes form an orthologous group with chordate kainate receptor subunits: Aplysia KA2, Tribolium GRIK1, and Limulus GRIK2. Identification of kainate receptor genes is unclear for many Ecdysozoan species due to extensive divergence of these subunits from Deuterostomia. Many predicted kainate receptor genes from Limulus, Priapulus, and Tribolium form an independent clade without a clear relationship to chordate kainate receptor genes (to the left of the tree, Fig. 2.2). This suggests they have a slightly different function, or that they work through a different mechanism than the kainate genes that have been studied. Additionally, Priapulus, Limulus, Daphnia, and Tribolium have several genes that appear to be very distant to all other genes in the tree (to the right of the tree, Fig. 2.2). Within vertebrates there are two ancient paralogs with a total of five kainate genes. The three paralogs GRIK_A1-3 form one orthologous type and the two paralogs 26 GRIK_B4-5 form together with the lancelet GRIK2 sequence the second type. Therefore, in comparison to the four paralogs expected under the 2R duplications, GRIK_A1-3 and GRIK_B4-5 were respectively losing one and two paralogs. In contrast to Grin and AMPAR genes, extensive divergence of kainate receptor genes in Deuterostomia and Protostomia suggests that they are the least conserved iGluR subtype across Bilateria. Orthologous groups of each iGluR subtype were further analyzed in the attempt to obtain a better resolution of the evolutionary relationships in these parts of the tree (Supplementary Figs. S2.1-S2.3, pg. 136-138). In this analysis the relationship between protostome and chordate iGluR were the same as in the full phylogeny, providing further support for the relationships found in the full phylogeny. An additional phylogenetic tree was built using only Aplysia iGluR to search for subunits that form monophyletic clades and thus may form complete receptors (Supplementary Fig. S2.4, pg. 139). The Aplysia-only tree confirms the findings of the full phylogeny, with subunits corresponding to currently predicted NMDAR and AMPAR subunits forming monophyletic clades, while the four predicted kainate receptor subunits do not show a monophyletic relationship. Number of iGluR genes in Aplysia and other bilaterians Through genomic searches, the number of iGluR genes identified in chordates was greater than the number of genes in any protostome (Table 2.3), although it must be noted that for some protostome species the number of iGluR genes may not be accurate due to limited availability of genomic information and annotation. As described in the phylogenetic analysis, extensive gene gain and loss of AMPAR genes in protostomes has 27 also contributed to the variable number of iGluR genes in different protostome species. The increased number of genes in chordates is likely due to retention of many paralogs after 2R genome duplications. We found that in both Grin1 and Grin3 orthologous groups three of the four paralogous genes from the 2R genome duplications have not been retained. Genomic searches using SMART (Letunic, et al. 2015) and InterPro tools (Mitchell, et al. 2014) revealed 12 unique iGluR genes and one splice variant of Grin1 in Aplysia. The translated Aplysia iGluR sequences showed a highly conserved SYTANLAAF motif (Wollmuth and Sobolevsky 2004) near the second transmembrane loop, which was used to identify them as iGluR (Fig. 2.3). This motif contributes to formation of the channel outer pore and its activation gate (Dai and Zhou 2013), and amino acid substitutions in this motif are reported to alter channel gating and permeability (Kohda, et al. 2000; Low, et al. 2003). Aplysia showed a greater number of iGluR subunits compared to other protostomes, primarily due to several gene duplications of AMPAR genes, as discussed above. cDNAs corresponding to all 12 Aplysia iGluR were isolated from the nervous system and confirmed to be transcribed in vivo. Sequence similarity and conserved domains of Grin1 in Aplysia and vertebrates Conserved domains of Grin1 proteins in Aplysia and the vertebrates studied were analyzed to elucidate whether similarities in these parts of the sequences might predict conservation of function. The first ~400 amino acids of NMDAR proteins, known as the N-terminal domain (NTD), show high sequence divergence, with only 19% sequence identity in this region within a species. Mutant Grin2 subunits lacking the entire NTD were shown to assemble into receptors functionally similar to complete receptors, thus it 28 is not surprising that this region showed low sequence similarity between species (Fayyazuddin, et al. 2000; Gielen, et al. 2009). Comparatively, agonist binding domains (ABDs) and transmembrane domains (TMDs) must show greater conservation to maintain iGluR functionality, and indeed, within a species these regions show 63% and 73% identity, respectively (Paoletti 2011). In comparing the Grin1 subunits of H. sapiens and Aplysia we found 66% sequence identity between these subunits after removal of the NTD, similar to the sequence similarity of ABDs and TMDs within a species. The three TMDs of Grin1 together form the iGluR ion channel and these regions show very high protein sequence similarity throughout the Bilateria. However, Aplysia sequences consistently showed greater sequence similarity to H. sapiens than did either D. melanogaster or C. elegans (Table 2.4). The TMDs of the vertebrate genes compared had close to 100% conservation, on average, and exhibited much greater amino acid similarity than the rest of the protein, suggesting high selective pressure on these sites to maintain function. The hydrophobicity of amino acids in a protein influences the folding and structure of the molecule, and protein regions with similar hydrophobicity profiles are predicted to maintain structural stability. TMDs are well conserved, but substitutions that result in different amino acids occur, as shown in Table 2.4. A plot of the hydrophobicity of the canonical Grin1 sequences in Aplysia and H. sapiens shows that amino acid substitutions in the TMD’s have been tolerated only when the substituted residue has a similar hydrophobicity, and thus is predicted not to significantly alter protein folding (Fig. 2.4). This result suggests that the structure and function of the transmembrane domains and ion channel have been maintained. Conversely, protein sequences for the 29 glycine and NMDA binding sites show high sequence divergence and numerous changes in hydrophobicity compared to H. sapiens sequences. Hydrophobicity plays a crucial role in receptor binding domains (De Loof, et al. 1986), and this lack of conservation may reflect the diminished role for glycine as a co-agonist in Aplysia NMDAR (Carlson, et al. 2012). Representative hydrophobicity plots of AMPAR and kainate receptor subunits can be found in Supplementary Figs. S2.5-S2.6, pgs. 140-141. For both AMPAR and kainate receptor subunits, hydrophobicity of the TMDs appears to be well conserved with H. sapiens, however, they are not as conserved as the TMDs of Grin1 subunits. This supports the phylogenetic evidence that AMPAR subunits in Aplysia are closely related to H. sapiens subunits and may perform similar functions. In the phylogeny, kainate receptor subunits of the prostostomes were highly divergent from vertebrates, however the TMDs are a notable exception and appeared to be well-conserved with H. sapiens in this analysis. Evaluation of Grin2 genes in Aplysia genomic DNA While a search of the recently revised A. californica genome revealed only one gene similar to Grin2 subunits in Chordata, the presence of gaps in this genome sequence raises the possibility that additional genes may have been missed. A Southern blot analysis of genomic DNA of Aplysia demonstrated only a single band homologous to a conserved region of Grin2 (data not shown). This band was at the expected molecular weight for the previously identified Aplysia Grin2, suggesting that no other Grin2 genes are present in this genome. This result was not surprising considering that there is only 1 Grin2 gene found in other Protostomia species (Teng, et al. 2010). 30 Quantification of iGluR gene expression in the nervous system The expression of identified iGluR in all Aplysia nervous system ganglia is shown in Fig. 2.5. All twelve subunits, as well as the splice variant Grin1-2, were expressed in all ganglia, with differences both within and between different iGluR subtypes. The subtype with the highest expression varied across the different ganglia (Fig. 2.5A). In the pleural and buccal ganglion, kainate-like receptor genes were expressed at the highest levels, followed by NMDAR-like genes, with AMPAR-like genes expressed at the lowest levels. The highest expression of all 3 iGluR subtypes was found in the pedal ganglion, suggesting that this ganglion has the greatest reliance on L-Glu-mediated neurotransmission. The buccal ganglion showed the lowest expression for both NMDAR and AMPAR genes and showed significantly greater (p£0.05) expression of kainate receptor subunits. Expression of NMDAR genes was primarily due to high expression of Grin1-1, representing >60% of NMDAR expression in nearly all nervous system tissues (Fig. 2.5B). These data suggest that Grin1 may be the subunit that contributes the most to NMDAR-mediated physiological responses to L-Glu in the nervous system of Aplysia. The lone exception was the buccal ganglion, where the splice variant Grin1-2 was the most highly expressed NMDAR subunit gene in the majority of the 6 animals studied. A similar pattern of high expression of 1 gene was observed with kainate receptor genes, where the GluR7 subunit comprised the majority of this subtype’s gene expression. The distribution of expression of AMPAR subunits favored no single subunit gene dominating the expression of this subtype. 31 2.5 Discussion iGluR mediated responses in the nervous system are of particular interest due to their role in synaptic plasticity associated with learning and memory. Protostomes have a long history of use as model organisms for studies of iGluR mediated synaptic plasticity due to their simple nervous systems and well-defined neural circuits. Many plasticityrelated changes in the vertebrate nervous system were first discovered in protostomes, and were subsequently shown to be conserved in chordates (Kandel 2001). Further studies in vertebrates have identified the NMDAR and AMPAR subtypes of iGluR underlying these processes. Despite the utility of protostomes for iGluR mediated responses in the nervous system, identification of iGluR subunits that are orthologous to chordate iGluR has so far not been thoroughly studied. The application of phylogenetic methods to specific protein families is a useful procedure to shed light on their functional status. We conducted a phylogenetic analysis of iGluR spanning all three superclades of bilaterians; Deuterostomia (including Chordata), and their sisterclade Protostomia, consisting of Ecdysozoa and Lophotrochozoa. We found that the evolutionary history of iGluR in Bilateria is distinct for each iGluR subtype. It was possible to identify protostome iGluR subunits most closely related to chordate iGluR, with possibly similar functional roles, for both NMDAR and AMPAR genes. The best supported conclusion from the analysis presented is that clear orthologous groups are formed for both Grin1 and Grin2 genes. This is an extension of the previous report that protostome Grin1 and Grin2, including Aplysia, are more closely related to chordate Grin1 and Grin2, respectively, than to each other (Ha, et al. 2006) and 32 further supports the close kinship of all known bilaterian NMDAR subunits. This study confirms an ancient duplication that generated two orthologous NMDAR genes present in the common bilaterian ancestor. Both orthogroups have been under high selective constraint that has maintained the function of these genes at least since the divergence of deuterostomes and protostomes, thus making protostome model organisms excellent study systems for NMDAR. The number of NMDAR subunits varies greatly between deuterostomes and protostomes. Chordate species in this study express 3-8 Grin1 splice variants, while only 1 Grin1 splice variant was identified in Aplysia. The chordates also express a greater number of Grin2 genes and splice variants. Two independent rounds of whole genome duplication in the deuterostome lineage since its split from the bilaterian ancestor, that also gave rise to Lophotrochozoa (Dehal and Boore 2005), are hypothesized to be the origin of these duplicated Grin2 genes (Teng, et al. 2010; Ohno 2013). The regulatory Cterminus contains most of the sequence variability in different vertebrate Grin2 genes. We hypothesize that the retention of duplicated Grin2 genes is explained by the subfunctionalization model, whereby each duplicate gene copy maintains a subset of its original function (Force, et al. 1999). The NMDAR subunits involved in receptor assembly have been shown to strongly alter pharmacological and biophysical properties of the channel, including sensitivity to allosteric modulators, single channel conductance, and activation/deactivation kinetics (Stern, et al. 1992; Rumbaugh, et al. 2000; Cull-Candy and Leszkiewicz 2004). The greater number of NMDAR subunits available for complete receptors in chordates increases the different physiological responses possible, and likely 33 underlies the more complex and nuanced signaling capability observed in vertebrates. Furthermore, activation of NMDAR has been shown to be an essential component of synaptic plasticity and memory formation in vertebrates (Hunt and Castillo 2012), and knockdown of NMDAR can disrupt learning in protostome species (Xia, et al. 2005; Esdin, et al. 2010; Müßig, et al. 2010). In spite of the greater diversity of NMDAR subunits in chordates, homology of Grin1 and Grin2 across bilaterian animals suggests that a wealth of discovery in the mechanisms of NMDAR-induced plasticity is nevertheless possible in simpler nervous systems such as Aplysia, with relevance to vertebrate NMDAR physiology, including that of humans. AMPAR subtype genes also form an orthologous group, with evidence of a single ancestral AMPAR gene in the common bilaterian ancestor. Unlike NMDAR, the number of protostome genes in this orthogroup is highly variable and has been subject to extensive lineage and taxon specific gene duplications. For example, in the Ecdysozoan superclade, Pripaulus, Drosophila, and C. elegans each have two inparalogous AMPAR genes most closely related to genes in their own species, suggesting that each gene is a paralog that arose independently in each lineage. Interestingly, the other two Ecdysozoan species studied, Daphnia and Tribolium, do not have any genes within the AMPAR orthogroup. In Lophotrochozoans expansion of AMPAR genes is particularly prevalent in Aplysia, where multiple gene duplication events have resulted in six AMPAR genes. Whether AMPAR physiology in protostomes will bear the same resemblance to chordate AMPAR, as is the case with NMDAR, is uncertain. The name AMPAR for this subtype is likely inaccurate in protostomes. Isolated Aplysia sensory neurons do not respond to exogenously applied AMPA, and AMPAR antagonists do not block whole cell 34 currents elicited by the L-Glu analog D-aspartate (D-Asp; (Carlson, et al. 2012)). Yet Aplysia AMPAR have physiological roles with relevance to vertebrate learning, with AMPAR antagonists inhibiting facilitation (Chitwood, et al. 2001; Li, et al. 2005). Synaptic plasticity in vertebrates is also AMPAR-dependent (Zamanillo, et al. 1999; Sanderson, et al. 2008). AMPAR antagonists such as 6,7-dinitroquinoxaline-2,3-dione disodium (DNQX) and 6-cyano-7-nitroquinoxaline-2,3-dione disodium (CNQX) act on sites that are different than the agonist binding site. Similarity to chordate AMPAR suggests that Aplysia and other protostome AMPAR may bind agonists and antagonists differently than do those of vertebrates, yet operate similarly physiologically once activated. Extensive gene gain and loss of AMPAR suggests less functional constraint compared to NMDAR genes, and may underlie these dissimilarities in AMPAR physiology in different taxonomic groups. Different numbers of AMPAR genes in different protostomes may result in unique receptor assemblages conveying different physiological properties in each taxon, thus the exact role of AMPAR may be unique in different lineages. Kainate receptors are the least conserved subtype, with no strongly supported monophyletic group, but were clearly present in the common bilaterian ancestor. Kainate receptors play a minor role in synaptic signaling (Song and Huganir 2002), and are not believed to be as involved in learning and memory as NMDAR and AMPAR (Contractor, et al. 2000). Therefore, it is unsurprising that there may be less functional constraint on these receptors than there is on AMPAR and NMDAR. Large divergence of kainate 35 receptor genes makes it difficult to predict the functional relevance of these genes to iGluR mediated excitability in the nervous system of protostomes. In agreement with previous studies, vertebrate iGluR subunits were arranged into three distinct clades corresponding to the 3 different agonists (Lipsky and Goldman 2003). Our analysis of Aplysia only iGluR genes indicated that NMDAR and AMPAR subunits form well-defined clades and thus may be capable of forming functional iGluR channels. In contrast, the makeup of kainate receptors is less clear, due to the lack of monophyletic groups in the protostomes. All 12 identified iGluR subunits were expressed in all Aplysia ganglia, extending the results of an in situ hybridization study concluding that Grin1 was expressed throughout the nervous system (Ha, et al. 2006). This attests to the importance of L-Glu mediated fast synaptic transmission in all parts of the Aplysia nervous system. Grin1 expression was highest in nearly all Aplysia ganglia, with its two splice variants Grin1-1 and Grin1-2 together comprising ~75% of total NMDAR expression. A study using a nuclease protection assay showed that expression of Grin1 comprised 67-88% of the total Grin expression in rat brain (Goebel and Poosch 1999). When combined with homology discussed earlier, this suggests that the regulation and function of NMDAR in Aplysia are highly conserved with those of vertebrates. Variable and spatially distinct expression of the three iGluR subtypes was observed in the ganglia of the Aplysia brain. Pedal ganglion had the highest iGluR expression, with the glutamatergic nature of pedal motoneuron transmission corroborated by physiological studies. (Walters, et al. 1983a, b). Variations in the frequency and amplitude of ionic currents activated by the iGluR agonists L-Glu and D-Asp have been 36 documented in neurons isolated from different ganglia (Carlson and Fieber 2011; Kempsell and Fieber 2014), lending support to the non-uniformity of the receptor expression patterns. Studies in mammalian brains have shown both spatial and developmental variations in patterns of expression of NMDAR and AMPAR subunits, with some subunits specific for certain brain regions, or variable expression dependent on the stage of development (Akazawa, et al. 1994; Mansour, et al. 2001). To place bilaterian iGluR into a larger evolutionary perspective, insights about the deep origins of iGluR have recently emerged from discoveries on ctenophores. Studies of candidate iGluR genes in the ctenophore Mnemiopsis leidyi, which is currently thought basal to bilaterian animals (Moroz, et al. 2014), revealed that ctenophore iGluR form a monophyletic clade separate from, and ancestral to, chordates (Ryan, et al. 2013). Thus it appears that subunit types emerged after the ctenophores split, but before the divergence of Deuterostomia, Ecdysozoa, and Lophotrochozoa. This is suggestive of individual AMPAR and NMDAR subunits evolving prior to the last common bilaterian ancestor, but after divergence from ctenophores. Furthermore, two studies using expressed sequence tags (EST), genome organization, gene structure and functional content found lower amino acid substitution rates in Lophotrochozoa than Ecdysozoa relative to chordates. These findings suggest that genes in Lophotrochozoa are more likely to have greater sequence similarity to chordates than Ecdysozoan genes, and hence may be more likely to be functionally equivalent to chordates. NMDAR and AMPAR subtypes of iGluR are vital to synaptic plasticity associated with vertebrate learning. This study confirms the ancestral origins of NMDAR and AMPAR genes and also, but less strongly supported, ancestral kainate receptor genes 37 in Bilateria. These findings underscore the utility of Aplysia and other protostome models for the studies of AMPAR and NMDAR mediated responses in the nervous system. This is the first analysis of the phylogenetic relationships between subtypes of iGluR genes across Bilateria. For decades, model organisms from the Protostomia have been used as models of nervous system function, and we show that AMPAR and NMDAR subtypes were present in the common bilaterian ancestor and have been maintained as orthologous groups. Functional constraint preventing amino acid substitutions in pore regions of NMDAR suggests a highly conserved function of these subunits and potentially a conserved mechanism of learning. Kainate receptor subunits are the least conserved and may not play the same role in protostomes and deuterostomes. qPCR results demonstrate that iGluR are expressed ubiquitously throughout the nervous system of Aplysia, underscoring the importance of this model to understanding iGluR mediated nervous system function. 38 Table 2.1 Primers used for qPCR. Gene Primer Sequences for qPCR (5’-3’) GluR1 F R F R F R F R F R F R F R F R F R F R F R F R F R GluR2 GluR3 GluR4 GluR5 GluR6 GluR7 GluR8 KA1 KA2 Grin1-1 Grin1-2 Grin2 Amplicon length (bp) GCTTTGTGGACAACACCAGC GGTTCTGCCATGATGATCGAC 116 GACTTAAAGGTGTCCAACGC CGTCATACGACAAGCTCTTC GAACCTTGACCCCAAGTTCTG CCTATTCACGAGAGCTTTCG CGTACTTGAGTTTGCGGTCC GTGACCATATCGCTGCAGAC CTTCACGGAGGAGTCAAAGTC CTTCAGACGGATGCAGCACT GAGATACGTGCTGGATCAGG GGCTCATGATGGACTACAAC CGAGACTGCTTTAGCCTACG CCATTAGCCTGCCTTGTGAC TCAAGTCGCTCAATCTCTCC CCACGTCAACACACTTCTAGTC GCTACCGAAACCAGCTCAATC GCAGTCACACGTACTTAAGAGG GATCTGGCTCTACGTCATTGG CTTGTAACCCAGACACGGAC GGGAAACATTGAGAAGC GCTCCCAATGCAAACACAGC 145 GACACCAACGAAATAGACCTC GCTCCCAATGCAAACACAGC AGTTCACCTGCGACTCTGAC CCGAGGTTCTCATCCTGAATC 146 101 121 146 154 100 110 145 115 135 150 39 Table. 2.2. Placement of Protostomia iGluR into subtypes based on phylogenies. Species Aplysia californica AMPA GluR1 GluR2 GluR3 GluR4 GluR5 GluR8 GluR1 GluR2 Kainate GluR7 KA2 NMDA Grin1 Grin2 Orphans KA1 GluR6 none Grin1 Grin2 Drosophila melanogaster GluR1A GluR1B GluR1C GluR2D GluRIIE Grin1 Grin2 GluR3 GluR4a GluR5 GluR6 GluR7 GluRIIA GluRIIB GluRIIC Daphnia magna none GRIK1 Limulus polyphemus GluR1 GRIK2 GRIK3 Lingula anatina GluR2 GluR4 GluR GRIK2 Priapulus caudatus GluR3 GluR4 none Grin1 GluR1 Grin2 Grin3 Grin1 Grin2 Grin3 Grin1 Grin2 Grin1 Grin3 Grin1 Grin2 Grin3 Tribolium castaneum none GRIK1 GluR1 Caenorhabditis elegans Octopus bimaculoides none Grin1 Grin2 GRIK2 GRIK1 GluR4 GluR2 GluR1 GRIK2 GRIK3 GluR1 GluR2 GRIK4 GRIK2 GRIK3 GRIK5 GluR2 iGluR genes in each protostome species were categorized into subtypes based on their phylogenetic relationship with chordate iGluR genes. Many protostome iGluR genes do not have a clear relationship with chordate subtypes and are thus identified as orphan receptors. Protostome orphan receptors are divergent from chordate genes and thus unlikely to perform the same subtype specific functions. 40 Table 2.3. Number of iGluR genes in the protostomes and vertebrates. Species Total iGluR genes Homo sapiens 14 Caenorhabditis elegans 9 Mus musculus 14 Ciona intestinalis 6 Rattus norvegicus 14 Daphnia magna 7 Danio rerio 14 Limulus Polyphemus 11 Aplysia california 12 Priapulus caudatus 11 Lingula anatina 6 Tribolium castaneum 10 6 Branchiostoma belcheri 10 Octopus bimaculoides Drosophila melanogaster 10 Species Total iGluR genes 41 Table 2.4. Sequence similarities in transmembrane domains of Grin1 between different protostomes and H. sapiens. Species A. californica TMD1 TMD2 TMD3 85% 100% 87% D. melanogaster 75% 100% 61% C. elegans 87% 48% 65% Compared to H. sapiens, Aplysia had fewer amino acid substitutions in transmembrane domains (TMDs) than did D. melanogaster and C. elegans. This points to a higher likelihood of conserved channel function between vertebrates and Aplysia. 42 Rattus norvegicus Rattus norvegicus Mus musculus Mus musculus Chordata Homo sapiens Homo sapiens Deuterostomia Danio Danio reriorerio Branchiostoma lancelolatum Branchiostoma lanceolatum Ciona intestinalis Ciona intestinalis Lingula anatina Lingula anatina Lophotrocozoa Cephalopoda Mollusca Gastropoda Octopus bimaculoides Octopus bimaculoides Aplysia californica Aplysia californica Priapulus caudatus Priapulus caudatus Daphnia Daphnia magnamagna Protostomia Drosophila melanogaster Drosophila melanogaster Arthropoda Ecdysozoa Tribolium castaneum Tribolium castaneum Limulus polyphemus Limulus polyphemus Nematoda Caenorhabditis Caenorhabditis elegans elegans 3.0 Fig. 2.1. Phylogeny of species used for iGluR analysis. Tree of the evolutionary relationships between species used in this study. The tree was built using the NCBI taxonomy browser (Sayers, et al. 2011). 43 NMDA Kainate Grin2 _Dro soph ila 1 m Gr in2 _T rib oliu ib ol iu m Gr in 1_ Tr im ul us G ri n3 _L ia n _Daph G ri n 3 pus _ O c to t G ri n 3 cele n h lu s _La is A f ri a p u rin3 e bra G e 3_P Z ous G ri n 3A_ _M t ia G r in r in 3 A A_Ra hn a G 3 ns ap n i n ga _D p h l a s ma Grin H u Cele luR1 _ D a p h i u l u lus A_ _ G rin1 o s o L i m pu pus i n 3 rin1 Gr G _ D r 1 _ ria to a G in P c i i n 1 G r 1_ _O lys rin 1 p G rin _A G n1 ri G Grin2_Daphnia 1 0 .9 9 an um se _H ou D M 2 _ rin D G rin2 lus s pu G l a let ulu ria u e iap 1 _ P r n g anc i a P _ L _L 2_ D e l t u s uR R1 4 t ul l u IK le G l us iap G GR nce P r iapul _ a 2 L a Pr t _ l _ e 4 s D RIK K3 lu I u G GR _Lim s R2 lu Glu _ L im u 1 a lt D e pulus Pria nia h n ia R1_ Glu 2_Dap 2 _ D a p h GRIK D e lt a L im u lu s 2_ a lt e D m pus Tr ib ol iu _ O ct o D el ta 1_ D e lt a 1 ibo liu m GR IK 5_ Tr Glu R2 _Tr ibo lium Delta2 _Tribo lium GluRIC_Drosophila GluRIB_Drosophila Glu R1A _Dr oso phi la GluR1_Lim ulus GluR3_ P GluR4 riapulus _ GluR _O Priapulus GluR ctopus 3_Ap ly GluR sia 4_A Glu p R2_ GluR1_ lysia A A Glu Gl GluR plysia plysia uR 5_A R GR Gl 4_L 8_A plys uR ing ply ia G GR IA sia u 2 G RIAIA1 1_M Gl _ L i n l a R 1 _R o uR g u I A _ a us 1_ l a 1 _ Hu t e Gl Ce Ze m uR le b r an 2_ ga af Ce ns is leg h an s se ou M at an ish 2_ R u m r a f e s h IA _ H b s f i R A2 _ e ou ra G RI IA2 2_Z _M Zeban G R I A A3 3 _ m G R RI I A H u ish G G R _ at raf h 3 G IA _R Zeb rafis GRRIA3IA4_ _Zeb G G R A4B t a I n GRIA4_RHuma e GRRIA4__Mous G RIA4 a G 1_Cionphnia ulus GluR 1_Da 3_Lim GRIK GRIK liu m R 2D _D ro so ph ila _ T ri b o G lu G lu R 1 rosophila GluRIIE_D GluR7_Aplysia GRIK2_Lingula G GR RI IK K GR 1 _ 2 _ Z IK Z e b e b GR G 1 r IK RIK GR _M rafi afis 3_ 1 IK o sh h _ GR Zeb Hu 1_Ruse IK3 raf ma at GR IK2 GR GR _M ish n _La IK IK ou nc 3_ 3_ s GR GRIK elet Hum Rate an GRIKIK5_M5_Rat GRIK 5 _ H ouse GRIK 5_Zebrauman 4_Ze fi s brafis h h GR GRIK4_IK4_Rat Mouse GRIK4_Hu man KA2_Aplysia GRIK 1_Tr iboliu m GRIK2_Limulus 0 .8 0 .9 9 Gr G r G in1 rin _L 1 in G _Ze gu rin b la 1 ra Gr G _Hu fish ri i Gl u R GR n 1 _ n 1 _ m a 2 _ IA M o R a n GR C i o 2 _ u s t G IK2 n a L a n c e G R R I K 2 _Cio ele I _ T na K t GR 1 IK3 _Lim ribo _P ulu lium G R IK GluR7 riapulu s _C s 3_ GluR 4_Lim T r ib o li elegan s ulus u m D e lt a 2_Oc to p u Delt s D el ta 1_a1_Rat M ou se De lta 2_ Hu De lta 1_ Ze bra m an fis h Delt a2_Z ebra fish Delta1_Hum an 1 Delta2_Rat 1 Delt a2_M ouse na De lta 2_ Cio a on et De lta 1_ Cielta1_Lancel D celet 1_Lan GRIK ancelet 1_L t s GRIA Lancele ctopu sia a2_ RIK2_O1_Aply Delt G KA sia ly _Ap R6 a Glu ophil phila a s se il o o s ph o Dr oumanat ns B_ A_Dr roso s s _M u _R ga RII I n Glu luRI IIC_Dlega Cele ns gan IK22_HIK2 G uR e 5 _ g a l e GR I K R R G G l 6_C luR Cele _Ce G uR G a_ R3 Gl 4 lu uR G l G Grin2_L imulus Grin2_Celegans Gr in2 _A ply sia G rin 2_ Li ng ul a G ri n 2 _ Grin2 P ri a p u lu s _Lan celet G ri n G r in 2A_C 2A_ io n a Hum Grin G r in 2A_Ra a n t 2 Gri A_Mo u n se Gr in 2 A _ Gr 2 B _ Z e b r a H Gr i n 2 B _ u m a f i s h in2 n Mo B_ Ra u s e G Grin t Gr rin2 2B_ C_ Ze G in2 H b G rin2 C _ M u m r a f i s rin C a h 2 C _R o u s n e G a rin G r _ Z e t 2D i n 2 b r _R D _ a f i s at Z e h br af is h AMPA 0.4 Fig. 2.2. Bootstrap consensus phylogenetic tree. iGluR protein sequences for six deuterostomes and nine protostomes with iGluR information available were aligned using MUSCLE in MEGA7 and the tree was constructed using maximum likelihood and 1000 bootstrap replicates. iGluR subtypes are indicated by different colors. Numbers indicate 0.4 bootstrap support and are displayed for boostrap values >0.6. The scale bar represents 0.4 substitutions per site. All species contain homologs for Grin1 and Grin2 subunits. An ancestral AMPAR and kainate receptor gene was present in the common bilaterian ancestor, but in the case of the kainate receptor, the monophyly is only weakly supported by bootstrap values. Fig2.3.ConservedmotifsinAplysiaandH.sapiens.BoxedregiondenotesSYTANLAAFLmotif vitalforformationofthechannelporeandactivationgate,andinallsubunitsthishighselective pressurehasresultedinhighlyconservedaminoacidsequencesoverlargeevolutionary distance.Thismotifisalsoconservedinallothervertebratesinthisstudy. 44 45 "# NMDA Binding Glycine Binding TM2 TM1 TM3 $# %# &# Grin1&Aplysia Grin1&Human -./0〹/6# %&'# %%&# %$%# %"$# %("# %)(# %*)# %+*# %,+# $',# $%'# $$&# $"%# $($# $)"# $*(# $+)# $,*# "'+# "&,# "$'# ""&# "(%# ")$# "*"# "+(# ",)# ('*# (&+# (%,# ("'# ((&# ()%# (*$# (+"# (,(# )')# )&*# )%+# )$,# )('# ))&# )*%# )+$# ),"# *'(# *&)# *%*# *$+# *",# *)'# **&# *+%# *,$# +'"# +&(# +%)# +$*# +"+# +(,# +*'# ++&# +,%# ,'$# ,&"# ,%(# ,$)# ,"*# '# -./0𓑰# !&# !%# !$# !"# Fig. 2.4. Hydrophobicity plot of Grin1 in H. sapiens and Aplysia. Kite/Doolittle scale was used to determine the hydrophobicity of each amino acid for the last 450 amino acids, containing the ligand binding, transmembrane (TM), and intracellular C-terminal domains. Despite large evolutionary distance between Aplysia and H. sapiens, only substitutions with similar hydrophobicity are tolerated in TMs to maintain proper folding of the ion channel. Predicted NMDA and glycine binding sites show many more substitutions that result in changes in hydrophobicity. 46 A Calculated number of transcripts of iGluR genes 14000 AMPAR NMDAR 12000 Kainate Orphans 10000 * 8000 6000 4000 2000 0 Pedal Pleural B Cerebral Abdominal Ganglion AMPA NMDA Buccal Kainate/orphans GluR1 GluR2 GluR6 Grin1-1 GluR3 GluR7 Grin1-2 GluR4 KA1 Grin2 GluR5 KA2 GluR8/10 Fig. 2.5. iGluR expression levels in Aplysia ganglia. The expression levels of each subtype of iGluR per 100 ng of RNA were determined by quantitative real-time PCR. Absolute copy number was calculated using standard curves. A) Comparisons of total expression of each iGluR subtype in different Aplysia ganglia, with calculated number of transcripts on the y-axis. Total kainate receptor subunit expression was significantly greater than other subtypes in the pleural ganglia (one-way ANOVA, Tukey’s post-hoc, p£0.05). B) Pie charts showing each subunit’s contribution to the total expression of its subtype. Chapter 3: Altered expression of ionotropic L-Glutamate receptors in aged sensory neurons of Aplysia 3.1 Summary Changes in mRNA expression of ionotropic glutamate receptors (iGluR) were studied in two cohorts of Aplysia at both sexual maturity and advanced age. Behavioral aging of both cohorts was confirmed via the righting and tail withdrawal reflexes. Previous studies have shown reduced behavioral performance in aged Aplysia is correlated with reduced L-Glu current densities in buccal S cluster sensory neurons, as well as in pleural ventral caudal sensory neurons that are involved in tail withdrawal. Aging in both sets of sensory neurons resulted in significantly decreased expression of iGluR subunits, particularly of the NMDA receptor subtype. This implies that age-related declines in L-Glu responsiveness may be linked with reduced expression of receptors that bind L-Glu. The NMDA receptor subtype was also expressed at significantly greater levels than other iGluR subtypes. Thus, NMDA receptors may be a major contributor to L-Glu mediated responses in Aplysia sensory neurons. 3.2 Background The relatively simple nervous system of the model organism Aplysia californica (Aplysia) allows for detailed studies of neuronal circuits that underlie many behaviors (Rattan and Peretz 1981; Walters, et al. 1983a, b; Kandel 2001). These simple neuronal circuits have been used to study changes in neuronal physiology that accompany aging in 47 48 the Aplysia nervous system (Hirsch and Peretz 1984; Fieber, et al. 2010; Kempsell and Fieber 2014, 2015b). There is evidence that many neuronal circuits in Aplysia use Lglutamate (L-Glu) as their primary excitatory neurotransmitter (Dale and Kandel 1993; Ha, et al. 2006). In this Chapter, previously described changes in L-Glu physiology in the aging Aplysia nervous system were correlated with altered mRNA expression of L-Glu receptors. L-Glu binding to its constituent receptors on the postsynaptic membrane activates both metabotropic and ionotropic L-Glu receptors. Metabotropic L-Glu receptors (mGluR) act through activation of G-protein coupled second messenger cascades and can induce changes in neuronal excitability and alterations in synaptic transmission (Niswender and Conn 2010). Ionotropic L-Glu receptors (iGluR) mediate the majority of fast synaptic transmission and regulate many processes in the central nervous system (Dingledine, et al. 1999; Traynelis, et al. 2010a). iGluR are classified into three subtypes based on selective agonists in vertebrates: N-methyl-D-aspartate (NMDA), α-Amino-3hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA), and kainate. Functional iGluR are formed as heterotetramers, consisting of iGluR subunits belonging to the same subtype. Each iGluR subtype has distinct electrophysiological characteristics, and there is evidence that analogues of each of these vertebrate subtypes exists in Aplysia, as described below. The composition of Aplysia iGluR tetramers, if indeed its functional iGluR are formed as in vertebrates, is unknown. In vertebrates, NMDA iGluR (NMDAR) have relatively slow activation and deactivation kinetics and are highly permeable to Ca2+ (Mayer and Westbrook 1987). 49 Activation of NMDAR and the opening of Ca2+ channels activates second messenger cascades that have been implicated in the induction of neuronal plasticity (Riedel, et al. 2003). NMDAR are the most well-conserved iGluR subtype across bilaterian species, including Aplysia (Chapter 2). This suggested that function of NMDAR may be conserved between Aplysia and other bilaterian clades, including vertebrates (Chapter 2). Application of NMDA to Aplysia buccal sensory neurons (BSC) in culture elicits small amplitude currents and implies that NMDAR-like receptors are physiologically relevant in Aplysia (Carlson and Fieber 2012; Fieber, et al. 2013). The AMPA subtype of L-Glu receptors (AMPAR) mediate the majority of fast excitatory transmission in the nervous system of vertebrates and have fast activation and deactivation kinetics compared to NMDAR. AMPA application did not induce currents in Aplysia BSC neurons in culture or in motoneurons of the abdominal ganglion (Trudeau and Castellucci 1993; Carlson and Fieber 2012). However, the AMPAR antagonist 6cyano-7-nitroquinoxaline-2,3-dione (CNQX)-induced block of synaptic transmission, suggesting that AMPA-like receptors exist in Aplysia (Trudeau and Castellucci 1993). An ancestral AMPAR subtype gene was present in the bilaterian ancestor (Chapter 2), but the Aplysia lineage has been subjected to several rounds of gene duplications. This suggested that AMPAR may exist in Aplysia but have diverged to the point where AMPA no longer binds as an agonist. Kainate iGluR are present in both pre and post-synaptic terminal in vertebrates, however their role is less clear than NMDAR and AMPAR. They are the most evolutionarily recent iGluR subtype, and unlike NMDAR and AMPAR, vertebrate and invertebrate kainate subunits do not form a clear monophyletic group (Chapter 2). 50 Kainate analogs elicited depolarizations and non-desensitizing inward currents in glutamatergic BSC neurons of Aplysia, similar to L-Glu-induced effects in these neurons (Wang, et al. 2013). Kainate is also an effective agonist in abdominal ganglion neurons (Trudeau and Castellucci 1993). Thus, although they may not be well-conserved, kainatelike receptors may also execute physiologically relevant processes in Aplysia. In Chapter 2, Aplysia iGluR subunits were placed into subtypes based on their phylogenetic relationships with vertebrate iGluR subunits (See Table 2.2, in Chapter 2). Grin1 and Grin2 subunits in Aplysia formed orthologous groups with vertebrate NMDAR subunits, providing strong support for these subunits as NMDAR-like. Aplysia GluR1GluR5 and GluR8 formed a monophyletic clade with vertebrate AMPAR subunits, and thus are predicted to play AMPAR-like roles. Kainate subunits formed only a weakly supported monophyletic clade, with GluR7 and KA2 included in this clade. Aplysia KA1 and GluR6 were identified as orphan subunits with no clear relationship to vertebrate iGluR subtypes. Several behavioral responses in Aplysia decline incrementally as the animal ages. Reduced function of iGluR in the underlying neuronal circuits has been implicated as a contributor to these behavioral deficits. An example is the tail-withdrawal reflex (TWR). During TWR, a stimulus applied to the tail causes withdrawal of the tail as a defensive mechanism. Time to withdraw the tail increased significantly in animals of advanced age (Kempsell and Fieber 2014). Pleural ventral caudal (PVC) sensory neurons in the pleural ganglion make strong connections to tail motoneurons in the pedal ganglion and together they comprise the neural circuit controlling TWR (Walters, et al. 1983a). PVC neurons are highly glutamatergic, with 52% of PVC neurons responding to L-Glu application in 51 young animals (Kempsell and Fieber 2014). In aged animals with reduced TWR behaviors, PVC neurons showed reduced L-Glu current density and the number of PVC neurons responding to L-Glu application significantly decreased (Kempsell and Fieber 2014). Thus, reduced L-Glu responses in PVC sensory neurons were correlated with reduced TWR and are believed a hallmark of aging in this neural circuit. The biting response also exhibited significant behavioral deficits with age. Biting amplitude decreased significantly in aged animals, while latency increased (Kempsell and Fieber 2014). BSC sensory neurons innervating the biting response showed reduced frequency of L-Glu responses and declines in L-Glu current densities, similar to those observed in TWR (Kempsell and Fieber 2014). Together, these experiments implied that aging of TWR and the biting response are due, at least in part, to reduced iGluR function of their sensory neurons. This study was an examination of iGluR in Aplysia sensory neurons as well as a correlation of reduced L-Glu current density in PVC and BSC sensory neurons of aged animals with altered transcription of iGluR subunits. We addressed which iGluR subunits are most highly expressed in sensory neurons and may be involved in L-Glu mediated responses. Also addressed was whether changes in transcription underlie reduced iGluR current amplitude in aged animals. Changes in transcription can be used as a proxy for the number of receptors expressed (Vogel and Marcotte 2012), and changes in iGluR expression may be involved in reduced function of iGluR in aged Aplysia. iGluR transcription in the abdominal ganglion, a heterogeneous ganglion containing both sensory and motoneurons, was compared to iGluR transcription in the highly glutamatergic clusters of PVC and BSC sensory neurons. 52 3.3 Materials and Methods Aplysia from the University of Miami National Resource for Aplysia were reared from egg masses of wild-caught animals. Two cohorts were reared and maintained throughout their lifespan at a temperature of 13-15°C. Once the animals reached age 4 months they were held at a maximum of 5 per cage to allow for sufficient room to grow large enough to reach sexual maturity. Animals were fed an ad lib diet of Agardhiella subulata as previously described (Gerdes and Fieber 2006) throughout their lifespan. Reflex completion times were measured in animals at sexual maturity (M; age 8.5 months) and advanced age (age 12 months) to quantify behavioral changes with age. Sexual maturity for each cohort was designated as the first day an egg mass was laid. Time to complete TWR and righting was recorded in 8 animals from each cohort, at each stage of age. Mean reflex completion times were compared to Kempsell and Fieber (2014) and determined that age 12 month animals in these two cohorts were Aged II (AII). Righting Righting was measured at M and then again at AII in 16 animals (8 from each cohort). Animals were placed in individual 48x27x20 cm cages with flowing seawater and acclimated for five minutes before measurements of righting. To record righting, the animal was picked up from the bottom of the cage and released upside down from the top of the water column. Time was recorded from the when the animal was released until it was able to right itself, adhere to the bottom of the cage, and take its first crawling step. The measurement was recorded 3 times per individual with a minimum of 5 minutes between trials, and the times averaged. Data was combined for all animals at each time 53 point and a Student’s t-test was used to determine significant differences in righting reflex between age groups. TWR TWR was measured at both time points in the same individuals from both cohorts that were analyzed for the righting reflex (n=8 for each cohort). After righting reflex trials animals were allowed to rest for 5 minutes before the start of TWR measurements. To record TWR the animal was first placed on its foot in the center of the cage, allowed to adhere to the bottom, and was lightly held in place to measure initial resting length. TWR was initiated by pressing a blunted 21-gauge needle to the tip of the animal’s tail for approximately 1 sec to initiate withdrawal of the tail. To measure TWR, time was recorded from the initiation of the reflex until the animal began to relax the tail after withdrawal. Final retracted length was measured with a ruler to calculate percent tail withdrawn. Measurements of duration and amplitude of TWR were recorded 3 times with a minimum of 5 minutes between trials and the measurements were averaged. For analysis, a Student’s t-test was used to calculate differences in TWR time and fraction of body length withdrawn between age groups. Weight Weight measurements were recorded monthly for a random selection of 15 animals from each cohort beginning at age 3 months until the end of life, approximately age 12.5 months in both cohorts. Excess water was drained from the parapodial cavity prior to weight measurements. 54 RNA extraction of select ganglia for qPCR RNA was extracted from six animals from each cohort at M and six animals from each cohort at AII. The animals chosen were a randomly selected subset of the animals tested for TWR and the righting at each age. Animals were first anesthetized by injection of 1/6th body weight (by volume) of isotonic MgCl2. The CNS was removed by severing the connectives and removal of all nervous system ganglia. Ganglia were immediately rinsed in artificial sea water (ASW: 417mM NaCl, 10mM KCl, 10mM CaCl2, 55mM MgCl2, and 15mM HEPES-NaOH, pH 7.6) after removal. PVC neurons were separated from the remainder of the pleural ganglion by pinning the ganglion tightly in a Sylgarded dish filled with ASW, as follows. Connective tissue surrounding the ganglion was removed, and the sensory neuron clusters were removed with a pair of fine forceps. Neuron clusters were immediately place in Qiagen lysis buffer and both left and right PVC hemiganglia were pooled into a single sample. The same procedure was used to remove BSC neurons from the buccal ganglion. Total RNA from PVC and BSC neurons were extracted with the Qiagen RNeasy Micro Kit following the manufacturer’s instructions. Genomic DNA contamination was removed during RNA extraction according as manufacturer’s instructions (Qiagen). The abdominal ganglia were rinsed with ASW and immediately placed in Trizol. The tissue was ground in a bead homogenizer to break cells out of the connective sheath prior to extraction RNA extraction. Total RNA was then extracted using Trizol Plus RNA Purification kit following the manufacturer’s instructions (ThermoFisher). Quantity of RNA from all tissues was analyzed using a NanoDrop (Model ND1000). One hundred ng of total RNA from each sample was reversed transcribed into 55 cDNA using SuperScript III reverse transcriptase (Invitrogen) with random hexamer primers following the manufacturer’s suggestions. The resulting cDNA was diluted 1:5 with nuclease free water to provide a final concentration of 1 ng RNA/1 µl cDNA for qPCR. qPCR qPCR reactions were executed on a Stratagene Mx3005P with SYBR Green master mix (Invitrogen). Each reaction consisted of 10 µl of SYBR Green master mix, 2 µl of cDNA and 0.2 µM of forward and reverse primers in a total volume of 20 µl. Thermal cycler conditions were as follows: 95°C for 10 minutes, followed by 40 cycles of 95°C for 15 sec, 58°C for 30 sec, 72°C for 30 sec. All 12 iGluR subunits and the splice variant of Grin1 identified were tested (see Table 2.2, Chapter 2 for list of iGluR genes). Primer sequences and amplicon lengths for each gene have been previously described (see Table 2.1, Chapter 2). For cohort 1, these ganglia and sensory neuron clusters from 6 animals at M and AII were analyzed with qPCR to quantify changes in iGluR expression in the nervous system. Two technical replicates were performed for each biological replicate. Absolute copy number for each replicate was calculated using serial dilutions of a known quantity of plasmid containing the target amplicon. PCR products for each gene were gel purified with a Qiagen gel purification kit and cloned into electrocompetent E coli cells using a TOPO TA cloning kit (ThermoFisher). Transformed bacterial colonies were plated onto LB agar plates with ampicillin and blue/white screened with X-gal to identify colonies containing the desired plasmid. Five bacterial colonies were picked from each plate and bacteria grown in a shaker overnight at 37°C. Plasmids were then 56 purified with a Qiagen mini-prep plasmid kit and quantified on a Nanodrop to determine plasmid concentration. Serial dilutions from 106-100 plasmid copies were run in triplicate on each plate to calculate a standard curve for absolute copy number calculation. For cohort 2, PVC and BSC neuron clusters from 4 animals were tested at each time point. These samples were run in technical triplicate instead of duplicates as were run for cohort 1. In this cohort we tested the expression of several of the most highly expressed iGluR subunits: Grin1-1, Grin1-2, Grin2, GluR1, GluR5, and GluR7. Instead of absolute quantification using a standard curve on each plate, quantification of target transcripts was normalized to GAPDH using the Gene Expression’s CT Difference (GED) formula (Schefe, et al. 2006). Briefly, this calculation is a modified version of the DDCT method previously described (Pfaffl 2001). The DDCT method relies on calculating the efficiency of the reaction based on the slope of one cycle in the exponential phase of the reaction. It has been shown that the efficiency of the qPCR reaction may vary in different parts of the qPCR plate and varying salt concentrations in different samples (Schefe, et al. 2006). The GED formula accounts for different efficiencies in each reaction by using calculating the efficiency of each qPCR reaction independently using Real-time PCR miner (Zhao and Fernald 2005). These efficiencies were then used to calculate relative transcription for each sample as previously described (Pfaffl 2001; Schefe, et al. 2006). 57 3.4 Results Behavioral aging in two cohorts Average weight declined beginning approximately two months prior to AII measurements in both cohorts, consistent with previously studies in aged Aplysia of reduced mass during advanced stages of aging (Fig. 3.1A). Time to right significantly increased in cohort 1 (Fig. 3.1B, 13.3±0.39 sec (mean±standard error) at M, 22.0±0.59 sec at age 12 months, p£0.05, Student’s t-test). TWR also increased significantly in AII animals compared to M in cohort 1 (Fig. 3.1C, 13.7±0.58 sec at M, 24.2 ± 0.58 sec at age 12 months, p£0.05, Student’s t-test). Significantly increased righting and TWR times in AII were also observed in cohort 2 (Fig. 3.1B and C; Righting: 13.3±0.47 sec at M, 23.7±0.68 sec age 12 months; TWR: 13.6±0.33 sec at M, 27.2±0.37 sec at age 12 months, p£0.05, Student’s t-test). The observed TWR ³20.1 sec and righting ³18.1 sec for age 12 month animals of both cohorts corresponded to AII based on published stages of aging in Aplysia (Kempsell and Fieber 2014). The amplitude of TWR, measured as the fraction of body length withdrawn in response to tail touch, decreased significantly in AII animals in both cohorts (Fig. 3.1D; cohort 1: 26.1±1.2% at M, 13.7±0.7% at AII; cohort 2: 27.5±2.0% at M, 15.8±1.0% at AII, p£0.05, one-way ANOVA on a logistic-regression with quasibinomial distribution). Thus, reflex times increased in AII while the amplitude of TWR decreased. The most highly expressed iGluR subunits in three nervous system tissues Differential expression of iGluR subunits was first studied in abdominal and sensory neuron clusters separately, without accounting for differences in age. These data were analyzed using data from cohort 1 because calculation of absolute copy number in 58 this cohort allowed for direct comparisons of genes from different qPCR plates. Grin1-1 showed significantly greater expression than all other iGluR subunits in each of the three tissues tested (Fig. 3.2A-C, 30,779±4,301 copies (mean±standard error) in PVC neurons, 27,095±3,572 copies in BSC neurons, 20,646±2,504 copies in abdominal ganglia, p£0.05, one-way ANOVA, Tukey’s post-hoc test). In PVC sensory neurons, GluR1 was the next most highly expressed subunit after Grin1-1, and GluR1 showed significantly greater expression than all less-expressed subunits except GluR7. GluR7 in PVC neurons was expressed at significantly greater copies than only GluR2 and KA2 (Fig. 3.2A). In BSC sensory neurons, GluR1 and Grin1-2 were expressed at significantly higher copies than all other subunits after Grin1-1 (Fig. 3.2B). GluR7, GluR1, and Grin1-2 subunits were expressed significantly higher in the abdominal ganglia than were other iGluR subunits, except Grin1-1 (Fig. 3.2C). Differential expression of iGluR subunits in different parts of the nervous system Next, each subunit was compared across the ganglia to reveal patterns of differential subunit expression in the different ganglia. This analysis was calculated for cohort 1 only, and expression from M and AII animals was pooled. Apparent once more was the dominant expression in all 3 tissues of Grin1-1, expressed at a minimum of >2 times the expression of other subunits in all tissues (Fig. 3.3, p£0.05 for each tissue, oneway ANOVA, Tukey’s post-hoc). Several differences in expression of other, individual subunits across tissues, however, were clear. GluR5 was expressed at significantly greater copies in PVC neurons than in either BSC neurons or the abdominal ganglia (Fig. 3.3, 4,015±643 copies (mean±standard error) in PVC neurons, 386±73 copies in BSC neurons, 1282±194 copies in abdominal ganglia, p£0.05, one-way ANOVA, Tukey’s 59 post-hoc). Grin1-2, GluR2, and GluR4 were expressed significantly more in BSC sensory neurons than the other tissues (Grin1-2: 5,170±719 copies in PVC neurons, 11,263±1350 copies in BSC neurons, 5,269±536 copies in abdominal ganglia; GluR2: 349±90 copies in PVC neurons, 864±185 copies in BSC neurons, 400±108 copies in abdominal ganglia; GluR4: 739±87 copies in PVC neurons, 2,253±265 copies in BSC neurons, 385±33 copies in abdominal ganglia, p£0.05 for each gene, one-way ANOVA, Tukey’s posthoc). Grin2 was expressed at significantly lower levels in BSC neurons than in any other ganglia. (2,970±264 copies in PVC neurons, 1,407±144 copies in BSC neurons, 3490±342 copies in abdominal ganglia, p£0.05, one-way ANOVA with Tukey’s posthoc) In the abdominal ganglia, GluR7 expression was significantly greater than in the other tissues, while KA1 and KA2 expression was lower in the abdominal ganglia than in other tissues (GluR7: 6,772±753 copies in PVC neurons, 6,258±574 copies in BSC neurons, 8,741±770 copies in abdominal ganglia; KA1: 1,559±199 copies in PVC neurons, 2,036±120 copies in BSC neurons, 1,036±96 copies in abdominal ganglia; KA2: 541±45 copies in PVC neurons, 451±51 copies in BSC neurons, 368±26 copies in abdominal ganglia, p£0.05 for each gene, one-way ANOVA with Tukey’s post-hoc). Expression of iGluR subunits with age Decreased TWR in AII was previously correlated with reduced L-Glu current density in PVC sensory neurons for the reflex (Kempsell and Fieber 2014). Whether declines in TWR were correlated with altered expression of iGluR subunits was investigated by comparing iGluR expression in PVC neurons of M and AII animals. In cohort 1, AII PVC sensory neurons exhibited significantly decreased expression of several iGluR subunits; these were both splice variants of Grin1: Grin1-1 and Grin1-2 60 (Fig. 3.4A; Grin1-1 AII:M expression 0.56±0.08, Grin1-2: AII:M expression 0.60±0.13, p£0.05 for each gene, Student’s t-test). Grin2, the only remaining NMDAR subunit, showed no change in AII compared to M. No changes in expression of any AMPAR subunits in aged PVC neurons were found. KA1 and GluR6, orphan receptors without a clear relationship to any iGluR subtype, had significantly decreased expression in AII (Fig. 3.4A; KA1 AII:M expression 0.47±0.10; GluR2 AII:M expression 0.46±0.09, p£0.05 for each gene, Student’s t-test). Select subunits were analyzed in cohort 2 to attempt to substantiate differential expression observed in cohort 1 (Fig 3.4B). Distinct differences were apparent in cohort 2 PVC neurons. Although Grin1-1 was significantly down-regulated in cohort 2 AII PVC neurons (AII:M expression 0.38±0.04, p£0.05, Student’s t-test), as had been observed in cohort 1, Grin1-2 expression was not significantly changed, representing a departure from the difference between M and AII Grin1-2 in cohort 1. GluR1, an AMPAR subunit, exhibited reduced expression in cohort 2 AII PVC neurons, but this subunit had not been differentially expressed with age in cohort 1 (AII:M expression 0.41±0.03, p£0.05, Student’s t-test). Transcript copy numbers per 100 ng of RNA were also calculated for cohort 1, and age-related changes in iGluR expression described above are displayed as copy numbers in Fig. 3.5. BSC sensory neurons innervate the biting reflex and were characterized by reduced L-Glu current density in AII animals (Kempsell and Fieber 2014). Expression of iGluR genes in BSC neurons of cohorts 1 and 2 were compared (Fig. 3.6). Grin1-1 and Grin1-2 were significantly downregulated in AII BSC sensory neurons of cohort 1, just as in PVC neurons of this cohort (Fig. 3.6A, Grin1-1 AII:M expression 0.59±0.08, Grin1-2 61 AII:M expression 0.57±0.05, p£0.05 for each gene, Student’s t-test). These subunits were also significantly down-regulated in cohort 2 (Fig. 3.6B, Grin1-1 AII:M expression 0.64±0.04, Grin1-2 AII:M expression 0.56± 0.06, p£0.05, Student’s t-test), whereas only Grin1-1 had been significantly downregulated in cohort 2 PVC neurons. In addition to Grin1 subunits, the AMPAR subunit GluR1 had significantly decreased expression in cohort 2, just as in the AII PVC neurons of cohort 2 (Fig. 3.6B, AII:M expression 0.43±0.02, p£0.05, Student’s t-test). AMPAR subunit GluR3 and orphan subunit GluR6 had significantly reduced expression in cohort 1 AII BSC neurons (Fig. 3.6A, GluR3 AII:M expression 0.61±0.05, GluR6 AII:M expression 0.69±0.05, p£0.05 for each gene, Student’s t-test). Transcript copy numbers per 100 ng of RNA were also calculated for cohort 1, and age-related changes in iGluR expression described above for BSC neurons are displayed as copy numbers in Fig. 3.7. Cohort 1 AII abdominal ganglion are displayed as transcript copy numbers in Fig. 3.8. The abdominal ganglion showed no significant changes in expression of any iGluR subunits (Fig. 3.8). The abdominal ganglion was not tested in cohort 2. Subunit composition Percent contribution of each subunit to the total expression for each subtype was calculated to gain insights about possible components of complete functional iGluR in Aplysia. These data were calculated using cohort 1 only, because calculation of absolute copy number in this cohort allowed for comparison of expression data across different qPCR runs. It has previously been shown that Aplysia NMDAR and AMPAR subunits form monophyletic groups and thus may form complete receptors (Chapter 2). The 62 makeup of kainate-like receptors in Aplysia is less clear, therefore we were unable to calculate subunit composition for this subtype. In vertebrates, NMDAR are comprised of a heterotetramer of two Grin1 subunits and two Grin2 subunits. Thus, it is to be expected that they are expressed in approximately equal amounts. In both PVC and BSC sensory neurons, the two splice variants of Grin1 together comprised >90% of total NMDAR subtype expression (Fig. 3.9A and B). This was true in M as well as AII PVC and BSC neurons. There were no statistically significant changes in the percent contribution of any NMDAR subunit between M and AII (Fig. 3.9A and B, p£0.05, one-way ANOVA on a logistical regression). For example, Grin1-1 contributed ~80% of NMDAR expression in both M and AII PVC neurons. Similarly, expression of the two Grin1 splice variants comprised ~85% of total NMDAR subunit expression in abdominal ganglia, and did not vary with age (Fig. 3.9C). Thus, Grin1 contributed the majority of NMDAR mRNA expression in the ganglia tissues studied, with NMDAR subunit percentages not significantly altered during aging. AMPAR subunits were analyzed similarly. GluR1 was the predominately expressed subunit in PVC and BSC sensory neurons, representing ~60% of total AMPAR expression in both sensory neuron clusters (Fig. 3.10A and B). GluR1 was also the most highly expressed subunit in the abdominal ganglia at ~70% of total AMPAR expression (Fig. 3.10C). This was a significantly higher percentage of total AMPAR expression than observed in PVC and BSC neurons (p£0.05, one-way ANOVA on a logistical regression, Tukey’s post-hoc). In all three tissues there were no significant differences in percent composition of any AMPAR subunit in AII animals (Fig. 3.10A-C). 63 3.5 Discussion In this study, mRNA expression of iGluR subunits in mature and aged sensory neuron clusters of the Aplysia nervous system was quantified. Changes in mRNA expression can be used as a proxy for decreased in protein synthesis that results in phenotypic changes (Vogel and Marcotte 2012). Therefore, decreased expression of iGluR subunits observed in aged sensory neurons in this study may directly influence the amount of iGluR receptor protein, and decrease the number of receptors available for LGlu mediated transmission. There is also the possibility that transcriptional changes may not be directly correlated with changes in protein, such as changes in turnover rate requiring altered transcription to maintain the same function. However, there is ample evidence to suggest that altered transcription of receptor genes is correlated with altered physiology. For example, in the plainfin midshipman, males produce a vocal hum as a courting behavior only during the breeding season. Females with mature eggs exhibited a strong phonotaxis to male humming frequencies that is only present during this time, and is lost the remainder of the year (McKibben and Bass 1998; Sisneros and Bass 2003; Zeddies, et al. 2010). Increased sensitivity to male vocal hums during the mating season correlated with an increased number of calcium-activated potassium (BK) channels in auditory hair cells (Coffin, et al. 2012). Expression levels of two isoforms of BK channels quantified by qPCR, showed upregulation in reproductive compared to nonreproductive females (Rohmann, et al. 2013). Hence, increased channel expression during mating season is correlated with increased transcription of BK channel genes. 64 Altered expression in euryhaline fish, such as the Mozambique tilapia, is correlated with the drastic physiological changes necessary to maintain osmotic balance in rapidly changing salinity regimes. The hormone prolactin plays an important osmoregulatory role when euryhaline fish encounter freshwater, by preventing water uptake and excessive loss of ions (Manzon 2002). Expression of prolactin receptors in the gills of Mozambique tilapia quantified by qPCR were significantly increased when exposed to freshwater, consistent with its required induction for freshwater adaptation (Pierce, et al. 2007). In human cognitive disorders, particularly schizophrenia, reduced NMDAR function is a major component of the disease (Tsai and Coyle 2002; Paoletti and Neyton 2007). Reducing NMDAR expression in transgenic mice recapitulates the symptoms, suggesting that reduced NMDAR expression leads to observed physiological dysfunction involving NMDAR that is prevalent in the disease (Mohn, et al. 1999). Together these studies show that reduced expression of constituent receptors is be correlated with altered physiology during major life history events. Thus, qPCR results on iGluR in this study can be expected to have physiological relevance in the Aplysia nervous system. Grin1-1, a subunit of the NMDAR subtype, was expressed at greater than twofold more than any other iGluR subunit in PVC sensory neurons, BSC sensory neurons, and the abdominal ganglion. Grin1 mRNA previously has been demonstrated to be expressed throughout the Aplysia nervous system (Ha, et al. 2006), and is also widely expressed in Lymnaea neurons and mammalian brains (Akazawa, et al. 1994; Goebel and Poosch 1999; Ha, et al. 2006). Ubiquitous expression of Grin1-1 in the nervous system suggests an important role for NMDAR in iGluR physiology across Bilateria. The greater 65 expression of Grin1-1 compared to other subunits suggests that iGluR in Aplysia is mostly comprised of Grin 1-1, and that most L-Glu-mediated responses studied in Aplysia may be Grin1-1-mediated. Other iGluR subunits were expressed in significantly higher abundance in one ganglion, and therefore may be more greatly involved in mediating iGluR transmission in that ganglion. GluR5 was expressed at significantly greater copies in PVC neurons than in other ganglia, and Grin1-2, GluR2, and GluR4 were all more highly expressed in BSC neurons. Thus, different parts of the Aplysia nervous system may emphasize different iGluR subunits in iGluR-mediated neurotransmission. Not only were Grin1-1 subunits highly expressed in this study, they also exhibited reduced expression in AII PVC and BSC neurons. If Grin1-1 plays the largest role is LGlu mediated responses, as the expression data indicate, its down-regulation would be predicted to have an impact on L-Glu-induced current densities in aged animals. Fieber et al. (2010) noted decreased density and frequency of iGluR-activated whole cell currents in PVC neurons from Aplysia near the end of life, without changes in BSC neuron iGluR currents. Later, Kempsell and Fieber (2014) demonstrated that AII PVC and BSC sensory neurons both had reduced L-Glu-induced current density. This suggested that the reflex declines noted in aging are correlated with either impaired iGluR function or iGluR absence in the neurons that underlie the behaviors. Reduced Grin1-1 expression in AII sensory neurons observed in this study is proposed as a contributor to deficits in L-Glumediated excitability that, in turn, contribute to the impaired TWR and biting reflex. Declines in mRNA and protein expression of Grin1 subunits have been described in the brain of aged rats, mice, and macaque monkeys, suggesting that Grin1 subunits are 66 highly susceptible to the effects of aging (Gazzaley, et al. 1996; Hof, et al. 2002; Shi, et al. 2007; Newton, et al. 2008). Reduced NMDAR mRNA expression in aged mice resulted in reduced protein expression that was greater than observed reductions in mRNA (Magnusson, et al. 2002). Therefore, it is reasonable to propose that reduced expression of NMDAR subunits observed here led to a decreased number of NMDAR proteins and fewer iGluR in sensory neurons. Significantly reduced expression of other subunits was also found, and may contribute to impaired L-Glu-mediated reflex pathways and impaired sensory neuron excitability in AII. KA1 and GluR6 were reduced in AII PVC neurons while GluR3 and GluR6 were reduced in BSC neurons. No subunits had increased expression in AII. Agerelated reductions in iGluR expression have been described in several vertebrate species (Wardas, et al. 1996; Hof, et al. 2002; Shi, et al. 2007), with reduced binding of L-Glu in aged neurons linked to reduced numbers of functional iGluR (Wenk and Barnes 2000). Thus reduced expression is likely a reasonable proxy for reduced glutamatergic neurotransmission and reflex function in aging in these models. The effect of aging on iGluR expression varied between PVC and BSC neurons, and in the two different cohorts. Some expression differences, such as reduced Grin1-1 expression, were observed in both cohorts and both tissues. Other iGluR genes with reduced expression varied between different cohorts and ganglia. For example, GluR1 was significantly down-regulated in AII of both sensory neurons clusters of cohort 2, yet was not differentially expressed in cohort 1. The discrepancy in GluR1 expression in PVC neurons between cohorts did not manifest as a change in the behavioral aging of TWR. Different rates of aging of iGluR physiology also occurred in different cohorts and 67 ganglia of Aplysia. Fieber et al. (2010) found no iGluR current differences in BSC neurons of three cohorts with age, while these same cohorts showed differences in iGluR currents in PVC neurons. In contrast, Kempsell and Fieber (2014) did observe iGluR current differences with age in BSC neurons of a different cohort. Thus, there are both similarities and differences in molecular and physiological aging of L-Glu, depending on the cohort and neuronal clusters examined. Physiological outcomes and behaviors may not need to map 1:1 with differences in expression to contribute to declines in nervous system function with age. Other neurons in the Aplysia nervous system showed age-related physiological deficits correlated with reduced receptor mRNA. The cholinergic R15 bursting neuron in the abdominal ganglion that showed reduced responsiveness to acetylcholine in aged animals was correlated with its decreased expression of acetylcholine receptors (Akhmedov, et al. 2013) These results parallel those found in the sensory neurons in this study, with reduced neurotransmitter actions in isolated neurons coinciding with decreased expression of its receptors. While several iGluR subunits decreased expression in AII PVC and BSC neurons, there were zero changes in iGluR expression in the abdominal ganglion. This suggests that iGluR transmission may not have been affected in the abdominal ganglia and that aging of iGluR occurred differently in different parts of the nervous system. Specific gene expression changes that vary between different individual neurons during aging has also been described in other neurons of the Aplysia nervous system (Moroz and Kohn 2010; Kadakkuzha, et al. 2013). Moroz and Kohn (2010) found that in two cholinergic neurons, R2 in the abdominal ganglia and LPl1 in the left pleural ganglion, 68 approximately half of the genes significantly affected by aging occurred in either R2 or LPl1, but not in both. For example, potassium channel subunits increased expression in aged R2 neurons, but reduced expression in aged LPl1 neurons. Inhibitor of apoptosis homolog protein was highly expressed in aged R2 and was relatively absent in LPl1 (Moroz and Kohn 2010). Transcription factors CREB1 and CREB2 also show highly variable patterns of expression in aging and showed increased, decreased, or no changes in expression in different individual neurons of the abdominal ganglia (Kadakkuzha, et al. 2013). Hence, differential aging of different neurons is common in Aplysia, and we have extended these findings to include differential aging of iGluR subunits. Differential aging of iGluR also occurs in the vertebrate brain. Reduced density of NMDAR-mediated currents during aging has been reported in many regions of the vertebrate brain (Magnusson and Cotman 1993; Nicolle, et al. 1996; Wardas, et al. 1996; Mitchell and Anderson 1998). In those studies, receptor density decreased 20-50% in aged animals. It has also been reported that in other regions of the vertebrate brain, NMDAR subunits may be upregulated or experience no changes in expression during aging (Jouvenceau, et al. 1998; Potier, et al. 2000). It would be expected that complete NMDAR in Aplysia are formed from two Grin1 and two Grin2 subunits in Aplysia, based on vertebrate NMDAR that consist of obligatory heterotetramers, formed from two Grin1 subunits and either two Grin2 subunits or two Grin3 subunits (Traynelis, et al. 2010a; Paoletti 2011). NMDAR are the most well-conserved iGluR subtype across bilaterian species, including Aplysia (Chapter 2), and it is reasonable to assume that functional Aplysia iGluR are formed from similar heterotetramers. Grin3 subunits have not been characterized in Aplysia and many other 69 protostomes (Chapter 2). Grin1 was expressed ~10 times higher than Grin2 in all three Aplysia ganglia. If Aplysia NMDAR are formed as heterotetramers of Grin1 and Grin 2, it may be that Grin1 turnover rates are much higher than Grin2, requiring greater transcription to maintain higher turnover rates. Another possibility is that Grin1 may form homotetramers in Aplysia. However, expression of Aplysia Grin1 only in Xenopus oocytes did not induce L-Glu currents (Ha, et al. 2006), casting doubt on the hypothesis that high expression of Grin1-1 may indicate homotetramer formation. These results suggest that if Aplysia NMDAR are tetrameric, the makeup of complete receptors may not be able to be predicted based on expression levels. Altered composition of the subunits comprising complete tetrameric AMPAR and NMDAR is associated with decreased L-Glu-mediated transmission in the aging vertebrate brain. Variations in the subunits that make up a complete receptor alter the biophysical, pharmacological, and signaling properties of the receptor and can confer different messages to the same stimulus (Paoletti 2011; Paoletti, et al. 2013). No changes in the percent contribution of any subunit were found during aging for NMDAR or AMPAR in PVC and BSC sensory neurons. If complete tetrameric receptors in Aplysia are formed from subunits within each subtype as predicted, these results suggest that reduced L-Glu responsiveness of aged PVC and BSC neurons is likely due directly to reduced expression of iGluR subunits, rather than altered receptor composition. In conclusion, quantification of mRNA can be used as a powerful tool to examine underlying changes in the nervous system. During major life history events, such as aging, changes in mRNA expression in the nervous system can be correlated with and can be a proxy for alterations in neuronal physiology. To follow up on this idea, Chapter 70 4 examines whole-transcriptome mRNA in M and AII PVC neurons for changes that may be correlated with overall reduced neuronal activity of aged sensory neurons. 71 Cohort 1 Cohort 2 30 400 300 200 100 0 2 Time to relax tail (s) C 4 20 15 10 5 * 25 * 20 15 10 5 0 08 6 10 12 8.5 Age (mos) Age (mos) 30 * 25 20 15 10 5 0 8.5 12 Age (mos) * D Percent tail withdrawn Weight (g) 500 30 B 25 Time to right (s) Time to relax tail (s) 600 A 12 35 30 25 20 15 10 5 0 8.5 12 Age (mos) * * 8.5 12 Age (mos) Fig 3.1. Behavioral and weight correlates of aging in two cohorts of Aplysia. All error bars represent ±SE of the mean. A) Animal mass estimated from wet weight decreased beginning approximately 2 months before AII measurements (n=15). Dotted lines represent time points animals were sacrificed for qPCR. B) Mean time to right increased significantly in AII. C) TWR increased significantly in AII animals compared to M animals in both cohorts. D) Amplitude of tail withdrawal as fraction of the initial length withdrawn decreased significantly in AII. * denotes significantly different than M (p£0.05, Student’s t-test) 72 A B BSC Abdominal GluR1 Grin1-1 Grin1-2 GluR7 GluR1 Grin1-1 Grin1-2 GluR7 PVC GluR1 Grin1-2 Grin2 GluR2 GluR3 GluR4 GluR5 GluR6 GluR8 KA1 KA2 Grin1-1 GluR7 Grin1-2 GluR2 Grin2 KA2 GluR1 GluR2 GluR3 GluR4 GluR5 GluR6 GluR7 GluR8 KA1 KA2 C Grin2 GluR2 GluR3 GluR4 GluR5 GluR6 GluR8 KA1 KA2 Grin1-2 Grin2 GluR1 GluR2 GluR3 GluR4 GluR5 GluR6 GluR7 GluR8 KA1 KA2 Grin2 GluR5 GluR2 GluR8 GluR3 KA2 GluR4 GluR5 GluR6 GluR8 KA1 KA2 Grin1-2 Grin2 GluR2 GluR3 GluR4 GluR5 GluR6 GluR8 KA1 KA2 Grin1-2 Grin2 GluR1 GluR2 GluR3 GluR4 GluR5 GluR6 GluR7 GluR8 KA1 KA2 Grin2 GluR2 GluR3 GluR4 GluR5 GluR6 GluR8 KA1 KA2 Grin2 GluR2 GluR3 GluR4 GluR5 GluR6 GluR8 KA1 KA2 Fig. 3.2. Differential expression of iGluR subunits in sensory neuronal clusters and abdominal ganglia of both Mature and Aged II animals. Blue circles represent subunits that were expressed at significantly greater copies than all subunits listed underneath them in A) PVC sensory neurons B) BSC sensory neurons and C) whole abdominal ganglia. Grin1-1 was the most highly expressed subunit in all three ganglia tested and was expressed at significantly greater copies than any other iGluR subunits, including other significantly differentially expressed subunits (all other blue circled subunits; p£0.05, one-way ANOVA, Tukey’s post-hoc). *Subunits listed underneath are ordered by subtype (NMDAR, AMPAR, kainate/orphans) 73 PVC 35000 BSC Abdominal 30000 Calculated number of transcripts 25000 20000 15000 * * 10000 * 5000 * * * * * * NMDA AMPA GluR6 KA1 KA2 GluR7 GluR8 GluR5 GluR4 GluR3 GluR2 GluR1 Grin2 Grin1-2 Grin1-1 0 Kainate and orphans Subunit Fig. 3.3. Calculated number of iGluR transcripts in different ganglia. Many iGluR subunits were significantly differentially expressed across different ganglia. Number of transcripts is calculated per 100 ng of RNA. This data does not account for age-related changes in expression and includes both M and AII animals. *denotes significantly differentially expressed compared to all other ganglia (p£0.05, one-way ANOVA, Tukey’s post-hoc test) 74 A Relative expression compared to Mature 2.5 2 1.5 1 * * * * 0.5 0 AMPA NMDA Kainate/orphans Subunit B Relative expression compared to Mature 1.4 1.2 1 0.8 0.6 * * 0.4 0.2 0 AMPA NMDA Kainate Subunit Fig. 3.4. Relative expression of iGluR subunits in PVC sensory neurons of Mature and Aged II animals from two cohorts. A) Relative expression for all iGluR subunits in AII compared to M (dotted line) in PVC sensory neurons of cohort 1. Several NMDAR and kainate/orphan receptor subunits were significantly down-regulated in AII PVC neurons. There were no changes in expression to any AMPAR subunits. B) Relative expression of iGluR subunits in PVC sensory neurons of AII animals compared to M in cohort 2. Only select subunits with the greatest expression in cohort 1 were analyzed in cohort 2. Grin1-1 and GluR1 had significantly reduced expression in AII compared to M. AII expression level for each gene was normalized against expression in M PVC neurons (dotted line). *denotes significantly reduced expression compared to Mature (Student’s t-test, p£0.05; n=6 for cohort 1, n=4 for cohort 2) 75 Mature Calculated number of transcripts 50000 Aged II 40000 30000 * 20000 10000 * * KA1 GluR6 * NMDA AMPA KA2 GluR7 GluR8 GluR5 GluR4 GluR3 GluR2 GluR1 Grin2 Grin1-2 Grin1-1 0 Kainate and orphans Subunit Fig. 3.5. Absolute transcript quantification of iGluR subunits in PVC sensory neurons of Mature and Aged II animals from cohort 1. Expression changes with age of iGluR genes from cohort 1 displayed in Fig. 3.4, shown as the number of transcripts per 100 ng of RNA. *denotes significantly reduced expression compared to Mature 76 A Relative expression compared to Mature 1.2 1 0.8 * * * * 0.6 0.4 0.2 0 AMPA NMDA Kainate/orphans Subunit B Relative expression compared to Mature 1 0.9 0.8 0.7 * * 0.6 0.5 * 0.4 0.3 0.2 0.1 0 AMPA NMDA Kainate Subunit Fig. 3.6. Relative expression of iGluR subunits in BSC sensory neurons of Mature and Aged II animals from two cohorts. A) Relative expression for all iGluR genes in AII BS sensory neurons compared to M in cohort 1. Grin1-1 and Grin1-2 NMDAR subtype genes were significantly down-regulated in AII. GluR3 and GluR6 also had significantly reduced expression in AII BSC neurons. B) Relative expression of iGluR subunits in AII BSC sensory neurons in cohort 2. Grin1-1 and Grin-2 were significantly down-regulated, which was also observed in cohort 1. GluR1 was also down-regulated in AII. AII expression levels were normalized against expression in M PVC neurons (dotted line). *denotes significantly reduced expression compared to M (Student’s t-test, p£0.05; n=6 for cohort 1, n=4 for cohort 2) 77 50000 Mature Aged II 45000 40000 Calculated number of transcripts 35000 30000 25000 * 20000 15000 * 10000 5000 * * GluR6 KA1 KA2 GluR7 GluR8 GluR5 GluR4 GluR3 GluR2 GluR1 Grin2 Grin1-2 Grin1-1 0 Fig 3.7. Absolute transcript quantification of iGluR subunits in BSC sensory neurons of Mature and Aged II animals from cohort 1. Expression changes with age of iGluR genes from cohort 1 displayed in Fig. 3.6, shown as the number of transcripts per 100 ng of RNA. *denotes significantly reduced expression compared to Mature 78 50000 Mature Aged II 45000 40000 Calculated number of transcripts 35000 30000 25000 20000 15000 10000 5000 NMDA AMPA GluR6 KA1 KA2 GluR7 GluR8 GluR5 GluR4 GluR3 GluR2 GluR1 Grin2 Grin1-2 Grin1-1 0 Kainate and orphans Subunit Fig. 3.8. Calculauted number of iGluR transcripts in Mature and Aged II whole abdominal ganglion of cohort 1. In the abdominal ganglion, zero iGluR subunits exhibited altered expression in AII animals. 79 Mature 2000 B 1800 Calculated number of transcripts % of total NMDAR expression 100 1600 90 1400 80 70 60 1200 1000 800 50 40 30 20 10 600 400 200 0 Grin1 Grin2 Subunit 0 100 % of total NMDAR expression A Aged II 90 80 70 60 50 40 30 20 10 0 Grin1-1 C Grin1-2 Subunit Grin2 Grin1-1 Grin1-2 Subunit Grin2 % of total NMDAR expression 100 90 80 70 60 50 40 30 20 10 0 Grin1-1 Grin1-2 Subunit Grin2 Fig. 3.9. Percent composition of NMDAR subtype subunits. A) Expression of NMDAR subunits in PVC sensory neurons showed no significant change in percent composition during aging. B) Percent composition also did not change significantly with age in BSC sensory neurons. C) Composition of NMDAR was not significantly altered in abdominal ganglia of AII animals. p>0.05 for all comparisons 80 Mature 2000 Aged II 1800 30 600 400 20 Subunit Grin2 GluR8 GluR5 GluR3 Grin1 GluR4 0 GluR1 0 GluR2 10 200 Subunit C 60 50 40 30 20 10 0 GluR8 40 800 70 GluR5 50 1000 80 GluR4 60 1200 90 GluR3 70 1400 GluR2 80 100 % of total AMPAR expression 90 Calculated number of transcripts % of total AMPAR expression 100 B 1600 GluR1 A Subunit % of total AMPAR expression 100 90 80 70 60 50 40 30 20 10 GluR8 GluR5 GluR4 GluR3 GluR2 GluR1 0 Subunit Fig. 3.10. Percent composition of AMPAR subtype subunits. The percent composition of AMPAR was not altered in AII compared to Mature in A) PVC sensory neurons B) BSC sensory neurons and C) whole abdominal ganglion. one-way ANOVA on a logistical regression, p>0.05 for all comparisons Chapter 4: Transcriptional changes during aging in sensory neurons innervating TWR in Aplysia 4.1 Summary Whole-transcriptome expression changes in sensory neurons of aging Aplysia were studied in two cohorts at maturity and advanced age. Behavioral aging of both the righting and tail withdrawal reflexes was confirmed in 2 cohorts of animals. Sensory neurons located in the pleural ventral caudal region of the pleural ganglion that control tail withdrawal were then tested for differential expression of genes between maturity and advanced age using the technique known as RNASeq. Many genes and gene ontology categories that were significantly altered in aged animals compared to maturity code for genes related to ion channels and neuronal function. This implies that aging has a profound effect on PVC neurons at the transcriptional level and corroborates physiological experiments that found reduced PVC neuron function with age. Several iGluR subunits were significantly downregulated in aged PVC neurons, corroborating qPCR evidence in Chapter 3. Many upregulated genes in Aged II PVC neurons are associated with stress response and defense against oxidative damage. Increased stress response during aging has been characterized in many other species and thus may be a ubiquitous effect on the aging nervous system of all animals. 4.2 Background Aging is a biological process resulting in progressive declines in physiological function, and is the result of a multitude of changes at the cellular and molecular levels. 81 82 Accumulation of reactive oxygen species (ROS) has been proposed to play a major role in age-related loss of function via oxidative damage. ROS are oxygen-based radicals produced as a byproduct of cellular metabolism (Braunersreuther and Jaquet 2012; Nayernia, et al. 2014). In healthy neurons, scavenging molecules and enzyme levels match ROS production, preventing ROS accumulation and oxidative damage to neurons (Circu and Aw 2010). During aging, increased ROS production and decreased clearance can lead to oxidative stress that contributes to DNA damage and cellular apoptosis observed in aging and other neurodegenerative diseases (Uttara, et al. 2009). Accumulation of ROS in the mitochondria leads to significant damage to the mitochondrial membrane, further exacerbating ROS production (Shigenaga, et al. 1994; Toescu, et al. 2000; Floyd and Hensley 2002). Large-scale gene expression changes occur in the nervous systems of both vertebrates and invertebrates during aging. In the human cortex, genes that mediate synaptic plasticity such as protein kinase C, synapsin, and calmodulin had reduced expression throughout aging (Lu, et al. 2004). These researchers suggested that DNA damage in promoter regions of these genes may underlie reduced expression. Reduced expression of genes involved in synaptic plasticity and neuronal function has also been described in Rhesus macaques, rats, and mice (Bishop, et al. 2010). Increased stress response in aged animals appears to be ubiquitous and has been characterized in many species (Yankner, et al. 2008). In the invertebrate model Drosophila melanogaster, both oxidative stress and aging are characterized by similar changes in gene expression, including upregulation of stress response genes such as heat shock proteins and antioxidant genes (Landis, et al. 2004). 83 In addition to transcriptional changes, alterations in neuronal morphology and physiology have also been implicated in age-related cognitive impairment in the vertebrate nervous system. Age-related performance decreases in tests of the memory of the location of a reward in Rhesus macaques may be due to loss of dendritic spines on glutamatergic pyramidal neurons and decreased density of synapses in the prefrontal cortex (Dumitriu, et al. 2010). Declines in recognition memory in aged rats has also been associated with reduced density of dendritic spines (Wallace, et al. 2007). Rhesus macaque displayed reduced frequency of glutamate receptor-mediated postsynaptic currents in pyramidal cells that has been linked with impaired memory performance (Luebke, et al. 2004). Impaired working memory performance in aged rats was significantly correlated with dopamine deficiencies in the frontal cortex (Luine, et al. 1990). Cumulatively, morphological, physiological, and transcriptional changes with age can result in whole organism declines in learning and memory, locomotion, and reflex systems (Chung, et al. 2005; Kumar and Foster 2007; Bordner, et al. 2011; Kempsell and Fieber 2015b) Using vertebrate model systems, behavioral deficits with age can be attributed to neuronal changes in specific brain regions. However, analysis of changes in the exact neurons underlying behavioral deficits is difficult due to the complexity of the mammalian brain and complicated neural networks. Aplysia californica (Aplysia) is a marine mollusk that has been used extensively in research correlating physiological, behavioral, and molecular changes in neuronal circuits during aging (Rattan and Peretz 1981; Peretz, et al. 1982; Peretz, et al. 1984; Kandel 2001; Liu, et al. 2011; Kempsell and Fieber 2014, 2015b). Due to a less complex 84 nervous system than vertebrates, behavioral alterations in aged Aplysia can be studied at the physiological and molecular levels in the individual neurons that control the behaviors. The siphon-gill withdraw reflex (GWR) is one such behavior in which agedrelated changes in behavior have been correlated with physiological and molecular alterations in the underlying neural circuit. In GWR, a weak stimulus applied to the siphon causes the siphon and gill to withdraw into the mantle for protection (Pinsker, et al. 1970). This circuit consists of 24 mechanosensory neurons that make monosynaptic connects to six motoneurons in the abdominal ganglion (Byrne, Castellucci, Carew, et al. 1978). In aged animals, the reflex shows reduced responsiveness to low intensity stimuli and altered habituation compared to young animals (Rattan and Peretz 1981). These behavioral deficits with age have been linked to alterations in morphology of the neuromuscular junction in the motoneurons of GWR, as well as decreased input resistance and decreased size of post synaptic potentials evoked by gill stimulation, both of which reduced excitability and synaptic communication (Peretz, et al. 1984). These studies on GWR provided evidence that behavioral perturbations with age were a consequence of alterations in the precise neurons that innervate the reflex. Recent molecular work on aging of GWR has focused on the L7 motoneuron, one of the six responsible for moving the tail. L7 is one of the largest neurons in the animal kingdom and contains sufficient RNA for single neuron molecular studies. High mRNA abundance of amyloid precursor protein (APP) was observed in L7 using in situ hybridization (Moroz and Kohn 2010). Increased APP production with age has been associated with neurodegenerative disease in mammals, including humans (Hardy and 85 Selkoe 2002). In healthy neuronal tissue, production of APP is matched by its clearance from the tissue (Castellano, et al. 2011). In aging, however, increased expression of APP in brain tissue and reduced clearance leads to deposition of amyloid beta plaques. Amyloid plaques reduce synaptic connections and overall neuronal function, causing memory impairments (Shankar, et al. 2008). Thus, APP in aged Aplysia may lead to buildup of amyloid plaques in L7, contributing to decreased function and synaptic transmission between L7 and pinnule muscles of GWR (Peretz, et al. 1982). APP was also found to be highly expressed in other motoneurons in the abdominal ganglion, as well as in R2, a cholinergic neuron (Moroz and Kohn 2010). Thus, APP production may be a consequence of aging in many parts of the Aplysia nervous system that contributes to reduced neuronal function with age. Other genes with altered expression in L7 of aged Aplysia include transcription factors and transcriptional regulators CREB1, CREB2, and S6K (Kadakkuzha, et al. 2013). Transcription factors are proteins that bind to specific DNA sequences and are responsible for controlling the rate of gene transcription of many other genes. Alterations in the expression of transcription factors may in turn affect gene expression of other downstream genes. Hence, changes in transcription factor abundance may be a contributing factor in age-associated changes in gene expression. Aging in the nervous system of Aplysia has been shown to result in ~10% of transcripts with ³ 2 fold change in expression (Kadakkuzha, et al. 2013), which may be due in part to transcription factor regulation. Together, these studies have shown that age-related alterations in GWR behavior are correlated with a multitude of changes in morphology, physiology, and transcription in the underlying neuronal circuitry. 86 Physiological and molecular correlates of aging have also been described in R15, a bursting neuron in the abdominal ganglion. R15 is believed to be involved in egglaying. Alterations in acetylcholine-induced action potentials occurred in R15 during aging, as well as reduced expression of some acetylcholine receptors (Akhmedov, et al. 2013). Transcriptional changes in biological pathways for cell signaling, cellular function, and neurological diseases in aged R15 neurons correlated with observed physiological declines (Kadakkuzha, et al. 2013). Although there are ample physiological and molecular data with aging, the behavioral correlates for these changes is an understudied area of research. In this study, a different, well-defined neuronal circuit in Aplysia, the tail withdrawal reflex (TWR), was used. TWR shows predictable behavioral and physiological aging. Direct tail stimulation initiating TWR activates mechanosensory neurons in the pleural ventral caudal region of the left and right pleural ganglion (PVC). PVC neurons make monosynaptic connections to three motoneurons of the pedal ganglion (P7-P9), which release neurotransmitter at the neuromuscular junction to trigger movement of the tail muscle (Walters, et al. 1983a). Behavioral studies have shown characteristic, predictable increases in TWR time to completion as the animal ages (Kempsell and Fieber 2014). Aplysia can be classified as Mature (M), Aged I (AI), and Aged II (AII) based on several factors: time since the onset of maturity, diminished reflex behaviors, as well as physiological changes, discussed below. Behavioral aging of TWR in AII animals is correlated with physiological changes in both PVC sensory neurons and pedal motoneurons of the reflex. AII PVC sensory neurons exhibited decreased frequency and density of currents induced by the both L- 87 glutamate (L-Glu) and D-aspartate (D-Asp) compared to M (Fieber, et al. 2010; Kempsell and Fieber 2014). AII pedal motoneurons of TWR exhibited reduced excitatory postsynaptic potential amplitude (EPSP) after tail touch compared to M (Kempsell and Fieber 2015b). Together, reduced excitability of PVC sensory neurons and decreased EPSP amplitude in PVC motoneurons suggest that synaptic transmission is negatively affected by age. Reduced synaptic transmission contributes to aging of TWR and increased times to complete the reflex in AII animals. Alterations in sensitization, a form of nonassociative learning, also occurred in TWR of AII animals. During sensitization, repetition of a stimulus results in a progressive amplification of the response via serotonin (5-HT) released by facilitory interneurons onto the sensorimotor synapse (Walters, et al. 1983b; Glanzman, et al. 1989; Mackey, et al. 1989). 5-HT application to TWR neurons resulted in increased excitability of PVC sensory neurons and increased transmission between PVC neurons and tail motoneurons (Kempsell and Fieber 2015b). In M animals, repeated tail shocks also produced sensitization of the reflex, via increased amplitude and duration of TWR to a subsequent stimulus (Walters, et al. 1983b). In AII animals, tail shocks and 5-HT application failed to increase excitatory responses in PVC neurons, and EPSP amplitude in tail motoneurons was significantly smaller compared to M (Kempsell and Fieber 2015b). Thus, reduced facilitation of TWR in AII is observed the level of both behavior and the neural circuit. Studies on the behavioral and physiological correlates of aging of TWR have greatly enhanced our understanding of the neuronal mechanisms that underlie diminished responses and reduced synaptic facilitation in AII. Diminished excitability of PVC 88 sensory neurons and reduced EPSP amplitude evoked in tail motoneurons both contribute to aging of TWR. It has been proposed that decreased EPSP in tail motoneurons may be the direct result of decreased excitability in PVC sensory neurons (Kempsell and Fieber 2014). The cause of altered PVC function in AII is a critical question that can provide insights into aging of TWR. The goal of this study was to characterize molecular correlates of altered physiology and behavior in aged sensory neurons innervating TWR. Wholetranscriptome changes were analyzed in PVC sensory neurons from sexually mature and old Aplysia, using RNASeq. First, behavioral experiments were conducted to confirm aging of TWR. Correlates of aging, including behavioral assessments, were compared against Kempsell and Fieber (2014) to designate the stage of aging of age 12 month animals as AII. Then, RNASeq was used to analyze whole-transcriptome changes in expression in PVC of M and AII Aplysia. 4.3 Materials and Methods Animal rearing and behavioral assessments of aging Two cohorts of Aplysia, that originated from differently parented egg masses of wild-caught animals, were reared at the University of Miami National Resource for Aplysia. Both cohorts were maintained at a maximum of 5 per cage and fed an ad lib diet of Agardhiella subulata throughout their life as previously described (Gerdes and Fieber 2006). Righting and TWR assessments were executed beginning at M on 6-8 randomly selected animals from each cohort using previously described protocols (Chapter 3). M 89 was defined as the day the first egg mass was found and occurred at age 8 months in both cohorts. To quantify behavioral aging, behavioral assessments were continued monthly on randomly selected animals (n=6-8) from each cohort through the end of life, age ~12.5 months in both cohorts. RNA extraction and sequencing preparation RNA was extracted from PVC neurons from six animals from each cohort at M and six animals from each cohort at AII. The animals chosen for dissection of ganglia and RNA extraction from neural clusters were a randomly selected subset of the animals tested for TWR and righting at M and AII. This allowed for direct correlation of an individual animals’ behavioral assessments with its expression profile. To remove PVC neurons for RNA extraction, animals were first anesthetized by injection of 1/6th body weight of isotonic MgCl2. The CNS was then removed by severing the connectives and removal of all nervous system ganglia. PVC neurons were separated from the remainder of the pleural ganglion by pinning the pedal and pleural ganglia tightly in a sylguarded 35mm plastic dish filled with artificial sea water (ASW: 417mM NaCl, 10mM KCl, 10mM CaCl2, 55mM MgCl2, and 15mM HEPES-NaOH, pH 7.6). PVC neurons were peeled away from the remainder of the pleural ganglion with a pair of fine forceps after removal of the surrounding connective tissue. PVC neurons from both hemiganglion were pooled as a single sample. Total RNA was extracted with the Qiagen RNeasy Micro Kit (Cat. #74004) following the manufacturer’s instructions. Samples were treated with DNase to remove any contaminating DNA. Purified RNA samples were stored in 2.5 volumes of ethanol with 0.3M sodium acetate at -20°C until further processing. Samples in ethanol and 90 sodium acetate were spun in a microfuge at 4°C at full speed for 20 minutes, ethanol removed, and the RNA pellet resuspended in 15µl of TE buffer before library preparation. RNA quantity and quality were assessed with both a Nanodrop (model ND1000) and an Agilent 2100 Bioanalyzer prior to library preparation. **Two hundred ng of RNA from the 3 highest quality RNA samples from each cohort at M and AII were used in downstream library preparation. The highest quality RNA samples were defined as having sufficient RNA quantity and 260/280nm ratio ~2 to ensure no DNA or protein contamination. Libraries were prepared using the Illumina TruSeq Stranded Total RNA Low-Throughput Library Prep Kit (Cat. #RS-122-2201) following to the manufacturer’s instructions. Beckman Coulter Agencourt RNAClean XP (cat# A63987) and Beckman Coulter Agencourt AMPure XP (cat# A63881) beads were used for recovery of RNA and DNA, respectively. Following library preparation, samples were assessed on the bioanalyzer for quantification and to verify 200-300bp fragment sizes. Libraries were sequenced as 100bp paired-end reads by Elim Biopharmaceuticals (Hayward, CA) in one lane on a HiSeq 2500 high-throughput sequencer (Illumina). Annotation information Aplysia exon information (ref_AplCal3.0_gnomon_scaffolds.gff3) and genome annotation (version AplCal3.0 from July, 2015) were downloaded from the NCBI ftp server database. Data processing Raw reads were quality filtered using the fastaxtoolkit (<http://hannonlab.cshl.edu/fastx_toolkit/>). When a base pair quality score of <20 was encountered in a read, the remainder of the read was cut off and removed. Reads shorter 91 than 20 base pairs after trimming and removal of barcodes were also removed. Quality score and nucleotide distributions for each library were visualized using the Galaxy web server (Blankenberg, et al. 2010). rRNA was removed during sequencing preparation, however M sample #6 still contained many rRNA reads in its library. In order to assess an accurate number of total reads used for downstream analysis it was necessary to remove rRNA reads from this library prior to analysis. rRNA reads were removed by mapping all reads to the Aplysia rRNA annotation with the STAR aligner (parameters described below) and discarding reads that aligned to the rRNA annotation. Therefore, downstream analysis did not contain rRNA reads. Mapping Reads to the genome with Tophat2 and STAR After quality filtering, reads were mapped to the Aplysia genome using both the Tophat2 and STAR aligners. The indexed genome for use with Tophat2 was built using Bowtie 2.2.6.0 (Langmead and Salzberg 2012) with default parameters from the AplCal3.0 genome fasta file. The quality controlled reads were then aligned using Tophat2 v2.1.0 (Kim, et al. 2013) to the indexed genome. Additional options beyond the default that were used during alignments were --no-discordant and changes to --mateinner-dist, as described below. The –no-discordant option removes discordant pairs from the alignments. While some of the discordant pairs may be true fusion transcripts, the Tophat2 authors found that most were mis-mapping artifacts. Because fragment lengths used for sequencing ranged from 250bp to 300bp, the alignments were run with the expected distance between mate pairs at both 50bp and 100bp. We found that the number of uniquely mapped reads 92 increased ~1-2% with the mate pair distance at 50bp in 3 libraries, and this option was then used for the remaining libraries. Uniquely mapped reads were assigned to Aplysia RefSeq transcripts using featureCounts (Liao, et al. 2014). Genome index files for use with STAR were generated using the --runMode genomeGenerate option using both the Aplysia AplCal3.0 genome and the ref_AplCal3.0_gnomon_scaffolds.gff3 transcript annotation. Quality controlled reads were aligned using the STAR aligner (Dobin, et al. 2013). The --quantMode Genecounts option was utilized to count the number of reads uniquely mapping to each transcript using the HTSeq-count program (Anders, et al. 2014). Mapping and counting for pairedend and singletons were executed separately, and total counts for each transcript was determined by summing the counts for the paired and singletons reads for each sample. Differential expression analysis Statistical testing for differential expression (DE) of transcript counts was performed using DeSeq2 (Love, et al. 2014), a method based on the negative binomial distribution and performed in the R statistical environment (Team 2014). Only reads that uniquely mapped to the genome were counted and used for analysis. Reads that were expressed in at least two biological replicates were analyzed for DE. Raw read counts were normalized in DeSeq2 to adjust for differences in library sizes. Significant DE was defined as p£0.05 after false discovery rate (FDR) correction. Data were normalized using the regularized log transformation for principal component analysis (PCA). Of the top 50 most significantly DE genes, 17 were classified as uncharacterized proteins. Each uncharacterized protein was analyzed in a BLAST search against human proteins and for conversed domains in InterPro (Hunter, et al. 2011) to attempt to 93 annotate these genes. Three genes were annotated in this way: CCAP1, DNA ligase-like, and glycine N-acyltransferase-like. None of these three genes were previously annotated in the Aplysia transcriptome database. Gene ontology enrichment Transcripts found to be DE were first translated into their corresponding protein sequences prior to gene ontology analysis. Translated sequences were blasted against the Homo sapiens refseq protein database (version 9606.9558) using Blast2GO (ver. 3.3, Evalue£1.0E-3), to identify proteins with human homologs for gene ontology (GO) analysis (Götz, et al. 2008). This allowed for mapping of DE genes to the better-studied and annotated human GO pathways. GO annotation terms with E-value≤1.0E-6 were used in downstream analysis. This process was repeated for the full Aplysia transcriptome (AplCal3.0) as a reference set for enrichment analysis. GO terms of proteins translations from DE transcripts were tested against the reference set of GO terms for the full Aplysia genome annotation to test for biological processes and molecular functions that were over-represented (enriched). Using a Fisher’s exact test, GO terms that were over-represented at an FDR corrected pvalue£0.05 were identified. We were particularly interested in the effects of aging on neuroexcitability, and we found many GO categories involved in ion channel activity. To reduce redundancy in these very similar categories, these were combined into a single GO category during the analysis which we called “ion channel associated proteins.” P-value for ion channel associated proteins was calculated by using all unique genes in both the DE genes and the full genome with ion channel function and a Fisher’s exact test in R. To calculate an FDR 94 corrected p-value for this category, p-values from all GO categories from DESeq2 analysis were FDR corrected using the fdrtools package in R (Strimmer 2008). Categories combined for ion channel associated proteins are highlighted in the complete list of enriched GO terms (Supplementary File 1). qPCR validation of selected genes qPCR was used to verify the expression status of selected genes identified from the RNASeq analysis. For these experiments, different animals from the same two cohorts as the RNASeq experiment were used. Four animals at M and four animals at AII, thus 8 total animals, were tested for the righting and TWR, then sacrificed for RNA extraction as previously described for RNASeq. RNA was quantified on a Nanodrop (model ND-1000), and 100 ng of RNA was reverse transcribed to cDNA using the SuperScript III First-Strand Synthesis System with random hexamer primers (Invitrogen). The resulting cDNA was diluted 1:5 with nuclease-free H2O to provide the working concentration of cDNA. Primers designed for each gene were 18-22 bp and generated target amplicons of 75-125 bp. Primer sequences and amplicon length for each gene can be found in Table 4.1. To assess amplification of the desired target sequences, each set of primers was tested by PCR amplification of cDNA. PCR products were separated on a 1% agarose gel to confirm the appropriate size for each amplicon. These bands from the agarose gel were then cut out and the DNA purified using a Qiagen Gel Purification kit and purified gel products were sequenced by Genewiz (South Plainfield, NJ) to confirm amplification of the desired target sequence. 95 qPCR reactions were carried out on a Stratagene Mx3005P with SYBR Green master mix under the following conditions: 95°C for 10 minutes, followed by 40 cycles of 95°C for 15 sec, 58°C for 30 sec, 72°C for 30 sec. For each biological replicate two technical replicates were performed. All run were normalized to GAPDH control and relative standard curves were used for analysis. Expression ratios were calculated for genes between M and AII, as described in Chapter 3. 4.4 Results Behavioral changes in TWR and righting with age There were no significant differences observed in either TWR or righting between the two cohorts at any age (Student’s t-test, p£ 0.05). Therefore, times were combined from the two cohorts for analysis of behavioral aging. TWR significantly increased with age in both cohorts (Fig. 4.1A; 14.0±0.4 sec at age 8-10 mos (mean±SE), 19.6±0.6 sec at 11 mos, and 26.4±0.7 sec at 12 mos; one-way ANOVA p£ 0.05; Tukey’s post-hoc analysis). Time to right also increased significantly with age (Fig. 4.1B; 13.6±0.4 sec at age 8-10 mos, 17.1±0.5 sec at age 11 mos, 20.1s±0.5 sec at age 12 mos; one-way ANOVA p£0.05; Tukey’s post-hoc analysis). Age>11 months, TWR time>20.1 sec, and righting time > 18 sec in age 12 month animals of this study corresponded to AII based on previously defined stages of aging in Aplysia (Kempsell and Fieber 2014). Read depth and mapping After quality control filtering, a total of ~210 million reads were used for downstream analysis (an average of ~17.5 million reads/individual). Quality score 96 boxplots of the reads showed that base quality scores were high for all 100bp, indicating high quality sample preparation and sequencing (data not shown). Genomic mapping of reads was performed using both STAR and Tophat2. 5070% of reads per individual uniquely mapped to the NCBI Aplysia californica Annotation Release 101 genome using either aligner (assembly AplCal3.0). In all samples STAR yielded a higher percentage of uniquely mapped reads compared to Tophat2 (~5-10%, data not shown). A summary of read depth and mapping for each library with the STAR aligner can be found in Table 4.2. Identifying DE genes in PVC neurons with age RNA from PVC neurons from six biological replicates at M and AII were isolated and analyzed for DE. The correlation between the biological replicates within each group was calculated with a principal component analysis (PCA) to determine if overall gene expression patterns differ between M and AII animals (Fig. 4.2). This analysis showed that the first two principal components accounted for 57% of the variability in the dataset. Most biological replicates clustered together, indicating that expression profiles were most similar between individual M animals on the one hand and individual AII animals on the other. One exception was a single M animal that showed an intermediate expression profile that did not clearly belong to either M or AII (Fig. 4.2, circled). There were 1202 and 1132 DE genes identified with STAR and Tophat2, respectively. Of these DE transcripts, >80% were identified with both analyses (Fig. 4.3). Due to high agreement between the methods, downstream gene ontology analysis used only the STAR aligner data. A full list of DE transcripts from STAR at p≤0.05 can be found in Supplementary File 2. 97 In aging, approximately as many genes were upregulated as were downregulated in PVC neurons (Fig. 4.4). Six hundred sixty-five of the 1202 DE genes (55.3%) increased expression in AII animals, with 537 genes (46.7%) having decreased expression. A heatmap of the top 50 genes with the lowest p-values can be found in Fig. 4.5. Several of the genes in this list that exhibited decreased expression in AII animals play a role in neuronal excitability, including voltage-gated potassium channels and ionotropic glutamate receptors (iGluR). Other genes with decreased expression are involved in induction of long-term memory, including adenylate cyclase (Fig. 4.5) and the catalytic subunit of protein kinase A (PKA; Supplementary File 2). Several transcripts that were upregulated in AII play a role in cell protection and oxidative stress response, including major vault protein, heat shock protein, and multi-drug resistance protein. To determine if batch effects from using two cohorts played a role in DE expression, DE was analyzed between the two cohorts, without accounting for age. We found 25 DE transcripts between the two cohorts (p≤0.05, FDR corrected). None of these 25 genes were also DE in the aging analysis. This is a very low number of transcripts compared to >1200 DE transcripts between age groups, and indicated that there were few differences in gene expression due to the biological variability of the two cohorts. Therefore, differences in expression in this study are likely attributed to transcriptional changes between M and AII. Expression of iGluR subunits iGluR subunits were analyzed to test if reduced expression of iGluR coincided with previously observed declines in L-Glu mediated responses of AII PVC neurons. STAR was unable to differentiate between the two splice variants of Grin1, Grin1-1 and 98 Grin1-2, therefore Grin1 expression in this study encompasses both splice variants. There were no changes in expression of NMDAR subunits Grin1 or Grin2 (Fig 4.6). GluR2, GluR3, belonging to the AMPA subtype, and KA2, a kainate subunit, were significantly down-regulated in AII PVC neurons compared to M PVC neurons. GluR4 and KA1 subunits were not analyzed due to a low number of counts for these two genes. These results are similar to qPCR results of Chapter 3, in that some iGluR genes were downregulated in AII PVC neurons, and zero subunits were upregulated (See Figs. 3.4 and 3.5, in Chapter 3). However, all down-regulated iGluR genes observed with RNASeq were different than those found using qPCR. The percent contribution of each subunit to the total expression for NMDAR and AMPAR subtypes was calculated to determine if receptor makeup may have changed during aging. There were no statistically significant changes in the percent contribution of either Grin1 or Grin2 NMDAR subunits between M and AII (Fig. 4.7A). In AMPAR subunits, GluR1 comprised a significantly greater percentage of the total AMPAR expression in AII than at M, and GluR3 contributed a significantly smaller percentage to total AMPAR expression (Fig. 4.7B). These findings largely corroborated the aging results of Chapter 3, where no significant changes in subunit composition were found. The apparent substitution in AMPAR of GluR3 for GluR1 found here was the exception. Gene ontology analysis DE genes were mapped to gene ontology pathways using Blast2GO to determine the biological processes and molecular functions that were most significantly affected during aging. BLAST hits matching human proteins were identified for 1054 of the 1202 DE Aplysia proteins (E-value<1E-3). Approximately 1000 gene ontology terms were 99 found to be over-represented in the analysis (p£0.05, FDR corrected), suggesting that many functions were significantly altered in AII animals. A complete list of enriched ontology terms can be found in Supplementary File 1. Fig. 4.8 presents the most significantly enriched GO categories. The GO category ion channel associated genes was the most significantly enriched molecular function in PVC neurons (p£0.05, Fig. 4.8). Furthermore, >80% of DE ion channel associated genes showed reduced expression in aged animals, suggesting decreased ion channel function occurred in aged animals (Fig 4.9). Synaptic transmission was also a significantly enriched biological process, further emphasizing alterations in neuronal transmission in AII (Supplementary File 1). Response to stress was also found to be an enriched biological process (Supplementary File 1). Approximately 70% of stress response genes DE showed increased expression in aged animals, consistent with findings of increased stress response with age in many other species (Fig. 4.10, Bishop, et al. 2010). qPCR verification of RNASeq A subset of nine genes important to neuronal excitability in humans were quantified via qPCR at M and AII, using new RNA samples from the same two cohorts sampled for RNASeq. Six of the 9 genes selected for qPCR showed significant DE in RNASeq (Fig. 4.11). Six of the genes coded for ion channel proteins: voltage-gated potassium channel Shaw (Shaw), voltage-gated potassium channel Shaker (Shaker), voltage-dependent calcium channel subunit alpha-2/delta-3 (Ca2+ subunit alpha), and sodium channel alpha-subunit SCAP1. CREB3 was chosen for its role in cell 100 communication and calcium ion transport. CCAAT/enhancer-binding protein (C/EBP) is a transcription factor important for transcriptional regulation of many genes. Three other genes tested via qPCR, CREB2, amyloid protein b, and huntingtin were not found to be DE in RNASeq, but were included for qPCR as genes previously tested in Aplysia nervous tissue. Two genes had significantly different expression in both RNASeq and qPCR: Shaw and Ca+ subunit alpha (Fig. 4.11, p£0.05). There were no significant changes in expression for the remaining 7 genes via qPCR. 4.5 Discussion Aplysia is an emerging model of aging in the nervous system due to a relatively compact life span of one year and well-mapped neural circuits (Frazier, et al. 1967; Carew, et al. 1981; Rattan and Peretz 1981; Stommes, et al. 2005). The use of two cohorts of hatchery reared Aplysia provided the advantage of studying 2 families of sibling animals with reduced biological variability compared to wild caught animals. PVC neurons are the primary source of sensory information for TWR, and their relative homogeneity makes them ideal for studies of aging in the nervous system. Many Aplysia genes have homologs in higher vertebrates, making it a relevant model for age-related neuronal changes that may be important in aging vertebrate brains (Moroz, et al. 2006). Furthermore, Aplysia contain homologs for many genes that are relevant to neurological diseases associated with age in humans (Moroz and Kohn 2010). We found that >85% of the DE transcripts aligned to human transcripts in BLAST searches (E-value£1E-3). This allowed for mapping most DE genes to the much bettercharacterized human gene ontologies for pathway enrichment, and allowed for 101 predictions of pathways that may be important in aging vertebrates. Another similarity we found with the vertebrate brain that became an advantage was the number of expressed transcripts. In mice, ~11,000-13,000 genes are expressed per brain region (Ramsköld, et al. 2009). We found that a similar ~10,000 transcripts in PVC neurons expressed at least 5 copy per biological replicate. Previous transcriptome studies on aging in Aplysia and other model organisms have found genes to be both up- and downregulated with age (Lee, et al. 2000; Moroz and Kohn 2010; Kadakkuzha, et al. 2013). Thus, it was unsurprising that aging of PVC neurons exhibited bidirectional changes in gene expression, with ~55% of DE genes displaying increased expression in Aged II. This is approximately the same percentage of genes up-regulated in the R15 bursting neuron of the Aplysia abdominal ganglion (Kadakkuzha, et al. 2013). In AII Aplysia, reduced performance of aged PVC sensory neurons has been proposed to significantly contribute to reduced TWR behaviors. PVC neurons exhibited reduced excitability to both tail taps and intracellular current injection in AII animals compared to M animals (Kempsell and Fieber 2014). We examined GO categories to functionally profile genes that were DE in AII PVC sensory neurons, and found that reduced excitability coincided with DE of excitation-related genes and gene ontology categories that underlie the behavior, as follows: The most significantly enriched molecular function GO category was ion channel associated proteins, with 87 out of 619 genes in this category exhibiting altered expression in AII PVC neurons. This result suggested that the aging process in sensory neurons has profound impacts on regulation of neuronal physiology. Furthermore, despite 102 overall bidirectionality of all gene expression changes in AII PVC neurons, 83% of DE ion channel associated genes were down-regulated, indicating that ion channel function is reduced in AII. Ten of the ion channel associated genes encode for voltage-gated potassium channel proteins, all of which were downregulated in AII PVC neurons. Reduced expression of voltage-gated K channels Shaw and Shaker were also examined using qPCR. Shaw showed significantly reduced expression in qPCR, while Shaker did not. During an action potential, voltage-gated Na+- and Ca2+ channels in the plasma membrane open to depolarize excitable cells such as neurons. Depolarization opens a broad assortment of voltage-gated K+ channels with a delay, and causes outward movement of K+ ions until the neuron is restored to its resting potential. Ca2+ and K+ channel activation are thus key to the reference level excitability of each neuron. While ion channel abundances were not measured directly here, mRNA abundance is often employed as a proxy for amount of protein, and thus receptor number (Vogel and Marcotte 2012). Therefore, our results that voltage-gated K+ and Ca2+ channel mRNAs were decreased may mean fewer of these channels in the membrane. A reduction in both could produce the reduced excitability demonstrated in AII PVC neurons (Kempsell and Fieber 2015b). Fieber et al (2010) did not observe reduced Ca2+ current density in aged PVC neurons studied with whole cell voltage clamp. Thus, reduced voltage-gated Ca2+ channel expression found here may only be a minor contributor to reduced excitability that was undetectable by whole cell voltage clamp. This may suggest that reduced K+ channel function is the major contributor to reduced excitability in AII PVC neurons. 103 Reduced excitability during aging has also been described in several other neurons of the Aplysia nervous system. In sensory neurons of the buccal ganglion, DAsp- and L-Glu-induced currents declined significantly with age (Fieber, et al. 2010; Kempsell and Fieber 2014). In L7 motoneurons, spiking was depressed and postsynaptic potential amplitude was decreased in response to siphon stimulation in aged animals (Rattan and Peretz 1981). The R15 cholinergic neuron in the abdominal ganglion showed reduced responsiveness to acetylcholine in aged animals (Akhmedov, et al. 2013). Together, these results suggest that reduced excitatory responses are prevalent in the aging Aplysia nervous system in both sensory and motoneurons. Reduced neuronal excitability also occurs in the aging mammalian hippocampus and is correlated with memory declines (Barnes 1990; Barnes, et al. 1992; Foster and Norris 1997; Jouvenceau, et al. 1998; Kumar and Foster 2007). Declines in neuronal excitability appear to be widespread, and are a hallmark of aging in the nervous system of both mammals and Aplysia. In addition to reduced excitability to tail tap and intracellular current injection, AII PVC neurons also exhibited reduced density of L-Glu-induced excitatory currents (Kempsell and Fieber 2014). Three iGluR subunits, GluR2, GluR3, belonging to the AMPA subtype, and KA2, a kainate subunit, showed significantly reduced expression in AII PVC neurons compared to M via RNASeq. There were zero iGluR subunits with significantly increased expression in AII. This suggested that reduced L-Glu current amplitude in AII PVC sensory neurons (Kempsell and Fieber 2014) may be due to decreased expression of iGluR. These findings further support results in Chapter 3 using qPCR, where several iGluR subunits had significantly reduced expression in AII PVC 104 neurons compared to M, and zero subunits were significantly upregulated (see Fig 3.4, in Chapter 3). The iGluR subunits observed to be DE in this study are not the same subunits that were DE in Chapter 3. This could be due to using different cohorts of animals in this study, or evidence that transcriptional changes in iGluR vary during aging. Regardless, both methods confirm reduced expression of iGluR occurs in AII PVC and may underlie reduced L-Glu-induced excitability in AII. In Chapter 3, Grin1-1 was shown to be expressed at significantly greater copies than any other iGluR subunit in PVC sensory neurons, present at a minimum of 2-fold higher copy number than any other iGluR subunits. The STAR aligner was unable to differentiate between the two splice variants of Grin1 in the study described here, thus Grin1 expression encompasses both Grin1-1 and Grin1-2. Grin1 subunits were expressed approximately three times higher than any other subunit. These findings are confirmatory of the expression status of Grin1 found in qPCR as the largest contributor to iGluR mRNA expression in PVC neurons. Further, these independent confirmations of Grin1 as overwhelmingly dominant strongly suggest that L-Glu-induced excitatory currents in PVC neurons are mediated by Grin1. Grin1, however, was not DE in this study, although Grin1-1 and Grin 1-2 were significantly DE in aging in the qPCR results of Chapter 3. It is possible that maintenance of Grin1 is necessary for neuronal function, and aged PVC neurons are attempting to compensate for loss of iGluR function by maintaining Grin1 expression levels. We did not test the L-Glu physiology of AII animals in qPCR and RNASeq cohorts, and instead used behavioral proxies. Declines in physiology with age occur at different rates in different cohorts of Aplysia (Fieber, et al. 2010; Kempsell and Fieber 2014). The cohorts studied in chapter 3 with qPCR may have been at a more 105 advanced stage of aging that was unable to compensate with increased Grin1 expression. Detecting this later stage of aging may not have been possible with behavioral proxies. A consequence of aging is increased oxidative damage from ROS. Accumulation of ROS can be detrimental to cell survival by inducing DNA mutations and can disrupt many cellular processes by altering the structure and function of many macromolecules (Stadtman 2001). Downregulation of neuronal genes in humans is associated with DNA damage that can repress transcription (Loerch, et al. 2008). Additionally, oxidation of ion channels is wide-spread during aging, and can modulate functional changes in voltagegated Na+, K+, and Ca2+ channels (Annunziato, et al. 2002; Patel and Sesti 2016). In response to oxidative stress, the transcription factor Nrf2 is translocated from the cytoplasm to the nucleus to bind DNA promoters and increase transcription of stress response genes (Nguyen, et al. 2009). Consistent with the idea of increased stress response during aging, GO for regulation of response to stress (GO:0080134) was significantly enriched in AII PVC neurons (Supplementary File 1). Increased expression of stress response genes during aging has previously been described in humans, monkey, rats, mice, fruit flies, and roundworms (Bishop, et al. 2010). In Aplysia, stress response genes are also upregulated in several neurons in the abdominal ganglion (Moroz and Kohn 2010; Kadakkuzha, et al. 2013), and therefore stress response is an evolutionarily conserved feature of aging in many parts of the Aplysia CNS. Heat shock proteins were upregulated in response to stress in vertebrates, and work to refold or degrade severely damaged proteins caused by ROS (Murshid, et al. 2013; Rodgers, et al. 2013; Leak 2014). Heat shock protein 70, in particular, protected 106 cells from cell death caused by increased cellular damage in mice (Calderwood, et al. 2009). Four heat shock proteins were upregulated in AII PVC neurons, including heat shock protein 70, and zero heat shock proteins had decreased expression. Other genes with increased expression in AII PVC neurons that have been implicated in chemoprotection and cell survival include major vault protein and multi-drug resistance protein. Major vault protein has been suggested to play a role in cell signaling and prevention of stress-induced apoptosis. Increased expression of major vault protein also occurred in aged humans (Ryu, et al. 2008), and its protein sequence is highly conserved between Aplysia and humans (Moroz and Kohn 2010). Multi-drug resistance protein has an important role in response to stressors such as ROS (Eldakak, et al. 2010), and was also upregulated in aged PVC neurons. Taken together, increased expression of heat shock protein and other chemoprotective genes in AII PVC neurons may be a compensatory mechanism in an effort to maintain proper homeostasis by preventing ROS and other oxidative damage accumulated during aging. Aplysia has homologs for many genes implicated in neurological diseases and reduced neuronal function in aged humans (Moroz, et al. 2006). Two of these genes, amyloid b protein and huntingtin, were tested with qPCR for DE in aged PVC neurons. Amyloid b is formed by proteolysis of amyloid precursor protein and is the major component of amyloid plaques. Increased deposition of amyloid plaques in aged neurons is associated with neuronal impairments related to Alzheimer’s disease (Schenk, et al. 1999). We did not find amyloid b, a derivative of amyloid precursor, to be DE in PVC neurons with either qPCR or RNASeq. We did, however, find it to be one of the most highly expressed transcripts in our study. In the abdominal ganglion of Aplysia, amyloid 107 precursor protein was highly expressed in many motoneurons, but DE in only some of these neurons (Moroz and Kohn 2010). Thus, high expression of amyloid proteins appears to be prevalent in the Aplysia CNS, but DE is apparently neuron specific, and was not observed in AII PVC neurons. Huntingtin is a protein critical for induction of learning in Aplysia (Choi, et al. 2014), and expression of mutant huntingtin induced neurological deficits in Huntington’s disease mouse models (Nguyen, et al. 2005; Bradford, et al. 2009). We found no change in huntingtin expression with either qPCR or RNASeq, indicating that altered huntingtin expression may not be involved in age-related neuronal deficits of PVC neurons. Together, these results suggest that aging and impaired physiology of PVC neurons in Aplysia may be due to normal aging and aging in this model may not predict roles for genes involved in neurological diseases. Induction of both short and long-term memory is compromised in aged Aplysia, including in TWR (Bailey, et al. 1983; Kempsell and Fieber 2015b). Adenylate cyclase and the catalytic subunit of PKA, both necessary for the induction of sensitization in sensory neurons, were downregulated in AII PVC neurons. During facilitation, sensitization induces release of 5-HT, which binds to G-protein coupled receptors on the sensory neurons to activate intracellular adenylate cyclase (Kandel 2001). Activated adenylate cyclase converts ATP to cyclic AMP (cAMP), and thus increased adenylate cyclase enhances intracellular cAMP concentrations. cAMP removes the two regulatory subunits from cAMP-dependent protein kinase A (PKA) to release the catalytic subunit of PKA. The PKA catalytic subunit closes K+ channels via phosphorylation of the channel. This, in turn, increases action potential duration, thereby increasing the amount 108 of free Ca2+intracellularly. Increased intracellular Ca2+ results in an increase in neurotransmitter release at the sensory to motoneuron synapse and increased synaptic strength characteristic of memory formation. Direct injection of PKA into AII PVC neurons induced facilitation, suggesting that processes upstream of PKA activation may be compromised during aging (Kempsell and Fieber 2015a). Reduced transcription of adenylate cyclase found in this study may result in less adenylate cyclase proteins available to be activated by 5-HT in aged neurons. This would in turn have downstream effects on cAMP activation. In addition to potentially reduced cAMP, reduced availability of the catalytic subunit of PKA could result in less PKA available for activation by cAMP. With fewer active PKA subunits phosphorylation of K+ channels would be decreased. Thus, increased action potential duration and increased excitability necessary for Ca2+ influx and enhanced neurotransmitter release may be affected. These molecular mechanisms may in part explain reduced excitability of PVC and reduced EPSP amplitude via reduced activation of proteins necessary for enhanced neurotransmitter release. Cognitive declines in memory tasks with age have been described in many species (Wallace, et al. 2007; Dumitriu, et al. 2010). Decreased expression during aging of genes involved in synaptic plasticity have been proposed to underlie many of these declines in mammals (Lu, et al. 2004). Reduced expression of genes involved in learning and memory systems appears to be a ubiquitous effect of aging in the nervous system of animals via reduced transcription of components necessary for induction of facilitation. This study examined whole-transcriptome changes that occur during aging in PVC sensory neurons. Gene expression changes during aging in Aplysia are variable 109 depending on the neurons studied, and this is the first study to analyze sensory neuron aging. The results suggested that the aging process in PVC sensory neurons coincides with decreased expression of many plasma membrane proteins important to excitability, such as voltage-gated K+ channels, voltage-gated Ca2+ channels, and iGluR. The summation of altered receptor expression with age may result in reduced PVC neuronal performance observed in AII Aplysia. Although changes in expression vary in the Aplysia nervous system and in different species, upregulation of stress response genes and decreased expression of genes important for synaptic plasticity are evolutionarily conserved changes with age observed in many vertebrates. Aplysia genes have many homologs in vertebrates, and conserved molecular pathways of aging and simple neural circuits together make Aplysia a powerful model for the effects of aging in the nervous system. 110 Table 4.1. Primer sequences for qPCR. Gene CCAAT/enhancer-binding protein zeta-like Primer Sequences for qPCR (5’-3’) F ATCAACTCAGCCGACGTCTC R GGGACTGCTTCTTCGTCTTGA Amplicon length (bp) 114 sodium channel alpha-subunit SCAP1 F ATCGGCATGCAGCTCTACTC R ACCCAATCGTTCCACTCGTC 104 potassium voltage-gated channel protein Shaw-like CREB2 F R F R F R F R F R F R F R 104 Huntingtin voltage-dependent calcium channel subunit alpha-2/delta-3 CREB3B amyloid beta A4 potassium voltage-gated channel protein Shaker-like TGGACATAAAGGCGTGTTGC GACTGTCATCTGTTCCCGCT TTGTTGGGTGGCATGGAACT AGTTCTTCAGCACCGCCAAT TGGACACTCAGACCACCAGT CTCTAATAACGCTGCACGGA ACTTATGAGGGCCTGGGTCT ACGTCCGTTCCTACAACACC TGCCAACGAACTTACCCCTC CCTTTTGCGGCTCTCTTTCG ATCAGAATAGACGACGCGGG CTGCCTCTCTCGTAGTCGTG GCCACTGGGGAAGCTCATAG CCTCTTCCAGTCTTGTGCGA 107 76 122 101 101 120 111 Table 4.2. Read depth and mapping statistics using STAR. Unpaired reads had a read removed during quality control. Sample Mature1 Mature2 Mature3 Mature4 Mature5 Mature6 Aged1 Aged2 Aged3 Aged4 Aged5 Aged6 paired Total Number of reads after Trimming 19,608,062 % reads mapped to genome by STAR 66.76% unpaired 1,037,824 79.24% paired unpaired paired unpaired paired unpaired paired unpaired paired unpaired paired unpaired paired unpaired paired unpaired paired unpaired paired unpaired paired unpaired 14,783,156 956,055 13,576256 918,426 14,511,446 627,032 25,116,528 1,502,596 28,406,435 1,886,646 18,443,688 1,113,916 18,051,002 960,838 17,130,158 1,157,305 15,336,780 710,853 14,527,360 656,870 10,089,056 832,514 70.54% 84.06% 72.74% 81.13% 61.82% 60.07% 72.62% 75.38% 50.64% 77.29% 59.36% 77.29% 63.18% 57.74% 60.77% 75.96% 58.36% 45.04% 60.41% 56.38% 52.91% 57.39% Total # transcripts counted by htSEQ 3,090,792 2,482,607 2,582,177 2,187,782 4,549,233 3,108,397 2,664,852 2,525,682 2,768,254 1,911,932 1,464,547 1,525,506 112 A Time to relax tail (s) 30 * 25 * 20 15 10 5 0 8 B 9 10 11 Age (months) 12 25 Time to right (s) * 20 * 15 10 5 0 8 9 10 11 Age (months) 12 Fig. 4.1. Behavioral assessments of aging in cohorts used for RNASeq. Two cohorts of Aplysia were tested monthly for behavioral declines associated with aging. For each behavior there were no significant differences between cohorts during any month, therefore data were combined from both cohorts at each time point. (A) TWR increased significantly with age (all data presented as mean ± standard error). B) Righting also increased significantly with age. Reflex completion times for age 12 month animals used for RNASeq corresponded to AII based on previously published stages of aging in Aplysia (Kempsell and Fieber 2014). *denotes significant difference compared to other ages, p£ 0.05, one-way ANOVA 113 Mature Mature Aged Aged II PC2 PC2 1010 00 −10 -10 -20 −20 -10 −10 0PC1 0 10 10 20 20 30 30 PC1 Fig. 4.2. Principal component analysis (PCA) of the first two principal components in M and AII PVC neurons. Gene expression profiles of M and AII PVC neurons were plotted against the first two principal components. Clustering of M and AII PVC neurons separately indicates expression profiles were most similar within each age group. One M animal exhibited an expression profile that did not clearly belong to either age group (circled). Colored circles represent 95% confidence interval. 114 Fig. 4.3. Overlap of DE genes identified with STAR and Tophat2. Weighted Venn diagram comparing the number of genes found to be DE in DeSeq2 (p£0.05, FDR corrected) after genome alignment with STAR and Tophat2. High overlap indicates good agreement between the transcripts found to be DE with two aligners. 115 3 log 2 fold change 2 1 0 -1 -2 -3 1e-01 1e+01 1e+03 1e+05 Mean expression Fig. 4.4. Scatterplot of mean expression of genes analyzed by STAR vs. their log-fold change. Points in red indicate significant DE at p£0.05. Negative log-fold changes along the y-axis corresponds to transcripts that showed lower expression in AII animals. There were approximately equal numbers of genes up and down regulated in AII PVC neurons. 116 Expression Aged II Expression Aged II Expression Mature Aged II Mature Expression Aged II Mature Mature glutathione S-transferase 7 XM_013079611.1 potassium voltage-gated channel protein Shaker-like XM_005089856.2 fatty acid-binding protein XM_005097581.2 glutathione S-transferase 7 glutathione S-transferase 7 XM_013079611.1 XM_013079611.1 CCAP-1 XM_005103331.1 potassium voltage-gated channel protein Shaker-like aquaporin-5-like potassium voltage-gated channel protein Shaker-like XM_005089856.2 XM_005111003.2 XM_005089856.2 brain-specific domain protein 3-like XM_005112117.2 homeobox/POU fatty acid-binding protein fatty acid-binding protein XM_005097581.2 XM_005097581.2 uncharacterized LOC101862347 XM_013082188.1 CCAP-1 XM_005103331.1 CCAP-1 XM_005103331.1 SKI family transcriptional corepressor 1 homolog-B-like XM_005096911.2 aquaporin-5-like XM_005111003.2 XM_005111003.2 adenylate cyclase aquaporin-5-like NM_001204659.1 brain-specific 3-like brain-specific domain protein 3-like XM_005112117.2 homeobox/POU domain protein glutamate receptor ionotropic, kainatehomeobox/POU 2-like XM_005112117.2 XM_005097743.2 stimulated by retinoic acid gene 6 protein LOC101862347 homolog uncharacterized LOC101862347 XM_005110741.2 uncharacterized XM_013082188.1 XM_013082188.1 uncharacterized LOC101862112 XM_013082201.1 SKI family transcriptional corepressor 1 homolog-B-like SKI family transcriptional corepressor 1 homolog-B-like XM_005096911.2 XM_005096911.2 glutamate receptor 2-like XM_005097744.2 adenylate cyclase adenylate cyclase NM_001204659.1 NM_001204659.1 homeobox protein abdominal-A homolog XM_005110906.2 glutamate receptor ionotropic, kainate 2-like uncharacterized glutamate receptor ionotropic, kainate 2-like XM_005097743.2 XM_005097743.2 XM_013082232.1 LOC101851731 stimulated by retinoic acid gene 6 protein homolog stimulated XM_013082230.1 XM_005110741.2 G-protein coupled receptor 64-likeby retinoic acid gene 6 protein homolog XM_005110741.2 XM_013082215.1 delta and Notch-likeuncharacterized epidermal growth factor-related receptor uncharacterized LOC101862112 LOC101862112 XM_013082201.1 XM_013082201.1 uncharacterized LOC106011702 XM_013082231.1 glutamate receptor 2-like XM_005097744.2 glutamate receptor 2-like XM_005097744.2 G2/mitotic-specific cyclin-B3-like XM_005100278.2 homeobox protein abdominal-A homolog XM_005110906.2 homeobox protein abdominal-A homolog XM_005110906.2 cAMP-specific XM_013086739.1 3',5'-cyclic phosphodiesterase 4D-like XM_013082232.1 uncharacterized LOC101851731 XM_013082232.1 LOC101851731 XM_013081720.1 protein enabled-likeuncharacterized XM_013083348.1 pro-neuregulin-2, membrane-bound isoformreceptor 64-like XM_013082230.1 XM_013082230.1 G-protein coupled receptor 64-like G-protein coupled XM_005103388.2 dihydroxyacetone acyltransferase-like XM_013082215.1 XM_013082215.1 delta and Notch-like epidermal growth factor-related receptorphosphate delta and Notch-like epidermal growth factor-related receptor XM_013080819.1 hemicentin-1 uncharacterized LOC106011702 uncharacterized LOC106011702 XM_013082231.1 XM_013082231.1 XM_005103777.2 heat shock protein 70 B2-like G2/mitotic-specific cyclin-B3-like G2/mitotic-specific cyclin-B3-like XM_005100278.2 XM_005100278.2 XM_005108566.2 major vault protein XM_005104526.2 cAMP-specific XM_013086739.1 3',5'-cyclic phosphodiesterase 4D-like cAMP-specific 4D-like XM_013086739.1 3',5'-cyclic phosphodiesterase uncharacterized LOC101864002 XM_005099220.2 calcium store-operated entry-associated regulatory factor XM_013081720.1 XM_013081720.1 protein enabled-like protein enabled-like XM_005113313.2 LOC101864506 XM_013083348.1 XM_013083348.1 pro-neuregulin-2, membrane-bound isoform pro-neuregulin-2, membrane-bound isoform uncharacterized XM_005106432.2 uncharacterized LOC101846602 XM_005103388.2 XM_005103388.2 dihydroxyacetone phosphate acyltransferase-like dihydroxyacetone phosphate acyltransferase-like XM_013086971.1 multidrug resistance-associated protein 1-like XM_013080819.1 XM_013080819.1 hemicentin-1 XM_013087976.1 hemicentin-1 elongation factor 2-like XM_005100777.2 XM_005103777.2 DNA ligase-like XM_005103777.2 heat shock protein 70 B2-like heat shock protein 70 B2-like XM_005104934.2 cat eye syndrome critical region 5 homolog XM_005108566.2 XM_005108566.2 major vaultprotein protein major vault protein XM_013085278.1 glutamine synthetase-like XM_005104526.2 XM_005104526.2 uncharacterized LOC101864002 uncharacterized LOC101864002 XM_005096327.2 myb-like protein I XM_005099220.2 calcium entry-associated regulatory factor XM_005099220.2 calcium entry-associated regulatory store-operated store-operated factor XM_013089182.1 putative eggshell protein XM_005113313.2 XM_013079144.1 XM_005113313.2 uncharacterized LOC101864506 uncharacterized LOC101864506 glycine N-acyltransferase-like XM_005103166.2 XM_005106432.2 XM_005106432.2 zinc finger protein 704-like uncharacterized LOC101846602 uncharacterized LOC101846602 XM_005097221.2 receptor-like XM_013086971.1 XM_013086971.1 multidrug resistance-associated protein 1-like multidrug resistance-associated protein 1-likeFMRFamide XM_005105439.2 uncharacterized LOC101855585 XM_013087976.1 XM_013087976.1 elongation factor 2-like elongation factor 2-like XM_005090908.2 uncharacterized LOC101861258 XM_005100777.2 XM_005100777.2 DNA ligase-like XM_013083177.1 DNA ligase-like uncharacterized LOC101856027 XM_005105964.2 XM_005104934.2 XM_005104934.2 cat eyedomain-containing syndrome critical region major facilitator superfamily protein 1-like protein 5 homolog cat eye syndrome critical region protein 5 homolog XM_013080424.1 XM_013085278.1 uncharacterized LOC106011289 XM_013085278.1 glutamine synthetase-like glutamine synthetase-like XM_005093424.2 uncharacterized LOC101856247 XM_005096327.2 XM_005096327.2 myb-like protein I myb-like protein I XM_013087517.1 inter-alpha-trypsin inhibitor heavy chain H4-like XM_013089182.1 XM_013089182.1 putative eggshell protein XM_005104831.2 putative eggshell protein uncharacterized LOC101854594 XM_013083237.1 glycine N-acyltransferase-like insulin-like growth factor-binding protein complex acid labile subunit XM_005099123.2 XM_005103166.2 M6 M5 M1 M2 M3 M4 M6 A3 M6 M5 A1 M5 M1 A2 M1 M2 A4 M2 M3 M3 A5 M4 XM_005103166.2 XM_005103166.2 uncharacterized LOC101858410 zinc finger protein 704-like zinc finger protein 704-like zinc finger protein 704-like XM_005097221.2 XM_005097221.2 XM_005097221.2 FMRFamide receptor-like FMRFamide receptor-like FMRFamide receptor-like XM_005105439.2 XM_005105439.2 XM_005105439.2 uncharacterized LOC101855585 uncharacterized LOC101855585 uncharacterized LOC101855585 XM_005090908.2 XM_005090908.2 XM_005090908.2 uncharacterized LOC101861258 uncharacterized LOC101861258 uncharacterized LOC101861258 XM_013083177.1 XM_013083177.1 XM_013083177.1 uncharacterized LOC101856027 uncharacterized LOC101856027 uncharacterized LOC101856027 XM_005105964.2 XM_005105964.2 XM_005105964.2 facilitator superfamily domain-contain major facilitator superfamily domain-containing protein 1-like major facilitator superfamily domain-containing protein 1-like major XM_013080424.1 XM_013080424.1 XM_013080424.1 uncharacterized LOC106011289 uncharacterized LOC106011289 uncharacterized LOC106011289 XM_005093424.2 XM_005093424.2 XM_005093424.2 uncharacterized LOC101856247 uncharacterized LOC101856247 uncharacterized LOC101856247 XM_013087517.1 XM_013087517.1 XM_013087517.1 inter-alpha-trypsin inhibitor heavy chain H4-like inter-alpha-trypsin inhibitor heavy chain H4inter-alpha-trypsin inhibitor heavy chain H4-like XM_005104831.2 XM_005104831.2 XM_005104831.2 uncharacterized LOC101854594 uncharacterized LOC101854594 uncharacterized LOC101854594 XM_013083237.1 XM_013083237.1 XM_013083237.1 insulin-like growth factor-binding protein complex acid labile subunit insulin-like growth factor-binding protein com insulin-like growth factor-binding protein complex acid labile subunit XM_005099123.2 XM_005099123.2 XM_005099123.2 uncharacterized LOC101858410 uncharacterized LOC101858410 uncharacterized LOC101858410 M4 A6 A3 A3 M6 A2 A1 A1 M5 A4 A2 M1 A5 A4 M2 A6 A5 M3 A6 M4 A3 A1 A2 A4 A5 XM_013079144.1 glycine N-acyltransferase-like XM_013079144.1 XM_013079144.1 glycine N-acyltransferase-like glutathione S-transferase 7 XM_013079611.1 potassium voltage-gated channel protein Sh XM_005089856.2 fatty acid-binding protein XM_005097581.2 CCAP-1 XM_005103331.1 aquaporin-5-like XM_005111003.2 brain-specific XM_005112117.2 homeobox/POU domain prote uncharacterized LOC101862347 XM_013082188.1 SKI family transcriptional corepressor 1 hom XM_005096911.2 adenylate cyclase NM_001204659.1 glutamate receptor ionotropic, kainate 2-like XM_005097743.2 stimulated by retinoic acid gene 6 protein ho XM_005110741.2 uncharacterized LOC101862112 XM_013082201.1 glutamate receptor 2-like XM_005097744.2 homeobox protein abdominal-A homolog XM_005110906.2 XM_013082232.1 uncharacterized LOC101851731 XM_013082230.1 G-protein coupled receptor 64-like XM_013082215.1 delta and Notch-like epidermal growth facto uncharacterized LOC106011702 XM_013082231.1 G2/mitotic-specific cyclin-B3-like XM_005100278.2 cAMP-specific XM_013086739.1 3',5'-cyclic phosphodiesteras XM_013081720.1 protein enabled-like XM_013083348.1 pro-neuregulin-2, membrane-bound isoform XM_005103388.2 dihydroxyacetone phosphate acyltransferas XM_013080819.1 hemicentin-1 XM_005103777.2 heat shock protein 70 B2-like XM_005108566.2 major vault protein XM_005104526.2 uncharacterized LOC101864002 XM_005099220.2 calcium entry-associated reg store-operated XM_005113313.2 uncharacterized LOC101864506 XM_005106432.2 uncharacterized LOC101846602 XM_013086971.1 multidrug resistance-associated protein 1-lik XM_013087976.1 elongation factor 2-like XM_005100777.2 DNA ligase-like XM_005104934.2 cat eye syndrome critical region protein 5 ho XM_013085278.1 glutamine synthetase-like XM_005096327.2 myb-like protein I XM_013089182.1 putative eggshell protein Fig 4.5 Heatmap of the top 50 most significantly DE genes from PVC neuron clusters of M and AII animals. Yellow indicates higher expression compared to the average across all replicates, and red indicates reduced expression. The role of underlined genes in the Aplysia nervous system is addressed in results and discussion to illustrate the aging process. This includes upregulation in AII animals of several stress related genes such as heat shock protein and major vault protein, and downregulation of many transcripts related to nervous system function. 117 1.2 Relative expression compared to Mature 1 0.8 * 0.6 * 0.4 * 0.2 NMDA AMPA KA2 GluR7 GluR6 GluR8 GluR5 GluR3 GluR2 GluR1 Grin2 Grin1 0 Kainate/orphans Subunit Fig. 4.6. Expression of iGluR subunits in M and AII PVC neurons based on RNASeq. Relative expression of iGluR transcripts in AII compared to M in PVC sensory neurons showed down-regulation of some iGluR genes in AII. There were no iGluR subunits with increased expression during aging. KA1 and GluR4 subunits did not meet the threshold for number of transcripts to be analyzed. *denotes significantly decreased expression compared to M (p£0.05, Student’s t-test, FDR corrected) 118 Mature 2000 Aged II 1800 100 1400 90 1200 1000 B 80 70 800 60 600 50 400 40 200 30 0 20 Grin1 Grin2 Subunit 10 0 100 % of total AMPAR expression 1600 % of total NMDA expression Calculated number of transcripts A 90 80 70 60 50 40 30 20 10 0 Grin1 Grin2 Subunit * * GluR1 GluR2 GluR3 GluR5 GluR8 Subunit Fig. 4.7. Composition of NMDAR and AMPAR in M and AII PVC neurons. The percent contribution of each subunit to total expression of each subtype was calculated. A) No significant changes in relative expression of NMDAR subunits Grin1 and Grin2 in AII PVC neurons were observed. B) For AMPAR subunits, GluR1 comprised a significantly greater percentage of total AMPAR expression in AII PVC neurons. GluR3 made up a significantly smaller percentage of AMPAR expression in AII PVC neurons. *denotes significantly different percentage of total subtype expression (p£0.05, one-way ANOVA on a logistic-regression with quasibinomial distribution) 119 Molecular Functions Molecular Functions Biological Processes Biological Processes activation of Activation of protein protein kinase A activity kinase A activity Ion channel ion channel associated associated nucleotide Nucleotide metabolic metabolic process process Protein kinase protein kinase bindingbinding nucleoside Nucleosidephosphate phosphate metabolic metabolic processprocess Nucleobase-containing nucleobase−containing small molecule small molecule metabolic metabolic process process Ubiquitin-like ubiquitin−like protein ligase protein ligase binding binding carbohydrate Carbohydrate derivative derivative metabolic process metabolic process Activation of activation of phospholipase C activity phosolipase C activity 1e−01 Adenyl adenyl ribonucleotide ribonucleotide binding binding 1e−04 1e−07 1e−10 1e−13 p−value (FDR corrected) p-value (FDR corrected) 1e−02 1e−05 1e−08 1e−11 p−value (FDR corrected) p-value (FDR corrected) Fig. 4.8. The pathways and activities most affected by aging. The most significantly affected GO terms are listed on the y-axis, with FDR-corrected p-values on the x-axis. The most significantly enriched molecular function is for ion channel associated genes, suggesting that many ion channels were affected by aging. Activation of protein kinase A activity implies that second messenger systems may be altered in aging. 120 2 2 0 0 -1 −1 log 2 fold change 1 1 −2 -2 −3 -3 1e−01 1e-01 1e+01 1e+01 1e+03 1e+03 Mean expression Fig. 4.9. Mean expression vs. log2 fold change for ion channel associated genes. Points in red indicate ion channel genes that were DE between M and AII animals (p£0.05, Student’s t-test, FDR corrected). Negative log-fold changes along the y-axis corresponds to transcripts that showed decreased expression in AII animals. Despite an approximately equal number of genes up and down-regulated in the full dataset, >80% of DE ion channel genes exhibited decreased expression in AII. 121 3 3 1 1 0 0 -1 −1 log 2 fold change 2 2 −2 -2 −3 -3 1e−01 1e-01 1e+01 1e+01 1e+03 1e+03 Mean expression Fig. 4.10. Mean expression vs. log2 fold change for stress response genes. Points in red indicate stress response genes that were DE between M and AII animals (p£0.05, Student’s t-test, FDR corrected). Positive log-fold changes along the y-axis corresponds to transcripts that showed increased expression in AII animals. Many species have shown increased stress response with age, and >70% of DE stress response genes in Aplysia were upregulated in AII PVC neurons. 122 1.5 RNAseq qPCR * log 2 fold change 1 0.5 0 -0.5 * * * * -1 * * -1.5 1 2 3 4 5 6 7 8 * 9 Fig 4.11. Comparison of DE genes in RNASeq and qPCR analyses. Expression status of genes important for neuronal excitability and genes implicated in aging from RNASeq were further tested via qPCR. Log2 fold change in expression in AII animals is displayed on the y-axis for both RNAseq and qPCR results. Negative log-fold change indicates reduced expression in AII animals. K+ Shaw and Ca+ subunit alpha were the only two genes significantly DE in both analyses (qPCR n=4, RNASeq n=6). *denotes significantly different expression compared to M Chapter 5: Conclusions Invertebrate species such as Aplysia californica (Aplysia), Drosophila melanogaster, and Caenorhabditis elegans have been used for decades as models of nervous system function. This dissertation focused on studies of the receptors for LGlutamate (L-Glu). L-Glu is the primary neurotransmitter for fast synaptic transmission in nearly all animals. The binding of L-Glu to ionotropic L-Glu receptors (iGluR) on the postsynaptic membrane opens transmembrane ion channels that allow positively charged ions to flow into the cell, leading to depolarization that can trigger an action potential and transmit information. The simplified nervous systems of invertebrates compared to vertebrates allows for detailed studies of iGluR to further understand processes relevant in vertebrates, such as humans. A better understanding of the relationship between iGluR subunits in Bilateria both clarifies and strengthens the power of animal models of iGluR-mediated neurotransmission. In Chapter 2, the evolutionary relationship of iGluR subunits was examined to identify invertebrate iGluR subunits most similar to the three vertebrate iGluR subtypes: N-methyl-d-aspartate receptors (NMDAR), α-Amino-3-hydroxy-5methyl-4-isoxazolepropionic acid receptors (AMPAR), and kainate receptors. The strongest conclusion of this analysis was that the NMDAR subtype is the most well-conserved across bilaterian species analyzed. Despite large evolutionary distance, both Grin1 and Grin2 NMDAR subunits formed orthologous groups across all species. This suggests that there is high conservation of individual NMDAR subunits. High selective constraint implies that NMDAR function has been conserved throughout 123 124 Bilateria, and thus NMDAR function is necessary for proper nervous system function in all species studied. This is unique to NMDAR; it is the only iGluR subtype that contained more than one gene prior to the divergence of protostomes and deuterostomes. One goal of this research was to predict subunits that form functional iGluR in Aplysia. In a phylogenetic analysis of vertebrate iGluR subunits, NMDAR, AMPAR, and kainate receptor subunits each form a monophyletic clade, with functional iGluR formed as obligatory tetramers composed only of subunits within each clade (Dingledine, et al. 1999). The phylogenetic analysis in Aplysia identified a well-defined clade for the NMDAR subunits that were supported in the full phylogeny. An Aplysia NMDAR-like iGluR must employ Grin1-1 or Grin1-2, and Grin2, if similar to the vertebrate heterotetrameric structure requiring two Grin1 and 2 Grin2 subunits (Paoletti and Neyton 2007). Since expression of the Aplysia Grin1-1 subunit alone did not form functional iGluR in Xenopus laevis oocytes (Ha, et al. 2006), the NMDAR-like iGluR cannot be a homotetramer. This suggests that tetrameric Aplysia NMDAR are similar to vertebrate NMDAR in the requirement of both Grin1 and Grin2 subunits for functional receptors. All 12 iGluR subunits were expressed in all Aplysia nervous system tissues. Prior to this dissertation, the breadth and reach of the role of iGluR, and of L-Glu as a neurotransmitter in Aplysia, was unclear. The ubiquitous expression of numerous iGluR subunits throughout the nervous system attests to the importance of glutamatergic neurotransmission in Aplysia, as in mammals. High conservation of individual NMDAR subunits and preliminary indications that Aplysia NMDAR may require both Grin1 and Grin2 subunits suggests that many 125 aspects of NMDAR physiology are conserved in bilaterians. This contention supports the use of invertebrate species as models for studies of NMDA-mediated physiology. There are other aspects of NMDAR physiology, however, that are likely different between vertebrates and at least, marine, invertebrates. In vertebrates, glycine is a coagonist that binds Grin1 subunits and is required for opening of NMDAR. Seawater contains large amounts of glycine (Hubberten, et al. 1994) that make it an unlikely coagonist in Aplysia NMDAR; 1 mm added glycine potentiated presumable iGluR currents activated by D-Aspartate (D-Asp) only at -30 mV. The finding that hydrophobicity of the Grin1 glycine binding site is not well-conserved in Aplysia further suggests glycine is unlikely to act as co-agonist. Further, vertebrate NMDAR are subject to allosteric modulation by extracellular Mg2+, which blocks channel opening constitutively near the resting potential (Mayer, et al. 1984), where the driving force is low. Mg2+ only weakly blocked L-Glu-induced excitatory currents in Aplysia motoneurons (Dale and Kandel 1993), while Carlson and Fieber (2011) found that Mg2+ attenuated D-Asp-activated currents in buccal sensory neurons equally at both hyperpolarized and depolarized voltages. These studies suggest that any modulatory role of Mg2+ is reduced or that its role is different in Aplysia NMDAR. Although other small molecules or ions, such as H+ and Zn2+, can act as allosteric modulators of vertebrate NMDAR (Paoletti 2011), there is currently no evidence that any of these molecules perform modulatory functions in NMDAR-like-mediated neurophysiology in Aplysia; however, these are understudied areas. In summary, some of the actions of vertebrate NMDAR allosteric modulators may not be relevant in Aplysia. 126 NMDAR in vertebrates are primarily Ca2+ channels, and thus their function is tightly regulated to control intracellular Ca2+ concentration. Carlson and Fieber (2011) found that presumable iGluR activated by D-Asp did not conduct Ca2+. If Aplysia NMDAR are not Ca2+-permeable, it may explain why they are not as tightly controlled through allosteric modulators. It is also possible that other molecules, not yet identified, play the modulatory role in Aplysia NMDAR, with Mg2+- and glycine-dependence, as well as Ca2+ permeability, evolving in vertebrates after divergence from invertebrates. NMDAR is the only iGluR subtype in vertebrates with these unique characteristics. The similarity of NMDAR in Aplysia and vertebrates emphasizes the important, ancestral role for NMDAR that has been maintained through functional constraint throughout Bilateria. While Grin1 and Grin2 were identified as the likely eponymous NMDAR subunits in Aplysia, more research is necessary to verify that these mediate observed NMDAR currents. New genome and transcriptome databases have identified Grin1-2 and Grin2 NMDAR subunits since the Xenopus expression study of Grin1 described above (Ha, et al. 2006). Co-expression of vertebrate Grin1 and Grin2 subunits can form complete functional receptors in Xenopus (Kuner and Schoepfer 1996; Laube, et al. 1998). By injecting various combinations of Aplysia NMDAR subunits, it could be tested if these elicit inward currents with pharmacological similarity to those recorded in Aplysia neurons in previous studies (Carlson and Fieber 2012; Kempsell and Fieber 2014). If Grin1 and Grin2 subunits together form functional receptors, it would further support the high conservation of NMDAR suggested by the full phylogeny. AMPAR subunits also form an orthologous group, with one AMPAR gene in the common bilaterian ancestor. There have been multiple rounds of gene duplication of 127 AMPAR genes in Aplysia, which has likely changed some of the receptor structure. This may explain why AMPA is not an effective agonist in Aplysia neurons (Trudeau and Castellucci 1993; Carlson and Fieber 2012). Block of synaptic transmission by AMPAR antagonists CNQX and DNQX suggests that AMPAR are relevant in Aplysia (Chitwood, et al. 2001). It is possible that the AMPA agonist-specific binding site of Aplysia AMPAR has been altered compared to vertebrates, causing AMPA to no longer act as an agonist. Returning to and expanding the importance of NMDAR, both qPCR and RNASeq showed Grin1-1 was expressed at significantly greater copies than all other iGluR subunits. This suggested that NMDAR, and Grin1-1 in particular, mediate glutamatergic responses in PVC and BSC sensory neurons of Aplysia. Grin1-1 expression was much higher in PVC and BSC sensory neuron clusters in (Chapter 3) than in whole ganglia (Chapter 2). In situ hybridization staining for Grin1 in PVC and BSC sensory neurons was high compared to whole ganglia (Ha, et al. 2006). Moreover, L-Glu–induced excitability was a hallmark of PVC neurons in culture (Kempsell and Fieber 2014). Together the results presented in this dissertation support the literature on NMDAR-like physiology and expression in Aplysia, and they suggest that Grin1-1 mediates the majority of iGluR activity in sensory neurons, and by extension, that NMDAR-like receptor channels form the basis of Aplysia iGluR-mediated neurotransmission in these neurons. In other tissues in which Grin1 expression is not as high, contributions from other non-NMDAR subunits may contribute more to L-Glu-induced neurotransmission. While Grin1 may compose iGluR in Aplysia, the characteristics of the currents are less classically NMDAR, both electrophysiologically and pharmacologically. For 128 example, NMDA-induced currents were rare in BSC sensory neurons, and, when present, were of smaller density compared to those induced by the agonists D-Asp or L-Glu (Carlson and Fieber 2012). Large evolutionary distance between Aplysia and vertebrates makes it unlikely that these receptors would be exactly the same across Bilateria. Binding sites for the agonists L-Glu and NMDA are distinct and, in vivo, L-Glu is the likely agonist at NMDAR. Therefore, selective constraint on the NMDA binding site is low compared to the L-Glu binding site. In Chapter 2, the NMDA binding site was observed to be less similar in hydrophobicity than transmembrane domains (TMDs) that form the ion channel. Therefore, NMDA may not effectively bind Aplysia NMDAR. The expression studies showed no change in subunit composition with age occurred for either NMDAR or AMPAR, suggesting that reduced L-Glu currents in sensory neurons of aged animals is caused by a loss of functional iGluR. Several iGluR subunits had significantly decreased expression in AII animals, and zero subunits had significantly increased expression via qPCR. Grin1-1, the dominant subunit, showed reduced expression in aged PVC and BSC neurons of both cohorts studied. This was an important result. As the subunit that likely mediates most iGluR currents, downregulation and loss of Grin1-1 receptors is the basis of reduced iGluR-mediated excitability in aged sensory neurons, if receptor loss equals decline in receptor physiology. Although their role is not as clear, reduced expression of other iGluR subunits, variable in identity among cohorts, may contribute to decline in receptor physiology in some way. Altered composition of functional iGluR, changing the subunits that make up the receptor, has been implicated as a contributor to reduced iGluR physiology with age in vertebrates 129 (Barnes, et al. 1992; Barnes, et al. 1997; Jouvenceau, et al. 1998), so in this sense Aplysia aging is unique. Yet in other ways, loss of Aplysia Grin1-1 with age echoes the pattern in other aging models. Vertebrate NMDAR may be more vulnerable to aging effects compared to other iGluR subtypes. Many layers of the mouse brain showed greater declines in NMDA binding than AMPA binding in aged animals (Magnusson and Cotman 1993; Magnusson 1995, 1997). Similar results have also been observed in dogs, monkeys, and rats (Tamaru, et al. 1991; Magnusson 2000; Hof, et al. 2002). The whole-transcriptome changes in aging Aplysia sensory neurons suggest the critical loss of iGluR is only part of the aging story. Genes associated with baseline neuronal excitability were significantly affected. The most striking observation was that >80% of the ion channel genes that were differentially expressed had reduced expression, with many coding for voltage-gated K+ and Ca2+ channels. The reduced baseline excitability that would result likely contributes to aging declines in the function of these sensory neurons. Interestingly, gene expression may be a sensitive index of aging changes in excitability, since no changes in Ca2+ currents (Fieber, et al. 2010), or in depolarization amplitude, afterhyperpolarization, or duration of AP (Kempsell and Fieber 2015b) were found in aged Aplysia sensory neurons. Reduced neuronal excitability as a hallmark of aging in the mammalian hippocampus is correlated with memory declines (Barnes 1990; Barnes, et al. 1992; Foster and Norris 1997; Jouvenceau, et al. 1998; Kumar and Foster 2007). Reduced intrinsic excitability in hippocampal pyramidal neurons occurred by alterations in afterhyperpolarization (AHP), which is primarily mediated by voltage-gated K+ channels 130 (Disterhoft and Oh 2007). Thus, K+ channel attrition may be a factor in memory and cognitive declines, and also may be a conserved process that occurs in all animals, including Aplysia. Here is another area of research in which the identified contributions of Aplysia neurons to specific reflex behaviors can be leveraged as a model. A ubiquitous mechanism of aging in vertebrate models is DNA damage caused by excessive oxygen radicals. These reactive oxygen species (ROS) accumulate in tissues of old animals, causing oxidative damage. Oxidation of ion channels is prevalent in aging vertebrate brains and can alter functional properties of voltage-gated Na+, K+, and Ca2+ channels (Annunziato, et al. 2002; Patel and Sesti 2016). The cellular response to mitigate the effects of ROS is increased transcription of stress response genes. Stress response genes can refold or degrade proteins that have been damaged by ROS. In aged PVC sensory neurons there was increased transcription of many stress response genes. Several that were upregulated are conserved between vertebrates and Aplysia, including heat shock proteins, major vault protein, and multi-drug resistance protein. Increased stress response has been correlated with increased ROS production. Thus, increased ROS in aged Aplysia, resulting in increased stress response gene transcription, may be a major component of aging effects in Aplysia, as it is in vertebrate models. Neurodegenerative diseases are a common cause of reduced physiology and cognitive declines in old humans. For example, abnormal protein accumulation that results in reduced neural function has been linked to diseases such as Alzheimer’s, Parkinson’s, and Huntington’s. Altered transcription specific genes, such as amyloid proteins and huntingtin, have been correlated with several of these diseases. There was no 131 change in expression for amyloid proteins or huntingtin in aged PVC neurons. These observations, when coupled to the reduced neurophysiological function observed in Aplysia near the end of its annual lifespan, provide some evidence that Aplysia models the normal aging process, rather than neurological disease. The aging process results in large scale gene expression changes that could be responsible for significant changes in neural function in aged animals. Many of these pathways that occur in vertebrates, such as reduced neural excitability, reduced response to neurotransmitters such as L-Glu, and increases in stress response gene transcription, were found to also occur in Aplysia. The findings of this dissertation suggest that many aspects of aging in the nervous system are conserved in Aplysia and vertebrates. Thus, Aplysia provide a relevant model system for the effects of aging with relevance to the vertebrate nervous system and is promising for the continued use of Aplysia for the neurobiology of aging with relevance to vertebrates. References Akazawa C, Shigemoto R, Bessho Y, Nakanishi S, Mizuno N. 1994. Differential expression of five N-methyl-D-aspartate receptor subunit mRNAs in the cerebellum of developing and adult rats. Journal of Comparative Neurology 347:150-160. Akhmedov K, Rizzo V, Kadakkuzha BM, Carter CJ, Magoski NS, Capo TR, Puthanveettil SV. 2013. Decreased response to acetylcholine during aging of aplysia neuron R15. PloS one 8:e84793. Alberini CM, Ghirardl M, Metz R, Kandel ER. 1994. C/EBP is an immediate-early gene required for the consolidation of long-term facilitation in Aplysia. Cell 76:10991114. Alivisatos AP, Chun M, Church GM, Greenspan RJ, Roukes ML, Yuste R. 2012. The brain activity map project and the challenge of functional connectomics. Neuron 74:970-974. Anders S, Pyl PT, Huber W. 2014. HTSeq—a Python framework to work with highthroughput sequencing data. Bioinformatics:btu638. Annunziato L, Pannaccione A, Cataldi M, Secondo A, Castaldo P, Di Renzo G, Taglialatela M. 2002. Modulation of ion channels by reactive oxygen and nitrogen species: a pathophysiological role in brain aging? Neurobiology of aging 23:819-834. Bacskai BJ, Hochner B, Mahaut-Smith M, Adams SR, Kaang B-K, Kandel ER, Tsien RY. 1993. Spatially resolved dynamics of cAMP and protein kinase A subunits in Aplysia sensory neurons. Science 260:222-222. Bailey CH, Bartsch D, Kandel ER. 1996. Toward a molecular definition of long-term memory storage. Proceedings of the National Academy of Sciences 93:1344513452. Bailey CH, Castellucci VF, Koester J, Chen M. 1983. Behavioral changes in aging Aplysia: a model system for studying the cellular basis of age-impaired learning, memory, and arousal. Behavioral and neural biology 38:70-81. Bailey CH, Chen M. 1988a. Long-term memory in Aplysia modulates the total number of varicosities of single identified sensory neurons. Proceedings of the National Academy of Sciences 85:2373-2377. Bailey CH, Chen M. 1988b. Long-term sensitization in Aplysia increases the number of presynaptic contacts onto the identified gill motor neuron L7. Proceedings of the National Academy of Sciences 85:9356-9359. 132 133 Barnes CA. 1990. Effects of aging on the dynamics of information processing and synaptic weight changes in the mammalian hippocampus. Progress in brain research 86:89-104. Barnes CA, Rao G, Foster T, McNaughton B. 1992. Region-specific age effects on AMPA sensitivity: Electrophysiological evidence for loss of synaptic contacts in hippocampal field CA1. Hippocampus 2:457-468. Barnes CA, Rao G, Shen J. 1997. Age-related decrease in the N-methyl-D-aspartate Rmediated excitatory postsynaptic potential in hippocampal region CA1. Neurobiology of aging 18:445-452. Bartsch D, Ghirardi M, Skehel PA, Karl KA, Herder SP, Chen M, Bailey CH, Kandel ER. 1995. Aplysia CREB2 represses long-term facilitation: relief of repression converts transient facilitation into long-term functional and structural change. Cell 83:979-992. Bishop NA, Lu T, Yankner BA. 2010. Neural mechanisms of ageing and cognitive decline. Nature 464:529-535. Blankenberg D, Gordon A, Von Kuster G, Coraor N, Taylor J, Nekrutenko A. 2010. Manipulation of FASTQ data with Galaxy. Bioinformatics 26:1783-1785. Bordner KA, Kitchen RR, Carlyle B, George ED, Mahajan MC, Mane SM, Taylor JR, Simen AA. 2011. Parallel declines in cognition, motivation, and locomotion in aging mice: association with immune gene upregulation in the medial prefrontal cortex. Experimental gerontology 46:643-659. Boxer PA, Morales FR, Chase MH. 1988. Alterations of group Ia-motoneuron monosynaptic EPSPs in aged cats. Experimental neurology 100:583-595. Bradford J, Shin J-Y, Roberts M, Wang C-E, Li X-J, Li S. 2009. Expression of mutant huntingtin in mouse brain astrocytes causes age-dependent neurological symptoms. Proceedings of the National Academy of Sciences 106:22480-22485. Braunersreuther V, Jaquet V. 2012. Reactive oxygen species in myocardial reperfusion injury: from physiopathology to therapeutic approaches. Current pharmaceutical biotechnology 13:97-114. Brockie PJ, Madsen DM, Zheng Y, Mellem J, Maricq AV. 2001. Differential expression of glutamate receptor subunits in the nervous system of Caenorhabditis elegans and their regulation by the homeodomain protein UNC-42. The Journal of Neuroscience 21:1510-1522. Brunelli M, Castellucci V, Kandel E. 1976. Synaptic facilitation and behavioral sensitization in Aplysia: possible role of serotonin and cyclic AMP. Science 194:1178-1181. 134 Byrne JH, Castellucci VF, Carew TJ, Kandel ER. 1978. Stimulus-response relations and stability of mechanoreceptor and motor neurons mediating defensive gillwithdrawal reflex in Aplysia. Journal of Neurophysiology 41:402-417. Byrne JH, Castellucci VF, Kandel ER. 1978. Contribution of individual mechanoreceptor sensory neurons to defensive gill-withdrawal reflex in Aplysia. Journal of Neurophysiology 41:418-431. Cai D, Chen S, Glanzman DL. 2008. Postsynaptic regulation of long-term facilitation in Aplysia. Current Biology 18:920-925. Calderwood SK, Murshid A, Prince T. 2009. The shock of aging: molecular chaperones and the heat shock response in longevity and aging–a mini-review. Gerontology 55:550-558. Capella-Gutiérrez S, Silla-Martínez JM, Gabaldón T. 2009. trimAl: a tool for automated alignment trimming in large-scale phylogenetic analyses. Bioinformatics 25:19721973. Capo TR, Fieber LA, Stommes DL, Walsh PJ. 2003. Reproductive output in the hatchery-reared California sea hare at different stocking densities. Journal of the American Association for Laboratory Animal Science 42:31-35. Carew TJ, Castellucci VF, Kandel ER. 1971. An analysis of dishabituation and sensitization of the gill-withdrawal reflex in Aplysia. International Journal of Neuroscience 2:79-98. Carew TJ, Walters ET, Kandel ER. 1981. Classical conditioning in a simple withdrawal reflex in Aplysia californica. The Journal of Neuroscience 1:1426-1437. Carlson SL, Fieber LA. 2011. Physiological evidence that D-aspartate activates a current distinct from ionotropic glutamate receptor currents in Aplysia californica neurons. Journal of Neurophysiology 106:1629-1636. Carlson SL, Fieber LA. 2012. Unique ionotropic receptors for D-aspartate are a target for serotonin-induced synaptic plasticity in Aplysia californica. Comparative Biochemistry and Physiology Part C: Toxicology & Pharmacology 155:151-159. Carlson SL, Kempsell AT, Fieber LA. 2012. Pharmacological evidence that D-aspartate activates a current distinct from ionotropic glutamate receptor currents in Aplysia californica. Brain Behav 2:391-401. Castellano JM, Kim J, Stewart FR, Jiang H, DeMattos RB, Patterson BW, Fagan AM, Morris JC, Mawuenyega KG, Cruchaga C. 2011. Human apoE isoforms differentially regulate brain amyloid-β peptide clearance. Science translational medicine 3:89ra57. 135 Castellucci V, Blumenfeld H, Goelet P, Kandel E. 1989. Inhibitor of protein synthesis blocks longterm behavioral sensitization in the isolated gill-withdrawal reflex of Aplysia. Journal of neurobiology 20:1-9. Castellucci VF, Kandel ER, Schwartz JH, Wilson FD, Nairn AC, Greengard P. 1980. Intracellular injection of t he catalytic subunit of cyclic AMP-dependent protein kinase simulates facilitation of transmitter release underlying behavioral sensitization in Aplysia. Proceedings of the National Academy of Sciences 77:7492-7496. Cedar H, Schwartz JH. 1972. Cyclic adenosine monophosphate in the nervous system of Aplysia californica II. Effect of serotonin and dopamine. The Journal of general physiology 60:570-587. Chalfie M, Sulston JE, White JG, Southgate E, Thomson JN, Brenner S. 1985. The neural circuit for touch sensitivity in Caenorhabditis elegans. The Journal of Neuroscience 5:956-964. Chitwood RA, Li Q, Glanzman DL. 2001. Serotonin facilitates AMPA-type responses in isolated siphon motor neurons of Aplysia in culture. The Journal of physiology 534:501-510. Choi Y-B, Kadakkuzha BM, Liu X-A, Akhmedov K, Kandel ER, Puthanveettil SV. 2014. Huntingtin Is Critical Both Pre-and Postsynaptically for Long-Term Learning-Related Synaptic Plasticity in Aplysia. PloS one 9:e103004. Chung SG, van Rey EM, Bai Z, Rogers MW, Roth EJ, Zhang L-Q. 2005. Aging-related neuromuscular changes characterized by tendon reflex system properties. Archives of physical medicine and rehabilitation 86:318-327. Circu ML, Aw TY. 2010. Reactive oxygen species, cellular redox systems, and apoptosis. Free Radical Biology and Medicine 48:749-762. Coffin AB, Mohr RA, Sisneros JA. 2012. Saccular-specific hair cell addition correlates with reproductive state-dependent changes in the auditory saccular sensitivity of a vocal fish. The Journal of Neuroscience 32:1366-1376. Collingridge G, Herron C, Lester R. 1988. Synaptic activation of N-methyl-D-aspartate receptors in the Schaffer collateral-commissural pathway of rat hippocampus. The Journal of physiology 399:283-300. Contractor A, Swanson GT, Sailer A, O'Gorman S, Heinemann SF. 2000. Identification of the kainate receptor subunits underlying modulation of excitatory synaptic transmission in the CA3 region of the hippocampus. The Journal of Neuroscience 20:8269-8278. Cull-Candy SG, Leszkiewicz DN. 2004. Role of distinct NMDA receptor subtypes at central synapses. sci STKE 2004:re16. 136 Dai J, Zhou H-X. 2013. An NMDA receptor gating mechanism developed from MD simulations reveals molecular details underlying subunit-specific contributions. Biophysical journal 104:2170-2181. Dale N, Kandel ER. 1993. L-glutamate may be the fast excitatory transmitter of Aplysia sensory neurons. Proceedings of the National Academy of Sciences 90:71637167. Danysz W, Zajaczkowski W, Parsons C. 1995. Modulation of learning processes by ionotropic glutamate receptor ligands. Behavioural pharmacology 6:455-474. De Loof H, Rosseneu M, Brasseur R, Ruysschaert JM. 1986. Use of hydrophobicity profiles to predict receptor binding domains on apolipoprotein E and the low density lipoprotein apolipoprotein BE receptor. Proceedings of the National Academy of Sciences 83:2295-2299. Dehal P, Boore JL. 2005. Two rounds of whole genome duplication in the ancestral vertebrate. PLoS biology 3:1700. Deupree DL, Bradley J, Turner DA. 1993. Age-related alterations in potentiation in the CA1 region in F344 rats. Neurobiology of aging 14:249-258. Dingledine R, Borges K, Bowie D, Traynelis SF. 1999. The glutamate receptor ion channels. Pharmacological reviews 51:7-62. Disterhoft JF, Oh MM. 2007. Alterations in intrinsic neuronal excitability during normal aging. Aging cell 6:327-336. Dobin A, Davis CA, Schlesinger F, Drenkow J, Zaleski C, Jha S, Batut P, Chaisson M, Gingeras TR. 2013. STAR: ultrafast universal RNA-seq aligner. Bioinformatics 29:15-21. Drake TJ, Jezzini S, Lovell P, Moroz LL, Tan W. 2005. Single cell glutamate analysis in Aplysia sensory neurons. Journal of neuroscience methods 144:73-77. Dumitriu D, Hao J, Hara Y, Kaufmann J, Janssen WG, Lou W, Rapp PR, Morrison JH. 2010. Selective changes in thin spine density and morphology in monkey prefrontal cortex correlate with aging-related cognitive impairment. The Journal of Neuroscience 30:7507-7515. Dunn CW, Hejnol A, Matus DQ, Pang K, Browne WE, Smith SA, Seaver E, Rouse GW, Obst M, Edgecombe GD. 2008. Broad phylogenomic sampling improves resolution of the animal tree of life. Nature 452:745-749. Edgar RC. 2004. MUSCLE: multiple sequence alignment with high accuracy and high throughput. Nucleic Acids Res 32:1792-1797. 137 Eldakak A, Rancati G, Rubinstein B, Paul P, Conaway V, Li R. 2010. Asymmetrically inherited multidrug resistance transporters are recessive determinants in cellular replicative ageing. Nature cell biology 12:799-805. Esdin J, Pearce K, Glanzman DL. 2010. Long-term habituation of the gill-withdrawal reflex in aplysia requires gene transcription, calcineurin and L-type voltage-gated calcium channels. Front in behavioral neuroscience 4:109-117. Fayyazuddin A, Villarroel A, Le Goff A, Lerma J, Neyton J. 2000. Four residues of the extracellular N-terminal domain of the NR2A subunit control high-affinity Zn 2+ binding to NMDA receptors. Neuron 25:683-694. Fieber LA, Carlson SL, Capo TR, Schmale MC. 2010. Changes in D-aspartate ion currents in the Aplysia nervous system with aging. Brain research 1343:28-36. Fieber LA, Carlson SL, Kempsell AT, Greer JB, Schmale MC. 2013. Isolation of sensory neurons of Aplysia californica for patch clamp recordings of glutamatergic currents. JoVE (Journal of Visualized Experiments):e50543. Floyd RA, Hensley K. 2002. Oxidative stress in brain aging: implications for therapeutics of neurodegenerative diseases. Neurobiology of aging 23:795-807. Force A, Lynch M, Pickett FB, Amores A, Yan Y-l, Postlethwait J. 1999. Preservation of duplicate genes by complementary, degenerative mutations. Genetics 151:15311545. Foster TC, Norris CM. 1997. Age-associated changes in Ca2+-dependent processes: Relation to hippocampal synaptic plasticity. Hippocampus 7:602-612. Fox LE, Lloyd PE. 1999. Glutamate is a fast excitatory transmitter at some buccal neuromuscular synapses in Aplysia. Journal of neurophysiology 82:1477-1488. Frazier WT, Kandel ER, Kupfermann I, Waziri R, Coggeshall RE. 1967. Morphological and functional properties of identified neurons in the abdominal ganglion of Aplysia californica. Journal of Neurophysiology 30:1288-1351. Gazzaley A, Siegel S, Kordower J, Mufson E, Morrison J. 1996. Circuit-specific alterations of N-methyl-D-aspartate receptor subunit 1 in the dentate gyrus of aged monkeys. Proceedings of the National Academy of Sciences 93:3121-3125. Gerdes R, Fieber LA. 2006. Life history and aging of captive-reared California sea hares (Aplysia californica). Journal of the American Association for Laboratory Animal Science 45:40-47. Gielen M, Retchless BS, Mony L, Johnson JW, Paoletti P. 2009. Mechanism of differential control of NMDA receptor activity by NR2 subunits. Nature 459:703707. 138 Glanzman DL, Kandel ER, Schacher S. 1990. Target-dependent structural changes accompanying long-term synaptic facilitation in Aplysia neurons. Science 249:799-802. Glanzman DL, Mackey SL, Hawkins RD, Dyke AM, Lloyd PE, Kandel ER. 1989. Depletion of serotonin in the nervous system of Aplysia reduces the behavioral enhancement of gill withdrawal as well as the heterosynaptic facilitation produced by tail shock. The Journal of Neuroscience 9:4200-4213. Goebel DJ, Poosch MS. 1999. NMDA receptor subunit gene expression in the rat brain: a quantitative analysis of endogenous mRNA levels of NR1 com, NR2A, NR2B, NR2C, NR2D and NR3A. Molecular Brain Research 69:164-170. Götz S, García-Gómez JM, Terol J, Williams TD, Nagaraj SH, Nueda MJ, Robles M, Talón M, Dopazo J, Conesa A. 2008. High-throughput functional annotation and data mining with the Blast2GO suite. Nucleic acids research 36:3420-3435. Greger IH, Ziff EB, Penn AC. 2007. Molecular determinants of AMPA receptor subunit assembly. Trends in neurosciences 30:407-416. Gundersen V, Fonnum F, Ottersen OP, Storm-Mathisen J. 2001. Redistribution of Neuroactive Amino Acids in Hippocampus and Striatum During Hypoglycemia: A Quantitative Immunogold Study. Journal of Cerebral Blood Flow & Metabolism 21:41-51. Ha TJ, Kohn AB, Bobkova YV, Moroz LL. 2006. Molecular characterization of NMDAlike receptors in Aplysia and Lymnaea: relevance to memory mechanisms. The Biological Bulletin 210:255-270. Halanych KM. 2004. The new view of animal phylogeny. Annual Review of Ecology, Evolution, and Systematics:229-256. Hansen KB, Yuan H, Traynelis SF. 2007. Structural aspects of AMPA receptor activation, desensitization and deactivation. Current opinion in neurobiology 17:281-288. Hardy J, Selkoe DJ. 2002. The amyloid hypothesis of Alzheimer's disease: progress and problems on the road to therapeutics. Science 297:353-356. Hart AK, Fioravante D, Liu R-Y, Phares GA, Cleary LJ, Byrne JH. 2011. Serotoninmediated synapsin expression is necessary for long-term facilitation of the Aplysia sensorimotor synapse. The Journal of Neuroscience 31:18401-18411. Herculano-Houzel S. 2009. The human brain in numbers: a linearly scaled-up primate brain. Frontiers in human neuroscience 3:31. 139 Hirsch HR, Peretz B. 1984. Survival and aging of a small laboratory population of a marine mollusc, Aplysia californica. Mechanisms of ageing and development 27:43-62. Hof PR, Duan H, Page TL, Einstein M, Wicinski B, He Y, Erwin JM, Morrison JH. 2002. Age-related changes in GluR2 and NMDAR1 glutamate receptor subunit protein immunoreactivity in corticocortically projecting neurons in macaque and patas monkeys. Brain research 928:175-186. Hof PR, Morrison JH. 2004. The aging brain: morphomolecular senescence of cortical circuits. Trends in neurosciences 27:607-613. Hu J-Y, Glickman L, Wu F, Schacher S. 2004. Serotonin regulates the secretion and autocrine action of a neuropeptide to activate MAPK required for long-term facilitation in Aplysia. Neuron 43:373-385. Hubberten U, Lara R, Kattner G. 1994. Amino acid composition of seawater and dissolved humic substances in the Greenland Sea. Marine Chemistry 45:121-128. Hunt DL, Castillo PE. 2012. Synaptic plasticity of NMDA receptors: mechanisms and functional implications. Current opinion in neurobiology 22:496-508. Hunter S, Jones P, Mitchell A, Apweiler R, Attwood TK, Bateman A, Bernard T, Binns D, Bork P, Burge S. 2011. InterPro in 2011: new developments in the family and domain prediction database. Nucleic acids research 40:D306-D312. Jouvenceau A, Dutar P, Billard J. 1998. Alteration of NMDA receptor-mediated synaptic responses in CA1 area of the aged rat hippocampus: Contribution of GABAergic and cholinergic deficits. Hippocampus 8:627-637. Kadakkuzha BM, Akhmedov K, Capo TR, Carvalloza AC, Fallahi M, Puthanveettil SV. 2013. Age-associated bidirectional modulation of gene expression in single identified R15 neuron of Aplysia. BMC genomics 14:880. Kanda K, Hashizume K, Nomoto E, Asaki S. 1986. The effects of aging on physiological properties of fast and slow twitch motor units in the rat gastrocnemius. Neuroscience research 3:242-246. Kandel ER. 2001. The molecular biology of memory storage: a dialogue between genes and synapses. Science 294:1030-1038. Kempsell AT, Fieber LA. 2015a. Age-related deficits in synaptic plasticity rescued by activating PKA or PKC in sensory neurons of Aplysia californica. Frontiers in aging neuroscience 7:173. Kempsell AT, Fieber LA. 2015b. Aging in sensory and motor neurons results in learning failure in Aplysia californica. PloS one 10:e0127056. 140 Kempsell AT, Fieber LA. 2014. Behavioral aging is associated with reduced sensory neuron excitability in Aplysia californica. Frontiers in aging neuroscience 6:84. Kew JN, Kemp JA. 2005. Ionotropic and metabotropic glutamate receptor structure and pharmacology. Psychopharmacology 179:4-29. Kim D, Pertea G, Trapnell C, Pimentel H, Kelley R, Salzberg SL. 2013. TopHat2: accurate alignment of transcriptomes in the presence of insertions, deletions and gene fusions. Genome Biol 14:R36. Kimura S, Kawasaki S, Takashima K, Sasaki K. 2001. Physiological and Pharmacological Characteristics of Quisqualic Acid-Induced K+-Current Response in the Ganglion Cells of Aplysia. The Japanese journal of physiology 51:511-521. Klein M, Kandel ER. 1980. Mechanism of calcium current modulation underlying presynaptic facilitation and behavioral sensitization in Aplysia. Proceedings of the National Academy of Sciences 77:6912-6916. Kohda K, Wang Y, Yuzaki M. 2000. Mutation of a glutamate receptor motif reveals its role in gating and δ2 receptor channel properties. Nature neuroscience 3:315-322. Kosinski R, Zaremba M. 2007. Dynamics of the model of the Caenorhabditis Elegans neural network. Acta Physica Polonica B 38:2201-2210. Kumar A, Foster TC. 2007. 10 Neurophysiology of Old Neurons and Synapses. Brain aging: models, methods, and mechanisms:229-250. Kumar S, Stecher G, Tamura K. 2016. MEGA7: Molecular Evolutionary Genetics Analysis version 7.0 for bigger datasets. Molecular biology and evolution 33:1870-1874. Kuner T, Schoepfer R. 1996. Multiple structural elements determine subunit specificity of Mg2+ block in NMDA receptor channels. The Journal of Neuroscience 16:3549-3558. Landfield PW, Pitler TA, Applegate MD. 1986. The effects of high Mg2+-to-Ca2+ ratios on frequency potentiation in hippocampal slices of young and aged rats. Journal of Neurophysiology 56:797-811. Landis GN, Abdueva D, Skvortsov D, Yang J, Rabin BE, Carrick J, Tavaré S, Tower J. 2004. Similar gene expression patterns characterize aging and oxidative stress in Drosophila melanogaster. Proceedings of the National Academy of Sciences of the United States of America 101:7663-7668. Langmead B, Salzberg SL. 2012. Fast gapped-read alignment with Bowtie 2. Nature methods 9:357-359. 141 Laube B, Kuhse J, Betz H. 1998. Evidence for a tetrameric structure of recombinant NMDA receptors. The Journal of Neuroscience 18:2954-2961. Laurie DJ, Seeburg PH. 1994. Regional and developmental heterogeneity in splicing of the rat brain NMDAR1 mRNA. The Journal of Neuroscience 14:3180-3194. Le SQ, Gascuel O. 2008. An improved general amino acid replacement matrix. Molecular biology and evolution 25:1307-1320. Leak RK. 2014. Heat shock proteins in neurodegenerative disorders and aging. Journal of cell communication and signaling 8:293-310. Lee C-K, Weindruch R, Prolla TA. 2000. Gene-expression profile of the ageing brain in mice. Nature genetics 25:294-297. Lee H-K, Takamiya K, Han J-S, Man H, Kim C-H, Rumbaugh G, Yu S, Ding L, He C, Petralia RS. 2003. Phosphorylation of the AMPA receptor GluR1 subunit is required for synaptic plasticity and retention of spatial memory. Cell 112:631643. Letunic I, Doerks T, Bork P. 2012. SMART 7: recent updates to the protein domain annotation resource. Nucleic acids research 40:D302-D305. Letunic I, Doerks T, Bork P. 2015. SMART: recent updates, new developments and status in 2015. Nucleic Acids Res 43:D257-D260. Levenson J, Endo S, Kategaya LS, Fernandez RI, Brabham DG, Chin J, Byrne JH, Eskin A. 2000. Long-term regulation of neuronal high-affinity glutamate and glutamine uptake in Aplysia. Proceedings of the National Academy of Sciences 97:1285812863. Levenson J, Sherry DM, Dryer L, Chin J, Byrne JH, Eskin A. 2000. Localization of glutamate and glutamate transporters in the sensory neurons of Aplysia. Journal of Comparative Neurology 423:121-131. Li Q, Roberts AC, Glanzman DL. 2005. Synaptic facilitation and behavioral dishabituation in Aplysia: dependence on release of Ca2+ from postsynaptic intracellular stores, postsynaptic exocytosis, and modulation of postsynaptic AMPA receptor efficacy. The Journal of neuroscience 25:5623-5637. Liao Y, Smyth GK, Shi W. 2014. featureCounts: an efficient general purpose program for assigning sequence reads to genomic features. Bioinformatics 30:923-930. Lipsky RH, Goldman D. 2003. Genomics and Variation of Ionotropic Glutamate Receptors. Annals of the New York Academy of Sciences 1003:22-35. Lisman JE, Fellous J-M, Wang X-J. 1998. A role for NMDA-receptor channels in working memory. Nature neuroscience 1:273-275. 142 Liu R-Y, Shah S, Cleary LJ, Byrne JH. 2011. Serotonin-and training-induced dynamic regulation of CREB2 in Aplysia. Learning & memory 18:245-249. Loerch PM, Lu T, Dakin KA, Vann JM, Isaacs A, Geula C, Wang J, Pan Y, Gabuzda DH, Li C. 2008. Evolution of the aging brain transcriptome and synaptic regulation. PloS one 3:e3329. Love MI, Huber W, Anders S. 2014. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol 15:550. Low C-M, Lyuboslavsky P, French A, Le P, Wyatte K, Thiel WH, Marchan EM, Igarashi K, Kashiwagi K, Gernert K. 2003. Molecular determinants of proton-sensitive Nmethyl-D-aspartate receptor gating. Molecular pharmacology 63:1212-1222. Lu T, Pan Y, Kao S-Y, Li C, Kohane I, Chan J, Yankner BA. 2004. Gene regulation and DNA damage in the ageing human brain. Nature 429:883-891. Luebke J, Chang Y-M, Moore T, Rosene D. 2004. Normal aging results in decreased synaptic excitation and increased synaptic inhibition of layer 2/3 pyramidal cells in the monkey prefrontal cortex. Neuroscience 125:277-288. Luine V, Bowling D, Hearns M. 1990. Spatial memory deficits in aged rats: contributions of monoaminergic systems. Brain research 537:271-278. Mackey S, Kandel E, Hawkins R. 1989. Identified serotonergic neurons LCB1 and RCB1 in the cerebral ganglia of Aplysia produce presynaptic facilitation of siphon sensory neurons. The Journal of Neuroscience 9:4227-4235. Magnusson KR. 2000. Declines in mRNA expression of different subunits may account for differential effects of aging on agonist and antagonist binding to the NMDA receptor. The Journal of Neuroscience 20:1666-1674. Magnusson KR. 1995. Differential effects of aging on binding sites of the activated NMDA receptor complex in mice. Mechanisms of ageing and development 84:227-243. Magnusson KR. 1997. Influence of dietary restriction on ionotropic glutamate receptors during aging in C57B1 mice. Mechanisms of ageing and development 95:187202. Magnusson KR, Cotman CW. 1993. Age-related changes in excitatory amino acid receptors in two mouse strains. Neurobiology of aging 14:197-206. Magnusson KR, Nelson SE, Young AB. 2002. Age-related changes in the protein expression of subunits of the NMDA receptor. Molecular Brain Research 99:4045. 143 Malenka RC, Bear MF. 2004. LTP and LTD: an embarrassment of riches. Neuron 44:521. Mansour M, Nagarajan N, Nehring RB, Clements JD, Rosenmund C. 2001. Heteromeric AMPA receptors assemble with a preferred subunit stoichiometry and spatial arrangement. Neuron 32:841-853. Manzon LA. 2002. The role of prolactin in fish osmoregulation: a review. General and comparative endocrinology 125:291-310. Marrus SB, Portman SL, Allen MJ, Moffat KG, DiAntonio A. 2004. Differential localization of glutamate receptor subunits at the Drosophila neuromuscular junction. The Journal of Neuroscience 24:1406-1415. Mayer M, Westbrook G. 1987. Permeation and block of N-methyl-D-aspartic acid receptor channels by divalent cations in mouse cultured central neurones. The Journal of physiology 394:501. Mayer ML, Westbrook GL, Guthrie PB. 1984. Voltage-dependent block by Mg2+ of NMDA responses in spinal cord neurones. Nature 309:261-263. McKibben JR, Bass AH. 1998. Behavioral assessment of acoustic parameters relevant to signal recognition and preference in a vocal fish. The Journal of the Acoustical Society of America 104:3520-3533. Meldrum BS. 2000. Glutamate as a neurotransmitter in the brain: review of physiology and pathology. The Journal of nutrition 130:1007S-1015S. Mitchell A, Chang H-Y, Daugherty L, Fraser M, Hunter S, Lopez R, McAnulla C, McMenamin C, Nuka G, Pesseat S. 2014. The InterPro protein families database: the classification resource after 15 years. Nucleic acids research 43:213-221. Mitchell JJ, Anderson KJ. 1998. Age-related changes in [3 H] MK-801 binding in the Fischer 344 rat brain. Neurobiology of aging 19:259-265. Mohn AR, Gainetdinov RR, Caron MG, Koller BH. 1999. Mice with reduced NMDA receptor expression display behaviors related to schizophrenia. Cell 98:427-436. Moroz LL, Edwards JR, Puthanveettil SV, Kohn AB, Ha T, Heyland A, Knudsen B, Sahni A, Yu F, Liu L. 2006. Neuronal transcriptome of Aplysia: neuronal compartments and circuitry. Cell 127:1453-1467. Moroz LL, Kocot KM, Citarella MR, Dosung S, Norekian TP, Povolotskaya IS, Grigorenko AP, Dailey C, Berezikov E, Buckley KM, et al. 2014. The ctenophore genome and the evolutionary origins of neural systems. Nature 510:109-114. 144 Moroz LL, Kohn AB. 2010. Do different neurons age differently? Direct genome-wide analysis of aging in single identified cholinergic neurons. Frontiers in aging neuroscience 2:1-18. Murshid A, Eguchi T, Calderwood SK. 2013. Stress proteins in aging and life span. International Journal of Hyperthermia 29:442-447. Müßig L, Richlitzki A, Rößler R, Eisenhardt D, Menzel R, Leboulle G. 2010. Acute disruption of the NMDA receptor subunit NR1 in the honeybee brain selectively impairs memory formation. Journal of Neuroscience 30:7817-7825. Naur P, Hansen KB, Kristensen AS, Dravid SM, Pickering DS, Olsen L, Vestergaard B, Egebjerg J, Gajhede M, Traynelis SF. 2007. Ionotropic glutamate-like receptor δ2 binds D-serine and glycine. Proceedings of the National Academy of Sciences 104:14116-14121. Nayernia Z, Jaquet V, Krause K-H. 2014. New insights on NOX enzymes in the central nervous system. Antioxidants & redox signaling 20:2815-2837. Newton IG, Forbes ME, Linville MC, Pang H, Tucker EW, Riddle DR, Brunso-Bechtold JK. 2008. Effects of aging and caloric restriction on dentate gyrus synapses and glutamate receptor subunits. Neurobiology of aging 29:1308-1318. Nguyen T, Hamby A, Massa SM. 2005. Clioquinol down-regulates mutant huntingtin expression in vitro and mitigates pathology in a Huntington's disease mouse model. Proceedings of the National Academy of Sciences of the United States of America 102:11840-11845. Nguyen T, Nioi P, Pickett CB. 2009. The Nrf2-antioxidant response element signaling pathway and its activation by oxidative stress. Journal of Biological Chemistry 284:13291-13295. Nicholson DA, Yoshida R, Berry RW, Gallagher M, Geinisman Y. 2004. Reduction in size of perforated postsynaptic densities in hippocampal axospinous synapses and age-related spatial learning impairments. The Journal of Neuroscience 24:76487653. Nicoll R, Malenka R. 1999. Expression Mechanisms Underlying NMDA ReceptorDependent Long-Term Potentiation. Annals of the New York Academy of Sciences 868:515-525. Nicolle M, Bizon J, Gallagher M. 1996. In vitro autoradiography of ionotropic glutamate receptors in hippocampus and striatum of aged Long–Evans rats: relationship to spatial learning. Neuroscience 74:741-756. Niswender CM, Conn PJ. 2010. Metabotropic glutamate receptors: physiology, pharmacology, and disease. Annual review of pharmacology and toxicology 50:295-322. 145 Ohno S. 2013. Evolution by gene duplication: Springer Science & Business Media. Paoletti P. 2011. Molecular basis of NMDA receptor functional diversity. European Journal of Neuroscience 33:1351-1365. Paoletti P, Bellone C, Zhou Q. 2013. NMDA receptor subunit diversity: impact on receptor properties, synaptic plasticity and disease. Nature Reviews Neuroscience 14:383-400. Paoletti P, Neyton J. 2007. NMDA receptor subunits: function and pharmacology. Current opinion in pharmacology 7:39-47. Paps J, Baguñà J, Riutort M. 2009. Bilaterian phylogeny: a broad sampling of 13 nuclear genes provides a new Lophotrochozoa phylogeny and supports a paraphyletic basal Acoelomorpha. Molecular Biology and Evolution 26:2397-2406. Patel R, Sesti F. 2016. Oxidation of ion channels in the aging nervous system. Brain research 1639:174-185. Peretz B, Ringham G, Wilson R. 1982. Age-diminished motor neuronal function of central neuron L7 in Aplysia. Journal of neurobiology 13:141-151. Peretz B, Romanenko A, Markesbery W. 1984. Functional history of two motor neurons and the morphometry of their neuromuscular junctions in the gill of Aplysia: evidence for differential aging. Proceedings of the National Academy of Sciences 81:4232-4236. Peters A. 2002. The effects of normal aging on myelin and nerve fibers: a review. Journal of neurocytology 31:581-593. Petralia RS, Wang YX, Zhao HM, Wenthold RJ. 1996. Ionotropic and metabotropic glutamate receptors show unique postsynaptic, presynaptic, and glial localizations in the dorsal cochlear nucleus. Journal of Comparative Neurology 372:356-383. Pfaffl MW. 2001. A new mathematical model for relative quantification in real-time RT– PCR. Nucleic acids research 29:e45-e45. Philippe H, Lartillot N, Brinkmann H. 2005. Multigene analyses of bilaterian animals corroborate the monophyly of Ecdysozoa, Lophotrochozoa, and Protostomia. Molecular biology and evolution 22:1246-1253. Pierce A, Fox B, Davis L, Visitacion N, Kitahashi T, Hirano T, Grau E. 2007. Prolactin receptor, growth hormone receptor, and putative somatolactin receptor in Mozambique tilapia: tissue specific expression and differential regulation by salinity and fasting. General and comparative endocrinology 154:31-40. Piggott BJ, Liu J, Feng Z, Wescott SA, Xu XS. 2011. The neural circuits and synaptic mechanisms underlying motor initiation in C. elegans. Cell 147:922-933. 146 Pinsker H, Kupfermann I, Castellucci V, Kandel E. 1970. Habituation and dishabituation of the GM-withdrawal reflex in Aplysia. Science 167:1740-1742. Potier B, Poindessous-Jazat F, Dutar P, Billard J. 2000. NMDA receptor activation in the aged rat hippocampus. Experimental gerontology 35:1185-1199. Putnam NH, Butts T, Ferrier DE, Furlong RF, Hellsten U, Kawashima T, RobinsonRechavi M, Shoguchi E, Terry A, Yu J-K. 2008. The amphioxus genome and the evolution of the chordate karyotype. Nature 453:1064-1071. Rambaut A. 2012. FigTree version 1.4.0. Available at http://tree.bio.ed.ac.uk/software/figtree. Ramsköld D, Wang ET, Burge CB, Sandberg R. 2009. An abundance of ubiquitously expressed genes revealed by tissue transcriptome sequence data. PLoS Comput Biol 5:e1000598. Rattan KS, Peretz B. 1981. Age-dependent behavioral changes and physiological changes in identified neurons in Aplysia californica. Journal of neurobiology 12:469-478. Riedel G, Platt B, Micheau J. 2003. Glutamate receptor function in learning and memory. Behavioural brain research 140:1-47. Roche KW, Tingley WG, Huganir RL. 1994. Glutamate receptor phosphorylation and synaptic plasticity. Current opinion in neurobiology 4:383-388. Rodgers KJ, Ford JL, Brunk UT. 2013. Heat shock proteins: keys to healthy ageing? Redox Report 14:147-153. Rohmann KN, Fergus DJ, Bass AH. 2013. Plasticity in ion channel expression underlies variation in hearing during reproductive cycles. Current Biology 23:678-683. Rumbaugh G, Prybylowski K, Wang JF, Vicini S. 2000. Exon 5 and spermine regulate deactivation of NMDA receptor subtypes. Journal of neurophysiology 83:13001306. Ryan JF, Pang K, Schnitzler CE, Nguyen A-D, Moreland RT, Simmons DK, Koch BJ, Francis WR, Havlak P, Smith SA. 2013. The genome of the ctenophore Mnemiopsis leidyi and its implications for cell type evolution. Science 342:1242592. Ryu S, An H, Oh Y, Choi H, Ha M, Park S. 2008. On the role of major vault protein in the resistance of senescent human diploid fibroblasts to apoptosis. Cell Death & Differentiation 15:1673-1680. Sanderson DJ, Good MA, Seeburg PH, Sprengel R, Rawlins JNP, Bannerman DM. 2008. The role of the GluR-A (GluR1) AMPA receptor subunit in learning and memory. Progress in brain research 169:159-178. 147 Sayers EW, Barrett T, Benson DA, Bolton E, Bryant SH, Canese K, Chetvernin V, Church DM, DiCuccio M, Federhen S. 2011. Database resources of the national center for biotechnology information. Nucleic acids research 39:D38-D51. Schefe JH, Lehmann KE, Buschmann IR, Unger T, Funke-Kaiser H. 2006. Quantitative real-time RT-PCR data analysis: current concepts and the novel “gene expression’s CT difference” formula. Journal of molecular medicine 84:901-910. Schenk D, Barbour R, Dunn W, Gordon G, Grajeda H, Guido T, Hu K, Huang J, Johnson-Wood K, Khan K. 1999. Immunization with amyloid-β attenuates Alzheimer-disease-like pathology in the PDAPP mouse. Nature 400:173-177. Schneggenburger R. 1996. Simultaneous measurement of Ca2+ influx and reversal potentials in recombinant N-methyl-D-aspartate receptor channels. Biophysical journal 70:2165-2174. Schwartz JH, Castellucci VF, Kandel ER. 1971. Functioning of identified neurons and synapses in abdominal ganglion of Aplysia in absence of protein synthesis. Journal of Neurophysiology 34:939-953. Shankar GM, Li S, Mehta TH, Garcia-Munoz A, Shepardson NE, Smith I, Brett FM, Farrell MA, Rowan MJ, Lemere CA. 2008. Amyloid-β protein dimers isolated directly from Alzheimer's brains impair synaptic plasticity and memory. Nature medicine 14:837-842. Shi L, Adams MM, Linville MC, Newton IG, Forbes ME, Long AB, Riddle DR, BrunsoBechtold JK. 2007. Caloric restriction eliminates the aging-related decline in NMDA and AMPA receptor subunits in the rat hippocampus and induces homeostasis. Experimental neurology 206:70-79. Shigenaga MK, Hagen TM, Ames BN. 1994. Oxidative damage and mitochondrial decay in aging. Proceedings of the National Academy of Sciences 91:10771-10778. Simpson JH. 2009. Mapping and manipulating neural circuits in the fly brain. Advances in genetics 65:79-143. Sisneros JA, Bass AH. 2003. Seasonal plasticity of peripheral auditory frequency sensitivity. The Journal of Neuroscience 23:1049-1058. Sobolevsky AI, Rosconi MP, Gouaux E. 2009. X-ray structure, symmetry and mechanism of an AMPA-subtype glutamate receptor. Nature 462:745-756. Song I, Huganir RL. 2002. Regulation of AMPA receptors during synaptic plasticity. Trends in neurosciences 25:578-588. Stadtman ER. 2001. Protein oxidation in aging and age-related diseases. Annals of the New York Academy of Sciences 928:22-38. 148 Stern P, Behe P, Schoepfer R, Colquhoun D. 1992. Single-channel conductances of NMDA receptors expressed from cloned cDNAs: comparison with native receptors. Proceedings of the Royal Society of London B: Biological Sciences 250:271-277. Stommes D, Fieber LA, Beno C, Gerdes R, Capo TR. 2005. Temperature effects on growth, maturation, and lifespan of the california sea hare (Aplysia californica). Journal of the American Association for Laboratory Animal Science 44:31-35. Strimmer K. 2008. fdrtool: a versatile R package for estimating local and tail area-based false discovery rates. Bioinformatics 24:1461-1462. Tamaru M, Yoneda Y, Ogita K, Shimizu J, Nagata Y. 1991. Age-related decreases of the N-methyl-D-aspartate receptor complex in the rat cerebral cortex and hippocampus. Brain research 542:83-90. Team RC. 2014. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. 2013. In. Teng H, Cai W, Zhou L, Zhang J, Liu Q, Wang Y, Dai W, Zhao M, Sun Z. 2010. Evolutionary mode and functional divergence of vertebrate NMDA receptor subunit 2 genes. PLoS One 5:e13342. Tichelaar W, Safferling M, Keinänen K, Stark H, Madden DR. 2004. The threedimensional structure of an ionotropic glutamate receptor reveals a dimer-ofdimers assembly. Journal of molecular biology 344:435-442. Toescu E, Myronova N, Verkhratsky A. 2000. Age-related structural and functional changes of brain mitochondria. Cell calcium 28:329-338. Traynelis SF, Wollmuth LP, McBain CJ, Menniti FS, Vance KM, Ogden KK, Hansen KB, Yuan H, Myers SJ, Dingledine R. 2010a. Glutamate receptor ion channels: structure, regulation, and function. Pharmacological reviews 62:405-496. Traynelis SF, Wollmuth LP, McBain CJ, Menniti FS, Vance KM, Ogden KK, Hansen KB, Yuan H, Myers SJ, Dingledine R. 2010b. Glutamate receptor ion channels: structure, regulation, and function. Pharmacol Rev 62:405-496. Trudeau L, Castellucci VF. 1995. Postsynaptic modifications in long-term facilitation in Aplysia: upregulation of excitatory amino acid receptors. The Journal of Neuroscience 15:1275-1284. Trudeau L-E, Castellucci VF. 1993. Excitatory amino acid neurotransmission at sensorymotor and interneuronal synapses of Aplysia californica. Journal of Neurophysiology 70:1221-1230. Tsai G, Coyle JT. 2002. Glutamatergic mechanisms in schizophrenia. Annual review of pharmacology and toxicology 42:165-179. 149 Uttara B, Singh AV, Zamboni P, Mahajan R. 2009. Oxidative stress and neurodegenerative diseases: a review of upstream and downstream antioxidant therapeutic options. Current neuropharmacology 7:65-74. Villareal G, Li Q, Cai D, Glanzman DL. 2007. The role of rapid, local, postsynaptic protein synthesis in learning-related synaptic facilitation in Aplysia. Current Biology 17:2073-2080. Vogel C, Marcotte EM. 2012. Insights into the regulation of protein abundance from proteomic and transcriptomic analyses. Nature Reviews Genetics 13:227-232. Wallace M, Frankfurt M, Arellanos A, Inagaki T, Luine V. 2007. Impaired recognition memory and decreased prefrontal cortex spine density in aged female rats. Annals of the New York Academy of Sciences 1097:54-57. Walters ET, Byrne J, Carew T, Kandel E. 1983a. Mechanoafferent neurons innervating tail of Aplysia. I. Response properties and synaptic connections. Journal of Neurophysiology 50:1522-1542. Walters ET, Byrne J, Carew T, Kandel E. 1983b. Mechanoafferent neurons innervating tail of Aplysia. II. Modulation by sensitizing stimulation. Journal of Neurophysiology 50:1543-1559. Wang W, Simovic DD, Di M, Fieber L, Rein KS. 2013. Synthesis, receptor binding and activity of iso and azakainoids. Bioorganic & medicinal chemistry letters 23:1949-1952. Wardas J, Pietraszek M, Schulze G, Ossowska K, Wolfarth S. 1996. Age-related changes in glutamate receptors: an autoradiographic analysis. Polish journal of pharmacology 49:401-410. Wenk GL, Barnes CA. 2000. Regional changes in the hippocampal density of AMPA and NMDA receptors across the lifespan of the rat. Brain research 885:1-5. Wollmuth LP, Sobolevsky AI. 2004. Structure and gating of the glutamate receptor ion channel. Trends Neurosci 27:321-328. Xia S, Miyashita T, Fu T-F, Lin W-Y, Wu C-L, Pyzocha L, Lin I-R, Saitoe M, Tully T, Chiang A-S. 2005. NMDA receptors mediate olfactory learning and memory in Drosophila. Current Biology 15:603-615. Yankner BA, Lu T, Loerch P. 2008. The aging brain. Annu. Rev. pathmechdis. Mech. Dis. 3:41-66. Zamanillo D, Sprengel R, Hvalby Ø, Jensen V, Burnashev N, Rozov A, Kaiser KM, Köster HJ, Borchardt T, Worley P. 1999. Importance of AMPA receptors for hippocampal synaptic plasticity but not for spatial learning. Science 284:18051811. 150 Zeddies DG, Fay RR, Alderks PW, Shaub KS, Sisneros JA. 2010. Sound source localization by the plainfin midshipman fish, Porichthys notatus. The Journal of the Acoustical Society of America 127:3104-3113. Zhao S, Fernald RD. 2005. Comprehensive algorithm for quantitative real-time polymerase chain reaction. Journal of computational biology 12:1047-1064. Zhao Y, Hegde AN, Martin KC. 2003. The ubiquitin proteasome system functions as an inhibitory constraint on synaptic strengthening. Current Biology 13:887-898. APPENDIX A CHAPTER 2 SUPPLEMENTARY FIGURES 1 0.278Grin2A_Human Grin2A_Rat Grin2A_Mouse Grin2A_Zebrafish Grin2B_Zebrafish 0.972 1 Grin2B_Rat 0.308 Grin2B_Human Grin2B_Mouse Grin2C_Zebrafish 0.844 Grin2C_Mouse 0.998 1 Grin2C_Rat 0.564 Grin2C_Human 0.934 0.97 1 1 0.966 1 0.886 Grin2D_Zebrafish Grin2D_Mouse 0.27Grin2D_Human 0.948 1 0.324 0.106 0.158 1 1 1 0.928 0.894 0.996 0.528 0.382 0.982 0.36 1 0.948 0.536 0.824 Grin2D_Rat Grin2A_Ciona Grin2_Lancelet Grin2_Aplysia Grin2_Celegans Grin2_Priapulus Grin2_Lingula Grin2_Limulus Grin2_Daphnia Grin2_Drosophila 1 1 Grin1_Tribolium Grin2_Tribolium Grin1_Celegans Grin1_Lingula Grin1_Drosophila Grin1_Daphnia Grin1_Limulus Grin1_Priapulus Grin1_Aplysia Grin1_Octopus Grin1_Zebrafish 0.59 Grin1_Human 0.818Grin1_Mouse Grin1_Rat Grin3_Daphnia Grin3_Limulus Grin3_Octopus 0.908 Grin3_Priapulus 1 1 1 Grin3A_Lancelet Grin3A_Zebrafish Grin3A_Human 0.998 Grin3A_Mouse Grin3A_Rat 0.2 Fig. S2.1. NMDAR subtype only tree. Phylogeny using sequences identified in the full phylogeny as NMDAR. Protostome sequences show monophyletic relationships with chordates for each subunit as seen in the full phylogeny. After alignment trimming there were 467 positions used for analysis. Numbers indicated bootstrap support and the scale bar represent 0.2 substitutions per site. 151 152 GluR3_Aplysia 0.86 GluR4_Aplysia 0.52 GluR_Octopus GluR3_Priapulus 0.95 0.96 GluR4_Priapulus 0.62 GluR1_Limulus 1 GluR1A_Drosophila 1 GluRIB_Drosophila 0.16 GluR5_Aplysia 1 GluR8_Aplysia 0.72 GluR4_Lingula 0.44 GluR1_Aplysia 1 0.18 GluR2_Aplysia GluR2_Lingula 0.12 GluR1_Celegans 1 GluR2_Celegans 0.88 GRIA1_Mouse 1 GRIA1_Rat 1 GRIA1_Human GRIA1_Zebrafish 0.41 GRIA2_Zebrafish 1 1 GRIA2_Human 0.71GRIA2_Mouse GRIA2_Rat 1 1 GRIA3_Human 1 1 0.96 GRIA3_Rat GRIA3_Mouse GRIA3_Zebrafish 1 GRIA4_Zebrafish 1 0.91 GRIA4_Mouse 0.02GRIA4_Human GRIA4_Rat GluR1_Ciona 0.2 Fig. S2.2. AMPAR subtype only tree. Phylogeny of sequences identified as AMPAR genes in the full phylogeny. The tree indicates a single AMPAR gene in the common bilaterian ancestor as seen in the full phylogeny. After alignment trimming there were 665 positions used for analysis. Numbers indicated bootstrap support and the scale bar represent 0.2 substitutions per site. 153 KA2_Aplysia 0.96 GRIK2_Limulus 1 GRIK1_Tribolium 0.94 GRIK2_Lingula GluR7_Aplysia 0.53 GluR2D_Drosophila 0.83 1 GluRIIE_Drosophila 0.84 GRIK3_Limulus 0.72 GRIK1_Daphnia 0.57 GluR1_Tribolium GRIK2_Human 0.42 GRIK2_Rat 1 GRIK2_Mouse 0.99 GRIK2_Zebrafish 0.62 GRIK3_Zebrafish 1 GRIK3_Human 1 GRIK3_Mouse 0.84 GRIK3_Rat 1 GRIK1_Zebrafish 1 GRIK1_Human 1 0.94 GRIK1_Mouse 0.99 GRIK1_Rat GRIK2_Lancelet GRIK4_Mouse 0.92 GRIK4_Rat 0.63 1 1 GRIK4_Human GRIK4_Zebrafish 1 GRIK5_Zebrafish 0.96 GRIK5_Human 1 GRIK5_Mouse 0.89 GRIK5_Rat 0.2 Fig. S2.3. Kainate receptor subtype only tree. Phylogeny of sequences identified as kainate receptor genes in the full tree. There were 687 positions in the alignment after trimming. Scale bar represents 0.2 amino acid substitutions per site. 154 GluR1_Aplysia 0.994 GluR2_Aplysia 0.873 GluR5_Aplysia 0.995 GluR8_Aplysia 0.9 GluR3_Aplysia 0.788 0.99 GluR4_Aplysia GluR7_Aplysia 0.975 KA2_Aplysia GluR6_Aplysia 0.941 KA1_Aplysia Grin2_Aplysia 1 Grin1-1_Aplysia 1 Grin1-2_Aplysia 0.2 Fig. S2.4. Aplysia iGluR protein tree. Subunits within a clade can form functional receptors in vertebrates, and clades formed by both AMPAR and NMDAR-type subunits of Aplysia suggest subunits within these clades may form complete receptors. Kainate receptor subunits do not form a clade. GluR7 and KA2 are the more likely kainate receptor subunit candidates based on their positions full tree. Numbers indicate bootstrap support and the scale bar represents 0.2 substitutions per site. 155 3.5 TMD1 TMD2 TMD3 3 Hydrophobicity score 2.5 2 Aplysia GluR1 Human GRIA1 1.5 1 0.5 0 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 Fig. S2.5. Representative AMPA subunit TMD Hydrophobicity. Aplysia GluR1 is monophyletic with vertebrate AMPAR subunits in the phylogeny, and shows similar hydrophobicity in its TMD1 and TMD3 to Human GRIA1. This suggests similarities in the protein structure of the ion channel of these two subunits. 156 3.5 TMD1 TMD2 TMD3 Hydrophobicity Score 3 2.5 2 Aplysia GluR7 Human GRIK1 1.5 1 0.5 0 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 Fig. S2.6. Representative kainate subunit TMD Hydrophobicity. Aplysia GluR7 is monophyletic with vertebrate kainate receptor subunits in the phylogeny, and shows similar hydrophobicity in its TMDs to Human GRIK1. This suggests similarities in the protein structure of the ion channel of these two subunits. APPENDIX B Chapter 4 Supplementary Data Files Description: The accompanying Excel spreadsheets show the full datasets for complete gene ontology terms (Supplementary File 1) and differential expression results (Supplementary File 2) from RNASeq analysis. Filenames: Chapter_4_Supp_File_1.xlsx Chapter_4_Supp_File_2.xlsx 157