Selective and Localized Marking of Activated Dendritic Spines Using a Novel Genetic Construct, Slayr

The brain's ability to store memory is as remarkable as it is complex. The mechanism behind memory storage is still not fully understood and has become an essential question in the neuroscience field. Long-term memories are thought to form via the strengthening of specific connections between neurons in the brain. At the onset of experience or event, an electrochemical signal causes an explicit circuit of neurons to display highly correlated electrical activity. When that same set of neurons is reactivated in mice, they are able to recall the original experience. However, it has been shown that one neural network can encode more than one memory. This suggests a synapse-specific memory engram model. A memory is allocated and consolidated via miniscule molecular and structural modifications that occur between the synapses of adjoining neurons. These modifications strengthen the connections between synapses and thereby form a plausible mechanism for memory storage. This ever-changing modulation of synaptic connections is known as synaptic plasticity. In order to effectively study the exact modifications that take place at the synaptic level, activated dendritic spines need to be identified. Dendritic spines are the structural correlate of excitatory synapses, and no reliable or reproducible system for this specific type of localized marking currently exists. I helped to develop a Synaptically Localized Activity Reporter (SLAyR) that will allow for the specific marking of active synapses within neurons. Three versions of SLAyR were cloned by combining a variety of different genetic elements that target the reporter to the activated synapse and mark it with a fluorescent protein. The most operative version of SLAyR will be used as a visualization tool to better elucidate the molecular mechanisms of plasticity and long-term memory consolidation.

SELECTIVE AND LOCALIZED MARKING OF ACTIVATED DENDRITIC SPINES USING A NOVEL GENETIC CONSTRUCT, SLAYR by Jenna Novy A Senior Honors Thesis Submitted to the Faculty of The University of Utah In Partial Fulfillment of the Requirements for the Honors Degree in Bachelor of Science In Biomedical Engineering Approved: ______________________________ Jason Shepherd, PhD Thesis Faculty Supervisor _____________________________ David W. Grainger, PhD Chair, Department of Biomedical Engineering _______________________________ Kelly Broadhead, PhD Honors Faculty Advisor _____________________________ Sylvia D. Torti, PhD Dean, Honors College April 2020 Copyright © 2020 All Rights Reserved ABSTRACT The brain’s ability to store memory is as remarkable as it is complex. The mechanism behind memory storage is still not fully understood and has become an essential question in the neuroscience field. Long-term memories are thought to form via the strengthening of specific connections between neurons in the brain. At the onset of experience or event, an electrochemical signal causes an explicit circuit of neurons to display highly correlated electrical activity. When that same set of neurons is reactivated in mice, they are able to recall the original experience. However, it has been shown that one neural network can encode more than one memory. This suggests a synapse-specific memory engram model. A memory is allocated and consolidated via miniscule molecular and structural modifications that occur between the synapses of adjoining neurons. These modifications strengthen the connections between synapses and thereby form a plausible mechanism for memory storage. This ever-changing modulation of synaptic connections is known as synaptic plasticity. In order to effectively study the exact modifications that take place at the synaptic level, activated dendritic spines need to be identified. Dendritic spines are the structural correlate of excitatory synapses, and no reliable or reproducible system for this specific type of localized marking currently exists. I helped to develop a Synaptically Localized Activity Reporter (SLAyR) that will allow for the specific marking of active synapses within neurons. Three versions of SLAyR were cloned by combining a variety of different genetic elements that target the reporter to the activated synapse and mark it with a fluorescent protein. The most operative version of SLAyR will be used as a visualization tool to better elucidate the molecular mechanisms of plasticity and long-term memory consolidation. ii TABLE OF CONTENTS ABSTRACT ii INTRODUCTION 1 BACKGROUND 4 METHODS 99 RESULTS 18 DISCUSSION 22 REFERENCES 28 iii 1 INTRODUCTION Alzheimer’s Disease is a neurodegenerative disease that affects 5.8 million people and is the 6th leading cause of death in the United States alone [12]. Memory is strongly linked to an individual’s identity, and an urgency to develop a therapy for Alzheimer’s is at the heart of considerable neuroscience research [13]. Understanding the neuronal mechanisms of long-term memory and the key molecular drivers of memory consolidation is vital for advancing new therapies. Memory consolidation is known to be driven by the process of synaptic plasticity [14]–[18]. Synaptic plasticity is the strengthening and weakening of synaptic connections along the dendrites of neurons [18]. There are small protrusions along neuronal dendrites known as dendritic spines that undergo these synaptic structural changes [19]–[21]. Dendritic spines are the structural correlate of excitatory synapses [21]. The specific interconnected network of neurons and their dendritic spines activated during a learning experience ultimately form an engram [22]. This explanation of engrammatic memory formation offers a broad overview but belies the hundreds of molecular mechanisms that function to drive synaptic plasticity. One protein known to be vital for memory consolidation to occur is Activityregulated- cytoskeletal protein (Arc) [23]. Arc protein helps mediate synaptic plasticity through the regulation of synaptic strength via receptor recruitment [24],[25]. Its expression is necessary for the consolidation of a novel learning experience to long-term memory [26]. Arc facilitates memory consolidation as a central regulator of AMPA glutamate-type receptor trafficking [25]. It is activity-dependent because Arc is within a subset of genes known as immediate-early-genes (IEGs) [27]. As an IEG, Arc mRNA is 2 rapidly transcribed upon the instigation of a learning experience. Uniquely, Arc mRNA is then shuttled to the activated dendrite where it is localized at the base of dendritic spines [27], [28]. The localization of Arc to the base of dendritic spines is an integral and primary step in the consolidation of an experience to a memory [23]. Although Arc mRNA is known to localize at the base of dendritic spines, further research is necessary to understand the protein’s trafficking patterns. Additionally, knowing which spines are activated during a learning experience would be an asset to memory researchers. Currently, fluorescent labeling techniques only allow for the tagging of activated neurons that form connected circuits in response to a specific learning experience [29]. These labeling techniques do not extend down to the level of the dendritic spine. Activated dendritic spines need to be identified and marked as they are activated in response to a learning experience in order to effectively study the exact modifications that take place at the synaptic level. No reliable or reproducible system to fluorescently tag these specific dendritic spines currently exists [20]. To meet the need for this localized activity-dependent reporter, we developed a genetic construct that would allow for the specific marking of activated dendritic spines of rat hippocampal neurons in vitro. This construct is called Synaptically Localized Activity Reporter (SLAyR). SLAyR was developed for the visualization of activated dendritic spines with fluorescent microscopy. SLAyR was cloned using specific transcriptional and targeting elements of Arc protein, genetic enhancer elements, repressor regions, and a recombinant probe that functions as the visualization element for the construct. Three versions of SLAyR and two control constructs were cloned to determine the most effective 3 combination of genetic elements. The final version of SLAyR must meet four criteria deemed necessary and sufficient for exclusive marking of active synapses: 1) be expressed in response to neuronal activity, 2) exhibit local translation at active synapses, 3) stably localize to synapses, and 4) not interfere with synaptic function. With the in vitroexpressing construct cloned and analyzed, SLAyR is now being developed for use within in vivo rat models. The final in vivo version of SLAyR will be used as a tool to better elucidate the molecular mechanisms of plasticity and subsequent long-term memory consolidation. New discoveries in memory formation are vital for fortifying advancements in Alzheimer’s research and therapeutic development. 4 BACKGROUND 1. Competing Theories of Neuronal Connectivity Understanding the structure of neurons is paramount to discovering the mechanisms that drive memory consolidation. A Spanish scientist named Santiago Ramon y Cajal was one of the most influential contributors to neuronal structure research [1]. In the late 1800s, Cajal framed a hypothesis about the structure of the nervous system known as the Neuronal Doctrine. The Neuronal Doctrine theorized that neurons are not physically connected to one another, but instead are individual structures with synaptic spaces between neighboring cells [1], [2]. The nervous system was not a singular network, but instead a sum of neurons with anatomical and functional distinction [3]. The Neuronal Doctrine is the accepted structural theory of today, and it was Cajal’s diligent and precise documentation and experimentation that led to its acceptance. Interestingly, Cajal was able to support his own theory using a new staining technique developed by his rival, Camillo Golgi. The staining technique was known as the Black Reaction, and Golgi had developed it in an attempt to validate the Reticular Theory. This theory offered a more holistic overview of the nervous system and was the prevalently accepted structural theory for the nervous system at the time. However, it conflicted with Cajal’s Neuronal Doctrine. The Reticular Theory posited that neurons are components of an intricate and interconnected nexus of fibers that serve as the fundamental ‘organ’ of the nervous system. This vast, singular ‘organ’ anatomically and functionally connected all parts of the brain, spinal cord, and peripheral nerves [3], [4]. In order to support the Neuronal Doctrine, Cajal worked meticulously to optimize the Black Reaction and confirm his findings with other staining techniques. He became 5 renowned for his exceedingly thorough and well-documented experimentation techniques. Cajal worked diligently to convince the scientific community that his theory was plausible and eventually the Neuronal Doctrine became an accepted theory in the field. Upon the invention of the electron microscope in the late 1930s, the existence of neuronal synapses was confirmed and the Neuronal Doctrine is still the predominant theory today [5]. In addition to the identification of the individuality of neurons, Cajal’s methodical experimentation and documentation led to the discovery of small structures along dendrites he labeled ‘dendritic spines’. These miniscule structures were previously dismissed by Golgi and other scientists as staining artifacts of the Black Reaction technique. However, Cajal performed a series of experiments and different histological approaches that verified their presence [6]. Although the physiological significance of dendritic spines was not yet identified, this discovery was foundational for future research involving the relation of dendritic spines to synaptic plasticity and long-term memory consolidation. Cajal is often referred to in literature as the “Father of Neuroscience,” and his contributions to the field provided a strong basis for new developments. He was awarded the Nobel Prize in Physiology and Medicine for their work on the structure of the nervous system in 1906 [3], [5]. 2. Pioneering the Engram Complex Around this same time, German scientist Richard Wolfgang Semon was pioneering several theories regarding memory storage and recall. His book, Die Mneme, was published in 1904 [7]. It offered an in-depth explanation of his newest ideas in memory research and outlined the fundamental concept of psycho-physiological parallelism [7] – [9]. This idea stipulated that physiological phenomena are the determining factors for every 6 psychological state of being. The relationship between mechanistic changes in the brain and their psychological manifestations is the foundation of current neuroscience [8]. The concepts presented in Die Mneme are utilized today to generalize the relationship between synaptic plasticity and memory consolidation. Die Mneme led to the coining of the term “engram,” which Semon generally defined as "... the enduring though primarily latent modification in the irritable substance produced by a stimulus..." [7]. An engram is now defined as a series of lasting connections between neurons that result from simultaneous excitations during a learning experience [10], [11] It is often referred to as a physical memory trace [8]. The engram was used by Semon to introduce a foundational law within memory research. The Law of Engraphy is defined by Semon in the following way: "All simultaneous excitations...within our organisms form a connected simultaneous complex of excitations which, as such, acts engraphically, that is to say leaves behind it a connected and, to that extent, unified engram-complex" [7]. The Law of Engraphy thus outlines two critical points. Firstly, each perception that humans experience elicits various neural stimulations that are all simultaneous and interconnected. Secondly, this entire interconnected network of stimulated neurons is what undergoes modification and acts as a whole to form an engram. Semon’s language and explanation of engrams has since been refined and explored to a greater extent and is still important to memory research today [8], [9]. 3. Confirming the Existence of the Engram using Optogenetics In 2014, a pivotal study was published by the Tonegawa Lab at MIT. Two researchers, Steve Ramirez and Xu Liu, provided exciting new evidence that confirmed the 7 presence of engrams. Their paper identified specific neurons involved in the encoding of a single memory in mice and demonstrated that memories can be manipulated or created at the cellular level. The study utilized optogenetics, a technique to operate light-activated proteins within neurons to generate a particular cellular response. There were three parts involved in the experimentation: formation of a fear memory, reactivation of the fear memory, and manipulation of a fear memory [11]. Formation of a fear memory occurs when a mouse explores a new box and then receives a negative stimulus in the form of a foot shock. Upon receiving the shock, the mouse’s neurons fire and encode a fear memory of the box within their hippocampus. When the mouse is then placed back in the box, they react to the fear memory by freezing in place. In order to attempt reactivation of the fear memory, Ramirez and Liu genetically engineered neurons to express channelrhodopsin-2, a light-gated ion channel that can drive electrical excitation of neurons in response to light stimulus. Channelrhodopsin-2 is driven by an activity dependent promoter, which means that only the neurons activated during the specific learning experience would express the channel. With neurons that express channelrhodopsin-2, the mouse again explores a new box and received a foot shock. This resulted in the subsequent encoding of a fear memory. Ramirez and Liu then attempted to reactivate the precise neural network that comprised the fear memory engram. They used thin, fiber optic cables to introduce laser stimulus to the mouse’s hippocampus near the area where the engram was thought to have formed. The laser stimulus reactivated the engram complex, and the mouse froze in characteristic fear behavior. The last sector of the study involved the manipulation of fear memories. A mouse 8 with neurons expressing channelrhodopsin-2 explores a novel box A. It is a safe box, with no negative stimuli, and the mouse encodes the memory as such. Then, the mouse explores a new box, box B, and receives a foot shock. However, the memory of box A is simultaneously reactivated using optogenetic stimulus as the shock occurs. Now, the mouse falsely links the memory of the safe box A with the negative stimulus. This is confirmed when the mouse is placed back in box A and exhibits the fear-freezing behavior [11]. This study shows that the identified engram complex encodes the fear memory and has become an important basis for future memory manipulation and engram- complex research. The research of Cajal, Semon, Ramirez, and Liu all hold importance in the field of neuroscience and are foundational for the development and use of SLAyR. Cajal’s work elucidating the individuality of neurons and the presence of dendritic spines has led to an interest in labeling synapses that are activated during memory formation. This is the role of the SLAyR construct. Once these engrams have been identified, they can potentially be used for new research projects through the optogenetic activation of specific neuronal pathways. This will help scientists understand the precise pathways of learning behaviors and direct research regarding the mechanistic workings of memory consolidation. 9 METHODS 1. Description of Genetic Elements within SLAyR a. Activity-Regulated Cytoskeletal Associated Protein (Arc) The localization of Arc to the base of dendritic spines is an integral and primary step in the consolidation of an experience to a memory [26], [29], [30]. Activated spines are what SLAyR targets for marking, so the targeting region of the Arc gene is of particular interest and use. Previous studies have determined that the 3’ Untranslated Region (UTR) of Arc is the genetic element that is necessary and sufficient for the localization of mRNA to activated dendritic spine bases [30], [31]. Specifically, there is a microdomain within the 3’ UTR of Arc known as the dendritic targeting element (DTE) that is the essential component to active synapse specific localization [32], [33]. When cloning the SLAyR constructs, the DTE was explicitly removed from the entire 3’ UTR because the DTE is the vital component that causes Arc mRNA localization to dendritic spines. By reducing the 3’ UTR by hundreds of nucleotides, the size of the SLAyR construct was significantly reduced. This size reduction will be beneficial for future in vivo experimentation, when SLAyR will be packaged into an Adeno Associated Virus (AAV). We tested a version of SLAyR that contained the whole 3’ UTR as well as two versions that contained only the DTE. b. Activity- Dependent Enhancer Region The Arc DTE targets the SLAyR construct to the base of dendritic spines upon expression of the construct. In order to induce activity-dependent expression of SLAyR, the specific enhancer region within Arc that drives activity-dependent gene expression was added to the construct. This enhancer region is a small critical element named Synaptic 10 Activity Responsive Element (SARE) and is responsible for the induction of the Arc gene [34]. The SARE sequence is tightly driven by neuronal activity and triggers a high level of Arc transcription. SARE is a 104 base pair element that contains binding sites for numerous transcription factors whose enzymatic activity is driven by neuronal stimulation [35]. However, the basal expression levels of Arc induced by SARE are not adequate to effectively express SLAyR in transfected neurons due to transfection inefficiency. Therefore, a synthetically modulated version of SARE, an Enhanced Synaptic ActivityResponsive Element (E-SARE), was inserted into two of the SLAyR constructs instead. ESARE was obtained from a collaboration with the Bito Lab. It is created by replicating, or multiplexing, SARE five times. This modulation allows for reporter expression to be enhanced 30-fold and the dynamic range to be increased 20-fold [36]. E-SARE was cloned into two of the SLAyR construct versions to view its efficacy as an enhancer element within the synaptic reporter. A third version of the SLAyR construct was cloned using a different activityregulated enhancer element in order to determine which version would induce the most effective expression of the genetic construct. This second element is known as the Robust Activity Marking (RAM) system that was developed by the Lin Lab for testing of activity dependent transcription. RAM, like E-SARE, is an enhancer element that is switched on in response to neuronal activity. However, unlike E-SARE, RAM is smaller in size and prevents high labeling background through its lowered sensitivity [37]. c. Recombinant Synaptic Probe All versions of SLAyR utilize a recombinant protein probe for visualization of the active synapses. This probe is comprised of a Fibronectin intrabody generated by mRNA 11 display (FingR) [38]. FingR binds to an endogenous scaffolding protein called PSD-95 that is located at all excitatory synapses in the dendrites [39]. FingRs do not modify expression of the target protein and localize to PSD-95 with a high degree of precision. The recombinant protein probe also codes for Green Fluorescent Protein (GFP) [38]. GFP is a protein commonly used in molecular biology that was isolated from the jellyfish Aequorea victoria [40], [41]. It exhibits green fluorescence when exposed to wavelengths in the blue to ultraviolet range and is used in SLAyR for visualization of synapses under the confocal microscope [42]. When FingR is used in association with the fluorescent protein GFP, it allows for SLAyR to bind to endogenous scaffolding protein PSD-95 and mark it for visualization. Additionally, because FingRs bind to PSD-95 in coordination with expression levels of the protein, the intensity of GFP can be quantified to determine activity levels of dendritic spines [38]. This recombinant synaptic probe is an important component of the SLAyR construct and vital for creating a reproducible system for localized marking of active synapses. A basic diagram of SLAyR and the possible interchangeable components are depicted below (Fig. 1). 12 Fig. 1. Genetic Elements of SLAyR. Depiction of the basic genetic components of the SLAyR construct and the elements that can be swapped in order to create various versions of SLAyR. 2. Cloning of Constructs and Controls Each version of SLAyR was cloned using traditional restriction cloning or Gibson cloning techniques [43], [44]. There are two primary genetic components of interest throughout the cloning process: the plasmid backbone and insert. The backbone contains an antibiotic resistance gene to allow for the growth of bacteria that have our new plasmid of interest later in the process. The inserts have the desired genetic components that will make up SLAyR. The basic steps to the cloning process are as follows, and each step often requires optimization and subsequent troubleshooting depending upon the characteristics of the template DNA (Fig. 2). 13 Fig. 2. Traditional Restriction Cloning versus Gibson Cloning [45]. The figure depicts some of the key differences between the two cloning techniques and the basic flow of experimental steps. First, A Plasmid Editor (ApE) software was utilized to view the sequenced DNA of the SLAyR backbone to identify the insertion location of the enhancer and recombinant probe elements. Restriction enzyme sites surrounding the elements were identified using ApE as well. Plasmid vectors that contain the RAM, E-SARE, and Recombinant Probe elements to be inserted into SLAyR backbone were identified and sequenced. ApE was then used to create primers to amplify the particular elements of interest within SLAyR. For traditional cloning, appropriate restriction enzyme sites were added to the ends of the primers and for Gibson cloning, overhangs were generated using the forward and reverse primers. The primers for each construct’s inserts were introduced to the DNA and run through the Polymerase Chain Reaction (PCR) in order to amplify the region of interest as determined by the added primers. The PCR products were then digested following New 14 England Biolab® DPN1 protocol in order to remove unwanted methylated DNA template and run through agarose gel electrophoresis to separate the PCR amplicons by size. The Azure® c300 Gel Imaging System was used to identify the appropriately sized bands and extract them from the gel. These bands were purified using the Qiagen® Gel Extraction kit and the concentration of eluted insert DNA was determined by measuring absorbance using the NanoDrop® 2000 system. Finally, the insert for each construct was digested with the appropriate restriction enzymes that matched those of their respective backbone. Using these same restriction enzymes, each SLAyR backbone was digested and followed by a dephosphorylation reaction using the Quick Calf Intestinal Phosphatase (CIP) from New England BioLabs®. These steps led to the generation of the SLAyR backbone with small overhangs (sticky ends) that could not self-ligate and would match with the sticky ends of their insert elements. These sticky ends allow the backbone and insert of each construct to be ligated together. Ligation of each construct’s backbone and insert was completed using the New England BioLabs® Quick Ligase protocol. This step combined the linear backbone and insert into a single plasmid DNA. Following ligation, each construct was transformed into competent E. coli cells (TOP10 of DH5α) from New England BioLabs® and grown overnight on agar plates with antibiotic selection. Several of the isolated colonies (each representing clonal populations of bacteria) were picked and grown in liquid cell culture media in order to amplify the number of bacteria with the cloned plasmid DNA. The bacteria were lysed and plasmid DNA was isolated using the Qiagen® Midi- Prep Kit. The eluted DNA was sent for sequencing to Genewiz® Sequencing Services to determine if the cloning was successful. This process was completed for each of the three SLAyR versions, 15 SLAyR 1, SLAyR 2, and SLAyR 3. Additionally, two control constructs that express at all dendrites regardless of activity level were obtained from a collaborator, Control 1/2 and Control 3. 3. Transfection into Neurons and Treatment Rat hippocampal neurons were transfected with the five SLAyR genetic constructs (2 controls and 3 constructs). We used lipid-based transfection, or lipofection, to introduce individual groups of neurons with one of the two FingR control constructs and one of the three SLAyR constructs. Next, we treated half of the neurons transfected with the six SLAyR constructs with 1μM tetrodotoxin (TTX) for 24 hours [46]. The other half of the neurons were left at basal expression levels. Following initial analysis, a later experiment involved the treatment of the most optimal version of SLAyR with 10 μM bicuculline for 24 hours [47]. 4. mCherry Expression In addition to the transfection of SLAyR, another protein, mCherry was transfected into neurons. The mCherry protein is a cell-filler that allows for the visualization of dendrites and dendritic spines. It is expressed throughout the entire neuron when exposed to wavelengths of 550-650 nm [48]. mCherry was transfected into all neurons to serve as a control. It was used to identify healthy neurons under the microscope while avoiding preferential selection of cells strongly expressing GFP that would lead to bias. 5. Imaging of Neurons After TTX and bicuculline treatment, the activity-dependent expression of SLAyR in both treated and untreated neurons was imaged. This was done via confocal microscopy on the Olympus® FV1000 Fluoview System. This optical imaging technique allows for the imaging of fluorescent proteins in cells. Two channels were used during imaging, 16 Fluorescein Isothiocyanate (FITC) and Tetramethyl Rhodamine Iso-Thiocyanate (TRITC). The FITC channel was used for the visualization of GFP at a fluorescence excitation wavelength of 490 nm and the TRITC channel was used for visualization of the mCherry fluorescence at a fluorescence excitation wavelength of 530 nm. Ten neurons from each of the five groups were imaged at 60x magnification. The mCherry TRITC channel was used to select neurons for imaging neurons so as not to bias the data in favor of neurons expressing high levels of GFP (Fig 3). 6. Image Analysis The program ImageJ was utilized for analysis of the neurons. Dendritic spine brightness in untreated neurons (which should exhibit basal activity) was compared to those treated with TTX and bicuculline. GFP expression in neurons treated with TTX, bicuculline, and controls were quantified by particle analysis of integrated density in 2x25 micron sections of dendrites from 8-10 neurons per group using ImageJ. Integrated density is the product of the region of interest’s area and the sum of the pixel values divided by the number of pixels. Additionally, the number of GFP expressing spines were counted and calculated as a percentage of the total number of dendritic spines within each the 2x25 micron region of interest. These percentages were averaged into a single value for each construct and control. 7. Statistical Methods All data collected along the 2x25 micron regions of interest were averaged and the standard error of the mean was taken for each of the SLAyR controls and constructs. An unpaired student’s t-test was conducted with assumed unequal variance. These tests were performed with an α-value of 0.05 with a sample size of n=8-12 neurons per group. This statistical method was used to determine whether the change in fluorescence levels changed 17 significantly between the basal and treated neurons for each SLAyR construct version. 18 RESULTS 1. Imaging of Neurons Figure 3 depicts examples of images taken using the confocal microscope. These images were loaded onto ImageJ, and 2x25 µm regions of interest (ROIs) were selected. The bright protrusions along the neuronal shaft are GFP tagged dendritic spines. Through visual inspection, it was seen that the two control constructs exhibit an increased number of dendritic spine fluorescence compared to the three versions of SLAyR. This observation is congruent with the localized activity reporting of the SLAyR construct. Only selectively activated dendritic spines are expected to exhibit fluorescence. In all of the images, additional green fluorescence along the neuron shaft is due to autofluorescence, light scattering, and bleed-through. The same neuron regions at the mCherry-expression wavelength show the ubiquitous expression of mCherry protein along the neuron dendritic spines and shaft. 10 μm Fig. 3. Neuron Visualization of GFP and mCherry Fluorescence. These images are representative of the neurons that were imaged from each construct and control group using FITC and TRITC channels to view GFP and mCherry fluorescence, respectively. mCherry is a cell filling protein used to identify baseline activity [48]. GFP is expressed by SLAyR and is hypothesized to selectively localize to activated spines. Control 1/2 is the control construct that is compared to SLAyR constructs 1 and 2. Control 3 is compared to SLAyR construct 3. The SLAyR transfected neurons exhibit more selective localization of the construct as evidenced by the decrease in number of fluoresced dendritic spines compared with the controls. 19 2. Percentage of GFP Expressing Spines in all SLAyR Constructs and Controls The average percentages of spines expressing GFP for each SLAyR construct and control are displayed in Figure 4. The average percentages of spines expressing GFP in neurons at basal level versus those treated with tetrodotoxin (TTX) are also displayed for each construct and control. In general, the percentage of fluorescently tagged spines was lower in both the basal and TTX groups for the SLAyR constructs as compared to their controls. SLAyR 2 and SLAyR 3, which both contain the 3’ dendritic targeting element (DTE) were far lower than their controls. This indicates increased selective localization of SLAyR constructs to dendritic spines compared with their controls. Additionally, only SLAyR 3 was observed to have a statistically significant (*p=0.0129) difference between its basal and TTX treated groups. This indicates that SLAyR 3 is sensitive to changes in neuronal activity levels and expresses GFP accordingly. Fig. 4. Average Percentages of Spines expressing GFP along a 2x25 µm ROI. This figure depicts the differences in numbers of spines expressed at basal and TTX treated levels. The number of spines were manually counted and divided by the total number of spines along the ROI for 8-12 neurons per group. SLAyR 3 was found to have a statistically significant difference in the percentages between the basal and treated groups. This was determined with a Student’s t-test with α=0.05 and resulting p-value of *p=0.0129. Error bars indicate standard error of the mean with n=8-12 samples for each SLAyR version and its control. 20 3. Percentage of GFP Expressing Spines in SLAyR 3 and its Control SLAyR 3 was identified as the only construct that showed a significant difference between basal and TTX treated groups, so more experimentation was done for this construct. The experiment was repeated, but neurons transfected with SLAyR 3 and its control were treated with Bicuculline in addition to their TTX and basal expression levels. The results of this experiment are shown in Figure 5. Again, the average percentage of GFP expressing dendritic spines along a 2x25 µm ROI is presented for each treatment of SLAyR 3 and Control 3. All SLAyR 3 averages are less than their respective control averages. This indicates selective localization of the construct compared with its control. Additionally, SLAyR 3 was tested with Bicuculline to assess its sensitivity in response to increased levels of neuronal activity. This was compared with the fluorescent tagging of TTX and basal groups to assess the synaptic reporter’s activity-dependent expression. A statistically significant (*p=0.03) difference between the TTX treated and Bicuculline treated groups was determined using a student’s t-test. However, no statistically significant difference was observed between fluorescence of the treated groups and the basal group for SLAyR 3. 21 Fig. 5. Average Percentage of GFP Expressing Spines along a 2x25 µm ROI for SLAyR 3. sensitivity to changes in activity levels of the neurons. SLAyR 3 displays more selective localization compared to the control. Expression is decreased in neurons treated with TTX and increased in neurons treated with Bicuculine. There is a statistically significant (*p=0.03) difference between the TTX and Bicuculine groups as determined by a Student’s t-test (α=0.05). Error bars indicate standard error of the mean with n= 12 samples for SLAyR 3 and its control. 22 DISCUSSION We aimed to develop in vitro SLAyR to visualize selectively activated dendritic spines via fluorescent confocal microscopy. There are currently no other reliable or reproducible systems for this level of fluorescent tagging. Identifying and marking activated dendritic spines will inform future research about the exact neuronal modifications that take place at the synaptic level in response to a learning experience. As depicted in Figure 1, SLAyR was cloned using various elements of Arc DNA, enhancer elements, repressor regions, and a recombinant probe. SLAyR 3 was found to be most sensitive to changes in neuronal activity levels and engaged in selective localization when compared to its control (Figure 5). With these promising results, it is now being developed and optimized for in vivo testing. Prior to the cloning of SLAyR, research was conducted to assess which genetic elements could be used to cooperatively function as an effective synaptic reporter. The 3’ UTR of Arc protein was identified as an element that could be used to localize of Arc mRNA to activated dendritic spine bases [30], [31]. Additionally, just the 374 base pair dendritic targeting element microdomain within the 3’ UTR was tested for its localization efficacy in SLAyR 2 and SLAyR 3 [32], [33]. The Arc 5’ UTR paired with two different enhancer elements, RAM and ESARE, were tested to assess the activity dependent transcription of the RNA [23],[36], [37]. The SLAyR transcript also encodes a recombinant synaptic probe fused to GFP to allow for visualization of dendritic spines. This probe is a fusion of DNA encoding protein to anchor SLAyR to the base of dendritic spines, FingR, and DNA encoding for GFP to allow for the visualization of SLAyR after translation [38], [40]. FingR and GFP function together so that SLAyR can be imaged via fluorescent 23 confocal microscopy and analyzed. The first test done on the three initial SLAyR versions and their controls was to assess their sensitivity to changes in neuronal activity levels and selective localization. This was done with pharmacological manipulation of neuron activity using the neurotoxin TTX known to decrease activity levels in neurons [46]. For each of the three SLAyR versions and controls, half of the transfected neurons were treated with TTX and the others were left at basal expression levels. Analysis of SLAyR localization to dendritic spines was conducted on 2x25 µm regions of dendrites with a calculation of the average percentage of GFP- expressing spines among 8-12 samples. From the analyzed samples, it was found that only SLAyR 3 exhibited a significant decrease in the percentage of GFP-expressing dendritic spines in the TTX treated neurons versus those at basal activity levels (Fig. 4). The decrease in expression level indicates that SLAyR 3 is activity-dependent and is expressed less when neurons exhibit decreased levels of activity. SLAyR 3 is sensitive to changes in neuronal activity levels and expresses GFP accordingly. In addition to testing for activity-dependent sensitivity, each SLAyR version was also analyzed for its selective localization ability. Each control indiscriminately marks all dendritic spines upon expression. Conversely, the SLAyR versions were designed to selectively mark only dendritic spines that are activated. While Figure 4 depicts the counted number of fluorescent spines for each construct, other experiments are underway to ascertain the specificity of SLAyR’s labeling to specifically active dendritic spines. Most of the SLAyR versions exhibited decreased average GFP-expressing spines compared with their respective controls. This indicates that selective localization is, in fact, occurring. 24 Following this initial experiment, SLAyR 3 was identified as a primary candidate for further analysis. The activity-dependence of the SLAyR 3 construct was again tested through pharmacological manipulation of neuronal activity levels. This time, 1/3 of the SLAyR 3-transfected neurons were treated with the neurotoxin TTX, another 1/3 were treated with the inhibitory blocker bicuculine, and the last 1/3 were left at basal expression levels. The same procedure was carried out on the SLAyR 3 control. The results of this experiment demonstrated a significant decrease in average GFP-expression in dendritic spines for the TTX and bicuculine-treated neurons (Fig. 5). These results indicate that SLAyR 3 exhibits activity-dependent expression for both decreased and increased levels of neuronal activity. Additionally, SLAyR 3 exhibits selective localization to dendritic spines when compared to its control. These findings highlight SLAyR 3 as a primary candidate for further development of SLAyR for in vivo use. The results of the SLAyR experiments corroborate the activity-dependent nature of Arc mRNA [23], [27]. Activity-dependent expression of SLAyR 3 was evidenced by the results of this experiment (Fig. 5). Additionally, the localized targeting of SLAyR to the base of dendritic spines due to the inclusion of the Arc DTE was observed in all of the SLAyR constructs (Fig. 4) [28]. To set SLAyR apart from other synaptic reporters, SLAyR must be expressed in response to neuronal activity, exhibit local translation at active synapses, stably localize to synapses, and not interfere with synaptic function. Currently, other published synaptic reporters that are used to detect learning-induced synaptic plasticity do not effectively meet the four requirements assigned to assess SLAyR. A method known as Synaptic Proximity Ligation Assay (SYNPLA) was developed that allows for the fluorescent tagging of a 25 nanoball of DNA amplified at synaptic clefts. This method ensures specificity and bright positive signal, but the large nanoball of DNA is far larger than the synaptic cleft and is not known whether it stably localizes to synapses or interferes with synaptic function [49]. Labelling of recently potentiated spines has also been done using optogenetics. In vivo imaging was conducted using a synaptic optoprobe called AS-PaRac1 (activated synapse targeting photoactivatable Rac1) to specifically label spines using the lightsensitive Rac1 protein. However, this method invokes the shrinking of AS-PaRac1 containing spines. Although this is useful for some synaptic ensemble formation experiments, it is unwanted when studying synapse development and memory consolidation [50]. The fully developed SLAyR construct will fill the gaps in current active synapse labelling methods. In the initial in vitro studies done with SLAyR, it has been shown to express in response to neuronal activity and exhibit local translation at active synapses. Further development and experimentation will need to be carried out to in in vivo hippocampal neurons to verify the other two requirements. The in vitro nature of these experiments determines the primary limitation of the previous SLAyR experiments. Because SLAyR was transfected within in vitro rat hippocampal neurons, all activity-dependence was determined on pharmacologically manipulated neurons. Activity levels in neurons treated with TTX or bicuculine are more exaggerated than the fine-tuned activity expression within in vivo neurons. Additionally, the stable and unreactive nature of SLAyR cannot be fully determined using only fixed in vitro experimentation. The current SLAyR construct is serviceable for the identification of genetic 26 elements that collectively create a selectively localized, activity-dependent synaptic reporter. This is an integral first step towards optimization of SLAyR for in vivo use. Once an in vivo version of SLAyR is fully developed and analyzed, it will be used to identify real-time dendritic spine activation during the initiation of a novel learning experience. This would be an asset to memory researchers hoping to identify the engrammatic networks of memory formation or the protein trafficking patterns of Arc. There are currently only fluorescent labeling techniques for the tagging of activated neurons that form connected circuits in response to a specific learning experience [29]. The tagging of selective dendritic spines would allow for a much more fine-tuned examination of memory pathways or synaptic labeling. Once SLAyR is developed for use in vivo, it will be used as a tool to better understand developmental brain activity and memory network formation. We have plans to analyze SLAyR expression in rat visual cortex to test for activity-dependent synaptic labeling following dark rearing and monocular deprivation. Additionally, SLAyR expression can be used to test synaptic expression in the hippocampi of mice in homecages versus enriched environments. These future directions would inform research on developmental synaptic pruning and help to link behavioral activities with mechanistic neuronal changes. Further, SLAyR can be paired with stimulation techniques like optogenetics or 2photon excitation to identify neuronal circuits for stimulation [11], [51], [52]. Expression of SLAyR in mouse hippocampi and stimulation of the perforant path can be conducted to test marking of a defined subset of spines and synapses involved in memory consolidation. This level of circuit identification and subsequent stimulation is just in the beginnings of 27 development within the neuroscience field. These new potential developments would not have been possible without the foundation laid by scientists like Semon and Ramon y Cajal. They pioneered scientific theory and methods before neuroscience was even a recognized field. It is exciting to see Semon’s engrammatic theory confirmed 100 years later by scientists like Ramirez and Liu using optogenetic technology. And even in its early stages, SLAyR allows for the visualization of dendritic spines that were once only careful sketches in Ramon y Cajal’s lab notebook. With the development of new tools to identify activated dendritic spine circuitry, scientists become one step closer to elucidating the molecular mechanisms and patterns of synaptic plasticity. Synaptic plasticity is the driving force behind long-term memory consolidation. A better understanding of how synaptic structural changes occur will galvanize progress in the field of memory research. With an improved understanding long-term memory, more expertise can be dedicated to developing new therapies for patients suffering from neurodegenerative disorders such as Alzheimer’s Disease. 28 REFERENCES [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] F. López-Muñoz, J. Boya, and C. Alamo, “Neuron theory, the cornerstone of neuroscience, on the centenary of the Nobel Prize award to Santiago Ramón y Cajal,” Brain Res. Bull., vol. 70, no. 4, pp. 391–405, 2006. M. Glickstein, “Golgi and Cajal: The neuron doctrine and the 100th anniversary of the 1906 Nobel Prize,” Curr. Biol., vol. 16, no. 5, pp. R147–R151, Mar. 2006. G. Cimino, “Reticular theory versus neuron theory in the work of Camillo Golgi.,” Physis. Riv. Int. Stor. Sci., vol. 36, no. 2, pp. 431–472, 1999. E. PANNESE, “The Black Reaction,” Brain Res. Bull., vol. 41, no. 6, pp. 343– 349, 1996. J. A. De Carlos and J. Borrell, “A historical reflection of the contributions of Cajal and Golgi to the foundations of neuroscience,” Brain Res. Rev., vol. 55, no. 1, pp. 8–16, 2007. R. Yuste, “The discovery of dendritic spines by Cajal,” Front. Neuroanat., vol. 9, p. 18, Apr. 2015. R. W. Semon, The Mneme. London: Aleen & Unwin, 1921. D. L. Schacter, J. E. Eich, and E. Tulving, “Richard Semon’s theory of memory,” J. Verbal Learning Verbal Behav., vol. 17, no. 6, pp. 721–743, 1978. S. A. Josselyn, S. Köhler, and P. W. Frankland, “Heroes of the Engram,” J. Neurosci., vol. 37, no. 18, pp. 4647 LP – 4657, May 2017. M. Poo et al., “What is memory? The present state of the engram,” BMC Biol., vol. 14, no. 1, p. 40, 2016. X. Liu et al., “Optogenetic stimulation of a hippocampal engram activates fear memory recall,” Nature, vol. 484, no. 7394, pp. 381–385, Mar. 2012. “Alzheimers and Dementias Facts and Figures,” Alzheimer’s Association, 2019. [Online]. Available: https://www.alz.org/alzheimers-dementia/facts-figures. [Accessed: 08-Oct2019]. An essay concerning human understanding. London: London, 1793. T. V Bliss and T. Lomo, “Long-lasting potentiation of synaptic transmission in the dentate area of the anaesthetized rabbit following stimulation of the perforant path,” J. Physiol., vol. 232, no. 2, pp. 331–356, Jul. 1973. S. B. Hofer, T. D. Mrsic-Flogel, T. Bonhoeffer, and M. Hübener, “Experience leaves a lasting structural trace in cortical circuits,” Nature, vol. 457, no. 7227, pp. 313–317, Jan. 2009. S. J. Martin, P. D. Grimwood, and R. G. M. Morris, “Synaptic Plasticity and Memory: An Evaluation of the Hypothesis,” Annu. Rev. Neurosci., vol. 23, no. 1, pp. 649–711, Mar. 2000. T. Takeuchi, A. J. Duszkiewicz, and R. G. M. Morris, “The synaptic plasticity and memory hypothesis: encoding, storage and persistence,” Philos. Trans. R. Soc. Lond. B. Biol. Sci., vol. 369, no. 1633, p. 20130288, Dec. 2013. T. Rogerson et al., “Synaptic tagging during memory allocation,” Nat. Rev. Neurosci., vol. 15, p. 157, Feb. 2014. E. G. Jones, “Santiago Ramón y Cajal and the Croonian Lecture, March 1894,” Trends Neurosci., vol. 17, no. 5, pp. 190–192, 1994. M. Bosch and Y. Hayashi, “Structural plasticity of dendritic spines,” Curr. Opin. Neurobiol., vol. 22, no. 3, pp. 383–388, 2012. 29 [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32] [33] [34] [35] [36] H. Kasai, M. Fukuda, S. Watanabe, A. Hayashi-Takagi, and J. Noguchi, “Structural dynamics of dendritic spines in memory and cognition,” Trends Neurosci., vol. 33, no. 3, pp. 121–129, 2010. D. S. Roy, A. Arons, T. I. Mitchell, M. Pignatelli, T. J. Ryan, and S. Tonegawa, “Memory retrieval by activating engram cells in mouse models of early Alzheimer’s disease,” Nature, vol. 531, no. 7595, pp. 508–512, Mar. 2016. C. R. Bramham et al., “The Arc of synaptic memory,” Exp. brain Res., vol. 200, no. 2, pp. 125–140, Jan. 2010. K. Minatohara, M. Akiyoshi, and H. Okuno, “Role of Immediate-Early Genes in Synaptic Plasticity and Neuronal Ensembles Underlying the Memory Trace,” Front. Mol. Neurosci., vol. 8, p. 78, Jan. 2016. S. Chowdhury et al., “Arc/Arg3.1 interacts with the endocytic machinery to regulate AMPA receptor trafficking,” Neuron, vol. 52, no. 3, pp. 445–459, Nov. 2006. C. Chia and T. Otto, “Hippocampal Arc (Arg3.1) expression is induced by memory recall and required for memory reconsolidation in trace fear conditioning,” Neurobiol. Learn. Mem., vol. 106, pp. 48–55, 2013. C. R. Bramham, P. F. Worley, M. J. Moore, and J. F. Guzowski, “The immediate early gene arc/arg3.1: regulation, mechanisms, and function,” J. Neurosci., vol. 28, no. 46, pp. 11760–11767, Nov. 2008. J. Dynes and O. Steward, “Arc mRNA Docks Precisely at the Base of Individual Dendritic Spines Indicating the Existence of a Specialized Microdomain for Synapse-Specific mRNA Translation,” J. Comp. Neurol., vol. 520, pp. 3105–3119, Oct. 2012. T. Kawashima, H. Okuno, and H. Bito, “A new era for functional labeling of neurons: activity-dependent promoters have come of age,” Front. Neural Circuits, vol. 8, p. 37, Apr. 2014. O. Steward, C. S. Wallace, G. L. Lyford, and P. F. Worley, “Synaptic Activation Causes the mRNA for the IEG Arc to Localize Selectively near Activated Postsynaptic Sites on Dendrites,” Neuron, vol. 21, no. 4, pp. 741–751, Oct. 1998. S. Farris, G. Lewandowski, C. D. Cox, and O. Steward, “Selective localization of arc mRNA in dendrites involves activity- and translation-dependent mRNA degradation,” J. Neurosci., vol. 34, no. 13, pp. 4481–4493, Mar. 2014. Y. Gao, V. Tatavarty, G. Korza, M. K. Levin, and J. H. Carson, “Multiplexed dendritic targeting of alpha calcium calmodulin-dependent protein kinase II, neurogranin, and activity-regulated cytoskeleton-associated protein RNAs by the A2 pathway,” Mol. Biol. Cell, vol. 19, no. 5, pp. 2311–2327, May 2008. H. Kobayashi, S. Yamamoto, T. Maruo, and F. Murakami, “Identification of a cis- acting element required for dendritic targeting of activity-regulated cytoskeleton- associated protein mRNA,” Eur. J. Neurosci., vol. 22, no. 12, pp. 2977–2984, Dec. 2005. T. Kawashima et al., “Synaptic activity-responsive element in the Arc/Arg3.1 promoter essential for synapse-to-nucleus signaling in activated neurons,” Proc. Natl. Acad. Sci. U. S. A., vol. 106, no. 1, pp. 316–321, Jan. 2009. M. Inoue, N. Yagishita-Kyo, M. Nonaka, T. Kawashima, H. Okuno, and H. Bito, “Synaptic activity-responsive element (SARE): A unique genomic structure with an unusual sensitivity to neuronal activity,” Commun. Integr. Biol., vol. 3, no. 5, pp. 443–446, Sep. 2010. T. Kawashima et al., “Functional labeling of neurons and their projections using the 30 [37] [38] [39] [40] [41] [42] [43] [44] [45] [46] [47] [48] [49] [50] [51] [52] synthetic activity–dependent promoter E-SARE,” Nat. Methods, vol. 10, p. 889, Jul. 2013. A. T. Sørensen et al., “A robust activity marking system for exploring active neuronal ensembles,” Elife, vol. 5, p. e13918, Sep. 2016. G. G. Gross et al., “Recombinant Probes for Visualizing Endogenous Synaptic Proteins in Living Neurons,” Neuron, vol. 78, no. 6, pp. 971–985, Jun. 2013. K.-O. Cho, C. A. Hunt, and M. B. Kennedy, “The rat brain postsynaptic density fraction contains a homolog of the drosophila discs-large tumor suppressor protein,” Neuron, vol. 9, no. 5, pp. 929–942, 1992. M. Chalfie, Y. Tu, G. Euskirchen, W. W. Ward, and D. C. Prasher, “Green fluorescent protein as a marker for gene expression,” Science (80-. )., vol. 263, no. 5148, pp. 802 LP – 805, Feb. 1994. J. Marshall, R. Molloy, G. W. J. Moss, J. R. Howe, and T. E. Hughes, “The jellyfish green fluorescent protein: A new tool for studying ion channel expression and function,” Neuron, vol. 14, no. 2, pp. 211–215, 1995. C. Combs, “Fluorescence Microscopy: A Concise Guide to Current Imaging Methods,” Curr. Protoc. Neurosci., vol. Chapter 2, p. Unit2.1, Jan. 2010. “Traditional Cloning,” New England BioLabs, 2019. [Online]. Available: https://www.neb.com/applications/cloning-and-synthetic-biology/traditionalcloning. [Accessed: 11-Apr-2019]. “Gibson Assembly,” Barrick Lab, 2018. [Online]. Available: https://barricklab.org/twiki/bin/view/Lab/ProtocolsGibsonCloning. [Accessed: 11- Apr2019]. L. Liner, “How Does Gibson Assembly® Cloning Compare to Traditional Cloning Methods?,” SGI-DNA, 2016. [Online]. Available: http://blog.sgidna.com/2016/08/howdoes-gibson-assembly-cloning.html. [Accessed: 11-Apr-2019]. T. Narahashi, J. W. Moore, And W. R. Scott, “Tetrodotoxin blockage of sodium conductance increase in lobster giant axons,” J. Gen. Physiol., vol. 47, no. 5, pp. 965–974, May 1964. D. R. Curtis, A. W. Duggan, D. Felix, And G. A. R. Johnston, “GABA, Bicuculline and Central Inhibition,” Nature, vol. 226, no. 5252, pp. 1222–1224, 1970. N. C. Shaner, R. E. Campbell, P. A. Steinbach, B. N. G. Giepmans, A. E. Palmer, and R. Y. Tsien, “Improved monomeric red, orange and yellow fluorescent proteins derived from Discosoma sp. red fluorescent protein,” Nat. Biotechnol., vol. 22, no. 12, pp. 1567–1572, 2004. K. Dore et al., “SYNPLA, a method to identify synapses displaying plasticity after learning,” Proc. Natl. Acad. Sci., p. 201919911, Jan. 2020. A. Hayashi-Takagi et al., “Labelling and optical erasure of synaptic memory traces in the motor cortex,” Nature, vol. 525, no. 7569, pp. 333–338, Sep. 2015. A. Guru, R. J. Post, Y.-Y. Ho, and M. R. Warden, “Making Sense of Optogenetics,” Int. J. Neuropsychopharmacol., vol. 18, no. 11, pp. pyv079– pyv079, Jul. 2015. R. K. P. Benninger and D. W. Piston, “Two-photon excitation microscopy for the study of living cells and tissues,” Curr. Protoc. cell Biol., vol. Chapter 4, p. Unit- 4.11.24, Jun. 2013.