Citation
Exploration of serotonin distribution and activity within the pavement ant brain (Tetramorium Caespitum)

Material Information

Title:
Exploration of serotonin distribution and activity within the pavement ant brain (Tetramorium Caespitum)
Creator:
Schumann, William ( author )
Place of Publication:
Denver, Colo.
Publisher:
University of Colorado Denver
Publication Date:
Language:
English
Physical Description:
1 electronic file : ;

Subjects

Subjects / Keywords:
Ants ( lcsh )
Serotoniin ( lcsh )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Review:
Ants have miniaturized brains, yet they exhibit surprisingly complex behaviors. In social insect colonies, individuals gather information, integrate it, and compare that information to an inherent set of rules to make behavioral decisions. Each individual decision can lead to complex, self-organized behaviors. However, little is known about the proximate mechanisms behind these collective behaviors. The study of neurotransmitter monoamines, such as serotonin, provide a possible explanation for such complex behaviors. Serotonin is associated with reproductive dominance, colony foundation, aggression, trophallaxis, behavioral development, division of labor, repertoire expansion, and nestmate recognition in ants. This study used pavement ants (Tetramorium caespitum) as a model species to explore the distribution and activity of serotonin within the neural architecture of the ant brain. Ants were exposed to a variety of contexts: social interaction, aggression, food excitement, and antenectomy. After exposure whole brains were dissected and underwent immunohistochemistry (IHC) to stain for the monoamine serotonin and the genetic marker for neuronal activity, c-Fos. Serotonin immunoreactivity was found in the antennal lobes (AL), subesophageal ganglion (SOG), optic lobes (OL), mushroom bodies (MB), and the MB calyces. Serotonergic processes were seen in the MB calyces, terminating in the lip of the calyx. Neuronal symmetry was observed in the AL, SOG, and OL. A maximum estimate of 78 serotonergic neurons were stained. This study provides further information for the serotonergic architecture within the ant brain. Fos colocalization with serotonin was seen in the AL, SOG, OL, and MB calyces in an ant engaged in aggression. This suggest that serotonin is active in these locations during this behavior. However, only one brain showed positive staining for Fos. The location and activation of serotonergic neurons in the AL and MB calyces suggest that serotonin is released from the AL and shuttled to the MB calyx’s lip where it is processed and a decision to aggress is made. If the Fos technique provided in this study can be replicated, it offers a new method to better understand the underlying mechanisms of behavior in ants.
Thesis:
Thesis (M.S.)--University of Colorado Denver
Bibliography:
Includes bibliographical references.
System Details:
System requirements: Adobe Reader.
Statement of Responsibility:
by Willia, Schumann.

Record Information

Source Institution:
University of Colorado Denver
Holding Location:
Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
on10611 ( NOTIS )
1061150505 ( OCLC )
on1061150505

Downloads

This item has the following downloads:


Full Text
EXPLORATION OF SEROTONIN DISTRIBUTION AND ACTIVITY WITHIN THE
PAVEMENT ANT BRAIN (TETRAMORIUM CAESPITUM)
by
WILLIAM SCHUMANN B.S., Pacific University, 2008
A thesis submitted to the Faculty of the Graduate School of the University of Colorado in partial fulfilment of the requirements for the degree of Master of Science Biology Program
2018


This thesis for the Master of Science degree by William Paul Schumann has been approved for the Biology Program by
Michael Greene, Chair John Swallow Benjamin Greenwood Sondra Bland
Date: May 12, 2018
ii


Schumann, William (M.S., Biology Program)
Exploration of Serotonin Distribution and Activity within the Pavement Ant Brain
{Tetramorium caespitum)
Thesis directed by Associate Professor Michael Greene and Professor John Swallow
ABSTRACT
Ants have miniaturized brains, yet they exhibit surprisingly complex behaviors. In social insect colonies, individuals gather information, integrate it, and compare that information to an inherent set of rules to make behavioral decisions. Each individual decision can lead to complex, self-organized behaviors. However, little is known about the proximate mechanisms behind these collective behaviors. The study of neurotransmitter monoamines, such as serotonin, provide a possible explanation for such complex behaviors. Serotonin is associated with reproductive dominance, colony foundation, aggression, trophallaxis, behavioral development, division of labor, repertoire expansion, and nestmate recognition in ants. This study used pavement ants {Tetramorium caespitum) as a model species to explore the distribution and activity of serotonin within the neural architecture of the ant brain. Ants were exposed to a variety of contexts: social interaction, aggression, food excitement, and antenectomy. After exposure whole brains were dissected and underwent immunohistochemistry (IHC) to stain for the monoamine serotonin and the genetic marker for neuronal activity, c-Fos. Serotonin immunoreactivity was found in the antennal lobes (AL), subesophageal ganglion (SOG), optic lobes (OL), mushroom bodies (MB), and the MB calyces. Serotonergic processes were seen in the MB calyces, terminating in the lip of the calyx. Neuronal symmetry was observed in the AL, SOG, and OL. A maximum estimate of 78 serotonergic neurons were stained. This study provides further information for the serotonergic architecture within the ant brain. Fos colocalization with serotonin was seen in the AL, SOG, OL, and MB calyces in an ant engaged in aggression. This suggest that serotonin is


active in these locations during this behavior. However, only one brain showed positive staining for Fos. The location and activation of serotonergic neurons in the AL and MB calyces suggest that serotonin is released from the AL and shuttled to the MB calyxs lip where it is processed and a decision to aggress is made. If the Fos technique provided in this study can be replicated, it offers a new method to better understand the underlying mechanisms of behavior in ants.
The form and content of this abstract are approved. I recommend publication.
Approved by: Michael Greene John Swallow
IV


TABLE OF CONTENTS
I. INTRODUCTION 1
II. METHODS 15
III. RESULTS 19
IV. DISCUSSION 27
V. REFERENCES 32
v


CHAPTER 1
INTRODUCTION
Ant colonies are regulated as non-hierarchal, distributed system where individual decisions drive colony behavior. The queen does not command her workers. Rather,
individuals asses local informational cues, integrate them in their simple, miniaturized brains, and compare that information to an inherent set of rules. Often these cues are based on the interaction rates they have with their nestmates (Greene and Gordon 2007, Davidson and Gordon 2017, Hoover et at. 2016, Bubak et at. 2016, Pratt 2005), the density of pheromone trails (Sumpter 2005, Muscedere et at. 2012), and other olfactory information like cuticular hydrocarbons (Hu et at. 2017, Sano et at. 2018 ). Ants share this information locally, and interact with a small, limited network of nestmates. The network is limited to a small fraction of the colony because one ant could not possibly interact with every other individual in the colony. However, these small networks extend to other small networks causing information to spread and a collective behavior to be expressed Therefore, individuals make decisions given their limited information (Sasaki and Pratt 2013, Edwards and Pratt 2009). The many decisions made by individuals lead to the emergence of colony behaviors including foraging, nest construction, brood care, colony defense, and colony maintenance. The colonys survival is dependent on these decisions made by individuals (Pruitt and Riechert 2011). Out of the repeated interactions of individuals a system emerges: self-organization to a collective decision.
Complex emergent behavior can arise from self-organization stemming from a simple set of rules (Sumpter 2005). These collective behaviors are frequently made via consensus from individuals and these collective decisions shape the fitness of the colony (Cronin 2015). Consensus decisions provide a way for individuals to coordinate information and, once a
1


threshold is reached, elicit a group-level behavior (Cronin 2015). The colony is, in essence, a society that functions as an information processing unit distributing its cognitive tasks to individual ants (Pratt et al. 2002). Just as an individuals decision is made from complex interactions between a network of neurons so too does an ant colonys decision emerge from a network of interactions among individual ants (Marshall et al. 2009, Edwards and Pratt 2009).
Ants have miniaturized brains, less than 1/1000 the size of a honey bee, yet collectively they exhibit amazingly complex behaviors including farming fungus, raising aphids for honeydew, stealing non-nestmate workers (Alloway 1979), building structures with their own bodies (Lioni and Deneubourg 2004), and warfare (Hoover et al. 2016). Relative to their brain volume, they possess enlarged antennal lobes (odor processing) and mushroom bodies (cognition, learning, and memory) (Fahrbach 2006). Using these enlarged structures, individual ants interpret local stimulus using intrinsic rules to make complex behavioral decisions. These decisions lead to the collective behavior of the colony.
Within Hymenoptera (wasps, bees, ants, and sawflies), we see that brain structures associated sensory collection and integration are enlarged. Apis mellifera have large optic lobes (OL) that improve vision which allows them to among other things, discriminate colors (von Frisch 1914) and even human faces (Dyer et al. 2005). However, most ants see a reduction in the size of their optic lobes because they mainly rely on olfaction (Cronenberg 1996). Since ants primary sense organ is their antennae, the ant brain has antennal lobes (AL) that are larger, relative to body and brain size, and more complex than in other social Hymentoptera (Cronenberg 1996). Honey bees brains contain approximately 166 AL glomeruli while ants possess an estimated 200 some of which are larger and more complex than in other insects. Glomeruli are associated with odor detection and larger and more complex glomeruli suggest the
2


importance of olfaction for social behavior in ants (Cronenberg 1996). Lastly, most social Hymenoptera have enlarged mushroom bodies (MB) a region responsible for higher order processing like sensory integration, learning, memory, and spatial orientation, all functions important to the social life of ants and bees (Cronenberg 1996). Within ants we see larger MB when compared to other insects and other Hymentopterans (Cronenberg 1996). In ants and bees, the mushroom bodies are divided into two calyces: the medial and lateral calyx. The role of the MB calyces in bees and ants is different from other insects as well. While other insects primarily use the MB calyces for odor processing, bee and ant MB calyces are a center for multimodal sensory integration (Rossler and Groh 2012). For example, the MB in A. mellifera work in tandem with the AL and the subesopheagel ganglion (SOG) during olfactory learning (Gauthier and Grunewald 2012). These three regions appear to play important roles within social insects and their behavior.
Social insects, like ants and other Hymenoptera, have relatively simple brains yet they can exhibit extremely complex behaviors. Hoyle (1985) established the orchestration hypothesis as a means to explain the coordination of complex behaviors by neurotransmitters. He discovered that direct manipulation of monoamines could elicit specific behaviors and that monoamines worked on the neural circuitry of organisms. With the development of the orchestration hypothesis, researchers had a new possible mechanism to study, monoamines. Octopamine (OA), dopamine (DA), and serotonin (5-HT) are commonly studied to determine the animergic effects on behavior (Kamhi and Traniello 2013, Hoyer et al. 2005, Tsuji et al. 2007, Muscedere et al. 2012, Muscedere et al. 2012, Muscedere et al. 2016, Bubak et al. 2016, Hoover et al. 2015). Ant researchers began to study these monoamines to understand the mechanisms underlying social behaviors (Kamhi and Traniello 2013). Exploring monoamines offers novel explanations for the
3


modulation of complex social behaviors. By examining neurochemistry and neural architecture, we can determine the underpinnings of social behavior, colony-level division of labor, and collective intelligence (Kamhi and Traniello 2013).
Octopamine (OA), DA, and 5-HT have been implicated in a wide range of ant behaviors including colony foundation, interspecies and predatory aggression, learning, development, trophallaxis, and nestmate recognition (Kamhi and Taniello 2013, Table 1). Kamhi et al. (2015) showed that OA is instrumental in aggression in the Australian weaver ant, Oecophylla smaragdina. This species has different worker castes; large majors engage in aggressive territorial defense while minors care for brood and collect food. For example, Kamhi et al.
(2015) found that majors contained higher brain titer levels of OA. To see if OA influenced aggression they manipulated OA levels within minors mid saw an increase in aggressive behaviors. Smith et al. (2013) found that in the ant, Achromyrmex echinatior, DA, OA, and 5-HT brain levels differed based on castes suggesting that these monoamines influence worker specialization. In a recent study, Wada-Katsumata et al. (2011) show that DA and OA are linked to the social behaviors of grooming and trophallaxis in Formica japonica. In this study ants where starved while DA and OA levels were monitored. Starving ants displayed low levels of DA, which was rescued after trophallaxis. Ants were also isolated from social interaction. This increased OA levels. When introduced to nestmates these ants increased their trophallaxis duration, allogrooming, and self-grooming behaviors suggesting that high levels of OA influences social behaviors. Queens from fie species, Veromessor pergandei, have varying colony foundation strategies, founding a colony singly, or working cooperatively with other queens. Muscedere et al. (2016) examined the role of DA, OA, and 5-HT in these different strategies. Queens that found colonies together often engage in aggressive conflict with each
4


other to establish dominance after worker eclosion. Early in colony foundation, young queens that founded singly had heightened brain levels of 5-HT and queens that founded together had heightened levels after workers eclosed. This suggests that 5-HT modulates aggression during colony foundation (Muscedere et al. 2016). Serotonin was found to influence foraging behaviors in Pheidole dentata (Muscedere et al. 2012). This study pharmacologically decreased serotonin in individuals. These individuals were less likely to follow pheromone trails and if oriented to trails only followed for short distances compared to control ants. These studies illustrate just how many behaviors are modulated by monoamines.
Although, monoamines have been shown to have many important functions in regulating social insect behavior, little is known about the underlying mechanisms that differentiate how monoamines modulate these complex behaviors. Most of our current methods for studying monoaminergic effects on behavior rely on manipulations of whole brain levels which offer much insight into how levels of monoamines can affect behaviors, but lack resolution at the level of brain architecture (Muscedere et al. 2016, Bubak et al. 2016, Penick et al. 2014, Smith et al. 2013, Kamhi et al. 2015, Muscedere et al. 2012, Wada-Katsumata et al. 2011). These studies typically employ High Powered Liquid Chromatography (HPLC) to measure the monoamine content within a whole brain. They also incorporate pharmacological manipulations to see how increases and decreases in whole brain levels of monoamines affect behaviors. These studies are limited in how they explore the possible mechanisms behind a given behavior. Pharmacological manipulations flood the brain with monoamines, higher levels than those found naturally. This blunt approach makes it easy to elicit a behavioral response and correlate a monoamine with a given behavior. However, this does little to show how these monoamines work within the brain to generate the behavior. If a deluge of monoamines floods the brain, how can you determine
5


what, where, when, and how these molecules modulate behavior. One can only make a general assumption that having high levels of monoamines causes a change in behavior. Unfortunately, these types of studies only look at the whole brain titers of the monoamines rather than looking at the neural architecture or the location of the monoamine within the brain. How then do we differentiate the specific effects of each monoamine? Do these monoamines work in concert or are they antagonistic? Does one modulate the others? While we have determined that these monoamines influence these specific behaviors, it remains unclear how exactly monoamines modulate behaviors (Hoyer et al. 2005).
As stated above there are limitations to overriding the brain with monoamines. To gain better insight into the relevant cells and structures involved in behavior, I used immunohistochemistry (IHC). This tool allows for the exploration of possible monoaminergic architecture of the brain to provide insights into the mechanisms that contribute to the generation of behavior (Giraldo et al. 2013, Hoyer et al. 2005, Smith et al. 2013, Kamhi et al. 2015). IHC is used commonly to explore social insect brain architecture (Hoyer et al. 2005, Galizia et al. 2012, Gronenberg 1996, Giraldo et al. 2013). Research often focuses on monoaminergic systems within the brain (Hoyer et al. 2005, Tsuji et al. 2007, Giraldo et al. 2013). By understanding the architecture within the brain, these studies link the systems within the brain and its architecture to specific behaviors including aggression (Hoyer et al. 2005), division of labor and repitoire expansion (Giraldo et al. 2013, Muscedere and Traniello 2012, Galizia et al. 2012), and sensory integration (Galizia et al. 2012).
Immunohistochemistry is a widely used technique to study the distribution and activity of monoamines. It involves detection of specific epitopes in the cells or tissues of interest using antibodies that bind to those specific epitopes. Additionally, fluorescence proteins are added for
6


better resolution of the target molecules. Studies that use this technique can pinpoint neurons and neuronal pathways associated with specific behaviors such as how 5-HT regulates division of labor in Pheidole dentata (Giraldo et al. 2013), or the role DA and 5-HT play in intraspecific aggression in Harpegnathos saltator (Hoyer et al. 2005). Giraldo et al. (2013) used IHC to determine structural differences between castes as they age as a possible explanation for behavioral repertoire expansion. Using IHC they were able to identify serotonergic neurons and their arborization within the mushroom bodies for all castes. Giraldo et al. (2013) discovered that only the major castes displayed structural changes as they matured as well as the majors had significantly more complex arborization within the mushroom bodies. They concluded that these structural differences were a possible explanation for the differences in caste behavior. Hoyer et al. (2005) investigated the serotonergic and dopaminergic neuronal systems as they relate to intraspecies aggression and to determine any differences in neural anatomy between castes and sex by employing dual staining IHC. They hoped to find structural differences between these monoaminergic systems during aggression and between castes and sex. The study was able to describe the serotonergic and dopaminergic systems within the brain, but they could not find any anatomical changes during aggression. However, they did find that males had smaller brain volumes proportionally to females that resulted in reduced size of the mushroom bodies. Additionally, the two female castes possessed more serotonergic processes while the males displayed higher dopaminergic processes. These differences could explain the differences in behavioral repertoire between caste and sex (Hoyer et al. 2005). Again, this study was able to describe monoaminergic neural anatomy to better understand behavior. These studies use this technique to gives us a better picture about the mechanisms of ant behaviors (Hoyer et al. 2005,
7


Giraldo et al. 2013). Using this technique can give a detailed picture of the origin, location, and mechanism of a behavior.
Since IHC has a variety of applications that extend to neuronal tissues, it becomes an excellent tool when trying to determine the mechanism behind specific behaviors (Hoyer et al. 2005, Tsuji et al. 2007). This technique allows you to visualize the tissue of interest to locate particular molecules of interest, in this case monoamines. Hoyer et al. (2005) used IHC to determine if the distribution of DA and 5-HT was different between castes and sexes in H. saltator since brain morphology did not change among these groups. The researchers found that there was deep innervation of serotoninergic neurons within mushroom bodies (responsible for cognition, learning, and memory) with serotonin neurons making connections between the mushroom bodies and antennal lobes (responsible for olfaction). Additionally, they were able to show that sterile workers and males had different distributions of serotonergic processes as compared to other castes and sexes. Sterile workers dopaminergic processes were also different from other castes. Hoyer et al. (2005) showed that different brain architecture exists in ants based on behavioral repertoire. They illustrated that monoamines must work differently within castes and sex. This is a possible explanation for the difference of behaviors among different castes of workers and between sexes. Tsuji et al. (2007) was able to provide the complete serotonergic network found in the antennal lobes of C. japonicas. This network could be used to determine how serotonin can orchestrate the social behaviors of C.japonicas. In both these studies, IHC provided a detailed map of the location of these monoamines in the brain. Knowing the location of these monoamines informs us of its function. If monoamines are in a brain region responsible for olfaction, they are likely responsible for modulating behaviors requiring olfaction
8


and chemo sensation. Since each region of the brain is responsible for different functions, this technique shows the possible mechanism that monoamines employ to elicit behavior.
IHC provides a map showing regions where monoamines are present. The information gathered from these maps can help us determine how these monoamines work in the brain and what kind of behaviors they may modulate (Hoyer et al. 2005, Tsuji et al. 2007). IHC gives the exact location of these monoamines during specific behaviors (Hoyer et al. 2005). Since the function of each brain region is known, we can use the location of these monoamines within those regions to explain the mechanisms behind behaviors like aggression and social interaction. But the method is limited since behavior relies on the firing of action potentials within neurons, which is difficult to detect with IHC.
To address these concerns with IHC, the methods proposed in this paper will incorporate the early activator gene, c-Fos. c-Fos can shows us the neuronal activity within the brain mid help to pinpoint areas that are active during a specific behavior. This can illuminate possible mechanisms including which serotonergic neurons are active during specific behavioral contexts. c-Fos is an early activator gene that is expressed when an action potential fires mid has become a marker for neuronal activity (Day et al. 2008, Dragunow and Faull 1989). c-Fos is visualized in IHC by targeting the proteins expressed by c-Fos. Dragunow and Robertson (1988) found that c-Fos was expressed in recently activated neurons in the rat brain and began using it as a high resolution marker for synaptic pathways in the mammalian brain (Dragunow and Faull 1989). C-Fos became used in vertebrates to understand the neural mechanisms and pathways behind behaviors (Guzowski et al. 2001, Shu 2002, Neumaier et al. 2001, Bastle et al. 2016). Bastle et al. (2016) used c-Fos to determine an interaction between the social and nicotine reward system in the brains of adult male rats. Additionally, Shu (2002) used cFos to explain the mechanism of
9


learning in rats. C-Fos allowed Neumairer et al. (2001) to visualize the location 5-HT7 receptors in rat brains which provided insight into their role in the circadian rhythm and potential enhancement of serotonergic pharmaceuticals. As these studies show, c-Fos is a powerful tool to explain neural mechanisms in animals.
c-Fos has been important tool for the understanding of neural mechanisms of behavior. Rats learning to locate platforms submerged under water saw increased expression of c-Fos RNA in the hippocampus, entorhinal cortex, and visual cortex when compared to controls.
Additionally, the act of learning the trail increased the level of c-Fos RNA expression in the dorsal hippocampi when compared to controls. These data suggest that Fos is expressed during spatial learning and that the regions showing Fos activity are responsible for the behavior (Guzowksi et al. 2001). To determine the efficacy of a specific neuropeptide thought to regulate hunger, NMUR2, c-Fos was used to show regions of Fos expression when mice were given a NMUR2 agonist. Fos was measured after the agonist, NMU-7005, was given to obese and fasted mice. Obese mice saw increased Fos expression in the lateral part of the hypothalamus when given the agonist when compared to controls given a saline solution. While fasted mice showed increased expression in the medial part of the hypothalamus compared to obese mice. Finally, obese mice given the agonist had increased Fos expression in the medulla oblongata. Again, this study shows where this drug is active providing a possible mechanism for the agonist (Kaisho et al 2017). c-Fos provides insight into the mechanism for a given behavior. It pinpoints the regions that express it which highlights where the behavior is likely occurring.
While this technique is widely used with vertebrate models, we rarely see it used in invertebrates (Renucci et al. 2000, Ghosal et al. 2010). There is little reason not to use c-Fos since studies have shown that invertebrates express c-Fos and c-Fos related antigens (Ghosal et
10


al. 2010, Renucci et al. 2000). Ghosal et al. (2010) measured the immunoreactivity for c-Fos and Fos related antigens (FRAs) in male crickets using IHC. They identified specific brain regions that expressed c-Fos and FRA proteins during aggression. They found FRA and Fos immunoreactivity in the ventromedial region of the brain, the deutocerebrum, and among the Kenyon cells of the cortex of the MB. Additionally, Fos was located in the nuclei of brain cells. Renucci et al. (2000) also measured the immunoreactivity of Fos and FRAs using IHC with the cricket brain. They saw staining within the nuclei of neurons of female crickets confirming the presence of Fos and FRAs in the insect brain. Still many invertebrate studies do not take advantage of this tool to understand the neural mechanisms of behavior. c-Fos can bring a higher resolution of the neural pathways behind behaviors to insects especially ants. As shown above c-Fos has aided in understanding complex behaviors in rats. We can use this tool to understand the complex, collective behaviors of ants.
As mentioned previously, monoamines are of growing interest as a possible explanation for ant collective behavior (Kamhi and Traniello 2013). This is especially relevant since these neurotransmitters are highly conserved across taxa (Iyer et al. 2004, Kang et al. 2009). Marshall et al. (2009) points out that neural mechanisms behind primate decision making bear a striking similarity to the self- organized decisions made by ant colonies. If ant collective behavior can inform how we understand how our neurons behave during the decision-making process, then perhaps learning how monoamines drive social behaviors in ants could explain role these monoamines play our own social collective behavior. In this way we could create a foundation that develops Hoyles orchestration hypothesis as the basis for complex social behavior. Studies investigation these monoamines typically look at serotonin, dopamine, and octopamine when
11


exploring the possible neural mechanisms behind collective behavior (Kamhi and Traniello 2013). For ease of analysis and technique this study focused on a single monoamine, serotonin.
(5-HT).
Behavior Monoamine
5-HT
S. invicta [Boulay et al., 2001]; H. saltator [Hoyer et al., 2005];
Reproductive dominance and colony foundation S. peetersi [Cuvillier-Hotand Lenoir, 2006]; F.japonica [Aonuma and Watanabe, 2012a]
P. dentata [Giraldo et al., 2013; Giraldo
Subcaste-related division of labor and Traniello, unpubl. obs.]; A. echinatior [Smith et al., 2013]
P. dentata [Seid and Traniello, 2005; Seid etal., 2008; Muscedere
Worker behavioral development, repertoire expansion, and temporal polyethism et al., 2012; Giraldo etal., 2013]; S. peetersi [Cuvillier-Hotand Lenoir, 2006]; F. polyctena [Wnuk et al., 2010]
C. mus [Falibene et al., 2012];
Social food flow (trophallaxis) P. dentata [Muscedere et al., 2013]
F. japonica [Aonuma and Watanabe, 2012]; P. dentata [Giraldo et
Aggression al., unpubl. obs.] T. caespitum [Hoover et al., 2015; Bubak et al. 2016]
C. fellah [Boulay et al., 2000]; O. smaragdina [Kamhi and
Nestmate recognition Traniello, unpubl. obs.] T. caespitum [Hoover et al., 2015; Bubak et al. 2016]
Table 1. Serotonin and the social behaviors associated with serotonin*
12


*Table modified from Kamhi and Traniello 2013.
Serotonin is an intracellular signaling molecule that can be found in all organisms with a central nervous system (Vleugels et al. 2015). This molecule is so widely conserved across taxa that it is found in plants and fungi (Kang et al. 2009). 5-HT is derived from tryptophan and is associated with the regulation of aggression, feeding/appetite, learning, memory, and social behaviors in insects (Vleugels et al. 2015). In ants, this monoamine has a wide variety of effects on their social behaviors. Studies have shown that 5-HT plays a role in colony foundation, interspecies and predatory aggression, learning, development, trophallaxis, and nestmate recognition (Kamhi and Traniello 2013, Table 1). Table 1 displays the species and papers that show a link between 5-HT and a specific social behavior. While this research shows that 5-HT is responsible for a long list of behaviors, we do not yet know how 5-HT modulates each of these behaviors. Thus, we need techniques like IHC to determine the mechanisms by which 5-HT influences specific behavior. This study will attempt to use IHC to establish 5-HTs role in modulating aggression in pavement ants.
Pavement ants (Tetramorium caespitum) are a ubiquitous invasive species found throughout the urban environment. They prefer to live under slabs, or flat rocks, hence the name pavement ants. These qualities make them an easy subject to obtain and study. In addition, pavement ants display a very conspicuous collective behavior, war. Pavement ants engage in conflict with other species of ants as well as conspecifics. Wars with other species are fought to the death, but not when pavement ants encounter non-nestmate conspecifics. Pavement ant war is ritualized with little to no causalities incurred by both colonies. Individual ants will use their mandibles to lock onto an opponent and contest in a tug-o-war battle to push the other colony out of their territory. Wars will last for hours until a victor is decided. Recent interactions with
13


nestmates and the frequency of those interactions influence an individual to decide to fight when encountering a non-nestmate (Hoover et al. 2016, Bubak et al. 2016). These interactions between nest-mates cause an increase in 5-HT and OA levels which prime the brain for aggressive confrontation with non-nestmates (Bubak et al. 2016, Hoover et al. 2016). Each interaction spikes OA and 5-HT levels in the brain after which the levels of monoamines start to decline. So repeated interactions are necessary to sustain monoamines at their threshold to encourage a decision to fight (Hoover et al. 2016). It is unclear what levels of monoamines are necessary for a decision to fight nor what the rate of decline of monoamines is. The methods of this paper were developed to better understand the mechanisms behind this decision to fight.
The methods in this paper were developed to further our understanding of the neural architecture and to create a technique in which we capture the neural activity of pavements ant based on different behavioral concepts. With these methods I will describe the serotoninergic architecture in pavement ants, show how this method can be used to measure neuronal activity, and provide an example on how these methods can be employed to make behavioral comparisons
14


CHAPTER 2
METHODS
Materials
Animals'. Pavement ants were collected in Denver, CO by aspirating individuals from bait trails. Ants were baited with peanut butter. Ants were housed in aspirator tubes with grass to provide a level of humidity. Ants were housed in these containers for 24hrs prior to experimentation and sacrifice.
Special Equipment:
The following lists contains all the equipment used to conduct the experiment. The specificity of the list is to account for all possible contributions to the images generated by IHC.
Horizontal shaker (Belly Dancer, Stovall Life Science, Greensboro, NC USA).
Cold room horizontal shaker (2314FS Lab Rotator Stirrer Shaker, Fisher Scientific, Waltham, MAUSA).
Sterilized 48 well cell culture plates (Greiner Bio-One, Kremsmunster, Austria)
Cold room (Environmental Growth Chambers, Chagrin Falls, OH USA)
For dissections, cold plate (CaterWare Round Aluminum Cooling Plate 11-inch x 1 3/4 inch, TableCraft, Gurnee, ILUSA)
Brains were imaged using Ziess LSM 700 confocal microscope.
Some images were imaged on a 31 Marianas inverted Spinning Disk confocal microscope
Chemicals
The following list contains all chemicals used in the immohistochemical reactions. I provide the specific chemicals used to ensure that the procedure laid out in this section can be replicated accurately. The specificity of the list accounts for all possible contributions to the success of the IHC.
4% Formaldehyde in phosphate buffered saline (pH 7.40.2) (BM-154, Lot# E1RF25R
Boston BioProducts Ashland, MA USA)
15


Gultaraldehyde, 50% aqueous solution (LOT: P27A045, Alfa Aesar, Ward Hill, MA USA)
Phosphate buffered saline (pH 7.4, IX, Gibco-Thermo Fisher Scientific, Waltham, MA USA).
Normal goat serum (S26-100ML, Lot#2855574, EMD Millipore, Temecula, CA USA)
Primary anti-5HT rabbit polyclonal IgG antibody (20080, Lot#l650001 hnmunostar Hudson, WI USA)
Primary anti-Fos mouse monoclonal IgG antibody (E-8, sc-166940, Santa Cruz Biotechnology, Dallas, TX USA)
Secondary goat anti-rabbit IgG Alexa Flour 594 (A21121, Lot#l 889303, Thermo Fisher Scientific, Waltham, MA USA). Secondary goat anti-mouse IgG .Alexa Flour
488 (A11012, Lot# 1745478, Thermo Fisher Scientific, Waltham, MAUSA).
Triton X-100 (Lot# SLBJ0812V, Sigma Life Science, St. Louis, MO)
Vectashield mounting medium for fluorescence with DAPI (H-1200, Vector Laboratories, Burlingame, CAUSA)
Detailed Procedure
Behavioral trails: Prior to dissections ant were placed into different behavioral contexts. This was to determine their brain states under differing conditions for comparison.
Aggression: Ants from two different colonies were collected. 50 ants from each colony were placed in 11cm X 11cm plastic container. Ants were allowed to acclimate for 15 minutes. After 15min ants that engaged in aggressive behavior (defined as biting, locking mandibles) were selected and sacrificed.
Social (control): Ants from the same colony were placed in 11cm X 11cm plastic container. Ants were allowed to acclimate for 15 minutes. After the acclimation period ants engaged in social behavior (defined as attenuating another individual) were sampled and sacrificed.
Dissection: All ants were dissected on a cold plate kept stored at -20C. Prior to dissection the cold plate was allowed to warm for 20 minutes. Once ants were selected they were placed on ice for 15 secs to anesthetize them. Once anesthetized they were transferred to the cold plate for
16


dissection. Using micro dissection tools ant heads were removed. The ant heads had their mandibles removed to allow fixative into the cranial cavity.
Immunohistochemsitry. All wash and incubation steps used a horizontal shaker. All steps were performed at room temperature (21C) except fixation and primary antibodies incubations.
Dissected ant heads were fixed in 4% formaldehyde with 0.1% Gluteralaldehyde for 5 hours. After 5 hours the brains were removed from the skull exoskeleton. The brains were then washed in PBS (3 X lOmin) Following fixation ant brains were placed in 0.3% Triton X-100 PBS for 2 hours at to permeabilize the tissue. To reduce non-specific bindings brains were incubated in 10% Normal Goat Serum (NGS) in PBS for 1 hour. Then each brain underwent standard secondary IHC staining for serotonin and c-Fos. Serotonin antibody used was ImmunoStar rabbit anti-5-HT. c-Fos antibody used Santa Cruz mouse anti-Fos. Ant brains were incubated in a dilution of rabbit anti-5-HT primary antibody (1:500, 0.5% NGS in PBS) and mouse anti-Fos primary antibody (1:50, 0.5% NGS in PBS) for 3 days at 4C. After the incubation period in primary antibodies the brains were washed in PBS (3 X lOmin). Following the washes, the tissue was incubated in secondary antibodies, goat anti-rabbit IgG Alexa Flour 594 for serotonin and goat anti-mouse IgG .Alexa Flour 488 for c-Fos (1:500 dilution in a solution of PBS). This was followed by a wash in PBS (3 X lOmin). Finally, the brains were mounted on positively charged slides using Vectashield anti-fading solution with DAPI. The slides were prepared using a piece of black tape with a window cut into it as the mounting field.
Confocal Microscopy. Images created from IHC were generated using Ziess LSM 700 confocal microscope. All images were created using the 20X objective lens. Red and green lasers were used at 594nm and 488nm. The red channel lasers power was set to 1.9% and the green channel lasers power was set to 3.0% for each image. Gain was left at default setting. Each image was
17


generated using z-stack. Imaging was set to the optimal slice thickness by using the built-in Ziess imaging software Optimal feature.
Image processing. Images were processed using FIJI. Images are 3d max projections of z-stack images. All images were compared to an atlas created by Tsuji et al. (2007) to identify morphological structures.
Controls. Controls were run on the secondary antibodies to ensure lack of nonspecific binding. These brains underwent the entire protocol, but without primary antibodies.
Due to issues with the techniques ability replicate c-Fos staining several different experiments were run to generate Fos expression.
Sucrose excitement. Ants were starved for 24 hours. After the starvation period, ants were given a solution of 10% sucrose prior to dissection. Dissection was performed the same as above. Ants were dissected at three different time points: immediately after sucrose ingestions, 15min after sucrose ingestion, and lhr after ingestion.
Antennal removal experiment: Ants were divided into three groups: controls, right antenna removed, or both antenna removed. Dissections were performed as described above, however prior to dissection each group was placed on ice for 15 secs, then placed on the cold plate where their antennae were removed. Controls underwent the same process, but did not have their antennae cut. All antennae were cut using micro dissection scissors. Ants from each group were sacrificed at three different time points: 5min, 6hrs, and 24hrs after antennal removal.
Analysis. The brains analyzed in the results are the best selected from each experimental group. Thus, all brains used in results are pooled from all the experimental contexts.
18


CHAPTER 3
RESULTS
Figure 1. Images of a whole T. caespitum brain stained for serotonin including ROI (regions of interest) and serotonin (5-HT) positive neurons. The figure is made up the best representations of each ROI. 5-HT is stained in red. a) Serotonin (5-HT) staining in a whole T.caespitum brain. Imaged at 20x magnification. AL = antennal lobes, MB = mushroom bodies, OL = optic lobe, and SOG = subesophageal ganglion. Brain in ventral orientation, b) SOG shows three pairs of serotonergic neuronal clusters highlighted by the blue arrows, c) A zoomed view of the serotonergic neuronal clusters in the antennal lobes, d) Highlighted serotonergic neurons within the optic lobe. 5-HT saturated medulla cells are present in the enclosed within the circle.
19


Figure 2. Images of the mushroom body (MB) structure in T. caespitum brains. The figure is made of the best representation of the MB. Serotonin (5-HT) is stained in red. Images produced at 20X magnification. MB = mushroom body, ICa = lateral calyx, and mCa = medial calyx, a) Zoomed image showing a MB. The arrows point to neurons in the MB and the dashed circle highlights serotonergic processes, b) Image showing the different calyces in the MB. The dashed circle highlights serotonergic processes.
Serotonin anatomy in T.caespitum brain. Our protocol showed excellent staining for serotonin (5-HT) (figure l.a). In figure 1 we saw strong presence of serotonergic neurons in the antennal lobes (AL), optic lobes (OL), and the subesophagael ganglion (SOG). There appears to be bilateral symmetry in number of neurons in those the regions (figure l.a). The AL showed an estimate of 10 neurons in each lobe (figurel.c). OL contained an estimate of 22 neurons per lobe.
20


The OL medulla appears saturated with 5-HT (figure l.d). Within the SOG there appears to be three clusters of neurons. A pair of clusters located in the anterior, toward the AL (n = 3 per cluster), a pair on lateral side (n = 3 per cluster), and a pair located laterally in the posterior (n =
3 per cluster) (figure l.b). The mushroom bodies (MB) also showed presence of serotonergic neurons (Figure 2.a). There is an estimate of 9 neurons in the MB (figure 2.b). The calyces of the MB have serotonergic processes suggesting that 5-HT is shuttled into the calyces from neurons elsewhere in the brain. The processes appear to terminate at the lip of each calyx.
Neuron count. Neurons were counted in images in which morphology could be identified (n=32). The neuron count represents an estimate of the number of neurons within in a T. caespitum brain. Due to the limitations of the technique, the numbers displayed the distribution in Figure 3 represent our best estimates of whole brain 5-HT neurons. Our max total count for 5-HT neurons was 78, our minimum was 1. The mean for the total number of neurons is 23.29. The median values for each region in the brain are close with their means very close to the median value
21


except for the OL. These neurons counted were the only neurons that were detected by our technique, meaning we can only provide an estimate on count.
t/3
d
£Z
.O
rr'
C
15
z
CD 0
O i o
1 1^1
40 o
AViV^WAS J
20 -H j
i i i j j fWWWWW1! # 1 1 1 1
H-H PWW'VWWV I 3 I
J, ,J, i: i
o 1 < 1 *
\ MB \ a; OL \ SOG 1 1 Other 1 1 Total
Type of Neurons
Figure 3. Boxplot showing the distribution of 5-HT neurons within T.caespitum. The black bar represents the median and the red diamond is the mean. Open circles are outlying points. AL = antennal lobes, MB =mushroom bodies, OL = optic lobe, and SOG = subesophageal ganglion.
22


Other represents any neurons that were present, but could not be placed within a specific morphological structure.
o
co
o
(£>
O
o
04
O
O O
MB Al OL SOG Other Type of Neurons
Figure 4. Boxplot showing the distribution of proportions of the total neuron count that are in each morphological structure. The proportions of each region were calculated using the total number of neurons within each imaged brain. The black bar represents the median and the red diamond is the mean. AL = antennal lobes, MB =mushroom bodies, OL = optic lobe, and SOG = subesophageal ganglion.
<5
o
o

m
c
o
5
Z
kf-
o
o
o
CL
o
We see a wide variability in each region of how many neurons are contained within that region (Figure 4) Mean MB = 39%, Mean AL = 43%, Mean OL = 52%, Mean SOG = 45%, Mean Other = 36%. We see the greatest proportion of neurons within the OL, mean proportion = 52%. The other regions are relatively similar in their proportion of the total.
23


Figure 5. linages of a whole T. caespitum brain stained for serotonin and c-Fos including ROI (regions of interest), serotonin (5-HT), c-Fos positive neurons. 5-HT is in red and c-Fos is in green. Yellow indicates co-localization of Fos and 5-HT. AL = antennal lobes, ICa = lateral calyx, and mCa = medial calyx, OL = optic lobe, and SOG = subesophageal ganglion. Brain in ventral orientation. Brain is in ventral orientation. All images are from the same brain, a) This image illustrates the whole brain with dual staining for Fos and 5-HT. The yellow staining shows serotonergic neurons that are active. Activity is located in the SOG, OL, and AL. Arrows point to neuronal clusters. The dashed box highlights both serotonergic and Fos processes, a closer view is represented in figure 3x.c Image is at 20x magnification, b) A closer view of the clusters to illustrate co-localization. Yellow indicates active serotonergic neurons. Arrows point to active neuronal clusters. Image is at 40x magnification, c) Zoomed view of Fos and 5-HT processes within the mushroom body calyces. Arrows highlight points of co-localization. Yellow (colocalization) indicates active 5-HT processes. Image is at 20x magnification. Ant brain dissected from an ant during aggression.
24


c-Fos staining. Using the method laid out previously, we were able to see distinct c-Fos staining (Figure 5). We see similar neuronal clusters and bilateral symmetry as in Figure 1, however there are distinct Fos neurons stained green (Figure 5). Those neurons stained red show presence of 5-HT and those colored yellow indicate co-localization of c-Fos and 5-FIT. Co-localization with Fos indicates activity within those neurons, specifically serotonergic activity (Figure 5). We see that there is 5-FIT activity within the AL, OL, SOG, and MB calyces. Within the AL we see unique 5-FIT and Fos neurons. Again, we see processes within the mushroom body calyces, but now we see Fos activation as well (Figure 5.c). These processes also have co-localization at points indicating serotonergic activity in the calyces (Figure 5.c).
Looking at Table 2, of the 51 neurons present the majority are serotonergic (91.49%). 50% of the neurons show the presence of c-Fos and 44.68% show co-localization. At the time of the staining 44.68% of the serotonergic neurons within this individual were active. All areas of the brain showed 5-HT activation (62.50% in the AL, 25% in the OL, and 46.67% in the SOG). Most of the serotonergic activity was in the AL and SOG (62.50% and 46.67% of neurons showing co-localization respectively). 100% of the neurons within the OL and SOG are serotonergic, but only 25% and 46.67% showed activity. Only the AL showed presence of Fos only neurons (Table 2).
25


Table 2. Number, type, and proportion of neurons within co-stained brain. All counts are based on Figure 3X. Neurons were counted as co-localized if the neuron contained any amount of yellow. Numbers were rounded to two decimal places.
Type of Neuron Total Number of Neurons Total c-Fos positive neurons T otal 5-HT positive Neurons Co- localized neurons %c- Fos % 5-HT % colocalized
AL 16 13 12 10 81.25 75.00 62.50
OL 16 4 16 4 25.00 100.00 25.00
SOG 15 7 15 7 46.67 100 46.67
Total 47 24 43 21 51.0 91.49 44.68
26


CHAPTER 4
DISCUSSION
Serotonergic staining. The technique employed in this paper provides an estimate for the number of 5-HT neurons within T. caespitum brains. It also shows where these neurons are likely located and the location of their activity. Serotonergic neurons were found in the AL, MB, SOG, and OL. Additionally, serotonergic processes were identified branching into the MB calyces. Both if. saltator (Hoyer et al. 2005) and C. japonicus (Tsuji et al. 2007) were found to have similar serotonin immunoreactivity. Both studies showed serotonergic processes and arborization within the MB calyces similar to Figure 2. However, the number of neuronal cells found in the AL (Fig. 1, Fig.3) differs from the single AL neuron found previously (Tsuji et al. 2007). The processes within the MB calyces appear to terminate in the lip of the calyx (Fig. 2). The termination of 5-HT processes in the lip of the MB calyx (Fig. 2) was also found in Tsuji et aids (2007) results. In ants, the lip of the MB calyces contains afferents from the antennal lobes and is associated with olfactory input (Gronenberg 2001, Fahrbach 2006). Therefore the 5-HT processes highlighted could be those afferents. If so, serotonin may influence higher order processing of olfactory information (Hoyer et al. 2005). Muscedere et al. (2012) blocked 5-HT in P. dentata and found that workers failed to locate pheromone trails. Without 5-HT they struggled to locate and follow olfactory information. This also suggests the importance of 5-HT in sensory integration. Since ants primarily rely on their antennae for sensory information, 5-HT provides a mechanism for how that information is processed within the MB. The SOG is responsible for control over an insects mouthparts, salivary glands and neck muscles. The presence of serotonin in the SOG suggests 5-HTs role in modulating mandible control. For T. caespitum this is quite important in regard to aggression against conspecifics. T. caespitum lock
27


mandibles during conflict with conspecifics so 5-HT is likely active within those neurons during aggression. We also see presence of 5-HT within the optic lobes and medulla of the optic lobes, which confirms others findings (Seid et al. 2008, Hoyer et al. 2005, Schafer and Bicker 1986). The serotonergic immunoreactivity in the medulla of the OL (Figure 1) is similar to the findings described by Hoyer et al (2005). However, most ants do not rely heavily on sight. The presence of 5-HT within the optic lobe could be a remnant of similar architecture found in other hymenopterans (Schafer and Bicker 1986) or the high number of serotonergic neurons is necessary for foragers outside the nest to process visual information to successfully collect food, subdue prey, and defend territory (Seid et al. 2008).
c-Fos staining. This technique shows the first example of Fos staining in an ant species. Additionally, it highlights the activity of 5-HT within a T. caespitum individual engaged in aggression. 5-HT neurons within the AL, OL, and SOG are shown to be active. We also see active serotonergic processes within the MB calyces. This activity within the antennal lobes and the lip of the MB calyces suggests that 5-HT is sent to the MB from the AL during aggression. Again, this supports 5-HT role in modulating higher order processing of olfactory information (Hoyer et al. 2005). We also see serotonergic activation in the SOG. This would correspond with the locking of mandibles that T. caespitum engage in during aggression. There was minimal 5-HT activation of the OL. Serotonin may not modulate visual information during aggression.
T. caespitum primarily relies on olfactory information thus it is not surprising that there is minimal activation in the OL. Our technique helps illustrate how 5-HT is being used during aggression.
Neuron count. We were able to obtain an estimate of 5-HT neurons using our technique. The maximum count of 5-HT neurons was 78. Honeybees have about 75 (Shurmann and Klemm
28


1984). Other have reported 130 to 200 in C. japonicus and C. herculeanus, respectively (Tsuji et al. 2007, Gronenberg 1996). Given this range our count seems feasible. Brain size is social insects is tied to social organization It is hypothesized that eusocial insects brains are smaller or miniaturized as social complexity increases. Individual ants cognition is limited, but as a whole their collective intelligence allows for complex group behaviors (Feinermen and Traniello 2016). The low number of 5-HT neurons helps support a limited cognitive ability within an individual ant. T.caespitum is able to perform a wide repertoire of behaviors using a limited number of neuronal cells.
Conceptual Model for 5-HT modulated aggression. Using the data from this technique I would like to propose a conceptual model for 5-HT modulated aggression in T. caespitum. Pretend you are an individual ant. You use a random walk to explore your colonies territorial space looking for resources. While walking you encounter other ants. You antentate them to determine their identity by reading their cuticular hydrocarbons and assessing the relative abundance of methyl-branched alkanes and n-alkenes (Sano et al. 2018). We know that when individuals interact with their nestmates 5-HT increases (Bubak et al. 2016, Hoover et al. 2015) so the interactions likely increase 5-HT production in the AL. The large number of 5-HT neurons found in the AL would support this (Fig.3). The 5-HT produced in the AL is sent to the MB lip along the processes found in Figure 5. From there the information is processed and the you recognize the other as nestmate. This memory slowly decays, but as you explore you encounter more nestmates refreshing the 5-HT in your brain. As you explore you begin to encounter other ants. These ants cuticular hydrocarbons to not match that of your nestmates. Since you have been in constant contact with your nestmates your 5-HT has been constantly refreshed. This constant refreshing of 5-HT by nestmate interaction primes the brain for aggression (Hoover et al. 2016). During
29


aggression we see a rise in brain titer levels of 5-HT (Bubak et al. 2016). The rise we see in brain titer levels of 5-HT during aggression could be explained by an increase in production of serotonin from the antennal lobes being shuttled to the MB (Fig.5). Once the 5-HT is integrated within the MB you decide to aggress. You lock mandibles with you opponent. This results in sustained serotonin release in the SOG to keep the mandibles locked with your adversary. (Fig.
5). The ants around you do the same since they share a similar brain state. After some time, a large portion of your colony is with you attacking the enemy. After some monoaminergic threshold is reached the fight ends and you follow the pheromone trail back to your nest.
Limitations. While this study is informative and descriptive of the pavement ant brain, the technique does have limitations. This data collected is based off the best examples from IHC; meaning only the images created that displayed detailed morphology and staining were chosen. Thus, all measurements and interpretations made are estimates. Using whole brains with IHC can make it difficult to achieve uniform orientation so brains in this study had varying orientations. This could have influenced interpretations of morphology and neuron counts. Some neurons tagged by the technique might not have been able to be counted due the orientation of the brain and the limitations of the microscope to penetrate the tissue. This reduces confidence in this techniques ability to establish a true neuron count. Also, the orientations of each brain may lead to inaccurate comparisons between experimental groups. Finally, since the images analyzed were selected as the best examples of the IHC technique, these brains come from a variety of individuals exposed to different behavioral contexts. It would be difficult to draw conclusions to a behavior or experimental setting.
All c-Fos measurements and estimates are based on a single brain. We were unable to replicate c-Fos staining. This limits the conclusions we can draw on the efficacy of our
30


technique. However, this might be due to the use of Formaldehyde instead of Para-formaldehyde (PFA). The antibodies used were described as working best with PFA as a fixative. Future studies that use this technique should try a variety of fixatives to determine which works best for Fos staining in their subjects.
Future Directions. This study shows that c-Fos can be used within ants to determine potential mechanisms for aggressive behavior in pavement ants. Additionally, once our c-Fos technique is perfected and replicated, we can use this technique to understand other behaviors in other species. The data collected helps build our conceptual model of collective behavior. By looking at an individuals brain we illustrate potential neural mechanisms that influence individual behavior. Those individual behaviors magnified to the whole colony could give insight into how collective decisions are made. This study highlights the proximate mechanism for the beginnings of complex, social behavior.
31


CHAPTER 5
REFERENCES
Alloway, T. M. (1979). Raiding behaviour of two species of slave-making ants, Harpagoxenus americanus (Emery) and Leptothorax duloticus wesson (Hymenoptera: Formicidae). Animal Behaviour, 27(PART 1), 202-210. https://doi.org/10.1016/0003-3472(79)90140-4
Bastle, R. M., Peartree, N. A., Goenaga, J., Hatch, K. N., Henricks, A., Scott, S., ...
Neisewander, J. L. (2016). Immediate early gene expression reveals interactions between social and nicotine rewards on brain activity in adolescent male rats. Behavioural Brain Research, 313, 244-254. https://doi.Org/10.1016/j.bbr.2016.07.024
Bubak, A. N., Yaeger, J. D. W., Renner, K. J., Swallow, J. G., Sc Greene, M. J. (2016).
Neuromodulation of nestmate recognition decisions by pavement ants. PLoS ONE, 77(11), 1-16. https://doi.org/10.1371/ioumal.pone.0166417
Cronin, A. L. (2015). Individual and Group Personalities Characterise Consensus Decision-Making in an Ant. Ethology, 121(1), 703-713. https://doi.Org/10.l 111/eth. 12386
Davidson, J. D., Sc Gordon, D. M. (2017). Spatial organization and interactions of harvester ants during foraging activity. Journal of The Royal Society Interface, 14. https ://doi. org/10.1098/rsif. 2017.0413
Day H.E., Kryskow E.M., Nyhuis T. J., Herlihy L., Campeau S. (2008). Conditioned fear inhibits
c-fos mRNA expression in the central extended amygdala. Brain Research, 1229, 137-46.
Dragunow, M., Sc Faull, R. (1989). The use of c-fos as a metabolic marker in neuronal pathway tracing. Journal of Neuroscience Methods, 29(3), 261-265. https://doi.org/10.1016/0165-0270(89)90150-7
Dragunow, M., Sc Robertson, H. A. (1988). Localization and induction of c-fos protein-like immunoreactive material in the nuclei of adult mammalian neurons. Brain Research,
440(2), 252-260. https://doi.org/10.1016/0006-8993(88)90993-6
Dyer, A. G. (2005). Honeybee (Apis mellifera) vision can discriminate between and recognise images of human faces. Journal of Experimental Biology, 208(24), 4709-4714. https://doi.org/10.1242/jeb.01929
Edwards, S. C., Sc Pratt, S. C. (2009). Rationality in collective decision-making by ant colonies.
Proceedings of the Royal Society B: Biological Sciences, 276(1613), 3655-3661. https://doi.org/10.1098/rspb.2009.0981
Fahrbach, S. E. (2006). Stmcture of the Mushroom Bodies of the Insect Brain. Annual Review of Entomology, 51(1), 209-232. https://doi.Org/10.l 146/annurev.ento.51.110104.150954
32


Gauthier, M. and Grunewald, B. (2012). Neurotransmitter Systems in the Honey Bee Brain: Functions in Learning and Memory. In: Galizia, C. G., Eisenhardt, D., Sc Giurfa, M., editors. Honey Bee Neurobiology and Behaviour. A Tribute to RandolfMenzel. Honeybee Neurobiology and Behavior. Dordrecht: Springer, https://doi.org/10.1007/978-94-007-2099-2
Galizia, C. G., Eisenhardt, D., Sc Giurfa, M. (2012). Honey Bee Neurobiology and Behaviour. A Tribute to RandolfMenzel. Honeybee Neurobiology and Behavior. https://doi.org/10.1007/978-94-007-2099-2
Ghosal, K., Naples, S. P., Rabe, A. R., Sc Killian, K. A. (2010). Agonistic behavior and electrical stimulation of the antennae induces fos-like protein expression in the male cricket brain.
Archives of Insect Biochemistry and Physiology, 74(1), 38-51. https ://doi.org/10.1002/arch.20360
Giraldo, Y. M., Patel, E., Gronenberg, W., Sc Traniello, J. F. A. (2013). Division of labor and structural plasticity in an extrinsic serotonergic mushroom body neuron in the ant Pheidole dentata. Neuroscience Letters, 534( 1), 107-111. https://doi.Org/10.1016/j.neulet.2012.l 1.057
Greene, M. J., Sc Gordon, D. M. (2007). Interaction rate informs harvester ant task decisions. Behavioral Ecology, 18(2), 451-455. https://doi.org/10.1093/beheco/arll05
Gronenberg, W. (1996). Neuroethology of ants. Naturwissenschaften, <83(1), 15-27. https://doi.org/10.1007/sQQ 1140050240
Gronenberg, W., 2001. Subdivisions of the hymenopteran mushroom body calyces by their afferent supply. Journal of Comparative Neurology, 436, 474-489.
Guzowski, J. F., Setlow, B., Wagner, E. K., Sc McGaugh, J. L. (2001). Experience-dependent gene expression in the rat hippocampus after spatial learning: a comparison of the immediate-early genes .Arc, c-fos, and zif268. The Journal of Neuroscience: The Official Journal of the Society for Neuroscience, 27(14), 5089-5098. https://doi.org/21/14/5089 [pii]
Hoover, K. M., Bubak, A. N., Law, I. J., Yaeger, J. D. W., Renner, K. J., Swallow, J. G., Sc
Greene, M. J. (2016). The organization of societal conflicts by pavement ants Tetramorium caespitum: An agent-based model of amine-mediated decision making. Current Zoology. https ://doi.org/10.1093/cz/zow041
Hoyer, S. C., Liebig, J., Sc Rossler, W. (2005). Biogenic amines in the ponerine ant
Harpegnathos saltator: Serotonin and dopamine immunoreactivity in the brain. Arthropod Structure and Development. https://doi.Org/10.1016/j.asd.2005.03.003
33


Hoyle G (1985): Generation of motor activity and control of behaviour: the role of the neuromodulator octopamine and the orchestration hypothesis; in Kerkut GA, Gilbert L (eds): Comparative Insect Physiology, Biochemistry and Pharmacology. Toronto, Pergamon Press, vol 5, 607-621
Hu, L., Vander Meer, R. K., Porter, S. D., Sc Chen, L. (2017). Cuticular Hydrocarbon Profiles Differentiate Tropical Fire Ant Populations (Solenopsis geminata, Hymenoptera: Formicidae). Chemistry and Biodiversity, 77(11), 1-8. https://doi.org/10.1002/cbdv.20170Q192
Iyer, L. M., Aravind, L., Coon, S. L., Klein, D. C., Sc Koonin, E. V. (2004). Evolution of cell-cell signaling in animals: Did late horizontal gene transfer from bacteria have a role? Trends in Genetics, 20(1), 292-299. https://doi.Org/10.1016/j.tig.2004.05.007
Kaisho, T., Nagai, H., Asakawa, T., Suzuki, N., Fujita, H., Matsumiya, K., ... Takekawa, S.
(2017). Effects of peripheral administration of a Neuromedin U receptor 2-selective agonist on food intake and body weight in obese mice. International Journal of Obesity, 41(12), 1790-1797. https://doi.org/10.1038/ijo.2017.176
Kamhi, J. F., Nunn, K., Robson, S. K. A., Sc Traniello, J. F. A. (2015). Polymorphism and division of labour in a socially complex ant: neuromodulation of aggression in the Australian weaver ant, Oecophylla smaragdina. Proceedings of the Royal Society B: Biological Sciences, 252(1811), 20150704. https://doi.org/10.1098/rspb.2015.0704
Kamhi, J. F., Sc Traniello, J. F. A. (2013). Biogenic amines and collective organization in a
superorganism: Neuromodulation of social behavior in ants. Brain, Behavior and Evolution, 82(4), 220-236. https://doi.org/10.1159/000356Q91
Kang, K., Park, S., Kim, Y. S., Lee, S., Sc Back, K. (2009). Biosynthesis and biotechnological
production of serotonin derivatives. Applied Microbiology and Biotechnology, 53(1), 27-34.
Lioni, A, Sc Deneubourg, J. L. (2004). Collective decision through self-assembling. Naturwissenschaften, 91(5), 237-241. https://doi.org/10.1007/s00114-004-0519-7
Marshall, J. A. R., Bogacz, R., Dornhaus, A., Planque, R., Kovacs, T., Sc Franks, N. R. (2009). On optimal decision-making in brains and social insect colonies. Journal of The Royal Society Interface, 6(40), 1065-1074. https://doi.org/10.1098/rsif.2008.0511
Muscedere, M. L., Cahan, S. H., Helms, K. R., Sc Traniello, J. F. A. (2016). Behavioral Ecology Geographic and life-history variation in ant queen colony founding correlates with brain amine levels. https://doi.org/10.1093/beheco/anT52
34


Muscedere, M. L., Johnson, N., Gillis, B. C., Kamhi, J. F., Sc Traniello, J. F. A. (2012).
Serotonin modulates worker responsiveness to trail pheromone in the ant Pheidole dentata. Journal of Comparative Physiology A: Neuroethology, Sensory, Neural, and Behavioral Physiology. https://doi.org/10.1007/sQ0359-011-0701-2
Muscedere, M. L., Sc Traniello, J. F. A. (2012). Division of labor in the hyperdiverse ant genus Pheidole is associated with distinct subcaste- and age-related patterns of worker brain organization. PLoS ONE, 7(2). https://doi.org/10.1371/joumal.pone.0031618
Neumaier, J. F., Sexton, T. J., Yracheta, J., Diaz, A. M., Sc Brownfield, M. (2001). Localization of 5-HT7 receptors in rat brain by immunocytochemistry, in situ hybridization, and agonist stimulated cFos expression. Journal of Chemical Neuroanatomy, 27(1), 63-73. https ://doi.org/10.1016/S0891 -0618(00)00092-2
Penick, C. A, Brent, C. S., Dolezal, K., Sc Liebig, J. (2014). Neurohormonal changes associated with ritualized combat and the formation of a reproductive hierarchy in the ant
Harpegnathos saltator. The Journal of Experimental Biology, 27 7(Pt 9). https://doi.org/10.1242/ieb.Q98301
Pratt, S. C. (2005). Quomm sensing by encounter rates in the ant Temnothorax albipennis.
Behavioral Ecology, 16(2), 488-496. https://doi.org/10.1093/beheco/ari020
Pratt, S. C., Mallon, E. B., Sumpter, D. J. T., Sc Franks, N. R. (2002). Quomm sensing, recmitment, and collective decision-making during colony emigration by the ant
Leptothorax albipennis. Behavioral Ecology and Sociobiology, 52(2), 117-127. https://doi.org/10.1007/s00265-002-0487-x
Pruitt, J. N., Sc Riechert, S. E. (2011). How within-group behavioural variation and task
efficiency enhance fitness in a social group. Proceedings of the Royal Society B: Biological Sciences, 275(1709), 1209-1215. https://doi.org/10.1098/rspb.2010.1700
Renucci, M., Tirard, A, Charpin, P., Augier, R., Sc Strambi, A. (2000). c-Fos-related antigens in the central nervous system of an insect, Acheta domesticus. Archives of Insect Biochemistry and Physiology, 45(4), 139-48. https://doi.org/10.1002/1520-6327(200012)45:4<139::AID-ARCH1>3.0.CO;2-N
Rossler, W. and Groh, C. (2012). Plasticity of Synaptic Microcircuits in the Mushroom-Body Calyx of the Honey Bee. In: Galizia, C. G., Eisenhardt, D., Sc Giurfa, M., editors. Honey Bee Neurobiology and Behaviour. A Tribute to RandolfMenzel. Honeybee Neurobiology and Behavior. Dordrecht: Springer. https://doi.org/10.1007/978-94-007-2Q99-2
Sano, K., Bannon, N., and Greene, M.J. 2018. Pavement ant workers (Tetramorium caespitum) assess cues coded in cuticular hydrocarbons to recognize conspecific and heterospecific non-nestmate ants. Journal of Insect Behavior, DOI: 10.1007/sl0905-017-9659-4.
35


Sasaki, T., & Pratt, S. C. (2013). Ants learn to rely on more informative attributes during decision-making. Biology Letters, 9(6), 20130667-20130667. https://doi.org/10.1098/rsbl.2013.0667
Schafer, S., Bicker, G., (1986). Common projection areas of 5-HT-like and GABA-like
immunoreactive fibers and the visual-system of the honeybee. Brain Research, 380, 368-370
Schurmann, F.W., Klemm, N., (1984). Serotonin-immunoreactive neurons in the brain of the honeybee. The Journal of Comparative Neurology, 225, 570-580.
Seid, M. A., Goode, K., Li, C., & Traniello, J. F. A. (2008). Age- and subcaste-related patterns of serotonergic immunoreactivity in the optic lobes of the ant Pheidole dentata. Developmental Neurobiology, <5<8(11), 1325-1333. https://doi.org/10.1002/dneu.20663
Smith, A. R., Muscedere, M. L., Seid, M. A, Traniello, J. F. A., & Hughes, W. O. H. (2013). Biogenic amines are associated with worker task but not patriline in the leaf-cutting ant
Acromyrmex echinatior. Journal of Comparative Physiology A: Neuroethology, Sensory, Neural, and Behavioral Physiology, https://doi.org/10.1007/s00359-013-0854-2
Sumpter, D. J. T. (2006). The principles of collective animal behaviour. Philosophical Transactions of the Royal Society B: Biological Sciences, 367(1465), 5-22. https://doi.org/10.1098/rstb.2005.1733
Tsuji, E., Aonuma, H., Yokohari, F., & Nishikawa, M. (n.d.). Serotonin-immunoreactive
Neurons in the .Antennal Sensory System of the Brain in the Carpenter .Ant, Camponotus japonicus. https://doi.org/10.2108/zsj.24.836
von Frisch K. (1914). Der Farbensinn und Formensinn der Biene. Zool Jahrb Abt AUg Zool Physiol Tiere, 37,1-238.
Vleugels, R., Verlinden, H., & Broeck, J. V. (2015). Serotonin, serotonin receptors and their actions in insects. Neurotransmitter, 2(314): doi: 10.14800/nt.314.
Wada-Katsumata, A, Yamaoka, R., & Aonuma, H. (2011). Social interactions influence
dopamine and octopamine homeostasis in the brain of the ant Formica japonica. Journal of Experimental Biology, 274(10), 1707-1713. https://doi.org/10.1242/jeb.051565
36


Full Text

PAGE 1

i EXPLORATION OF SEROTONIN DISTRIBUTION AND ACTIVITY WITHIN THE PAVEMENT ANT BRAIN ( TETRAMORIUM C A E SPITUM ) by WILLIAM SCHUMANN B.S. Pacific University 2 008 A thesis submitted to the Faculty of the Graduate School of the University of Colorado in partial fulfilment of the requirements for the degree of Master of Science Biology Program 2018

PAGE 2

ii This thesis for the Master of Sc ience degree by William Paul Schumann has been approved for the Biology Program b y Michael Greene, Chair John Swallow Benjamin Greenwood Sondra Bland Date: May 12 2018

PAGE 3

iii Schumann, William (M.S. Biology Program ) Exploration of Serotonin Distribut ion and Activity within the Pavement Ant Brain ( Te tram orium caespitum ) Thesis directed by Associa te Professor Michael Greene and Professor John Swallow ABSTRACT Ants have miniaturized brains, yet they exhibit surprisingly complex behaviors. In social insect colonies, individuals gather information, integrate it, and compare that information to an in herent set of rules to make behavioral decisions. Each individual decision can lead to complex, self organized behaviors. However, l ittle is known about the proximate mechanisms behind these collective behaviors. The study of neurotransmitter monoamines, s uch as serotonin, provide a possible explanation for such complex behaviors. Serotonin is associated with reproductive dominance, colony foundation, aggression, trophallaxis, behavioral development, division of labor, repertoire expansion, and nestmate rec ognition in ants. This study used pavement ants ( Tetramorium caespitum ) as a model species to explore the distribution and activity of serotonin within the neural architecture of the ant brain. Ants were exposed to a variety of contexts: social interaction, aggression, food excitement, and antenectomy. After exposure whole bra ins were dissected and underwent immunohistochemistry (IHC) to stain for the monoamine serotonin and the genetic marker for neuronal activity, c Fos. Serotonin immunoreactivity was found in the antennal lobes (AL), subesophageal ganglion (SOG), optic lobes (OL), mushroom bodies (MB), and the MB calyces. Serotonergic processes were seen in the MB calyces, terminating in the lip of the calyx. Neuronal symmetry was observed in the AL, SOG, and OL. A maximum estimate of 78 serotonergic neurons were stained. Thi s study provides further information for the serotonergic architecture within the ant brain. Fos colocalization with serotonin was seen in the AL, SOG, OL, and MB calyces in an ant engaged in aggression. This suggest that serotonin is

PAGE 4

iv active in these locat ions during this behavior. However, only one brain showed positive staining for Fos. The location and activation of serotonergic neurons in the AL and MB calyces suggest that serotonin is released from the AL and shuttled to the MB cal y rocessed and a decision to aggress is made. If the Fos technique provided in this study can be replicated, it offers a new method to better understand the underlying mechanisms of behavior in ants. The form and content of this abstract are approved. I reco mmend publication. Approved by: Mic hael Greene John Swallow

PAGE 5

v TABLE OF CONTENTS I. INTRODUCTION 1 II. METHODS 1 5 III. RESULTS 19 IV. DISCUSSION 27 V. REFERENCES 3 2

PAGE 6

1 CHAPTER 1 INTRODUCTION Ant colonies are regulated as non hierarchal, distributed system where individual R ather i ndividuals asses local informational cues, integrate them in their simple miniaturized brains, and compare that information to an inherent set of rules. Often these cues are based on the interaction rates they have with their nestmates (Greene and Gordon 2007 Davi dson and Gordon 2017, Hoover et al 201 6 Bubak et al 201 6 Pratt 2005 ) the density of pheromone trails (Sumpter 2005 Muscedere et al 2012 ) and other olfactory information like cuticular hydrocarbons ( Hu et al 2017, Sano et al 2018 ) Ants share this information locally, and interact with a small, limited network of nestmates. The network is limited to a small fraction of the colony because o ne ant could not possibly interact with every other individual in the colony. However, these small network s extend to other small networks causing information to spread and a collective behavior to be expressed Therefore, individuals make decisions given their limited information (Sas a ki and Pratt 2013, Edwards and Pratt 2009). The many decisions made by individuals lead to the emergence of colony behaviors including foraging, nest construction, brood care, colony defense, and colony maintenance. these decisions made by individuals (Pruitt and Riechert 2011). Out of t he repeated interactions of individuals a system emerges: self organization to a collective decision. Complex emergent behavior can arise from self organization stemming from a simple set of rules (Sumpter 2005). These collective behaviors are frequently m ade via consensus from individuals and th ese collective decisions shape the fitness of the colony (Cronin 2015). Consensus decisions provide a way for individuals to coordinate information and, once a

PAGE 7

2 threshold is reached, elicit a group level behavior (Cr onin 2015). The colony is, in essence, a society that functions as an information processing unit distributing its cognitive tasks to individual ants (Pratt et al interactions between a network network of interactions among individual ants (Marshall et al 2009, Edwards and Pratt 2009). Ants have miniaturized brains, less than 1/1000 the size of a honey bee, yet collectively they exhibit amazingly complex behaviors including farming fungus, raising aphids for honeydew, stealing non nestmate workers (Alloway 1979) building structures with their own bodies (Lioni and Deneubourg 2004) and warfare (Hoover et al 2016) Relative to th eir brain volume, they possess enlarged antennal lobes (odor processing) and mushroom bodies (cognition, learning, and memory) (Fahrbach 2006). Using these enlarged structures, individual ants interpret local stimulus using intrinsic rules to make complex behavioral decisions. These decisions lead to the collective behavior of the colony. Within Hymenoptera (wasps, bees, ants, and sawflies), we see that brain structures associated sensory collection and integration are enlarged Apis mellifera have large optic lobes (OL) that improve vision which allows the m to among other things discriminate colors (von Frisch 1914) and even human faces (Dyer et al 2005). However, most ants see a reduction in the size of their optic lobes because the y mainly rely on olfaction (Gronenberg 1996). Since primary sense organ is their antennae, the ant brain has antennal lobes (AL) that are larger relative to body and brain size, and more complex than in other social Hymentoptera (Gronenberg 1996). Honey bee possess an estimated 200 some of which are larger and more complex than in other insects. Glomeruli are associated with odor detection and larger and more complex glomeruli suggest the

PAGE 8

3 importance of olfaction for social behavior in ants (Gronenberg 1996). Lastly, m ost social Hymenoptera have enlarged mushroom bodies (MB) a region responsible for higher order processing like sensory integration, learning, memory, and spatial orientation all functions important to the social life of ants and bees (Gronenberg 1996). Within ants we see larger MB when compared to other insects and other Hymentopterans (Gronenber g 1996). In ants and bees the mushroom bodies are divided into two calyces: the medial and lateral calyx. The role of the MB calyces in bees and ants is different from other insects as well. While other insects primarily use the MB calyces for odor processing, bee and ant MB calyces are a center for multimodal sensory integration ( Rossler and Groh 2012). For example, the MB in A. mellifera work in tandem with the AL and the subesopheagel ganglion (SOG) during olfactory learning (Gauthier and Grunewald 2012) These three regions appear to play important roles within social insects an d their behavior Social insects, like ants and other Hymenoptera have relatively simple brains yet they can exhibit extremely complex behaviors. Hoyle (1985) established the orchestration hypothesis as a means to explain the coordination of complex behaviors by neurotransmitters. He discovered that direct manipulation of monoamines could elicit specific behaviors and that monoamines worked on the neural circuitry of organisms. With the development of the orchestration hypothesis, researchers had a new possible mechanism to study, monoamines. Octopamine (OA), dopamine ( DA), and serotonin (5 HT) are commonly studied to determine the animergic effects on behavior (Kam h i and Traniello 2013 Hoyer et al 2005, Tsuji et al 2007, Muscedere et al 2012, Muscedere et al 2012, Muscedere et al 2016, Bubak et al 2016, Hoover et al 2015 ). Ant researchers began to study these monoamines to understand the mechanisms underlying social behaviors (Kam h i and Traniello 2013). Exploring monoamines offers novel explanations for the

PAGE 9

4 modulation of complex social behaviors. By examining neurochemistry and neural architecture we can determine the underpinnings of social behavior, colony level division of labor, and collective intelligence (Kamhi and Traniello 2013). Octopamine ( OA ) DA, and 5 HT have been implicated in a wide range of ant behaviors including colony foundation, interspecies and predatory aggression, learning, development, trophallaxis, and nestmate recognition ( Kamhi and Taniello 2013 Table 1 ). Kamhi et al ( 2015 ) showed that OA is instrumental in aggression in the Australian weaver ant, Oecophylla smaragdina. This species has different worker castes; large majors engage in aggressive territorial defense while minors care for brood and collect food. For example, Kamhi et al (2015 ) found that majors contained higher brain titer levels of OA. To see if OA influenced aggression they manipulated OA levels within minors and saw an increase in aggressive behaviors. Smith et al (2013) found that in the ant, Achromyrmex echinatior DA, OA, and 5 HT brain levels differed based on castes suggesting that these monoamines influence worker specialization In a recent study Wada Katsumata et al (2011) show that DA and OA are linked to the social behaviors of grooming and trophallaxis in Formica japonica In this study ants where starved while DA and OA levels were monitored. Starving ants displayed low levels of DA, which was rescued after trophallaxis. Ants were also isolated from social interaction. This increased OA levels. When introduced t o nestmates these ants increased their trophallaxis duration, allogrooming, and self grooming behaviors suggesting that high levels of OA influences social behaviors. Queens from the species, Veromessor pergandei have varying colony foundation strategies, founding a colony singly, or working cooperatively with other queens. Muscedere et al (2016) examined the role of DA, OA, and 5 HT in these different strategies. Queens that found colonies together often engage in aggressive conflict with each

PAGE 10

5 other to establish dominance after worker eclosion Early in colony foundation, young queens that founded singly had heightened brain levels of 5 HT and queens that founded together had heightened levels after workers eclos ed. This suggests that 5 HT modulates aggression during colony foundation (Musceder e et al 2016). Serotonin was found to influence foraging behaviors in Pheidole dentata (Muscedere et al 2012). This study pharmacologically decreased serotonin in individuals. These individuals were less likely to follow pheromone trails and if oriented to trails only followed for short distances compared to control ants. These studies illustrate just how many beha viors are modulated by monoamines. Although, m onoamines have been shown to have many important functions in re gulating social insect behavior little is known about the underlying mechanisms that differentiate how monoamines modulate these complex behaviors Most of our current methods for studying monoaminergic effects on behavior rely on manipulations of whole brain levels which offer much insight into how levels of monoamines can affect behaviors, but lack resolution at the level of brain archite cture ( Muscedere et al 2016, Bubak et al 201 6 Penick et al 2014, Smith et al 2013, Kamhi et al 2015, Muscedere et al 2012 Wada Katsumata et al 2011 ). These studies typically employ High Powered Liquid Chromatography (HPLC) to measure the monoamine content within a whole brain. They also incorporate pharmacological manipulations to see how increases and decrease s in whole brain levels of monoamines affect behavior s. These studies are limited in how they explore the possible mechanisms behind a given behavior. Pharmacological manipulations flood the brain with monoamines, higher levels than those found naturally. T his blunt approach makes it easy to elicit a behavioral response and correlate a monoamine with a given behavior. However, this does little to show how these monoamines work within the brain to generate the behavior. If a deluge of monoamines floods the br ain, how can you determine

PAGE 11

6 what, where, when, and how these molecules modulate behavior. One can only make a general assumption that having high levels of monoamines causes a change in behavior. Unfortunately, these types of studies only look at the whole brain titers of the monoamine s rather than looking at the neural architecture or the location of the monoamine within the brain. How then do we differentiate the specific effects of each monoamine? Do these monoamines work in concert or are they antagonist ic? Does one modulate the others? While we have determined that these monoamines influence these specific behaviors, it remains unclear how exactly monoamines modulate behaviors (Hoyer et al 2005). As stated above there are limitations to overriding the brain with monoamines. To gain better insight into the relevant cells and structures involved in be havior, I used immunohistochemistry (IHC) This tool allows for the exploration of possible monoamine rgic architecture of the brain to provide insights into the mechanisms that contribute to the generation of behavior (Giraldo et al 2013, Hoyer et al 2005, Smith et al 2013, Kamhi et al 2015). IHC is used commonly to explore social insect brain archite cture (Hoyer et al 2005, Galizia et al 2012, Gronenberg 1996, Giraldo et al 2013 ). Research often focuses on monoaminergic systems within the brain (Hoyer et al 2005, Tsuji et al 2007 Giraldo et al 2013 ) By understanding the architecture within the brain, these studies link the systems within the brain and its architecture to specific behaviors including aggression (Hoyer et al 2005), division of labor and repitoire expansion (Giraldo et al 2013, Muscedere and Traniello 2012, Galizia et al 2012), and sensory integration (Galizia et al 2012). Immunohistochemistry is a widely used technique to study the distribution and activity of monoamines. It involves detection of specific epitopes in the cells or tissues of interest using antibodies that bind to those specific epitopes. Additionally, fluorescence proteins are added for

PAGE 12

7 better resolution of the target molecules. Studies that use this technique can pinpoint neurons and neuronal pathways associate d with specific behaviors such as how 5 HT regulates division of labor in Pheidole dentata (Giraldo et al 2013), or the role DA and 5 HT play in intraspecific aggression in Harpegnathos saltator (Hoyer et al 2005). Giraldo et al (2013) used IHC to determine structural differences between castes as the y age as a possible explanation for behavioral repertoire expansion. Using IHC they were able to identify serotonergic neurons and their arborization within the mushroom bodies for all castes Giraldo et al (2013) discovered that only the major castes displayed structural changes as they matured as well as the majors had significantly more complex arborization within the mushroom bodies. They concluded that these structural differences were a possible explanation for the differences in caste behavior. Hoyer et al (2005) investigated the serotonergic and dopaminergic neuronal systems as they relate to intraspecies aggression and to dete rmine any differences in neural anatomy between castes and sex by employing dual staining IHC They hoped to find structural differences between these monoaminergic systems during aggression and between castes and sex The study was able to describe the serotonergic and dopaminergic systems within the brain, but they could not find any anatomical changes during aggression. However, they did find that males had smaller brain volumes proportionally to females that resulted in reduced size of the mushroom bodies. Additionally, the two female castes possessed more seroton ergic processes while the males displayed higher dopaminergic processes. These differences could explain the differences in behavioral repertoire between caste and sex (Hoyer et al 2005). Again, this study was able to describe monoaminergic neural anatomy to better understand behavior. These studies use this technique to gives us a better picture about the mechanisms of ant behavior s (Hoyer et al 2005,

PAGE 13

8 Giraldo et al 2013). Using this technique can give a detailed picture of the origin, location, and mechanism of a behavior. Since IHC has a variety of applications that extend to neuronal tissues, it becomes an excellent tool when trying to determine the mechanism beh ind specific behaviors (Hoyer et al 2005, Tsuji et al 2007). This technique allows you to visualize the tissue of interest to locate particular molecules of interest, in this case monoamines. Hoyer et al (2005) used IHC to determine if the distribution of DA and 5 HT was different between castes and sexes in H. saltator since brain morphology did not change among these groups. The researchers found that there was deep innervation of serotoninergic neurons within mushroom bodies (responsible for cognition, learning, and memory) with serotonin neurons making connections be tween the mushroom bodies and antennal lobes (responsible for olfaction). Additionally, they were able to show that sterile workers and males had different distributions of serotonergic processes as minergic process es were also different from other castes. Hoyer et al (2005) showed that different brain architecture exists in ants based on behavioral repertoire. They illustrated that monoamines must work differently within castes and sex. This is a po ssible explanation for the difference of behaviors among different castes of workers and between sexes. Tsuji et al. (2007) was able to provide the complete serotonergic network found in the antennal lobes of C. japonicas This network could be used to det ermine how serotonin can orchestrate the social behaviors of C.japonicas In both these studies, IHC provided a detailed map of the location of these monoamines in the brain. Knowing the location of these monoamines informs us of its function. If monoamines are in a brain region responsible for olfaction, they are likely responsible for modulating behaviors requiring olfaction

PAGE 14

9 and chemosensation. Since each region of the brain is responsible for different functions, this technique shows the possible mechanism that monoamines employ to elicit behavior. IHC provides a map show ing regions where monoamines are present. The information gathered from these maps can help us determine how these monoamines work in the brain and what kind of behaviors they may modulate (Hoyer et al 2005, Tsuji et al 2007). IHC gives the exact location of these monoamines during specific behaviors (Hoyer et al 2005). Sin ce the function of each brain region is known, we can use the location of these monoamines within those regions to explain the mechanisms behind behaviors like aggression and social interaction. But the method is limited since behavior relies on the firing of action potentials within neurons, which is difficult to detect with IHC. To address these concerns with IHC, the methods proposed in th is paper will incorporate the early activator gene, c Fos. c Fos can shows us the neuronal activity within the brai n and help to pinpoint areas that are active during a specific behavior. This can illuminate possible mechanisms including which serotonergic neurons are active during specific behavioral contexts c Fos is an early activator gene that is expressed when an action potential fires and has become a marker for neuronal activity (Day et al 2008, Dragunow and Faull 1989). c Fos is visualized in IHC by targeting the proteins expressed by c Fos. Dragunow and Robertson (1988) found that c Fos was expressed in recently activated neurons in the rat brain and began using it as a high resolution marker for synaptic pathways in the mammalian brain (Dragunow and Faull 1989). C Fos became used in vertebrates to understand the neural mechanisms and pathways behind behaviors (Guzowski et al 2001, Shu 2002, Neumaier et al 2001, Bastle et al 2016). Bastle et al. (2016) used c Fos to determine an interaction between the social and nicotine reward system in the brains of adult male rats. Additionally, Shu (2002) used cFos to explain the mechanism of

PAGE 15

10 learning in rats. C Fos allowed Neumairer et al (2001) to visualize the location 5 HT7 receptors in rat brains which provided insight into their role in the circadian rhythm and potential enhancement of se rotonergic pharmaceuticals. As these studies show, c Fos is a powerful tool to explain neural mechanisms in animals. c Fos has been important tool for the understanding of neural mechanisms of behavior. Rats learning to l ocate platforms submerged under water saw increased expression of c Fos RNA in the hippocampus, entorhinal cortex, and visual cortex when compared to controls. Additionally the act of learning the trail increased the level of c Fos RNA expression in the dorsal hippocampi when compared t o controls. These data suggest that Fos is expressed during spatial learning and that the regions showing Fos activity are responsible for the behavior (Guzowksi et al 2001). To determine the efficacy of a specific neuropeptide thought to regulate hunger, NMUR2, c Fos was used to show regions of Fos expression when mice were given a NMUR2 agonist. Fos was measured after the agonist, NMU 7005, was given to obese and fasted mice. Obese mice saw increased Fos expression in the lateral part of the hypothalamus when given the agonist when compared to controls given a saline solution. While fasted mice showed increased expression in the medial part of the hypothalamus compared to obese mice. Finally, obese mice given the agonist had increased Fos expression in th e medulla oblongata. Again, this study shows where this drug is active providing a possible mechanism for the agonist (Kaisho et al 2017). c Fos provides insight into the mechanism for a given behavior. It pinpoints the regions that express it which highl ights where the behavior is likely occurring. While this technique is widely used with vertebrate models, we rarely see it used in invertebrates (Renucci et al 2000, Ghosal et al 2010). There is little reason not to use c Fos since studies have shown that invertebrates express c Fos and c Fos related antigens (Ghosal et

PAGE 16

11 al 2010, Renucci et al 2000). Ghosal et al (2010) measure d the immunoreactivity for c Fos and Fos related antigens (FRA in male crickets using IHC. They identified specific brain regions that expressed c Fos and FRA proteins during aggression. They found FRA and Fos immunoreactivity in the ventromedial region of the brain, the deutocerebrum, and among the Kenyon cells of the cortex of the MB. Additionally, Fos was l ocated in the nuclei of brain cells. Renucci et al cricket brain. They saw staining within the nuclei of neurons of female crickets confirming the the insect brain. Still many invertebrate studies do not take advantage of this tool to understand the neural mechanisms of behavior. c Fos can bring a higher resolution of the neural pathways behind behaviors to insects especially ants. As shown above c F os has aided in understanding complex behaviors in rats. We can use this tool to understand the complex, collective behaviors of ants. As mentioned previously, m onoamines are of growing interest as a possible explanation for ant collective behavior ( Kamhi and Traniello 2013). This is especially relevant since these neurotransmitters are highly conserved across taxa (Iyer et al. 2004, Kang et al. 2009). Marshall et al (2009) points out that neural mechanisms behind primate decision making bear a striking s imilarity to the self organized decisions made by ant colonies. If ant collective behavior can inform how we understand how our neurons behave during the decision making process, then perhaps learning how monoamines drive social behaviors in ants could ex plain role these monoamines play our own social collective behavior. In this way we could create a foundation Studies investigation these monoamines typically look at serotonin, dopamine, and octopamine when

PAGE 17

12 exploring the possible neural mechanisms behind collective behavior ( Kamhi and Traniello 2013). For ease of analysis and technique this study focused on a single monoamine, serotonin (5 HT). Table 1. Serotonin and the social behaviors associated with serotonin Behavior Monoamine 5 HT Reproductive dominance and colony foundation S. invicta [Boulay et al., 2001]; H. saltator [Hoyer et al., 2005]; S. peetersi [Cuvillier Hot and Lenoir, 2006]; F. japonica [Aonuma and Watanabe, 2012a] Subcaste related division of labor P. dentata [Giraldo et al., 2013; Giraldo and Traniello, unpubl. obs.]; A. echinatior [Smith et al., 2013] Worker behavioral development, repertoire expansion, and temporal polyethism P. dentata [Seid and Traniello, 2005; Seid et al., 2008; Muscedere et al., 2012; Giraldo et al., 2013]; S. peetersi [Cuvillier Hot and Lenoir, 2006]; F. polyctena [Wnuk et al., 2010] Social food flow (trophallaxis) C. mus [Falibene et al., 2012]; P. dentata [Muscedere et al., 2013] Aggression F. japonica [Aonuma and Watanabe, 2012]; P. dentata [Giraldo et al., unpubl. obs.] T. caespitum [Hoover et al., 2015; Bubak et al. 2016] Nestmate recognition C. fellah [Boulay et al., 2000]; O. smaragdina [Kamhi and Traniello, unpubl. obs.] T. caespitum [Hoover et al., 2015; Bubak et al. 2016]

PAGE 18

13 Table modified from Kamhi and Traniello 2013. Serotonin is an intracellular signaling molecule that can be found in all organisms with a central nervous system (Vleugels et al 2015). This molecule is so widely conserved across taxa that it is found in plants and fungi (Kang et al. 2009). 5 HT is derived from tryptophan and is associated with the regulation of aggression, feeding/appetite, learning, memory, and social behaviors in insects (Vleugels et al. 2015). In ants, this monoamine has a wide variety of effects on their social behaviors. Studies have shown that 5 HT plays a role in colony foundation, interspecies and predatory aggression, learning, development, trophallaxis, and nestmate recognition ( Kamhi and Traniello 2013 Table 1 ) Table 1 displays the species and papers that show a link between 5 HT and a specific social behavior. While this research shows that 5 HT is responsible for a long list of behaviors, we do not yet know how 5 HT modulates each of these behaviors. Thus, we need techniques like IHC to determine the mechanisms by which 5 HT influences specific behavior. This study will attempt to use IHC to establish 5 modulating aggression in pavement ants. Pavement ants ( Tetramorium caespitum ) are a ubiquitous invasive species found throughout the urban environment. They prefer to live under slabs, or flat rocks, hence the name pavement ants. These qualities make them an easy subject to obtain and stu dy. In addition, pavement ants display a very conspicuous collective behavior, war. Pavement ants engage in conflict with other species of ants as well as conspecifics. Wars with other species are fought to the death, but not when pavement ants encounter n on nestmate conspecifics. Pavement ant war is ritualized with little to no causalities incurred by both colonies. Individual ants will use their o out of their te rritory. Wars will last for hours until a victor is decided. Recent interactions with

PAGE 19

14 nestmates and the frequency of those interactions influence an individual to decide to fight when encountering a non nestmate ( Hoover et al 2016, Bubak et al 201 6 ). These interactions between nest mates cause an increase in 5 HT and OA levels which prime the brain for aggressive confrontation with non nestmates (Bubak et al 201 6 Hoover et al 2016). Each interaction spikes OA and 5 HT levels in the brain after whic h the levels of monoamines start to decline. So repeated interactions are necessary to sustain monoamines at their threshold to encourage a decision to fight (Hoover et al 2016). It is unclear what levels of monoamines are necessary for a decision to figh t nor what the rate of decline of monoam ines is. The methods of this paper were developed to better understand the mechanism s behind this decision to fight. The methods in this paper were developed to further our understanding of the neural architecture a nd to create a technique in which we capture the neural activity of pavements ant based on different behavioral concepts. With these methods I will describe the serotoninergic architecture in pavement ants, show how this method can be used to measure neuro nal activity, and provide an example on how these methods can be employed to make behavioral comparisons

PAGE 20

15 CHAPTER 2 METHODS Material s Animals : Pavement ants were collected in Denver, CO by aspirating individuals from bait trails. Ants were baited with peanut butter. Ants were housed in aspirator tubes with grass to provide a level of humidity. Ants were housed in these containers for 24hrs prior to experimentation and sacrifice. Special Equipment : The following lists contains all the equipment used to conduct the experiment. The specificity of the list is to account for all possible contributions to the images generated by IHC. H orizontal shaker (Belly Dancer, Stovall Life Science, Gree nsboro, NC USA). Cold room horizontal shaker ( 2314FS Lab Rotator Stirrer Shaker, Fisher Scientific, Waltham, MA USA). Sterilized 48 well cell culture plates (Greiner Bio One Kremsmnster, Austria ) Cold room (Environmental Growth Chambers, Chagrin Falls, OH USA) For dissections, cold plate ( CaterWare Round Aluminum Cooling Plate 11 inch x 1 3/4 inch TableCraft, Gurnee, IL USA) Brains were imaged using Ziess LSM 700 confocal microscope. Some images were imaged on a 3I Marianas inverted Spinning Disk confocal microscope Chemicals The following list contains all chemicals used in the immohistochemical reactions. I provide the specific chemicals used to ensure that the procedure laid out in this section can be replicated accurately. The specificity of the list accounts for all possible contributions to the success of the IHC. 4% Formaldehyde in phosphate buffered s aline (pH 7.40.2) (BM 154, Lot# E1RF25R Boston BioProducts Ashland, MA USA )

PAGE 21

16 Gultaraldehyde, 50% aqueous solution (LOT: P27A045, Alfa Aesar, Ward Hill, MA USA) Phosphate buffered saline ( pH 7.4, 1X, Gibco Thermo Fisher Scientific, Waltham, MA USA). Normal goat serum ( S26 100ML, Lot#2855574 EMD Millipore, Temecula, CA USA) Primary anti 5HT rabbit polyclonal IgG antibody ( 20080, Lot#1650001 Immunostar Hudson, WI USA) Primary anti Fos mouse monoclonal IgG antibody ( E 8, sc 166940, Santa Cruz Biotechnology, Dallas, TX USA) Secondary goat anti rabbit IgG Alexa Flour 594 ( A21121, Lot#18 89303, Thermo Fisher Scientific, Waltham, MA USA). Secondary goat anti mouse IgG Alexa Flour 488 ( A11012, Lot#1745478, Thermo Fisher Scientific, Waltham, MA USA). Triton X 100 (Lot# SLBJ0812V, Sigma Life Science, St. Louis, MO) Vectashield mounting medium for fluorescence with DAPI ( H 1200, Vector Laboratories, Burlingame, CA USA) Detailed Procedure Behavioral trails : Prior to dissections ant were placed into different behavioral contexts. This was to determine their brain states under differing conditions for comparison. Ag gression : Ants from two different colonies were collected. 50 ants from each colony were placed in 11cm X 11cm plastic container. Ants were allowed to acclimate for 15 minutes. After 15min ants that engaged in aggressive behavior (defined as biting locking mandibles ) were selected and sacrificed. Social ( control) : Ants from the same colony were placed in 11cm X 11cm plastic container Ants were allowed to acclimate for 15 minutes. After the acclim ation period ants engaged in social behavior ( defined as attenuating another individual) were sampled and sacrificed. Dissection : All ants were dissected on a cold plate kept stored at 20C. Prior to dissection the cold plate was allowed to warm for 20 minutes. Once ants were selected they were placed on ice for 15 secs to anesthetize them. Once anesthetized they were transferred to the cold plate for

PAGE 22

17 dissection. Using micro dissection tools ant heads were removed. The ant heads had their mandibles removed to allow fixative into the cranial cavity. Immunohistochemsitry : All wash and incubation steps used a horizontal shaker. All s teps were performed at room temperature (21C) except fixation and primary antibodies incubations. Dissected ant heads were fixed in 4% formaldehyde with 0.1% Gluteralaldehyde for 5 hours. After 5 hours the brains were removed from the skull exoskeleton. The brains were then washed in PBS (3 X 10min) Following fixation ant brains were placed in 0.3% Trit on X 100 PBS for 2 hours at to permeabilize the tissue To reduce non specific bindings brains were incubated in 10% Normal Goat Seru m (NGS) in PBS for 1 hour Then each brain underwent standard secondary IHC staining for serotonin and c Fos. Serotonin antibody used was ImmunoStar rabbit anti 5 HT c Fos antibody used Santa Cruz mouse anti Fos Ant brains were incubated in a dilution of rabbit anti 5 HT pri mary antibody (1:500, 0.5% NGS in PBS) and mouse anti Fos primary antibody (1:50, 0.5% NGS in PBS) for 3 days at 4C After the incubation period in primary antibodi es the brains were washed in PBS ( 3 X 10min). Following the washes, the tissue was incubated in secondary antibodies goat anti rabbit IgG Alexa Flour 594 for serotonin and goat anti mouse IgG Alexa Flour 488 for c Fos (1:500 dilution in a solution of PBS) This was followed by a wash in PBS (3 X 1 0min). Finally, the brains were mounted on positively charged slides using Vectashield anti fading solution with DAPI. The slides were prepared using a piece of black tape with a window cut into it as the mounting field. Confocal Microscopy Images created from I HC were generated using Ziess LSM 700 c onfocal microscope All images were created using the 20X objective lens. Red and green lasers were used at 594nm and 488nm. Gain was left at default setting. Each image was

PAGE 23

18 generated using z stack. Imaging was set to the optimal slice thickness by using the built i n Ziess Image processing Images were processed using FIJI. Images are 3d max projections of z stack images. All images were compared to an atlas created by Tsuji et al ( 2007 ) to identify morphological structures. Controls Controls were run on the secondary antibodies to ensure lack of nonspecific binding. These brains underwent the entire protocol, but without primary antibodies. Fos staining several different experiments were run to generate Fos expression. Sucrose excitement Ants were starved for 24 hours. After the starvation period, a nts were given a solution of 1 0% sucr ose prior to dissection. Dissection was performed the same as above. Ants were dissected at three different time points: immediately after sucrose ingestions, 15min after sucrose ingestion, and 1hr after ingestion. Antennal removal experiment : Ants were di vided into three groups: controls, right antenna removed, or both antenna removed. Dissections were performed as described above, however prior to dissection each group was placed on ice for 15 secs, then placed on the cold plate where their antennae were removed. Controls underwent the same process, but did not have their antennae cut. All antennae were cut using micro dissection scissors. Ants from each group were sacrificed at three different time points: 5min 6hrs, and 24hrs after antennal removal Analysis The brains analyzed in the results are the best selected from each experimental group. Thus, all brains used in results are pooled from all the experimental contexts

PAGE 24

19 CHAPTER 3 RES ULTS Figure 1 Images of a whole T. caespitum brain stained for serotonin including ROI (regions of interest) and serotonin (5 HT) positive neurons. The figure is made up the best representations of each ROI. 5 HT is stained in red. a) Serotonin (5 HT) staining in a whole T.caespitum brain. Imaged at 20x magnification AL = antennal lobes, MB = mushroom bodies, OL = optic lobe, and S O G = subesophageal ganglion. Brain in ventral orientation. b) S O G shows three pairs of serotonergic neuronal clusters highlighted by the blue arrows. c) A zoomed view of t he serotonergic neuronal clusters in the antennal lobes. d) Highlighted serotonergic neurons within the optic lobe. 5 HT saturated medulla cells are present in the enclosed within the circle. SOG

PAGE 25

20 Figure 2 Images of the mushroom body (MB) structure in T. caespitum brains. The figure is made of the best representation of the MB. Serotonin (5 HT) is stained in red. Images produced at 20X magnification. MB = mushroom body, lCa = lateral calyx and mCa = medial calyx. a) Zoomed image showing a MB. The arrows point to neurons in the MB and the dashed circle highlights serotonergic processes. b) Image showing the different calyces in the MB. The dashed circle highlights serotonergic processes. S erotonin anatomy in T.caespitum brain Our protocol showed excellent staining for serotonin (5 HT) ( figure 1.a). In figure 1 we saw strong presence of serotonergic neurons in the antennal lobes (AL), optic lobes (OL), and the subesophagael ganglion (S O G). There appears to be bilateral symmetry in number of neurons in those the regions (figure 1.a). The AL showed an estimate of 10 neurons in each lobe (figure1.c). OL contained an estimate of 22 neurons per lobe.

PAGE 26

21 The OL medulla appears saturated with 5 HT ( fi gure 1.d). Within the SOG there appears to be three clusters of neurons. A pair of clusters located in the anterior, toward the AL (n = 3 per cluster), a pair on lateral side (n = 3 per cluster), and a pair located laterally in the posterior (n = 3 per clu ster) (figure 1.b). The mushroom bodies (MB) also showed presence of serotonergic neurons (Figure 2.a). There is an estimate of 9 neurons in the MB (figure 2.b). The calyces of the MB have serotonergic processes suggesting that 5 HT is shuttled into the ca lyces from neurons elsewhere in the brain. The processes appear to terminate at the lip of each calyx. Neuron count Neurons were counted in images in which morphology could be identified (n=32). The neuron count represents an estimate of the number of neurons within in a T. caespitum brain. Due to the limitations of the technique, the numbers displayed the distributio n in Figure 3 represent our best estimates of whole brain 5 HT neurons. Our max total count for 5 HT neurons was 78, our minimum was 1. The mean for the total number of neurons is 23.29. The median values for each region in the brain are close with their m eans very close to the median value

PAGE 27

22 except for the OL. These neurons counted were the only neurons that were detected by our technique, meaning we can only provide an estimate on count. Figure 3 Boxplot showing the distribution of 5 HT neurons wi thin T.caespitum The black bar represents the median and the red diamond is the mean. Open circles are outlying points. AL = antennal lobes, MB = mushroom bodies, OL = optic lobe, and SOG = subesophageal ganglion

PAGE 28

23 O ther represents any neurons that were p resent, but could not be placed within a specific m orphological structure. Figure 4 Boxplot showing the distribution of proportions of the total neuron count that are in each morphological structure. The proportions of each region were calculated using the total number of neurons within each imaged brain. The black bar represents the median and the red diamond is the mean. AL = antennal lobes, MB = mushroom bodies, OL = optic lobe, and SOG = subesophageal ganglion. We see a wi de variability in each regi on of how many neurons are contained within that region (Figure 4 ) Mean MB = 39%, Mean AL = 43%, Mean OL = 52%, Mean SOG = 45%, Mean Other = 36%. We see the greatest proportion of neurons within the OL, mean proportion = 52%. The other regions are relatively similar in their proportion of the total.

PAGE 29

24 Figure 5 Images of a whole T. caespitum brain stained for serotonin and c Fos including ROI (regions of interest), serotonin (5 HT), c Fos positive neurons. 5 HT is in red and c Fos is in green. Yellow indicates co localization of Fos and 5 HT. AL = antennal lobes, lCa = lateral calyx, and mCa = medial calyx, OL = optic lobe, and SOG = subesophageal ganglion. Brain in ventral orientation. Brain is in ventral orientation. All images are from the same brain. a) This image illustrates the whole brain with dual staining for Fos and 5 HT. The yellow staining shows serotonergic neurons that are active. Activity is located in the SOG, OL, and AL. Arrows point to neuronal clusters. The dashed box highlights both serotonergic and Fos processes, a closer view is represented in figure 3x.c Image is at 20x m agnification. b) A closer view of the clusters to illustrate co localization. Yellow indicates active serotonergic neurons. Arrows point to active neuronal clusters. Image is at 40x magnification. c) Zoomed view of Fos and 5 HT processes within the mushro om body calyces Arrows highlight points of co localization. Yellow (co localization) indicates active 5 HT processes. Image is at 20x magnification. Ant brain dissected from an ant during aggression.

PAGE 30

25 c Fos staining Using the method laid out previously, we were able to see distinct c Fos staining (Figure 5 ). We see similar neuronal clusters and bilateral symmetry as in Figure 1, however there are distinct Fos neurons stained green (Figure 5 ). Those neurons stained red show presence of 5 HT and those col ored yellow indicate co localization of c Fos and 5 HT. Co localization with Fos indicates activity within those neurons, specifically serotonergic activity (Figure 5 ). We see that there is 5 HT activity within the AL, OL, SOG, and MB calyces. Within the A L we see unique 5 HT and Fos neurons. Again we see processes within the mushroom body calyces, but now we see Fos activation as well (Figure 5 .c). These processes also have co localization at points indicating serotonergic activity in the calyces (Figure 5 .c). Looking at Table 2, of the 51 neurons present the majority are serotonergic (91.49%). 50% of the neurons show the presence of c Fos and 44.68% show co localization. At the time of the staining 44.68% of the serotonergic neurons within this indivi dual were active. All areas of the brain showed 5 HT activation (62.50% in the AL, 25% in the OL, and 46.67% in the SOG). Most of the serotonergic activity was in the AL and SOG (62.50% and 46.67% of neurons showing co localization respectively). 100% of t he neurons within the OL and SOG are serotonergic, but only 25% and 46.67% showed activity. Only the AL showed presence of Fos only neurons (Table 2)

PAGE 31

26 Table 2 Number, type, and proportion of neurons within co stained brain. All counts are based on Figure 3X. Neurons were counted as co localized if the neuron contained any amount of yellow. Numbers were rounded to two decimal places. Type of Neuron Total Number of Neurons Total c F os positive neurons Total 5 HT positive Neurons Co localized neurons % c Fos % 5 HT % co localized AL 16 13 12 10 81.25 75 .00 62.5 0 OL 16 4 16 4 25 .00 100 .00 25 .00 SOG 15 7 15 7 46.67 100 46.67 Total 47 24 43 21 51.0 91.49 44.68

PAGE 32

27 CHAPTER 4 DISCUSSION Serotonergic staining The technique employed in this paper provides an estimate for the number of 5 HT neurons within T. caespitum brains. It also shows where these neurons are likely located and the location of their activity. S erotonergic neurons were found in the AL, MB, SOG, and OL. Additionally, serotonergic processes were identified branching into the MB calyces Both H. saltator (Hoyer et al 2005) and C. japonicus (Tsuji et al 2007) were found to have similar serotonin immunoreactiv ity. Both studies showed serotonergic processes and a r borization within the MB calyces similar to Figure 2. However, the number of neuronal cells found in the AL (Fig.1, Fig.3) differs from the single AL neuron found previously (Tsuji et al 2007). The pr ocesses within the MB calyces appear to terminate in the lip of the calyx (Fig.2) The termination of 5 HT processes in the lip of the MB calyx (Fig. 2) was also found in Tsuji et al (2007) results. In ants, t he lip of the MB calyces contains afferents from the antennal lobes and is associated with olfactory input (Gronenberg 200 1, Fahrbach 2006 ). Therefore the 5 HT processes highlighted could be those afferents. If so, serotonin may influence higher order processing of olfactory information (Hoyer et a l 2005) Muscedere et al (2012) blocked 5 HT in P. dentata and found that workers failed to locate pheromone trails. Without 5 HT they struggled to locate and follow olfactory information. This also suggests the importance of 5 HT in sensory integration Since ants primarily rely on their antennae for sensory information, 5 HT provides a mechanism for how that information is processed within the MB. The SOG is muscles. The prese nce of serotonin in the SOG suggests 5 role in modulating mandible control. For T. caespitum this is quite important in regard to aggression against conspecifics. T. cae s pitum lock

PAGE 33

28 mandibles during conflict with conspecifics so 5 HT is likely active within those neurons during aggression. We also see presence of 5 HT within the optic lobes and medulla of the optic lobes which confirms other s findings (Seid et al 2008, Hoyer et al 2005, Schafer and Bicker 1986). The serotonergic immunoreactivity in the medulla of the OL (Figure1) is similar to the findings described by Hoyer et al (2005 ). However, most ants do not rely heavily on sight. The presence of 5 HT within the optic l obe could be a remnant of similar architecture found in other hymenopterans ( Schafer and Bicker 1986) or the high number of serotonergic neurons is necessary for foragers outside the nest to process visual information to successfully collect food, subdue prey, and defend territory (Seid et al 2008) c Fos staining This technique shows the first example of Fos staining in an ant species. Additionally, it highlights the activity of 5 HT w ithin a T. caespitum individual engaged in aggression. 5 HT neurons within the AL, OL, and SOG are shown to be active. We also see active serotonergic processes within the MB calyces. This activity within the antennal lobes and the lip of the MB calyces sugges ts that 5 HT is sent to the MB from the AL during aggression. Again, this supports 5 HT role in modulating higher order processing of olfactory information (Hoyer et al 2005). We also see serotonergic activation in the SOG. This would correspond wit h the locking of mandibles that T. caespitum engage in during aggression. There was minimal 5 HT activation of the OL. Serotonin may not modulate visual information during aggression. T.caespitum primarily relies on olfactory information thus it is not sur prising that there is minimal activation in the OL. Our technique helps illustrate how 5 HT is being used during aggression. Neuron count We were able to obtain an estimate of 5 HT neurons using our technique. The maximum count of 5 HT neurons was 78. Honeybees have about 75 (Shurman n and Klemm

PAGE 34

29 1984). Other have reported 130 to 200 in C. jap onicus and C herculeanus respectively (Tsuji et al 2007 Gronenberg 1996 ). Given this range our count seems feasible. Brain size is social insects is tied to social organization It is hypothesized that miniaturized as social complexity increases. Individual ant s cognition is limited, but as a whole their collective intelligence allows for co mplex group behaviors (Feinermen and Traniello 2016) The low number of 5 HT neurons helps support a limited cognitive ability within an individual ant T caespitum is able to perform a wide repertoire of behaviors using a limited number of neuronal cells Conceptual Model for 5 HT modulated aggression Using the data from this technique I would like to propose a conceptual model for 5 HT modulated aggression in T. caespitum Pretend you are an individual ant. You use a random walk to explore your colonies territorial space looking for resources While walking you encounter other ants You a n tentate them to determine their identity by reading their cuticular hydrocarbons and assessing the relative abundance of methyl branched alkanes and n alkenes (Sano et al 2018) W e k now that when individuals interact with their nestmates 5 HT increases (Bubak et al 2016, Hoover et al 2015) so the interactions likely increase 5 HT p roduction in the AL The large number of 5 H T neurons found in the AL would support this (Fig.3) Th e 5 HT produced in the AL is sent to the MB lip along the processes found in Figure 5 From there the information is processed and the you recognize the other as nestmate. This memory s lowly decays but as you explore you encounter more nestmates refreshing the 5 HT in your brain. cuticular hydrocarbons to not match that of your n estmates. Since you have been in constant contact with your nestmates your 5 HT has been constantly refreshed. This constant refreshing of 5 HT by nestmate interaction primes the brain for aggression (Hoover et al 2016). During

PAGE 35

30 aggression we see a rise in brain titer levels of 5 HT (Bubak et al 2016). The rise we see in brain titer levels of 5 HT during aggression could be explained by an increase in production of serotonin from the antennal lobes being shuttled to the MB (Fig.5) Once the 5 HT is integrated within the MB you decide to aggress. You lock mandibles with you opponent. This results in sustained serotonin release in the SOG to keep the mandibles locked with your adversary. (Fig. 5) The ants around you do the same sinc e they share a similar brain state. After some time, a large portion of your colony is with you attacking the enemy. After some monoaminergic threshold is reached the fight ends and you follow the pheromone trail back to your nest Limitations While this study is informative and descriptive of the pavement ant brain, the technique does have limitations. This data collected is based off the best examples from IHC ; m eaning only the images created that displayed detailed morphology and staining were chosen. Thus, a ll measurements and interpretations made are estimates Using whole brains with IHC can make it difficult to achieve uniform orientation so brain s in this study had varying orientations This could have influenced interpretations of morphology and n euron counts. Some neurons tagged by the technique might not have been able to be counted due the orientation of the brain and the limitations of the microscope to penetrate the tissue. This reduces confidence in this techniques ability to establish a true neuron count. Also, the orientations of each brain may lead to inaccurate comparisons between experimental groups. Finally, since the images analyzed were selected as the best examples of the IHC technique, these brains come from a variety of individuals exposed to different behavioral contexts. It would be difficult to draw conclusions to a behavior or experimental setting. All c Fos measurements and estimates are based on a single brain. We were unable to replicate c Fos staining This limits the conclusions we can draw on the efficacy of our

PAGE 36

31 technique. However, this might be due to the use of Formaldehyde instead of Para formaldehyde (PFA). The antibodies used were described as working best with PFA as a fixative. Future studies t hat use this technique should try a variety of fixatives to determine which works best for Fos staining in their subjects. Future Directions Th is study shows that c Fos can be used within ants to determine potential mechanisms for aggressive behavior in pavement ants. Additionally, once our c Fos technique is perfected and replicated, we can use this technique to understand other behaviors in other species. The data collected helps build our conceptual model of collective behavior. By looking at an indivi illustrate potential neural mechanisms that influence individual behavior. Those individual behaviors magnified to the whole colony could give insight into how collective decisions are made. This study highlights the proximate mechanism for the beginnings of complex, social behavior.

PAGE 37

32 C HAPTER 5 REFERENCES Alloway, T. M. (1979). Raiding behaviour of two species of slave making ants, Harpagoxenus americanus (Emery) and Leptothorax duloticus wesson (Hyme noptera: Formicidae). Animal Behaviour 27 (PART 1), 202 210. https://doi.org/10.1016/0003 3472(79)90140 4 Neisewander, J. L. (2016). Immediate early gene expression revea ls interactions between social and nicotine rewards on brain activity in adolescent male rats. Behavioural Brain Research 313 244 254. https://doi.org/10.1016/j.bbr.2016.07.024 Bubak, A. N., Yaeger, J. D. W., Renner, K. J., Swallow, J. G., & Greene, M. J (2016). Neuromodulation of nestmate recognition decisions by pavement ants. PLoS ONE 11 (11), 1 16. https://doi.org/10.1371/journal.pone.0166417 Cronin, A. L. (2015). Individual and Group Personalities Characterise Consensus Decision Making in an Ant. Ethology 121 (7), 703 713. https://doi.org/10.1111/eth.12386 Davidson, J. D., & Gordon, D. M. (2017). Spatial organization and interactions of harvester ants during foraging activity. Journal of The Royal Society Interface 14 https://doi.org/10.1098/rsif.2017.0413 Day H E Kryskow E M Nyhuis T J Herlihy L., Campeau S. (2008). C ondit ioned fear inhibits c fos mRNA expression in the central extended amygdala Brain Res earch 1229 137 46. Dragunow, M., & Faull, R. (1989). The use of c fos as a metabolic marker in neuronal pathway tracing. Journal of Neuroscience Methods 29 (3), 261 265. https://doi.org/10.1016/0165 0270(89)90150 7 Dragunow, M., & Robertson, H. A. (1988). Localization and induction of c fos protein like immunoreactive material in the nuclei of adult mammalian neurons. Brain Research 4 40 (2), 252 260. https://doi.org/10.1016/0006 8993(88)90993 6 Dyer, A. G. (2005). Honeybee (Apis mellifera) vision can discriminate between and recognise images of human faces. Journal of Experimental Biology 208 (24), 4709 4714. https://doi.org/10.1242/jeb.01929 Edwards, S. C., & Pratt, S. C. (2009). Rationality in collective decision making by ant colonies. Proceedings of the Royal Society B: Biological Sciences 276 (1673), 3655 3661. https://doi.org/10.1098/rspb.2009.0981 Fahrba ch, S. E. (2006). Structure of the Mushroom Bodies of the Insect Brain. Annual Review of Entomology 51 (1), 209 232. https://doi.org/10.1146/annurev.ento.51.110104.150954

PAGE 38

33 Gauthier M. an d Grunewald, B. (2012). Neurotransmitter Systems in the Honey Bee Brain: Functions in Learning and Memory In: Galizia, C. G., Eisenhardt, D., & Giurfa, M. editors Honey Bee Neurobiology and Behaviour. A Tribute to Randolf Menzel Honeybee Neurobiology and Behavior Dordrecht: Springer. https://doi.org/10.1007/978 94 007 2099 2 Galizia, C. G., Eisenhardt, D., & Giurfa, M. (2012). Honey Bee Neurobiology and Behaviour. A Tribute to Randolf Menzel Honeybee Neurobiology and Behavior https://doi.org/10.1007/978 94 007 2099 2 Ghosal, K., Naples, S. P., Rabe, A. R., & Killian, K. A. (2010). Agonistic behavior and electrical stimulation of the antennae induces fos like protein expression in the male cricket brain. Archives of Insect Bioch emistry and Physiology 74 (1), 38 51. https://doi.org/10.1002/arch.20360 Giraldo, Y. M., Patel, E., Gronenberg, W., & Traniello, J. F. A. (2013). Division of labor and structural plasticity in an extrinsic serotonergic mushroom body neuron in the ant Pheid ole dentata Neuroscience Letters 534 (1), 107 111. https://doi.org/10.1016/j.neulet.2012.11.057 Greene, M. J., & Gordon, D. M. (2007). Interaction rate informs harvester ant task decisions. Behavioral Ecology 18 (2), 451 455. https://doi.org/10.1093/beheco/arl105 Gronenberg, W. (1996). Neuroethology of ants. Naturwissenschaften 83 (1), 15 27. https://doi.org/10.1007/s001140050240 Gronenberg, W., 2001. Subdivisions of the hymenopteran mushroom body calyces by their afferent supply. Journal of Comparative Neurology 436 474 489. Guzowski, J. F., Setlow, B., Wagner, E. K., & McGaugh, J. L. (2001) Experience dependent gene expression in the rat hippocampus after spatial learning: a comparison of the immediate early genes Arc, c fos, and zif268. Journal of the Society for Neuroscience 21 (14), 5089 5098. h ttps://doi.org/21/14/5089 [pii] Hoover, K. M., Bubak, A. N., Law, I. J., Yaeger, J. D. W., Renner, K. J., Swallow, J. G., & Greene, M. J. (2016). The organization of societal conflicts by pavement ants Tetramorium caespitum : An agent based model of amine mediated decision making. Current Zoology https://doi.org/10.1093/cz/zow041 Hoyer, S. C., Liebig, J., & Rssler, W. (2005). Biogenic amines in the ponerine ant Harpegnathos saltator : Serotonin and dopamine immunoreactivity in the brain. Arthropod Structure and Development https://doi.org/10.1016/j.asd.2005.03.003

PAGE 39

34 Hoyle G (1985): Generation of motor activity and control of behaviour: the role of the neuro modulator octopamin e and the orchestration hypothesis; in Kerkut GA, Gilbert L (eds): Comparative Insect Physiology, Biochemistry and Pharmacology. Toronto, Pergamon Press, vol 5, 607 621 Hu, L., Vander Meer, R. K., Porter, S. D., & Chen, L. (2017). Cuticular Hydrocarbon Pro files Differentiate Tropical Fire Ant Populations ( Solenopsis geminata Hymenoptera: Formicidae ). Chemistry and Biodiversity 14 (11), 1 8. https://doi.org/10.1002/cbdv.201700192 Iyer, L. M., Aravind, L., Coon, S. L., Klein, D. C., & Koonin, E. V. (2004). Evolution of cell cell signaling in animals: Did late horizontal gene transfer from bacteria have a role? Trends in Genetics 20 (7), 292 299. https://doi.org/10.1016/j.tig.2004.05 .007 (2017). Effects of peripheral administration of a Neuromedin U receptor 2 selective agonist on food intake and body weight in obese mice. International Journal o f Obesity 41 (12), 1790 1797. https://doi.org/10.1038/ijo.2017.176 Kamhi, J. F., Nunn, K., Robson, S. K. A., & Traniello, J. F. A. (2015). Polymorphism and division of labour in a socially complex ant: neuromodulation of aggression in the Australian weaver ant, Oecophylla smaragdina Proceedings of the Royal Society B: Biological Sciences 282 (1811), 20150704. https://doi.org/10.1098/rspb.2015.0704 Kamhi, J. F., & Traniello, J. F. A. (2013). Biogenic amines and collective organization in a superorganism: Neuromodulation of social behavior in ants. Brain, Behavior and Evolution 82 (4), 220 236. https:/ /doi.org/10.1159/000356091 Kang, K., Park, S., K im, Y. S., Lee, S., & Back, K. (2009) Biosynthesis and biotechnological production of serotonin derivatives. Applied Microbiology a nd Biotechnology 83 (1), 27 34. Lioni, A., & Deneubourg J. L. (2004). Collective decision through self assembling. Naturwissenschaften 91 (5), 237 241. https://doi.org/10.1007/s00114 004 0519 7 Marshall, J. A. R., Bogacz, R., Dornhaus, A., Planque, R., Kovacs, T., & Franks, N. R. (2009). On optimal decision m aking in brains and social insect colonies. Journal of The Royal Society Interface 6 (40), 1065 1074. https://doi.org/10.1098/rsif.2008.0511 Muscedere, M. L., Cahan, S. H., Helms, K. R., & Traniello, J. F. A. ( 2016 ). Behavioral Ecology Geographic and life history variation in ant queen colony founding correlates with brain amine levels. https://doi.org/10.1093/beheco/arv152

PAGE 40

35 Muscedere, M. L., Johnson, N., Gillis, B. C., Kamhi, J. F., & Traniello, J. F. A. (2012). Serotonin modulates worker responsivenes s to trail pheromone in the ant Pheidole dentata Journal of Comparative Physiology A: Neuroethology, Sensory, Neural, and Behavioral Physiology https://doi.org/10.1007/s00359 011 0701 2 Muscedere, M. L., & Traniello, J. F. A. (2012). Division of labor in the hyperdiverse ant genus Pheidole is associated with distinct subcaste and age related patterns of worker brain organization. PLoS ONE 7 (2). https://doi.org/10.1371/journal.pone.0031618 Neumaie r, J. F., Sexton, T. J., Yracheta, J., Diaz, A. M., & Brownfield, M. (2001). Localization of 5 HT7 receptors in rat brain by immunocytochemistry, in situ hybridization, and agonist stimulated cFos expression. Journal of Chemical Neuroanatomy 21 (1), 63 73. https://doi.org/10.1016/S0891 0618(00)00092 2 Penick, C. A., Brent, C. S., Dolezal, K., & Liebig, J. (2014). Neurohormonal changes associated with ritualized combat and the formation of a reproductive hierarchy in the ant Harpegnathos saltator The Journa l of Experimental Biology 217 (Pt 9). https://doi.org/10.1242/jeb.098301 Pratt, S. C. (2005). Quorum sensing by encounter rates in the ant Temnothorax albipennis. Behavioral Ecology 16 (2), 488 496. https: //doi.org/10.1093/beheco/ari020 Pratt, S. C., Mallon, E. B., Sumpter, D. J. T., & Franks, N. R. (2002). Quorum sensing, recruitment, and collective decision making during colony emigration by the ant Leptothorax albipennis Behavioral Ecology and Sociobiology 52 (2), 117 127. https://doi.org/10.1007/s00265 002 0487 x Pruitt, J. N., & Riechert, S. E. (2011). How within group behavioural variation and task efficiency enhance fitness in a social group. Proceedings of the Royal Society B: Biological Sc iences 278 (1709), 1209 1215. https://doi.org/10.1098/rspb.2010.1700 Renucci, M., Tirard, A., Charpin, P., Augier, R., & Strambi, A. (2000). c Fos related antigens in the central nervous system of an insect, Acheta domesticus. Archives of Insect Biochemistry and Physiology 45 (4), 139 48. https://doi.org/10.1002/1520 6327(200012)45:4<139::AID ARCH1>3.0.CO;2 N Rossler, W. and Groh, C. (2012). Plasticity of Synaptic Microcircuits in the Mushroom Body Calyx of the Honey Bee In: Galizia, C. G., Eisen hardt, D., & Giurfa, M. editors Honey Bee Neurobiology and Behaviour. A Tribute to Randolf Menzel Honeybee Neurobiology and Behavior Dordrecht: Springer. https://doi.org/10.1007/978 94 007 2099 2 Sano, K., Bannon, N., and Greene, M.J. 2018. Pavement ant workers ( Tetramorium caespitum ) assess cues coded in cuticular hydrocarbons to recognize conspecific and heterospecific non nestmate ants. Journal of Insect Behavior DOI: 10.1007/s10905 017 9659 4.

PAGE 41

36 Sasaki, T., & Pratt, S. C. (2013). Ants learn to rely on more informative attributes during decision making. Biology Letters 9 (6), 20130667 20130667. https://doi.org/10.1098/rsbl.2013.0667 Schafer, S., Bicker, G., ( 1986 ) Common projection areas of 5 HT like and GABA like immunoreactive fibers and the visual system of the honeybee. Brain Research 380 368 370 Schurmann, F.W., Klemm, N., ( 1984 ) Serotonin immunoreactive neurons in the brain of the honeybee. The Journal of Comparative Neurology 225 570 580. Seid, M. A., Goode, K., Li, C., & Traniello, J. F. A. (2008). Age and subcaste related patterns of serotonergic immunoreactivity in the optic lobes of the ant Pheidole dentata. Developme ntal Neurobiology 68 (11), 1325 1333. https://doi.org/10.1002/dneu.20663 Smith, A. R., Muscedere, M. L., Seid, M. A., Traniello, J. F. A., & Hughes, W. O. H. (2013). Biogenic amines are associated with worker task but not patriline in the leaf cutting ant Acromyrmex echinatior Journal of Comparative Physiology A: Neuroethology, Sensory, Neural, and Behavioral Physiology https://doi.org/10.1007/s00359 013 0854 2 Sumpter, D. J. T. (2006). The principles of collective animal behaviour. Philosophical Transactions of the Royal Society B: Biological Sciences 361 (1465), 5 22. https://doi.org/10.1098/rstb.2005.1733 Tsuji, E., Aonuma, H., Yokohari, F., & Nishikawa, M. (n.d.). Serotonin immunoreactive N eurons in the Antennal Sensory System of the Brain in the Carpenter Ant, Camponotus japonicus. https://doi.org/10.2108/zsj.24.836 von Frisch K. (1914). Der Farbensinn und Formensinn der Biene. Zool Jahrb Abt Allg Zool Physiol Tiere 37 1 238. Vleugels, R., Verlinden, H., & Broeck J. V. (2015) Serotonin, serotonin receptors and their actions in insects. Neurotransmitter 2 (314) : doi: 10.14800/nt.314. Wada Katsumata, A., Yamaoka, R., & Aonuma, H. (2011). Social interactions influence dopamine and octopamine homeostasis in the brain of the ant Formica japonica Journal of Experimental Biology 214 (10), 1707 1713. https://doi.org/10.1242/jeb.051565