Master's thesis of Science: Characterising microbial and neuroimmune interactions in a mouse model of autism

Characterising Microbial and Neuroimmune Interactions in a Mouse Model of Autism

A thesis submitted in fulfilment of the requirements for the degree of Master of Science

Samiha Sayed Sharna

BSc in Microbiology (North South University, Bangladesh)

School of Health and Biomedical Science

College of Science, Engineering and Health

RMIT University

March 2020

Declaration

I certify that except where due acknowledgement has been made, the work is that of the

author alone; the work has not been submitted previously, in whole or in part, to qualify for

any other academic award; the content of the thesis is the result of work which has been

carried out since the official commencement date of the approved research program; any

editorial work, paid or unpaid, carried out by a third party is acknowledged; and, ethics

procedures and guidelines have been followed.

Samiha Sayed Sharna

2nd March 2020

Publications arising from this research

Original research articles:

1. Sharna, S. S., Balasuriya, G. K., Hosie, S., Nithianantharajah, J., Franks, A. E., & Hill-Yardin, E.

L. (2020). Altered Caecal Neuroimmune Interactions in the Neuroligin-3R451C Mouse

Model of Autism. Frontiers in cellular neuroscience, 14(85). doi:10.3389/fncel.2020.00085

2. Sharna SS, Balasuriya G, Jen Wood, Howard Habtom, Hill-Yardin EL*, Franks AE*. Altered

Microbial composition in the Neuroligin-3R451C mouse model of autism.

For submission to Scientific Reports; IF 4.011

Review article:

* Sharna SS, *Abo-Shaban T, *Hosie S, Franks AE, and Hill-Yardin EL. Gut-associated lymphoid tissue:

A functional comparison of Peyer’s and caecal patches in mice and implications for neurological

disease.

For submission to Frontiers in Cellular and Infection Microbiology IF 4.3

Conference Abstracts:

1. Australasian Neuroscience Society (ANS) 2019 (2-5th December, Adelaide)

Poster presentation:

Samiha S. Sharna, Jennifer L. Wood, Howard Habtom, Rachele Gore, Tanya Abo-Shaban, Ashley E.

Franks*, Elisa L. Hill-Yardin*“Changes in caecal neuroimmune interactions in the Neuroligin-3R451C

mouse model of autism”

2. Federation of Neurogastroenterology & Motility (FNM) 2020 (25-28th March, Adelaide)

Abstract submitted:

Samiha S. Sharna1, Jennifer L. Wood2, Howard Habtom2, Gayathri K Balasuriya1, Ashley E. Franks2*,

Elisa L. Hill-Yardin1*“Altered caecal neuroimmune interactions in the Neuroligin-3R451C mouse model

of autism”

*equal contribution

Acknowledgements

Firstly, I would like to express my utmost gratitude towards my supervisors, Associate Professor Elisa

Hill-Yardin and Professor Ashley Franks, to whom I am deeply in debt. Elisa, your consistent

mentorship, constructive criticism and endless support has been truly been inspirational. I appreciate

how you are always concerned about my well-being and being present regardless of time. Ashley, I

am deeply thankful for in the manner in which you have introduced microbial ecology to me. Though

I am a microbiologist, through your guidance I am now well-versed in the principals of ecology. It is

with their positive encouragement that I am able to produce this thesis. Thank you so much for giving

me the opportunity to create unforgettable memories with my peers.

Secondly, I would like to thank Dr. Gayathri Balasuriya for warmly welcoming me to the lab. Your

endless optimism and friendliness never cease to amaze me. You have made sure my transition to the

lab environment was comfortable and exciting. Additionally, to Dr. Jen Wood and Dr. Howard Habtom,

I thank you for remaining patient while teaching me the R software and statistical knowledge.

Moreover, I display gratefulness towards Dr. Suzanne Hosie, who has always provided me feedback

on literature reviews. Also, to my lab members: Tanya, Kevin, Mansour, Mitti, Rachele, Shani, Sarah,

Ellie and Josh, thank you for always reassuring and helping me with the thesis. I am very lucky to be

able to work alongside such amazing peers.

Furthermore, I would like to acknowledge my parents as they have provided me with unconditional

support and love through this journey. Without their trust and belief in me, it would have been

impossible for me to undertake this pathway. To my sister, Mina Nasrin, and brother-in-law, Delwar

Hossain, I would like to express my sincere appreciation as they have consoled me whenever I felt

homesick. To my brother, Thofayl Mahmud, and sister-in-law, Shahnaz Parvin, thank you for cheering

me on all the way from Bangladesh. Also, I am grateful towards my nieces and nephews, Labiba (my

cheerleader), Namira, Arham, Ahnaf and Abrar for providing me with words of encouragement during

biology. Lastly, I would like to show my deep appreciation towards my husband Shahid. Regardless of

this period. In Addition, I thank my schoolteacher Biltu Kumar Singha for quietly nurturing my love of

his own PhD, he ensured I was always motivated and optimistic.

Declaration

Publications arising from this research

Acknowledgments

List of figures

List of abbreviations and units

Abstract

Chapter 1: Introduction

1.1 Overview

1.2 Genetics of ASD

1.2.1 The R451C mutation in Neuroligin-3

1.3 The enteric nervous system

1.3.1 Altered neuro-immune interactions in NL3R451C mice

1.4 Gut associated lymphoid tissue (GALT)

1.4.1 Caecal Patch

1.5 Microbial influences on gut-associated lymphoid tissue

1.6 Microbiome-neuroimmune signatures in ASD

1.7 Microbial changes in NL3R451C mice

1.8 Project Rationale

1.8.1 Abnormal neuroimmune pathways in NL3R451C mice

1.8.2 Alterations in gut microbes in ASD

1.9 Hypothesis and aims

Chapter 2: Altered caecal neuroimmune interactions in the Neuroligin-3R451C mouse model of autism

2.1 Abstract

2.2 Introduction

2.3 Methodology

2.4 Results

2.5 Discussion

2.6 Conclusion

2.7 References

Chapter 3: Altered Microbial composition in the Neuroligin-3R451C mouse model of autism

3.1 Abstract

3.2 Introduction

3.3 Methodology

3.4 Results

3.5 Discussion

3.6 Conclusion

3.7 References

Chapter 4: Discussion

4.1 The NL3 R451C mutation causes a reduction in caecal weight

4.2 The NL3 R451C mutation leads to an altered enteric neuronal composition

4.3 Caecal macrophages are altered due to the NL3R451C mutation

4.4 The NL3 R451C mutation causes caecal microbial dysbiosis

Conclusion

Future directions

References

Appendices

Appendix 1: Ethics approval letter

Appendix 2: Haematoxylin & Eosin staining

Appendix 3: Caecal patch analysis using ImageJ to determine the cell density

Appendix 4: Using ImageJ, analysis of number of enteric neurons/ganglia

Appendix 5: Iba1 immunoreactive cell analysis using Imaris7

Appendix 6: Mouse ID used for microbial analysis

Appendix 7: R scripts for data preparation

Appendix 8: R scripts for making bar charts

Appendix 9: R scripts for NMDS

Appendix 10: R scripts for Statistics (PERMANOVA)

Appendix 11: R scripts for compositional analysis

Appendix 12: R scripts for differential analysis

List of figures

Chapter 1

1.1

Myenteric ganglia of proximal jejunum in WT and NL3R451C

1.2

The effect of genotype on faecal microbial communities at five and nine weeks of age of WT and NL3R451C mice.

Chapter 2

2.1

Body weight, caecal weight, and caecal tissue area.

2.2

Cell density of the caecal patch.

2.3

Caecal submucosal neuronal numbers and proportions.

2.4

Caecal myenteric neuronal numbers and proportions.

2.5

Analysis of Immunofluorescent staining in caecal patches of WT and NL3R451C mice.

Chapter 3

3.1

Relative abundance of total bacterial OTUs (p < 0.05) from region specific samples of WT and NL3 R451C mice.

2.2

NMDS ordinations of bacterial communities of region-specific gut samples

3.3

Microbial community composition changes among different regions of WT and NL3 R451C

3.4

Microbial community change in NL3 R451C mice compared with WT

3.5

Compositional dissimilarity throughout the gut within WT and NL3 R451C mice using Bray- Curtis Dissimilarity.

List of abbreviations and units

ASD: Autism Spectrum Disorder BSA: Bovine Serum Albumin CAM: Cell adhesion molecule CNS: Central nervous system CP: Caecal patch DC: Dendritic cells ENS: Enteric nervous system FAE: follicle associated epithelium FGID: Functional gastrointestinal disorder GALT: Gut associated lymphoid tissues GI: Gastrointestinal GM-CSF: Granulocyte-macrophages colony-stimulating factor GRO: growth related oncogene H&E: Haematoxylin & Eosin IBD: Inflammatory bowel disease IFN: Interferon IgA: Immunoglobulin A IgM: Immunoglobulin M IL-10: Interleukin 10 IL1β: Interleukin 1 beta KI: Knock-in mice LPS: Lipopolysaccharides M cells: Microfold cells MAMPs: microbe-associated molecular patterns MRPP: Multi-response permutation procedure NL3: Neuroligin 3 NMDS: non-metric multidimensional scaling NOS: Nitric Oxide Synthase NT: Neurotypical OTUs: Operational Taxonomic Unit PBS: Phosphate Buffered Saline PERMANOVA: Permutational multivariate analysis of variance PGP: poly-glycoprotein Poly-IC: polyinosinic- polycytidylic acid SCFA: short chain fatty acids SFB: Segmented filamentous bacteria SPF: Specific pathogen free Th17: T-helper 17 cells VEGF: Vascular endothelial growth factor WT: Wild type

Abstract

Gastrointestinal dysfunction and changes in the microbial community of the gut are commonly

reported in children with autism spectrum disorder (ASD) and have been proposed to contribute to

behavioural impairment. Gut microbial communities interact with the immune system and also

produce neuro-active molecules that modulate the central and enteric nervous systems. Many rare

gene mutations implicated in autism, including the well-studied neuroligin-3 R451C missense

mutation, influence neuronal communication. In addition to changes in the nervous system, people

with autism are more likely to have immune disorders. The caecum is involved in generating immune

responses and acts as a repository of intestinal microorganisms, but it is unknown whether this organ

plays a role in autism. At the blind end of the caecum, gut lymphoid aggregates known as caecal patch

(CP) are found which contain various immune cells such as macrophages and dendritic cells, which are

crucial for mucosal immunity. To assess whether the autism-associated R451C mutation in the Nlgn3

gene influences the caecum, we first assessed for changes in caecal weight in NL3R451C mice. Using

immunofluorescence in wholemount preparations, we quantified the total number of enteric neurons

per ganglion in the myenteric and submucosal plexus. In frozen cross sections of the caecal patch, we

examined the density and morphology of Iba1-expressing macrophages. We also assessed caecal

microbial community composition using Illumina deep sequencing. NL3R451C mice have significantly

reduced caecal weight compared to wild-type (0.65±0.02g, 0.54±0.01g, WT and NL3R451C respectively;

p=0.0001). NL3R451C caecal myenteric plexus had enlarged ganglia (12.12 ± 0.02 and 17.19 ± 0.05 mm2

per ganglion; WT and NL3R451C respectively; p= 0.008) and more neurons per ganglion (WT:11±1,

NL3R451C:15±1; p=0.005). An increase in the number of neurons per ganglion was also observed in the

submucosal plexus (WT:5±1 and NL3: 6±1; p=0.04). 3D image analysis revealed a higher density of

Iba1-expressing macrophages in the caecal patch (11±1 and 14±1 cells/100 µm2, WT and NL3R451C

respectively; p=0.04) and smaller cell volume (946.9 ± 0.4µm3 and 559.7 ± 0.4µm3, WT and NL3R451C

respectively; p=0.004) in NL3R451C mice. Moreover, distinctly different caecal microbial community

structures were observed in NL3R451C compared to wild-type mice. These findings suggest that the

autism-associated R451C mutation in Neuroligin-3 impacts caecal neuronal structure, immune cell

(369 words)

properties, and microbial populations.

Chapter 1

1.0 INTRODUCTION

1.1 Overview

Autism Spectrum disorder (ASD; autism) is a highly prevalent neurodevelopmental disorder affecting

1 in 59 children, with a 4:1 male to female ratio (Loomes, Hull, & Mandy, 2017). Many individuals with

autism experience immune dysfunction and gastrointestinal (GI) disorders which have significant

health, developmental, social, and educational impacts (Coury et al., 2012; Wing & Gould, 1979).

Individuals with autism are three times more susceptible to developing GI symptoms compared to

children with typical development (Chaidez, Hansen, & Hertz-Picciotto, 2014) including alternating

diarrhoea and constipation, and abdominal pain (McElhanon et al., 2014). Increased permeability of

the GI tract along with altered motility are commonly correlated with core autism traits (Horvath &

Perman, 2002; Kohane et al., 2012; Neuhaus et al., 2018; Parracho et al., 2005). Moreover, a study of

over 14,000 patients reported that the prevalence of inflammatory bowel disease (IBD) was

significantly higher in children with ASD (Kohane et al., 2012).

Many gene mutations associated with autism affect nervous system function including a point

mutation in the NLGN3 gene (NL3) encoding the synaptic adhesion protein, Neuroligin-3. The gut has

its own nervous system known as the enteric nervous system (ENS). Recent research from our

laboratory demonstrated for the first time that the R451C mutation in NLGN3 associated with autism

causes altered gut function via changes in the ENS (Hosie et al., 2019). In the current study, we

assessed whether this R451C mutation in the NLGN3 gene affects ENS structure as well as immune

cell populations in mice (NL3R451C mice).

One of the focal points of this study is to investigate caecal neuro-immune interactions in the caecum

which is an understudied region of GI tract. The precise role of the caecum is unclear; however, it is

thought to be associated with the regulation of immune function (Kooij et al., 2016). The caecal-

appendix or caecum was initially thought to be an evolutionary remnant from primate ancestors, as

the equivalent intestinal compartment is enlarged in herbivores such as rabbits and cows (Furness,

2012). In humans, the appendix is a finger-like, blind-ended extension at the junction between the

small and large intestine and contains multiple aggregates of lymphoid follicles (the caecal patch)

located within the submucosa and lamina propria of the appendiceal wall (discussed in section 1.4

‘Gut-associated lymphoid tissue’; Kooij et al., 2016; Spencer, Finn, & Isaacson, 1985). In gut lymphoid

follicles, B cells, T cells, dendritic cells (DC) and macrophages are the most abundant population of

leukocytes and play a vital role in maintaining homeostasis (Liu et al., 2018; Ratcliffe, 2002; Kühl et al.,

2015). In this study, we screened for changes in caecal neuronal proportions and a subset of

monocytes or macrophages within the caecal patch immunoreactive for Iba-1 in NL3R451C mice.

Microbes and the metabolites they produce are important for maintaining gut physiology and

maintain a symbiotic relationship with the host (Johansson et al., 2015). In addition, mucosa-

associated microbes play a role in the initiation of immune responses via T cell activation, thus

contributing to cell mediated immunity. Any disruption of these interactions between host and

microbes can cause abnormalities in physiological homeostasis (Reigstad et al., 2015; Wang et al.,

2014) and thus may contribute to several human disorders including Inflammatory bowel disease

(IBD), cardiovascular disease, obesity, and ASD (Blumberg & Powrie, 2012, Luna et al., 2017). Different

compositions of bacterial species have been observed in faecal samples from children with autism (De

Angelis et al., 2015; Kang et al., 2013; Li et al., 2017). However, whether microbial populations change

at different locations along the gastrointestinal tract is largely unknown.

Little is known about the innervation of the mouse caecum and its role in autism-related gut

dysfunction. Therefore, a focal point of this research is to understand the role of the autism-associated

R451C mutation in the NLGN3 gene on enteric neurons and immune function in the mouse caecum.

It was also of interest to determine if microbial populations differed along the gastrointestinal tract in

wildtype and NL3R451C mutant mice.

The aims of this study are threefold.

To characterize the influence of the R451C mutation on caecal enteric neuron i)

organization,

To characterize the influence of the R451C mutation on caecal patch macrophage density ii)

and morphology

iii) To assess for alterations in gut microbes in different regions of the GI tract in the NL3R451C

mouse model of autism.

1.2 Genetics in ASD

It is now well established that ASD has a robust genetic impact. It is proposed that more than 50% of

factors contributing to the development of ASD are due to genetic variation (De Rubeis & Buxbaum,

2015; Hallmayer et al., 2011). Approximately 1000 genes are associated with ASD and most of these

genes play a vital role in biological pathways (De Rubeis et al., 2014; Iossifov et al., 2014; Iossifov et

al., 2012; Sanders et al., 2012). Among this large group of genes, rare single gene mutations can cause

impaired protein production and result in autism (Gottfried et al., 2015; Krumm et al., 2014) including

mutations altering synaptic adhesion proteins (However, due to phenotypic heterogeneity and

variable penetrance, only approximately 1% of autism patients exhibit single gene mutations

(Geschwind & State, 2015). Nevertheless, the study of these rare mutations can assist in

understanding the biological mechanisms underlying the core symptoms of ASD.

Mutations in synaptic genes are the most consistently stated genetic abnormalities in ASD and include

mutations in the neuroligin (NLGN) gene family (Gauthier et al., 2005; Jamain et al., 2003; Laumonnier

et al., 2004; Lawson-Yuen et al., 2008; Mackowiak, Mordalska, & Wedzony, 2014). There are five

neuroligin genes in humans: NLGN-1, 2, 3, 4X and 4Y (Bolliger et al., 2001), each with different

expression profiles. NLGN-1 is expressed on excitatory synapses, NLGN-2 on inhibitory and NLGN-3 is

found in both excitatory and inhibitory synapses. NLGN3 and NLGN4 are X-linked neuroligin genes and

NLGN4Y is located on the Y chromosome (Betancur et al., 2009; Jamain et al., 2003). The neuroligin

genes have been highlighted in contributing to neurodevelopmental disorders, as different studies

found NLGN3 and NLGN4 gene mutations in patients with familial autism, Asperger syndrome and X-

linked mental retardation (Jamain et al., 2003; Laumonnier et al., 2004; Yan et al., 2005).

1.2.1 The R451C mutation in Neuroligin-3

A missense mutation in NLGN3 gene (R451C) that occurs at a conserved amino acid sequence and

converts an arginine to a cysteine residue at position 451 was identified in two brothers diagnosed

with autism and with severe GI dysfunction (Hosie et al., 2019; Jamain et al., 2003). The R451C

mutation hinders the intracellular trafficking of NL3 and decreases NL3 protein expression at the cell

membrane by 90% (Ulbrich et al., 2016).

When introduced to mice, the Neuroligin-3R451C mutation causes a range of ASD relevant behaviours

including impaired social interactions, increased repetitive behaviour and elevated aggression

(Burrows et al., 2015; Hosie et al., 2019), enhanced spatial learning, along with a reduced startle

response compared to wild type (WT) mice (Chadman et al., 2008; Tabuchi et al., 2007). Therefore,

NL3R451C mice are well established as a useful preclinical model for studying ASD (Argyropoulos, Gilby,

& Hill-Yardin, 2013; Betancur et al., 2009).

1.3 The enteric nervous system

The enteric nervous system (ENS) of the GI tract comprises a network of 200-600 million neurons in

two plexuses; the myenteric plexus which is located between the longitudinal and circular muscle

layers of the intestine, and the sub-mucosal plexus which lies beneath the submucosa (Furness, 2012).

The ENS plays an important role in modulating the integrity of the mucosa of the gut lumen and cells

within the gut wall. The ENS communicates with the central nervous system (CNS) in a bidirectional

manner; however, it can regulate intestinal functions in the absence of external neural connections

from the CNS. The myenteric plexus predominantly regulates gut motility, whereas the sub-mucosal

plexus mainly contributes to the control of water movement and regulating electrolyte secretion in

the GI tract. However, interconnections between the two plexuses are important to modulate both GI

motility and secretion (Gwynne & Bornstein, 2007).

It is important to note that the ENS is a complex nervous system containing more than 20 subtypes of

neurons based on their various combinations of neurochemical content, firing activity and

morphology (reviewed in Furness, 2012). A common approach to visualise enteric neurons is to

fluorescently label the pan-neuronal marker protein; Hu in wholemount preparations. Nitric oxide

(NO) is the major inhibitory neurotransmitter in the ENS and approximately 45% of myenteric neurons

in the colon contain neuronal nitric oxide synthase (nNOS), a rate-limiting enzyme involved in the

synthesis of NO. Neuronal nitric oxide synthase immunoreactive neurons are also present in the

submucosal plexus in mice. In addition, enteric glial cells (EGCs) are traditionally thought to provide

mechanical support for enteric neurons and most importantly maintain intestinal homeostasis (Yu &

Li, 2014). Enteric glial cells are commonly labelled in the mouse gastrointestinal tract using antisera

targeting S100 beta and glial fibrillary acidic protein (GFAP). In both ENS and CNS, glial dysfunction

occurs in neurodegenerative diseases such as in Parkinson’s, ASD and Alzheimer’s disease (Edmonson,

Ziats, & Rennert, 2014; Fonseca et al., 2012; Joe et al., 2018; Lima et al., 2007). Changes in enteric glial

networks are correlated with gut inflammation (Geboes et al., 1992; Morales-Soto & Gulbransen,

2019). In glial cell injury as well as in irritable bowel disease (IBD) or infectious colitis (for example,

involving infection by Clostridium difficile bacteria) the expression of glial proteins can be altered,

causing glial cell dysfunction (Coelho-Aguiar Jde et al., 2015; von Boyen et al., 2011).

1.3.1 Altered neuro-immune interactions in NL3 R451C mice

The NL3R451C mutation found in ASD patients may alter ENS circuitry and neuronal communication

within the GI tract (Hosie et al., 2019). Therefore, it is important to explore potential neural

mechanisms that might contribute to GI disorders in animal models in order to understand the

etiology of this complex disorder. Using video imaging techniques and pharmacology to assess colonic

function ex vivo, data from the current laboratory showed subtle changes in colonic motility in NL3R451C

mice (Hosie et al., 2019). NL3R451C mouse proximal jejunal tissue labelled for the pan-neuronal marker,

Hu shows a 30% increase in the number of myenteric neurons per ganglion (Figure 1.1 ; Hosie et al.,

2019). NL3R451C mutant mice also show faster small intestinal transit compared to WT littermates

(Hosie et al., 2019).

The changes in the nervous system in NL3R451C mice and the potential for changes in caecal function

to affect immune responses highlights the need to investigate the histological structure and enteric

nervous system organisation of the caecum in mice expressing autism-associated R451C mutation in

NLGN3.

Figure 1.1 Myenteric ganglia of proximal jejunum in WT and NL3R451C. Confocal images (x 20) of Hu (red) and

nitric oxide synthase (NOS; green) staining. Density of neurons (labelled by the pan-neuronal marker, Hu) in the

proximal jejunum of wild-type (A) and NL3R451C mice (B). The graph below shows a higher density of total neurons

inhibitory neurotransmitter, nitric oxide, in the proximal jejunum of NL3R451C mice compared to wild-type mice

(Hosie et al., 2019).

1.4 Gut associated lymphoid tissue (GALT)

The intestine contains up to 70% of the body’s immunocytes that protect the host from microbial

invasion and is the largest mucosal surface of the body (Neutra, Mantis, & Kraehenbuhl, 2001; Jung,

Hugot, & Barreau, 2010). The mucosa-associated lymphoid tissues that line the gut are commonly

known as gut-associated lymphoid tissue (GALT) which is interfaced with the ENS (Mayer, 2011;

Mazzoli & Pessione, 2016). GALT includes lymphoid aggregates in the small intestine (Peyer’s patches),

the appendix (or caecum; caecal patches) and the large intestine and rectum (Koboziev, Karlsson, &

Grisham, 2010). In the healthy intestinal mucosa, mononuclear phagocytes comprising both

macrophages and dendritic cells (DC) are the most abundant population of leukocytes and play an

important role in maintaining homeostasis (Kühl et al., 2015). However, little is known about the

macrophages associated with gut-associated lymphoid tissue in the intestine such as the caecal patch

(Den Haan & Martinez-Pomares, 2013). Studies have reported altered numbers and morphology of

intestinal macrophages in diseased conditions such as IBD (Bain & Mowat, 2014; Mowat & Bain, 2011)

and macrophages are also known to play an important role in the pathogenesis of Crohn’s disease

(Smith et al., 2011). In this study, we aimed to investigate the effect of the autism-associated R451C

mutation in the Neuroligin-3 gene on macrophages in caecal patch tissue using the pan-macrophage

marker, Iba-1.

1.4.1 Caecal Patch

Caecal patches (CP) are lymphoid aggregates located at the blind end of the caecum and contain

enterocytes as well as microfold cells (M cells) and immunoglobulin A (IgA) secretory cells (Laurin,

Everett, & Parker, 2011). M cells within the caecal patch display longer and more irregular microvilli

than the surrounding enterocytes (Gebert & Bartels, 1991; Jepson et al., 1993; Jepson et al., 1992).

The primary role of M cells is to uptake and present antigens along with microorganisms to the

immune cells of lymphoid follicles to initiate an immune response (Corr, Gahan, & Hill, 2008). In the

peripheral region of the caecal patch (i.e. the follicle-associated epithelium; FAE), a significant

proportion of caecal patch M cells are not associated with lymphocytes (Bye, Allan, & Trier, 1984;

Gebert, Hach, & Bartels, 1992). Compared to the FAE of well-studied gut lymphoid aggregates of small

intestine known as Peyer’s patches, there are more goblet cells and a relatively diffuse distribution of

lymphocytes and M cells in the caecal patch FAE. The irregular nature of the caecal patch M cell

microvilli highlights potential variations in the interaction of microorganisms with M cells between the

caecal patch and Peyer’s patch (Jepson et al., 1992).

IgA plays an important role in maintaining gut homeostasis. Although both caecal and Peyer’s patches

contain IgA secreting cells, the destination of the secreted IgA differs. A study of germfree mice

colonized with intestinal commensal bacteria demonstrated that IgA secreted from Peyer’s patches is

restricted to the small intestine, whereas IgA secreted from the caecal patch travels to both large and

small intestine (Masahata et al., 2014).

It is thought that the caecum acts as a “safe house” for beneficial bacteria and provides protection in

case of contamination caused by pathogens. During gut dysfunctions resulting in diarrhoea, beneficial

bacteria are washed out along with pathogens as part of the diarrhoeal response. Microbes from the

caecum are thought to subsequently assist in rebuilding healthy microbial population and contribute

to maintaining host homeostasis (Laurin et al., 2011). In a study comparing germ free mice and specific

pathogen free mice, germ free mice showed a reduction in the size of the caecal patch, indicating that

the development of the caecal patch is influenced by commensal microbiota (Masahata et al., 2014).

1.5 Microbial influences on gut-associated lymphoid tissue

Bacterial colonization in the gut is crucial for the development and maturation of GALT (Vossenkämper

et al., 2013). Segmented filamentous bacteria are commensal bacteria that affect the development of

the immune system in rodents (Ericsson et al., 2014). These microbes promote immune tolerance via

polysaccharide signalling along with microbe-associated molecular patterns (MAMPs) and produce

short chain fatty acids (SCFA; Belkaid & Hand, 2014). Short chain fatty acids prevent colonization of

pathogens by stimulating the secretion of IgA, which plays an important role in immune responses at

mucosal surfaces of the GI tract (Belkaid & Hand, 2014; Kamada et al., 2013). Moreover, commensal

bacteria promote the expression of epithelial intestinal alkaline phosphatase (IAP) that plays a role in

detoxification of luminal lipopolysaccharides to prevent intestinal inflammation (Fawley & Gourlay,

2016; Ohland & Jobin, 2015).

Segmented filamentous bacteria (SFB) also play crucial roles in the development of T helper (Th) 17

cells via production of epithelial cytokines and the presentation of antigens to dendritic cells (Goto et

al., 2014). Th17 cells are an essential component of the host defence system via their role in the

eradication of pathogenic bacteria by producing proinflammatory cytokines (Curtis & Way, 2009;

Volpe, Battistini, & Borsellino, 2015). Each of these pathways are important in maintaining gut-

associated lymphoid tissue function in physiological conditions (Ohland & Jobin, 2015).

1.6 Microbiome-neuroimmune signatures in ASD

Various studies analysing stool specimens from patients with ASD revealed alterations in microbial

community composition compared to neurotypical individuals (De Angelis et al., 2013; Finegold et al.,

2010; Gondalia et al., 2012; Williams et al., 2011). Since ASD individuals commonly report GI disorders,

various studies have focused on the association between microbes and abdominal pain (Wasilewska

& Klukowski, 2015). A study of patient mucosal microbes showed that patients with both ASD and

FGID had increased Clostridiales and they are known to play a major role in regulating gut derived T

cell immunity (Atarashi et al., 2015; Ohnmacht et al., 2015).

Increased levels of inflammatory cytokines from peripheral blood have been reported in association

with ASD-relevant behaviours such as aggression, repetitive behaviour, language impairment and

social deficits (Ashwood et al., 2011). In a mouse model of maternal immune activation, inflammatory

cytokines IL17 and IL6 has been implicated and offspring showed ASD relevant phenotypes (Hsiao et

al., 2012; Wu et al., 2015). In addition, expression of IL6 was increased in samples from ASD patients

with FGID and abdominal pain (Luna et al., 2016).

Two Clostridium species (C. disporicum and C. tertium; that are reportedly associated with ASD and

abdominal pain are also significantly correlated with increased IL6 (Luna et al., 2016). IFN-ү is also

reportedly increased in ASD and associated with elevated levels of Clostridium bacteria. Overall, this

work suggested a potential link between clostridial species, abdominal pain in ASD and increased

inflammation (Luna et al., 2016).

1.7 Microbial changes in NL3R451C mice

Analysis of faecal samples from a small group of WT (n=4) and NL3R451C (n=5) mice cohoused in mixed

genotype groups showed changes in microbial content in mutant mice (Figure 1.2; Hosie et al., 2019).

Unique operational taxonomic units (OTUs; which give an indication of microbial species diversity)

characterized as those present in one genotype but not the other, were identified at both five and

nine weeks of age (Figure 1.2). At 5 weeks of age (Figure 1.2A and C), presence/absence analyses

showed significant grouping of microbial populations between WT and NL3 R451C mice (Figure 1.2C)

whereas this structural change was not evident at 9 weeks of age in the same mice (Figure 1.2B and

D), which could be due to the fact that these cohoused mice are coprophagic, therefore microbial

profiles between genotypes became more similar over time. Next generation sequencing of 16S

ribosomal RNA was utilised to identify individual microbial species contributing to differences in the

microbial community. These next-generation sequencing data revealed that key OTUs belonging to

the family Lachnospiraceae (phylum Firmicutes) drive the structural differences observed between

WT and NL3R451C mice at five weeks of age (data not shown, (Hosie et al., 2019), however further

research to confirm these changes in a larger dataset and at different regions along the

gastrointestinal tract are needed.

Figure 1.2 The effect of genotype on faecal microbial communities at five and nine weeks of age of WT and

NL3R451C mice. Presence and absence analysis (A-B) and non-metric multidimensional scaling (NMDS) ordinations

(C-D) of ARISA (automated method of ribosomal intergenic spacer analysis, for the rapid estimation of the

microbial diversity of environmental samples) generated microbial communities at five (A-C) and nine (B-D)

weeks of age. Significant grouping was observed between WT and NL3R451C at five weeks of age (p < 0.01). Dashed

line showing genotype effect; decreasing stress values indicate increasing goodness of fit (Hosie et al., 2019).

Each data point in the NMDS ordinations represents an individual animal.

1.8 PROJECT RATIONALE

1.8.1 Abnormal neuroimmune pathways in NL3R451C mice

Preliminary data from the current laboratory revealed a significant reduction in the caecal weight of

NL3R451C mutant mice bred on a mixed genetic background (B6; 129-NLGN3tm1Sud/J mice maintained

in excess of 10 generations on a hybrid Sv129/C57Bl6 background) compared to wild type littermates.

Based on the proposed role of the caecum in replenishing beneficial bacteria in the context of

infection, this primary finding suggests an alteration in inflammatory pathways could be present in

mice expressing the R451C mutation in the gene encoding Neuroligin-3. Preliminary evidence from

RNASeq (n=3 WT and n=3 NL3R451C mouse colon and jejunum samples, not shown) and histological

analyses (increased numbers of broken villi in the jejunum; Seger et al., in preparation for publication)

also suggest an altered neuronal and microbial status in the GI tract of NL3R451C mice bred on a mixed

genetic background. In addition, increased neuronal numbers were observed in the small intestine of

NL3R451C mice compared with wild type littermates but the structure of the enteric nervous system in

caecal tissue was not examined. It is important to study the effects of genetic mutations on mice bred

from different genetic background strains as the presence and severity of phenotypes can differ

significantly. Whether the observed increase in small intestinal neuronal numbers, the reduction in

caecal weight and altered villus structure phenotypes persist when NL3R451C mice are bred on a pure

C57/Bl6 genetic background is not known.

1.8.2 Alterations in gut microbes in ASD

It has been suggested that alterations in inflammatory pathways may impact gut microbes and

subsequently disrupt the symbiotic relationship between the host and bacterial species (e.g. Fung,

Olson, & Hsiao, 2017). Therefore, this study will assess for microbial alterations at different GI regions

including the ileum, caecum and colon in wild type and NL3R451C mice. This approach will serve to

confirm if the NL3R451C mutation specifically affects microbial content in the mucosa of the GI tract and

will further provide an overview of interactions of the enteric nervous system with microbiota.

1.9 HYPOTHESIS AND AIMS

The overall hypothesis of this project is that the distribution of neurochemical and immune pathway

markers is altered in the caecum and that gut microbial populations differ in NL3R451C compared to wild

type mice.

The main aims of the project are:

Aim 1: To assess whether the number and proportions of enteric neurons are altered in the NL3R451C

caecum compared to wild type mice using immunofluorescence techniques.

Aim 2: To determine if the density and morphology of macrophages in caecal patch gut-associated

lymphoid tissue (GALT) is altered in NL3R451C mice using immunohistochemical and histological

techniques.

Aim 3: To assess for changes in microbial communities in NL3R451C mice by using deep sequencing

analysis techniques.

Findings from this study will contribute to the general understanding of microbial-host interactions

and will assist in identifying potential therapeutic targets within the gut-brain axis of ASD patients.

Chapter 2

Altered caecal neuroimmune interactions in the

Neuroligin-3R451C mouse model of autism

2.1 Abstract

The intrinsic nervous system of the gut interacts with the gut-associated lymphoid tissue (GALT) via

bidirectional neuroimmune interactions. The caecum is an understudied region of the gastrointestinal

(GI) tract that houses a large supply of microbes and is involved in generating immune responses. The

caecal patch is a lymphoid aggregate located within the caecum that regulates microbial content and

immune responses. People with Autism Spectrum Disorder (ASD; autism) experience serious GI

dysfunction, including inflammatory disorders, more frequently than the general population. Autism

is a highly prevalent neurodevelopmental disorder defined by the presence of repetitive behaviour or

restricted interests, language impairment, and social deficits. Mutations in genes encoding synaptic

adhesion proteins such as the R451C missense mutation in neuroligin-3 (NL3) are associated with

autism and impair synaptic transmission. We previously reported that NL3R451C mice, a well-

established model of autism, have altered enteric neurons and GI dysfunction. However, whether the

autism-associated R451C mutation alters the caecal enteric nervous system and immune function is

unknown. We assessed for gross anatomical changes in the caecum and quantified the proportions of

caecal submucosal and myenteric neurons in wild-type and NL3R451C mice using immunofluorescence.

In the caecal patch, we assessed total cellular density as well as the density and morphology of Iba-1

labelled macrophages to identify whether the R451C mutation affects neuro-immune interactions.

NL3R451C mice have significantly reduced caecal weight compared to wild-type mice, irrespective of

background strain. Caecal weight is also reduced in mice lacking Neuroligin-3. NL3R451C caecal ganglia

contain more neurons overall and increased numbers of nitric oxide (NO) producing neurons (labelled

by nitric oxide synthase; NOS) per ganglion in both the submucosal and myenteric plexus. Overall

caecal patch cell density was unchanged. However, NL3R451C mice have an increased density of Iba-1

labelled enteric macrophages. Macrophages in NL3R451C were smaller and more spherical in

morphology. Here, we identify changes in both the nervous system and immune system caused by an

autism-associated mutation in NLGN3 encoding the postsynaptic cell adhesion protein, Neuroligin-3.

These findings provide further insights into the potential modulation of neural and immune pathways.

2.2 Introduction

Emerging evidence suggests that altered communication between the nervous system and

inflammatory pathways is associated with multiple diseases including autism. Both altered

inflammatory activity (Wei et al., 2011) and a maternal history of autoimmune diseases, such as

rheumatoid arthritis and celiac disease, is associated with an increased risk of autism (Atladottir et al.,

2010). The gut-associated lymphoid tissue (GALT) plays a crucial role in mucosal immunity and

microbial populations. Caecal patches are lymphoid aggregates located at the blind end of the caecum

and contain various immune cells such as macrophages and dendritic cells (Masahata et al., 2014).

The precise role of the caecum is unclear, but it has been suggested that the appendix in humans

houses a ‘reserve population’ of commensal microbes (Randal et al., 2007). The caecal patch

contributes to gut homeostasis and is a major site for the generation of IgA-secreting cells that

subsequently migrate to the large intestine (Masahata et al., 2014). Secretory IgA plays an important

role in regulating the activities and compositions of commensal bacteria populations in animal models

(Fagarasan et al., 2002; Peterson et al., 2007; Strugnell & Wijburg, 2010; Suzuki et al., 2004). However,

whether caecal innervation and immune function is altered in preclinical models of neural disorders

is unknown.

Autism is a neurodevelopmental disorder affecting 1 in 59 children (Baio et al., 2018; Loomeset al.,

2017). In many autism patients core features such as impairments in social interaction,

communication and repetitive and/or restrictive behaviours are present along with immunological

dysfunction (Marchezan et al., 2018) and gastrointestinal (GI) disorders (Buie et al., 2010; Coury et al.,

2012; Valicenti-McDermott et al., 2006). Individuals with autism are four times more likely to

experience frequent GI symptoms including alternating diarrhoea and constipation, and abdominal

pain compared to children with typical development (McElhanon et al., 2014). Interestingly,

Inflammatory bowel disease (IBD) is present at significantly higher rates in people with autism than

the general public (Kohane et al., 2012). Autism-associated GI dysfunction includes increased GI

permeability along with altered motility (Horvath & Perman, 2002; Kohane et al., 2012; Neuhaus et

al., 2018; Parracho et al., 2005). Mice expressing the Neuroligin-3 R451C mutation exhibit autism-

relevant behaviours including impaired social interaction (Etherton et al., 2011; Tabuchi et al., 2007),

a heightened aggression phenotype (Burrows et al., 2015; Hosie et al., 2019), impaired communication

(Chadman et al., 2008) and increased repetitive behaviours (Rothwell et al., 2014). Furthermore, the

robust aggression phenotype in these mice is rescued by a clinically relevant antipsychotic, risperidone

(Burrows et al., 2015), highlighting that this model is useful for preclinical studies. These mice also

show altered GI motility, in line with the notion that alterations in the nervous system may also affect

the ENS to result in GI dysfunction (Gershon & Ratcliffe, 2004; Hosie et al., 2019).

Most research to date in animal models of autism has focused on replicating the core traits of ASD, in

addition to using invasive techniques to highlight changes in neural network activity in the brain

(Etherton et al., 2011; Halladay et al., 2009; Hosie et al., 2019; Lonetti et al., 2010; Patterson, 2011;

Schmeisser et al., 2012; Tabuchi et al., 2007; Varghese et al., 2017). Using these approaches, it is well

established that many gene mutations identified in autism patients affect neuronal function. Here we

assessed whether the autism-associated R451C mutation in Neuroligin-3 affects gross caecal

morphology, enteric neuronal populations or immune cells within the caecal patch.

2.3 Methodology

Animals

Adult male NL3R451C mice (8-14 weeks old) and wild type (WT) littermate controls from 2 different

colonies were used in this study. Neuroligin 3 knockout mice (NL3-/-; 12 weeks old) were also

examined. NL3R451C mutant mice (B6;129-NLGN3tm1Sud/J) were originally obtained from Jackson

Laboratories (Bar Harbour, Maine USA) and maintained on a mixed background (mbNL3R451C) at the

Biomedical Sciences Animal Facility, The University of Melbourne (Hosie et al., 2019). These mice

were then backcrossed onto a C57BL6 background for more than 10 generations (i.e. B6NL3R451C mice)

and maintained at the animal facility at RMIT University, Bundoora, Australia. In contrast, NL3-/- mice

(Radyushkin et al., 2009; Leembruggen et al., 2019) were bred on a C57Bl/6NCrl background at the

Florey Institute of Neurosciences and Mental Health. All NL3R451C mice were culled by cervical

dislocation in accordance with RMIT University and The University of Melbourne animal ethics

guidelines (AEC# 1727, AEC#1513519). NL3-/- mice were anesthetized, and fresh brain tissue was

immediately collected for other applications (AEC# 14095), therefore body weight data for these

animals were unavailable. Mouse body weights for B6NL3R451C and mbNL3R451C mice were recorded

prior to dissection. All data from mutant mice were compared with matched WT littermate controls

from the respective cohorts to remove environmental and additional genetic factors (i.e. data from

mbNL3R451C animals were compared with mbWT mice; B6NL3R451C versus B6WT mice and C57Bl/6NCrl

NL3-/- mice versus C57Bl/6NCrl WT littermates).

Caecal collection

The caecum containing its content was collected and weighed from B6NL3R451C, mbNL3R451C, NL3-/-, and

WT mice. The caecum from each mouse was then opened and pinned with the mucosa facing upwards

and submerged in 0.1M PBS on a petri dish lined with sylgard (Sylgard Silicone Elastomer, Krayden

Inc., USA), enabling visualization of the lymphoid patch (i.e. the caecal patch). Images of caecal tissue

with a measuring scale were captured and caecal area measured using ImageJ software (ImageJ 1.52a,

NIH, USA).

Wholemount tissue preparation

Caecal myenteric and submucosal plexus neurons were revealed by microdissection using fine forceps

and dissecting spring scissors. The submucosal plexus was revealed by removing the mucosal layer

and carefully exposing neurons adjacent to the circular muscle within the caecal tissue. To obtain the

myenteric plexus, the circular muscle was then peeled away from the remaining caecal tissue. A small

area of tissue (approximately 0.5 cm2) adjacent to the caecal patch of the caecal tip containing

myenteric and submucosal plexuses was transferred into a small Petri dish, submerged in 0.1M PBS

for labelling by immunofluorescence.

Wholemount immunofluorescence for neuronal populations

Immunofluorescence staining was performed on wholemount caecal tissue samples to assess for

potential differences in neuronal cell numbers between NL3R451C and WT mice. Wholemount samples

of myenteric and submucosal plexus were permeabilized in 0.01% TritonTM X-100 and blocked with

10% CAS-block (Invitrogen Australia, Mt Waverley, Australia) for 30 min at room temperature (RT) to

reduce non-specific binding of antibodies. Tissues were then incubated with 30 µl primary antisera;

human anti Hu (1:5000, a pan-neuronal marker; a gift from Dr. V. Lennon, Mayo Clinic, USA) and sheep

anti- neuronal nitric oxide synthase (NOS; 1:400; Abcam, USA) and kept at 4°C overnight in a sealed

container. After incubation, caecal tissues were washed with 0.1 M PBS (3 washes of 10 min duration).

Then, 30 µl of secondary antisera (Donkey anti-human, 1:750; Jackson, ME, USA and Donkey anti

sheep, 1:400; Invitrogen) were applied to the samples and left for 2.5h at RT on a shaker incubator

(Digital Shaking Incubator OM11, Ratek, Australia). Caecal tissues were mounted using fluorescence

mounting medium (DAKO Australia Pty. Ltd.; Botany, Australia).

Imaging of caecal neuronal populations

Images of caecal tissue containing the submucosal and myenteric plexuses were obtained using a

confocal electron microscope (Nikon Confocal Microscope: A1; Version 4.10) at 10X and 60X Apo

(water) magnification and later analysed using ImageJ (ImageJ 1.52a, NIH, USA) and Imaris software

(Imaris x64 9.1.0; Bitplane AG, UK). 10 myenteric ganglia and 10 submucosal ganglia were selected

from each wholemount caecal tissue sample (n=5 NL3R451C and n=5 WT samples). From each ganglion,

the number of Hu and NOS stained cells were counted.

Caecal patch tissue collection

Caecal tissues including caecal patch samples were fixed in 4% formaldehyde solution at 4°C overnight.

The next day, tissue samples were washed three times (10 min per wash) with filtered 0.1M PBS. The

caecal patch was excised from the caecal tissue using spring scissors. Caecal patch samples were

subsequently placed into a 30% sucrose solution in distilled water overnight at 4°C for cryoprotection.

Caecal patches were placed in a cryomold (Tissue-Tek Cryomold, Sakura, Finetek, USA) filled with

optimal cutting temperature compound (Tissue-Tek, OCT compound, Sakura, Finetek, USA).

Cryomolds containing caecal patch samples were then snap frozen using liquid nitrogen and tissue

blocks stored at -80°C. Frozen caecal patch samples were sectioned at 6-micron thickness using a

cryostat (Leica CM1950 Clinical Cryostat, Leica Biosystems Nussloch GmbH, Germany) and collected

on poly-lysin coated microscopic slides (Thermo Scientific, Menzel-Glaser, SuperfrostR plus, New

Hampshire, USA and stained with Haematoxylin & Eosin (H&E) to assess for overall cell density.

Caecal patch image analysis

Images were obtained using an Olympus slide scanner microscope (VS120-S5; Olympus Australia Pty.

Ltd.; Melbourne, Australia) and the cell density within the caecal patch was analysed using ImageJ

software (ImageJ v1.52a, NIH, USA). The entire area of each caecal patch was selected to calculate

the area of the caecal patch and cell numbers within that area. The total number of cells was then

divided by the area of interest to calculate the number of cells per 100 µm2.

Caecal patch immunofluorescence

Immunofluorescence was also performed on cross sections of caecal patch tissue samples to assess

for altered density and morphology of macrophages. To observe a subpopulation of immune cells

within the caecal patch, immunofluorescence for the immune cell marker Iba-1 (1:3000, Abcam, USA)

was conducted (Mikkelsen et al., 2017). Sections were incubated for 30 min with 0.1% triton and 10%

CAS-block at RT. 30 µl of primary antibody was subsequently applied to each section and kept at 4°C

overnight in a moisture sealed container. After incubation, caecal patch sections were washed with

0.1 M PBS (3 x 10 min washes). Secondary antiserum (30 µl; Donkey anti-rabbit, 1:400) was applied to

the samples and left for 2.5h at RT on a shaker incubator. Caecal sections were mounted using

fluorescence mounting medium (DAKO Australia Pty. LTD.; Botany, Australia) containing DAPI (4’, 6-

diamidino-2-phenylindole) and stored at 4°C overnight. Tissue samples were imaged using a confocal

electron microscope (Nikon Confocal Microscope: A1; Version 4.10). A Z-series of images of caecal

patch sections (30 µm thickness) were captured and saved in ND2 file format. Imaris software (Imaris

x64 9.1.0; Bitplane AG, UK) was used for 3D cellular reconstruction of Iba-1 labelled macrophages.

Statistical analysis

Potential statistical differences between groups were identified using Student’s t-tests in GraphPad

Prism 8.1.2.

2.4 Results

Mouse body weight, caecal weight and caecal tissue area were assessed to determine if anatomical

changes occur in the presence of the autism-associated R451C mutation in mice. To address whether

the R451C mutation and the NLGN3 gene itself plays a broader role in caecal weight, ceacae from

NL3R451C mice bred on two different background strains were weighed, and caecal weights from mice

lacking NLGN3 compared to WT littermates were also compared.

The average body weight of WT (n=39) and NL3R451C (n=34) mice was similar (26.38 ± 0.4 g and 26.46

± 0.4 g, WT and NL3R451C respectively; p = 0.88; Figure 2.1A). To determine if the R451C mutation

affects caecal structure in mice, the fresh caecal weight from 38 WT and 36 NL3R451C mice was

recorded. NL3R451C caecae were significantly lighter than WT (0.65 ± 0.02 g and 0.54 ± 0.01 g, WT and

NL3R451C respectively; p=0.0001; Figure 2.1B). To determine if the reduction in NL3R451C caecal weight

was due to a reduction in the size of the caecum itself, total caecal tissue area was measured. No

difference between the caecal area of WT (n=15) and NL3R451C (n=16) mice was observed (7.99 ± 0.36

and 7.75 ± 0.5 cm2, respectively; p=0.51; Figure 2.1C). A role for the NLGN3 gene in influencing caecal

weight is supported by similar observations in NL3R451C mice bred on a mixed background strain and in

NLGN3-/- (NL3-/-) mice in which the NLGN3 gene is deleted. In mice expressing the R451C mutation

bred on a mixed background (mb) strain, the average body weight was similar (28.11 ± 1.01g and 27.1

± 0.9 g, WT and mbNL3R451C n=16 and n=21, respectively; p = 0.30; Figure 2.1D). Caecal weight was

also reduced in mb strain mutant littermates (0.69 ± 0.11 g, 0.49 ± 0.28 g; WT (n = 14) and mbNL3R451C

(n = 21), respectively; p<0.0001; Figure 2.1E). Similar to data from both the C57/Bl6 and mb strains of

NL3 R451C mice, KO (NL3-/-) mice also revealed a reduction in caecal weight (1.16 ± 0.5 g and 0.61 ± 0.53

g; WT and NL3-/-, respectively, n = 8 in each group; p=0.02; Figure 2.1F). These findings suggest a role

for the NLGN3 gene in regulating caecal weight in mice.

Figure 2.1 Body weight, caecal weight, and caecal tissue area. A: Pure C57/Bl6 background WT and

NL3R451C mice (red) show similar body weights. WT (n=39) and NL3R451C (n=34) mice; p=0.88. B: Similar

caecal tissue area for WT (n=15) and NL3R451C (n=16) mice. C: Caecal weight is reduced in NL3R451C pure

background mice; WT (n=38) and NL3R451C (n=36) mice. D: In mixed background mice (orange), no

differences in body weight were found. WT (n=16) and NL3R451C (n=21) mice; p=0.30. E: Caecal weight

is also reduced in NL3R451C (orange) mixed background mice. WT (n=14) and NL3R451C (n=21) mice). F:

Reduced caecal weight in NL3-/- (green) mice. WT (n=8) and NL3-/- (n=8) mice. Students t-test *p<0.05;

****p<0.0001. Each symbol indicates an individual mouse. Mixed background mice were bred on a

mixed Imj/Sv129/C57/Bl6 genetic background; KO: NL3-/- mice (bred on C57Bl/6NCrl mice).

To investigate whether the NL3R451C mutation alters neural populations in the caecal submucosal and

myenteric plexus, immunofluorescence for the pan-neuronal marker Hu; and NOS (which labels a

significant subset of enteric neurons) was conducted. Wholemount preparations of WT (Figure 2.2A-

C) and NL3R451C (Figure 2.2D-F) submucosal plexus were labelled with Hu and NOS to quantify neuronal

subpopulations. The total number of neurons (i.e. labelled by Hu) per submucosal ganglion was

increased in NL3R451C mice (5 ± 1 and 6 ± 1 neurons, WT and NL3R451C, respectively; p=0.04; Figure 2.2G).

Similarly, NL3R451C mice showed increased numbers of NOS immunoreactive neurons per ganglion (2 ±

1 and 3 ± 1 cells; WT and NL3R451C respectively, n=5 in each group; p=0.02; Figure 2.2H). In submucosal

neurons, there was a trend for an increased percentage of NOS neurons per ganglion in WT and

NL3R451C mice (44 ± 0.4% and 56 ± 0.4%; WT and NL3R451C respectively; p=0.06; Figure 2.2I).

Figure 2.2 Caecal submucosal neuronal numbers and proportions. WT caecal submucosal plexus ganglia

labelled for Hu (A) NOS (B); and overlap illustrated in merged image (C). NL3R451C caecal submucosal plexus ganglia labelled for Hu (D) and NOS (E); merge (F). Scale bar = 20µm. G: The total number of Hu labelled neurons per ganglion. H: The total number of NOS stained cells per ganglion. I: Proportions of NOS stained neurons/ ganglion. Each symbol indicates an individual mouse. Immuno-labelled neurons were shown by

arrows. Bars in boxplots indicate the mean and range of the data. Students t-test *p<0.05.

Wholemount preparations of WT (Figure 2.3A-C) and NL3R451C (Figure 2.3D-F) myenteric plexus were

labelled with Hu and NOS. Similar to findings in the submucosal plexus, more myenteric neurons

(labelled for Hu) were seen in NL3R451C mice (11 ± 1 and 15 ± 1 neurons/ganglion, WT and NL3R451C

respectively, p=0.005; Figure 2.3G). The number of NOS stained caecal myenteric neurons per ganglion

was also increased in NL3R451C mice (5 ± 1 and 8 ± 1 neurons/ganglion, WT and NL3R451C, respectively,

n=5 in each group; p=0.008; Figure 2.3H). The percentage of NOS stained neurons per myenteric

ganglion was also increased in NL3R451C mice (42 ± 0.7 % and 57 ± 0.2 %; WT and NL3R451C respectively;

p=0.008; Figure 2.3I). These data show that the R451C mutation results in increased numbers of caecal

submucosal and myenteric neurons in mice.

Figure 2.3 Caecal myenteric neuronal numbers and proportions. WT caecal submucosal plexus

ganglia labelled for A: Hu (green), B: NOS (red) and C: merge. NL3R451C caecal submucosal plexus

ganglia labelled for D: Hu (green) E: NOS (red), F: merge. Scale bar = 20 µm. G: The number of Hu+

neurons/ganglion. H: The number of NOS immunoreactive neurons/ganglion. I: The percentage of

NOS neurons/ganglion. Each symbol indicates an individual mouse. Immuno-labelled neurons

indicated by arrows. Bars in boxplots indicate the means and range of the data. Students t-test

*p<0.05; **p<0.005.

To assess whether the R451C mutation alters GALT structure, we measured total cell density in

Haematoxylin and Eosin stained cross sections of the caecal patch of WT (Figure 2.4A, B) and NL3R451C

(Figure 2.4C, D) mice. Caecal patch cellular density was similar in both genotypes (1276 ± 48 and 1428

± 22 cells/100 µm2, WT and NL3R451C mice respectively; p=0.28; Figure 2.4E).

Figure 2.4 Caecal patch cell density. Haematoxylin and Eosin stained transverse section of caecal

patches from WT (A-B) and NL3R451C (C, D) mice. E: There was no difference in overall caecal patch cell

density in WT (n=8) and NL3R451C mice (n=8). Each symbol indicates an individual mouse. Bars in

boxplots indicate the mean and range of the data.

Caecal patch samples were labelled with the pan-nuclear marker, DAPI, and a pan-macrophage

antiserum targeting the ionised calcium-binding adaptor molecule 1 (Iba-1) to determine whether the

R451C mutation affects these immune cells in WT (Figure 2.5A-D) and NL3R451C (Figure 2.5E-H) within

the caecal patch. NL3R451C caecal patch tissue had a higher density of Iba-1 stained cells (29 ± 1

cells/100 µm2) compared to WT mice (24 ± 1 cells/100 µm2; p=0.032; Figure 2.5I). The volume of Iba-

1 stained cells in WT was larger than in NL3R451C mice (946.9 ± 0.4 µm3 and 559.7 ± 0.4 µm3; WT and

NL3R451C, respectively; p=0.004; Figure 2.5J). Iba-1 stained cells in NL3R451C mice showed a trend

towards increased sphericity (0.6 ± 0.02 and 0.8 ± 0.05 arbitrary units; WT and NL3R451C, respectively;

p=0.06; Figure 2.5K). These results suggest that the autism-associated R451C mutation in NLGN3

alters macrophage density and morphology within the caecal gut-associated lymphoid tissue.

Figure 2.5 Caecal patch macrophage density and morphology. WT caecal patch tissue labelled for A:

DAPI and B: Iba-1, C: merge; D: 3-D reconstruction of Iba-1 labelled cell morphology. NL3R451C caecal

patch tissue labelled for E: DAPI, F: Iba-1, G: merge, H: 3-D reconstruction of Iba-1 labelled cell

morphology. I: Density of Iba-1 stained cells in WT and NL3R451C caecal patch tissue. J: Volume of Iba-1

stained cells in WT and NL3R451C caecal patch tissue. K: Sphericity of Iba-1 stained cells in WT and

NL3R451C mice. Each symbol indicates an individual mouse. Bars in boxplots indicate the mean and

range of the data. Students t-test *p<0.05; **p<0.005. Scale bars = 20 µm.

2.5 Discussion

The nervous system and the immune system are in constant bidirectional communication (reviewed

in Margolis et al., 2016). Altered immune responses and gut dysfunction commonly occur in individuals

genetically susceptible to autism (Coury et al., 2012). Altered neuronal communication in autism

(Betancur et al., 2009; Grabrucker et al., 2011; Huguet et al., 2016), likely contributes to changes in

the peripheral nervous system, and therefore GI function (Hosie et al., 2019; Leembruggen et al.,

2019).

A main finding from this study is the clear reduction of caecal weight in mice expressing the Neuroligin-

3 R451C mutation. Importantly, in addition to our findings in mice bred on a pure C57/Bl6 genetic

background, caecal weight was also reduced in mice bred on a mixed background in a different animal

facility. These findings therefore confirm a persistent effect of the gene mutation and rule out genetic

susceptibility due to background strain or environment. Furthermore, mice lacking Neuroligin-3

expression (NL3-/- mice) that were bred in a third animal facility, and therefore experienced a different

environment to the two NL3R451C strains, also had reduced caecal weight. Together, these findings

suggest that the NLGN gene plays a role in caecal neuroimmune physiology and that the reduction in

weight is unlikely solely due to diet, microbial populations, and other environmental factors. A

reduction in caecal weight has also been reported in mouse models of obesity. Obese mice fed a high

fat diet had caecal weights approximately 50% less than controls, and this reduction was restored by

antibiotic treatment (Soto et al., 2018). Since obesity is associated with increased inflammation, our

observations in NL3R451C mice might also indicate elevated inflammatory cytokine levels, which remain

to be assessed.

The reduced caecal weight in NL3R451C mice may indicate changes in caecal mucus thickness. The

hydrophilic mucus layer that coats the GI tract plays an important role in innate host defence (Mowat,

2003). Changes in the mucus thickness could contribute to an altered immune response in the host

organism (Lievin-Le Moal & Servin, 2006; McGuckin, et al., 2009). Accordingly, altered mucus thickness

along the GI tract may contribute to GI dysfunction which is commonly observed in children with

autism. Based on studies in preclinical models of other disorders, aberrant mucus production may be

present alongside other phenotypic traits. For example, caecal tissue sampled from a mouse model of

stroke (72h after brain injury) showed decreased numbers of mucus-producing goblet cells compared

to sham-treated mice (Houlden et al., 2016). Reductions in goblet cell number and size were also

reported in mice during the development of ulcerative colitis (Johansson et al., 2008; Van der Sluis et

al., 2006). Although potential changes in caecal weight were not correlated with these observations,

a thinning of the adherent mucus layer and reduced total mucus volume within the caecum may

contribute to the significant reduction in caecal weight in NL3R451C mutant mouse strains identified

here.

The enteric nervous system (ENS) regulates GI motility and secretion, as well as nutrient uptake and

gut immune and inflammatory processes (Goyal & Hirano, 1996). The two main cell populations of the

ENS are neurons and enteric glial cells (EGCs) (Jessen, 2004). Many studies have identified enteric

neuron pathologies in the context of inflammatory disease (Boyer et al., 2007; Li et al., 2016; Marlow

& Blennerhassett, 2006; Rahman et al., 2015; Talapka et al., 2014; Winston, Li, & Sarna, 2013), but

how alterations in the ENS might affect inflammatory pathways remains largely unknown.

Nevertheless, altered neuronal activity has previously been implicated in altering immune function,

where reports investigating NO levels in human colonic and rectal mucosal biopsies in active ulcerative

and Crohn’s disease showed elevated expression of nitric oxide synthase (NOS) (Ljung et al., 2006;

Rachmilewitz et al., 1995).

Changes in enteric neuron numbers are reported in animal models demonstrating GI dysfunction (de

Fontgalland et al., 2014; Schneider et al., 2001; Hosie et al., 2019). Our findings that both submucosal

and myenteric neuron numbers are increased in NL3R451C mice caecal tissue indicate that the R451C

mutation likely alters neuron populations during development. These results are in agreement with

our previous report showing increased jejunal neuron numbers in adult NL3R451C mice bred on a mixed

genetic background (Hosie et al., 2019). In addition to a potential developmental effect, these findings

suggest that the NL3R451C mutation may influence caecal function. Specifically, we speculate that the

R451C mutation could alter the rhythmic caecal ‘churning’ of waste that occurs post digestion and

prior to expulsion via the colon. However, this hypothesis remains to be investigated. The contractile

activity of the GI tract is neurally regulated so given that the R451C mutation is expressed in the gut

in these models (Hosie et al., 2019), it would indeed be of interest to assess whether NL3R451C mice

show altered caecal motility.

In addition to characterising changes in enteric neuronal populations in NL3R451C mice, we investigated

effects of the autism-associated R451C mutation on macrophages in caecal tissue using the pan-

macrophage marker, Iba-1. NL3R451C mice showed increased numbers of Iba-1 stained cells in caecal

patch tissue compared to WT mice. In addition, the volume of Iba-1 immunoreactive cells was

decreased and more spherical in NL3R451C mutant mice compared to WT littermates. These findings

could indicate that macrophages within NL3R451C caecal patch tissue are present in a more reactive

state compared with in WT mice, with potential implications for immune pathways in this model.

Similar observations were reported in disease conditions such as IBD, where both the number and

morphology of intestinal macrophages are altered (Bain & Mowat, 2014; Mowat & Bain, 2011).

Moreover, macrophages are integral to the pathogenesis of Crohn’s disease (Smith et al., 2011).

2.6 Conclusions

This is the first study to assess the impact of the Neuroligin-3 R451C mutation on caecal structure at

both an anatomical and cellular level in mice. The observation that the Neuroligin-3 gene plays a role

in regulating caecal weight across multiple genetic backgrounds and environments identifies a new

role for the NLGN3 gene in mice. This work also highlights the caecum as a region of interest within

the GI tract that may play a central role in modulating neuro-immune interactions. In the context of

neurodevelopmental disorders, our findings that an autism-associated mutation that affects nervous

system function also impacts gut-associated lymphoid tissue have implications for identifying novel

interactions between the enteric nervous system, microbes located within the gut lumen, immune

pathways and potential therapeutic targets for GI dysfunction.

2.7 References

Atladottir, H. O., Thorsen, P., Ostergaard, L., Schendel, D. E., Lemcke, S., Abdallah, M., & Parner, E. T. (2010). Maternal infection requiring hospitalization during pregnancy and autism spectrum disorders. J Autism Dev Disord, 40(12), 1423-1430. doi:10.1007/s10803-010-1006-y Bain, C. C., & Mowat, A. M. (2014). Macrophages in intestinal homeostasis and inflammation. Immunological reviews, 260(1), 102-117. doi:10.1111/imr.12192

Baio, J., Wiggins, L., Christensen, D. L., Maenner, M. J., Daniels, J., Warren, Z., . . . Dowling, N. F. (2018). Prevalence of Autism Spectrum Disorder Among Children Aged 8 Years - Autism and Developmental Disabilities Monitoring Network, 11 Sites, United States, 2014. MMWR Surveill Summ, 67(6), 1-23. doi:10.15585/mmwr.ss6706a1

Betancur, C., Sakurai, T., & Buxbaum, J. D. (2009). The emerging role of synaptic cell-adhesion pathways in the pathogenesis of autism spectrum disorders. Trends Neurosci, 32(7), 402-412. doi:10.1016/j.tins.2009.04.003

Boyer, L., Sidpra, D., Jevon, G., Buchan, A. M., & Jacobson, K. (2007). Differential responses of VIPergic and nitrergic neurons in paediatric patients with Crohn's disease. Auton Neurosci, 134(1-2), 106-114. doi:10.1016/j.autneu.2007.03.001

Buie, T., Campbell, D. B., Fuchs, G. J., Furuta, G. T., Levy, J., VandeWater, J., . . . Winter, H. (2010). Evaluation, Diagnosis, and Treatment of Gastrointestinal Disorders in Individuals With ASDs: A Consensus Report. Pediatrics, 125(Supplement 1), S1. doi:10.1542/peds.2009-1878C Burrows, E. L., Laskaris, L., Koyama, L., Churilov, L., Bornstein, J. C., Hill-Yardin, E. L., & Hannan, A. J. (2015). A neuroligin-3 mutation implicated in autism causes abnormal aggression and increases repetitive behavior in mice. Mol Autism, 6, 62. doi:10.1186/s13229-015-0055-7 Bush, T. G. (2002). Enteric glial cells. An upstream target for induction of necrotizing enterocolitis and Crohn's disease? Bioessays, 24(2), 130-140. doi:10.1002/bies.10039 Cabarrocas, J., Savidge, T. C., & Liblau, R. S. (2003). Role of enteric glial cells in inflammatory bowel disease. Glia, 41(1), 81-93. doi:10.1002/glia.10169

Chadman, K. K., Gong, S., Scattoni, M. L., Boltuck, S. E., Gandhy, S. U., Heintz, N., & Crawley, J. N. (2008). Minimal aberrant behavioral phenotypes of neuroligin-3 R451C knockin mice. Autism Res, 1(3), 147-158. doi:10.1002/aur.22

Chaidez, V., Hansen, R. L., & Hertz-Picciotto, I. (2014). Gastrointestinal problems in children with autism, developmental delays or typical development. Journal of Autism and Developmental Disorders, 44(5), 1117-1127. doi:10.1007/s10803-013-1973-x

Coury, D. L., Ashwood, P., Fasano, A., Fuchs, G., Geraghty, M., Kaul, A., . . . Jones, N. E. (2012). Gastrointestinal conditions in children with autism spectrum disorder: developing a research agenda. Pediatrics, 130 Suppl 2, S160-168. doi:10.1542/peds.2012-0900N

de Fontgalland, D., Brookes, S. J., Gibbins, I., Sia, T. C., & Wattchow, D. A. (2014). The neurochemical changes in the innervation of human colonic mesenteric and submucosal blood vessels in ulcerative colitis and Crohn's disease. Neurogastroenterol Motil, 26(5), 731-744. doi:10.1111/nmo.12327 den Haan, J. M., & Martinez-Pomares, L. (2013). Macrophage heterogeneity in lymphoid tissues. Semin Immunopathol, 35(5), 541-552. doi:10.1007/s00281-013-0378-4

Etherton, M., Foldy, C., Sharma, M., Tabuchi, K., Liu, X., Shamloo, M., . . . Sudhof, T. C. (2011). Autism- linked neuroligin-3 R451C mutation differentially alters hippocampal and cortical synaptic function. Proc Natl Acad Sci U S A, 108(33), 13764-13769. doi:10.1073/pnas.1111093108

Fagarasan, S., Muramatsu, M., Suzuki, K., Nagaoka, H., Hiai, H., & Honjo, T. (2002). Critical Roles of Activation-Induced Cytidine Deaminase in the Homeostasis of Gut Flora. Science, 298(5597), 1424. doi:10.1126/science.1077336

Gershon, M. D., & Ratcliffe, E. M. (2004). Developmental biology of the enteric nervous system: pathogenesis of Hirschsprung's disease and other congenital dysmotilities. Seminars in pediatric surgery, 13(4), 224-235.

Goyal, R. K., & Hirano, I. (1996). The enteric nervous system. N Engl J Med, 334(17), 1106-1115. doi:10.1056/nejm199604253341707

Grabrucker, A. M., Schmeisser, M. J., Schoen, M., & Boeckers, T. M. (2011). Postsynaptic ProSAP/Shank in the cross-hair of synaptopathies. Trends Cell Biol, 21(10), 594-603. scaffolds doi:10.1016/j.tcb.2011.07.003

Halladay, A. K., Amaral, D., Aschner, M., Bolivar, V. J., Bowman, A., DiCicco-Bloom, E., . . . Threadgill, for (2009). Animal models of autism spectrum disorders: information

D. W. neurotoxicologists. Neurotoxicology, 30(5), 811-821. doi:10.1016/j.neuro.2009.07.002 Horvath, K., & Perman, J. A. (2002). Autism and gastrointestinal symptoms. Curr Gastroenterol Rep, 4(3), 251-258.

Hosie, S., Ellis, M., Swaminathan, M., Ramalhosa, F., Seger, G. O., Balasuriya, G. K., . . . Hill-Yardin, E. L. (2019). Gastrointestinal dysfunction in patients and mice expressing the autism-associated R451C mutation in neuroligin-3. Autism Research, 0(0). doi:10.1002/aur.2127

production. immunity, behavior, Brain, and 57, Hosie, S., Malone, D. T., Liu, S., Glass, M., Adlard, P. A., Hannan, A. J., & Hill-Yardin, E. L. (2018). Altered Amygdala Excitation and CB1 Receptor Modulation of Aggressive Behavior in the Neuroligin- 3(R451C) Mouse Model of Autism. Front Cell Neurosci, 12, 234. doi:10.3389/fncel.2018.00234 Houlden, A., Goldrick, M., Brough, D., Vizi, E. S., Lénárt, N., Martinecz, B., . . . Denes, A. (2016). Brain injury induces specific changes in the caecal microbiota of mice via altered autonomic activity and mucoprotein 10-20. doi:10.1016/j.bbi.2016.04.003

Huguet, G., Benabou, M., & Bourgeron, T. (2016). The Genetics of Autism Spectrum Disorders. In P. Sassone-Corsi & Y. Christen (Eds.), A Time for Metabolism and Hormones (pp. 101-129). Cham (CH): Springer Copyright 2016, The Author(s). Int J Biochem Cell Biol, 36(10), 1861-1867. Jessen, K. R. (2004). Glial cells. doi:10.1016/j.biocel.2004.02.023

Johansson, M. E., Phillipson, M., Petersson, J., Velcich, A., Holm, L., & Hansson, G. C. (2008). The inner of the two Muc2 mucin-dependent mucus layers in colon is devoid of bacteria. Proc Natl Acad Sci U S A, 105(39), 15064-15069. doi:10.1073/pnas.0803124105

Kohane, I. S., McMurry, A., Weber, G., MacFadden, D., Rappaport, L., Kunkel, L., . . . Churchill, S. (2012). The Co-Morbidity Burden of Children and Young Adults with Autism Spectrum Disorders. PLOS ONE, 7(4), e33224. doi:10.1371/journal.pone.0033224

Kühl, A. A., Erben, U., Kredel, L. I., & Siegmund, B. (2015). Diversity of Intestinal Macrophages in 613-613. immunology, Diseases. Frontiers Bowel 6, in Inflammatory doi:10.3389/fimmu.2015.00613

Leembruggen, A. J., Balasuriya, G. K., Zhang, J., Schokman, S., Swiderski, K., Bornstein, J. C., ... & Hill‐ Yardin, E. L. (2019). Colonic dilation and altered ex vivo gastrointestinal motility in the neuroligin‐3 knockout mouse. Autism Research.

Li, S., Fei, G., Fang, X., Yang, X., Sun, X., Qian, J., . . . Ke, M. (2016). Changes in Enteric Neurons of Small Intestine in a Rat Model of Irritable Bowel Syndrome with Diarrhea. J Neurogastroenterol Motil, 22(2), 310-320. doi:10.5056/jnm15082

Lievin-Le Moal, V., & Servin, A. L. (2006). The front line of enteric host defense against unwelcome intrusion of harmful microorganisms: mucins, antimicrobial peptides, and microbiota. Clin Microbiol Rev, 19(2), 315-337. doi:10.1128/cmr.19.2.315-337.2006

Ljung, T., Lundberg, S., Varsanyi, M., Johansson, C., Schmidt, P. T., Herulf, M., . . . Hellstrom, P. M. (2006). Rectal nitric oxide as biomarker in the treatment of inflammatory bowel disease: responders versus nonresponders. World journal of gastroenterology, 12(21), 3386-3392. doi:10.3748/wjg.v12.i21.3386

Lonetti, G., Angelucci, A., Morando, L., Boggio, E. M., Giustetto, M., & Pizzorusso, T. (2010). Early environmental enrichment moderates the behavioral and synaptic phenotype of MeCP2 null mice. Biol Psychiatry, 67(7), 657-665. doi:10.1016/j.biopsych.2009.12.022

Loomes, R., Hull, L., & Mandy, W. P. L. (2017). What Is the Male-to-Female Ratio in Autism Spectrum Disorder? A Systematic Review and Meta-Analysis. J Am Acad Child Adolesc Psychiatry, 56(6), 466-474. doi:10.1016/j.jaac.2017.03.013

Marchezan, J., Winkler dos Santos, E. G. A., Deckmann, I., & Riesgo, R. S. (2018). Immunological for Therapy. in Autism Spectrum Disorder: A Potential Target Dysfunction Neuroimmunomodulation, 25(5-6), 300-319. doi:10.1159/000492225 Margolis KG, Gershon MD, Bogunovic M. Cellular Organization of Neuroimmune Interactions in the Gastrointestinal Tract. Trends Immunol. 2016;37(7):487–501. doi:10.1016/j.it.2016.05.003

Marlow, S. L., & Blennerhassett, M. G. (2006). Deficient innervation characterizes intestinal strictures in a rat model of colitis. Exp Mol Pathol, 80(1), 54-66. doi:10.1016/j.yexmp.2005.04.006 Masahata, K., Umemoto, E., Kayama, H., Kotani, M., Nakamura, S., Kurakawa, T., . . . Takeda, K. (2014). Generation of colonic IgA-secreting cells in the caecal patch. Nature Communications, 5(1), 3704. doi:10.1038/ncomms4704

McElhanon, B. O., McCracken, C., Karpen, S., & Sharp, W. G. (2014). Gastrointestinal symptoms in 872-883. a meta-analysis. Pediatrics, disorder: 133(5), autism spectrum doi:10.1542/peds.2013-3995

in McGuckin, M. A., Eri, R., Simms, L. A., Florin, T. H., & Radford-Smith, G. (2009). Intestinal barrier Inflamm Bowel Dis, 15(1), 100-113. inflammatory bowel diseases. dysfunction doi:10.1002/ibd.20539

(Hoboken, N.J. Mikkelsen, H. B., Huizinga, J. D., Larsen, J. O., & Kirkeby, S. (2017). Ionized calcium-binding adaptor molecule 1 positive macrophages and HO-1 up-regulation in intestinal muscularis resident macrophages. Anatomical Record : 2007), 300(6), 1114-1122. doi:10.1002/ar.23517 Mowat, A. M. (2003). Anatomical basis of tolerance and immunity to intestinal antigens. Nature reviews. Immunology, 3(4), 331-341. doi:10.1038/nri1057 Mowat, A. M., & Bain, C. C. (2011). Mucosal macrophages in intestinal homeostasis and inflammation. J Innate Immun, 3(6), 550-564. doi:10.1159/000329099

Neuhaus, E., Bernier, R. A., Tham, S. W., & Webb, S. J. (2018). Gastrointestinal and Psychiatric Symptoms Among Children and Adolescents With Autism Spectrum Disorder. Frontiers in psychiatry, 9, 515-515. doi:10.3389/fpsyt.2018.00515

Parracho, H. M., Bingham, M. O., Gibson, G. R., & McCartney, A. L. (2005). Differences between the gut microflora of children with autistic spectrum disorders and that of healthy children. J Med Microbiol, 54(Pt 10), 987-991. doi:10.1099/jmm.0.46101-0 Patterson, P. H. (2011). Modeling autistic features in animals. Pediatric research, 69(5 Pt 2), 34R-40R. doi:10.1203/PDR.0b013e318212b80f

Peterson, D. A., McNulty, N. P., Guruge, J. L., & Gordon, J. I. (2007). IgA Response to Symbiotic Bacteria as a Mediator of Gut Homeostasis. Cell Host & Microbe, 2(5), 328-339. doi:https://doi.org/10.1016/j.chom.2007.09.013

Rachmilewitz, D., Stamler, J. S., Bachwich, D., Karmeli, F., Ackerman, Z., & Podolsky, D. K. (1995). Enhanced colonic nitric oxide generation and nitric oxide synthase activity in ulcerative colitis and Crohn's disease. Gut, 36(5), 718-723. doi:10.1136/gut.36.5.718

Radyushkin, K., Hammerschmidt, K., Boretius, S., Varoqueaux, F., El-Kordi, A., Ronnenberg, A., . . . Ehrenreich, H. (2009). Neuroligin-3-deficient mice: model of a monogenic heritable form of autism with an olfactory deficit. Genes Brain Behav, 8(4), 416-425. doi:10.1111/j.1601- 183X.2009.00487.x

Rahman, A. A., Robinson, A. M., Jovanovska, V., Eri, R., & Nurgali, K. (2015). Alterations in the distal colon innervation in Winnie mouse model of spontaneous chronic colitis. Cell Tissue Res, 362(3), 497-512. doi:10.1007/s00441-015-2251-3

Rothwell, P. E., Fuccillo, M. V., Maxeiner, S., Hayton, S. J., Gokce, O., Lim, B. K., . . . Sudhof, T. C. (2014). impair striatal circuits to boost Autism-associated neuroligin-3 mutations commonly repetitive behaviors. Cell, 158(1), 198-212. doi:10.1016/j.cell.2014.04.045

Ruhl, A. (2005). Glial cells in the gut. Neurogastroenterol Motil, 17(6), 777-790. doi:10.1111/j.1365- 2982.2005.00687.x

Schmeisser, M. J., Ey, E., Wegener, S., Bockmann, J., Stempel, A. V., Kuebler, A., . . . Boeckers, T. M. (2012). Autistic-like behaviours and hyperactivity in mice lacking ProSAP1/Shank2. Nature, 486(7402), 256-260. doi:10.1038/nature11015

Schneider, J., Jehle, E. C., Starlinger, M. J., Neunlist, M., Michel, K., Hoppe, S., & Schemann, M. (2001). Neurotransmitter coding of enteric neurones in the submucous plexus is changed in non- inflamed rectum of patients with Crohn's disease. Neurogastroenterol Motil, 13(3), 255-264. Smith, P. D., Smythies, L. E., Shen, R., Greenwell-Wild, T., Gliozzi, M., & Wahl, S. M. (2011). Intestinal macrophages and response to microbial encroachment. Mucosal immunology, 4(1), 31-42. doi:10.1038/mi.2010.66

Soto, M., Herzog, C., Pacheco, J. A., Fujisaka, S., Bullock, K., Clish, C. B., & Kahn, C. R. (2018). Gut microbiota modulate neurobehavior through changes in brain insulin sensitivity and metabolism. Molecular Psychiatry, 23(12), 2287-2301. doi:10.1038/s41380-018-0086-5 Strugnell, R. A., & Wijburg, O. L. C. (2010). The role of secretory antibodies in infection immunity. Nature Reviews Microbiology, 8(9), 656-667. doi:10.1038/nrmicro2384

Suzuki, K., Meek, B., Doi, Y., Muramatsu, M., Chiba, T., Honjo, T., & Fagarasan, S. (2004). Aberrant expansion of segmented filamentous bacteria in IgA-deficient gut. Proceedings of the National Academy of Sciences, 101(7), 1981. doi:10.1073/pnas.0307317101

Tabuchi, K., Blundell, J., Etherton, M. R., Hammer, R. E., Liu, X., Powell, C. M., & Sudhof, T. C. (2007). A neuroligin-3 mutation implicated in autism increases inhibitory synaptic transmission in mice. Science, 318(5847), 71-76. doi:10.1126/science.1146221

Talapka, P., Nagy, L. I., Pal, A., Poles, M. Z., Berko, A., Bagyanszki, M., . . . Bodi, N. (2014). Alleviated mucosal and neuronal damage in a rat model of Crohn's disease. World journal of gastroenterology, 20(44), 16690-16697. doi:10.3748/wjg.v20.i44.16690

Valicenti-McDermott, M., McVicar, K., Rapin, I., Wershil, B. K., Cohen, H., & Shinnar, S. (2006). Frequency of gastrointestinal symptoms in children with autistic spectrum disorders and association with family history of autoimmune disease. J Dev Behav Pediatr, 27(2 Suppl), S128- 136. doi:10.1097/00004703-200604002-00011

for is Van der Sluis, M., De Koning, B. A., De Bruijn, A. C., Velcich, A., Meijerink, J. P., Van Goudoever, J. B., . . . Einerhand, A. W. (2006). Muc2-deficient mice spontaneously develop colitis, indicating that MUC2 colonic protection. Gastroenterology, 131(1), 117-129. critical doi:10.1053/j.gastro.2006.04.020

Varghese, M., Keshav, N., Jacot-Descombes, S., Warda, T., Wicinski, B., Dickstein, D. L., . . . Hof, P. R. spectrum disorder: neuropathology and animal models. Acta (2017). Autism neuropathologica, 134(4), 537-566. doi:10.1007/s00401-017-1736-4

Villanacci, V., Bassotti, G., Nascimbeni, R., Antonelli, E., Cadei, M., Fisogni, S., . . . Geboes, K. (2008). Enteric nervous system abnormalities in inflammatory bowel diseases. Neurogastroenterol Motil, 20(9), 1009-1016. doi:10.1111/j.1365-2982.2008.01146.x

Wei, H., Zou, H., Sheikh, A. M., Malik, M., Dobkin, C., Brown, W. T., & Li, X. (2011). IL-6 is increased in the cerebellum of autistic brain and alters neural cell adhesion, migration and synaptic formation. J Neuroinflammation, 8, 52. doi:10.1186/1742-2094-8-52

Winston, J. H., Li, Q., & Sarna, S. K. (2013). Paradoxical regulation of ChAT and nNOS expression in animal models of Crohn's colitis and ulcerative colitis. Am J Physiol Gastrointest Liver Physiol, 305(4), G295-302. doi:10.1152/ajpgi.00052.2013

Chapter 3

Altered Microbial composition in the Neuroligin-3R451C mouse

model of autism

3.1 Abstract

Gastrointestinal dysfunction is commonly reported in children with autism spectrum disorder which

may also contribute to behavioural impairment. It is thought that the crosstalk between gut

microbiota and the central nervous system has a vital influence in neurodevelopmental processes.

This study investigated the impact of the R451C missense mutation in the gene encoding the synaptic

protein, Neuroligin-3 on gut microbial communities in mice using Illumina 16S rRNA profiling. We

hypothesized that the distribution of gut microbial populations is altered in the gastrointestinal tract

of NL3R451C mice compared to wild type mice. Meta-barcoding analysis of the region-specific gut

microbial contents of wild type and NL3 R451C mice suggested compositional dysbiosis, expressed as an

increased abundance of Firmicutes and decreased abundance of Bacteroidetes and Actinobacteria.

Further work is required to understand how this mutation alters microbial populations in the

gastrointestinal tract. Findings from this study could assist in the understanding of the effects of an

autism-associated gene mutation in the host on the gut bacterial community structure in mice.

3.2 Introduction

Autism Spectrum disorder (ASD; autism) is a neurodevelopmental disorder which affects 1 in 59

children. This disorder is characterized by social deficits, language impairment and repetitive

behaviours (Baio et al., 2018; Loomes et al., 2017). Meta-analyses of clinical research on functional

gastrointestinal disorders (FGID) show that GI dysfunction is more frequent in children with ASD than

in neurotypical children (McElhanon et al., 2014). Gastrointestinal dysfunctions include functional

constipation, abdominal pain and irritable bowel syndrome (Hyman et al., 2006). It has been suggested

that altered communication within the gut-brain axis (Mayer & Tillisch, 2011) and more recently the

microbe-gut-brain axis (Levy et al., 2006; Mayer et al., 2014) may contribute to FGIDs.

The intestinal bacterial community and associated metabolites play an important role in maintaining

gut physiology and the symbiotic relationship with the host (Hooper et al., 2012; Johansson et al.,

2015). Any disruption of this communication between host and microbes can cause abnormalities in

physiological homeostasis (Reigstad et al., 2015; Wang et al., 2014). Numerous studies have suggested

a role for gut microbiota in the severity of autism-associated symptoms in children (Bolte, 1998; Luna

et al., 2017; Parracho et al., 2005; Shaw et al., 1995). Different compositions of bacterial species have

been observed in faecal samples from autistic children compared to controls (De Angelis et al., 2015;

Kang et al., 2013; Li et al., 2017). These changes are proposed to alter immune responses in children

with ASD compared to neurotypical (NT) children with Functional gastrointestinal disorder (FGID; Luna

et al., 2017).

Gut microbiota contribute to homeostasis during central nervous system (CNS) development and

maturation including via neural, immune, and circulatory pathways (Tremlett et al., 2017). Studies

examining germ free animals and animals treated with broad-spectrum antibiotics show microbiota

can influence the neurochemistry and physiology of the CNS (Smith et al., 2011). Alterations in the

nervous system are now known to affect the enteric nervous system (ENS) to result in gastrointestinal

dysfunction (Gershon & Ratcliffe, 2004; Heuckeroth & Pachnis, 2006; Hosie et al., 2019).

In this study, we studied mice expressing the Neuroligin-3 R451C mutation, which exhibit ASD-relevant

behaviours including impaired social interaction (Etherton et al., 2011; Tabuchi et al., 2007), a

heightened aggression phenotype (Burrows et al., 2015; Hosie et al., 2018); impaired communication

(Chadman et al., 2008), and increased repetitive behaviours (Rothwell et al., 2014). The robust

aggression phenotype in these mice was rescued by a clinically relevant antipsychotic, risperidone

(Burrows et al., 2015; Hosie et al., 2018), justifying the use of this model for preclinical studies.

Numerous studies have reported differences in the composition of various bacterial species in children

with ASD compared to neurotypical children (De Angelis et al., 2015; Finegold et al., 2010; Gondalia

et al., 2012; Kang et al., 2013; Wang et al., 2013; Williams et al., 2011). Although microbial abundance

and composition change at different locations within the GI tract, to date, gut microbial analyses have

mostly been performed on stool specimens. In the current study, microbial samples were collected

from multiple regions throughout the gut (i.e. the duodenum, ileum, caecum, and colon content) to

assess for regional changes in abundance and composition. Here we explored whether an autism-

associated mutation in the NLGN3 gene encoding the Neuroligin-3 synaptic adhesion protein affects

gastrointestinal bacterial composition in a mouse model of autism.

3.3 Method and materials

Animals

Neuroligin 3R451C mice were originally obtained from the Jackson Laboratories (Bar Harbour, Maine

USA). Mice were bred on a C57BL6 background for more than 10 generations and maintained at the

animal facility at RMIT University since 2017. Adult (8-14 weeks-old) male wild type (WT, n=17) and

NL3R451C (n=17) mice were used. These mice were obtained by mating NL3R451C male mice with

heterogenous females resulting in 50:50 WT and NL3R451C male offspring (Y/+ and Y/R451C) (Tabuchi

et al., 2007). The experimental mice were housed in mixed genotype cages. Mice were culled by

cervical dislocation in accordance with RMIT University animal ethics guidelines (AEC# 1727). Mouse

body weights were recorded prior to opening the abdomen using small dissecting scissors. To observe

the microbial distribution throughout the intestinal contents from NL3R451C and WT mice, lumenal

contents from mouse small intestine, caecum and colon were collected in eppendorf tubes; snap

frozen in liquid nitrozen and stored immediately at -80°C. Mucosa-associated microbes from the small

intestine were collected by scraping the content using a sterile transfer pipette and these samples

were subsequently transferred into a sterile labeled eppendorf tube. Before dissecting the caecum,

the microbial content was collected. Faecal pellets were also collected from the distal colon of each

animal.

DNA extraction and sequencing

Using standard DNA extraction techniques (DNeasy PowerSoil Kit: Lot: 157028918; QIAGEN GmbH,

Germany), 50 µl DNA was extracted from stored gut samples (i.e. duodenal-Jejunal junction, ileum,

caecum, and colon of WT (n=18) and NL3 R451C (n=20 mice), at La Trobe University, Bundoora, Australia.

The extracted DNA was kept at 4°C overnight before DNA concentration was checked using QubitTM

Fluorometer (InvitrogenTM QubitTM 4 fluorometer) and QubitTM dsDNA BR assay kit (Invitrogen, REF:

Q32850; QubitTM dsDNA BR assay kit; Lot: 1987262). The broad range (BR) assay kit was used instead

of the high sensitivity (HS) assay kit to account for the broad variation of the microbial population at

different regions of intestine. A volume of 5 µl of extracted DNA was mixed with Qubit buffer and

reagent of BR assay kit to measure the initial DNA concentration. PCR primers 338F (5’-

ACTCCTACGGGAGGCAGCAG-3’) and 806R (5’-GGACTACHVGGGTWTCTAAT-3’) were used to target the

V3 and V4 hypervariable regions of the 16S rRNA gene (Klindworth et al., 2013; Zhang et al., 2014).

PCR was performed in 25 µl reactions using 2.5 µl genomic DNA templates (5 µl/ng), 5µl of each

forward and reverse primer (10 µM) and 12.5 µl 2X KAPA HiFi HotStar Ready Mix (KAPA Biosystems,

Boston, MA, USA). The settings for PCR Cycle of the bacterial V3 and V4 regions were as follows:

denaturation at 95°C for 3 min, then 28 (bacterial) cycles of 30s at 95° C, 30s at 55°C and 30s at 72°C,

followed by an extension step at 72°C for 5 min. The DNA concentration of PCR reactions was

measured using broad range Qubit and then the libraries were normalised using 10M Tris (pH 8).

Samples were adjusted to the same molarity (4nM), pooled, and paired end sequenced (2 X 300 bp)

on an Illumina MiSeq platform. The MiSeq run was conducted at La Trobe University (Melbourne,

Australia).

Raw, demultiplexed, fastq files obtained from the Miseq run were re-barcoded, joined and quality-

filtered using UPARSE OTU clustering pipeline (Edgar, 2013). Merged reads were discarded if the

merged region contained > 15 bp differences. Reads were quality filtered by discarding reads with

total expected errors > 1.0. Operational taxonomic units (OTUs) were created at 97 % similarity with

a minimum threshold of 2 reads per OTU. Taxonomy was assigned using the Ribosomal Database

Project (RDP) 16S reference dataset. The minimum percentage identity required for an OTU to

consider a database match a hit was 97%. Phylogenetic trees were created using a neighbour-joining

algorithm via the multiple alignment software MUSCLE (Edgar, 2004a, 2004b). OTUs identified as

chloroplast and mitochondrial DNA were removed from the data prior to downstream analysis.

Microbiome analysis

All analysis of the sequenced metagenomic DNA was performed by using statistical software programs

in R (statistical computing and graphics; R development core team, 2014; R version 3.6.0) and PC-ORD

(version 6, Wild Blueberry Media LLC, USA; McCune & Grace, 2002) in order to observe and compare

the microbial diversity between NL3R451C and WT mice.

The UPARSE pipeline was used for quality filtering, trimming the reads to a fixed length, and discarding

singleton reads. Afterwards, the clustering and chimera filtering of next generation reads was

conducted (Edgar, 2013).

To observe the normality and variance homogeneity of the data, the ‘Shapiro test, and ‘Bartlett test’

functions were used. If normality and variance homogeneity assumptions were not met, a

nonparametric analysis method known as the MRPP (Multi-response permutation procedure) was

conducted using PC-ORD software (version 6) which does not require distribution assumptions such

as multivariate normality and homogeneity variance. MRPP tests show pairwise analysis of different

GI regions between WT and NL3R451C mice and provides aggregation values by A (effect size). The p

value produced by this test relies on both A and T (Test statistics) for more accuracy.

Beta-diversity analyses were performed on rarefied OTU-matrices using the ‘phyloseq’ (McMurdie &

Holmes, 2013) ‘vegan’ (Oksanen et al., 2015) modules of the R statistical package. Community beta-

diversity of region-specific gut microbial community was examined using Unifrac dissimilarity matrices

and non-metric multidimensional scaling (NMDS) ordination (Clarke & Ainsworth, 1993; Lozupone &

Knight, 2005). To support the NMDS analysis output, Permutational multivariate analysis of variance

(PERMANOVA) was used to test for significant differences in the mean group centroids.

Using the relatedness approach of Bray-Curtis Dissimilarities, compositional analysis was performed

using the R statistical software package to observe microbial community changes within different

regions of WT and NL3R451C mice. Additionally, differential analysis of OTUs was performed using the

DESeq2 extension within the ‘phyloseq’ package (Love et al., 2014) which shows relative abundance

changes of microbial taxa between WT and NL3R451C mice where negative values represent a reduction

in abundance and positive values represent an increase in abundance.

3.4 Results

This study identified 14,117,060 OTUs from four different gastrointestinal regions (duodenum, ileum,

caecum, and colon) of WT (n=16) and NL3R451C (n=17) mice. The highest number of OTUs were

obtained from the caecum of NL3R451C mice (986,804 OTUs). The second highest number of OTUs were

obtained from NL3R451C faecal pellets (822,159 OTUs). The least number of OTUs were identified from

NL3R451C duodenal samples (149,807OTUs).

The relative abundance of the bacterial community represents a proportional analysis of predominant

orders within the microbial content of WT and NL3R451C mice (Figure 3.1). The most predominant

orders were Bacteroidales, Candidatus, Clostridiales, Desulfovibrionales, Lactobacillales,

Verrucomicrodiales, Bifidobacteriales and Erysipelotrichales. Overall, the abundance of the order

Bacteroidales ranges from 30% to 60% throughout the gut. However, the relative abundance of

Bacteroidales was 11% higher in WT mice compared to NL3R451C mice. The second most predominant

order was Clostridiales for which the abundance ranges from 10% to 30% across all samples. However,

the relative abundance was 16% higher in NL3R451C mice throughout the GI tract compared to WT mice.

Microbial content from the ileum of both WT and NL3R451C mice revealed higher abundance of

Lactobacillales; but overall, NL3R451C mice showed an 11% increase in relative abundance of this order.

The relative abundance of Candidatus was increased in the caecum of NL3R451C mice compared to WT

mice. In NL3R451C mice both the ileum and caecum show an increase in the relative abundance of

bacteria of the order Desulfovibrionales. Interestingly, an increase in the relative abundance of

Bifidobacteriales and Erysipelotrichales was observed in the WT ileum samples. The OTUs of

Parcubacteria were only observed in the duodenum of both WT and NL3R451C mice. The duodenum of

NL3R451C mice showed a major increase in the presence of Mycoplasmatales compared with samples

derived from other regions of the gut. Another bacterial order of considerable presence was

Verrucomicrodiales, which was also observed throughout the GI tract in samples from both WT and

NL3R451C mice.

Figure 3.1 Relative abundance of total bacterial OTUs (p < 0.05) from region specific samples of WT and NL3

R451C mice. OTUs are clustered by order. Only orders that represented > 0.2% of the total community are

plotted.

The pairwise analysis of MRPP revealed that the microbial community separated gradually and

progressively along the GI tract within WT and NL3R451C mice (Table 3.1). The p value in the MRPP test

was dependent on Test statistics and the power of aggregation which was represented by A (effect

size). The results suggested that beginning from the ileum region, the microbial composition (i.e.

abundance and diversity) started to shift significantly, and in the distal colon, the difference between

the composition of microbial populations in each genotype was highest.

Table 3.1 MRPP (Multi-response permutation procedure) Test. A is effect size and represents the

heterogeneity of the groups and T (Test statistic) represents separation of the groups. The more

negative the T value is, the more separation of microbial populations is present between the groups.

Groups Compared (identifiers) T A p-value

WT Duodenum vs. NL3 Duodenum -1.89 0.014 0.0531

WT ileum vs. NL3 ileum -4.23 0.034 0.0019

WT caecum vs. NL3 caecum -6.29 0.053 0.0005

WT colon vs. NL3 colon -7.12 0.062 0.0002

The findings from the NMDS ordination analysis of samples from NL3R451C mice showed less grouping

in the microbial community whereas WT shows more aggregation suggesting similar microbial

communities (Figure 3.2).

Figure 3.2 NMDS ordinations of bacterial communities of region-specific gut samples (WT, n=16 and NL3

R451C, n=17) using weighted Unifrac distance. Each coloured symbol (WT=red, NL3 R451C =green) indicates

individual mice. The stress value of NMDS for all region-specific analysis were less than 0.2 (Duodenum: 0.17;

ileum: 0.16; Caecum: 0.13; colon: 0.13), which suggests microbial community differences between WT and NL3

R451C mice.

Permutational multivariate analysis of variance (PERMANOVA) revealed the effect of the R451C

mutation on gastrointestinal microbial community using both weighted (Table 3.2; R2=0.013, p<0.05)

and unweighted (i.e. binary transformed) Unifrac distances (Table 3.2; R2=0.028, p=0.001).

Furthermore, the effect of location on the microbial community structure was highly significant as

indicated by both weighted (Table 3.2; R2=0.175, p=0.001) and unweighted (i.e. binary transformed)

Unifrac distances (Table 3.2; R2=0.146, p=0.001). The test statistic R2 values for location-specific

PERMANOVA analysis were higher, indicating higher correlations compared to the genotype-specific

PERMANOVA analysis. When both variables (location and genotype) were considered, only data

resulting from the unweighted Unifrac analysis revealed statistically significant differences between

regions and genotype (p<0.05).

Table 3.2 Permutational multivariate analysis of variance (PERMANOVA). R2 represents effect size.

Significance values (*p<0.05, **p<0.01, ***p<0.001)

R2 p-value p-value Weighted Unifrac Sums Of Sqs Mean Sqs Unweighted Unifrac R2 Mean Sums Sqs Of Sqs

Location 4.318 1.439 0.175 0.001 *** 1.751 0.584 0.146 0.001 ***

Genotype 0.310 0.310 0.013 0.037 * 0.331 0.332 0.028 0.001 ***

0.459 0.153 0.019 0.624 0.438 0.146 0.037 0.025 * Location: Genotype

Using Bray-Curtis Dissimilarities, the compositional relatedness of OTUs was assessed, and it was

observed that the microbial community is highly divergent in caecum (66%) of WT and NL3R451C mice

which was supported by similar observations in the NMDS ordination analysis of caecum (Figure 3.2).

In both duodenum and ileum, the microbial composition differed by 53% whereas faecal pellet

microbial composition of WT and NL3R451C mice differed by 48% with respect to microbial community

composition (Figure 3.3).

Figure 3.3 Microbial community composition changes among different regions of WT and NL3 R451C by using

the relatedness approach of Bray-Curtis Dissimilarities. Error bars represent standard error of the mean. Y-axis

represents Bray-Curtis Dissimilarities between WT and NL3R451C mice within each gastrointestinal tract region.

The findings of Bray-Curtis Dissimilarities between WT and NL3R451C mice were supported by the

presence and absence analysis (Figure 3.4) where each square datapoint represented a single OTU of

WT mice. These OTUs of WT mice were used to observe a shift in microbial diversity in NL3R451C mice.

This analysis revealed that more OTUs disappeared (indicated by a negative value) in NL3R451C mice.

Figure 3.4 The graph shows microbial community change in NL3 R451C mice compared with WT; data points

located at the value of “0” show no change, negative values represent a reduction in abundance and positive

values represent increased abundance. Each square datapoint represents a single operational transcriptional

unit (OTU).

Moreover, we wanted to observe the ongoing bacterial community changes at different gut regions

from the proximal to distal GI tract and compare between WT and NL3R451C mice. Using Bray-Curtis

Dissimilarity analysis, we compared the microbial composition of samples derived from one gut region

with other regions (i.e. from duodenum, ileum, caecum, and distal colon) within mice of the same

genotype. The results from these analyses suggested that when comparing microbial populations

from different gut regions of mice of the same genotype, WT samples show more diversity in microbial

composition compared to those from NL3R451C mice. This was illustrated in that the microbial

compositional dissimilarity from duodenum to ileum in NL3R451C mice was 39% whereas microbial

compositional changes in WT for similar regions was 52%. (Figure 3.5). Given these findings and

considering the NMDS ordination results (Figure 3.2), WT mice had specific microbial communities in

each region of GI tract; therefore, when we compared communities from different regions, we

observed higher microbial diversity in WT mice. However, in NL3R451C mice, these microbial

communities were disrupted, thereby when different gut regions within similar genotype were

compared, less microbial diversity was found. In this experiment only exception was when we

compared caecal microbial community with colonic microbial community within similar genotype,

both WT and NL3R451C mice showed similar microbial diversity.

Figure 3.5 Compositional dissimilarity throughout the gut within WT and NL3R451C mice analysed using Bray-

Curtis Dissimilarity testing. Microbial community composition changes along the length of the GI tract and WT

samples show more microbial diversity throughout the gut compared to NL3R451C mice with one exception, as

the comparison between caecal and colonic microbial community showed similar microbial diversity in WT and

NL3R451C mice. Error bars represent standard error of the mean.

3.5 Discussion

Since microbiota influences CNS function via various immunological pathways, microbes may be

involved in the progression of neurodevelopmental disorders such as ASD (autism; Sharon et al.,

2016). In this study, we assessed for changes along the microbe-gut axis in a mouse model expressing

a mutation in the neuroligin-3 synaptic membrane protein. It is now well established that microbes

interact with the nervous system, largely based on studies of germ-free mice that show behavioural

changes and neurological deficiencies in memory, learning, and recognition (Foster et al., 2017;

Gareau et al., 2011).

A main finding of this study is the alteration in the gut microbial composition of WT and NL3R451C mice.

A previous report by Kang and colleagues assessed faecal samples from 20 neurotypical and autistic

children (3 to 16 years of age) and similarly revealed decreased diversity in the microbial community

of children with autism (Kang et al., 2013). An elevated diversity of gut bacteria may permit enhanced

mucosal barrier integrity and an increased ability to protect the GI tract from environmental stresses;

thus, maintaining gut homeostasis (Stecher et al., 2007). By conducting the pairwise analysis of MRPP

test, we observed that the microbial community became more divergent between WT and NL3R451C

mice at the distal region of the GI tract. In the GI tract, optimal pH and nutrient levels increase

gradually; therefore the bacterial communities in the small intestine and stomach (i.e. the proximal

region of the GI tract) differ to those in the large intestine and faeces (Berg, 1996; Gu et al., 2013). The

least separation in microbial composition between WT and NL3R451C was in the duodenum, and the

degree of separation between genotypes gradually increased from ileum to distal colon. However,

results arising from Bray-Curtis analysis, which represented only the compositional value or the

proportion of the composition that was changed, indicated that the microbial community was highly

divergent in the caecum (66%) and less divergent in the distal colon (48%) of WT and NL3R451C mice.

Previous meta-barcoding analysis of the bacterial communities has confirmed that Firmicutes and

Bacteroidetes are the predominant phyla in the GI tract of mammals (de Oliveira et al., 2013; Han et

al., 2015; Ursell et al., 2014). In the current study, although we observed a similar pattern, the gut

microbial content of NL3R451C mice showed an increase in the relative abundance of Firmicutes and a

decrease in the phylum Bacteroidetes compared to the microbial population of WT mice. Within the

Firmicutes phyla, order level reads indicated a significant increase in Clostridiales and Lactobacillales

and a decrease in Erysipelotrichales in the gut microbial content of NL3R451C mice. While studying faecal

samples of children with ASD, Finegold et al observed a decrease in the Firmicutes to Bacteroidetes

ratio (Finegold et al., 2010). In contrast, another study analysing ileum and caecum biopsy samples of

neurotypical and autistic children (both with GI dysfunction) reported an opposite trend (Williams et

al., 2011). In addition, a faecal sample study by Kang et al observed no differences in the ratio of

Firmicutes and Bacteroidetes (Kang et al., 2013). Reasons for these differences across studies could

include methodological variations or population differences, such as host genetics, diet, and

geography (Wang et al., 2014). In patients experiencing other GI dysfunctions such as IBD, a reduction

in the phyla of Firmicutes and Bacteroidetes was observed compared to healthy individuals

(Manichanh et al., 2006; Sokol et al., 2008). In colonic Crohn’s disease (CD) patients, increased levels

of Firmicutes, Actinobacteria, and Terenicutes were observed, whereas ileal CD patients showed a

decrease in Firmicutes and increase in Proteobacteria levels (Sokol et al., 2008). The results from the

current work differed to these patient studies. We observed a decrease in Actinobacteria and

Proteobacteria phyla in the gut content of NL3R451C mice. In contrast, order level reads of

Desulfovibrionales showed an increase in the ileal and caecal content of NL3R451C mice. Moreover,

samples derived from the duodenum of NL3R451C mice revealed an increased population of

Mycoplasmatales which is within the phylum of Terenicutes. Another predominant gut microbial

phylum observed in this study was Verrucomicrobia which was found in high abundance in gut

microbial content collected from different regions. However, this phylum was more abundant in WT

mice compared to NL3R451C mice. Interestingly, data from a faecal sample study assessing children with

ASD also revealed a decrease in Verrucomicrobia (Kang et al., 2013).

Studies examining the metabolic interactions between the host and gut microflora suggest that

microflora can influence brain development, behaviour and immune responses (Diaz Heijtz et al.,

2011; Macpherson & Harris, 2004; O'Hara & Shanahan, 2006; Sudo et al., 2004; Sudo et al., 1997).

Most gut microbial composition studies of samples from ASD patients were performed using faecal

samples (Finegold et al., 2010; Kang et al., 2013; Son et al., 2015). However, it is well known that the

small and large intestines are functionally distinct, so it is important to sample gut microbial content

from a range of gut regions. Previous work form our laboratory has shown a reduction in the caecal

weight and alteration in the caecal neuro-immune interactions (Chapter 2) which formed the rationale

for the current work which included the analyses of the caecal microbial composition in NL3R451C mice.

For this reason, we sampled the duodenum, ileum, caecum, and distal colon in the current study.

Although the function of the caecum is unclear, it is thought to act as a “safe house” for beneficial

bacteria and assist in rebuilding healthy microbial populations during GI dysfunction such as diarrhoea,

thus maintaining host homeostasis (Laurin et al., 2011). Importantly, we found altered caecal

microbial composition in NL3R451C compared to WT mice which suggests that the R451C mutation

affects gut microflora.

In summary, we demonstrated that the autism-associated R451C mutation in the neuroligin-3 gene

appeared to interact with different regions of the gastrointestinal tract to generate a distinct gut

microflora profile that could be characterized by altered composition of microbial community.

3.6 Conclusions

To our knowledge, this is the first study of microbial populations from different regions of the

gastrointestinal tract in NL3R451C mice. These findings assist in improving the understanding of the

crosstalk between gut microbiota and the nervous system in autism patients. This work forms the

basis for identifying specific molecular and microbial signatures in children with ASD and sets the stage

for future research to define the pathophysiology and identify targets to treat GI symptoms in children

with ASD.

3.7 References

Baio, J., Wiggins, L., Christensen, D. L., Maenner, M. J., Daniels, J., Warren, Z., . . . Dowling, N. F. (2018).

Prevalence of Autism Spectrum Disorder Among Children Aged 8 Years - Autism and Developmental

Disabilities Monitoring Network, 11 Sites, United States, 2014. MMWR Surveill Summ, 67(6), 1-23.

doi:10.15585/mmwr.ss6706a1

Berg, R. D. (1996). The indigenous gastrointestinal microflora. Trends Microbiol, 4(11), 430-435.

doi:10.1016/0966-842x(96)10057-3

Bolte, E. R. (1998). Autism and Clostridium tetani. Med Hypotheses, 51(2), 133-144. doi:10.1016/s0306-

9877(98)90107-4\

Burrows, E. L., Laskaris, L., Koyama, L., Churilov, L., Bornstein, J. C., Hill-Yardin, E. L., & Hannan, A. J. (2015). A

neuroligin-3 mutation implicated in autism causes abnormal aggression and increases repetitive

behavior in mice. Mol Autism, 6, 62. doi:10.1186/s13229-015-0055-7

Chadman, K. K., Gong, S., Scattoni, M. L., Boltuck, S. E., Gandhy, S. U., Heintz, N., & Crawley, J. N. (2008).

Minimal aberrant behavioral phenotypes of neuroligin-3 R451C knockin mice. Autism Res, 1(3), 147-

158. doi:10.1002/aur.22

Clarke, K., & Ainsworth, M. (1993). A method of linking multivariate community structure to environmental

variables (Vol. 92).

De Angelis, M., Piccolo, M., Vannini, L., Siragusa, S., De Giacomo, A., Serrazzanetti, D. I., . . . Francavilla, R.

(2013). Fecal microbiota and metabolome of children with autism and pervasive developmental

disorder not otherwise specified. PLOS ONE, 8(10), e76993. doi:10.1371/journal.pone.0076993

de Oliveira, M. N., Jewell, K. A., Freitas, F. S., Benjamin, L. A., Totola, M. R., Borges, A. C., . . . Suen, G. (2013).

Characterizing the microbiota across the gastrointestinal tract of a Brazilian Nelore steer. Vet

Microbiol, 164(3-4), 307-314. doi:10.1016/j.vetmic.2013.02.013

Diaz Heijtz, R., Wang, S., Anuar, F., Qian, Y., Bjorkholm, B., Samuelsson, A., . . . Pettersson, S. (2011). Normal

gut microbiota modulates brain development and behavior. Proc Natl Acad Sci U S A, 108(7), 3047-

3052. doi:10.1073/pnas.1010529108

Edgar, R. C. (2004a). MUSCLE: A multiple sequence alignment method with reduced time and space

complexity. BMC Bioinformatics, 5. doi:10.1186/1471-2105-5-113

Edgar, R. C. (2004b). MUSCLE: Multiple sequence alignment with high accuracy and high throughput. Nucleic

Acids Research, 32(5), 1792-1797. doi:10.1093/nar/gkh340

Edgar, R. C. (2013). UPARSE: highly accurate OTU sequences from microbial amplicon reads. Nature Methods,

10, 996. doi:10.1038/nmeth.2604

Edgar, R. C. (2013). UPARSE: Highly accurate OTU sequences from microbial amplicon reads. Nature Methods,

10(10), 996-998. doi:10.1038/nmeth.2604

Etherton, M., Foldy, C., Sharma, M., Tabuchi, K., Liu, X., Shamloo, M., . . . Sudhof, T. C. (2011). Autism-linked

neuroligin-3 R451C mutation differentially alters hippocampal and cortical synaptic function. Proc

Natl Acad Sci U S A, 108(33), 13764-13769. doi:10.1073/pnas.1111093108

Finegold, S. M., Dowd, S. E., Gontcharova, V., Liu, C., Henley, K. E., Wolcott, R. D., . . . Green, J. A. (2010).

Pyrosequencing study of fecal microflora of autistic and control children. Anaerobe, 16(4), 444-453.

doi:https://doi.org/10.1016/j.anaerobe.2010.06.008

Finegold, S. M., Dowd, S. E., Gontcharova, V., Liu, C., Henley, K. E., Wolcott, R. D., . . . Green, J. A., 3rd. (2010).

Pyrosequencing study of fecal microflora of autistic and control children. Anaerobe, 16(4), 444-453.

doi:10.1016/j.anaerobe.2010.06.008

Foster, J. A., Rinaman, L., & Cryan, J. F. (2017). Stress & the gut-brain axis: Regulation by the microbiome.

Neurobiol Stress, 7, 124-136. doi:10.1016/j.ynstr.2017.03.001

Gareau, M. G., Wine, E., Rodrigues, D. M., Cho, J. H., Whary, M. T., Philpott, D. J., . . . Sherman, P. M. (2011).

Bacterial infection causes stress-induced memory dysfunction in mice. Gut, 60(3), 307-317.

doi:10.1136/gut.2009.202515

Gershon, M. D., & Ratcliffe, E. M. (2004). Developmental biology of the enteric nervous system: pathogenesis

of Hirschsprung's disease and other congenital dysmotilities. Seminars in pediatric surgery, 13(4), 224-

235.

Gondalia, S. V., Palombo, E. A., Knowles, S. R., Cox, S. B., Meyer, D., & Austin, D. W. (2012). Molecular

Characterisation of Gastrointestinal Microbiota of Children With Autism (With and Without

Gastrointestinal Dysfunction) and Their Neurotypical Siblings. Autism Research, 5(6), 419-427.

doi:10.1002/aur.1253

Gu, S., Chen, D., Zhang, J.-N., Lv, X., Wang, K., Duan, L.-P., . . . Wu, X.-L. (2013). Bacterial Community Mapping

of the Mouse Gastrointestinal Tract. PLOS ONE, 8(10), e74957. doi:10.1371/journal.pone.0074957

Han, X., Yang, Y., Yan, H., Wang, X., Qu, L., & Chen, Y. (2015). Rumen bacterial diversity of 80 to 110-day-old

goats using 16S rRNA sequencing. PLOS ONE, 10(2), e0117811. doi:10.1371/journal.pone.0117811

Heuckeroth, R. O., & Pachnis, V. (2006). Getting to the guts of enteric nervous system development.

Development, 133(12), 2287. doi:10.1242/dev.02418

Hooper, L. V., Littman, D. R., & Macpherson, A. J. (2012). Interactions between the microbiota and the immune

system. Science, 336(6086), 1268-1273. doi:10.1126/science.1223490

Hosie, S., Ellis, M., Swaminathan, M., Ramalhosa, F., Seger, G. O., Balasuriya, G. K., . . . Hill-Yardin, E. L. (2019).

Gastrointestinal dysfunction in patients and mice expressing the autism-associated R451C mutation in

neuroligin-3. Autism Research, 0(0). doi:10.1002/aur.2127

Hyman, P. E., Milla, P. J., Benninga, M. A., Davidson, G. P., Fleisher, D. F., & Taminiau, J. (2006). Childhood

Functional Gastrointestinal Disorders: Neonate/Toddler. Gastroenterology, 130(5), 1519-1526.

doi:https://doi.org/10.1053/j.gastro.2005.11.065

Johansson, M. E. V., Jakobsson, H. E., Holmén-Larsson, J., Schütte, A., Ermund, A., Rodríguez-Piñeiro, A. M., . . .

Hansson, G. C. (2015). Normalization of Host Intestinal Mucus Layers Requires Long-Term Microbial

Colonization. Cell Host & Microbe, 18(5), 582-592. doi:10.1016/j.chom.2015.10.007

Kang, D. W., Park, J. G., Ilhan, Z. E., Wallstrom, G., Labaer, J., Adams, J. B., & Krajmalnik-Brown, R. (2013).

Reduced incidence of Prevotella and other fermenters in intestinal microflora of autistic children.

PLOS ONE, 8(7), e68322. doi:10.1371/journal.pone.0068322

Klindworth, A., Pruesse, E., Schweer, T., Peplies, J., Quast, C., Horn, M., & Glöckner, F. O. (2013). Evaluation of

general 16S ribosomal RNA gene PCR primers for classical and next-generation sequencing-based

diversity studies. Nucleic acids research, 41(1), e1-e1. doi:10.1093/nar/gks808

Kruskal, W. H., & Wallis, W. A. (1952). Use of Ranks in One-Criterion Variance Analysis. Journal of the American

Statistical Association, 47(260), 583-621. doi:10.1080/01621459.1952.10483441

Laurin, M., Everett, M. L., & Parker, W. (2011). The Cecal Appendix: One More Immune Component With a

Function Disturbed By Post-Industrial Culture. The Anatomical Record, 294(4), 567-579.

doi:10.1002/ar.21357

Levy, R. L., Olden, K. W., Naliboff, B. D., Bradley, L. A., Francisconi, C., Drossman, D. A., & Creed, F. (2006).

Psychosocial Aspects of the Functional Gastrointestinal Disorders. Gastroenterology, 130(5), 1447-

1458. doi:https://doi.org/10.1053/j.gastro.2005.11.057

Li, Q., Han, Y., Dy, A. B. C., & Hagerman, R. J. (2017). The Gut Microbiota and Autism Spectrum Disorders.

Frontiers in cellular neuroscience, 11, 120-120. doi:10.3389/fncel.2017.00120

Lindgreen, S., Adair, K. L., & Gardner, P. P. (2016). An evaluation of the accuracy and speed of metagenome

analysis tools. Sci Rep, 6, 19233. doi:10.1038/srep19233

Loomes, R., Hull, L., & Mandy, W. P. L. (2017). What Is the Male-to-Female Ratio in Autism Spectrum Disorder?

A Systematic Review and Meta-Analysis. J Am Acad Child Adolesc Psychiatry, 56(6), 466-474.

doi:10.1016/j.jaac.2017.03.013

Love, M. I., Huber, W., & Anders, S. (2014). Moderated estimation of fold change and dispersion for RNA-seq

data with DESeq2. Genome Biol, 15(12), 550. doi:10.1186/s13059-014-0550-8

Lozupone, C., & Knight, R. (2005). UniFrac: A new phylogenetic method for comparing microbial communities.

Applied and Environmental Microbiology, 71(12), 8228-8235. doi:10.1128/AEM.71.12.8228-8235.2005

Luna, R. A., Oezguen, N., Balderas, M., Venkatachalam, A., Runge, J. K., Versalovic, J., . . . Williams, K. C. (2017).

Distinct Microbiome-Neuroimmune Signatures Correlate With Functional Abdominal Pain in Children

With Autism Spectrum Disorder. Cellular and Molecular Gastroenterology and Hepatology, 3(2), 218-

230. doi:https://doi.org/10.1016/j.jcmgh.2016.11.008

Macpherson, A. J., & Harris, N. L. (2004). Interactions between commensal intestinal bacteria and the immune

system. Nature reviews. Immunology, 4(6), 478-485. doi:10.1038/nri1373

Manichanh, C., Rigottier-Gois, L., Bonnaud, E., Gloux, K., Pelletier, E., Frangeul, L., . . . Dore, J. (2006). Reduced

diversity of faecal microbiota in Crohn’s disease revealed by a metagenomic approach. Gut, 55(2),

205. doi:10.1136/gut.2005.073817

Mayer, E. A., Savidge, T., & Shulman, R. J. (2014). Brain–Gut Microbiome Interactions and Functional Bowel

Disorders. Gastroenterology, 146(6), 1500-1512. doi:https://doi.org/10.1053/j.gastro.2014.02.037

Mayer, E. A., & Tillisch, K. (2011). The Brain-Gut Axis in Abdominal Pain Syndromes. Annual Review of

Medicine, 62(1), 381-396. doi:10.1146/annurev-med-012309-103958

McCune, B., & Grace, J. (2002). Analysis of ecological communities. MJM Software Design, Gleneden Beach, OR.

McElhanon, B. O., McCracken, C., Karpen, S., & Sharp, W. G. (2014). Gastrointestinal symptoms in autism

spectrum disorder: a meta-analysis. Pediatrics, 133(5), 872-883. doi:10.1542/peds.2013-3995

McMurdie, P. J., & Holmes, S. (2013). Phyloseq: An R Package for Reproducible Interactive Analysis and

Graphics of Microbiome Census Data. PLoS ONE, 8(4). doi:10.1371/journal.pone.0061217

O'Hara, A. M., & Shanahan, F. (2006). The gut flora as a forgotten organ. EMBO reports, 7(7), 688-693.

doi:10.1038/sj.embor.7400731

Oksanen, J., Blanchet, F. G., Kindt, R., Legendre, P., Minchin, P., O’Hara, R. B., . . . Wagner, H. (2015). Vegan:

community ecology package. R package vegan, vers. 2.2-1.

Parracho, H. M., Bingham, M. O., Gibson, G. R., & McCartney, A. L. (2005). Differences between the gut

microflora of children with autistic spectrum disorders and that of healthy children. J Med Microbiol,

54(Pt 10), 987-991. doi:10.1099/jmm.0.46101-0

Radyushkin, K., Hammerschmidt, K., Boretius, S., Varoqueaux, F., El-Kordi, A., Ronnenberg, A., . . . Ehrenreich,

H. (2009). Neuroligin-3-deficient mice: model of a monogenic heritable form of autism with an

olfactory deficit. Genes Brain Behav, 8(4), 416-425. doi:10.1111/j.1601-183X.2009.00487.x

Reigstad, C. S., Salmonson, C. E., Rainey, J. F., 3rd, Szurszewski, J. H., Linden, D. R., Sonnenburg, J. L., . . .

Kashyap, P. C. (2015). Gut microbes promote colonic serotonin production through an effect of short-

chain fatty acids on enterochromaffin cells. Faseb j, 29(4), 1395-1403. doi:10.1096/fj.14-259598

Reiss, P. T., Stevens, M. H., Shehzad, Z., Petkova, E., & Milham, M. P. (2010). On distance-based permutation

tests for between-group comparisons. Biometrics, 66(2), 636-643. doi:10.1111/j.1541-

0420.2009.01300.x

Rothwell, P. E., Fuccillo, M. V., Maxeiner, S., Hayton, S. J., Gokce, O., Lim, B. K., . . . Sudhof, T. C. (2014). Autism-

associated neuroligin-3 mutations commonly impair striatal circuits to boost repetitive behaviors. Cell,

158(1), 198-212. doi:10.1016/j.cell.2014.04.045

Sharon, G., Sampson, T. R., Geschwind, D. H., & Mazmanian, S. K. (2016). The Central Nervous System and the

Gut Microbiome. Cell, 167(4), 915-932. doi:10.1016/j.cell.2016.10.027

Shaw, W., Kassen, E., & Chaves, E. (1995). Increased urinary excretion of analogs of Krebs cycle metabolites

and arabinose in two brothers with autistic features. Clin Chem, 41(8 Pt 1), 1094-1104.

Smith, P. D., Smythies, L. E., Shen, R., Greenwell-Wild, T., Gliozzi, M., & Wahl, S. M. (2011). Intestinal

macrophages and response to microbial encroachment. Mucosal immunology, 4(1), 31-42.

doi:10.1038/mi.2010.66

Sokol, H., Lay, C., Seksik, P., & Tannock, G. W. (2008). Analysis of bacterial bowel communities of IBD patients:

What has it revealed? Inflammatory Bowel Diseases, 14(6), 858-867. doi:10.1002/ibd.20392

Son, J. S., Zheng, L. J., Rowehl, L. M., Tian, X., Zhang, Y., Zhu, W., . . . Li, E. (2015). Comparison of Fecal

Microbiota in Children with Autism Spectrum Disorders and Neurotypical Siblings in the Simons

Simplex Collection. PLOS ONE, 10(10), e0137725. doi:10.1371/journal.pone.0137725

Stecher, B., Robbiani, R., Walker, A. W., Westendorf, A. M., Barthel, M., Kremer, M., . . . Hardt, W. D. (2007).

Salmonella enterica serovar typhimurium exploits inflammation to compete with the intestinal

microbiota. PLoS Biol, 5(10), 2177-2189. doi:10.1371/journal.pbio.0050244

Sudo, N., Chida, Y., Aiba, Y., Sonoda, J., Oyama, N., Yu, X. N., . . . Koga, Y. (2004). Postnatal microbial

colonization programs the hypothalamic-pituitary-adrenal system for stress response in mice. J

Physiol, 558(Pt 1), 263-275. doi:10.1113/jphysiol.2004.063388

Sudo, N., Sawamura, S., Tanaka, K., Aiba, Y., Kubo, C., & Koga, Y. (1997). The requirement of intestinal bacterial

flora for the development of an IgE production system fully susceptible to oral tolerance induction. J

Immunol, 159(4), 1739-1745.

Tabuchi, K., Blundell, J., Etherton, M. R., Hammer, R. E., Liu, X., Powell, C. M., & Sudhof, T. C. (2007). A

neuroligin-3 mutation implicated in autism increases inhibitory synaptic transmission in mice. Science,

318(5847), 71-76. doi:10.1126/science.1146221

Tremlett, H., Bauer, K. C., Appel-Cresswell, S., Finlay, B. B., & Waubant, E. (2017). The gut microbiome in

human neurological disease: A review. Ann Neurol, 81(3), 369-382. doi:10.1002/ana.24901

Ursell, L. K., Haiser, H. J., Van Treuren, W., Garg, N., Reddivari, L., Vanamala, J., . . . Knight, R. (2014). The

intestinal metabolome: an intersection between microbiota and host. Gastroenterology, 146(6), 1470-

1476. doi:10.1053/j.gastro.2014.03.001

Wang, L., Christophersen, C. T., Sorich, M. J., Gerber, J. P., Angley, M. T., & Conlon, M. A. (2013). Increased

abundance of Sutterella spp. and Ruminococcus torques in feces of children with autism spectrum

disorder. Molecular Autism, 4(1), 42. doi:10.1186/2040-2392-4-42

Wang, L., Conlon, M. A., Christophersen, C. T., Sorich, M. J., & Angley, M. T. (2014). Gastrointestinal microbiota

and metabolite biomarkers in children with autism spectrum disorders. Biomark Med, 8(3), 331-344.

doi:10.2217/bmm.14.12

Wang, W., Chen, L., Zhou, R., Wang, X., Song, L., Huang, S., . . . Xia, B. (2014). Increased proportions of

Bifidobacterium and the Lactobacillus group and loss of butyrate-producing bacteria in inflammatory

bowel disease. J Clin Microbiol, 52(2), 398-406. doi:10.1128/jcm.01500-13

Wilcoxon, F. (1945). Individual Comparisons by Ranking Methods. Biometrics Bulletin, 1(6), 80-83.

doi:10.2307/3001968

Williams, B. L., Hornig, M., Buie, T., Bauman, M. L., Cho Paik, M., Wick, I., . . . Lipkin, W. I. (2011). Impaired

carbohydrate digestion and transport and mucosal dysbiosis in the intestines of children with autism

and gastrointestinal disturbances. PLOS ONE, 6(9), e24585. doi:10.1371/journal.pone.0024585

Zhang, Q.-z., Dijkstra, F. A., Liu, X.-r., Wang, Y.-d., Huang, J., & Lu, N. (2014). Effects of Biochar on Soil Microbial

Biomass after Four Years of Consecutive Application in the North China Plain. PLOS ONE, 9(7),

e102062. doi:10.1371/journal.pone.0102062

Chapter 4

Discussion

The gut–brain axis is an important bidirectional pathway of communication between two major

nervous systems. It is hypothesised that ASD is primarily a disorder of neuronal communication and

that subtle changes in neural function may cause many of the systemic and neurological issues of ASD

(Betancur 2009; Grabrucker et al., 2011; Huguet, Benabou, & Bourgeron, 2016). This project

investigated the microbial and neuroimmune interactions in the NL3R451C mouse model of autism in

order to identify potential causes of altered gastrointestinal physiology in autism.

4.1 The NL3 R451C mutation causes a reduction in caecal weight

A main finding from this study is the clear reduction of caecal weight in mice expressing the Neuroligin-

3R451C mutation that were bred on a pure C57/Bl6 genetic background. This finding supports previous

observations of NL3R451C mice bred on a mixed genetic background (ImJ/Sv129/C57Bl6) at The

University of Melbourne and NL3-/- mice bred on a C57Bl/6NCrl background at the Florey Institute of

Neurosciences and Mental Health (Hosie et al., 2019). Hence, the finding that caecal weight is reduced

in the current study demonstrates that this is a persistent effect of the gene mutation rather than

genetic background or environment. This reduction could be due to changes in the mucus layer, water

content within the caecum, or muscle hypertrophy for example. Changes in bacterial composition

could also lead to altered mucus content and possibly alter the caecal weight. As discussed in Chapter

3, this mutation causes a shift in bacterial composition in the caecum. Alternatively, the R451C

mutation could change caecal motility or downstream colonic motility. Our lab has previously

demonstrated that colonic motility is altered in NL3R451C mice bred on a mixed background (Hosie et

al., 2019). However, controlled experiments need to be carried out to investigate these possible

underlying mechanisms in future experiments.

4.2 The NL3 R451C mutation leads to an altered enteric neuronal composition

To investigate whether the caecal enteric neural population is altered by the R451C neuroligin-3

mutation, wholemount preparations of WT and NL3R451C caecal submucosal and myenteric plexus

were labelled with Hu (a pan-neuronal marker, labelling all neurons) and NOS (neuronal nitric oxide

synthase, a marker of approximately 50% of myenteric neurons). Total caecal submucosal and

myenteric neuronal numbers were increased in NL3R451C mice. These data indicate that the R451C

mutation alters neural populations in agreement with preliminary data from jejunal myenteric

preparations in NL3R451C mice bred on a mixed genetic background (Hosie et al., 2019). These

alterations in enteric neurons could be due to compensatory mechanisms occurring in these mice due

to the R451C mutation in NLGN3.

4.3 Caecal macrophages are altered due to the R451C mutation in Neuroligin-3

The intestine is the largest mucosal surface of the body, thus the biggest compartment of the immune

system. In the healthy intestinal mucosa, mononuclear phagocytes comprising both macrophages and

dendritic cells (DC) are the most abundant population of leukocytes and play an important role in

maintaining homeostasis (Bain & Mowat, 2014). In the current study, we aimed to investigate the

density and morphology of Iba-1 immunoreactive cells in the caecal patch of WT and NL3R451C mice. In

caecal patch tissue, NL3R451C mice showed increased numbers of Iba-1 stained cells compared to WT

mice. Moreover, the volume of Iba-1 immunoreactive cells was decreased and appeared more

spherical in NL3R451C mutant mice compared to WT littermates. These findings could suggest that

macrophages within NL3R451C caecal patch tissue are present in a more reactive state compared with

in WT mice. Based on the discussion included in Chapter 2 and 3, we observed that the R451C mutation

in NL3 leads to altered GI physiology which likely then leads to changes in bacterial composition. These

changes may then cause an alteration in the density and morphology of caecal macrophage

populations in this model.

4.4 The NL3 R451C mutation causes caecal microbial dysbiosis

Changes in gut microbiota, inappropriate immune responses and increased intestinal permeability

occur in individuals genetically susceptible to autism (Coury et al., 2012). As evident from the

literature, the gut microbiota influences brain function via neuroimmune, neuroendocrine, and the

autonomic nervous system (Dinan & Cryan, 2017; Grenham t al., 2011; Margolis & Gershon, 2016). In

the current study, we aimed to investigate components of the complex biological interplay within the

microbe-gut -brain axis and to address how changes in this communication may cause GI dysfunction.

A main finding of this study is that differences in the gut microbial community composition between

WT and NL3R451C mice progressively increased along the GI tract. The pairwise comparison between

WT and NL3R451C mice indicated that the microbial community is shifting significantly from the ileal

region and that this shift increases at the distal colon where the composition change is highest. These

changes indicate that the NLGN3 R451C mutation affects gut microflora.

Taking into account the results of the current project it is evident that the NL3R451C mutation, which

affects synapse function in the central nervous system, changes the composition of the caecal enteric

nervous system, caecal macrophage density and morphology as well as the caecal bacterial

composition. Although the specific sequence in which these incidents occur is unclear, these changes

to the gut-brain axis could play a vital role in gut dysfunction in autism.

Conclusion

Microorganisms are immersed in our biological systems and play crucial roles in maintaining

physiological homeostasis. Disruption of this symbiotic relationship between the host and microbes

can contribute to gastrointestinal dysfunction, which is a common comorbidity among individuals with

autism. Very little is known about the function of the caecum and nothing is known about how the

NL3R451C mutation affects the structure of the caecal ENS in mice. The findings of this study suggested

an altered caecal neuro-immune interaction in the NL3R451C mice. We also demonstrated that the ASD-

associated gene mutation in neuroligin-3 is closely associated with an altered composition of the

microbial community.

Overall, when expressed in C57/Bl6 mice, the NL3 R451C mutation leads to alterations in the

gastrointestinal tract through an altered caecal ENS and differences in caecum-associated immune

cells and caecal microbiota. We observed gradual changes in microbial community composition from

the duodenum to the distal colon between WT and NL3 R451C mice; and these changes were greatest

in the region from the ileum to distal colon. It is therefore important to understand neuro-immune

and microbial interactions in different gut regions which might assist in identifying potential

therapeutic targets leading to improved quality of life for people with autism. In summary, the

physiological changes observed in this thesis, including anatomical abnormalities occurring in the

caecum, could be a significant cause of the gut dysfunction seen in autism.

Future directions

1. A significant finding of this study was a clear reduction in the caecal weight in NL3R451C mutant

mice. This reduction might be due to reduced caecal mucus thickness (currently under

investigation using histological and image analysis methods), which could also contribute to an

altered immune response.

2. Another possible explanation for reduced caecal weight could be that the caecum is emptied

more frequently due to increased caecal motility. This hypothesis is currently being assessed

by studying caecal motility using video imaging techniques (Lee et al., preliminary data from

the current laboratory).

3. An assessment of mucosal permeability and barrier function in NL3R451C mice is also being

undertaken to investigate whether gap junction proteins involved in barrier function are

altered which will assist in understanding how microbial interactions occur in caecum.

4. We observed gradual changes in microbial community composition from the duodenum to

the distal colon between WT and NL3 R451C mice; and these changes were highest in the region

from the ileum to distal colon. This finding indicates the importance of understanding neuro-

immune and microbial interactions in the ileum which might assist in identifying potential

therapeutic targets that may lead to improved quality of life for people with autism.

References

Argyropoulos, A., Gilby, K. L., & Hill-Yardin, E. L. (2013). Studying autism in rodent models:

reconciling endophenotypes with comorbidities. Front Hum Neurosci, 7, 417. doi:10.3389/fnhum.2013.00417

Ashwood, P., Krakowiak, P., Hertz-Picciotto, I., Hansen, R., Pessah, I. N., & Van de Water, J. (2011). Associations of impaired behaviors with elevated plasma chemokines in autism spectrum disorders. J Neuroimmunol, 232(1-2), 196-199. doi:10.1016/j.jneuroim.2010.10.025 Ashwood, P., Wills, S., & Van de Water, J. (2006). The immune response in autism: a new frontier for autism research. J Leukoc Biol, 80(1), 1-15. doi:10.1189/jlb.1205707 Atarashi, K., Tanoue, T., Ando, M., Kamada, N., Nagano, Y., Narushima, S., . . . Honda, K. (2015). Th17

Cell Induction by Adhesion of Microbes to Intestinal Epithelial Cells. Cell, 163(2), 367-380. doi:10.1016/j.cell.2015.08.058 Bain, C. C., & Mowat, A. M. (2014). Macrophages in intestinal homeostasis and inflammation. Immunological reviews, 260(1), 102-117. doi:10.1111/imr.12192 Belkaid, Y., & Hand, T. W. (2014). Role of the microbiota in immunity and inflammation. Cell, 157(1), 121-141. doi:10.1016/j.cell.2014.03.011 Betancur, C., Sakurai, T., & Buxbaum, J. D. (2009). The emerging role of synaptic cell-adhesion

pathways in the pathogenesis of autism spectrum disorders. Trends Neurosci, 32(7), 402- 412. doi:10.1016/j.tins.2009.04.003 Blumberg, R., & Powrie, F. (2012). Microbiota, disease, and back to health: a metastable journey.

Science translational medicine, 4(137), 137rv137-137rv137. doi:10.1126/scitranslmed.3004184

Bolliger, M. F., Frei, K., Winterhalter, K. H., & Gloor, S. M. (2001). Identification of a novel neuroligin in humans which binds to PSD-95 and has a widespread expression. The Biochemical journal, 356(Pt 2), 581-588. doi:10.1042/0264-6021:3560581 Burrows, E. L., Laskaris, L., Koyama, L., Churilov, L., Bornstein, J. C., Hill-Yardin, E. L., & Hannan, A. J.

(2015). A neuroligin-3 mutation implicated in autism causes abnormal aggression and increases repetitive behavior in mice. Mol Autism, 6, 62. doi:10.1186/s13229-015-0055-7 Bye, W. A., Allan, C. H., & Trier, J. S. (1984). Structure, distribution, and origin of M cells in Peyer's patches of mouse ileum. Gastroenterology, 86(5 Pt 1), 789-801. Chadman, K. K., Gong, S., Scattoni, M. L., Boltuck, S. E., Gandhy, S. U., Heintz, N., & Crawley, J. N.

(2008). Minimal aberrant behavioral phenotypes of neuroligin-3 R451C knockin mice. Autism Res, 1(3), 147-158. doi:10.1002/aur.22 Chaidez, V., Hansen, R. L., & Hertz-Picciotto, I. (2014). Gastrointestinal problems in children with

autism, developmental delays or typical development. Journal of Autism and Developmental Disorders, 44(5), 1117-1127. doi:10.1007/s10803-013-1973-x

Coelho-Aguiar Jde, M., Bon-Frauches, A. C., Gomes, A. L., Verissimo, C. P., Aguiar, D. P., Matias, D., . . . Moura-Neto, V. (2015). The enteric glia: identity and functions. Glia, 63(6), 921-935. doi:10.1002/glia.22795

Corr, S. C., Gahan, C. C., & Hill, C. (2008). M-cells: origin, morphology and role in mucosal immunity and microbial pathogenesis. FEMS Immunol Med Microbiol, 52(1), 2-12. doi:10.1111/j.1574- 695X.2007.00359.x

Coury, D. L., Ashwood, P., Fasano, A., Fuchs, G., Geraghty, M., Kaul, A., . . . Jones, N. E. (2012).

Gastrointestinal conditions in children with autism spectrum disorder: developing a research agenda. Pediatrics, 130 Suppl 2, S160-168. doi:10.1542/peds.2012-0900N Curtis, M. M., & Way, S. S. (2009). Interleukin-17 in host defence against bacterial, mycobacterial and fungal pathogens. Immunology, 126(2), 177-185. doi:10.1111/j.1365-2567.2008.03017.x

De Angelis, M., Francavilla, R., Piccolo, M., De Giacomo, A., & Gobbetti, M. (2015). Autism spectrum

disorders and intestinal microbiota. Gut Microbes, 6(3), 207-213. doi:10.1080/19490976.2015.1035855 De Angelis, M., Piccolo, M., Vannini, L., Siragusa, S., De Giacomo, A., Serrazzanetti, D. I., . . .

Francavilla, R. (2013). Fecal microbiota and metabolome of children with autism and pervasive developmental disorder not otherwise specified. PLOS ONE, 8(10), e76993. doi:10.1371/journal.pone.0076993 De Rubeis, S., & Buxbaum, J. D. (2015). Genetics and genomics of autism spectrum disorder:

embracing complexity. Hum Mol Genet, 24(R1), R24-31. doi:10.1093/hmg/ddv273 De Rubeis, S., He, X., Goldberg, A. P., Poultney, C. S., Samocha, K., Cicek, A. E., . . . Buxbaum, J. D. (2014). Synaptic, transcriptional and chromatin genes disrupted in autism. Nature, 515(7526), 209-215. doi:10.1038/nature13772 den Haan, J. M., & Martinez-Pomares, L. (2013). Macrophage heterogeneity in lymphoid tissues.

Semin Immunopathol, 35(5), 541-552. doi:10.1007/s00281-013-0378-4 Dinan, T. G., & Cryan, J. F. (2017). The Microbiome-Gut-Brain Axis in Health and Disease. Gastroenterol Clin North Am, 46(1), 77-89. doi:10.1016/j.gtc.2016.09.007 Edmonson, C., Ziats, M. N., & Rennert, O. M. (2014). Altered glial marker expression in autistic post-

mortem prefrontal cortex and cerebellum. Molecular Autism, 5(1), 3. doi:10.1186/2040- 2392-5-3

Ericsson, A. C., Hagan, C. E., Davis, D. J., & Franklin, C. L. (2014). Segmented filamentous bacteria: commensal microbes with potential effects on research. Comp Med, 64(2), 90-98. Fawley, J., & Gourlay, D. M. (2016). Intestinal alkaline phosphatase: a summary of its role in clinical disease. J Surg Res, 202(1), 225-234. doi:10.1016/j.jss.2015.12.008 Finegold, S. M., Dowd, S. E., Gontcharova, V., Liu, C., Henley, K. E., Wolcott, R. D., . . . Green, J. A.,

3rd. (2010). Pyrosequencing study of fecal microflora of autistic and control children. Anaerobe, 16(4), 444-453. doi:10.1016/j.anaerobe.2010.06.008

Fonseca, A. C., Romao, L., Amaral, R. F., Assad Kahn, S., Lobo, D., Martins, S., . . . Lima, F. R. (2012). Microglial stress inducible protein 1 promotes proliferation and migration in human glioblastoma cells. Neuroscience, 200, 130-141. doi:10.1016/j.neuroscience.2011.10.025 Furness, J. B. (2012). The enteric nervous system and neurogastroenterology. Nat Rev Gastroenterol Hepatol, 9(5), 286-294. doi:10.1038/nrgastro.2012.32 Gauthier, J., Bonnel, A., St-Onge, J., Karemera, L., Laurent, S., Mottron, L., . . . Rouleau, G. A. (2005).

NLGN3/NLGN4 gene mutations are not responsible for autism in the Quebec population. Am J Med Genet B Neuropsychiatr Genet, 132b(1), 74-75. doi:10.1002/ajmg.b.30066 Gebert, A., & Bartels, H. (1991). Occluding junctions in the epithelia of the gut-associated lymphoid tissue (GALT) of the rabbit ileum and caecum. Cell Tissue Res, 266(2), 301-314. Gebert, A., Hach, G., & Bartels, H. (1992). Co-localization of vimentin and cytokeratins in M-cells of rabbit gut-associated lymphoid tissue (GALT). Cell Tissue Res, 269(2), 331-340.

Geboes, K., Rutgeerts, P., Ectors, N., Mebis, J., Penninckx, F., Vantrappen, G., & Desmet, V. J. (1992). Major histocompatibility class II expression on the small intestinal nervous system in Crohn's disease. Gastroenterology, 103(2), 439-447. doi:10.1016/0016-5085(92)90832-j

Geschwind, D. H., & State, M. W. (2015). Gene hunting in autism spectrum disorder: on the path to precision medicine. Lancet Neurol, 14(11), 1109-1120. doi:10.1016/s1474-4422(15)00044-7 Gondalia, S. V., Palombo, E. A., Knowles, S. R., Cox, S. B., Meyer, D., & Austin, D. W. (2012).

Molecular characterisation of gastrointestinal microbiota of children with autism (with and without gastrointestinal dysfunction) and their neurotypical siblings. Autism Res, 5(6), 419- 427. doi:10.1002/aur.1253 Goto, Y., Panea, C., Nakato, G., Cebula, A., Lee, C., Diez, M. G., . . . Ivanov, II. (2014). Segmented

filamentous bacteria antigens presented by intestinal dendritic cells drive mucosal Th17 cell differentiation. Immunity, 40(4), 594-607. doi:10.1016/j.immuni.2014.03.005

Gottfried, C., Bambini-Junior, V., Francis, F., Riesgo, R., & Savino, W. (2015). The Impact of

Neuroimmune Alterations in Autism Spectrum Disorder. Frontiers in psychiatry, 6, 121. doi:10.3389/fpsyt.2015.00121 Grenham, S., Clarke, G., Cryan, J. F., & Dinan, T. G. (2011). Brain-gut-microbe communication in health and disease. Front Physiol, 2, 94. doi:10.3389/fphys.2011.00094

Gwynne, R. M., & Bornstein, J. C. (2007). Local inhibitory reflexes excited by mucosal application of nutrient amino acids in guinea pig jejunum. Am J Physiol Gastrointest Liver Physiol, 292(6), G1660-1670. doi:10.1152/ajpgi.00580.2006

Hallmayer, J., Cleveland, S., Torres, A., Phillips, J., Cohen, B., Torigoe, T., . . . Risch, N. (2011). Genetic heritability and shared environmental factors among twin pairs with autism. Arch Gen Psychiatry, 68(11), 1095-1102. doi:10.1001/archgenpsychiatry.2011.76 Horvath, K., & Perman, J. A. (2002). Autism and gastrointestinal symptoms. Curr Gastroenterol Rep, 4(3), 251-258. Hosie, S., Ellis, M., Swaminathan, M., Ramalhosa, F., Seger, G. O., Balasuriya, G. K., . . . Hill-Yardin, E.

L. (2019). Gastrointestinal dysfunction in patients and mice expressing the autism-associated R451C mutation in neuroligin-3. Autism Research, 0(0). doi:10.1002/aur.2127 Hsiao, E. Y., McBride, S. W., Chow, J., Mazmanian, S. K., & Patterson, P. H. (2012). Modeling an

autism risk factor in mice leads to permanent immune dysregulation. Proc Natl Acad Sci U S A, 109(31), 12776-12781. doi:10.1073/pnas.1202556109

Iossifov, I., O'Roak, B. J., Sanders, S. J., Ronemus, M., Krumm, N., Levy, D., . . . Wigler, M. (2014). The contribution of de novo coding mutations to autism spectrum disorder. Nature, 515(7526), 216-221. doi:10.1038/nature13908 Iossifov, I., Ronemus, M., Levy, D., Wang, Z., Hakker, I., Rosenbaum, J., . . . Wigler, M. (2012). De

novo gene disruptions in children on the autistic spectrum. Neuron, 74(2), 285-299. doi:10.1016/j.neuron.2012.04.009

Jamain, S., Quach, H., Betancur, C., Rastam, M., Colineaux, C., Gillberg, I. C., . . . Bourgeron, T. (2003). Mutations of the X-linked genes encoding neuroligins NLGN3 and NLGN4 are associated with autism. Nat Genet, 34(1), 27-29. doi:10.1038/ng1136 Jepson, M. A., Clark, M. A., Simmons, N. L., & Hirst, B. H. (1993). Epithelial M cells in the rabbit caecal

lymphoid patch display distinctive surface characteristics. Histochemistry, 100(6), 441-447. Jepson, M. A., Mason, C. M., Bennett, M. K., Simmons, N. L., & Hirst, B. H. (1992). Co-expression of vimentin and cytokeratins in M cells of rabbit intestinal lymphoid follicle-associated epithelium. Histochem J, 24(1), 33-39. Joe, E.-H., Choi, D.-J., An, J., Eun, J.-H., Jou, I., & Park, S. (2018). Astrocytes, Microglia, and

Parkinson's Disease. Experimental neurobiology, 27(2), 77-87. doi:10.5607/en.2018.27.2.77 Johansson, M. E. V., Jakobsson, H. E., Holmén-Larsson, J., Schütte, A., Ermund, A., Rodríguez-Piñeiro,

A. M., . . . Hansson, G. C. (2015). Normalization of Host Intestinal Mucus Layers Requires Long-Term Microbial Colonization. Cell Host & Microbe, 18(5), 582-592. doi:10.1016/j.chom.2015.10.007

Jung, C., Hugot, J.-P., & Barreau, F. (2010). Peyer's Patches: The Immune Sensors of the Intestine. International journal of inflammation, 2010, 823710-823710. doi:10.4061/2010/823710 Kamada, N., Chen, G. Y., Inohara, N., & Núñez, G. (2013). Control of pathogens and pathobionts by the gut microbiota. Nature Immunology, 14(7), 685-690. doi:10.1038/ni.2608

Kang, D. W., Park, J. G., Ilhan, Z. E., Wallstrom, G., Labaer, J., Adams, J. B., & Krajmalnik-Brown, R. (2013). Reduced incidence of Prevotella and other fermenters in intestinal microflora of autistic children. PLOS ONE, 8(7), e68322. doi:10.1371/journal.pone.0068322 Koboziev, I., Karlsson, F., & Grisham, M. B. (2010). Gut-associated lymphoid tissue, T cell trafficking,

and chronic intestinal inflammation. Ann N Y Acad Sci, 1207 Suppl 1, E86-93. doi:10.1111/j.1749-6632.2010.05711.x

Kohane, I. S., McMurry, A., Weber, G., MacFadden, D., Rappaport, L., Kunkel, L., . . . Churchill, S.

(2012). The Co-Morbidity Burden of Children and Young Adults with Autism Spectrum Disorders. PLOS ONE, 7(4), e33224. doi:10.1371/journal.pone.0033224 Kooij, I. A., Sahami, S., Meijer, S. L., Buskens, C. J., & Te Velde, A. A. (2016). The immunology of the

vermiform appendix: a review of the literature. Clin Exp Immunol, 186(1), 1-9. doi:10.1111/cei.12821 Krumm, N., O'Roak, B. J., Shendure, J., & Eichler, E. E. (2014). A de novo convergence of autism

genetics and molecular neuroscience. Trends Neurosci, 37(2), 95-105. doi:10.1016/j.tins.2013.11.005 Kühl, A. A., Erben, U., Kredel, L. I., & Siegmund, B. (2015). Diversity of Intestinal Macrophages in

Inflammatory Bowel Diseases. Frontiers in immunology, 6, 613-613. doi:10.3389/fimmu.2015.00613 Laumonnier, F., Bonnet-Brilhault, F., Gomot, M., Blanc, R., David, A., Moizard, M. P., . . . Briault, S.

(2004). X-linked mental retardation and autism are associated with a mutation in the NLGN4 gene, a member of the neuroligin family. Am J Hum Genet, 74(3), 552-557. doi:10.1086/382137 Laurin, M., Everett, M. L., & Parker, W. (2011). The Cecal Appendix: One More Immune Component

With a Function Disturbed By Post-Industrial Culture. The Anatomical Record, 294(4), 567- 579. doi:10.1002/ar.21357 Lawson-Yuen, A., Saldivar, J. S., Sommer, S., & Picker, J. (2008). Familial deletion within NLGN4

associated with autism and Tourette syndrome. Eur J Hum Genet, 16(5), 614-618. doi:10.1038/sj.ejhg.5202006 Li, Q., Han, Y., Dy, A. B. C., & Hagerman, R. J. (2017). The Gut Microbiota and Autism Spectrum Disorders. Frontiers in cellular neuroscience, 11, 120-120. doi:10.3389/fncel.2017.00120

Liu, Y.-H., Ding, Y., Gao, C.-C., Li, L.-S., Wang, Y.-X., & Xu, J.-D. (2018). Functional macrophages and gastrointestinal disorders. World journal of gastroenterology, 24(11), 1181-1195. doi:10.3748/wjg.v24.i11.1181 Lima, F. R., Arantes, C. P., Muras, A. G., Nomizo, R., Brentani, R. R., & Martins, V. R. (2007). Cellular

prion protein expression in astrocytes modulates neuronal survival and differentiation. J Neurochem, 103(6), 2164-2176. doi:10.1111/j.1471-4159.2007.04904.x

Luna, R. A., Oezguen, N., Balderas, M., Venkatachalam, A., Runge, J. K., Versalovic, J., . . . Williams, K. C. (2016). Distinct Microbiome-Neuroimmune Signatures Correlate With Functional Abdominal Pain in Children With Autism Spectrum Disorder. Cellular and molecular gastroenterology and hepatology, 3(2), 218-230. doi:10.1016/j.jcmgh.2016.11.008 Luna, R. A., Oezguen, N., Balderas, M., Venkatachalam, A., Runge, J. K., Versalovic, J., . . . Williams, K. C. (2017). Distinct Microbiome-Neuroimmune Signatures Correlate With Functional Abdominal Pain in Children With Autism Spectrum Disorder. Cellular and Molecular Gastroenterology and Hepatology, 3(2), 218-230. doi:https://doi.org/10.1016/j.jcmgh.2016.11.008 Mackowiak, M., Mordalska, P., & Wedzony, K. (2014). Neuroligins, synapse balance and

neuropsychiatric disorders. Pharmacol Rep, 66(5), 830-835. doi:10.1016/j.pharep.2014.04.011 Margolis, K. G., & Gershon, M. D. (2016). Enteric Neuronal Regulation of Intestinal Inflammation. Trends in neurosciences, 39(9), 614-624. doi:10.1016/j.tins.2016.06.007 Masahata, K., Umemoto, E., Kayama, H., Kotani, M., Nakamura, S., Kurakawa, T., . . . Takeda, K.

(2014). Generation of colonic IgA-secreting cells in the caecal patch. Nature Communications, 5(1), 3704. doi:10.1038/ncomms4704

Mayer, E. A. (2011). Gut feelings: the emerging biology of gut–brain communication. Nature Reviews Neuroscience, 12(8), 453-466. doi:10.1038/nrn3071 Mazzoli, R., & Pessione, E. (2016). The Neuro-endocrinological Role of Microbial Glutamate and

GABA Signaling. Frontiers in Microbiology, 7(1934). doi:10.3389/fmicb.2016.01934 McElhanon, B. O., McCracken, C., Karpen, S., & Sharp, W. G. (2014). Gastrointestinal symptoms in

autism spectrum disorder: a meta-analysis. Pediatrics, 133(5), 872-883. doi:10.1542/peds.2013-3995 Morales-Soto, W., & Gulbransen, B. D. (2019). Enteric Glia: A New Player in Abdominal Pain. Cellular

and Molecular Gastroenterology and Hepatology, 7(2), 433-445. doi:10.1016/j.jcmgh.2018.11.005

Mowat, A. M., & Bain, C. C. (2011). Mucosal macrophages in intestinal homeostasis and inflammation. J Innate Immun, 3(6), 550-564. doi:10.1159/000329099 Neuhaus, E., Bernier, R. A., Tham, S. W., & Webb, S. J. (2018). Gastrointestinal and Psychiatric

Symptoms Among Children and Adolescents With Autism Spectrum Disorder. Frontiers in psychiatry, 9, 515-515. doi:10.3389/fpsyt.2018.00515 Neutra, M. R., Mantis, N. J., & Kraehenbuhl, J. P. (2001). Collaboration of epithelial cells with

organized mucosal lymphoid tissues. Nat Immunol, 2(11), 1004-1009. doi:10.1038/ni1101- 1004

Ochoa-Cortes, F., Turco, F., Linan-Rico, A., Soghomonyan, S., Whitaker, E., Wehner, S., . . . Christofi, F. L. (2016). Enteric Glial Cells: A New Frontier in Neurogastroenterology and Clinical Target for Inflammatory Bowel Diseases. Inflammatory bowel diseases, 22(2), 433-449. doi:10.1097/MIB.0000000000000667 Ohland, C. L., & Jobin, C. (2015). Microbial activities and intestinal homeostasis: A delicate balance

between health and disease. Cellular and Molecular Gastroenterology and Hepatology, 1(1), 28-40. doi:10.1016/j.jcmgh.2014.11.004

Ohnmacht, C., Park, J. H., Cording, S., Wing, J. B., Atarashi, K., Obata, Y., . . . Eberl, G. (2015). MUCOSAL IMMUNOLOGY. The microbiota regulates type 2 immunity through RORgammat(+) T cells. Science, 349(6251), 989-993. doi:10.1126/science.aac4263 Parracho, H. M., Bingham, M. O., Gibson, G. R., & McCartney, A. L. (2005). Differences between the

gut microflora of children with autistic spectrum disorders and that of healthy children. J Med Microbiol, 54(Pt 10), 987-991. doi:10.1099/jmm.0.46101-0 Ratcliffe, M. J. (2002). B cell development in gut associated lymphoid tissues. Vet Immunol Immunopathol, 87(3-4), 337-340. doi:10.1016/s0165-2427(02)00061-2

Reigstad, C. S., Salmonson, C. E., Rainey, J. F., 3rd, Szurszewski, J. H., Linden, D. R., Sonnenburg, J. L., .

. . Kashyap, P. C. (2015). Gut microbes promote colonic serotonin production through an effect of short-chain fatty acids on enterochromaffin cells. Faseb j, 29(4), 1395-1403. doi:10.1096/fj.14-259598

Sanders, S. J., Murtha, M. T., Gupta, A. R., Murdoch, J. D., Raubeson, M. J., Willsey, A. J., . . . State, M. W. (2012). De novo mutations revealed by whole-exome sequencing are strongly associated with autism. Nature, 485(7397), 237-241. doi:10.1038/nature10945 Smith, P. D., Smythies, L. E., Shen, R., Greenwell-Wild, T., Gliozzi, M., & Wahl, S. M. (2011). Intestinal

macrophages and response to microbial encroachment. Mucosal immunology, 4(1), 31-42. doi:10.1038/mi.2010.66 Spencer, J., Finn, T., & Isaacson, P. G. (1985). Gut associated lymphoid tissue: a morphological and

immunocytochemical study of the human appendix. Gut, 26(7), 672-679. doi:10.1136/gut.26.7.672 Tabuchi, K., Blundell, J., Etherton, M. R., Hammer, R. E., Liu, X., Powell, C. M., & Sudhof, T. C. (2007).

A neuroligin-3 mutation implicated in autism increases inhibitory synaptic transmission in mice. Science, 318(5847), 71-76. doi:10.1126/science.1146221 Ulbrich, L., Favaloro, F. L., Trobiani, L., Marchetti, V., Patel, V., Pascucci, T., . . . De Jaco, A. (2016). Autism-associated R451C mutation in neuroligin3 leads to activation of the unfolded protein

response in a PC12 Tet-On inducible system. The Biochemical journal, 473(4), 423-434. doi:10.1042/BJ20150274

Volpe, E., Battistini, L., & Borsellino, G. (2015). Advances in T Helper 17 Cell Biology: Pathogenic Role and Potential Therapy in Multiple Sclerosis. Mediators of inflammation, 2015, 475158- 475158. doi:10.1155/2015/475158 von Boyen, G. B. T., Schulte, N., Pflüger, C., Spaniol, U., Hartmann, C., & Steinkamp, M. (2011).

Distribution of enteric glia and GDNF during gut inflammation. BMC gastroenterology, 11, 3- 3. doi:10.1186/1471-230X-11-3

Vossenkämper, A., Blair, P. A., Safinia, N., Fraser, L. D., Das, L., Sanders, T. J., . . . Spencer, J. (2013). A role for gut-associated lymphoid tissue in shaping the human B cell repertoire. The Journal of experimental medicine, 210(9), 1665-1674. doi:10.1084/jem.20122465 Wang, W., Chen, L., Zhou, R., Wang, X., Song, L., Huang, S., . . . Xia, B. (2014). Increased proportions

of Bifidobacterium and the Lactobacillus group and loss of butyrate-producing bacteria in inflammatory bowel disease. J Clin Microbiol, 52(2), 398-406. doi:10.1128/jcm.01500-13

Wasilewska, J., & Klukowski, M. (2015). Gastrointestinal symptoms and autism spectrum disorder: links and risks - a possible new overlap syndrome. Pediatric health, medicine and therapeutics, 6, 153-166. doi:10.2147/PHMT.S85717

Williams, B. L., Hornig, M., Buie, T., Bauman, M. L., Cho Paik, M., Wick, I., . . . Lipkin, W. I. (2011). Impaired carbohydrate digestion and transport and mucosal dysbiosis in the intestines of children with autism and gastrointestinal disturbances. PLOS ONE, 6(9), e24585. doi:10.1371/journal.pone.0024585 Wing, L., & Gould, J. (1979). Severe impairments of social interaction and associated abnormalities in

children: Epidemiology and classification. Journal of Autism and Developmental Disorders, 9(1), 11-29. doi:10.1007/BF01531288 Wu, W. L., Adams, C. E., Stevens, K. E., Chow, K. H., Freedman, R., & Patterson, P. H. (2015). The

interaction between maternal immune activation and alpha 7 nicotinic acetylcholine receptor in regulating behaviors in the offspring. Brain Behav Immun, 46, 192-202. doi:10.1016/j.bbi.2015.02.005 Yan, J., Oliveira, G., Coutinho, A., Yang, C., Feng, J., Katz, C., . . . Sommer, S. S. (2005). Analysis of the

neuroligin 3 and 4 genes in autism and other neuropsychiatric patients. Mol Psychiatry, 10(4), 329-332. doi:10.1038/sj.mp.4001629

Yu, Y. B., & Li, Y. Q. (2014). Enteric glial cells and their role in the intestinal epithelial barrier. World journal of gastroenterology vol. 20,32, 11273-80. doi:10.3748/wjg.v20.i32.11273

Appendices

Appendix 1 Ethics approval letter

Appendix 2: Haematoxylin & Eosin staining

Slides containing the tissue sections were submerged in Mayer’s haematoxylin (constitution: choral

hydrate (100g); potassium aluminum sulphate (100g); haematoxylin (4g); citric acid (2g); sodium

iodate (0.4g); distilled water (2l) and glacial acetic acid (20ml)). Scott’s tap water (constitution:

magnesium sulphate (40g); sodium bicarbonate (7g) and distilled water (2L)) was used as a ‘blueing

reagent’ to aid in visualizing samples under a microscope.

0.1% Eosin (supplemented with 2g/100ml of CaCl2) is also needed to complete the staining. Alcohol

at two different concentrations (70% and 100%) was then used to clean excess staining from slides.

Samples were then cleared using histolene (Trajan Scientific Australia, Ringwood, VIC, Australia) and

mounted in DPX medium (Millipore, Bayswater, VIC, Australia).

Steps Method Duration

Wash in running tap water 10sec 1

Submerge samples in Mayer’s Haemotoxylin 60 sec 2

Wash in running tap water 10 sec 3

Wash in Scott’s tap water 30 sec 4

Wash in running tap water 10 sec 5

Immerse in 0.1% Eosin 30 sec 6

Wash in running tap water 10 sec 7

Immerse in 70% alcohol 2 dips 8

Immerse in absolute alcohol 2 dips 9

Immerse in absolute alcohol 4 dips 10

Immerse in absolute alcohol x2 60 sec each 11

Place samples in Histolene 120 sec 12

Place samples in Histolene 120 sec 13

Mount slides in DPX and coverslip Air dry under fume hood overnight 14

Appendix 3: Caecal patch analysis using ImageJ to determine the cell density

Appendix 4: Using ImageJ, analysis of number of neurons/ganglia in whole mount preparations

Appendix 5: Iba1 immunoreactive cell analysis using Imaris

To measure the volume and sphericity of the cells:

Surpass from the object tool bar

Region of interest -> select the channel -> threshold adjustment -> filtration to remove unwanted selection

File needs to be converted using Imaris file converter.

Threshold adjustment -> scale bar -> export all statistics (will contain volume, area, etc)

Appendix 6: Mouse ID used for microbial analysis

Parents

Genotype

Animal ID

Sex

Colour

Age

♂

♀

WT/0

64A

Black

11 weeks

15A

♂

Ki/0

67A

Black

11 weeks

15A

♂

Ki/0

75A

Black

12 weeks

11A

♂

Ki/0

77A

Black

12 weeks

11A

♂

WT/0

123A

Black

9 weeks

15A

♂

WT/0

124A

Black

9 weeks

15A

♂

WT/0

127A

Black

9 weeks

12A

♂

WT/0

128A

Black

9 weeks

12A

♂

WT/0

140A

Black

10 weeks

13A

♂

WT/0

141A

Black

10 weeks

14A

♂

Ki/0

142A

Black

10 weeks

14A

♂

Ki/0

131A

Black

9 weeks

11A

♂

Ki/0

132A

Black

9 weeks

11A

♂

Ki/0

133A

Black

9 weeks

11A

♂

WT/0

134A

Black

9 weeks

11A

♂

WT/0

135A

Black

9 weeks

13A

♂

Ki/0

137A

Black

9 weeks

13A

♂

WT/0

143A

Black

10 weeks

13A

♂

WT/0

144A

Black

10 weeks

13A

♂

Ki/0

145A

Black

10 weeks

13A

♂

WT/0

146A

Black

10 weeks

13A

♂

Ki/0

148A

Black

10 weeks

15A

10A

♂

WT/0

149A

Black

10 weeks

15A

10A

♂

WT/0

151A

Black

11 weeks

15A

10A

♂

Ki/0

162A

Black

8 weeks

15A

10A

♂

WT/0

165A

♂

Black

8 weeks

14A

Ki/0

174A

Black

9 weeks

11A

33A

♂

Ki/0

175A

Black

9 weeks

11A

33A

♂

Ki/0

179A

Black

10 weeks

11A

33A

♂

Ki/0

185A

Black

10 weeks

12A

36A

♂

#install.packages("BiocManager")

#BARCHARTS script created for analysis of metagenomic (16S amplicon) data ##accompianing files: metadata.txt, miseq_work_file/OTU_table_SSS.csv,09R_files/tax_table.csv, 09R_files/tree_seqs.phy ##script created by Samiha Sayed Sharna 18.05.2019 ##First: change working directory to top folder of metagenomic project ## on first run we need to open the .txt and remove the '#' from the first row ## otherwise R will not read it. also convert 'OTU ID' to 'OTU_ID' ## make a excel file and read in as a .csv otu_table <- subset(as.data.frame(read.csv("miseq_work_file_s/OTU_table_SS.csv", row.names=1, header=TRUE))) ## to look at the total reads per sample and decide on a rarefaction depth rare<-colSums(otu_table) rare plot(rare) ## to observe the max, min and median max(rare) min(rare) median(rare) ##next we need a tree - we use it for the ordinations ##this will generated in the UNIX script and placed in the miseq work_file folder for us ##install the package ape library(ape) TREE <- read.tree("miseq_work_file/tree_seqs_nj_file.phy") #### now a taxonomy table ## make this file in excel from HPC outputs ## headings in the .csv file should be: 'OTU_ID' "Domain' 'Phylum' 'Class' 'Order' 'Family' 'Genus' 'Species' taxmat <- as.matrix(read.csv("miseq_work_file_s/work_taxonomy_file_s.csv", row.names=1, header=TRUE)) ##row names for taxmat and otu table MUST be the same rownames(taxmat) rownames(otu_table) ##Load and create a metadata data frame ##run make this in excel:first row should be sample names, subsequent rows are treatment aspects ##IMPORTANT : must have more than one column in your treatment file ##(the first column will be pulled out to be used as labels so more than 2 columns is needed) treat <- as.data.frame(read.csv("miseq_work_file_s/metadata_sheet_file.csv", row.names=1, header=TRUE)) ##OK time to make our phyloseq object #to intall the phyloseq package we will need to follow instructions on #https://bioconductor.org/packages/release/bioc/html/phyloseq.html # copy and paste the function below to install phyloseq #if(!requireNamespace("BiocManager", quietly = TRUE)) #BiocManager::install("phyloseq") library(phyloseq) OTU = otu_table(otu_table, taxa_are_rows = TRUE) TAX = tax_table(taxmat) TREAT = sample_data(treat) # we will combine all tables using phyloseq OBJ1 = phyloseq(OTU,TAX,TREAT,TREE) sample_names(OTU) sample_names(TAX) sample_names(TREAT) ## to confirm correct labels/treatments ect have been assigned sample_data(OBJ1) ## to remove mitochondria and chloroplasts you can use this script library(magrittr) OBJ1 <- OBJ1 %>% subset_taxa( Domain == "Bacteria" & Family != "mitochondria" & Class != "Chloroplast") ##the following walkthrough detail many preprocessing options #####here is a few typical way of subseting phyloseq data: levels(sample_data(OBJ1)) otu_table(OBJ1) tax_table(OBJ1)

Appendix 7: R scripts for data preparation

OBJ1_Duo <- subset_samples(OBJ1, location == "Duodenum") plot_richness(OBJ1_Duo,x="location",measur=c("Simpson","Chao1","Shannon")) rare_Dou <- colSums(otu_table(OBJ1_Duo)) OBJ1_Ile <- subset_samples(OBJ1, location == "Ileum ") plot_richness(OBJ1_Ile,x="location",measur=c("Simpson","Chao1","Shannon")) rare_Ile <- colSums(otu_table(OBJ1_Ile)) ## in my metadata file after the name ileum I have as space that's why had to put an space after the name Ileum OBJ1_Cae <- subset_samples(OBJ1, location == "Caecum ") plot_richness(OBJ1_Cae,x="location",measur=c("Simpson","Chao1","Shannon")) rare_Cae <- colSums(otu_table(OBJ1_Cae)) ## same as Ileum had to add a space OBJ1_Faeces <- subset_samples(OBJ1, location == "Colon") plot_richness(OBJ1_Faeces,x="location",measur=c("Simpson","Chao1","Shannon")) rare_Faeces <- colSums(otu_table(OBJ1_Faeces)) sample_data(OBJ1_Duo) rare<-sample_data(OBJ1_Duo) rare plot(rare) OBJ1_Clostri = subset_taxa(OBJ1, Order=="Clostridiales") ##lets rarefy otu data to do some alpha and beta diversity set.seed(8385) #there is an element of randomness in rarfying. this eliminates that OBJ1_r <- rarefy_even_depth(OBJ1, sample.size = 4000, rngseed = TRUE, replace = FALSE, trimOTUs = TRUE) OBJ1_rDuo <- rarefy_even_depth(OBJ1_Duo, sample.size = 3000, rngseed = TRUE, replace = FALSE, trimOTUs = TRUE) OBJ1_rIle <- rarefy_even_depth(OBJ1_Ile, sample.size = 8000, rngseed = TRUE, replace = FALSE, trimOTUs = TRUE) OBJ1_rCae <- rarefy_even_depth(OBJ1_Cae, sample.size = 10000, rngseed = TRUE, replace = FALSE, trimOTUs = TRUE) OBJ1_rFaeces <- rarefy_even_depth(OBJ1_Faeces, sample.size = 10000, rngseed = TRUE, replace = FALSE, trimOTUs = TRUE)

# Functions that are used here # transform_sample_counts: Using this function the abundance values of the data have been transformed according to the transformation specifiecd by the function (sample-wise) # tax_glom: This function merges species that have the same taxonomy # facet_grid:This function is used to split up data by one or two variables based on horizontal and/or vertical direction plot_bar(OBJ1_r, "location", fill="Phylum", facet_grid=~genotype) OBJ1_phy <- tax_glom(OBJ1,taxrank = "Phylum") plot_bar(OBJ1_phy, "location", fill="Phylum", facet_grid=~genotype) plot_bar(OBJ1_Duo, "location", fill="Phylum", facet_grid=~genotype) OBJ1_phy_Duo <- tax_glom(OBJ1_Duo,taxrank = "Phylum") plot_bar(OBJ1_phy_Duo, "location", fill="Phylum", facet_grid=~genotype) OBJ1_phy_Ile <- tax_glom(OBJ1_Ile,taxrank = "Phylum") plot_bar(OBJ1_phy_Ile, "location", fill="Phylum", facet_grid=~genotype) OBJ1_phy_Faeces <- tax_glom(OBJ1_Faeces,taxrank = "Phylum") plot_bar(OBJ1_phy_Faeces, "location", fill="Phylum", facet_grid=~genotype) OBJ1_phy_Cae <- tax_glom(OBJ1_Cae,taxrank = "Phylum") plot_bar(OBJ1_phy_Cae, "location", fill="Phylum", facet_grid=~genotype) ##transform to relative abundance: OBJ1_r_phy <- transform_sample_counts(OBJ1_phy, function(x) x / sum(x) ) plot_bar(OBJ1_r_phy, "location", fill="Phylum", facet_grid=~genotype) OBJ1_r_phy_Duo <- transform_sample_counts(OBJ1_phy_Duo, function(x) x / sum(x) ) OBJ1_rr_phy_Duo <- tax_glom(OBJ1_r_phy_Duo,taxrank = "Phylum") plot_bar(OBJ1_rr_phy_Duo, "location", fill="Phylum", facet_grid=~genotype) OBJ1_r_phy_Ile <- transform_sample_counts(OBJ1_phy_Ile, function(x) x / sum(x) ) OBJ1_rr_phy_Ile <- tax_glom(OBJ1_r_phy_Ile,taxrank = "Phylum") plot_bar(OBJ1_rr_phy_Ile, "location", fill="Phylum", facet_grid=~genotype) OBJ1_r_phy_Cae <- transform_sample_counts(OBJ1_phy_Cae, function(x) x / sum(x) ) OBJ1_rr_phy_Cae <- tax_glom(OBJ1_r_phy_Cae,taxrank = "Phylum") plot_bar(OBJ1_rr_phy_Cae, "location", fill="Phylum", facet_grid=~genotype) OBJ1_r_phy_Faeces <- transform_sample_counts(OBJ1_phy_Faeces, function(x) x / sum(x) ) OBJ1_rr_phy_Faeces <- tax_glom(OBJ1_r_phy_Faeces,taxrank = "Phylum") plot_bar(OBJ1_rr_phy_Faeces, "location", fill="Phylum", facet_grid=~genotype) ##remove rare OTUs OBJ1_rr_phy <- transform_sample_counts(OBJ1_r_phy, function(x) x / sum(x) ) OBJ1_phy1 <- tax_glom(OBJ1_rr_phy,taxrank = "Phylum") OBJ1_phy2 <- filter_taxa(OBJ1_phy1, function(x) mean(x) > 0.005, TRUE) plot_bar(OBJ1_phy1, "location", fill="Phylum", facet_grid=~genotype) plot_bar(OBJ1_phy2, "location", fill="Phylum", facet_grid=~genotype)

Appendix 8: R scripts for making bar charts

plot_bar(OBJ1_phy2, "location", fill="Phylum", facet_grid=location~genotype) plot_bar(OBJ1_ord, fill="Order", facet_grid=location~genotype) ###### remove rare OTUs of caecum OBJ1_phy1_Cae<- filter_taxa(OBJ1_rr_phy_Cae, function(x) mean(x) > 0.005, TRUE) plot_bar(OBJ1_phy1_Cae, "location", fill="Phylum", facet_grid=~genotype) ##again at lower tax rank OBJ1_rr_phy <- transform_sample_counts(OBJ1_r_phy, function(x) x / sum(x) ) OBJ1_ord <- tax_glom(OBJ1,taxrank = "Order") OBJ1_ord <- filter_taxa(OBJ1_ord, function(x) mean(x) > 0.005, TRUE) plot_bar(OBJ1_ord, "Phylum", fill="Order", facet_grid=~location) plot_bar(OBJ1_ord, "Phylum", fill="Order", facet_grid=~genotype) #######again at lower tax rank for OBJ1_ord <- tax_glom(OBJ1,taxrank = "Order") OBJ1_ord <- filter_taxa(OBJ1_ord, function(x) mean(x) > 0.005, TRUE) OBJ1_Duo <- subset_samples(OBJ1_ord, location == "Duodenum") OBJ1_Duo<- filter_taxa(OBJ1_Duo, function(x) mean(x) > 0.005, TRUE) plot_bar(OBJ1_Duo, "Phylum", fill="Order", facet_grid=~location) plot_bar(OBJ1_Duo, "Phylum", fill="Order", facet_grid=~genotype) OBJ1_Ile<- subset_samples(OBJ1_ord, location == "Ileum ") OBJ1_Ile<- filter_taxa(OBJ1_Ile, function(x) mean(x) > 0.005, TRUE) plot_bar(OBJ1_Ile, "Phylum", fill="Order", facet_grid=~location) plot_bar(OBJ1_Ile, "Phylum", fill="Order", facet_grid=~genotype) OBJ1_Cae <- subset_samples(OBJ1_ord, location == "Caecum ") OBJ1_cae<- filter_taxa(OBJ1_Cae, function(x) mean(x) > 0.005, TRUE) plot_bar(OBJ1_cae, "Phylum", fill="Order", facet_grid=~location) plot_bar(OBJ1_cae, "Phylum", fill="Order", facet_grid=~genotype) OBJ1_Faeces<- subset_samples(OBJ1_ord, location == "Colon") OBJ1_Faeces<- filter_taxa(OBJ1_Faeces, function(x) mean(x) > 0.005, TRUE) plot_bar(OBJ1_Faeces, "Phylum", fill="Order", facet_grid=~location) plot_bar(OBJ1_Faeces, "Phylum", fill="Order", facet_grid=~genotype) #custom barchart #this method will remove the segmentation in the bar plots library(microbiome) install.packages("microbiome") library(ggplot2) install.packages("ggplot2") install.packages("ggExtra") OBJ1_ord <- tax_glom(OBJ1,taxrank = "Order") OBJ1_ord_g <- merge_samples(OBJ1_ord, "WT_KI_location") ##### to group the samples sample_data(OBJ1_ord) #### allows us to observe the data set OBJ1_ord_gr <- transform_sample_counts(OBJ1_ord_g, function(x) x/ sum(x)) #####relative abundance OBJ1_ord2 <- filter_taxa(OBJ1_ord_gr, function(x) mean(x) > 0.00005, TRUE) #### filtering since there are more than 90 order levels(sample_data(OBJ1_ord)$WT_KI_location) #### allow us to observe how much order are removed #OBJ1_rr <- transform(OBJ1_r, "compositional")#transform to relative abundance #OBJ1_ord <- tax_glom(OBJ1_rr,taxrank = "Order")#concatenate to tax level we want to plot ###OBJ1_ord2 <- filter_taxa(OBJ1_ord, function(x) mean(x) > 0.005, TRUE) #if we want remove really rare stuff ####make a giant colour vector: library(RColorBrewer) colvec1 <- brewer.pal(11, "RdYlBu") colvec4 <- brewer.pal(10, "PuOr") colvec3 <- brewer.pal(10, "BrBG") colvec2 <- brewer.pal(10, "PiYG") colvec5 <- brewer.pal(10, "PiYG") colvec6 <- brewer.pal(10, "BrBG") colvec <- rep(c(colvec1, colvec2,colvec3,colvec4,colvec5,colvec6),2) melt<-psmelt(OBJ1_ord2)# we are extracting the information from the phyloseq object head(melt) #check column headings melt<-melt[sort.list(melt[,9]), ] #reorder 'melt' by the column we will be plotting by (in this cae 'order is in col 10) ####ordering the x axis labels: melt$Sample<- as.character(melt$Sample) ####Then turn it back into an ordered factor melt$Sample <- factor(melt$Sample, levels=unique(melt$Sample)) #make a vetor that puts the grouping factor in the order you want. Write the grouping factors as the appear in main file #phyloseq object. use melt$location (or melt$location etc) to fien out what teh grouping factes appear as melt$Sample <- factor(melt$Sample, levels=c("WT_Duodenum","WT_Ileum ","WT_Caecum ","WT_Colon","KI_Duodenum","KI_Ileum ","KI_Caecum ","KI_Colon")) #these are grouping factors #optinoal make a vector of the labels you want to apper along the x-axis. #below vector is empty, fill it with labels as we want #so make sure you labes correpond the the order thing will appear in whcih you determined above lab <- c("WT_Duodenum","WT_Ileum ","WT_Caecum ","WT_Colon","KI_Duodenum","KI_Ileum ","KI_Caecum ","KI_Colon") t1<- ggplot(melt, aes(Sample, Abundance, fill=Order))+ geom_bar(stat = "identity",) t1 <- t1 + theme(axis.text.x = element_text(angle = 45, size = 9,vjust = 0.7))+ scale_fill_manual(values = colvec) t1 <- t1+ labs(title = NULL, x= "Microbiome samples" , y= "Relative abundance") t1

Appendix 9: R scripts for NMDS ##some ordinations first library("ggplot2") UNI1 <- ordinate(OBJ1_rDuo, "PCoA", distance="unifrac", weighted=TRUE, parallel=TRUE) p1<-plot_ordination(OBJ1_rDuo, UNI1, color="genotype", shape="location", label=NULL) print(p1) #look at eigen/stress values for PCoA/NMDS UNI1 ######## NMDS UNI1 <- ordinate(OBJ1_r, "NMDS", distance="unifrac", weighted=TRUE, parallel=TRUE) p1<-plot_ordination(OBJ1_r, UNI1, color="genotype", shape="location", label=Null) print(p1) UNI1 colvec<-c("green4","red3", "darkorange1","green4", "red3","darkorange1") #this should reflect the number of groups you compare i.e 2 not 6 outline <- c(rep("black",6)) #this should reflect the number of groups you compare i.e 2 not 6 ###lab<- c("WT_Duodenum","WT_Ileum","WT_Caecum","WT_Colon","KI_Duodenum","KI_Ileum","KI_Caecum","KI_Colon") ## NMDS for Duodenum: UNI1 <- ordinate(OBJ1_rDuo, "NMDS", distance="unifrac", weighted=TRUE, parallel=TRUE) p1<-plot_ordination(OBJ1_rDuo, UNI1, color="genotype", shape="location", label=NULL) print(p1) p1 <- p1 + theme(axis.title.y = element_text(angle = 90, hjust = 0.5)) p1 <- p1 + geom_point(aes(colour=factor(genotype), fill = factor(genotype), shape= factor(location)), size = 3.5)+ ggtitle(NULL)+ theme(legend.position="none") p1 <- p1 + scale_shape_manual(values = c(21,24)) #### other shape IDs for this chosoe any number correspondign to a blue shape in the image in this link http://www.sthda.com/english/wiki/ggplot2-point-shapes p1 <- p1 + scale_colour_manual(values = outline) #####this is shape outline colour (i.e. black) p1 <- p1 + scale_fill_manual(values = colvec) ######this is shape fill colour p1 ## NMDS for Ileum: UNI1 <- ordinate(OBJ1_rIle, "NMDS", distance="unifrac", weighted=TRUE, parallel=TRUE) p1<-plot_ordination(OBJ1_rIle, UNI1, color="genotype", shape="location", label=NULL) print(p1) p1 <- p1 + theme(axis.title.y = element_text(angle = 90, hjust = 0.5)) p1 <- p1 + geom_point(aes(colour=factor(genotype), fill = factor(genotype), shape= factor(location)), size = 3.5)+ ggtitle(NULL)+ theme(legend.position="none") p1 <- p1 + scale_shape_manual(values = c(21,24)) #####choose other shape IDs for this chosoe any number correspondign to a blue shape in the image in this link http://www.sthda.com/english/wiki/ggplot2-point-shapes p1 <- p1 + scale_colour_manual(values = outline) #####this is shape outline colour (i.e. black) p1 <- p1 + scale_fill_manual(values = colvec) ######this is shape fill colour p1 ## NMDS for Caecum: UNI1 <- ordinate(OBJ1_rCae, "NMDS", distance="unifrac", weighted=TRUE, parallel=TRUE) p1<-plot_ordination(OBJ1_rCae, UNI1, color="genotype", shape="location", label=NULL) print(p1) p1 <- p1 + theme(axis.title.y = element_text(angle = 90, hjust = 0.5)) p1 <- p1 + geom_point(aes(colour=factor(genotype), fill = factor(genotype), shape= factor(location)), size = 3.5)+ ggtitle(NULL)+ theme(legend.position="none") p1 <- p1 + scale_shape_manual(values = c(21,24)) ####choose other shape IDs for this chosoe any number correspondign to a blue shape in the image in this link http://www.sthda.com/english/wiki/ggplot2-point-shapes p1 <- p1 + scale_colour_manual(values = outline) #####this is shape outline colour (i.e. black) p1 <- p1 + scale_fill_manual(values = colvec) ######this is shape fill colour p1 ## NMDS for Colon: UNI1 <- ordinate(OBJ1_rFaeces, "NMDS", distance="unifrac", weighted=TRUE, parallel=TRUE) p1<-plot_ordination(OBJ1_rFaeces, UNI1, color="genotype", shape="location", label=NULL) print(p1) p1 <- p1 + theme(axis.title.y = element_text(angle = 90, hjust = 0.5)) p1 <- p1 + geom_point(aes(colour=factor(genotype), fill = factor(genotype), shape= factor(location)), size = 3.5)+ ggtitle(NULL)+ theme(legend.position="none") p1 <- p1 + scale_shape_manual(values = c(21,24)) #####you can choose other shape IDs for this chosoe any number correspondign to a blue shape in the image in this link http://www.sthda.com/english/wiki/ggplot2-point-shapes p1 <- p1 + scale_colour_manual(values = outline) #####this is shape outline colour (i.e. black) p1 <- p1 + scale_fill_manual(values = colvec) ######this is shape fill colour p1

Appendix 10: R scripts for Statistics (PERMANOVA) ##anosims are good for one-way beta diversity analysis: Group_ano <- anosim(distance(OBJ1_r, "wunifrac"), Location) Group_ano Group_ano <- anosim(distance(OBJ1_r, "unifrac"), Location) Group_ano ##adonis is similar but can handle more complex designs ##make distance matricies (these are weighted and unweighted unifrac. can also use 'bray' OBJ1_r_wu <- distance(OBJ1_r, "wunifrac") OBJ1_r_u <- distance(OBJ1_r, "unifrac") Group_ado = adonis(OBJ1_r_wu ~ Location * Genotype) Group_ado Group_ado = adonis(OBJ1_r_u ~ Location * Genotype) Group_ado plot(OBJ1_r_wu) plot(OBJ1_r_u) Appendix 11: R scripts for differential analysis

###### differential analysis data<-read.csv("Differential_Analysis.csv") attach(data) par(mfrow = c(1,1), mar = c(4,10,1.0,1),mgp = c(2,0.7,0)) stripchart(log(R.Change)~Gut_Region, data = data,xlim=c(-3,3),ylim=c(1,5),vertical = "FALSE", method = "stack", jitter = 0.01, las = 1,cex.lab=1.2,cex.axis=1,xlab = "relative abundance change (Wild type-->knockout)") Appendix 12: R scripts for compositional analysis

###compositional analysis within similar group to observe the microbial shift along the GI tract data<-read.csv("miseq_work_file_s/compositional_Analysis.csv") attach(data) library(ggplot2) library(plyr) Mdata <- ddply(data, c("location", "Genotype"), summarise, N = length(Bray.Curtis), mean = mean(Bray.Curtis), sd = sd(Bray.Curtis), se = sd / sqrt(N)) Mdata str(Mdata) # Error bars represent standard error of the mean ggplot(Mdata, aes(x=location, y=mean, fill=Genotype))+ geom_bar(position=position_dodge(), stat="identity") + geom_errorbar(aes(ymin=mean-se, ymax=mean+se), width=.2, position=position_dodge(.9)) ####compositional analysis between groups (WT and NL3) data<-read.csv("miseq_work_file_s/compositional_WT_KI.csv") attach(data) library(ggplot2) library(plyr) Mdata <- ddply(data, c("location"), summarise, N = length(Bray.Curtis), mean = mean(Bray.Curtis), sd = sd(Bray.Curtis), se = sd / sqrt(N)) Mdata str(Mdata) # Error bars represent standard error of the mean ggplot(Mdata, aes(x=location, y=mean, fill=location))+ geom_bar(position=position_dodge(), stat="identity") + geom_errorbar(aes(ymin=mean-se, ymax=mean+se), width=.2, position=position_dodge(.9))

Master's thesis of Science: Characterising microbial and neuroimmune interactions in a mouse model of autism

Characterising Microbial and Neuroimmune Interactions in a Mouse Model of Autism

Samiha Sayed Sharna

School of Health and Biomedical Science

College of Science, Engineering and Health

RMIT University

March 2020

Declaration

Publications arising from this research

Original research articles:

Review article:

Conference Abstracts:

Acknowledgements

Table of contents

Declaration

Publications arising from this research

Acknowledgments

List of figures

List of abbreviations and units

Abstract

Chapter 1: Introduction

Chapter 2: Altered caecal neuroimmune interactions in the Neuroligin-3R451C mouse model of autism

Chapter 3: Altered Microbial composition in the Neuroligin-3R451C mouse model of autism

Chapter 4: Discussion

Conclusion

Future directions

References

Appendices

Appendix 1: Ethics approval letter

Appendix 2: Haematoxylin & Eosin staining

Appendix 3: Caecal patch analysis using ImageJ to determine the cell density

Appendix 4: Using ImageJ, analysis of number of enteric neurons/ganglia

Appendix 5: Iba1 immunoreactive cell analysis using Imaris7

Appendix 6: Mouse ID used for microbial analysis

Appendix 7: R scripts for data preparation

Appendix 8: R scripts for making bar charts

Appendix 9: R scripts for NMDS

Appendix 10: R scripts for Statistics (PERMANOVA)

Appendix 11: R scripts for compositional analysis

Appendix 12: R scripts for differential analysis

List of figures

Chapter 1

1.1

1.2

Chapter 2

2.1

2.2

2.3

2.4

2.5

Chapter 3

3.1

2.2

3.3

3.4

3.5

List of abbreviations and units

Abstract

Chapter 1

1.0 INTRODUCTION

1.1 Overview

1.2 Genetics in ASD

1.2.1 The R451C mutation in Neuroligin-3

1.3 The enteric nervous system

1.3.1 Altered neuro-immune interactions in NL3 R451C mice

Figure 1.1 Myenteric ganglia of proximal jejunum in WT and NL3R451C. Confocal images (x 20) of Hu (red) and

1.4 Gut associated lymphoid tissue (GALT)

1.4.1 Caecal Patch

1.5 Microbial influences on gut-associated lymphoid tissue

1.6 Microbiome-neuroimmune signatures in ASD

Figure 1.2 The effect of genotype on faecal microbial communities at five and nine weeks of age of WT and

NL3R451C mice. Presence and absence analysis (A-B) and non-metric multidimensional scaling (NMDS) ordinations

1.8 PROJECT RATIONALE

1.8.1 Abnormal neuroimmune pathways in NL3R451C mice

1.8.2 Alterations in gut microbes in ASD

1.9 HYPOTHESIS AND AIMS

Chapter 2

Altered caecal neuroimmune interactions in the

Neuroligin-3R451C mouse model of autism

2.1 Abstract