Master's thesis of Science: Intestinal permeability and functional properties of duodenal enteric neurons in a mouse model of autism

Intestinal Permeability and Functional Properties of Duodenal Enteric Neurons in a Mouse Model of Autism. A thesis submitted in fulfilment of the requirements for the degree of Master of Science Joshua Kenneth Williams BSc in Biomedical science and Biotechnology (RMIT University, Melbourne) School of Health and Biomedical Science College of Science, Technology, Engineering and Math RMIT University August 2022

Thesis declaration

I certify that except where due acknowledgement has been made, this research is that of the author

alone; the content of this research submission 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.

In addition, I certify that this submission contains no material previously submitted for award of any

qualification at any other university or institution, unless approved for a joint-award with another

institution, and acknowledge that no part of this work will, in the future, be used in a submission in

my name, for any other qualification in any university or other tertiary institution without the prior

approval of the University, and where applicable, any partner institution responsible for the joint-

award of this degree.

I acknowledge that copyright of any published works contained within this thesis resides with the

I give permission for the digital version of my research submission to be made available on the web,

via the University’s digital research repository, unless permission has been granted by the University

to restrict access for a period of time.

I acknowledge the support I have received for my research through the provision of an Australian

Government Research Training Program Scholarship.

Joshua Kenneth Williams.

02 August 2022.

Acknowledgements

I would like to express my deepest gratitude to my supervisors A/Prof. Elisa Hill and Dr Suzanne Hosie.

Both Elisa and Suzanne were encouraging, inspiring and supportive throughout my candidature.

Thank you both for providing me with constructive criticism, along with skills in electrophysiology,

permeability, scientific presentations, and scientific writing. I would also like to acknowledge my lab

members: Jackson, Tanya, Chalystha, Pasindu, Rachele, Samantha and Miti for providing me with

invaluable feedback on my presentations and troubleshooting problems with experiments that arose

during my candidature. I would like to acknowledge my partner Latasha who provided constant

emotional support and encouragement throughout my candidature.

Table of Contents

Thesis declaration .............................................................................................................................. i

Acknowledgements .......................................................................................................................... ii

List of Figures .................................................................................................................................. vi

List of Tables ................................................................................................................................... vii

List of abbreviations and units......................................................................................................... vii

Abstract............................................................................................................................................ 1

Chapter 1: Introduction ............................................................................................................... 3

1.0 Autism Spectrum Disorder (ASD) overview .................................................................................. 3 1.1 Gastrointestinal disturbances in Autism patients. ...................................................................................................... 4 1.2 The gastrointestinal tract ............................................................................................................................................ 5 1.3 The enteric nervous system ........................................................................................................................................ 8 1.4 The submucosal plexus................................................................................................................................................ 9 1.5 The myenteric plexus .................................................................................................................................................. 9

2.0 The gastrointestinal mucosal barrier .......................................................................................... 10 2.1 Intestinal epithelial cells ............................................................................................................................................ 12 2.2 Apical junction protein complex ............................................................................................................................... 13 2.2.1 Tight junctions............................................................................................................................................... 14 2.2.2 Paracellular permeability .............................................................................................................................. 16 2.3 Gastrointestinal distress and ASD ............................................................................................................................. 17 2.4 Restoration of intestinal permeability ...................................................................................................................... 18 2.4.1 The impact of L-glutamine on intestinal permeability ................................................................................. 20 2.4.2 The impact of caffeine on intestinal permeability ........................................................................................ 23

3.0 Classification of enteric neurons ................................................................................................ 26 3.1. Morphological classification of enteric neurons ...................................................................................................... 26 3.2 Electrophysiological classification of enteric neurons. ............................................................................................. 28 3.2.1 Synaptic-neurons (S-neurons) ...................................................................................................................... 28 3.2.2 AH (after-hyperpolarisation) neurons .......................................................................................................... 28 3.3 Functional classification of enteric neurons.............................................................................................................. 29 3.3.1 Intrinsic primary afferent neurons (IPANs) ................................................................................................... 30 3.3.2 Interneurons ................................................................................................................................................. 31 3.3.3 Muscle motor neurons.................................................................................................................................. 32 3.4 Neurochemical classification of myenteric neurons ................................................................................................. 33

4.0 ASD genetic mutations .............................................................................................................. 35 4.1 ASD and synaptic cell adhesion molecules................................................................................................................ 35 4.2 Neuroligins ................................................................................................................................................................. 36 4.3 Neuroligin-3 overview ............................................................................................................................................... 37 4.4 Neuroligin functionality ............................................................................................................................................. 37 4.5 NL3R451C mutation ...................................................................................................................................................... 38 4.6 Expression of Nlgn3 in the gastrointestinal tract ...................................................................................................... 39

5.0 Project rationale ....................................................................................................................... 41 5.1 Assessing permeability in an autism mouse model. ................................................................................................. 41 5.2 Action potential characteristics in duodenal myenteric neurons using an autism mouse model ........................... 42 5.3 Aims and hypotheses ................................................................................................................................................ 43

Chapter 2: Measuring intestinal permeability in the Neuroligin-3R451C mouse model of autism. ... 44

1.0 Introduction .............................................................................................................................. 44

2.0 Methods and materials ............................................................................................................. 45

iii

2.1 Animals ...................................................................................................................................................................... 45 2.2 Segmentation of the small and large intestine ......................................................................................................... 45 2.3 Preparation of L-glutamine and caffeine stock solutions ......................................................................................... 46 2.4 Injection of FITC-Dextran 4. ....................................................................................................................................... 47 2.5 Time course permeability experiment ...................................................................................................................... 47 2.6 Construction of standard curve using log serial dilutions ......................................................................................... 49 2.7 Statistical analysis ...................................................................................................................................................... 50

3.0 Results ...................................................................................................................................... 51 3.1 Small intestinal permeability in non-fasted wild-type and mutant mice ................................................................. 51 3.2 Permeability effects in wild-type and mutant fasted mice ....................................................................................... 53 3.3 Effects of fasting on NL3R451C mice ............................................................................................................................ 54 3.4 Effects of fasting on wild-type mice .......................................................................................................................... 56 3.5 Effects of L-glutamine on fasted NL3R451C mice ......................................................................................................... 58 3.6 Effects of L-glutamine on fasted wild-type mice ....................................................................................................... 60 3.7 Effects of caffeine on fasted NL3R451C mice ............................................................................................................... 62 3.8 Effects of caffeine on fasted wild-type mice ............................................................................................................. 64

4.0 Discussion ................................................................................................................................. 66 4.1 Understanding how NL3R451C mutation and feeding conditions effect paracellular permeability ........................... 66 4.2 Understanding how L-glutamine and caffeine impact paracellular permeability .................................................... 70

5.0 Conclusion ................................................................................................................................ 77

Chapter 3: Optimisation of the patch-clamp recording technique in the enteric nervous system to examine action potential characteristics in mouse duodenal myenteric neurons. ....................... 78

1.0 Introduction .............................................................................................................................. 78

2.0 Methods and materials ............................................................................................................. 80 2.1 Animals ...................................................................................................................................................................... 80 2.2 General perfusion/dissecting Krebs solution ............................................................................................................ 80 2.3 Microdissection ......................................................................................................................................................... 81 2.4 Identification of myenteric ganglion. ........................................................................................................................ 83 2.5 Protease solution ....................................................................................................................................................... 83 2.6 Whole-cell patch recording ....................................................................................................................................... 83 2.7 External patching solution ......................................................................................................................................... 84 2.8 Current clamp recording protocol ............................................................................................................................. 85 2.9 Statistical analysis ...................................................................................................................................................... 87

3.0 Results ...................................................................................................................................... 88 3.1 Comparison of success rate of neuronal recordings for older versus younger mice................................................ 88 3.2 Recording action potentials in younger mice. ........................................................................................................... 90 3.2.1 Comparison of firing properties in 3 myenteric neurons ............................................................................. 92 3.2.2 Action potential characteristics in younger mice ......................................................................................... 92 3.3 Action potential characteristics observed in Neurons 1, 2 and 3 ............................................................................. 93 3.4 Neuronal profiles in younger mice ............................................................................................................................ 97 3.4.1 Comparison of action potentials full trace ................................................................................................... 98 3.4.2 Comparison of current-voltage (IV) curves ................................................................................................. 100 3.4.3 Comparison of action potentials and current curves ................................................................................. 100 3.5 Discussion ................................................................................................................................................................ 102 3.5.1 Success rate of whole-cell patch clamp recording in older versus younger mice. ..................................... 102 3.5.2 Future directions for patch-clamp electrophysiology in the ENS ............................................................... 104 3.6 Conclusion ............................................................................................................................................................... 107

References ................................................................................................................................... 108

Appendices ................................................................................................................................... 118 Appendix: Ethics approval letter ................................................................................................................................... 118 Appendix: Current clamp protocol used for whole-cell patch clamping. ..................................................................... 118

Electrophysiology appendices ....................................................................................................... 119 Appendix Table 1. .......................................................................................................................................................... 119 Appendix Table 2. .......................................................................................................................................................... 119 Appendix Table 3 ........................................................................................................................................................... 120 Appendix Table 4. .......................................................................................................................................................... 120

Permeability appendices ............................................................................................................... 121 Appendix Table 5. .......................................................................................................................................................... 121 Appendix Table 6. .......................................................................................................................................................... 121 Appendix Table 7. .......................................................................................................................................................... 121 Appendix Table 8 ........................................................................................................................................................... 122 Appendix Table 9. .......................................................................................................................................................... 122 Appendix Table 10. ........................................................................................................................................................ 122 Appendix Table 11. ........................................................................................................................................................ 122 Appendix Table 12. ........................................................................................................................................................ 123 Appendix Table 13. ........................................................................................................................................................ 123 Appendix Table 14. ........................................................................................................................................................ 123 Appendix Table 15. ........................................................................................................................................................ 123 Appendix Table 16. ........................................................................................................................................................ 124 Appendix Table 17. ........................................................................................................................................................ 124 Appendix Table 18. ........................................................................................................................................................ 124 Appendix Table 19. ........................................................................................................................................................ 125 Appendix Table 20. ........................................................................................................................................................ 126 Appendix Table 21. ........................................................................................................................................................ 127 Appendix Table 22. ........................................................................................................................................................ 128 Appendix Table 23. ........................................................................................................................................................ 129 Appendix Table 24. ........................................................................................................................................................ 130 Appendix Table 25. ........................................................................................................................................................ 131 Appendix Table 26. ........................................................................................................................................................ 132 Appendix Table 27. ........................................................................................................................................................ 133 Appendix Table 28. ........................................................................................................................................................ 134 Appendix Table 29. ........................................................................................................................................................ 135 Appendix Table 30. ........................................................................................................................................................ 136 Appendix Table 31. ........................................................................................................................................................ 136 Appendix Table 32. ........................................................................................................................................................ 137 Appendix Table 33. ........................................................................................................................................................ 137 Appendix table 34.......................................................................................................................................................... 137

Figure 1: Organization of the mouse gastrointestinal tract ............................................................................................... 6 Figure 2: The organization of the enteric nervous system. ................................................................................................ 8 Figure 3: The physiology of the mucus layer in the colon is different to the small intestine .......................................... 11 Figure 4: Cross-sectional image of small intestine depicting the major cell types .......................................................... 12 Figure 5: Tight junctions, desmosomes and adherens junctions, make up apical junctional complexes. ...................... 13 Figure 6: Intestinal epithelial cells utilise both transcellular and paracellular pathways. ............................................... 16 Figure 7: Potential mechanisms for L-glutamine restoring paracellular permeability in IEC’s. ....................................... 21 Figure 8: Overeating and duration of food deprivation influences intestinal mucus composition. ................................ 24 Figure 9: Three major functional classes of enteric neurons. .......................................................................................... 28 Figure 10: Setup for measuring the permeability of FITC through the paracellular route. ............................................. 46 Figure 11: Standard curve of known concentrations of FITC with absorbance ............................................................... 47 Figure 12: Small intestinal paracellular permeability in non-fasted NL3R451C and WT mice............................................. 49 Figure 13: Small and large intestinal paracellular permeability fasted NL3R451C and WT mice ........................................ 51 Figure 14: Small intestinal paracellular permeability for fasted and non-fasted NL3R451C mice ...................................... 52 Figure 15: Effect of fasting and non-fasting on intestinal paracellular permeability in WT mice.................................... 54 Figure 16: L-Glutamine restored paracellular permeability in NL3R451C mice to WT levels in small and large intestinal regions .............................................................................................................................................................................. 56 Figure 17: Small and large intestinal paracellular permeability on wild-type mice treated with L-glutamine ............... 58 Figure 18: Caffeine restored paracellular permeability in NL3R451C mice to WT concentrations in small and large intestinal regions .............................................................................................................................................................. 60 Figure 19: Caffeine decreased paracellular permeability in WT mice ............................................................................. 62 Figure 20: Visual representation of obtaining an LMMP preparation ............................................................................. 78 Figure 21: Visual representation of the general set up of patch-clamping...................................................................... 81 Figure 22: Custom-made hair cell attached to a glass micropipette ............................................................................... 81 Figure 23: A myenteric neuron action potential with measured characteristics ............................................................. 83 Figure 24: Number of times whole cell seal configuration was obtained ........................................................................ 84 Figure 25: Number of times whole cell seal obtained and/or AP firing recorded in myenteric neurons using younger and older mice .................................................................................................................................................................. 85 Figure 26: Three duodenal myenteric neurons exhibited action potentials in response to current steps ..................... 87 Figure 27: Action potential characteristics of three neurons........................................................................................... 92 Figure 28: Characterisation of action potential profiles recorded throughout the full current injection protocol for three duodenal myenteric neurons.................................................................................................................................. 93

List of Figures

Table 1: Myenteric neurons are classified according to their neurochemical coding Table 2: Known concentrations of FITC and absorbance produced Table 3: Action potential parameters for three duodenal myenteric neurons

34 50 93

List of Tables

List of abbreviations and units

AH: After-hyperpolarisation

AHP: After-hyperpolarisation potential

ASD: Autism Spectrum Disorder

ATP: Adenosine triphosphate

BA: Bile acid

cAMP: cyclic adenosine monophosphate

ChAT: Choline acetyltransferase

CM: Circular muscle

Cm: Membrane capacitance

CNS: Central Nervous System

DMEM: Dulbecco’s modified eagle medium

EEC: Enteroendocrine cell

EGF: Epidermal growth factor

ENS: Enteric Nervous System

ER: Endoplasmic reticulum

mTOR: Mechanistic target of rapamycin

FEPSPs: Fast excitatory post-synaptic potential

FITC: Fluorescein Isothiocyanate

GABA: Gamma-Aminobutyric acid

vii

GI: Gastrointestinal

HFD: High-fat diet

HSP: Heat shock proteins.

HSF: Heat shock factor

ICC: Interstitial cells of Cajal

IEC: Intestinal epithelial cell

IGF: Insulin-like growth factor

IPAN: Intrinsic primary afferent neurons

ISN: Intrinsic sensory neuron

I-V curve: Current – Volt curve

JAM: Junctional adhesion protein

KO: Knock-out

LM: Longitudinal-muscle

LMMP: Longitudinal-muscle-myenteric-plexus

LP: Lamina propria

LPS Lipopolysaccharide

MAPK: Mitogen-activated protein kinases

ML: Mucosa layer

MM: Muscularis mucosae

MP: Myenteric plexus

NO: Nitric oxide

Neuroligin: NLGN3, Nlgn3 or NL3

NRXN: Neuroexin

NOS: Nitric Oxide Synthase

viii

NF-B: nuclear factor-B

pA: Picoamp

PYY: Peptide YY

Ra: Access resistance

Rm: Membrane resistance

RMP: Resting membrane potential

SCFA: Short chain fatty acids

S-neurons: Synaptic-neurons

SMP: Submucosal plexus

Tau: Membrane time constant

TGF: Transforming growth factor

TGF- α -: Transforming growth factor α

TJ: Tight junction

TMAO: trimethylamine N-oxide

VAchT: Vesicular acetylcholine transporter

VIP: Vaso-intestinal peptide

WT: Wild-type

ZO: Zonulin

Abstract

Autism Spectrum Disorder (ASD) is a complex neurodevelopmental disorder characterised by

impaired social communication and the presence of repetitive behaviours. In addition to these

behavioural characteristics, as many as ninety percent of individuals with ASD exhibit gastrointestinal

(GI) issues. A missense mutation at position R451C in the synaptic adhesion protein, neuroligin-3

(NLGN3) was previously identified in patients with ASD. The Nlgn3 gene is present in both the brain

and the intrinsic neural network of the gut, the enteric nervous system (ENS). Within the ENS, the

submucosal plexus (SMP) regulates mucosal barrier functions such as secretion and permeability,

while the myenteric plexus predominantly regulates gut motility. Clinical studies have shown that

individuals with ASD who experience GI disturbances have elevated blood concentrations of simple

sugars such as mannitol and lactulose, which is a marker for increased intestinal permeability.

Previous data from the current laboratory showed faster small intestinal transit time in NL3R451C mice

compared to wild-type littermates. We hypothesise that the R451C mutation affects intestinal

function (i.e., permeability and motility) in mice. Duodenal, jejunal, ileal and colonic paracellular

permeability was assessed using an ex-vivo whole tissue assay in non-fasted and fasted mice. The

effects of the R451C mutation on functional characteristics of duodenal myenteric neurons were also

investigated in wild type mice. No significant differences in intestinal permeability were observed

between non-fasted NL3R451C mice and wild-type littermates in any intestinal region. In a fasted state,

however, NL3R451C mice showed increased intestinal permeability in the duodenum, jejunum, ileum

and colon when compared to wild-type littermates. The addition of L-glutamine and caffeine rescued

the increased intestinal permeability in the duodenum, jejunum, ileum and colon in NL3R451C mice.

To characterise functional neuronal properties in the ENS, action potentials and baseline cellular

parameters were recorded from myenteric neurons using the whole-cell patch clamping technique

on Longitudinal Muscle Myenteric Plexus Preparations (LMMPs). The probability of obtaining a

recording was greater in younger (2-4-week-old) Swiss white and C57BL/6 mice in comparison to

older (6-12-week-old) C57Bl/6 mice. Different action potential firing patterns were observed which

may correlate with different functional subgroups of myenteric neurons. Characterising intestinal

permeability and categorizing enteric neurons by their functional profiles may be useful for

understanding and reducing GI symptoms, chronic low-grade systemic inflammation, and core

neurological symptoms in people with ASD.

Chapter 1: Introduction

1.0 Autism Spectrum Disorder (ASD) overview

ASD is a complex neurodevelopmental disorder that results in a range of clinical traits that can differ

in type and severity in each individual (Vahia, 2013). These clinical traits must meet specific criteria

outlined in The American Psychiatric Association’s Diagnostic and Statistical Manual, Fifth Edition

(Vahia, 2013). For instance, an individual with ASD will often display persistent deficits in social-

emotional reciprocity, nonverbal communicative behaviours (hand gestures and eye contact), verbal

communication and developing, maintaining and understanding relationships. In addition, the

individual must have persistent deficits in at least two types of repetitive or restricted behaviour

(Vahia, 2013). These include motor movements that are repetitive or stereotyped, difficulties in

managing distress at small changes, fixating on interests and hypo- or hyperactivity (Vahia, 2013).

The neuropathology underlying the behavioural abnormalities are proposed to include disrupted

neurocircuit connectivity, altered synaptic transmission, inhibitory and excitatory imbalance and

altered neurochemical signalling. Brain regions in which these abnormalities have been observed

include the posterior temporal sulcus, amygdala, adjacent anterior cingulate cortex, medial

prefrontal cortex and the temporal poles (Etherton et al., 2011, Ha et al., 2015, Tabuchi et al., 2007).

By the age of 8, 1 in 54 United States children are diagnosed with ASD (Baio et al., 2018). An increase

in the prevalence of ASD is thought to be due to implementation of new diagnostic criteria and better

monitoring systems (Baio et al., 2018). Although ASD is not specific to ethnic, racial and

socioeconomic groups, a recent meta-analysis (Loomes et al., 2017) outlines that ASD is more

common in males than females with a ratio of 3:1. Although potential genetic and environmental

causes for ASD have been explored; no unifying mechanism responsible for the aetiology has been

identified (Hodges et al., 2020).

1.1 Gastrointestinal disturbances in Autism patients.

Historically, ASD research has been overwhelmingly limited to psychological assays and

understanding alterations to the central nervous system (CNS). However, more recent advances have

identified that gastrointestinal disturbances are commonly experienced by ASD patients. The most

frequently reported disturbances in Individuals with ASD are diarrhoea, abdominal pain and

constipation (Buie et al., 2010, Coury et al., 2012). Other less frequently reported GI disturbances

include: bloody stool, flatulence, gastroesophageal reflux, celiac disease, esophagitis, belching and

Chron’s disease (Coury et al., 2012). These symptoms are frequently uncomfortable and disturbing

to a person's day-to-day life, and people with ASD have a fourfold increased risk of being hospitalised

due to gastrointestinal symptoms compared to the general population (McElhanon et al., 2014). In

addition, there is an association between the severity of GI symptoms and core features of autism

including social and language impairment (Gorrindo et al., 2012). Alleviating GI disturbances could

improve quality of life for patients and carers. For example, improved gut health could improve

participation in behavioural therapies and sleep patterns leading to potential improvements in mood

and behaviour. Although GI disturbances are disruptive and more commonly experienced by people

with ASD, astonishingly, the awareness of GI dysfunction in ASD is limited and there are no specific

treatments available (Rao and Bhagatwala, 2019).

1.2 The gastrointestinal tract

The major aims of this dissertation are to assess for changes in gut function by analysing permeability

in a mouse model of autism and to contribute to enhancing the classification system for enteric

neurons in the mouse myenteric plexus based on their action potential firing characteristics. The GI

tract consists of the oesophagus, stomach, small intestine, and large intestine. From the pylorus to

the ileocecal valve, the small intestine is divided into three sections: duodenum, jejunum, and ileum

(Furness, 2012). The small intestine's major role is nutrient absorption. The caecum is the most

proximal section of the large intestine, and its exact function in humans (i.e., the appendix) is

unknown. The large intestine (i.e., the colon), extends from the ileocecal valve to the rectum, and its

primary role is in water and electrolyte absorption (Furness, 2012). The small intestine's major role

is nutrient absorption. The gut is divided into functional regions, each of which has specific

anatomical characteristics. Although the general anatomical structure of the mammalian GI tract is

highly conserved, the anatomy and physiology of different species differ significantly. This could be

related to a variety of factors such as diet, metabolic needs, feeding patterns, and body size (Nguyen

et al., 2015). The mouse GI tract, for example, differs from the human GI tract in terms of

morphology, physiology, cellular structure, and genetics, despite several commonalities. As a result,

mouse models are widely used in GI research, as they are a good tool for assessing preliminary

differences caused by genetic mutations and as an indication for human research.

Figure 1: Organization of the mouse gastrointestinal tract Distinct regions of the gastrointestinal tract of a mouse. The stomach is proximal to the small intestine which is separated into three distinct regions; the duodenum, jejunum and ileum. The large intestine is also separated into three distinct regions. The caecum is distal to the Ileum with the colon being proximal to the rectum (Furness, 2012). The GI tract is comprised of distinctly different cellular layers. The innermost layers (i.e., in close

proximity to the host organs) are the longitudinal muscle (LM), adjacent to the myenteric plexus

(MP), with the next layer being the circular muscle (CM) alongside the SMP muscularis mucosae (MM)

and the mucosa layer (Etherton et al.) (Figure 2). The intestinal mucosa is a single layer of epithelial

cells acting as the interface between the external gut environment and the internal environment of

the body. Epithelial cells absorb nutrients, water, and electrolytes, produce a variety of digestive

secretions and act as a physical barrier to potentially hazardous luminal substances. Enteroendocrine

cells (EECs) and other modified epithelial cell subtypes produce regulatory proteins that have local

paracrine or neurocrine effects. Paneth cells, for example, release antimicrobial mediators which

prevent invasions of potential pathogenic microbes into the epithelium (Allaire et al., 2018, Elphick

and Mahida, 2005). Similarly, mucus is secreted by goblet cells to lubricate the epithelial lining and

form a protective barrier against some microorganisms in the gut. The distribution of secretory cell

types varies by gut area and is intimately linked to the function of each gut region. Within the SMP is

a network of cells that coordinate the absorption, secretion, and immune activity of the mucosa. The

SMP consists of blood vessels, smooth muscle cells, neurons, and glial cells (Allaire et al., 2018).

The GI tract, in broad terms, protects the host from harmful substances that pass through the

intestinal epithelium. Neurons from the SMP mostly innervate the mucosa (Neunlist et al., 2013).

Immune cells and neurons work together to orchestrate inflammatory responses and evoke neuronal

signalling pathways that help the body eliminate infections. Neural activity regulates gut contractile

function as well as impacting permeability and secretion (e.g., of ions and water). Running the entire

length of the GI tract are three clearly defined smooth muscle layers, the CM, LM and the MM. As

their names suggests, the cells of the CM layer are oriented in a circular direction around the gut and

change gut diameter, while the cells in the outer LM layer are oriented along the gut's longitudinal

direction and change the gut length. Outside the lamina propria (Elphick and Mahida, 2005), the MM

is a thin layer of muscle that serves to separate it from the submucosa. Coordinated contractions and

relaxation of the CM and LM layers shape the complex motility contractions of the GI tract. These

patterns range from non-propulsive mixing movements to highly propulsive peristaltic contractions

(Costa et al., 2000, Furness, 2012, Furness et al., 2014). Embedded within the muscle layers of the GI

tract are two ganglionated plexus known as the SMP and the MP. Together these make up the ENS,

an intrinsic neural network that can regulate intestine activities independently of the CNS (Furness

et al., 2014). The ENS governs practically all gastrointestinal activities, including mucosal barrier

function, secretion, motility, and blood flow, for efficient food digestion and absorption. Although

the ENS can function independently of the CNS, it receives extrinsic innervation from the brain and

spinal cord to coordinate vital GI activities (Furness et al., 2014).

Figure 2: The organization of the enteric nervous system. In humans and large animals, the mucosa is the intestine's exterior covering, which separates the lumen from the internal structures. The submucosal and myenteric plexuses are the two ganglionic plexuses of the ENS. The sub-mucosal plexus (SMP) runs between the mucosa layer and the circular muscle (CM) layer, and its ganglia are divided into one or three layers. The SMP innervates the mucosa and regulates mucosal barrier functions as secretion and permeability (Furness et al., 2014).

1.3 The enteric nervous system

The ENS is the largest division of the autonomic nervous system, with more than 100 million neurons

and 400 million neuron-supporting glial cells. This intricate neural system is crucial for maintaining

proper digestive function (Furness, 2012). Enteric nerve cells and glia are grouped together to form

ganglia in the ENS. A neural plexus is formed when ganglia are connected by nerve fiber bundles. The

SMP and the MP are two ganglionated plexuses in the ENS. The MP connects the LM and CM layers

of the external musculature, while the SMP connects the CM layer to the mucosa. The MP regulates

gut motility, while the SMP regulates mucosal barrier function, however, both plexuses work

together to ensure optimal GI functionality. Even though the ENS can perform functions independent

of the CNS, it is not autonomous. To govern local enteric reflexes, the integrated network of the ENS

and CNS mediates neuronal regulation of the GI system. Vagal nerve routes, the spinal thoracolumbar

spinal cord, and the pelvic pathway all play a role in neural communication between the CNS and the

ENS (Furness et al., 2013).

1.4 The submucosal plexus

The SMP is predominantly responsible for regulating water and electrolyte secretion along with

regulating localised blood flow (Furness, 2012). These processes are governed by a variety of enteric

neurons and elucidating the function of specific neuronal subtypes can offer insight into therapeutic

treatments for GI disorders. In the SMP of mice and guinea pigs, there are two pharmacologically and

neurochemically separate populations of neurons. Cholinergic neurons express choline

acetyltransferase (ChAT), the enzyme that synthesizes acetylcholine. Non-cholinergic neurons

possess vasoactive intestinal peptide (VIP) but lack the ability to express ChAT. Many VIP neurons in

humans and rats, on the other hand, express ChAT. SMP neurons express an array of neurochemicals

that provide them with a unique neurochemical code in addition to VIP and ChAT as two primary

neurotransmitters (Bornstein and Foong, 2018).

1.5 The myenteric plexus

The MP is located between the LM and CM layers and predominantly regulates GI motility including

peristaltic contractions of the smooth muscles to facilitate the transit of luminal contents (Furness,

2012). The MP comprises approximately 16 enteric neuronal subtypes which release a range of

excitatory (e.g. Acetylcholine and tachykinins) or inhibitory (i.e. nitric oxide, ATP (adenosine tri-

phosphate)-like transmitters and VIP) neurotransmitters to contract or relax the intestinal smooth

muscle, respectively (Furness, 2012).

2.0 The gastrointestinal mucosal barrier

The intestinal epithelial wall of the human Gl tract covers an estimated 400 m2 of mucosal surface

area in an adult individual. Microorganisms, digestive enzymes and acids, digested food particles,

microbial by-products, and food-associated toxins all penetrate the mucus layer, which serves as

the GI tract's first line of defence. This mucus layer coats the surface of the GI tract, lubricates the

luminal contents, and acts as a physical barrier to bacteria and other antigenic chemicals present in

the lumen. The moist, nutrient-rich mucus layer next to the epithelial barrier of the GI tract is also

critical for intestinal homoeostasis because it contains a robust biofilm including both beneficial

and pathogenic bacteria species, reviewed by Herath and co-authors (Herath et al., 2020). The GI

tract is continually under attack by pathogens, medications, nutrients, and bacterial toxins. Not only

must the host distinguish between commensal bacteria and potential pathogens, but the host must

also prevent these species and secreted molecules from crossing the epithelial barrier while

allowing nutrients to be absorbed. Thus, the intestinal epithelium functions as a selective barrier to

luminal substances. This is accomplished in part by the innate epithelial defence system of the

mucosa, which operates via a responsive biological system composed of constitutive and inducible

mechanisms. Therefore, if the intestinal epithelial barrier's function is impaired, the host may be

more vulnerable to a variety of GI disorders. Consequently, this thesis will discuss how increased

paracellular permeability may be a possible explanation for why the NL3R451C mouse model exhibits

a dysfunctional epithelial barrier and how natural therapeutic treatments/agents such as, fasting

from solid foods, L-glutamine and caffeine supplementation may restore a compromised epithelial

barrier in (ASD).

Figure3: The physiology of the mucus layer in the colon is different to the small intestine (A) the small intestine contains a single layer of mucus that is loosely linked to the epithelium and easily penetrable. Anti-microbial modulators effectively repel microorganisms from the small intestinal epithelium. (B) The distal colon has two mucus layers including a stratified adhering inner mucus layer (which is not penetrable by molecules greater than 1 μm diameter). The inner mucus layer of the colon is virtually sterile, whereas the outer mucus layer contains intestinal bacteria (Herath et al., 2020).

2.1 Intestinal epithelial cells

The intestinal mucosal barrier comprises a variety of extrinsic and intrinsic components that regulate

mucosal homeostasis. The intestinal epithelial cell layer and the tight junctions that connect these

cells to form a continuous single layer throughout the GI tract are essential components of the

mucosal barrier (Peterson and Artis, 2014). The intestinal epithelial cells (IECs) sense and respond to

microbial stimuli to strengthen barrier function and coordinate appropriate immune responses,

ranging from anti-pathogen immunity to tolerance. As a result, IECs play a critical immunoregulatory

role in the formation and homeostasis of mucosal immune cells. The intestinal epithelial cell layer

and the tight junctions that connect these cells to form a continuous single layer are essential

components of the mucosal barrier throughout the GI tract.

Figure 4: Cross-sectional image of small intestine depicting the major cell types The intestinal epithelium is made up of secretory and absorptive cell lineages; enterocytes perform the absorption function, while three secretory cell types, including hormone-releasing EECs, antimicrobial peptide- producing Paneth cells and mucus-producing goblet cells, are responsible for secretion (Kong et al., 2018)

2.2 Apical junction protein complex

Apical junction protein complexes in the lateral membranes of neighbouring cells maintain epithelial cells in a tightly packed confirmation. The apical junctional complex is formed by three types of junctions (Figure 5). intercellular junctions: desmosomes, tight junctions and adherens

desmocollin and desmoglein, combine keratin to Figure 5: Tight junctions, desmosomes and adherens junctions, make up apical junctional complexes. At the tight junction, claudin, junctional adhesion molecule (JAM), occludin, Zonular occluden and F-actin interact to merge the lateral and apical plasma membranes of two adjacent junctions are formed when E-cadherin, β-catenin α-Catenin and F-actin cells. Adherens interact. Desmoplakin, produce desmosomes (Neunlist et al., 2013).

2.2.1 Tight junctions

The primary driver of mucosal permeability is the apical membrane tight junction complex. Tight

junctions act as a rate-limiting factor for paracellular permeability by selectively limiting solute flux

through the intestinal epithelium. There are two major paths for transporting substances through

the epithelium in the intestine: the transcellular pathway, which allows material to flow through

cellular membranes, and the paracellular pathway, which allows substances to pass via tight junction

complexes (Shen et al., 2011). This project is focused on understanding the paracellular pathway in

the GI tract in a mouse model of autism.

The paracellular pathway is selectively permeable due to the presence of different types of proteins

at tight junction complexes, which separate molecules traveling through intercellular spaces based

on their size and charge. The three main types of tight junctional proteins are transmembrane

proteins, scaffolding proteins such as Zonular occudens and regulatory proteins (Odenwald and

Turner, 2017). Transmembrane proteins comprise the pore-forming elements of tight junctions. The

well-studied adhesion molecules occludin, claudins and tricelluin are integral transmembrane

proteins (Furuse et al., 1998, Furuse et al., 1993, Ikenouchi et al., 2005). Claudins are considered the

most critical transmembrane proteins contributing to mucosal permeability, as they form a pore

between neighbouring cells to govern tight junctional ion selectivity (Van Itallie and Anderson, 2006).

Scaffolding proteins (including occludins) play an important role in cell signalling pathways by

bringing together many binding partners to enhance their coordinated interactions and functions.

The zonula occludens (ZO) proteins, which have three different forms in mammals, are some of the

most well-studied scaffold proteins located on the cytoplasmic side of the epithelial tight junction

(ZO-1, ZO-2, ZO-3) (Stevenson et al., 1986). Through binding to claudins and other occludins, as well

as phosphoinositides located within the plasma membrane, ZO proteins are targeted to the plasma

membrane and assist in regulating pore size (Stevenson et al., 1986). Although the precise function

of occludin proteins in barrier modulation is unknown, both cell culture and animal model studies

show that occludin is critical for tight junction construction and barrier function (Findley and Koval,

2009). This is evident from recent research using siRNA to reduce translation of all tight junction

occludin mRNA in male C57BL/6 mice. This enabled intestinal permeability to be investigated which

demonstrated a selective increase in intestinal epithelial cell paracellular permeability in the absence

of occludins. These findings suggest that occludins play a role in maintaining tight junctions (Al-Sadi

et al., 2011).

2.2.2 Paracellular permeability

The paracellular route is selectively permeable due to tight junction proteins that differentiate

molecules by size and charge. The apical plasma membrane of the epithelial cell lining is impermeable

to most hydrophilic solutes, hence material is predominantly transported across the epithelial barrier

via intercellular junctions (Turner, 2009). Paracellular permeation allows passage of specific ions and

small uncharged solutes that range from 2.5 – 4.0 Å and larger macromolecules that range from 7.5

– 40 Å across the mucosal barrier (Shen et al., 2011). The creation of an ion gradient to support the

passive paracellular transport of solutes requires charge selectivity (Shen et al., 2011). Charge

selectivity, size of molecule and an ion gradient is dependent on the composition of epithelial apical,

lateral and basolateral proteins (Shen et al., 2011). Importantly, the composition of the epithelial,

apical, lateral and basolateral proteins determines the charge selectivity and size of molecule

permitted to cross the paracellular pathway. This dissertation will examine whether the paracellular

permeability route is altered in a mouse model of autism.

Figure 6: Intestinal epithelial cells utilise both transcellular and paracellular pathways. Highlights the two main pathways that metabolites can translocate in gut epithelial cells. Finger-like projections represent microvilli on the luminal side of the cellular membrane. The flat basolateral membrane is illustrated at the base of this diagram. Note that the transcellular pathway enables small molecules such as mannitol (182 Da) to pass across the epithelium whereas the paracellular pathway is permeable to larger molecules (e.g., lactulose which is approximately 340 Da in size) (Vojdani, 2013).

2.3 Gastrointestinal distress and ASD

Increased intestinal permeability has been observed in both Individuals with ASD and animal

models. To date, the majority of clinical evidence comes from sugar permeability tests, which

examine the urine collection of two sugars (lactulose and mannitol) with distinct molecular sizes and

absorption pathways (D'Eufemia et al., 1996, de Magistris et al., 2010, Souza et al., 2012). Mannitol

is excreted via the transcellular pathway, which involves the sugar passing through aqueous holes in

cellular membranes. Lactulose, on the other hand, travels through extrusion zones at the villus tips

of the small intestine before translocating through apical junction complexes. In ASD patients,

increased urine recovery of lactulose has been documented, implying that altered gut permeability

is due to abnormal levels of intestinal tight junction proteins and a dysfunctional paracellular

pathway (de Magistris et al., 2010). Even in the absence of GI problems, children with autism

reportedly have relatively high intestinal permeability (i.e., compared to age-matched controls)

(D'Eufemia et al., 1996). Sugar permeability experiments also reveal changes in intestinal

permeability in first-degree relatives of people with ASD, implying the presence of genetic factors in

these families that could affect tight junction protein levels (de Magistris et al., 2010). ASD patients

and animal models of ASD have altered expression levels of genes that encode tight junction

proteins, suggesting that this contributes to a disruption in mucosal barrier functionality at the

cellular level. For example, ASD patients have been found to have altered expression of tight junction

proteins in the blood-brain barrier and in the intestinal epithelial apical junction complexes

(Fiorentino et al., 2016). Claudin-12 and claudin-5 levels were higher in cerebellar and post-mortem

cortex tissue samples from those patients, while MMP9, Claudin-3 and tricellulin were higher in

cerebral cortex tissue. Tricellulin, claudin-1 and occludin expression levels were decreased, while

claudin-15, claudin-2 and claudin-10 expression levels were elevated in the gastrointestinal tract

(Fiorentino et al., 2016). A decrease in expression of genes encoding occludin, zonulin-2, claudin-8,

and zonulin-581, as well as increased claudin-15 expression, was observed in offspring of a maternal

inflammatory activation mouse model of autism, similar to findings in patient tissue samples (Hsiao

et al., 2013). Despite reports of hyperpermeability, no research has been done to determine which

region of the small or large intestine may demonstrate increased permeability in transgenic mouse

models of ASD. Understanding which region of the GI tract contributes to alterations in permeability

can shed light on biological mechanisms contributing to GI dysfunction (for example in the context

of ASD), given that each gut region has different anatomical and physiological characteristics.

2.4 Restoration of intestinal permeability

The mucosa of the digestive tract is lined with multifunctional, rapidly growing epithelial cells. They

serve as the principal gateway between luminal contents and interstitial tissue. These cells receive

nourishment from both luminal and systemic sources and are influenced by intra- and extra-luminal

nutrient intake as reviewed in (Herath et al., 2020). During a typical lifetime, 60 tonnes of food travel

through the gastrointestinal tract, providing a constant threat to the integrity of the gastrointestinal

tract and the host organism overall (Kårlund et al., 2021). Proteases, dietary components,

medications, bacteria, intestinal ischemia, bacterial toxin exposure, microbial degradation, cytotoxic

agents and the presence of pro-inflammatory cytokines such as IFN and TNF typically cause mild

damage to the tight junction proteins (Anderson and Van Itallie, 1995, Kårlund et al., 2021, Mitic and

Anderson, 1998). An inability to increase tight junction protein expression can be harmful as it can

result in multiple organ dysfunction, chronic low-grade inflammation and sepsis (Blikslager et al.,

2007, Gonzalez et al., 2015). Thus, to identify potential therapeutic targets, it is critical to understand

the mechanisms that regulate tight junction complexes and gene expression during intestinal

epithelium repair. Numerous innovative medications have recently been investigated relevant to

restoration of tight junction protein expression. For example, monoclonal antibodies with high

affinity and specificity for claudin receptors promote claudin protein production (Singh et al., 2017).

Commonly used, safe, and affordable supplements have also been assessed to ascertain if these can

reduce intestinal permeability. For instance, decaffeinated coffee has been shown to stimulate the

expression of tight junction proteins and aid in weight loss in a rat model of obesity (Caporaso et al.,

2016). Additionally, the herb known as marshmallow root (Althaea officinalis) that is indigenous to

Europe, Northen Africa and Western Asia has been utilised as a natural treatment for the healing of

gastrointestinal, skin and respiratory disorders. Another relevant compound is Gelatin tannate (GT),

comprising tannic acid and gelatin which has astringent, anti-inflammatory and antibacterial

properties and forms a protective barrier in the GI tract. Both marshmallow root and GT create a

protective GI tract barrier through increased mucus production (Ruszczyński et al., 2014, Zaghlool et

al., 2019). Specifically, it is unknown if mice expressing the R451C mutation encoding Neuroligin-3

have enhanced intestinal permeability or if this is restricted to specific gut regions.

2.4.1 The impact of L-glutamine on intestinal permeability

L-glutamine is the most prevalent amino acid in human blood, and skeletal muscle (Achamrah et al.,

2017, Decker, 2002, Deters and Saleem, 2021, Rao and Samak, 2012b). It is involved in numerous

physiologically significant metabolic processes; as an intermediary in energy metabolism and as a

substrate for the production of peptides and non-peptides such as nucleotide bases, glutathione, and

neurotransmitters (Albrecht et al., 2010, Amores-Sánchez and Medina, 1999). Under normal

physiological conditions, L-glutamine is a non-essential amino acid and the body produces sufficient

amounts. In the case of severe infections, physical trauma, specific disease states, radiation-induced

damage, and serious burns, however, physiological L-glutamine levels are insufficient and should be

supplemented with dietary L-glutamine. Once glutamine reserves are exhausted, the gastrointestinal

lining becomes more susceptible to injury (Rao and Samak, 2012a) Additionally, L-glutamine aids in

the elimination of ammonia and in maintaining a healthy acid-base balance in the body (Patience,

1990). The gut consumes around 30% of total L-glutamine (Wu, 1998), demonstrating that it is a

critical nutrition source for the intestine. Three-quarters of enterally administered L-glutamine is

absorbed into splanchnic tissues, and the majority of absorbed L-glutamine is processed in the

small intestine (Newsholme and Carrié, 1994). When plasma L-glutamine passes through the small

intestine, one-fourth of it is absorbed (Hankard et al., 1995). The intestine competes with other

tissues for L-glutamine obtained from the body's amino acid pool and dietary sources (Evans and

Shronts, 1992). L-glutamine protects tight junction proteins in three key ways including: i) preserving

intestinal tissue integrity, ii) producing anti-inflammatory mediators, and iii) preventing apoptosis

and cellular stress (Figure 7) (Kim and Kim, 2017). As previously described, tight junctions connect

adjacent epithelial cells to form a physical barrier between epithelial and endothelial cells (Bjerknes

and Cheng, 2005). Evidence suggests that a sub-population of individuals with ASD have increased

intestinal permeability, however drugs or supplements that may restore increased intestinal

permeability in this population are understudied and correlations of increased GI permeability in

different intestinal regions in animal models of ASD are not well established.

Figure 7: Potential mechanisms for L-glutamine restoring paracellular permeability in IEC’s. Glutamine preserves intestinal tissue integrity by facilitating enterocyte proliferation, activating mitogen-activated protein kinases (MAPKs) (JNK and ERK1/2), optimisation of growth factors insulin- like growth factor (IGF), (TGF)-α), transforming growth factor (TGF), (epidermal growth factor (EGF) and stimulating the expression of tight-junction (TJ) proteins (zonula occludins (ZO)-1, ZO-2, and ZO- 3, claudin-1, occludin and claudin-4. Glutamine inhibits pro-inflammatory signal transduction factors such nuclear factor-B (NF-B) and signal transducers and activators of transcription Glutamine inhibits widespread apoptosis by contributing to the production of glutathione (GSH) and by modulating the heat shock factor-1-(HSF-1) induced expression of heat shock proteins (HSPs). Glutamine reduces endoplasmic reticulum (ER) stress and induces autophagy by suppressing the mechanistic target of rapamycin (mTOR) pathway, thereby protecting enterocytes against stressful cellular conditions. The T bars denote inhibition, whereas the arrows reflect activation Kim and Kim, (2007).

2.4.2 The impact of caffeine on intestinal permeability

Caffeine is a common component in coffee, energy beverages, and food supplements. Caffeine

stimulates the CNS, increasing alertness and producing anxiety and restlessness in susceptible

individuals. It relaxes smooth muscle, promotes cardiac muscle contraction, and improves athletic

performance (Nehlig et al., 1992). Caffeine stimulates gastrointestinal motility and stomach acid

secretion (Liszt et al., 2017). In vitro research (e.g., in animal models or cell systems) has mostly

examined the complex pharmacology of caffeine's effects and identified that caffeine is an

adenosinergic antagonist via a non-selective pathway (Institute of Medicine Committee on Military

Nutrition, 2001). In addition, caffeine raises the amount of cyclic adenosine monophosphate (cAMP)

in tissue via decreasing the action of phosphodiesterases in numerous cell types (Institute of

Medicine Committee on Military Nutrition, 2001). Caffeine has been used to investigate the

contractile and/or electrical properties of the various gut wall components involved in motor

function along the gastrointestinal tract including the myenteric plexus (both neurons and glial cells),

smooth muscle cells and ICCs, as well as their dependence on intracellular calcium dynamics (Ito et

al., 1974). Various approaches, including organ baths and electrical recordings of single cultured

smooth muscle cells, have been used to examine the in vitro effects of caffeine on the gastrointestinal

tract smooth muscle in mice (Tokutomi et al., 2001). It has been demonstrated that caffeine

promotes the secretion of anions by enterocytes. This occurs via the RyR/Orai1/Ca2+ signalling

pathway. These findings show components of the RyR/Orai1/Ca2+ pathway regulated by caffeine

may provide novel potential therapeutic targets for the regulation of intestinal anion secretion (Wei

et al., 2018, Zhang et al., 2019). To date, however, no research has examined how caffeine affects

paracellular permeability in mouse models of autism.

2.4.3 The impact of fasting on intestinal permeability

Short-term fasting has comparatively few negative side effects and can be advantageous for persons

who are generally healthy and wish to control their weight (Alscher et al., 2001). During short-term

fasting by utilising the nutrients and electrolytes received by the intestinal epithelium, the

requirements of the body's metabolism are met (Ferraris and Carey, 2000). However, fasting deprives

the body of essential nutrients and can cause electrolyte imbalances, both of which affect

physiological homoeostasis. The digestive system is the first organ system to be impacted by changes

in nutrient intake, and undergoes the most rapid and severe adaptations in response to dietary

shortage (Ferraris and Carey, 2000). These modifications may increase or decrease intestinal

permeability. Before evaluating the effect of fasting on gastrointestinal permeability in humans, it is

necessary to ascertain the duration of abstinence from eating. Recent reports indicate that the

amount of protective mucus coating the lumen of the digestive tract is correlated to the duration of

time spent fasting (Alscher et al., 2001, Ferraris and Carey, 2000, Mohr et al., 2021). In contrast, it is

commonly accepted that permeability is increased when the mucus barrier is weakened (Figure 8),

such as following overfeeding or after consuming a high-fat diet which results in elevated

inflammatory effects and alterations in the mucus barrier (Mohr et al., 2021). If a person fasts for

less than two days, mucus production has been shown to increase. Consequently, under these

conditions, IECs will be better shielded from physical damage and potential pathogenic invasion and

GI permeability will be reduced (Mohr et al., 2021). When an individual fasts for more than two days,

endogenous bacteria in the stomach begin to break down mucus in order to meet the energy needs

of microorganisms, resulting in increased intestinal permeability (Mohr et al., 2021).

Figure 8: Overeating and duration of food deprivation influences intestinal mucus composition. The potential impact of fasting/feeding cycles on gut function include increased permeability as a result of high fat diets and overeating whereas short term fasting and potentially longer term fasting can lead to reduced GI permeability. SCFA: short chain fatty acids; HFD: high-fat diet; LPS: lipopolysaccharide; TJ: tight junction; LPS: lipopolysaccharide; BA: bile acid; TMAO: trimethylamine N-oxide; GI: gastrointestinal Mohr et al., (2021).

3.0 Classification of enteric neurons

The enteric nervous system is the largest division of the autonomic nervous system, with over 400

million neuron-supporting glial cells and over 100 million neurons that lie within the walls of the

gallbladder, biliary tree, small and large intestines, oesophagus, stomach, pancreas as well as nerve

fibres that link these nerve fibres ganglia and that supply the mucosal epithelium, gut wall muscle

and arterioles (Furness, 2006). The normal functioning of the digestive system is dependent on this

complex neuronal system. Enteric neurons are classified based on their morphology,

electrophysiology, function and neurochemistry (Furness, 2006).

3.1. Morphological classification of enteric neurons

Enteric neurons were initially characterised by morphological traits as three types of neurons: Dogiel

type I, II and III neurons (Brehmer et al., 1999, Dogiel, 1895). Dogiel type I neurons have small cell

bodies with short dendrites and a single axon and include ascending and descending interneurons,

as well as inhibitory and excitatory motor neurons (Furness, 2006). In mice, Dogiel type II neurons

make up 10-20% of myenteric neurons and are multiaxonal. Dogiel type II neurons could be absent

from the sub-mucosal plexus (Mongardi Fantaguzzi et al., 2009) which could impact intestinal

permeability due to the several innervations. However, other authors suggest Dogiel type II neurons

are present in the submucosal plexus (Furness et al., 2003). Dogiel type III neurons feature large

smooth cell bodies that are oval or spherical, with numerous axons that run circumferentially

(Bornstein et al., 1991, Foong et al., 2012, Furness, 2006, Nurgali et al., 2004). Use of other methods

such as dye injections and immunohistochemistry enabled the identification of additional neuronal

morphologies, thus expanding on Dogiel's classification to include type IV, V, VI, and VII neurons, as

well as mini neurons (Brehmer et al., 1999). Type IV neurons are uni-axonal and have

short, branching, tapering dendrites that extend vertically. Type V neurons are uni-axonal and have

long branched dendrites, forming clusters. Dogiel neurons of type VI have a single axon and fine

dendrites (Furness, 2006).

3.2 Electrophysiological classification of enteric neurons.

Enteric neurons are divided into two broad groups based on their electrical features: synaptic

neurons (S-neurons) and after-hyperpolarisation (AH) neurons. S-neurons display short action

potentials lacking slow after-hyperpolarizing potentials and rapid excitatory postsynaptic potentials.

AH neurons are characterised by large action potentials with an inflection on the falling phase,

followed by extended after-hyperpolarizing potentials (Hirst et al. (1974).

3.2.1 Synaptic-neurons (S-neurons)

S-neurons exhibit fast excitatory post-synaptic potentials (fEPSPs) (Hirst et al., 1974). Other

characteristics that distinguish S-neurons include: i) a larger proportion of membrane potential

depolarisation events than AH-neurons ii) hyperexcitability (due to lack of afterhyperpolarisation) iii)

when depolarising pulses are applied, lengthy spike trains of action potentials are elicited that are

amplitude dependent (Nurgali 2009).

3.2.2 AH (after-hyperpolarisation) neurons

Microelectrodes were initially used to study the electrical properties of sensory neurons in the ENS

which were named intrinsic primary afferent neurons (IPANs, or intrinsic sensory neurons; ISNs).

IPANs were dubbed AH-type neurons due to their action potentials exhibiting a significant AHP

component. Functional and expression features typical of AH neurons include: i) a greater

hyperpolarized membrane potential than S-neurons ii) expression of IK channels, and a iii) hump in

the action potential falling phase (Li, 2022))In AH neurons the rising phase of action potentials are

carried by Na+ and Ca2+ ions and a shoulder are evident in the falling phase (North 1973). Similar to

many neuron subtypes in the CNS, the falling phase is modulated by voltage- and time-dependent K+

channels. The presence of the AHP component, which stops continued excitation of the neuron for

up to 30 seconds, is critical to the action potential properties of AH neurons.

3.3 Functional classification of enteric neurons.

There are approximately 20 distinct types of enteric neurons, with the numbers varying slightly

between intestinal regions and animal species. Each type is defined by a combination of

characteristics (projections to targets, functional roles, morphology, neurochemical properties, cell

physiology and electrophysiological properties). Three functional classifications can be distinguished

among the twenty types; i) IPANs (or ISNs), ii) motor neurons and iii) interneurons. IPANs can detect

both the physical state of organs (for instance, tension in the intestinal walls) and the chemical

composition of luminal contents. IPANs respond to these signals by initiating reflex regulation of

blood flow, motility and secretion. IPANs communicate with one another and directly with motor

neurons and interneurons. Interneurons interact with motor neurons and other interneurons.

Muscle motor neurons, motor neurons to enteroendocrine cells (EECs) and neurons innervating

lymphoid follicles, secretomotor neurons, secretomotor/vasodilator neurons are all examples of

motor neurons (Furness et al., 2014). This research project aimed to assess the functional

characteristics of myenteric neurons of the mouse duodenum using whole cell patch clamp recording

as an effort to contribute to future more detailed classification of neuronal subtypes using multiple

methodological approaches.

Figure9: Three major functional classes of enteric neurons. (Intrinsic primary afferent neurons; also known as Intrinsic Sensory Neurons (ISNs), interneurons, and motor neurons. To influence GI function, IPANs synapse with other interneurons, motor neurons, and interneurons. Interneurons and motor neurons also communicate with one another via synapses (Furness, 2012).

3.3.1 Intrinsic primary afferent neurons (IPANs)

Intrinsic primary afferent neurons (IPANs) or intrinsic sensory neurons (ISNs) sense the state of the

GI tract. By encoding information about the chemical environment of the gut lumen and the state of

the tissue they innervate, these neurons communicate with the enteric neuronal circuitry to

modulate gut function. IPANs detect a number of stimuli, including luminal chemicals, intestinal wall

force/strain, mucosa deformation, and activate intrinsic reflex pathways to regulate physiological

activities such as gut motility, secretion, and blood flow (Furness et al., 2014). IPANs have a large

smooth cell body and numerous axons and correspond to a Dogiel type II morphology.

Electrophysiological features of these neurons include extended after hyperpolarizing potentials

(AHPs) (Kunze et al., 1995, Bertrand et al., 1997). IPANs form synaptic connections with various types

of myenteric neurons, including interneurons and motor neurons, as well as with other IPANs

(Furness et al., 2004). Within the submucosal plexus, Dogiel type II neurons communicate with

neurons in the myenteric plexus and other submucosal plexus neurons. According to a study using

combined electrophysiological and neurochemical techniques, IPANs are not detected in the mouse

submucosal plexus (Mongardi Fantaguzzi et al., 2009). The axons of myenteric IPANs project

circumferentially throughout the small intestine of guinea pigs, as well as the colon and duodenum

of mice, and send synaptic outputs to surrounding myenteric ganglia (Bornstein et al., 1991, Foong

et al., 2012, Nurgali et al., 2004). Each villus in the guinea-pig ileum receives axons from around 65

different myenteric IPANs, and each neuron innervates between 20 and 60 villi (Bertrand et al.,

1998).

3.3.2 Interneurons

Four subtypes of interneurons have been identified through immunohistochemical studies in the

small intestine of guinea pigs (Costa et al., 1996a). Three subtypes of descending interneurons have

a lengthy descending projection that connects to myenteric and submucosal neurons. In most cases,

these three subtypes of descending interneurons establish long neuronal chains by making synaptic

synapses with neurons of the same class (Furness et al., 2014). The fourth subtype are ascending

interneurons which were identified in the ileum of guinea pigs (Furness et al., 2014). This type of

ascending interneuron uses acetylcholine and tachykinins as the main neurotransmitters.

Structurally, these ascending interneurons have a single orally directed axon and form cellular chains.

Ascending interneurons tend to be involved in local motility reflexes and create synaptic connections

with excitatory muscle motor neurons and other ascending interneurons (Furness et al., 2014,

Portbury et al., 1995, Stebbing and Bornstein, 1996).

3.3.3 Muscle motor neurons

Motor neurons innervate the circular muscles, longitudinal muscles, and muscularis mucosa

throughout the GI tract. Depending on the effects exerted on the muscles they innervate, these

neurons are classed as excitatory or inhibitory motor neurons (Furness, 2006). Throughout the GI

tract, motor neurons innervate the circular muscle (CM) and longitudinal muscle (LM). Motor

neurons are classed as either excitatory or inhibitory depending on the effect exerted on the muscles

they innervate (Furness, 2006). In excitatory motor neurons, the primary neurotransmitters are

acetylcholine and tachykinins. The principal neurotransmitter of inhibitory motor neurons is nitric

oxide (NO), however these neurons also express ATP and VIP (Furness, 2006).

Motor neurons that innervate muscles are uni-axonal and show S-type electrophysiological traits

(Furness, 2006). The myenteric plexus contains most of the cell bodies of CM motor neurons. In the

guinea pig, all CM motor neurons are located in the myenteric plexus (Wilson et al., 1987).

Conversely, a small percentage of CM motor neurons are found in the SMP in other species such as

the rat, dog, pig, and human (Ekblad et al., 1987, Furness et al., 1990). NOS (nitric oxide-synthase) or

VIP immunoreactivity is found in inhibitory circular muscle motor neurons, while tachykinin or the

vesicular acetylcholine transporter (VAChT) is present in excitatory circular muscle motor neurons

(Furness, 2006). Inhibitory and excitatory neurotransmitters have comparable expression patterns in

the small intestines of mice and guinea pigs (Qu et al., 2008).

3.4 Neurochemical classification of myenteric neurons

Enteric neurons are also classified according to the expression of neurochemicals. For instance, an

IPAN located in the guinea pig duodenum may express a subset of neurochemicals unique to that

enteric neuron subtype, although this approach holds significant complexity. For example,

neurochemical coding varies between animal species, between different intestinal regions and for

neurons located in different plexuses embedded in the GI wall. Due to this complexity, a

comprehensive approach to characterising neurochemical coding, for example in the mouse, is

required. One aim of the current project is to contribute to the classification of myenteric neurons in

the duodenum in C57BL/6 mice by collating cellular functional profiles using patch clamp

electrophysiological recordings. Table 1 outlines how myenteric neurons have historically been

classified according to gene expression, functional and morphological characteristics (Costa et al.,

1996b). Recent research has utilised single-cell sequencing to uncover 21 enteric neuron subtypes

and 3 enteric glial subtypes that span across 5 broad groupings (Drokhlyansky et al., 2020). These 5

groups of enteric neurons are: excitatory motor neurons, inhibitory motor neurons, sensory neurons

secretomotor/vasodilator neurons and interneurons. This work serves as the foundation for an

enteric neuron “library” that can be utilised to correlate individual enteric neuron subtypes with

electrical recordings and morphological data. Since this project focused on the small intestine, the

action potentials recorded from duodenal enteric neurons may not correlate precisely to

Drokhlyansky's research, as this research was conducted in the colon.

Table 1: Myenteric neurons are classified according to their neurochemical coding Dogiel type classification, and functional class (Costa et al, 1996).

% Total neurons

Chemical coding

Morphology

Functional class

VIPINOS/dynorphin/GRP/NFP/ ± AP

Type I

≤5 12

VIP/NOS/ dynorphin/enkephalin/±NFP/±AP SP/ChAT/enkephalin/NFP

Type I

7

Long descending inhibitory motor neurons to circular muscle Short descending inhibitory motor neurons to circular muscle Long ascending excitatory motor neurons to circular muscle

SP/ChAT

Type I

5

Short ascending excitatory motor neurons to circular muscle

ChAT/SP/calretinin

Type I

4 17-24

ChAT/ ± SP/calretinin

Type I

Excitatory motor neurons longitudinal muscle Excitatory motor neurons longitudinal muscle Ascending interneurons

Type I

ChAT/SP

3

SP/ChAT/enkephalin/calretinin/NFP

Type I

Ascending interneurons

5

5-HT/ChAT/NFP

Type I

Descending interneurons

2

Somatostatin/ChAT

Filament

Descending interneurons

4

VIP/ChAT/±NFP

Type I

Descending interneurons

3

VIP/NOS/dynorphin/GRP/NFP/ ± AP

Type I

Descending interneurons

Type II

Sensory neurons

≤5 30

± Calbindin/ ± SP/ ± ChAT/novel/ intermediate filament

VIP/dynorphin

Type III

Secreto- and/or vasomotor neurons

1

NPY/ChAT/somatostatin/CGRP/CCK

Type III

Secreto- and/or vasomotor neurons

1

(to

prevertebral

≤1

Combinations of ± VIP/ ± enkephalin/± dynorphin/± GRPI ± CCK

Intestinofugal ganglia)

4.0 ASD genetic mutations

Genetic factors are a significant contributor to ASD diagnosis. Twin and family studies indicate that

ASD is one of the most frequently occurring hereditary neuropsychiatric disorders. The chance of

ASD recurrence among ASD children's siblings is approximately 45 times higher in comparison to the

general population (Constantino and Todd, 2003, Ronald et al., 2006). Classifying and modelling

genetic predisposition has proven difficult due to contributions from a widely varied

heterogeneous genetic pool and the presence of substantial gene-environment and gene-

gene interactions in individuals diagnosed with ASD (Chaste and Leboyer, 2012, de la Torre-Ubieta et

al., 2016). ASD has been linked to several gene interactions or rare mutations, complicating such

research even further. There are at least 1000 genes associated with ASD; and within this group at

least 250 genes are specifically implicated in ASD. Therefore, over 700 genes are implicated in

individuals with ASD that have at least one other comorbid condition and the majority of these genes

are implicated in synaptic cell adhesion molecule pathways (David et al., 2016). Given these findings,

the Neuroligin-3R451C mouse model of autism which involves a missense mutation in the gene

encoding the neuroligin-3 cell adhesion molecule located at neuronal synapses is an ideal model to

study changes occurring in the nervous system in ASD.

4.1 ASD and synaptic cell adhesion molecules

Many ASD-related gene mutations encode proteins involved in neural transmission, such as synaptic

adhesion molecules which form homophilic or heterophilic connections between pre and

postsynaptic membranes. These proteins have essential roles in the plasticity, development

and function of synapses (Bourgeron, 2015, Dalva et al., 2007). Synaptic adhesion molecules

primarily bind with cytoplasmic elements such as cell signalling molecules, scaffolding proteins

and cytoskeleton proteins facilitating downstream signalling cascades (Iida et al., 2004). The

neuroligin-neurexin complex is a well-studied synaptic cell adhesion molecular pathway

complex linked to ASD (Südhof, 2008). This pathway is the focus of the current study, which aims to

assess for changes in the enteric nervous system in the Neuroligin-3R451C mouse model of autism.

4.2 Neuroligins

Neuroligins (NLGNs) are postsynaptic membrane proteins that bind to presynaptic neurexins

(NRXNs). A broad extracellular domain containing a single transmembrane domain, an esterase-like

sequence and a short cytoplasmic tail make up the neuroligin protein. Neuroligins form constitutive

oligomers via interactions at the extracellular domain. The cytoplasmic tail of NLGNs has a type I PDZ-

domain binding domain that binds synaptic proteins to facilitate signalling processes and intracellular

protein-protein interactions (Kim and Sheng, 2004, Bolliger et al., 2001, Ichtchenko et al., 1995).

Mammals express four neuroligin isoforms (Neuroligin 1-4). Genetic sequencing reveals neuroligin

1,3 and 4 are similar, however, the evolution of neuroligin-2 is divergent (Bolliger et al., 2001,

Ichtchenko et al., 1996). Despite significant sequence homology, the cellular localization and function

of each neuroligin isoform is different. NLGN3, the focus of the current project, is involved at both

excitatory and inhibitory synapses in the brain (Budreck and Scheiffele, 2007). Even though

neuroligins can form homodimers, they act in a dimeric state at the postsynaptic membrane (Araç et

al., 2007, Comoletti et al., 2003, Fabrichny et al., 2007, Poulopoulos et al., 2012). This dimerisation

occurs early during secretion of neuroligins and is required for neuroligins to reach the postsynaptic

membrane (Poulopoulos et al., 2012).

4.3 Neuroligin-3 overview

Nlgn3 is located on at Xq13.1 on chromosome X and was first discovered in neurons within the central

nervous system (Chih et al., 2005, Gilbert et al., 2001, Varoqueaux et al., 2006). The NLGN3 protein

binds with NLGN1 (neuroligin-1) in excitatory synapses (Budreck and Scheiffele, 2007, Poulopoulos

et al., 2012) suggesting that synapse function in the mouse brain requires selectivity of contacts

between neuroligin isoforms. Furthermore, electrophysiological investigations in mice demonstrate

that neuroligin knockdown impacts synaptic transmission in the cerebellum, striatum and the

hippocampus (Baudouin et al., 2012, Etherton et al., 2011, Földy et al., 2013, Rothwell et al., 2014).

These findings suggest that NLGN3 is important in synapse structure and function in the brain,

however similar studies in the enteric nervous system of the GI tract are lacking.

4.4 Neuroligin functionality

NLGNs are located at the postsynaptic membrane and participate in the development and

consolidation of excitatory and inhibitory synaptic connections, depending on their distinct subtypes.

During development, it has been suggested that once a growing axon reaches its target, NLGNs are

hypothesised to induce synaptogenesis by forming a trans-synaptic connection with presynaptic

neurexins (NRXNs) (i.e., the generation of functional synapses) (Südhof, 2008). NRXNs and NLGNs are

therefore thought to play a crucial role in defining the specialisation and differentiation of synaptic

connections (Scheiffele et al., 2000, Südhof, 2008). Neuroligin overexpression increases neuronal

synapse density, (Boucard et al., 2005, Chih et al., 2005, Chubykin et al., 2007) whereas knocking

down neuroligins (i.e., using microRNA treatments) drastically decreases spines and synapses in the

hippocampus (de Wit et al., 2009, Shipman et al., 2011). These findings imply that neuroligins are

involved in the plasticity of synapses.

Neuroligins are required for synaptic function but not synapse development, according to evidence

from neuroligin knockout mice. The triple deletion of Neuroligin-1, NL3, and neuroligin-2 in mice

results in neonatal death due to substantially defective synaptic transmission (Varoqueaux et al.,

2006). Nonetheless, a triple neuroligin knockdown had little effect on overall synaptic ultrastructure

or synaptic numbers synapse, indicating that NLGNs are likely primarily involved in synaptic

organisation rather than synapse production (Varoqueaux et al., 2006). According to these studies,

NLGNs are essential for synaptic function and maturation but not for the developmental generation

of synapses.

4.5 NL3R451C mutation

Abnormalities in the NLGN3 protein resulting from point mutations and deletions in Nlgn3 are

associated with ASD. The R451C mutation in the Nlgn3 gene (the focus of the current study) is the

most thoroughly researched mutation relating to autism. The R451C mutation in Nlgn3 refers to a

cysteine residue being substituted for an arginine at position 451 in the NLGN3 protein. The mutation

was initially discovered in two brothers, one with a diagnosis of severe autism and the other with

Asperger's syndrome (Jamain et al., 2003). This mutation is located within the esterase domain,

which is known to provide Ca2+ dependent activities and structural integrity (Südhof, 2008). Because

this interaction occurs in the presence of Ca2+ ions, it is hypothesized that the R451C mutation alters

the binding of NLGN3 to neurexins at the presynaptic membrane (Jamain et al., 2003). In the brain,

the R451C mutation has been well characterized functionally and biochemically. As a result, it has

been reported that the R451C mutation causes trafficking and folding abnormalities in the NLGN3

protein. Nlgn3 mRNA expression levels are thought to be unaffected by the R451C mutation in the

brain, however levels of the mutated NLGN3 protein exported from the endoplasmic reticulum are

drastically reduced (to only 10% of control levels) and integrated into the synapse (Comoletti et al.,

2004, Tabuchi et al., 2007). Nevertheless, whole cell patch clamp recordings in brain slices from

Neuroligin-3R451C mice revealed increased inhibitory synaptic transmission (Tabuchi et al., 2007).

Therefore, the R451C substitution mutation is considered to be a 'gain-of-function' mutation. Mice

expressing the R451C mutation in Nlgn3 show phenotypes relevant to ASD such as altered social

interaction, increased aggressive behaviour and repetitive behaviour (Birchenough et al., 2015,

Burrows et al., 2015, Etherton et al., 2011, Tabuchi et al., 2007). Electrophysiological studies in mice

expressing the NL3R451C mutation also revealed altered synaptic activity in other brain regions

including the dorsal striatum, cortex, basolateral amygdala, and hippocampus (Földy et al., 2013,

Gogolla et al., 2009, Hosie et al., 2019, Martella et al., 2018, Rothwell et al., 2014, Tabuchi et al.,

2007). Although a broad range of studies have established electrophysiological changes in neural

networks of the brain, nothing is known about how this mutation impacts the enteric nervous system

at a cellular functional level.

4.6 Expression of Nlgn3 in the gastrointestinal tract

Although NLGN3 protein expression has been widely studied in the brain, limited studies have

observed its presence in the gastrointestinal system (Leembruggen et al., 2020). In human tissue,

Nlgn3 is expressed in myenteric interstitial cells of Cajal (Kulkarni et al.), which have been reported

as myogenic pacemaker cells, and abnormal expression levels of Nlgn3 are present in Hirschsprung's

disease (Wang et al., 2013, Zhang et al., 2013). The presence of Nlgn3 mRNA was also identified

within the cytoplasm of peptide YY (PYY)-containing entero-endocrine cells in mice (Bohórquez et al.,

2015). RT-PCR testing with primers designed specifically for the NL3R451C construct demonstrates that

in the GI tract, this mutation is expressed in mice that carry the NL3R451C mutation (Hosie et al., 2019).

Although the expression of Nlgn3 is relatively well characterised in brain tissue, it is unknown if

NLGN3 is found in specific neuron subtypes within the GI tract. Furthermore, the precise impact of

the R451C mutation on Nlgn3 mRNA cellular expression in mice is only beginning to be investigated

(Herath et al., 2022).

4.7 Effects Neuroligin-3 R451C mutation on the function of the GI system

Individuals with the R451C mutation have diarrhea, oesophageal regurgitation, and persistent gut

pain, suggesting that the R451C mutation may have a role in GI dysfunction (Hosie et al., 2019). In

Neuroligin-3R451C mice, the R451C mutation increased the density of myenteric neurons in the

jejunum. In the proximal jejunum of these mutant mice, there were 30% more myenteric neurons

per ganglion than wild-type mice (Hosie et al., 2019). In addition, the number of NOS immunoreactive

neurons per ganglion is higher in NL3R451C mouse jejunal tissues (Hosie et al., 2019). In the

NL3R451C mouse caecum, an increase in neuronal populations in the submucosal and myenteric

enteric ganglia was also observed (Sharna et al., 2020) suggesting a widespread change in ENS

structure as a result of this mutation. Specifically, in the caecal submucosal and myenteric plexus,

these mice similarly had an increased proportion of NOS-expressing neurons (Sharna et al., 2020). In

Neuroligin-3R451C mutant mice, however, immunocytochemical investigations revealed no change in

neuronal populations in the colon. Furthermore, when mutant mice were compared to littermate

controls, there was no change in baseline colonic motility (Hosie et al., 2019). Still, subtle changes in

colon function were present when antagonists for the GABA (gamma-aminobutyric acid) A receptor

(i.e., gabazine and bicuculline) were applied to motility assays. In these conditions, tissue from

NL3R451C mice showed decreased colonic motility, indicating that the R451C mutation enhances

sensitivity to GABA through the GABAA receptor (Hosie et al., 2019). These experiments were

undertaken in the GI tract because multiple investigations identified that the R415C mutation alters

GABAergic neurotransmission in many brain regions (Etherton et al., 2011, Rothwell et al., 2014,

Tabuchi et al., 2007) and the possibility of these changes being mirrored in the gut was explored.

Although not a major inhibitory neurotransmitter, GABA is found in a subset of motor and

interneurons in the colon of mice (Li et al., 2011, Sang and Young, 1996), and it does play an important

role in excitatory synaptic transmission in the ENS (Krantis, 2000). These findings suggest that in the

mouse colon, the R451C mutation affects enteric neuronal circuitry including GABAergic transmission

in the mouse.

5.0 Project rationale

GI dysfunction is common in people diagnosed with ASD but the cause is unknown. Many gene

mutations associated with autism affect synaptic function in the brain and recent research has shown

that these same mutations can alter enteric nervous system function and gut motility. The well-

established NL3R451C mouse model of autism provides an ideal tool to assess how changes in the

nervous system affect gut function.

5.1 Assessing permeability in an autism mouse model.

In both ASD patients and animal models of ASD, increased intestinal permeability has been

documented. To date, the majority of clinical evidence comes from sugar permeability tests, which

assess the urine collection of two sugars (lactulose and mannitol) with significantly different

molecular sizes and absorption pathways (D'Eufemia et al., 1996, de Magistris et al., 2010, Souza et

al., 2012). Increased lactulose recovery in the urine has been observed in ASD patients, implying that

altered gut permeability is caused by excessive levels of intestinal tight junction proteins (de Magistris

et al., 2010). Even in the absence of gastrointestinal difficulties, children with autism exhibit a

significantly high intestinal permeability (in comparison to age-matched controls) D'Eufemia et al.,

1996). Sugar permeability studies reveal variations in intestinal permeability in first-degree relatives

of individuals with ASD, revealing the presence of genetic factors affecting tight junction protein

levels in these families (de Magistris et al., 2010). ASD patients and animal models of ASD have

altered expression levels of genes encoding tight junction proteins, indicating a disruption in the

mucosal barrier’s biological integrity. Tight junction proteins are reportedly overexpressed in the

blood-brain barrier and the intestinal epithelial apical junction complexes of ASD patients (Fiorentino

et al., 2016). However, it remains unknown what proportion of ASD patients have a defective

intestinal barrier or whether specific intestinal regions are differentially susceptible to altered

permeability. My research will examine whether the autism-associated Neuroligin-3R451C mutation

affects intestinal permeability in different gut regions in mice.

5.2 Action potential characteristics in duodenal myenteric neurons using an autism mouse model

Since GI motility changes have been reported in isolated preparations of the GI tract from these mice,

future research is needed to investigate which myenteric neuronal subtypes are involved in these

changes. To date, studies involving the NL3R451C model have shown significant increases in the

number of NOS-containing neurons compared to wild-type mice (Hosie et al., 2019; Sharna et al.,

2020). Based on these findings, the current project aims to first characterise functional parameters

in myenteric neurons in mice at different ages and in different strains. This work will contribute to

generating a comprehensive database of electrophysiological, morphological and gene expression

data in the mouse small intestine which will be useful to the research field as a reference but will also

enable future work in the field of cellular dysfunction in gastrointestinal disorders in autism. It is

anticipated that these data will be used to identify cellular targets for future therapeutic design to

reduce the prevalence of GI disturbances in individuals with ASD. Currently there is a gap in

understanding of the pathophysiological mechanisms underlying GI dysfunction in autism. In this

project, I aim to characterise cellular functional properties in duodenal myenteric neurons using

whole-cell patch clamping which is anticipated to contribute to new modes of classification for

enteric neurons in the myenteric plexus.

5.3 Aims and hypotheses

Hypothesis: NL3R451C mice show increased paracellular permeability compared to wild-type

littermates and neurons involved in regulating motility in the small intestine have different functional

characteristics.

Aim 1: To compare paracellular permeability in fasted and non-fasted NL3R451C and wild-type mice.

Aim 2: To assess for effects of L-glutamine and caffeine supplementation on permeability in NL3R451C

and wild-type mice.

Aim 3: To optimize electrophysiological recording protocols and characterise duodenal neuronal

properties in wild-type mice.

intestinal permeability

in the

Chapter 2: Measuring Neuroligin-3R451C mouse model of autism.

1.0 Introduction

GI dysfunction is highly prevalent in individuals with ASD (Buie et al., 2010). GI problems can

exacerbate the severity of ASD-related behaviour, especially in people who are unable to

communicate their distress. Although the specific aetiology for GI difficulties is not known, clinical

evidence suggests that mucosal barrier impairment may play a role (D'Eufemia et al., 1996, Horvath

and Perman, 2002). The mucosa of the intestine separates the potentially harmful luminal

environment from the body’s interior and when luminal metabolites diffuse into the circulatory

system they can translocate to the CNS, potentially worsening ASD-related behavioural symptoms

(Fowlie et al., 2018). The intestinal epithelium is a key element of the intestinal mucosa barrier and

is designed to absorb nutrients while also protecting the host from pathogens. Tight junction protein

complexes tightly seal epithelial cells to one other, limiting the passage of materials via the

paracellular route. Both individuals with ASD and animal models of autism have been found to have

impaired paracellular pathways (D'Eufemia et al., 1996, de Magistris et al., 2010, Hsiao et al., 2013,

Wei et al., 2017). Crucially, these alterations occur alongside changes in tight junction protein

expression (de Magistris et al., 2010, Fiorentino et al., 2016, Hsiao et al., 2013, Wei et al., 2017).

Although the effects of the NLGN3 R451C mutation and NLGN3 ablation have been investigated on

myenteric plexus and GI motility using ex-vivo assays (Hosie et al., 2019, Leembruggen et al., 2020),

the impact of the R451C mutation on paracellular permeability has not been fully examined. This

chapter describes how the NL3R451C mutation impacts paracellular permeability in the small and large

intestines and the effects of L-glutamine and caffeine on permeability in mice.

2.0 Methods and materials

2.1 Animals

The original source of NL3R451C mice was Jackson Laboratories (Bar Harbour, Maine USA). For more

than 11 generations, experimental mice were grown on a C57BL/6 background at the University of

Melbourne's animal facility (Parkville, Victoria, Australia). We utilised male WT and NL3R451C mice.

These were produced by mating heterogeneous females with WT male mice producing in 50:50

NL3R451C and WT offspring (Tabuchi et al., 2007). Experimental mice were co-housed and sacrificed

by cervical dislocation in compliance with RMIT University's animal rules and guidelines (AEC# 22647

formerly classified as #1727). Mice either received free access to food (non-fasted) or were deprived

access to food (fasted) for approximately 18 hours prior to experimentation. The body weight of

experimental mice was recorded immediately after cervical dislocation. Following culling via cervical

dislocation, mice were placed on a dissecting bench and their extremities pinned out exposing the

abdominal cavity. Dissecting scissors (FST scientific, Canada) were used to cut open the abdominal

cavity, exposing the internal organs. The small and large intestines were identified and separated by

cutting away connective and mesentery tissue. The small intestine was separated from the large

intestine at the proximal end of the caecum.

2.2 Segmentation of the small and large intestine

The ex-vivo gut region assay for permeability used in the current project was based on the method

described in (Mateer et al., 2016). The mouse intestinal tract was removed using micro dissection

spring scissors (FST Scientific, Canada) and laid out on a plastic tray to be measured using a flexible

measuring tape. Excess mesentery was removed using microdissection spring scissors to obtain ex-

vivo sac gastrointestinal tissue preparations; i.e., approximately 7 cm lengths of fresh duodenum and

jejunum, 4 cm lengths of distal ileum, and 3 cm length sections of colon. Sections of mouse

duodenum were obtained immediately distal to the stomach. Jejunum sections were measured and

obtained from the distal end of the duodenum. Distal ileum tissue was obtained from the distal end

of the jejunum and at the oral end of the caecum, whereas the entire colon length was used for

experimentation. For each small intestinal segment, the luminal contents were immediately flushed

with 2 ml of 1 x PBS and a suture was placed as close to the oral end of the segments as possible. A

second suture was tied 4-5 cm in the anal direction depending on the intestinal segment. For the

colon, the luminal contents were immediately flushed from the anal end with 2 ml of 1 x PBS and a

suture was placed as close to the anal end of the segment as possible. A second suture was tied 3 cm

in the oral direction.

2.3 Preparation of L-glutamine and caffeine stock solutions

L-Glutamine stock solutions of 30 mM and 90 mM (molecular weight 146.14g/mol; Sigma-Aldrich,

Australia, CAS number: 56-85-9) and 10 mM caffeine (molecular weight 194.19 g/mol; Sigma-Aldrich,

Australia, Cas number: 58-08-02) were prepared in DMEM culture media (Dulbecco's Modified Eagle

Medium; Thermo Fisher Scientific, Catalogue number: 21063045). These concentrations were

selected based on previous gut permeability studies that also measured the movement of FITC across

the GI tract in an Ussing chamber (Abely et al., 2000, Yang et al., 1999, Zhang et al., 2019). Since

DMEM already contained 4mM L-glutamine, 38.0 mg of L-glutamine was measured and added to 10

mL of DMEM, resulting in a final concentration of 30 mM L-glutamine (26 mM of L-glutamine powder

and 4 mM of L-glutamine dissolved in DMEM culture medium). Additionally, 125.7 mg of L-glutamine

was measured and added to 10 mL of DMEM resulting in a final concentration of 90 mM L-glutamine

(86 mM of L-glutamine powder and 4 mM of L-glutamine dissolved in DMEM culture medium).

Finally, 19.4 mg of caffeine was added to 10ml of DMEM resulting in a final concentration of 10mM.

To 1 ml of DMEM supplemented with 30 mM, 90 mM L-glutamine or 10mM caffeine, a fluorescent

marker, FITC-Dextran 4 (Fluorescein isothiocyanate-dextran molecular weight 4000 Da; Sigma-

Aldrich, Australia) was added.

2.4 Injection of FITC-Dextran 4.

FITC-Dextran (4kDa) is a fluorescent marker used to study paracellular permeability in the

gastrointestinal tract (Volynets et al., 2016) . A stock solution of 100 mg FITC with 1 mL of DMEM

culture medium was used. A working solution of 1 mg FITC with 1 mL of DMEM was prepared daily

from the stock solution (DMEM was either enriched with 26 mM of L-glutamine, 86 mM of L-

glutamine, 10mM caffeine or lacked supplemental enrichment). A 1 ml syringe with an attached

needle (23 g blunt SN-23 Cat:64-1490 Warner Instruments, United States) was gently inserted into

the lumen of the ligated intestinal sample and held in place with a suture. The average length of the

entire small and large intestine was between 17-22cm for the mice used in these experiments. The

following consistent gut preparation lengths for each region were used in all experiments to control

for potential variations in results due to differences in preparation volume, duodenum: 5 cm,

jejunum: 5cm, distal ileum: 4cm and colon: 3cm. For every 1 cm of GI tissue, 40 µl FITC (1 mg/ml)

was injected into the lumen of the fresh GI tissue sample and the needle gently removed. Using

forceps, the intestinal opening was clamped shut and the suture was then fully tied off. The serosal

side of the ex-vivo sac tissue preparation was washed with 1x PBS to ensure the removal of any

residual FITC.

2.5 Time course permeability experiment

Following dissection of the tissue, the ex-vivo sac tissue preparation was immediately placed in a 50

ml falcon tube containing 20 ml of 1xDMEM with a gas line containing 95% 02 and 5% CO2 feeding

directly into the falcon tube. 100 µl triplicate samples were obtained from solution surrounding the

ex-vivo sac every 30 minutes for 2 hours. To ensure FITC had not accumulated at the bottom of the

falcon tube, the falcon tube was inverted slowly and held for 4 seconds before samples were

collected.

Figure10: Setup for measuring the permeability of FITC through the paracellular route. The ex-vivo intestine segment containing injected FITC is inserted into an unsealed 50ml falcon tube containing 95% O2 and 5% CO2. Red arrow indicates the diffusion of FITC molecules via the paracellular route.

2.6 Construction of standard curve using log serial dilutions

To assess whether a significant difference in FITC concentration (mg/ml) existed between NL3R451C

and WT mice at 0, 30, 60, 90, or 120-minute time points during the permeability assay, a repeated

measures two-way ANOVA with a Sidak’s (Tukey’s for more than two groups) multiple comparisons

test was used. To determine if a significant difference existed, a p-value <0.05 was used. Data was

expressed as mean ± SEM. Additionally, a simple linear regression test was used to determine the

gradient value of the concentration-time curves which was used to compare the concentration-time

curves in different gut regions.

Figure 11: Standard curve of known concentrations of FITC with absorbance Table 2: Known concentrations of FITC and absorbance produced Concentration (mg/ml)

0.0065

0.001

0.003

0

0.017 Relative fluorescence units (RFU) 0

63.8

206.4

487.9 1458.1

2.7 Statistical analysis

A repeated measures two-way ANOVA that used a Sidak’s (Tukey’s for more than two groups)

multiple comparisons test was used to assess whether a significant difference in FITC concentration

(mg/ml) existed between NL3R451C and WT mice at 0, 30, 60, 90, or 120-minute time points during

the permeability assay. To determine if a significant difference existed, a p-value <0.05 was used.

Data was expressed as mean ± SEM. Additionally, a simple linear regression test was used to

determine the gradient value of the concentration-time curves which was used to compare the

concentration-time curves in different gut regions.

3.0 Results

This chapter investigated whether the autism-associated R451C mutation in the Nlgn3 gene impacts

paracellular permeability. This was explored by comparing concentrations of FITC in the small and

large intestine of Neuroligin-3R451C and wild-type mice using an ex-vivo permeability assay approach

(Mateer et al., 2016).

3.1 Small intestinal permeability in non-fasted wild-type and mutant mice

Paracellular permeability was first measured in mice fed a standard ad libitum diet by injecting FITC

into duodenum, jejunum, distal ileum and colon ex-vivo sac preparations and subsequently

determining the quantity of FITC that passed through intestinal layers via the paracellular pathway.

Because it is known that the permeability of the gut fluctuates according to the fed state the change

in permeability in standard fed mice was first assessed prior to investigating for changes in fasted

mice. Initially, only small intestinal preparations were compared due to limitations of the

experimental equipment in terms of the number of tissue preparations that could be run

simultaneously. For some later studies described in this thesis, colon preparations were also

assessed.

For small intestinal preparations, the concentration of FITC increased linearly with time in all

gastrointestinal regions and did not plateau (Figure 12). In the duodenum, jejunum and distal ileum

of non-fasting mice, no change in paracellular permeability was observed in NL3R451C mice compared

to wild-type mice. Interestingly, NL3R451C and wild-type mice small intestinal gradient values were

higher when compared to the colon (Appendix Table 5). These findings show that when mice were

fed a normal chow diet and were not fasting, different GI regions have different permeability rates,

however no significant differences in small intestinal paracellular permeability were found between

NL3R451C mice and wild-type littermates.

Figure12: Small intestinal paracellular permeability in non-fasted NL3R451C and WT mice Effect of non-fasting on NL3R451C and wild-type mice. Repeated measures two-way ANOVA, correcting for multiple comparisons with Sidak post hoc. Data are expressed as mean ± SEM. (Two-way ANOVA results are shown in Appendix Tables 6-8).

3.2 Permeability effects in wild-type and mutant fasted mice

Paracellular permeability was determined in fasted mice by injecting FITC into ex-vivo sac from

duodenum, jejunum, distal ileum, and colon preparations and quantifying the amount of FITC present

in the surrounding solution (i.e. that passed through intestinal layers via the paracellular pathway).

In all gastrointestinal regions, the concentration of FITC increased linearly with time and did not

plateau and gradient values were different between gastrointestinal regions (Appendix Table 5). In

the duodenum, fasted NL3R451C mice had increased paracellular permeability compared to wild-type

littermates at the 60, 90 and 120-minute timepoints (Figure 13A). In the jejunum, fasted NL3R451C

mice also had increased paracellular permeability compared to wild-type littermates, however this

effect took longer to become evident (i.e. increased permeability was seen at 90 and 120-minute

timepoints but not at 60 minutes after FITC addition) (Figure 13B). In the distal ileum, fasted NL3R451C

mice similarly had increased paracellular permeability compared to wild-type littermates which was

observed at 60, 90 and 120-minute timepoints (Figure 13C). In the colon, fasted NL3R451C mice had

increased paracellular permeability in comparison with wild-type littermates, however this was only

observed at the 120-minute timepoint (Figure 13D), Additionally, the amount of FITC that permeated

through the colon was decreased, compared to the duodenum, jejunum and distal ileum for NL3R451C

mice and their wild-type littermates. These findings show that although no differences were seen

when mice were fed standard chow ad libitum, when NL3R451C mice are fasted, significant differences

in small and large GI paracellular permeability are found compared to wild-type littermates.

Figure 13: Small and large intestinal paracellular permeability fasted NL3R451C and WT mice Effect of fasting on NL3R451C and wildtype mice. Repeated measures two-way ANOVA, correcting for multiple comparisons with Sidak post hoc. Data are expressed as mean ± SEM. p= * <0.05, ** <0.01, *** <0.001, ****<0.0001. (Two-way ANOVA results are included in Appendix tables 9-12).

3.3 Effects of fasting on NL3R451C mice

In addition to comparing permeability in samples from wild-type and mutant mice, the paracellular

pathway was compared in fasting and non-fasting mice of the same genotype to ascertain the effect

of fasting. In NL3R451C mice, for all gastrointestinal areas, the concentration of FITC grew linearly with

time and did not plateau and gradient values were different between gastrointestinal regions

(Appendix Table 5). In the duodenum of fasted and non-fasted NL3R451C mice there was no significant

difference in paracellular permeability (Figure 14A). In the jejunum, fasted NL3R451C mice had

increased paracellular permeability compared to non-fasted NL3R451C mice at the 120-minute

timepoint only (Figure 14B). In the distal ileum, fasted NL3R451C mice had increased paracellular

permeability compared to non-fasted NL3R451C mice at the 90 minute and 120-minute timepoints.

Interestingly, in the duodenum, jejunum and distal ileum, non-fasting NL3R451C mice showed a larger

variance in paracellular permeability between animals at 90 and 120-minute timepoints compared

to fasted NL3R451C mice (Figure 14C). These findings show that fasting in NL3R451C mice results in

significant differences in jejunal and distal ileal paracellular permeability compared to non-fasted

NL3R451C mice.

Figure 14: Small intestinal paracellular permeability for fasted and non-fasted NL3R451C mice Effect of fasting compared to non-fasting NL3R451C mice. Repeated measures two-way ANOVA, correcting for multiple comparisons with Sidak post hoc. Data are expressed as mean ± SEM. p= *<0.05, **<0.01, ***<0.001, **** <0.0001. (Two-way ANOVA results are included in Appendix tables 13-15).

3.4 Effects of fasting on wild-type mice

Paracellular permeability was also compared in fasted and non-fasted wild-type mice. In all

gastrointestinal areas, the concentration of FITC grew linearly with time and did not plateau and

gradient values were different between gastrointestinal regions (Appendix Table 5). In the

duodenum, fasted wild-type mice had decreased paracellular permeability compared to non-fasting

wild-type mice at 90 and 120-minute timepoints (Figure 15A). In the jejunum, there was no significant

difference in paracellular permeability between fasted and non-fasted wild-type mice (Figure 15B).

In the distal ileum, fasted wild-type mice had decreased permeability compared to non-fasted

NL3R451C mice at 90 and 120-minute timepoints (Figure 15C). Additionally, in the jejunum, non-fasted

wild-type mice showed a wider variance in permeability measures (i.e., a larger range of SEM values)

at 30, 60, 90 and 120-minute timepoints compared to fasted wild-type mice (Figure 15B). These

findings show that fasting wild-type mice causes a significant decrease in paracellular permeability in

the duodenum compared to non-fasting wild-type mice. Interestingly, in the distal ileum, fasted wild-

type mice showed increased paracellular permeability compared to non-fasted mice.

Figure 15: Effect of fasting and non-fasting on intestinal paracellular permeability in WT mice Effect of fasting compared to non-fasting wild-type mice. Repeated measures two-way ANOVA, correcting for multiple comparisons with Sidak post hoc, p < 0.05. Data are expressed as mean ± SEM. p= * <0.05, ** <0.01, *** <0.001, **** <0.0001. (Two-way ANOVA results are shown in Appendix Tables 16-18).

3.5 Effects of L-glutamine on fasted NL3R451C mice

Permeability was assessed in fasted Neuroligin-3R451C mice with or without L-glutamine (30 or 90 mM)

treatment. In all gastrointestinal areas, the concentration of FITC grew linearly with time and did not

plateau and gradient values were different between gastrointestinal regions (Appendix Table 9).

Application of 30 mM of L-glutamine treatment decreased duodenal paracellular permeability at 60,

90 and 120 minutes post FITC application in NL3R451C preparations (Figure 16A). Similarly, 90 mM of

L-glutamine treatment decreased duodenal paracellular permeability at 60, 90 and 120 minutes in

NL3R451C mice compared to untreated NL3R451C mice (Figure 16A). In the jejunum, 30 mM of L-

glutamine treatment decreased paracellular permeability at 90, and 120 minutes in NL3R451C mice

compared to untreated NL3R451C mice (Figure 16B). A similar effect (although after a shorter duration;

i.e., from the 60 minute timepoint) of 90 mM of L-glutamine treatment was observed at 60, 90, and

120 minutes as a decrease in jejunal paracellular permeability in NL3R451C mice compared to

untreated NL3R451C mice (Figure 16B). In ileum samples, 30mM of L-glutamine treatment decreased

distal ileal paracellular permeability at 90 and 120 minutes post FITC application in NL3R451C mice

compared to untreated NL3R451C mice (Figure 16C). Similarly, 90mM of L-glutamine treatment

decreased distal ileal paracellular permeability at the 90 and 120 minute timepoints in NL3R451C mice

compared to untreated NL3R451C mice (Figure 16C). In the colon, 30mM of L-glutamine treatment

decreased colonic paracellular permeability in NL3R451C mice at 120 minutes post FITC addition

compared to untreated NL3R451C mice (Figure 16D). 90mM of L-glutamine treatment also decreased

colonic paracellular permeability in NL3R451C mice but after a shorter duration; i.e., from 90 minutes

(and at 120 minutes) compared to untreated NL3R451C mice (Figure 16D).

In summary, L-glutamine restored paracellular permeability in the duodenum and jejunum from 60

minutes after application, and in the distal ileum and colon at 90 and 120 minutes following

application. Of interest, less FITC permeated the colon than the duodenum, jejunum, and distal ileum

in both NL3R451C and wild-type littermates (data included in Appendix Table 5).

Figure 16: L-Glutamine restored paracellular permeability in NL3R451C mice to WT levels in small and large intestinal regions Effect of L-glutamine on NL3R451C mice. Repeated measures two-way ANOVA, correcting for multiple comparisons with Tukey’s post hoc. Data are expressed as mean ± SEM.* denotes a significantly different treatment (NL3 90mM vs NL3) effect. # denotes a significantly different treatment effect (NL3 30mM vs NL3). p= * <0.05, ** <0.01, *** <0.001, ****<0.0001. p= # <0.05, ## <0.01, ### <0.001. (Two-way ANOVA results are shown in Appendix Tables 19-22).

3.6 Effects of L-glutamine on fasted wild-type mice

Permeability was assessed in fasted wild-type mice with or without L-glutamine (30 or 90 mM)

treatment. In all gastrointestinal areas, the concentration of FITC grew linearly with time and did not

plateau and gradient values were different between gastrointestinal regions (Appendix Table 9).

Application of 30 mM of L-glutamine treatment increased duodenal paracellular permeability at 60,

90 and 120 minutes post FITC application in wild-type preparations (Figure 17A). In contrast, 90 mM

of L-glutamine treatment had no significant difference in paracellular permeability (Figure 17A). In

the jejunum, 30 mM of L-glutamine treatment had no significant difference in paracellular

permeability (Figure 17B). In the jejunum, 90 mM of L-glutamine treatment had no significant

difference in paracellular permeability (Figure 17B). In ileum samples, 30mM of L-glutamine

treatment had no significant difference in paracellular permeability (Figure 17C). Similarly, 90mM of

L-glutamine treatment had no significant difference in paracellular permeability (Figure 17C). In the

colon, 30mM of L-glutamine treatment decreased paracellular permeability in wild-type mice at 120

minutes post FITC addition compared to untreated wild-type mice (Figure 17D). In contrast, 90mM

of L-glutamine treatment had no significant difference in paracellular permeability (Figure 17D).

In summary, L-glutamine increased paracellular permeability in the duodenum from 60 minutes after

application and decreased paracellular permeability at 120 minutes following application. (data

included in Appendix Table 5)

Figure 17: Small and large intestinal paracellular permeability on wild-type mice treated with L- glutamine Effect of L-glutamine on wild-type mice. Repeated measures two-way ANOVA, correcting for multiple comparisons with Sidak post hoc. Data are expressed as mean ± SEM. * denotes a significantly different treatment effect (30mM glutamine). p= * <0.05, ** <0.01, *** <0.001, **** <0.0001. (Two-way ANOVA results are shown in Appendix Tables 23-26).

3.7 Effects of caffeine on fasted NL3R451C mice

Paracellular permeability was measured by injecting FITC into duodenum, jejunum, distal ileum and

colon ex-vivo sac preparations representing the small and large intestines, subsequently determining

the quantity of FITC that passed through intestinal layers via the paracellular pathway in fasting

NL3R451C mice and fasting NL3R451C mice treated with caffeine. In all gastrointestinal areas, the

concentration of FITC grew linearly with time and did not plateau and gradient values were different

between gastrointestinal regions (Appendix Table 5). At 60, 90, and 120 minutes, 10mM of caffeine

treatment decreased duodenal paracellular permeability in NL3R451C mice compared to untreated

NL3R451C mice (Figure 18A). At 60, 90, and 120 minutes, 10mM of caffeine treatment decreased

jejunal paracellular permeability in NL3R451C mice compared to untreated NL3R451C mice (Figure 18B).

At 60, 90 and 120 minutes, 10mM of caffeine treatment decreased distal ileal paracellular

permeability in NL3R451C mice compared to untreated NL3R451C mice (Figure 18C). At 90 and 120

minutes, 10mM of caffeine treatment decreased colonic paracellular permeability in NL3R451C mice

compared to untreated NL3R451C mice (Figure 18D). Wild-type mice had less FITC permeate the colon

than the duodenum, jejunum, and distal ileum (Appendix Table 9). These findings show that in the

small and large intestine, caffeine significantly decreases paracellular permeability of NL3R451C mice

when compared to un-treated NL3R451C mice.

Figure 18: Caffeine restored paracellular permeability in NL3R451C mice to WT concentrations in small and large intestinal regions Effect of caffeine on NL3R451C mice. Repeated measures two-way ANOVA, correcting for multiple comparisons with the Sidak post hoc test. Data are expressed as mean ± SEM. * denotes a significantly different treatment effect (10mM caffeine) p= * <0.05, ** <0.01, *** <0.001, ****<0.0001. (Two-way ANOVA results are shown in Appendix Tables 27-30).

3.8 Effects of caffeine on fasted wild-type mice

Paracellular permeability was measured in the duodenum, jejunum, distal ileum and colon ex-vivo

sac preparations in fasted wild-type mice (controls) and fasted wild-type mice treated with 10 mM

caffeine. In all gastrointestinal areas, the concentration of FITC grew linearly with time and did not

plateau and gradient values were different between gastrointestinal regions (Appendix Table 5). In

the duodenum, 10mM caffeine treatment had no effect on permeability in wild-type mice (Figure

19A). In the jejunum, caffeine treatment decreased jejunal paracellular permeability at the 90 and

120 minute timepoints in wild-type mice (Figure 19B). The effect in the ileum was delayed in

comparison; at 120 minutes, 10mM of caffeine treatment decreased distal ileal paracellular

permeability in wild-type mice (Figure 19C). In the colon, caffeine had no effect on paracellular

permeability in wild-type mice (Figure 19D). Interestingly, overall, less FITC permeated the colon than

the duodenum, jejunum, and distal ileum in wild-type mice (Appendix Table 5). Overall, these findings

show that caffeine significantly decreases paracellular permeability in the jejunum and distal ileum

of treated wild-type mice when compared to un-treated wild-type mice.

Figure 19: Caffeine decreased paracellular permeability in WT mice Effect of caffeine (10mM) in the A) duodenum, B) jejunum, C) distal ileum and D) colon. Repeated measures two-way ANOVA, correcting for multiple comparisons with Sidak post hoc. Data are expressed as mean ± SEM. p= * <0.05, ** <0.01, *** <0.001, ****<0.0001. (Two-way ANOVA results are shown in Appendix Tables 31-34).

4.0 Discussion

This chapter explores gut permeability in different gastrointestinal regions in mice. Specifically, data

examining paracellular permeability in control fed and fasted wild-type and Neuroligin-3R451C mice

are discussed. Subsequently the effects of L-glutamine and caffeine treatment on GI permeability in

these mouse groups is discussed.

4.1 Understanding how NL3R451C mutation and feeding conditions effect paracellular permeability

We hypothesised that NL3R451C mice show increased paracellular permeability compared to wild-type

littermates. When fed standard chow ‘ad libitum’ no change in paracellular permeability in the

NL3R451C mouse model of autism compared to wild-type littermates was identified. In contrast, in

both small and large intestinal preparations, increased paracellular permeability was observed in

fasted NL3R451C mice compared to fasted wild-type littermates. Fasting in NL3R451C mice increased

paracellular permeability in jejunal and distal ileal preparations compared to non-fasted NL3R451C

mice. In contrast, fasting in wild-type mice reduced paracellular permeability in the duodenum and

increased permeability in the distal ileum. Given that the current study is the first to assess the

potential impact of an ASD-associated mutation (specifically the NL3R451C mutation) on GI paracellular

permeability, no previous studies are available for comparison. Several reports suggested that fasting

can impact paracellular permeability in both mice and humans although some studies showed no

effects. For example, one clinical study administered lactulose for a duration of 5 days and found no

significant change in blood lactulose levels (marker for paracellular permeability) however, a major

limitation for this study was the small sample size of 3-5 (Elia et al., 1987). A later study in 16-hour

fasted mice showed no difference in paracellular permeability (Alscher et al., 2001). On the other

hand, mice fasted for 1-3 days showed paracellular translocation of lipopolysaccharides (LPS) from

the gut lumen into the bloodstream compared to non-fasted mice (Faggioni et al., 2000). Excessive

levels of LPS in the bloodstream can lead to chronic low-grade systemic inflammation which is also

linked to increased paracellular permeability (Faggioni et al., 2000). Other studies show contradictory

results in terms of fasting and GI permeability in mice. For example, Ferraris and others showed that

fasting mice for less than two days decreases paracellular permeability whereas Mohr and colleagues

demonstrated that fasting mice for more than two days results in increased paracellular permeability

(Ferraris and Carey, 2000, Mohr et al., 2021). These studies do not identify the specific intestinal

location where paracellular permeability changes are taking place or if fasting is beneficial or

detrimental for paracellular permeability in animal models of ASD.

ASD phenotypes and an elevated risk of developing ASD are thought to be related to changes in the

gut microbiome (during early development (Li et al., 2017)). Therefore, dietary modifications like

intermittent fasting (IF) may be able to control ASD-like behaviour. IF restored fear conditioning in

an ASD mouse model with a Pten (Phosphatase and tensin homolog; a gene linked to ASD in humans)

haploinsufficiency however, whether this was due to restoration of the intestinal wall was not

explored (Cabral-Costa et al., 2018). In addition, IF has been shown to alter Brain Derived

Neurotrophic Factor (BDNF), ketone levels, and mTOR (mammalian target of rapamycin) pathway

activity andmay impact ASD-like behaviour. Increased mTOR signalling in the brain is a factor in the

pathogenesis of (ASD). In mice with mTOR-associated single gene mutations, inhibiting the mTOR

pathway by supplementation of amino acids, improves neuropathology and behaviour in an ASD

animal model (Han et al., 2013; Huber et al., 2015 ;Wu et al., 2017). Future clinical research is

required to determine whether IF has a positive impact on the symptoms of ASD”.

Based on the literature, we hypothesised that the NL3R451C mutation increased paracellular

permeability and fasting Neuroligin-3R451C mice for 16-18 hours would decrease paracellular

permeability. The current study found that the Neuroligin-3R451C mutation did not significantly alter

paracellular permeability in the small intestine in standard fed mice. This finding could be explained

by the rate of paracellular permeability not being affected by the Neuroligin-3R451C mutation and/or

the presence of luminal contents. Individuals with ASD expressing the NL3R451C mutation therefore

may fall into the subcategory of Individuals with ASD that are not prone to increased paracellular

permeability.

However, the current study also showed that following fasting, mice expressing the Neuroligin-3R451C

mutation had increased small and large intestinal paracellular permeability compared to fasted wild-

type littermates. This is the first finding indicating a difference in paracellular permeability between

Neuroligin-3R451C and wild-type littermate mice in an ex-vivo permeability assay. Although the

underlying mechanisms are unknown, tight junctions may play a role in this alteration in permeability

following fasting. Since tight junction proteins are the rate limiting factor for paracellular

permeability (Anderson and Van Itallie, 1995, Mitic and Anderson, 1998, Shen et al., 2011, Zihni et

al., 2016) the NL3R451C mutation may influence the composition of tight junction proteins, which could

alter the rate of paracellular permeability in the presence of fasting.

These results show that the Neuroligin-3R451C mutation increased paracellular permeability in all

intestinal regions and provides the first evidence to suggest that genetic mutations linked to ASD

could impair the rate of paracellular permeability in the gut. Such changes in permeability could allow

microbes, endotoxins, faeces, and nutrients to more readily translocate into the bloodstream, which

could contribute to chronic low-grade inflammation (Rönnbäck and Hansson, 2019).

Another important finding in this study showed increased permeability in the jejunum and distal

ileum of mice that were fasted when compared to non-fasted mice. Surprisingly, most studies

suggest that fasting mice for less than two days will result in decreased paracellular permeability,

however in the current experiments, however, I observed increased paracellular permeability when

food was removed for 16-18 hours.

Another important finding is that the effect of fasting in the duodenum of wild-type mice appeared

to have a restorative effect on paracellular permeability whilst in the distal ileum of wild-type mice

fasting increased paracellular permeability. These results from wild-type mice suggest that food

deprivation for 16-18 hours can differentially impact various regions of the gut. According to Mohr

and others (Mohr et al., 2021), the composition of the mucus layer is dependent on the duration of

time animals have fasted. Interestingly, preliminary data suggests that in the ileum, mucus density is

reduced in Neuroligin-3R451C mice and that the bacterial composition is also changed in this region

(Herath et al., 2022). Given that mucus density is inherently reduced in Neuroligin-3R451C mice, fasting

for 16-18 hours could have reduced mucus density further and exposed the paracellular pathway to

result in an increased rate of permeability. In contrast, fasting could increase duodenal mucosal

density and reduce ileal mucosal density. Changes in mucus can influence the rate of paracellular

permeability and could be a contributing factor to the current findings in fasted Neuroligin-3R451C and

wild-type mice. It would be interesting to know if mice expressing the Neuroligin-3R451C mutation have

an impaired mucus layer, since a shorter food deprivation duration could cause mucus layer

degradation and expose the paracellular pathway to damage from luminal contents.

The scope of this study was limited by several factors. In non-fasted mice, it was difficult at times to

flush out all luminal contents when preparing the ex-vivo sacs. As a result, FITC molecules may have

had a strong affinity for lipophilic substances such as faecal matter or potentially could not reach all

areas of the ex-vivo sac which may have resulted in regional luminal pressure differences altering the

rate of paracellular permeability. Although the data from the current study were supported by robust

statistical evidence, the animal numbers for each group were limited, leading to the possibility

that increasing the animal numbers would result in a greater statistical power.

Despite these promising outcomes, questions remain. Future studies could utilise larger molecules

of FITC molecule to further determine changes in permeability in NL3R451C mice. Future experiments

could analyse the transcriptome in different regions of the intestinal tract to determine protein

expression levels, with a particular emphasis on tight junction proteins, which are known to be

correlated with paracellular permeability. Additionally, future experiments could examine

physiological changes that occur when fasting time is increased or decreased and if these are

beneficial or detrimental in impacting paracellular permeability in the small and large intestine of

mice.

Practical implications drawn from these data are that specialised ASD clinicians could advise

individuals with ASD and their caregivers that increased gut permeability has been linked with autism

and to potentially avoid substances that are damaging to the gut wall or to consume nutrients and

supplements that promote a healthy gut. Fasting could be a viable therapeutic strategy for healing

intestinal permeability. By reversing increased intestinal permeability, systemic inflammation could

be reduced and that could have therapeutic implications for GI symptoms However, clinically, there

is currently no clear evidence of a positive or negative impact fasting has on GI dysfunction in

individuals diagnosed with ASD.

4.2 Understanding how L-glutamine and caffeine impact paracellular permeability

After identifying increased permeability in fasted mutant mice compared to wild type littermates,

this was rescued in the presence of L-glutamine and caffeine. L-glutamine and caffeine were selected

for this experiment do to being the most extensively examined supplements for lowering paracellular

permeability and being quickly translatable into the clinic. Multiple animal studies have revealed that

L-glutamine can protect tight junction proteins in enterocytes via protection from apoptosis and

stress, anti-inflammatory properties, and the direct maintenance of intestinal tissue integrity (Kim

and Kim, 2017). This study is the first evidence of L-glutamine reducing paracellular permeability in a

mouse model of ASD so further research to clarify mechanisms of action are needed. In the clinic,

critically ill individuals revealed no evidence of improvement in patient outcome following L-

glutamine treatment, but patients diagnosed with ASD were excluded from this research (Andrews

et al., 2011, Heyland et al., 2013). It would be of interest to determine whether L-glutamine

supplementation can decrease paracellular permeability in individuals diagnosed with ASD. Given

that ninety percent of children with ASD are selective eaters, which can lead to nutritional deficits

(Ahearn et al., 2001, Kral et al., 2013), L-Glutamine supplements could be more effective in this

subpopulation compared to the general public. For example, children with ASD commonly have

deficiencies in critical plasma amino acids due to inadequate protein consumption (Arnold et al.,

2003). In line with this, Glutamine concentrations were decreased in individuals with ASD when

plasma levels of 25 amino acids in children with high-functioning autism and healthy controls were

assessed (Shimmura et al., 2011). Nevertheless, this study did not include patients expressing the

Neuroligin-3R451C mutation which is the focus of the current project.

In light of the fact that L-glutamine is a non-essential amino acid and that amino acid levels fluctuate

depending on specific disease states, it is possible that the NL3R451C mutation reduces plasma

concentrations of L-glutamine and that L-glutamine supplementation restores bioaccessibility of L-

glutamine in the small and large intestines. Due to the closely related metabolic pathways of

glutamine, glutamate and GABA, any disruption to the glutamine-glutamate cycle can disrupt the

regulation of brain glutamate and GABA concentrations which are essential for normal functioning.

These neurotransmitters are also known to regulate synapse formation and play a crucial role in the

development of brain circuits (Kawada et al., 2021). Dysregulation of these pathways can contribute

to excitotoxicity and neurodegeneration such as that triggered by excess glutamate levels due to

glutamate crossing the blood-brain barrier (Sheldon and Robinson, 2007). It has also been

hypothesised that excess glutamate levels may contribute to ASD symptoms due to increased

cerebral inflammation (Blaylock and Strunecka, 2009). Incidentally, a known cause of cerebral

inflammation is increased paracellular permeability, and previous work from the current laboratory

has shown a low level baseline increase in parameters indicating inflammatory activity in both the

brain and the gut of these mice (Matta et al., 2020, Sharna et al., 2020). Therefore, it is possible that

providing ASD patients with L-glutamine supplements could reduce excitotoxicity,

neurodegeneration, and cerebral inflammation if present.

Although the current study did not evaluate if L-glutamine influences the levels of excitotoxicity,

neurodegeneration, cerebral inflammation, or GI dysfunction, we found that L-glutamine

supplementation decreases paracellular permeability in the small and large intestine in mice carrying

the NL3R451C mutation. In the current study, L-glutamine rescued an increase in small and large

intestinal paracellular permeability in NL3R451C mice. Surprisingly, L-glutamine decreased paracellular

permeability in the colon of fasted NL3R451C mice. This result was unexpected because it is reported

that the primary fuel source of enterocytes and colonocytes is L-glutamine and that L-glutamine

binds to the basal surface of colonocytes accessible by the bloodstream and not via the apical cell

surface (Blachier et al., 2009) which is the primary location of paracellular tight junction complexes.

It is therefore unclear how L-glutamine exerts its effects in the colon in the current experiments since

it was applied on the luminal surface. However, it is possible that L-glutamine may traverse the

mucosal barrier and interact further with other components of the host GI tissue prior to exerting its

effects in these assays.

An association of L-glutamine with decreased paracellular permeability in animal models and cellular

assays has been reviewed by (Kim and Kim, 2017) and several possible explanations exist for how L-

glutamine directly and indirectly protects tight junction proteins. For example, L-glutamine may

function in i) increasing the expression of tight junction proteins ii) producing anti-inflammatory

mediators, and iii) preventing enterocytes from apoptosis and cellular stress (Kim and Kim, 2017).

Therefore, it is conceivable that (via an unknown mechanism) the Neuroligin-3R451C mutation

depletes L-glutamine levels such that the intestinal barrier function is recovered with L-glutamine

supplementation.

Furthermore, we found that L-glutamine increases paracellular permeability in the duodenum. This

was unexpected since L-glutamine supplements are reportedly overwhelmingly beneficial to the

small intestinal paracellular pathway as reviewed by (Kim and Kim, 2017). On the other hand, a study

of a rat model of anorexia showed that oral supplementation of L-glutamine decreased paracellular

permeability in the colon (L'Huillier et al., 2019). Another study showed an increase in jejunal

paracellular permeability in fasted rats supplemented with L-glutamine (Yang et al., 1999). These rats

were fasted for 48 hours, whereas the current study fasted mice for 16-18 hours and observed

increase paracellular permeability in the duodenum. Due to the fact that the duodenum is the

principal location of L-glutamine absorption, additional L-glutamine could be toxic to duodenocytes

if consumed by wild-type mice.

Exposure to caffeine decreased paracellular permeability in duodenal, jejunal, ileal and colonic

preparations from fasted Neuroligin-3R451C mice. Previous research found that caffeine promoted

intestinal anion production via the RyR/Orai1/Ca2+ signalling pathway (Wei et al., 2018, Zhang et al.,

2019). Intestinal anion secretion is necessary for mucosal protection against a variety of factors,

including gastric acid, which can stimulate physiological anion secretion in the duodenum

(Flemstrom & Isenberg, 2001). Caffeine may therefore stimulate duodenal anion secretion and

provide intestinal mucosa protection in Neuroligin-3R451C mice. Similarly, a recent report that found

that caffeine induced gastric acid secretion in humans (Liszt et al., 2017). Based on this work, novel

potential therapeutic targets for the regulation of intestinal anion secretion and paracellular

permeability may include the RyR/Orai1/Ca2+ pathway regulated by caffeine (Wei et al., 2018, Zhang

et al., 2019). To date, however, no research has examined how caffeine affects paracellular

permeability in mice carrying the NL3R451C mutation.

Another important finding from the current work was that caffeine decreased small and large

paracellular permeability in Neuroligin-3R451C and wild-type littermates. As noted above, two

previous studies have shown that caffeine can stimulate the secretion of anions in the GI tract of

mice (Wei et al., 2018, Zhang et al., 2019). The production of anions due to caffeine stimulation is

evidence of neurally-evoked secretion, which also includes the production of mucus and water. It is

possible that the secretion of anions, water, and mucus blocks luminal contents from directly

accessing the paracellular pathway which could be an indirect approach of restoring paracellular

permeability. In our experiments, caffeine did not influence paracellular permeability in the

duodenum or colon in wild-type mice, showing that its action was specific to the jejunum and distal

ileum.

Overall, these findings may have limitations due to the small number of animals used which can

reduce statistical power. In addition, because a dose-response curve was not generated for L-

glutamine and caffeine, it is unknown what the optimal concentration for effects on permeability

would be. Altogether, these findings suggest that L-glutamine and caffeine are beneficial in restoring

paracellular permeability in mice expressing the R451C mutation in the nervous system adhesion

molecule, Neuroligin-3. It is imperative that further work is carried out to investigate molecular

mechanisms underlying the potential therapeutic properties of L-glutamine and caffeine on intestinal

epithelial cells (IECs). Based on our data, L-glutamine and caffeine are effective in specific regions of

the small and large intestine; therefore, capsules could be manufactured to dissolve and release their

bio-active cargo into specific regions of the digestive tract. This is an important consideration because

the oral route of absorption does not allow for the absorption of L-glutamine or caffeine in the colon

due to absorption occurring exclusively in the small intestine.

In addition, dose-response curves for L-glutamine and caffeine should be constructed to establish

optimal doses. To increase the statistical power of the study, the number of animals in each group

should be increased to enable all components of the study to be compared simultaneously. Plateau

control tests were not regularly carried out for any intestinal region due to time constraints (COVID

-19). However, a trial experiment (n=1) was carried out for a 3-hour preparation of the duodenum.

The concentration of FITC appeared to plateau after three hours (results not shown). Drugs or

supplements may cause the concentration of FITC to plateau quicker than it would under control,

thus knowing this information would be helpful.

These findings could have a significant impact on the GI symptoms associated with ASD and

potentially reduce the core neurological symptoms correlated with ASD via actions impacting the

gut-brain axis which is now known to transmit information from the gut to the brain and vice versa.

These findings have practical implications for clinicians and show that L-glutamine and/or caffeine

may be useful for gut health in people diagnosed with ASD. Clinicians could determine the plasma

concentration of L-glutamine and other amino acids in individuals with ASD to assess for impaired

gut permeability. Physiological L-glutamine concentrations could be determined, and L-glutamine

potentially recommended as a supplement for this patient group. Caffeine, however, as a CNS

stimulator, could have a negative impact on the core neurological deficits associated with ASD. On

the other hand, our data that caffeine could be beneficial for reducing gut permeability. Targeting

paracellular permeability with L-glutamine and caffeine may prove to be an innovative future

strategy for reducing gastrointestinal symptoms, chronic low-grade systemic inflammation, and

modifying core neurological symptoms in ASD patients as well as in the general population.

5.0 Conclusion

In this investigation, mice carrying the R451C mutation in Neuroligin-3 were assessed for increased

paracellular permeability compared to wild-type littermates. Additionally, we sought to determine if

natural supplements could modulate the rate of paracellular permeability in NL3R451C mice and wild-

type littermates. Two significant findings emerged from this study. One was that fasted NL3R451C mice

displayed increased paracellular permeability when compared to wild-type littermates and secondly,

that L-glutamine and caffeine restored the observed increase in paracellular permeability in the

mutant mice. This project is the first comprehensive investigation of paracellular permeability in a

pre-clinical mouse model of ASD and it has the potential to pave the way for future investigations

into novel treatments using dietary supplements for gastrointestinal symptoms as well as potentially

impacting the core neurological symptoms correlated with ASD. The most significant limitation of

this investigation was the lack of information to estimate optimal dosages of L-glutamine and caffeine

along with the small number of animals used due to time and COVID-19 related constraints. Whilst

this study did not confirm the mechanism for increased paracellular permeability in NL3R451C mice or

the mechanism for the restorative effects observed from L-glutamine and caffeine, our data

demonstrated that the Neuroligin-3R451C mutation impacts the paracellular pathway and provides a

novel area of research for assessing therapeutic targets for gut permeability.

Chapter 3: Optimisation of the patch-clamp recording technique in the enteric nervous system to examine action potential characteristics in mouse duodenal myenteric neurons.

1.0 Introduction

Synapses are specialised intercellular connections that enable neurons to communicate. Synaptic cell

adhesion molecules help align the membrane components of the synapse and aid precise signal

transduction events during synapse formation (Südhof, 2008). Neuroligin-3 (NLGN3) is a synaptic cell

adhesion molecule located at the postsynaptic membrane of specific synapses. NLGN3 binds to

neurexisn (NRXNs), which are located on the presynaptic membrane and influence synaptic function

during development (Craig and Kang, 2007). Genetic mutations causing the deletion of NLGN3 as well

as the R451C point mutation in the Nlgn3 gene (Jamain et al., 2003) are associated with ASD in

patients. Both NL3R451C and NLGN3 knockout (KO) mice show GI dysfunction due to alterations in

NLGN3 expression (Hosie et al., 2019; Leembruggen et al., 2020). GI dysfunction is seen in patients

with the NL3R451C mutation, including chronic intestinal pain post-meal regurgitation, diarrhoea,

faecal incontinence, oesophageal inflammation, chronic intestinal pain, bladder and stool control

problems (Hosie et al., 2019). The current laboratory has also demonstrated that mice expressing the

Neuroligin-3R451C mutation have faster small intestinal transit and increased numbers of myenteric

neurons in the jejunum of the small intestine (Hosie et al., 2019). In addition to these findings in

NL3R451C mice, mice lacking NLGN3 have dilated colons and faster colonic muscle contractions

(Leembruggen et al., 2020), demonstrating that Neuroligin-3 is required for healthy gut function. This

knowledge about GI dysfunction at the organ level in the presence of the R451C mutation is

important, but changes at the cellular level remain to be clarified. Although the R451C mutation

influences GI function, little is known about how this mutation impacts neuronal subtypes in the

enteric nervous system (ENS). Therefore, by examining cellular function including action potential

characteristics, as well as cellular morphology and neurochemical profiles of enteric neurons, a

clearer understanding of changes at the cellular level due to ASD-associated mutations (such as the

NL3R451C mutation) can be obtained. Future applications of this work can identify differences in

cellular mechanisms between NL3R451C and wild-type mice and assist in contributing to a

comprehensive database of enteric neuronal profile data using recent methodological techniques in

preclinical disease models. In the context of ASD, this research may contribute to the identification

of cellular therapeutic targets to reduce GI disturbances in individuals with autism.

2.0 Methods and materials

2.1 Animals

The original source of NL3R451C mice was Jackson Laboratories (Bar Harbour, Maine USA). For more

than 11 generations, experimental mice were grown on a C57BL/6 background at the University of

Melbourne's animal facility (Parkville, Victoria, Australia). We utilised male WT and NL3R451C mice.

These were produced by mating heterogeneous females with WT male mice producing in 50:50

NL3R451C and WT offspring (Tabuchi et al., 2007). Experimental mice were co-housed and culled by

cervical dislocation in compliance with RMIT University's animal guidelines (approved RMIT animal

ethics project AEC# 22647 formerly classified as #1727). The body weights of all experimental mice

were immediately recorded after cervical dislocation. Additionally, the small intestine of 2-4-week-

old Swiss white (imported from Animal Resource Centre, Western Australia) mice were used as an

age comparison for electrophysiological recording studies. Mice were placed on a dissecting bench

and their extremities pinned out. Dissecting scissors were used to cut open the abdominal cavity and

expose the internal organs. The small and large intestines were identified and separated by cutting

away connective and mesentery tissue. The duodenum tissue was identified as the 4cm of tissue

immediately distal to the stomach. Jejunum segments were excised when required as an additional

4cm of tissue located distal to the duodenum.

2.2 General perfusion/dissecting Krebs solution

A 4 cm segment of duodenal tissue was placed into a small glass petri dish lined with 3 mm of Sylgard

184 (Dow Corning). One day prior to the experimental day, the physiological Krebs solution consisting

of 11 mM glucose, 25 mM NaHCO3, 118 mM NaCl, 4.8 mM KCL and 1 mM NaH2PO4 was dissolved in

1 litre of MilliQ H20. On the experimental day, 5% CO2 and 95% O2 was added to the above Krebs

physiological solution for approximately 15 minutes at a flow rate of 0.5 litre/min as per (MacConaill,

1985) before adding 1.2 mM MgCl2, 2.5 mM CaCl2. This additional step prevents precipitation of

divalent ions such as MgCl2 and CaCl2 (MacConaill, 1985). Immediately after MgCl2 and CaCl2 were

added, 1 ml of 6 μl nicardipine and 2 μl atropine was added to reduce tissue contractility by blocking

Ca2+ channels and equilibrated for at least 20 minutes prior to the solution being oxygenated with

carbogen (5% CO2 and 95% O2). 50 mL of dissecting Krebs solution was placed inside the glass petri

dish, along with a 4 cm duodenal segment of GI tract tissue.

2.3 Microdissection

Myenteric neurons were revealed by peeling away the small intestinal layers by microdissection. A 4

cm duodenal segment (Figure 20A) was transferred to a custom-made glass petri dish lined with 3

mm depth of Sylgard. Custom-made 2-4 mm long pins (from 0.1 mm diameter fine wire, California

Fine Wire Company) were used to secure the duodenal segment to the Sylgard in the petri dish. Using

straight scissors with blunt tips (Fine Science tools, cat. No. 15025-10) the duodenal segment was cut

open along the mesenteric border (Figure 20B-C) and additional pins were used to fully secure the

intestinal segment to the Sylgard. Fine forceps (Dumont no.5) were used to peel back the mucosa

layer (Figure 20D-E). Ultra-fine forceps (Dumont no.4) were used to peel back the sub-mucosa and

subsequently the circular muscle to reveal the myenteric plexus (Figure 20F). The remaining layers

(i.e., the LM layer and the MP) formed a LMMP preparation.

Figure 20: Visual representation of obtaining an LMMP preparation A) Two incisions were required to obtain a duodenal segment. B) Duodenal segment with attached mesenteric border pinned to 3mm of Sylgard. C) Duodenal segment being cut with micro scissors along the mesenteric border, removing excess mesentery tissue. D) Pinned and stretched duodenal segment showing mucosal layer. E) Peeled mucosal, sub-mucosal and circular muscle layers. F) Peeled myenteric plexus and circular muscle layers (Osorio et al., 2011)

2.4 Identification of myenteric ganglion.

LMMP preparations were transferred to an Olympus IX-50 inverted microscope. Using 10x

magnification, an appropriate ganglion suitable for recording (i.e., based on infrared imaging

suggesting that the ganglion was healthy and had no bright/reflective areas) was selected. By

switching to 40x magnification a further assessment of the viability of a myenteric neuron to be

targeted for recording was made. The first step in this process was to visually distinguish between

glial cells and neurons. Typically, myenteric neurons are larger than glial cells. The cellular

membranes of healthy myenteric neurons are elongated and devoid of dark intracellular granules. If

the neuron takes on a circular shape, this may indicate that it is unhealthy due to the cell membrane

becoming perforated or unhealthy which leads to the cell swelling. Dark intracellular granules

typically indicate apoptosis which is a sign the membrane is not healthy and cannot be used for

electrophysiological studies.

2.5 Protease solution

Once an appropriate ganglion was selected, 2-3 mL of a 0.02% wt/vol protease solution was locally

applied to the ganglion of interest using the local perfusion outlet (Figure 21) in order to degrade the

overlying connective tissue to enable penetration of the recording electrode. Simultaneously, a

custom-made glass patch pipette (with a hair/eyelash fixed with glue at the tip) (Figure 22) was used

to brush away the surface connective tissue that had undergone proteolytic digestion. This process

is essential to enable the patch pipette tip to facilitate seals more readily with the cell membrane (as

described in Osorio et al., 2011).

2.6 Whole-cell patch recording

Once the connective tissue overlying a ganglion of interest had been digested and healthy myenteric

neurons were visualised under 40x magnification, a patch pipette was freshly pulled from the pipette

puller (Brown-Flaming P-97, Sutter Instrument Company) with a pipette resistance ranging from 2.0

to 3.3 mega ohms. The patch pipette was then filled with an internal solution to facilitate current

clamp recordings. The neuronal microenvironment requires optimization of three major factors: pH,

osmolarity and concentration of solutes. The internal solution for current clamp recordings consisted

of 0.4 mM Na-GTP, 1mM MgCl2, 110mM KCl, 30mM KF 1mM CaCl2, 10mM HEPES and 2mM EGTA. To

reach a pH of 7.4, 1-3ml of 2M KOH was used the osmolarity adjusted to 300 mOsm using 4-5ml of

2M KCL. 2mg/ml of biocytin dye was added to the patch pipette, prior to recording to enable the

morphology of the neurons to be visualized.

2.7 External patching solution

To further increase the amount of time a tight seal between the glass patch pipette is maintained

with the membrane of a myenteric neuron, an external physiological patching solution is locally

perfused over the myenteric ganglion. This solution contains the same solutes as the dissecting

solution, with the addition of HEPES. Furthermore, the pH is adjusted to 7.4 and the osmolarity is in

the appropriate range of 310mOsm and 320mOsm. By having a higher osmolarity in the external

patching solution in comparison to the internal solution tighter seals are facilitated and the duration

of holding a tight seal between the glass patch clamp pipette and the myenteric neuron is increased.

The external solution is comprised of: 1mM MgCl2, 11mM glucose, 4mM KCL, 140mM NaCl, 2.5mM

CaCl2 and 10mM HEPES. Using 2M NaOH, the pH was adjusted to 7.4. Subsequently, the osmolarity

was adjusted to within the range of 310-320 mOsm using 2M NaCl.

Figure 21: Visual representation of the general set up of patch-clamping General set up for patch clamping and visual representation of the manifold which locally perfuses 0.02% wt/vol protease onto the selected myenteric ganglion (Osorio et al., 2011)

Figure 22: Custom-made hair cell attached to a glass micropipette The hair cell and protease are used simultaneously to remove surface connective tissue and cell surface proteins to facilitate seal formation on the cell membrane (Osorio et al., 2011).

2.8 Current clamp recording protocol

The whole-cell patch clamp technique allows cellular functional characteristics to be recorded using

two configurations: current clamp or voltage clamp mode. In this study, the current clamp

configuration was utilised to record evoked action potentials in response to a series of current steps.

Using a MultiClamp 700B amplifier (Molecular Devices Corporation, California) connected to an

AxonDigidata 1550B digital to analog converter (DAC; Molecular Devices Corporation, California), a

series of current steps were injected into the recorded cell. Specifically, thirteen current steps in 10

pA increments (ranging from a minimum of -30 pA to a maximum step of 90 pA) and 600 ms in

duration were applied to each recorded cell. Cells were held at -70 mV and cell responses were

recorded in current clamp configuration using PClamp 10 software (Molecular Devices Corporation,

California) and analysed using ClampFit10 (Molecular Devices Corporation, California).

Cellular parameters were recorded and saved in a digital lab book using ClampFit10. Cellular

parameters included i) Rm (Membrane Resistance): the resistance of a cell membrane to an external

force which is inversely proportional to the conductivity of the membrane ii) Cm (Membrane

Capacitance): the capacitance of a neuron reflecting the amount of charge required to change its

potential (larger cells with a larger membrane surface area have a higher capacitance and therefore

require more charge to change their voltage) iii) Tau (Membrane Time Constant): the voltage change

rate of a cell in response to currents flowing across its membrane is expressed by the value of tau iv)

Ra (Access Resistance): access resistance is generated by the open tip at the end of the patch pipette

and an optimal resistance of between 2-3.5 mOhms was selected for whole-cell patch clamp

recordings based on a previous study (Taylor, 2012).

Action potential traces and corresponding characteristics were obtained using the "transfer trace"

and "analyse statistics" functions from the ClampFit10.0 software package. Action potential

characteristics obtained included peak (mV), overshoot (mV), resting membrane potential (RMP;

mV), firing threshold (mV), after-hyperpolarisation peak (AH; mV), duration elapsed below RMP (ms)

and half-width (ms). Additionally, a graphical output of an action potential was designed to illustrate

how characteristics were calculated in ClampFit10.0 (Figure 23).

2.9 Statistical analysis

Action potential parameter data were imported from ClampFit10.0 spreadsheets into GraphPad

Prism 9.0 (University of California, San Diego) and analysed using two-tailed unpaired students t-test

and expressed as mean plus or minus standard error of the mean. Additionally, when experimental

outcomes of younger and older mice were compared, a difference between proportions using a Chi-

square test was performed to determine the effect size and a Fisher’s exact test was used to compute

the P-value.

Figure 23: A myenteric neuron action potential with measured characteristics AH: After- hyperpolarisation, RMP: Resting membrane potential.

3.0 Results

3.1 Comparison of success rate of neuronal recordings for older versus younger mice.

Obtaining recordings from myenteric neurons using the whole-cell patch clamp technique is

inherently technically challenging. To increase the number of successful recordings, we examined if

animal age affects the recording success rate. In this study, we compared outcomes of recording

attempts of myenteric neurons from 2-5- week-old male C57BL/6 mice, 2-4-week-old Swiss white

female mice and 6-12-week-old male C57BL/6 mice. A successful recording was defined as achieving

whole cell configuration and visualizing a trace in real time during the recording protocol which was

applied using pClamp 10.0 (Figure 24). Of a total of 47 neurons (obtained from 45 mice) where

recording was attempted, the overall success rate of obtaining whole cell configuration was 27%

((13/47 neurons) shown in Appendix Tables 1-3)). These 47 duodenal myenteric neuron recordings

were obtained from (31 C57BL/6 mice and 14 Swiss mice).

Figure24: Number of times whole cell seal configuration was obtained Attempted recording of mouse myenteric neurons at different ages and in Swiss white and C57BL/6 female mouse strains.

The success rate for 2-5-week-old male C57BL/6 mice was 37.5% (3 of 8 attempted recordings). The

success rate for 2-4-week-old Swiss white mice was 50% (7 of 14 attempted recordings). In contrast,

the success rate for 6-12-week-old mice was 12% (3 of 25 attempted recordings).

Figure25: Number of times whole cell seal obtained and/or AP firing recorded in myenteric neurons using younger and older mice Statistical comparisons were made using a Chi-square test and effect size was used in a Fisher’s exact test to determine a P-value. p= **** <0.0001.

Recordings from young mice of both Swiss white and C57BL/6 strains were combined and the success

rate compared for younger (2-5 weeks of age) versus older (6-12 weeks of age) mice. The success

rate for achieving whole cell configuration in myenteric neurons from 2-5-week-old male or female

mice Swiss white or C57BL/6 mice was 45% (i.e., 10 of 22 attempted recordings). This success rate

was higher in younger mice compared to that achieved for enteric neuron recording attempts in older

mice (i.e., 12%; 3 of 25 attempted recordings in 6-12-week-old male mice). In total, 13 recordings

were obtained from 13 cells (10 cells from 2-5 week-old and 3 cells from 6-12 week-old mice, data

included in Appendix Tables 1-8). Chi-square test statistical data are shown in Appendix Table 4.

3.2 Recording action potentials in younger mice.

To compare functional properties for recorded duodenal myenteric neurons, action potential

parameters from three neurons were compared. The remaining ten neuronal recordings were

omitted from a comprehensive examination because the myenteric neuron did not generate an

action potential or because there was too much electrical interference visible on the recording trace,

which prevented a precise analysis. The three cells analysed identified the minimum current required

to initiate and sustain action potentials, as well as the number of action potentials that occurred in

response to increasing current steps in these neurons. These 3 duodenal myenteric neurons were

recorded from a 4-week-old Swiss white female mouse (Neuron 1), a 4-week-old Swiss white female

mouse (Neuron 2) and a 4-week-old C57BL/6 male mouse (Neuron 3). From the full current step

protocol consisting of thirteen 10 picoAmp (pA) current steps (i.e., from -30 pA to +90 pA), the cellular

response to five current steps (i.e., 30, 0, 30, 60, 90 pA) was compared for the three duodenal

myenteric neurons as a visually representative way of presenting these data (Figure 26).

Figure 26: Three duodenal myenteric neurons exhibited action potentials in response to current steps A 10 pA current step was applied to myenteric neurons to stimulate action potentials using whole-cell patch clamping. Across these three neurons examined in detail, differences in the recording profiles were

determined. Neuron 1 exhibited properties typical of an S-neuron, with well-defined action

potentials in response to currents of 30, 60, and 90 pA (including 2 action potentials generated in

response to each of the 60 and 90 pA current steps). A minimum current of 30 pA was required to

trigger an action potential in this cell. Neuron 2 also exhibited typical features of an S-neuron,

including a well-defined single action potential in response to current step injections between 60 and

90 pA. Neuron 3 in contrast, exhibited AH-neuron-like characteristics. Neuron 3 showed an extended

refractory period and a single well-defined action potential following 30, 60, and 90 pA of injected

current. For neuron 3, similar to neuron 1, but in contrast to neuron 2, the minimal amount of current

required to initiate an action potential was 30 pA.

3.2.1 Comparison of firing properties in 3 myenteric neurons

In comparison to Neurons 1 and 3, Neuron 2 had a smaller action potential peak and a higher

minimum current was required to stimulate an action potential (60 pA). Additionally, Neuron 2 did

not exhibit a negatively projecting profile (i.e., afterhyperpolarization) when -30 pA current was

injected, whereas Neurons 1 and 3 exhibited afterhyperpolarization durations as a negative curve on

the recording trace when a -30 pA current step was applied. Interestingly, Neuron 2 and 3 fired a

single action potential at all current steps, whereas Neuron 1 fired multiple action potentials in

response to 60 and 90 pA current steps only.

3.2.2 Action potential characteristics in younger mice

GraphPad Prism 9.0 was used to plot action potentials triggered at 60 pA for each of the three

characterised neuronal recordings ((i.e., Neurons 1,2, and 3 (Figure 26)). All action potentials of the

3 selected neurons show minimal electrical interference, evident by minimal high frequency

oscillations. Initial qualitative observations were complemented by quantifying seven action

potential parameters for each of the neurons shown in Figure 26 (Table 3).

Table 3: Action potential parameters for three duodenal myenteric neurons Neuron 3

Parameter

Neuron 1

Neuron 2

Number of action potentials 13 5 7

Peak (mV) 41.44 17.82 48.86

Half width duration (ms) 3.76 2.09 2.71

RMP (mV) -49.39 -50.76 -42.72

Threshold (mV) -19.54 -24.04 -19.06

AH peak (mV) Duration below RMP (ms) -64.45 31.07 -51.15 0 -61.28 48.40

Input resistance (Ri)

5.0

4.61

3.75

Overshoot (mV) 60.98 41.86 67.92

3.3 Action potential characteristics observed in Neurons 1, 2 and 3

The total number of action potentials produced in response to the current injection protocol differed

for each of the three neurons. Neuron 1 generated 13 action potentials in total over the full protocol,

Neuron 2 generated 5 action potentials and Neuron 3 generated 7 action potentials (Figure 27) and

raw data outlined in (Appendix Tables 1-7).

The average peak height differed between all 3 neurons (Figure 27A). Neuron 1 showing a higher

peak height compared to Neuron 2 (Neuron 1 peak; 31.76 mV +/- 2.4 mV, Neuron 2 peak; 18.30 mV

+/-0.9 mV, p=0.0034). Neuron 3 showed a higher peak height compared to Neuron 1 (Neuron 1; 31.76

mV +/- 2.4 mV, Neuron 3; 47.94 mV +/- 0.6 mV, p=0.0001). When comparing Neurons 2 and 3, Neuron

3 showing a higher peak height (Neuron 2; 18.30 mV +/-0.9 mV vs Neuron 3; 47.94 mV +/- 0.6 mV,

p=<0.0001). When comparing action potential half width durations (Figure 27B). Neuron 2 had a

shorter half width than Neuron 1 (Neuron 1; 3.00 ms +/- 0.1 ms, Neuron 2; 2.20 ms +/- 0.04 ms,

p=0.0027). Average action potential half-width values were similar for Neurons 1 and 3 (Neuron 1;

3.00 ms +/- 0.1 ms, Neuron 3; 2.7 ms +/- 0.2 ms, p=0.127). Neuron 2 had a shorter average half width

compared to Neuron 3 (Neuron 2; 2.2 ms +/- 0.04 ms, vs Neuron 3; 2.7 ms +/- 0.2 ms, p=<0.0001).

The resting membrane potential was also calculated for these three neurons (Figure 27C). Neuron 2

had a lower resting membrane potential than Neuron 1 (Neuron 1; -46.2 mV +/- 0.6 mV, Neuron 2; -

51.1 mV +/- 0.4 mV, p=0.0003). Similarly, the resting potential of Neuron 1 was lower than that of

Neuron 3 (Neuron 1; -46.2 mV +/- 0.6 mV, Neuron 3; -42.2 mV +/- 0.3 mV, p=0.000). Neuron 2 had a

lower resting potential than Neuron 3 (Neuron 2; -51.1 mV +/- 0.4 mV, Neuron 3; -42.2 mV +/- 0.3

mV, p=<0.0001).

Action potential thresholds were also compared for the three neurons (Figure 27D). The average

action potential threshold was higher for Neuron 1 compared to Neuron 2 (Neuron 1; -17.8 mV +/-

0.5 mV, Neuron 2; -26.2 mV +/- 1.2 mV, p=<0.0001). No difference in action potential threshold was

found for Neurons 1 and 3 (Neuron 1; -17.8 mV +/- 0.5 mV, Neuron 3; -18.0 mV +/- 0.5 mV, p=0.7775).

Neuron 3 had a higher threshold for firing than Neuron 2 (Neuron 2; -26.2 mV +/- 1.2 mV, Neuron 3;

-18.0 mV +/- 0.5 mV, p=<0.0001).

When the AH peak was compared, Neuron 1 had a larger peak than Neuron 2 (Neuron 1; -61.2 mV

+/- 0.6 mV, Neuron 2; -50.1 mV +/- 1.8 mV, p=<0.0001) (Figure 27E). AH peaks for Neurons 1 and 3

were similar (Neuron 1; -61.2 mV +/- 0.6 mV, Neuron 3; -60.9 mV +/- 0.5 mV, p=0.7227). Neuron 3

had a larger AH peak than Neuron 2 (Neuron 2; -50.1 mV +/- 1.8 mV, Neuron 3; -60. 9 mV +/- 0.5 mV,

p=<0.0001).

The duration of time spent below resting membrane potential (RMP) differed for each of the 3

recorded neurons (Figure 27F). The duration of time spent below RMP was longer for Neuron 1

compared to Neuron 2 (Neuron 1; 35.1 ms +/- 3.1 ms, Neuron 2; 0.00 ms +/- 0.00 ms, p=<0.0001).

The duration spent below RMP was longer for Neuron 3 compared to Neuron 1 (Neuron 1; 35.1 ms

+/- 3.1 ms, Neuron 3; 52.3 ms +/- 4.7 ms, p=0.005). Neuron 3 also spent a longer duration below RMP

compared to Neuron 2 (Neuron 2; 0.00 ms +/- 0.00 ms, Neuron 3; 52.3 ms +/- 4.7 ms, p=<0.0001).

Three neuron comparisons of overshoot values were performed, and two comparisons showed

significant differences (Figure 27G). Overshoot values of Neuron 1 and 2 were similar; Neuron 1; 56.3

mV +/- 2.2 mV, Neuron 2; 44.5 mV +/- 2.0 mV, p=0.1169). Neuron 3 had a larger overshoot value than

Neuron 1. (Neuron 1; 56.3 mV +/- 2.2 mV, Neuron 3; 65.9 mV +/- 1.1 mV, p=0.0003. Neuron 3 had a

larger overshoot value than Neuron 2. (Neuron 2; 44.5 mV +/- 2.0 mV, Neuron 3; 65.9 mV +/- 1.1 mV,

p=<0.0001)

Figure 27: Action potential characteristics of three neurons A: Action potential peak height. B: Action potential half-width. C: Action potential resting membrane potential. D: Action potential threshold. E: After hyperpolarization peak. F: time below RMP. G: Action potential overshoot.

3.4 Neuronal profiles in younger mice

In addition to assessing the characteristics of individual action potentials of duodenal myenteric

neurons, neuronal profiles representing activity throughout the full duration of the current injection

protocol were generated.

Figure 28: Characterisation of action potential profiles recorded throughout the full current injection protocol for three duodenal myenteric neurons Action potential profile (A-C): All action potentials generated from a current clamp recording protocol. Current versus volts profile (D-F): The relationship between the current injected into a neuron (pA) and the resulting voltage (mV). Current versus number of action potentials (G-I): The relationship between the number of action potentials that are generated in response to an injected current (pA).

3.4.1 Comparison of action potentials full trace

By visualising the full trace recorded from each neuron, the change in cell membrane (mV) in

response to the injected current steps was assessed. This approach generated an overall view of the

cell response in terms of the number of action potentials fired and changes in membrane voltage

over a range of current steps in comparison to analysing the detailed parameters of a single action

potential for each cell (Figure 28A-C).

Neuron 1 responded to the full current step protocol with multiple action potentials over the

duration of the recording. Neuron 1 showed a clear voltage sag in response to hyperpolarisation

when negative current steps were applied (visible as a “sag” in the traces at the base of Figure 28A).

Neuron 2 also responded to the full current step protocol with multiple action potentials over the

duration of the recording. In contrast with Neuron 1, Neuron 2 showed no evidence of voltage sag in

response to hyperpolarisation when negative current steps were applied (Figure 28B). Neuron 3

responded to the full current step protocol with multiple action potentials over the duration of the

recording. Neuron 3 showed a voltage sag in response to hyperpolarisation when negative current

steps were applied (visible as a “sag” in the traces corresponding to negative injected current steps)

(Figure 28C). Due to the fast afterhypolarisation, all three neurons could be S neurons.

In comparison to Neurons 2 and 3, Neuron 1 fired more action potentials over the duration of the

recording. In response to negative current steps, neurons 1 and 3 exhibited a similar after-

hyperpolarisation. On the other hand, Neuron 2 exhibited no after-hyperpolarisation curve in

response to negative current steps. At the beginning of the recording, all neurons fired a cluster of

action potentials; however, only Neuron 1 continued to fire action potentials following the initial

cluster. Based on the action potential profile data, the membrane potential increased throughout

the protocol for all cells when increasing current was injected and all three neurons displayed

different cellular profiles when IV curves were compared throughout the protocol.

3.4.2 Comparison of current-voltage (IV) curves

An IV curve illustrates the relationship between the current injected into a neuron (pA) and the

resulting voltage (mV). Based on the current-voltage (I-V) curve data, the membrane potential

increased throughout the protocol for all cells in response to increasing current. When Neuron 1 was

stimulated at -30 pA, a minimum of -58 mV was generated; at 90 pA, a maximum of -35 mV was

generated; and the IV curve plateaued at 20 pA (Figure 28D). When Neuron 2 was stimulated at -30

pA, a minimum of -52 mV was generated; at 90 pA, a maximum of -30 mV was generated; and the IV

curve did not plateau throughout the duration of the recording (Figure 28E). When Neuron 3 was

stimulated at -30 pA, a minimum of -65 mV was generated; at 20 pA, a maximum of -30 mV was

generated; and the IV curve did not plateau throughout the duration of the recording (Figure 28F)

Compared to Neuron 1 and 2, Neuron 3 had a maximum value of -30 pA when 20 pA of current was

injected. In contrast, Neuron 1 and 2 reported similar maximum values of -58 mV and -52 mV

respectively, which occurred at 90 pA for Neuron 1 and 2. Interestingly, the IV curve of Neuron 1

plateaued, whereas Neuron 2 and 3 did not. Nevertheless, each of the three neurons studied in detail

displayed different cellular profiles when IV curves were compared.

3.4.3 Comparison of action potentials and current curves

An action potential and current curve shows the relationship between the number of action

potentials that are generated in response to injected current. Neuron 1 generated one action

potential when injected with 30 pA, two action potentials at 50 pA, and three action potentials at 70

pA. Interestingly, when 80 and 90 pA were injected into Neuron 1, two action potentials were

produced (Figure 28G). Neuron 2 generated one action potential when injected with 50 pA and one

action potential at 60 pA, 70 pA, 80 pA and 90 pA (Figure 28H). Neuron 3 generated one action

potential when injected with 30 pA and one action potential at 40, 50, 60, 70, 80 and 90 pA (Figure

100

28I). In contrast with the firing profiles of Neurons 2 and 3, Neuron 1 produced multiple action

potentials when 50 pA was injected. Neuron 2 and 3, however, did not fire multiple action potentials

irrespective of the amount of current injected into Neuron 2 or 3. Interestingly, Neuron 2 and 3

showed similar action potential current curves except that Neuron 2 fired the first action potential

later (i.e., following the 50 pA stimulation step) than Neuron 2 (which fired the first action potential

following the 30 pA stimulation step). Nevertheless, as noted above for the I-V curve analyses, each

of the three neurons studied in detail displayed different cellular profiles when action potential

current curves were compared.

101

3.5 Discussion

This study aimed to optimise the methodology for quantifying action potential characteristics and

neuronal firing profiles of myenteric neurons in C57BL/6 and Swiss white mice utilising whole-cell

patch clamp electrophysiology.

3.5.1 Success rate of whole-cell patch clamp recording in older versus younger mice.

In this investigation, recorded neurons displayed differences in action potential properties. In this

study, seven action potential characteristics were examined, as well as an action potential firing

profile, I-V curve, and the number of action potentials produced when increasing amounts of current

were injected into the myenteric neuron. These key features were assessed and comprised the

neuronal profile as reported in this thesis. Compared to the present work, previous evaluations of

action potential characteristics and neuronal profiles in the myenteric plexus have investigated the

AHP magnitude and duration, action potential firing profile, and resting membrane potential of CNS-

derived stem cells that developed into enteric neurons when implanted into the small intestine

(Kulkarni et al., 2011). The transplanted neurons exhibited similar properties to those of native

enteric neurons, but the authors did not identify if these neurons belonged to an established

subpopulation of enteric neurons. In contrast, Melo and others characterised the I-V curve and an

action potential firing profile, but not the number of action potentials produced at specific current

steps or action potential properties IPANS in the mouse duodenum (Melo et al., 2020). In yet another

study, Cai and colleagues investigated the effect of diffusing nitric oxide onto an LMMP

preparation on the action potential firing profile of ICCs (Intestinal cells of Cajal) in the myenteric

plexus (Cai et al., 2017). This study had a different focus and did not include analyses of I-V curves,

the number of action potentials, or other action potential features. Similarly, West and colleagues

examined IPANS in the myenteric plexus but did not report the same features as the current study

102

with respect to action potential profile parameters (West et al., 2020). Mao and others reported

mouse myenteric IPAN excitability by measuring action potential thresholds, action potential number

and resting membrane potential of in the jejunum when exposed to a common bacterium (Mao et

al., 2013). Mao and colleagues determined that the action potential threshold of the myenteric IPANs

was on average 5.9 mV +/-2.9, the number of action potentials was 1.6 +/- 0.7, and the average

membrane potential at rest was -63 mV. Although the number of action potentials reported by Mao

and others is comparable to Neuron 1 in the current investigation, the resting membrane potential

and action potential threshold differ. Based on their findings it is possible that Neuron 1 could be an

IPAN, however further assessment of morphological characteristics is needed in the current study.

Another study examined the characteristics of adult IPANS in the duodenum (Hao et al., 2012). Their

findings revealed a mean half-width duration of 2.1ms +/-1.1, which is comparable to that of the

three neurons in the present investigation. Nevertheless, their data demonstrated an average action

potential peak height of 105 mV +/- 1.8 , which differs from the three neurons examined in the

present investigation. Similar half width lengths could be the result of detecting the same

subpopulation of myenteric neurons, whilst variable action potential peak heights could be the result

of applying different recording procedures.

Although the number of neurons assessed in the current study is low, our approach is useful for

contributing to a comprehensive examination of the action potential characteristics and neuronal

profiles of myenteric neurons in mice. These data must be interpreted with caution, however, as

neuronal recordings were distributed across two distinct mouse strains and observed differences

could be attributed to these different mouse species. These findings have important implications for

the development of a more comprehensive resource for myenteric neuron profiling; however,

additional research is required to accumulate a large resource for useful analysis.

103

3.5.2 Future directions for patch-clamp electrophysiology in the ENS

Future experiments will aim to better understand the electrophysiological properties of submucosal

neurons to determine if submucosal neurons have altered functionality in the NL3R451C mouse

model which could impact the rate of intestinal permeability. Currently, it is known that in humans,

VIPergic pathways utilised by submucosal neurons to regulate intestinal permeability and the

production of the tight junction-associated protein ZO-1 (Neunlist et al., 2003). VIP promotes the

expression of important tight junction proteins and other connective components that affect

enterocyte paracellular permeability (Wu et al., 2015). Cholinergic pathways through muscarinic

and nicotinic receptors play a major role in mediating enhanced macromolecular permeability in

rats (Gareau et al., 2007). In rabbits, direct activation of the muscarinic receptors on the jejunal

epithelium results in an increase in paracellular permeability (Greenwood et al., 1992). Therefore, it

is possible to conclude that intestinal epithelial cells do impact permeability. The current study's

design did not explore mechanisms underlying why the R451C mutation enhanced intestinal

permeability but left open the possibility that additional factors could be involved. Since the focus

of the functional components of these experiments was on myenteric neurons rather than

submucosal neurons, the whole-cell patch clamp technique was not used to determine whether

submucosal neurons are involved in increased intestinal permeability. By combining

electrophysiological and morphological methods, the functional types of neurons in the ENS, such

as motorneurons, interneurons, and intrinsic primary afferent neurons, can be examined. Neuronal

characteristics, such as the expression profiles for ion channel currents and firing patterns, can be

examined in both healthy and diseased conditions. The ability to record via the patch clamp

technique from neurons while these cells maintain synaptic connections is a significant benefit of

our preparation. During patch clamp recording, electrical nerve bundle stimulation can be utilised

to record synaptic activity and examine the function of each functional class of neurons in intestinal

104

neural circuits. In a variety of mouse models, patch clamp investigations can be used to examine

the synaptic characteristics of myenteric neurons and the alterations that take place during

functional abnormalities (Osorio et al.,2011). The creation of therapeutics requires a thorough

understanding of the electrical characteristics of enteric neurons, synaptic responses, and the

changes connected to enteric neuropathies. While the use of patch-clamping techniques can record

currents, sharp-electrode recording methods cannot be used due to the inherent limitations of the

technique. The input resistance of cells decreases by 20–40% when sharp electrodes break through

the neuronal membrane (Li et al., 2004). Additionally, compared to patch-clamping, the resting

membrane potential (RMP) is much higher when sharp electrode recording is conducted. When the

membrane reseals, the RMP will stabilise, but the baseline RMP is higher. As a result, when

compared to sharp-cell electrode recordings, patch-clamp recordings may be more precise when

capturing input resistance and RMP (Li et al., 2004). Patch-clamping in the gut is intrinsically

challenging and requires a high degree of precision to expose myenteric neurons in mice. The

success of the patch-clamp technique, however, was improved using younger animals (2-4 weeks)

(Figure 24 and 25). To improve the value of using this low throughput technique the

electrophysiological data gained should be paired with the collection of single-cell genomic, and

cellular morphological data. Together these data can be used to create a “catalogue” of enteric

neurons that can be used for a variety of applications. Additionally, the ability to record a variety of

intracellular currents such as Na, K, IPSC, and EPSCs in enteric neurons is another benefit of whole-

cell patch clamping”. Post synaptic potentials may be examined more closely in the future. To

capture IPSC and EPSC instead, which is consistent with the NL3R451C mouse model demonstrating

increased IPSCs in the brain, (Hosie et al., 2019) we instead opted to use the whole-cell patch

clamping technique. We predict that duodenal myenteric neurons will have elevated IPSC, which

may help to explain GI dysfunction. The ENS is affected by the NL3R451C gene mutation, which

105

results in altered small intestinal transit and a shift in colonic motility mediated by the GABAA

receptor. Although it is unclear exactly how GABA affects colonic motility by way of GABAA

receptors, a recent study found that GABAA receptor subunits play a crucial role in rodents'

inflammatory reactions to psychological stress [Seifi, Rodaway, Rudolph, & Swinny, 2018].

Individuals with autism who frequently receive prescriptions for a variety of drugs including

benzodiazepines, which affect GABAergic pathways in the brain to treat anxiety, agitation, and

nervousness, are also very relevant to GABA-mediated changes in GI motility. Therefore, it would

be interesting to see how a GABAA antagonist such as bicuculline would influence the action

potential characteristics of myenteric neurons.

Further work is also needed to investigate the action potential characteristics and neuronal profiles

of specific subpopulations of myenteric neurons in Neuroligin-3R451C compared to wild-type

littermate mice. This research will assist in determining if aberrant cellular functions exist in

myenteric neurons in the Neuroligin-3R451C mouse model of ASD and support the need to identify

cellular therapeutic targets to reduce gastrointestinal symptoms in ASD.

106

3.6 Conclusion

The purpose of this study was to increase the probability of successfully recording myenteric neurons

using whole-cell patch clamp technology and to compile a resource of myenteric neuronal cellular

and action potential characteristics. Using younger mice resulted in a higher rate of successful

recordings compared to older mice. Due to time limitations and the low throughput of the patch

clamp recording technique, it was not possible to correlate the observed action potential

characteristics and neuronal profiles with other parameters contributing to the categorisation of

subpopulations of myenteric neurons in mice. Furthermore, recorded neurons were derived from

two distinct mouse strains, so any observed differences could be attributed to this variable. Finally,

the longer-term aim of this project was to characterise myenteric neurons in the Neuroligin-3R451C

mouse model of ASD, however this was not completed due to whole-cell patch clamping in the gut

being inherently difficult to perform and a longer than expected optimisation phase (in addition to

the occurrence of COVID-19 delays and restricted access to the laboratory). Nevertheless, this study

outlines a comprehensive approach for analysing cellular functional profiles of myenteric neurons in

LMPP preparations. It is anticipated that this study will pave the way for future investigations for pre-

clinical mouse models of ASD to develop novel and targeted therapies with the goal of reducing

gastrointestinal symptoms in ASD patients.

107

References

ABELY, M., DALLET, P., BOISSET, M. & DESJEUX, J. F. 2000. Effect of cholera toxin on glutamine metabolism and transport in rabbit ileum. Am J Physiol Gastrointest Liver Physiol, 278, G789-96. ACHAMRAH, N., DÉCHELOTTE, P. & COËFFIER, M. 2017. Glutamine and the regulation of intestinal

permeability: from bench to bedside. Curr Opin Clin Nutr Metab Care, 20, 86-91. AHEARN, W. H., CASTINE, T., NAULT, K. & GREEN, G. 2001. An assessment of food acceptance in

children with autism or pervasive developmental disorder-not otherwise specified. J Autism Dev Disord, 31, 505-11. AL-SADI, R., KHATIB, K., GUO, S., YE, D., YOUSSEF, M. & MA, T. 2011. Occludin regulates

macromolecule flux across the intestinal epithelial tight junction barrier. Am J Physiol Gastrointest Liver Physiol, 300, G1054-64. ALBRECHT, J., SIDORYK-WĘGRZYNOWICZ, M., ZIELIŃSKA, M. & ASCHNER, M. 2010. Roles of glutamine in neurotransmission. Neuron Glia Biol, 6, 263-76. ALLAIRE, J. M., CROWLEY, S. M., LAW, H. T., CHANG, S. Y., KO, H. J. & VALLANCE, B. A. 2018. The

Intestinal Epithelium: Central Coordinator of Mucosal Immunity. Trends Immunol, 39, 677- 696.

ALSCHER, K. T., PHANG, P. T., MCDONALD, T. E. & WALLEY, K. R. 2001. Enteral feeding decreases gut apoptosis, permeability, and lung inflammation during murine endotoxemia. American Journal of Physiology-Gastrointestinal and Liver Physiology, 281, G569-G576. AMORES-SÁNCHEZ, M. A. I. & MEDINA, M. Á. 1999. Glutamine, as a Precursor of Glutathione, and Oxidative Stress. Molecular Genetics and Metabolism, 67, 100-105. ANDERSON, J. M. & VAN ITALLIE, C. M. 1995. Tight junctions and the molecular basis for regulation of paracellular permeability. Am J Physiol, 269, G467-75.

ANDREWS, P. J., AVENELL, A., NOBLE, D. W., CAMPBELL, M. K., CROAL, B. L., SIMPSON, W. G., VALE, L. D., BATTISON, C. G., JENKINSON, D. J. & COOK, J. A. 2011. Randomised trial of glutamine, selenium, or both, to supplement parenteral nutrition for critically ill patients. Bmj, 342, d1542. ARAÇ, D., BOUCARD, A. A., OZKAN, E., STROP, P., NEWELL, E., SÜDHOF, T. C. & BRUNGER, A. T.

2007. Structures of neuroligin-1 and the neuroligin-1/neurexin-1 beta complex reveal specific protein-protein and protein-Ca2+ interactions. Neuron, 56, 992-1003. ARNOLD, G. L., HYMAN, S. L., MOONEY, R. A. & KIRBY, R. S. 2003. Plasma amino acids profiles in

children with autism: potential risk of nutritional deficiencies. J Autism Dev Disord, 33, 449- 54. BAIO, J., WIGGINS, L., CHRISTENSEN, D. L., MAENNER, M. J., DANIELS, J., WARREN, Z., KURZIUS-

SPENCER, M., ZAHORODNY, W., ROBINSON ROSENBERG, C., WHITE, T., DURKIN, M. S., IMM, P., NIKOLAOU, L., YEARGIN-ALLSOPP, M., LEE, L. C., HARRINGTON, R., LOPEZ, M., FITZGERALD, R. T., HEWITT, A., PETTYGROVE, S., CONSTANTINO, J. N., VEHORN, A., SHENOUDA, J., HALL-LANDE, J., VAN NAARDEN BRAUN, K. & 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, 1-23. BAUDOUIN, S. J., GAUDIAS, J., GERHARZ, S., HATSTATT, L., ZHOU, K., PUNNAKKAL, P., TANAKA, K. F.,

SPOOREN, W., HEN, R., DE ZEEUW, C. I., VOGT, K. & SCHEIFFELE, P. 2012. Shared synaptic pathophysiology in syndromic and nonsyndromic rodent models of autism. Science, 338, 128-32.

108

BERTRAND, P. P., KUNZE, W. A., BORNSTEIN, J. C. & FURNESS, J. B. 1998. Electrical mapping of the

projections of intrinsic primary afferent neurones to the mucosa of the guinea-pig small intestine. Neurogastroenterol Motil, 10, 533-41. BERTRAND, P. P., KUNZE, W. A., BORNSTEIN, J. C., FURNESS, J. B. & SMITH, M. L. 1997. Analysis of

the responses of myenteric neurons in the small intestine to chemical stimulation of the mucosa. Am J Physiol, 273, G422-35. BIRCHENOUGH, G. M. H., JOHANSSON, M. E., GUSTAFSSON, J. K., BERGSTRÖM, J. H. & HANSSON, G.

C. 2015. New developments in goblet cell mucus secretion and function. Mucosal Immunology, 8, 712-719. BJERKNES, M. & CHENG, H. 2005. Gastrointestinal stem cells. II. Intestinal stem cells. Am J Physiol Gastrointest Liver Physiol, 289, G381-7.

BLACHIER, F., BOUTRY, C., BOS, C. & TOMÉ, D. 2009. Metabolism and functions of L-glutamate in the epithelial cells of the small and large intestines. Am J Clin Nutr, 90, 814s-821s. BLIKSLAGER, A. T., MOESER, A. J., GOOKIN, J. L., JONES, S. L. & ODLE, J. 2007. Restoration of barrier function in injured intestinal mucosa. Physiol Rev, 87, 545-64. BOHÓRQUEZ, D. V., SHAHID, R. A., ERDMANN, A., KREGER, A. M., WANG, Y., CALAKOS, N., WANG, F.

& LIDDLE, R. A. 2015. Neuroepithelial circuit formed by innervation of sensory enteroendocrine cells. J Clin Invest, 125, 782-6. 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. Biochem J, 356, 581-8.

BORNSTEIN, J. C. & FOONG, J. P. P. 2018. Chapter 19 - Enteric Neural Regulation of Mucosal Secretion. In: SAID, H. M. (ed.) Physiology of the Gastrointestinal Tract (Sixth Edition). Academic Press.

BORNSTEIN, J. C., HENDRIKS, R., FURNESS, J. B. & TRUSSELL, D. C. 1991. Ramifications of the axons of AH-neurons injected with the intracellular marker biocytin in the myenteric plexus of the guinea pig small intestine. J Comp Neurol, 314, 437-51. BOUCARD, A. A., CHUBYKIN, A. A., COMOLETTI, D., TAYLOR, P. & SÜDHOF, T. C. 2005. A splice code

for trans-synaptic cell adhesion mediated by binding of neuroligin 1 to alpha- and beta- neurexins. Neuron, 48, 229-36. BOURGERON, T. 2015. From the genetic architecture to synaptic plasticity in autism spectrum disorder. Nat Rev Neurosci, 16, 551-63. BREHMER, A., SCHRODL, F. & NEUHUBER, W. 1999. Morphological classifications of enteric neurons--100 years after Dogiel. Anat Embryol (Berl), 200, 125-35. BUDRECK, E. C. & SCHEIFFELE, P. 2007. Neuroligin-3 is a neuronal adhesion protein at GABAergic and glutamatergic synapses. Eur J Neurosci, 26, 1738-48.

BUIE, T., FUCHS, G. J., 3RD, FURUTA, G. T., KOOROS, K., LEVY, J., LEWIS, J. D., WERSHIL, B. K. & WINTER, H. 2010. Recommendations for evaluation and treatment of common gastrointestinal problems in children with ASDs. Pediatrics, 125 Suppl 1, S19-29. 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. Molecular Autism, 6, 62. CABRAL-COSTA, J. V., ANDREOTTI, D. Z., MELLO, N. P., SCAVONE, C., CAMANDOLA, S. &

KAWAMOTO, E. M. 2018. Intermittent fasting uncovers and rescues cognitive phenotypes in PTEN neuronal haploinsufficient mice. Scientific Reports, 8, 8595.

CAI, Y., TANG, H., JIANG, F. & DONG, Z. 2017. Slow Wave Activity and Modulations in Mouse Jejunum Myenteric Plexus In Situ. J Neurogastroenterol Motil, 23, 117-123.

109

CAPORASO, N., G, M., LEMBO, V., D'ARGENIO, G., VITAGLIONE, P., A, R., GUARINO, M. & MORISCO, F. 2016. Decaffeinated coffee induces expression of tight junction proteins in rats with NAFLD. Functional Foods in Health and Disease, 6, 602. CHASTE, P. & LEBOYER, M. 2012. Autism risk factors: genes, environment, and gene-environment interactions. Dialogues Clin Neurosci, 14, 281-92. CHIH, B., ENGELMAN, H. & SCHEIFFELE, P. 2005. Control of excitatory and inhibitory synapse formation by neuroligins. Science, 307, 1324-8. CHUBYKIN, A. A., ATASOY, D., ETHERTON, M. R., BROSE, N., KAVALALI, E. T., GIBSON, J. R. &

SÜDHOF, T. C. 2007. Activity-dependent validation of excitatory versus inhibitory synapses by neuroligin-1 versus neuroligin-2. Neuron, 54, 919-31.

COMOLETTI, D., DE JACO, A., JENNINGS, L. L., FLYNN, R. E., GAIETTA, G., TSIGELNY, I., ELLISMAN, M. H. & TAYLOR, P. 2004. The Arg451Cys-neuroligin-3 mutation associated with autism reveals a defect in protein processing. J Neurosci, 24, 4889-93. COMOLETTI, D., FLYNN, R., JENNINGS, L. L., CHUBYKIN, A., MATSUMURA, T., HASEGAWA, H.,

SÜDHOF, T. C. & TAYLOR, P. 2003. Characterization of the interaction of a recombinant soluble neuroligin-1 with neurexin-1beta. J Biol Chem, 278, 50497-505. CONSTANTINO, J. N. & TODD, R. D. 2003. Autistic traits in the general population: a twin study. Arch Gen Psychiatry, 60, 524-30. COSTA, M., BROOKES, S. J. & HENNIG, G. W. 2000. Anatomy and physiology of the enteric nervous system. Gut, 47 Suppl 4, iv15-iv26. COSTA, M., BROOKES, S. J., STEELE, P. A., GIBBINS, I., BURCHER, E. & KANDIAH, C. J. 1996a.

Neurochemical classification of myenteric neurons in the guinea-pig ileum. Neuroscience, 75, 949-67. COSTA, M., BROOKES, S. J. H., STEELED, P. A., GIBBINS, I., BURCHER, E. & KANDIAH, C. J. 1996b.

Neurochemical classification of myenteric neurons in the guinea-pig ileum. Neuroscience, 75, 949-967. COURY, D. L., ASHWOOD, P., FASANO, A., FUCHS, G., GERAGHTY, M., KAUL, A., MAWE, G.,

PATTERSON, P. & JONES, N. E. 2012. Gastrointestinal conditions in children with autism spectrum disorder: developing a research agenda. Pediatrics, 130 Suppl 2, S160-8. CRAIG, A. M. & KANG, Y. 2007. Neurexin-neuroligin signaling in synapse development. Curr Opin Neurobiol, 17, 43-52. D'EUFEMIA, P., CELLI, M., FINOCCHIARO, R., PACIFICO, L., VIOZZI, L., ZACCAGNINI, M., CARDI, E. &

GIARDINI, O. 1996. Abnormal intestinal permeability in children with autism. Acta Paediatr, 85, 1076-9. DALVA, M. B., MCCLELLAND, A. C. & KAYSER, M. S. 2007. Cell adhesion molecules: signalling functions at the synapse. Nat Rev Neurosci, 8, 206-20.

DAVID, M. M., ENARD, D., OZTURK, A., DANIELS, J., JUNG, J. Y., DIAZ-BELTRAN, L. & WALL, D. P. 2016. Comorbid Analysis of Genes Associated with Autism Spectrum Disorders Reveals Differential Evolutionary Constraints. PLoS One, 11, e0157937. DE LA TORRE-UBIETA, L., WON, H., STEIN, J. L. & GESCHWIND, D. H. 2016. Advancing the

understanding of autism disease mechanisms through genetics. Nat Med, 22, 345-61. DE MAGISTRIS, L., FAMILIARI, V., PASCOTTO, A., SAPONE, A., FROLLI, A., IARDINO, P., CARTENI, M., DE ROSA, M., FRANCAVILLA, R., RIEGLER, G., MILITERNI, R. & BRAVACCIO, C. 2010. Alterations of the intestinal barrier in patients with autism spectrum disorders and in their first-degree relatives. J Pediatr Gastroenterol Nutr, 51, 418-24.

110

DE WIT, J., SYLWESTRAK, E., O'SULLIVAN, M. L., OTTO, S., TIGLIO, K., SAVAS, J. N., YATES, J. R., 3RD, COMOLETTI, D., TAYLOR, P. & GHOSH, A. 2009. LRRTM2 interacts with Neurexin1 and regulates excitatory synapse formation. Neuron, 64, 799-806.

DECKER, G. M. 2002. Glutamine: indicated in cancer care? Clin J Oncol Nurs, 6, 112-5. DETERS, B. J. & SALEEM, M. 2021. The role of glutamine in supporting gut health and neuropsychiatric factors. Food Science and Human Wellness, 10, 149-154.

DOGIEL, A. S. 1895. Zur Frage über den feinere Bau des sympathischen Nerven-systems bei den Säugethieren. Archiv für mikroskopische Anatomie, 46, 305-344. EKBLAD, E., WINTHER, C., EKMAN, R., HÅKANSON, R. & SUNDLER, F. 1987. Projections of peptide- containing neurons in rat small intestine. Neuroscience, 20, 169-88.

ELIA, M., GOREN, A., BEHRENS, R., BARBER, R. W. & NEALE, G. 1987. Effect of total starvation and very low calorie diets on intestinal permeability in man. Clin Sci (Lond), 73, 205-10. ELPHICK, D. A. & MAHIDA, Y. R. 2005. Paneth cells: their role in innate immunity and inflammatory disease. Gut, 54, 1802-1809. ETHERTON, M., FÖLDY, C., SHARMA, M., TABUCHI, K., LIU, X., SHAMLOO, M., MALENKA, R. C. &

SÜDHOF, T. C. 2011. Autism-linked neuroligin-3 R451C mutation differentially alters hippocampal and cortical synaptic function. Proc Natl Acad Sci U S A, 108, 13764-9. EVANS, M. A. & SHRONTS, E. P. 1992. Intestinal fuels: glutamine, short-chain fatty acids, and dietary fiber. J Am Diet Assoc, 92, 1239-46, 1249. FABRICHNY, I. P., LEONE, P., SULZENBACHER, G., COMOLETTI, D., MILLER, M. T., TAYLOR, P.,

BOURNE, Y. & MARCHOT, P. 2007. Structural analysis of the synaptic protein neuroligin and its beta-neurexin complex: determinants for folding and cell adhesion. Neuron, 56, 979-91. FAGGIONI, R., MOSER, A., FEINGOLD, K. R. & GRUNFELD, C. 2000. Reduced leptin levels in starvation increase susceptibility to endotoxic shock. Am J Pathol, 156, 1781-7. FERRARIS, R. P. & CAREY, H. V. 2000. Intestinal transport during fasting and malnutrition. Annu Rev Nutr, 20, 195-219. FINDLEY, M. K. & KOVAL, M. 2009. Regulation and roles for claudin-family tight junction proteins. IUBMB Life, 61, 431-7.

FIORENTINO, M., SAPONE, A., SENGER, S., CAMHI, S. S., KADZIELSKI, S. M., BUIE, T. M., KELLY, D. L., CASCELLA, N. & FASANO, A. 2016. Blood-brain barrier and intestinal epithelial barrier alterations in autism spectrum disorders. Mol Autism, 7, 49. FÖLDY, C., MALENKA, R. C. & SÜDHOF, T. C. 2013. Autism-associated neuroligin-3 mutations commonly disrupt tonic endocannabinoid signaling. Neuron, 78, 498-509.

FOONG, J. P., NGUYEN, T. V., FURNESS, J. B., BORNSTEIN, J. C. & YOUNG, H. M. 2012. Myenteric neurons of the mouse small intestine undergo significant electrophysiological and morphological changes during postnatal development. J Physiol, 590, 2375-90. FOONG, J. P., TOUGH, I. R., COX, H. M. & BORNSTEIN, J. C. 2014. Properties of cholinergic and non-

cholinergic submucosal neurons along the mouse colon. J Physiol, 592, 777-93. FOWLIE, G., COHEN, N. & MING, X. 2018. The Perturbance of Microbiome and Gut-Brain Axis in Autism Spectrum Disorders. Int J Mol Sci, 19. FURNESS, J. B. 2006. The organisation of the autonomic nervous system: peripheral connections. Auton Neurosci, 130, 1-5. FURNESS, J. B. 2012. The enteric nervous system and neurogastroenterology. Nat Rev Gastroenterol Hepatol, 9, 286-94.

111

FURNESS, J. B., ALEX, G., CLARK, M. J. & LAL, V. V. 2003. Morphologies and projections of defined

classes of neurons in the submucosa of the guinea-pig small intestine. The Anatomical Record Part A: Discoveries in Molecular, Cellular, and Evolutionary Biology, 272A, 475-483. FURNESS, J. B., CALLAGHAN, B. P., RIVERA, L. R. & CHO, H. J. 2014. The enteric nervous system and gastrointestinal innervation: integrated local and central control. Adv Exp Med Biol, 817, 39- 71. FURNESS, J. B., JONES, C., NURGALI, K. & CLERC, N. 2004. Intrinsic primary afferent neurons and nerve circuits within the intestine. Prog Neurobiol, 72, 143-64. FURNESS, J. B., KURAMOTO, H. & MESSENGER, J. P. 1990. Morphological and chemical

identification of neurons that project from the colon to the inferior mesenteric ganglia in the guinea-pig. J Auton Nerv Syst, 31, 203-10. FURNESS, J. B., RIVERA, L. R., CHO, H. J., BRAVO, D. M. & CALLAGHAN, B. 2013. The gut as a sensory organ. Nat Rev Gastroenterol Hepatol, 10, 729-40.

FURUSE, M., FUJITA, K., HIIRAGI, T., FUJIMOTO, K. & TSUKITA, S. 1998. Claudin-1 and -2: novel integral membrane proteins localizing at tight junctions with no sequence similarity to occludin. J Cell Biol, 141, 1539-50. FURUSE, M., HIRASE, T., ITOH, M., NAGAFUCHI, A., YONEMURA, S., TSUKITA, S. & TSUKITA, S. 1993.

Occludin: a novel integral membrane protein localizing at tight junctions. J Cell Biol, 123, 1777-88. GILBERT, M., SMITH, J., ROSKAMS, A. J. & AULD, V. J. 2001. Neuroligin 3 is a vertebrate gliotactin

expressed in the olfactory ensheathing glia, a growth-promoting class of macroglia. Glia, 34, 151-64. GOGOLLA, N., LEBLANC, J. J., QUAST, K. B., SÜDHOF, T. C., FAGIOLINI, M. & HENSCH, T. K. 2009.

Common circuit defect of excitatory-inhibitory balance in mouse models of autism. J Neurodev Disord, 1, 172-81. GONZALEZ, L. M., MOESER, A. J. & BLIKSLAGER, A. T. 2015. Animal models of ischemia-reperfusion-

induced intestinal injury: progress and promise for translational research. Am J Physiol Gastrointest Liver Physiol, 308, G63-75. GORRINDO, P., WILLIAMS, K. C., LEE, E. B., WALKER, L. S., MCGREW, S. G. & LEVITT, P. 2012.

Gastrointestinal dysfunction in autism: parental report, clinical evaluation, and associated factors. Autism Res, 5, 101-8. HA, S., SOHN, I. J., KIM, N., SIM, H. J. & CHEON, K. A. 2015. Characteristics of Brains in Autism

Spectrum Disorder: Structure, Function and Connectivity across the Lifespan. Exp Neurobiol, 24, 273-84. HANKARD, R. G., DARMAUN, D., SAGER, B. K., D'AMORE, D., PARSONS, W. R. & HAYMOND, M.

1995. Response of glutamine metabolism to exogenous glutamine in humans. Am J Physiol, 269, E663-70. HAO, M. M., LOMAX, A. E., MCKEOWN, S. J., REID, C. A., YOUNG, H. M. & BORNSTEIN, J. C. 2012.

Early development of electrical excitability in the mouse enteric nervous system. J Neurosci, 32, 10949-60. HERATH, M., BORNSTEIN, J. C., HILL-YARDIN, E. L. & FRANKS, A. E. 2022. The autism-associated

Neuroligin-3 R451C mutation alters mucus density and the spatial distribution of bacteria in the mouse gastrointestinal tract. bioRxiv, 2022.06.27.497808. HERATH, M., HOSIE, S., BORNSTEIN, J. C., FRANKS, A. E. & HILL-YARDIN, E. L. 2020. The Role of the

Gastrointestinal Mucus System in Intestinal Homeostasis: Implications for Neurological Disorders. Front Cell Infect Microbiol, 10, 248.

112

HEYLAND, D., MUSCEDERE, J., WISCHMEYER, P. E., COOK, D., JONES, G., ALBERT, M., ELKE, G., BERGER, M. M. & DAY, A. G. 2013. A randomized trial of glutamine and antioxidants in critically ill patients. N Engl J Med, 368, 1489-97. HIRST, G. D., HOLMAN, M. E. & SPENCE, I. 1974. Two types of neurones in the myenteric plexus of duodenum in the guinea-pig. J Physiol, 236, 303-26. HODGES, H., FEALKO, C. & SOARES, N. 2020. Autism spectrum disorder: definition, epidemiology, causes, and clinical evaluation. Translational pediatrics, 9, S55-S65. HORVATH, K. & PERMAN, J. A. 2002. Autism and gastrointestinal symptoms. Curr Gastroenterol Rep, 4, 251-8. HOSIE, S., ELLIS, M., SWAMINATHAN, M., RAMALHOSA, F., SEGER, G. O., BALASURIYA, G. K.,

GILLBERG, C., RÅSTAM, M., CHURILOV, L., MCKEOWN, S. J., YALCINKAYA, N., URVIL, P., SAVIDGE, T., BELL, C. A., BODIN, O., WOOD, J., FRANKS, A. E., BORNSTEIN, J. C. & HILL- YARDIN, E. L. 2019. Gastrointestinal dysfunction in patients and mice expressing the autism- associated R451C mutation in neuroligin-3. Autism Res, 12, 1043-1056. HSIAO, E. Y., MCBRIDE, S. W., HSIEN, S., SHARON, G., HYDE, E. R., MCCUE, T., CODELLI, J. A., CHOW,

J., REISMAN, S. E., PETROSINO, J. F., PATTERSON, P. H. & MAZMANIAN, S. K. 2013. Microbiota modulate behavioral and physiological abnormalities associated with neurodevelopmental disorders. Cell, 155, 1451-63.

ICHTCHENKO, K., HATA, Y., NGUYEN, T., ULLRICH, B., MISSLER, M., MOOMAW, C. & SÜDHOF, T. C. 1995. Neuroligin 1: a splice site-specific ligand for beta-neurexins. Cell, 81, 435-43. ICHTCHENKO, K., NGUYEN, T. & SÜDHOF, T. C. 1996. Structures, alternative splicing, and neurexin binding of multiple neuroligins. J Biol Chem, 271, 2676-82. IIDA, J., HIRABAYASHI, S., SATO, Y. & HATA, Y. 2004. Synaptic scaffolding molecule is involved in the synaptic clustering of neuroligin. Mol Cell Neurosci, 27, 497-508. IKENOUCHI, J., FURUSE, M., FURUSE, K., SASAKI, H., TSUKITA, S. & TSUKITA, S. 2005. Tricellulin constitutes a novel barrier at tricellular contacts of epithelial cells. J Cell Biol, 171, 939-45. INSTITUTE OF MEDICINE COMMITTEE ON MILITARY NUTRITION, R. 2001. Caffeine for the

Sustainment of Mental Task Performance: Formulations for Military Operations. Washington (DC): National Academies Press (US)

Copyright 2001 by the National Academy of Sciences. All rights reserved. ITO, Y., OSA, T. & KURIYAMA, H. 1974. Topical differences of caffeine action on the smooth muscle cells of the guinea pig alimentary canal. Jpn J Physiol, 24, 217-32.

JAMAIN, S., QUACH, H., BETANCUR, C., RÅSTAM, M., COLINEAUX, C., GILLBERG, I. C., SODERSTROM, H., GIROS, B., LEBOYER, M., GILLBERG, C. & BOURGERON, T. 2003. Mutations of the X-linked genes encoding neuroligins NLGN3 and NLGN4 are associated with autism. Nat Genet, 34, 27-9. KÅRLUND, A., PAUKKONEN, I., GÓMEZ-GALLEGO, C. & KOLEHMAINEN, M. 2021. Intestinal Exposure

to Food-Derived Protease Inhibitors: Digestion Physiology- and Gut Health-Related Effects. Healthcare (Basel, Switzerland), 9, 1002. KAWADA, K., KURAMOTO, N. & MIMORI, S. 2021. Possibility that the Onset of Autism Spectrum

Disorder is Induced by Failure of the Glutamine-Glutamate Cycle. Curr Mol Pharmacol, 14, 170-174.

KIM, E. & SHENG, M. 2004. PDZ domain proteins of synapses. Nat Rev Neurosci, 5, 771-81. KIM, M.-H. & KIM, H. 2017. The Roles of Glutamine in the Intestine and Its Implication in Intestinal Diseases. International journal of molecular sciences, 18, 1051. KONG, S., ZHANG, Y. H. & ZHANG, W. 2018. Regulation of Intestinal Epithelial Cells Properties and Functions by Amino Acids. BioMed Research International, 2018, 2819154.

113

KRAL, T. V., ERIKSEN, W. T., SOUDERS, M. C. & PINTO-MARTIN, J. A. 2013. Eating behaviors, diet

quality, and gastrointestinal symptoms in children with autism spectrum disorders: a brief review. J Pediatr Nurs, 28, 548-56.

KRANTIS, A. 2000. GABA in the Mammalian Enteric Nervous System. News Physiol Sci, 15, 284-290. KULKARNI, S., ZOU, B., HANSON, J., MICCI, M.-A., TIWARI, G., BECKER, L., KAISER, M., XIE, X. S. & PASRICHA, P. J. 2011. Gut-derived factors promote neurogenesis of CNS-neural stem cells and nudge their differentiation to an enteric-like neuronal phenotype. American journal of physiology. Gastrointestinal and liver physiology, 301, G644-55. KUNZE, W. A., BORNSTEIN, J. C. & FURNESS, J. B. 1995. Identification of sensory nerve cells in a peripheral organ (the intestine) of a mammal. Neuroscience, 66, 1-4.

L'HUILLIER, C., JARBEAU, M., ACHAMRAH, N., BELMONTE, L., AMAMOU, A., NOBIS, S., GOICHON, A., SALAMEH, E., BAHLOULI, W., DO REGO, J. L., DÉCHELOTTE, P. & COËFFIER, M. 2019. Glutamine, but not Branched-Chain Amino Acids, Restores Intestinal Barrier Function during Activity-Based Anorexia. Nutrients, 11. LEEMBRUGGEN, A. J. L., BALASURIYA, G. K., ZHANG, J., SCHOKMAN, S., SWIDERSKI, K., BORNSTEIN,

J. C., NITHIANANTHARAJAH, J. & HILL-YARDIN, E. L. 2020. Colonic dilation and altered ex vivo gastrointestinal motility in the neuroligin-3 knockout mouse. Autism Res, 13, 691-701. LI, Z., CHALAZONITIS, A., HUANG, Y. Y., MANN, J. J., MARGOLIS, K. G., YANG, Q. M., KIM, D. O.,

CÔTÉ, F., MALLET, J. & GERSHON, M. D. 2011. Essential roles of enteric neuronal serotonin in gastrointestinal motility and the development/survival of enteric dopaminergic neurons. J Neurosci, 31, 8998-9009. LISZT, K. I., LEY, J. P., LIEDER, B., BEHRENS, M., STÖGER, V., REINER, A., HOCHKOGLER, C. M., KÖCK,

E., MARCHIORI, A., HANS, J., WIDDER, S., KRAMMER, G., SANGER, G. J., SOMOZA, M. M., MEYERHOF, W. & SOMOZA, V. 2017. Caffeine induces gastric acid secretion via bitter taste signaling in gastric parietal cells. Proceedings of the National Academy of Sciences of the United States of America, 114, E6260-E6269. 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, 466-474.

MACCONAILL, M. 1985. Calcium precipitation from mammalian physiological salines (Ringer solutions) and the preparation of high [Ca] media. J Pharmacol Methods, 14, 147-55. MAO, Y. K., KASPER, D. L., WANG, B., FORSYTHE, P., BIENENSTOCK, J. & KUNZE, W. A. 2013.

Bacteroides fragilis polysaccharide A is necessary and sufficient for acute activation of intestinal sensory neurons. Nat Commun, 4, 1465.

MARTELLA, G., MERINGOLO, M., TROBIANI, L., DE JACO, A., PISANI, A. & BONSI, P. 2018. The neurobiological bases of autism spectrum disorders: the R451C-neuroligin 3 mutation hampers the expression of long-term synaptic depression in the dorsal striatum. Eur J Neurosci, 47, 701-708.

MATEER, S. W., CARDONA, J., MARKS, E., GOGGIN, B. J., HUA, S. & KEELY, S. 2016. Ex Vivo Intestinal Sacs to Assess Mucosal Permeability in Models of Gastrointestinal Disease. J Vis Exp, e53250.

MATTA, S. M., MOORE, Z., WALKER, F. R., HILL-YARDIN, E. L. & CRACK, P. J. 2020. An altered glial phenotype in the NL3R451C mouse model of autism. Scientific Reports, 10, 14492. MCELHANON, B. O., MCCRACKEN, C., KARPEN, S. & SHARP, W. G. 2014. Gastrointestinal symptoms in autism spectrum disorder: a meta-analysis. Pediatrics, 133, 872-83. MELO, C. G. D. S., NICOLAI, E. N., ALCAINO, C., CASSMANN, T. J., WHITEMAN, S. T., WRIGHT, A. M., MILLER, K. E., GIBBONS, S. J., BEYDER, A. & LINDEN, D. R. 2020. Identification of intrinsic

114

primary afferent neurons in mouse jejunum. Neurogastroenterology and motility : the official journal of the European Gastrointestinal Motility Society, 32, e13989-e13989. MITIC, L. L. & ANDERSON, J. M. 1998. Molecular architecture of tight junctions. Annu Rev Physiol, 60, 121-42. MOHR, A. E., GUMPRICHT, E., SEARS, D. D. & SWEAZEA, K. L. 2021. Recent advances and health

implications of dietary fasting regimens on the gut microbiome. American Journal of Physiology-Gastrointestinal and Liver Physiology, 320, G847-G863. MONGARDI FANTAGUZZI, C., THACKER, M., CHIOCCHETTI, R. & FURNESS, J. B. 2009. Identification

of neuron types in the submucosal ganglia of the mouse ileum. Cell Tissue Res, 336, 179-89. NEHLIG, A., DAVAL, J. L. & DEBRY, G. 1992. Caffeine and the central nervous system: mechanisms of

action, biochemical, metabolic and psychostimulant effects. Brain Res Brain Res Rev, 17, 139-70. NEUNLIST, M., VAN LANDEGHEM, L., MAHÉ, M. M., DERKINDEREN, P., DES VARANNES, S. B. &

ROLLI-DERKINDEREN, M. 2013. The digestive neuronal-glial-epithelial unit: a new actor in gut health and disease. Nat Rev Gastroenterol Hepatol, 10, 90-100. NEWSHOLME, E. A. & CARRIÉ, A. L. 1994. Quantitative aspects of glucose and glutamine metabolism by intestinal cells. Gut, 35, S13-7. NGUYEN, T. L. A., VIEIRA-SILVA, S., LISTON, A. & RAES, J. 2015. How informative is the mouse for human gut microbiota research? Disease models & mechanisms, 8, 1-16. NURGALI, K., STEBBING, M. J. & FURNESS, J. B. 2004. Correlation of electrophysiological and

morphological characteristics of enteric neurons in the mouse colon. J Comp Neurol, 468, 112-24. ODENWALD, M. A. & TURNER, J. R. 2017. The intestinal epithelial barrier: a therapeutic target? Nat Rev Gastroenterol Hepatol, 14, 9-21. OSORIO, N., DELMAS, P. & JONES, P. A. 2011. Patch clamp recording from enteric neurons in situ. Nat Protoc, 6, 15-27. PATIENCE, J. F. 1990. A review of the role of acid-base balance in amino acid nutrition. J Anim Sci, 68, 398-408. PETERSON, L. W. & ARTIS, D. 2014. Intestinal epithelial cells: regulators of barrier function and immune homeostasis. Nat Rev Immunol, 14, 141-53.

PORTBURY, A. L., POMPOLO, S., FURNESS, J. B., STEBBING, M. J., KUNZE, W. A., BORNSTEIN, J. C. & HUGHES, S. 1995. Cholinergic, somatostatin-immunoreactive interneurons in the guinea pig intestine: morphology, ultrastructure, connections and projections. J Anat, 187 ( Pt 2), 303- 21. POULOPOULOS, A., SOYKAN, T., TUFFY, L. P., HAMMER, M., VAROQUEAUX, F. & BROSE, N. 2012.

Homodimerization and isoform-specific heterodimerization of neuroligins. Biochem J, 446, 321-30. QU, Z. D., THACKER, M., CASTELUCCI, P., BAGYÁNSZKI, M., EPSTEIN, M. L. & FURNESS, J. B. 2008.

Immunohistochemical analysis of neuron types in the mouse small intestine. Cell Tissue Res, 334, 147-61. RAO, R. & SAMAK, G. 2012a. Role of Glutamine in Protection of Intestinal Epithelial Tight Junctions. Journal of epithelial biology & pharmacology, 5, 47-54. RAO, R. & SAMAK, G. 2012b. Role of Glutamine in Protection of Intestinal Epithelial Tight Junctions. J Epithel Biol Pharmacol, 5, 47-54. RAO, S. S. C. & BHAGATWALA, J. 2019. Small Intestinal Bacterial Overgrowth: Clinical Features and Therapeutic Management. Clin Transl Gastroenterol, 10, e00078.

115

RONALD, A., HAPPÉ, F., PRICE, T. S., BARON-COHEN, S. & PLOMIN, R. 2006. Phenotypic and genetic overlap between autistic traits at the extremes of the general population. J Am Acad Child Adolesc Psychiatry, 45, 1206-1214.

RÖNNBÄCK, C. & HANSSON, E. 2019. The Importance and Control of Low-Grade Inflammation Due to Damage of Cellular Barrier Systems That May Lead to Systemic Inflammation. Frontiers in Neurology, 10.

ROTHWELL, P. E., FUCCILLO, M. V., MAXEINER, S., HAYTON, S. J., GOKCE, O., LIM, B. K., FOWLER, S. C., MALENKA, R. C. & SÜDHOF, T. C. 2014. Autism-associated neuroligin-3 mutations commonly impair striatal circuits to boost repetitive behaviors. Cell, 158, 198-212.

RUSZCZYŃSKI, M., URBAŃSKA, M. & SZAJEWSKA, H. 2014. Gelatin tannate for treating acute gastroenteritis: a systematic review. Annals of gastroenterology, 27, 121-124. SANG, Q. & YOUNG, H. M. 1996. Chemical coding of neurons in the myenteric plexus and external muscle of the small and large intestine of the mouse. Cell Tissue Res, 284, 39-53. SCHEIFFELE, P., FAN, J., CHOIH, J., FETTER, R. & SERAFINI, T. 2000. Neuroligin expressed in nonneuronal cells triggers presynaptic development in contacting axons. Cell, 101, 657-69. 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-3(R451C) Mouse Model of Autism. Front Cell Neurosci, 14, 85.

SHELDON, A. L. & ROBINSON, M. B. 2007. The role of glutamate transporters in neurodegenerative diseases and potential opportunities for intervention. Neurochem Int, 51, 333-55. SHEN, L., WEBER, C. R., RALEIGH, D. R., YU, D. & TURNER, J. R. 2011. Tight junction pore and leak pathways: a dynamic duo. Annual review of physiology, 73, 283-309.

SHIMMURA, C., SUDA, S., TSUCHIYA, K. J., HASHIMOTO, K., OHNO, K., MATSUZAKI, H., IWATA, K., MATSUMOTO, K., WAKUDA, T., KAMENO, Y., SUZUKI, K., TSUJII, M., NAKAMURA, K., TAKEI, N. & MORI, N. 2011. Alteration of plasma glutamate and glutamine levels in children with high-functioning autism. PLoS One, 6, e25340. SHIPMAN, S. L., SCHNELL, E., HIRAI, T., CHEN, B. S., ROCHE, K. W. & NICOLL, R. A. 2011. Functional dependence of neuroligin on a new non-PDZ intracellular domain. Nat Neurosci, 14, 718-26. SINGH, P., TOOM, S. & HUANG, Y. 2017. Anti-claudin 18.2 antibody as new targeted therapy for advanced gastric cancer. Journal of Hematology & Oncology, 10, 105. SOUZA, N. C., MENDONCA, J. N., PORTARI, G. V., JORDAO JUNIOR, A. A., MARCHINI, J. S. &

CHIARELLO, P. G. 2012. Intestinal permeability and nutritional status in developmental disorders. Altern Ther Health Med, 18, 19-24. STEBBING, M. J. & BORNSTEIN, J. C. 1996. Electrophysiological mapping of fast excitatory synaptic

inputs to morphologically and chemically characterized myenteric neurons of guinea-pig small intestine. Neuroscience, 73, 1017-28.

STEVENSON, B. R., SILICIANO, J. D., MOOSEKER, M. S. & GOODENOUGH, D. A. 1986. Identification of ZO-1: a high molecular weight polypeptide associated with the tight junction (zonula occludens) in a variety of epithelia. J Cell Biol, 103, 755-66. SÜDHOF, T. C. 2008. Neuroligins and neurexins link synaptic function to cognitive disease. Nature, 455, 903-11.

TABUCHI, K., BLUNDELL, J., ETHERTON, M. R., HAMMER, R. E., LIU, X., POWELL, C. M. & SÜDHOF, T. C. 2007. A neuroligin-3 mutation implicated in autism increases inhibitory synaptic transmission in mice. Science, 318, 71-6. TAYLOR, A. L. 2012. What we talk about when we talk about capacitance measured with the voltage-clamp step method. Journal of computational neuroscience, 32, 167-175.

116

TOKUTOMI, Y., TOKUTOMI, N. & NISHI, K. 2001. The properties of ryanodine-sensitive Ca2+ release in mouse gastric smooth muscle cells. British Journal of Pharmacology, 133, 125-137. TURNER, J. R. 2009. Intestinal mucosal barrier function in health and disease. Nat Rev Immunol, 9, 799-809. VAHIA, V. N. 2013. Diagnostic and statistical manual of mental disorders 5: A quick glance. Indian journal of psychiatry, 55, 220-223. VAN ITALLIE, C. M. & ANDERSON, J. M. 2006. Claudins and epithelial paracellular transport. Annu Rev Physiol, 68, 403-29.

VAROQUEAUX, F., ARAMUNI, G., RAWSON, R. L., MOHRMANN, R., MISSLER, M., GOTTMANN, K., ZHANG, W., SÜDHOF, T. C. & BROSE, N. 2006. Neuroligins determine synapse maturation and function. Neuron, 51, 741-54. VOJDANI, A. 2013. For the assessment of intestinal permeability, size matters. Altern Ther Health Med, 19, 12-24.

VOLYNETS, V., REICHOLD, A., BÁRDOS, G., RINGS, A., BLEICH, A. & BISCHOFF, S. C. 2016. Assessment of the Intestinal Barrier with Five Different Permeability Tests in Healthy C57BL/6J and BALB/cJ Mice. Dig Dis Sci, 61, 737-46. WANG, J., MOU, Y., ZHANG, Q., ZHANG, F., YANG, H., ZHANG, W. & LI, A. 2013. Expression and

significance of neuroligins in myenteric cells of Cajal in Hirschsprung's disease. PLoS One, 8, e67205.

WEI, S. C., YANG-YEN, H. F., TSAO, P. N., WENG, M. T., TUNG, C. C., YU, L. C. H., LAI, L. C., HSIAO, J. H., CHUANG, E. Y., SHUN, C. T., NI, Y. H., XAVIER, R. J., PODOLSKY, D. K., YEN, J. J. Y. & WONG, J. M. 2017. SHANK3 Regulates Intestinal Barrier Function Through Modulating ZO-1 Expression Through the PKCε-dependent Pathway. Inflamm Bowel Dis, 23, 1730-1740. WEI, X., LU, Z., YANG, T., GAO, P., CHEN, S., LIU, D. & ZHU, Z. 2018. Stimulation of Intestinal Cl-

Secretion Through CFTR by Caffeine Intake in Salt-Sensitive Hypertensive Rats. Kidney Blood Press Res, 43, 439-448. WEST, C. L., MAO, Y. K., DELUNGAHAWATTA, T., AMIN, J. Y., FARHIN, S., MCQUADE, R. M.,

DIWAKARLA, S., PUSTOVIT, R., STANISZ, A. M., BIENENSTOCK, J., BARBUT, D., ZASLOFF, M., FURNESS, J. B. & KUNZE, W. A. 2020. Squalamine Restores the Function of the Enteric Nervous System in Mouse Models of Parkinson's Disease. J Parkinsons Dis, 10, 1477-1491. WILSON, A. J., LLEWELLYN-SMITH, I. J., FURNESS, J. B. & COSTA, M. 1987. The source of the nerve fibres forming the deep muscular and circular muscle plexuses in the small intestine of the guinea-pig. Cell Tissue Res, 247, 497-504.

WU, G. 1998. Intestinal mucosal amino acid catabolism. J Nutr, 128, 1249-52. YANG, H., SÖDERHOLM, J., LARSSON, J., PERMERT, J., OLAISON, G., LINDGREN, J. & WIRÉN, M.

1999. Glutamine effects on permeability and ATP content of jejunal mucosa in starved rats. Clin Nutr, 18, 301-6. ZAGHLOOL, S. S., ABO-SEIF, A. A., RABEH, M. A., ABDELMOHSEN, U. R. & MESSIHA, B. A. S. 2019.

Gastro-Protective and Anti-Oxidant Potential of Althaea officinalis and Solanum nigrum on Pyloric Ligation/Indomethacin-Induced Ulceration in Rats. Antioxidants (Basel, Switzerland), 8, 512. ZHANG, F., WAN, H., YANG, X., HE, J., LU, C., YANG, S., TUO, B. & DONG, H. 2019. Molecular

mechanisms of caffeine-mediated intestinal epithelial ion transports. British Journal of Pharmacology, 176, 1700-1716.

ZHANG, Q., WANG, J., LI, A., LIU, H., ZHANG, W., CUI, X. & WANG, K. 2013. Expression of neurexin and neuroligin in the enteric nervous system and their down-regulated expression levels in Hirschsprung disease. Mol Biol Rep, 40, 2969-75.

117

ZIHNI, C., MILLS, C., MATTER, K. & BALDA, M. S. 2016. Tight junctions: from simple barriers to multifunctional molecular gates. Nat Rev Mol Cell Biol, 17, 564-80.

Appendices

Appendix: Ethics approval letter

Appendix: Current clamp protocol used for whole-cell patch clamping.

118

Electrophysiology appendices

Appendix Table 1. Action potential characteristics for Neuron 1

3.28 2.62 2.19 3.03 3.76 2.42 2.84 3.52 3.26 2.99 2.33 3.41

23.26 35.19 38.21 26.58 41.44 41.47 29.44 21.73 42.96 26.89 41.35 22.06

-43.83 -44.55 -48.27 -48.27 -49.39 -46.09 -46.09 -46.09 -46.54 -46.54 -42.85 -42.85

-18.05 -18.49 -13.94 -17.52 -19.54 -19.87 -17.93 -18.77 -17.73 -17.11 -14.91 -18.95

-60.91 -61.58 -63.16 -60.03 -64.45 -64.24 -61.16 -59.54 -63.66 -59.11 -62.16 -56.55

AP number Peak (mV) RMP (mV) Threshold (mV) AH peak (mV) Overshoot (mV) Half width (ms)

56.24 53.68 58.15 44.1 60.98 61.34 47.37 40.55 60.7 44.0 56.26 41.01

Time below RMP (mS) 65.61 46.07 31.61 30.24 31.07 31.3 47.21 29.37 29.96 26.88 28.79 32.3 1 2 3 4 5 6 7 8 9 10 11 12

Appendix Table 2. Action potential characteristics for Neuron 2

119

2.15 2.09 2.18 2.27 2.32

15.23 17.82 18.9 20.23 19.31

-52.11 -50.76 -51.32 -51.36 -49.75

-22.71 -24.04 -27.32 -28.15 -29.05

-53.99 0 -51.15 0 -53.47 0 -47.33 0 -44.77 0

37.94 41.86 46.22 48.38 48.36

AP number Peak (mV) RMP (mV) Threshold (mV) Overshoot (mV) Half width (ms) AH peak (mV) Time below RMP (mS)

Threshold (mV)

1 2 3 4 5

AP number

RMP (mV)

Peak (mV)

Overshoot (mV)

Half width (ms)

AH peak (mV)

Appendix Table 3. Action potential characteristics for Neuron 3 Time below RMP (mS) 61.43 76.3 62.37 59.1 65.8 53.09 67.92 48.4 68.12 47.81 -59.36 40.95 -58.84 40.72 61.43 76.3

-62.29 -62.29 -62.01 -61.28 -60.18 -18.66 -19.28 -62.29

-16.39 -15.62 -17.95 -19.06 -19.08 -43.15 -42.68 -16.39

45.04 46.75 47.85 48.86 49.04 49.04 48.98 45.04

2.72 2.6 2.71 2.71 2.69 2.7 2.76 2.72

1 2 3 4 5 6 7 8

-41.47 -42.1 -41.21 -42.72 -42.21 67.7 68.26 -41.47

Appendix Table 4. Statistical parameters for younger vs older electrophysiology experiment Effect size

compute

95% CI

Value

to

confidence

Method intervals Newcombe/Wilson with CC

Attributable risk (P1- P2)

0.3300

0.2143 to 0.4595

120

Permeability appendices

Appendix Table 5. Simple linear regression calculated gradient values ((mg/ml)/time)) for all paracellular permeability-concentration time curves

Genotype

Intestinal region

Treatment / Condition Fasting Non-fasting

L-

NL3R451C

Wild-type

NL3R451C

Jejunum

Wild-type

Jejunum

NL3R451C

Wild-type

NL3R451C

Duodenum 8.609e-005 Duodenum 9.068e-005 1.046e-004 1.101e-004 6.345e-005 4.389e-005 N/A

Distal Ileum Distal ileum Colon

Wild-type

Colon

N/A

30mM glutamine 7.444e-005 9.252e-005 7.369e-005 7.358e-005 5.896e-005 7.145e-005 1.857e-005 1.466e-005

90mM glutamine 7.604e-005 8.112e-005 7.007e-005 8.459e-005 5.597e-005 4.647e-005 1.956e-005 1.909e-005

10mM Caffeine 4.829e-005 5.671e-005 4.660e-005 7.684e-005 5.384e-005 6.820e-005 1.381e-005 2.533e-005

1.161e-004 5.671e-005 1.046e-004 7.684e-005 9.852e-005 6.820e-005 3.682e-005 2.533e-005

Appendix Table 6. Statistical parameters for duodenal permeability for NL3R451C and WT mice utilising a two-way ANOVA with a Sidak multiple comparisons test.

Šídák's multiple comparisons test Mean Diff. Duodenum (non-fasting) NL3 n=9 - WT n=9 0 minute 1 minute 30 minutes 60 minutes 90 minutes 120 minutes

-0.00027 -0.00027 -0.00019 0.00016 -0.00087 -0.00075

95.00% CI of diff. -0.0048 to 0.0042 -0.0048 to 0.0042 -0.0047 to 0.0043 -0.0044 to 0.0047 -0.0054 to 0.0036 -0.0053 to 0.0038

Summary ns ns ns ns ns ns

Adjusted P Value >0.99 >0.99 >0.99 >0.99 0.99 0.99

Appendix Table 7. Statistical parameters for jejunum permeability for NL3R451C and WT mice utilising a two-way ANOVA with a Sidak multiple comparisons test.

Šídák's multiple comparisons test Mean Diff. Jejunum (non-fasting) NL3 n=12 - WT n=8 0 minute 1 minute 30 minutes 60 minutes 90 minutes 120 minutes

0.00015 0.00030 0.0014 0.0027 0.0038 0.0047

95.00% CI of diff. -0.0064 to 0.0067 -0.0062 to 0.0069 -0.0051 to 0.0080 -0.0038 to 0.0093 -0.0027 to 0.010 -0.0019 to 0.011

Summary Adjusted P Value ns ns ns ns ns ns

>0.99 >0.99 0.99 0.84 0.53 0.30

121

Appendix Table 8. Statistical parameters for distal ileum permeability for NL3R451C and WT mice utilising a two-way ANOVA with a Sidak multiple comparisons test.

Šídák's multiple comparisons test Mean Diff. Distal Ileum (non-fasting) NL3 n=13 - WT n=10 0 minute 1 minute 30 minutes 60 minutes 90 minutes 120 minutes

-2.1e-005 -0.00013 0.00018 0.0014 0.0017 0.0020

95.00% CI of diff. -0.0040 to 0.0040 -0.0041 to 0.0038 -0.0038 to 0.0042 -0.0025 to 0.0054 -0.0022 to 0.0058 -0.0019 to 0.0060

Summary Adjusted P Value ns ns ns ns ns ns

>0.99 >0.99 >0.99 0.90 0.80 0.68

Appendix Table 9. Statistical parameters for duodenal permeability for of NL3R451C and WT fasting mice utilising a two-way ANOVA with a Sidak multiple comparisons test.

Šídák's multiple comparisons test Duodenum (fasting) NL3 n=10 - WT n=8 0 minute 1 minute 30 minutes 60 minutes 90 minutes 120 minutes

Mean Diff. -0.00026 -0.00027 -0.0015 -0.0043 -0.0064 -0.0066

95.00% CI of diff. -0.0028 to 0.0023 -0.0028 to 0.0023 -0.0041 to 0.0010 -0.0070 to -0.0017 -0.0090 to -0.0038 -0.0092 to -0.0040

Summary ns ns ns * **

Adjusted P Value >0.99 0.99 0.52 0.0001 <0.0001 <0.0001

Appendix Table 10. Statistical parameters for jejunum permeability for NL3R451C and WT fasting mice utilising a two-way ANOVA with a Sidak multiple comparisons test.

Šídák's multiple comparisons test Jejunum (fasting) NL3 n=10 - WT n=11 0 minute 1 minute 30 minutes 60 minutes 90 minutes 120 minutes

Mean Diff. -0.00012 -0.00010 -0.00053 -0.0016 -0.0026 -0.0033

95.00% CI of diff. -0.0027 to 0.0024 -0.0026 to 0.0024 -0.0031 to 0.0020 -0.0042 to 0.00091 -0.0052 to -7.4e-005 -0.0058 - to -0.00071

Summary ns ns ns ns * **

Adjusted P Value >0.99 >0.99 0.99 0.41 0.040 0.0052

Appendix Table 11. Statistical parameters for distal ileum permeability for NL3R451C and WT fasting mice utilising a two-way ANOVA with a Sidak multiple comparisons test.

Šídák's multiple comparisons test Distal ileum (fasting) NL3 n=9 - WT n=8 0 minute 1 minute 30 minutes 60 minutes 90 minutes 120 minutes

Mean Diff. -0.00023 -0.00024 -0.0017 -0.0024 -0.0032 -0.0038

95.00% CI of diff. -0.0025 to 0.0020 -0.0025 to 0.0020 -0.0040 to 0.00061 -0.0048 to -0.00016 -0.0055 to -0.00092 -0.0061 to -0.0014

Summary ns ns ns * ** ***

Adjusted P Value >0.99 0.99 0.26 0.02 0.0018 0.00020

122

Appendix Table 12. Statistical parameters for colon permeability for NL3R451C and WT fasting mice utilising a two-way ANOVA with a Sidak multiple comparisons test.

Šídák's multiple comparisons test Colon (fasting) NL3 n=9 - WT n=8 0 minute 1 minute 30 minutes 60 minutes 90 minutes 120 minutes

Mean Diff. -3.6e-005 -6.0e-005 0.00010 -0.00041 -0.00090 -0.0014

95.00% CI of diff. -0.0013 to 0.0013 -0.0014 to 0.0012 -0.0012 to 0.0014 -0.0017 to 0.00092 -0.0022 to 0.00043 -0.0027 to -8.2e-005

Summary ns ns ns ns ns *

Adjusted P Value >0.99 >0.99 >0.99 0.95 0.36 0.03

Appendix Table 13. Statistical parameters for duodenal permeability for NL3R451C non-fasting and fasting mice utilising a two-way ANOVA with a Sidak multiple comparisons test.

Šídák's multiple comparisons test Duodenum NL3 n=10 (Fasting) - NL3 n=9 (non-fasting) 0 minute 1 minute 30 minutes 60 minutes 90 minutes 120 minutes

Mean Diff. 0.00035 0.00019 0.00062 0.0020 0.0036 0.0033

95.00% CI of diff. -0.0037 to 0.0044 -0.0039to 0.0043 -0.0035 to 0.0047 -0.0021 to 0.0061 -0.00044 to 0.0078 -0.00083 to 0.0074

Summary ns ns ns ns ns ns

Adjusted P Value >0.99 >0.99 0.99 0.71 0.11 0.18

Appendix Table 14. Statistical parameters for jejunum permeability for NL3R451C non-fasting and fasting mice utilising a two-way ANOVA with a Sidak multiple comparisons test.

Šídák's multiple comparisons test Jejunum NL3 n=10 (Fasting) - NL3 n=12 (non-fasting) 0 minute 1 minute 30 minutes 60 minutes 90 minutes 120 minutes

Mean Diff. 0.00043 0.00040 0.00086 0.0023 0.0035 0.0041

95.00% CI of diff. -0.0031 to 0.0040 -0.0031 to 0.0040 -0.0027 to 0.0044 -0.0012 to 0.0059 -8.1e-005 to 0.0071 0.00051 to 0.0077

Summary ns ns ns ns ns *

Adjusted P Value 0.99 0.99 0.98 0.42 0.06 0.02

Appendix Table 15. Statistical parameters for distal ileum permeability for NL3R451C non-fasting and fasting mice utilising a two-way ANOVA with a Sidak multiple comparisons test.

Šídák's multiple comparisons test Distal ileum NL3 n=9 (Fasting) - NL3 n=13 (non-fasting) 0 minute 1 minute 30 minutes 60 minutes 90 minutes 120 minutes

Mean Diff. 0.00042 0.00039 0.0025 0.0032 0.0040 0.0044

95.00% CI of diff. -0.0036 to 0.0045 -0.0037 to 0.0044 -0.0015 to 0.0066 -0.00085 to 0.0073 3.8e-006 to 0.0081 0.00039 to 0.0085

Summary ns ns ns ns * *

Adjusted P Value 0.99 >0.99 0.47 0.19 0.040 0.02

123

Appendix Table 16. Statistical parameters for duodenal permeability for WT non-fasting and fasting mice utilising a two-way ANOVA with a Sidak multiple comparisons test.

Šídák's multiple comparisons test Duodenum WT n=8 (Fasting) - WT n=9 (non-fasting) 0 minute 1 minute 30 minutes 60 minutes 90 minutes 120 minutes

Mean Diff. -0.00017 -0.00035 -0.0011 -0.0022 -0.0036 -0.0041

95.00% CI of diff. -0.0032 to 0.0029 -0.0034 to 0.0027 -0.0041 to 0.0019 -0.0052 to 0.00089 -0.0067 to -0.00054 -0.0071 to -0.0010

Summary ns ns ns ns * **

Adjusted P Value >0.99 0.99 0.91 0.30 0.012 0.0035

Appendix Table 17. Statistical parameters for jejunum permeability for WT non-fasting and fasting mice utilising a two-way ANOVA with a Sidak multiple comparisons test.

Šídák's multiple comparisons test Jejunum WT n=11 (Fasting) - WT n=8 (non-fasting) 0 minute 1 minute 30 minutes 60 minutes 90 minutes 120 minutes

Mean Diff. 0.00014 -1.0e-005 -0.0011 -0.0020 -0.0029 -0.0038

95.00% CI of diff. -0.0060 to 0.0063 -0.0062 to 0.0061 -0.0073 to 0.0050 -0.0082 to 0.0041 -0.0091 to 0.0031 -0.010 to 0.0023

Summary ns ns ns ns ns ns

Adjusted P Value >0.99 >0.99 0.99 0.93 0.73 0.45

Appendix Table 18. Statistical parameters for distal ileum permeability for WT non-fasting and fasting mice utilising a two-way ANOVA with a Sidak multiple comparisons test.

Šídák's multiple comparisons test Distal ileum WT n=8 (Fasting) - WT n=10 (non-fasting) 0 minute 1 minute 30 minutes 60 minutes 90 minutes 120 minutes

Mean Diff. 0.00017 1.0e-005 0.0010 0.0022 0.0026 0.0027

95.00% CI of diff. -0.0024 to 0.0027 -0.0025 to 0.0026 -0.0015 to 0.0036 -0.00037 to 0.0048 4.3e-005 to 0.0052 0.00014 to 0.0053

Summary Adjusted P Value ns ns ns ns * *

>0.99 >0.99 0.88 0.13 0.044 0.033

124

Appendix Table 19. Statistical parameters for duodenal permeability for fasting NL3R451C mice versus

fasting NL3R451C mice treated with L-glutamine utilising a two-way ANOVA with a Tukeys multiple

comparisons test. (Green fill represents glutamine restoring NL3 FITC concentrations to WT

concentrations).

Tukey's multiple comparisons test Duodenum 0 minutes NL3 + 30mM Glutamine n=5 vs. NL3 n=10 NL3 + 30mM Glutamine n=5 vs. NL3 + 90mM Glutamine n=7 NL3 + 30mM Glutamine n=5 vs. WT n=8 NL3 n=10 vs. NL3 + 90mM Glutamine n=7 NL3 n=10 vs. WT n=8 NL3 + 90mM Glutamine n=7 vs. WT n=8 1 minute NL3 + 30mM Glutamine n=5 vs. NL3 n=10 NL3 + 30mM Glutamine n=5 vs. NL3 + 90mM Glutamine n=7 NL3 + 30mM Glutamine n=5 vs. WT n=8 NL3 n=10 vs. NL3 + 90mM Glutamine n=7 NL3 n=10 vs. WT n=8 NL3 + 90mM Glutamine n=7 vs. WT n=8 30 minutes NL3 + 30mM Glutamine n=5 vs. NL3 n=10 NL3 + 30mM Glutamine n=5 vs. NL3 + 90mM Glutamine n=7 NL3 + 30mM Glutamine n=5 vs. WT n=8 NL3 n=10 vs. NL3 + 90mM Glutamine n=7 NL3 n=10 vs. WT n=8 NL3 + 90mM Glutamine n=7 vs. WT n=8 60 minutes NL3 + 30mM Glutamine n=5 vs. NL3 n=10 NL3 + 30mM Glutamine n=5 vs. NL3 + 90mM Glutamine n=7 NL3 + 30mM Glutamine n=5 vs. WT n=8 NL3 n=10 vs. NL3 + 90mM Glutamine n=7 NL3 n=10 vs. WT n=8 NL3 + 90mM Glutamine n=7 vs. WT n=8 90 minutes NL3 + 30mM Glutamine n=5 vs. NL3 n=10 NL3 + 30mM Glutamine n=5 vs. NL3 + 90mM Glutamine n=7 NL3 + 30mM Glutamine n=5 vs. WT n=8 NL3 n=10 vs. NL3 + 90mM Glutamine n=7 NL3 n=10 vs. WT n=8 NL3 + 90mM Glutamine n=7 vs. WT n=8 120 minutes NL3 + 30mM Glutamine n=5 vs. NL3 n=10 NL3 + 30mM Glutamine n=5 vs. NL3 + 90mM Glutamine n=7 NL3 + 30mM Glutamine n=5 vs. WT n=8 NL3 n=10 vs. NL3 + 90mM Glutamine n=7 NL3 n=10 vs. WT n=8 NL3 + 90mM Glutamine n=7 vs. WT n=8

Mean difference -0.00021 -4.8e-006 4.4e-005 0.00021 0.00026 4.9e-005 -0.00022 -2.6e-006 5.0e-005 0.00021 0.00027 5.3e-005 -0.00096 0.00059 0.00057 0.0015 0.0015 -2.4e-005 -0.0029 0.00059 0.0014 0.0035 0.0043 0.00081 -0.0046 -0.00059 0.0017 0.0040 0.0064 0.0023 -0.0046 0.00014 0.0020 0.0047 0.0066 0.0018

95% CI of diff. -0.0031 to 0.0027 -0.0031 to 0.0031 -0.0030 to 0.0031 -0.0024 to 0.0028 -0.0022 to 0.0028 -0.0027 to 0.0028 -0.003162 to 0.0027 -0.003148 to 0.0031 -0.003012 to 0.0031 -0.002430 to 0.0028 -0.002277 to 0.0028 -0.002727 to 0.0028 -0.0039 to 0.0019 -0.0025 to 0.0037 -0.0024 to 0.0036 -0.0010 to 0.0042 -0.0010 to 0.0040 -0.0028 to 0.0027 -0.0059 to -3.8e-005 -0.0025 to 0.0037 -0.0016 to 0.0044 0.00092 to 0.0062 0.0018 to 0.0069 -0.0019 to 0.0035 -0.0076 to -0.0017 -0.0037 to 0.0025 -0.0012 to 0.0048 0.0014 to 0.0067 0.0039 to 0.0089 -0.00040 to 0.0051 -0.0075 to -0.0016 -0.0029 to 0.0032 -0.0010 to 0.0050 0.0021 to 0.0074 0.0040 to 0.0091 -0.00089 to 0.0046

Summary ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns * ns ns ns * ns ns * ns * ns ns ns

Adjusted P Value 0.99 >0.99 >0.99 0.99 0.99 >0.99 0.99 >0.99 >0.99 0.99 0.99 >0.99 0.83 0.96 0.96 0.42 0.40 >0.99 0.040 0.96 0.62 0.00 <0.0001 0.87 0.0004 0.96 0.43 0.0006 <0.0001 0.12 0.0004 0.99 0.31 <0.0001 <0.0001 0.29

125

Appendix Table 20. Statistical parameters for jejunum permeability for fasting NL3R451C mice versus fasting NL3R451C mice treated with L-glutamine utilising a two-way ANOVA with a Tukeys multiple comparisons test. (Green fill represents glutamine restoring NL3 FITC concentrations to WT concentrations).

Tukey's multiple comparisons test Jejunum 0 minutes NL3 + 30mM Glutamine n=6 vs. NL3 n=9 NL3 + 30mM Glutamine n=6 vs. NL3 + 90mM Glutamine n=7 NL3 + 30mM Glutamine n=6 vs. WT n=11 NL3 n=9 vs. NL3 + 90mM Glutamine n=7 NL3 n=9 vs. WT n=11 NL3 + 90mM Glutamine n=7 vs. WT n=11 1 minute NL3 + 30mM Glutamine n=6 vs. NL3 n=9 NL3 + 30mM Glutamine n=6 vs. NL3 + 90mM Glutamine n=7 NL3 + 30mM Glutamine n=6 vs. WT n=11 NL3 n=9 vs. NL3 + 90mM Glutamine n=7 NL3 n=9 vs. WT n=11 NL3 + 90mM Glutamine n=7 vs. WT n=11 30 minutes NL3 + 30mM Glutamine n=6 vs. NL3 n=9 NL3 + 30mM Glutamine n=6 vs. NL3 + 90mM Glutamine n=7 NL3 + 30mM Glutamine n=6 vs. WT n=11 NL3 n=9 vs. NL3 + 90mM Glutamine n=7 NL3 n=9 vs. WT n=11 NL3 + 90mM Glutamine n=7 vs. WT n=11 60 minutes NL3 + 30mM Glutamine n=6 vs. NL3 n=9 NL3 + 30mM Glutamine n=6 vs. NL3 + 90mM Glutamine n=7 NL3 + 30mM Glutamine n=6 vs. WT n=11 NL3 n=9 vs. NL3 + 90mM Glutamine n=7 NL3 n=9 vs. WT n=11 NL3 + 90mM Glutamine n=7 vs. WT n=11 90 minutes NL3 + 30mM Glutamine n=6 vs. NL3 n=9 NL3 + 30mM Glutamine n=6 vs. NL3 + 90mM Glutamine n=7 NL3 + 30mM Glutamine n=6 vs. WT n=11 NL3 n=9 vs. NL3 + 90mM Glutamine n=7 NL3 n=9 vs. WT n=11 NL3 + 90mM Glutamine n=7 vs. WT n=11 120 minutes NL3 + 30mM Glutamine n=6 vs. NL3 n=9 NL3 + 30mM Glutamine n=6 vs. NL3 + 90mM Glutamine n=7 NL3 + 30mM Glutamine n=6 vs. WT n=11 NL3 n=9 vs. NL3 + 90mM Glutamine n=7 NL3 n=9 vs. WT n=11 NL3 + 90mM Glutamine n=7 vs. WT n=11

95% CI of diff. -0.0031 to 0.0025 -0.0031 to 0.0028 -0.0028 to 0.0026 -0.0025 to 0.0028 -0.0022 to 0.0025 -0.0026 to 0.0026 -0.0030 to 0.0026 -0.0030 to 0.0029 -0.0028 to 0.0026 -0.0025 to 0.0028 -0.0023 to 0.0025 -0.0026 to 0.0026 -0.0033 to 0.0023 -0.0024 to 0.0036 -0.0025 to 0.0029 -0.0016 to 0.0038 -0.0017 to 0.0031 -0.0030 to 0.0021 -0.0048 to 0.00090 -0.0022 to 0.0038 -0.0027 to 0.0027 3.3e-005 to 0.0055 -0.00049 to 0.0043 -0.0034 to 0.0018 -0.0059 to -0.00018 -0.0023 to 0.0036 -0.0029 to 0.0025 0.00098 to 0.0064 0.00040 to 0.0052 -0.0034 to 0.0017 -0.0067 to -0.0010 -0.0028 to 0.0031 -0.0032 to 0.0022 0.0013 to 0.0068 0.00097 to 0.0058 -0.0032 to 0.0019

Summary ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns * ns ns * ns ns ** * ns ns ns * ** ns

Adjusted P Value 0.99 0.99 0.99 0.99 0.99 >0.99 0.99 >0.99 0.99 0.99 0.99 >0.99 0.96 0.95 0.99 0.71 0.88 0.97 0.2917 0.8958 >0.9999 0.0460 0.1662 0.8494 0.031 0.939 0.99 0.0030 0.015 0.82 0.0026 0.99 0.96 0.0009 0.0021 0.91

Mean difference -0.00027 -0.00013 -0.00013 0.00013 0.00014 3.6e-006 -0.00021 -7.0e-005 -9.0e-005 0.00014 0.00012 -2.0e-005 -0.00050 0.00061 0.00017 0.0011 0.00068 -0.00043 -0.0019 0.00081 -1.8e-006 0.0027 0.0019 -0.00081 -0.0030 0.00066 -0.00020 0.0037 0.0028 -0.00087 -0.0039 0.00015 -0.00052 0.0040 0.0034 -0.00067

126

Appendix Table 21. Statistical parameters for distal ileum permeability for fasting NL3R451C mice versus fasting NL3R451C mice treated with L-glutamine utilising a two-way ANOVA with a Tukeys multiple comparisons test. (Green fill represents glutamine restoring NL3 FITC concentrations to WT concentrations).

Tukey's multiple comparisons test Distal ileum 0 minutes NL3 + 30mM Glutamine n=5 vs. NL3 n=9 NL3 + 30mM Glutamine n=5 vs. NL3 + 90mM Glutamine n=7 NL3 + 30mM Glutamine n=5 vs. WT n=8 NL3 n=9 vs. NL3 + 90mM Glutamine n=7 NL3 n=9 vs. WT n=8 NL3 + 90mM Glutamine n=7 vs. WT n=8 1 minutes NL3 + 30mM Glutamine n=5 vs. NL3 n=9 NL3 + 30mM Glutamine n=5 vs. NL3 + 90mM Glutamine n=7 NL3 + 30mM Glutamine n=5 vs. WT n=8 NL3 n=9 vs. NL3 + 90mM Glutamine n=7 NL3 n=9 vs. WT n=8 NL3 + 90mM Glutamine n=7 vs. WT n=8 30 minutes NL3 + 30mM Glutamine n=5 vs. NL3 n=9 NL3 + 30mM Glutamine n=5 vs. NL3 + 90mM Glutamine n=7 NL3 + 30mM Glutamine n=5 vs. WT n=8 NL3 n=9 vs. NL3 + 90mM Glutamine n=7 NL3 n=9 vs. WT n=8 NL3 + 90mM Glutamine n=7 vs. WT n=8 60 minutes

Summary ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns

Adjusted Value 0.99 >0.99 0.99 0.99 0.99 >0.99 0.99 >0.99 >0.99 0.99 0.99 >0.99 0.88 0.99 0.93 0.78 0.42 0.95

Mean difference -0.00035 -7.2e-005 -0.00012 0.00028 0.00023 -5.1e-005 -0.00031 -2.0e-005 -6.5e-005 0.00029 0.00024 -4.4e-005 -0.00093 0.00015 0.00076 0.0010 0.0017 0.00061

NL3 + 30mM Glutamine n=5 vs. NL3 n=9 NL3 + 30mM Glutamine n=5 vs. NL3 + 90mM Glutamine n=7 NL3 + 30mM Glutamine n=5 vs. WT n=8

ns ns ns

0.23 0.98 >0.99

-0.0024 0.00045 5.3e-005

NL3 n=9 vs. NL3 + 90mM Glutamine n=7

0.06

0.0028

NL3 n=9 vs. WT n=8 NL3 + 90mM Glutamine n=7 vs. WT n=8 90 minutes

ns ns

0.12 0.98

0.0024 -0.00040

to

-

NL3 + 30mM Glutamine n=5 vs. NL3 n=9 NL3 + 30mM Glutamine n=5 vs. NL3 + 90mM Glutamine n=7 NL3 + 30mM Glutamine n=5 vs. WT n=8 NL3 n=9 vs. NL3 + 90mM Glutamine n=7 NL3 n=9 vs. WT n=8 NL3 + 90mM Glutamine n=7 vs. WT n=8 120 minutes NL3 + 30mM Glutamine n=5 vs. NL3 n=9 NL3 + 30mM Glutamine n=5 vs. NL3 + 90mM Glutamine n=7 NL3 + 30mM Glutamine n=5 vs. WT n=8 NL3 n=9 vs. NL3 + 90mM Glutamine n=7 NL3 n=9 vs. WT n=8 NL3 + 90mM Glutamine n=7 vs. WT n=8

95% CI of diff. -0.0037 to 0.0029 -0.0035 to 0.0034 -0.0035 to 0.0032 -0.0027 to 0.0033 -0.0026 to 0.0031 -0.0031 to 0.0030 -0.0036 to 0.0030 -0.0035 to 0.0034 -0.0034 to 0.0033 -0.0027 to 0.0033 -0.0026 to 0.0031 -0.0031 to 0.0030 -0.0042 to 0.0024 -0.0033 to 0.0036 -0.0026 to 0.0041 -0.0019 to 0.0041 -0.0012 to 0.0046 -0.0024 to 0.0037 -0.0057 0.00091 -0.0030 to 0.0039 -0.0033 to 0.0034 -0.00013 to 0.0059 -0.00043 0.0054 -0.0035 to 0.0027 -0.0072 0.00076 -0.0033 to 0.0036 -0.0042 to 0.0025 0.0012 to 0.0072 0.00032 to 0.0061 -0.0041 to 0.0021 -0.0081 to -0.0014 -0.0031 to 0.0038 -0.0044 to 0.0024 0.0021 to 0.0081 0.00088 to 0.0067 -0.0044 to 0.0017

ns ns * ns ns ns * ** ns

0.0092 0.99 0.91 0.0021 0.02 0.83 0.0016 0.99 0.87 0.0001 0.0050 0.67

-0.0041 0.00012 -0.00087 0.0042 0.0032 -0.00099 -0.0047 0.00035 -0.00099 0.0051 0.0038 -0.0013

127

Appendix Table 22. Statistical parameters for colon permeability for fasting NL3R451C mice versus fasting NL3R451C mice treated with L-glutamine utilising a two-way ANOVA with a Tukeys multiple comparisons test. (Green fill represents glutamine restoring NL3 FITC concentrations to WT concentrations).

95% CI of diff. -0.0018 to 0.0018 -0.0019 to 0.0020 -0.0018 to 0.0019 -0.0012 to 0.0013 -0.0011 to 0.0011 -0.0013 to 0.0013 -0.0018 to 0.0018 -0.0018 to 0.0020 -0.0018 to 0.0019 -0.0012 to 0.0014 -0.0010 to 0.0012 -0.0013 to 0.0013 -0.0022 to 0.0014 -0.0019 to 0.0019 -0.0023 to 0.0013 -0.00092 to 0.0017 -0.0012 to 0.0010 -0.0018 to 0.00085 -0.0029 to 0.00077 -0.0020 to 0.0018 -0.0025 to 0.0012 -0.00035 to 0.0022 -0.00073 to 0.0015 -0.0018 to 0.00080 -0.0034 to 0.00025 -0.0020 to 0.0018 -0.0025 to 0.0011 0.00016 to 0.0027 -0.00024 to 0.0020 -0.0019 to 0.00077 -0.0040 to -0.00032 -0.0019 to 0.0020 -0.0026 to 0.0011 0.00088 to 0.0035 0.00027 to 0.0025 -0.0021 to 0.00056

Summary ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns * ns ns * ns ns * ns

Adjusted P Value >0.99 0.99 >0.99 0.99 0.99 0.99 >0.99 0.99 0.99 0.99 0.99 0.99 0.94 >0.99 0.89 0.86 0.99 0.77 0.43 0.99 0.79 0.23 0.77 0.71 0.11 0.99 0.77 0.021 0.17 0.68 0.014 >0.99 0.72 0.0002 0.0087 0.43

Tukey's multiple comparisons test Colon 0 minutes NL3 + 30mM Glutamine n=2 vs. NL3 n=9 NL3 + 30mM Glutamine n=2 vs. NL3 + 90mM Glutamine n=5 NL3 + 30mM Glutamine n=2 vs. WT n=8 NL3 n=9 vs. NL3 + 90mM Glutamine n=5 NL3 n=9 vs. WT n=8 NL3 + 90mM Glutamine n=5 vs. WT n=8 1 minute NL3 + 30mM Glutamine n=2 vs. NL3 n=9 NL3 + 30mM Glutamine n=2 vs. NL3 + 90mM Glutamine n=5 NL3 + 30mM Glutamine n=2 vs. WT n=8 NL3 n=9 vs. NL3 + 90mM Glutamine n=5 NL3 n=9 vs. WT n=8 NL3 + 90mM Glutamine n=5 vs. WT n=8 30 minutes NL3 + 30mM Glutamine n=2 vs. NL3 n=9 NL3 + 30mM Glutamine n=2 vs. NL3 + 90mM Glutamine n=5 NL3 + 30mM Glutamine n=2 vs. WT n=8 NL3 n=9 vs. NL3 + 90mM Glutamine n=5 NL3 n=9 vs. WT n=8 NL3 + 90mM Glutamine n=5 vs. WT n=8 60 minutes NL3 + 30mM Glutamine n=2 vs. NL3 n=9 NL3 + 30mM Glutamine n=2 vs. NL3 + 90mM Glutamine n=5 NL3 + 30mM Glutamine n=2 vs. WT n=8 NL3 n=9 vs. NL3 + 90mM Glutamine n=5 NL3 n=9 vs. WT n=8 NL3 + 90mM Glutamine n=5 vs. WT n=8 90 minutes NL3 + 30mM Glutamine n=2 vs. NL3 n=9 NL3 + 30mM Glutamine n=2 vs. NL3 + 90mM Glutamine n=5 NL3 + 30mM Glutamine n=2 vs. WT n=8 NL3 n=9 vs. NL3 + 90mM Glutamine n=5 NL3 n=9 vs. WT n=8 NL3 + 90mM Glutamine n=5 vs. WT n=8 120 minutes NL3 + 30mM Glutamine n=2 vs. NL3 n=9 NL3 + 30mM Glutamine n=2 vs. NL3 + 90mM Glutamine n=5 NL3 + 30mM Glutamine n=2 vs. WT n=8 NL3 n=9 vs. NL3 + 90mM Glutamine n=5 NL3 n=9 vs. WT n=8 NL3 + 90mM Glutamine n=5 vs. WT n=8

Mean difference -5.6e-006 7.2e-005 3.1e-005 7.8e-005 3.6e-005 -4.3e-005 -8.3e-006 9.3e-005 5.2e-005 0.00010 6.0e-005 -4.0e-005 -0.00039 -7.4e-006 -0.00049 0.00039 -0.00010 -0.00049 -0.0010 -0.00011 -0.00065 0.00096 0.00041 -0.00054 -0.0015 -0.00011 -0.00068 0.0015 0.00090 -0.00057 -0.0021 2.7e-005 -0.00075 0.0022 0.0014 -0.00077

128

Appendix Table 23. Statistical parameters for duodenal permeability for fasting WT mice versus fasting WT mice treated with L-glutamine utilising a two-way ANOVA with a Sidak multiple comparisons test.

Sidak multiple comparisons test Duodenum 0 minutes WT + 90mM Glutamine n=5 vs. WT n=8 WT + 90mM Glutamine n=5 vs. WT + 30mM Glutamine n=4 WT n=8 vs. WT + 30mM Glutamine n=4 1 minute WT + 90mM Glutamine n=5 vs. WT n=8 WT + 90mM Glutamine n=5 vs. WT + 30mM Glutamine n=4 WT n=8 vs. WT + 30mM Glutamine n=4 30 minutes WT + 90mM Glutamine n=5 vs. WT n=8 WT + 90mM Glutamine n=5 vs. WT + 30mM Glutamine n=4 WT n=8 vs. WT + 30mM Glutamine n=4 60 minutes WT + 90mM Glutamine n=5 vs. WT n=8 WT + 90mM Glutamine n=5 vs. WT + 30mM Glutamine n=4 WT n=8 vs. WT + 30mM Glutamine n=4 90 minutes WT + 90mM Glutamine n=5 vs. WT n=8 WT + 90mM Glutamine n=5 vs. WT + 30mM Glutamine n=4 WT n=8 vs. WT + 30mM Glutamine n=4 120 minutes WT + 90mM Glutamine n=5 vs. WT n=8 WT + 90mM Glutamine n=5 vs. WT + 30mM Glutamine n=4 WT n=8 vs. WT + 30mM Glutamine n=4

Mean difference 0.00025 0.00015 -0.00010 0.00021 0.00018 -3.1e-005 0.00087 -0.0013 -0.0022 0.0012 -0.0027 -0.0039 0.0029 -0.0015 -0.0044 0.0028 -0.00083 -0.0037

95% CI of diff. -0.0027 to 0.0032 -0.0033 to 0.0036 -0.0032 to 0.0030 -0.0027 to 0.0031 -0.0033 to 0.0036 -0.0032 to 0.0031 -0.0020 to 0.0038 -0.0048 to 0.0021 -0.0054 to 0.00097 -0.0016 to 0.0042 -0.0061 to 0.00078 -0.0071 to -0.00080 -2.7e-005 to 0.0059 -0.0050 to 0.0019 -0.0076 to -0.0012 -7.5e-005 to 0.0058 -0.0043 to 0.0026 -0.0069 to -0.00054

Summary Adjusted P Value ns ns ns ns ns ns ns ns ns ns ns * ns ns ** ns ns *

0.97 0.99 0.99 0.98 0.99 0.99 0.76 0.63 0.22 0.55 0.16 0.010 0.05 0.55 0.0035 0.057 0.83 0.017

129

Appendix Table 24. Statistical parameters for jejunum permeability for fasting WT mice versus fasting WT mice treated with L-glutamine utilising a two-way ANOVA with a Sidak multiple comparisons test.

Sidak multiple comparisons test Jejunum 0 minutes WT + 90mM Glutamine n=5 vs. WT n=11 WT + 90mM Glutamine n=5 vs. WT + 30mM Glutamine n=4 WT n=11 vs. WT + 30mM Glutamine n=4 1 minute WT + 90mM Glutamine n=5 vs. WT n=11 WT + 90mM Glutamine n=5 vs. WT + 30mM Glutamine n=4 WT n=11 vs. WT + 30mM Glutamine n=4 30 minutes WT + 90mM Glutamine n=5 vs. WT n=11 WT + 90mM Glutamine n=5 vs. WT + 30mM Glutamine n=4 WT n=11 vs. WT + 30mM Glutamine n=4 60 minutes WT + 90mM Glutamine n=5 vs. WT n=11 WT + 90mM Glutamine n=5 vs. WT + 30mM Glutamine n=4 WT n=11 vs. WT + 30mM Glutamine n=4 90 minutes WT + 90mM Glutamine n=5 vs. WT n=11 WT + 90mM Glutamine n=5 vs. WT + 30mM Glutamine n=4 WT n=11 vs. WT + 30mM Glutamine n=4 120 minutes WT + 90mM Glutamine n=5 vs. WT n=11 WT + 90mM Glutamine n=5 vs. WT + 30mM Glutamine n=4 WT n=11 vs. WT + 30mM Glutamine n=4

Mean difference 2.0e-005 0.00010 8.1e-005 -3.6e-005 0.00014 0.00018 -0.00018 -0.00014 3.4e-005 0.00022 0.00064 0.00041 0.00048 0.00091 0.00043 0.00094 0.0013 0.00044

95% CI of diff. -0.0027 to 0.0027 -0.0033 to 0.0035 -0.0028 to 0.0030 -0.0027 to 0.0027 -0.0032 to 0.0035 -0.0027 to 0.0031 -0.0029 to 0.0025 -0.0035 to 0.0032 -0.0029 to 0.0030 -0.0025 to 0.0029 -0.0027 to 0.0040 -0.0025 to 0.0033 -0.0022 to 0.0032 -0.0024 to 0.0043 -0.0025 to 0.0034 -0.0017 to 0.0036 -0.0020 to 0.0048 -0.0025 to 0.0034

Summary ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns

Adjusted P Value 0.99 0.99 0.99 0.99 0.99 0.98 0.98 0.99 0.99 0.97 0.89 0.94 0.90 0.79 0.93 0.68 0.59 0.93

130

Appendix Table 25. Statistical parameters for distal ileum permeability for fasting WT mice versus fasting WT mice treated with L-glutamine utilising a two-way ANOVA with a Sidak multiple comparisons test.

Sidak multiple comparisons test Distal ileum 0 minutes WT + 90mM Glutamine n=5 vs. WT n=8 WT + 90mM Glutamine n=5 vs. WT + 30mM Glutamine n=4 WT n=8 vs. WT + 30mM Glutamine n=4 1 minute WT + 90mM Glutamine n=5 vs. WT n=8 WT + 90mM Glutamine n=5 vs. WT + 30mM Glutamine n=4 WT n=8 vs. WT + 30mM Glutamine n=4 30 minutes WT + 90mM Glutamine n=5 vs. WT n=8 WT + 90mM Glutamine n=5 vs. WT + 30mM Glutamine n=4 WT n=8 vs. WT + 30mM Glutamine n=4 60 minutes WT + 90mM Glutamine n=5 vs. WT n=8 WT + 90mM Glutamine n=5 vs. WT + 30mM Glutamine n=4 WT n=8 vs. WT + 30mM Glutamine n=4 90 minutes WT + 90mM Glutamine n=5 vs. WT n=8 WT + 90mM Glutamine n=5 vs. WT + 30mM Glutamine n=4 WT n=8 vs. WT + 30mM Glutamine n=4 120 minutes WT + 90mM Glutamine n=5 vs. WT n=8 WT + 90mM Glutamine n=5 vs. WT + 30mM Glutamine n=4 WT n=8 vs. WT + 30mM Glutamine n=4

Mean difference -0.00018 -0.00020 -1.2e-005 -0.00019 -0.00013 6.1e-005 -0.00037 -0.0015 -0.0011 -0.0010 -0.0026 -0.0015 -0.0014 -0.0028 -0.0014 -0.0022 -0.0029 -0.00067

95% CI of diff. -0.0027 to 0.0024 -0.0032 to 0.0028 -0.0028 to 0.0027 -0.0027 to 0.0024 -0.0031 to 0.0029 -0.0027 to 0.0028 -0.0029 to 0.0022 -0.0046 to 0.0015 -0.0039 to 0.0016 -0.0037 to 0.0015 -0.0056 to 0.00045 -0.0043 to 0.0012 -0.0040 to 0.0011 -0.0059 to 0.00017 -0.0042 to 0.0013 -0.0048 to 0.00034 -0.0059 to 0.00013 -0.0034 to 0.0021

Summary Adjusted P Value ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns

0.98 0.98 >0.99 0.98 0.99 0.99 0.93 0.45 0.57 0.57 0.11 0.40 0.38 0.068 0.44 0.10 0.064 0.83

131

Appendix Table 26. Statistical parameters for colon permeability for fasting WT mice versus fasting WT mice treated with L-glutamine utilising a two-way ANOVA with a Sidak multiple comparisons test.

Sidak multiple comparisons test Colon 0 minutes WT + 90 mM Glutamine n=5 vs. WT n=9 WT + 90 mM Glutamine n=5 vs. WT + 30mM Glutamine n=3 WT n=9 vs. WT + 30mM Glutamine n=3 1 minute WT + 90 mM Glutamine n=5 vs. WT n=9 WT + 90 mM Glutamine n=5 vs. WT + 30mM Glutamine n=3 WT n=9 vs. WT + 30mM Glutamine n=3 30 minutes WT + 90 mM Glutamine n=5 vs. WT n=9 WT + 90 mM Glutamine n=5 vs. WT + 30mM Glutamine n=3 WT n=9 vs. WT + 30mM Glutamine n=3 60 minutes WT + 90 mM Glutamine n=5 vs. WT n=9 WT + 90 mM Glutamine n=5 vs. WT + 30mM Glutamine n=3 WT n=9 vs. WT + 30mM Glutamine n=3 90 minutes WT + 90 mM Glutamine n=5 vs. WT n=9 WT + 90 mM Glutamine n=5 vs. WT + 30mM Glutamine n=3 WT n=9 vs. WT + 30mM Glutamine n=3 120 minutes WT + 90 mM Glutamine n=5 vs. WT n=9 WT + 90 mM Glutamine n=5 vs. WT + 30mM Glutamine n=3 WT n=9 vs. WT + 30mM Glutamine n=3

Mean difference -2.4e-005 9.2e-005 0.00011 -2.6e-005 0.00013 0.00016 -0.00055 0.00022 0.00077 -0.00060 0.00035 0.00096 -0.00074 0.00044 0.0011 -0.00075 0.00067 0.0014

95% CI of diff. -0.0011 to 0.0011 -0.0013 to 0.0015 -0.0012 to 0.0014 -0.0011 to 0.0011 -0.0013 to 0.0015 -0.0011 to 0.0015 -0.0016 to 0.00058 -0.0012 to 0.0016 -0.00056 to 0.0021 -0.0017 to 0.00052 -0.0011 to 0.0018 -0.00038 to 0.0023 -0.0018 to 0.00038 -0.0010 to 0.0018 -0.00015 to 0.0025 -0.0018 to 0.00038 -0.00077 to 0.0021 8.1e-005 to 0.0027

Summary ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns *

Adjusted P Value 0.99 0.98 0.97 0.99 0.97 0.95 0.47 0.92 0.35 0.41 0.82 0.20 0.26 0.74 0.092 0.25 0.50 0.035

132

Appendix Table 27. Statistical parameters for duodenal permeability for fasting NL3 mice versus fasting NL3 mice treated with caffeine utilising a two-way ANOVA with a Sidak multiple comparisons test. (Green fill represents caffeine restoring NL3 FITC concentrations to WT concentrations).

Sidak multiple comparisons test Duodenum 0 minutes NL3 (10mM caffeine) n=5 vs. NL3 n=10 NL3 (10mM caffeine) n=5 vs. WT n=8 NL3 n=10 vs. WT n=8 1 minute NL3 (10mM caffeine) n=5 vs. NL3 n=10 NL3 (10mM caffeine) n=5 vs. WT n=8 NL3 n=10 vs. WT n=8 30 minutes NL3 (10mM caffeine) n=5 vs. NL3 n=10 NL3 (10mM caffeine) n=5 vs. WT n=8 NL3 n=10 vs. WT n=8 60 minutes NL3 (10mM caffeine) n=5 vs. NL3 n=10 NL3 (10mM caffeine) n=5 vs. WT n=8 NL3 n=10 vs. WT n=8 90 minutes NL3 (10mM caffeine) n=5 vs. NL3 n=10 NL3 (10mM caffeine) n=5 vs. WT n=8 NL3 n=10 vs. WT n=8 120 minutes NL3 (10mM caffeine) n=5 vs. NL3 n=10 NL3 (10mM caffeine) n=5 vs. WT n=8 NL3 n=10 vs. WT n=8

Mean difference -0.00018 7.5e-005 0.00026 -0.00021 5.3e-005 0.00027 -0.0017 -0.00021 0.0015 -0.0048 -0.00048 0.0043 -0.0067 -0.00033 0.0064 -0.0077 -0.0011 0.0066

95% CI of diff. -0.0038 to 0.0034 -0.0037 to 0.0038 -0.0028 to 0.0034 -0.0038 to 0.0034 -0.0037 to 0.0038 -0.0028 to 0.0034 -0.0053 to 0.0018 -0.0039 to 0.0035 -0.0016 to 0.0046 -0.0085 to -0.0012 -0.0042 to 0.0032 0.0012 to 0.0075 -0.010 to -0.0031 -0.0041 to 0.0034 0.0033 to 0.0095 -0.011 to -0.0041 -0.0049 to 0.0026 0.0035 to 0.0097

Summary ns ns ns ns ns ns ns ns ns ns * ns ns

Adjusted P Value >0.99 >0.99 >0.99 >0.99 >0.99 >0.99 0.93 >0.99 0.93 0.0014 >0.99 0.0007 <0.0001 >0.99 <0.0001 <0.0001 0.99 <0.0001

133

Appendix Table 28. Statistical parameters for jejunum permeability for fasting NL3 mice versus fasting NL3 mice treated with caffeine utilising a two-way ANOVA with a Sidak multiple comparisons test. (Green fill represents caffeine restoring NL3 FITC concentrations to WT concentrations).

Sidak multiple comparisons test Jejunum 0 minutes NL3 + 10mM caffeine n=5 vs. NL3 n=9 NL3 + 10mM caffeine n=5 vs. WT n=11 NL3 n=9 vs. WT n=11 1 minute NL3 + 10mM caffeine n=5 vs. NL3 n=9 NL3 + 10mM caffeine n=5 vs. WT n=11 NL3 n=9 vs. WT n=11 30 minutes NL3 + 10mM caffeine n=5 vs. NL3 n=9 NL3 + 10mM caffeine n=5 vs. WT n=11 NL3 n=9 vs. WT n=11 60 minutes NL3 + 10mM caffeine n=5 vs. NL3 n=9 NL3 + 10mM caffeine n=5 vs. WT n=11 NL3 n=9 vs. WT n=11 90 minutes NL3 + 10mM caffeine n=5 vs. NL3 n=9 NL3 + 10mM caffeine n=5 vs. WT n=11 NL3 n=9 vs. WT n=11 120 minutes NL3 + 10mM caffeine n=5 vs. NL3 n=9 NL3 + 10mM caffeine n=5 vs. WT n=11 NL3 n=9 vs. WT n=11

Mean difference -0.00023 -8.7e-005 0.00014 -0.00026 -0.00014 0.00012 -0.0017 -0.0010 0.00068 -0.0045 -0.0025 0.0019 -0.0062 -0.0033 0.0028 -0.0067 -0.0033 0.0034

95% CI of diff. -0.0040 to 0.0036 -0.0038 to 0.0036 -0.0029 to 0.0032 -0.0041 to 0.0035 -0.0038 to 0.0035 -0.0029 to 0.0032 -0.0056 to 0.0020 -0.0048 to 0.0026 -0.0024 to 0.0037 -0.0083 to -0.00067 -0.0063 to 0.0011 -0.0011 to 0.0050 -0.010 to -0.0023 -0.0070 to 0.00035 -0.00026 to 0.0059 -0.010 to -0.0029 -0.0070 to 0.00038 0.00030 to 0.0065

Summary Adjusted P Value ns ns ns ns ns ns ns ns ns ns ns ns ns ** ns *

>0.99 >0.99 >0.99 >0.99 >0.99 >0.99 0.96 0.99 >0.99 0.0088 0.49 0.66 <0.0001 0.11 0.10 <0.0001 0.12 0.019

134

Appendix Table 29. Statistical parameters for distal ileum permeability for fasting NL3 mice versus fasting NL3 mice treated with caffeine utilising a two-way ANOVA with a Sidak multiple comparisons test. (Green fill represents caffeine restoring NL3 FITC concentrations to WT concentrations).

Sidak multiple comparisons test Distal ileum 0 minutes NL3 + 10mM caffeine n=5 vs. NL3 n=9 NL3 + 10mM caffeine n=5 vs. WT n=8 NL3 n=9 vs. WT n=8 1 minute NL3 + 10mM caffeine n=5 vs. NL3 n=9 NL3 + 10mM caffeine n=5 vs. WT n=8 NL3 n=9 vs. WT n=8 30 minutes NL3 + 10mM caffeine n=5 vs. NL3 n=9 NL3 + 10mM caffeine n=5 vs. WT n=8 NL3 n=9 vs. WT n=8 60 minutes NL3 + 10mM caffeine n=5 vs. NL3 n=9 NL3 + 10mM caffeine n=5 vs. WT n=8 NL3 n=9 vs. WT n=8 90 minutes NL3 + 10mM caffeine n=5 vs. NL3 n=9 NL3 + 10mM caffeine n=5 vs. WT n=8 NL3 n=9 vs. WT n=8 120 minutes NL3 + 10mM caffeine n=5 vs. NL3 n=9 NL3 + 10mM caffeine n=5 vs. WT n=8 NL3 n=9 vs. WT n=8

Mean difference -0.00011 0.00011 0.00023 -0.00015 8.7e-005 0.00024 -0.0019 -0.00023 0.0017 -0.0033 -0.00084 0.0024 -0.0043 -0.0010 0.0032 -0.0054 -0.0016 0.0038

95% CI of diff. -0.0033 to 0.0030 -0.0031 to 0.0033 -0.0025 to 0.0030 -0.0033 to 0.0030 -0.0031 to 0.0033 -0.0025 to 0.0030 -0.0051 to 0.0012 -0.0035 to 0.0030 -0.0010 to 0.0044 -0.0065 to -0.00013 -0.0041 to 0.0024 -0.00029 to 0.0052 -0.0075 to -0.0011 -0.0043 to 0.0021 0.00045 to 0.0060 -0.0086 to -0.0022 -0.0049 to 0.0016 0.0010 to 0.0065

Summary ns ns ns ns ns ns ns ns ns * ns ns ns ** ns

Adjusted P Value >0.99 >0.99 >0.99 >0.99 >0.99 >0.99 0.70 >0.99 0.69 0.033 >0.99 0.12 0.0013 0.99 0.0099 <0.0001 0.91 0.0011

135

Appendix Table 30. Statistical parameters for colon permeability for fasting NL3 mice versus fasting NL3 mice treated with caffeine utilising a two-way ANOVA with a Sidak multiple comparisons test. (Green fill represents caffeine restoring NL3 FITC concentrations to WT concentrations).

Sidak multiple comparisons test Colon 0 minutes NL3 + 10mM caffeine n=3 vs. NL3 n=9 NL3 + 10mM caffeine n=3 vs. WT n=8 NL3 n=9 vs. WT n=8 1 minute NL3 + 10mM caffeine n=3 vs. NL3 n=9 NL3 + 10mM caffeine n=3 vs. WTn=8 NL3 n=9 vs. WTn=8 30 minutes NL3 + 10mM caffeine n=3 vs. NL3 n=9 NL3 + 10mM caffeine n=3 vs. WTn=8 NL3 n=9 vs. WTn=8 60 minutes NL3 + 10mM caffeine n=3 vs. NL3 n=9 NL3 + 10mM caffeine n=3 vs. WTn=8 NL3 n=9 vs. WTn=8 90 minutes NL3 + 10mM caffeine n=3 vs. NL3 n=9 NL3 + 10mM caffeine n=3 vs. WTn=8 NL3 n=9 vs. WTn=8 120 minutes NL3 + 10mM caffeine n=3 vs. NL3 n=9 NL3 + 10mM caffeine n=3 vs. WTn=8 NL3 n=9 vs. WTn=8

Mean difference -0.00020 -0.00016 3.6e-005 -0.00022 -0.00015 6.0e-005 -0.00075 -0.00085 -0.00010 -0.0014 -0.0010 0.00041 -0.0021 -0.0012 0.00090 -0.0030 -0.0015 0.0014

95% CI of diff. -0.0021 to 0.001771 -0.0021 to 0.001838 -0.0014 to 0.001475 -0.0021 to 0.001753 -0.0021 to 0.001844 -0.0013 to 0.001499 -0.0027 to 0.0012 -0.0028 to 0.0011 -0.0015 to 0.0013 -0.0033 to 0.00055 -0.0030 to 0.0010 -0.0010 to 0.0018 -0.0041 to -0.00018 -0.0032 to 0.00074 -0.00053 to 0.0023 -0.0049 to -0.0010 -0.0035 to 0.00041 -1.6e-005 to 0.0028

Summary ns ns ns ns ns ns ns ns ns ns ns ns * ns ns *** ns ns

Adjusted P Value >0.99 >0.99 >0.99 >0.99 >0.99 >0.99 0.99 0.97 >0.99 0.41 0.91 0.99 0.020 0.66 0.65 0.0002 0.27 0.055

Appendix Table 31. Statistical parameters for duodenal permeability for of fasting WT mice versus fasting WT mice treated with caffeine utilising a two-way ANOVA with a Sidak multiple comparisons test.

Šídák's multiple comparisons test Duodenum WT (10mM caffeine) n=5 - WT n=8 0 minute 1 minute 30 minutes 60 minutes 90 minutes 120 minutes

Mean Diff. 8.664e-005 7.340e-005 -0.0002189 -0.0002480 0.0007436 0.0007404

95.00% CI of diff. -0.002838 to 0.003012 -0.002852 to 0.002998 -0.003144 to 0.002706 -0.003173 to 0.002677 -0.002182 to 0.003669 -0.002185 to 0.003665

Summary ns ns ns ns ns ns

Adjusted P Value >0.9999 >0.9999 >0.9999 >0.9999 0.9830 0.9834

136

Appendix Table 32. Statistical parameters for jejunum permeability for fasting WT mice versus fasting WT mice treated with caffeine utilising a two-way ANOVA with a Sidak multiple comparisons test.

Šídák's multiple comparisons test Jejunum WT+ 10mM caffeine n=6 - WT n=11 0 minute 1 minute 30 minutes 60 minutes 90 minutes 120 minutes

Mean Diff. -0.0002265 -0.0002577 -0.001127 -0.002843 -0.003635 -0.003588

95.00% CI of diff. -0.003159 to 0.002706 -0.003190 to 0.002675 -0.004059 to 0.001805 -0.005775 to 8.947e-005 -0.006568 to -0.0007027 -0.006520 to -0.0006551

Summary ns ns ns ns ** **

Adjusted P Value >0.9999 >0.9999 0.8863 0.0623 0.0074 0.0085

Appendix Table 33. Statistical parameters for distal ileum permeability for fasting WT mice versus fasting WT mice treated with caffeine utilising a two-way ANOVA with a Sidak multiple comparisons test.

Šídák's multiple comparisons test Distal ileum WT (10mM caffeine) n=6 - WT n=8 0 minute 1 minute 30 minutes 60 minutes 90 minutes 120 minutes

Mean Diff. -0.0001776 -0.0001928 -0.0004516 -0.001400 -0.001972 -0.003133

95.00% CI of diff. -0.003176 to 0.002821 -0.003192 to 0.002806 -0.003450 to 0.002547 -0.004398 to 0.001599 -0.004970 to 0.001027 -0.006132 to -0.0001342

Summary ns ns ns ns ns *

Adjusted P Value >0.9999 >0.9999 0.9990 0.7584 0.3917 0.0360

Appendix table 34. Statistical parameters for colon permeability for fasting WT mice versus fasting WT mice treated with caffeine utilising a two-way ANOVA with a Sidak multiple comparisons test.

Šídák's multiple comparisons test Colon WT + 10mM caffeine n=5 - WT n=9 0 minute 1 minute 30 minutes 60 minutes 90 minutes 120 minutes

Mean Diff. -0.0001743 -0.0001610 -0.0007668 -0.0009190 -0.001096 -0.001389

95.00% CI of diff. -0.001563 to 0.001215 -0.001550 to 0.001228 -0.002156 to 0.0006222 -0.002308 to 0.0004700 -0.002485 to 0.0002932 -0.002778 to 3.687e-007

Summary ns ns ns ns ns ns

Adjusted P Value 0.9997 0.9998 0.5929 0.3829 0.1979 0.0501

137