The Anti-Inflammatory Effects of the Dietary Dipeptide, Gamma- Glutamylvaline in a Mouse Model of LPS-Induced Sepsis

MacKenzie Chee

A Thesis presented to The University of Guelph

In partial fulfillment of requirements for the degree of Master of Science in Food Science

Guelph, Ontario, Canada

ABSTRACT

THE ANTI-INFLAMMATORY EFFECTS OF THE DIETARY DIPEPTIDE, GAMMA- GLUTAMYLVALINE IN A MOUSE MODEL OF LPS-INDUCED SEPSIS

MacKenzie Chee Advisor: University of Guelph, 2016 Professor Y. Mine

This study endeavored to identify the anti-inflammatory activity and signaling mechanism of the bioactive dietary dipeptide gamma-L-glutamyl-L-valine (γ-EV) via calcium- sensing receptor (CaSR) activation in a lipopolysaccharide (LPS)-induced mouse model of sepsis. Sepsis occurs when the otherwise tightly-regulated immune and inflammatory response that eliminates invading pathogens and repairs injured tissues, switches to an uncontrolled response that causes an acute life-threatening inflammatory syndrome. γ-EV had anti- inflammatory activity in BALB/c mice with LPS-induced sepsis as measured by the reduction of the pro-inflammatory cytokines TNF-α, IL-6, and IL-1β in plasma and small intestine tissue.

There was a γ-EV-mediated decrease in the phosphorylation of the downstream signaling proteins JNK and IκBα in small intestine tissue, which indicates γ-EV likely has an effect on an upstream signaling protein common to both the AP-1 and NF-κB pathways. The study proposes that a direct interaction of CaSR-recruited β-arrestin2 with TRAF6, TAB1, and IκBα constitutes

γ-EV’s anti-inflammatory activity in LPS-induced inflammation.

This study provides evidence that γ-EV has health-promoting activity.

ii ACKNOWLEDGEMENTS

I would like to express my gratitude to my entire support system during my Graduate studies. My advisor, Dr. Yoshinori Mine, provided me with guidance in academia and research, and encouraged me to lead a balanced ‘bushido’ life. I would like to thank my lab peers Dr.

Toshihiko Fukuda, Dr. Fang Geng, Dr. Hua Zhang, and Dr. Kaustav Majumder for educating and training me in the lab. I would not have the lab skills set that I have today without these fine scientists teaching me along the way. I would also like to express my appreciation for Dr. Rong

Cao who provided me with guidance as a member of my Advisory Committee.

Thank you to my partner in life – Zachary Munroe, who has stood by me through my journey as a Graduate student. Thank you for always believing in me, and encouraging me to be the best person I can possibly be. Your support has been paramount during my Masters studies.

I would like to thank the Advanced Foods and Materials Network (AFMNet) and the

Natural Sciences and Engineering Research Council of Canada (NSERC) for the financial support for this project. As well, I would like to express my utmost gratitude for the OAC Class of 1950, Mary Frances Hucks and family, Kenneth W. Knox and family, as well as Martha Robb and family for their tremendous generosity.

iii TABLE OF CONTENTS

LIST OF TABLES……………………………………………………………………... vii

LIST OF FIGURES…………………………………………………………………….. viii

ABBREVIATIONS…………………………………………………………………….. ix

1. LITERATURE REVIEW 1.1 Gut Health, Intestinal Inflammation and Chronic Inflammatory Diseases of the Gut………………………………………………………. 1 1.2 Sepsis…………………………………………………………………….. 3 1.2.1 Definition of Sepsis……………………………………….. 3 1.2.2 Causative Pathogens and Models of Sepsis………………. 7 1.2.3 Pathogen-Associated Molecular Patterns and Their Receptors…………………………………………… 8 1.2.4 Pathogenesis………………………………………………. 12 1.2.5 Diagnostic Biomarkers of Sepsis…………………………. 14 1.2.6 Clinical Diagnostic Criteria……………………………….. 16 1.2.7 Treatment of Sepsis……………………………………….. 17 1.2.8 Vulnerable Populations…………………………………… 19 1.2.9 Prevalence and Healthcare Costs………………………….. 19 1.3 Immunology of Sepsis…………………………………….……………… 20 1.4 Bioactive Peptides……………………………………..…………………. 21 1.5 Gamma-glutamylvaline..……………….………………………………… 23 1.6 Calcium-Sensing Receptor (CaSR) ……………………………………… 28 1.6.1 Background…………………………………….………….. 28 1.6.2 Structure and Function of CaSR…………………………... 28 1.6.3 Agonists and Antagonists…………………………………. 30 1.6.4 Binding Site for L-Amino Acids…………………………... 31 1.6.5 CaSR Signaling…………………………..………………... 32

2. RATIONALE, HYPOTHESIS AND OBJECTIVES………………………………. 35 2.1. Rationale………………………….……………………………………… 35 2.2. Hypothesis………………………….……………………………………. 35 2.3. Objectives………………………….…………………………………….. 35

3. MATERIALS AND METHODS………………………….………………………… 36 3.1. Reagents………………………….………………………….…………... 36 3.2. Animal Study………………………….………………………….……… 36 3.2.1. Animals………………………….………………………… 36

iv 3.2.2. Treatment………………………….……………………... 37 3.2.3. Sepsis Induction………………………….……………..... 37 3.2.4. Tissue Collection………………………….…………..….. 38 3.2.5. Tissue Protein Extraction and Protein Concentration…..... 38 3.3. ELISA………………………….………………………….……….…… 39 3.4. Western blot………………………….…………………………..…….. 40 3.5. Histological Analysis………………………….……………………….. 41 3.6. Data Analysis………………………….…………………...….……….. 42

4. RESULTS 4.1. Effects of γ-EV on Acute Inflammatory Cytokine Secretion in Plasma and Small Intestine……………………….…...………………. 43 4.1.1. Effects of γ-EV on Acute Inflammatory Cytokine Secretion in Plasma………..…………………………….. 43 4.1.2. Effects of γ-EV on Acute Inflammatory Cytokine Secretion in Small Intestine……………………………... 46 4.2. Effects of γ-EV on Phosphorylation and Expression of Signaling Proteins………………………………………………….….. 48 4.2.1. p-IκBα/IκBα……………...……….…………………….. 48 4.2.2. p-JNK/JNK…………...………….………………………. 48 4.2.3. TRAF6/ β-actin……………………….………………….. 48 4.2.4. Ubiquitin/ β-actin……………………….………………... 50 4.3. Small Intestine Histology………………………….…………………… 53

5. DISCUSSION 5.1. γ-EV Suppressed Pro-Inflammatory Cytokine Secretion in Plasma…….. 55 5.2. γ-EV Suppressed Pro-Inflammatory Cytokine Secretion in Small Intestine………………………….………………………………… 56 5.3. γ-EV Inhibited Phosphorylation of IκBα………………………………… 57 5.4. γ-EV Inhibited Phosphorylation of JNK………………………………… 58 5.5. γ-EV Had an Effect on Background Ubiquitin Protein Levels that was Unrelated to TRAF6………………………….…………………………. 59 5.6. Protein Concentration of Total TRAF6 Decreased with LPS Stimulation………………………….………………………….….. 60 5.7. Cellular Signaling Crosstalk Mechanism………………………………. 60 5.8. Applications to Sepsis Prevention.……………………….…………….. 67

6. CONCLUSION AND FUTURE WORK………………………….………………. 69

v 7. REFERENCES………………………….………………………….……………… 70

vi LIST OF TABLES

Table 1: Diagnostic criteria for sepsis………………………….……………………. 5

Table 2: The Sepsis-related organ failure assessment (SOFA) ……………………... 6

Table 3: Diagnostic biomarkers of sepsis….………………………………………… 15

Table 4: Treatments for sepsis……………….………………………………………. 18

Table 5: Bioactive food-derived peptides and their anti-inflammatory activities.….... 24

vii LIST OF FIGURES

Figure 1: The structure of the gut barrier integrity……………….…………………… 2

Figure 2: The structure of lipopolysaccharide (LPS) ……………….……………….. 9

Figure 3: The Toll-like receptor 4 (TLR4) structure and its associated TIR signaling

domain……………….……………………………………………………… 11

Figure 4: The three phases of sepsis……………….…………………………………. 13

Figure 5: The structure of γ-EV……………….……………………………………… 27

Figure 6: Calcium-sensing receptor (CaSR): the structure of the receptor and its

intracellular signaling domain……….…………………………………….. 33

Figure 7: The TNF-α, IL-6 and IL-1β protein expression in the plasma…………….. 45

Figure 8: The TNF-α, IL-6, IL-1β, and IL-10 protein expression in the

small intestine…..………….………………………………………………. 47

Figure 9: The p-JNK/JNK and p-IκBα/ IκBα protein expression in the

small intestine…..………….………………………………………………. 49

Figure 10: The TRAF6/β-actin and Ubiquitin/β-actin protein expression in the

small intestine……………….…………………………………………….. 52

Figure 11: Small intestine histological slides stained with H&E…………………...… 54

Figure 12: Proposed signaling crosstalk interaction between β−arrestin2 and TRAF6.. 64

Figure 13: Proposed signaling crosstalk interaction between β−arrestin2 and TAB1… 65

Figure 14: Proposed signaling crosstalk interaction between β−arrestin2 and IκBα..... 66

viii LIST OF ABBREVIATIONS

1,25(OH)2D3 1,25-dihydroxyvitamin D3

AP-1 activator protein 1

APACHE acute physiology and chronic health evaluation

AS ankylosing spondylitis

APC antigen-presenting cell

BSA bovine serum albumin

JNK c-Jun N-terminal kinase

CRP C-reactive protein

CT calcitonin

CaSR calcium-sensing receptor

CD14 cluster of differentiation 14

Co-IP co-immunoprecipitation

CD Crohn’s disease

DAMP damage-associated molecular pattern

DC dendritic cells

ELISA enzyme-linked immunosorbent assay

ECD extracellular domain

γ-EV gamma-glutamylvaline

GI gastrointestinal (tract)

H&E hematoxylin and eosin

HLA-DR human leukocyte antigen – antigen D related

HUVEC human umbilical vein endothelial cell

ix IKK IκB kinase

IRAK IL-1R-associated kinase iNOS inducible nitric oxide synthase

IBD inflammatory bowel disease

ICU intensive care unit

ICAM-1 intercellular adhesion molecule-1

IFN interferon

IL interleukin

IEC intestinal epithelial cell

IP intraperitoneal

LP lamina propria

LPS lipopolysaccharide

LBP lipopolysaccharide-binding protein

LTA lipoteichoic acid

MAPK mitogen-activated protein kinase

MCP-1 monocyte chemoattractant protein-1

Mal MyD88-adaptor-like protein

MD-2 myeloid differentiation-2

NICU neonatal intensive care unit

NO nitric oxide

NF-κB nuclear factor kappa-light-chain-enhancer of activated B cells

IκBα nuclear factor of kappa light polypeptide gene enhancer in B-cells inhibitor, alpha

PTH parathyroid hormone

x PAMP pathogen-associated molecular pattern

PRR pathogen recognition receptor

PCT procalcitonin

RIP1 receptor-interacting protein 1

RA rheumatoid arthritis

RNase ribonuclease

SOFA sepsis-related organ failure assessment siRNA small interfering RNA

SEM standard error of the mean

SIRS systemic inflammatory response syndrome

TAK1 TGF-β-activated kinase 1

TAB1 TAK1-binding protein

TRIF TIR-domain-containing adapter-inducing interferon-β

TRAF6 TNF receptor associated factor 6

TIRAP toll-interleukin 1 receptor (TIR) domain containing adaptor protein

TLR4 toll-like receptor 4

TMD transmembrane domain

TNF-α tumour necrosis factor alpha

TNFR tumour necrosis factor receptor

Ub ubiquitin

UC ulcerative colitis

VCAM-1 vascular cell adhesion molecule-1

WB Western blot

xi 1. LITERATURE REVIEW

1.1 Gut Health, Intestinal Inflammation, and Chronic Inflammatory Diseases of the Gut

The gastrointestinal (GI) tract is in constant contact with the external environment through the ingestion of foodstuffs and their associated allergens and pathogens (Schneeman,

2002). Thus, the GI tract is responsible for maintaining a robust defense system against transient foreign agents in order to support health and prevent disease (Schneeman, 2002; Sukhotnik et al.,

2004). Maintenance of gut health is also required for the proper absorption of nutrients from ingested foodstuffs (Schneeman, 2002). There are several defense systems in place, both physical and immunological, that work concomitantly to protect the gut and whole body

(Salminen et al., 1998) (Figure 1). Immunological mechanisms include intraepithelial lymphocytes, Peyer’s patches, and mesenteric lymph nodes containing lymphoid tissues

(Salminen et al., 1992; Sukhotnik et al., 2004). The physical defense system includes gut mucosa, tight junctions, and intestinal motility (Salminen et al., 1998; Sukhotnik et al., 2004).

Additionally, the gut microflora or ‘microbiota’ has a significant role in gut health maintenance, immune support, and nutrient absorption through its metabolites and through direct competition

(Berg and Owens, 1979; Salminen et al., 1998; Schneeman, 2002; Sukhotnik et al., 2004).

Intestinal inflammation has become a significant health concern, especially within

Europe, the UK, and North America (Loftus, 2004). Inflammatory bowel diseases (IBDs) such as

Crohn’s disease (CD) and ulcerative colitis (UC) are chronic, relapsing inflammatory conditions that cause pain, malabsorption of nutrients, requirement for surgical resectioning, and impact quality of life (Sartor, 2006; Ghosh and Mitchell, 2007; Vatn and Sandvik, 2015). The loss of the intestinal barrier integrity and mucosal injury exhibited in IBD permits food particles and

Figure 1: The physical, immunological, and microbial defense mechanisms of the gut. Physical defenses include tight junctions (TJs), gut motility, and gut mucosa. Immunological defenses include cells of the innate immune system such as macrophages and dendritic cells (DCs). Microbial defenses include resident or transient commensal bacteria that protect the host by antagonism, as well as through microbial metabolites. (Adapted from Petersen and Artis, 2014 and Michielan and D’Incà, 2015.)

2 pathogens that would be otherwise benign if contained within the lumen, to slip through the intestinal epithelium into the lamina propria (LP) where they can mount an immune response

(Michielan and D’Incà, 2015).

There are many studies that purport the heritability of IBD, as demonstrated by a 58.3% concordance rate of CD and 18.2% for UC in monozygotic twins (Orholm et al., 2000). There are, however, various environmental influences that affect the chance of developing UC and CD.

For example, studies found that appendectomy increased the risk of developing CD, however played a protective role against the development of UC (Rutgeerts et al., 1994; Andersson,

2003). Similarly, cigarette smokers had an increased the risk of developing CD, yet had a decreased risk of UC (Logan et al., 1984; Tobin et al., 1987). IBD is commonly associated with inflammatory comorbidities such as ankylosing spondylitis (AS) and rheumatoid arthritis (RA), indicating that treatment of the underlying inflammatory processes rather than merely managing symptoms is more effective for whole-body health (Ghosh and Mitchell, 2007).

LPS endotoxemia leads to a more acute-onset of loss of intestinal barrier integrity in the gut and damage to the intestinal mucosa (Deitch et al., 1989; Sukhotnik et al., 2004). The compromised intestinal epithelial integrity leads to bacterial translocation, whereby luminal bacteria bypass the epithelial barrier to enter the LP, mesenteric lymph nodes, and other organs to subsequently mount an immune response (Berg and Owens, 1979; Sukhotnik et al., 2004).

1.2 Sepsis

1.2.1 Definition of Sepsis

Sepsis is defined as the systemic inflammatory response syndrome (SIRS) with the presence of an infection (Balk et al., 1992; Levy et al., 2003). This differs from SIRS without

3 infection that can arise from a noninfectious process present in the body such as pancreatitis, trauma, or ischemia which initiates the innate immune system through damage-associated molecular patterns (DAMPs) produced by host cells (Balk et al., 1992; Levy et al., 2003; Faix,

2013).

SIRS is diagnosed when a patient exhibits one or more of the following symptoms (Table

-1 -1 1): tachycardia (>90 min ), hyperventilation (respiratory rate >20 min or PaCO2 <32 mmHg), body temperature >38oC or <36oC, or a white blood cell (WBC) count of >12,000 cells µL-1 or

<4,000 µL-1 (Levy et al., 2003). Clinicians are able to diagnose sepsis despite not having definitive blood culture results to confirm infection, in which case, diagnosis is based on suspicion of infection and satisfaction of SIRS criteria (Levy et al., 2003). The unconfirmed diagnosis of sepsis is done in order to promptly initiate treatment to decrease the rate of morbidity and mortality (Rivers et al., 2001). Severe sepsis is diagnosed when there is evidence of organ dysfunction, sepsis-induced hypotension, and hypoperfusion abnormalities (Balk et al.,

1992; Levy et al., 2003). Examples of hypoperfusion abnormalities are oliguria, lactic acidosis, or an acute change in mental function (Balk et al., 1992). Organ dysfunction or failure in sepsis can be evaluated by the Sepsis-Related Organ Failure Assessment (SOFA) score (Table 2), developed in 1994 by the European Society of Intensive Care Medicine (ECISM) or general disease severity can be assessed using the Acute Physiology and and Chronic Health Evaluation

(APACHE) score (Vincent et al., 1996; Rivers et al., 2001). Ferreira et al. (2001) determined that the maximum and mean SOFA scores of patients during their stay in the intensive care unit

(ICU) were an accurate indicator of prognosis and risk of mortality, indicating that increased organ dysfunction was associated with increased mortality rate. Jones et al. (2009) additionally determined that the change in SOFA score (∆SOFA) between T0 (emergency department

4 Table 1: Diagnostic criteria for sepsis (Levy et al., 2003).

Infection, documented or suspected, and some of the following: General variables Fever core temperature >38.3oC Hypothermia core temperature <36 oC Tachycardia heart rate >90 min-1 or >2 SD above normal value for age Tachypnea abnormally rapid breathing Altered mental status Significant edema or positive fluid balance (>20 mL/kg over 24 hrs) Hyperglycemia plasma glucose >120 mg/dL or 7.7 mmol/L, in the absence of diabetes Inflammatory variables Leukocytosis WBC count > 12,000 µL-1 Leukopenia WBC count < 4000 µL-1 Normal WBC with >10% immature forms Plasma CRP >2 SD above the normal value Plasma PCT >2 SD above the normal value Hemodynamic variables Arterial hypotension systolic blood pressure (SBP) <90 mm Hg, mean arterial blood pressure (MAP) decrease > 40mm Hg in adults or <2 SD below normal for age Mixed venous oxygen saturation (S�O2) >70% Cardiac index >3.5 L•min-1•M-23 Organ dysfunction variables Arterial hypoxemia PaO2/FIO2 <300 Acute oliguria urine output < 0.5 mL•kg-1 or 45 mmol/L for at least 2 hours Creatinine increase >0.5 mg/dL Coagulation abnormalities international normalized ratio (INR) > 1.5 or activated partial thromboplastin time (aPTT) > 60s Ileus (absent bowel sounds) Thrombocytopenia Platelet count <100,000 µL-1 Hyperbilirubinemia Plasma total bilirubin >4 mg/dL or 70 mmol/L Hyperlactatemia >1 mmol/L Decreased capillary refill or mottling

5 Table 2: The Sepsis-Related Organ Failure Assessment (SOFA) Score Criteria (Adapted from

Vincent et al., 1996; Jones et al., 2009).

SOFA Score 1 2 3 4

Respiration

PaO2/FIO2, mm Hg <400 <300 <200 <100

with respiratory support

SaO2/FIO2 221-301 142-220 67-141 <67 Coagulation Platelets x103/mm3 <150 <100 <50 <20 Liver Bilirubin, mg/dL 1.2-1.9 2.0-5.9 6.9-11.9 >12.0 (µmol/L) (20-32) (33-101) (102-204) (<204) Cardiovascular Hypotension MAP<70 Dopamine≤5 or Dopamine >5 Dopamine >15 dobutamine (any or or dose) epinephrine ≤0.1 epinephrine >0.1 or norepinephrine≤0.0.1 or norepinephrine > 0.1 Central Nervous System Glasgow coma 13-14 10-12 6-9 <6 score Renal Creatinine, mg/dL 1.2-1.9 2.0-3.4 3.5-4.9 or <500 >5.0 or <200 (µmol/L)) urine (110-170) (171-299) (300-440) (>400) output or <500 mL/day or <200 mL/day

6 presentation with evidence of hypoperfusion) and T72 (72h after ICU admission) was positively associated with changes in mortality. Thus, the SOFA score is an accurate predictor of mortality and indicates that organ failure is directly associated with an increased mortality rate in sepsis.

Severe sepsis is an alarming health concern as it is the leading cause of death in ICUs and the second most common cause of death in neonatal ICUs (NICUs) (Levy et al., 2003; Angus et al.,

2001; Jacob et al., 2015). Septic shock is diagnosed when patients in addition to the criteria for severe sepsis, undergo cardiovascular collapse or arterial hypotension, despite intervention with vasopressive therapies and fluid resuscitation (Balk et al., 1992; Levy et al., 2003; Faix 2013).

The pathogenesis of sepsis is troublesome as it presents as a heterogeneous syndrome in different patients (Levy et al., 2003 and O’Brien et al., 2007), which makes a silver bullet approach to sepsis treatment difficult. A treatment approach that may significantly improve one patient’s health status may cause another patient to enter a severe immunosuppressive state that may ultimately lead to death (Fisher et al., 1996; Reinhart and Karzai, 2001). It is, however, important to note that in most cases of sepsis, the clinical presentation of the syndrome is similar regardless of the invading pathogen, as sepsis is a consequence of the body’s exaggerated response to infection rather than the actual infection itself (Faix, 2013).

1.2.2 Causative Pathogens and Models of Sepsis

Sepsis is caused by both Gram-negative and Gram-positive bacteria, as well as anaerobes, fungi, and polymicrobial infections (Martin et al., 2003). The most common sepsis-causing pathogens are Escherichia coli (E. coli), a Gram-negative bacterium, and Staphylococcus aureus

(S. aureus), a Gram-positive bacterium. However, there is a higher mortality rate associated with

Gram-negative Pseudomonas aeruginosa (P. aeruginosa) infection (Vincent et al., 2006). Sites

7 of infection include primary bacteremia, respiratory, device-related, genitourinary, endocarditis, and wound/soft tissue (Angus et al., 2001).

There are various models of sepsis induction for animal trials depending on the study objectives. For example, researchers may look to examine sepsis caused by a specific bacterium, or to examine polymicrobial sepsis. The caecal ligation and puncture (CLP) method to induce polymicrobial sepsis uses endogenous caecal contents, as well as necrosis from caecal ligation, and tissue trauma from surgery to induce a sepsis whose pathogenesis closely resembles human sepsis (Dejager et al., 2011). The caecal slurry sepsis model consists of collecting the caecal contents from a donor mouse and subsequently administering an intraperitoneal (IP) injection of a caecal slurry (CS) solution produced with the donor mouse’s caecal contents to induce a polymicrobial sepsis (Starr et al., 2014). The present project used the lipopolysaccharide (LPS), or endotoxin, sepsis model in which purified LPS from a Gram-negative bacteria of choice is administered orally or via injection. The present project used LPS from E. coli 0111:B4 dissolved in endotoxin-free water administered via IP injection as the project objectives were to evaluate the specific effects of LPS-induced sepsis, rather than a polymicrobial model. Thus, the

LPS-induced sepsis model was chosen specifically to match the present project’s objectives, as is the case when researchers choose other sepsis models.

1.2.3 Pathogen-Associated Molecular Patterns and their Receptors

Evolutionarily-conserved structural motifs on the surface of bacteria called pathogen- associated molecular patterns (PAMPs) are recognized by pathogen recognition receptors (PRRs) on cells of the innate immune system such as macrophages and dendritic cells (DCs) (Sriskandan and Altmann, 2008; van der Poll and Opal, 2008). Lipoteichoic acid (LTA) found on the surface

Figure 2: The structure of lipopolysaccharide (LPS). The pathogenic proximal lipid A moiety, followed by the non-repeating oligosaccharide core, and the distal polysaccharide O-antigen (Adapted from Sigma-Aldrich, 2011; Maeshima and Fernandez, 2013).

9 of Gram-positive bacteria binds and activates Toll-like receptor 2 (TLR2) which initiates the

TLR-IL-1R (TIR) inflammatory signaling pathway (Michelsen et al., 2001; Takeda and Akira,

2004). LPS, a Gram-negative PAMP, is the major component of the outer membrane of Gram negative bacteria, comprising 70% of its structure (Michelsen et al., 2001; van der Poll and Opal,

2007). LPS is composed of three parts: a hydrophobic lipid A moiety, a non-repeating oligosaccharide, and polysaccharide O-antigen (Figure 2) (Raetz and Whitfield, 2002). The lipid

A moiety confers to LPS its potent immune-stimulating properties. With the aid of the scaffolding proteins LPS-binding protein (LBP), cluster of differentiation 14 (CD14), and myeloid differentiation-2 (MD-2), LPS binds and activates Toll-like receptor 4 (TLR4) to activate the TIR signaling cascade (Figure 3) (Tobias et al., 1986; Tobias et al., 1989; Wright et al., 1990; Shimazu et al., 1999; Mogensen, 2009). The evolutionarily conserved TIR domain signals through an MyD88-dependent and an MyD88-independent, alternatively known as TIR- domain-containing adapter-inducing interferon-β (TRIF), pathway, both of which converge onto

TNF receptor associated factor 6 (TRAF6) and its subsequent downstream signaling (Akira and

Takeda, 2004; Mogensen, 2009). Upon receptor activation, the MyD88-dependent pathway initiates with the bridging of TLR4 and MyD88 by the MyD88-adaptor-like protein (Mal) (Akira and Takeda, 2004). MyD88 then recruits the IL-1R-associated kinases (IRAKs), of which

IRAK4 and IRAK1/2 are phosphorylated (Mogensen, 2009). Phosphorylated IRAK1/2 then associates with TRAF6, which is activated via polymerization and Lys-63-linked autoubiquitination (Mogensen, 2009). Polyubiquitinated TRAF6 facilitates the formation of the

TGF-β-activated kinase 1 (TAK1)-TAK-1-binding protein 1 (TAB1)-TAB2 complex, as well ubiquinates TAK1 (Mogensen, 2009). The activated TAK1-TAB1-TAB2 complex then stimulates the activator protein 1 (AP-1) and mitogen-activated protein kinase (MAPK) signaling

Figure 3: Intracellular signaling cascade of LPS-activated TLR4, showing the MyD88- dependent and MyD88-independent signaling cascades of the conserved TIR signaling pathway (Adapted from Akira and Takeda, 2004; Mogensen, 2009).

11 pathways (Mogensen, 2009). In the case of the AP-1 pathway, TAK-1 mediated phosphorylation of MAPK kinases (MKKs) leads to the phosphorylation of p38 and JNK to activate the AP-1 transcription factor (TF) (Mogensen, 2009). Additionally, TAK-1-mediated activation of the IκB kinases (IKKs) IKKα and IKKβ leads to the phosphorylation and subsequent proteasomal degradation of IκBα, which liberates the NF-κB TF to bind to DNA κB binding sites to induce the transcription of pro-inflammatory genes (Mogensen, 2009). Target genes of the NF-κB TF that are activated upon LPS stimulation include TNF-α, IL-6, and IL-1β (Shakhov et al., 1990;

Libermann and Baltimore, 1990; Hiscott et al., 1993).

TRAF6 and its associated downstream signaling are also activated through the MyD88- independent, or TRIF, pathway whereby TRIF recruits receptor-interacting protein 1 (RIP1) and

TRAF6, which form a complex that activates the subsequent events mentioned above (TAK1,

AP-1, and NF-κB activation) (Mogensen, 2009).

The inflammatory mediators that are upregulated during TLR4 activation by LPS exert further actions in the immune response of sepsis, which will be discussed in the section on the

Immunology of Sepsis.

1.2.4 Pathogenesis

Sepsis is a result of the body’s dysregulated and hyperamplified response to an invading pathogen rather than the pathogen itself (Thomas 1972; Faix 2013). It is not yet understood why the body inappropriately reacts with such an exaggerated, maladaptive inflammatory and immune response (Faix, 2013).

Sepsis has been described as a triphasic process as illustrated in Figure 4, with the first phase being the SIRS, followed by a recovery phase that attempts to restore homeostasis, which

12 Sepsis

Hyperimmune Pro-inflammatory

SIRS

≈ • ⇑Immune activation Attempt to • ⇑ Pro-inflammatory restore Recovery

mediators homeostasis

Immune Status

≈ Death Inflammatory Status

CARS • Immune cell death Immune Paralysis Hypoimmune • Anti-inflammatory • Immunosuppression • ⇓Pro-inflammatory mediators • ⇑Risk of nosocomial infection

Organ Dysfunction

Severe Sepsis Figure 4: The three phases of severe sepsis: the initial systemic inflammatory response syndrome (SIRS), followed by a recovery phase, and finally the compensatory anti-inflammatory response syndrome (CARS) (Hotchkiss and Karl, 2003; Faix, 2013).

13 then overshoots into the compensatory anti-inflammatory response syndrome (CARS) phase

(Bone et al., 1997; Faix, 2013). The SIRS phase is as described previously: a presentation of more than one conditions in the SIRS criteria. In the CARS stage, death of the immune cells B cells, CD4 T-cells, and follicular DCs leads to a dangerous condition of immunodepletion

(Hotchkiss and Karl, 2003). This marked immunosuppression places patients at increased risk of nosocomial infection (Faix, 2013). The immune response of sepsis is discussed in a following section.

1.2.5 Diagnostic Biomarkers of Sepsis

Upon suspicion of sepsis, physicians immediately measure blood lactate concentration as their first step in diagnosing sepsis. There are a few hypotheses as to why lactate levels increase during sepsis, with the prominent theory being that the anaerobic glucose metabolism that occurs in oxygen-deprived tissues produces lactate from pyruvate rather than oxidizing pyruvate, which would otherwise occur in an aerobic condition. The anaerobic glucose metabolism occurs as a result of sepsis-related hypotension and its subsequent hypoperfusion of tissues (Faix, 2013).

A concomitant increase in the production of the three pro-inflammatory cytokines TNF-α,

IL-6, and IL-1β produces a systemic inflammatory response which is a hallmark of early sepsis

(Faix, 2013). The three pro-inflammatory cytokines are produced by the cells of the innate immune system such as macrophages and DCs when activated by PAMPs on invading pathogens

(Hotchkiss and Karl, 2003; Faix, 2013). Of the three pro-inflammatory cytokines, an increase in plasma IL-6 concentration is the best indicator of sepsis prognosis such that increased IL-6 is associated with increased mortality and an increased risk of developing severe sepsis (Patel et al.,

1994; Faix, 2013). Although IL-6 independently serves as the most prognostic cytokine of the

14 Table 3: Diagnostic biomarkers of sepsis (McCulloh, 2012; Hotchkiss and Karl 2003; Faix, 2013; Pierrakos and Vincent; 2010)

Biomarker Stimulated by Action Sepsis specificity Anaerobic glucose Causes Used as a primary diagnostic biomarker in Lactate metabolism caused by organ SIRS with suspected infection, upon which hypoperfusion. dysfunction first-line treatments are initiated. and failure. Indicates organ dysfunction. Produced by macrophages, Activates Not sepsis-specific, TNF-α DCs, and T-cells via TNFR1 general inflammatory activation of PRRs, TNFR, and 2. biomarker, also IL-1R à NF-κB and AP-1 elevated in colitis. transcription factors. Elevated in first 24h Concerted elevation of sepsis. of all three is Produced by macrophages, Activates Not sepsis-specific, characteristic of IL-6 DCs, and T-cells via IL6R. general inflammatory early phases of activation of PRRs, TNFR, biomarker. sepsis IL-1R à NF-κB and AP-1 transcription factors. Increasing IL-6 levels indicate worsening infection. Produced by macrophages, Activates Not sepsis-specific, IL-1β DCs, and T-cells via IL-1R. general inflammatory activation of PRRs, TNFR, biomarker. IL-1R à NF-κB and AP-1 transcription factors. Lung and intestinal production Part of Not sepsis-specific, also elevated in trauma PCT of PCT stimulated by bacterial immune and major surgery. infection. mechanism that leads Able to differentiate bacterial infection vs. to severe noninfectious inflammation. sepsis. Good indicator of sepsis-related mortality and organ failure. Hepatic production of CRP Effect of Not sepsis-specific, also elevated in CRP stimulated by IL-6. CRP inflammatory conditions such as unknown. cardiovascular disease and atherosclerosis.

Considered a sepsis-specific biomarker within the acute onset (initial 24h) of sepsis.

Elevated CRP correlated with organ dysfunction and death.

15 three, TNF-α and IL-1β are both significantly elevated in endotoxin sepsis (Faix, 2013).

Procalcitonin (PCT), the precursor to the hormone calcitonin, has been identified as a sepsis biomarker capable of discriminating between sepsis and SIRS in critically ill patients and of determining whether the etiology of infection is bacterial versus viral (Karzai et al., 1997;

Harbarth et al., 2001). PCT is also an accurate biomarker of the severity of sepsis and serial PCT concentrations can help direct therapeutic strategies (Harbarth et al., 2001). However, PCT is also moderately elevated in other conditions such as trauma or pneumonia (Karzai et al., 1997).

The hepatic production of the acute phase reactant C-reactive protein (CRP) is increased with stimulation by IL-6 during sepsis (Faix, 2013). However, CRP is also elevated in other inflammatory conditions such as cardiovascular disease and atherosclerosis (Benzaquen et al.,

2002). Thus, as with many other biomarkers discussed in this section, CRP is not specific to sepsis but is an important indicator of sepsis when analyzed together with all of the other aforementioned sepsis biomarkers to give a complete “biomarker profile” upon which to base diagnosis and prognosis.

1.2.6 Diagnostic Criteria

Table 1 outlines the diagnostic criteria that physicians use to diagnose SIRS and sepsis

(Balk et al., 1992; Levy et al., 2003). Patients being admitted to the ICU, as well as at-risk patients being admitted in the emergency department are screened automatically using the SIRS criteria. It is important to note that the 2001 International Sepsis Definitions Conference developed the SIRS definition and criteria to be applicable in both human clinical settings and research laboratory settings (Levy et al., 2003). This supports the relevance of the present study’s findings in human sepsis research and development of potential therapeutic targets.

16 1.2.7 Treatment of Sepsis

Although by definition sepsis is caused by an infection, many patients presenting with clinical sepsis have negative blood cultures and, therefore, physicians cannot base their diagnosis upon this criterion, and instead base diagnosis on SIRS criteria (Balk et al., 1992; Levy et al.,

2003; O’Brien et al., 2007). An early diagnosis and initiation of treatment increases the rate of survival and reduces the risk of morbidities such as permanent organ damage (Rivers et al.,

2001; Kumar et al., 2006). When culture results do become available, the treatment course can be modified to treat the identified pathogen (O’Brien et al., 2007). There have been many experimental treatments tested clinically including those that target the first phase (SIRS) such as

TNF antagonists, corticosteroids, antiendotoxin antibodies, and IL-1R antagonists (Hotchkiss and Karl, 2003). However, it is known that SIRS transitions into the recovery and CARS phases and therefore patients became significantly immunosuppressed with these anti-inflammatory approaches and susceptible to nosocomial infections (Hotchkiss and Karl, 2003).

Treatments that are used currently include broad-spectrum antibiotics, vasopressive therapy, intensive insulin therapy, corticosteroids, dialysis, and fluid resuscitation (Balk et al.,

1992; Hotchkiss and Karl, 2003; Kumar et al., 2006; van der Poll and Opal, 2008). The anticoagulant human recombinant activated protein C, marketed by Eli Lilly as Xigris®, was determined to be an effective anti-inflammatory in sepsis (Hotchkiss and Karl 2013), however was pulled off the market in 2012, a mere 11 years after FDA approval, due its controversial and dubious clinical trials (Bernard et al., 2001; Ranieri et al., 2012; Faix, 2013; Hotchkiss and Karl,

2013). Thus, these treatments lack efficacy and there is a need for an effective treatment for sepsis to decrease its alarming rates of mortality and morbidity.

17 Table 4: Treatments for sepsis (Hotchkiss and Karl, 2003; van der Poll and Opal, 2008).

Treatment Action Broad-spectrum antibiotics To kill or inhibit the growth of bacteria prior to identification of pathogen Bacteria-specific antibiotic To kill or inhibit the growth of culture-identified bacteria Intensive insulin therapy To treat sepsis-induced hyperglycemia

Vasopressor therapy To increase blood pressure in hypotensive patients Fluid resuscitation To optimize cardiac preload, afterload, and contractility Corticosteroids Adrenal suppression

Use is controversial as it may lead to secondary infections. Dialysis To treat renal failure Recombinant human Anticoagulant activated protein C (Xigris®) Anti-inflammatory

Lacks efficacy, removed from market by FDA due to dubious research trials

18 1.2.8 Vulnerable Populations

Populations that are more susceptible to development of sepsis and a poor prognosis upon diagnosis are those with immunosuppressive conditions such as the geriatric population, neonatal patients, and the immunocompromised and critically ill, such as HIV positive individuals are at an increased risk of developing and succumbing to sepsis (Balk et al., 1992; Angus et al., 2001).

1.2.9 Prevalence and Cost of Treatment

There are an estimated 18 million cases of sepsis globally per year, 750,000 of which are in the United States (Angus et al., 2001; Slade et al., 2003). Studies estimate 1 million cases of sepsis will occur in the United States in the year 2020 (O’Brien et al., 2007).

The age distribution of sepsis patients includes a high incidence in infants less than 1 year of age (5.3 cases per 1000 individuals), lower incidence in children aged 5-14 (0.2/1000), higher in adults aged 60-64 (5.3/1000), and highest in geriatric patients greater than 85 years of age

(26.2/1000) (Angus et al., 2001).

The mortality rate of sepsis is 30% and is the leading cause of death in critically ill patients in the developed world (Angus et al., 2001; Slade et al., 2003; Pierrakos and Vincent,

2010). In the year 1995, the 215,000 deaths from sepsis in the United States accounted for 9.3% of all deaths (Angus et al., 2007). This is an alarming number, and with the estimated per annum increase in sepsis incidence of 1.5% and without any significant decrease in mortality rate, sepsis poses a grave health and economic concern (Angus et al., 2001). Various international committees have been formed with the commitment to decrease the rate of mortality, such as the

Surviving Sepsis campaign established 2003 which endeavored to decrease mortality to 25% over a five-year period (Slade et al., 2003).

19 The average cost of treating a patient with sepsis is $22,000 USD and requires an average of 19.6 days in hospital (values published in 2001, account for inflation) (Angus et al., 2001). In total, sepsis costs the United States health care system $16.7 billion annually, compared to the cost of cancer at approximately $157 billion and the cost of diabetes at $245 billion (Angus et al., 2001; Centre for Disease Control).

1.3 Immunology of Sepsis

Pathogenic microorganisms as well as those that are poorly pathogenic but opportunistic in a host with a compromised physical defense are capable of mounting an immune response. As previously described, structural motifs on invading pathogens called PAMPS are recognized by

PRRs expressed by cells of the innate immune system, such as macrophages and DCs (Hotchkiss and Karl, 2003; Leon et al., 2006; Faix, 2013). The activation of PRRs by PAMPs and the subsequent signaling produces the induced immune response which informs both the innate and adaptive immune responses. In Gram-negative bacteria, LPS found within the outer membrane is a PAMP recognized by the PRR toll-like receptor 4 (TLR4) and to a lesser degree, TLR2 (Yang et al., 1999; Takeda and Akira, 2004). LPS activation of TLR4 on the cells of the innate immune response activates the AP-1 and NF-κB inflammatory pathways and stimulates the production of the pro-inflammatory cytokines TNF-α, IL-6, and IL-1β (Kobayashi et al., 2004; Faix, 2013).

These three pro-inflammatory mediators mount pro-inflammatory effects, by binding to their own respective receptors (ex. TNF-α binding to TNFR and IL-1β binding to IL-1R) and by stimulating production of acute phase proteins in target organs (ex. IL-6-stimulated liver production of C-reactive protein), which promotes further inflammatory events. TLR4 activation also produces type 1 interferons (IFNs) and inducible nitric oxide synthase (iNOS), effectively

20 increasing nitric oxide (NO) concentrations, which induces vasodilation (Takeda and Akira,

2004; Sriskandan and Altmann, 2008)

The complement system is another element of the innate immune response that can directly kill invading pathogens as is the case with complement 5 – complement 9 (C5-C9), as well as attach to the surface of invading microbes to target them for destruction by neutrophils

(C3b) (Sriskandan and Altmann, 2008). Host mannose-binding proteins can detect glycosylated proteins that are expressed in low levels on the surface of bacteria and work cooperatively with the complement system to increase C3b attachment for neutrophil-mediated death (Sriskandan and Altmann, 2008). Small peptides called cathelicidins also serve in the first line of defense, in which their pore-forming activity has the ability to kill both Gram negative and Gram positive bacteria (Sriskandan and Altmann, 2008).

The cells of the adaptive immune response, B cells and T cells, are activated by both classical antigen presentation by APCs (such as DCs), but also directly by TLR which activates

T cells (Tregs) expansion (Sriskandan and Altmann, 2008). While the host’s ultimate goal is to clear the bacterial infection, its efforts to do so exert sinister and damaging effects.

1.4 Bioactive Peptides

Food-derived bioactive peptides are now being recognized for their health-promoting properties such as antimicrobial, antihypertensive, antithrombotic, and immunomodulatory effects (Nagaoaka et al., 2001; Katayama et al., 2006; Lee et al., 2006; Kovacs-Nolan et al.,

2006; Hartmann and Meisel, 2007; Young et al., 2012; Iskandar et al., 2013; Majumder et al.,

2015; Kodera and Nio, 2015). This review will examine bioactive peptides with anti- inflammatory properties, of which some are outlined in Table 5. Bioactive peptides are short

21 peptides 2-20 residues in length that are released from their native dietary proteins into their active forms by enzymatic hydrolysis, for example by the gastrointestinal enzymes trypsin and pepsin, or by microbial fermentation (Korhonen and Pihlanto, 2005; Hartmann and Meisel,

2007). Bioactive peptides have been isolated from many food sources including milk, eggs, soy, meat, and fish. (Kovacs-Nolan et al., 2006; Huang et al., 2010; Young et al., 2012; Iskandar et al., 2013; Shi et al., 2014; Ortega-González et al., 2014; Kobayashi et al., 2015; Kobayashi et al.,

2015). While many bioactive peptides and their anti-inflammatory properties continue to be identified, much remains to be reported in terms of their mechanisms of action.

Examples of bioactive peptides with anti-inflammatory activity include the PepT1- transportable soy-derived tripeptide VPY that decreased IL-8 secretion in TNF-α-stimulated

IECs and decreased TNF-α secretion in LPS-stimulated THP-1 macrophages. VPY also had anti- inflammatory activity in vivo in mice with DSS-induced colitis as measured by decreased MPO activity, a marker for neutrophil infiltration, colon histology, and gene expression levels of various pro- and anti-inflammatory mediators (Kovacs-Nolan et al., 2006). Poly-L-lysine, a basic polypeptide capable of allosterically activating CaSR had anti-inflammatory effects in TNF-α- stimulated Caco-2 and HT-29 cells as demonstrated by reduced IL-8 secretion, as well as reduced the expression of the pro-inflammatory genes IL-8, IL-6, TNF-α, and IL-1β. Poly-L- lysine also had anti-inflammatory activity in mice with DSS-induced colitis as measured by body weight, clinical symptoms, colon length, colon morphology, colonic secretion of IL-6 and TNF-

α, and the expression of pro-inflammatory genes (IL-6, TNF-α, IL-1β, IL-17) (Mine and Zhang,

2015). Another allosteric activator of CaSR, L-Trp has been reported to have anti-inflammatory activity in TNF-α-stimulated Caco-2 and HT-29 and in piglets with DSS-induced colitis (Kim et al., 2010; Mine and Zhang, 2015).

22 The egg-derived tripeptide IRW had anti-inflammatory activity in an in vitro model of vascular inflammation. IRW reduced the expression of the intercellular adhesion molecule-1

(ICAM-1), monocyte chemoattractant protein-1 (MCP-1), and vascular cell adhesion molecule-1

(VCAM-1) in TNF-α-stimulated human umbilical vein endothelial cells (HUVECs). Treatment with IRW was also reported to have an effect on the NF-κB pathway by inhibiting the nuclear translation of the NF-κB subunit p65 (Huang et al., 2010). Various other anti-inflammatory food- derived bioactive peptides are described in Table 5.

The identification of bioactive peptides and their mechanisms of action provide the opportunity to develop new bioactive peptide-containing functional foods, as well as a new angle for the marketing of existing products. Future research should focus largely on clarifying the exact mechanisms of food-derived anti-inflammatory peptides. This knowledge will help to direct the development of effective nutraceutical products, as well as aid in identifying more bioactive peptides.

1.5 γ-EV

Gamma-L-glutamyl-L-valine (γ-EV) is a dietary dipeptide comprised of glutamic acid and valine residues. The ‘gamma’ nomenclature denotes the peptide bond between the two amino acids involving the γ carbon in glutamic acid’s backbone, which is essential to its ability to bind and activate CaSR (Ohsu et al., 2010). The γ-peptide bond also confers γ-EV its stability throughout the gastrointestinal (GI) tract, which is important as it permits the dipeptide to maintain activity throughout its entire passage through the GI tract.

γ-EV is currently marketed by Ajinomoto (Japan) as a kokumi flavour-enhancing compound that enhances a variety of tastes including sweet, salty, and umami (Maruyama

23 Table 5: Food-derived bioactive peptides and their anti-inflammatory activities

Food source Peptide Anti-Inflammatory Activity Reference Egg Ile-Arg-Trp (IRW) Reduced ICAM-1, VCAM-1, and Huang et al., 2010. MCP-1 production in TNF-α- stimulated HUVECs.

Inhibited the nuclear translation of p65 in TNF-α-stimulated HUVECs. Soy Val-Pro-Tyr (VPY) Reduced IL-8 secretion in TNF-α- Kovacs-Nolan et al., stimulated Caco-2 cells. 2012

Reduced TNF-α secretion in LPS- stimulated THP-1 macrophages.

Anti-inflammatory in DSS-induced mouse colitis. Soy Soy-derived di- and Anti-inflammatory in DSS-induced Young et al., 2012 tripeptides porcine colitis.

Whey Hydrolysates of Decreased IL-8 secretion in LPS- Iskandar et al., 2013 pressurized and stimulated respiratory epithelial cell native whey protein lines CFTE29o- and 1HAEo-. Beans, brain, γ-EV, γ-EC In TNF-α-stimulated Caco-2 cells: Zhang et al., 2015 onions - Reduced IL-8 secretion. - Reduced expression of the pro- inflammatory genes IL-8, TNF-α, IL- 6, and IL-1β. - Increased expression of the anti- inflammatory gene IL-10. - γ-EC inhibited phosphorylation of JNK and IκB.

Anti-inflammatory in DSS-induced mouse colitis. L-Trp Reduced IL-8 secretion in TNF-α- Mine and Zhang, 2015 stimulated Caco-2 and HT-29 cells.

Reduced phosphorylation of JNK and IκBα in HT-29 cells.

Anti-inflammatory in DSS-induced Kim et al., 2010 porcine colitis. Bovine milk Glycomacropeptide Anti-inflammatory effects in Ortega-González et (GMP) lymphocyte-transfer model of colitis al., 2014 in mice. Produced by Poly-L-lysine Reduced IL-8 secretion in TNF-α- Mine and Zhang, 2015 Streptomyces stimulated Caco-2 and HT-29 cells. albulus during fermentation Reduced expression of the pro- inflammatory genes IL-8, IL-6, TNF- α, and IL-1β in TNF-α-stimulated IECs.

Anti-inflammatory in DSS-induced mouse colitis. Sardine muscle Trp-His Reduced IL-8 secretion in TNF-α- Kobayashi et al., 2015 hydrolysate stimulated HT-29 cells.

Inhibited activation of p38, ERK, JNK, and IκBα in TNF-α-stimulated HT-29.

Anti-inflammatory in DSS-induced colitis in mice. Egg white Ovotransferrin Anti-inflammatory in DSS-induced Kobayashi et al., 2015 colitis in mice. Eggshell Hydrolysate from Reduced IL-8 secretion in TNF-α- Shi et al., 2014 membrane eggshell membrane stimulated Caco-2 cells.

Anti-inflammatory activity in DSS- induced mouse colitis.

25 et al., 2012). As a kokumi substance, γ-EV itself has a slightly astringent taste independently but increases the mouthfulness, complexity, and palate length of the above-mentioned tastes (Dunkel et al., 2007). γ-EV can be found in beans (Phaseous vulgaris L.), onion (Allium cepa), bovine brain, and beech nut (Fagus sylvatica L.) seeds (Virtanen and Matikkala, 1960; Kanazawa et al.,

1965; Kristensen et al., 1974; Dunkel et al., 2007). γ-EV can be purified and identified from these foods using acid hydrolysis and ion-exchange chromatographic techniques (Kanazawa et al., 1965). γ-EV can also be synthesized as previously reported by Rowlands and Young (1952), or a more recent protocol by Suzuki et al. (2004) utilizes the bacterial enzyme γ-glutamyl- transpeptidase (GGT). The optimal conditions for Suzuki et al.’s GGT protocol require 20 mM glutamine (Gln) (as a cost-effective γ-glutamyl donor), 300 mM Val, and 0.04 U/mL GGT at pH

10 for a 3h incubation at 37oC, which produce 17.6 mM γ-EV (88% yield) (Suzuki et al., 2004).

γ-EV has recently been identified as a bioactive peptide, as described by Zhang et al.

(2015) in their study on a DSS-induced colitis mouse model and TNF-α-stimulated intestinal epithelial cells (IECs). In this study, mice with DSS-induced colitis were pretreated with either γ-

EC and γ-EV administered via drinking water. Both γ-EC and γ-EV had anti-colitis activity in mice as measured by colon length, body weight, colonic histological parameters such as inflammatory cell infiltration and mucosal integrity, IL-6 and TNF-α concentrations in colonic homogenates, as well as the mRNA expression profiles of various relevant inflammatory meditators in colon tissue. Zhang et al. (2015) also examined the anti-inflammatory activity of γ-

EC and γ-EV in vitro in TNF-α-stimulated IECs. Caco-2 cell monolayers were pretreated with γ-

EC and γ-EV and then stimulated with TNF-α. Compared to the TNF-α positive control group,

Caco-2 cells treated with γ-EC or γ-EV had a dose-dependent decrease in the secretion of the

Figure 5: Structure of γ-L-glutamyl-L-valine (From The Human Metabolome Database).

27 chemokine IL-8. As well, γ-EC and γ-EV treatment reduced the mRNA expression levels of IL-

8, IL-6, TNF-α, and IL-1β (Zhang et al., 2015). These recent findings of the anti-inflammatory activities of γ-EC and γ-EV in a DSS-induced colitis mouse model and TNF-α-stimulated IECs contribute greatly to the rationale for the present project.

1.6 Calcium-Sensing Receptor (CaSR)

1.6.1 Background

First discovered in the parathyroid gland for its role in calcium (Ca2+) homeostasis the calcium-sensing receptor (CaSR) is a class C G-protein-coupled receptor (GPCR) (Hofer and

Brown, 2003). CaSR has several functions including ion homeostasis, hormone regulation, and nutrient sensing (Hofer and Brown, 2003; Hebert et al., 2004). GPCRs are seven-transmembrane domain receptors with associated intracellular heterotrimeric G-proteins (Conigrave and Ward,

2013; Tfelt-Hansen and Brown, 2013). Other class C GPCRs include the glutamate metabotropic receptors (mGluRs) and γ aminobutyric acid type B receptor (GABAB) (Tfelt-Hansen and

Brown, 2013). The vital role of CaSR in calcium regulation was demonstrated in CaSR-null mice that developed severe hyperparathyroidism caused by the inability to regulate extracellular Ca2+- dependent PTH secretion (Ho et al., 1995).

1.6.2 Structure and Function of CaSR

CaSR has a large amino-terminal ECD consisting of 612 residues, 7 transmembrane- spanning domains (TMDs) made up of 1085 residues, and a carboxy-terminal intracellular domain of 126 residues (Hofer and Brown, 2003; Tfelt-Hansen and Brown, 2013). The ECD of

CaSR forms a homodimer, as well as a heterodimer with other GPCRs such as mGluR1 and

28 mGluR5 (Hofer and Brown, 2003). Its dimer orientation is stabilized by two covalent, disulfide bonds between cysteine residues 129 and 131, and noncovalent hydrophobic linkages (Tfelt-

Hansen and Brown, 2013). The two ECDs of the CaSR dimer form a bilobed Venus-flytrap

2+ domain (VFT) that undergoes conformational changes upon binding and activation by Ca

(Tfelt-Hansen and Brown, 2013). The VFT conformational changes, as well as, changes in the

TMD initiate its intracellular signaling events (Tfelt-Hansen and Brown, 2013).

2+ CaSR is capable of detecting minute fluctuations (~200 µM) in the physiological Ca o concentration of 1.1-1.3 mM (Hofer and Brown, 2003; Brown et al., 2013). CaSR is inactive

2+ 2+ when the [Ca o] is less than 0.2 mM and active when the [Ca o] is 0.5-2.0 mM (Conigrave and

2+ Ward, 2013). When Ca o exceeds 1.3 mM, CaSR suppresses the parathyroid gland’s release of parathyroid hormone (PTH), which activates calcitonin (CT) to stimulate increased excretion of calcium in the kidneys (Hofer and Brown, 2003). The PTH-stimulated calcium resorption in the

2+ kidney is reduced when [Ca o] >1.3mM, thus contributing to the excretion of calcium via the urine (calciuria) (Tfelt-Hansen and Brown, 2013). Hypercalcemia also decreases net release of calcium from bones through its inhibition of the active vitamin D metabolite 1,25- dihydroxyvitamin D3 (1,25(OH)2 D3) production, as well as via activation of the osteoblast PTH receptors (PTHR) (Hofer and Brown, 2003; Tfelt-Hansen and Brown, 2013). The reduction of net calcium release from bone is another component of CaSR-mediated calcium homeostasis.

2+ Low Ca o stimulates the secretion of PTH by the parathyroid gland which promotes

2+ 1,25(OH)2 D3-mediated Ca uptake in the intestine and suppresses calciuria (Hofer and Brown,

2003). The net release of calcium from the bone matrix is a consequence of continuous PTH secretion, rather than short bursts in PTH elevation, which is actually beneficial in the treatment of osteoporosis (Gowen et al., 1995; Hofer and Brown, 2003). Thus, it is evident that a tightly

29 2+ controlled Ca o concentration is necessary for the proper functioning of several integrated systems, and CaSR is an indispensable component of calcium control.

1.6.3 Agonists and Antagonists

CaSR is found in various tissues including the parathyroid gland, kidneys, bone, stomach, small intestine, colon, brain, and mouth (Conigrave et al., 2002; Maruyama et al., 2012).

Although its main ligand is Ca2+, it exhibits ligand promiscuity and ligand-based signaling

(Conigrave and Ward, 2013). Type I CaSR agonists include Ca2+, Mg2+, polylysine, spermine, gadolinium, and neomycin (Tfelt-Hansen and Brown, 2013; Brown et al., 2013). Type II agonists include L-amino acids, polyamines, and γ-glutamylpetides (Maruyama et al., 2012; Tfelt-Hansen and Brown, 2013; Brown et al., 2013). Type II agonists function as allosteric activators, such that their binding increases the receptor’s sensitivity to its ‘main’ type I ligands, effectively co-

2+ activating CaSR in subthreshold Ca o concentrations (Hofer and Brown, 2003; Mun et al.,

2004). Synthetic type II agonists known as calcimimetics are used to treat conditions such as

2+ o hypercalcemia in parathyroid carcinoma, regulation of Ca o in mild 1 hyperparathyroidism, and elevated serum PTH in stage V chronic kidney disease (CKD) (Tfelt-Hansen and Brown, 2013).

2+ While type I agonists bind and activate CaSR regardless of Ca o concentration, allosteric

2+ agonists require a tight range of [Ca o] to bind to the receptor (Tfelt-Hansen and Brown, 2013).

CaSR antagonists, or calcilytics, inhibit the receptor and its associated signaling (Tfelt-

Hansen and Brown, 2013). NPS-2143 is a calcilytic originally discovered by high-throughput methods and is used routinely as a CaSR antagonist in research lab protocols (Hofer and Brown,

2003). Calcilytics are also used in medical applications to stimulate endogenous PTH surges as a treatment for osteoporosis (Gowen et al., 1995; Tfelt-Hansen and Brown, 2013).

30 The role of CaSR in the GI tract is of particular interest to this project where it functions in nutrient sensing, protein metabolism, epithelial growth and differentiation, and regulation of ion homeostasis and gastric acid secretion (Hebert et al., 2004).

1.6.4 Binding Site for L-Amino Acids

The binding site for L-amino acids on CaSR, as determined by mutational and chimeric receptor analysis, is distinct from the Ca2+ binding sites (Mun et al., 2004). While the four Ca2+

57 90,116 binding sites reside in each CaSR monomer’s VFT and HH domains, the L-amino acid binding site exists in a highly conserved region within the VFT (Conigrave and Ward, 2013).

The conserved binding domain for L-amino acids within the VFT has been identified as residues

S147, S170, D190, Y218, and E297 (Conigrave and Hampson, 2006). The γ peptide bond in γ- glutamyl peptides is essential for VFT binding as it exposes the carboxy and amino groups making them available for binding (Ohsu et al., 2010). Allosteric activation of CaSR by

γ−glutamyl peptides requires a basal Ca2+ concentration of 0.75 mM (Ohsu et al., 2010), a requirement that is satisfied in physiologic conditions.

γ-EV’s flavour-enhancing properties arise from its allosteric activation of CaSR on the tongue, specifically on the posterior tongue on circumvallate taste cells (as observed in mice)

(Maruyama et al., 2012). Maruyama et al. (2012) showed that the CaSR-mediated kokumi taste effect is elicited through the PLC-mediated release of stored intracellular Ca2+ rather than Ca2+ influx.

31 1.6.5 CaSR Signaling

Upon binding Ca2+, the VFT undergoes conformational changes and a cysteine-rich domain within the ECD located between the VFT and the TMD aids in facilitating the signal transduction into the TMD (Brown et al., 2013). The signal transduction into the TMD causes a conformational change in its 7-transmembrane loops which causes G-protein activation within the cytosol (Brown et al., 2013). G-protein binding and activation is an essential part of CaSR, that is, of all class C GPCRs signaling (Conigrave and Ward, 2013). Specifically, CaSR signals through the following G-proteins: Gq/11, G12/13, and Gi/o, (Conigrave and Ward, 2013). Signaling through Gq/11 activates the PLC-activated inositol-1,4-5-triphosphate (IP3)-mediated release of intracellular Ca2+ stores into the cytoplasm (Conigrave and Ward, 2013). The activation of the

G12/13 protein stimulates phospholipase D (PLD) activation, actin polymerization, and membrane

ruffling (Conigrave and Ward, 2013). Lastly, CaSR signals through the Gi/ο protein to inhibit adenylate cyclase and to decrease cyclic AMP (cAMP) concentration (Conigrave and Ward,

2013). The Gi/ο-mediated recruitment of the scaffolding and desensitizing protein β-arrestin2 accounts for the anti-inflammatory activity by both γ-EC- and L-Trp-mediated CaSR allosteric activation in TNF-α-stimulated intestinal epithelial cells (Zhang et al., 2015; Mine and Zhang,

2015). In this signaling crosstalk scheme, activation by either γ-EC- and L-Trp leads to the direct interaction of β-arrestin2 with TAB1 which inhibits its complexation with its binding partners

TAK1 and TAB2 to effectively inhibit further signaling events and thus suppress inflammation

(Zhang et al., 2015; Mine and Zhang, 2015). β−Arrestin1 and 2 also interact directly with IκBα to inhibit its phosphorylation, which prevents the liberation of the NF-κB TF and its nuclear translocation, thus effectively downregulating TNF-α-stimulated NF-κB-mediated inflammation

(Witherow et al., 2004). Gao et al. (2004) also established the direct interaction of β-arrestin2

32 CaSR monomer Cys129, Cys131

2+ Ca o binding Extracellular site VFT domains

cysteine-rich region

7 TM-spanning loops

G12/13 Intracellular G Gi/o q/11 G-proteins

PLC PLD β−arrestin2 AC

IP3 Actin polymerization Membrane ruffling cAMP

Release of intracellular Ca2+ stores

Figure 6: The calcium-sensing receptor (CaSR), a G-protein-coupled receptor and its associated intracellular signaling pathways. The extracellular Venus flytrap (VFT) domain contains the 2+ binding sites for both Ca o and γ-EV. Cysteine 129 and Cysteine 131 in each CaSR monomer participate in disulfide bonds that stabilize the receptor’s dimer orientation. The recruitment and activation of the Gq/11 protein leads to the activation of phospholipase C (PLC) and the IP3- 2+ mediated mobilization of intracellular Ca stores. CaSR-mediated activation of the G12/13 protein leads to activation of phospholipase D (PLD) and the subsequent actin polymerization and membrane ruffling. CaSR-mediated activation of Gi/o leads to the inhibition of adenylyl cyclase (AC) which suppresses production of cAMP. Of particular interest to the present project is the CaSR-activated Gi/o-mediated recruitment of β-arrestin2 which interacts directly with proteins in other signaling domains. (Gao et al., 2004; Wang et al., 2006; Conigrave and Ward, 2013; Tfelt- Hansen and Brown, 2013; Zhang et al., 2015.)

33 with IκBα to inhibit its activation and subsequent detachment from NF-κB, and determined that

β-arrestin2-IκBα interaction actually stabilized IκBα thus increasing its inhibitory effect on NF-

κB nuclear translocation. Wang et al. (2005) identified TRAF6 as another binding partner of β- arrestins, in which β-arrestins inhibit the polymerization and autoubiquitination of TRAF6 in

LPS- and IL-1β-stimulated TIR signaling, effectively inhibiting the activation of TRAF6 and any further signaling events. It is clearly evident that β-arrestins are capable of exerting their effects at multiple levels of the TNFR and TIR signaling cascades by direct interactions with signaling proteins to downregulate the subsequent inflammatory effects.

34 2 RATIONALE, HYPOTHESIS AND OBJECTIVES

2.1 Rationale

γ-EV has shown promising potential as a bioactive peptide as demonstrated in its anti- inflammatory activity in a mouse model of DSS-induced colitis and in TNF-α-stimulated intestinal epithelial cells (IEC) (Zhang et al., 2015). The similarities in the intracellular signaling pathways of the TNFR and TLR4 receptors provide evidence that γ-EV may exert similar β- arrestin2-mediated anti-inflammatory crosstalk effects to suppress LPS-induced TLR4-mediated inflammatory signaling. In doing so, this project endeavours to not only identify alternative health-promoting properties of the bioactive dietary dipeptide γ-EV, but also a signaling mechanism that may provide insight into alternative approaches for the treatment of sepsis.

2.2 Hypothesis

Treatment with the dietary dipeptide γ-EV can suppress inflammation in BALB/c mice with

LPS-induced sepsis.

2.3 Objectives

Objective 1: To determine if γ-EV has anti-sepsis activity in an LPS-induced mouse model of sepsis.

Objective 2: To identify the cell signaling crosstalk mechanism between the LPS-activated

TLR4 and γ-EV-allosterically activated CaSR signaling pathways behind γ-EV’s anti-sepsis activity.

35 3. MATERIALS & METHODS

3.1 Reagents

γ-EV was generously provided by Ajinomoto Co, Inc. (Kawasaki, Japan).

Lipopolysaccharide (LPS) from Escherichia coli (E. coli) 0111:B4 was used to induce sepsis

(Sigma, Oakville, Canada). The following protease inhibitors were used while extracting protein from small intestine: phenylmethanesulfonyl fluoride (PMSF) (Sigma), aprotinin (Sigma), and pepstatin A (Sigma). TNF-α, IL-6, IL-1β, and IL-10 ELISA experiments were conducted with

Ready-SET-Go! ELISA kits as per manufacturer’s instructions (eBioscience, Affymetrix, Inc.,

San Diego, CA). The following antibodies were used for Western blot (WB) experiments of small intestine homogenate: SAPK/JNK rabbit pAb (Cell Signaling, Danvers/Beverly, MA), phospho-SAPK/JNK rabbit pAb (Cell Signaling), IκBα mouse mAb (Cell Signaling), phospho-

IκBα mouse mAb (Cell Signaling), TRAF6 rabbit pAb (Santa Cruz Biotechnology, Dallas, TX), ubiquitin mouse mAb (Santa Cruz Biotechnology), β-actin mouse mAb (Cell Signaling), β-actin rabbit mAb (Cell Signaling), HRP conjugated anti-rabbit IgG (H+L) (Promega, Madison, WI), and human adsorbed-HRP goat anti-mouse IgG (Southern Biotech, Birmingham, AL). Western blots were detected using AmershamTM ECLTM Prime Western Blotting Detection Reagent (GE

Healthcare, Mississauga, Canada).

3.2 Animal Trial

3.2.1 Animals

The animal study was approved by the University of Guelph Animal Care Committee and was performed in accordance with the Canadian Council on Animal Care Guide to the Care and

36 Use of Experimental Animals. The Animal Utilization Protocol (AUP) number for the animal study is AUP1362. The mice were housed in the Central Animal Facility (CAF) at the University of Guelph (Guelph, ON) for the duration of study.

3.2.2 Treatment

Female BALB/c mice (6-8 weeks, 18-20g, 8 mice/group; Charles River Laboratories,

Montreal, QC) were housed on a 12h light/dark cycle and permitted ad libitum access to water and standard mouse chow. γ-EV was administered daily via oral gavage for 8 days, with the 8th day being the day of LPS stimulation. There were 5 treatment groups with 8 mice per group

(n=8). The γ-EV low-dose (LD) group received a 100 µL bolus of 50 mg/kg bodyweight (BW) γ-

EV in filter-sterilized PBS, adjusted for physiological pH (7.0-7.4). The γ-EV high-dose (HD) group and γ-EV vehicle group both received a 100 µL bolus of 150 mg/kg BW γ-EV in filter- sterilized PBS. Negative control mice did not receive an oral gavage. The administration of γ-EV on the sepsis induction day (day 8) was done precisely 3h prior to the LPS injection. γ-EV solutions were prepared every morning based on the previous days’ mean body weight of each treatment group. Oral gavage was chosen for the administration of γ-EV to control the exact treatment dose.

3.2.3 Sepsis Induction

Sepsis was induced by an intraperitoneal (IP) injection of LPS (10 µg/mouse or 0.5 mg/kg BW, E. coli 0111:B4 in endotoxin-free water). Immediately following sepsis induction, mice were monitored for clinical signs and to ensure welfare. Mice were euthanized exactly 2h after sepsis induction by CO2 asphyxiation performed by a CAF technician. Blood was collected

37 by cardiac puncture by a CAF technician immediately after mice were euthanized.

3.2.4 Tissue Collection

Blood collected by cardiac puncture immediately after CO2 asphyxiation was temporarily stored in EDTA-coated tubes and then centrifuged (2500g, for 10 minutes) shortly thereafter.

Plasma was collected and stored at -80°C until further analysis by ELISA. The small intestine

o was flash-frozen in liquid nitrogen (LN2) and stored at -80 C for future analysis by ELISA and

WB.

2.2.5 Tissue Protein Extraction and Protein Concentration

Protein from small intestine tissue was extracted using the following protocol: small intestine tissue was homogenized on ice in five times volume ice-cold PBS containing protease inhibitors (1 mM PMSF, 10 µg/mL aprotinin, 10 µg/mL pepstatin A) for 2 min using a homogenizer (Polytron PT 1200, Kinematica, AG, Luzern, Switzerland). Protein-containing supernatant was collected after tissue homogenate was centrifuged at 8000g for 30 minutes at

4°C. Protein concentration was measured by DC Protein Assay (Bio-Rad, Mississauga, Canada), using bovine serum albumin (BSA) (Fisher Scientific, Whitby, Canada) as a standard, and bovine albumin (97% purity) (Sigma A-2153) as an internal reference standard. Protein samples used for cytokine detection by ELISA were stored at 4°C. Protein samples used for protein concentration semi-quantification by WB were immediately diluted using PBS and Sample

Buffer (2x Laemmli Concentration) (Sigma) for a final concentration of 2.5 µg/µL protein and stored at -30°C.

38 3.3 ELISA

TNF-α, IL-6, IL-1β, and IL-10 concentrations were measured by ELISA according to manufacturer’s instructions (eBioscience). 96-Well ELISA plates (Corning, Corning, NY) were incubated overnight with 100 µL per well of capture antibody diluted in 1x coating buffer

(eBioscience) at 4°C. Wells were washed 3x with PBS-0.05% Tween 20 (PBST) using a BioTek microplate washer (BioTek, Winooski, VT), as well as between each subsequent step. Wells were then blocked with 200 µL 1x assay diluent and incubated for 1h at 37°C with shaking.

Standards were diluted according to each cytokine’s standard detection range (TNF-α 15.625-

1000 pg/mL, IL-6 15.625-1000 pg/mL, IL-1β 15.625-1000 pg/mL, and IL-10 62.5-4000 pg/mL).

The sample protein/plasma and cytokine standards (100 µL) were added to the wells and incubated for 2h at 37°C with shaking. Wells were washed and then 100 µL detection antibody diluted in assay diluent was added to the wells and incubated for 1h at 37°C with shaking. Wells were washed and 100 µL avidin horseradish peroxidase (Av-HRP) diluted in assay diluent was added to wells and incubated for 30 min at 37°C with shaking. Wells were washed one last time and 100 µL substrate solution 3,3′,5,5′-tetramethylbenzidine (TMB) was added to the wells. The

ELISA plate was wrapped in aluminum foil and incubated for 15 min at 37°C with shaking, checking every few minutes for colour development. The TMB reaction was terminated with 50

µL 1.0M sulphuric acid (H2SO4). The sample and standard absorbance were then read at 450 nm using a microplate reader (iMark model 550, Bio-Rad). Cytokine concentrations are expressed as picogram cytokine per mililitre plasma (pg cytokine/mL plasma) and picogram cytokine per miligram protein (pg cytokine/mg protein). Each small intestine sample was tested using two wells per sample and standard, and assays were done in duplicate to confirm reproducibility.

Plasma TNF-α and IL-1β were tested using two wells per sample and standard. Plasma IL-6 was

39 tested using one well per sample (two per standard), however sample absorbance were pooled post-hoc.

3.4 Western Blotting

Small intestine protein samples were loaded into an SDS-PAGE gel consisting of a 9% resolving gel and a 4% stacking gel. Each lane was loaded with 25 µg (10 µL) of protein and run at 140V for 10 min and then at 120V for 60 min. Proteins were then transferred from the SDS-

PAGE gel to a nitrocellulose membrane for 90 min at 80V in ice-cold transfer buffer.

Nitrocellulose membranes were then blocked in 5% BSA in TBS for 2h at room temperature with shaking. Membranes were then incubated with the first primary antibody – either phospho-

IκBα or phospho-SAPK/JNK (1:4000 (v/v) diluted in 5% BSA in Tris-buffered saline solution containing Tween20 (TBST) overnight at 4oC on a rotating table. Membranes were washed 5x for 8 min each in TBST with shaking at room temperature. Membranes were then incubated with secondary antibody (1:8000 (v/v) in 5% BSA in TBST) for 1.5h with shaking at room temperature. Membranes were then washed 5x for 8 min each, then once with ddH2O for 5 min,

TM and then stored in ddH2O. Membranes were detected using ECL solution, ChemiDoc XRS+

Molecular Imager (Bio Rad) and Image Lab Software (Bio Rad). After membranes were washed with ddH2O once for 5 min, they were stripped using the following mild stripping protocol: mild stripping buffer 2x for 8 min, PBS 2x for 10 min and then TBST 2x for 6 min. Blots were then blocked for 2h with 5% BSA in TBS with shaking at room temperature. Blots were then incubated with primary antibodies for the corresponding total protein– either IκBα or SAPK/JNK

(1:4000 (v/v) diluted in 5% BSA in TBST) overnight at 4oC on a rotating table. The same protocol for washing, secondary antibody incubation, detection and mild stripping was repeated.

40 Finally, the membranes were incubated with β-actin primary antibody (1:4000 (v/v) in 5% BSA in TBST) overnight at 4oC on a rotating table. The previous washing, secondary antibody, and detecting (but not mild stripping) protocols were repeated. Protein concentrations were semi- quantified by calculating densities of protein bands using ImageJ software (Image Processing and Analysis in Java, National Institutes of Health). Protein concentrations are expressed as a ratio of phospho-protein to total protein, for example p-JNK/JNK. Small intestine protein was also tested for TRAF6 and ubiquitin protein concentration on a 15% resolving gel with a 4% stacking gel. The protocol was the same as above however the membrane was incubated with

TRAF6 and then ubiquitin, both diluted 1:1000 (v/v) in 5% BSA in TBST.

3.5 Histological Analysis

o Small intestine samples were initially flash-frozen in LN2 and then stored at -80 C for future analysis. Small intestine samples destined for histological analysis were brought to room temperature and then ~1 cm segments of small intestine were placed into 10% formalin for fixation. Histotechnologists in the Department of Pathobiology then embedded the formalin- fixed tissues into paraffin (as per the Formalin Fixation and Paraffin Embedding (FFPE) protocol). Microscope slides were then prepared with sections of FFPE small intestine stained with general hematoxylin and eosin (H&E) staining.

Preliminary histological analysis using light microscopy (Olympus Life Sciences,

Toronto, Canada) was conducted by myself, as well as Dr. Majumder Kaustav to determine if there was gross evidence of intestinal epithelial damage prior to investing in histological analysis by a Pathologist. A previous study by Sukhotnik et al. (2004) examined the effects of LPS (IP injection 10 mg/kg, 3-day trial) on gut mucosal injury and observed reduced villous height in the

41 small intestine, as well as decreased intestinal thickness and diameter. In Deitch et al.’s (1989) investigation into LPS-induced gut permeability, it was observed that LPS induces acute mucosal damage, in which the villous epithelium of the intestine separates from the basal lamina.

However, an appreciable difference in morphology was not observed by the two of us and it was decided not to move forward with scoring by a Pathologist.

It is important to note that the ideal condition for the histological analysis of tissue is

o immediate fixation upon tissue collection, rather than flash freezing on LN2 and storing at -80 C.

The freezing process risks damage to tissue due to the formation of ice crystals, which can create artefacts when attempting to analyze morphology.

3.6 Data Analysis

Results are expressed as mean values ± SEM. Statistical significance was determined using one-way analysis of variance (ANOVA) followed by Tukey’s multiply comparison test or a one-tailed, unpaired Student’s t-test (GraphPad Prism version 5.0). ANOVA with the Tukey’s post hoc test was used when analyzing the means of more than two groups. The Student’s t-test was used when analyzing the means of two specific groups. Results will be considered statistically significant if p < 0.05.

42 4. RESULTS

4.1 Effects of γ-EV on Acute Inflammatory Cytokine Secretion in Plasma and Small

Intestine

4.1.1 Effects of γ-EV on Acute Inflammatory Cytokine Secretion in Plasma

Due to the limited amount of plasma available from each mouse, plasma samples were pooled such that n=8 became n=4. All pooling, either by the actual pooling of plasma/tissue samples or pooling values, was done by pooling animal #1 of treatment cage 1 with animal #1 of treatment cage 2, animal #2 of treatment cage 1 with animal #2 of treatment cage 2, etc., within each treatment group. Thus, pooling of specific animals was consistent through all biomarker assays. IL-6 ELISA was performed with unpooled plasma samples (n=8) but calculations and data analyses were done by pooling absorbance values after ELISA assay (n=4). TNF-α and IL-

1β ELISA experiments were both done with pooled plasma (n=4). Data are reported for both γ-

EV LD (50 mg/kg BW) and γ-EV HD (150 mg/kg BW) groups. As shown in Figure 7, there was no significant difference in plasma TNF-α concentration between the positive control (PC) group

(1485.8 ± 168.8 pg/mL) and the γ-EV LD (1385.6 ± 186.4 pg/mL) treatment group using a one- tailed, unpaired Student’s t-test (p = 0.3520). There was a significant difference in plasma TNF-α concentration between the PC group and the γ-EV HD group (982.4 ± 73.5 pg/mL) using the

Student’s t-test (p = 0.0170). There was no significant difference in plasma IL-6 concentration between the PC group (21735.3 ± 2933.0 pg/mL) and the γ-EV LD group (16705.9 ± 2904.8 pg/mL) using the Student’s t-test (p = 0.1344). There was a significant difference in plasma IL-6 concentration between the PC group and the γ-EV HD group (12429.8 ± 1080.8 pg/mL) using the Student’s t-test (p = 0.0124). There was a significant difference in plasma IL-1β

43 concentration between the PC group (171.6 ± 34.8 pg/mL) and both of the γ-EV treatment groups (γ-EV LD 19.2 ± 2.2 pg/mL; γ-EV HD 21.6 ± 3.6 pg/mL) using the Student’s t-test (PC vs. γ-EV LD p=0.0024, PC vs. γ-EV HD p=0.0026). The concentration of the anti-inflammatory cytokine IL-10 was not measured in plasma due to the limited volume of plasma collected.

Plasma TNF-α Plasma IL-6 Plasma [IL-1β] 2000 * 30000 * 300 ns 1500 ns 20000 200

1000

10000 100 500 Concentration (pg/mL) Concentration Concentration (pg/mL) Concentration β α **** **** ****

IL-1 0 0 (pg/mL) Concentration IL-6

TNF- 0 NC PC NC PC NC PC -EV LD -EV HD -EV LD -EV HD γ γ -EV LD -EV HD γ γ γ γ -EV Vehicle -EV Vehicle γ -EV Vehicle γ γ Treatment Group Treatment Group Treatment Group

Figure 7: Anti-inflammatory effects of γ-EV on acute inflammatory cytokine secretion in plasma. The anti-inflammatory activity of low-dose γ-EV (50 mg/kg BW) and high-dose γ-EV (150 mg/kg BW) was evaluated in plasma by the secretion of the acute pro-inflammatory cytokines TNF-α, IL-6 and IL-1β, whose concerted upregulated secretion is characteristic of early sepsis. Data are represented as mean ± SEM of n=4 per group. Differences in means were considered significant when p<0.05.

45 4.1.2 Effects of γ-EV on Acute Inflammatory Cytokine Secretion in Small Intestine

All cytokine concentrations were calculated both as picogram cytokine per miligram total protein (pg cytokine/mg protein) and picogram cytokine per gram tissue (pg cytokine/g tissue) to verify consistency of trend. Data are reported for both γ-EV LD (50 mg/kg BW) and γ-EV HD

(150 mg/kg BW) groups. As shown in Figure 8, there was no significant difference in small intestine TNF-α concentration (n=8) between the PC group (40.8 ± 6.8 pg/mg) and the γ-EV LD group (33.9 ± 8.2 pg/mg) using the Student’s t-test (p=0.2643). There was a significant difference in small intestine TNF-α concentration between the PC group and the γ-EV HD group

(25.7 ± 4.6 pg/mg) using the Student’s t-test (p=0.0441). There was no significant difference in

IL-6 concentration (n=8) between the PC group (204.6 ± 55.3 pg/mg) and the γ-EV LD (179.3 ±

40.2 pg/mg) group using the Student’s t-test (p=0.3553). There was a significant difference in

IL-6 between the PC group and the γ-EV HD group (81.2 ± 25.3 pg/mg) using the Student’s t- test (p=0.0289). The small intestine IL-1β concentration graph below is constructed with the minimum and maximum data points from each group being excluded from the calculations

(n=6). There was no significant difference in small intestine IL-1β concentration between the PC group (73.17 ± 6.5 pg/mg) and the γ-EV LD group (57.3 ± 21.7 pg/mg) the Student’s t-test

(p=0.2491). There was a significant difference in small intestine IL-1β between the PC group and the γ-EV HD group (22.7 ± 5.1 pg/mg) using the Student’s t-test (p<0.0001). There was no significant difference in small intestine IL-10 concentration (n=8) between the PC group (259.8

± 56.3 pg/mg) and either of the γ-EV treatment groups (γ-EV LD 239.1 ± 53.6 pg/mg; γ-EV HD

211.1 ± 24.6 pg/mg) using the Student’s t-test (PC vs. γ-EV LD p=0.3973, PC vs. γ-EV HD p=0.2206).

Small Intestine [TNF-α] Small Intestine [IL-6] 300 * 60 * ns 200 ns 40

100 20 Concentration (pg/mg protein) (pg/mg Concentration

α 0 0

NC PC protein) (pg/mg concentration IL-6 NC PC TNF- -EV LD -EV HD -EV LD -EV HD γ γ γ γ -EV Vehicle -EV Vehicle γ γ Treatment Group Treatment Group

Small Intestine [IL-1β] Small Intestine [IL-10] 100 **** 400 ns 80 ns 300 ns 60 200 40 100 20

Concentration (pg/mg protein) (pg/mg Concentration 0

β 0

PC protein) (pg/mg concentration IL-10 NC PC IL-1 NC -EV LD -EV LD -EV HD γ -EV HD γ γ γ -EV Vehicle -EV Vehicle γ γ Treatment Group Treatment Group

Figure 8: Anti-inflammatory effects of γ-EV on acute inflammatory cytokine secretion in small intestine tissue. The anti-inflammatory activities of low-dose γ-EV (50 mg/kg BW) and high- dose γ-EV (150 mg/kg BW) were evaluated in small intestine tissue by the secretion of TNF-α, IL-6, IL-1β, and IL-10. Data are represented as mean ± SEM of TNF-α n=8, IL-6 n=8 IL-1β n=6, and IL-10 n=8. Differences in means were considered significant when p<0.05.

47 4.2 Effects of γ-EV on Phosphorylation and Expression of Signaling Proteins

4.2.1 phospho-IκBα/IκBα

The phosphorylation of IκBα, an event that leads to NF-κB activation, was significantly increased in the PC group compared to the NC group using the Student’s t-test (p<0.0001). As shown in Figure 9, there was a significant difference in the level of phosphorylation of IκBα between the PC group and the γ-EV HD group using the Student’s t-test (p=0.0408).

4.2.2 phospho-JNK/JNK

The phosphorylation of JNK, a downstream kinase that leads to the activation of the AP-

1 inflammatory pathway, was increased in the PC compared to the NC group, however this increase was not statistically significant using the Student’s t-test (p=0.0820). As shown in

Figure 9, JNK phosphorylation was reduced by γ-EV treatment, however this reduction was not statistically significant using the Student’s t-test (p=0.0666). Although a significant effect was not produced by γ-EV treatment, there is a clear trend to indicate that γ-EV suppressed JNK phosphorylation.

4.2.3 TRAF6/β-actin

The concentration of TRAF6 was measured in an attempt to identify the degree of ubiquitination of the upstream signaling protein as a way to evaluate if γ-EV treatment had an effect on TRAF6 activation. The experimental approach consisted of probing a Western blot with both TRAF6 and ubiquitin antibodies to first identify the TRAF6 band, and then to identify if there was significant binding of the ubiquitin antibody to the previously established TRAF6 region. This approach was supported by the fact that the molecular weight of the TRAF6 protein

p-JNK/JNK p-IκBα/IκBα 0.4 2.0 *

0.3 1.5 ns

0.2 Expression 1.0 α B κ /I

0.1 α 0.5 B κ -JNK/JNK Expression -JNK/JNK -I p p 0.0 0.0 NC PC γ-EV NC PC γ-EV p-JNK p-IκBα

JNK IκBα

Figure 9: γ-EV reduced the phosphorylation of JNK and IκBα. Data are represented as mean ± SEM of n=6. p-JNK/JNK results are from unreplicated experiments. p-IκBα/IκBα results are from duplicated experiments. Differences in means were considered significant when p<0.05.

49 is 60 kilodalton (kDa) while the molecular weight of the ubiquitin protein is 8.5 kDa. Therefore, it was assumed that ubiquitin binding in the 60 kDa region of the blot would likely indicate ubiquitin bound to TRAF6, rather than unbound ubiquitin. As well, since TRAF6 activation is a result of polyubiquination, ubiquitin binding over a range >60 kDa (60 kDa + 8.5 kDa+ 8.5 kDa

+ …) was expected rather than a discrete band of ubiquitin binding at ~68.5 kDa. While a clear

TRAF6 band was produced by Western blotting, probing with ubiquitin elicited a wide range of binding throughout each lane, indicating that ubiquitin was bound to several proteins in the background. This finding makes sense as ubiquitin is a ubiquitous protein throughout the cell.

Therefore, the TRAF6 and ubiquitin results are presented separately in this report. It should also be noted that this approach was not appropriate for identifying protein-protein interactions, but rather it would have provided evidence of a likely interaction between the two proteins. Co- immunoprecipitation (Co-IP) techniques would have provided robust results to definitively conclude an interaction between TRAF6 and ubiquitin.

As shown in Figure 10, there was no significant difference in TRAF6 concentration between the NC and PC groups using ANOVA. The concentration of the upstream signaling protein TRAF6 was significantly reduced in the γ-EV HD group compared to both the NC group

(****) and the PC group (**) using ANOVA.

4.2.4 Ubiquitin/β-actin

As described in the previous section, the concentration of ubiquitin was originally measured in an attempt to evaluate the effect of γ-EV on TRAF6 ubiquitination in the various control and treatment groups, however the ubiquitin WB results were non-specific, which is not

50 surprising as this protein is ubiquitous in the cell. As shown in Figure 10, the concentration of ubiquitin was significantly increased in the PC group compared to the NC group using ANOVA

(****). The concentration of ubiquitin was significantly different between the PC group and the

γ-EV HD group using ANOVA (****). There was no significant difference between the NC and

γ-EV HD group using ANOVA.

TRAF6/β-actin Ubiquitin/β-actin 1.5 **** 2.0 **** ** 1.5 1.0

1.0 -actin Expression β -actin Expression 0.5 β 0.5 TRAF6/ 0.0 Ubiquitin/ 0.0 NC PC γ-EV NC PC γ-EV

TRAF6 Ubiquitin

-actin β-actin β

Figure 10: Protein expression profiles of upstream signaling protein TRAF6 and the ubiquitous protein Ubiquitin. Data are represented as mean ± SEM of n=6 from duplicated experiments. Differences in means were considered significant when p<0.05.

52 4.3 Small Intestine Histology

LPS endotoxemia induced by either IP or intramuscular (IM) injections of LPS is known to cause gut mucosal injury by increasing gut permeability, which leads to bacterial translocation

(Deitch et al., 1989). Bacteria that cross the mucosal barrier and intestinal epithelium travel to and survive in the mesenteric lymph nodes, where they mount systemic immune responses

(Deitch et al., 1989). Acute mucosal injury was previously observed in LPS endotoxemia, whereby villous epithelium dissociated from the basal lamina to produce subepithelial spaces, which facilitated the translocation of bacteria (Deitch et al., 1989). Another study examining the effects of L-Arg in the gut of LPS-stimulated mice documented a significant decrease in villous height and a trend towards decreased crypt depth, an effect which decreased the absorptive surface area (Sukhotnik et al., 2004).

The histological analysis of the present study’s small intestine sections was compromised by the fact that the tissue was not immediately fixed upon collection, but rather was flash frozen

o on LN2 and then stored at -80 C. This introduced the possibility of morphological damage and artefacts that would otherwise not be present in properly fixed tissue. The slides were examined by two personnel, myself and Dr. Kaustav Majumder, for gross morphological differences between the treatments groups. As shown in Figure 11, no appreciable differences in epithelial integrity, crypt depth, and villous height were observed.

53 200x

Negative Control 100x

200x

Positive Control 100x

200x

γ-EV HD 100x

Figure 11: Small intestine histological slides stained with H&E. Images are representative of each group.

54 5. DISCUSSION

5.1 γ-EV Suppressed Pro-Inflammatory Cytokine Secretion in Plasma

The initial phase of sepsis is characterized by the elevation of the three pro-inflammatory cytokines TNF-α, IL-6 and IL-1β (Faix, 2013). The coordinated spike in these three pro- inflammatory mediators initiates the systemic inflammatory response seen in the acute phase of sepsis, which stimulates the adaptive immune response and the subsequent cytokine storm (Faix,

2013). TNF-α secreted as a result of TLR4 activation by LPS binds to the TNF receptor (TNFR) to further propagate the inflammatory response via NF-κB activation (Sriskandan and Altmannn,

2008). Activation of endothelial cells by TNF-α and IL-1β recruits polymorphonuclear leukocytes (PMNs) such as neutrophils, eosinophils, and basophils (Faix, 2013). Secreted IL-6 stimulates the liver to produce acute-phase proteins such as CRP and IL-6 levels have been associated clinically with increased risk of mortality from sepsis (Faix, 2013). As well, IL-1β binds to its receptor IL-1R, which shares the TIR pathway with TLR, mediated by MyD88 recruitment (Deng et al., 2000). Thus, TNF-α and IL-1β sustain the inflammatory response in sepsis via positive feedback loops to stimulate their NF-κB-mediated expression. Thus, as a hallmark of early sepsis, the concentrations of these three pro-inflammatory cytokines were measured in plasma to determine if γ-EV had anti-sepsis activity. It was observed that mice in the γ-EV high-dose group (150 mg/kg BW) had significantly suppressed plasma concentrations of TNF-α and IL-6, and mice in both the γ-EV low-dose (50 mg/kg BW) and high-dose groups had significantly suppressed plasma concentrations of IL-1β. Therefore, γ-EV did demonstrate anti-sepsis activity through the decreased secretion of pro-inflammatory cytokines in plasma. By modulating this early phase of sepsis, γ-EV effectively plays a role in inhibiting the progression

55 or decreasing the severity of sepsis by inhibiting the early mechanisms of sepsis. It is thought that by modulating sepsis in the acute phase response, patients will experience lower rates of morbidity and mortality. Morbidities such as permanent organ damage following the resolution of sepsis arise from damage incurred during severe sepsis caused by hypotension and its resultant hypoperfusion of tissues. Organ failure is known to occur predominately in the immunosuppressive CARS phase of severe sepsis, thus by inhibiting the progression of the SIRS phase into the CARS phase, organ dysfunction can be avoided. Similarly, death predominately occurs in the CARS phase of severe sepsis when immunosuppression places patients at significant risk of nosocomial infections. Thus, by preventing this phase altogether, the mortality rate of sepsis can be reduced.

5.2 γ-EV Suppressed Pro-Inflammatory Cytokine Secretion in Small Intestine

Small intestine tissue was also analyzed for both pro- and anti-inflammatory cytokines to determine the effects of γ-EV in the intestinal organ. The small intestine tissue sample was a combination of cells, including intestinal epithelial cells (IECs), immune cells, and transient cells within the intestinal lumen. We examined this organ as a potential target for novel anti-sepsis therapy, to determine if an ingested treatment could exert an anti-inflammatory effect potent enough to slow or halt the progression of sepsis. As well, the small intestine contributes considerably to the pathogenesis of the small intestine, as gut permeability induced by LPS leads to bacterial translocation to induce local and systemic immune responses (Deitch et al., 1989).

The concentration of the three pro-inflammatory cytokines that are increased during the acute phase of sepsis were measured. This project focused on the acute phase of sepsis since the stimulation of mice with LPS lasted 2h in length. Mice in the γ-EV high-dose group had

56 significantly decreased concentrations of TNF-α, IL-6, and IL-1β in the small intestine. The concentration of the anti-inflammatory cytokine IL-10 was also determined in the small intestine tissue and it was observed that there was no change in IL-10 secretion amongst the groups. The concentration of IL-10 was measured to determine if γ-EV treatment had an effect on anti- inflammatory mediators instead of, or as well as, pro-inflammatory mediators to help identify the anti-inflammatory mechanism of γ-EV. Therefore, these findings suggest that the anti-sepsis activity of γ-EV is via the suppression of pro-inflammatory mediators rather than the upregulation of anti-inflammatory mediators.

5.3 γ-EV Inhibited Phosphorylation of IκBα

In the LPS-activated TLR4 signaling cascade, the downstream inhibitory molecule

IκBα prevents the nuclear translocation of the NF-κB TF via physical association.

Phosphorylation of IκBα by IKK dissociates the inhibitory molecule from the NF-κB TF which then translocates and binds to the 10 base-pair (bp) DNA κB binding sites in the promoters and enhancers of numerous target genes, including several pro-inflammatory genes (Hiscott et al.,

1993; Pahl, 1999).

In the present study, the phosphorylation of IκBα was measured to determine if treatment with γ-EV had an effect on the activation of the NF-κB pathway, to help elucidate the anti-sepsis activity of γ-EV. Since the NF-κB pathway is inactive when IκBα is in its unphosphorylated state, by measuring the protein expression concentrations of p-IκBα/ IκBα, the degree of NF-κB activation of target genes can be evaluated. It was found that the γ-EV high-dose treatment group significantly decreased the phosphorylation of the IκBα in the small intestine tissue. Thus, it can be determined that the treatment of γ-EV in mice with LPS-induced sepsis can significantly

57 decrease the phosphorylation and activation of the downstream inhibitory molecule IκBα to inhibit the activation and expression of several inflammatory target genes. As previously discussed, the expression of TNF-α and IL-1β genes is controlled by NF-κB activation, however

TNFR activation by its ligand TNF-α and IL-1R activation by its ligand IL-1β also lead to the activation of the NF-κB pathway, thus producing an inflammatory positive feedback loop. By implementing γ-EV-mediated inhibition of IκBα phosphorylation, the positive feedback loop that contributes to the sepsis ‘cytokine storm’ can be halted.

5.4 γ-EV Inhibited Phosphorylation of JNK

The LPS-activated TLR4 signaling pathway also leads to the activation of the downstream MAPK pathway, in which MKK4/7-mediated phosphorylation of JNK activates the

AP-1 TF that binds AP-1 DNA binding sites to stimulate the expression of pro-inflammatory cytokines, such as TNF-α (Das et al., 2009; Miller et al., 2012). Thus, by preventing the phosphorylation of JNK by MKK4/7, the expression of various pro-inflammatory genes is inhibited. The present study used the protein expression concentration of p-JNK/JNK to indicate if treatment with γ-EV had an effect on the downstream MAPK/AP-1 pathway in mice with LPS- induced sepsis. Since the TIR intracellular signaling pathway leads to both the TAK-1-mediated activation of the NF-κB and AP-1 pathways, this project aimed to identify if γ-EV had an effect on one or both downstream pathways, prior to investigating target proteins upstream to IKK and

MKK4/7.

The protein concentrations of both p-JNK and JNK in small intestine tissue were measured by WB techniques. Although there was not a statistically significant effect by γ-EV, there was a clear trend to indicate that the phosphorylation of JNK was suppressed in the γ-EV

58 HD group compared to the PC group. Therefore, the current findings indicate that the anti- inflammatory activity of γ-EV was also a result of its modulation of the MAPK pathway, specifically the phosphorylation of JNK.

5.5 γ-EV had an Effect on Background Ubiquitin Protein Levels that were Unrelated to

TRAF6

The ubiquitin/β-actin WB results demonstrate that γ-EV had an effect on background ubiquitin protein levels that cannot be specifically ascribed to TRAF6 ubiquitination without the use of Co-IP techniques. Hence its name, ubiquitin is a ubiquitous protein in the cell and the process of ubiquination is associated with both the activation of signaling proteins through its service as a scaffold (Zeng et al., 2015) and proteasomal degradation of target proteins. Lys-48- linked ubiquitination is associated with target protein degradation, whereas Lys-63-linked ubiquitination is associated with the activation of signaling proteins (Zeng et al., 2015). A study by Zeng et al., found that a natural molecule called FMHM, from the Chinese herbal medicine

Radix Polygalae, increased Lys-48-linked ubiquitination of TRAF6 during LPS stimulation while also decreasing its Lys-63-linked ubiquitination, thus demonstrating that TRAF6 is a potential therapeutic target for downregulating TLR4 signaling (2015). However, there are other targets of ubiquitination in the TIR signaling pathway including IKK (by TRAF6), TRAF6

(autoubiquitination), as well as others involved in entirely different cellular processes. If this effect of γ-EV on ubiquitin was related to TRAF6 ubiquitination, we could have expected the following: if looking only at Lys-48-linked ubiquitin, then TRAF6 levels would be inversely related to ubiquitin levels. Or, if examining Lys-63-linked ubiquitin, then the phosphorylation of both JNK and IκBα would have been increased as a consequence of TRAF6 activation.

59 However, none of these phenomena were observed and thus no appreciable conclusion can be drawn, particularly in relevance to the present study.

While we cannot speculate specifically where the effect of γ-EV on ubiquitin protein expression is occurring within the cell, we can conclude that γ-EV has a significant effect on background ubiquitin levels that warrants further investigation.

5.6 Protein Concentration of Total TRAF6 Decreased with LPS Stimulation

As discussed in the previous section, the Lys-48-linked ubiquination of TRAF6 leads to its proteasomal degradation (Pickart, 1997). The decrease in total TRAF6 protein in the PC group compared to the NC group may be a result of the body’s, albeit comprised, immune system attempting to downregulate TIR signaling through Lys-48-linked ubiquination-mediated proteasomal degradation. This is supported by the ubiquitin/β-actin WB results showing total ubiquitin in the PC group was significantly increased compared to both the NC and γ-EV HD groups. It is, however, puzzling to interpret the decrease in total TRAF6 protein in the γ-EV HD group concurrently with the ubiquitin concentration being comparable to the ubiquitin concentration of the NC group. The LPS-induced reduction in total TRAF6 protein warrants further investigation.

5.7 Cellular Signaling Crosstalk Mechanism

The anti-inflammatory effect of γ-EV in mice with LPS-induced sepsis is hypothesized to be a result of a crosstalk signaling mechanism between the intracellular signaling domains of the

LPS-activated TLR4 and the γ-EV-allosterically-activated CaSR. The recruitment of β-arrestin2 by the allosterically-activated CaSR has been previously reported by Zhang et al. (2015). In this

60 previous study, IECs pre-treated with γ-EC and γ-EV and then subsequently stimulated with

TNF-α secreted significantly reduced concentrations of the chemokine IL-8. Through various methods including blocking CaSR with the CaSR antagonist NPS-2143, producing β-arrestin2 knockdown cells (β-arr2 KD) using β-arrestin2-specific small interfering RNA (siRNA), and Co-

IP to identify protein-protein interactions, it was determined that CaSR-recruited β-arrestin2 directly interacted with TAB1 within the TNFR signaling domain to prevent the formation of the

TAK1-TAB1-TAB2 complex. The blockade of the formation of the TAK1-TAB1-TAB2 complex inhibited further signaling events in the TNFR domain, which effectively suppressed the NF-κB and AP-1 TFs. The β-arrestin2-TAB1 crosstalk mechanism has also been confirmed to mediate the anti-inflammatory effects of γ-EC and γ-EV in DSS-induced mouse colitis and of

L-Trp in TNF-α-stimulated IECs (Zhang et al., 2015, Mine and Zhang, 2015). The fact that γ- glutamylpeptides and the L-amino acid were capable of inducing the same anti-inflammatory effects is due to the nature of their specific binding to CaSR, which involves the exposed carboxy and amino groups of L-amino acids or a peptide with a γ-peptide bond (Ohsu et al.,

2010). Thus, the specific nature of the allosteric activation of CaSR by its type II ligands confers the potent β-arrestin2-mediated anti-inflammatory activity.

The intracellular signaling domains of TLR4 and TIR are similar in that they both contain

TRAF6 and its associated downstream signaling proteins such as the TAK1-TAB1-TAB2 complex, the NF-κB and AP-1 pathways, and the activation of various pro-inflammatory mediators. Thus, because of the marked similarities between the two signaling pathways and the gene expression products that are a result of their activation, it can be strongly hypothesized that the anti-inflammatory activity of γ-EV in LPS-induced sepsis was a result of, or partially a result

61 of, the interaction of β-arrestin2 and TAB1 that suppressed TIR signaling downstream of the

TAK1-TAB1-TAB2 complex.

It has been determined that β-arrestin2 interacts directly with other signaling proteins including TRAF6 in TIR signaling and IκBα in TNFR signaling to negatively regulate signaling events and the activation of the NF-κB and AP-1 TFs (Wang et al., 2006; Gao et al., 2004). By negatively regulating the activation of the NF-κB and AP-1 TFs, the various β-arrestin2- mediated crosstalk events effectively suppress the expression of several inflammatory target genes that stimulate immune and inflammatory responses (Pahl, 1999).

Wang et al. (2006) have reported the direct interaction of β-arrestin2 with TRAF6, which prevents the oligomerization and ubiquination of TRAF6, which effectively halts TIR signaling upstream at TRAF6. β-arrestin2 interacts directly with the TRAF-N domain to inhibit the oligomerization required for its auto-polyubiqination, rather than functioning as a deubiquitinase

(Wang et al., 2006). The present project aimed to identify the degree of TRAF6 ubiquitination in order to determine if treatment with γ-EV prevented the oligomerization and ubiquination of

TRAF6. However, the research methods were limited to WB, rather than the ideal Co-IP approach, which would have definitively identified the interaction between the two proteins.

Unfortunately, the WB approach did not indicate the degree of TRAF6 activation. However, based on Wang et al.’s previous study that confirmed that interaction of β-arrestin2 with TRAF6, it can be strongly hypothesized that the anti-inflammatory activity of γ-EV in LPS-induced sepsis was a result of, or partially a result of, the interaction of β-arrestin2 and TRAF6 that suppressed

TIR signaling downstream of TRAF6.

62 The stabilization of IκBα via the direct interaction of β-arrestin2 with IκBα in TNFR signaling has previously been established by Gao et al. (2004). While it is proposed that the β- arrestin2-IκBα interaction is an element of the anti-inflammatory mechanism of γ-EV, it is clear that it is not the only mechanism. This is evidenced by the suppressed phosphorylation of JNK, which would otherwise not occur if treatment with γ-EV was downregulating only the NF-κB pathway. Therefore, the β-arrestin2-IκBα interaction is proposed to be one element of the present project’s crosstalk mechanism.

β-Arrestin2 has also been shown to interact with p38 MAPK to upregulate the expression of IL-10 (Li et al., 2014). However, based on the small intestine IL-10 ELISA results of the result project that show that γ-EV did not have an effect on the expression of IL-10, it can be concluded that γ-EV’s anti-inflammatory activity was not a result of interaction between β- arrestin2 and p38.

Therefore, the anti-inflammatory activity of γ-EV in LPS-induced sepsis is achieved by a crosstalk signaling mechanism between the CaSR-recruited β-arrestin2 with TRAF6, TAB1, and

IκBα within the TIR signaling domain. The redundancy of β-arrestin2 interactions with TIR components is suggested by Wang et al. (2006) to be a “double-checking” feature to ensure the

TIR pathway does not go awry and result in the exaggerated inflammatory response seen in sepsis.

Figure 12: The proposed signaling crosstalk interaction between β−arrestin2 and TRAF6. The red bidirectional arrow indicates the previously reported direct interaction of β−arrestin2 with the TRAF-N domain of TRAF6. γ-EV is proposed to inhibit TIR signaling downstream of TRAF6 via its allosteric activation of CaSR. Through the direction interaction of CaSR-recruited β−arrestin2 with the TRAF-N domain of TRAF6, β−arrestin2 effectively blocks TRAF6 activation by blocking its polymerization and autoubiquitination (Adapted from Wang et al., 2006).

Figure 13: The proposed signaling crosstalk interaction between β−arrestin2 and TAB1. The red bidirectional arrow indicates the previously reported direct interaction between β-arrestin2 and TAB1. Through β-arrestin2 interfering with the formation of the TAK1-TAB1-TAB2 complex, γ-EV is hypothesized to inhibit TIR signaling downstream of TAB1 (Adapted from Zhang et al., 2015).

Figure 14: The proposed signaling crosstalk interaction between β−arrestin2 and IκBα. The red bidirectional arrow indicates the direct interaction of β-arrestin2 and IκBα. By stabilizing IκBα by direct interaction, CaSR-recruited β-arrestin2 inhibits the activation of the NF-κB pathway. The direct interaction of β−arrestin2 and IκBα prevents the phosphorylation and proteasomal degradation of IκBα, thus preventing the nuclear translocation of NF-κB and expression of NF- κB-target genes (Adapted from Gao et al., 2014).

66 5.8 Applications to Sepsis Prevention

Findings from the current study provide potential novel targets for the treatment of sepsis.

While the anti-inflammatory effects of γ-EV in LPS-induced sepsis likely will not completely abrogate the sinister pathogenesis and effects of sepsis in human patients, the findings from this study can help to inform researchers on cell signaling targets for sepsis prevention.

Food research and developers can implement the use of other γ-glutamylpeptides and L- amino acids that have the confirmed ability to allosterically activate CaSR and the Gi/o-mediated recruitment of β-arrestin2, since γ-EC and L-Trp also exert their anti-inflammatory effects with the same mechanism as γ-EV. The use of γ-EV and other allosteric activators of CaSR for the prevention of sepsis could be developed into functional nutraceuticals. Since γ-EV already boasts a strong marketing feature – its kokumi flavour, the bioactive peptide can be introduced, or reintroduced, to the market in various forms. While not in a particularly potent form, foods containing γ-EV in their native forms such as beans and onions could be advertised as having natural anti-inflammatory properties. This approach would likely benefit agricultural stakeholders as it would aim to increase sales of their products. A second approach to introducing γ-EV as a functional nutraceutical would essentially be to re-introduce Ajinomoto’s flavour-enhancing γ-EV ingredient product, with a new marketing angle. Consumers could be educated on the anti-inflammatory and sepsis prevention properties of a product that may already be present in their pantries. By having the γ-EV functional food widely available, it has the potential to reach consumers of various demographics, including individuals at high-risk for sepsis. Parallel outcomes of this functional food being on the market would be a decreased rate of sepsis as well as an increase in sales for the food industry. Lastly, researchers can make the health-promoting effects of γ-EV available to consumers in the form of a supplement.

67 Nutraceutical supplements are in high demand amongst lay health and fitness advocates. Overall, there is a great potential for introducing γ-EV as nutraceutical to the general population, as it would benefit many stakeholders including the healthcare budget, at-risk individuals, and the food industry.

68 6. CONCLUSION AND FUTURE WORK

The present study has determined that the food-derived bioactive dipeptide γ-EV has anti- inflammatory activity in a mouse model of LPS-induced sepsis. The anti-sepsis activity of γ-EV is a result of the dipeptide’s allosteric activation of CaSR which subsequently activates Gi/o- mediated recruitment of β-arrestin2, which interacts directly with TRAF6, TAB1, and/or IκBα in the LPS-activated TLR4 signaling domain. These crosstalk events between the two receptors’ signaling domains leads to the suppression of the NF-κB and AP-1 pathways to inhibit the production of pro-inflammatory mediators, including TNF-α, IL-6, and IL-1β.

Future work to continue this project should focus on using experimental techniques to confirm the proposed crosstalk mechanism, such as the use of a CaSR antagonist to confirm the activation of CaSR by γ-EV, the creation of β-arr2 KD cells or β-arr2 KD mice to confirm the role of β-arrestin2, as well as, and probably most importantly, Co-IP to confirm the direct interaction of β-arrestin2 with its three proposed binding partners: TRAF6, TAB1, and IκBα.

Future work should also investigate the effect of γ-EV on gut permeability in a mouse model of LPS-induced sepsis. Endotoxin-induced gut permeability is considered to be a significant event in endotoxin-induced sepsis because it leads to bacterial translocation (Deitch et al., 1989). In order to investigate the effect of γ-EV on gut permeability, researchers can use the lactulose/mannitol test for intestinal permeability as well as the histological analysis of small intestine tissue that has been fixed immediately upon collection (Sequeira et al., 2014; Deitch et al., 1989).

69 7. REFERENCES

1) Andersson R. 2003. Appendectomy is followed by an increased risk of Crohn’s disease. Gastroentereology 124: 40-46.

2) Akira S, Takeda K. 2004. Toll-like Receptor Signaling. Nature Reviews Immunology 4: 499-511.

3) Angus DC, Linde-Zwirble WT, Lidicker J, Clerment G, Carcillo J, Pinsky MR. 2001. Epidemiology of severe sepsis in the United States: Analysis of incidence, outcome, and associated costs of care. Critical Care Medicine 29: 1303-1310.

4) Balk RA, Bone RC, Cerra FB, Dellinger RP, Fein AM, Knaus A. 1992. Definitions for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis. Chest 10: 1644-1655.

5) Benzaquen LR, Yu H, Rifai N. 2002. High Sensitivity C-reactive Protein: An Emerging Role in Cardiovascular Risk Assessment. Critical Reviews in Clinical Laboratory Sciences 39: 459-497.

6) Berg RD, Owens WE. 1979. Inhibition of translation of viable Escherichia coli from the gastrointestinal tract of mice by bacterial antagonism. Infection and Immunity 25: 820- 827.

7) Bernard GR, Vincent JL, Laterre PF, LaRosa SP, Dhainaut JF, Lopez-Rodriguez A, Steingrub JS, Garber GE, Helterbrand JF, Ely EW, Fisher CJ. 2001. Efficacy and Safety of Recombinant Human Activated Protein C for Severe Sepsis. The New England Journal of Medicine 344: 699-709.

8) Bone RC, Grodzin CJ, Balk RA. 1997. Sepsis: a new hypothesis for pathogenesis of the disease process. Chest 112: 235-243.

9) Brown EM, Hebert SC, Riccardi D, Geibel JP. 2013. The Calcium-Sensing Receptor. In Seldin and Giebiech’s The Kidney, (Fifth Edit, Vol. 1, pp. 2187-2224). Elsevier Inc.

10) Centers for Disease Control and Prevention (CDC): Chronic Diseases Overview. Available at: http://www.cdc.gov/chronicdisease/overview/#ref17. Accessed February 2016.

11) Chen XH, Yin YJ, Zhang JX. Sepsis and immune response. World Journal of Emergency Medicine 2: 88-92.

12) Conigrave AD, Franks AH, Brown EM, Quinn SJ. 2002. L-Amino acid sensing by the calcium-sensing receptor: a general mechanism for coupling protein and calcium metabolism? European Journal of Clinical Nutrition 56: 1072-1080.

13) Conigrave AD, Hampson DR. 2006. Broad-spectrum L-amino acid sensing by class 3 G- protein-coupled receptors. Trends in Endocrinology and Metabolism 17: 398-407.

14) Conigrave AD, Ward DT. 2013. Calcium-sensing receptor (CaSR): Pharmacological properties and signaling pathways. Best Practice & Research Clinical Endocrinology & Metabolism 27: 315–331.

15) Dalmasso G, Charrier-Hisamuddin L, Nguyen HTT, Yan Y, Sitaraman S, Merlin D. 2008. PepT1-mediated tripeptide KPV uptake reduces intestinal inflammation. Gastroenterology 134: 166–178.

16) Das M, Sabio G, Jiang F, Rincón M, Flavell RA, Davis RJ. 2009. Induction of Hepatitis by JNK-Mediated Expression of TNF-α. Cell 136: 249-260.

17) Dejager L, Pinheiro I, Dejonckheere E, Libert C. 2011. Cecal ligation and puncture: the gold standard model for polymicrobial sepsis? Trends in Microbiology 19: 198-208.

18) Deitch EA, Ma L, Ma WJ, Grisham MB, Granger DN, Specian RD, Berg RD. 1989. Inhibition of endotoxin-induced bacterial translocation in mice. The Journal of Clinical Investigation 84: 36-42.

19) Deng L, Wang C, Spencer E, Yang L, Braun A, You J, Slaughter C, Pickart C, Chen ZJ. 2000. Activation of the IκB Kinase Complex by TRAF6 Requires a Dimeric Ubiquitin- Conjugating Enzyme Complex and a Unique Polyubiquitin Chain. Cell 103: 351–361.

20) Dunkel A, Köster J, Hofmann T. 2007. Molecular and sensory characterization of γ- glutamyl peptides as key contributors to the kokumi taste of edible beans (Phaseolus vulgaris L.). Journal of Agricultural and Food Chemistry 55: 6712–6719.

21) Faix JD. 2013. Biomarkers of sepsis. Critical Reviews in Clinical Laboratory Sciences 50: 23-26.

22) Ferreira FL, Bota DP, Bross A, Mélot C, Vincent JL. 2001. Serial Evaluation of the SOFA Score to Predict Outcome in Critically Ill Patients. The Journal of the American Medical Association 286: 1754-1758.

23) Fisher CJ, Agosti JM, Opal SM, Lowry SF, Balk RA, Sadoff JC, Abrahm E, Schein RMH, Benjamin E. 1996. Treatment of septic shock with the tumour necrosis factor: Fc fusion protein. The Soluble TNF Receptor Sepsis Study Group. The New England Journal of Medicine 334: 1697-1702.

24) Gao H, Sun Y, Wu Y, Luan B, Wang Y, Qu B, & Pei G. 2004. Identification of β- arrestin2 as a G protein-coupled receptor-stimulated regulator of NF-κB pathways. Molecular Cell 14: 303–317.

71 25) Ghosh S, Mitchell R. 2007. Impact of inflammatory bowel disease on quality of life: Results of the European Federation of Crohn’s and Ulcerative Colitis Associations (EFCCA) patient survey. Journal of Crohn’s & Colitis 1: 10–20.

26) Gowen M, Stroup GB, Dodds RA, James IE, Votta BJ, Smith BR, Bhatnagar PK, Lago AM, Callahan JF, DelMar EG, Miller MA, Nemeth EF, Fox J. 1995. Antagonizing the parathyroid calcium receptor stimulates parathyroid hormone secretion and bone formation in osteopenic rats. The Journal of Clinical Investigation 105: 1595-1604.

27) Harbarth S, Holeckova K, Froidevaux C, Pittet D, Ricou B, Grau GE, Vadas L, Pugin J, the Geneva Sepsis Network. 2001. Diagnostic Value of Procalcitonin, Interleukin-6, and Interleukin-8 in Critically Ill Patients Admitted with Suspected Sepsis. American Journal of Respiratory and Critical Care Medicine 164: 396-402.

28) Hartmann R, Meisel H. 2007. Food-derived peptides with biological activity: from research to food applications. Current Opinion in Biotechnology 18: 163–169.

29) Hebert SC, Cheng S, Geibel J. 2004. Functions and roles of the extracellular Ca2+- sensing receptor in the gastrointestinal tract. Cell Calcium 35: 239–247.

30) Hiscott J, Marois J, Garoufalis J, D’Addario M, Roulsto A, Kwan Y, Pepin N, Lacoste J, Nguyen H, Bensi G, Fenton M. 1993. Characterization of a Functional NF-κB site in the Human Interleukin 1β Promoter: Evidence for a Positive Autoregulatory Loop. Molecular and Cellular Biology 13: 6231-6240.

31) Ho C, Conner DA, Pollak MR, Ladd DJ, Kifor O, Warren HB, Brown EM, Seidman JG, Seidman CE. 1995. A mouse model of human familial hypocalciuric hypercalcemia and neonatal severe hyperparathyroidism. Nature Genetics 11: 389-394.

32) Hofer AM, Brown EM. 2003. Extracellular calcium sensing and signalling. Nature Reviews Molecular Cell Biology 4: 530–538.

33) Hotchkiss RS and Karl IE. 2002. The Pathophysiology and Treatment of Sepsis. The New England Journal of Medicine. 348(2): 138–150.

34) Huang W, Chakrabarti S, Majumder K, Jiang Y, Davidge ST, Wu J. 2010. Egg-Derived Peptide IRW Inhibits TNF-a-Induced Inflammatory Response and Oxidative Stress in Endothelial Cells. Journal of Agricultural and Food Chemistry 58: 10840-10846.

35) Iskandar MM, Dauletbaev N, Kubow S, Mawji N, Lands LC. 2013. Whey protein hydrolysates decrease IL-8 secretion in lipopolysaccharide (LPS)-stimulated respiratory epithelial cells by affecting LPS binding to Toll-like receptor 4. British Journal of Nutrition 110: 58-68.

72 36) Jacob J, Kamitsuka M, Clark RH, KelleherAS, Spitzer AR. 2015. Etiologies of NICU Deaths. Pediatrics 135: 159-165.

37) Jones AE, Trzeciak S, Kline JA. 2009. The Sequential Organ Failure Assessment score for predicting outcome in patients with severe sepsis and evidence of hypoperfusion at the time of emergency department presentation. Critical Care Medicine 37: 1649-1654.

38) Kanazawa A, Kakimoto Y, Nakajima T, Sano I. 1965. Identification of γ-glutamylserine, γ-glutamylalanine, γ-glutamylvaline and S-methylglutathione of bovine brain. Biochimica et Biophysica Acta 111: 90–95.

39) Karzai W, Oberhoffer M, Meier-Hellmann A, Reinhart K. 1997. Procalcitonin – A New Indicator of the Systemic Response to Severe Infections. Infection 25: 329-334.

40) Katayama S, Xu X, Fan MZ, Mine Y. 2006. Antioxidative Stress Activity of Oligophosphopeptides Derived from Hen Egg Yolk Phosvitin in Caco-2 Cells. Journal of Agriculture and Food Chemistry 54: 774-778.

41) Kim CJ, Kovacs-Nolan JA, Yang C, Archbold T, Fan MZ, Mine Y. 2010. L-Tryptophan exhibits therapeutic function in a porcine model of dextran sodium sulfate (DSS)-induced colitis. Journal of Nutritional Biochemistry 21: 468-475.

42) Kobayashi Y, Kovacs-Nolan J, Matsui T, Mine Y. 2015. The Anti-atherosclerotic Dipeptide, Trp-His, Reduces Intestinal Inflammation through the Blockade of L-Type Ca2+ Channels. Journal of Agriculture and Food Chemistry 63: 6041-6050.

43) Kobayashi Y, Rupa P, Kovacs-Nolan J, Turner PV, Matsui T, Mine Y. 2015. Oral Administration of Hen Egg White Ovotransferrin Attenuates the Development of Colitis Induced by Dextran Sodium Sulfate in Mice. Journal of Agriculture and Food Chemistry 63: 1532–1539.

44) Kobayashi T, Walsh MC, Choi Y. 2004. The role of TRAF6 in signal transduction and the immune response. Microbes Infect 6: 1333–1338.

45) Kodera T, Nio N. 2006. Identification of an Angiotensin I-converting Enzyme Inhibitory Peptides from Protein Hydrolysates by a Soybean Protease and the Antihypertensive Effects of Hydrolysates in Spontaneously Hypertensive Model Rats. Journal of Food Science 71: C164-173.

46) Korhonen H, Pihlanto A. 2006. Bioactive peptides: Production and functionality. International Dairy Journal 16: 945-960.

47) Kovacs-Nolan J, Zhang H, Ibuki M, Nakamori T, Yoshiura K, Turner PV, Matsui T, Mine Y. 2012. The PepT1-transportable soy tripeptide VPY reduces intestinal inflammation. Biochimica et Biophysica Acta 1820: 1753–1763.

73 48) Kristensen I, Larsen PO, Sørensen H. 1974. Free amino acids and γ-glutamyl peptides in seeds of Fagus silvatica. Phytochemistry 13: 2803–2811.

49) Kumar A, Roberts D, Wood KE, Light B, Parrillo JE, Sharma S, Suppes R, Feinstein D, Zanotti S, Taiberg L, Gurka D, Kumar A, Cheang M. 2006. Duration of hypotension before initiation of effective antimicrobial therapy is the critical determinant of survival in human septic shock. Critical Care Medicine 34: 1589-1596.

50) Lamothe B, Besse A, Campos AD, Webster WK, Wu H, Darney BG. 2007. Site-Specific Lys 63-Linked Tumour Necrosis Factor Receptor-Associated Factor 6 Auto-Ubiquination is a Critical Determinant of IKK Activation. Journal of Biological Chemistry 282: 4102– 4112.

51) Lee NY, Cheng JT, Enomoto T, Nakano Y. 2006. One Peptide Derived from Hen Ovotransferrin as Pro-Drug to Inhibit Angiotensin Converting Enzyme. Journal of Food and Drug Analysis 14: 31-35.

52) Leon F, Smythies LE, Smith PD, Kelsall BL. 2006. Involvement of Dendritic Cells in the Pathogenesis of Inflammatory Bowel Disease. In Immune Mechanisms of Inflammatory Bowel Disease 579: 117–132.

53) Levy MM, Fink MP, Marshall JC, Abraham E, Angus D, Cook D, Cohen J, Opal SM, Vincent JL, Ramsay G. 2003. 2001 SCCM/ESICM/ACCP/ATS/SIS International Sepsis Definitions Conference. Critical Care Medicine 31: 1250-1256.

54) Li H, Hu D, Fan H, Zhang Y, LeSage GD, Caudle Y, Stuart C, Liu Z, Yin D. 2014. β- Arrestin 2 negatively regulates Toll-like receptor 4 (TLR4)-triggered inflammatory signaling via targeting p38 MAPK and interleukin 10. Journal of Biological Chemistry 289: 23075–23085.

55) Libermann TA, Baltimore D. 1990. Activation of Interleukin-6 Gene Expression through the NF-κB Transcription Factor. Molecular and Cellular Biology 10: 2327-2334.

56) Loftus EV. Clinical epidemiology of inflammatory bowel disease: incidence, prevalence, and environmental influences. Gastroenterology 126: 1504–1517.

57) Logan RF, Edmond M, Somerville KW, Langman MJ. 1984. Smoking and ulcerative colitis. British Medical Journal (Clinical Research Ed.) 288: 751-753.

58) Lomaga MA, Yeh WC, Sarosi I, Duncan GS, Furlonger C, Ho A, Morony S, Capparelli C, Van G, Kaufman S, van der Hiden A, Itie A, Wakeham A, Khoo W, Sasaki T, Cao Z, Penninger JM, Paige CJ, Lacey DL, Dunstan CR, Boyle WY, Goeddel DV, & Mak TW. 1999. TRAF6 deficiency results in osteopetrosis and defective interleukin-1, CD40, and LPS signaling. Genes & Development 13: 1015–1024.

74 59) Maeshima M, Fernandez RC. 2013. Recognition of lipid Avariants by the TLR4-MD-2 receptor complex. Frontiers in Cellular and Infection Microbiology 3:1-13.

60) Majumder K, Chakrabarti S, Morton JS, Panahi S, Kaufman S. Davidge ST, Wu J. 2013. Egg-derived tri-peptide IRW exerts antihypertensive effects in spontaneously hypertensive rats. PLoS ONE 8: 1–14.

61) Martin GS, Mannino DM, Eaton S, Moss M. 2003. The Epidemiology of Sepsis in the United States from 1979 through 2000. The New England Journal of Medicine 348: 1546-1554.

62) Maruyama Y, Yasuda R, Kuroda M, Eto Y. 2012. Kokumi Substances, Enhancers of Basic Tastes, Induce Responses in Calcium-Sensing Receptor Expressing Taste Cells. PLoS ONE 7: e34489.

63) McCulloh RJ. 2012. Biomarkers in Sepsis and Severe Infection: Where Immunology Meets Diagnostics. Journal of Immunodeficiency & Disorders 1: 1-2.

64) Michelsen KS, Aicher A, Mohaupt M, Hartung T, Dimmeler S, Kirschning CJ, Schumann RR. 2001. The Role of Toll-like Receptors (TLRs) in Bacteria-iduced Maturation of Murine Dendritic Cells (DCs). The Journal of Biological Chemistry 276: 25680-25686.

65) Michielan A, D’Incà R. 2015. Intestinal Permeability in Inflammatory Bowel Disease: Pathogenesis, Clinical Evaluation, and Therapy of Leaky Gut. Mediators of Inflammation 2015: 1-10.

66) Miller YI, Choi, SH, Wiesner P, Bae YS. 2012. The SYK side of TLR4: signalling mechanisms in response to PLS and minimally oxidized LDL. British Journal of Pharmacology 167: 990-999.

67) Mine Y, Zhang H. 2015. Anti-inflammatory Effects of Poly-L-lysine in Intestinal Mucosal System by Calcium-Sensing Receptor Activation. Journal of Agricultural and Food Chemistry 63: 10437-10447.

68) Mine Y, Zhang H. 2015. Calcium-Sensing Receptor (CaSR)-Mediated Anti- inflammatory Effects of L-Amino Acids in Intestinal Epithelial Cells. Journal of Agricultural and Food Chemistry 63: 9987–9995.

69) Mogensen TH. 2009. Pathogen Recognition and Inflammatory Signaling in Innate Immune Defenses. Clinical Microbiology Reviews 2: 240-273.

70) Mun HC, Franks AH, Culverston EL, Krapcho K, Nemeth EF, Conigrave AD. 2004. The 2+ Venus Fly Trap Domain of the Extracellular Ca -sensing Receptor Is Required for L- Amino Acid Sensing. The Journal of Biological Chemistry 279: 51739-51744.

75 71) Nagaoka S, Futamura Y, Miwa K, Awano T, Yamauchi K, Kanamaru Y, Tadashi K, Kuwata T. 2001. Identification of Novel Hypocholesterolemic Peptides Derived from Bovine Milk β-Lactoglobulin. Biochemical and Biophysical Research Communications 281: 11-17.

72) O’Brien JM, Ali NA, Aberegg SK, Abraham E. 2007. Sepsis. The American Journal of Medicine 120: 1012-1022.

73) Ohsu T, Amino Y, Nagasaki H, Yamanaka T, Takeshita S, Hatanaka T, Maruyama Y, Miyamura N, Eto Y. 2010. Involvement of the Calcium-sensing Receptor in Human Taste Perception. Journal of Biological Chemistry 285: 1016–1022.

74) Orholm M, Binder V, Sørensen TIA, Rasmussen LP, Kyvik KO. 2000. Concordance of Inflammatory Bowel Disease among Danish Twins. Scandinavian Journal of Gastroenterology 25: 1075-1081.

75) Ortega-González, M, Capitán-Cañadas F, Requena P, Ocón B, Romero-Calvo I, Aranda C, Suárez MD, Zarzuelo A, Sánchez de Medina F, Martínez-Augustin O. 2014. Validation of bovine glycomacropeptide as an intestinal anti-inflammatory nutraceutical in the lymphocyte-transfer model of colitis. British Journal of Nutrition 111: 1202-1212.

76) Pahl HL. 1999. Activators and target genes of Rel/NF-κb transcription factors. Oncogene 18: 6853-6866.

77) Patel RT, Deen KI, Youngs D, Warwick J, Keighley MRB. 1994. Interleukin 6 is a prognostic indicator of outcome in severe intra-abdominal sepsis. The British Journal of Surgery 81: 1306-1308.

78) Petesen LW, Artis D. 2014. Intestinal epithelial cells: regulators of barrier function and immune homeostasis. Nature Reviews Immunology 14: 141-153.

79) Pickart CM. 1997. Targeting of substrates to the 26S proteasome. Federation of American Societies for Experimental Biology Journal 11: 1055–1066.

80) Pierrakos C, Vincent JL. 2010. Sepsis biomarkers: a review. Critical Care 14: R15.

81) Raetz CRH, Whitfield C. 2002. Lipopolysaccharide Endotoxins. Annual Review of Biochemistry 71: 635-700.

82) Ranieri VM, Thompson BT, Barie PS, Dhainaut JF, Douglas IS, Finfer S, Gårdlung B, Marshall JC, Rhodes A, Artigas A, Payen D, Tenhunen J, Al-Khalidi HR, Thompson V, Janes J, Macias WL, Vangerow B, Williams MD. 2012. Drotrecogin Alfa (Activates) in Adults with Septic Shock. The New England Journal of Medicine 366: 2055-2064.

83) Reinhart K, Karzai W. 2001. Anti-tumour necrosis factor therapy in sepsis: Update on clinical trials and lessons learned. Critical Care Medicine 29: 121-125.

84) Rivers E, Nguyen B, Havstad S, Ressler J, Muzzin A, Knoblich B, Peterson E, Tomlanovich M. 2001. Early Goal-Directed Therapy in the Treatment of Severe Sepsis and Septic Shock. The New England Journal of Medicine 345: 1368-1377.

85) Rutgeerts P, D’Haens G, Hiele M, Geboes K, Vantrappen G. 1994. Appendectomy protects against ulcerative colitis. Gastroenterology 106: 1251-1253.

86) Salminen S, Bouley C, Boutron-Ruault MC, Cummings JH, Franck A, Gibson GR, Isolauri E, Moreau MC, Roberfroid M, Rowland I. 1998. Functional food science and gastrointestinal physiology and function. British Journal of Nutrition 80: S147-S171.

87) Sartor RB. 2006. Mechanisms of Disease: pathogenesis of Crohn’s disease and ulcerative colitis. Nature Clinical Practice Gastroenterology & Hepatology 3: 390-407.

88) Schneeman BO. 2002. Gastrointestinal physiology and functions. The British Journal of Nutrition 88: S159–S163.

89) Sequeira IR, Lentle RG, Kruger MC, Hurts RD. 2014. Standardising the Lactulose Mannitol Test of Gut Permeability to Minimise Error and Promote Comparability. PLoS One 9: 1-12.

90) Shakhov AN, Collart MA, Vassalli P, Nedospasov SA, Jongeneel CV. 1990. κB-Type Enhancers are Involved in Lipopolysaccharide-Mediated Transcriptional Activation of the Tumour Necrosis Factor α Gene in Primary Macrophages. The Journal of Experimental Medicine 171: 35-47.

91) Shi Y, Rupa P, Bo J, Mine Y. 2014. Hydrolysate from eggshell membrane ameliorates intestinal inflammation in mice. International Journal of Moelcular Sciences 15: 22728- 22742.

92) Shimazu R, Akashi S, Ogata H, Nagai Y, Fukudome K, Miyake K, Kimoto M. 1999. MD-2, a Molecule that Confers Lipopolysaccharide Responsiveness on Toll-like Receptor 4. The Journal of Experimental Medicine 189: 1777-1782.

93) Sigma-Aldrich: The Challenge of Manufacturing Fermentation Derived Lipopolysaccharides (LPSs). Available at: http://www.sigmaaldrich.com/safc/news-and- resource-center/technical-reference-library/fermentation-derved- lipopolysaccharides.html. Accessed February 2016.

94) Slade E, Tamber P, Vincent JL. 2003. The Surviving Sepsis Campaign: raising awareness to reduce mortality. Critical Care 7, 1–2.

95) Sriskandan S, Altmann D. 2008. The immunology of sepsis. The Journal of Pathology 214: 211–223.

77 96) Starr MS, Steele AM, Saito M, Hacker BJ, Evers BM, Saito H. 2014. A New Cecal Slurry Preparation Protocol with Improved Long-Term Reproducibility for Animal Models of Sepsis. PLoS One 9: e115705.

97) Sukhotnik I, Mogilner J, Krausz MM, Lurie M, Hirsh M, Coran AG, Shiloni E. 2004. Oral arginine reduces gut mucosal injury caused by lipopolysaccharide endotoxemia in rat. The Journal of Surgical Research 122: 256–62.

98) Suzuki H, Kato K, Kumagai H. 2004. Enzymatic Synthesis of γ-Glutamylvaline to Improve the Bitter Taste of Valine. Journal of Agricultureal and Food Chemistry 52: 577- 580.

99) Takeda K, Akira S. 2004. TLR signaling pathways. Seminars in Immunology 16: 3-9.

100) Tfelt-Hansen J, & Brown EM. 2013. Calcium-Sensing Receptor (CaSR). In Encyclopedia of Biological Chemistry (2nd ed., pp. 326–330). Elsevier Inc.

101) The Human Metabolome Database (HMDB): Gamma-glutamyl-valine. Available at: http://www.hmdb.ca/metabolites/HMDB29162. Accessed February 2016.

102) Thomas L. 1972. Germs. The New England Journal of Medicine 287: 553-555.

103) Tobias PS, Soldau K, Ulevitch RJ. 1986. Isolation of a Lipopolysaccharide- Binding Acute Phase Reactant from Rabbit Serum. The Journal of Experimental Medicine 164: 777-793.

104) Tobias PS, Soldau K, Ulevitch RJ. 1989. Identification of a Lipid A Binding Site in the Acute Phase Reactant Lipopolysaccharide Binding Protein. The Journal of Biological Chemistry 264: 10867-10871.

105) Tobin MV, Logan RF, Langman MJ, McConnell RB, Gilmore IT. 1987. Cigarette smoking and inflammatory bowel disease. Gastroenterology 93: 316-321.

106) van der Poll T, Opal SM. 2007. Host-pathogen interactions in sepsis. The Lancet Infectious Diseases 8: 32-43.

107) Vatn MH, Sandvik AK. 2015. Inflammatory bowel disease. Scandinavian Journal of Gastroenterology 50: 748–762.

108) Vincent JL, Moreno R, Takala J, Willatts S, De Mendonça A, Bruining H, Reinhart CK, Suter PM, Thijs LG. 1996. The SOFA (Sepsis-related Organ Failure Assessment) score to describe organ dysfunction/failure. Intensive Care Medicine 22: 707-710.

109) Vincent JL, Sakr Y, Sprung CL, Ranieri VM, Reinhart K, Gerlach H, Moreno R, Carlet J, Le Gall JR, Payen D. 2006. Sepsis in European intensive care units: results of

78 the SOAP study. Critical Care Medicine 34: 344-353.

110) Virtanen AI, Matikkala EJ. New γ-glutamylpeptides in onion (Allium cepa). Hopper-Seyler’s Z. Physiol. Chem 322: 8-20. (VERIFY)

111) Wang Y, Tang Y, Teng L, Wu Y, Zhao X, & Pei G. 2006. Association of β- arrestin and TRAF6 negatively regulates Toll-like receptor-interleukin 1 receptor signaling. Nature Immunology 7: 139–147.

112) Witherow DS, Garrison TR, Miller WE, Lefkowitz RJ. 2004. β-arrestin inhibits NF-κB activity by means of its interaction with the NF-κB inhibitor IκBα. Proceedings of the National Academy of Sciences 101: 8603-8607.

113) Wright SD, Ramos RA, Tobias PS, Ulevitch RJ, Mathison JC. 1990. CD14, a Receptor for Complexes of Lipopolysaccharide (LPS) and LPS Binding Protein. Science 249: 1431-1433.

114) Wu H, Arron JR. 2003. TRAF6, a molecular bridge spanning adaptive immunity, innate immunity and osteoimmunology. BioEssays 25: 1096–1105.

115) Yang RB, Mark MR, Gurney AL, Godowski PJ. 1999. Signaling events induced by lipopolysaccharide-activated toll-like receptor 2. The Journal of Immunology 163: 639–643.

116) Young D, Ibuki M, Nakamori T, Fan M, Mine Y. 2012. Soy-Derived Di- and Tripeptides Alleviate Colon and Ileum Inflammation in Pigs with Dextran Sodium Sulfate-Induced Colitis. The Journal of Nutrition 142: 363–368.

117) Zeng KW, Liao LX, Lv HN, Song FJ, Yu Q, Dong X, Li J, Jiang Y, Tu PF. 2015. Natural small molecule FMHM inhibits lipopolysaccharide-induced inflammatory response by promoting TRAF6 degradation via K48-linked polyubiquitination. Scientific Reports 5: 14715.

118) Zhang H, Kovacs-Nolan J, Kodera T, Eto Y, Mine Y. 2015. γ-Glutamyl cysteine and γ-glutamyl valine inhibit TNF-α signaling in intestinal epithelial cells and reduce inflammation in a mouse model of colitis via allosteric activation of the calcium sensing receptor. Biochimica et Biophysica Acta 1852: 792–804.