Gosaccharides (Gos) Supplementation: a Randomized, Double-Blind Controlled Study in Healthy Aged Adults

Home , Galactooligosaccharide

IMMUNOLOGICAL AND GASTROINTESTINAL EFFECTS OF GALACTOOLIGOSACCHARIDES (GOS) SUPPLEMENTATION: A RANDOMIZED, DOUBLE-BLIND CONTROLLED STUDY IN HEALTHY AGED ADULTS

STEPHANIE-ANNE GIRARD

A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA

2012

To my parents, Yves and Danielle, and to all of those that have supported me during this incredible and life changing experience

ACKNOWLEDGMENTS

Firstly, I would like to thank my mentor Dr. Bobbi Langkamp-Henken who has successfully guided me through this journey where many obstacles were encountered but all surpassed. She was always there to motivate and encourage me whilst helping me acquire new skills along the way. I would also like to acknowledge all of the members of my committee; Dr. Wendy Dahl, Dr. Susan Percival and Dr. Volker Mai who have challenged me and helped me attain the level of knowledge that I possess today.

I give my thanks to my parents, Yves and Danielle, who have been with me, although not physically but rather emotionally and mentally, and without their support I surely could not have completed this degree. They have encouraged me throughout these three years and helped me reach my ultimate goal. I would certainly not have completed my PhD without the continual support from my lifelong friend and companion

Kaunteya Nundy. I have to share this doctoral degree with him as we have together achieved this huge success. As we prepare to build a life together, I feel that we have accomplished this steppingstone to our prosperous, both with love and success, future together.

A special thanks go to all of the current and former members of my laboratory; CJ

Nieves, Gretchen Specht, Sam Spaiser, Ally Radford and Christine Hughes. They certainly made my days sunnier and kept my hopes up during hard times. I must also acknowledge all of the undergraduate students that have worked with me and helped me execute all of the laboratory assays as well as conduct this incredible study; Elaine

Tan, Christina Vitale, Nicholas Morrell, Vivek Iyer, and Paula Beers.

Additionally, fellow graduate students of collaborating laboratories such as

Rebecca Creasy and Tyler Culpepper have always been there for me when I needed a 4 friend. They have helped me understand concepts and master techniques and without their support and friendship, I could not have excelled and achieved my goals.

TABLE OF CONTENTS

page

ACKNOWLEDGMENTS ...... 4

LIST OF TABLES ...... 9

LIST OF FIGURES ...... 10

LIST OF ABBREVIATIONS ...... 11

ABSTRACT ...... 14

CHAPTER

1 INTRODUCTION AND SPECIFIC AIMS ...... 16

2 REVIEW OF THE LITERATURE ...... 18

Introductory Notes ...... 18 Galactooligosaccharides ...... 19 Properties ...... 19 GRAS Status and Health Claims ...... 20 Research Involving Galactooligosaccharides ...... 21 Stool improvement ...... 23 Mineral absorption ...... 24 Lipid metabolism ...... 25 Immunomodulation ...... 26 Carcinogenesis ...... 28 Common Cold Symptoms and Complementary Health Practices ...... 28 Cold Assessment in Research ...... 30 Nutritional Status, Immune Function and Common Cold ...... 31 Nutritional Modulation of Common Cold ...... 33 Echinacea ...... 34 Ascorbic acid ...... 35 Zinc ...... 36 Multivitamins and minerals ...... 39 Garlic ...... 39 Probiotics ...... 41 Prebiotics ...... 42 Age and Common Cold ...... 43 Immune System ...... 45 Leukocytes ...... 45 Serum Acute Phase Proteins ...... 46 Peripheral Blood Mononuclear Cell (PBMC) Cytokine Production to Mitogens...... 47 Immunosenescence ...... 50 6

Mucosal Immunity ...... 52 Salivary and Fecal Secretory IgA (sIgA) ...... 54 Digestive Health ...... 56 Microbiota ...... 56 Gastrointestinal Symptom Rating Scale ...... 57 GOS and the Microbiota ...... 58 Aged Adults and the Microbiota ...... 59

3 METHODS ...... 66

Participants ...... 66 Experimental Design ...... 67 GOS Administration Protocol ...... 68 Influenza Vaccination ...... 69 Study Questionnaires...... 69 Mini Nutritional Assessment ...... 71 Food Frequency Questionnaire (FFQ) ...... 73 Reagents ...... 74 Blood Sample Collection ...... 74 Serum Collection ...... 75 Phenotypic Determination of Lymphocyte Population ...... 75 CRP Assay ...... 77 PBMC Isolation ...... 77 PBMC Stimulation with Mitogens (PHA and LPS) Assay ...... 78 Multiplex Bead Based Assay ...... 78 Saliva Collection ...... 79 Salivary sIgA Assay ...... 80 Fecal Collection ...... 80 Fecal sIgA Assay ...... 81 Microbiota Analysis ...... 81 Statistical Analysis ...... 82

4 RESULTS ...... 87

Self-Reported Cold Days, Symptoms and Symptom Intensities ...... 88 Quality of Life Measurements ...... 89 Lymphocytes Subpopulation Cell Numbers ...... 90 Cytokine Produced Following PHA Stimulation of Peripheral Blood Mononuclear Cells...... 91 Cytokine Produced Following LPS Stimulation of Peripheral Blood Mononuclear Cells...... 92 Salivary and Fecal Secretory Immunoglobulin A and C-Reactive Protein...... 92 GSRS Questionnaire ...... 93 Food Frequency Questionnaire ...... 93 Bifidobacteria Genome Equivalent (GE) Proportion Changes ...... 94

5 DISCUSSION AND CONCLUSIONS ...... 113

Effect of GOS on Symptom Intensity and Quality of Life...... 116 GOS and the Immune System ...... 117 Increase in Immunogenic Bacteria ...... 118 GOS and Digestive Health ...... 120 Preventing Attachment of Pathogenic Bacteria ...... 122 How GOS through its Influence on Proliferation of Beneficial Bacteria within the Colon May Affect URTI ...... 122 Limitations of the Study...... 125 Conclusions ...... 126

APPENDIX

A IRB APPROVAL LETTER ...... 128

B IRB INFORMED CONSENT ...... 130

C QUESTIONNAIRES ...... 142

D DESCRIPTIVE STATISTICS OF PLACEBO AND GALACTOOLIGOSACCHARIDES BY AGE GROUP ...... 144

LIST OF REFERENCES ...... 150

BIOGRAPHICAL SKETCH ...... 170

LIST OF TABLES

Table page

1-1 Age-related changes in production of cytokines following stimulation of isolated peripheral blood mononuclear cells with mitogens ...... 64

4-1 Demographics and compliance ...... 100

4-2 Self-reported cold days ...... 102

4-3 Quality of life measurements on days of self-reported cold ...... 104

4-4 Lymphocytes subpopulations cell numbers per microliter of whole blood by treatment and responder group, and blood draw ...... 105

4-5 Cytokines produced following phytohemagglutinin stimulation of peripheral blood mononuclear cells by treatment and responder group, and blood draw . 107

4-6 Cytokines produced following lipopolysaccharide stimulation of peripheral blood mononuclear cells by treatment and responder group, and blood draw . 109

4-7 Salivary and fecal secretory immunoglobulin A and C-reactive protein by treatment and responder group, and blood draw ...... 111

D-1 Descriptive statistics of placebo and galactooligosaccharides supplementation by age group ...... 145

LIST OF FIGURES

Figure page

1-1 General structure of a β-(1-4) linked galactooligosaccharide...... 63

1-2 Effects of aging on the microbiota and the subsequent inflammatory response ...... 63

3-1 Study design ...... 86

4-1 Flow chart of participant recruitment, allocation and analysis ...... 95

4-2 Effect of treatment group and age group on the probability of self-reporting a cold over the 6-month intervention ...... 96

4-3 Effect of treatment group and age group on concentration of interferon-γ produced following stimulation of peripheral blood mononuclear cells ...... 97

4-4 Effect of treatment group and age group the percentage of lymphocytes that were identified as natural killer cells ...... 98

4-5 Monthly gastrointestinal symptoms in participants receiving the placebo or galactooligosaccharides ...... 99

LIST OF ABBREVIATIONS

AGE Aged Garlic Extract

ANOVA Analysis of Variance

AP Autologous Plasma

APC Antigen Presenting Cells

B cell B Lymphocyte

BMI Body Mass Index

CC Calf Circumference

CD Cluster Designation

DC Dendritic cells

DGGE Denaturing Gradient Gel Electrophoresis

EFSA European Food Safety Authority

ELISA Enzyme-Linked Immuno-Sorbent-Assay

FBS Fetal Bovine Serum

FDA U.S. Food and Drug Administration

FOS Fructooligosaccharides

GALT Gut-Associated Lymphoid Tissue

GE Genome Equivalents

GI Gastrointestinal Tract

GOS Galactooligosaccharides

GSRS Gastrointestinal Symptom Rating Scale

HDL High Density Lipoprotein

HEPES 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid

HIV Human Immunodeficiency Virus

IBD Inflammatory Bowel Disease

ICAM-1 Intracellular Cell Adhesion Molecule-1

IFN-ɣ Interferon-ɣ

IL Interleukin

Ig Immunoglobulin

IRB Institutional Review Board

LAB Lactic Acid Bacteria

LPS Lipopolysaccharide

MAC Mid-Arm Circumference mg Milligram mL Milliliter

MHC Major Histocompatibility Complex

MLN Mesenteric Lymph Node

MNA Mini Nutritional Assessment

NK cell Natural Killer cell

PAMP Pathogen Associated Molecular Pattern

PRR Pattern Recognition Receptor

PBMC Peripheral Blood Mononuclear Cell

PHA-L Phytohemagglutinin-Leukocyte

RBC Red Blood Cell rpm Revolution Per Minute

RPMI Roswell Park Memorial Institute

RT Room Temperature

SCFA Short Chain Fatty Acid

SEM Standard Error of the Mean

SI Symptom Intensity

12 sIgA Secretory Immunoglobulin A

T cell T lymphocyte

TCR T Cell Receptor

Th T helper

TNF-α Tumor Necrosis Factor-α

TOS Transgalactooligosaccharides

Treg T regulatory

U Units

μg Microgram

μL Microliter

μM Micromolar

URTI Upper Respiratory Tract Infection

WURSS Wisconsin Upper Respiratory Syndrome Survey

Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy

IMMUNOLOGICAL AND GASTROINTESTINAL EFFECTS OF GALACTOOLIGOSACCHARIDES (GOS) SUPPLEMENTATION: A RANDOMIZED, DOUBLE-BLIND CONTROLLED STUDY IN HEALTHY AGED ADULTS

Stephanie-Anne Girard

August 2012

Chair: Bobbi Langkamp-Henken Major: Nutritional Sciences

Galactooligosaccharides (GOS) are resistant to digestion and selectively stimulate the growth and activity of beneficial bacteria within the colon conferring health benefits and are therefore classified as prebiotics. In healthy older adults, GOS supplementation improved natural killer (NK) cell activity and altered cytokine production but no studies examine whether these are associated with a reduction in probability of upper respiratory tract infection (URTI). The purpose of this work was to examine the ability of

GOS in maintaining digestive health and immune strength over cold/flu season. It was hypothesized that participants consuming GOS would have decreased diarrhea, indigestion, constipation and reflux syndromes and a lower probability of URTI due to alterations in the microbiota and cytokine production. In a randomized, double-blind study, adults received 0 g (n=38) or 5 g (n=43) of GOS daily for 24 weeks. Blood was drawn at week 0, 3 (seasonal influenza vaccination) and 5 (2 weeks post-vaccination).

Stool samples were collected at week 0 and during week 2. Gastrointestinal and cold/flu symptoms were collected daily over the 24-week study. Peripheral blood mononuclear cells were stimulated in vitro with mitogens and cytokine production was

14 quantified. Lymphocyte subpopulations were determined. Differences were observed between treatment groups where GOS significantly reduced digestive symptoms of constipation (p=0.01) and abdominal pain (p=0.02). The probability of URTI was not different between GOS and placebo group. However, participants receiving GOS were significantly older (p=0.04); therefore participants were sub-divided by age, Younger

Aged (˂65 y) and Older Aged (65 y), and the data were reanalyzed. Older Aged participants receiving GOS had a significantly lower probability of self-reporting a cold

(33±10%) compared to Younger Aged participants (77±9%) receiving GOS (p=0.0332).

Older Aged participants in the GOS group had a significantly higher percentage of NK cells (17.9±1.3%) than Younger Aged participants (12.3±1.3%) receiving GOS

(p=0.0022). Older Aged participants receiving GOS had a significantly (p=0.006) higher in vitro production of interferon (IFN)-ɣ (11100±1520 pg/mL) compared to the placebo

(4460±1210pg/mL). The lower probability of URTI in the Older Aged participants receiving GOS may be due to a higher percentage of NK cells or an indirect effect of

IFN-ɣ.

CHAPTER 1 INTRODUCTION AND SPECIFIC AIMS

Factors such as various diseases, genetics and nutrition can influence the immune system. In recent years, an increase in interest of products exhibiting immune- modulatory properties has been observed and this is reflected by a significant increase in foods or beverages with structure/function claims (1). One such category of dietary supplements are prebiotics which are defined as food ingredients resistant to digestion in the stomach that selectively stimulate the growth and activity of select bacteria within the colon which then confers health benefits upon the host (2). One such prebiotic recognized by experts for its clinical evidence of digestive and immune health is the non-digestible carbohydrate galactooligosaccharides (GOS). Because of their structural similarity to some of the oligosaccharides found in breast milk which have been found to have bifidogenic effects, GOS have been added to infant formulas. Although GOS have received increased interest, especially in the pediatric field where GOS has been shown to exert immune enhancing effects in infants (3), studies are still lacking as to whether

GOS helps maintain immune and digestive health in adults. Galactooligosaccharides have previously been shown to reduce days of upper respiratory tract infection (URTI) in academically stressed undergraduate students (4). Additionally, studies have shown improved natural killer cell activity and altered cytokine production with the addition of

GOS to the diets of healthy older adults (5); however, no studies examine whether these changes in immune parameters are associated with a reduction in probability of

URTI. Since the immune system of aged adults exhibits diminished functions; perhaps supplementation of GOS could improve immune function which would enhance nonspecific resistance of the host to infections such as the common cold.

The purpose of this study was to determine whether the functional fiber GOS can help maintain immune strength and digestive health in free-living older adults. More specifically, this study was conducted to determine if GOS would decrease the probability of self-reporting a cold through modulation of the microbiota, increasing the amounts of bifidobacteria genome equivalents and immune parameter such as NK cells.

It is hypothesized that older adults consuming GOS daily for 24 weeks over cold and flu season will have a decreased probability of self-reporting a cold due to a proliferation of beneficial bacteria within the colon enhancing immune parameters and altering cytokine production. It is also hypothesized that GOS will decrease the symptom intensity (SI) score of cold symptoms on self-reported cold days and days with cold symptoms as a proportion of total study days. Following supplementation with

GOS, immune cells such as natural killer (NK) cells will increase in total cell numbers and as a percentage of total lymphocytes. It is believed that changes in lymphocyte subpopulations will subsequently increase the production of certain cytokines. It is also hypothesized that through modulation of the microbiota, GOS will improve gastrointestinal symptoms such as constipation and diarrhea syndromes.

CHAPTER 2 REVIEW OF THE LITERATURE

Introductory Notes

Dietary supplementation with micro and macronutrients is believed to maintain good health and decrease the symptoms and duration of debilitating illnesses such as the common cold. Galactooligosaccharides are classified as prebiotics and have been known to possess immune enhancing effects in infants (3). These non-digestible carbohydrates are generally recognized as safe (GRAS) and added to infant formula to more closely resemble breast milk; however, studies are lacking as to whether they help maintain health in adults. Days of URTI have previously been shown to be reduced with

GOS in academically stressed undergraduate students (4). Additionally, a study conducted by Vulevic and colleagues has shown improved natural killer cell activity and altered cytokine production with the addition of GOS to the diets of healthy older adults

(5); however, no studies examine whether these changes in immune parameters in aged adults are associated with a reduction in probability of URTI. Topics covered in this literature review are GOS, their GRAS status and various studies involving their use, cold and flu symptoms, cold assessment in research, the effects of nutritional status, nutrients and age on cold and flu. The immune system will be reviewed; cells of the immune system, cytokines produced following stimulation by mitogens, immunosenescence as well the mucosal immune system. Finally, the methodology used to evaluate gastrointestinal (GI) symptoms, and factors affecting the microbiota such as GOS and age are also included.

Galactooligosaccharides

Properties

Galactooligosaccharides are galactose-containing non-digestible carbohydrates naturally found in human milk or synthesized from lactose. Collectively, galactose- containing oligosaccharides are found in breast milk at a concentration of approximately

1 g/L. Galactooligosaccharides, in addition to a wide variety of carbohydrates, are considered prebiotics. Candidate prebiotics comprise carbohydrates such as manno- oligosaccharides, pectic-oligosaccharides, soybean-oligosaccharides, isomalto- oligosaccharides, xylo-oligosaccharides and lactulose which can be further classified according to their chemical structure and degree of polymerization (3). However, oligosaccharides such as inulin, fructooligosaccharides (FOS) – naturally found in fruits, vegetables and grains, and GOS are the most commonly studied (6). Since FOS and

GOS are considered as being commercially safe, historically, they have been widely used in food products. According to a report from the Global Industry Analysts, the popularity of prebiotic ingredients is on the rise on the US market and projected sales are thought to reach nearly 225 million dollars by the year 2015 (7). The promising future of the prebiotic market, mainly through heightened public awareness, has triggered increased interest in prebiotic research and more information and strict guidelines are now becoming available on these useful dietary tools.

In order to be characterized as a prebiotic, a food ingredient must satisfy the three following criteria; resist metabolic fate (resisting gastric acidity, hydrolysis by mammalian enzymes and absorption in the upper gastrointestinal tract), allow for fermentation (fermented by the intestinal microbiota) and have an effect on gastrointestinal microbiota (stimulating the growth and activity of intestinal bacteria

19 leading to potential health and well-being) (2). The prebiotic used in this study,

PurimuneTM, was obtained from GTC Nutrition and is a mixture of β-linked GOS in various configurations β(1-3), β(1-4) and β(1-6) as well as differing degrees of oligomerization which range between 3 and 5 (Figure 1-1). Galactooligosaccharides contain at least 90.4% of GOS, on a dry weight basis, with the remaining material being principally lactose, water as well as small amounts of two monosaccharides – dextrose and galactose (8). Galactooligosaccharides are manufactured from food grade lactose by means of a transgalactosylation reaction which is catalyzed by a β-galactosidase enzyme obtained from the non-toxigenic and non-pathogenic bacteria Bacillus circulans

LOB 377. The transgalactosylation results in the formation of 4’- or 6’- galactosylactose, longer oligosaccharides, transgalactosylated disaccharides and non- reducing oligosaccharides (9).

GRAS Status and Health Claims

Inulin, FOS and GOS are recognized by the U.S. Food and Drug Administration

(FDA) as being safe (2). In 2009, following scientific procedures executed by GTC

Nutrition, PurimuneTM received the GRAS status and this notice is available online (8).

Since GOS are traditionally found in yogurts in Japan and because they can be produced from ingested lactose by the resident intestinal microbiota (10), they are designated as Food for Specific Health Use (FOSHU) by Japanese authorities and have a non-novel food status in the EU (11).

At present, many countries do not have the premarket restrictions by which regulatory agencies approve or disapprove of prebiotics because they lack an established system for health claims. As a general rule, scientific validation of health claims attributed to marketed products should still be available in the event that 20 authorities request it (2). In the United States, most marketing claims used for prebiotics are structure/function claims due to the fact that health claims and nutrient content claims cannot currently be used; the FDA has yet to approve such claims and a daily content value for prebiotics has still not been established (2). Recently the

European Food Safety Authority (EFSA) rejected a health claim, attributing a reduction in disease risk (traveler’s diarrhea) for BiMuno, a B-GOS marketed in the UK.

Furthermore, two health claims, which were based on newly developed scientific evidence, and a request for the protection of proprietary data for BiMuno “helps maintain a healthy gastrointestinal function” and “supports your natural defenses” were also rejected (12) despite the fact that a double-blind, placebo controlled study looking at the effect of BiMuno on traveler’s diarrhea (13) was published a few months earlier.

Nonetheless, in countries like Japan, oligosaccharides and various dietary fibers are present in foods which are approved for claims such as “modifies gastrointestinal conditions” and all products that contain fiber and/or oligosaccharides are allowed to display the health claim “suitable for improving the regulation and control of the gastrointestinal tract” or “improves bowel movement”. In regards to prebiotics, additional information obtained from clinical trial results, is required to assist regulatory agencies in coming to a consensus as to what marketing claims should be allowed.

Research Involving Galactooligosaccharides

Since the beginning of the last century, it was thought that “factors” present in breast milk promoted the growth of certain types of intestinal bacteria (14). Breastfed infants seemed to have an intestinal microbiota largely dominated by bifidobacteria resulting from their ability to utilize oligosaccharides found in breast milk (15). For this reason, many studies involving GOS have investigated its effect on microbiological 21 changes. Although many studies have examined the effects of a mixture of 90% GOS and 10% long chain FOS (inulin) at a 9:1 ratio to simulate the composition of human milk (16, 17), for the purpose of this literature review, studies relating to GOS research; stool improvement, mineral absorption, lipid metabolism, immunomodulation and carcinogenesis, use GOS as the sole substrate unless stated otherwise.

A placebo controlled study executed by Ito et al. (18) showed that a daily dose of

10 g of GOS fed to 12 healthy male volunteers for 8 weeks increased fecal bifidobacteria and lactobacilli when compared to a placebo. A second study executed by these same investigators looked at the effects of feeding 2.5 g of GOS daily (rather than

10 g) for 3 weeks (instead of 8 weeks). The results obtained from 12 healthy males suggested that feeding a smaller amount of GOS over a shorter period of time still increased bifidobacteria in the feces as well as decreased numbers of clostridia and bacteroides (19). An additional study examining changes in the microbiota was conducted in 30 participants. Each participant was randomized to receive 8.1 g of GOS

(n=8), 8.1 g of GOS with 3 x 1010 Bifidobacterium lactis Bb-12 (n=10) or 3 x 1010

Bifidobacterium lactis alone (n=10) (20). Unlike the previous studies, little change in fecal bifidobacteria was observed in the group receiving GOS alone. An explanation of these results is the method of detection used (polymerase chain reaction-enzyme-linked immunosorbent assay; PCR-ELISA) which gives relative amounts of bifidobacteria rather than total amounts.

After ingestion, non-digestible poly and oligosaccharides are used as substrates for microbial fermentation and selectively stimulate the growth of bifidobacteria in complex microbial communities that exist in the large intestine. The major end-products

22 of prebiotic fermentation by bacteria are short chain fatty acids (SCFA) (i.e., acetate, propionate and butyrate) (11). Since more than 95 percent of SCFA are rapidly absorbed by the colonocytes in the large intestine, fecal measurements of SCFA do not provide useful information about the fermentability of different prebiotics. Although in vivo methods of measurements of SCFA still remain to be developed, Bouhnik et al.

(21) studied the fermentation profile of GOS using fecal homogenate in an in vitro study model. The investigators showed that the addition of 10 g of GOS to batch cultures inoculated with human fecal homogenates resulted in increased acetate and lactate production which was not observed in control fermenters. In fact, acetate and lactate formation in the previous study is consistent with other findings demonstrating that these two SCFA are produced from bifidobacteria and lactobacillus metabolism and that butyrate does not seem to be an end-product. In terms of in vivo conditions, it is likely that lactate produced by the fermentation of GOS is in turn utilized by butyrate producing bacteria which increases butyrate production as an indirect consequence.

Additionally, microbial fermentation of GOS should liberate small amounts of carbon dioxide, hydrogen, and methane gas (22).

Stool improvement

Many studies involving the use of lactulose, FOS and GOS have demonstrated that these prebiotics possess laxative effects. The gastroenterological effects conferred by prebiotics have been exhaustively studied and tools are available to assess their effects on constipation. Although frequency of defecation is generally used as a crude measurement of constipation, additional questionnaires such as the Gastrointestinal

Symptom Rating Scale (GSRS) can be used to assess this syndrome. The GSRS comprises of questions addressing constipation and defines it as a reduced ability to 23 empty the bowels, being bothered by hardened stool, and the feeling that after finishing a bowel movement, there are still some stools that need to be passed (23). A study by

Deguchi et al. examined the outcome of giving 75 women with a propensity to suffer from constipation either 2.5 g or 5 g of GOS daily for seven days (24). The investigators concluded that 5 g of GOS significantly improved the defecation frequency. A double blind, randomized, crossover controlled trial conducted in fourteen constipated elderly women investigated whether a yoghurt containing 9 g of GOS could relieve constipation

(25). The results, although not significant, suggested that GOS seemed to make defecation easier (p=0.07). Additionally, when participants were consuming the GOS containing yogurt, the mean defecation frequency per week (7.1stools/week) was higher than during consumption of the placebo (5.9 stools/week).

Mineral absorption

The vast bulk of mineral absorption occurs in the small intestine and the principal route is through the paracellular route (26). Since some of the determinants of mineral absorption are mineral solubility, permeability of the intestinal epithelium and luminal transit time, administration of prebiotics and its effects on mineral absorption have been studied. Minerals in the gastrointestinal (GI) tract are present in forms that are poorly absorbable and production of SCFA by the microbial fermentation of prebiotics acidifies the gut lumen and increases the solubility and absorption of minerals such as calcium.

The majority of studies looking at mineral metabolism and GOS have been executed using animal models. In experiments conducted in rats, GOS supplementation seemed to stimulate calcium absorption and this resulted in an improvement in the efficiency of magnesium absorption (27-30). Human studies of mineral absorption have shown inconsistent results. A study conducted with healthy male volunteers supplemented 24 with 15 g of GOS, FOS or inulin looked at calcium and iron absorption through 24 hours urine collections and measurements of calcium isotopes and found no effect on calcium and iron absorption with any of the prebiotics (31). A later study by the same group, though executed with healthy postmenopausal women, looked at the effect of 20 g of transgalactooligosaccharides (TOS) on calcium absorption. The results from this study showed by measuring administered calcium isotopes in urine collections a significant increase in calcium absorption in this population (32).

Lipid metabolism

To reiterate, the ingestion and subsequent fermentation of prebiotics lead to the formation of SCFA. In turn, SCFA such as propionate and acetate can be absorbed along the colon and reach the liver through the portal vein (33). Once acetate enters the liver and through enzymatic activation, it can enter the cholesterogenesis and lipogenesis pathways. It is thought that acetate can downregulate the rate-limiting enzyme in the cholesterol synthesis pathways hence this could explain the hypocholesterolemic effect of prebiotics, such as lactulose, and partly explain the link from diet to atherosclerosis and cardiovascular disease (34). Several animal studies have suggested that prebiotics, such as inulin and FOS, could positively affect serum lipids by lowering blood cholesterol levels (35). However, results from studies executed in human clinical trials are still conflicting. Thus far, two studies have been completed with human participants, both comparing GOS to FOS, inulin or to a placebo and measuring serum cholesterol levels. The first study by van Dokkum et al. (36) was a double-blind, randomized and controlled trial where 12 young healthy male participants were given 15 g per day of GOS, FOS and inulin (during different treatment periods of 3 weeks for each prebiotic) and the authors observed no significant difference in blood 25 lipids or glucose absorption. The second study was completed in healthy aged adults where participants were either given 5.5 g of a GOS mixture or a placebo (5). The results from this study also showed no significant difference in total serum cholesterol or high-density lipoprotein (HDL) cholesterol levels. The commonality between these two studies is that they were both executed in normolipidemic participants. Further studies could potentially show a decrease in total serum cholesterol and/or an increase in HDL cholesterol if they are done in participants with initial elevated serum cholesterol levels.

Immunomodulation

The gastrointestinal tract has been known to be a significant component of the immune system. This is perhaps evolutionarily attributable to the fact that this organ is in continuous interaction with foreign antigenic stimuli from food and commensal or pathogenic bacteria (37). The gut-associated lymphoid tissue (GALT), the mucosal immune system found in the intestine is comprised of innate and adaptive immune cells.

The GALT is presumed to be made up of aggregated lymphoid tissue (solitary lymphoid follicles and Peyer’s Patches) as well as non-aggregated lymphoid tissue (lamina propria lymphoid cells and intraepithelial lymphocytes) (38). Since the GALT is in continuous interaction with the commensal bacteria within our gut lumen as well as their metabolic byproducts, dietary substrates such as prebiotics, which are able to be fermented by and influence the microbiota, should in turn indirectly affect the GALT

(11).

Several means by which prebiotics are thought to impact the immune system have been hypothesized. An example being attenuation of inflammation in the colon through increased SCFA production. Short chain fatty acids can not only provide energy to colonic cells but also inhibit tumor formation (39) by inhibiting DNA synthesis and 26 inducing cell differentiation (40) as well as increasing the numbers of immunomodulatory bacteria such as lactobacilli and bifidobacteria (3). An important immunological role of butyrate is its ability to suppress lymphocyte proliferation through inhibition of cytokine production by T helper 1 (Th1) lymphocytes (such as interferon

(IFN)-ɣ and interleukin (IL)-2) (41). A study by Vulevic et al. (5) looked at the effect of supplementing 5.5 g of GOS to the diets of healthy aged adults. This was a double- blind, placebo controlled crossover study where participants consumed both the placebo and GOS for 10 weeks each. Results from this study showed significant effects on the immune response where there was a significant increase in phagocytosis of bacteria by monocytes and neutrophils, in NK cell activity and in the secretion of the anti-inflammatory cytokine IL-10 by peripheral blood mononuclear cells (PBMCs). A clear positive correlation was observed between the number of bifidobacteria and both phagocytosis and NK cell activity. Recently, a study in Smad3-deficient mice, an animal model linked to gastric neoplasia (42) and increased systemic inflammation, reported the effect of supplementation with GOS on immune variables (43). The investigators observed that GOS supplementation increased fecal Bifidobacterium spp. by 1.5 fold, significantly increased the percentage of NK cells in the spleen and in the mesenteric lymph node, stimulated NK expression of certain chemokine receptors (CCR9) and finally stimulated colonic IL-15 production. Additionally GOS, less so FOS and inulin, can mimic eukaryotic cell surface receptors. An in vitro study by Shoaf et al. demonstrated that GOS can inhibit the attachment of enteropathogenic Escherichia coli to Caco-2 cells (44). Virulent bacteria adhere to the glycoconjugate-like structures on these oligosaccharides in lieu of epithelial cells in the gastrointestinal tract.

Carcinogenesis

Fermentation of GOS and FOS can lead to the production of butyrate. This SCFA has been shown to stimulate apoptosis making it a potential chemopreventive factor in carcinogenesis (45). Additionally, selective fermentation of GOS by specific bacteria can lead to the production of propionate (46), which has also been shown to be anti- inflammatory in terms of colon cancer cells (47). Although in vitro studies and studies executed in animals are encouraging, the anti-tumorigenic activity of prebiotics has not been examined thoroughly in humans and there are very few human studies examining specific anti-cancer effects of fermentable carbohydrates and colonic microbiota.

Epidemiological studies have been inconsistent in regards to the correlation between diet and colorectal cancer risk (48). Data collected from the Nurses’ Health Study, where 121,700 female registered nurses aged 30 to 55 years old and residing in the

United States completed a mailed questionnaire on known or suspected risk factors for cancer and coronary heart disease, showed no correlation between dietary fiber consumption and colon cancer (49). Additionally, human intervention studies using either FOS, a combination of oligosaccharides or even a mixture of prebiotics and probiotics (synbiotics) have shown conflicting results in terms of their effects on the colon environment which may or may not impact colorectal neoplasia (50-52).

Common Cold Symptoms and Complementary Health Practices

The common cold still continues to exert a great financial burden on society despite tremendous advances in the scientific field. Additionally, it is a nuisance in regards to unpleasant and bothersome symptoms (nasal stuffiness and discharge, sneezing, sore throat and cough) (53). The economic burden from the common cold arises not only from direct costs such as healthcare resource use (visits to doctors) but

28 also indirect costs reflected by productivity losses related to absences from work.

According to the United States Department of Health and Human Services, it is estimated that in 2008, there were over one billion common cold cases. Each year in the United States, these cold episodes result in about 25 million visits to family doctors, approximately 20 million days of absence from work and 22 million days of absence from school (54). It has been estimated that the common cold affects adults two to four times per year (55) and children can suffer up to five to six colds each year (56, 57).

The common cold can be caused by a multitude of different viruses that have different pathogenic mechanisms and this is one of the reasons there is still no universal treatment for this illness. The rhinovirus (30 to 50% of the total cases) and coronavirus

(10 to 15% of the total cases) are the most common viruses responsible for the common cold (53), however approximately 200 other viruses are also the cause of the common cold and about 23% of colds are caused by unknown and non-infectious agents (58, 59). Unfortunately, although antibiotics are not effective against viruses, they seem to be widely used for treatment of upper respiratory tract viral infections (60).

The Agency for Healthcare Research and Quality recently reported that overuse of antibiotics is still a concern; though prescribing patterns are slowly improving (61). In

2007, the overall annual average of antibiotics prescribed at visits where there was a diagnosis of the common cold stood at 1,728 per 100,000 persons which is still above the Healthy People 2020 objective of 864 courses of antibiotics per 100,000 population

(62). For these reasons and in order to manage this prevalent and expensive illness, effective antiviral therapies or dietary means to prevent or reduce symptom severity and duration should be examined.

Cold Assessment in Research

Although colds and flu are considered one of the most common infections in humans (63), currently, there are no infallible tools for assessing colds (64). In the literature, certain laboratory parameters such as virus identification, counts of neutrophils or other leukocytes, as well as quantitative assays of different cytokines including IL-1, IL-6 and IL-8 have been used as biomarkers of this illness (65). The problem with these laboratory measures is that they do not correlate very well with specific symptoms and do not predict important health outcomes. As for questionnaires that could help in assessing common cold severity in adults, only one is commonly recognized and accepted; the Jackson scale (66). However, the Jackson scale, which was developed in the 1950s, does not include quality of life measures rather, it includes eight symptoms (sneezing, nasal obstruction, nasal discharge, sore throat, cough, headache, chilliness and malaise) which are rated as either absent, mild, moderate or severe and the rating of these symptoms is either self-assessed or assessed by a clinician or research assistant (67). Another debate emerges; whether clinical assessment of a cold should be a subjective (self-assessed) or objective (outsider’s observation) measure of health. In the past, studies comparing self-reports to physicians’ reports have claimed that self-reports are extremely unreliable (68).

Explicably, it was believed that lay people assessed cold symptoms much differently than clinicians because the disability caused by this infection was primarily due to the symptoms and the subjective perception of the illness (69). An additional questionnaire called the Wisconsin Upper Respiratory Syndrome Survey (WURSS) has been developed to assess health values related to cold illnesses and measure change over time (improvements or deteriorations compared to the previous day) in areas which are 30 important to the participants experiencing the cold. This questionnaire is a 44-item index which consists of impairments to daily living, breathing, sleeping, working, and interpersonal relationships. It associates more strongly with health-related quality of life lacking in the Jackson scale. Research conducted by Barrett and colleagues used data from an earlier study where nearly four hundred participants were inoculated with a rhinovirus. Biomarkers such as IL-8, nasal neutrophil counts, mucus weight, viral titer, and seroconversion were measured. In this reported study, both the Jackson cold scale and WURSS were administered (67). Although the study was initially conducted to test the effects of echinacea (70), the results were reported as being negative. When

Barrett et al. strictly looked at the correlation with biomarkers, although a correlation with laboratory-assessed measures was observed, neither the Jackson scale or the

WURSS was a good predictor of induced infection. This is not unexpected since, generally, only one-third of participants exposed to viruses will develop the cold and only one-third of those infected individuals will then develop symptoms (69, 71).

Currently, improved questionnaires based on self-assessment are commonly used to clinically measure common cold episodes and this is supported by the presence of a good concordance between the participant’s self-assessment of symptoms through these questionnaires and evaluation by a physician (72). The use of good questionnaires combined with immunological markers in intervention studies should be the gold-standard in order to better understand changes in immunological markers and their clinical significance (73).

Nutritional Status, Immune Function and Common Cold

It is well recognized that nutritional status highly influences the immune system and immune dysfunction can result from single nutrient deficiencies (74, 75). 31

Deficiencies in trace elements such as copper and iron may negatively impact neutrophil function. In both humans and rats, copper deficiency has been shown to impair neutrophil circulating numbers and function (76, 77). Given that neutrophils are usually the first immune cells to arrive at sites of tissue damage or infection; this observed loss of function negatively impacts the innate immune system. A deficiency in iron has also been shown to decrease the ability of neutrophils to kill bacteria (78).

Vitamin A deficiency in rats (79) and vitamin E (80) deficiency in humans have both shown to negatively impact NK cell numbers and/or activity. Nutritional inadequacies such as the ones mentioned above as well as protein-energy malnutrition induce a weak immune response to vaccination (81). Undernutrition as well as protein deficiency have both been negatively associated with specific and non-specific immunity (82).

Nutritional assessment in aged adults can be determined using the Mini Nutritional

Assessment (MNA) questionnaire. This validated assessment tool has been shown to be highly specific (98%) and sensitive (96%) and shows high correlation with other measures of nutritional status such as vitamin status (83). A study by Hudgens and colleagues conducted with nursing home elders suffering from pressure ulcers showed that an MNA score indicative of malnutrition was associated with an impaired immune function (84). Participants that were identified as being malnourished (with an MNA score of less than 17 out of a possible score of 30) exhibited reduced neutrophil respiratory burst and decreased whole blood lymphocyte proliferation in response to the mitogens concanavalin A and pokeweed. This study demonstrated that a low MNA score was associated with an impaired immune response validating the use of the MNA questionnaire to link nutritional status and immune function.

Since malnutrition has such an effect on the immune system, it is also a major factor in the occurrence of infections as the two interact synergistically in worsening the health outcome (75). Atrophy and loss of function of lymphoid organs are associated with malnutrition and increase susceptibility to pathogens, reactivation of viral infections and development of opportunistic infections (85). Consequently, improvement of nutritional status could potentially decrease the incidence of infections such as the common cold.

Nutritional Modulation of Common Cold

It has been known for quite some time that nutrients found in foods, such as vitamins, can influence immunity (86). In fact, the Japanese were the first to observe that food could exert a role beyond gastronomic pleasure and energy supply (87) and that nutrients in foods could enhance certain physiological functions leading to the prevention or management of diseases (88). As previously indicated, very few effective pharmacological therapies exist for the treatment of the common cold. The majority of drugs prescribed against this illness are effective at reducing the pain symptoms (non- steroidal anti-inflammatory drugs) but have no apparent effect on overall duration and severity of the cold. A systematic review by D’Agostino and colleagues (89) looked at the effectiveness of antihistamines in reducing the severity of nasal symptoms (running nose and sneezing) and concluded that chlorphenamine and doxylamine conferred a small reduction in symptoms compared to the placebo. A meta-analysis of controlled clinical trials on the effect of nasal decongestants on the common cold found that there was a small significant decrease (4%) in subjective nasal symptoms after a single dose of decongestant compared to the placebo and that this slight decrease in symptoms is also observed over a longer period of three to five days (90). It is evident from these 33 studies that there exists a need for safe and efficacious alternate therapies to treat the common cold and that the demand is undoubtedly present. In 2007, the National

Center for Complementary and Alternative Medicine published a report where, according to the National Health Interview Survey (NHIS), US adults spent approximately 14.8 billion dollars on non-vitamin and non-mineral natural products and nearly 2.9 billion dollars on homeopathic medicine (91). Additionally, a telephone survey of medication use in the United States was conducted from February 1998 through December 1999. This survey named the Slone Survey reported that prevention of colds and influenza as well as immune enhancement were among the top 10 reasons participants took vitamins and herbal supplements (92). This suggests that Americans do remediate to alternative and complementary medicine for enhancement of immune health.

Echinacea

Although echinacea, also known as “purple coneflower”, has been used medicinally by Indigenous peoples for quite some time, its popularity grew considerably when it was first used in Europe against infections. The most common species of echinacea recognized for their medicinal value are Echinacea angustifolia, Echinacea pallida and Echinacea purpurea (93) and frequently used plant parts in different products are the flower, the stem and the root. Additionally, all species of echinacea contain water-soluble polysaccharides, a lipophilic fraction (alkylamides and polyacetylenes) and caffeoyl conjugates (chicoric acid and caffeic acid) (58). In vitro studies with echinacea have demonstrated its ability to modulate certain immune cells.

A study by Goel et al. demonstrated that a formulation prepared from Echinacea purpurea plants decreased total daily cold symptom scores in participants when 34 compared to a placebo and increased the total numbers of circulating leukocytes, monocytes, neutrophils and NK cells in fasted blood samples (94). Potential mechanisms by which echinacea exerts its immunomodulatory effects have been investigated by Woelkart and colleagues (95). These investigators looked at the effect of Echinacea angustifolia roots on the endocannabinoid system of a rodent model and more specifically their ability to bind to the cannabinoid receptors CB1 and CB2. They found that alkylamides, a major lipophilic constituent of Echinacea angustifolia roots, showed selective affinity to CB2 receptors suggesting that alkylamides may modulate immune suppression, induction of apoptosis and cell migration. These results support findings from an earlier study by Gertsch et al. reporting that alkylamides from

Echinacea purpurea tincture induced de novo synthesis of tumor necrosis factor α

(TNF- α) messenger RNA (mRNA) and that this was mediated by the CB2 receptors

(96). Although in vitro studies seem to suggest that echinacea can modulate immune responses, whether it can efficiently prevent or treat the common cold is still controversial. Recently, a meta-analysis was conducted to evaluate the ability of echinacea in modulating duration and incidence of the common cold. This meta- analysis included 14 trials comprising 16 analyses and suggested that echinacea brings an additional benefit in the prevention as well as the treatment of a cold (58).

Nonetheless, it must be noted that clinical trial results have been difficult to interpret partly because there is a lack of consistency in the use of different species of echinacea and plant parts.

Ascorbic acid

Ascorbic acid, commonly referred to as vitamin C, is a powerful antioxidant required for the inhibition of membrane lipid peroxidation and it also plays a major role 35 in the prevention of cardiovascular disease, cancer and cataracts (87). Several experimental studies suggest that vitamin C might effectively be capable of “boosting” the immune system (97). In fact, vitamin C has been shown to be involved in delayed- type-hypersensitivity skin tests, antibody production as well as lymphocyte proliferation most particularly in aged adults (98). As for vitamin C use in the treatment of common cold, data seem to indicate that taking a daily vitamin C dose of 200 mg or more at the onset of cold symptoms does not significantly decrease symptom severity or duration

(99). Indeed, a Cochrane systematic review (100) found seven clinical trials using vitamin C as a treatment and evaluated its effects cold duration. Only one of the seven trials found that participants consuming 8 g of vitamin C at the onset of symptoms had a positive effect on cold duration (less than a day shorter in cold duration) compared to those receiving 4 g. Conflictingly, the remaining six studies showed no effect of vitamin

C as a treatment of the common cold. The same review looked at the effect of ascorbic acid when taken prophylactically and found that when it is taken as a preventative measure, a very slight decrease in the number of colds is observed however, the same decrease is not detected in cold severity. Additionally, symptom duration decreased by approximately 8% in trials that used a greater than 1 g daily dose of vitamin C. When considering that the common cold has a mean duration of symptoms of seven to ten days and that in certain cases symptoms can still be present after three weeks (53), an

8% decrease in symptom duration can represent up to one to two fewer cold symptoms days.

Zinc

The effects of zinc on the immune system have been well characterized. In fact, zinc deficiency has led us to understand more about the important role this essential 36 mineral plays on the immune system. In humans, during mild zinc deficiency, there is a decrease in NK cell activity, lymphocyte proliferation, IL-2, IFN-ɣ and TNF-α production as well as a lowered CD4+ to CD8+ T lymphocyte ratio (101). In regards to the common cold, zinc has been shown to inhibit viral replication by interfering with rhinovirus protein cleavage or capsid binding to the intracellular cell adhesion molecule-1 (ICAM-1) on nasal epithelium (102). Following the discovery of the mechanisms responsible for rhinovirus attachment to cellular components, researchers have attempted to use pharmacological means to block the attachment of the virus to the ICAM-1 receptor by using soluble decoys as well as different viral capsid binders (plecoranil; antirhinoviral drug). However, both of these approaches have resulted in only modest reductions in severity and duration of illness (103). There has been some evidence proposing that zinc may help in reducing the severity of cold symptoms through its effect as an astringent on the trigeminal nerve (104). A drawback regarding the use of zinc in the context of prevention or treatment of the common cold is the different preparations of zinc which are currently being used. In clinical trials, caution must be exercised when choosing the formulation in which to administer zinc as some modes of administration such as citric or tartaric acids, sorbitol, or mannitol are responsible for binding to and subsequently inactivating elemental zinc (105, 106). The formulation must be able to release the ionic form of zinc in order to be efficacious. A study by Mossad and colleagues comparing zinc lozenges (containing 13.3 mg of zinc gluconate; which has been shown not to bind to zinc as tightly as citrate) to a placebo and its effect on the symptom duration reported that zinc had positive effects in significantly reducing the duration of common cold symptoms (107). However, a different study conducted by

Smith et al. also looking at the effect of 11.5 mg of zinc gluconate lozenges showed that the duration of illness was not affected by zinc but that on days four to seven of treatment the participants taking zinc had significantly lower symptom severity scores than the placebo group (108). An additional mode of delivery of zinc sulfate is through nasal sprays. Results from a study using a nasal spray, delivering a maximum daily dose of 0.044 mg of elemental zinc, showed that participants receiving zinc had a significantly lower nasal symptom score on the first day of a cold but this effect was not observed on the cold days that followed (109). The zinc nasal spray had no effect on the duration of cold symptoms. A more recent study conducted by the Mossad group looked at the ability of a zinc nasal gel, delivering a maximum daily dose of 2.1 mg of elemental zinc, in shortening the duration and reducing the severity of the common cold in healthy adults (110). Throughout the entire duration of the study, participants were to apply the zinc nasal gel only when they thought they had a cold. The results from this study indicated that zinc was effective in significantly reducing the duration and significantly decreasing the absolute scores of the cold symptoms from the second day to the seventh day of the cold. These results suggest that direct delivery of zinc to the nasal mucosa, which is the area of viral replication, may provide an added benefit compared to the use of lozenges. A very recent meta-analysis evaluating eight studies and the efficacy and safety of zinc in the treatment of the common cold concluded that oral zinc formulations could shorten the duration of cold symptoms in adults (111). The authors also found that this reduction in duration of symptoms was not statistically significant among children.

Multivitamins and minerals

A randomized, double-blind, placebo controlled trial conducted in the Netherlands investigated the effect of physiological doses of multivitamin and minerals, 200 mg of vitamin E, both, or placebo on incidence and severity of self-reported acute respiratory tract infection in non-institutionalized healthy adults aged 60 years or older (112). The results from this study indicated that the severity of symptoms was not influenced by multivitamin and mineral supplementation and that when vitamin E was used compared to no vitamin E, the severity was actually worsened. No effect was observed for multivitamin and mineral or vitamin E supplementation on the incidence of illness.

Garlic

Other than for its traditional culinary purposes, historically, garlic (Allium sativum) has been used for its medicinal effects (113). Garlic has been shown to provide cardiovascular benefits (114) as well as offer a protective effect against stomach and colorectal cancer (115, 116). Additionally, it has been purported that garlic exhibits antimicrobial, antifungal and antiviral features although the main mechanism by which garlic exerts these effects is still unknown. Several compounds present within the garlic preparation itself may impact its effectiveness. Allicin, an organosulfur compound found in garlic, has been demonstrated to have antibacterial (117) and antiviral (118) properties in vitro. However, it is important to keep in mind that although a quarter of a fresh clove of garlic, weighing approximately 1 g (119), contains 4.38 to 4.65 mg of allicin (120), the latter is inactivated by cooking and therefore the process used to formulate the product containing the garlic compounds must be considered. This may account for differences in the release of allicin from those different products (121). A

2012 Cochrane review by Lissiman et al. (122) looked at randomized controlled trials

39 which compared garlic to a placebo, no treatment or standard treatment of the common cold. As a primary outcome, this review focused on both prevention trials where the effects of garlic on occurrences of cold were measured and treatment trials looking at duration of symptoms. The investigators also took into consideration the number of days until recovery. Interestingly, of the six trials that were identified as potentially applicable, five were rejected, being left with only a single study which met their stringent selection criteria. In this study, Josling and colleagues randomized 146 adult participants to either receive a capsule containing 180 mg of allicin powder or a placebo for 12 weeks between the months of November and February (123). The participants were instructed to record in a daily diary the number of occurrences of the common cold. Participants had to self-report the cold duration, the number of days when they felt challenged or that a cold was coming on, and the number of days needed to recover from the cold. The authors reported a significant reduction in the number of colds in participants receiving the garlic suggesting that garlic may have a preventative effect on the common cold. However, the number of days to recover from a cold did not differ between the groups. Few adverse effects were encountered with only smell and skin rash being reported. Since garlic consumption has been associated with unpleasant gastric side effects (belching and strong breath odor) and body odor following consumption, further studies are interested in aged garlic extract (AGE) supplementation. In fact, unlike traditional garlic, AGE is nearly odorless and is produced from minced garlic which is incubated in aqueous alcohol (15 to 20%) for 20 months and then concentrated (124). AGE lacks the lipid-soluble compound allicin which was the main component previously mentioned. A recent study conducted by

Nantz et al. looked at supplementation with AGE and its effect on immune function and health outcomes (125). More specifically, this randomized, double-blind, placebo- controlled study conducted in healthy adults looked at the impact of AGE on immune cell function and proliferation and cold and flu symptoms. The results from this study demonstrated that the group receiving the garlic supplement showed significantly higher proliferation index of a T lymphocyte population (γδ-T cells) and NK cells as well as a reduction in the number of symptoms and consequently less missed school/work days.

These results suggest that perhaps different components alone or in combination within this specialized growth bud may be responsible for its numerous and diverse beneficial effects.

Probiotics

In recent years, the growing numbers of randomized clinical trials executed using prebiotics and probiotics demonstrate the considerable interest in this field. Several of these pre- or probiotics have shown significant physiological benefits against disease state (126). In brief, probiotics have first been described almost 60 years ago by Kollath

(127) and are now defined as “live microorganisms which, when administered in adequate amounts, confer a beneficial physiological effect on the host” (128). A recent

Cochrane review published in 2011 investigated the effects of probiotics on the prevention of acute upper respiratory tract infections (URTI) (129). The authors included single or mixture of strains of probiotics as well as several dosages and routes of administration. As a primary outcome, the investigators were interested in the number as well as the duration of acute episodes of URTIs. Data from 10 trials were included in the review and used in the analysis where the mean age of the participants was of about 40 years. As a summary, it was concluded that in regards to the number 41 of participants that experienced episodes of URTIs, antibiotics used and the rate ratio of episodes of acute URTIs, probiotics may provide more benefit than a placebo. In terms of duration of the episodes of URTI, no advantage was observed with the probiotics.

Prebiotics

As for prebiotics and the common cold, a study conducted in infants by Tschernia et al. (130) examined the effect of oligofructose supplementation on general health.

This was a double-blind, randomized, controlled study where infants attending day-care and aged between four and 24 months were either given 1.1 g per day of oligofructose in a standard weaning cereal or just the cereal (i.e. no oligofructose) for six months.

The consumption of the prebiotic cereal was associated with a decrease in occurrence of fever during diarrhea, decrease in number of days absent from the day-care because of diarrhea, as well as a significant decrease in occurrence of fever and antibiotic use during respiratory illness. Another study also conducted in infants, aged 15 to 120 days, who were randomized to either a standard formula containing GOS/FOS in a 9 to

1 ratio or to a control standard formula (131). This one year open trial looked at the outcome of the prebiotic mixture in reducing the incidence of intestinal and respiratory infections. The results showed that in the GOS/FOS group, the number of children with more than three episodes of URTI was significantly lower and that significantly fewer infants received more than two antibiotic courses per year.

A study conducted in both assisted- and independent-living adults aged 65 years old or older looked at the effects of an experimental nutritional formula containing FOS on days of symptoms of URTI (132). The results from this randomized, double-blind, controlled study demonstrated that the participants receiving the FOS containing formula for a total of 183 days had a significant decrease in the median days of URTIs. 42

However, the FOS formula given also contained antioxidants (vitamins C, E and β carotene), B vitamins, zinc, selenium, other vitamins and minerals, as well as structured triacylglycerol which could have acted synergistically and potentiated the beneficial effects observed. More recently, a study conducted by Hughes and colleagues looked at the effects of the prebiotic GOS on the percentage of days with a cold or flu in 427 academically stressed undergraduate students (4). The main findings from this randomized, double-blind, placebo-controlled study were that acute psychological stress, from academic exams, was directly related to gastrointestinal dysfunction and days of cold or flu and that supplementation with either 2.5 or 5 g of GOS reduced some symptoms of intestinal discomfort as well as the number of cold days. As for the effect on the average cold symptom intensity score, GOS was more effective at significantly reducing the symptom intensity scores at lower levels of stress.

Several studies, such as a study by Puccio et al., have examined the effects of a synbiotic on cold episodes (131). These investigators randomized full-term newborns who were not breastfed after the 14th day of life to either receive a control formula or a formula containing a mixture of GOS/FOS at a 9 to 1 ratio plus Bifidobacterium longum

BL999. The results from this study, although not significantly different, showed a trend in infants receiving the synbiotic formula to have fewer URTIs. There seems to be an increasing number of studies demonstrating positive effects of prebiotics, probiotics and synbiotics on days of URTI and this warrants further investigation.

Age and Common Cold

Since the proportion of aged individuals (60 years of age or older) has tripled in the second half of the 20th century (133), it is not surprising that there is increased

43 popular and scientific interest in the aging process and its consequences. Although the common cold affects persons of all ages (55), exposure to an infectious agent is a major cause of morbidity and mortality in susceptible populations such as the very young and the very old as well as in individuals undergoing psychological stress (4, 69). As stress increases, it has been demonstrated that susceptibility to the common cold also increases in a continuous manner similar to a dose-response effect (71). In fact, both

T- and B-lymphocytes, mast cells (134) and neutrophils (135) are increased in the nasal mucosa of patients with colds. Consequently, an appropriate inflammatory response is necessary for the clearance of the infection.

When compared to the general population, Americans aged 65 years old or older are twice as likely to visit the doctor and three times more likely to be hospitalized (136).

Other than the well described effects of age on the immune system which predisposes the aging population to common cold episodes, living conditions, such as assisted or independent living, may also render them more vulnerable to infections. In the United

States, since 37% of aged adults will be admitted to a long-term nursing home care facility and since infections occur more frequently in nursing home residents than independent-living elders (137), living conditions may increase the risk and occurrence of the common cold. Additionally, aging not only increases the risk illnesses but in aged adults that contract a rhinovirus infection, practically two thirds of them will subsequently develop lower respiratory complications such as otitis media, sinus infection, exacerbation of asthma and chronic bronchitis and pneumonia (138). Healthy yet vulnerable populations should be examined when conducting nutritional studies and

44 examining health outcomes since this population would greatly benefit from such interventions.

Immune System

Leukocytes

Leukocytes, also known as white blood cells, are a group of immune cells circulating in the blood and lymph, populating the lymphoid organs. Leukocytes are composed of neutrophils (50 to 70% of total leukocytes), lymphocytes (20 to 25%)

(139), monocytes (1 to 6%), eosinophils (1 to 3%) and basophils (˂1%) (140).

Lymphocytes are responsible for adaptive immunity, memory and self/non-self recognition which makes them central players of the immune response. Of the peripheral blood lymphocytes, approximately 85 to 90% are either T or B cells, however the remaining 10 to 15% are referred to as NK cells and are part of the innate immune system. T and B lymphocytes are essential for adaptive immunity because of their highly diverse repertoire of antigen-specific cell receptors which can easily recognize some 1011 different antigens (141). To distinguish these diverse populations of lymphocytes, specific proteins can be detected and these cellular markers are molecules known as cluster of differentiation (CD). In fact, all lymphocytes exhibit the signal transduction molecule CD45. To further distinguish lymphocyte subpopulations,

T lymphocytes express CD3 and all B lymphocytes selectively express CD19. A specific subset of T lymphocytes, T-helper cells (Th cells), express an adhesion molecule that binds to class II major histocompatibility complex (MHC) molecules called

CD4. Cytotoxic T Lymphocytes (CTL) express CD8 an adhesion molecule that binds to class I MHC molecules. Although CTL secrete very few cytokines, these cells have the capacity to recognize and subsequently eliminate altered-self cells. On the other hand,

NK cells also express very specific adhesion molecule; CD56 and the low affinity receptor for immunoglobulin (Ig) G, CD16. These different CD markers are important as they can be recognized by particular monoclonal antibodies and differentiate different lineage or stages of maturation of lymphocytes (e.g., T lymphocytes that harbor both

CD4 and CD8 are still immature and yet to terminally differentiate). The analysis of the different lymphocyte subsets by flow cytometry can provide useful information exemplified by the fact that triggers such as fasting or refeeding can elicit a redistribution of the leukocyte pool (142).

Serum Acute Phase Proteins

Acute phase responses are defined as changes accompanied by inflammation and involving many organ systems at distant sites from the source of inflammation (143).

This concept emerged from the discovery of C-reactive protein (CRP), specifically named because of its reaction with the pneumococcal C-polysaccharide in the plasma of patients suffering from pneumococcal pneumonia (144). Since circulating concentrations of CRP are changing rapidly, it is a useful tool in differentiating inflammatory from non-inflammatory conditions (e.g., response of the host to infection) however; it is of low suitability as a marker of immune function. Other acute phase proteins such as α-glycoprotein, albumin, pre-albumin, fibrinogen, mannose-binding protein and complement components are found in the serum or plasma and are also indicators of ongoing clinical infections or other diseases (73). Although these proteins may be useful to understand results of functional assays, they cannot be extrapolated to immune function or resistance to infection and more immune parameter must be measured for such inferences.

Peripheral Blood Mononuclear Cell (PBMC) Cytokine Production to Mitogens

Clinical trials have investigated the effectiveness of immune components to act as biomarkers with the ultimate goal to be able to determine the participant’s health status

(145), however, a determinant of one’s health through a single biomarker has yet to have been discovered and it is needed. Since cytokines and their production have been recognized to be important in many disease states, cytokine production has been utilized as an assessment of immune function. Specifically, measurement of cytokine production following mitogenic stimulation of short-term and in-vitro cultures of PBMCs is widely used in human clinical trials (146). Following isolation and separation of

PBMCs from human blood samples, by means of a density gradient centrifugation, T and B lymphocytes, monocytes, NK cells, basophils, and dendritic cells are then stimulated in culture. This assay partly reveals their functional capabilities as the function of Th cells is typically assessed by the level and production of cytokines. While these assays are extremely sensitive in measuring cytokine concentrations, information on the functional activities of the cytokines measured is lacking (146).

The choice of mitogen employed to stimulate cells in culture is also as important.

De Groote et al. (147) showed that lipopolysaccharide (LPS) stimulated PBMCs produce IL-1β, TNF-α and IL-6. LPS has previously been shown to stimulate B cells, monocytes and to a lesser extent dendritic cells (DC). Dendritic cells of aged adults have been shown to have a significantly increased response to LPS and double stranded RNA-induced secretion of TNF-α and IL-6 (148). LPS selectively induces many genes encoding for a set of cytokines and chemokines (149). As quickly as 30 minutes following exposure to LPS, cells have been shown to synthesize and secrete pro-inflammatory cytokines such as IL-1β, IL-6 and TNF-α (150). On the other hand, 47

PBMCs stimulated with the mitogen phytohemagglutinin (PHA) produce IL-2, IFN-γ as well as GM-CSF (147). The response from T lymphocytes to PHA has been demonstrated and Th1 cells exclusively produces IFN-γ, IL-12 and IL-2, whereas Th2 cells secrete IL-4, IL-5, IL-6 and IL-10 (151).

When planning experiments, determining the duration of the in-vitro stimulation of cells by the different mitogens is equally important. A study executed by McHugh and colleagues looked at the temporal production profile of different cytokines following the activation of PBMCs by various mitogens (151). The investigators stimulated PBMCs by adding 10 μg/mL of PHA to the culture media and observed the proportion between the highest (100%) and lowest (0%) amounts recorded for multiple cytokines. Following

48 hours of mitogenic stimulation, the authors observed a maximum production of approximately 25% for IL-2, 100% for IL-5, 50% for IL-6, 95% for IL-10 and 40% for IFN-

γ. Generally, cytokine responses were fast and vigorous and seemed to peak between

24 and 96 hours after stimulation. Similarly, a study conducted by Rittig et al. investigated the cytokine release from human monocytes following 60 minutes of stimulation with brucellae displaying smooth or rough LPS at a multiplicity ratio of infection of 20 to 1, or with LPS from E. coli serotype 055:B5 as a positive control (152).

The supernatant was harvested at different intervals (0, 4, 8, 24 and 48 hours) and it was observed that the concentrations of IL-6, IL-8 and TNF-α were the highest after 48 hours and were also the highest when stimulated by the strains Brucella melitensis

B3B2 (rough LPS phenotype) and Brucella suis VIG4 (rough LPS phenotype).

When observing changes in immune parameters, it is imperative to consider that immune function can easily be influenced by many factors such as age, gender, obesity,

48 health status, disease, and use of medication (139). Hence, when evaluating age- related changes in immunity, selection of participants and their inherent characteristics is of upmost importance. The SENIEUR protocol (153) is employed by many when developing studies where a link between age and immune function must be made without attributing any effects to underlying diseases. This protocol, developed by the

European Economic Community Concerted Action on Aging (EURAGE), uses rigorous health criteria for selection of participants to exclude any possible pathological changes in the immune system and to only describe changes that are caused by age (154). A weakness emerging from this study design is that the obtained results may really only apply to exceptional individuals that are unusually healthy and have survived age- related illnesses. Other researchers have decided to conduct clinical trials with generally healthy aged adults following the assumption that results from these experimental protocols will be more applicable to the general aged population. Results from age-related changes in cytokine production by PBMCs in response to PHA stimulation are summarized below (Table 1-1). Outcomes from these studies seem to indicate that following strong mitogens stimulation with the use of PHA, IL-2 production seems to be lower in the aged adults in comparison to the younger population.

Demonstrated in fewer studies and to a lesser extent, IFN-γ production has also been shown to be lower in aged adults. Although IL-6 has previously been considered to be the “gerontology cytokine” (155) in that greater concentration was associated with increasing age, recent reports have failed to validate such results and no difference seems to be observed in comparison to younger adults.

Immunosenescence

It has been known for quite some time that aged adults are at increased risk of infectious diseases and this is thought to be an age-related loss of immune function leading to a reduced numbers of lymphoid cell in peripheral circulation and a decline in function (156). Dysregulated immune functions associated with aging, known as immunosenescence, has been characterized by a decrease in protective immune responses to infectious agents and anti-cancerous immune defenses (157).

Additionally, modifications in different aspects of cellular immunity are observed with immunosenescence. The term “inflamma-aging” was generated by Claudio Franceschi

(158) due to the now recognized low grade systemic inflammation observed with immunosenescence. This pro-inflammatory status is thought to be mediated through the NF-κB factor illustrated in Figure 1-2 (159).

A hallmark of immunosenescence is the decreased numbers of mature CD3+ lymphocytes in peripheral circulation exemplified primarily with a decline in the CD8+

(cytotoxic T cells) subset whereas the CD4+ subset seems to be maintained (160) as long as nutritional status remains normal (73). However, some evidence suggests that these changes are not necessarily present in all aged adults. The decrease in cytotoxic

T cells is not significant in healthy elderly participants when CD2+CD3- immature T cell subsets represent only 10 to 15% of the total peripheral T lymphocyte pool (161). An additional feature of immunosenescence is a decreased pool of naïve (CD45RA+) T cells and a predisposition toward a T helper 2 (Th2) phenotypic expression (154). This decrease in T cells, reported with age, could be attributable to a decreased size and cellularity of the thymus and a subsequent reduction in the T cell compartment (162).

By the age of 60, the tissue that constitutes the thymus is almost completely replaced 50 by adipose tissue rendering T cell selection and maturation practically impossible (163).

A decline in thyroid hormones is also observed progressively following puberty which could contribute to this physiological change. Not only are there observable changes in cell types but also rodent studies have shown that there is an age related shift from a cell mediated immune response a Th1-like response (e.g., IL-2 and IFN-γ) to an antibody-mediated Th2-like (e.g., IL-6 and IL-10) cytokine response. This change is thought to be due to the lifespan accumulation of antigenic pressure (164). A shift to a

Th2 antibody-mediated response may explain why healthy elderly participants seem to have similar antibody responses than younger adults following antigenic exposure and why serum titers of immunoglobulins are increased (73). Nevertheless, the impaired response of the Th1 cell subset is surely exemplified by the lower IFN- γ production in elder participants explaining the reduced cytotoxic activity of lymphocytes and increased incidence of infections in the older population (165).

Ageing has been associated with an impaired proliferation of T-cells caused either by a dysfunction of T cells themselves or a defective activation of T cells by monocytes

(166). Previous studies have demonstrated that during an illness, monocytes of aged individuals release more IL-1, IL-6 or TNF and this activation of monocytes is thought to help stimulate the depressed lymphocytes to reach a sufficient level of efficiency (167).

Gillis and colleagues demonstrated more than three decades ago that with increasing age, there are alterations in the production of cytokines and their recognition by cells of the immune system. In more detail, the investigators showed a decrease in IL-2 production in aged adults compared to young participants (168).

Another important characteristic of the immune system in aged adults is the change in the NK cell population. As previously mentioned, NK cells are major contributors to cell-mediated immunity by functioning as immunomodulatory cells through the secretion of IFN- γ (169). Older adults seem to exhibit a deficiency in NK cell activity despite their increase in numbers (170, 171).

Mucosal Immunity

External surfaces of the body such as the skin and the gastrointestinal, respiratory and genitourinary tracts are important first lines of defense by preventing the invasion of pathogens and their products (172). Despite the fact that the immune system of the mucosal surfaces is thought to be the largest immune component of the body, no viable method exists to sample the immune system in the intestine; only indirect and invasive techniques such as endoscopies and biopsies are available (73). Unlike the systemic immune system, where blood sampling can be done with relative ease and is considered to be non-invasive, techniques allowing for the examination of the immune system in the GI track are still lacking. A test useful in determining the integrity of the intestinal barrier, the first line of defense, is the use of markers (such as non- metabolisable sugars; lactulose or mannitol) given orally and then later measured in the urine (173). Despite risk, a double-blind, placebo-controlled intervention study by

Bovee-Oudenhoven et al. was conducted where 32 healthy men ingested a live but attenuated enterogenic Escherichia coli strain that was able to induce mild but short- lived symptoms (174). While the study was primarily looking at the protective effects of a milk product containing a high concentration of calcium (1100 mg/day) versus low calcium concentration (60 mg/day) on the E. coli infection, such an approach could potentially be used to measure the mucosal immune system, more specifically the 52 intestinal barrier. Considering the fact that mucosal surfaces are the main interface between the host and its external environment, protection against potentially harmful bacteria is imperative. For this reason, within the entire immune system, the gastrointestinal mucosa is the most abundant production site of IgA antibody (175).

This explains why, thus far, the most useful marker of the mucosal immune system is

IgA. However, care must be taken when extrapolating these measures of the adaptive immune system to the rest of the mucosal immune system.

In addition to the GI tract’s ability to act as a first line of protection reducing the need for a systemic immune response, the gut possesses immunosuppressive mechanisms and tolerogenic responses to food, commensal microbiota and self antigens. Oral tolerance is an essential adaptive immune function which is defined as the systemic non-responsiveness to ingested antigens (176). In the event that this immunosuppressive mechanism is disrupted, occurrence of pathologies such as food allergy, celiac disease and inflammatory bowel disease are observed (177). The constant interaction of the mucosal immune system of the gut with the commensal microbiota facilitates the establishment and maintenance of tolerance to foreign dietary antigen. Lack of tolerance to the non-pathogenic microbiota contributes to the development of inflammatory bowel diseases such as Crohn’s disease and ulcerative colitis. It has also been shown that tolerogenic antigen presenting cells (APCs), such as dendritic cells, can avoid local hyperactivation of the immune system by carrying dietary and microbial antigens, that have somehow penetrated the epithelial barrier, away from the mucosa (178). Epithelial cells and DC can secrete tolerance signals such as

53 transforming growth factor (TGF) β, IL-10 and retinoic acid leading to the development of regulatory T (Treg) cells and maintenance of immune tolerance (179).

The intestinal mucosal immune system is composed of three main compartments; the surface epithelium of the mucosa, the underlying connective tissue (lamina propria) and the secondary lymphoid tissue. The lamina propria is thought to constitute the effector site where activated CD4+ T lymphocytes, activated B cells, macrophages, dendritic cells, eosinophils, mast cells, and IgA secreting plasma cells are primarily found. The secondary lymphoid tissue comprises the GALT and the mesenteric lymph node (MLN). As previously mentioned, the GALT includes the Peyer’s Patches, which are found predominantly in the small intestine, and the isolated lymphoid follicles, found throughout the intestine as well as the appendix (180). The GALT is the main site where antigens are presented to T and B lymphocytes and is considered to be the inductive site of the GI tract responsible for the adaptive phase of the intestinal immunity.

Salivary and Fecal Secretory IgA (sIgA)

In the lamina propria of the GI tract, it has been shown that 90% of the plasma cells (mature B cells) secrete IgA (181). This is equivalent to more than 2.5 x 1010 IgA- secreting plasma cells. Most of the IgA in the gut lumen is found in the secretory form and is distinct from IgA present in the serum. IgA is the predominant immunoglobulin secreted at mucosal surfaces (73) and is found in secretions such as breast milk, saliva, tears and mucus. In fact, secretory IgA (sIgA) is composed of polymers of 2 to 4 monomeric units of IgA molecules which are joined by a J chain and attached to a secretory component that protects it from proteolysis. The secretory component of the sIgA antibody is acquired from the receptor after IgA binds to and is transcytosed 54 across the mucosal surface. The daily production of secretory IgA is greater than that of any other immunoglobulin. Every day, a human can secrete anywhere from 5 to 15 g of secretory IgA into mucous secretions (140). IgA is extremely important as a first line of defense. It can immobilize microorganisms at mucosal surfaces and since sIgA is polymeric, it can bind to large antigens with multiple epitopes and prevent their attachment to the epithelial cells of the mucosa. The intestinal microbiota balance has recently been attributed to the presence of sIgA. This has been demonstrated by experiments from van der Waaj et al. where commensal bacteria were found to be coated with anti-commensal sIgA (182). Following coating by sIgA, the intestinal bacteria adhere to M cells (present in the Peyer’s Patches) and are subsequently internalized where naïve B cells are induced and differentiate into IgA-committed plasma cells. After neutralization of the bacteria a decrease in pro-inflammatory cytokines is observed (183). Secretory IgA is believed to be one of the most useful markers of the mucosal immune system (73), whether it be the respiratory, gastrointestinal or vaginal tract. For this reason, studies have explored its effect on

URTI. There seems to be some evidence that low levels of salivary sIgA are associated with increased incidence of URTI (184) and periods of chronic physical and psychological stress (73). Inversely, lactobacilli which are part of the commensal intestinal microbiota, have been reported to enhance IgA responses in 14 children with

Crohn’s disease (185). The relationship between sIgA present in the mucosal surfaces and the occurrence of URTI still need to be studied in more details as this immunoglobulin could potentially be an accessible biomarker of health.

Digestive Health

Microbiota

There has been a rapid increase towards our understanding that commensal microbes, collectively known as the microbiota, have on the mammalian gut. It is estimated that in 2009, more than five hundred articles related to the intestinal microbiota were published (186); double the number of publications released in 2004.

It is estimated that there are about 1014 bacterial cells in the human GI tract and more than 2000 different species of commensal microorganisms (187). The bacteria residing the GI tract are either facultative anaerobes, aerobes or strict anaerobes, the latter dominating by two to three orders of magnitude. Several studies have demonstrated that the estimated ratio of anaerobic to facultatively anaerobic bacteria is approximately 300 to 1 (188). In the large intestine, particularly the colon, the microbiota consists primarily of anaerobic, Gram-negative nonsporing bacteria and

Gram-positive, spore and nonsporing rods. The human gut microbiota is mainly composed of two phyla; the Bacteroidetes, which comprises Flavobacteria and

Sphingobacteria, and the Firmicutes, comprising known butyrate producers of the

Clostridia class (189).

Inter-individual variations in the microbial makeup of the GI tract are quite significant. Although numerous studies were conducted to elucidate the factors responsible for these observed differences, the verdict is still out. Some potential sources of variation are the modes of child birth (vaginal or C-section) (190), the type of infant feeding (breastfed versus formula fed), antibiotic use in the first few years of life

(191), genetics (192), diet/environment (193), and medication usage throughout ones lifespan (194). Significant debate surrounds the question - does the environment or

56 genetic makeup have a greater influence on a person’s microbiota? Multiple studies have been conducted to answer this question, with conflicting results. Zoetendal and colleagues recruited individuals that were either related (monozygotic or dizygotic twins) or not and their microbiota composition was examined using DGGE (denaturing gradient gel electrophoresis) (195). What the authors found was that related individuals

(mono or dizygotic twins) had a greater similarity in their microbiota profile than non- related individuals as well as marital couples despite the fact that these twins had been living apart for years (the range was anywhere from 5 to 15 or more years living apart).

In fact, married couples had a greater diversity in their microbiota than twins, suggesting that genetic rather than environmental factors had more of an impact on microbiota composition. The microbiota has also been thought to be implicated in certain diseases from the more obvious inflammatory bowel disease (IBD) (196) to the unexpected activation of the chronic human immunodeficiency virus (HIV) (197).

Gastrointestinal Symptom Rating Scale

The Gastrointestinal Symptom Rating Scale was developed alongside the

Comprehensive Psychopathological Rating Scale (CPRS) to assess gastrointestinal symptoms in patients suffering from irritable bowel syndrome (IBS) (198). This questionnaire was created in order to evaluate gastrointestinal symptoms from the patient’s perspective, emphasizing on subjective discomfort and the impact this has on the patient’s wellbeing. Subsequent studies conducted using this questionnaire, have shown that the GSRS contains adequate psychometric characteristics and can also be used on patients with GI disorders or suffering from GI-related side effects where GI symptoms are not necessarily the chief complaint (23, 199, 200). The GSRS demonstrates good reliability and validity in different populations whether the GI 57 symptoms are caused by immunosuppressive regimens seen in renal transplant patients (201) or by academic stress as observed in undergraduate university students

(4). Additionally, this quality of life questionnaire has been evaluated in the general population providing researchers with norm values (202). The GSRS is a 15 item questionnaire that is divided into five symptom clusters; reflux (heartburn and acid regurgitation), abdominal pain (abdominal pain, hunger pains, and nausea), indigestion

(rumbling, bloating, burping and gas), diarrhea (diarrhea, loose stools, and urgent need for defecation) and constipation (constipation, hard stools, and feeling of incomplete evacuation). This questionnaire uses a 7-point graded Likert-type scale where 1 represents no discomfort and conversely very severe discomfort is graded 7.

GOS and the Microbiota

Prebiotics have been shown to exert positive effects on the microbiota by selectively promoting the growth of bacteria that are considered beneficial to the human host (e.g., bifidobacteria and lactobacilli), but also by preventing or inhibiting colonization of potentially pathogenic bacteria (e.g., Clostridium) (203-205).

For the past two decades, the effects of prebiotics (e.g., GOS and FOS) on the colonic microbiota have been investigated and it has been found that these oligosaccharides have the potential to increase the concentrations of bifidobacteria in the colon (206). Multiple studies have examined the effect of GOS on the microbiota and, for the most part, a bifidogenic effect of GOS seems to be observed. A Japanese cross-over study, mentioned previously, was conducted in 12 healthy male volunteers who were randomized to receive 0.0, 2.5, 5.0 and 10.0 g of GOS (18). The results from this study showed a linear relationship between the amount of GOS ingested and the number of bifidobacteria per gram of feces. Additionally, this daily amount of prebiotic 58 did not have a statistically significant effect on abdominal pain. A study by Bouhnik and colleagues demonstrated that healthy participants given 10 g of transgalactooligosaccharides for 21 days had a significant increase of bifidobacteria in their feces. Similarly to the Japanese study, the participants reported that they did not experience any gastrointestinal discomfort following the ingestion of 10 g of TOS (21).

A more recent study by the same investigators looked at different non-digestible carbohydrates and their effect on fecal bifidobacteria (206). The findings from this double-blind, randomized, placebo-controlled study suggest that GOS, given at a dose of 10 g per day for 7 days, significantly increased bifidobacteria concentration in the feces and that this bifidogenic effect conferred by GOS was stronger than that of FOS and a debranched retrograded tapioca maltodextrin (type III resistant starch). Vulevic and colleagues examined the effect of a trans-galactooligosaccharide mixture on the fecal microbiota profile, however, the study was executed in free-living aged adults that were 64 to 79 years old (5). In this protocol, participants were randomized to receive a placebo or 5.5 g of B-GOS for ten weeks, followed by a four week washout period and finally the participants were crossed-over to receive the other treatment for an additional ten weeks. The results suggest that the consumption of B-GOS significantly increased fecal bifidobacteria when compared to baseline levels as well as placebo supplementation levels.

Aged Adults and the Microbiota

Aging causes many physical changes in the GI tract and this is exemplified by the reduction in physiological processes (207). Reduction in gastric acid production, gastric emptying rate, gut motility, gut blood flow, absorption surface as well as decreased uptake and transport across epithelial cells have been observed in the GI tract of aged 59 adults. Additionally, a loss in adaptability following challenges such as changes in the diet or nutritional stress is seen in the older population. Other important changes observed with aging are a marked decline in the bifidobacteria population within the intestinal microbiota (208) as well as a change in species diversity (209). A study by

Hopkins and colleagues suggests that the levels of bifidobacteria and lactobacilli appear to be lower in elderly participants when compared to younger adults whereas the levels of enterococci, enterobacteria and clostridia appears to be similar in both groups of participants (210). A later study showed slightly different results where, when compared to younger adults, the elder population exhibited higher levels of Rhuminococcus and lower levels of Eubacterium and Bacteriodetes (211). With advancing age, variation in ones microbiota seems to be observed; however, there does not appear to be a definitive compositional change as factors such as geographic region, medication and genetics, as previously defined, can also influence the microbiota. (194).

With age comes an increased incidence of diseases which in turn will cause a greater medication use negatively impacting the microbiota. According to a report published by the Center for Disease Control and prevention (CDC), in 2008, 76% of older Americans (60 years of age or older) used two or more prescription drugs and

37% used five or more (212). Prescription drugs are known to have varied effects on the digestive tract and consequently the microbiota. It has been established that many drugs cause, as an undesirable side-effect, dry-mouth or xerostomia (213). This dampened ability to produce sufficient saliva and therefore a reduction in salivary flow rate impacts the antimicrobial oral defenses by decreasing lysozymes, amylases, myeloperoxidase, IgA, IgG, and IgM in the oral cavity (214) and may consequently have

60 adverse effects on the gut mucosa. In the elderly population, commonly prescribed drugs are proton pump inhibitors recommended to treat ulcers. Lowering the gastric pH, the main mechanism of action of proton pump inhibitors, may promote colonization of the gastrointestinal tract by pathogenic bacteria who would not have been able to survive otherwise, under normal conditions (215). It is also important to mention that with increased age, there is a reduction in the total SCFA (butyrate, propionate, and acetate) production especially in antibiotic treated older adults (216). With advancing age, changes in the microbiota, modifications in the total numbers of bacteria, and differences in species diversity of certain genera of bacteria all lead to changes in microbial activity and their end products (194). As previously mentioned, SCFA produced by certain bacteria exert many beneficial effects on the large intestine and changes in these fermentation products could be detrimental to the GI tract of aged adults. These organic acids serve as the preferred energy source of colonic epithelial cells, increase blood flow to the colon mucosa (217) and increase colonic motility (218).

A rodent study by Suzuki et al. using isolated caecal mucosa, anaesthetized rats, and cultured Caco-2 cells demonstrated that SCFA, particularly acetate, enhance mucosal barrier function (219). The barrier function of the GI tract is crucial in maintaining a beneficial relationship between the host and the microbiota. This especially hold true in the colon where approximately 1.5 kg of microbes reside (194). The mucus in the colon and the tight junctions between the colonic epithelial cells are the primary physical barriers of the intestines. This barrier function is essential to prevent the bacteria themselves and their toxic compounds from translocating through the epithelial layer and potentially causing septic shock, systemic inflammatory response syndrome and

61 ultimately organ failure (220). A plausible mechanism where, with increasing age, an altered microbiota could lead to intestinal permeability and could explain the chronic inflammatory status seen in aged adults is shown in Figure 1-2. One of the most studied interactions between the innate immune system and the infectious non-self is the interface between the pathogen-associated molecular patterns (PAMPs), which are conserved microbial components, and the host pattern recognition receptors (PRRs), expressed on cells of the innate immune system. Toll-like receptors (TLRs), a class of

PRR, are expressed on dendritic cells, macrophages, neutrophils, mast cells and epithelial cells. Toll-like receptor-4 specifically recognizes the LPS component of Gram- negative bacteria (157). Once LPS binds to TLR-4 on an epithelial cell of the GI tract, the transcription factor NF-κB is activated resulting in its nuclear translocation and activation of pro-inflammatory genes. This continual interaction between LPS and TLR-

4 results in the production of inflammatory cytokines by activated genes and could be responsible for the chronic low-grade inflammation observed in the older population

(176).

Figure 1-1. General structure of a β-(1-4) linked galactooligosaccharide molecule where [Galactose]n Glucose. Figure provided with permission by Christina M. Vitale.

Figure 1-2. Effects of aging on the microbiota and the subsequent inflammatory response (157)

Table 1-1. Age-related changes in production of cytokines following stimulation of isolated peripheral blood mononuclear cells (PBMCs) with mitogens Mitogen Health used and Participant status of Results (aged time of number (age the aged Cytokine Function versus young stimulation range or adults; adult) for reported mean±SEM) Healthy or results SENIEUR Secreted exclusively by T lymphocytes (CD4+ are the main Lower PHA aged adults; producers) production in Cells 16 (79.6±7.5) IL-2 SENIEUR (188) the aged adults stimulated young adults; Function: (221) for 48 hours 16 (24.6±3.1) Activation/ Proliferation of T lymphocytes (146) Lower PHA aged adults; production in Cells 16 (79.6±7.5) SENIEUR Secreted the aged adults stimulated young adults; mainly by T (221) for 48 hours 16 (24.6±3.1) IFN-γ lymphocytes Lower PHA aged adults; and NK cells production in Cells 15 (66-84) Healthy the aged adults stimulated young adults; (222) for 48 hours 12 (21-40) PHA and Fetal Bovine No difference in Serum Secreted by T production (FBS) or aged adults; lymphocytes, between the PHA and 26 (65-85) SENIEUR macrophages, aged and Autologous young adults; monocytes. young adults plasma (AP) 21 (18-35) (223) Cells IL-6 Induces T stimulated cells for 48 hours activation and No difference in differentiation production PHA aged adults between the Cells 5 (66-87) Healthy aged and stimulated young adults young adults for 48 hours 9 (20-54) (224)

Table 1-1. Continued Mitogen Health used and Participant status of Results (aged time of number (age the aged Cytokine Function versus young stimulation range or adults; adult) for reported mean±SEM) Healthy or results SENIEUR No difference at 24 hours but higher Induces B cell aged adults production in differentiation 13 (80.8±2.1) IL-6 the aged adults PHA SENIEUR and mucosal young adults at 48 and 72 IgA responses 13 (26.8±0.8) hours compared to the young (225) Produced by T and B No difference in lymphocytes, No production aged adults monocytes stimulation between the 12 (75-102) IL-10 and Results SENIEUR aged and young adults macrophages. taken after young adults 12 (19-34) Inhibits 24 hours (226) macrophages activity (188)

CHAPTER 3 METHODS

Participants

Participants were healthy older adults aged 60 years or older living independently and who had suffered at least one cold in the previous year. Participants were recruited via listservs, flyers, posters, postcards and newspaper advertisements at the end of the summer of 2010. Potential participants were excluded if they i) had chronic allergies involving the upper respiratory tract, milk allergy or immunosuppressive illnesses or treatments in the previous year; ii) were current smokers; iii) had received antibiotic therapy two months prior to the start of the study; iv) were unwilling to discontinue any fiber or potentially immune enhancing dietary supplements (e.g., prebiotics, probiotics, echinacea, fish oil, vitamin E [˃100% of the RDA or ˃15mg/day]); v) had a cold on the first day of the study; vi) were being treated for Alzheimer’s disease; vii) were taking anti-inflammatory drugs or medication for constipation or diarrhea on a regular basis; viii) were receiving supplemental oxygen; vix) were not eligible or unwilling to receive the fall influenza vaccination for the year 2010-2011 as part of the study protocol or x) were unable to take foods or the study fiber without the aid of another person.

Additionally, potential participants had to be willing to i) complete daily and monthly questionnaires; ii) receive the fall influenza vaccination; iii) provide three blood and saliva and two stool samples; iv) answer a food frequency questionnaire towards the end of the study; v) take the study fiber twice daily for 24 weeks; and vi) provide a social security number to receive study payment (it is important to note that if the participants were unwilling to provide their social security number, they could still participate in the study, however no financial reimbursement would be provided). Each participant gave

66 written informed consent and all study procedures were in accordance with the ethical standards of the University of Florida Institutional Review Board.

Experimental Design

Participants (n=84) were randomly assigned to a supplement group before the cold and flu season at the beginning of October of 2011. Participants took part in the study for a total of 24 weeks during which they were monitored. This study was a prospective, randomized, parallel, double-blind, placebo controlled trial (Figure 3-1).

The participants were proportionally stratified based on BMI (≤25 and >25) and were randomized via sealed envelopes to receive either 0 g or 5 g of GOS (PurimuneTM GTC

Nutrition, Golden, CO) each day. The stratification and randomization schemes were generated by the study statistician who did not have direct contact with any of the participants.

The following is a list of study activities that were required in order to take part of the study. The week prior to the study start, height and weight were measured for each of the participants; additionally a baseline stool sample was collected. On the first day of the study, anthropometric measurements were taken such as height, weight, and arm and calf circumferences. Nutritional status was assessed and a short questionnaire about GI symptoms and level of activity was administered. On that same day, blood was drawn and saliva was collected. Every day of the study, participants were required to fill out a questionnaire about GOS intake, stress level, bowel habits and cold symptoms and participants were asked to consume the study supplement twice daily

(Appendix C). Between weeks two and three of supplementation, participants were asked to provide a final stool sample. After three weeks of supplementation, blood was drawn and saliva was collected for a second time, and participants were given the 67 influenza vaccine. Each month, participants were asked to complete the questionnaire about GI symptoms (GSRS) and level of activity questionnaire. Following five weeks of supplementation, or two weeks post-vaccination, blood was drawn and saliva was collected for a third time. On the last day of the study, 24 weeks post-supplementation, nutritional status was assessed using the MNA. Questionnaires on typical food intake,

GI symptoms and level of physical activity as well as a final questionnaire about which treatment group the participants thought that they were on were administered.

Throughout the study, participants were contacted by study coordinators, via telephone calls, for follow up in order to encourage compliance and sustain interest in the study.

Additionally, study coordinators met with participants a few times during the study to provide them with additional study supplements.

GOS Administration Protocol

The GOS supplements were provided in coded packets that were similar, both in size and shape, to commercially available single-serve “On the Go” drink mixes. The placebo consisted of Baker’s sugar, or sucrose, and both the GOS and placebo packets were similar in the way they looked and weighed. Silicon dioxide was added to all packets as a flow agent to improve the emptying of package contents. The supplement contained 86% GOS with 21% Degree of Polymerization (DP) 2 and 65% DP3 and above. Participants were instructed to pour the content of the packages into beverages, preferentially into hot beverages such as coffee or tea or into soft foods such as yogurt or pudding and to consume the supplement in its entirety twice daily, every day, for the entire 24 week study for a total intake of either 0 g or 5 g of GOS per day. Both the

GOS and the placebo had a slight sweet taste to decrease the probability of participants being able to distinguish the GOS packets from the placebo packets. 68

Influenza Vaccination

As part of the study protocol, participants received the 2010-2011 influenza flu vaccine. Vaccination was done during the second blood draw appointment which took place between Monday November 1st 2010 and Thursday November 4th 2010. The influenza vaccination was administered by a licensed nurse. The pre-filled syringes of vaccine were purchased from the Student Health Care Center of the University of

Florida and came from the same lot 111691 4P with an expiration date of 04/2011. The name of the vaccine was Fluvirin® manufactured by Novartis. Each dose of this vaccine

(0.5 mL) is made up of 15 μg hemagglutinin from three influenza virus types;

A/California/7/2009, NYMC X-181 (H1N1)-like virus; A/Perth/16/2009 (H3N2)-like virus; and B/Brisbane/60/2008-like virus. This vaccine is intended for intramuscular use only.

Fluvirin® is prepared from the extraembryonic fluid of embryonated chicken egg inoculated with a specific type of influenza virus suspension which also contains neomycin and polymyxin.

Study Questionnaires

On the day of randomization (first day of the study) participants met with the investigators and had their height and weight measured, gave blood and saliva samples, answered an MNA questionnaire, were instructed on the study procedures and obtained the supplement packets. Additionally, each participant was given a month’s worth of daily questionnaires identified with their study number in a paper folder with brass fasteners. The folder also contained a monthly GSRS questionnaire.

Participants were instructed on how to complete the questionnaires and when study coordinators met the participants three weeks later (for their second blood draw) they

69 ensured that questionnaires were filled out correctly. Questions regarding participant characteristics were contained in a short questionnaire at baseline. Daily questionnaires asked about consumption of the supplement to the nearest quarter packet, level of stress on a scale from 0 (no stress) to 10 (extremely stressed), number of bowel movements, antibiotic use and any cold and flu symptoms (Appendix C). If the participants were experiencing cold and flu symptoms they were asked to rate the symptom intensity (no symptom, mild symptoms, moderate symptoms, severe symptoms) for nasal symptom (running nose, stuffed nose, blowing the nose, yellow secretion, bloody secretion), pharyngeal symptoms (scratchy throat, sore throat, hoarseness), bronchial symptoms (couch, secretion, yellow secretion), headache, achiness (muscle pain), conjunctivitis (reddish eyes), fatigue, ear discomfort and stiffness/chills. The cold/flu symptom intensity (SI) score was calculated by attributing a score of 0 for no symptom, 1 for mild symptoms, 2 for moderate symptoms and 3 for severe symptoms and summing the individual symptoms intensities (107). The conjunctivitis symptom was not included in the total SI score. The daily questionnaire also contained “Yes”, “No” or “Does not apply” responses about sneezing, loss of appetite, fever (if “Yes”, the temperature in °F was required), use of prescription or non- prescription medication to relieve cold or flu symptoms, doctor’s visit and quality of life questions related to cold or flu symptoms such as problems with work, bathing/washing, getting dressed in the morning, visiting or socializing with friends, shopping or running errands, taking a vacation and finally if the participant thought they had the cold or the flu. These questions were administered as a generic health-status instrument for assessing quality of life and were adapted from the Nottingham Health Profile (227).

The baseline and monthly questionnaires asked about gastrointestinal symptoms and used the GSRS (23). The final questionnaire administered on the last day of the study asked participants to guess if 5.0 g of GOS or the placebo was in their daily packets and the reason behind their response.

Mini Nutritional Assessment

Nutritional status was assessed using the long form of the MNA questionnaire

(Nestlé Nutrition Institute), which is the most widely used and best validated nutritional assessment tool for aged adults. It was developed by Vellas and Guigoz in 1989 (228).

Study coordinators assigned to administer the MNA were trained using the information provided in the “Guide to Completing the Mini Nutritional Assessment MNA” and this ensured consistency in the administration of the questionnaire. The study coordinators further interpreted ambiguous questions. For example, Question A asks about a decrease in food intake caused by a loss of appetite, digestive problems, chewing, or swallowing difficulties. The answer options are severe, moderate, and no loss of appetite, and lead to a score of 0, 1, or 2, respectively. This question investigates decreased food intake, while all the answer options involve loss of appetite. Since a participant may have a decrease in appetite without having a decrease in food intake, or vice versa the investigators opted to ask the question “Over the past three months, have you eaten less than normal?”. According to the investigators, a change in nutrient intake, rather than a change in appetite, will have a more direct impact on an individual’s nutritional status. Question B is related to loss of weight over the past three months. Additionally, study coordinators recorded whether the loss of weight was intentional, from dieting for example. Question C, which inquires about mobility,

Question G, independent living, and Question N, mode of feeding, were not asked 71 during the administration of the MNA. In fact, the inclusion/exclusion criteria of our study called for participants to be mobile, live independently, not in assisted-living facilities such as a nursing home or hospital, and feed themselves without the help of another person. For this reason, all of the participants were automatically given the maximum score for these three questions. Question D relates to psychological stress and investigators asked: “Have you been under a lot of stress recently (for example, stress from the loss of a loved one, a recent illness or moving from your home)?”. Question E asks about neuropsychological problems and study coordinator formulated the question:

“Have you had any problems with depression?”. Additional detail is required if the participant’s response was “yes” and the resulting question was “Would you consider your depression to be mild or severe?”. Question F inquires about BMI, and required the use of a weighing scale (SECA Scale model 7701321134) and stadiometer (SECA model 213). The BMI used for the MNA was the baseline measurement. Question H asks about the number of prescription drugs taken on a daily basis. The researchers included anything found over the counter which was prescribed by their physician, such as aspirin or calcium supplements, as a prescription drug. Question I asks about bed sores or pressure ulcers. Question J inquires about the number of full meals the participant eats in general each day. Since this question is ambiguous in regards to what constitutes a “full meal”, study coordinators defined a full meal as eating at least three items of a different food group in one sitting. Question K asks about the daily consumption of dairy products, servings of beans or eggs per week and whether or not the participant consumes meat, fish or poultry every day. Regarding what a serving of dairy constitutes in this particular question, study coordinators gave examples such as

72 one cup of milk, three cheese cubes or half a cup of low fat cottage cheese. When talking about bean or egg servings, the examples of one serving of legumes was ½ cup cooked lentils or beans, ½ cup of raw/regular tofu, ¼ cup peanut or 1 cup plain soy milk.

Question L asks about how many servings of fruits and vegetables are consumed each day and examples of one serving given by the study coordinators were one piece of fruit, one medium cup of fruit or vegetable juice, or one cup of raw/cooked vegetables.

Question M asks about how much fluid is consumed daily and the study coordinators asked the question mentioning to participants that fluids can include water, coffee, juice, soda, tea and milk and that a cup is eight ounces. Rather than asking participants to choose from the list of possible answers, study coordinators asked participants to answer how many cups of fluid they drank every day and coordinators selected the multiple choice answer according to the participants’ response. Question O asks the participant to self-report his/her nutritional status, more specifically if they would describe it as being pretty good or not so good. Question P asks about the self-view of the participant’s health in comparison to people his/her age. Questions Q and R, which account for mid-arm circumference and calf-circumference, were measured according to the “Guide to Completing the Mini Nutritional Assessment MNA” with a Gulick tape measure calibrated for tightness.

Food Frequency Questionnaire (FFQ)

This self-administered questionnaire was developed to estimate usual and customary intake of a wide array of nutrients and food groups. It contains approximately 110 food items and takes 30 to 40 minutes to complete. The food list was developed from NHANES 1999-2002 dietary recall data and the nutrient database was developed using the USDA Food and Nutrient Database for Dietary Studies 73

(FNDDS) version 1.0. Individual portion size was inquired for each food and pictures were provided as a visual help and to increase the quantification exactness. The FFQ was administered at the end of the 24-week study to reflect the dietary habits of the participants throughout the study. This was to ensure that no dietary discrepancies were observed between the two treatment groups (placebo and GOS) which could account for observed differences.

Reagents

RPMI-1640 complete with L-glutamine and without phenol red was used as cell culture media (Mediatech, Manassas, VA). HEPES buffer at 25mM (Mediatech,

Manassas, VA), penicillin at 100U/mL and streptomycin at 100 µg/mL (Sigma Aldrich,

St. Louis, MO) were added to the RPMI-1640 complete.

Sodium Chloride solution (0.9%, 0.2% and 1.6%). Using the Mettler Toledo

AB104 analytical balance (Sigma #Z311774), NaCl was weighed and added to a 4000 mL graduated cylinder (Thermo Scientific Nalgene, T.C./T.D. 25˚C, Cat# S-06135-90) along with ultrapure H2O of Type I to provide the appropriate concentration. The NaCl was dissolved using a stir plate and a stirring rod. Under a laminar flow hood, the solution was transferred into three different 1 L Vacuum Filter Systems (0.22 μm,

Corning # 430517) and the solution was drawn into the collection bottle using a vacuum system. The filter funnel was then removed and the bottle was sealed with the sterile screw cap. The lid was sealed with Parafilm-M and the solution was removed from the

Laminar Flow Hood. The solution was stored at room temperature.

Blood Sample Collection

Blood samples were collected at baseline (on the randomization day), during the vaccination appointment (3 weeks post-supplementation) and 2 weeks post-vaccination 74

(5 weeks post-supplementation). The evening prior to blood sample collection, participants were contacted by telephone to remind them of the appointment as well as to restrain from eating after going to bed and on the morning of the blood draw as they needed to be fasted. Emphasis was made for them to stay hydrated and drink plenty of water. The following morning, between 6 a.m. and 10:30 a.m. the participants met with the study coordinators at the study site where blood samples were collected by a trained phlebotomist. Blood was drawn via a venipuncture and was collected into a serum separator (5 mL), L-heparin (10 mL), and two 6 mL purple top tubes (BD

Vacutainer K2 EDTA 10.8 mg). The two purple top EDTA tubes were then shipped to the Department of Food Science & Human Nutrition at Michigan State University where subsequent quantification of lymphocyte phenotypes was done.

Serum Collection

The serum tubes were centrifuged for 20 minutes at room temperature (RT) in a

Jouan CR312 centrifuge at 800 x g (2,000 rpm) to collect the serum. Serum (400 µl) was added to tubes containing 1.6 mL of RPMI-1640 (Cellgor; Mediatech, Herndon, VA) complete medium (100,000U/L penicillin; 100mg/L streptomycin; 2mM/L L-glutamine;

25mM/L HEPES buffer) to prepare culture medium with 10% autologous serum for each participant. Serum was also aliquoted for use during in vitro assays or stored at -70°C until assayed for C-reactive protein (CRP, ALPCO Immunoassays, Cat# 30-9710S) and endotoxin assay (PyrogeneTM, Recombinant Factor C Lonza endotoxin Detection Assay

50-658U).

Phenotypic Determination of Lymphocyte Population

In order to assess percent and absolute numbers of lymphocytes per microliter of blood, the TBNK Reagent and True Count tubes (BD Biosciences cat no. 337166) were

75 used. The Human TBNK Reagent kit provides a pre-mixed antibody cocktail with the antibodies (BD Biosciences cat no. 335775); anti-CD3 (clone SK7)-FITC, anti-CD19

(clone SJ25C1)-APC, anti-CD16 (clone 73.1)-PE, anti-56 (clone NCAM16.2)-PE, anti-

CD4 (clone SK3)-PE-Cy7, anti-CD8 (clone SK1)-APC-Cy7, and anti-CD45 (clone 2D1)-

PerCp-Cy5.5. Samples were run through a flow cytometer (FACS Canto flow) to quantify subpopulations of CD45+ cells and the clinical software FACS Canto was used for analysis. The software automatically creates gates for the lymphocyte populations, however these were the gates used; lymphocytes; low FSC/SSC + CD45+, TruCount

Beads; CD19-APC bright, T cells; CD3+/CD45+, T helper cells; CD45+/CD3+/CD4+,

Cytotoxic T cells; CD45+/CD3+/CD8+, B cells; CD45+/CD3-/CD19+/CD16-CD56-, and

NK cells; CD45+/CD3-/CD19-/CD16+CD56+. In brief, one TruCount tube was labeled for each sample. Twenty μL of TBNK reagent were pipetted into the bottom of each

TruCount tube, care was taken not to pipette any of the reagent on the side of the tube.

To these tubes, 50μL of well-mixed anticoagulated whole blood from the 6 mL purple top tubes (BD Vacutainer K2 EDTA 10.8 mg) were added to the bottom of the TruCount tubes. The tubes were then capped and vortexed gently. A dark incubation step followed where the tubes were left at room temperature. All subsequent procedures were executed in the dark. Following the fifteen minutes incubation, 450 μL of the

FACS Lysing solution (BD Biosciences cat no. 349202) was added to the tubes. The tubes were capped and vortexed gently. The tubes were then incubated at room temperature for an additional 15 minutes. Then the samples were run on the BD FACS

Canto II Flow cytometer using the clinical software. The gates were checked for

76 accuracy and adjusted if needed and the results were exported into an Excel

Spreadsheet.

CRP Assay

Stored serum samples were assayed for C-reactive protein using a commercially available kit (CRP, ALPCO Immunoassays, Cat# 30-9710S). Samples were diluted at a

1:100 using sample buffer (ALPCO Sample buffer, Cat# K 9710sPV). Data were reported in mg/L.

PBMC Isolation

Peripheral Blood Mononuclear cells (PBMC) were isolated from the blood in the heparinized tubes using a polysaccharide gradient under a laminar flow hood. Among the PBMCs obtained through this procedure, T cells are the most abundant representers consisting 70% of the PBMCs (70%) (229). The blood was diluted 1:2 with

0.9% room temperature (RT) 0.22 μm filtered NaCl and layered over Lympholyte® -H

(1.0770 ± 0.0001 g/cm3, Cedarlane Labs, Ontario, Canada, Product # CL5020) Cell

Separation Media. The gradient tubes were centrifuged (Jouan CR312) at 800 x g

(2,000 rpm) without a brake at RT for 20 minutes.

The PBMC band was removed from the gradient tube into a corresponding pre- labeled 50 mL conical tube that contained 20 mL of sterile RPMI-Wash solution and centrifuged in the Jouan CR312 centrifuge at RT for 10 minutes at 1,250 x g (2,500 rpm).

Red blood cells (RBC) within each pellet were lysed with the addition of 10 mL of

0.2% NaCl to each tube. After 30 seconds of brief gentle mixing, 10 mL of 1.6% NaCl was added to each tube followed by a gentle vortex. The tubes were centrifuged

(Jouan CR312) centrifuged for 10 minutes at 1,250 x g (2,500 rpm). The supernatant

77 was poured off and the cells were resuspended in RPMI wash (20 mL) and recentrifuged (Jouan CR312) for 10 minutes at 1,250 x g (2,500 rpm). Following the second wash, the medium was poured off and the pellets were resuspended in 1 mL of the RPMI-1640 complete Media (without phenol red), and the tubes were vortexed.

The PBMCs were counted at a 1:20 dilution with trypan blue on a hemocytometer.

A preformatted Excel Spreadsheet was used for calculations. The equation used to calculate concentration was = Average of all 4 corners x Dilution x 104 = cells/mL. Once the initial dilution of 2.0 x 106 cells/mL with 20% autologous serum was obtained using data from the spreadsheet, 7.25 x 106 cells were transferred into a total volume of 2.9 mL of RPMI-1640 complete (without phenol red) and 0.725 mL of autologous serum.

After a 1:1 dilution of cells with RPMI, or diluted mitogen, the final concentration was of

1.0 x 106 cells/mL in 10% autologous serum.

PBMC Stimulation with Mitogens (PHA and LPS) Assay

6 The PBMCs were plated at 1.0 x 10 cells/mL in RPMI-1640 complete media with

10% autologous serum, stimulated with LPS (Sigma # L3012, 20 mg/L) or PHA-L isolectin (MP Biomedicals # 08780361, 10 mg/L). The plates were incubated for 48 hours at 37C in a humidified atmosphere of 95% humidity and 5% CO2. After incubation, cells and supernatant fluid were centrifuged (500 x g, 10 min, RT) and the supernatant fluids were removed and frozen at -70C until cytokine production was quantified.

Multiplex Bead Based Assay

Different technologies exist to measure cytokines in biological samples.

Technologies such as ELISA (Enzyme-Linked ImmunoSorbent Assay), ELISPOT

(Enzyme-Linked ImmunoSorbent SPOT), multiplex cytokine bead arrays (CBAs;

Luminex or other), intracellular cytokine staining (ICS) by flow cytometry and gene expression platforms (230). Luminex allows for the rapid analysis of multiple analytes in small volumes of specimen. In vitro cytokine production in response to mitogens was quantified using a Human Cytokine Multiplex Immunoassay kit, according to the manufacturer’s directions (Bio-Rad, Group I). More specifically, interleukin (IL)-2, IL-5,

IL-10, IL-17A and interferon-γ (IFN-γ) concentrations were measured in PHA-stimulated cultures. Tumor necrosis (TNF)-α, IL-1β, IL-6, IL-8 and IL-10 were measured in LPS- stimulated cultures. The samples were analyzed using a Luminex® 200 instrument

(Austin, Texas) with xPONENT 3.1 software.

Saliva Collection

Unstimulated saliva collections were done on three separate occasions, concurring with the three different blood sample collections, between 6:00 a.m. and

10:30 a.m. Saliva collections were done at weeks 0, 3 (at vaccination) and 5 (2 weeks post-vaccination) of supplementation. Before saliva collection, participants were asked to fast for 12 hours, this is supported by other studies that recommend that unstimulated saliva collections should be made at rest in the fasted state and at specific times during the day to limit circadian rhythm variations (231). Participants were given a timer with a setting of ten minutes and asked to place these timers around their necks. They were then prompted to rinse out their mouth with water for at least ten seconds and then spit the water out. After rinsing their mouths, the timer set for ten minutes was started.

When ten minutes had elapsed, the timer was set for two minutes while the participant used half of a straw to passively pass their saliva into a 15 mL conical tube which was pre-labeled with the participant’s study number. After saliva collection, the conical tubes 79 were place in a cooler containing two ice-packs and brought back to the laboratory for processing shortly after. The 15 mL conical tubes were then centrifuged in a Jouan

CR312 centrifuge at 1,250 x g (3,500 rpm) for ten minutes and total volume of each sample was measured and recorded to the closest tenth of a microliter. The supernatant was divided into two equal portions and transferred using a 5 mL serological pipette into 1.5 mL amber microcentrifuge tubes. At this time, notes about strange textures and “gel-like” saliva samples were made into a study binder. Finally, the samples were stored at -70°C until assayed.

Salivary sIgA Assay

Saliva samples were thawed, vortexed, and centrifuged following storage.

Secretory immunoglobulin A (sIgA) was measured in saliva samples via ELISA using a commercially available kit (ALPCO Immunoassays, Cat# 30-8870). Samples were diluted at a 1:4000 dilution using wash buffer (ALPCO ELISA Wash Buffer, Cat# K

8870WP). Data were reported in µg/min. Previous studies have demonstrated that salivary sIgA could be measured using the ELISA method and this method yielded coefficient of variations (CV) of 5 to 10% (73).

Fecal Collection

Fecal samples were collected from participants during pre-baseline (1 week before the start of the study) and between weeks 2 and 3 of the study, post-supplementation.

Participants were instructed to drop off stool sample in a cooler between 7 a.m. and 6 p.m. Monday through Friday in accordance with the protocol and the instructions given during the consenting appointments when the stool collection kits were also provided. It is important to note that although participants were given the option of dropping off their samples throughout the day, most of the participants brought their samples with them at

80 their first and second blood draw appointments. Once collected, stool samples were mixed, divided into aliquots and frozen at -70°C within 6 hours of defecation.

Fecal sIgA Assay

Prior to analysis, all fecal samples were freeze-dried and then weighed for a moisture content value. Secretory immunoglobulin A (sIgA) was measured in freeze- dried fecal samples via Enzyme-Linked Immuno-Sorbent-Assay (ELISA) using a commercially available kit (ALPCO Immunoassays, Cat# 30-8870). Results for sIgA concentration were given in micrograms per milliliter (µg/mL).

Microbiota Analysis

Once fecal samples were collected, they were processed and analyzed at the

Emerging Pathogen’s Institute of the University of Florida. Quantitative PCR (qPCR) analysis was used to determine the bifidobacteria genome equivalents (GE) in fecal samples. Briefly, the qPCR reactions were first carried out with an initial melting step at

95C for 10 minutes followed by 40 cycles of 95C for 30 seconds, 58C for 60 seconds

(for bifidobacteria, 56C for the V3 universal primer set) and finally 72C for 1 minute.

The primer sequences used for bifidobacteria were forward 5’-

GATTCTGGCTCAGGATGAACG-3’ and reverse 5’-CGGGTGCTICCCCACTTTCATG-3’ and for the V3 universal primer forward 5’-CCTACGGGAGGCAGCAG-3’ and V3 reverse 5’-ATTACCGCGGCTGCTGG-3’ where Y = C or T and I = inosine. The V3 primer detects all bacterial genome equivalents and so it was used to determine the proportion of bifidobacteria genome equivalents compared with all bacterial genome equivalents. For the purpose of this study, the response to GOS supplementation was determined by results from the bifidobacteria genome equivalents. Participants that

81 were considered a “responder” had a two fold increase (≥double the baseline amount) of the bifidobacteria genome equivalents in the final stool sample when compared to the baseline stool sample. Participants were characterized as a “non-responder” if they did not have a 2-fold increase (˂double or decrease compared to baseline) of the bifidobacteria genome equivalents in the final versus the baseline stool sample.

Statistical Analysis

Compliance was determined by summing the daily reported proportion of the supplement packets consumed and dividing this number by the number of days that each participant participated in the study (most of the participants were involved in the study for 168 days). On the daily questionnaire, when the question about supplement intake was left unanswered, an intake of 0 packet was assumed. A lag time between the start of the supplement and its effect on study outcomes was expected and for this reason, data that was collected during the first week of the study was not included when evaluating self-reported cold and flu symptoms. Hence data for only 161 days were used in the statistical analyses. Additionally, the data collected on the first day of the study (week 0) was used to control for variations in individual responses for the monthly

GSRS questionnaires.

When analyzing the self-reported cold data, it was decided that a day of cold represented when the participant answered “yes” to the last question on the daily questionnaire “based on how you feel today, would you say that you have a cold or the flu?”. If a participant left the cold self-assessment question blank but answered questions about cold symptoms, an SI score above six, representing at least three different symptoms of which two symptoms would have to be reported with severe intensity (SI of 3) was considered a self-reported day of cold. The conjunctivitis cold 82 symptom was not included in the SI score. The probability of self-reporting a cold was determined using a fixed effect model with treatment group and age group as main effects and the treatment group by age group interaction.

Concentration of IFN-γ produced following PHA stimulation of PBMCs was determined using a mixed model with the following main effects in the main model; draw

(baseline, at vaccination and 2 weeks post-vaccination), treatment group, age group, responder group (≥2-fold change in fecal bifidobacteria genome equivalents), gender and MNA (as a continuous variable) with two and three way interactions and the random effect of participants to account for repeated observations on the same participants. Non-significant interactions were removed hierarchically until the final model was obtained. If necessary, data were transformed by taking the square root or natural log to correct for a skewed distribution. Back-transformed least squares means and approximate SEMs for the back transformed least squares means were used and obtained by averaging over all of the other effects and reporting at the average Mini

Nutritional Assessment (MNA) for the 81 participants of 27.

Percentage of lymphocytes that were positively identified as NK cells was obtained by averaging over the gender and draw (baseline, at vaccination and 2 weeks post- vaccination). Data were analyzed using a mixed model with the following main effects in the main model; draw (baseline, at vaccination and 2 weeks post-vaccination), treatment group, age group, responder group (≥2-fold change in fecal bifidobacteria genome equivalents), gender and MNA (as a continuous variable) with two and three way interactions and the random effect of participants to account for repeated

83 observations on the same participants. Non-significant interactions were removed hierarchically until the final model was obtained. Data represent the LSmean±SEM.

Five main categories (diarrhea syndrome, constipation syndrome, abdominal pain, indigestion syndrome, and reflux syndrome) comprising of fifteen gastrointestinal symptoms were measured on a monthly basis. Each of the fifteen symptoms were rated from 1 to 7 where 1 = no discomfort and 7 = severe discomfort. Scores within each category were summed and the gastrointestinal symptoms were adjusted for baseline responses. The baseline response was an individual’s response on the first

GSRS questionnaire which was on the day of randomization (day 1 of the study). A generalized linear mixed model was used.

When presenting participants’ characteristic and compliance, the differences between the two treatment groups were calculated using the non-parametric Mann-

Whitney Rank Sum test since the data was not normally distributed. Categorical data were compared using the chi-square statistics and data are reported as mean±SEM.

Cold symptom data were analyzed using a two-way analysis of variance. The average score for each treatment group was obtained by averaging each proportion of cold days with each of the specific symptoms within a cold and averaging the colds within a participant. Symptom Intensity (SI) score for each of the symptoms on a self- reported cold day ranged from 0 to 3. The average score for each treatment group was obtained by averaging each symptom score within a cold and averaging the colds within a participant. Total SI score on self-reported cold days was obtained by summing the

LSmean for each of the nine symptoms where the average score for each treatment

84 group was obtained by averaging each symptom within a cold and averaging the colds within a participant.

Quality of life measurements were obtained from the daily questionnaire. Average proportion of cold days was obtained by averaging proportion of cold days within a cold and averaging colds within each participant. Differences between treatment groups were calculated using a two-way analysis of variance.

For lymphocyte subpopulation cell numbers, salivary and fecal sIgA and CRP concentrations, cytokine production following mitogenic (LPS and PHA) stimulation of

PBMCs, a mixed model was used with the main effects (age group, treatment group, gender, draw [baseline, vaccination, post-vaccination], responder group [≥2-fold change in fecal bifidobacteria genome equivalents], and Mini Nutritional Assessment [MNA] as a continuous variable), all 2-way interactions, and the random effect for participant to account for the repeated observations. Non-significant effects and interactions were hierarchically eliminated from the model. To correct for a skewed distribution, data were transformed by taking the natural log of salivary sIgA, fecal sIgA and serum CRP, PHA- induced IL-2, IL-5, and IL-10 and LPS-induced IL-1β, IL-6, IL-8, IL-10 and TNF-α.

Analyses were executed with the use of SigmaPlot for Windows (Build 12.1.0.15 version 12.0, 2011, Systat Software, Inc., San Jose, CA) and the generalized linear mixed models were fitted using SASv9.2 (SAS Institute, Cary, NC). Unless stated otherwise, data are reported as LSmean±SEM.

Figure 3-1. Study design

CHAPTER 4 RESULTS

Ninety-seven participants were consented and assessed for eligibility (Figure 4-1).

Of this group, 13 were excluded of which six did not meet the inclusion criteria and seven declined to participate. Of the 84 participants remaining, 40 were randomized to the placebo group and 44 were randomized to the GOS group. In the placebo group, two participants withdrew from the study (one participant withdrew on the second day of the study because of allergy-like symptoms and another participant withdrew after 9 days for unknown reasons) and in the GOS group, one participant withdrew after 19 days of being part of the study due to a prolonged illness that she described as

“pneumonia” leading to a halt in GOS supplementation during the first few days of the study (Figure 4-1). Data from these three participants were not included in any of the analyses since only one out of the three blood draws was completed for these participants. Data was analyzed using intent to treat from 81. The study blinding was successful with 61% of the participants in the placebo group and 59% of the participants in the GOS group, correctly identifying the supplement that they were receiving

(p=0.993).

Both study groups were compliant with an average intake of more than 1.8 supplement packets per day (Table 4-1). There were no significant differences in the number of supplement packets consumed (total number or percentage) or participant characteristics between the two groups with the exception of age and the dietary score of the initial MNA (Table 4-1). Participants in the GOS group were on average 2.8 years older than participants in the placebo group and scored 0.5 points higher out of a possible 8 points in the dietary sub-score category of the MNA. However, when looking

87 at the total MNA score, no difference was observed between the placebo and GOS groups.

Self-Reported Cold Days, Symptoms and Symptom Intensities

There was no significant difference in the probability of self-reporting a cold on the daily questionnaire between the GOS and placebo group (data not shown). Due to the significant difference in age between supplement groups, data were subdivided into age groups that allowed for balanced group numbers (i.e., “Younger Aged” group [60 to 64 y] and “Older Aged” group [65 y or above]). With data divided by treatment group and age group, there was a significant effect of age group (p=0.0257) and a trend for the interaction between treatment group and age group (p=0.0846) on the probability of self-reporting a cold; consequently, all subsequent analyses were completed based on treatment and age group. Probability of self-reporting a cold using these age groups was significantly different with a lower probability (p=0.0332) of self-reporting a cold in

Older Aged compared to the Younger Aged participants in the GOS group. The Older

Aged or Younger Aged adults did not differ within the placebo group or between the placebo and GOS groups (Figure 4-2).

The proportion of participants with one or more self-reported cold days, the duration of cold episodes and number of cold days for participants with one or more self-reported cold days were not significantly different between treatment groups or age groups (Table 4-2). Significant differences were observed when looking at the proportion of cold days in respect to total days, where total participants in the GOS group had a significantly lower proportion of self-reported cold days in the Older Aged group (p=0.003) when compared to the placebo group. This same effect was not observed in the Younger Aged group where participants receiving the GOS 88 supplementation had a significantly higher proportion of self-reported cold days

(p=0.049). The average days of cold for participants who experienced at least one day of cold over the entire study averaged across both treatment and age groups, were not different. Average proportions of cold days with cold symptoms as well as symptom intensities for each of the symptoms for each participant were examined. A significant interaction (p=0.03) between treatment group and age group was observed when looking at the proportion of cold days with the stuffed/running nose cold symptom.

Younger Aged participants receiving the GOS supplementation had a significantly higher proportion of cold days when they experienced nasal symptoms when compared to the Younger Aged participants receiving the placebo. This significant difference was not observed in the Older Aged participants. No other significant difference was observed between both the age groups within each treatment group and the treatment group within each age group for the other cold symptoms as well as the total symptom intensity score on days of self-reported colds (Table 4-2).

When dividing participants only by treatment group and not by age groups, participants receiving the GOS supplementation had a significantly lower fatigue symptom score (p=0.045) on self-reported cold days and a significantly lower proportion of cold days with fatigue (p=0.0275) (data not shown).

Quality of Life Measurements

On self-reported cold days information on health-related quality of life was collected. When looking at quality of life questions on days of self-reported cold, no significant difference was observed between both the age groups within each treatment group and the treatment group within each age group (Table 4-3). Additionally, no difference between treatment groups or age groups was found when looking at the 89 proportion of cold episodes when participants visited the doctor because of the symptoms.

Lymphocytes Subpopulation Cell Numbers

There was an age group by treatment group interaction (p=0.0511) for the percentage of NK cells the percentage of NK cells was significantly higher (p=0.0022) in the Older Aged participants in the GOS group when compared to Younger Aged participants in the same group (Figure 4-4). Across age and supplement groups there was a significant draw effect where the percentage of NK cells was significantly lower at vaccination blood draw versus baseline (p=0.0485). There was no difference in NK cell percentages at vaccination and two weeks post-vaccination (data not shown).

For lymphocyte subpopulations and specifically cell numbers per microliter of whole blood, no differences were observed between treatment or age group or blood draw (i.e., baseline, vaccination, or 2-weeks post-vaccination Table 4-4).

A significant draw effect (p=0.04) was found for the total cell numbers of the immature T lymphocyte subset. A significant difference (p=0.0281), regardless of treatment group, was seen when comparing the total cell counts at baseline (31±4) and at 2 weeks post-vaccination (5 weeks post-supplementation) (26±4) averaged over all of the participants. Vaccination alone did not affect immune cell phenotypes or numbers in either the GOS or placebo group (Appendix D). When the data were further subdivided based on changes in fecal bifidobacteria genome equivalents (i.e., ≥2-fold increase or responder group versus ˂2-fold increase or non-responder group) there was no effect of treatment group or age by treatment group interaction for any of the lymphocyte subpopulations.

Cytokine Produced Following PHA Stimulation of Peripheral Blood Mononuclear Cells

Older Aged participants in the placebo group produced significantly lower concentrations of mitogen-induced IFN-γ than Younger Aged participants who received the placebo (p=0.0008), and IFN-γ concentrations were significantly higher with GOS supplementation of the Older Aged participants (p=0.006, Figure 4-3). Concentrations of IFN-γ were also influenced by draw (p=0.0010) and the interaction between MNA and gender (p=0.0360) and MNA and treatment group (p=0.0128) where concentrations of

IFN-γ decreased with increasing MNA score in participants in the placebo group.

Cytokine production following PHA stimulation of peripheral blood mononuclear cells was different between participants who were classified as responders versus non- responders based on changes in fecal bifidobacteria. A significant interaction was seen between responder group effect and treatment group effect for the concentration levels of IL-2 (p=0.0057, Table 4-5). Participants who were in the non-responder group and receiving GOS had significantly higher concentrations of IL-2 than non-responders receiving placebo (p=0.0182). Additionally, participants in the GOS group who were responders had significantly lower concentrations of IL-2, than participants from the same treatment group who were non-responders (p=0.0353). As for the concentration of IL-5 produced following PHA stimulation, a significant interaction was observed between the responder group effect and treatment group effect (p=0.0448). Participants in the responder group that were receiving GOS had significantly lower concentrations of IL-5 than responders receiving the placebo (p=0.0092). Finally, concentrations of IL-

10 showed a significant interaction between the responder group effect and the treatment group effect (p=0.0014). In fact, many significant differences were observed.

For participants in the placebo group, responders had a significantly higher concentration of IL-10 then non-responders (p=0.0097). Inversely, when only looking at participants in the GOS group, responders had a significantly lower concentration of IL-

10 then non responders (0.0471). When comparing participants who were responders; those in the GOS group had significantly lower concentrations of IL-10 following PHA stimulation than responders in the placebo group (p=0.0205). As for non-responders, when receiving the GOS supplementation, IL-10 concentrations were significantly higher than participants receiving the placebo (p=0.022).

Cytokine Produced Following LPS Stimulation of Peripheral Blood Mononuclear Cells

Following LPS stimulation of PBMCs, a significant interaction between responder group effect and treatment group was effect observed (p=0.048) for concentrations of

IL-10 (Table 4-6). Participants in the responder group receiving the placebo had a significantly higher concentration of IL-10 when compared to the participants supplemented with GOS (p=0.0124). Participants in the placebo group that were classified as responders also had a significantly higher concentration of IL-10 when compared to non-responders in that same group (p=0.0185). A significant effect of treatment group was observed for IL-1β and TNF-α; however there were no interactions between treatment group and blood draw. These data suggest that the group differences were present at baseline.

Salivary and Fecal Secretory Immunoglobulin A and C-Reactive Protein

No significant differences were observed for concentrations of salivary sIgA amongst the two treatment groups and throughout the three time points. However, a significant responder effect was observed where participants who saw their fecal

92 bifidobacteria increase more than two fold, regardless of the treatment group or draw, had higher concentrations of salivary sIgA then the non-responders (Table 4-7).

Fecal sIgA yielded no differences between treatment and responder groups in both baseline and final stool samples. Since general inflammation can be a confounder for tests of nutrient status and because it can also affect immune response, serum C- reactive protein, a measure of acute phase reaction, was measured. There was no significant difference in serum CRP across the three time points (baseline, at vaccination and 2 weeks post-vaccination) and amongst the two treatment groups

(Table 4-7).

GSRS Questionnaire

Gastrointestinal symptom scores were significantly lower with GOS supplementation for constipation (p=0.0101) syndrome and abdominal pain (p=0.0207)

(Figure 4-5). There was no difference between treatment groups for diarrhea syndrome, indigestion syndrome or reflux syndrome. Participants reported daily bowel movements/stools and when comparing stool output between the placebo group and the GOS group, there was no effect of treatment group (p=0.76), week (p=0.24) or interaction of treatment group and week (p=0.42) on the daily number of stool. On average, participants in the GOS group had 1.6±0.1 bowel movements per day and participants in the placebo group had 1.5±0.1 bowel movements per day.

Food Frequency Questionnaire

In addition no significant differences were observed between placebo responders, placebo non-responders, GOS responders and GOS non-responders in terms of intake of dietary fiber, dietary soluble fiber, average daily servings of whole grains, daily

93 servings of grains, dietary fiber from -beans, -vegetables and fruits, or -grains as well as the daily vegetable servings (data not shown).

Bifidobacteria Genome Equivalent (GE) Proportion Changes

Results following qPCR analysis showed significant differences between GOS and placebo supplementation when comparing baseline stool samples and samples obtained during week 2 of supplementation (post-treatment). Proportion of bifidobacteria GE was obtained by normalizing bifidobacteria GE with the universal primer set and change in proportion was obtained by subtracting baseline proportion of bifidobacteria GE from post-treatment proportion of bifidobacteria GE. In fact, it was observed that participants consuming 5 g of GOS daily had a significantly (p=0.013) greater positive change (i.e. increase) in the proportion of bifidobacteria GE

(0.078±0.023) when compared to changes in proportions of bifidobacteria in the placebo group (0.002±0.007).

Preliminary screen (n=435)

Declined to participate or self-excluded (n=33 )

Consented and assessed for eligibility (n= )

Not meeting inclusion criteria (n=6) Declined to participate (n= )

Randomized (n= 4)

Allocated to placebo (n=40) Allocated to GOS (n=44) Received placebo (n=40) Received GOS (n=44)

Discontinued intervention after  1 day due to allergy-like Discontinued intervention after symptoms (n=1)  1 days due to prolonged  days due to undisclosed illness (n=1) reason (n=1)

Analyzed (n=3 ) Analyzed (n=43)

Figure 4-1. Flow chart of participant recruitment, allocation and analysis. GOS, galactooligosaccharides

Figure 4-2. Effect of treatment group and age group on the probability of self-reporting a cold over the 6-month intervention. Data were analyzed using a fixed effects model with treatment group and age group as main effects and with the treatment group by age group interaction. Data are presented as LSmean±SEM. Bars with different letters are significantly different (p˂0.05).

Figure 4-3. Effect of treatment group and age group on concentration of interferon (IFN)-γ produced following stimulation of peripheral blood mononuclear cells with a T cell mitogen. Data were transformed by taking the square root (IFN- γ) to correct for a skewed distribution. Data were obtained by averaging over all of the other effects and reporting at the average Mini Nutritional Assessment (MNA) score for the 81 participants (27). Data were analyzed using a mixed model with the following main effects; blood draw (baseline, at vaccination and 2 weeks post-vaccination), treatment group, age group, responder group (≥2-fold change in fecal bifidobacteria genome equivalents), gender and MNA (as a continuous variable) with two and three way interactions and the random effect of participants to account for repeated observations on the same participants. Non-significant interactions were removed hierarchically until the final model was obtained. Data for IFN-γ concentration are back transformed least squares means and approximate SEMs for the back transformed least squares means. Bars with different letters are significantly different (p˂0.05).

Figure 4-4. Effect of treatment group and age group the percentage of lymphocytes that were identified as natural killer (NK) cells. Data were obtained by averaging over the gender and blood draw (baseline, at vaccination and 2 weeks post- vaccination). Data were analyzed using a mixed model with the following main effects; draw, treatment group, age group, responder group (≥2-fold change in fecal bifidobacteria genome equivalents), gender and Mini Nutritional Assessment score (as a continuous variable) with two and three way interactions and the random effect of participants to account for repeated observations on the same participants. Non-significant interactions were removed hierarchically until the final model was obtained. Data are reported as LSmean±SEM. Bars with different letters are significantly different (p˂0.05).

Placebo (n=38) GOS (n=43) 7 a a

a 5 a a b a 4 b

a 3 a

Sum of GI symptom scores ofsymptomGISum 2

1 Diarrhea Constipation Abdominal Indigestion Reflux Syndrome Syndrome Pain Syndrome Syndrome

Figure 4-5. Monthly gastrointestinal symptoms in participants receiving the placebo or galactooligosaccharides (GOS). Gastrointestinal (GI) symptoms were scored during the baseline period and then monthly during the 24 week intervention. Symptom scores from the baseline period were used to control for variation in individual responses for the subsequent monthly questionnaires. Individual symptoms were scored from 1 (no discomfort at all) to 7 (very severe discomfort) and summed for each symptom within each syndrome. Questions within each syndrome were as follows: diarrhea syndrome - diarrhea, loose stools, urgent need for defecation (possible scores 3 to 21); constipation syndrome - constipation, hard stools, and feeling of incomplete evacuation (possible scores 3 to 21); abdominal pain - abdominal pain, hunger pains, and nausea (possible scores 3 to 21); indigestion syndrome - rumbling, bloating, burping, and gas (possible scores 4 to 28); reflux syndrome - heartburn and acid reflux (possible scores 2 to 14). A generalized linear mixed model was used to analyze the data. Data represent the LSmean±SEM. Within each GI symptoms, bars with different letters are significantly different (p˂0.05).

Table 4-1. Demographics and compliance1 Placebo GOS P value2 n (%) 38 (47%) 43 (53%) NS Gender [n, (%)] Male 15 (39%) 16 (37%) NS Female 23 (61%) 27 (63%) Race/Ethnicity [n, (%)] White 34 (89%) 36 (84%) NS Hispanic/Latino 1 (3%) 1 (2%) African American 3 (8%) 6 (14%) Age (years) 64.8±1.0 67.6±1.1 0.04 BMI (kg/m2) 28.9±0.8 28.4±0.8 NS BMI category [n, (%)] Normal Weight (18.5 - 24.9)3 10 (26%) 13 (30%) NS Overweight (25.0 – 29.9) 11 (29%) 17 (40%) Obese (30.0 and above) 17 (45%) 13 (30%) Average packets consumed per d Mean ± SEM 1.88±0.04 1.91±0.02 NS Median (25th, 75th) 1.96 (1.85, 1.99) 1.96 (1.88, 1.99) Percentage of supplement packets 94±2 % 96±1 % NS consumed per d Daily stress level 2.3±0.2 1.8±0.2 NS (0=no stress, 10=extreme stress) Total initial MNA score 26.8±0.3 27.3±0.4 NS Anthropometric subscore4 7.4±0.2 7.7±0.1 NS Global sub-score5 10.6±0.1 10.4±0.2 NS Dietary sub-score6 5.4±0.2 5.9±0.2 0.04 Subjective sub-score7 3.5±0.1 3.3±0.1 NS MNA category [n, (%)] At risk of malnutrition9 3 (8%) 6 (14%) NS Well-nourished10 35 (92%) 37 (86%) 1Abbreviations: galactooligosaccharides, GOS; body mass index, BMI; day, d; Mini Nutritional Assessment, MNA. 2Differences between the two treatment groups were calculated using the non- parametric Mann-Whitney Rank Sum test. Categorical data were compared using the chi-square statistics. Data are reported as mean±SEM unless stated otherwise. 3One of the participants in the GOS group was underweight with a BMI of 18.4, but was analyzed within the normal weight category. 4Anthropometric sub-score: questions related to weight loss during the last 3 months, BMI, and mid-arm and calf circumference (possible score from 0 to 8). 5Global sub-score: mobility, questions related to psychological stress or acute disease in the past 3 months, neuropsychological problems, living circumstances (independent or not), prescription drugs per day, pressure sores or skin ulcers, and mode of feeding (possible score from 0 to 11). 6Dietary sub-score: questions related to decline in food intake over the past 3 months due to loss of appetite, digestive problems, chewing or swallowing difficulties, number of

100 full meals eaten daily, selected consumption markers for protein intake, servings of fruit or vegetables per day, and fluid consumed per day (possible score from 0 to 7). 7Subjective sub-score: questions related to self-view of nutritional and health status (possible score from 0 to 4). 8Individuals with a score between 17 and 23 out of a maximum of 30 are considered to be “at risk of malnutrition”. 9Individuals with a score of 23.5 or greater out of a maximum of 30 are considered to be “well-nourished”.

101

Table 4-2. Self-reported cold days1 Younger Aged Older Aged

(60 to 64 y) (65 y and older) Placebo GOS Placebo GOS P value Participants with >1 d of cold/total 11/22 17/22 7/16 7/21 Younger Aged: NS participants [n/n, (%)] (50%) (77%) (44%) (33%) Older Aged: NS 103/3689 134/3696 71/2669 55/3528 Younger Aged:0.049 Total cold days/Total days [n/n, (%)] (2.8%) (3.6%) (2.7%) (1.6%) Older Aged:0.003 Duration of each cold (consecutive d) 4.5±0.9 4.8±0.8 7.1±1.3 5.0±1.2 T:NS A:NS I:NS Cold d/participants with >1 cold days 9.4±2.0 7.9±1.6 10.1±2.5 7.9±2.5 T:NS A:NS I:NS Proportion of cold d with each symptom2 Stuffed/Running Nose 79.2±6.1%a 97.1±4.9%b 93.6±7.6%ab 81.0±7.6%ab T:NS A:NS I:0.03 Scratchy/Sore Throat 61.7±10.2% 69.8±8.2% 66.2±12.8% 60.9±12.8% T:NS A:NS I:NS Cough 60.4±11.0 68.8±8.9% 73.0±13.8% 65.5±13.8% T:NS A:NS I:NS Headache 34.5±11.5% 35.5±9.3% 44.2±14.4% 45.3±14.4% T:NS A:NS I:NS Achiness 25.2±9.8% 28.5±7.9% 50.3±12.3% 27.9±12.3% T:NS A:NS I:NS Fatigue 69.2±11.2% 48.6±9.1% 68.8±14.1% 55.0±14.1% T:NS A:NS I:NS Ear Discomfort 21.0±9.4% 11.5±7.5% 18.8±11.7% 20.0±11.7% T:NS A:NS I:NS Stiffness/Chills 21.3±10.0% 20.2±8.1% 32.6±12.6% 25.6±12.6% T:NS A:NS I:NS Fever 8.2±4.2% 4.4±3.4% 0.0±5.3% 5.6±5.3% T:NS A:NS I:NS Symptom Intensity (SI) score on cold d3 Stuffed/Running Nose 1.18±0.13 1.22±0.11 1.33±0.17 1.16±0.17 T:NS A:NS I:NS Scratchy/Sore Throat 0.95±0.17 1.00±0.14 0.94±0.22 0.78±0.22 T:NS A:NS I:NS Cough 0.87±0.17 0.86±0.14 1.04±0.22 0.93±0.22 T:NS A:NS I:NS Headache 0.56±0.19 0.48±0.15 0.47±0.23 0.68±0.23 T:NS A:NS I:NS Achiness 0.43±0.12 0.30±0.10 0.55±0.15 0.37±0.15 T:NS A:NS I:NS Fatigue 0.95±0.16 0.60±0.13 0.97±0.20 0.67±0.20 T:NS A:NS I:NS Ear Discomfort 0.23±0.10 0.15±0.08 0.19±0.13 0.24±0.13 T:NS A:NS I:NS Stiffness/Chills 0.33±0.15 0.29±0.12 0.38±0.19 0.38±0.19 T:NS A:NS I:NS Fever4 0.13±0.11 0.13±0.09 0.00±0.13 0.06±0.13 T:NS A:NS I:NS Total SI score on cold d5 5.6±0.8 5.0±0.7 5.9±1.1 5.3±1.1 T:NS A:NS I:NS

102

1Self-reported cold days are defined as days when participants reported “YES” to “Based on how you feel today, would you say that you have the cold or the flu?”. When the question was left unanswered but the symptom intensity score was >6 on that day, then it was counted as a day of cold and included in the analysis. Abbreviations: Galactooligosaccharides, GOS; days, d; symptom intensity, SI; treatment group, T; age group, A; interaction, I. Categorical data were compared using the chi-square statistics. Symptom data were analyzed using a two-way analysis of variance. Data are reported as LSmean±SEM 2Proportion of cold days when the participant with a self-reported cold was experiencing each specific symptom. The average score for each treatment group was obtained by averaging each proportion of cold days with each of the symptoms within a cold and averaging the colds within a participant. 3Symptom Intensity score for each of the symptoms (ranging from 0 to 3) on a self-reported cold day. Average score for each treatment group was obtained by averaging each symptom score within a cold and averaging the colds within a participant. 4Where no fever = 0; Yes with no °F reported would get a score of 1, a fever of <100°F would =1, 100-102°F=2, and >102°=3. 5Data obtained by summing the LSmean for each of the nine symptoms where the average score for each treatment group was obtained by averaging each symptom within a cold and averaging the colds within a participant.

103

Table 4-3. Quality of life measurements on days of self-reported cold1 Younger aged Older aged

(60 to 64 y) (65 y and older) Placebo GOS Placebo GOS P value Proportion of cold episodes Saw a doctor because of the cold/flu 19.7±9.5% 5.9±7.6% 14.3±11.9% 14.3±11.9% T:NS A:NS I:NS symptoms Proportion of cold days where; Cold/Flu Symptoms interfered with work2 9.3±5.5% 3.0±4.7% 9.8±6.7% 12.5±8.2% T:NS A:NS I:NS Used medication to relieve cold/flu 51.4±11.5% 56.6±9.3% 54.3±14.4% 51.8±14.4% T:NS A:NS I:NS symptoms3 Cold/Flu Symptoms interfered with 1.7±5.4% 5.9±4.1% 0.0±6.4% 5.2±6.4% T:NS A:NS I:NS bathing/washing Cold/Flu Symptoms interfered with getting 2.2±7.4% 6.4±5.7% 5.9±8.9% 21.9±8.9% T:NS A:NS I:NS dressed in the morning Cold/Flu Symptoms interfered with visiting 20.3±8.2% 11.0±6.0% 30.8±9.3% 19.0±10.1% T:NS A:NS I:NS or socializing friends/family Cold/Flu Symptoms prevented from 18.4±8.6% 10.4±6.6% 29.4±10.3% 26.2±11.2% T:NS A:NS I:NS shopping/running errands Cold/Flu Symptoms prevented from taking 4.0±5.3% 2.8±3.4% 0.0±4.8% 12.5±5.9% T:NS A:NS I:NS a winter vacation or weekend away 1Self-reported cold days are defined as days when participants reported “YES” to “Based on how you feel today, would you say that you have the cold or the flu?”. When the question was left unanswered but the symptom intensity score was >6 on that day, then it was counted as a day of cold and included in the analysis. Available Yes responses to questions are reported when applicable. Average proportion of days for each treatment group was obtained by averaging each proportion of cold days within a cold and averaging the colds within a participant. Differences between treatment groups were calculated using a two-way analysis of variance. Data are expressed as LSmean±SEM Abbreviations: Galactooligosaccharides, GOS; treatment group, T; age group, A; interaction, I. 2This question referred to paid employment. 3This question referred to non- and/or prescription medication.

104

Table 4-4. Lymphocytes subpopulations cell numbers per microliter of whole blood by treatment and responder group, and blood draw (baseline, at vaccination and 2 weeks post-vaccination)1 Non-Responders Responders (≤2-fold increase in (≥2-fold increase in fecal bifidobacteria fecal bifidobacteria genome equivalents) genome equivalents) Placebo GOS P value2 Placebo GOS Placebo GOS P value3 Total T lymphocytes D1-3: 38 D1-2: 434 D1-3: 17 D1-3: 17 D1-3: 19 D1-2: 26 D3: 42 T: NS D3: 25 T: NS Baseline 1560±99 1620±96 D: NS 1530±158 1610±142 1600±133 1620±132 R: NS At vaccination 1570±114 1590±85 I: NS 1620±181 1560±109 1550±159 1600±123 I: NS Post-vaccination 1540±96 1510±81 1550±132 1580±111 1550±150 1470±115 Total cytotoxic T cells D1-3: 38 D1-2: 43 D1-3: 17 D1-3: 17 D1-3: 19 D1-2: 26 D3: 42 T: NS D3: 25 T: NS Baseline 481±54 510±56 D: NS 484±70 470±68 486±90 537±82 R: NS At vaccination 487±60 482±49 I: NS 508±72 419±51 472±101 523±73 I: NS Post-vaccination 463±49 440±39 480±61 427±46 454±81 449±58 Total T helper cells D1-3: 38 D1-2: 43 D1-3: 17 D1-3: 17 D1-3: 19 D1-2: 26 D3: 42 T: NS D3: 25 T: NS Baseline 1040±68 1090±63 D: NS 1020±111 1140±95 1080±88 1060±84 R: NS At vaccination 1050±74 1090±60 I: NS 1080±125 1140±81 1040±98 1050±85 I: NS Post-vaccination 1050±71 1060±64 1040±96 1150±92 1070±113 994±87 Total immature T cells D1-3: 38 D1-2: 43 D1-3: 17 D1-3: 17 D1-3: 19 D1-2: 26 D3: 42 T: NS D3: 25 T: NS Baseline 23±3 38±8 D: 0.045 22±4 39±10 25±6 37±11 R: NS At vaccination 21±2 34±7 I: NS 22±4 34±10 19±2 35±10 I: NS Post-vaccination 22±3 30±7 21±3 33±9 23±5 28±11 Total leukocytes D1-3: 38 D1-2: 43 D1-3: 17 D1-3: 17 D1-3: 19 D1-2: 26 D3: 42 T: NS D3: 25 T: NS Baseline 2280±163 2170±107 D: NS 2360±308 2170±175 2280±171 2160±139 R: NS At vaccination 2310±222 2110±99 I: NS 2510±436 2070±136 2190±211 2140±140 I: NS Post-vaccination 2250±166 2060±94 2410±309 2140±129 2170±182 2010±132

105

Table 4-4. Continued Non-Responders Responders (≤2-fold increase in (≥2-fold increase in fecal bifidobacteria fecal bifidobacteria genome genome equivalents) equivalents) Placebo GOS P value2 Placebo GOS Placebo GOS P value3 Total B cells D1-3: 38 D1-2: 43 D1-3: 17 D1-3: 17 D1-3: 19 D1-2: 26 D3: 42 T: NS D3: 25 T: NS Baseline 353±90 210±14 D: NS 441±199 199±22 297±26 218±18 R: NS At vaccination 380±128 211±14 I: NS 519±284 198±22 277±28 220±18 I: NS Post-vaccination 361±98 219±15 477±217 213±27 280±22 223±19 1Abbreviations: Galactooligosaccharides, GOS; treatment group, T; blood draw, D; interaction, I; responder group, R. Data were analyzed using a mixed model with the main effects (age group, treatment group, gender, draw [baseline, vaccination, post-vaccination], responder group, and Mini Nutritional Assessment [MNA] as a continuous variable), all 2- way interactions, and the random effect for participant to account for the repeated observations. Non-significant effects and interactions were hierarchically eliminated from the model. Unless stated otherwise, data represents the sample mean±SEM. 2P value for the treatment group effect (T); blood draw (D; baseline versus at vaccination versus 2 weeks post- vaccination), and interaction (I) between blood draw and treatment group. 3P value for the treatment group effect (T), the responder group effect (R), and the interaction between treatment group and responder group (I). 4The number of participants varies between blood draws (baseline, D1; at vaccination, D2; 2 weeks post-vaccination, D3) due to limited volume of blood obtained from each of the participants. 5Reported means are averaged over age group and reported at the average MNA where MNA*age group (p=0.0445).

106

Table 4-5. Cytokines produced following phytohemagglutinin stimulation of peripheral blood mononuclear cells by treatment and responder group, and blood draw (baseline, at vaccination and two weeks post-vaccination)1 Non-Responders Responders (≤2-fold increase in (≥2-fold increase in fecal bifidobacteria fecal bifidobacteria genome equivalents) genome equivalents) Placebo GOS P value2 Placebo GOS Placebo GOS P value3 PHA Stimulated D1-3: 37 D1-2: 434 D1-3: 16 D1-3: 17 D1-3: 19 D1-2: 26 IL-2 (ng/mL)5 D2: 42 T: NS D3: 25 T: NS Baseline 0.68±0.12 0.49±0.08 D: 0.004 0.63±0.23 0.49±0.08 0.78±0.13 0.49±0.12 R: 0.018 At vaccination 0.33±0.06 0.25±0.04 I: NS 0.29±0.09 0.30±0.07 0.39±0.09 0.21±0.05 I: 0.0066 Post-vaccination 0.30±0.06 0.23±0.04 0.28±0.10 0.18±0.04 0.33±0.07 0.27±0.06 D1-3: 38 D1-2: 43 D1-3: 17 D1-3: 17 D1-3: 19 D1-2: 26 IL-5 (ng/mL)7 D2: 42 T: NS D3: 25 T: NS Baseline 0.18±0.04 0.11±0.02 D:˂0.0001 0.14±0.03 0.15±0.03 0.22±0.08 0.09±0.02 R: NS At vaccination 0.16±0.04 0.09±0.01 I: NS 0.10±0.03 0.12±0.02 0.21±0.07 0.07±0.01 I: 0.0458 Post-vaccination 0.14±0.03 0.10±0.02 0.09±0.03 0.12±0.03 0.19±0.05 0.08±0.02 D1-3: 38 D1-2: 43 D1-3: 17 D1-3: 17 D1-3: 19 D1-2: 26 IL-10 (ng/mL) D2: 42 T: NS D3: 25 T: NS Baseline 0.21±0.02 0.21±0.02 D: NS 0.15±0.02 0.25±0.04 0.26±0.04 0.18±0.03 R: NS At vaccination 0.21±0.02 0.20±0.02 I: NS 0.17±0.03 0.24±0.03 0.25±0.03 0.18±0.03 I: 0.0019 Post-vaccination 0.23±0.02 0.25±0.03 0.19±0.03 0.31±0.07 0.26±0.03 0.22±0.03 1Abbreviations: Galactooligosaccharides, GOS; peripheral blood mononuclear cell, PBMC; phytohemagglutinin, PHA; interleukin, IL; treatment group, T; blood draw, D; interaction, I; responder group, R. Data were analyzed using a mixed model with the main effects (age group, treatment group, gender, draw [baseline, vaccination, post-vaccination], responder group, and Mini Nutritional Assessment [MNA] as a continuous variable), all 2-way interactions, and the random effect for participant to account for the repeated observations. Non-significant effects and interactions were hierarchically eliminated from the model. Data were transformed by taking the natural log (PHA-induced IL-2, IL-5, and IL- 10) to correct for a skewed distribution. Unless stated otherwise, data represents the sample mean±SEM. 2P value for the treatment group effect (T), blood draw (D; baseline versus at vaccination versus 2 weeks post- vaccination) and the interaction (I) between draw and treatment group.

107

3P value for the treatment group effect (T), the responder group effect (R), and the interaction between treatment group and responder group (I). 4The number of participants varies between blood draws (baseline, D1; at vaccination, D2; 2 weeks post-vaccination, D3) due to limited volume of blood obtained from each of the participants. 5For IL-2; reported means are averaged over age group and reported at the average MNA where the main effects of age group (p=0.005 ), MNA*draw (p=0.000 ) and MNA*responder group (p=0.01 5) were significant. 6For participants in the non-responder group receiving placebo versus GOS (p=0.0182) and for participants in the GOS group that are responders versus non-responders (p=0.0353). 7For IL-5; reported means are averaged over age group and gender where the main effects of age group (p=0.045 ) and gender (p=0.0014) were significant. 8For participants in the responder group receiving placebo versus GOS (p=0.0092). For participants the responder group receiving placebo versus GOS (p=0.0205), for participants in the non-responder group receiving placebo versus GOS (p=0.0222), for participants in the placebo group that are responders versus non- responders (p=0.0097) and for participants in the GOS group that are responders versus non-responders (p=0.0471).

108

Table 4-6. Cytokines produced following lipopolysaccharide stimulation of peripheral blood mononuclear cells by treatment and responder group, and blood draw (baseline, at vaccination and two weeks post-vaccination)1 Non-Responders Responders (≤2-fold increase in fecal (≥2-fold increase in fecal bifidobacteria genome bifidobacteria genome equivalents) equivalents) Placebo GOS P value2 Placebo GOS Placebo GOS P value3 LPS Stimulated D1-3: 36 D1-2: 434 D1-3: 17 D1-3: 17 D1-3: 17 D1-2: 26 IL-1β (ng/mL) D2: 42 T: 0.045 D3: 25 T: 0.045 Baseline 3.34±0.30 2.78±0.20 D: NS 3.07±0.38 2.76±0.29 3.66±0.51 2.79±0.28 R: NS At vaccination 3.92±0.39 2.87±0.22 I: NS 4.01±0.58 2.78±0.38 3.94±0.61 2.93±0.27 I: NS Post-vaccination 3.52±0.33 2.95±0.21 3.72±0.60 2.81±0.29 3.45±0.35 3.04±0.30 D1-3: 38 D1-2: 43 D1-3: 17 D1-3: 17 D1-3: 19 D1-2: 26 IL-6 (ng/mL) D2: 42 T: NS D3: 25 T: NS Baseline 28±2 24±2 D: NS 30±4 23±4 28±2 25±3 R: NS At vaccination 28±2 23±2 I: NS 24±2 21±3 33±3 24±2 I: NS Post-vaccination 29±2 26±2 25±3 25±4 32±3 26±3 D1-3: 37 D1-2: 39 D1-3: 17 D1-3: 16 D1-3: 18 D1-2: 23 IL-8 (ng/mL) D2: 38 T: NS D3: 22 T: NS Baseline 175±20 150±17 D: NS 204±39 162±28 156±16 142±22 R: NS At vaccination 180±20 144±11 I: NS 143±21 152±20 216±35 139±13 I: NS Post-vaccination 161±15 158±17 134±15 181±32 187±25 142±17 D1-3: 38 D1-2: 43 D1-3: 17 D1-3: 17 D1-3: 19 D1-2: 26 IL-10 (ng/mL)5 D2: 42 T: NS D3: 25 T: NS Baseline 285±27 268±30 D: NS 246±40 297±52 329±39 249±37 R: NS At vaccination 295±24 239±22 I: NS 223±29 257±45 348±34 227±21 I: 0.0486 Post-vaccination 327±33 323±48 273±47 399±110 357±44 272±30 D1-3: 38 D1-2: 43 D1-3: 17 D1-3: 17 D1-3: 19 D1-2: 26 TNF-α (ng/mL) D2: 42 T: 0.019 D3: 25 T:0.019 Baseline 4020±394 2590±314 D: ˂0.0001 4240±666 2940±618 3980±512 2370±330 R: NS At vaccination 3130±373 1720±206 I: NS 2630±384 1830±413 3680±646 1650±214 I: NS Post-vaccination 3210±414 2070±268 2910±526 2570±583 3590±682 1740±202

109

1Abbreviations: Galactooligosaccharides, GOS; peripheral blood mononuclear cell, PBMC; lipopolysaccharide, LPS; interleukin, IL; treatment group, T; blood draw, D; interaction, I; responder group, R. Data were analyzed using a mixed model with the main effects (age group, treatment group, gender, blood draw [baseline, vaccination, post-vaccination], responder group, and Mini Nutritional Assessment [MNA] as a continuous variable), all 2-way interactions, and the random effect for participant to account for the repeated observations. Non-significant effects and interactions were hierarchically eliminated from the model. Data were transformed by taking the natural log (LPS-induced IL-1β, IL-6, IL-8, IL-10 and TNF-α) to correct for a skewed distribution. Unless stated otherwise, data represents the sample mean±SEM. 2P value for the treatment group effect (T), blood draw (D; baseline versus at vaccination versus 2 weeks post- vaccination), and the interaction (I) between draw and treatment group. 3P value for the treatment group effect (T), the responder group effect (R), and the interaction between treatment group and responder group (I). 4The number of participants varies between blood draws (baseline, D1; at vaccination, D2; 2 weeks post-vaccination, D3) due to limited volume of blood obtained from each of the participants. 5For IL-10; reported means are averaged over age group where the main effect of age group (p=0.04) was significant. 6For participants in the responder group receiving placebo versus GOS (p=0.0124) and for participants in the placebo group that are responders versus non-responders (p=0.0185).

110

Table 4-7. Salivary and fecal secretory immunoglobulin A and C-reactive protein by treatment and responder group, and blood draw (baseline, at vaccination and 2 weeks post-vaccination)1 Non-Responders Responders (≤2-fold increase in (≥2-fold increase in fecal fecal bifidobacteria bifidobacteria genome genome equivalents) equivalents) Placebo GOS P value2 Placebo GOS Placebo GOS P value3 5 D1-3: 34 D1-2: 41 D1-3: 15 D1-3: 16 D1-3: 17 D1-2: 25

Salivary sIgA (μg/min)4 D3: 40 D3: 24 T: NS T: NS Baseline 476±47 779±96 435±78 566±99 516±66 916±138 D: NS b b a a R: 0.050 At vaccination 551±68 738±92 463±76 618±111 633±115 815±133 I: NS I: NS Post-vaccination 526±64 756±143 472±72 523±127 589±111 911±220

Fecal sIgA (ng/mg) B & F: 366 B & F: 40 T: NS B & F: 16 B & F: 17 B & F: 19 B & F: 23 T: NS

Baseline 3660±703 2670±462 D: NS 3930±1340 2880±715 3560±736 2510±617 R: NS

Final 3510±574 2940±413 I: NS 3370±836 3270±667 3750±837 2690±530 I: NS D1: 34 D1-3: 35 D1-3: 17 D1-3: 16 D1: 15 D1-3: 19

Serum CRP (mg/L) D2-3: 35 D2-3: 16 T: NS T: NS Baseline 3.4±0.8 2.9±0.5 4.8±1.4 2.7±0.7 2.0±0.5 3.2±0.7 D: NS R: NS At vaccination 3.5±1.1 3.5±0.8 3.9±2.1 4.1±1.5 3.4±0.8 2.9±0.6 I: NS I: NS Post-vaccination 3.1±0.9 2.9±0.6 3.7±1.7 2.7±0.8 2.8±0.7 3.0±0.8 1Abbreviations: Galactooligosaccharides, GOS; secretory immunoglobulin A, sIgA; treatment group, T; blood draw, D; interaction, I; responder group, R; C-reactive protein, CRP. Data were analyzed using a mixed model with the main effects (age group, treatment group, gender, draw [baseline, vaccination, post-vaccination], responder group, and Mini Nutritional Assessment [MNA] as a continuous variable), all 2-way interactions, and the random effect for participant to account for the repeated observations. Non-significant effects and interactions were hierarchically eliminated from the model. Data were transformed by taking the natural log (salivary sIgA, stool sIgA and serum CRP) to correct for a skewed distribution. Unless stated otherwise, data represents the sample mean±SEM. 2P value for the treatment group effect (T), blood draw (D; baseline versus at vaccination versus 2 weeks post vaccination), and for the interaction (I) between draw and treatment group. 3P value for the treatment group effect (T), the responder group effect (R), and the interaction between treatment group and responder group (I). 4For salivary sIgA; reported means are averaged over gender where the main effect of gender was significant (p=0.001 ).

111

5The number of participants varies between blood draws (baseline, D1; at vaccination, D2; 2 weeks post-vaccination, D3) due to limited volume of blood obtained from each of the participants. 6The number of participants varies between stool collections (Baseline, B; Final, F) because of inability to produce a stool sample.

112

CHAPTER 5 DISCUSSION AND CONCLUSIONS

The primary aim of this study was to determine the effect of GOS on days of cold and flu. Since a decrease in cold days was observed with the same prebiotic in a previous study (4), it was hypothesized that GOS would significantly decrease days of illness; however, this outcome was not observed. However, there was a significant age difference (p=0.04) between the placebo group and the GOS group where participants in the GOS group were significantly older (Table 4-1). The next logical step was to determine if this significant age difference was also biologically significant, specifically if differences in age could influence certain immune parameters. Supporting this biological difference, a study by McNerlan and colleagues demonstrated that healthy participants aged 0 to years had significantly higher (p˂0.0001) circulating percentage of NK cells (19%) when compared to younger participants aged 40 to 69 years (13%) (232). A study by Vulevic et al. conducted in healthy aged adults demonstrated that a trans-GOS had significant effects on the immune system evidenced by an improvement of NK cell activity. It is possible that GOS could be exerting its beneficial effects by priming NK cells. For these reasons a post-hoc analysis was done separating the participants into two age groups.

After separating participants into a “Younger Aged” group (60 to 64 y) and “Older

Aged” group (65 y or older), a significant difference in percentage of NK cells at 2 weeks post-vaccination (5 weeks post-supplementation) was seen in the GOS group between the “Younger Aged” and the “Older Aged”. Participants in the GOS treatment group that were in the “Older Aged” category had a significantly higher percentage of NK cells

(Figure 4-4). These results are consistent with a study by Yang et al. (233), which

113 looked at age-related changes of peripheral blood lymphocytes and compared the percentage of NK cells between middle-aged (45 to 64 y) and elders (65 y or above).

This study also demonstrated that participants in the elder group had a significantly higher percentage of NK cells.

Considering the probability of self-reporting a cold and ignoring the age of the participants, it was unclear whether the treatment or the age effect was being observed.

In order to uncouple age from the treatment effect, a model which included both effects was executed and this was possible because both age groups were present in both treatment groups. There appears to be a treatment group as well as an age group and treatment group interaction (Figure 4-2). The probability of reporting a cold by participants receiving the GOS supplement seemed to be dependent upon which age group they were part of. In the GOS group, a significant difference was observed between the “Younger Aged” and “Older Aged” participants. This suggests that by dividing the data into age categories, we successfully separated the treatment from the age effect. It was found that participants aged 65 years of age or older in the GOS group had a significantly lower probability of reporting a cold than participants aged 60 to 64 years old in that same treatment group. Although NK cell function was not measured in this study, a study executed by Gills et al. demonstrated that after supplying aged adults with lactic acid bacteria (LAB) the most vulnerable ages within the elder population were the ones benefiting the most from the probiotic supplementation (156). This study was also executed in healthy aged adults where the age range was 60 to 84 years old and it was shown that participants aged over 70 years experienced a significantly higher increase in NK cell function following LAB

114 consumption when compared to the younger participants less than 70 years old. These results are consistent with our findings suggesting that the extremes of age (i.e. “Older

Aged” age group) are benefiting the most from GOS supplementation and that this could be partly due to an inherent suppressed NK cell function which is rescued by the intervention.

It was also found that participants in the “Older Aged” group receiving GOS had a significantly higher concentration of IFN-γ following mitogen stimulation of peripheral blood mononuclear cells than participants receiving the placebo in that same age group

(Figure 4-3). This result is supported by a study from Holma and colleagues where a combination of probiotics and GOS given to 18 healthy men for two weeks resulted in a significant increase in the selective secretion of IFN-γ by PHA stimulated PBMCs (234).

This also suggests that IFN-γ does not have a direct influence on the probability of self- reporting a cold. If this was the case, there would also be a significant difference in self- reported colds between participants in the GOS group and those that are in the placebo group in the “Older Aged” age group. Interferon-γ does not seem to be entirely explaining what is happening because, unlike what is observed with IFN-γ, there are no differences in self-reported colds between the “Younger Aged” and the “Older Aged” receiving the placebo. Interferon-γ has been shown to possess antiviral properties and confer protection against influenza infections (235). Although this was not observed in this study, the influenza virus was not specifically considered; rather, common cold infections in general, caused by more than 200 different viruses, were investigated.

Interferon-γ is known to stimulate and be secreted by NK cells and may explain the significantly higher percentage of NK cells observed in the GOS “Older Aged” group

115 which could in turn help interpret the difference seen in self-reported colds. A study by

Arunachalam and colleagues demonstrated that in healthy elderly participants, the oral consumption of a probiotic (Bifidobacterium lactis HN019) was shown to enhance the capacity of peripheral blood cells to secrete this cytokine (236). In fact, by modulating the microbial population in the GI tract (e.g. increasing the immunoregulatory bifidobacteria), GOS could enhance cellular immunity and lead to a decreased probability of self-reporting a cold.

Effect of GOS on Symptom Intensity and Quality of Life

Although many studies involving nutraceuticals or functional foods have been unable to demonstrate an additional benefit in regards to reducing the occurrence of

URTIs, many showed an improvement in the severity and duration of cold symptoms.

This was not the case in our study where GOS supplementation did not have a significant effect on the proportion of self-reported cold days with different cold symptoms or symptom intensity score on those cold days (Table 4-2). The reason behind this lack of effect may be explained by the fact that of the 81 participants over the entire 24 week study, only about half self-reported a day of cold. The population taking part in our study seemed remarkably healthy considering that it was conducted during peak cold and flu season. This explanation may also clarify why we did not capture any differences between the groups and perhaps a larger population sample is needed to increase the likelihood of self-reported cold days. The only significant difference observed in terms of symptoms on self-reported cold days was the proportion of cold days with running or stuffed nose where participants in the Younger Aged group receiving GOS had a significantly higher proportion of cold days with this symptom than participants receiving the placebo in the same age group, additionally there were 116 significantly more cold days as a percentage of total study days in these participants with GOS supplementation.

Quality of life measurements during cold days were monitored throughout the duration of the study. The objective of these health-related questions was to determine the impact of the common cold on physical, social an emotional functioning. On a self- reported cold day, quality of life questions were analyzed in order to determine if GOS could positively impact participants by decreasing the degree of interference of cold symptoms on activities of daily living (Table 4-3). Since the economic burden resulting from the common cold is partly caused by a decrease in income following absences from paid employment, whether GOS could reduce interference of cold symptoms with work was of interest. Although no significant differences were observed between both, treatment groups and age groups, when focusing on the Younger Aged participants, which would be participants more likely still working and not retired, there seems to be a trend with GOS showing a lower proportion of self-reported cold days where symptoms interfered with work. A possible explanation as to why no significant differences were observed between the GOS and placebo group in regards to quality of life questions could be the questions themselves. The questions employed were adapted from the

Multidimensional Assessment of Fatigue (MAF) scale and are predominantly used to measure four dimensions of fatigue rather than cold symptoms.

GOS and the Immune System

To determine the mechanism by which GOS could have an effect on the probability of self-reporting a cold, the ability of GOS to modulate the immune system was examined. When comparing GOS to the placebo in terms of immune parameters, we did not observe many significant differences. Besides the increase in NK cells as a 117 percentage of lymphocytes following five weeks of supplementation (Table D-1), GOS did not seem to have modified total cell numbers in terms of lymphocyte subpopulations per microliter of whole blood (Table 4-4). These results are consistent with previous findings where supplementation with either Bifidobacterium bifidum Bb12 or

Lactobacillus acidophilus strain LA1 did not have an effect on lymphoid cell populations in peripheral blood (237). In human clinical trials, investigators are limited in their ability to obtain samples; blood and external secretions such as saliva are generally the most typical samples obtained. It is important to keep in mind that the majority of immune cells do not reside in the blood stream and that at any given time only 2% of total lymphocytes are circulating in the blood (238). Perhaps, what is occurring locally in the

GALT is not being captured when measuring lymphocytes in the systemic circulation. A murine study by Gopalakrishnan and colleagues showed that GOS supplementation significantly increased the percentage of NK cells in the spleen and in the mesenteric lymph node (43). This suggests that GOS does affect lymphocytes subpopulation cell numbers in specific organ systems and that, for the time being, human clinical trials are unable to provide us with such insightful information.

Increase in Immunogenic Bacteria

The study by Vulevic and colleagues showed that a trans-GOS mixture had significant effects on the immune system of healthy aged adults demonstrated by an improvement in NK cell activity. These investigators also demonstrated that GOS significantly promoted the growth of Bifidobacterium bifidum and Bifidobacterium longum and hypothesized that this change in the microbiota enhanced gastrointestinal tract physiological functions and subsequently the immune system (5). No link was established between these changes in immune parameters and the probability of URTI. 118

To our knowledge, this was the first prebiotic study where a correlation between changes in bifidobacteria and changes in cytokine production by PBMCs following mitogenic stimulation was investigated. As it was hypothesized that GOS was acting predominantly by altering the microbiota, through an increase in bifidobacteria genome equivalents, and that this was consequently changing the cytokine production profile of immune cells, it would be expected that placebo responders (that somehow also increased by two-fold their fecal bifidobacteria) and GOS responders to have similar cytokine production promoting a less inflammatory state; but this was not the case.

When strictly comparing the responder participants either receiving GOS or placebo, we found that GOS responders had a significantly lower concentration of IL-5 and IL-10 following PHA stimulation of PBMCs (Table 4-5) as well as a significantly lower concentration of IL-10 following LPS stimulation of PBMCs (Table 4-6). This implies that GOS may be modifying the response to an increase in bifidobacteria or, perhaps, these significant differences observed between placebo responders and GOS responders might be purely coincidental. These results suggest that bifidobacteria are probably not the only players exerting their beneficial effects and that maybe other bacterial species are also altering the microflora-dependent effects and this was not captured in our study. Conceivably, if we would have examined different bacterial species that have also changed following GOS supplementation, we could have subdivided the groups using fold changes in these bacterial species and differences in immune parameters could have been observed. A change in bifidobacteria genome equivalents seems to be a piece of the puzzle but not the whole story by itself.

Administration of prebiotics, such as oligofructose, was shown to increase lactate and

119 acetate production following fermentation by lactobacilli and bifidobacteria but in turn, these SCFA can be subsequently degraded by other bacteria such as Anaerostipes caccae or Roseburia intestinalis, both known butyrate producers (239).

What we might be observing is a chain of events whereby GOS → microbiota → bifidogenic effect → lactate produced by bifidobacteria is normally utilized by other bacterial species → the bacteria utilizing lactate are butyrate producers → butyrate increases the barrier integrity of the gut → tight junctions decrease translocation of bacteria → lessening LPS systemic load. The increase in butyrate producing bacteria following GOS consumption may also confer growth inhibition of pathogens in the upper respiratory tract. Stimulation of T and B lymphocytes in the GALT by immunostimulatory bacteria will cause migration of these immune cells to related mucosal-effector sites such as the trachea and bronchi and lessen the risk of bacterial infection in the lungs.

GOS and Digestive Health

We investigated the effect of GOS on GI symptoms because the use of prebiotics has previously been associated with GI discomfort, more specifically; gas production in the gut has often been reported in human prebiotic feeding studies (240). Production of gases such as CO2 and hydrogen is inevitable following fermentation of certain prebiotics and it is one of the major disincentives for consumption of these non- digestible carbohydrates. In this present study, all participants, regardless of their treatment group, were not bothered by GI symptoms. Participants consuming GOS had significantly lower symptoms of constipation and abdominal pain than participants consuming the placebo (Figure 4-5). These results are consistent with findings from a previous study where healthy undergraduate students undergoing academic stress 120 receiving either 2.5 or 5.0 g of GOS had significantly lower GI symptom scores for all symptom categories (i.e. diarrhea syndrome, constipation syndrome, abdominal syndrome, and indigestion) except reflux syndrome than students receiving the placebo

(4). A possible explanation for the reduced abdominal pain syndrome observed with

GOS supplementation may be due to their beneficial effect on the microbiota. Previous studies have shown that probiotics can relieve abdominal symptoms related to abnormal colonic transit and motility. A study conducted in mice and rats demonstrated that L. acidophilus NCFM upregulated μ-opiod and cannabinoid receptors on intestinal cells which conferred analgesic effects through the upregulation of the enteric nervous system (241). The modulation of the intestinal microbiota, whether it be through the administration of probiotics or prebiotics, seems to contribute to visceral sensitivity and gut-brain interactions. A study conducted by Diop and colleagues demonstrated that administration of L. acidophilus R-52 and B. longum R-175 significantly reduced 2 stress-induced GI symptoms; abdominal pain and nausea/vomiting (242). This supports the hypothesis that probiotics or modulation of the gut microbiome by prebiotics such as

GOS can modulate the gut-brain axis and improve gastrointestinal symptoms and visceral sensitivity.

Another hypothesis as to how GOS may be indirectly modulating the immune system could be through increased barrier function of the gut protecting the host against inflammation and infection. Although the tight junctions located between the epithelial cells of the GI tract were not specifically observed in our study, it is known that butyrate production in addition to providing fuel for the colonocytes also suppresses expression of the transcription factor NK-κB and the consequent production of pro-inflammatory

121 cytokines (243). This could explain why participants in the GOS group that saw at least a two-fold increase in their bifidobacteria genome equivalents had a significantly lower concentration of IL-5 and IL-8 following stimulation of peripheral blood mononuclear cells with PHA and LPS respectively (Table 4-5 and 4-6). Two ways by which prebiotics were shown to modulate the immune system are by increasing SCFA production and increasing the numbers of immunogenic bacteria such as bifidobacteria and lactobacilli

(244).

Preventing Attachment of Pathogenic Bacteria

Additionally, GOS could be exhibiting their immunomodulatory potential by protecting the gut from infection through the inhibition of pathogenic bacteria attachment to the colonic epithelium. On epithelial cells, glycoconjugates on glycoproteins and lipids permit attachment of bacteria to the microvillus membrane.

Galactooligosaccharides have similar structures to those glycoconjugates found on the epithelium and because they are capable of binding to bacterial receptors, they prevent bacterial attachment to the GI tract. This could prevent the translocation of pathogenic bacteria to other organs (244) and may explain how GOS is positively altering the immune status of the host.

How GOS through its Influence on Proliferation of Beneficial Bacteria within the Colon May Affect URTI

Since consumption of prebiotics can ultimately lead to an increase in bifidobacteria and these latter can interact with the immune system, it is of interest to look at the effects of these bacteria and how, synergistically with GOS, they might decrease the incidence of URTI. Winkler (245) and colleagues and de Vrese (246) and colleagues looked at mechanisms by which probiotics may reduce the occurrence or the duration of

122 respiratory tract infections and found that probiotic supplementation increased total cell counts per microliter of blood (when looking at the change between study day 14 and study day 0) of leukocytes, lymphocytes, CD4+ Th cells, cytotoxic CD8+ T cells and monocytes. However, when looking at the results from our study and executing either a t-test or a Mann-Whitney Rank Sum test when data were not normally distributed, no such results were found (Table D-1). The total lymphocytes, CD4+ Th cells, cytotoxic T cells, B cells, and NK cells were not different when looking at the change between baseline and three weeks post-supplementation as well as baseline and five weeks post-supplementation.

A study by Gluck et al. conducted in healthy adults demonstrated that a combination of probiotics decreased the nasal colonization of pathogenic bacteria and this was thought to be due to stimulation of B cells and the production of antibody in the

GALT and subsequent lymphocyte migration to the upper respiratory tract immune system (247). In this present study, concentration of sIgA in the saliva of participants was investigated and no difference between the two treatment groups was observed

(Table 4-7). Perhaps measuring salivary sIgA concentrations on days of self-reported colds, rather than at three different time points at the beginning of the study (baseline, at vaccination and 2 weeks post-vaccination) would have yielded different results.

In a rodent study by Yasui and colleagues, oral administration of probiotics reduced the presence of viral titers in the nasal washing of mice that were infected with an influenza virus suggesting that probiotics induced the activation of innate immunity in the lungs (248). The investigators hypothesized that this effect was through the observed enhancement of pulmonary NK-cell activity in mice receiving the probiotic.

123

Our results suggest a possible decrease in the inflammatory process in aged adults after supplementation with GOS although this decrease in general inflammation was not observed when looking at systemic CRP concentration levels (Table 4-7). A decrease in inflammation, more specifically the reduction in IL-6 mRNA in peripheral blood monocytes, with the use of FOS was previously observed (249). This study conducted with an elderly population showed that by modulating the microbiota and increasing bifidobacteria counts in the feces; a decrease in the inflammatory process was observed.

A surprising outcome following the microbiota analysis was that a two-fold increase in bifidobacteria genome equivalents was also observed in the placebo group.

There were nearly as many responders in the placebo group (53%) than the GOS

(60%) treatment group. A study by Alles et al. demonstrated the importance of a placebo and the influence of basal bifidobacteria counts on the response to non- digestible carbohydrate consumption (250). These investigators conducted a study utilizing the prebiotic TOS and the results showed no significant increase in bifidobacteria in the feces when compared to a placebo where both, the placebo and prebiotic, caused an increase in bifidobacteria. These results, although surprising, are not impossible as many factors can influence the microbiota makeup. The diet, which was not controlled for in our study, could have played a role in this bifidobacteria increase although no differences in fiber intake were captured between the two treatment groups when a food frequency questionnaire was administered at the end of the study. This could have been a plausible explanation elucidating the increase in bifidobacteria in the placebo group. Perhaps participating in a nutritional study itself

124 could have prompted participants to follow a “healthier lifestyle” and although study coordinators urged participants not to change usual dietary habits, this could have occurred.

Davis and colleagues demonstrated that the initial levels of bifidobacteria between the responders and the non-responders were not different and that even when GOS was administered at high doses (up to 10 g a day) for many weeks, a bifidogenic response did not occur in certain individuals (251). This is supported by a crossover study by Depeint et al. where it was demonstrated that two GOS mixtures had different effects on the microbiota (252). It was reported that, in healthy humans, one week consumption of a GOS mixture containing mainly β1→3, as well as β1→4 and β1→6 linkages had a more bifidogenic effect than consumption of a GOS mixture containing mainly a β1→4, as well as β1→6. This suggests that in addition to the purity of the prebiotic, the types of linkages within a prebiotic, which can vary due to the difference in enzymes sources, can also affect its utilization by the microbiota. This might explain why some participants in the GOS group did not show a two-fold increase, sometimes exhibiting a decrease, in fecal bifidobacteria genome equivalents.

Limitations of the Study

A limitation of this study is that no NK cell activity measurements were made.

Since it was suspected that GOS exerted its immunomodulatory effects through NK cells and that Vulevic et al. showed increased NK cell activity with GOS supplementation (5), it would have been interesting to see if NK cell activity was higher in the GOS group in our aged adult population. As mentioned previously, although aging is associated with an increase in NK cell numbers, these leukocytes seem to

125 exhibit lower activity levels. Additionally, when measuring relative proportions as well as total numbers of NK cells, we specifically looked at CD45+/CD3-/CD19-/CD16-56+ which are the non-thymic NK cells. It would be relevant to also measure the dually labeled CD3/CD56 (i.e. NK T cells) which might have been partly responsible for the observed effects.

When categorizing responders and non-responders, we decided on an arbitrary twofold increase, which was an arbitrary delineation and if we would have considered a threefold change instead, perhaps we would have obtained different results. In fact, this twofold change was chosen because of the values of the replicate which were rarely exactly the same, so a cut-off of acceptable variance was selected. Another possible explanation as to why we are not observing similar results to comparable GOS studies is that this was not a crossover study. A crossover study could have accounted for inter-individual differences as each participant would have acted as his/her own control.

Finally, Yasui and colleagues conducted a study in mice where, following the supplementation with a probiotic (Bifidobacterium breve YIT4064), they saw an increase in serum anti-influenza virus IgG antibodies to oral influenza vaccination (253). Since we did administer the influenza vaccine to all of our participants, it would have been interesting to look at the effects of GOS on the antibody response to the influenza vaccine by looking at antibody titers.

Conclusions

Galactooligosaccharides had more of an effect at the “extremes” of age regarding the probability of self-reporting a cold. In this Older Aged population (65 years of age); participants receiving GOS had a lower probability of self-reporting a cold when

126 compared to Younger Aged participants in the same treatment group and IFN-γ concentrations were also higher with GOS for Older Aged participants when compared to the placebo in the same age group. In regards to NK cells, participants receiving

GOS in the Older Aged group had a higher percentage of NK cells as a percentage of lymphocytes than Younger Aged participants in the GOS group.

Galactooligosaccharides supplementation had positive effects on digestive health by reducing certain symptoms of GI dysfunction. Supplementation with GOS did not have a significant effect on cold symptoms or quality of life. Additional studies are needed where the effect of GOS on different strains of bacteria should be investigated as well as functional NK cell assays. These results are very promising in terms of providing a safe alternative for the improvement of cellular immunity in aged adults.

127

APPENDIX A IRB APPROVAL LETTER

128

129

APPENDIX B IRB INFORMED CONSENT

130

131

132

133

134

135

136

137

138

139

140

141

APPENDIX C QUESTIONNAIRES

142

143

APPENDIX D DESCRIPTIVE STATISTICS OF PLACEBO AND GALACTOOLIGOSACCHARIDES BY AGE GROUP

144

Table D-1. Descriptive statistics of placebo and galactooligosaccharides supplementation by age group Younger Aged (60 to 64 y) Older Aged (65 y or older) Placebo GOS Placebo GOS P value n 22 22 16 21 0.656 Gender Male 11 8 4 8 0.474 Female 11 14 12 13 T: 0.259 # Cold days SI>6 1.8 ± 0.6 1.6 ± 0.4 2.1 ± 0.9 0.9 ± 0.5 A: 0.787 I: 0.402 T: 0.259 % Days when SI>6 1.1 ± 0.4 1.0 ± 0.3 1.3 ± 0.6 0.6 ± 0.3 A: 0.787 I: 0.402 T: 0.264 MNA Initial Total 27.3 ± 0.5 27.3 ± 0.5 26.2 ± 0.6 27.3 ± 0.5 A: 0.277 Score I: 0.264 T: 0.911 % NK Draw 1 15.1 ± 1.5 12.2 ± 1.5 14.5 ± 1.7 17.7 ± 1.5 A: 0.121 I: 0.055 T: 0.510 % NK Draw 2 14.9 ± 1.4ab 10.6 ± 1.4a 13.9 ± 1.6a 16.3 ± 1.4b A: 0.110 I: 0.022 T: 0.768 % NK Draw 3 15.1 ± 1.5a,b 12.1 ± 1.1a,b 14.0 ± 1.9a 17.9 ± 1.6b A: 0.129 I: 0.028 T: 0.398 % Cytotoxic T cells 21.5 ± 2.1 22.3 ± 2.1 20.2 ± 2.4 23.0 ± 2.1 A: 0.893 Draw 1 I: 0.653

145

Table D-1. Continued Younger Aged (60 to 64 y) Older Aged (65 y or older) Placebo GOS Placebo GOS P value T: 0.591 % Cytotoxic T cells 21.8 ± 1.9 22.3 ± 1.9 20.1 ± 2.3 21.8 ± 2.0 A: 0.604 Draw 2 I: 0.771 T: 0.865 % Cytotoxic T cells 21.9 ± 1.9 20.4 ± 1.9 19.5 ± 2.2 21.6 ± 1.9 A: 0.770 Draw 3 I: 0.377 T: 0.011 Salivary sIgA Draw 1 (μg/min) A: 0.966 486 ± 111a 763 ± 108b 463 ± 133a 796 ± 111b Mean ± SEM I: 0.814 Median (25th, 75th) 495 (235, 697) 718 (329, 1219) 387 (224, 621) 590 (302, 1029)

Salivary sIgA Draw 2 T: 0.137 (μg/min) 529 ± 116 717 ± 113 583 ± 139 760 ± 116 A: 0.690 Mean ± SEM 507 (177, 754) 570 (305, 850) 428 (234, 989) 522 (331, 923) I: 0.961 Median (25th, 75th) Salivary sIgA Draw 3 T: 0.180 (μg/min) 525 ± 162 726 ± 162 526 ± 193 786 ± 162 A: 0.859 Mean ± SEM I: 0.862 Median (25th, 75th) 576 (369, 674) 490 (173, 862) 343 (228, 786) 494 (309, 1120) T: 0.315 Fecal sIgA Baseline (ng/mg) 4567 ± 776 3031 ± 776 2397 ± 918 2268 ± 815 A: 0.079 I: 0.395 T: 0.465 Fecal sIgA Post- treatment (ng/mg) 3854 ± 666 3240 ± 666 3027 ± 788 2602 ± 700 A: 0.304 I: 0.893 T: 0.570 CRP Draw 1 (mg/L) 3.0 ± 0.8 2.5 ± 0.9 4.0 ± 1.1 3.4 ± 0.9 A: 0.326 I: 0.975 T: 0.793 CRP Draw 2 (mg/L) 2.4 ± 1.2 2.6 ± 1.3 5.3 ±1.4 4.4 ± 1.3 A: 0.081 I: 0.662

146

Table D-1. Continued Younger Aged (60 to 64 y) Older Aged (65 y or older) Placebo GOS Placebo GOS P value T: 0.695 CRP Draw 3 (mg/L) 2.5 ± 0.9 2.7 ± 1.0 4.1 ± 1.1 3.1 ± 1.0 A: 0.328 I: 0.576 PHA Stimulated

PBMC T: 0.209 IL-2 Draw 1 (pg/mL) 716 ± 134 606 ± 134 618 ± 163 367 ± 137 A: 0.241 I: 0.624 T: 0.294 IL-2 Draw 2 (pg/mL) 381 ± 65a 318 ± 65a 254 ± 79b 171 ± 66b A: 0.050 I: 0.881 T: 0.467 IL-2 Draw 3 (pg/mL) 354 ± 62a 306 ± 64a 207 ± 75b 158 ± 64b A: 0.030 I: 0.991 T: 0.152 IL-5 Draw 1 (pg/mL) 199 ± 43 104 ± 43 161 ± 51 125 ± 44 A: 0.857 I: 0.522 T: 0.077 IL-5 Draw 2 (pg/mL) 168 ± 35 83 ± 35 145 ± 41 98 ± 36 A: 0.915 I: 0.604 T: 0.180 IL-5 Draw 3 (pg/mL) 141 ± 31 81 ± 31 142 ± 36 115 ± 31 A: 0.586 I: 0.607 T: 0.941 IL-10 Draw 1 199 ± 31 230 ± 31 214 ± 36 187 ± 31 A: 0.665 (pg/mL) I: 0.378 T: 0.875 IL-10 Draw 2 223 ± 28 215 ± 28 194 ± 32 194 ± 28 A: 0.387 (pg/mL) I: 0.886

147

Table D-1. Continued Younger Aged (60 to 64 y) Older Aged (65 y or older) Placebo GOS Placebo GOS P value T: 0.455 IL-10 Draw 3 245 ± 39 276 ± 39 201 ± 45 231 ± 39 A: 0.278 (pg/mL) I: 0.993 T: 0.999 IFN-γ Draw 1 15751 ± 1887a 11224 ± 1887ab 9404 ± 2233b 13938 ± 1887ab A: 0.362 (pg/mL) I: 0.025 T: 0.425 IFN-γ Draw 2 12007 ± 1557a 10892 ± 1557a 6044 ± 1842b 9780 ± 1557b A: 0.033 (pg/mL) I: 0.142 T: 0.740 IFN-γ Draw 3 16673 ± 2213a 11958 ± 2213ab 6026 ± 2556b 12268 ± 2160ab A: 0.027 (pg/mL) I: 0.019 LPS Stimulated

PBMC T: 0.138 IL-1β Draw 1 3565 ± 353 2743 ± 336 3061 ± 394 2809 ± 344 A: 0.542 (pg/mL) I: 0.428 T: 0.023 IL-1β Draw 2 4382 ± 424a 3009 ± 404b 3349 ± 474a 2727 ± 414b A: 0.130 (pg/mL) I: 0.385 T: 0.175 IL-1β Draw 3 3982 ± 370 2941 ± 362 2946 ± 414 2952 ± 362 A: 0.178 (pg/mL) I: 0.170 T: 0.220 IL-6 Draw 1 27943 ± 3144 22496 ± 3144 28957 ± 3687 26226 ± 3218 A: 0.475 (pg/mL) I: 0.682 T: 0.038 IL-6 Draw 2 27258 ± 2591a 21643 ± 2591b 29775 ± 3038a 23912 ± 2652b A: 382 (pg/mL) I: 0.964

148

Table D-1. Continued Younger Aged (60 to 64 y) Older Aged (65 y or older) Placebo GOS Placebo GOS P value T: 0.274 IL-6 Draw 3 29054 ± 3088 24886 ± 3160 29567 ± 3621 26546 ± 3160 A: 0.740 (pg/mL) I: 0.861 T: 0.392 IL-8 Draw 1 177378 ± 24331 126191 ± 25519 170339 ± 29466 176022 ± 26182 A: 0.421 (pg/mL) I: 0.286 T: 0.178 IL-8 Draw 2 195799 ± 21250 125553 ± 22287 156728 ± 25735 164094 ± 22866 A: 0.991 (pg/mL) I: 0.097 T: 0.882 IL-8 Draw 3 157599 ± 20562 143279 ± 22126 165808 ± 24902 173457 ± 22126 A: 0.396 (pg/mL) I: 0.627 T: 0.564 IL-10 Draw 1 247 ± 39 239 ± 39 337 ± 46 298 ± 40 A: 0.071 (pg/mL) I: 0.716 T: 0.063 IL-10 Draw 2 265 ± 30a 207 ± 30a 335 ± 36b 272 ± 31b A: 0.038 (pg/mL) I: 0.929 T: 0.810 IL-10 Draw 3 270 ± 57 299 ± 58 405 ± 66 347 ± 58 A: 0.132 (pg/mL) I: 0.466 T: 0.007 TNF-α Draw 1 4188 ± 482a 2401 ± 482b 3794 ± 565a 2794 ± 493b A: 0.999 (pg/mL) I:0.440 T: 0.001 TNF-α Draw 2 3367 ± 395a 2050 ± 395b 2797 ± 463a 1368 ± 404b A: 0.136 (pg/mL) I: 0.893 T: 0.029 TNF-α Draw 3 3426 ± 464a 2306 ± 475b 2908 ± 544a 1841 ± 475b A: 0.320 (pg/mL) I:0.958

149

LIST OF REFERENCES

1. Nunes K. An ounce of prevention. Products promoting immune system support are gaining traction. Food Business News: Sosland Publishing Co.; 2010.

2. Brownawell AM, Caers W, Gibson GR, Kendall CW, Lewis KD, Ringel Y, Slavin JL. Prebiotics and the health benefits of fiber: current regulatory status, future research, and goals. J Nutr. 2012;142:962-74.

3. Macfarlane G, Steed H, Macfarlane S. Bacterial metabolism and health-related effects of galacto-oligosaccharides and other prebiotics. J Appl Microbiol. 2008;104:305-44.

4. Hughes C, Davoodi-Semiromi Y, Colee JC, Culpepper T, Dahl WJ, Mai V, Christman MC, Langkamp-Henken B. Galactooligosaccharide supplementation reduces stress-induced gastrointestinal dysfunction and days of cold or flu: a randomized, double-blind, controlled trial in healthy university students. Am J Clin Nutr. 2011;93:1305-11.

5. Vulevic J, Drakoularakou A, Yaqoob P, Tzortzis G, Gibson G. Modulation of the fecal microflora profile and immune function by a novel trans- galactooligosaccharide mixture (B-GOS) in healthy elderly volunteers. Am J Clin Nutr. 2008;88:1438-46.

6. Macfarlane S, Macfarlane G, Cummings J. Review article: prebiotics in the gastrointestinal tract. Aliment Pharmacol Ther. 2006;24:701-14.

7. Lamsal BP. Production, health aspects and potential food uses of dairy prebiotic galactooligosaccharides. J Sci Food Agric. 2012.

8. Nutrition G. Galactooligosaccharide GRAS Notice. Golden; 2009.

9. Angus F, Smart S, Shortt C. Prebiotic ingredients with emphasis on galactooligosaccharides and fructo-oligosaccharides: Oxford: Blackwell Publishing; 2005.

10. Torres DPM, Goncalves MPF, Teixeira JA, Rodrigues LR. Galacto- Oligosaccharides: Production, Properties, Applications, and Significance as Prebiotics. Comprehensice Reviews in Food Science and Food Safety. 2010;9:438-54.

11. Charalampopoulos D, Rastall RA. Prebiotics and Probiotics: Springer; 2009.

12. EUROPA. European Union Register of nutrition and Health claims made on food:rejected health claims;2010. 2010 [cited 2012 05-03]; Available from: http://ec.europa.eu/nuhclaims/?event=search&status_ref_id=5

150

13. Drakoularakou A, Tzortzis G, Rastall RA, Gibson GR. A double-blind, placebo- controlled, randomized human study assessing the capacity of a novel galactooligosaccharide mixture in reducing travellers' diarrhoea. Eur J Clin Nutr. 2010;64:146-52.

14. Moro E. Morphologische und bakterogische untersuchungen uber die Dambakterien des Sauglings: die bacterium flora des normalen frauenmilch stuhls. Jahb Kinderheilkd. 1900:686-734.

15. Harmsen HJ, Wildeboer-Veloo AC, Raangs GC, Wagendorp AA, Klijn N, Bindels JG, Welling GW. Analysis of intestinal flora development in breast-fed and formula-fed infants by using molecular identification and detection methods. J Pediatr Gastroenterol Nutr. 2000;30:61-7.

16. Boehm G, Stahl B, Jelinek J, Knol J, Miniello V, Moro GE. Prebiotic carbohydrates in human milk and formulas. Acta Paediatr Suppl. 2005;94:18-21.

17. Haarman M, Knol J. Quantitative real-time PCR assays to identify and quantify fecal Bifidobacterium species in infants receiving a prebiotic infant formula. Appl Environ Microbiol. 2005;71:2318-24.

18. Ito M, Deguchi Y, Miyamori A, Matsumote K, Kikuchi H, Matsumoto K, Yajima T, Kan T. Effects of administration of galactooligosaccharides on the human faecal microflora, stool weigh and abdominal sensation. Microb Ecol Health Dis. 1990;3:285-92.

19. Ito M, Deguchi Y, Matsumoto K, Kimura M, Onodera N, Yajima T. Influence of galactooligosaccharides on the human fecal microflora. J Nutr Sci Vitaminol (Tokyo). 1993;39:635-40.

20. Malinen E, Mättö J, Salmitie M, Alander M, Saarela M, Palva A. PCR-ELISA II: Analysis of Bifidobacterium populations in human faecal samples from a consumption trial with Bifidobacterium lactis Bb-12 and a galacto-oligosaccharide preparation. Syst Appl Microbiol. 2002;25:249-58.

21. Bouhnik Y, Flourié B, D'Agay-Abensour L, Pochart P, Gramet G, Durand M, Rambaud JC. Administration of transgalacto-oligosaccharides increases fecal bifidobacteria and modifies colonic fermentation metabolism in healthy humans. J Nutr. 1997;127:444-8.

22. Chonan O, Shigehara-Sone H, Takahashi R, Ikeda M, Kikuchi-Hayakawa H, Ishikawa F, Kimura K, Matsumoto K. Undigestibility of Galactooligosaccharides. Journal of the Japanese Society for Food Science and Technology. 2004;51:28- 33.

23. Revicki DA, Wood M, Wiklund I, Crawley J. Reliability and validity of the Gastrointestinal Symptom Rating Scale in patients with gastroesophageal reflux disease. Qual Life Res. 1998;7:75-83.

151

24. Deguchi Y, Matsumoto K, Ito A, Watanuki M. Effects of b1-4 galactooligosaccharides administration on defecation of healthy volunteers with constipation tendency. Jpn J Nutr. 1997;55:13-22.

25. Teuri U, Korpela R. Galacto-oligosaccharides relieve constipation in elderly people. Ann Nutr Metab. 1998;42:319-27.

26. Bronner F. Calcium absorption--a paradigm for mineral absorption. J Nutr. 1998;128:917-20.

27. Chonan O, Matsumoto K, Watanuki M. Effect of galactooligosaccharides on calcium absorption and preventing bone loss in ovariectomized rats. Biosci Biotechnol Biochem. 1995;59:236-9.

28. Chonan O, Watanuki M. Effect of galactooligosaccharides on calcium absorption in rats. J Nutr Sci Vitaminol (Tokyo). 1995;41:95-104.

29. Chonan O, Watanuki M. The effect of 6'-galactooligosaccharides on bone mineralization of rats adapted to different levels of dietary calcium. Int J Vitam Nutr Res. 1996;66:244-9.

30. Chonan O, Takahashi R, Watanuki M. Role of activity of gastrointestinal microflora in absorption of calcium and magnesium in rats fed beta1-4 linked galactooligosaccharides. Biosci Biotechnol Biochem. 2001;65:1872-5.

31. van den Heuvel EG, Schaafsma G, Muys T, van Dokkum W. Nondigestible oligosaccharides do not interfere with calcium and nonheme-iron absorption in young, healthy men. Am J Clin Nutr. 1998;67:445-51.

32. van den Heuvel E, Schoterman M, Muijs T. Transgalactooligosaccharides stimulate calcium absorption in postmenopausal women. J Nutr. 2000;130:2938- 42.

33. Demigne C, Remesy C, Morand C. Short Chain Fatty Acids. In: G.R. G, M.B. R, editors. Colonic Microbiota, Nutrition and Health. The Netherlands: Springer; 1999. p. 312.

34. Delzenne NM, Cani PD, Neyrinck AM. Prebiotics and Lipid Metabolism. Therapeutic Microbiology: Probiotics and Related Strategies: ASM Press; 2008. p. 183-92.

35. Fiordaliso M, Kok N, Desager JP, Goethals F, Deboyser D, Roberfroid M, Delzenne N. Dietary oligofructose lowers triglycerides, phospholipids and cholesterol in serum and very low density lipoproteins of rats. Lipids. 1995;30:163- 7.

152

36. van Dokkum W, Wezendonk B, Srikumar TS, van den Heuvel EG. Effect of nondigestible oligosaccharides on large-bowel functions, blood lipid concentrations and glucose absorption in young healthy male subjects. Eur J Clin Nutr. 1999;53:1-7.

37. Schley PD, Field CJ. The immune-enhancing effects of dietary fibres and prebiotics. Br J Nutr. 2002;87 Suppl 2:S221-30.

38. Langkamp-Henken B, Glezer JA, Kudsk KA. Immunologic structure and function of the gastrointestinal tract. Nutr Clin Pract. 1992;7:100-8.

39. Gropper SAS, Smith JL, Groff JL. Advanced nutrition and human metabolism. 5th ed. Australia ; United States: Wadsworth/Cengage Learning; 2009.

40. Kruh J, Defer N, Tichonicky L. Effects of butyrate on cell proliferation and gene expression. In: J.H. C, J.L. R, T. S, editors. Physiological and Clinical Aspects of Short-Chain Fatty Acids. Cambridge: Cambridge University Press; 1995. p. 275- 88.

41. Cavaglieri CR, Nishiyama A, Fernandes LC, Curi R, Miles EA, Calder PC. Differential effects of short-chain fatty acids on proliferation and production of pro- and anti-inflammatory cytokines by cultured lymphocytes. Life Sci. 2003;73:1683- 90.

42. Nam KT, O'Neal R, Lee YS, Lee YC, Coffey RJ, Goldenring JR. Gastric tumor development in Smad3-deficient mice initiates from forestomach/glandular transition zone along the lesser curvature. Lab Invest. 2012.

43. Gopalakrishnan A, Clinthorne JF, Rondini EA, McCaskey SJ, Gurzell EA, Langohr IM, Gardner EM, Fenton JI. Supplementation with Galacto-Oligosaccharides Increases the Percentage of NK Cells and Reduces Colitis Severity in Smad3- Deficient Mice. J Nutr. 2012.

44. Shoaf K, Mulvey GL, Armstrong GD, Hutkins RW. Prebiotic galactooligosaccharides reduce adherence of enteropathogenic Escherichia coli to tissue culture cells. Infect Immun. 2006;74:6920-8.

45. Scheppach W, Weiler F. The butyrate story: old wine in new bottles? Curr Opin Clin Nutr Metab Care. 2004;7:563-7.

46. Hernot DC, Boileau TW, Bauer LL, Middelbos IS, Murphy MR, Swanson KS, Fahey GC. In vitro fermentation profiles, gas production rates, and microbiota modulation as affected by certain fructans, galactooligosaccharides, and polydextrose. J Agric Food Chem. 2009;57:1354-61.

47. Nurmi JT, Puolakkainen PA, Rautonen NE. Bifidobacterium Lactis sp. 420 up- regulates cyclooxygenase (Cox)-1 and down-regulates Cox-2 gene expression in a Caco-2 cell culture model. Nutr Cancer. 2005;51:83-92.

153

48. Fund WCR. Food, Nutrition and the prevention of cancer: a global perspective. Wahington, DC: World Cancer research Fund; 1997.

49. Fuchs CS, Giovannucci EL, Colditz GA, Hunter DJ, Stampfer MJ, Rosner B, Speizer FE, Willett WC. Dietary fiber and the risk of colorectal cancer and adenoma in women. N Engl J Med. 1999;340:169-76.

50. Boutron-Ruault MC, Marteau P, Lavergne-Slove A, Myara A, Gerhardt MF, Franchisseur C, Bornet F, Group ES. Effects of a 3-mo consumption of short-chain fructo-oligosaccharides on parameters of colorectal carcinogenesis in patients with or without small or large colorectal adenomas. Nutr Cancer. 2005;53:160-8.

51. Rafter J, Bennett M, Caderni G, Clune Y, Hughes R, Karlsson PC, Klinder A, O'Riordan M, O'Sullivan GC, et al. Dietary synbiotics reduce cancer risk factors in polypectomized and colon cancer patients. Am J Clin Nutr. 2007;85:488-96.

52. Limburg PJ, Mahoney MR, Ziegler KL, Sontag SJ, Schoen RE, Benya R, Lawson MJ, Weinberg DS, Stoffel E, et al. Randomized phase II trial of sulindac, atorvastatin, and prebiotic dietary fiber for colorectal cancer chemoprevention. Cancer Prev Res (Phila). 2011;4:259-69.

53. Heikkinen T, Järvinen A. The common cold. Lancet. 2003;361:51-9.

54. Adams PF, Hendershot GE, Marano MA, Statistics CfDCaPNCfH. Current estimates from the National Health Interview Survey, 1996. Vital Health Stat 10. 1999:1-203.

55. Monto AS. Epidemiology of viral respiratory infections. Am J Med. 2002;112 Suppl 6A:4S-12S.

56. Diseases NIoAaI. The Common Cold Overview. Health & Research A to Z [cited 2012 05-07]; Available from: http://www.niaid.nih.gov/topics/commonCold/Pages/overview.aspx

57. Arroll B. Common cold. Clin Evid (Online). 2011;2011.

58. Shah SA, Sander S, White CM, Rinaldi M, Coleman CI. Evaluation of echinacea for the prevention and treatment of the common cold: a meta-analysis. Lancet Infect Dis. 2007;7:473-80.

59. Monto AS, Sullivan KM. Acute respiratory illness in the community. Frequency of illness and the agents involved. Epidemiol Infect. 1993;110:145-60.

60. Gonzales R, Malone DC, Maselli JH, Sande MA. Excessive antibiotic use for acute respiratory infections in the United States. Clin Infect Dis. 2001;33:757-62.

61. National Health Care Quality Report 2008. 2008:77-85.

154

62. Immunization and Infectious Diseases. Healthy People 2020 Summary of Objective 2012 [cited 2012 March 15th]; Available from: http://healthypeople.gov/2020/topicsobjectives2020/objectiveslist.aspx?topicId=23

63. Prior L, Evans MR, Prout H. Talking about colds and flu: the lay diagnosis of two common illnesses among older British people. Soc Sci Med. 2011;73:922-8.

64. Barrett B, Brown RL, Mundt MP, Thomas GR, Barlow SK, Highstrom AD, Bahrainian M. Validation of a short form Wisconsin Upper Respiratory Symptom Survey (WURSS-21). Health Qual Life Outcomes. 2009;7:76.

65. Barrett B, Brown R, Mundt M, Safdar N, Dye L, Maberry R, Alt J. The Wisconsin Upper Respiratory Symptom Survey is responsive, reliable, and valid. J Clin Epidemiol. 2005;58:609-17.

66. Jackson GG, Dowling HF, Spiesman IG, Boand AV. Transmission of the common cold to volunteers under controlled conditions. I. The common cold as a clinical entity. AMA Arch Intern Med. 1958;101:267-78.

67. Barrett B, Brown R, Voland R, Maberry R, Turner R. Relations among questionnaire and laboratory measures of rhinovirus infection. Eur Respir J. 2006;28:358-63.

68. Kirscht JP. Social and psychological problems of surveys on health and illness. Soc Sci Med. 1971;5:519-26.

69. Takkouche B, Regueira C, Gestal-Otero JJ. A cohort study of stress and the common cold. Epidemiology. 2001;12:345-9.

70. Turner RB, Bauer R, Woelkart K, Hulsey TC, Gangemi JD. An evaluation of Echinacea angustifolia in experimental rhinovirus infections. N Engl J Med. 2005;353:341-8.

71. Cohen S, Tyrrell DA, Smith AP. Psychological stress and susceptibility to the common cold. N Engl J Med. 1991;325:606-12.

72. Macintyre S, Pritchard C. Comparisons between the self-assessed and observer- assessed presence and severity of colds. Soc Sci Med. 1989;29:1243-8.

73. Albers R, Antoine JM, Bourdet-Sicard R, Calder PC, Gleeson M, Lesourd B, Samartín S, Sanderson IR, Van Loo J, et al. Markers to measure immunomodulation in human nutrition intervention studies. Br J Nutr. 2005;94:452- 81.

74. Beisel WR, Edelman R, Nauss K, Suskind RM. Single-nutrient effects on immunologic functions. Report of a workshop sponsored by the Department of Food and Nutrition and its nutrition advisory group of the American Medical Association. JAMA. 1981;245:53-8.

155

75. Scrimshaw NS, SanGiovanni JP. Synergism of nutrition, infection, and immunity: an overview. Am J Clin Nutr. 1997;66:464S-77S.

76. Percival SS. Neutropenia caused by copper deficiency: possible mechanisms of action. Nutr Rev. 1995;53:59-66.

77. Babu U, Failla ML. Copper status and function of neutrophils are reversibly depressed in marginally and severely copper-deficient rats. J Nutr. 1990;120:1700-9.

78. Yetgin S, Altay C, Ciliv G, Laleli Y. Myeloperoxidase activity and bactericidal function of PMN in iron deficiency. Acta Haematol. 1979;61:10-4.

79. Ross AC. Vitamin A deficiency and retinoid repletion regulate the antibody response to bacterial antigens and the maintenance of natural killer cells. Clin Immunol Immunopathol. 1996;80:S63-72.

80. Adachi N, Migita M, Ohta T, Higashi A, Matsuda I. Depressed natural killer cell activity due to decreased natural killer cell population in a vitamin E-deficient patient with Shwachman syndrome: reversible natural killer cell abnormality by alpha-tocopherol supplementation. Eur J Pediatr. 1997;156:444-8.

81. Lesourd BM. [Immunologic aging. Effect of denutrition]. Ann Biol Clin (Paris). 1990;48:309-18.

82. Lesourd BM. Nutrition and immunity in the elderly: modification of immune responses with nutritional treatments. Am J Clin Nutr. 1997;66:478S-84S.

83. Guigoz Y. The Mini Nutritional Assessment (MNA) review of the literature--What does it tell us? J Nutr Health Aging. 2006;10:466-85; discussion 85-7.

84. Hudgens J, Langkamp-Henken B, Stechmiller JK, Herrlinger-Garcia KA, Nieves C. Immune function is impaired with a mini nutritional assessment score indicative of malnutrition in nursing home elders with pressure ulcers. JPEN J Parenter Enteral Nutr. 2004;28:416-22.

85. Cunningham-Rundles S, Lin DH. Nutrition and the immune system of the gut. Nutrition. 1998;14:573-9.

86. Gross RL, Newberne PM. Role of nutrition in immunologic function. Physiol Rev. 1980;60:188-302.

87. López-Varela S, González-Gross M, Marcos A. Functional foods and the immune system: a review. Eur J Clin Nutr. 2002;56 Suppl 3:S29-33.

88. Roberfroid MB. Prebiotics and probiotics: are they functional foods? Am J Clin Nutr. 2000;71:1682S-7S; discussion 8S-90S.

156

89. D'Agostino RB, Weintraub M, Russell HK, Stepanians M, Cantilena LR, Graumlich JF, Maldonado S, Honig P, Anello C. The effectiveness of antihistamines in reducing the severity of runny nose and sneezing: a meta-analysis. Clin Pharmacol Ther. 1998;64:579-96.

90. Taverner D, Latte GJ. WITHDRAWN: Nasal decongestants for the common cold. Cochrane Database Syst Rev. 2009:CD001953.

91. Nahin RL, Barnes PM, Stussman BJ, Bloom B. Costs of Complementary and Alternative Medicine (CAM) and Frequency of Visits to CAM Practitioners: United States, 2007: Division of Health Interview Statistics; 2009.

92. Kaufman DW, Kelly JP, Rosenberg L, Anderson TE, Mitchell AA. Recent patterns of medication use in the ambulatory adult population of the United States: the Slone survey. JAMA. 2002;287:337-44.

93. Huntley AL, Thompson Coon J, Ernst E. The safety of herbal medicinal products derived from Echinacea species: a systematic review. Drug Saf. 2005;28:387-400.

94. Goel V, Lovlin R, Chang C, Slama JV, Barton R, Gahler R, Bauer R, Goonewardene L, Basu TK. A proprietary extract from the echinacea plant (Echinacea purpurea) enhances systemic immune response during a common cold. Phytother Res. 2005;19:689-94.

95. Woelkart K, Xu W, Pei Y, Makriyannis A, Picone RP, Bauer R. The endocannabinoid system as a target for alkamides from Echinacea angustifolia roots. Planta Med. 2005;71:701-5.

96. Gertsch J, Schoop R, Kuenzle U, Suter A. Echinacea alkylamides modulate TNF- alpha gene expression via cannabinoid receptor CB2 and multiple signal transduction pathways. FEBS Lett. 2004;577:563-9.

97. Hemilä H. Vitamin C supplementation and respiratory infections: a systematic review. Mil Med. 2004;169:920-5.

98. Grimble RF. Effect of antioxidative vitamins on immune function with clinical applications. Int J Vitam Nutr Res. 1997;67:312-20.

99. Simasek M, Blandino DA. Treatment of the common cold. Am Fam Physician. 2007;75:515-20.

100. Douglas RM, Hemilä H, Chalker E, Treacy B. Vitamin C for preventing and treating the common cold. Cochrane Database Syst Rev. 2007:CD000980.

101. Hoyles L, Vulevic J. Diet, immunity and functional foods. Adv Exp Med Biol. 2008;635:79-92.

157

102. Nahas R, Balla A. Complementary and alternative medicine for prevention and treatment of the common cold. Can Fam Physician. 2011;57:31-6.

103. Turner RB, Wecker MT, Pohl G, Witek TJ, McNally E, St George R, Winther B, Hayden FG. Efficacy of tremacamra, a soluble intercellular adhesion molecule 1, for experimental rhinovirus infection: a randomized clinical trial. JAMA. 1999;281:1797-804.

104. Novick SG, Godfrey JC, Godfrey NJ, Wilder HR. How does zinc modify the common cold? Clinical observations and implications regarding mechanisms of action. Med Hypotheses. 1996;46:295-302.

105. Godfrey JC. Zinc for the common cold. Antimicrob Agents Chemother. 1988;32:605-6.

106. Zarembo JE, Godfrey JC, Godfrey NJ. Zinc(II) in saliva: determination of concentrations produced by different formulations of zinc gluconate lozenges containing common excipients. J Pharm Sci. 1992;81:128-30.

107. Mossad S, Macknin M, Medendorp S, Mason P. Zinc gluconate lozenges for treating the common cold. A randomized, double-blind, placebo-controlled study. Ann Intern Med. 1996;125:81-8.

108. Smith DS, Helzner EC, Nuttall CE, Collins M, Rofman BA, Ginsberg D, Goswick CB, Magner A. Failure of zinc gluconate in treatment of acute upper respiratory tract infections. Antimicrob Agents Chemother. 1989;33:646-8.

109. Belongia EA, Berg R, Liu K. A randomized trial of zinc nasal spray for the treatment of upper respiratory illness in adults. Am J Med. 2001;111:103-8.

110. Mossad SB. Effect of zincum gluconicum nasal gel on the duration and symptom severity of the common cold in otherwise healthy adults. QJM. 2003;96:35-43.

111. Science M, Johnstone J, Roth DE, Guyatt G, Loeb M. Zinc for the treatment of the common cold: a systematic review and meta-analysis of randomized controlled trials. CMAJ. 2012.

112. Graat JM, Schouten EG, Kok FJ. Effect of daily vitamin E and multivitamin-mineral supplementation on acute respiratory tract infections in elderly persons: a randomized controlled trial. JAMA. 2002;288:715-21.

113. Rivlin RS. Historical perspective on the use of garlic. J Nutr. 2001;131:951S-4S.

114. Reinhart KM, Talati R, White CM, Coleman CI. The impact of garlic on lipid parameters: a systematic review and meta-analysis. Nutr Res Rev. 2009;22:39-48.

158

115. Fleischauer AT, Poole C, Arab L. Garlic consumption and cancer prevention: meta-analyses of colorectal and stomach cancers. Am J Clin Nutr. 2000;72:1047- 52.

116. Ngo SN, Williams DB, Cobiac L, Head RJ. Does garlic reduce risk of colorectal cancer? A systematic review. J Nutr. 2007;137:2264-9.

117. Cavallito CJ, Bailey JH. Allicin, the Antibacterial Principle of Allium sativum. I. Isolation, Physical Properties and Antibacterial Action. Journal of the American Chemical Society. 1944;66:1950-1.

118. Weber ND, Andersen DO, North JA, Murray BK, Lawson LD, Hughes BG. In vitro virucidal effects of Allium sativum (garlic) extract and compounds. Planta Med. 1992;58:417-23.

119. Staba EJ, Lash L, Staba JE. A commentary on the effects of garlic extraction and formulation on product composition. J Nutr. 2001;131:1118S-9S.

120. Ruddock PS, Liao M, Foster BC, Lawson L, Arnason JT, Dillon JA. Garlic natural health products exhibit variable constituent levels and antimicrobial activity against Neisseria gonorrhoeae, Staphylococcus aureus and Enterococcus faecalis. Phytother Res. 2005;19:327-34.

121. Lawson LD, Wang ZJ, Papadimitriou D. Allicin release under simulated gastrointestinal conditions from garlic powder tablets employed in clinical trials on serum cholesterol. Planta Med. 2001;67:13-8.

122. Lissiman E, Bhasale AL, Cohen M. Garlic for the common cold. Cochrane Database Syst Rev. 2012;3:CD006206.

123. Josling P. Preventing the common cold with a garlic supplement: a double-blind, placebo-controlled survey. Adv Ther. 2001;18:189-93.

124. World Health Organization. WHO monographs on selected medicinal plants. Geneva: World Health Organization; 1999.

125. Nantz MP, Rowe CA, Muller CE, Creasy RA, Stanilka JM, Percival SS. Supplementation with aged garlic extract improves both NK and γδ-T cell function and reduces the severity of cold and flu symptoms: A randomized, double-blind, placebo-controlled nutrition intervention. Clin Nutr. 2012.

126. Marteau P, Boutron-Ruault MC. Nutritional advantages of probiotics and prebiotics. Br J Nutr. 2002;87 Suppl 2:S153-7.

127. Kollath W. [Nutrition and the tooth system; general review with special reference to vitamins]. Dtsch Zahnarztl Z. 1953;8:Suppl 7-16.

159

128. Reid G, Sanders ME, Gaskins HR, Gibson GR, Mercenier A, Rastall R, Roberfroid M, Rowland I, Cherbut C, Klaenhammer TR. New scientific paradigms for probiotics and prebiotics. J Clin Gastroenterol. 2003;37:105-18.

129. Hao Q, Lu Z, Dong BR, Huang CQ, Wu T. Probiotics for preventing acute upper respiratory tract infections. Cochrane Database Syst Rev. 2011:CD006895.

130. Tschernia A, Moore N, Abi-Hanna A. Effects of long-term supplementation of a weaning food supplemented with oligofructose, a prebiotic, on general infant health status. J Pediatr Gastroenterol Nutr. 1999;29.

131. Bruzzese E, Volpicelli M, Salvini F, Bisceglia M, Lionetti P, Cinquetti M, Iacono G, Guarino A. Early administration of GOS/FOS prevents intestinal and respiratory infections in infants. Journal of Pediatric Gastroenterology and Nutrition. 2006;42.

132. Langkamp-Henken B, Bender BS, Gardner EM, Herrlinger-Garcia KA, Kelley MJ, Murasko DM, Schaller JP, Stechmiller JK, Thomas DJ, Wood SM. Nutritional formula enhanced immune function and reduced days of symptoms of upper respiratory tract infection in seniors. J Am Geriatr Soc. 2004;52:3-12.

133. Population Division D. World Population Ageing: 1950-2050. United Nations; 2010.

134. Alho OP, Karttunen TJ, Karttunen R, Tuokko H, Koskela M, Uhari M. Lymphocyte and mast cell counts are increased in the nasal mucosa in symptomatic natural colds. Clin Exp Immunol. 2003;131:138-42.

135. Wat D. The common cold: a review of the literature. Eur J Intern Med. 2004;15:79- 88.

136. High KP. Nutritional strategies to boost immunity and prevent infection in elderly individuals. Clin Infect Dis. 2001;33:1892-900.

137. Meydani SN, Leka LS, Fine BC, Dallal GE, Keusch GT, Singh MF, Hamer DH. Vitamin E and respiratory tract infections in elderly nursing home residents: a randomized controlled trial. JAMA. 2004;292:828-36.

138. Nicholson KG, Kent J, Hammersley V, Cancio E. Risk factors for lower respiratory complications of rhinovirus infections in elderly people living in the community: prospective cohort study. BMJ. 1996;313:1119-23.

139. Cummings J, Antoine J, Azpiroz F, Bourdet-Sicard R, Brandtzaeg P, Calder P, Gibson G, Guarner F, Isolauri E, et al. PASSCLAIM--gut health and immunity. Eur J Nutr. 2004;43 Suppl 2:II118-II73.

140. Kindt TJ, Goldsby RA, Osborne BA, Kuby J. Kuby immunology. 6th ed. New York: W.H. Freeman; 2007.

160

141. Sompayrac L. How the immune system works. 2nd ed. Malden, Mass.: Blackwell Pub.; 2003.

142. Walrand S, Moreau K, Caldefie F, Tridon A, Chassagne J, Portefaix G, Cynober L, Beaufrère B, Vasson MP, Boirie Y. Specific and nonspecific immune responses to fasting and refeeding differ in healthy young adult and elderly persons. Am J Clin Nutr. 2001;74:670-8.

143. Gabay C, Kushner I. Acute-phase proteins and other systemic responses to inflammation. N Engl J Med. 1999;340:448-54.

144. Tillett WS, Francis T. Serological reactions in pneumonia with a non-protein somatic fraction of pneumococcus. J Exp Med. 1930;52:561-71.

145. Williamson S, Munro C, Pickler R, Grap MJ, Elswick RKJ. Comparison of Biomarkers in Blood and Saliva in Healthy Adults. Nursing Research and Practice. 2012;2012:1-4.

146. Gardner EM, Murasko DM. Age-related changes in Type 1 and Type 2 cytokine production in humans. Biogerontology. 2002;3:271-90.

147. De Groote D, Zangerle PF, Gevaert Y, Fassotte MF, Beguin Y, Noizat-Pirenne F, Pirenne J, Gathy R, Lopez M, Dehart I. Direct stimulation of cytokines (IL-1 beta, TNF-alpha, IL-6, IL-2, IFN-gamma and GM-CSF) in whole blood. I. Comparison with isolated PBMC stimulation. Cytokine. 1992;4:239-48.

148. Agrawal A, Agrawal S, Cao JN, Su H, Osann K, Gupta S. Altered innate immune functioning of dendritic cells in elderly humans: a role of phosphoinositide 3- kinase-signaling pathway. J Immunol. 2007;178:6912-22.

149. Guha M, Mackman N. LPS induction of gene expression in human monocytes. Cell Signal. 2001;13:85-94.

150. Demas GE, Zysling DA, Beechler BR, Muehlenbein MP, French SS. Beyond phytohaemagglutinin: assessing vertebrate immune function across ecological contexts. J Anim Ecol. 2011.

151. McHugh S, Deighton J, Rifkin I, Ewan P. Kinetics and functional implications of Th1 and Th2 cytokine production following activation of peripheral blood mononuclear cells in primary culture. Eur J Immunol. 1996;26:1260-5.

152. Rittig MG, Kaufmann A, Robins A, Shaw B, Sprenger H, Gemsa D, Foulongne V, Rouot B, Dornand J. Smooth and rough lipopolysaccharide phenotypes of Brucella induce different intracellular trafficking and cytokine/chemokine release in human monocytes. J Leukoc Biol. 2003;74:1045-55.

161

153. Ligthart GJ, Corberand JX, Geertzen HG, Meinders AE, Knook DL, Hijmans W. Necessity of the assessment of health status in human immunogerontological studies: evaluation of the SENIEUR protocol. Mech Ageing Dev. 1990;55:89-105.

154. Rink L, Cakman I, Kirchner H. Altered cytokine production in the elderly. Mech Ageing Dev. 1998;102:199-209.

155. Ershler WB, Sun WH, Binkley N, Gravenstein S, Volk MJ, Kamoske G, Klopp RG, Roecker EB, Daynes RA, Weindruch R. Interleukin-6 and aging: blood levels and mononuclear cell production increase with advancing age and in vitro production is modifiable by dietary restriction. Lymphokine Cytokine Res. 1993;12:225-30.

156. Gill HS, Rutherfurd KJ, Cross ML. Dietary probiotic supplementation enhances natural killer cell activity in the elderly: an investigation of age-related immunological changes. J Clin Immunol. 2001;21:264-71.

157. Schiffrin E, Morley J, Donnet-Hughes A, Guigoz Y. The inflammatory status of the elderly: The intestinal contribution. Mutat Res. 2009.

158. Franceschi C, Bonafè M, Valensin S, Olivieri F, De Luca M, Ottaviani E, De Benedictis G. Inflamm-aging. An evolutionary perspective on immunosenescence. Ann N Y Acad Sci. 2000;908:244-54.

159. Salminen A, Huuskonen J, Ojala J, Kauppinen A, Kaarniranta K, Suuronen T. Activation of innate immunity system during aging: NF-kB signaling is the molecular culprit of inflamm-aging. Ageing Res Rev. 2008;7:83-105.

160. Lesourd BM, Mazari L, Ferry M. The role of nutrition in immunity in the aged. Nutr Rev. 1998;56:S113-25.

161. Lesourd BM, Meaume S. Cell mediated immunity changes in ageing, relative importance of cell subpopulation switches and of nutritional factors. Immunol Lett. 1994;40:235-42.

162. Steinmann GG. Changes in the human thymus during aging. Curr Top Pathol. 1986;75:43-88.

163. Wick G, Grubeck-Loebenstein B. Primary and secondary alterations of immune reactivity in the elderly: impact of dietary factors and disease. Immunol Rev. 1997;160:171-84.

164. Cakman I, Rohwer J, Schütz RM, Kirchner H, Rink L. Dysregulation between TH1 and TH2 T cell subpopulations in the elderly. Mech Ageing Dev. 1996;87:197-209.

165. Saltzman RL, Peterson PK. Immunodeficiency of the elderly. Rev Infect Dis. 1987;9:1127-39.

162

166. Chen WH, Kozlovsky BF, Effros RB, Grubeck-Loebenstein B, Edelman R, Sztein MB. Vaccination in the elderly: an immunological perspective. Trends Immunol. 2009;30:351-9.

167. Lesourd BM, Mazari L. Immune responses during recovery from protein-energy malnutrition. Clin Nutr. 1997;16 Suppl 1:37-46.

168. Gillis S, Kozak R, Durante M, Weksler ME. Immunological studies of aging. Decreased production of and response to T cell growth factor by lymphocytes from aged humans. J Clin Invest. 1981;67:937-42.

169. Trinchieri G. Biology of natural killer cells. Adv Immunol. 1989;47:187-376.

170. Shaw AC, Joshi S, Greenwood H, Panda A, Lord JM. Aging of the innate immune system. Curr Opin Immunol. 2010;22:507-13.

171. Krishnaraj R, Blandford G. Age-associated alterations in human natural killer cells. 2. Increased frequency of selective NK subsets. Cell Immunol. 1988;114:137-48.

172. Moens E, Veldhoen M. Epithelial barrier biology: good fences make good neighbours. Immunology. 2012;135:1-8.

173. Bjarnason I, MacPherson A, Hollander D. Intestinal permeability: an overview. Gastroenterology. 1995;108:1566-81.

174. Bovee-Oudenhoven IM, Lettink-Wissink ML, Van Doesburg W, Witteman BJ, Van Der Meer R. Diarrhea caused by enterotoxigenic Escherichia coli infection of humans is inhibited by dietary calcium. Gastroenterology. 2003;125:469-76.

175. Mestecky J, Russell MW. Specific antibody activity, glycan heterogeneity and polyreactivity contribute to the protective activity of S-IgA at mucosal surfaces. Immunol Lett. 2009;124:57-62.

176. Murphy K, Travers P, Walport M, Janeway C. Janeway's immunobiology. 8th ed. New York: Garland Science; 2012.

177. Brandtzaeg P. Mechanisms of gastrointestinal reactions to food. Environ Toxicol Pharmacol. 1997;4:9-24.

178. Milling SW, Cousins L, MacPherson GG. How do DCs interact with intestinal antigens? Trends Immunol. 2005;26:349-52.

179. Bron PA, van Baarlen P, Kleerebezem M. Emerging molecular insights into the interaction between probiotics and the host intestinal mucosa. Nat Rev Microbiol. 2012;10:66-78.

163

180. Brandtzaeg P, Farstad IN, Johansen FE, Morton HC, Norderhaug IN, Yamanaka T. The B-cell system of human mucosae and exocrine glands. Immunol Rev. 1999;171:45-87.

181. Laissue JA, Gebbers JO. The intestinal barrier and the gut-associated lymphoid tissue. Curr Stud Hematol Blood Transfus. 1992:19-43.

182. van der Waaij LA, Limburg PC, Mesander G, van der Waaij D. In vivo IgA coating of anaerobic bacteria in human faeces. Gut. 1996;38:348-54.

183. Boullier S, Tanguy M, Kadaoui KA, Caubet C, Sansonetti P, Corthésy B, Phalipon A. Secretory IgA-mediated neutralization of Shigella flexneri prevents intestinal tissue destruction by down-regulating inflammatory circuits. J Immunol. 2009;183:5879-85.

184. Isaacs D, Webster AD, Valman HB. Immunoglobulin levels and function in pre- school children with recurrent respiratory infections. Clin Exp Immunol. 1984;58:335-40.

185. Malin M, Suomalainen H, Saxelin M, Isolauri E. Promotion of IgA immune response in patients with Crohn's disease by oral bacteriotherapy with Lactobacillus GG. Ann Nutr Metab. 1996;40:137-45.

186. Sekirov I, Russell SL, Antunes LC, Finlay BB. Gut microbiota in health and disease. Physiol Rev. 2010;90:859-904.

187. Neish AS. Microbes in gastrointestinal health and disease. Gastroenterology. 2009;136:65-80.

188. Kuby J. Immunology. 3rd edition (February 1997) ed. New York: W H Freeman & Co (Sd); 1997.

189. Eckburg PB, Bik EM, Bernstein CN, Purdom E, Dethlefsen L, Sargent M, Gill SR, Nelson KE, Relman DA. Diversity of the human intestinal microbial flora. Science. 2005;308:1635-8.

190. Huurre A, Kalliomäki M, Rautava S, Rinne M, Salminen S, Isolauri E. Mode of delivery - effects on gut microbiota and humoral immunity. Neonatology. 2008;93:236-40.

191. Penders J, Thijs C, Vink C, Stelma FF, Snijders B, Kummeling I, van den Brandt PA, Stobberingh EE. Factors influencing the composition of the intestinal microbiota in early infancy. Pediatrics. 2006;118:511-21.

192. Khachatryan ZA, Ktsoyan ZA, Manukyan GP, Kelly D, Ghazaryan KA, Aminov RI. Predominant role of host genetics in controlling the composition of gut microbiota. PLoS One. 2008;3:e3064.

164

193. De Filippo C, Cavalieri D, Di Paola M, Ramazzotti M, Poullet JB, Massart S, Collini S, Pieraccini G, Lionetti P. Impact of diet in shaping gut microbiota revealed by a comparative study in children from Europe and rural Africa. Proc Natl Acad Sci U S A. 2010;107:14691-6.

194. Tiihonen K, Ouwehand AC, Rautonen N. Human intestinal microbiota and healthy ageing. Ageing Res Rev. 2010;9:107-16.

195. Zoetendal EG, Ben-Amor K, Akkermans AD, Abee T, de Vos WM. DNA isolation protocols affect the detection limit of PCR approaches of bacteria in samples from the human gastrointestinal tract. Syst Appl Microbiol. 2001;24:405-10.

196. Hooper LV. Do symbiotic bacteria subvert host immunity? Nat Rev Microbiol. 2009;7:367-74.

197. Hofer U, Speck RF. Disturbance of the gut-associated lymphoid tissue is associated with disease progression in chronic HIV infection. Semin Immunopathol. 2009;31:257-66.

198. Svedlund J, Sjödin I, Dotevall G. GSRS--a clinical rating scale for gastrointestinal symptoms in patients with irritable bowel syndrome and peptic ulcer disease. Dig Dis Sci. 1988;33:129-34.

199. Eypasch E, Williams JI, Wood-Dauphinee S, Ure BM, Schmülling C, Neugebauer E, Troidl H. Gastrointestinal Quality of Life Index: development, validation and application of a new instrument. Br J Surg. 1995;82:216-22.

200. Kulich KR, Madisch A, Pacini F, Piqué JM, Regula J, Van Rensburg CJ, Ujszászy L, Carlsson J, Halling K, Wiklund IK. Reliability and validity of the Gastrointestinal Symptom Rating Scale (GSRS) and Quality of Life in Reflux and Dyspepsia (QOLRAD) questionnaire in dyspepsia: a six-country study. Health Qual Life Outcomes. 2008;6:12.

201. Kleinman L, Kilburg A, Machnicki G, Faull R, Walker R, Prasad R, Ambuehl P, Bahner U, Margolis MK. Using GI-specific patient outcome measures in renal transplant patients: validation of the GSRS and GIQLI. Qual Life Res. 2006;15:1223-32.

202. Dimenäs E, Carlsson G, Glise H, Israelsson B, Wiklund I. Relevance of norm values as part of the documentation of quality of life instruments for use in upper gastrointestinal disease. Scand J Gastroenterol Suppl. 1996;221:8-13.

203. Kolida S, Tuohy K, Gibson GR. Prebiotic effects of inulin and oligofructose. Br J Nutr. 2002;87 Suppl 2:S193-7.

204. Hentges DJ. Human intestinal microflora in health and disease. New York: Academic Press; 1983.

165

205. Cummings JH, Macfarlane GT. Colonic microflora: nutrition and health. Nutrition. 1997;13:476-8.

206. Bouhnik Y, Raskine L, Simoneau G, Vicaut E, Neut C, Flourié B, Brouns F, Bornet F. The capacity of nondigestible carbohydrates to stimulate fecal bifidobacteria in healthy humans: a double-blind, randomized, placebo-controlled, parallel-group, dose-response relation study. Am J Clin Nutr. 2004;80:1658-64.

207. Thomson AB. Small intestinal disorders in the elderly. Best Pract Res Clin Gastroenterol. 2009;23:861-74.

208. Mitsuoka T. Bifidobacteria and their role in human health. J Ind Microbiol. 1990:263-7.

209. Maukonen J, Mättö J, Kajander K, Mattila-Sandholm T, Saarela M. Diversity and temporal stability of fecal bacterial populations in elderly subjects consuming galacto-oligosaccharide containing probiotic yogurt. Int Dairy J. 2008:386-95.

210. Hopkins MJ, Sharp R, Macfarlane GT. Age and disease related changes in intestinal bacterial populations assessed by cell culture, 16S rRNA abundance, and community cellular fatty acid profiles. Gut. 2001;48:198-205.

211. He T, Harmsen HJM, Raangs GC, Welling GW. Composition of faecal microbiota of elderly people. Microb Ecol Health Dis. 2003;15:153-9.

212. Gu Q, Dillon CF, Burt VL. Prescription drug use continues to increase: U.S. prescription drug data for 2007-2008. Hyattsville, MD; 2010.

213. Guggenheimer J, Moore PA. Xerostomia: etiology, recognition and treatment. J Am Dent Assoc. 2003;134:61-9; quiz 118-9.

214. Lähteenmäki MT, Salo MS, Tenovuo JO, Helminen AV, Vilja PJ, Huupponen RK. The effects of glycopyrrolate on oral mucous host defenses in healthy volunteers. Anesth Analg. 2000;91:467-72.

215. Williams C, McColl KE. Review article: proton pump inhibitors and bacterial overgrowth. Aliment Pharmacol Ther. 2006;23:3-10.

216. Woodmansey EJ. Intestinal bacteria and ageing. J Appl Microbiol. 2007;102:1178- 86.

217. Aaronson PI, McKinnon W, Poston L. Mechanism of butyrate-induced vasorelaxation of rat mesenteric resistance artery. Br J Pharmacol. 1996;117:365- 71.

166

218. Dass NB, John AK, Bassil AK, Crumbley CW, Shehee WR, Maurio FP, Moore GB, Taylor CM, Sanger GJ. The relationship between the effects of short-chain fatty acids on intestinal motility in vitro and GPR43 receptor activation. Neurogastroenterol Motil. 2007;19:66-74.

219. Suzuki T, Yoshida S, Hara H. Physiological concentrations of short-chain fatty acids immediately suppress colonic epithelial permeability. Br J Nutr. 2008;100:297-305.

220. Yu LC, Wang JT, Wei SC, Ni YH. Host-microbial interactions and regulation of intestinal epithelial barrier function: From physiology to pathology. World J Gastrointest Pathophysiol. 2012;3:27-43.

221. Born J, Uthgenannt D, Dodt C, Nünninghoff D, Ringvolt E, Wagner T, Fehm HL. Cytokine production and lymphocyte subpopulations in aged humans. An assessment during nocturnal sleep. Mech Ageing Dev. 1995;84:113-26.

222. Lio D, D'Anna C, Gervasi F, Scola L, Potestio M, Di Lorenzo G, Listì F, Colombo A, Candore G, Caruso C. Interleukin-12 release by mitogen-stimulated mononuclear cells in the elderly. Mech Ageing Dev. 1998;102:211-9.

223. Beharka AA, Meydani M, Wu D, Leka LS, Meydani A, Meydani SN. Interleukin-6 production does not increase with age. J Gerontol A Biol Sci Med Sci. 2001;56:B81-8.

224. Candore G, Di Lorenzo G, Melluso M, Cigna D, Colucci AT, Modica MA, Caruso C. gamma-Interferon, interleukin-4 and interleukin-6 in vitro production in old subjects. Autoimmunity. 1993;16:275-80.

225. Fagiolo U, Cossarizza A, Scala E, Fanales-Belasio E, Ortolani C, Cozzi E, Monti D, Franceschi C, Paganelli R. Increased cytokine production in mononuclear cells of healthy elderly people. Eur J Immunol. 1993;23:2375-8.

226. Llorente L, Richaud-Patin Y, Alvarado C, Vidaller A, Jakez-Ocampo J. Autoantibody production in healthy elderly people is not promoted by interleukin- 10 although this cytokine is expressed in them by a peculiar CD8+CD3+ large granular cell subpopulation. Scand J Immunol. 1997;45:401-7.

227. Németh G. Health related quality of life outcome instruments. Eur Spine J. 2006;15 Suppl 1:S44-51.

228. Morley JE. Assessment of malnutrition in older persons: a focus on the Mini Nutritional Assessment. J Nutr Health Aging. 2011;15:87-90.

229. Gad SC. In vitro toxicology. 2nd ed. New York: Taylor & Francis; 2000.

230. Siebert JC, Walker EB. Monitoring cytokine profiles during immunotherapy. Immunotherapy. 2010;2:799-816.

167

231. Gleeson M. Mucosal immune responses and risk of respiratory illness in elite athletes. Exerc Immunol Rev. 2000;6:5-42.

232. McNerlan SE, Alexander HD, Rea IM. Age-related reference intervals for lymphocyte subsets in whole blood of healthy individuals. Scand J Clin Lab Invest. 1999;59:89-92.

233. Jiao Y, Qiu Z, Xie J, Li D, Li T. Reference ranges and age-related changes of peripheral blood lymphocyte subsets in Chinese healthy adults. Sci China C Life Sci. 2009;52:643-50.

234. Holma R, Kekkonen RA, Hatakka K, Poussa T, Vaarala O, Adlercreutz H, Korpela R. Consumption of Galactooligosaccharides Together with Probiotics Stimulate the in vitro Peripheral Blood Mononuclear Cell Proliferation and IFNγ Production in Healthy Men. ISRN Immunology. 2011.

235. Bunout D, Hirsch S, Pía de la Maza M, Muñoz C, Haschke F, Steenhout P, Klassen P, Barrera G, Gattas V, Petermann M. Effects of prebiotics on the immune response to vaccination in the elderly. JPEN J Parenter Enteral Nutr. 2002;26:372-6.

236. Arunachalam K, Gill HS, Chandra RK. Enhancement of natural immune function by dietary consumption of Bifidobacterium lactis (HN019). Eur J Clin Nutr. 2000;54:263-7.

237. Schiffrin EJ, Rochat F, Link-Amster H, Aeschlimann JM, Donnet-Hughes A. Immunomodulation of human blood cells following the ingestion of lactic acid bacteria. J Dairy Sci. 1995;78:491-7.

238. Calder PC. Immunological parameters: what do they mean? J Nutr. 2007;137:773S-80S.

239. Falony G, Vlachou A, Verbrugghe K, De Vuyst L. Cross-feeding between Bifidobacterium longum BB536 and acetate-converting, butyrate-producing colon bacteria during growth on oligofructose. Appl Environ Microbiol. 2006;72:7835-41.

240. Cummings JH, Macfarlane GT. Gastrointestinal effects of prebiotics. Br J Nutr. 2002;87 Suppl 2:S145-51.

241. Rousseaux C, Thuru X, Gelot A, Barnich N, Neut C, Dubuquoy L, Dubuquoy C, Merour E, Geboes K, et al. Lactobacillus acidophilus modulates intestinal pain and induces opioid and cannabinoid receptors. Nat Med. 2007;13:35-7.

242. Diop L, Guillou S, Durand H. Probiotic food supplement reduces stress-induced gastrointestinal symptoms in volunteers: a double-blind, placebo-controlled, randomized trial. Nutr Res. 2008;28:1-5.

168

243. Inan MS, Rasoulpour RJ, Yin L, Hubbard AK, Rosenberg DW, Giardina C. The luminal short-chain fatty acid butyrate modulates NF-kappaB activity in a human colonic epithelial cell line. Gastroenterology. 2000;118:724-34.

244. Sangwan V, Tomar SK, Singh RR, Singh AK, Ali B. Galactooligosaccharides: novel components of designer foods. J Food Sci. 2011;76:R103-11.

245. Winkler P, de Vrese M, Laue C, Schrezenmeir J. Effect of a dietary supplement containing probiotic bacteria plus vitamins and minerals on common cold infections and cellular immune parameters. Int J Clin Pharmacol Ther. 2005;43:318-26.

246. de Vrese M, Winkler P, Rautenberg P, Harder T, Noah C, Laue C, Ott S, Hampe J, Schreiber S, et al. Effect of Lactobacillus gasseri PA 16/8, Bifidobacterium longum SP 07/3, B. bifidum MF 20/5 on common cold episodes: a double blind, randomized, controlled trial. Clin Nutr. 2005;24:481-91.

247. Glück U, Gebbers JO. Ingested probiotics reduce nasal colonization with pathogenic bacteria (Staphylococcus aureus, Streptococcus pneumoniae, and beta-hemolytic streptococci). Am J Clin Nutr. 2003;77:517-20.

248. Yasui H, Kiyoshima J, Hori T. Reduction of influenza virus titer and protection against influenza virus infection in infant mice fed Lactobacillus casei Shirota. Clin Diagn Lab Immunol. 2004;11:675-9.

249. Guigoz Y, Rochat F, Perruisseau-Carrier G, Rochat I, Schiffrin EJ. Effects of oligosaccharide on the faecal flora and non-specific immune system in elderly people Nutrition Research. 2002:13-25.

250. Alles MS, Hartemink R, Meyboom S, Harryvan JL, Van Laere KM, Nagengast FM, Hautvast JG. Effect of transgalactooligosaccharides on the composition of the human intestinal microflora and on putative risk markers for colon cancer. Am J Clin Nutr. 1999;69:980-91.

251. Davis LM, Martínez I, Walter J, Hutkins R. A dose dependent impact of prebiotic galactooligosaccharides on the intestinal microbiota of healthy adults. Int J Food Microbiol. 2010;144:285-92.

252. Depeint F, Tzortzis G, Vulevic J, I'anson K, Gibson GR. Prebiotic evaluation of a novel galactooligosaccharide mixture produced by the enzymatic activity of Bifidobacterium bifidum NCIMB 41171, in healthy humans: a randomized, double- blind, crossover, placebo-controlled intervention study. Am J Clin Nutr. 2008;87:785-91.

253. Yasui H, Kiyoshima J, Hori T, Shida K. Protection against influenza virus infection of mice fed Bifidobacterium breve YIT4064. Clin Diagn Lab Immunol. 1999;6:186- 92.

169

BIOGRAPHICAL SKETCH

Stephanie-Anne Girard was born in Montreal, Canada in 1982 where she has resided most of her life. After completing her high school education with honors, she attended McGill University for her undergraduate studies. She finished her Bachelor of

Science degree in microbiology and immunology in April 2005. After working in the pharmacy field for two years, she decided to pursue her education and obtained her master’s degree in pharmacology from the Department of Medicine at Université de

Montréal. At that time, her main research focus was the effects of probiotics on post- myocardial infarction depression. More specifically, her interest was the kinetic of biochemical and physiological cerebral changes following a myocardial infarction. Her exposure to the probiotic field served as a platform to her current research topic; prebiotics. She received her doctorate from the College of Agricultural and Life

Sciences in the summer of 2012, focusing on oligosaccharides and their effect on mucosal immunity and health outcomes.

170