sign in sign up
<<
Home , Cholecalciferol

University of Tennessee, Knoxville TRACE: Tennessee Research and Creative Exchange

Doctoral Dissertations Graduate School

5-2019

Encapsulation of D3 in Gum Arabic to Improve Application in Beverages and Enhance Bioavailability

Mary Ross Lamsen University of Tennessee, [email protected]

Follow this and additional works at: https://trace.tennessee.edu/utk_graddiss

Recommended Citation Lamsen, Mary Ross, "Encapsulation of Vitamin D3 in Gum Arabic to Improve Application in Beverages and Enhance Bioavailability. " PhD diss., University of Tennessee, 2019. https://trace.tennessee.edu/utk_graddiss/5403

This Dissertation is brought to you for free and open access by the Graduate School at TRACE: Tennessee Research and Creative Exchange. It has been accepted for inclusion in Doctoral Dissertations by an authorized administrator of TRACE: Tennessee Research and Creative Exchange. For more information, please contact [email protected]. To the Graduate Council:

I am submitting herewith a dissertation written by Mary Ross Lamsen entitled "Encapsulation of Vitamin D3 in Gum Arabic to Improve Application in Beverages and Enhance Bioavailability." I have examined the final electronic copy of this dissertation for form and content and recommend that it be accepted in partial fulfillment of the equirr ements for the degree of Doctor of Philosophy, with a major in Food Science.

Qixin Zhong, Major Professor

We have read this dissertation and recommend its acceptance:

Guoxun Chen, Vermont P. Dia, Doris D'Souza

Accepted for the Council:

Dixie L. Thompson

Vice Provost and Dean of the Graduate School

(Original signatures are on file with official studentecor r ds.)

Encapsulation of Vitamin D3 in Gum Arabic to Improve Application in Beverages and Enhance Bioavailability

A Dissertation Presented for the Doctor of Philosophy Degree The University of Tennessee, Knoxville

Mary Ross Llonillo Lamsen May 2019

Copyright © 2019 by Mary Ross Llonillo Lamsen All rights reserved.

ii Dedication

This work is dedicated to my husband, Leo, and our daughter, Sofia.

Leo, I have known you for more than 24 years. You and I come from humble beginnings, and to see how we have grown together is a testament to what it means to have a supportive significant other.

My little Sofia, You will be turning 6 yrs old by the time I graduate. I hope that what I am doing will enable you to see the importance of always finishing what you have started. I always want you to be truthful in everything you do, always take the time to help others in need, choose to be kind, and do the best you can always.

iii Acknowledgements I would like to express my gratitude towards my major advisor, Dr. Qixin Zhong. Dr.

Zhong, you have been patient in guiding me for almost a decade now as you were also my major advisor on my master’s degree. I could only imagine the considerable time you spend reading our abstracts, dissertations, posters, application for research awards and much more. You never get mad, you always keep your composure, and provide feedback as best as you can. Please know that I appreciate you and the comments and suggestions you give. I was lost for a long time, and

I am glad that you helped re-direct my focus in research. Thank you for being an integral part of my training.

I am thankful to all my doctoral committee members for always having the door open for me to ask questions. Dr. D’Souza, you are one of the few people I have known the longest in our department. You guided me even during my masters training up to now. I remember practicing some of my presentations in front of you and you provided so much constructive advice.

Whenever I am not sure of something, and just need another person to converse my thoughts with, you have always been there. Thank you for being you, and please do not ever change the way you are. Dr. Chen, you have always exhibited a wonderful positive energy. I will always remember how you cut up some pineapples and offered it to Tiannan and I when we were taking a break during our work with the animals. Besides this fatherly gesture, you helped me so much with the animal study. Thank you for teaching me how to do animal procedures and guiding me in how to best decipher the data collected from the animal research. Dr. Dia, you have always extended your doors open for consultation. When I was first trying to test out the ELISA kits, you were my go- to person in trying to make sure I was doing it right. It was so intimidating at first, because I know

iv the kits were expensive and I did not want to make a mistake. Thank you for letting me share my concerns with you and for answering all my questions.

I would also like to express my gratitude towards my previous PI, Dr. Zivanovic and committee members, Dr. Penfield and Dr. Kopsell. Although they are no longer with the

University, I am grateful for the time they had given me. I know that they advised and guided me the best they can.

I want to thank Tiannan Wang for being the main hands on person who helped me in the entire duration of the animal study. I don’t think I would have been able to successfully gavage the rats if you did not hold them for me. Thank you for helping me withdraw blood and do animal procedures. You played a critical role in ensuring all requirements are accomplished with our research. Thank you for your time, I will forever be grateful.

There was a time during the pharmacokinetic study, where I could not be in two places at the same time. I want to thank Lianger “Felix” Dong, Melody Fagan, Jennifer Vuia-Riser, Andrea

Nieto, Shan Hong, and Dara Smith for assisting me during blood collection by centrifuging the blood and separating out the serum. Thank you for being there for me and keeping me company for many long hours. Also, I want to express my gratitude to Dr. Philipus Pangloli, for not only helping me on my in vivo research, but for being my mentor and friend.

I want to thank my mom and mother-in-law who have always been a constant part of my life. Mom, thank you for visiting me from the Philippines and cooking food for Leo and Sofia while I was away working long hours at the animal facility. I want to thank my mother-in-law

(Mama Pina) for always being there for me. I know I can always call you when I need help, thank you for being a strong support. To my sister Margie, and my younger brother Raymond, you both are my rock. Thank you for being a part of my journey.

v Abstract

Delivery of vitamin D3 (VD3) in foods should exhibit desirable physicochemical characteristics and enhance nutritional significance. To incorporate lipophilic VD3 in beverages, micro/nanoencapsulation of VD in food biopolymers stable at a wide pH range is a potential strategy. Gum arabic (GA), a highly branched arabinogalactan with a conjugated peptide, was investigated as a potential VD3 carrier in this dissertation with objectives to determine (1) optimum conditions of encapsulating VD3 in GA to maximize loading and characterize physicochemical properties, (2) in vitro bioaccessibility of encapsulated VD3 in optimized formulation, and (3)

-1 absorption of VD3 after a single high-dose (300 µg) and daily-dose (60 µg d , 2-wks) oral supplementation by measuring 25(OH)D levels in Sprague-Dawley rats. To encapsulate VD3, GA solution was prepared at 5% (w/v) in deionized water, VD3 solutions were prepared separately in

5 mL ethanol corresponding to 0.03-6.0% mass of GA. VD3 solution was added dropwise into 50 mL of the GA solution while blending at 8,000 rpm for 3 min and then 10,000 rpm for 2 min and freeze-dried. The optimized formulation had an encapsulation efficiency of 71.3% and a loading capacity of 3.44%. As pH increased from 2.0-7.4, the average hydrodynamic diameter decreased from 238.1 to 81.3 nm; whereas the polydispersity index increased from 0.372-0.478, and zeta- potential increased its magnitude from -3.09 to -30.97 mV. No precipitation was observed for all dispersions during 100-d storage at 3 °C. In vitro bioaccessibility of encapsulated VD3 was

95.76%, compared to 68.98% for non-encapsulated VD3 (p < 0.05). The in vivo pharmacokinetic study after oral gavage of 300 µg VD3 showed that the area-under-curve in 48 h of the encapsulation treatment was 4.32-fold of the non-encapsulated VD3 and more than twice higher

-1 than the physical mixture. For rats given a daily-dose of 60 µg VD3 d for 2 weeks, the serum

25(OH)D levels of rats that received capsules versus physical mixture showed an increase by 50%

vi and 65%, respectively, when compared with the non-encapsulated VD3 group. The current work demonstrates that the studied encapsulation system may be used to fortify VD3 in beverages with a wide pH range.

Keywords: cholecalciferol, gum acacia, encapsulation, physicochemical properties, absorption

vii Table of Contents

Chapter 1 Introduction and literature review...... 1

1.1 ...... 3

1.1.1 Discovery, structure, source and physicochemical characteristics ...... 3 1.1.2 Biological functions ...... 4 1.1.3 Recommended dietary intake, status assessment, metabolism ...... 7 1.1.4 Tolerable upper intake level ...... 8 1.1.5 Challenges in ...... 9 1.1.6 Encapsulated vitamin D for food or pharmaceutical use ...... 10 1.1.7 Delivery and controlled release in vitro and in vivo studies ...... 11 1.2 Gum arabic ...... 12

1.2.1 Source and chemical structure ...... 12 1.2.2 Current food use and application ...... 13 1.2.3 Carrier for labile compounds in food systems ...... 14 1.2.4 Digestion and absorption ...... 15 1.2.5 Delivery and controlled release ...... 15 1.3 Concluding remarks ...... 17 1.4 Hypothesis and overview of dissertation research ...... 17

References ...... 18

Chapter 2 Encapsulation of vitamin D3 in gum arabic: physicochemical properties and bioaccessibility ...... 28

2.1 Abstract ...... 29 2.2 Introduction ...... 30 2.3 Materials and methods ...... 32

2.3.1 Materials ...... 32 2.3.2 Preparation of capsules ...... 32

2.3.3 Determination of total and non-bound vitamin D3 in freeze-dried powder ...... 33 2.3.4 Determination of encapsulation efficiency and loading capacity ...... 34

viii 2.3.5 Differential scanning calorimetry ...... 34 2.3.6 Physicochemical properties of capsules and storage stability of reconstituted dispersions ...... 35 2.3.7 Assessment of bioaccessibility ...... 35 2.3.8 Data and statistical analysis ...... 36 2.4 Results and discussion ...... 37

2.4.1 Encapsulation properties ...... 37 2.4.2 Physicochemical properties of capsules before and after freeze-drying ...... 38 2.4.3 Physicochemical properties during 100 d storage at 3 °C ...... 39

2.4.4 Bioaccessibility of vitamin D3 encapsulated in gum arabic ...... 40 2.5 Conclusions ...... 40 2.6 Acknowledgments...... 41

References ...... 42 Appendix ...... 47

Chapter 3 Encapsulation of vitamin D3 in gum arabic: in vivo bioavailability in Sprague- Dawley rats ...... 59

3.2 Abstract ...... 60 3.3 Introduction ...... 61 3.4 Materials and methods ...... 64

3.4.1 Materials ...... 64 3.4.2 Animals and diets ...... 64 3.4.3 Preparation of treatments for gavage administration in phase I and phase II ...... 65 3.4.4 Blood collection and serum separation ...... 67 3.4.5 Measurement of serum 25(OH)D concentration ...... 68 3.4.6 Pharmacokinetic data analysis ...... 68 3.4.7 Data and statistical analysis ...... 68 3.5 Results and discussion ...... 69

3.5.1 Effect of diet on survival, overall health, and body weight of Sprague-Dawley rats .. 69 3.5.2 Effect of dietary induced or sufficiency ...... 70

3.5.3 Pharmacokinetics following a dose of 300 µg vitamin D3 ...... 72

ix 3.5.4 Serum 25(OH)D levels during 2-week study with 60 µg vitamin D3 daily dose ...... 74 3.6 Conclusion ...... 76 3.7 Acknowledgements ...... 77

References ...... 78 Appendix ...... 83

Chapter 4 Concluding remarks and future directions ...... 95

4.1 Conclusions ...... 96 4.2 Future work ...... 97

References ...... 98

Vita ...... 99

x List of Tables

Table 2.1. Compositions of simulated saliva and gastrointestinal juices adapted from Brandon and others (2006) and Oomen and others (2003)...... 47

Table 2.2. Total and free vitamin D3 in freeze-dried powder. * ...... 48

Table 2.3. Encapsulation efficiency (EE%) and loading capacity (LC%) of treatments with VD3 used at 0.03-6.00 % of mass of gum arabic (GA).* ...... 49

Table 2.4. Vitamin D3 detected in the precipitate after simulated mouth, gastric and intestinal phase digestions...... 58 Table 3.1. Dietary composition of vitamin D sufficient (VDS) pellet food a (TD.96348, Envigo, Teklad Custom Diet, Madison, WI)...... 84 Table 3.2. Dietary composition of vitamin D deficient (VDD) pellet food a (TD.87095, Envigo, Teklad Custom Diet, Madison, WI)...... 85 Table 3.3. The designed and actual masses of ingredients used in Phase I in vivo pharmacokinetics study.* ...... 86 Table 3.4. The designed and actual masses of ingredients used in Phase II in vivo daily-dose for 2 wks study.* ...... 87 Table 3.5. Mean body weight of SD rats at age 21, 78, and 108 day in corresponding treatment groups.* ...... 89 Table 3.6. Pharmacokinetic data in 25(OH)D levels (ng/mL) for different treatments (same data in Fig. 3.4).* ...... 92

Table 3.7. In vivo pharmacokinetic parameters following 300 µg VD3 oral administration to Sprague-Dawley rats...... 93 Table 3.8. Serum 25(OH)D (ng/mL) level of Sprague-Dawley rats at weaning age (day 21), during

2-week daily-dose of 60 µg VD3 treatments (day 85-98, measured on days 89, 92, 99), post- treatment (day 106), and on euthaenization (day 108).* ...... 94

xi List of Figures

Figure 2.1. Differential scanning calorimetry (DSC) thermograms of A) non-encapsulated VD3, and B) gum arabic...... 50

Figure 2.2. DSC thermograms of capsules prepared with VD3 at A) 0.03, B) 0.06, C) 0.3, D) 0.6, E) 3.0, and F) 6.0% mass of gum arabic...... 51

Figure 2.3. Mean hydrodynamic diameter (Dh) of fresh dispersion with VD3 corresponding to 6% mass of gum arabic, dispersion reconstituted with the corresponding freeze-dried powder, and gum arabic at pH 2.0-7.4. Error bars are SD (n=3). Different letters above bars indicate statistical difference among samples at the same pH (P < 0.05)...... 52

Figure 2.4. Polydispersity index (PDI) of fresh dispersion with VD3 corresponding to 6% mass of gum arabic, dispersion reconstituted with the corresponding freeze-dried powder, and gum arabic at pH 2.0-7.4. Error bars are SD (n=3). Different letters above bars indicate statistical difference among samples at the same pH (P < 0.05)...... 53

Figure 2.5. Zeta potential of fresh dispersion with VD3 corresponding to 6% mass of gum arabic, dispersion with the corresponding freeze-dried powder, and gum arabic at pH 2.0-7.4. Error bars are SD (n=3)...... 54

Figure 2.6. Mean hydrodynamic diameter (Dh) of A) gum arabic, and B) dispersion reconstituted with freeze-dried powder prepared with VD3 corresponding to 6% mass of gum arabic during 100- day storage at 3 °C. Error bars are SD (n=3)...... 55 Figure 2.7. Polydispersity index (PDI) of A) gum arabic, and B) dispersion reconstituted with freeze-dried powder prepared with VD3 corresponding to 6% mass of gum arabic during 100-day storage at 3 °C. Error bars are SD (n=3)...... 56 Figure 2.8. Zeta-potential of A) gum arabic, and B) dispersion reconstituted with freeze-dried powder prepared with VD3 corresponding to 6% mass of gum arabic during 100-day storage at 3 °C. Error bars are SD (n=3)...... 57 Figure 3.1. General schematic design of the in vivo study...... 83 Figure 3.2. General schematic on oral gavage administration of formulations, animal and blood collection, serum separation and analysis...... 88 Figure 3.3. Mean serum 25(OH)D of SD rats on weaning (day 21) and after 8 wk dietary treatment (day 77, baseline) for the two controls and 4 treatment groups...... 90

xii Figure 3.4. Pharmacokinetic profile in 25(OH)D concentrations after administration of 300 µg

VD3 in the capsules (dispersed in phosphate buffer), physical mixture (VD3 pre-dissolved in propylene glycol and gum arabic in phosphate buffer), or non-encapsulated (pre-dissolved in propylene glycol) forms. The control treatments are fed with the same amount of gum arabic dissolved in phosphate buffer. Rats were fed with VD deficient (VDD) or sufficient (VDS) diets. Error bars are S.E.M. (n=10)...... 91

xiii Chapter 1 Introduction and literature review

1 Encapsulation of lipophilic nutrients has the aim to improve stability in final products and release in the small intestines for absorption. Encapsulation is simply defined as a process to entrap one or more substances, (e.g., active agent), within another substance, commonly referred to as a carrier material (Fang and Bhandari 2010). Vitamin D (VD) is one of four fat-soluble (e.g., A, D, E, and K). VD is unique in that it is not strictly defined as a vitamin because it can be synthesized in the skin upon exposure to B (UVB; 290-315 nm) rays from the sun, thereby, classifying VD more as a prohormone (a precursor of an active hormone) (Chen and others 2007; Holick and Chen 2008). However, VD deficiency is increasingly being reported and may be attributed to various factors including not getting enough sunlight exposure and/or limited food choices (Calvo and Whiting 2013; Dwyer and others 2014).

The importance of getting adequate VD in the diet has prompted the US Food and Drug

Administration (FDA) to implement changes in the nutrition labels to replace vitamins A and C with D. Beginning January 1, 2020, the VD level (zero or added amount) in the Nutrition Facts

Label of packaged foods must be clearly labeled (www.fda.gov). Therefore, improving delivery of VD by encapsulation in a hydrophilic carrier material to enhance dispersibility in beverages may increase VD accessibility.

Beverages are one of the fastest growing food sectors and may provide a good medium for

VD fortification. Of the two forms of VD (VD2 vs VD3), VD3 is 2-3 times more effective than

VD2 at raising and maintaining circulating 25(OH)D levels (Tripkovic and others 2012). The focus of this study is to encapsulate VD3 in gum arabic (GA), a water-soluble matrix. GA is currently not used alone as an encapsulating matrix for VD. This literature review will therefore include an introduction to VD, its structure, stability, and types of carrier materials that have been used to encapsulate VD. Similarly reviewed will be the encapsulation matrix (GA), its structure,

2 physicochemical properties, practical applications, and studies that utilizes GA alone or as a co- encapsulant will be presented.

1.1 Vitamin D

1.1.1 Discovery, structure, source and physicochemical characteristics

Discovery of VD was credited to a select few individuals who made major contributions.

Alleviation of like symptoms was first demonstrated by the work of Sir Edward Mellanby in 1918 in feeding sunlight deprived dogs with cod liver oil (Holick 2003; DeLuca 2004).

However, it was the work of Adolf Windausin on elucidating the structure of several including VD3, and 7-dehydrocholesterol, the precursor of VD3 in the skin that earned him a Nobel

Prize in 1928 (DeLuca 2004).

Structurally, VD are secosteroids with a structural resemblance to (four- ring core structure); however, unlike cholesterol, VD has a break between carbons 9 and 10 in the

B ring, with rings A, C, and D remaining intact (Gropper and others 2009). This break in C9-C10 bond is structurally important as it allows flexibility of the A-ring for rearrangement, resulting in the synthesis of VD (Gropper 2009). This A-ring structure has been reported to play a significant role in receptor binding, and the side-chain at C20 critical for metabolism (Boullata 2010). The side-chain structure determines VD affinity for a transport protein referred to as, VD binding protein (VDBP) (Gropper and others 2009). It is through the action of VDBP that transports VD metabolites for tissue disposition (Gropper and others 2009).

There are two main dietary forms of VD (VD2 and VD3) that are structurally differentiated by a double bond in C22 and C23 in VD2, and in VD3 a methyl group at C24 lacking (Boullata

2010). Natural VD dietary sources are either plant derived from irradiated

(, referred as VD2), or animal derived (cholecalciferol, referred as VD3) (IOM,

3 1997). Natural VD sources are limited and include, egg yolks, -liver oils, fatty fish, liver, and mushrooms (Fixsen and Roscoe 1940; Mattila and others 1999; Nakamura and others 2002). VD fortified foods include milk, margarine, cereal, and orange juice.

VD in foods is sensitive to light and oxygen which is attributed to the conjugated set of double bonds in ring B (Gropper and others 2009). Luo and others (2012) demonstrated that when non-encapsulated VD3 was exposed to 352 nm UV light, only 30% of the VD3 remains after 9.5 h. In a different study, when VD was packed under oxygen, only 10% of original potency remained after four-week storage; whereas, 60% potency remained under nitrogen conditions (Fritz and others 1942).

1.1.2 Biological functions

The major biological function of VD was previously isolated in its action with (PTH) in the homeostasis of blood concentrations and phosphorus metabolism

(Dusso and others 2005). However, there is increasing evidence of VD as an essential nutrient not only for musculoskeletal health, but also in extra skeletal tissues (Jones and others 1998; Wang and others 2010). For example, cell membrane receptors for VD are found in many tissues including muscle, cardiac, pancreas (ß-cells), brain, and skin, thereby regulating various physiological functions (Munker and others 1996; Holick 2004; Wang and others 2008; Cross and

Kallay 2009; Nazarian and others 2011; Roosens and others 2011; Morello and others 2018;

Abdel-Rehim and others 2019). Furthermore, there are growing evidences that show a multimodal association linking circulating 25(OH)D levels in the blood with occurrence of diseases through physiological functions (Goncalves and others 2014; Di Somma and others 2017; Hoseini and others 2017; Zhang and others 2018b). The mechanism of disease development is complex and is beyond the scope of our research. However, a simplified explanation may be attributed to VD role

4 upon its activation by its cellular receptor, leading to transcription rates of target genes responsible for biological responses (Li and others 1997; Dusso and others 2005; Ben-Dov and others 2007;

Abbas 2017).

VD is inactive until it undergoes two processes, first in the liver, and second, in the kidney (Gropper and others 2009). In the liver, 25-hydroxylase functions at carbon 25 to form 25-hydroxyvitamin D (also referred as 25(OH)D3, calcidiol or 25-OH cholecalciferol) which represents the main form of VD released in the blood (Gropper and others 2009). The liver largely secretes most of the 25(OH)D3 into the blood (largest single pool storage site) with little of this metabolite taken up by the extrahepatic tissues (Gropper and others 2009). Most studies report circulating 25(OH)D levels as a marker on VD status. The 25(OH)D level is correlated to oral VD dosing from supplements and foods which is the first area of study because it precedes the mechanism of action of VD in target tissues (Boullata 2010). When the body is deprived of VD, release of VD from reservoirs including, skin and is taken up by tissues especially the kidney (Gropper and others 2009). In the kidney, 1-hydroxylase functions to hydroxylate

25(OH)D3 at position 1, resulting in the second hydroxylation forming 1,25(OH)2D3 (also, referred to as 1, 25 dihydroxy cholecalciferol, or ) (Gropper and others 2009). Calcitriol is synthesized in the kidney due to action of NADPH-dependent (1-hydroxylase) which is also present in various tissues (e.g., bone, intestine, skin, and macrophages) (Gropper and others

2009).

In order to understand why VD should be attained in sufficient levels, studies showing what happens when the body lacks VD are briefly discussed. There are several studies that demonstrated chronic implications resulting from decreased serum 25(OH)D. For example, 71.5%

(88/123) of elderly patients who are VD deficient (serum 25(OHD) < 12 ng mL-1) were associated

5 with development of cognitive dysfunction (Zhang and others 2018b). In animal studies and clinical trials there is an inverse correlation between low VD levels and increased diabetic nephropathy (DN) (Nazarian and others 2011; Prietl and others 2013; Hu and others 2019).

Although VD3 is selected to be used in our study, it is important to mention why different literature sources disagree on whether VD2 and VD3 have comparative efficacy or if one form is superior than the other. The basis of selecting VD3 for our study will be explained by discussing clinical trials that evaluated the efficacy of VD2 vs VD3. A randomized controlled trial of VD replacement strategies comparing 50,000 IU of VD2 twice weekly for 8 weeks versus 50,000 IU of VD3 given to children and young adults (6-21 yrs of age) with cystic fibrosis (CF) showed no difference between groups in change of 25(OH)D (p = 0 .65), leading to the study conclusion that both VD2 and VD3 are equally effective (Simoneau and others 2016). In contrast, participants from an ambulatory osteoporosis clinic (≥ 18 yrs old) receiving 50,000 IU of VD3 or VD2 twice weekly for 5 weeks showed VD3 to be more effective than VD2 in raising serum 25(OH)D levels

(+27.6 vs +12.2 ng mL-1) (Shieh and others 2016). Both studies have the same inclusion criterion of serum 25(OH)D baseline (< 30 ng mL -1) (Shieh and others 2016; Simoneau and others 2016).

Human clinical trials have shown VD3 to be more effective than VD2 in treating nutritional VD deficiency in demographics with chronic diseases (e.g., , alcoholic liver cirrhosis)

(Malham and others 2012; Daroux and others 2013; Mangoo-Karim and others 2015; Tripkovic and others 2017). Furthermore, comparison between VD3 versus VD2 fortification in animal feeds

(e.g., heifer) showed VD3 was more effective (e.g., increasing total beef VD content) than when

VD2 was used as a supplement (Duffy and others 2018).

6 1.1.3 Recommended dietary intake, status assessment, metabolism

There are various professional associations that provide level recommendations for VD consumption. For example, the Institute of Medicine (IOM) is a regulatory authority that determines the VD status and dietary recommendations of nutrients in North America (Fallon

2002). The IOM established the VD Recommended Dietary Allowance (RDA, covers ≥ 97.5 % of the population) at 600 IU/d (ages 1-70 yr) and 800 IU/d (ages 71 and older) (Ross and others

2011). Professional associations have a consensus that serum 25(OH)D is the functional indicator/biomarker of VD status and is expressed in units in ng mL-1 or nmol L-1. According to the IOM, optimal levels for bone health which corresponds to 15-20 µg/d (1 IU cholecalciferol =

40 µg) meet serum 25(OH)D of at least 20 ng mL-1 (50 nmol L-1) (Ross and others 2011). The

Endocrine Society recommends a preferred 25(OH)D blood level range between 40-60 ng mL-1

(100-150 nmol L-1) (Holick 2017). Deficiency in VD is classified when 25(OH)D levels fall to less than 20 ng mL-1 (50 nmol L-1) (Kim and others 2014).

Biochemical assessment to evaluate VD toxic status, in general, uses elevated calcium levels as a marker. To date, no systematic VD intoxication levels have been studied in humans due to ethical reasons (Jones 2008; Chakraborty and others 2015). Our understanding of VD toxicity is from case reports that inadvertently over-fortified foods and/or error in prescription recommendations. For example, over-fortified milk containing 70 to 600 times the recommended

2,000 IU VD gallon-1 resulted in patients (n = 56) having serum calcium level of 13.1 mg dL-1 exhibiting hypercalcemia symptoms (normal Ca levels = 9.8-10.8 mg dL-1) (Blank and others

1995); the 25(OH)D levels of these patients fit the laboratory determined minimum of 90 ng mL-

1 (225 nmol L-1) that is a default diagnosis of hypervitaminosis D. In a different study, a error resulted in a female taking in VD3 supplementation of 60,000 IU once every day for a total

7 of 4 mo alarmingly resulting in high 25(OH)D levels (746 ng mL-1 or 1,865 nmol L-1); however, the elevated serum 25(OH)D levels did not result in elevated calcium levels and the woman exhibited normal concentration of serum calcium (9.0 mg dL-1) (Chakraborty and others 2015).

The metabolism of VD in the body is also affected by amounts present in storage sites.

Combination of VD storage in adipose tissue and distribution mechanisms from liver and kidney sites results in a slow turnover in the body (half-life ~2 mo) (Jones 2008). The main transported metabolite, calcidiol (25(OH)D) has a half-life of ~15 d; whereas the hormone, calcitriol (1α,

25(OH)2D) has a half-life of ~15 h (Jones 2008). Calcitriol (from kidney) is the active form of

VD that targets tissues; however, calcidiol (from liver) reflects that largest indicator of circulating

VD in the blood (Gropper and others 2009). Calcitriol is “not suitable to assess vitamin D status because it is kept within reference limits as long as possible by hormonal mechanisms, especially parathyroid hormone (PTH) for stimulation and serum calcium and phosphate suppression” (Lips

2007). Therefore, serum 25(OH)D (calcidiol) is recognized as having direct effect on overall nutritional VD status.

1.1.4 Tolerable upper intake level

The Food and Nutrition Board (FNB) established a tolerable upper intake level (UL) for

VD (50 μg d-1, or 2,000 IU) to mitigate the prevailing concerns of VD toxicity (Hathcock and others 2007). VD supplementation dosage higher than the UL is reliable in raising serum 25(OH)D without any toxicological symptoms or side-effects (Hollis and others 2012). This has been demonstrated by clinical studies. For example, daily supplementation of 4000 IU VD3/day for 1 y to African American males was an effective strategy in raising serum levels > 40 ng mL-1 with levels of 67.7 ng mL-1 after 1 y (Hollis and others 2012). In another study, no toxicological side- effects were observed after a VD dose of 25,000 IU was given weekly to multiethnic obese children

8 (8-14 yrs age), and these children’s 25(OH)D baseline when compared to 9 weeks later was at adequate levels (< 50 to sufficiency > 50 nmol L-1) (Radhakishun and others 2014).

1.1.5 Challenges in food fortification

Natural sources of VD are limited and possibly infrequently consumed as evident in nutrition surveillance data from various countries (Jayaratne and others 2013; Vaes and others

2017; Cashman and Kiely 2018). Furthermore, increasing reports of vitamin D insufficiency has brought awareness on the need to consume more VD from the diet (Holick 2017; Jahn and others

2019).

It is important to take into consideration whether a product is appropriate to be fortified with VD. The reason is that studies have found a link between perceived appropriateness of a product being chosen to be fortified with VD with purchase intent (Jahn and others 2019). For example, sausage or liver paté were considered inappropriate for VD fortification; whereas, milk and cheese were considered neutral in appropriateness for fortification with VD (Jahn and others

2019).

As with any formulations, it is necessary to overcome several challenges. Practical problems of adding small amounts of powdered forms of VD directly into foods is to accurately measure and mix homogeneously with food. Liquid non-encapsulated VD are less of a problem when added in fats and oils. However, VD, being lipid-soluble has limitations preventing its compatibility with aqueous foods (e.g., beverages). Beverages are consumed regularly and is a viable product for VD fortification because of its greater capacity to deliver the nutrient (Ahmad and Ahmed 2019). VD needs to be encapsulated using water-soluble matrices that can entrap VD for application in aqueous systems. Special considerations on VD load and type of coating material

9 should provide easier dispersion throughout a product and yield to higher retention during processing, storage and in vivo efficacy.

1.1.6 Encapsulated vitamin D for food or pharmaceutical use

Several delivery systems have been used to encapsulate VD. For example, ß-lactoglobulin

(ß-Lg), a protein found in whey was shown to successfully encapsulate VD3 by triggering the self- aggregation of the ß-Lg/VD3 coagulum by mild acidification (pH 4.7) with target applications in dairy foods (e.g., cheese, yogurt) (Diarrassouba and others 2015b). The formulated ß-Lg/VD3 coagulum exhibited high encapsulation efficiency (EE) (94.5 ± 1.8%) and increased photochemical stability to UV light (254 nm, 15 W) (94.2 ± 4.1% of the VD3 remained)

(Diarrassouba and others 2015b).

High amylose corn starch nanoparticle was used to encapsulate VD3 (EE range from 37.06-

78.11%) by means of ultrasonication (Hasanvand and others 2015). The VD3 loaded in high amylose starch resulted in particle size dimensions < 40 nm, and food enrichment in milk showed no difference in homogeneity from a blank control making it a promising carrier for VD3 in dairy

(Hasanvand and others 2015). The method of VD3 entrapment in amylose may be attributed to the helical hydrophobic cavity which can be expanded upon alkali processing (Hasanvand and others 2015). Additionally, VD3 loaded in high amylose is solubilized in aqueous systems due to amorphous regions present in the polysaccharide chains and its relatively small size < 40 nm

(Hasanvand and others 2015). Other targeted delivery systems for VD3 formulation have included stabilization of VD3 using emulsifiers (e.g., sodium caseinate; Tweens 20, 40, 60, 80, and 85; soybean lecithin; decaglycerol monooleate; polysorbate 20 etc.) (Tippetts and others 2012; Guttoff and others 2015; Tipchuwong and others 2017; Golfomitsou and others 2018; Shu and others

2018). Nevertheless, expanding beyond dairy-based products for VD fortification may be

10 beneficial due to high prevalence of lactose intolerance or milk sensitivity allergies (Casellas and others 2016).

There are limited studies that aim to encapsulate VD3 for target applications in non-dairy beverages. Some of these VD3 + matrix formulations include lipid-based carrier systems (e.g., nanoliposomes) (Mohammadi and others 2017), liprotides which are complexes of lipids and partially denatured proteins (Pedersen and others 2016), soybean ß-conglycinin which is a globular storage protein (Levinson and others 2014), and ß-cyclodextrin (Szejtli and Bolla 1980; Szejtli and

Tardy-Lengyel 1984; Delaurent and others 1998). Thus, further studies on improving VD3 stabilization/encapsulation in non-dairy beverages may be considered as it offers an alternative for consumers.

1.1.7 Delivery and controlled release in vitro and in vivo studies

Maintenance of VD stability in the stomach and matrix release in the intestine are two critical steps that precede the mechanism of its absorption over the membrane (Guan and others

2019). Therefore, a widely known prerequisite for effective absorption of VD is its solubilization in the small intestinal fluids.

There are limited studies that evaluated the relationship between oral dose and absorption of encapsulated VD3 in vivo. Entrapped VD3 in ß-lactoglobulin-lysozyme microspheres (EE =

90.8 ± 4.8%) fed to rats at a dose of 77 µg (3080 IU VD3) significantly increased the 25(OH)D

3 3 levels compared to rats fed non-encapsulated VD3 (42.4 ± 8.17 x10 and 15.12 ± 8.17 x10 nmol

-1 L ) (Diarrassouba and others 2015a). In human studies, VD3 in re-assembled casein micelles resulted to comparable increase in serum 25(OH)D as that with VD3 in polysorbate-80

(PS80/Tween-80) (~ 8 ng mL-1) (Levinson and others 2016). Thus, research on release kinetics of

VD3 in other studies is needed if target health benefits are of importance.

11 1.2 Gum arabic

1.2.1 Source and chemical structure

GA is a polysaccharide purified from natural exudate produced from various species of

Acacia senegal trees (Williams and Phillips 2009). The average molecular weight of GA has been reported to be ~400,000 Da with a high degree of branching, spiral coils and polysaccharide side chains (Ward 2000). GA in nature is primarily composed of mixed calcium, magnesium and potassium salt of a polysaccharidic acid (arabic acid) (Glicksman 1983). GA has six carbohydrate moieties which include the following: ᴅ-galactose (~44%), arabinopyranose and arabinofuranose

(~27%), L-rhamnose (~13%), ᴅ-glucuronic acid (~14.5%), and 4-O-methylglucuronic acid ~1.5%

(Glicksman 1983; Williams and Phillips 2009). Structural composition of GA is widely referred to as an arabinogalactan-protein polysaccharide with fractions (as per total mass) being arabinogalactan (~90%), arabinogalactan protein (~10%), and glycoproteins (~1%) (Randall and others 1989).

The GA proteinaceous component gives it emulsifying functional characteristics (Williams and others 1990). Major amino acid composition isolated in fractions primarily includes aspartic acid, phenylalanine, serine, glutamic acid, glycine, valine, and leucine (Randall and others 1989).

In the whole gum, approximately 50% of its proteinaceous matter includes hydroxyproline (~41 nmol mg-1 gum,), serine (~22 nmol mg-1 gum) and proline residues (Randall and others 1989;

Phillips and others 2008).

GA solubility in water is attributed to its highly branched carbohydrate structure and resembles a packed globular molecule. It exhibits low-viscosity in solutions even at relatively high gum concentrations (Glicksman 1983; Islam and others 1997). For example, a 50% solution of gum arabic has a similar viscosity to xanthan gum at 1.5% (Islam and others 1997). Other

12 formulations from which GA was used with another compound (e.g., whey protein concentrate) resulted in reduction of apparent viscosity values than other linear chain polysaccharides (e.g., 80-

250 times less than sodium alginate-whey protein concentrate) (Erben and others 2019).

1.2.2 Current food use and application

GA is widely used in the food industry to control and improve rheological properties of food systems (Daoub and others 2018). The diversity of GA applications in food products includes its actions as stabilizers, film formers, thickeners, emulsifiers, flocculants, and suspending agents

(Williams and others 1990; Ward 2000; Daoub and others 2018). For example, in the beverage industry, GA is used in preparation of flavor emulsion concentrates (Tiss and others 2001). It is also used as a “clouding” agent in dry beverage mixes (Glicksman 1983). Clouding agent is meant to give aqueous solutions the turbid appearance of natural suspensions such as fruit juices or milk

(Wuhrmann and others 1975). Formulation in dry beverage products results in free-flowing granules, and when reconstituted with water, gives the appearance and texture of fresh fruit juice

(Asano 2017). In soft drinks, brominated vegetable oils have been used as a “cloud” agent, however, the use of GA to similarly mimic the effect have been proposed by Oppenheimer (1972).

Enrichment of non- or low-fat beverages utilizing GA as part of the formulation has resulted in improved physicochemical properties in the target delivery food system. In beer, the

“lace curtain” effect on the sides of the glass upon beer consumption was achieved by using GA as a foam stabilizer, demonstrated by Bavisotto and Hansen (1971). In confectionery, GA has been used as a glaze or coating (imparts smoothness, flexibility, and pliability) because the gum is cold-water soluble, forming clear solutions with viscosities approximately twice that of sugar

(Glicksman 1983).

13 1.2.3 Carrier for labile compounds in food systems

Improvements in thermal stability of food bioactives using GA has been demonstrated by multiple studies (Guan and Zhong 2015; Selig and others 2018; Zhang and others 2018a). In the presence of GA, anthocyanins, a group of water-soluble pigments showed better protection from thermal degradation after heating at 80 and 126 °C for 30 min, resulting in residual concentration

1.02 and 1.35 times higher than that without GA (Guan and Zhong 2015). Color stability of beet extract processed under thermal treatment (50 and 80 °C) exhibited greater stability when complexed with anionic polysaccharides, alginate and GA than with formulations without GA

(Selig and others 2018). Similarly, survival of Lactobacillus plantarum (LP) during isothermal heating at 90 °C was strongly influenced by whether GA or reconstituted skim milk was used; with

GA alone as an encapsulant leading to the highest survival of LP cells (Zhang and others 2018a).

The abundant hydroxyl groups and branched structure of GA provide the fundamental basis on its effectiveness in its strong adsorption properties (Kong and others 2014).

GA can stabilize bioactives across a wide pH range (pH > 2.0), which is particularly useful in stabilization of pH sensitive ingredients added to highly acidic foods. For example, anthocyanin pigments exhibit different chemical forms, which is dependent on the pH of the solution (Loypimai and others 2016). Red color predominates at pH 1.0, quinoidal blue occurs when pH is between

2.0 - 4.0, and between pH 5.0 - 6.0, two colorless compounds (carbinol pseudobase and chalcone) can be detected (Loypimai and others 2016). Guan and Zhong (2015) demonstrated that incorporation of anthocyanins in GA significantly improved its stability not only in temperatures up to 126 °C, but also at slightly acidic pH of 5.0. Similarly, ß- protection at low pH was shown to significantly improve upon encapsulation in bovine serum albumin and GA (pH 2.5 -

4.0) (Sheng and others 2018). Another example, (TOC) encapsulated within zein-GA

14 nanoparticles resulted in stable structures through a wide pH range between 3.0 and 9.0 (Li and others 2018). The stability at a wide pH range improved upon addition of GA shows promising applications to both acidic and neutral beverages.

1.2.4 Digestion and absorption

GA is not digestible in the upper gastrointestinal tract, and transit to the large intestine results in slow fermentation by intestinal bacteria (Nasir and others 2008; Alarifi and others 2018).

Digestion of GA specifically increases Bifidobacterium spp and Lactobacillus spp. in healthy subjects which is why GA is considered as a prebiotic and a natural (Alarifi and others

2018). Physiological-botanical description of dietary fibers is that of “remnants of plant components that are resistant to hydrolysis by human alimentary ” (Phillips and others

2008). Since GA is fermentable, its presence in the gut leads to production of beneficial short chain fatty acids with fecal absorbability at around 95% (Mariod 2018). Thus, higher levels of acetate, propionate, and butyrate that result from GA fermentation makes it a good prebiotic

(Alarifi and others 2018).

1.2.5 Delivery and controlled release

The use of GA as an emulsifier, encapsulating wall material, or as a co-ingredient in formulations have demonstrated significant improvements in nutrient and/or bioactive absorption.

For example, in the study of Ozaki and others (2010), incorporation of coenzyme Q10 (CoQ10) in

GA by emulsification enhanced the CoQ10 bioavailability in rats and humans showing at least a 3- fold increase when compared against non-emulsified GA with CoQ10 bulk formulation. In the presence of GA, formation of mixed micelles is facilitated, and CoQ10 is solubilized reassembling itself with chylomicrons. Lipophilic bioactives interaction with chylomicrons allows its uptake within the enterocyte cells and transfer into the epithelium cells. Once this occurs, the bioactive

15 lipophilic compound can be circulated through the portal vein and transported for metabolism in the liver, resulting in enhanced bioavailability (Ozturk 2017).

In another study, GA as a co-encapsulating matrix with whey protein isolate (WPI) and tuna oil (O) for a probiotic bacterium (L. casei) provided protection for the bacterium against acidic pH conditions (pH=3.7) (Eratte and others 2015). The results of the study show that using GA-

WPI-O as a matrix significantly improved the viability of L. casei when compared to the control regardless of whether L. casei were prepared as liquid suspension or spray dried, or freeze-dried microcapsules. Eratte and others (2015) elucidated the importance of preventing the degradation or inactivation of probiotics against acidic pH which may lead to a higher potential for desired site delivery into the small intestines to receive any health promoting benefits.

The use of co-encapsulating matrices to protect probiotics may serve as a physical barrier providing additional protection in gastric pH conditions. Protection of L. casei from gastric pH conditions by using multiple encapsulating matrices including GA was previously demonstrated by Gul (2017). The study showed that a combination of GA, maltodextrin, and reconstituted skim milk resulted in more viable cell survival when compared with free cells after 60 min in simulated gastric juices (pH 2.0). Viability of free cells were completely lost after processing at gastric pH conditions, whereas, encapsulated L. casei had approximately 48 - 75% survival rate. Formulation stability was further evaluated, and encapsulated L. casei resulted in 7-8 log cfu/g viable cell count even after 21 d storage in refrigerated conditions. Thus, it is evident that incorporation of GA either alone or as a co-encapsulant provides significant improvements in protection of otherwise labile compounds.

16 1.3 Concluding remarks

The use of GA in food systems shows a promising application in delivery of nutraceuticals in beverages showing improved physicochemical and biological properties of bioactives as discussed in preceding sections. Beverages are a good medium for delivery of nutraceuticals.

Addition of lipophilic vitamins in beverages will require dispersion and storage stability across a wide pH range. Furthermore, encapsulated VD3 would have higher impact if evaluated using in vitro and in vivo models to provide clinical relevance. This will allow us to understand the overall bioavailability of encapsulated VD3 in GA and link potential health promoting effects as an added ingredient in foods and/or beverages.

1.4 Hypothesis and overview of dissertation research

The overall hypothesis for this dissertation is that encapsulation of VD3 in GA will facilitate the incorporation of VD3 in aqueous systems at a wide pH range and enhance bioavailability. Specific objectives of the present dissertation are to (1) determine optimum conditions of encapsulating VD3 in GA to maximize loading and characterize physicochemical properties, (2) determine in vitro bioaccessibility of encapsulated VD3 in optimized formulation,

-1 and (3) determine absorption of VD3 after a single high-dose (300 µg) and daily-dose (60 µg d for 2-wks) oral supplementation by measuring 25(OH)D levels in Sprague-Dawley rats.

17 References

Abbas MA. 2017. Physiological functions of Vitamin D in adipose tissue. Journal of Biochemistry and Molecular Biology. 165:369-81. Abdel-Rehim WM, El-Tahan RA, El-Tarawy MA, Shehata RR, Kamel MA. 2019. The possible antidiabetic effects of vitamin D receptors agonist in rat model of type 2 diabetes. Molecular and Cellular Biochemistry. 450(1-2):105-12. Ahmad A, Ahmed Z. 2019. 3 - Fortification in Beverages. In: Grumezescu AM, Holban AM, editors. Production and Management of Beverages: Woodhead Publishing. p. 85-122. Alarifi S, Bell A, Walton G. 2018. In vitro fermentation of gum acacia - impact on the faecal microbiota. International Journal of Food Sciences and Nutrition. 69(6):696-704. Asano Y, inventor; Asahi Inryo Kk, assignee. 2017. Manufacture of yellow peach fruit juice- containing food and drink involves mixing emulsified dye formulation of fat-soluble dye containing yellow peach straight fruit juice and gum arabic as emulsifier. JP2017070261- A. Bavisotto VS, Hansen GL. 1971. Hop extract emulsion and preparation and use thereof. Google Patents. Ben-Dov IZ, Galitzer H, Lavi-Moshayoff V, Goetz R, Kuro-o M, Mohammadi M, Sirkis R, Naveh- Many T, Silver J. 2007. The parathyroid is a target organ for FGF23 in rats. Journal of Clinical Investigation. 117(12):4003-8. Blank S, Scanlon KS, Sinks TH, Lett S, Falk H. 1995. An outbreak of hypervitaminosis-D associated with the overfortification of milk from a home-delivery dairy. American Journal of Public Health. 85(5):656-9. Boullata JI. 2010. Vitamin D supplementation: a pharmacologic perspective. Current Opinion in Clinical Nutrition and Metabolic Care. 13(6):677-84. Calvo MS, Whiting SJ. 2013. Survey of current vitamin D food fortification practices in the United States and Canada. Journal of Steroid Biochemistry and Molecular Biology. 136:211-3. Casellas F, Aparici A, Perez MJ, Rodriguez P. 2016. Perception of lactose intolerance impairs health-related quality of life. European Journal of Clinical Nutrition. 70(9):1068-72. Cashman KD, Kiely M. 2018. Chapter 63 - Vitamin D and Food Fortification. In: Feldman D, editor. Vitamin D (Fourth Edition): Academic Press. p. 109-27.

18 Chakraborty S, Sarkar AK, Bhattacharya C, Krishnan P, Chakraborty S. 2015. A Nontoxic Case of . Labmedicine. 46(2):146-9. Chen TC, Chimeh F, Lu Z, Mathieu J, Person KS, Zhang A, Kohn N, Martinello S, Berkowitz R, Holick MF. 2007. Factors that influence the cutaneous synthesis and dietary sources of vitamin D. Archives of Biochemistry and Biophysics. 460(2):213-7. Cross HS, Kallay E. 2009. Regulation of the colonic vitamin D system for prevention of tumor progression: an update. Future Oncology. 5(4):493-507. Daoub RMA, Elmubarak AH, Misran M, Hassan EA, Osman ME. 2018. Characterization and functional properties of some natural Acacia gums. Journal of the Saudi Society of Agricultural Sciences. 17(3):241-9. Daroux M, Shenouda M, Bacri JL, Lemaitre V, Vanhille P, Bataille P. 2013. Vitamin D-2 versus vitamin D-3 supplementation in hemodialysis patients: a comparative pilot study. Journal of Nephrology. 26(1):152-7. Delaurent C, Siouffi AM, Pepe G. 1998. Cyclodextrin inclusion complexes with vitamin D-3: Investigations of the solid complex characterization. Chemia Analityczna. 43(4):601-16. DeLuca HF. 2004. Overview of general physiologic features and functions of vitamin D. American Journal of Clinical Nutrition. 80(6):1689S-96S. Di Somma C, Scarano E, Barrea L, Zhukouskaya VV, Savastano S, Mele C, Scacchi M, Aimaretti G, Colao A, Marzullo P. 2017. Vitamin D and Neurological Diseases: An Endocrine View. International Journal of Molecular Sciences. 18(11):26. Diarrassouba F, Garrait G, Remondetto G, Alvarez P, Beyssac E, Subirade M. 2015a. Food protein-based microspheres for increased uptake of vitamin D-3. Food Chemistry. 173:1066-72. Diarrassouba F, Garrait G, Remondetto G, Alvarez P, Beyssac E, Subirade M. 2015b. Improved bioavailability of vitamin D3 using a beta-lactoglobulin-based coagulum. Food Chemistry. 172:361-7. Duffy SK, O'Doherty JV, Rajauria G, Clarke LC, Hayes A, Dowling KG, O'Grady MN, Kerry JP, Jakobsen J, Cashman KD, Kelly AK. 2018. Vitamin D-biofortified beef: A comparison of cholecalciferol with synthetic versus UVB-mushroom- derived ergosterol as feed source. Food Chemistry. 256:18-24.

19 Dusso AS, Brown AJ, Slatopolsky E. 2005. Vitamin D. American Journal of Physiology-Renal Physiolology. 289(1):F8-F28. Dwyer JT, Woteki C, Bailey R, Britten P, Carriquiry A, Gaine PC, Miller D, Moshfegh A, Murphy MM, Edge MS. 2014. Fortification: new findings and implications. Nutrition Reviews. 72(2):127-41. Eratte D, McKnight S, Gengenbach TR, Dowling K, Barrow CJ, Adhikari BP. 2015. Co- encapsulation and characterisation of omega-3 fatty acids and probiotic bacteria in whey protein isolate-gum Arabic complex coacervates. Journal of Functional Foods. 19:882-92. Erben M, Perez AA, Osella CA, Alvarez VA, Santiago LG. 2019. Impact of gum arabic and sodium alginate and their interactions with whey protein aggregates on bio-based films characteristics. International Journal of Biological Macromolecules. 125:999-1007. Fallon H. 2002. The Institute of Medicine and its quality of healthcare in America reports. Transactions of the American Clinical and Climatological Association. 113:119-24; discussion 25. Fang Z, Bhandari B. 2010. Encapsulation of - a review. Trends in Food Science & Technology. 21(10):510-23. FDA.gov. Changes to the Nutrition Facts Panel. [Accessed 2019 10 April] Available from: https://www.fda.gov/food/guidanceregulation/guidancedocumentsregulatoryinformation/l abelingnutrition/ucm385663.htm. Fixsen MAB, Roscoe MH. 1940. Tables of the vitamin content of human and animal foods. Nutrition Abstracts and Reviews, A and B. 9(4):795-861. Fritz JC, Halpin JL, Hooper JH, Kramke EH. 1942. Oxidative destruction of vitamin D. Industrial and Engineering Chemistry. 34:979-82. Glicksman M. 1983. Food Hydrocolloids - Gum arabic (Gum acacia). In: Glicksman M, editor. Boca Raton, FL: CRC Press, Inc. p. 8-26. Golfomitsou I, Mitsou E, Xenakis A, Papadimitriou V. 2018. Development of food grade O/W nanoemulsions as carriers of vitamin D for the fortification of emulsion based food matrices: A structural and activity study. Journal of Molecular Liquids. 268:734-42. Goncalves JG, de Braganca AC, Canale D, Shimizu MHM, Sanches TR, Moyses RMA, Andrade L, Seguro AC, Volpini RA. 2014. Vitamin D Deficiency Aggravates Chronic Kidney Disease Progression after Ischemic Acute Kidney Injury. PLoS One. 9(9):13.

20 Gropper SAS, Smith JL, Groff JL. 2009. Advanced nutrition and human metabolism. In: Smith JL, Groff JL, editors. 5th ed. Australia; United State: Wadsworth/Cengage Learning. Guan L, Liu J, Yu H, Wu G, Liu B, Tian H, Dong P, Li J, Liang XJF. 2019. Water-dispersible -rich nanopowder: Preparation, oral safety andantioxidant activity in vivo. Food and Function. 10(3):1386-97. Guan Y, Zhong Q. 2015. The improved thermal stability of anthocyanins at pH 5.0 by gum arabic. LWT-Food Science and Technology. 64(2):706-12. Gul O. 2017. Microencapsulation of Lactobacillus casei Shirota by spray drying using different combinations of wall materials and application for probiotic dairy dessert. Journal of Food Processing and Preservation. 41(5):9. Guttoff M, Saberi AH, McClements DJ. 2015. Formation of vitamin D nanoemulsion-based delivery systems by spontaneous emulsification: Factors affecting particle size and stability. Food Chemistry. 171:117-22. Hasanvand E, Fathi M, Bassiri A, Javanmard M, Abbaszadeh R. 2015. Novel starch based nanocarrier for vitamin D fortification of milk: Production and characterization. Food and Bioproducts Processing. 96:264-77. Hathcock JN, Shao A, Vieth R, Heaney RJTAjocn. 2007. Risk assessment for vitamin D. The American Journal of Clinical Nutrition. 85(1):6-18. Holick MF. 2003. Evolution and function of vitamin D. Vitamin D Analogs in Cancer Prevention and Therapy: Springer. p. 3-28. Holick MF. 2004. Sunlight and vitamin D for bone health and prevention of autoimmune diseases, cancers, and cardiovascular disease. The American Journal of Clinical Nutrition. 80(6):1678S-88S. Holick MF. 2017. The vitamin D deficiency pandemic: Approaches for diagnosis, treatment and prevention. Reviews in Endocrine and Metabolic Disorders. 18(2):153-65. Holick MF, Chen TC. 2008. Vitamin D deficiency: a worldwide problem with health consequences. The American Journal of Clinical Nutrition. 87(4):1080S-6S. Hollis BW, Wagner CL, Garrett-Mayer E, Kindy MS, Gattoni-Celli S. 2012. Vitamin D3 supplementation (4000 IU/d for 1 y) eliminates differences in circulating 25- hydroxyvitamin D between African American and white men. The American Journal of Clinical Nutrition. 96(2):332-6.

21 Hoseini R, Damirchi A, Babaei P. 2017. Vitamin D increases PPAR gamma expression and promotes beneficial effects of physical activity in metabolic syndrome. Nutrition. 36:54-9. Hu X, Liu W, Yan Y, Liu H, Huang Q, Xiao Y, Gong Z, Du J. 2019. Vitamin D protects against diabetic nephropathy: Evidence-based effectiveness and mechanism. European Journal of Pharmacology. 845:91-8. Islam AM, Phillips GO, Sljivo A, Snowden MJ, Williams PA. 1997. A review of recent developments on the regulatory, structural and functional aspects of gum arabic. Food Hydrocolloids. 11(4):493-505. Jahn S, Tsalis G, Lähteenmäki L. 2019. How attitude towards food fortification can lead to purchase intention. Appetite 133:370-7. Jayaratne N, Hughes MCB, Ibiebele TI, van den Akker S, van der Pols JC. 2013. Vitamin D intake in Australian adults and the modeled effects of milk and breakfast cereal fortification. Nutrition. 29(7):1048-53. Jones G. 2008. Pharmacokinetics of vitamin D toxicity. The American Journal of Clinical Nutrition. 88(2):582S-6S. Jones G, Strugnell SA, DeLuca HF. 1998. Current understanding of the molecular actions of vitamin D. Physiological Reviews. 78(4):1193-231. Kim JJ, Kim SS, Yoon SJ, Jung JG, Kim JS. 2014. Vitamin d status and response to initial vitamin d supplementation in korean women with osteoporosis. Journal of Bone Metabolism. 21(4):257-62. Kong H, Yang J, Zhang Y, Fang Y, Nishinari K, Phillips GO. 2014. Synthesis and antioxidant properties of gum arabic-stabilized selenium nanoparticles. International Journal of Biological Macromolecules. 65:155-62. Levinson Y, Ish-Shalom S, Segal E, Livney YD. 2016. Bioavailability, rheology and sensory evaluation of fat-free yogurt enriched with VD3 encapsulated in re-assembled casein micelles. Food and Function. 7(3):1477-82. Levinson Y, Israeli-Lev G, Livney YD. 2014. Soybean beta-Conglycinin Nanoparticles for delivery of hydrophobic nutraceuticals. Food Biophysics. 9(4):332-40. Li J, Xu X, Chen Z, Wang T, Wang L, Zhong Q. 2018. Biological macromolecule delivery system fabricated using zein and gum arabic to control the release rate of encapsulated tocopherol during in vitro digestion. Food Research International. 114:251-7.

22 Li YC, Pirro AE, Amling M, Delling G, Baroni R, Bronson R, DeMay MB. 1997. Targeted ablation of the : An animal model of vitamin D-dependent rickets type II with alopecia. Proceedings of the National Academy of Sciences. U. S. A. 94(18):9831- 5. Lips P. 2007. Relative Value of 25(OH)D and 1,25(OH)2D Measurements. Journal of Bone and Mineral Research. 22(11):1668-71. Loypimai P, Moongngarm A, Chottanom P. 2016. Thermal and pH degradation kinetics of anthocyanins in natural food colorant prepared from black rice bran. Journal of Food Science and Technology. 53(1):461-70. Luo YC, Teng Z, Wang Q. 2012. Development of Zein Nanoparticles Coated with Carboxymethyl Chitosn for Encapsulation and Controlled Release of Vitamin D3. Journal of Agricultural and Food Chemistry. 60(3):836-43. Malham M, Jorgensen SP, Lauridsen AL, Ott P, Glerup H, Dahlerup JF. 2012. The effect of a single oral megadose of vitamin D provided as either ergocalciferol (D-2) or cholecalciferol (D-3) in alcoholic liver cirrhosis. European Journal of Gastroenterology and Hepatology. 24(2):172-8. Mangoo-Karim R, Abreu JD, Yanev GP, Perez NN, Stubbs JR, Wetmore JB. 2015. Ergocalciferol versus Cholecalciferol for Nutritional Vitamin D Replacement in CKD. Nephron. 130(2):99-104. Mariod AA. 2018. 20 - Gum Arabic Dietary Fiber. In: Mariod AA, editor. Gum Arabic: Academic Press. p. 237-43. Mattila P, Ronkainen R, Lehikoinen K, Piironen V. 1999. Effect of household cooking on the vitamin D content in fish, eggs, and wild mushrooms. Journal of Food Composition and Analysis. 12(3):153-60. Mohammadi M, Pezeshki A, Abbasi MM, Ghanbarzadeh B, Hamishehkar H. 2017. Vitamin D-3- Loaded Nanostructured Lipid Carriers as a Potential Approach for Fortifying Food Beverages; in Vitro and in Vivo Evaluation. Advanced Pharmaceutical Bulletin. 7(1):61- 71. Morello M, Landel V, Lacassagne E, Baranger K, Annweiler C, Feron F, Millet P. 2018. Vitamin D Improves Neurogenesis and Cognition in a Mouse Model of Alzheimer's Disease. Molecular Neurobiology. 55(8):6463-79.

23 Munker R, Kobayashi T, Elstner E, Norman AW, Uskokovic M, Zhang W, Andreeff M, Koeffler HP. 1996. A new series of vitamin D analogs is highly active for clonal inhibition, differentiation, and induction of WAF1 in myeloid leukemia. Blood. 88(6):2201-9. Nakamura K, Nashimoto M, Okuda Y, Ota T, Yamamoto M. 2002. Fish as a major source of vitamin D in the Japanese diet. Nutrition. 18(5):415-6. Nasir O, Artunc F, Saeed A, Kambal MA, Kalbacher H, Sandulache D, Boini KM, Jahovic N, Lang F. 2008. Effects of Gum Arabic (Acacia senegal) on Water and Electrolyte Balance in Healthy Mice. Journal of Renal Nutrition. 18(2):230-8. Nazarian S, St Peter JV, Boston RC, Jones SA, Mariash CN. 2011. Vitamin D3 supplementation improves insulin sensitivity in subjects with impaired fasting glucose. Translational Research. 158(5):276-81. Oppenheimer A. 1972. Gum arabic substitute saves 25 percent. Food Engineering 44(2):103-+. Ozaki A, Muromachi A, Sumi M, Sakai Y, Morishita K, Okamoto T. 2010. Emulsification of Coenzyme Q(10) Using Gum Arabic Increases Bioavailability in Rats and Human and Improves Food-Processing Suitability. Journal of Nutritional Science and Vitaminology. 56(1):41-7. Ozturk B. 2017. Nanoemulsions for food fortification with lipophilic vitamins: Production challenges, stability, and bioavailability. European Journal of Lipid Science and Technology. 119(7). Pedersen JN, Frislev HS, Pedersen JS, Otzen DE. 2016. Using protein-fatty acid complexes to improve vitamin D stability. Journal of Dairy Science. 99(10):7755-67. Phillips GO, Ogasawara T, Ushida K. 2008. The regulatory and scientific approach to defining gum arabic (Acacia senegal and Acacia seyal) as a dietary fibre. Food Hydrocolloids. 22(1):24-35. Prietl B, Treiber G, Pieber TR, Amrein K. 2013. Vitamin D and Immune Function. Nutrients. 5(7):2502-21. Radhakishun NN, van Vliet M, Poland DC, Weijer O, Beijnen JH, Brandjes DP, Diamant M, von Rosenstiel IA. 2014. Efficacy and tolerability of a high loading dose (25,000 IU weekly) vitamin D3 supplementation in obese children with vitamin D insufficiency/deficiency. Hormone Research in Paediatrics. 82(2):103-6.

24 Randall RC, Phillips GO, Williams PA. 1989. Fractionation and characterization of gum from Acacia senegal. Food Hydrocolloids. 3(1):65-75. Roosens B, Droogmans S, Hostens J, Somja J, Delvenne E, Hernot S, Bala G, Degaillier C, Caveliers V, Delvenne P, Lahoutte T, Van Camp G, Cosyns B. 2011. Integrated Backscatter for the In Vivo Quantification of Supraphysiological Vitamin D-3-Induced Cardiovascular Calcifications in Rats. Cardiovascular Toxicology. 11(3):244-52. Ross AC, Manson JE, Abrams SA, Aloia JF, Brannon PM, Clinton SK, Durazo-Arvizu RA, Gallagher JC, Gallo RL, Jones G, Kovacs CS, Mayne ST, Rosen CJ, Shapses SA. 2011. The 2011 report on dietary reference intakes for calcium and vitamin D from the Institute of Medicine: what clinicians need to know. The Journal of Clinical Endocrinology and Metabolism 96(1):53-8. Selig MJ, Celli GB, Tan C, La E, Mills E, Webley AD, Padilla-Zakour OI, Abbaspourrad A. 2018. High pressure processing of beet extract complexed with anionic polysaccharides enhances red color thermal stability at low pH. Food Hydrocolloids. 80:292-7. Sheng BL, Li L, Zhang X, Jiao WJ, Zhao D, Wang X, Wan LT, Li B, Rong H. 2018. Physicochemical Properties and Chemical Stability of beta-Carotene Bilayer Emulsion Coated with Bovine Serum Albumin and Arabic Gum Compared to Monolayer Emulsions. Molecules. 23(2):13. Shieh A, Chun RF, Ma C, Witzel S, Meyer B, Rafison B, Swinkels L, Huijs T, Pepkowitz S, Holmquist B, Hewison M, Adams JS. 2016. Effects of High-Dose Vitamin D2 Versus D3 on Total and Free 25-Hydroxyvitamin D and Markers of Calcium Balance. The Journal of Clinical Endocrinology and Metabolism. 101(8):3070-8. Shu GF, Khalid N, Tan TB, Zhao YG, Neves MA, Kobayashi I, Nakajima M. 2018. In vitro bioaccessibility of ergocalciferol in nanoemulsion-based delivery system: the influence of food-grade emulsifiers with different stabilising mechanisms. International Journal of Food Science and Technology. 53(2):430-40. Simoneau T, Sawicki GS, Milliren CE, Feldman HA, Gordon CM. 2016. A randomized controlled trial of vitamin D replacement strategies in pediatric CF patients. Journal of Cystic Fibrosis: Official Journal of the European Cystic Fibrosis Society. 15(2):234-41. Szejtli J, Bolla E. 1980. Stabilization of fat-soluble vitamins with beta-cyclodextrin. Starke. 32(11):386-91.

25 Szejtli J, Tardy-Lengyel M. 1984. Vitamin D-3 beta cyclodextrin inclusion complex. Drugs of the Future. 9(9):675-6. Tipchuwong N, Chatraporn C, Ngamchuachit P, Tansawat R. 2017. Increasing retention of vitamin D-3 in vitamin D-3 fortified ice cream with milk protein emulsifier. International Dairy Journal. 74:74-9. Tippetts M, Martini S, Brothersen C, McMahon DJ. 2012. Fortification of cheese with vitamin D- 3 using dairy protein emulsions as delivery systems. Journal of Dairy Science. 95(9):4768- 74. Tiss A, Carrière F, Verger R. 2001. Effects of Gum Arabic on Lipase Interfacial Binding and Activity. Analytical Biochemistry. 294(1):36-43. Tripkovic L, Lambert H, Hart K, Smith CP, Bucca G, Penson S, Chope G, Hypponen E, Berry J, Vieth R, Lanham-New S. 2012. Comparison of vitamin D2 and vitamin D3 supplementation in raising serum 25-hydroxyvitamin D status: a systematic review and meta-analysis. The American Journal of Clinical Nutrition. 95(6):1357-64. Tripkovic L, Wilson LR, Hart K, Johnsen S, de Lusignan S, Smith CP, Bucca G, Penson S, Chope G, Elliott R, Hypponen E, Berry JL, Lanham-New SA. 2017. Daily supplementation with 15 mug vitamin D2 compared with vitamin D3 to increase wintertime 25-hydroxyvitamin D status in healthy South Asian and white European women: a 12-wk randomized, placebo- controlled food-fortification trial. The American Journal of Clinical Nutrition. 106(2):481- 90. Vaes AMM, Brouwer-Brolsma EM, van der Zwaluw NL, van Wijngaarden JP, Berendsen AAM, van Schoor N, van der Velde N, Uitterlinden A, Lips P, Dhonukshe-Rutten RAM, de Groot LCPGM. 2017. Food sources of vitamin D and their association with 25-hydroxyvitamin D status in Dutch older adults. The Journal of Steroid Biochemistry and Molecular Biology. 173:228-34. Wang TJ, Pencina MJ, Booth SL, Jacques PF, Ingelsson E, Lanier K, Benjamin EJ, D'Agostino RB, Wolf M, Vasan RS. 2008. Vitamin D deficiency and risk of cardiovascular disease. Circulation. 117(4):503-11. Wang TJ, Zhang F, Richards JB, Kestenbaum B, van Meurs JB, Berry D, Kiel DP, Streeten EA, Ohlsson C, Koller DL, Peltonen L, Cooper JD, O'Reilly PF, Houston DK, Glazer NL, Vandenput L, Peacock M, Shi J, Rivadeneira F, McCarthy MI, Anneli P, de Boer IH,

26 Mangino M, Kato B, Smyth DJ, Booth SL, Jacques PF, Burke GL, Goodarzi M, Cheung CL, Wolf M, Rice K, Goltzman D, Hidiroglou N, Ladouceur M, Wareham NJ, Hocking LJ, Hart D, Arden NK, Cooper C, Malik S, Fraser WD, Hartikainen AL, Zhai GJ, Macdonald HM, Forouhi NG, Loos RJF, Reid DM, Hakim A, Dennison E, Liu YM, Power C, Stevens HE, Jaana L, Vasan RS, Soranzo N, Bojunga J, Psaty BM, Lorentzon M, Foroud T, Harris TB, Hofman A, Jonsson JO, Cauley JA, Uitterlinden AG, Gibson Q, Jarvelin MR, Karasik D, Siscovick DS, Econs MJ, Kritchevsky SB, Florez JC, Todd JA, Dupuis J, Hypponen E, Spector TD. 2010. Common genetic determinants of vitamin D insufficiency: a genome-wide association study. Lancet. 376(9736):180-8. Ward FM. 2000. Uses of gum arabic (Acacia sp.) in the food and pharmaceutical industries. Cell and Developmental Biology of Arabinogalactan-Proteins: Springer. p. 231-9. Williams P, Phillips G, Randall RC. 1990. Structure–function relationships of gum arabic. Gums and stabilisers for the food industry. 5:25-36. Williams PA, Phillips G. 2009. Gum arabic. Handbook of Hydrocolloids (Second Edition): Elsevier. p. 252-73. Wuhrmann J-J, Venries B, Buri R. 1975. Process for preparing a colored powdered edible composition. Google Patents. Zhang L, Chen XD, Boom RM, Schutyser MAI. 2018a. Survival of encapsulated Lactobacillus plantarum during isothermal heating and bread baking. LWT-Food Science and Technology. 93:396-404. Zhang Y, Shan GJ, Zhang YX, Cao SJ, Zhu SN, Li HJ, Ma D, Wang DX, Investigators S. 2018b. Preoperative vitamin D deficiency increases the risk of postoperative cognitive dysfunction: a predefined exploratory sub-analysis. Acta Anaesthesiologica Scandinavica. 62(7):924-35.

27 Chapter 2 Encapsulation of vitamin D3 in gum arabic: physicochemical properties and bioaccessibility

28 2.1 Abstract

To incorporate lipophilic vitamin D (VD2 or VD3) in beverages, micro/nanoencapsulation of VD in food biopolymers stable at a wide range of pH is a potential strategy. Gum arabic (GA), a highly branched arabinogalactan with a conjugated peptide was evaluated as a potential VD3 carrier. In this study, the two objectives were to (1) determine optimum conditions of encapsulating VD3 in GA to maximize loading and characterize physicochemical properties, and

(2) determine in vitro bioaccessibility of encapsulated VD3 in optimized formulation. For 5% w/v

GA, encapsulation properties were evaluated for formulations with VD3 corresponding to 0.03,

0.06, 0.3, 0.6, 3.0, and 6.0% mass of GA. The encapsulation efficiency decreased while the loading capacity increased with the increase in VD3 amount used in encapsulation. The treatment with

VD3 corresponding to 6% GA had the highest loading capacity (3.44%) and a satisfactory encapsulation efficiency (71.35%) without visual phase separation and therefore was chosen for further characterization. As pH increased from 2.0 to 7.4, the hydrodynamic diameter decreased from 238.1 to 81.3 nm, the polydispersity index increased from 0.372 to 0.478, and the magnitude of negative zeta potential increased from -3.1 to -31.0 mV. No precipitation was observed during the 100-d storage at 3°C. Bioaccessibility of capsules was 95.76% and was significantly higher than 68.98% of the non-encapsulated VD3. The capsules exhibited excellent stability at a wide pH range and promise for delivery of VD3 in beverages.

29 2.2 Introduction

Lipophilic vitamin D (VD2 or VD3) is widely known for maintaining bone health due to its role in calcium and phosphorus homeostasis. Physiological needs of VD are fulfilled by photo- conversion of VD in the skin through sunlight exposure or adequate dietary intake consumption at

15-20 µg d-1 (Del Valle and others 2011; Ross and others 2011). However, the average VD consumption in U.S. has declined by ~ 5.3 µg capita-1 (NHANES 2009-2010), and diet sources are receiving tremendous interest to overcome increasing reports of VD insufficiency (Calvo and

Whiting 2013; Jayaratne and others 2013; Dwyer and others 2014; Lopes and others 2014; Gomes and others 2019).

To incorporate lipophilic VD in beverages, encapsulation of VD in colloidal particles is needed. Dairy proteins are frequently investigated as a matrix for encapsulation of VD including casein micelles or sodium caseinate (Markman and Livney 2012; Livney and Dalgleish 2015), whey protein isolate (WPI) (Abbasi and others 2014) or individual whey proteins such as ß- lactoglobulin (Diarrassouba and others 2015b) and α-lactalbumin (Delavari and others 2015).

Other researchers have added other materials as a co-encapsulant with dairy proteins. For example,

ß-lactoglobulin was investigated with co-encapsulating matrices of lactose (Gorska and others

2012), pectin (Zimet and Livney 2009), lysozyme (Diarrassouba and others 2015a), and ß-casein

(Forrest and others 2005). However, the use of dairy proteins has drawbacks such as the dispersion instability at a pH close to the isoelectric point of proteins and allergenicity to certain consumers.

Other biopolymers investigated for VD encapsulation include chitosan (Rabelo and others 2018) and its water-soluble derivative, carboxymethyl chitosan (Luo and others 2013) used alone or crosslinked with other compounds e.g., zein or soy protein (Luo and others 2012; Teng and others

30 2013). Chitosan is not soluble at neutral pH and is currently not approved as a generally recognized as safe (GRAS) food additive in the USA (Kumar and others 2016).

In this study, Gum arabic (GA) was evaluated as a GRAS biopolymer for encapsulation of

VD3. GA is a water-soluble biopolymer widely used in the food industry (Garti and others 1993).

Functionality of GA in foods is attributed to its structure, which consists of arabinogalactan- proteins (Akiyama and others 1984). The proteinaceous moiety of GA allows it to adsorb strongly and effectively onto oil droplets (Garti and others 1993), and the adsorption of GA on oil droplet surface results in the stability of oil-in-water emulsions at a wide pH range by means of repulsive steric and electrostatic mechanisms (Garti and others 1993). The electrostatic force of GA is attributed to the negative charge of carboxylate groups with pKa of 2.3 (Gulao and others 2016).

The amphiphilic nature of GA has been used to encapsulate and stabilize insoluble substances such as natural colorants and (Pitalua and others 2010; Khazaei and others 2014; Sanchez and others 2015; Jian and others 2017). For example, GA improved the stability and solubility of

Monascus pigments (a natural colorant) in an acidic aqueous solution at pH 3.0-5.0 (Jian and others

2017). GA adsorbed on zein particles with encapsulated tocopherols improved the stability in a wide pH range between 3.0 and 9.0 (Li and others 2018). However, GA alone has not been studied for VD3 encapsulation.

We hypothesize that GA is a feasible biopolymer to encapsulate VD3 and the capsules have physicochemical and bioaccessibility properties feasible for incorporation in beverages with a wide range of pH. The specific objectives are to (1) encapsulate VD3 using gum arabic; (2) evaluate physicochemical properties including storage stability of capsules; and (3) determine bioaccessibility of VD3 in capsules after processing in simulated saliva, gastric, and intestinal juices.

31 2.3 Materials and methods

2.3.1 Materials

Cholecalciferol (VD3) crystalline powder (99% purity), Alfa Aesar™ was obtained from

Fisher Scientific (Pittsburgh, PA). GA powder, Pretested® Gum Arabic FT, was purchased from

TIC Gums (Belcamp, MD). HPLC grade solvents including water, hexane, methanol, butanol, isooctane and ethanol were purchased from Sigma-Aldrich Corp. (St. Louis, MO).

Dimethylsulfoxide (DMSO) and reagents and chemicals of analytical grade were obtained from either Fisher Scientific (Pittsburgh, PA) or Sigma-Aldrich Corp. (St. Louis, MO). Mucin, bovine serum albumin, lipase, and pepsin were purchased from Sigma-Aldrich Corp. (St. Louis, MO).

2.3.2 Preparation of capsules

Encapsulation of VD3 was performed at room temperature (RT, 21 °C). GA solution was prepared separately at 5% (w/v) in deionized water by blending at 9700 rpm for 4 min using a shear homogenizer (Polytron® System PT 10-35 GT, Kinematica, Luzern, CH) and centrifugation at 1912 g for 10 min (RC-5B centrifuge, SH-3000 rotor, Sorvall, DuPont, DE) to remove insoluble particulates, and the supernatant was transferred into a 100-mL glass beaker. VD3 stock solution was prepared by dissolving the VD3 powder in 5 mL ethanol (200 proof) at concentrations corresponding to 0.03, 0.06, 0.3, 0.6, 3.0, and 6.0% mass of GA, named for Formulation I, II, II,

IV, V, and VI, respectively. After vortexing for 30 s, VD3 stock solution was added dropwise into the GA solution while being blended at 8,000 rpm for 3 min using the above homogenizer. The mixture was further homogenized at 10,000 rpm for 2 min before freeze-drying (Model Advantage

Plus EL-85, VirTis Company, Inc., Gardiner, NY). The freeze-dried powders were collected into scintillation glass vials and stored at -20 °C in a freezer until analysis.

32 2.3.3 Determination of total and non-bound vitamin D3 in freeze-dried powder

To quantitate the total amount of VD3 in unit mass of freeze-dried powder, 200, 30 and 10 mg of a freeze-dried powder sample was used for Formulations I and II, III and IV, and V and VI, respectively. The powder was placed in a 50 mL test tube and mixed with solvents in a two-step process. First, 5 mL of a binary solvent of DMSO and deionized water at a volume ratio of 9:1 was added, followed by heating at 60 °C for 10 min in a water-bath (model Reciprocal Shaking

Bath, Fisher Scientific, Winchester, VA). Samples were immediately cooled in an ice-bath, and

10 mL of isooctane was added, followed by stirring overnight at RT. Next, samples were transferred to centrifuge tubes and 2 mL of deionized water was added to facilitate the separation of DMSO and isooctane. Samples were left to stand at RT for 15 mins, and the mixtures were centrifuged for 10 min at 1912 g (RC-5B centrifuge, SH-3000 rotor, Sorvall, DuPont, DE). The supernatant was filtered using a polyvinylidene fluoride (0.45 μm) syringe membrane. The permeate was used to quantify VD3 using Waters 2695 Series HPLC system (Waters Corp.,

Milford, MA, USA) equipped with a Waters 2998 model diode array detector set at 265 nm and an injection valve with a 20-µL sample loop. The normal phase separation with a Waters

Spherisorb NH2 (5 µm, 4.6 x 150 mm) with guard column was conducted in isocratic mode at a flow rate of 1 mL/min using a mobile phase composed of 98% isooctane, 1% butanol, and 1% ethanol (Pastore and others 1997). Quantification was based on external standard (VD3 dissolved in isooctane) with correlation coefficient of 0.999.

The amount of free VD3 was measured according to Luo and others (2012) with minor modifications. Briefly, 130, 20, and 10 mg freeze-dried powdered was used for Formulations I and II, III and IV, and V and VI, respectively. The powder was placed on a no. 1 Whatman filter paper and flushed with 1 mL of hexane three times. The permeate was combined and dried using

33 nitrogen and re-dissolved with methanol. After filtration through a 0.45 μm polytetrafluoroethylene Whatman syringe filter (Sigma-Aldrich Corp., St. Louis, MO, USA), the permeate was analyzed with a Waters 2695 Series HPLC system (Waters Corp., Milford, MA,

USA) equipped with a Waters 2998 model diode array detector set at 265 nm and an injection valve with a 20-µL sample loop. Compounds were separated on a reversed-phase Zorbax Eclipse

Plus C18 column (4.6 mm x 150 mm, 5 µm pore size) protected by Zorbax Eclipse Plus C18 analytical guard column (Agilent Technologies, Santa Clara, CA, USA). An isocratic methanol

-1 mobile phase operating at a flow rate of 1 mL min was used to elute VD3. Quantification was based on an external standard (VD3 dissolved in methanol) with a correlation coefficient of 0.999.

2.3.4 Determination of encapsulation efficiency and loading capacity

The encapsulation efficiency (EE%) was calculated using Equation 1.

EE (%) = [1- (Free VD3 mass/Total VD3 mass used in encapsulation)] × 100 [1]

The loading capacity (LC%) was calculated based on total and free VD3 mass detected in freeze-dried powder using Equation 2.

LC (%) = [(Total VD3 mass – free VD3 mass)/Total mass of freeze-dried powder)] x 100 [2]

2.3.5 Differential scanning calorimetry

Differential scanning calorimetry (DSC) of samples was performed using a Q2000 instrument (TA Instruments, New Castle, DE). A freeze-dried sample (2.0 ± 0.05 mg) was weighed in an aluminum pan and sealed with a lid. An empty pan with lid was used as a reference.

Sample was equilibrated at 20 °C before a temperature ramp of 10 °C min-1 to 250 °C.

34 2.3.6 Physicochemical properties of capsules and storage stability of reconstituted dispersions

The physicochemical characteristics of fresh dispersion and the dispersion reconstituted with freeze-dried sample of Formulation VI were characterized in comparison to a dispersion rehydrated with freeze-dried powder prepared from the GA stock solution used in encapsulation.

The hydrodynamic diameter (Dh), polydispersity index (PDI), and zeta-potential were measured using a model Zetasizer Nano ZS series instrument (Malvern Instruments, Ltd., Worcestershire,

UK). The dynamic light scattering (DLS) experiments were conducted at a scattering angle of

173°. The fresh dispersion was adjusted to pH 2.0-7.4 using 0.1 M NaOH or 0.1 M HCl. To prepare reconstituted samples, a solution of 8 mM sodium phosphate dibasic solution with 0.01 % sodium azide was prepared and pH adjusted to 2.0-7.4 using 0.1 M NaOH or 0.1 M HCl.

Subsequently, freeze-dried capsules and GA were hydrated in the phosphate buffer at a concentration of 0.1 and 0.75 mg powder mass, respectively. All measurements were performed in triplicate at RT.

To evaluate storage stability, dispersion reconstituted with freeze-dried powder containing

0.01% sodium azide were kept at 3°C in a refrigerator. The Dh, PDI, and zeta-potential were measured periodically during storage for up to 100 days.

2.3.7 Assessment of bioaccessibility

Bioaccessibility of encapsulated VD3 in Formulation VI and free VD3 was evaluated.

Simulated saliva, gastric, and intestinal fluids (SF, GF, and IF, respectively) were prepared as described in previous reports (Oomen and others 2003; Brandon and others 2006) with modifications, outlined in Table 2.1. For the mouth phase, 6 mL of SF (pH 6.5 ± 0.2) was added to freeze-dried capsules or free VD3 powder, followed by vortexing for 30 s and incubation at 37

35 °C with continuous agitation (100 rpm) for 5 min. Second, 12 mL of GF was added, followed by vortexing for 30 s and incubation at 37 °C with continuous agitation (100 rpm) for 2 h. Third, IF represented as 12 mL of duodenal juice and 6 mL of bile juice was added with 2 mL sodium bicarbonate (84.7 g L-1), followed by vortexing for 30 s and incubation at 37 °C with continuous agitation (100 rpm) for 2 h. Lastly, to stop enzymatic activity, mixtures were incubated at 75 °C for 20 min and immediately cooled using an ice-bath. Incubation with agitation of samples was performed with a reciprocal shaking bath (Fisher Scientific, Winchester, VA).

After simulated digestions, mixtures were centrifuged at 26,892 g for 30 min (fixed angle

SS-34-10.70, Sorvall, DuPont, DE) to collect the micelle phase (supernatant). In preliminary experiments, VD3 in the micelle phase was not quantifiable using the above HPLC conditions, likely below the detection limit. Instead, the precipitate in the centrifuge flask was stored in -40

°C overnight and freeze-dried. The centrifuge flask was directly added with 5 mL of binary solvent with DMSO and deionized water at a volume ratio of 9:1, followed by heating at 60 °C for 10 min in a reciprocal shaking bath (Fisher Scientific, Winchester, VA). After immediate cooling in an ice-bath, 10 mL of isooctane was added, followed by stirring using magnetic stir plate overnight at RT. Next, samples were transferred to centrifuge tubes and 2 mL of deionized water was added to facilitate the separation of DMSO and isooctane. Samples were left to stand at RT for 15 min, and the mixtures were centrifuged for 10 min at 1912 g (RC-5B centrifuge, SH-3000 rotor, Sorvall,

DuPont, DE). The supernatant was filtered using polyvinylidene fluoride (0.45 μm) membrane filter prior to quantification using normal phase HPLC as described previously.

2.3.8 Data and statistical analysis

All experiments were conducted in triplicate with data reported as the mean ± standard deviation (SD). Statistical analysis was performed using SPSS v.25 for Windows software

36 (Chicago, IL) or SAS 9.4 for Windows 64x from SAS Institute. (Cary, NC). The significance level was set at a p value of 0.05. Analysis of the differences between treatments was performed using two-way ANOVA with Tukey’s multiple-comparison test.

2.4 Results and discussion

2.4.1 Encapsulation properties

Table 2.2 delineates total VD3, and free/unbound VD3 in freeze-dried powder that were used to determine EE and LC (Table 2.3). As the amount of VD3 used in encapsulation increased, a decline in EE (99.45 to 71.35%) and an increase in LC (0.02 to 3.44%) were observed. As GA has a small fraction (1.5-2.6%) of proteins (Idris and others 1998) to provide binding for VD3, a decrease in EE and an increase in the amount of VD3 extractable by hexane washing is expected.

Our results are comparable to some studies. VD3 encapsulated in zein nanoparticles coated with carboxymethyl chitosan had an EE of 87.9 % and an LC of 2.1 % (Luo and others 2012).

This same research group added soy protein isolate as a co-encapsulant with carboxymethyl chitosan to protect VD3 which resulted in an EE of 84.46% and an LC of 6.06% (Teng and others

2013). Other matrices used to encapsulate VD3 were reported for much higher EE, including fish oil (EE% = 96.95%) (Walia and others 2017), nanoliposomes (EE% = 93%) (Mohammadi and others 2017), niosomes (EE% = 95.9%) (Wagner and others 2016), and carboxymethyl chitosan beads (EE% = 96.9%) (Luo and others 2013). However, drawbacks using these matrices include the incompatibility with acid-based beverage systems, turbidity, and suitability and label friendliness for beverage applications.

Figure 2.1 shows the thermal properties from DSC analysis of non-encapsulated VD3 and

GA with corresponding melting point of 81.4 and 160 °C, respectively. The DSC thermogram of

VD3 is in agreement with Tsai and others (2017). The thermograms of freeze-dried powder

37 produced with 6 formulations of different VD3-GA amounts are shown in Figure 2.2. All 6 formulations showed a disappearance of the VD3 melting point with only the GA melting point observed at 160 °C which suggest an inclusion complex was formed.

Although the formulation VI had the lowest EE, it had the highest VD3 LC and had characteristics for beverage applications, as presented below. The high LC provides convenience in sample preparation when a relatively large amount of VD3 is needed, e.g., for animal feeding studies. Therefore, this formulation was characterized for physicochemical properties and bioaccessibility.

2.4.2 Physicochemical properties of capsules before and after freeze-drying

The Dh is shown in Figure 2.3 for fresh dispersions and 1 h after reconstituting with capsules or GA at pH 2.0-7.4. The Dh was independent on pH (p > 0.05). The fresh dispersion had smaller Dh (276.6 nm) than the reconstituted capsule dispersion (304.1 nm), which suggests incomplete hydration of capsules in less than 1 h. The bigger Dh of fresh and reconstituted capsule dispersions than that of GA (38.8 nm) suggests the encapsulation of VD3 increased particle dimension. The incomplete hydration of freeze-dried powder was also indicated by the relatively large PDI of capsules and GA (Figure 2.4), indicating heterogeneity in particle size distribution

(Navideh and others 2012). The PDI range of capsules observed in our study is comparable to the

PDI of astaxanthin stabilized by GA (Navideh and others 2012).

The zeta-potential is shown in Figure 2.5. The zeta-potential measurements for all 3 samples was negative over the pH range studied (2.0 – 7.4) and had a bigger magnitude at a higher pH. This is primarily attributed to the ionizable side chain carboxylate group of the GA molecule

(Islam and others 1997; Weinbreck and others 2003). Both pH and ionic strength of solutions influence stability of capsules dispersed in solutions (Weinbreck and others 2003). Addition of

38 acid lowers the surface charge of the GA molecule due to protonation of carboxylate groups resulting in a decrease in electrostatic repulsion between particles (Liu and others 2010).

Increasing the ionic concentration of dispersions screens charges and weakens electrostatic repulsion, leading to aggregation of particles (Ozturk and others 2015; Zha and others 2019).

Therefore, a balance between pH and ionic strength of solution needs to be maintained for capsule stability. In our study, the zeta-potential magnitude is small at low pH values, but no precipitation was observed. This is attributed to both electrostatic repulsion and steric hindrance resulting from polysaccharide chains of GA on particle surface preventing particle aggregation (Liu and others

2010; Guan and Zhong 2015; Zha and others 2019).

2.4.3 Physicochemical properties during 100 d storage at 3 °C

The GA control had a consistent Dh of 39 nm at pH 2.0 to 7.4 throughout 100 d storage

(Figure 2.6), which is in good agreement with other studies (Abdolkhalegh and others 2018; Lv and others 2019). The larger errors on Dh day 0 than later days, e.g., day 20, suggest freeze-dried powder was continuously hydrated during storage. On the same day, the Dh overall decreased as the pH was increased, which was most evident on day 100 showing Dh of 238.1 nm and 81.3 nm at pH 2.0 and 7.4, respectively.

The bigger PDI of capsules at a higher pH suggested the more polydispersed systems as pH increased (Figure 2.7). For example, the PDI was 0.649 and 0.376 at a pH of 7.4 and 2.0, respectively. The magnitude of zeta-potential of capsules increased from absolute -3.09 to -40.58 mV as pH increased from 2.0 to 7.4 during the 100-d storage (Figure 2.8). For all treatments, no precipitation was observed throughout the storage, and particle aggregation was not apparent

(Figure 2.6). It is widely accepted that stable emulsions exhibit a zeta-potential magnitude greater than 25 mV (Lamba and others 2015). As discussed previously, the stability of capsule dispersions

39 during storage is similarly attributed to repulsive electrostatic and steric forces originating from

GA molecules (Zhang and others 2015; Ozturk 2017).

2.4.4 Bioaccessibility of vitamin D3 encapsulated in gum arabic

The amount of VD3 that was bioaccessible after processing in simulated saliva, gastric, and intestinal juices was indirectly measured by evaluating the precipitate instead of the micelle portion. Processing of the micelle portion would require freeze-drying ~48 mL of digestive juices prior to quantification using HPLC. The higher volume resulted to more sample loss after freeze- drying due to transfer of contents for the multi-step extraction process. The loss was attributed to the dried-up micelles which adheres to the container walls, making it difficult to effectively transfer contents without significant sample loss. Therefore, it was more feasible to freeze-dry the precipitate instead of the micelle portion to prevent sample loss. Furthermore, solvent extraction using the binary solvent composed of DMSO and water at a 9:1 ratio was more effective in extracting VD3 from the precipitate because the sample mass of the precipitate was significantly less than that of the micelle portion.

In Table 2.4, the VD3 recovered from the precipitate of capsules after simulated digestions was significantly less than that of the non-encapsulated VD3 (4.2 and 31.0%, respectively). Thus, it is possible, that 95.8% of VD3 in capsules would have been bioaccessible, in contrast to only

69% in the non-encapsulated form. The results suggest VD3 encapsulated in GA may have a significantly higher bioavailability after digestion, which is to be validated in vivo.

2.5 Conclusions

In this study, we successfully encapsulated VD3 using GA as an encapsulation matrix. At chosen conditions, a high LC of 3.44% and a moderate EE of 71.35% were observed. These capsules exhibited good stability across a pH range between 2.0 and 7.4, attributed to a generally

40 small Dh and molecular characteristics of GA providing strong repulsive forces preventing particle aggregation during 100-d refrigerated storage. The higher bioaccessibility of encapsulated VD3 in

GA than the non-encapsulated form further suggests the benefit of encapsulation. Overall, the studied encapsulation system holds great potential as a value-added ingredient to supplement lipophilic VD3 in beverages with a wide pH range.

2.6 Acknowledgments

I would like to acknowledge and thank the Department of Food Science, the University of

Tennessee for financially supporting this study. Assistance of Dr. Philipus Pangloli is greatly appreciated in preparation of some of the simulated juices used in the in vitro study, and his assistance in HPLC methods.

41 References

Abbasi A, Emam-Djomeh Z, Mousavi MAE, Davoodi D. 2014. Stability of vitamin D-3 encapsulated in nanoparticles of whey protein isolate. Food Chemistry. 143:379-83. Abdolkhalegh G, Masoud Taghavi S, Aghili Dehnavi F. 2018. The emulsifying properties of Persian gum (Amygdalus scoparia Spach) as compared with gum Arabic. International Journal of Food Properties. 21(1):416-36. Akiyama Y, Eda S, Kato K. 1984. Gum arabic is a kine of arabinogalactan protein. Agricultural and Biological Chemistry. 48(1):235-7. Brandon EFA, Oomen AG, Rompelberg CJM, Versantvoort CHM, van Engelen JGM, Sips A. 2006. Consumer product in vitro digestion model: Bio accessibility of contaminants and its application in risk assessment. Regulatory Toxicology and Pharmacology. 44(2):161- 71. Calvo MS, Whiting SJ. 2013. Survey of current vitamin D food fortification practices in the United States and Canada. Journal of Steroid Biochemistry and Molecular Biology. 136:211-3. Del Valle HB, Yaktine AL, Taylor CL, Ross AC. 2011. Dietary reference intakes for calcium and vitamin D: National Academies Press. Institute of Medicine of the National Academies. Washington, D.C. Delavari B, Saboury AA, Atri MS, Ghasemi A, Bigdeli B, Khammari A, Maghami P, Moosavi- Movahedi AA, Haertle T, Goliaei B. 2015. Alpha-lactalbumin: A new carrier for vitamin D-3 food enrichment. Food Hydrocolloids. 45:124-31. Diarrassouba F, Garrait G, Remondetto G, Alvarez P, Beyssac E, Subirade M. 2015a. Food protein-based microspheres for increased uptake of vitamin D-3. Food Chemistry. 173:1066-72. Diarrassouba F, Garrait G, Remondetto G, Alvarez P, Beyssac E, Subirade M. 2015b. Improved bioavailability of vitamin D3 using a beta-lactoglobulin-based coagulum. Food Chemistry. 172:361-7. Dwyer JT, Woteki C, Bailey R, Britten P, Carriquiry A, Gaine PC, Miller D, Moshfegh A, Murphy MM, Edge MS. 2014. Fortification: new findings and implications. Nutrition Reviews. 72(2):127-41.

42 Forrest SA, Yada RY, Rousseau D. 2005. Interactions of vitamin D-3 with bovine beta- lactoglobulin A and beta-casein. Journal of Agricultural and Food Chemistry. 53(20):8003- 9. Garti N, Reichman D, Hendrickx H, Dickinson E, Jackson LK, Bergenstahl B. 1993. Hydrocolloids as food emulsifiers and stabilizers. Food Structure. 12(4):411-26. Gomes TLN, Fernandes RC, Vieira LL, Schincaglia RM, Mota JF, Nóbrega MS, Pichard C, Pimentel GD. 2019. Low vitamin D at ICU admission is associated with cancer, infections, acute respiratory insufficiency, and liver failure. Nutrition. 60:235-40. Gorska A, Ostrowska-Ligeza E, Szulc K, Wirkowska M. 2012. A differential scanning calorimetric study of beta-lactoglobulin and vitamin D-3 complexes. Journal of Thermal Analysis and Calorimetry. 110(1):473-7. Guan Y, Zhong Q. 2015. The improved thermal stability of anthocyanins at pH 5.0 by gum arabic. LWT-Food Science Technology. 64(2):706-12. Gulao ED, de Souza CJF, Andrade CT, Garcia-Rojas EE. 2016. Complex coacervates obtained from peptide leucine and gum arabic: Formation and characterization. Food Chemistry. 194:680-6. Idris OHM, Williams PA, Phillips GO. 1998. Characterisation of gum from Acacia senegal trees of different age and location using multidetection gel permeation chromatography. Food Hydrocolloids. 12(4):379-88. Islam AM, Phillips GO, Sljivo A, Snowden MJ, Williams PA. 1997. A review of recent developments on the regulatory, structural and functional aspects of gum arabic. Food Hydrocolloids. 11(4):493-505. Jayaratne N, Hughes MCB, Ibiebele TI, van den Akker S, van der Pols JC. 2013. Vitamin D intake in Australian adults and the modeled effects of milk and breakfast cereal fortification. Nutrition. 29(7):1048-53. Jian W, Sun Y, Wu J. 2017. Improving the water solubility of Monascus pigments under acidic conditions with gum arabic. Journal of the Science of Food and Agriculture. 97(9):2926- 33. Khazaei KM, Jafari SM, Ghorbani M, Kakhki AH. 2014. Application of maltodextrin and gum Arabic in microencapsulation of saffron petal's anthocyanins and evaluating their storage stability and color. Carbohydrate Polymers. 105:57-62.

43 Kumar A, Vimal A, Kumar A. 2016. Why Chitosan? From properties to perspective of mucosal drug delivery. International Journal of Biological Macromolecules. 91:615-22. Lamba H, Sathish K, Sabikhi L. 2015. Double Emulsions: Emerging Delivery System for Plant Bioactives. Food and Bioprocess Technology. 8(4):709-28. Li J, Xu X, Chen Z, Wang T, Wang L, Zhong Q. 2018. Biological macromolecule delivery system fabricated using zein and gum arabic to control the release rate of encapsulated tocopherol during in vitro digestion. Food Research International. 114:251-7. Liu S, Elmer C, Low NH, Nickerson MT. 2010. Effect of pH on the functional behaviour of pea protein isolate–gum Arabic complexes. Food Research International. 43(2):489-95. Livney YD, Dalgleish DG. 2007. Casein micelles for nanoencapsulation of hydrophobic compounds. Patent 2649788 A1. Lopes JB, Fernandes GH, Takayama L, Figueiredo CP, Pereira RMR. 2014. A predictive model of vitamin D insufficiency in older community people: From the Sao Paulo Aging & Health Study (SPAH). Maturitas. 78(4):335-40. Luo Y, Teng Z, Wang X, Wang Q. 2013. Development of carboxymethyl chitosan hydrogel beads in alcohol-aqueous binary solvent for nutrient delivery applications. Food Hydrocolloids. 31(2):332-9. Luo YC, Teng Z, Wang Q. 2012. Development of Zein Nanoparticles Coated with Carboxymethyl Chitosn for Encapsulation and Controlled Release of Vitamin D3. Journal of Agricultural and Food Chemistry. 60(3):836-43. Lv SS, Zhang YH, Tan HY, Zhang RJ, McClements DJ. 2019. Encapsulation within Oil-in-Water Emulsions: Impact of Emulsifier Type on Physicochemical Stability and Bioaccessibility. Journal of Agricultural and Food Chemistry. 67(5):1521-9. Markman G, Livney YD. 2012. Maillard-conjugate based core-shell co-assemblies for nanoencapsulation of hydrophobic nutraceuticals in clear beverages. Food and Function. 3(3):262-70. Mohammadi M, Pezeshki A, Abbasi MM, Ghanbarzadeh B, Hamishehkar H. 2017. Vitamin D-3- Loaded Nanostructured Lipid Carriers as a Potential Approach for Fortifying Food Beverages; in Vitro and in Vivo Evaluation. Advanced Pharmaceutical Bulletin. 7(1):61- 71.

44 Navideh A, Tan C, Nehdi IA, Ling T. 2012. Colloidal astaxanthin: preparation, characterisation and bioavailability evaluation. Food Chemistry. 135(3):1303-9. NHANES. What we eat in America. 2009-2010. [Accessed 2019 07 March] Available from: https://www.ars.usda.gov/ARSUserFiles/80400530/pdf/0910/tables_1-40_2009-2010.pdf Oomen AG, Rompelberg CJM, Bruil MA, Dobbe CJG, Pereboom D, Sips A. 2003. Development of an in vitro digestion model for estimating the bioaccessibility of soil contaminants. Archives of Environmental Contamination and Toxicology. 44(3):281-7. Ozturk B. 2017. Nanoemulsions for food fortification with lipophilic vitamins: Production challenges, stability, and bioavailability. European Journal of Lipid Science and Technology. 119(7). Ozturk B, Argin S, Ozilgen M, McClements DJ. 2015. Formation and stabilization of nanoemulsion-based vitamin E delivery systems using natural biopolymers: Whey protein isolate and gum arabic. Food Chemistry. 188:256-63. Pastore RJ, Dunnett RV, Webster GK. 1997. Dimethyl sulfoxide extraction method for the liquid chromatographic analysis of microencapsulated vitamin D-3. Journal of Agricultural and Food Chemistry. 45(5):1784-6. Pitalua E, Jimenez M, Vernon-Carter EJ, Beristain CI. 2010. Antioxidative activity of microcapsules with beetroot juice using gum Arabic as wall material. Food and Bioproducts Processing. 88(C2-3):253-8. Rabelo RS, Oliveira IF, da Silva VM, Prata AS, Hubinger MD. 2018. Chitosan coated nanostructured lipid carriers (NLCs) for loading Vitamin D: A physical stability study. International Journal of Biological Macromolecules. 119:902-12. Ross AC, Manson JE, Abrams SA, Aloia JF, Brannon PM, Clinton SK, Durazo-Arvizu RA, Gallagher JC, Gallo RL, Jones G, Kovacs CS, Mayne ST, Rosen CJ, Shapses SA. 2011. The 2011 report on dietary reference intakes for calcium and vitamin D from the Institute of Medicine: what clinicians need to know. The Journal of Clinical Endocrinology and Metabolism. 96(1):53-8. Sanchez V, Baeza R, Chirife J. 2015. Comparison of monomeric anthocyanins and colour stability of fresh, concentrate and freeze-dried encapsulated cherry juice stored at 38 degrees C. Journal of Berry Research. 5(4):243-51.

45 Teng Z, Luo Y, Wang Q. 2013. Carboxymethyl chitosan-soy protein complex nanoparticles for the encapsulation and controlled release of vitamin D-3. Food Chemistry. 141(1):524-32. Tsai SY, Lin HY, Hong WP, Lin CP. 2017. Evaluation of preliminary causes for vitamin D series degradation via DSC and HPLC analyses. Journal of Thermal Analysis and Calorimetry 130(3):1357-69. Wagner ME, Spoth KA, Kourkoutis LF, Rizvi SSH. 2016. Stability of niosomes with encapsulated vitamin D-3 and ferrous sulfate generated using a novel supercritical carbon dioxide method. Journal of Liposome Research. 26(4):261-8. Walia N, Dasgupta N, Ranjan S, Chen L, Ramalingam C. 2017. Fish oil based vitamin D nanoencapsulation by ultrasonication and bioaccessibility analysis in simulated gastro- intestinal tract. Ultrasonics Sonochemistry. 39:623-35. Weinbreck F, de Vries R, Schrooyen P, de Kruif CG. 2003. Complex Coacervation of Whey Proteins and Gum Arabic. Biomacromolecules. 4(2):293-303. Zha FC, Dong SY, Rao JJ, Chen BC. 2019. Pea protein isolate-gum Arabic Maillard conjugates improves physical and oxidative stability of oil-in-water emulsions. Food Chemistry. 285:130-8. Zhang J, Peppard TL, Reineccius GA. 2015. Preparation and characterization of nanoemulsions stabilized by food biopolymers using microfluidization. Flavour Fragrance Journal. 30(4):288-94. Zimet P, Livney YD. 2009. Beta-lactoglobulin and its nanocomplexes with pectin as vehicles for omega-3 polyunsaturated fatty acids. Food Hydrocolloids. 23(4):1120-6.

46 Appendix

Table 2.1. Compositions of simulated saliva and gastrointestinal juices adapted from Brandon and others (2006) and Oomen and others (2003).

Saliva juice Gastric juice Duodenal juice Bile juice Inorganic phase (500 mL) 896 ± 4 mg KCl 2,752 ± 14 mg NaCl 7,012 ± 35 mg NaCl 5,259 ± 26 mg NaCl 200 ± 4 mg KSCN 266 ± 3 mg NaH2PO4 3,388 ± 17 mg NaHCO3 5785 ± 29 mg NaHCO3 888 ± 4 mg NaH2PO4 824 ± 8 mg KCl 80 ± 4 mg KH2PO4 376 ± 4 mg KCl 570 ± 6 mg Na2SO4 400 ± 4 mg CaCl2.2H2O 564 ± 6 mg KCl 150 µL 37% HCl 298 ± 3 mg NaCl 306 ± 4 mg NH4Cl 50 ± 5 mg MgCl2.6H2O 1,694 ± 8 mg NaHCO3 6.5 mL 37% HCl 180 µL 37% HCl Organic phase (500 mL) 200 ± 2 mg urea 650 ± 7 mg glucose 100 ± 5 mg urea 250 ± 2.5 mg urea 20 ± 2 mg glucuronic acid 85 ± 1.5 mg urea 330 ± 3 mg glucosamine HCl Combine organic and inorganic phase in corresponding digestive juices and add the following components:

290 ± 3 mg amylase 1 ± 0.005 g BSA 200 ± 4 mg CaCl2.2H2O 222 ± 2 mg CaCl2.2H2O 15 ± 1.5 mg uric acid 2 ± 0.13 g pepsin 1 ± 0.005 g BSA 1.8 ± 0.009 g BSA 25 ± 2 mg mucin 3 ± 0.015 g mucin 9 ± 0.045 g pancreatin 30 ± 0.03 g bile 1.5 ± 0.0075 g lipase pHa 6.8 ± 0.1 1.3 ± 0.1 8.1 ± 0.1 8.2 ± 0.1 a Adjust pH with 1.0 M NaOH or 37% HCl.

47 Table 2.2. Total and free vitamin D3 in freeze-dried powder. *

Formulation Mass% of VD3 with respect to VD3% with respect to total mass used in encapsulation gum arabic Total detected Free/unbound

Control Non-encapsulated 98.1 ± 7.16ab vitamin D3

I 0.03 81.7 ± 0.26bc 0.9 ± 0.17b

II 0.06 88.3 ± 0.42b 0.5 ± 0.10b

III 0.3 95.6 ± 0.52ab 0.7 ± 0.12b

IV 0.6 93.5 ± 1.62ab 4.3 ± 0.23b

V 3.0 83.4 ± 1.45bc 20.8 ± 3.13a

VI 6.0 78.3 ± 2.95c 25.9 ± 10.85a

*Different superscript letters show significant difference within the column based on two-way ANOVA (P < 0.05).

48 Table 2.3. Encapsulation efficiency (EE%) and loading capacity (LC%) of treatments with VD3 used at 0.03-6.00 % of mass of gum arabic (GA).*

Formulation VD3 amount in mass% of GA EE% LC%

I 0.03 99.45 ± 0.11 a 0.02 ±0.00 d II 0.06 99.65 ± 0.06 a 0.05 ±0.00 d III 0.30 99.75 ± 0.04 a 0.29 ±0.00 cd IV 0.60 98.53 ± 0.07 a 0.55 ± 0.01 c V 3.00 86.62 ± 6.40 b 2.29 ± 0.05 b VI 6.00 71.35 ± 5.84 c 3.44 ± 0.38 a *Mean ± SD. Different letters represent significant difference within the column (P < 0.05).

49 A)

81.4 °C Vitamin D3 Melting Point

B)

160 °C Gum Arabic Melting Point

Figure 2.1. Differential scanning calorimetry (DSC) thermograms of A) non-encapsulated VD3, and B) gum arabic.

50 A) D)

81.4 °C Melting Point Non-observed (A to F) 160 °C Melting Point (A to F)

B) E)

C) F)

Figure 2.2. DSC thermograms of capsules prepared with VD3 at A) 0.03, B) 0.06, C) 0.3, D) 0.6, E) 3.0, and F) 6.0% mass of gum arabic.

51

Figure 2.3. Mean hydrodynamic diameter (Dh) of fresh dispersion with VD3 corresponding to 6% mass of gum arabic, dispersion reconstituted with the corresponding freeze-dried powder, and gum arabic at pH 2.0-7.4. Error bars are SD (n=3). Different letters above bars indicate statistical difference among samples at the same pH (P < 0.05).

52

Figure 2.4. Polydispersity index (PDI) of fresh dispersion with VD3 corresponding to 6% mass of gum arabic, dispersion reconstituted with the corresponding freeze-dried powder, and gum arabic at pH 2.0-7.4. Error bars are SD (n=3). Different letters above bars indicate statistical difference among samples at the same pH (P < 0.05).

53

Figure 2.5. Zeta potential of fresh dispersion with VD3 corresponding to 6% mass of gum arabic, dispersion with the corresponding freeze-dried powder, and gum arabic at pH 2.0-7.4. Error bars are SD (n=3).

54 A)

B)

Figure 2.6. Mean hydrodynamic diameter (Dh) of A) gum arabic, and B) dispersion reconstituted with freeze-dried powder prepared with VD3 corresponding to 6% mass of gum arabic during 100-day storage at 3 °C. Error bars are SD (n=3).

55 A)

B)

Figure 2.7. Polydispersity index (PDI) of A) gum arabic, and B) dispersion reconstituted with freeze-dried powder prepared with VD3 corresponding to 6% mass of gum arabic during 100-day storage at 3 °C. Error bars are SD (n=3).

56 A)

B)

Figure 2.8. Zeta-potential of A) gum arabic, and B) dispersion reconstituted with freeze-dried powder prepared with VD3 corresponding to 6% mass of gum arabic during 100-day storage at 3 °C. Error bars are SD (n=3).

57 Table 2.4. Vitamin D3 detected in the precipitate after simulated mouth, gastric and intestinal phase digestions.

Sample Vitamin D3 recovered in precipitate (%)*

a Non-encapsulated vitamin D3 31.0 ± 0.8

b Capsules prepared with vitamin D3 at 6% mass of gum arabic 4.2 ± 0.4

* Different superscript letters indicate significant differences at P < 0.05.

58 Chapter 3 Encapsulation of vitamin D3 in gum arabic: in vivo bioavailability in Sprague-

Dawley rats

59 3.2 Abstract

Vitamin D3 (VD3) is a lipophilic hormone precursor and improving its incorporation in water by encapsulation in colloidal particles may address its nutritional significance for VD3 deficient consumers. The aim of this study was to evaluate the efficacy of an encapsulation system previously developed in enhancing the bioavailability of VD3. We hypothesize that encapsulation in gum arabic (GA) will improve the absorption of VD3 when compared to unencapsulated VD3.

The objective was to determine absorption of VD3 after a single high-dose (300 µg) and daily-dose

(60 µg/d for 2-wks) oral supplementation by measuring 25(OH)D levels in Sprague-Dawley rats.

The rats at weaning age were randomly assigned to one of 6 groups (n=10/group, 60 rats total, male and female equally distributed). Rats received either a vitamin D sufficient (VDS) (2200

IU/kg diet) or a vitamin D deficient (VDD) (0 IU/kg diet) pellet food and water ad libitum. The two control groups were in a VDS or a VDD group given an oral gavage of GA alone. The four treatment groups included rats fed with a VDD diet and gavaged with non-encapsulated VD3, a physical mixture of GA and VD3, or encapsulated VD3, and rats fed with a VDS and gavaged with encapsulated VD3. The 25(OH)D level in serum was determined by competitive enzyme-linked immunosorbent assay. The in vivo pharmacokinetic study results after oral gavage of 300 µg VD3 showed that the serum 25(OH)D level area-under-curve in 48 h of the encapsulation treatment was

4.32-fold of the non-encapsulated VD3 and more than twice higher than the treatment with simple mixture of GA and VD3. At 2-wks after the oral gavage of 60 µg VD3/d, rats that received capsules or physical mixture had 25(OH)D levels that were at least 81 ng mL-1 higher than that of the non- encapsulated VD3 group. The results of this study show encapsulating VD3 in GA is an effective strategy to improve absorption of lipophilic VD3.

60 3.3 Introduction

Foods enriched and/or fortified with vitamin D (VD) e.g., milk, orange juice, margarine, and cereal contribute to a proportion of VD intake; but food choice patterns influence whether people can get adequate VD (Dwyer and others 2014). Consequently, recommended levels of

Adequate Intake (AI) of VD for adults at 600 – 800 IU/day or 15 – 20 µg/day is not being met resulting to decreased levels of serum 25-hydroxyvitamin D [25(OH)D], also known as calcidiol, a status biomarker (Holick and Chen 2008; Holick 2017). Therefore, there is a desired need to obtain nutrients from fortified food products, and for busy, on-the go consumers a continued demand for convenient foods (Zimet and Livney 2009; Casey and others 2018). A potential strategy is to encapsulate VD in a food grade water-soluble matrix to increase its solubilization in aqueous foods. Furthermore, encapsulated VD must be stable against gastric pH degradation

(Cohen and others 2017) for target delivery in the small intestines where VD is absorbed.

Determining oral VD uptake by pharmacokinetic and daily dose studies are linked to uptake and retention in target tissues (Chow and others 2013; Apaydin and others 2018; Neme and others 2018). Oral administration of encapsulated VD is evaluated by assessing serum 25(OH)D levels because it represents the best indicator of VD status not only from food, but also from supplements or endogenous production of VD from sunlight (Norman 2008). There are limited studies that systematically assessed efficacy of encapsulated VD in vivo in animal (Diarrassouba and others 2015) or human subject (Levinson and others 2016). In Wistar rats, VD3 prepared with bovine ß-lactoglobulin (ß-Lg) with a final VD3 concentration of 77 µg (3080 IU/d) (1 IU cholecalciferol = 40 µg) as a coagulum was administered every day for 3-wks via oral gavage

(Diarrassouba and others 2015). After 3-wks, VD3 in the mixture was better absorbed when compared with non-encapsulated VD3 (42.4 vs. 15.1 µM serum 25(OH)D) (Diarrassouba and

61 others 2015). In humans, bioavailability of 1,250 µg or 50,000 IU encapsulated VD3 in re- assembled casein micelles (VD3- rCMS) and/or in polysorbate-80 (PS80/Tween-80) (VD3-PS80), a commonly used synthetic emulsifier, was evaluated in fat-free yogurt and compared against a control group receiving a VD3-free placebo control (Levinson and others 2016). Results showed

-1 mean serum ∆25(OH)D levels increased similarly (~8 ng ml ) for both VD3- rCMS and VD3-PS80 against the VD3-free placebo group (Levinson and others 2016). Nevertheless, the encapsulated

VD3 ingredients (Diarrassouba and others 2015; Levinson and others 2016) are targeted for applications in semi-solid foods e.g., cheese, yogurt due to rheological impact on final product e.g., increase in viscosity and elasticity. Therefore, improvements in VD3 encapsulation for delivery in other food systems with enhanced bioavailability are needed.

Gum arabic (GA) may be used as a potential VD3 carrier in food systems. GA is a hydrocolloid derived from exudates of Acacia senegal or Acacia seyal trees widely used in the food and pharmaceutical industry as an emulsifier and stabilizer (Nasir and others 2008; Phillips and others 2008). GA as an emulsifier has both hydrophilic and lipophilic portions which promote the mixing of lipids in water (Rolfes and others 2014) which may facilitate the incorporation of

VD in aqueous systems. Emulsification of fats with bile is an important step that precedes absorption. Bile which is stored in the gallbladder is not an enzyme but contains emulsifiers that brings lipids into suspension in water to facilitate enzymes (pancreatic and intestinal lipases) to hydrolyze tri- and di- acylglycerols (Rolfes and others 2014). These hydrolyzed lipids dissolving lipophilic nutrients will then be absorbed within the small intestinal wall (Insel and Wardlaw 1996;

Gropper 2009). These nutrients then undergo transport via vessels that merge into the portal vein, which conducts all absorbed materials into the liver (Insel and Wardlaw 1996; Gropper 2009).

62 GA alone and/or used as a co-encapsulant with another matrix increased the bioavailability of lipophilic compounds (Zhou and others 2018). For example, droplets with esterified astaxanthin emulsified within whey protein/GA showed significantly higher concentrations of astaxanthin in plasma with a peak at 9 h (tmax) after ingestion when compared with esterified astaxanthin oleoresin

(control) with a tmax of 12 h (Zhou and others 2018). Similarly, encapsulation of co-enzyme Q10

(CoQ10) with GA significantly improved its bioavailability when compared to non-emulsified

CoQ10 bulk powder (Ozaki and others 2010). Specifically, oral administration of emulsified CoQ10 showed significantly greater serum and/or plasma CoQ10 levels with tmax of 2 and 6 h for rats and humans, respectively; whereas, the CoQ10 bulk powder control showed the absence of peak in serum and/or plasma CoQ10 levels (Ozaki and others 2010). GA has also been studied as a wall material to protect lipophilic nutraceuticals e.g., ß-carotene, , and and improve physicochemical properties (Montenegro and others 2007; Qv and others 2011; Boiero and others

2014); however, bioavailability via in vivo studies was not assessed in these studies.

In Chapter 2, VD3 was incorporated successfully in GA as particulates with a hydrodynamic diameter (Dh) of 242.54 nm, and particles remained dispersed in a wide pH range between 2.0 to 7.4 during 100-day storage at 3 °C. Additionally, bioaccessibility of VD3 after simulated mouth, gastric and intestinal phase digestions showed that 95.81% of encapsulated VD3 could be potentially absorbed, which was significantly higher than 68.98% of non-encapsulated

VD3. These preliminary studies suggest the feasibility of the studied food grade capsules to enhance VD3 bioavailability with promising applications in beverage systems.

Our research hypothesis is that encapsulation of VD3 in GA will enhance absorption of

VD3. The objective is to investigate the bioavailability of VD3 encapsulated in GA using Sprague-

Dawley rats after a single high-dose of 300 µg and a two-week daily lower dose of 60 µg/day.

63 This study marks the first in vivo investigation into the oral delivery performance of VD3 encapsulated in GA.

3.4 Materials and methods

3.4.1 Materials

Cholecalciferol, VD3, crystalline powder (99% purity), Alfa Aesar™ was obtained from

Fisher Scientific (Pittsburgh, PA). TIC Pretested® Gum Arabic FT Powder was purchased from

TIC Gums (Belcamp, MD). Propylene glycol (99.9% pure USP Food Grade) was sourced from

Amazon (Froggys Fog brand). Sodium phosphate dibasic and citric acid were obtained from

Sigma-Aldrich Corp. (St. Louis, MO). Isoflurane was provided by the Institutional Animal Care and Use Committee at the University of Tennessee, Knoxville (UTK-IACUC) for use in animal study. The 2% chlorhexidine antiseptic solution sourced from a local veterinary clinic was used to wipe tail of rats to prevent infection and to clean the wound after blood withdrawal. Sucrose was obtained from a local grocery store.

3.4.2 Animals and diets

All procedures were approved by the UTK-IACUC (protocol numbers: 2123 and 2547).

Treatments of animals were in accordance with the Guide to the Care and Use of Experimental

Animals. For protocol# 2123, three female and two male Sprague-Dawley (SD) rats (7 weeks of age) were purchased from Envigo (Envigo, Indianapolis, IN) and used for breeding. The SD rats underwent a one-week acclimation period. After acclimation, one male and no more than 2 female rats were grouped per breeding cage. Environmental conditions in the animal facility had an average daily temperature of 21 ± 3 °C, and humidity range of 45-65%, and 12 h of light alternating with 12 h of darkness. Cages were maintained and cleaned twice a week and placed with an

64 enrichment (PVC pipe or a paper cone). For the entire duration of the study, all SD rats were housed in colony cages with bedding and were given food and demineralized water ad libitum.

Breeding pairs were fed with Teklad rodent chow (#8640, Harlan Laboratories, Indianapolis, IN) which supports normal growth and reproduction in laboratory rodents. Males used for breeding were separated from the cage when females delivered new pups. Pups were weaned at Day 21 and transferred to protocol 2547 at 4 wks old.

In protocol# 2547, a total of 60 SD rats (30 male, 30 female) were obtained from the original SD breeding pairs. Each SD rat was randomly distributed into one of 6 groups (N=10 per group, 5 males and 5 females). Figure 3.1 delineates the overall general schematic of the in vivo study design. SD rats first underwent an 8-wk dietary manipulation. Weaned male and female

SD rats were assigned to groups to receive either a VD sufficient (VDS) (TD.96348, 2200 IU

Vitamin D/kg diet) (Table 3.1) or a VD deficient (TD.87095, 0 IU Vitamin D/kg diet) (Table 3.2) pelleted food. Both VDS and VDD diets contained 20% lactose, 2.0% calcium, and 1.25% phosphorus.

The duration of 8 weeks has been shown to be the length of time needed to successfully induce VDD in rats (Warner and Tenenhouse 1985; Benbrahim and others 1988) or mice (Kankova and others 1991; Malley and others 2009). Experiments began when the rats reached 12-wks of age.

3.4.3 Preparation of treatments for gavage administration in phase I and phase II

All formulations administered on Day 78 for Phase I pharmacokinetics study are outlined in Table 3.3. For the animals receiving capsules, the following procedures of sample preparation were performed. First, GA solution was prepared at 5% (w/v) in deionized water. Second, VD3 solutions was prepared separately in 5 mL ethanol corresponding to 6.0% mass of GA. Mixtures

65 of both components involved dropwise addition of the VD3 solution into 50 mL GA solution while blending at 8,000 rpm for 3 min, then, 10,000 rpm for 2 min before freeze-drying. This formulation represents the capsules used for delivery in treatment 3 (VDS- Capsules, T3), and treatment 4

(VDD- Capsules, T4). Preparation of test solutions containing these capsules for oral gavage administration first required creating a stock solution containing 8 mM sodium phosphate dibasic

(food-grade) adjusted to pH 7.0 using 0.1 M citric acid. For the single high-dose gavage samples used for pharmacokinetics, a target of 1 mL gavage volume containing 300 µg of VD3 present in the capsules was needed. The actual VD3 amount detected in freeze-dried powder (52.1 µg

VD3/mg powder), as presented in Chapter 2, was used to determine mass of capsules needed to deliver 300 µg VD3.

Non-encapsulated VD3 (T1) was prepared by dissolving VD3 in propylene glycol before being added to the phosphate buffer at pH 7.0. The physical mixture treatment (T4) consisted of both non-encapsulated VD3 and GA prepared by mixing the non-encapsulated VD3 dissolved in propylene glycol and GA dissolved in the phosphate buffer at concentrations equivalent to the capsule treatment. Lastly, the controls (C1 and C2) consisted of GA alone prepared by dissolving

GA in the phosphate buffer at a level comparable to the capsule treatment.

All formulations administered during phase II, with target daily dose amount of 60 µg/d for two-week period, are delineated in Table 3.4. Samples were dissolved similarly to those in

Phase I, except with different masses of compounds.

Test liquid dispersions were orally administered using a gavage needle, Type 304 (SS-304) curved 3-in stainless-steel feeding tube with a stainless-steel ball at the tip. To reduce stress on the animal, the rats were acclimatized by allowing them to lick a precoated/dipped sugary solution

(≤ 1 g mL-1) up to two weeks prior to the first gavage treatment. On treatment days, the tip of the

66 gavage needle was precoated/dipped in a sugary solution (≤ 1 g mL-1) prior to gavaging rats with test liquid dispersions. A general schematic on how the formulations were administered to the rat is shown in Figure 3.2. Briefly, a treatment solution was placed in a syringe attached to the 3-in stainless-steel feeding tube and contents released slowly while positioned at the last rib cage for direct delivery in the stomach.

3.4.4 Blood collection and serum separation

Blood withdrawal from the rat in the lateral tail vein was performed with a surflo winged infusion butterfly catheter (25 g x ¾”) equipped with a tube (Lee and Goosens 2015). The rat was first anesthetized using isoflurane in an open drop method. The rats were restrained using a conical decapicone or a rodent restrainer (max < 10 min), and the rat tail was placed in water at 41 ± 2 °C for 40 - 50 s (Lee and Goosens 2015) and dried with a paper towel, followed by insertion of a butterfly catheter into the tail for blood withdrawal. The butterfly catheter includes a tube that can be manually attached to a syringe for stable blood withdrawal. The use of a butterfly catheter in conjunction with giving isoflurane for anesthesia was less stressful for the rats, which was similarly demonstrated by Lee and Goosens (2015).

To enable blood clot formation, blood (without anticoagulant) was placed into 0.6 mL microcentrifuge tubes at room temperature for up to 30 min. The samples were placed in a centrifuge at refrigerated (3 °C) conditions and spun at 2,000 g for 10 min (Lee and Goosens 2015).

The supernatant (serum) was transferred and stored at -80 °C until analysis.

Baseline blood samples were drawn at days 21 and 77, respectively, before and after effects of a VDS diet and/or a VDD diet.

67 3.4.5 Measurement of serum 25(OH)D concentration

Serum 25(OH)D levels were analyzed using Mouse/Rat 25-OH Vitamin D ELISA Assay kit (Eagle Biosciences, Inc. Nashua, NH). Serum samples were diluted to fall within the range of the calibration curve (10.9-144.7 ng mL-1) using 25(OH)D free zero calibrator provided in the

ELISA kit. Internal controls using serum from SD rats in the VDS-control (C1) group were similarly diluted, used as representatives of the dilutions, and run in assays using the same kit

(intra-assay), and different kits (inter-assay) as shown in Figure 3.2.

3.4.6 Pharmacokinetic data analysis

For the pharmacokinetic data after a single high-dose of 300 µg, the corresponding Cmax, tmax, and area under the curve (AUC) were determined. The tmax represents the time it takes post

-1 dosing to reach a maximum 25(OH)D serum referred as Cmax (ng mL ). AUC represents the measure of systemic exposure to the drug in response to dose. The AUC of serum 25(OH)D concentrations versus time profiles (0-48 h) was calculated using the trapezoidal method. The VD3 relative bioavailability of a treatment was calculated with respect to non-encapsulated VD3.

3.4.7 Data and statistical analysis

Serum 25(OH)D data were analyzed using mixed model analysis with the treatment and gender as the between-subject effects while day or hour as the with-subject effects.

Diagnostic analysis was conducted on residuals to verify the model assumptions of normality and equal variance. Rank data transformation was applied when the model assumptions were violated. Multiple comparisons were performed with Tukey's adjust. Significance was identified at 0.05. All analysis was conducted in SAS9.4 for Windows (SAS Institute, Cary, NC).

68 3.5 Results and discussion

3.5.1 Effect of diet on survival, overall health, and body weight of Sprague-Dawley rats

The animals tolerated the diet (VDS or VDD), and no death was attributed to dietary manipulations. However, one male SD rat from VDD-capsules (T4) group was removed from the study due to complications in anesthesia overdose prior to blood withdrawal. All other animals (n

= 59) survived until they were euthanized at the end of the study (108 d).

Our results showed dietary manipulations in SD rats fed either a VDS or a VDD diet did not have any significant effect on body weight during and after the 8 wk feeding period (p > 0.05)

(Table 3.5). Furthermore, there were no obvious alterations in behavior (e.g., weakness or reduced activity) during and after the induction of VDD compared to rats receiving VDS diet. The results are in agreement with a similar study that used the same VDD diet (Stavenuiter and others 2015).

However, in the study by Stavenuiter and others (2015), the length of VDD induction through diet manipulations was 5 wk, shorter than 8 wk in the present study. Furthermore, Wistar rats instead of SD rats were used, and their rats additionally received intraperitoneal injections of 32 ng of 19- nor-1,25-dihydroxyvitamin D2 (; Zemplar from Abbvie). Stavenuiter and others (2015) had a more aggressive approach where the paricalcitol drug was used to accelerate the catabolism of endogenous 25D and calcitriol storage with no reported adverse effects except for the intended reduction on serum 25(OH)D.

VDD diet has compounding effects on mineral homeostasis in the body resulting to or hypophosphatemia (Holick 2003). Therefore, sufficient dietary mineral amounts probably maintained balance in calcium and phosphorus levels, which may have resulted in no significant difference in body weight between the VDS and VDD diet groups in our study. Both diets contained 20% lactose, 2% calcium, and 1.25% phosphorus, a rescue diet formulated to

69 counteract the absence of VD receptor (VDR) dependent intestinal Ca and P absorption, as demonstrated in VDR-null mice (Scholz-Ahrens and others 2001; Bouillon and others 2008) by enabling the parathyroid hormone to enhance bone formation (Toromanoff and others 1997;

Lieben and Carmeliet 2013).

3.5.2 Effect of dietary induced vitamin D deficiency or sufficiency

Comparison of serum 25(OH)D levels between treatment groups when SD rats were weaned (21 d) and after an 8-wk dietary induced VDD or VDS is delineated in Figure 3.3. Serum

25(OH)D levels at weaning age (21 d) were not significantly different among all treatment groups.

After an 8-wk dietary treatment, serum 25(OH)D levels in rats fed a VDS diet (C1 and T3) did not decrease on Day 77 in comparison to levels on Day 21, ranging between 72.18 to 83.93 ng mL-1.

In contrast, serum 25(OH)D levels in rats fed a VDD diet (C2, T1, T2, T4) decreased on Day 77 when compared to Day 21. Overall serum 25(OH)D level reductions from Day 21 to Day 77 on rats that received a VDD diet were observed at ~52.35% in control (C2), ~42.65% in non- encapsulated (T1), ~38.95% in physical mixture (T2), and ~52.43% in capsules (T4). Despite having significantly lower 25(OH)D levels than the VDS group (C1), the VDD rats (C2) were not

VD deficient according to the < 20 ng mL-1 (50 nmol L-1) cut-off established by the Institute of

Medicine (Ross and others 2011).

Other studies have successfully induced VDD within the 8 wk time frame. For example,

6-8 wks was sufficient in inducing VDD in mice corresponding to 25(OH)D levels of < 5 ng mL-

1 (Lagishetty and others 2010). Similarly, 6 wk induction of VDD in SD rats resulted to 10.9 ng mL-1 25(OH)D levels (Angeline and others 2014). There are two possible explanations as to why the VDD diet treatment groups did not have 25(OHD) levels fall below 20 ng mL-1. First, it is possible that light exposure of the rats during the 12-h light/dark cycles in the animal facility may

70 have influenced the VD serum levels. Second, endogenous breakdown of any VD stores may have not been facilitated due to diet effect. The only other study that used the same food source and rat light exposure was conducted by Stavenuiter and others (2015). In their study, rats fed a similar diet containing 20% lactose, 2% Ca, and 1.25% P resulted in 6 ng mL-1 of 25(OH)D; however, rats were also given an intraperitoneal injection of 32 ng of 19-nor-1,25-dihydroxyvitamin D2

(paricalcitol; Zemplar from Abbvie) which accelerated the catabolism of endogenous 25D and calcitriol stores (Stavenuiter and others 2015).

Currently, there are no generally accepted consensus on optimal serum 25(OH)D levels

(Moulas and Vaiou 2018). Optimum 25(OH)D levels have been evaluated on a case by case basis.

For example, levels greater than 30 ng mL-1 but less than 100 ng mL-1 25(OH)D for children with celiac disease have been recommended (Ahlawat and others 2017). In our study, the 25(OH)D levels observed in the VDS control group (C1) after weaning age (78.9 ± 4.1 ng mL-1) fall within the range of acceptable levels. The serum 25(OH)D level observed in the control group was high and may be attributed to several reasons. For example, all animals were exposed to a 12-h light and dark cycle in the animal facility. Furthermore, rats on average consumed 20 g of VDS pellet food containing 2.2 IU VD/g diet (TD.96348), which would have led to a total 44 IU VD/day.

The serum 25(OH)D levels of the 2 control groups from Day 21 (weaning) to Day 108

(euthanization) showed different results (Table 3.8). As expected, rats fed a VDS diet (C1) did not exhibit any reduction of serum 25(OH)D in the duration of the study (108 d) (p > 0.05).

However, the serum 25(OH)D levels of the VDD diet group (C2) decreased by 46.7 % from Day

21 to Day 108.

71 3.5.3 Pharmacokinetics following a dose of 300 µg vitamin D3

The 25(OH)D levels of treatments measured at 2, 4, 8, 24 and 48 h post administration are shown in Figure 3.4. Overall, capsule treatments (T3 and T4) resulted to better absorption of VD3 than non-encapsulated (T1) or physical mixture (T2) treatments. The significant increase from

42.09 ± 3.56 to 134.29 ± 39.13 ng mL-1 for the VDD-capsules treatment (T4) at the first 2-h time point indicate the fast absorption of encapsulated VD3 (Table 3.6). Next, both the VDS-capsules

(T3) and VDD-physical mixture (T2) treatments had a significant increase of 25(OH)D levels at

4 h, followed by the VDD-non-encapsulated group (T1), which occurred at 24 h. The faster absorption of VD by VDD rats than the VDS group was similarly observed by Boullata (2010).

The pharmacokinetic parameters are shown in Table 3.7. The relative bioavailability of

VD observed from the capsules treatments T3 and T4 was 4.54-fold and 4.32-fold of the non- encapsulated VD3 treatment (T1). However, T4 exhibited the shortest tmax (8 h) and Cmax of 659.13

-1 -1 ng mL ; whereas, tmax of T3 occurred at 24 h with corresponding Cmax of 659.25 ng mL . The

VDD-physical mixture (T2) treatment had tmax a similar time (8 h) as that of T4 but had a lower relative bioavailability, 2.05-fold of the non-encapsulated VD3 (T1). The increased absorption of

CoQ10 emulsified by GA based on AUC was also reported by Ozaki and others (2010). In their study, emulsified CoQ10 had a relative bioavailability being 3-fold of the non-emulsified form and had a tmax of 2 and 6 h after oral administration in SD rats and humans, respectively.

The relationship between oral dose of 300 µg to percent VD3 absorbed may be estimated based on pharmacokinetic data. The amount of 25(OH)D in the blood is increased with the continued VD3 absorption and is also decreased as it is cleared by the liver (Boullata 2010).

Considering 25(OH)D has a half-life of 15 days (Gropper and others 2009), the amount of

72 25(OH)D is expected to be similar between tmax and the next time point, and Cmax may be used to estimate the percentage of VD3 absorbed based on Equation 1.

VD3 absorption (%) = [(Cmax -Cbaseline) x (VD3 MW /25(OH)D MW) x Wrats x (7.2 mL/100 g)] x 100 [1] 300,000 ng

Where: Cbaseline is the 25(OH)D level reported at hour 0 during the pharmacokinetics study (Table

3.6). VD3,MW and 25(OH)DMW corresponds to the molecular weight of VD3 and 25(OH)D, respectively. Wrats is the average weight of rats on Day 78 (Table 3.5). The estimated blood volume is ~7.2 mL/100 g SD rat (Probst and others 2006). The highest VD3 absorption was observed in the VDD-capsule treatment (T4) corresponding to 4.05 %, followed by VDS-capsule

(T3) with 3.84%. VD3 absorption was lower in both the VDD-physical mixture treatment (T2) and VDD-non-encapsulated treatment (T1), corresponding to 1.69 and 1.09%, respectively.

Benefits of capsule formulations are evident upon consumption of a single high-dose (300 µg) when compared against efficacy of non-encapsulated or physical mixtures.

Pharmacologic effects of GA related to gastrointestinal absorption of nutrients have been demonstrated in various studies (Ibrahim and others 2004; Nasir and others 2008; Ozaki and others

2010). For example, our findings on early onset of absorption in VDD-capsules treatment (T4) at

2 h post administration was similar to results obtained by Ozaki and others (2010) who studied

CoQ10 emulsified with GA. In their study, mixture of both compounds (CoQ10 and GA) was performed under high-pressure homogenization at 900 kg cm2. The in vivo efficacy of emulsified

CoQ10 was performed on 6 wk old Sprague-Dawley rats (n=4) and tmax observed at 2 h post- administration; whereas, bulk CoQ10 had no peaks observed (Ozaki and others 2010). In another study on SD rats, oral administration of an isotonic solution containing glutamate or acetaminophen mixed with GA resulted in a rapid absorption of both compounds (tmax = 30 min), suggesting that GA has pro-absorptive capability (Codipilly and Wapnir 2004). Similarly, isotonic

73 mixture of zinc and GA yielded consistently higher blood zinc concentrations within 20 min after administration than with GA-free solutions (Ibrahim and others 2004).

Increased bioavailability and quick uptake of lipophilic nutrients may be attributed to several factors. First, increasing the water-solubility of lipophilic compounds through encapsulation allows for “increased compatibility with the aqueous nature of the lining of the digestive tract, thereby enhancing the bioavailability of the otherwise insoluble lipophilic compound” (Diarrassouba and others 2015). Second, entrapment of VD3 in GA may lead to enhanced interaction with bile salts with the compact GA structure opening at near neutral pH and exposing the VD3 resulting to mass exchange to bile surfactant micelles and absorption.

Furthermore, the small particle dimensions of capsules in neutral pH may result in potential direct absorption in the intestines (Inchaurraga and others 2019; Patel and others 2019).

3.5.4 Serum 25(OH)D levels during 2-week study with 60 µg vitamin D3 daily dose

To better understand the effect of VD supplementation, the same SD rats were given oral

-1 gavage solutions of 60 µg d for a period of 2 wks. A one-week “no VD3 treatment” period between phase I (300 µg, single-dose) and phase II (60 µg d-1, 2 wks) was adapted to allow the rats to have a resting period between the two-phases as illustrated in Figure 3.1.

Daily dose administration of 60 µg d-1 resulted in no significant change in 25(OH)D levels on Day 89, 92, and 99 for the T3 and T4 treatments, corresponding to 346.88 and 359.57 ng mL-1 respectively. This 25(OH)D level is high in comparison with other studies that orally administered higher VD3 amounts. For example, Wistrar rats fed 82.73 µg/d for 4 wks had plasma 25(OH)D levels at 224.4 ± 52.4 ng mL-1 (Mirhosseini and others 2016). The VD used by Mirhosseini and others (2016) was from a vegetarian source, and no other details was mentioned.

74 There may be several contributing factors on the higher 25(OH)D levels observed in SD rats that received capsules (T3 and T4). First, the absorption of VD3 is improved after encapsulation in GA (Figure 3.4). Second, it is also possible that the consistently high 25(OH)D levels may be attributed to the previously administered single-high dose of VD3 (300 µg) administered on Day 78 because Day 89 and Day 92 fall within the 25(OH)D circulating half-life of 15 days (Insel and Wardlaw 1996; Jones 2008). Specifically, 15 days after administration of the single-high dose of 300 µg given on Day 78 would fall on Day 93. If the highest observed

25(OH)D levels for VDD-capsules treatment (T4) on Day 78 was at 659.13 (8 h) (Table 3.6),

~329.5 ng mL-1 25(OH)D would be an approximate estimate on Day 92 based on the half-life of

15 d. In fact, 335.79 ng mL-1 25(OH)D was observed for the VDD-capsules (T4) group on Day

92. The VDS-capsules (T3) group did not differ significantly from VDD-capsules (T4) on Day

92 (p > 0.05) showing a potential threshold in absorbed VD. In contrast, on Day 92, non- encapsulated (T1) (273.92 ng mL-1) and physical mixture (T2) (254.48 ng mL-1) had higher

25(OH)D levels when compared with ~109 and 153 ng mL-1, respectively.

Limited data are available on the comparative effects of oral supplementation of different forms of VD (e.g., encapsulated vs non-encapsulated vs physical mixture) in rats. Therefore, assessment of expected range of 25(OH)D increase in the animal is challenging. In humans,

- additional of 100 IU VD3 per day raises serum 25(OH)D concentration by approximately 1 ng mL

1 (Heaney and others 2011). Specific examples of 25(OH)D level increase after a daily VD3 supplementation in humans include the following: (1) an increase from baseline 23.0 to 33.0 ng

-1 mL after 3 month administration of 3,000-4,000 IU (75-100 µg) VD3 daily for human subjects with an age of ≥ 60 yr (Ginde and others 2017); (2) an increase from baseline 20.1 to 65.7 ng mL-

1 after 3 month administration of 12,000 IU (300 µg) VD3 for human subjects with an age of 18 –

75 45 yr (Raja-Khan and others 2014); (3) an increase from baseline 19.16 to 39.88 ng mL-1 6 weeks post administration of a single oral dose of 200,000 IU (5,000 µg) VD3 for human subjects with an age of ≥ 18 yr (Slow and others 2018). Overall estimation of expected 25(OH)D level increase cannot be generalized even for healthy adults because other non-age factors e.g., diet, disease, and health status can influence the metabolism of orally consumed VD. In addition, matrices that provide protection against gastric stomach pH for target delivery in the small intestines (Cohen and others 2017) may also lead to higher absorbed VD in relation to dose response.

There was a period of ten days (Day 99 – Day 108) of “no treatment” after the last 60 µg

VD3 was administered to the SD rats. On Day 108, rats were euthanized and no significant differences in 25(OH)D levels were observed between VDD-non-encapsulated (T1), VDS- capsules (T3), and VDD-capsules groups (T4) (p > 0.05). In contrast, the VDD-physical mixture

(T2) group exhibited significantly higher 25(OH)D levels on Day 108 when compared with all other treatments (p < 0.05). According to pharmacokinetics (Figure 3.4), the encapsulated VD3 is expected to be absorbed better than the physical mixture of VD3 and GA. The mechanism leading to the unexpected observation is to be studied.

3.6 Conclusion

The development of VD3-GA capsules for use as a food-grade ingredient to enhance the absorption of VD3 was demonstrated in this study. The VD3 encapsulated in GA significantly increased the concentration of 25(OH)D in blood of SD rats compared to the non-encapsulated

VD3. Specifically, after a single high-dose oral gavage, the relative bioavailability of VD3 in capsules was at least 4-fold of the non-encapsulated VD3 and the absorption of VD3 encapsulated in GA was faster than the physical mixture or non-encapsulated VD3 formulations. In contrast, treatments of daily dose of 60 µg VD3 resulted in equal 25(OH)D levels between capsules and

76 physical mixture at 2 weeks supplementation, which may have been complicated by the long half- life of 25(OH)D and the insufficient delay between the two-phase studies. After 10 days since the last administration of the low-dose treatment, serum 25(OH)D levels were similar between all treatment groups except the physical mixtures of GA and VD3 with higher 25(OH)D. This study showed that GA is a good excipient for use as a bioactive absorption enhancer to improve the absorption of VD3.

3.7 Acknowledgements

I acknowledge the financial support received from both the Department of Food Science and the award from the Graduate School at the University of Tennessee to conduct this research.

I gratefully thank IACUC personnel, Chris Carter and Jerri O’Rourke for training in animal care, handling, and blood withdrawal. The assistance of Tiannan Wang is greatly appreciated, as she effectively restrained the animal, allowing me to be stable in insertion of the gavage needle directly to the stomach of the SD rat without complications. Tiannan Wang’s assistance in performing blood withdrawal from the tail vein and overall animal care is also greatly appreciated. I would also like to thank Dr. Xiaocun Sun for helping in statistical analysis of data obtained from the in vivo study. I am also grateful for the help of Dr. Philipus Pangloli, Lianger Dong, Shan Hong,

Melody Fagan, Andrea Nieto, Jennifer, Vuia-Riser, and Dara Smith for their assistance in centrifugation of the blood and separation of serum during busy the busy pharmacokinetic study.

77 References

Ahlawat R, Weinstein T, Pettei MJ. 2017. Vitamin D in pediatric gastrointestinal disease. Current Opinion in Pediatrics. 29(1):122-7. Angeline ME, Ma R, Pascual-Garrido C, Voigt C, Deng XH, Warren RF, Rodeo SA. 2014. Effect of Diet-Induced Vitamin D Deficiency on Rotator Cuff Healing in a Rat Model. American Journal of Sports Medicine. 42(1):27-34. Apaydin M, Can AG, Kizilgul M, Beysel S, Kan S, Caliskan M, Demirci T, Ozcelik O, Ozbek M, Cakal E. 2018. The effects of single high-dose or daily low-dosage oral colecalciferol treatment on vitamin D levels and muscle strength in postmenopausal women. BMC Endocrine Disorders. 18:8. Benbrahim N, Dube C, Vallieres S, Gasconbarre M. 1988. The calcium ionophore-A23187 is a potent stimulator of the vitamin-D3-25 hydroxylase in isolated from normocalcemic vitamin-D-depleted rats. Biochemistry Journal. 255(1):91-7. Boiero ML, Mandrioli M, Vanden Braber N, Rodriguez-Estrada MT, García NA, Borsarelli CD, Montenegro MA. 2014. Gum arabic microcapsules as protectors of the photoinduced degradation of in whole milk. Journal of Dairy Science. 97(9):5328-36. Bouillon R, Carmeliet G, Verlinden L, van Etten E, Verstuyf A, Luderer HF, Lieben L, Mathieu C, Demay M. 2008. Vitamin D and Human Health: Lessons from Vitamin D Receptor Null Mice. Endocrine Reviews. 29(6):726-76. Boullata JI. 2010. Vitamin D supplementation: a pharmacologic perspective. Current Opinion in Clinical Nutrition and Metabolic Care. 13(6):677-84. Casey C, Woodside JV, McGinty A, Young IS, McPeake J, Chakravarthy U, Rahu M, Seland J, Soubrane G, Tomazzoli L, Topouzis F, Vioque J, Fletcher AE. 2018. Factors associated with serum 25-hydroxyvitamin D concentrations in older people in Europe: the EUREYE study. European Journal of Clinical Nutrition. Chow ECY, Durk MR, Maeng HJ, Pang KS. 2013. Comparative effects of 1 alpha-hydroxyvitamin D-3 and 1,25-dihydroxyvitamin D-3 on transporters and enzymes in fxr(+/+) and fxr(-/-) mice. Biopharmaceutics and Drug Disposition. 34(7):402-16. Codipilly CN, Wapnir RA. 2004. Proabsorptive action of gum arabic in isotonic solutions orally administered to rats. II. Effects on solutes under normal and secretory conditions. Digestive Diseases and Sciences. 49(9):1473-8.

78 Cohen Y, Levi M, Lesmes U, Margier M, Reboul E, Livney YD. 2017. Re-assembled casein micelles improve in vitro bioavailability of vitamin D in a Caco-2 cell model. Food and Function. 8(6):2133-41. Diarrassouba F, Garrait G, Remondetto G, Alvarez P, Beyssac E, Subirade M. 2015. Improved bioavailability of vitamin D3 using a beta-lactoglobulin-based coagulum. Food Chemistry. 172:361-7. Dwyer JT, Woteki C, Bailey R, Britten P, Carriquiry A, Gaine PC, Miller D, Moshfegh A, Murphy MM, Edge MS. 2014. Fortification: new findings and implications. Nutrition Reviews. 72(2):127-41. Ginde AA, Blatchford P, Breese K, Zarrabi L, Linnebur SA, Wallace JI, Schwartz RS. 2017. High- Dose Monthly Vitamin D for Prevention of Acute Respiratory Infection in Older Long- Term Care Residents: A Randomized . Journal of the American Geriatrics Society. 65(3):496-503. Gropper SAS. 2009. Advanced nutrition and human metabolism. In: Smith JL, Groff JL, editors. 5th ed. Australia. Belmont, CA : Wadsworth Cengage Learning. Heaney RP, Holick MF. 2011. Why the IOM recommendations for vitamin D are deficient. Journal of Bone and Mineral Research. 26(3):455-7. Holick MF. 2003. Evolution and function of vitamin D. Vitamin D Analogs in Cancer Prevention and Therapy: Springer. p. 3-28. Holick MF. 2017. The vitamin D deficiency pandemic: Approaches for diagnosis, treatment and prevention. Reviews in Endocrine and Metabolic Disorders. 18(2):153-65. Holick MF, Chen TC. 2008. Vitamin D deficiency: a worldwide problem with health consequences. The American Journal of Clinical Nutrition. 87(4):1080S-6S. Ibrahim MA, Kohn N, Wapnir RA. 2004. Proabsorptive effect of gum arabic in isotonic solutions orally administered to rats: effect on zinc and other solutes. Journal of Nutritional Biochemistry. 15(3):185-9. Inchaurraga L, Martinez-Lopez AL, Cattoz B, Griffiths PC, Wilcox M, Pearson JP, Quincoces G, Penuelas I, Martin-Arbella N, Irache JM. 2019. The effect of -coating nanoparticles on their biodistribution and fate following oral administration. European Journal of Pharmaceutical Science. 128:81-90.

79 Insel P, Wardlaw G. 1996. Perspectives in nutrition 3rd Edition. St. Louis, Missouri: Mosby-Year Book, Inc. Jones G. 2008. Pharmacokinetics of vitamin D toxicity. The American Journal of Clinical Nutrition. 88(2):582S-6S. Kankova M, Luini W, Pedrazzoni M, Riganti F, Sironi M, Bottazzi B, Mantovani A, Vecchi A. 1991. Impairment of cytokine production in mice fed a vitamin-D3-deficient diet. Immunology. 73(4):466-71. Lagishetty V, Misharin AV, Liu NQ, Lisse TS, Chun RF, Ouyang Y, McLachlan SM, Adams JS, Hewison M. 2010. Vitamin D Deficiency in Mice Impairs Colonic Antibacterial Activity and Predisposes to Colitis. Endocrine Reviews. 31(3):S827-S. Lee G, Goosens KA. 2015. Sampling Blood from the Lateral Tail Vein of the Rat. JOVE-Journal of Visualized Experiments. (99). Levinson Y, Ish-Shalom S, Segal E, Livney YD. 2016. Bioavailability, rheology and sensory evaluation of fat-free yogurt enriched with VD3 encapsulated in re-assembled casein micelles. Food and Function. 7(3):1477-82. Lieben L, Carmeliet G. 2013. The delicate balance between vitamin D, calcium and bone homeostasis: Lessons learned from intestinal- and osteocyte-specific VDR null mice. Journal of Steroid Biochemistry and Molecular Biology. 136:102-6. Malley RC, Muller HK, Norval M, Woods GM. 2009. Vitamin D3 deficiency enhances contact hypersensitivity in male but not in female mice. Cell Immunology. 255(1-2):33-40. Mirhosseini NZ, Knaus SJ, Bohaychuk K, Singh J, Vatanparast HA, Weber LP. 2016. Both high and low plasma levels of 25-hydroxy vitamin D increase blood pressure in a normal rat model. The British Journal of Nutrition. 116(11):1889-900. Montenegro MA, Nunes IL, Mercadante AZ, Borsarelli CD. 2007. Photoprotection of vitamins in skimmed milk by an aqueous soluble lycopene-gum arabic microcapsule. Journal of Agricultural and Food Chemistry. 55(2):323-9. Moulas AN, Vaiou M. 2018. Vitamin D fortification of foods and prospective health outcomes. Journal of Biotechnology. 285:91-101. Nasir O, Artunc F, Saeed A, Kambal MA, Kalbacher H, Sandulache D, Boini KM, Jahovic N, Lang F. 2008. Effects of Gum Arabic (Acacia senegal) on Water and Electrolyte Balance in Healthy Mice. Journal of Renal Nutrition. 18(2):230-8.

80 Neme A, Seuter S, Malinen M, Nurmi T, Tuomainen T-P, Virtanen JK, Carlberg C. 2018. In vivo transcriptome changes of human white blood cells in response to vitamin D. The Journal of Steroid Biochemistry and Molecular Biology. 188:71-76. Norman AW. 2008. From vitamin D to hormone D: fundamentals of the vitamin D endocrine system essential for good health. The American Journal of Clinical Nutrition. 88(2):491S- 9S. Ozaki A, Muromachi A, Sumi M, Sakai Y, Morishita K, Okamoto T. 2010. Emulsification of Coenzyme Q(10) Using Gum Arabic Increases Bioavailability in Rats and Human and Improves Food-Processing Suitability. Journal of Nutritional Science and Vitaminology. 56(1):41-7. Patel M, Mundada V, Sawant K. 2019. Enhanced intestinal absorption of asenapine maleate by fabricating solid lipid nanoparticles using TPGS: elucidation of transport mechanism, permeability across Caco-2 cell line and in vivo pharmacokinetic studies. Artificial Cells Nanomedicine and Biotechnology. 47(1):144-53. Phillips GO, Ogasawara T, Ushida K. 2008. The regulatory and scientific approach to defining gum arabic (Acacia senegal and Acacia seyal) as a dietary fibre. Food Hydrocolloids. 22(1):24-35. Probst RJ, Lim JM, Bird DN, Pole GL, Sato AK, Claybaugh JR. 2006. Gender differences in the blood volume of conscious Sprague-Dawley rats. Journal of the American Association for Laboratory Animal Science: JAALAS. 45(2):49-52. Qv X-Y, Zeng Z-P, Jiang J-G. 2011. Preparation of lutein microencapsulation by complex coacervation method and its physicochemical properties and stability. Food Hydrocolloids. 25(6):1596-603. Raja-Khan N, Shah J, Stetter CM, Lott MEJ, Kunselman AR, Dodson WC, Legro RS. 2014. High- dose vitamin D supplementation and measures of insulin sensitivity in polycystic ovary syndrome: a randomized, controlled pilot trial. Fertility and Sterility. 101(6):1740-6. Rolfes SR, Pinna K, Whitney E. 2014. Understanding normal and clinical nutrition (5th Edition). Wadsworth Publishing Company. Belmont, CA. U.S.A. Ross AC, Manson JE, Abrams SA, Aloia JF, Brannon PM, Clinton SK, Durazo-Arvizu RA, Gallagher JC, Gallo RL, Jones G, Kovacs CS, Mayne ST, Rosen CJ, Shapses SA. 2011. The 2011 report on dietary reference intakes for calcium and vitamin D from the Institute

81 of Medicine: what clinicians need to know. The Journal of Clinical Endocrinology and Metabolism. 96(1):53-8. Scholz-Ahrens KE, Schaafsma G, van den Heuvel E, Schrezenmeir J. 2001. Effects of prebiotics on mineral metabolism. The American Journal of Clinical Nutrition. 73(2):459S-64S. Slow S, Epton M, Storer M, Thiessen R, Lim S, Wong J, Chin P, Tovaranonte P, Pearson J, Chambers ST, Murdoch DR, Vidcaps G. 2018. Effect of adjunctive single high-dose vitamin D-3 on outcome of community-acquired pneumonia in hospitalised adults: The VIDCAPS randomised controlled trial. Scientific Reports. 8:9. Stavenuiter AWD, Arcidiacono MV, Ferrantelli E, Keuning ED, Vila Cuenca M, ter Wee PM, Beelen RHJ, Vervloet MG, Dusso AS. 2015. A Novel Rat Model of Vitamin D Deficiency: Safe and Rapid Induction of Vitamin D and Calcitriol Deficiency without . BioMed Research International. (v2015)1-5. Toromanoff A, Ammann P, Mosekilde L, Thomsen JS, Riond JL. 1997. Parathyroid hormone increases bone formation and improves mineral balance in vitamin D-deficient female rats. Endocrinology. 138(6):2449-57. Warner M, Tenenhouse A. 1985. Regulation of renal vitamin-D hydroxylase-activity in vitamin- D deficient rats. Canadian Journal of Physiology and Pharmacology. 63(8):978-82. Zhou Q, Yang L, Xu J, Qiao X, Li Z, Wang Y, Xue C. 2018. Evaluation of the physicochemical stability and digestibility of microencapsulated esterified astaxanthins using in vitro and in vivo models. Food Chemistry. 260:73-81. Zimet P, Livney YD. 2009. Beta-lactoglobulin and its nanocomplexes with pectin as vehicles for omega-3 polyunsaturated fatty acids. Food Hydrocolloids. 23(4):1120-6.

82 Appendix

Figure 3.1. General schematic design of the in vivo study.

83 Table 3.1. Dietary composition of vitamin D sufficient (VDS) pellet food a (TD.96348, Envigo, Teklad Custom Diet, Madison, WI).

Formula b g/Kg Casein, “Vitamin-Free” Test 192.0 DL-Methionine 3.0 Sucrose 264.78 Lactose, monohydrate 200.0 Corn Starch 150.0 Corn Oil 50.0 Cellulose 50.0 Vitamin Mix, Teklad (40060) 10.0 Mineral Mix, AIN-76 (170915) 35.0 Calcium Phosphate, dibasic 31.17 Calcium Carbonate 14.05 Selected Nutritional Information c % by weight Protein 17.6 Carbohydrate 59.8 Fat 5.0 Kcal/g 3.5 a Irradiated b Referred to in the literature as a rescue diet for mice lacking a functional vitamin D receptor. Contains approximately 2200 IU vitamin D/kg diet, along with 20% lactose, 2.0% calcium, and 1.25% phosphorus. TD. 87095 is a version without vitamin D that is colored brown. c Values are calculated from ingredient analysis or manufacturer data

84 Table 3.2. Dietary composition of vitamin D deficient (VDD) pellet food a (TD.87095, Envigo, Teklad Custom Diet, Madison, WI).

Formula b g/Kg Casein, “Vitamin-Free” Test 192.0 DL-Methionine 3.0 Sucrose 264.78 Lactose, monohydrate 200.0 Corn Starch 151.28 Corn Oil 50.0 Cellulose 50.0 Mineral Mix, AIN-76 (170915) 35.0 Calcium Phosphate, dibasic 31.17 Calcium Carbonate 14.05 Vitamin Mix, w/o choline, A, D, E (83171) 5.0 Choline Dihydrogen Citrate 3.5 Vitamin E, DL-alpha tocopheryl acetate (1000 IU/g) 0.121 Palmitate (200,000 IU/g in oil) 0.0991 Brown Food Color 0.1 Selected Nutritional Information c % by weight Protein 17.6 Carbohydrate 59.8 Fat 5.0 Kcal/g 3.5 a Irradiated and does not include vitamin D b Values are calculated from ingredient analysis or manufacturer data

85 Table 3.3. The designed and actual masses of ingredients used in Phase I in vivo pharmacokinetics study.*

Diet Formulation Designed mass Actual mass

VD3 powder GA powder Freeze-dried VD3 GA Freeze-dried VD3- GA VD3- GA capsules capsules VDS

Control (C1) - 5.46 mg - 5.61 ± 0.10 mg -

Capsules (T3) - - 5.76 mg - - 5.87 ± 0.06 mg

VDD

Control (C2) - 5.46 mg - - 5.61 ± 0.10 mg -

Non-encapsulated 300 μg - - 312.0 ± 5.66 μg - -

(T1)

Physical Mixture 300 μg 5.46 mg - 313.67 ± 9.29 μg 5.61 ± 0.04 mg -

(T2)

Capsules (T4) - - 5.76 mg - - 5.87 ± 0.06 mg

* A dash (-) indicates not applicable. Actual mass represents mean ± SD (n = 10). Freeze-dried capsule powder contains 5.21% VD3 as quantified in normal-phase HPLC.

86 Table 3.4. The designed and actual masses of ingredients used in Phase II in vivo daily-dose for 2 wks study.*

Diet Formulation Designed mass Actual mass

VD3 powder GA powder Freeze-dried VD3 GA Freeze-dried VD3- GA VD3- GA capsules capsules VDS

Control (C1) - 1.11 mg - 1.13 ± 0.10 mg -

Capsules (T3) - - 1.15 mg - - 1.18 ± 0.06 mg

VDD

Control (C2) - 1.11 mg - - 1.13 ± 0.10 mg -

Non-encapsulated (T1) 60 μg - - 64.8± 5.66 μg - -

Physical Mixture (T2) 60 μg 1.11 mg - 64.2 ± 9.29 μg 1.19 ± 0.04 mg -

Capsules (T4) - - 1.15 mg - - 1.18 ± 0.06 mg

* A dash (-) indicates not applicable. Actual mass represents mean ± SD (n=10). Freeze-dried capsule powder contains 5.21%

VD3 as quantified in normal-phase HPLC.

87

Figure 3.2. General schematic on oral gavage administration of formulations, animal and blood collection, serum separation and analysis.

88 Table 3.5. Mean body weight of SD rats at age 21, 78, and 108 day in corresponding treatment groups.*

Rat age (day) VDS Diet VDD Diet Control 1 Capsules (T3) Control 2 (C2) Non- Physical Capsules (T4) (C1) encapsulated Mixture (T2)

VD3 (T1) 21 (weaning) 47.82 ± 6.10 47.21 ± 6.53 47.61 ± 7.03 43.11 ± 3.99 46.11 ± 5.05 51.64 ± 7.06

78 281.02 ± 15.24 289.64 ± 24.41 283.88 ± 23.13 269.55 ± 15.28 285.75 ± 15.92 284.97 ± 22.07 (pharmacokinetics study) 108 302.72 ± 18.80 318.16 ± 28.20 307.67 ± 28.06 301.35 ± 17.19 301.73 ± 17.72 312.66 ± 25.24 (euthanization) * Numbers are mean ± S.E.M. No significant difference in body weight between treatments on the same day (P > 0.05).

89

Figure 3.3. Mean serum 25(OH)D of SD rats on weaning (day 21) and after 8 wk dietary treatment (day 77, baseline) for the two controls and 4 treatment groups.

90

Figure 3.4. Pharmacokinetic profile in 25(OH)D concentrations after administration of 300 µg VD3 in the capsules (dispersed in phosphate buffer), physical mixture (VD3 pre-dissolved in propylene glycol and gum arabic in phosphate buffer), or non-encapsulated (pre-dissolved in propylene glycol) forms. The control treatments are fed with the same amount of gum arabic dissolved in phosphate buffer. Rats were fed with VD deficient (VDD) or sufficient (VDS) diets. Error bars are S.E.M. (n=10).

91 Table 3.6. Pharmacokinetic data in 25(OH)D levels (ng/mL) for different treatments (same data in Fig. 3.4).*

Diet and Time after gavage (h) Treatment 0 2 4 8 24 48 VDS Diet

Control (C1) 72.18 ± 4.96 A, a 78.28 ± 5.68 B, a 70.84 ± 6.82 B, a 80.36 ± 6.15 C, a 73.27 ± 6.44 C, a 70.77 ± 3.97 C, a Capsules (T3) 83.93 ± 5.52 A, c 119.53 ±14.94 A, c 329.24 ± 65.32 A, b 629.15 ± 46.06 A, a 659.25 ± 36.27 A, a 537.57 ± 29.61 A, a VDD Diet

Control 2 (C2) 38.33 ± 3.27 B, a 40.31 ± 3.30 C, a 49.56 ± 6.84 C, a 35.87 ± 2.09 D, a 38.96 ± 3.22 C, a 38.66 ± 2.36 D, a Non- 43.33 ± 2.83 B, d 47.20 ± 2.31 C, d, c 55.97 ± 6.99 C, b, c 62.90 ± 3.66 C, b 121.78 ± 18.78 B, a 218.64 ± 56.41 B, a encapsulated (T1) Physical 50.37 ± 3.07 B, d 83.30 ± 17.52 B, c 176.73 ± 57.15 B, b 306.80 ± 88.49 B, a 260.11 ± 88.94 B, a 274.28 ± 70.81 B, a Mixture (T2) Capsules (T4) 42.09 ± 3.56 B, e 134.29 ± 39.13 A, d 367.51 ± 62.50 A, c 659.13 ± 42.74 A, a 605.17 ± 52.65 A, b 482.83 ± 23.89 A, c * Numbers are mean ± S.E.M. (n=10). Different uppercase superscript letters in the same column are significantly different (P < 0.05). Different lowercase superscript letters in the same row are significantly different (P < 0.05).

92 Table 3.7. In vivo pharmacokinetic parameters following 300 µg VD3 oral administration to Sprague-Dawley rats.

Pharmacokinetic VDD Diet VDS Diet VDD Diet parameters Non-encapsulated (T1) Capsules (T3) Physical Mixture Capsules (T4) (T2) t max (h) 48 24 8 8 -1 C max (ng mL ) 218.64 ± 56.41 659.25 ± 36.27 306.80 ± 88.49 659.13 ± 42.74 AUC (ng ml-1.h) # 5,994 27,238 12,309 25,902 Relative 1.00 4.54 2.05 4.32 Bioavailability * # Area under curve of serum 25(OH)D concentrations versus time profiles (0-48 h), calculated using the trapezoidal method. * Ratio of the area under curve of 25(OH)D concentrations versus time profiles, calculated as AUC (0-48 h) of treatment / AUC (0-48 h) of non-encapsulated VD3.

93 Table 3.8. Serum 25(OH)D (ng/mL) level of Sprague-Dawley rats at weaning age (day 21), during 2-week daily-dose of 60 µg VD3 treatments (day 85-98, measured on days 89, 92, 99), post-treatment (day 106), and on euthaenization (day 108).*

Diet and Serum 25(OH)D level (ng/mL) Treatment Day 21 Day 89 Day 92 Day 99 Day 106 Day 108 VDS Diet Control (C1) 75.77 ± 4.36 A, a 76.78 ± 5.45 C, a 79.49 ± 6.10 C, a 79.32 ± 4.98 C, a 86.02 ± 5.03 B, a 77.32 ± 4.62 C, a Capsules (T3) 77.08 ± 3.02 A, c 350.96 ± 21.18 A, a 345.78 ± 19.27 A, a 343.90 ± 17.75 A, a 168.55 ± 15.70 A, b 143.21 ± 6.95 B, b VDD Diet Control (C2) 80.44 ± 4.46 A, a 34.02 ± 4.08 D, b 34.31 ± 3.93 D, b 59.44 ± 11.99 D, c 47.70 ± 5.65 C, c, d 41.56 ± 4.69 D, d Non-encapsulated 75.37 ± 4.25 A, e 200.48 ± 33.63 B, b 273.92 ± 30.70 B, a 262.88 ± 23.77 B, a 175.98 ± 13.07 A, c 144.29 ± 9.98 B, d (T1) Physical Mixture 82.51 ± 4.47 A, c 239.80 ± 54.06 B, b 254.48 ± 45.50 B, b 434.79 ± 43.37 A, a 209.32 ± 18.17 A, b 210.18 ± 25.33 A, b (T2) Capsules (T4) 88.48 ± 2.61 A, d 347.90 ± 8.93 A, a 335.79 ± 12.34 A, a 395.02 ± 24.01 A, a 182.67 ± 20.81 A, b 135.60 ± 9.69 B, c * Numbers are mean ± S.E.M. (n=10). Different uppercase superscript letters in the same column are significantly different (P < 0.05). Different lowercase superscript letters in the same row are significantly different (P < 0.05).

94 Chapter 4 Concluding remarks and future directions

95 4.1 Conclusions

In conclusion, this dissertation demonstrated that VD3 was successfully encapsulated in gum arabic (GA) with an encapsulation efficiency of 71.35% and a loading capacity of 3.44%.

Capsules were stable across a wide pH range between 2.0 and 7.4 even after 100-d storage at 3 °C.

In vitro digestion results showed capsules to be more bioaccessible than non-encapsulated VD3

(95.8% vs 69.0%, respectively). Oral administration of capsules leads to faster and improved absorption of VD3 in Sprague-Dawley rats. The fast absorption was significant and occurred within 2 hrs post oral gavage of capsules. The clinical significance of rapidly increasing 25(OH)D levels had a greater impact on SD rats on a VDD rather than a VDS diet. Interestingly, administering lower VD3 doses of 60 µg resulted to similar absorption for both capsules and that of the physical mixture after a 2 wks daily dose regimen.

This study provides valuable insight on potential enhancement of VD3-GA interaction with bile in the small intestinal mucosa. By using GA to encapsulate VD3, interaction with bile salts may have been facilitated leading to faster absorption of VD3. Once inside the intestinal cells, the vitamin is transported in the blood by interaction with chylomicrons where it then makes its way to the liver, resulting to the higher observed levels of 25(OH)D. The overall improvements in solubility, pH stability, and VD3 bioavailability when entrapped in a GA matrix makes this emulsifier a good excipient for delivery of hydrophobic bioactive compounds in foods and beverages.

96 4.2 Future work

The capsules in our study showed excellent physicochemical properties and high absorption in SD rats. However, effects of the high VD3 dose should be evaluated for potential toxicological effects. Toxicity is referred to as hypervitaminosis D with clinical symptoms that include nausea, vomiting, and weakness (Blank and others 1995). The SD rats did not exhibit any of these symptoms after the one-time high dose of 300 µg or the 2-wk daily dose of 60 µg VD3.

However, VD regulates calcium and phosphorus homeostasis and can therefore be measured to determine the dosing regimen effect on animal health. Excess VD promotes bone calcium to return to the blood. When there is excess blood calcium, precipitation in soft tissues, known as calcification occurs. Calcium deposits result to formation of kidney stones and hardening of blood vessels leading to irreversible damage (Gropper and others 2009). The VD dose administered to the rats is highly interconnected with the amount of phosphorus in the blood. An excess of VD will lead to an increase in blood phosphate levels. The relationship of elevated calcium to phosphorus is that ionized Ca concentration affects the amount of PTH secreted. At high levels of calcium, release of PTH hormone falls (Insel and Wardlaw 1996; Gil and others 2018).

For future studies, oral administration of encapsulated VD3 in GA should include determining the impact on gut microbiota as we regret that this was not examined in this study.

GA is a prebiotic, meaning it is a dietary substance that proliferates enteric bacteria (Saez-Lara and others 2015). Prebiotics are defined as a food ingredient that is not digested by the host but is fermented by intestinal microflora (Scholz-Ahrens and others 2001). GA is fermented by intestinal bacteria to produce short-chain fatty acids (SCFAs) in the large intestine and acts a prebiotic at a dose of 10 g/day (Phillips and others 2008).

97 References

Blank S, Scanlon KS, Sinks TH, Lett S, Falk H. 1995. An outbreak of hypervitaminosis-D associated with the overfortification of milk from a home-delivery dairy. American Journal of Public Health. 85(5):656-9. Gil A, Plaza-Diaz J, Mesa MD. 2018. Vitamin D: Classic and Novel Actions. Annals of Nutrition and Metabolism. 72(2):87-95. Gropper SAS, Smith JL, Groff JL. 2009. Advanced nutrition and human metabolism. Australia; United States: Wadsworth/Cengage Learning. Insel P, Wardlaw G. 1996. Perspectives in nutrition 3rd Edition. St. Louis, Missouri: Mosby-Year Book, Inc. Phillips GO, Ogasawara T, Ushida K. 2008. The regulatory and scientific approach to defining gum arabic (Acacia senegal and Acacia seyal) as a dietary fibre. Food Hydrocolloids. 22(1):24-35. Saez-Lara MJ, Gomez-Llorente C, Plaza-Diaz J, Gil A. 2015. The role of probiotic lactic acid bacteria and bifidobacteria in the prevention and treatment of inflammatory bowel disease and other related diseases: a systematic review of randomized human clinical trials. BioMed Research International. 2015:505-878. Scholz-Ahrens KE, Schaafsma G, van den Heuvel E, Schrezenmeir J. 2001. Effects of prebiotics on mineral metabolism. The American Journal of Clinical Nutrition. 73(2):459S-64S.

98 Vita

Mary Ross Llonillo Lamsen, known to her friends and family as “May,” was born on May

13, 1979 to Rogelio Viñeza Llonillo and Maria Pilar Cuevas Llonillo in Manila, Philippines. She received her elementary education in the Philippines at Maryknoll (Loyola Heights, Quezon City).

Her family immigrated to the United States in 1993 and resided in Michigan. She attended North

Farmington High School and later received her Bachelor of Science in Nutrition at Wayne State

University (Detroit, MI). Her work experience included Supervisor in the Nutrition Department at St. John Health (Southfield, MI), Quality Assurance Intern at Gordon Food Service (Grand

Rapids, MI), and R &D Formula and Documentation Systems Specialist at Dreyer’s Grand Ice

Cream (Bakersfield, CA), a division of Nestlé (Vevey, Switzerland). It was her experience at

Dreyer’s Grand Ice Cream that influenced her decision to go back to school to pursue a graduate degree in Food Science. She entered the graduate program at The University of Tennessee

(Knoxville, TN) on January 2008 where she received a Master of Science degree in Food Science in May 2010. Additionally, she also completed a ten-month Culinary Arts Instructional Program approved by the American Culinary Federation from The University of Tennessee that same year.

She was admitted to the PhD program at the University of Tennessee in August 2010. During her

PhD training, she accepted a position at Gerber (Fremont, MI), a company also referred as Nestlé

Nutrition North America. At Gerber, May was an intern in the R & D Sensory and Consumer

Insights group where her project was focused on method development to obtain liking and/or preference directly from toddlers age 24-42 months. Upon her return to the University, she served as a graduate assistant in Sensory Science, Applied Food Science, and Food Chemistry. Dr. Qixin

Zhong, was her main advisor for both her M.S. and PhD degrees. She passed her dissertation defense on March 28, 2019.

99