Water – soluble choline compounds in human milk: Their variation and impact of storage and diet. by Lynda S. Soberanes Garcia B.Sc. Universidad Iberoamericana Puebla, Puebla, Mexico, 2007. A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in The Faculty of Graduate and Postdoctoral Studies (Human Nutrition) THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) April, 2015 © Lynda S. Soberanes Garcia, 2015   ii ABSTRACT Choline is an essential nutrient with potentially important roles for infant neurodevelopment. During infancy, choline is usually provided by milk, where is present mostly as water-soluble cholines (WSC). However, published data suggest wide variability in the WSC content with unknown cause, but maternal diet has been suggested as an explanation. It is also unclear if variability results from methodological approach or stability in expressed milk. The objective was to determine the milk WSC content and composition, to determine if variability is present and results from storage, or from expressing WSC/ml, as well as to determine WSC variability within and across women and the potential role of maternal diet. Two studies were conducted: 1) Milk expression and storage on WSC (n=6). Complete milk expressions were analyzed immediately and after different storage conditions. 2) Milk WSC intra and inter-individual variability within one day (n=20), with collection of dietary data. WSC was analyzed using LC-MS/MS and kilocalorie content with a Human Milk Analyzer. ANOVA was used to determine changes in milk WSC following different storage. Correlation analysis was used to explore associations between WSC/ml or kcal, and between maternal diet and milk WSC. Concentration of milk WSC remained similar over short term storage, and only changed when milk was stored for 6 months at -80 °C. Glycerophosphocholine was the most stable WSC;, milk phosphocholine decreased 30% and free choline increased 53 % after 4 hours of storage at room temperature. WSC content and distribution of free choline, phosphocholine and glycerophosphocholine showed wide inter-individual variability, with some women showing high   iii variability at different times in one day. Maternal choline intake was not related to the concentration of milk water-soluble cholines. Only 20% of participants consumed the Adequate Intake=550 mg/d, with eggs as the major food source of choline in the maternal diet. In summary, individual WSC compounds are not stable in expressed milk. The milk WSC content and composition is highly variable across and within women and sampling from one point might not represent infants’ intake. This is the first study addressing the storage and maternal diet as causes for the milk WSC variability.   iv PREFACE This thesis contains the work accomplished by myself, under the supervision of Dr. Sheila M. Innis and the guidance from Dr. Eunice Li-Chan and Dr. Tim Green, and was prepared in accordance to the University of British Columbia and The Faculty of Graduate and Postdoctoral Studies requirements. This project was part of a larger study addressing the variability in fatty acids and other nutrients in human mature milk from healthy women. The study protocol was designed by principal investigator Dr. Sheila Innis and conducted according to guidelines laid down in the Declaration of Helsinki and all procedures involving human subjects were approved by the Committee for Ethical Review of Research Involving Human Subjects at the University of British Columbia (B.C.) and the B.C. Children’s and Women’s Hospital (UBC C&W Number: H12-03191). Written informed consent was obtained from all participants prior enrolment. The study included the work of different team members, including graduate student Sara Moukarzel and Roger Dyer as the expert laboratory technician. My roles in this project included subject enrolment, conducting participant interviews, and dietary intake data collection using food diaries and Food Frequency Questionnaires. I was responsible for entry of all of the dietary data into the nutrient bases, I estimated the choline intake for all diet diaries. I assisted in milk sample preparation for the analysis of water-soluble cholines analysis by LC-MS/MS, which was done by Roger Dyer. I was responsible for summarizing data, conducted data and statistical analysis in this thesis and wrote the thesis with supervision from Dr. Sheila Innis. Sections of this thesis will be submitted as a manuscript for publication in peer-reviewed academic journal.   v TABLE OF CONTENTS ABSTRACT…………………………………………………………………………………………………...ii PREFACE…………………………………………………………………………………………………….iv TABLE OF CONTENTS……………………………………………………………………………………..v LIST OF TABLES………………………………………………………………………………………….. vii LIST OF FIGURES…………………………………………………………………………………………. ix LIST OF ABBREVIATIONS………………………………………………………………………………... x AKNOWLEDGEMENTS…………………………………………………………………………………… xi DEDICATION……………………………………………………………………………………………….. xii CHAPTER 1: BACKGROUND…………………………………………………………………………….. 1 1.1 Introduction……………………………………………………………………………………………………………………….. 1 1.1.1 Choline importance during infancy…………………………………………………………………………………… 2 1.1.2 Differences in the digestion, absorption and metabolism for the different forms of choline.. 4 1.1.3 Deficiency………………………………………………………………………………………………………………………... 6 1.2 Choline in human milk………………………………………………………………………………………………………. 6 1.2.1 Variability of choline in human milk………………………………………………………………………………… 10 1.2.2 Sources of human milk cholines……………………………………………………………………………………...11 1.2.2.1 Endogenous synthesis in the mammary gland……………………………………………11 1.2.2.2 Uptake from maternal circulation……………………………………………………………………………... 12 1.2.3 Comparison of human milk choline to formula or intravenous solutions…………………….13 1.2.4 Implications of human milk storage………………………………………………………………………………… 14 1.3 Dietary requirements of choline……………………………………………………………………………………….15 1.3.1 Recommendations of choline intake during infancy…………………………………………...15 1.3.2 Recommendations of choline intake during lactation………………………………………….16 1.4 Maternal food sources………………………………………………………………………………. 16 1.4.1 Dietary intake of choline compounds among lactating woman……………………………… 17 1.4.2 Role of maternal diet in the milk choline content……………………………………………… 18 CHAPTER 2: RATIONALE, OBJECTIVES, HYPOTHESES AND SPECIFIC AIMS……………….20   vi 2.1 Rationale………………………………………………………………………………………………………………………….. 20 2.2 Objective………………………………………………………………………………………………………………………….. 22 2.3 Hypothesis………………………………………………………………………………………………………………………. 23 2.4 Specific aims…………………………………………………………………………………………………………………… 23 2.5 Methods……………………………………………………………………………………………………………………….. 25 2.5.1 Design and settings……………………………………………………………………………………………………….. 25 2.5.2 Subjects and milk collection……………………………………………………………………………………….. 25 2.5.2.1 Dietary information collection…………………………………………………………………………………... 29 2.5.2.2 Laboratory analysis…………………………………………………………………………………………………. 29 2.5.2.3 Statistical analysis…………………………………………………………………………………………………... 33 CHAPTER 3: RESULTS…………………………………………………………………………………...34 3.1 Demographics…………………………………………………………………………………………………………………. 34 3.2 Stability study…………………………………………………………………………………………………………………. 35 3.3 Variability study………………………………………………………………………………………………………………. 40 CHAPTER 4: DISCUSSION, LIMITATIONS AND FUTURE DIRECTIONS………………………… 64 4.1 Discussion………………………………………………………………………………………………………………………..64 4.2 Limitations……………………………………………………………………………………………………………………. 77 4.3 Future directions…………………………………………………………………………………………………………. 78 REFERENCES……………………………………………………………………………………………... 80 APPENDIX 1. Instruction for milk sample collection ………………………………………………….. 91 APPENDIX 2. Dietary diary form.………………………………………………………………………… 92 APPENDIX 3. Milk caloric (kcal/dl) and water-soluble choline content (umol/l) in 20 healthy breastfeeding women across one day.…………………………………………………………………... 93   vii LIST OF TABLES TABLE 1. Published studies of choline in human milk..………………………………………………… 8 TABLE 2. Choline content (mg/100g) for some commonly consumed foods……………………….. 17 TABLE 3. Temperatures and length of storage of human milk for analysis of water-soluble cholines……………………………………………………………………………………………………… 27 TABLE 4. Maternal and infant characteristics. ………………………………………………………….34 TABLE 5. Change in milk water-soluble choline concentration after different temperatures and storage conditions.…………………………………………………………………………………………. 35 TABLE 6. Change in milk glycerophosphocholine concentration after different temperatures and storage conditions………………………………………………………………………………………….. 36 TABLE 7. Change in milk free choline concentration after different temperatures and storage conditions…………………………………………………………………………………………………….39 TABLE 8. Change in milk phosphocholine concentration after different temperatures and storage conditions…………………………………………………………………………………………………….40 TABLE 9. Concentration of water-soluble cholines (µmol/l) in milk samples of healthy women….. 41 TABLE 10. Water-soluble cholines in 5 milk samples from different women, minimum and maximum concentration and the percent difference in the highest and lowest………………………44 TABLE 11. Free choline (µmol/l) in human milk in samples collected at 5 time points on a single day…………………………………………………………………………………………………………… 45 TABLE 12. Phosphocholine (µmol/l) in human milk samples collected at 5 time points on a single day…………………………………………………………………………………………………………… 46 TABLE 13. Glycerophosphocholine (µmol/l) in human milk samples collected at 5 time points on a single day……………………………….……………………………………………………………………47 TABLE 14. Total water-soluble cholines (µmol/l) in human milk samples collected at 5 time points on a single day……………………………………………………………………………………………… 48 TABLE 15. Caloric content (kcal/dl) of human milk samples collected at 5 time points on a single day…………………………………………………………………………………………………………… 50   viii TABLE 16. Range of water soluble-cholines (µmol/kcal) in human milk collected at 5 time points in a day.………………………………………………………………………………………………………… 51 TABLE 17. Dietary intakes of energy, macronutrient, fatty acid and cholesterol among lactating women.……………………………………………………………………………………………………….53 TABLE 18. Dietary choline intakes (mg) among lactating women, n=20……………………………. 53 TABLE 19. Contribution of different forms of cholines to the maternal diet…………………………. 54 TABLE 20. Food sources of choline in the maternal diet.……………………………………………...55 TABLE 21. Mean concentration of total choline in the maternal diet, from different from sources in all subjects and consumers.………………………………………………………………………………. 56 TABLE 22. Choline intake of lactating women with daily intakes of ≥2, 1 or no egg, and the participants who met current ai for choline intake………………………………………………………. 58 TABLE 23. Estimation of total phospholipids, phosphatidylcholine and sphingomyelin in milk samples of healthy breastfeeding women……………………………………………………………….. 61   ix LIST OF FIGURES FIGURE 1. Overview of the metabolism of choline.………………………………………………………5 FIGURE 2. Different forms of choline found in human milk and foods, and its structure…………….7 FIGURE 3. Choline (umol/L) compound in human milk.…………………………………………………9 FIGURE 4. Pathways for choline secretion in the milk.…………………………………………………11 FIGURE 5. Change in water-soluble cholines (umol/l) in milk left at room temperature for up to 4 h after expression.…………………………………………………………………………………………….37 FIGURE 6. Intra-individual variability in the milk water-soluble cholines……………………………..43 FIGURE 7. Distribution of total choline intake (mg/d) and choline/2000 kcal for lactating mothers.57 FIGURE 8. Free choline, phosphocholine, glycerophoshphocholine and total water-soluble cholines in milk of women meeting (n=5) and not meeting (n=15) the dietary choline recommendations….. 59 FIGURE 9. Relation between dietary phosphocholine and milk water soluble cholines……………60 FIGURE 10. Water soluble choline in human milk; mean estimate of lipid bound choline for each subject compared to the estimated milk choline content of 1500-2000 mg/ml needed to meet the choline requirements of the breastfed infant……………………………………………………………..62 FIGURE 11. Baseline concentration of free choline, phosphocholine, glycerophosphocholine and total water-soluble cholines of human milk of 6 healthy women during lactation, as umol/l……….. 68 FIGURE 12. Change in the concentration of the water-soluble cholines concentration in the milk, according to infant age, as umol/l………………………………………………………………………… 70   x LIST OF ABBREVIATIONS FC: Free choline GPC: Glycerophosphocholine MTHFR: Methyl-tetrahydrofolate reductase Pchol: Phosphocholine PC: Phosphatidylcholine SM: Sphingomyelin PE: Phosphatidylethanolamine PEMT: Phosphatidyl ethanolamine methyltransferase SNP: Single nucleotide polymorphism TMAO: Trimethylamine N-oxide WSC: Water-soluble cholines     xi AKNOWLEDGEMENTS I offer my sincere gratitude to my supervisor Dr. Sheila Innis for this great opportunity, for her supervision and for being an example of strength and love to her family and her work. I am also grateful to my committee members Dr. Tim Green and Dr. Li-Chan for all the guidance, support and help throughout my MSc studies. My graduate studies were financed by the National Council of Science and Technology (Consejo Nacional de Ciencia y Tecnologia, CONACYT), Mexico. I would also like to acknowledge the great people that has been part of the Innis lab and has made it a perfect place to learn among friends, Rong Yu, Brian Wu, Betina Rasmussen, Alejandra Wiedeman, Vinodha Chetty, Julie Matheson, Guillene Boyce, Dr. Cyrielle Garcia, Dr. Jie Yang and specially to Kelly Mulder for her constructive suggestions and for sharing all her knowledge and experience with me, and to Sara Moukarzel who helped me in each step during my studies, thank you for the constant support and friendship. Thanks to Roger Dyer for all his patience and help in the laboratory analysis, always with a smile in his face. Thank you to all the great women that were participants in this study, without whom this project would have not been possible. I would like to thank my loving family and friends for all their support.   xii DEDICATION I first dedicate this work to God, thanks to him I have guidance on each step I walk. To my father, who has always been an example of hard work, honesty, and who taught me to never give up. To my mother that has been a light in my way when everything else is dark, and who believed in me even when I didn’t do it. To Pau that has always been so close to me even in the physical distance, who has listen to the most insignificant or transcendental talk with love and patience, and for teaching me that no matter how big is your job, the importance is to do it with love. To Carlos for being always holding my hand in happy and difficult times, for looking at every presentation and reading every draft without complaining, for showing me how strong I am, and for being the best partner and the best friend. And last but not least, to Santiago and Ximena, whose smiles can give energy and peace even in the more stressful situation, you are the motor of my life. All my work and what I achieve is for you and because of you.   1 Chapter 1: Background 1.1 Introduction Choline, also known as trimethyl-beta-hydroxyethylammonium, is a quaternary ammonium compound with three methyl groups covalently attached to a nitrogen atom (Zeisel et al., 1991, 2000). The term choline often refers to the water-soluble forms of free choline (FC) phosphocholine (Pchol) and glycerophosphocholine (GPC), as well as the lipid-bound forms phosphatidylcholine (PC) and sphingomyelin (SM). Choline is essential for three main pathways. First, choline can be converted to betaine, which is a methyl donor required in the conversion of homocysteine to methionine, and acts as osmolyte (Zeisel, 1981, 1994, Niculescu et al., 2002). Second the production of the phospholipids PC and SM, important constituents of cell membranes that play important roles in trans-membrane signaling, function and structure, in cholesterol synthesis and metabolism, and export of triglycerides from the liver (Cole et al., 2012, Zeisel 2000 b). Finally is part of the neurotransmitter acetylcholine, which is important in the brain, as well as non-neural function in other tissues (Zeisel, 2000, Zeisel et al., 2004, 2006. Blusztajn, 1998). Recent evidence suggests dietary choline is important for the neurodevelopment of infants and young children. Breastmilk (or breastmilk substitutes) are recommended as the only source of choline for the first months of life. In human milk all choline forms are present; however, water-soluble cholines (WSC) account for more than 90% of the total choline, with Pchol as the predominant form followed by GPC (Zeisel et al., 1986, Holmes Mc-Nary et al., 1996, Holmes et al., 2000, Ilcol et al., 2005, Fischer et al., 2010). Lipid-bound forms of choline, PC and SM are present in low concentrations in human breastmilk and mainly as part of the milk fat globule membrane. As opposed to adulthood when lipid-bound cholines are major sources of dietary   2 choline, in infancy the WSC are the main contributors of the choline in the diet. This is possibly because newborns have a low production of bile salts and pancreatic phospholipases, which makes them more difficult to digest phospholipids such as PC and SM (Sian et al., 1992). A second possible reason is related to metabolic efficiency: having enough FC and Pchol allows creation of PC via cytidine diphosphate-choline (CDP-choline) without using methyl groups needed when PC is created via phosphatidylethanolamine. Of relevance, the concentration of WSC compounds in milk, reported based on milk volume, has been shown to be widely variable across women with no clear reason (Holmes Mc-Nary et al., 1996, Holmes et al., 2000, Ilcol et al., 2005, Fischer et al., 2010). The stability of WSC has also not been addressed yet, and there is no evidence that storage will not affect the quantity and balance of milk WSC. The relevance of storage is not only applicable to laboratory settings where milk samples are analyzed after months of storage, but also in common households where there is an increase in the practice of expressing and storing milk for feeding the infant later. Additionally, dietary choline intake has been proposed to be related to the concentration of choline in milk (Zeisel et al., 1986, Fischer et al., 2010), but there is no conclusive information on how the different forms of dietary choline can affect the amount and variability of milk WSC. 1.1.1 Choline importance during infancy SM is an important sphingolipid involved in regulation of cell growth and differentiation, intracellular signal transduction and the myelination of the developing central nervous system (Oshida et al., 2003, Zeisel et al., 2006a). In the brain, choline is needed for the myelination and synthesis of neuronal membranes, but also for the formation of the neurotransmitter, acetylcholine. The neonatal period is the stage of highest body growth rate in the extra-uterine life, with increased membrane creation rate, along with an important level of neurodevelopment. In the brain, there is   3 an increase in synaptic creation, myelination of central nervous system (CNS) and constant creation of neurotransmitters, including acetylcholine where choline can play important functions (Oshida et al., 2003, Albright et al., 1999, Zeisel, 2000a). During pregnancy, maternal dietary choline intake and thus transfer to the infant is highly important to deliver adequate amounts of choline, since there is an increased level of neurogenesis of cholinergic cells of the basal forebrain. A relationship between maternal choline status during pregnancy to cognitive development in 18 months old human infants has been investigated in both animals and human studies and a clear association has been concluded (Wu et al., 2012, Albright et al., 1999, Cheng et al., 2008). It is well known that the brain development continues after birth, with high post-natal synaptogenesis rate, and great capacity of incorporation of choline to the brain from the blood-brain barrier, as seen in animal studies (Braun et al., 1980, Zeisel et al., 1994). The enzyme phosphatidyl-ethanolamine methyltransferase (PEMT) in the neonatal brain is highly active, suggesting a definitive need of choline for the developing brain during infancy (Zeisel et al., 1994). At the same time, children no longer rely on the direct transfer of choline from the mother, as during pregnancy, but need food in order to get the choline they need. The first food that the infants are exposed to is milk, with current recommendations for the first 6 months of exclusive consumption of human milk for the breastfed infant (WHO, 2011). Human milk has been determined as to have increased amounts of total choline, especially the phosphorylated WSC forms Pchol and GPC, which along with FC account for up to 90% of the total choline in the human milk, however the lipid bound cholines are also present in less than 10%. The concentration of total choline as well as each one of the choline-containing compounds in milk has been suggested to vary, with little understanding on what factors intervene in that variability (Zeisel et al., 1986, Holmes Mc-Nary et al., 1996, Holmes et al., 2000, Ilcol et al., 2005, Fischer et al., 2010).   4 1.1.2 Differences in the digestion, absorption and metabolism for the different forms of choline Dietary choline is absorbed in the lumen of the small intestine via transporter proteins in the enterocyte, but some amount of choline is metabolized from bacteria in the gut to betaine and methylamines (not a methyl donor). The WSC compounds, such as FC, Pchol and GPC, are absorbed in the portal circulation to the liver; the lipid-soluble compounds SM and PC are absorbed as chylomicrons in the lymph via thoracic duct (Zeisel et al., 1991 & 2000b). The metabolism of choline is highly dynamic, with only the oxidizing step to form betaine being non- reversible (Michel et al., 2006). To date, there is no available information about the differences in the percent of absorption of the different choline containing compounds, but given the differences on the needed substrates to create new compounds, and their molecular structure, it is thought there might be differences on absorption and function from the different forms of choline.   5 Figure 1. Overview of the metabolism of choline. Adapted from Michel et al., 2006. One example of non-equivalence between the different forms of choline is presented in the brain, during pathological circumstances. In the normal brain, high proportions of PC and FC are found. In postmortem analysis of the brain from Alzheimer patients, those compounds were found in lower levels, with higher levels of GPC (Tayebati et al., 2013). The authors proposed the higher levels of GPC were due to an increase in the catabolism of membrane phospholipids and lower levels of phospholipid precursors, such as FC. The reasons are still under research, but it is suggestive of the importance of having adequate levels of choline, but also maintaining a balance across the several compounds of choline. In infants (humans or animals) no research has been done regarding the implication of having an imbalance of the choline compounds in human milk.   6 1.1.3 Deficiency Traditionally, studies focusing on health outcomes for adults, found that male, post-menopausal women and patient supported by total parenteral nutrition (TPN) with intravenous solutions deficient in choline, based on decreased choline stores in liver, with outcomes such as liver dysfunction and muscle damage, resolved when choline is reintroduced in the diet (Zeisel et al., 1991, Buchman et al., 1995). Pre-menopausal women appeared to be less susceptible to deficiency believed to be due to an increased endogenous choline synthesis from phosphatidylethanolamine (PE), linked to the estrogen secretion (Fischer et al., 2007). Susceptibility to choline deficiency in infancy or lactation has not been investigated. Some studies using cell culture in the media, have shown that brain and liver cells have absolute requirements of choline and die by apoptosis when they are deprived of choline (Holmes et al., 1997, Oh et al., 1997, Zeisel et al., 1997). At macro level, some health outcomes secondary to diets lacking choline in young animals were determined, some of those being growth retardation, renal dysfunction, hemorrhage and bone abnormalities (DRI choline, 1998, Fisher et al., 2001, Kular et al., 2014). Animal studies showed that when choline was not included in the maternal diet during pregnancy, newborns were in disadvantage in neurodevelopment when compared with peers born from a mother with supplemented intake of choline, the effects were life-lasting and were mainly in memory functions (Meck et al., 1989).   1.2 Choline in human milk. Human milk ad libitum is the recommended feeding practice for the first six months of life for the human infant (WHO, 2011), it will deliver all the nutrients and non-nutritive (i.e. immunoglobulin) bioactive factors that are needed to achieve optimum growth, development and immunity of the   7 infant (Newton, 2004, Walker, 2010). Human milk from healthy women is known to be well matched to the needs of the infant for almost all nutrients and to provide the necessary support for immune system development and function, and high growth and development of the brain, including enhancement of gastrointestinal function (Heiman et al., 2007, Lawrence et al., 2007, Sousa et al., 2014). Human milk is considered to be the gold standard for the nutrients required by young infants, and in most cases the average milk intake and milk nutrient content is used to set the dietary recommendations for infants from birth to 6 months of age. Choline, as part of the nutrients of human milk, is delivered in large amounts to the infants in different forms (DRI Choline, 1998), Figure 2. NAME STRUCTURE % CHOLINE1 Free Choline 100% Phosphocholine 56% Glycerophosphocholine 40% Phosphatidylcholine 13% Sphingomyelin 15% Figure 2. Different forms of choline found in human milk and foods, and their structure. 1 Estimated percent of net choline in the different compounds is based on the molecular weight of the different molecules.   8 Several studies have addressed the amount of each choline-containing compound and total choline in human milk, Table 1. These studies have shown that Pchol and GPC are the major choline compounds for which the average amounts are higher in milk than FC; PC and SM are in some situations found in quantities similar to FC, however, they are not found in the aqueous phase of milk but as part of the milk fat globule membrane and the proportion of choline coming from PC and SM is low. Table 1. Published studies of choline in human milk Data is shown as means of each compound in human milk, as (μmol/L), with the SD calculated from the sample size and SE published. TC: Total choline; FC: free choline; Pchol: phosphocholine; GPC: glycerophosphocholine; PC: phosphatidylcholine; SM: sphingomyelin. 1No SD or SE reported for total choline in Holmes McNary et al., 1996 and Fischer et al., 2010. Author, Year (Country) n Total choline FC Pchol GPC PC SM Zeisel et al., 1986 (USA) 10 73 ±21 140 ± 32 188 ± 31 Holmes McNary et al., 1996 (UK) 16 13101 98 ± 186 693 ± 487 379 ± 173 90 ± 54 104 ± 37 Holmes et al., 2000 (USA) 8 1280 ± 396 210 ± 141 480 ± 198 410 ± 226 100 ± 28 100 ± 28 Ilcol et al., 2005 (Turkey) 95 1476 ± 468 228 ± 97 551 ± 322 499 ± 155 104 ± 107 94 ± 88 Fischer et al., 2010 (USA) 103 11981 83 ± 54 553 ± 181 388 ± 168 107 ± 47 67 ± 27   9 As described, the different choline compounds have different structure, with not all of them having the same proportion of choline in the molecule. The WSC contribute to about 90% to the total choline compounds in human milk, however when analyzing the net choline from each one of the compounds, the WSC represent around 96% of the total choline in the milk; the water-soluble compounds FC, Pchol and GPC are small molecules with a higher proportion of choline on them, while the lipid bound cholines have low proportion of choline on their structures since they also contain fatty acids in their structure, Figure 3. Figure 3. Choline (µmol/L) compounds in human milk. A, mean compound concentration in the milk; B, net choline from different choline-containing compounds. Compounds are found in bars from top to bottom: sphingomyelin, phosphatidylcholine, glycerophosphocholine, phosphocholine, free choline. Choline from different sources μmol/L Mean milk choline compounds μmol/L B A   10 1.2.1 Variability of choline in human milk Choline is known to be present in human milk, but with wide variability (Holmes McNary et al., 1996, Holmes et al., 2000, Ilcol et al., 2005, Fischer et al., 2010). No significant differences in the concentrations of some choline compounds such as FC and lipid-bound cholines in foremilk compared to hind milk has been suggested (Zeisel et al., 1986), but no data is available regarding Pchol and GPC change in those two milk phases. The choline containing compounds are present in colostrum as well as in mature milk, with lower total choline in colostrum and higher FC, Pchol and GPC in mature milk (Holmes et al., 2000, Ilcol et al., 2005). The lipid-bound choline compounds have been reported to be present in similar amounts in colostrum and mature human milk (Zeisel et al., 1986, Holmes Mc-Nary et al., 1996, Holmes et al., 2000, Ilcol et al., 2005, Fischer et al., 2010). The published data on human milk cholines with SE and subject number (Zeisel et al., 1986, Holmes Mc-Nary et al., 1996, Holmes et al., 2000, Ilcol et al., 2005, Fischer et al., 2010) indicate a high variability, Table 1. Zeisel et al., (1986) and Fischer et al., (2010) investigated the variability with no clear conclusion and no coefficient of variations of milk choline provided. The mean and SD calculated from the SE reported in the papers, show high variability in milk cholines, suggesting skewness of milk choline data or presence of analytical problems. A potential biological variability in choline, as seen in other nutrients in the milk, with different concentrations and an imbalance in the choline containing compounds could be of great importance for the infant growth, development and health. This suggestion is based on the roles of choline in different metabolic pathways, such as cell membranes, lipoprotein and neural functions.   11 1.2.2 Sources of human milk cholines Some compounds in human milk are synthesized from precursors taken up from the maternal plasma, while others are taken into milk via specific transport pathways. For choline, animal studies found that there are two ways to get choline into milk, the first is the uptake of choline by the mammary epithelial cells, and the second is de novo synthesis of PC in the epithelial cell (Chiao et al., 1988, Yang et al., 1988, Rillema, 2004), Figure 4. Figure 4. Pathways for choline secretion in the milk. Mammary gland image modified from Breast normal anatomy cross-section by Patrick J. Lynch, licenced under Creative Commons 3.0; Artery image modified from Illustration depicting coronary artery disease by Blausen.com staff. “Blausen gallery 2014”. Wikiversity of Medicine, Creative Commons 3.0. 1.2.2.1 Endogenous synthesis in the mammary gland Although the liver is the main organ where de novo synthesis of PC takes place, the mammary gland has also been shown to have endogenous synthesis of PC. In this PC synthesis pathway as in the liver, phosphatidylethanolamine is sequentially methylated to PC, having S-   12 adenosylmethionine as the methyl donor and using the enzyme phosphatidylethanolamine-N-methyltransferase (PEMT). PEMT activity is influenced by several hormones including estrogen and gene polymorphisms, suggesting that not everybody have the same level of endogenous PC synthesis (Yang et al., 1988, Fisher et al., 2007). In vitro, epithelial cells had higher total choline content after 30 min of incubation without the addition of choline, suggesting the presence of the biosynthesis of choline (Chiao et al., 1988). However, PC content is low in milk, and it is currently unknown if the mammary gland synthesizes PC and uses it as a substrate for WSC forms. Additionally, the biosynthesis of PC via CDP-choline has been identified to be the principal source of PC in mammary glands from different animals, such as rabbit and cow both during pregnancy and lactation (Infante et al., 1976). 1.2.2.2 Uptake from maternal circulation Uptake of choline in the alveolar lumen of the mammary gland is an active transport of choline from maternal serum, which has considerably lower concentrations of FC when compared to the FC concentrations in human milk. The active choline uptake includes calcium or sodium dependent transporter, with findings from pregnant mice suggesting that insulin may play a role in increasing choline uptake, and prolactin stimulating the incorporation of choline into phospholipids (Rillema, 2004, Yang et al., 1988). In rats at basal levels, choline is 15 fold higher in milk than in serum, but after 6 hours of oral choline administration, the difference increased, with milk choline being 60 fold higher than in maternal serum, showing a possible implication of maternal intake of choline in the available choline in the maternal circulation for uptake and incorporation in the milk (Chiao et al., 1988).   13 1.2.3 Comparison of human milk choline to formula or intravenous solutions. It is not always possible to feed infants with human milk, and milk formulas are often used. Many formulas, soy or bovine milk-protein derived, have been developed trying to provide the infants with an adequate concentration of nutrients, currently based on human milk. Choline is included in infant formula, but choline compounds differ in amounts across formulas, and between formulas and human milk. Studies reported in 1986 and 1996 reported lower total choline in cow-milk based formula, higher proportions of FC and PC, but lower proportion of Pchol, compared to human milk, with much lower total choline, Pchol and GPC in soy-protein based formula (Zeisel et al., 1986, Holmes McNary et al., 1996). Later research showed cow-protein based formula had statistically lower FC, SM and Pchol compared to mature human milk, while soy-protein based formula had a higher proportion of FC, with lower PC and lower Pchol in some products, compared to human milk (Holmes et al., 2000, Ilcol et al., 2005). The impact of the differences in choline amounts or the type of choline are not yet understood, although exclusively breast-fed infants showed higher blood concentrations of free choline compared with formula-fed infants (Ilcol et al., 2010). Whether the source of choline impacts absorption or later metabolism is not clear yet. More critical is the situation of infants fed by intravenous nutrition since TPN lacks any WSC in its composition. In children with ongoing intravenous feeding, there is a decrease in plasma FC with higher PC compared to plasma concentrations previous to the TPN, with frequently associated liver disease thought to be the result of triglycerides accumulation in the liver. The triglycerides accumulation is perhaps secondary to the low synthesis and release of VLDL-cholesterol, thus a decreased export of lipids from the liver (Buchman et al., 2001). In TPN-fed infants, brain development could also be at risk if choline deficiency persists, as suggested by animal studies shown choline deficiency impacting brain development (Cheng et al., 2008).   14 1.2.4 Implications of human milk storage Human milk consumption is recommended for infants exclusively for the first 6 months of life (WHO, 2011), in recent years when this was not possible, mothers would feed their infants with formula. Two situations have been changing recently. Mothers are now interested in nutrition and many of them understand the importance of feeding their infants with their own milk, even if they are employed or are not able to feed them personally. At the same time, more accessible technology has been created to support the expression of the human milk and store it. However, little research has been done addressing the potential changes that the human milk and its nutrients would suffer once they are out of the human body and left at room temperature, refrigerated or frozen. Some vitamins are known to be light or temperature sensitive, such as vitamin C, which has been found to degrade after expression and storage, when compared to baseline. Similarly, when milk is frozen, proteins form micelles and lipids separate from the aqueous phase and usually are left on the walls if a proper homogenization process is not conducted (Garcia Lara et al., 2012). Human milk also contains several enzymes such as lipases, capable of hydrolysis of fats in the milk. This suggests that expressed human milk is not the same in nutrient quantity and quality, when compared to the milk at baseline, thus, infants would have different intake of nutrients. Scarce information is available for choline stability. Fresh human milk samples were compared with samples incubated at 37 °C for 15 min, and samples frozen at -10 °C for 72 hr., then incubated at 37 °C for 15 minutes. Levels of PC, SM and FC were determined in the milk, and no differences in the mean concentration of any of the choline compounds was found (n=3, Zeisel et al., 1986). However, they did not measure Pchol or GPC, or length of time and temperatures used commonly   15 in home settings, such as leaving the human milk in the refrigerator for up to wks., and in the freezer up to months. 1.3 Dietary requirements of choline Choline was accepted as an essential nutrient by the Institute of Medicine in 1998 for humans because de novo synthesis is not sufficient, and dietary recommendations were established for all age groups. As information of choline in humans was limited, an estimated average requirement level (EAR) could not be determined. Rather, an Adequate Intake (AI) was set (DRI choline, 1998). 1.3.1 Recommendations of choline intake during infancy The AI for infants 0-6 months of age was set as 125 mg/d (DRI Choline, 1998). For assessing the amount of choline needed, research should include different dosages in the diet and assessing the health outcomes, but research along this line has not been done for ethical reasons. For that reason, the AI was based on the mean amount of total choline concentration in human milk from healthy, well nourished mothers who exclusively breastfed infants. It assumes a mean milk output of 0.78 L/day, with a mean total choline content of 160-210 mg, or 1500-2000 μmol/L, as determined from two studies published in USA in 1986 and 1996, including 26 participants in total (Zeisel et al., 1986, Holmes McNary et al., 1996). Also, all choline forms are considered to be equivalent in the human milk, as only the concentration of total choline was considered. The information used to set this AI is limited, since only two studies were included, considering milk samples from only 26 women in the US in total (Table 1). Moreover, the results from those studies are inconclusive and due to the low number of participants, may not represent a higher proportion of the population. In addition, they do not consider not consuming adequate levels of choline, or producing/delivering different amounts of human milk.   16 1.3.2 Recommendations of choline intake during lactation The adequate choline intake for healthy women during lactation is considered to be 550 mg/d (DRI Choline, 1998), and was determined based on the large amount of choline delivered to her infant, via human milk. It is thought that there is a risk of depletion of the maternal hepatic stores of choline, which are mainly in the form of Pchol, especially if during pregnancy the dietary intake was not sufficient. (DRI Choline, 1998, Zeisel et al., 1995, 2006 a,b) To the AI, some assumptions were made regarding efficiency of the process (100%), the amount of milk production (set as an average of 0.78 L/day) and the choline concentration of the milk as being 156 mg/L or 1.5 mmol/L, increasing the amount of dietary choline needed in 125 mg/day (1.2 mmol/day). 1.4 Maternal food sources Animal studies showed that lactating rats can easily become deficient when compared to non-lactating rats. During lactation, the estrogen levels have reached a low point, after being really high during pregnancy, suggesting that the de novo synthesis would not be as high as in non-lactating or pregnant women, and the dietary requirements would be higher than non-lactating women, so the intake of choline would be important for preventing diseases in the women. For that, mothers must include in their diets some of the food rich in choline, enough to provide the infant and themselves with an adequate amount of choline for the optimal health, growth and development. Little information is available regarding the major food sources during lactation, however it is possible that eggs and dairy are the main contributors to choline during this stage as seen in different age groups, such as pregnant women and children (Lewis et al., 2014, Wu et al., 2012). Choline can be included in the diet easily, if an omnivorous diet is followed, since choline is mainly present in the animal food sources but in proportions different to human milk. The main   17 choline form found in animal-based foods is usually the lipid-bound PC with WSC found in lower amounts. Food sources rich in choline include liver, eggs and milk, with lower amounts found in vegetable sources such as fruits and vegetables (Patterson et al., 2008), Table 2. Table 2. Choline content (mg/100g) for some commonly consumed foods. Food Total Choline FC Pchol GPC PC SM Egg, fried 270 0.7 0.7 0.6 250 17 Liver 430 57 12 78 250 24 Milk 16 2.8 1.6 10 1.2 0.9 Yogurt 15 2.3 1.7 9.1 1 1.1 Fish 83 21 2.5 1.2 54 4.1 Poultry 62 3.2 2.1 1.6 46 8.9 Meat 85 2.2 0.4 3.3 72 7.6 Grains 2.1 0.7 0 1 0.4 0 Bread 27 18 0.3 4.9 3.3 0 Legumes 32 17 0.8 1.3 12 0 Olive oil 0.3 0 0 0.3 0 0 Lettuce 8.5 5.9 2.4 0 0.2 0 Data is from the USDA Database for the Choline Content of Common Foods Release Two (Patterson et.al. 2008), presented as mg/100 g food. FC, free choline; Pchol, phosphocholine; GPC, glycerophosphocholine; PC, phosphatidylcholine; SM, sphingomyelin. Each food group is represented by a common food consumed, and in all cases is similar to other foods within the same category as follows: Egg: whole, fried; milk: 2% fat; yogurt: plain low fat; fish: tilapia, cooked; poultry: chicken breast meat without skin, roasted; Grains: white long rice; bread: whole wheat; Legumes: beans, plain  1.4.1 Dietary intake of choline compounds among lactating woman There is a lack of information regarding the dietary intake of choline compounds of women during lactation. It is considered that when a low choline diet is consumed, the human milk choline content is lower, compared with the milk of women consuming an adequate diet in total choline levels,   18 being phosphocholine the main compound affected in human milk (Zeisel et al., 1986, Holmes-McNary et al., 1996). A small amount of research regarding choline intake has shown that choline in the maternal diet during lactation is present in inadequate amounts, compared to the AI. In 2010 in the US, women showed to be consuming around 350 mg/day of choline in their regular diets (Fischer et al., 2010). In 2014 in breastfeeding Canadian participants, the results were similar, showing that around two thirds of the subjects studied were consuming the AI, with the main dietary choline sources in the form of eggs and milk (Lewis et al., 2014). However, the impact of the different levels of maternal dietary choline intake is still under debate. 1.4.2 Role of maternal diet in the milk choline content Maternal dietary choline intake has been investigated in relation to the amount of choline in the milk in both animals and humans. In rats, FC, Pchol and total choline concentration in human milk increased in dams consuming adequate levels of dietary choline compared to the ones consuming choline deficient diets. The amounts in milk of these compounds increased further in supplemented animals (Zeisel et al., 1986). In humans, FC intake of mothers was related to the amount of PC in human milk, but the majority of the women studied were not consuming the AI for choline. During oral supplementation with high doses of PC in the form of capsules (PC supplementation= 5400 mg/d), human milk was shown to have higher levels of Pchol and FC (Fisher et al., 2010). However, these results have not been confirmed yet, and there is no conclusive information on how much maternal diet can affect the level and balance of the different choline compounds in milk, in normal and adequate choline intakes. In summary, choline is an important nutrient for the rapidly growing and developing infant and is largely available in human milk mainly in the WSC. A large intra-individual variability in WSC is   19 apparent in the data of all publications on choline in human milk, with no full understanding of this variability to date. Additionally, some variables such as the differences on the storage of milk, management of data, the differences on the caloric content in the milk, as well as the maternal dietary choline intake in normal levels (i.e. not from supplements) could be related to the content of milk WSC, but have not been studied yet.   20 Chapter 2: Rationale, Objectives, Hypotheses and Specific Aims 2.1 Rationale Some nutrients are highly variable in human milk due to the maternal diet, genetics and other factors. Among these, choline has been identified as one of the most variable nutrients in milk (Zeisel et al., 1986, Holmes Mc-Nary et al., 1996, Holmes et al., 2000, Ilcol et al., 2005, Fischer et al., 2010), however; the reason for the variability is unknown. Choline can be synthesized endogenously, but is also considered an essential dietary nutrient in 1998 by the Institute of Medicine because endogenous synthesis is insufficient to meet the body needs, and AI levels were estimated (Zeisel 1991, DRI Choline 1998). The AI for infants 0-6 mo. of 125 mg/d was based on the mean choline concentration of human milk, however, the choline content of human milk is variable in content of total choline and each one of the forms of choline (Holmes Mc-Nary et al., 1996, Holmes et al., 2000, Ilcol et al., 2005, Fischer et al., 2010), which may have implications for infants. It has been suggested that the maternal choline intake during lactation is associated with breast milk choline concentration (Zeisel et al., 1986, Fischer et al., 2010), and therefore could contribute to the variance in milk choline, suggesting that if dietary intake of choline in lactating women is inadequate and contributes to low milk choline, then adverse effects on the infant could occur. The actual AI for dietary choline for women during lactation is 550 mg/d. Choline is involved in 3 main pathways. Choline is needed for the formation of PC and SM, which are needed for the synthesis and turnover of cell membranes, secretory lipids such as surfactant, bile and plasma lipoproteins. Oxidation of choline results in the production of betaine, which is a major methyl donor and acts as an osmolyte protecting cells, proteins, and enzymes from stress. Choline is also needed for the synthesis of the neurotransmitter acetylcholine, which plays multiple   21 roles in the central nervous system, as well as systemic neurological system, and in non-neural inflammation pathways (Zeisel, 1981, Zeisel et al., 1994). Choline in human milk is known to be present in several different forms, the water-soluble forms FC, GPC and Pchol, and the lipid forms PC and SM. Interestingly WSC accounts for up to 90% of the total choline in human milk. PC and SM are present in human milk as part of the milk fat globule membrane, which surrounds the triglycerides that make up the fat globule (Giuffrida et al., 2013). Current available information on choline in human milk suggests variability both between and within mothers, but the factors contributing to the variability are not yet understood. One potential contributing factor may be the collection and storage conditions of human milk by researchers and their subjects prior to analysis, with the impact of milk expression and storage by mothers for later feeding of their infant on milk choline also not known. Expression of human milk is becoming a more common practice among lactating women, both for later use and for donation to milk banks. Little information exists on the stability of nutrients in expressed human milk, including the stability of choline and its multiple forms, which may be affected by storage temperature and time. In the home, human milk may be stored at room temperature, or in the refrigerator or freezer from days to months. In the laboratory, milk samples for choline analysis are generally frozen at -70 °C to -80 °C, with the assumption that no choline degradation occurs, although again this is unknown. A further problem with the measurement of choline in milk is how to express it. The volume of milk mothers produce varies considerably, with wide differences in the milk calories and fat concentrations. Infants control their human milk intake primarily through caloric density, not   22 volume. This raises the question of whether the variability observed in milk choline measured per volume (i.e. per ml of milk) also occurs if expressed per kcal of milk. Finally, as seen with other nutrients, such as n-6 and n-3 fatty acids, the wide variability of water-soluble cholines in human milk might be explained by differences in maternal dietary choline intake. Choline is incorporated into human milk mainly through an active transport of choline from the maternal circulation (Rillema, 2004, Chiao et al., 1988, Yang et al., 1988). To date, little has been published on the impact of dietary choline intake on milk choline during lactation, although it is thought that low intakes of choline in mothers can increase risk of maternal hepatic choline depletion, as seen in animal models and adults fed a choline deficient diet (Zeisel et al., 1995). Whether variation in dietary choline intake in lactating women contribute to variation or low choline in human milk, however, is not yet clear, and implications for the breastfed infant are similarly unknown. 2.2 Objectives 1. To determine the effect of storage duration and temperature of human milk on the concentration of different water-soluble choline compounds. 2. To determine the concentration and variability of the different water-soluble choline- containing compounds in milk collected from different women, using standard and validated methods. 3. To determine if variability in human milk water-soluble choline occurs in milk produced at different times in the day from individual women.   23 4. To compare the variability of water-soluble choline containing compounds in human milk when expressed as per ml milk or per kcal milk. 5. Begin to understand if maternal dietary choline intake contributes to the variability in the amount of water-soluble choline in human milk. 2.3 Hypotheses 1. Time and temperature of storage of expressed human milk will affect the concentration of the individual water-soluble containing compounds in milk. 2. The water-soluble choline-containing compounds in human milk will show high inter-individual variability. 3. The concentrations and distribution of water-soluble choline-containing compounds in human milk will be variable within a woman (intra-individual variability) when collected at different times within a day. 4. Expression of milk water-soluble choline per kcal will reduce intra and inter-individual variability in concentration when compared to expression of concentration per volume of milk. 5. Differences in dietary choline intake among women will contribute to the variability in the amount of water-soluble choline-containing compounds in human milk. 2.4 Specific Aims 1. To use LC-MS/MS to determine the effect of temperature and duration of storage of human milk on the amount of total and individual water-soluble choline-containing compounds in milk collected from 6 women.   24 2. To determine quantity and variability of water-soluble choline-containing compounds in milk in samples from 20 breastfeeding women. 3. To determine the intra-individual variability in the total amount and individual forms of water-soluble choline-containing compounds in mid-feed milk samples collected during 5 different feeds within one day for each one of 20 lactating women. 4. To determine the extent of variability of milk water-soluble choline within and among women when expressed per kcal compared to per ml milk. 5. To begin to explore if the amount and source of choline in the lactating women diet is related to human milk water-soluble choline in 20 women.   25 2.5 Methods 2.5.1 Design and settings The study was conducted at the Oak Street Campus of The University of British Columbia (UBC), in the Child and Family Research Institute (CFRI), which is on the site of BC Women’s and Children’s Hospital. The study protocol was approved by the Committee for Ethical Review of Research Involving Human Subjects at the University of British Columbia and the British Columbia Children’s and Women’s Hospital (UBC C&W Number: H12-03191). This project included two parts: first a stability study to investigate the effect of storage conditions of human milk on WSC content, and second, a variability study to address the variability in milk WSC within and between lactating women. To reduce variability identical methods were used to collect and process milk samples. All participants were recruited from the community via poster advertisements and direct contact in the Greater Vancouver area. Written informed consent was obtained for each participant before collection of milk samples. The study addressing the effect of storage on human milk WSC involved 6 healthy breastfeeding women, all of whom were feeding one healthy full-term infant. Each mother provided one fresh complete breast expression of milk in an early morning visit to CFRI. The study addressing inter- and intra-individual variability in milk WSC, and addressing the potential role of the mothers diet on milk water-soluble cholines, involved 20 healthy breastfeeding women. Each participant provided 5 milk samples collected at a different time within the same day. 2.5.2 Subjects and milk collection For the stability and variability study, healthy women, 19- 40 years of age, with infants 2 to 6 months of age were invited to participate. The inclusion criteria included: full term, single birth   26 infants, exclusive breastfeeding since infant birth, no known health problems in either the mother or the infant, and the ability to speak and understand English. The exclusion criteria included mothers giving birth preterm, not exclusive breastfeeding, and mothers and infants with chronic, genetic, metabolic, neurological, immune, hepatic or renal diseases. Women currently taking medications that may interfere with fat metabolism, hormones or choline or lecithin supplements were also excluded. Mothers interested in participating in the study, who met the inclusion criteria and none of the exclusion criteria, were enrolled. Signed informed consent was obtained from each participant. Each participant was assigned a random computer-generated code, which was used on all forms, milk samples and analyses, to protect the confidentiality of the volunteers. For the stability study, each subject made one single visit to the Clinical Research Education Unit (CREU) at CFRI. This was necessary to control exactly the timing and handling of milk expression. Socio-demographic and general health information, including maternal and infant’s age, ethnicity, consumption of supplements and medication, smoking status, current intake of alcohol, and general dietary pattern was collected. Each subject provided a fresh, complete expression of human milk from one breast using a commercial breast pump in plastic commercial containers. All visits took place in the early morning (9-10 am) during regular weekdays. Milk samples collected in the CREU were immediately transferred to the laboratory and handled as follows. Each milk sample was vigorously vortexed and divided into 19 aliquots of 20 μl each, in 1.5 ml eppendorf tubes. The first aliquot was analyzed immediately and is defined in this thesis as ‘baseline’ or time 0. The remaining milk aliquots were stored by temperature and set times, as shown in Table 3.   27 Table 3. Temperatures and duration of storage of human milk for analysis of water-soluble cholines. Temperature Duration until analysis Room (23˚ C) 30 min, 60 min, 90 min, 2 h, 3 h, 4 h 4˚ C 6 h, 12 h, 24 h -20˚ C 24 h, 48 h, 1 wk., 2 wk., -80˚ C 24 h, 48 h, 1 wk., 2 wk. and 6 mo. Human milk was immediately transferred to the laboratory after expression, with baseline analysis starting approx. 5-10 minutes after expression. The temperatures were chosen to address laboratory and home refrigerator and freezer temperatures: • Room temperature would mimic when a mother expresses milk and leaves it for a few hours (i.e. infant is not hungry and the milk is left out of the fridge/freezer until the next feeding). This practice has been recommended by the academy of breastfeeding medicine (Academy of Breastfeeding Medicine Protocol Committee et al., 2010) and several Internet websites for women during lactation (www.cdc.gov, kellymom.com, medelabreastfeedingus.com, http://www.llli.org/faq/milkstorage.html, www.mayoclinic.org). • 4°C could reflect common household refrigeration temperature. • -20°C resembles the normal household freezer temperature, although some common freezers can be as high as -10°C. • -80°C is a typical storage temperature at research laboratories. Additionally, one 10 ml milk sample per mother was stored for 1 week at -20°C, then moved to -80 °C until 6 months after expression. The purpose of this last sample was to consider possible effects if mothers were asked to express their milk, then freeze in their home, and transfer to a research laboratory in which analysis is done later as part of a group.   28 In the variability study, each participant was seen twice by the research assistant at the subject’s home. At the first visit, pre-labeled vials for milk collection were provided, together with the instructions on milk collection (Appendix 1). Each subject also completed a socio-demographic questionnaire, with a one-day dietary diary form (Appendix 2) completed on the day of milk sample collection. Instructions for the completion of the dietary intake diary with the preset forms were discussed with each mother by the research assistant. Participants were asked to collect their milk samples in their own home, but could choose the day to collect the milk samples at their own convenience, within 2 weeks of the first visit. Each participant was asked to place the milk samples in the vials provided, then immediately after collection transfer the milk samples to their freezer. Sampling times for milk collection were determined by the infant breastfeeding times and the mother’s activities. We requested a sample of milk collected midway through each infant breast-feeding; briefly, mothers were instructed to feed their baby for 3 minutes, then collect the 20 ml milk sample for us, then return to breastfeeding. The mothers were instructed to call when they had collected the samples. To determine intra-subject variation over time, participants were asked to provide five separate collections of 20 ml milk on the same day: 1) first feeding time in the morning, 2) before lunch, 3) 1-1.5 hours after lunch, 4) 1-1.5 hours after dinner, and 5) before going to bed. The women were provided with written instructions for milk expression and storage, as well as pre-labeled vials with their study number. Mothers were asked to write the date and time of milk expression, and instructed to freeze each milk sample immediately after collection of that sample. The milk samples were collected from the participant’s house the day after milk collection (i.e. within 24 hr.) by a member of the research team, then transported on ice to the laboratory and   29 transferred to the -80 °C freezer as quickly as possible. Within 2 weeks of the sample collection, milk was thawed, divided in aliquots and analyzed for choline (20 µl), and macronutrients/caloric content (1.5 ml). On the visit for milk collection, each subject also provided dietary intake information and completed socio-demographic questionnaires. 2.5.2.1 Dietary information collection For the stability study, mothers provided 24 hr. dietary recall, assisted by a nutritionist using the standardized 5 step method (Conway et. al., 2003, Blanton et al., 2006), covering the intake of all foods and beverages for the 24 hrs. prior to the milk sample collection. For the variability study, a food diary recording each food, drink, supplement and the amount of each were completed by each participant on the same day as the milk samples were collected. Written instructions and examples for completion of the diary were given to the participants by a nutritionist. The dietary information collected included the time of consumption of any foods and drinks, together with portion sizes, brands, cooking methods and ingredients in all foods and beverages. 2.5.2.2 Laboratory analysis Water-soluble cholines analysis FC, Pchol, GPC and betaine were analyzed using isotope dilution liquid chromatography tandem mass spectrometry (LC-MS/MS). The LC-MS/MS is a Waters ACQUITY UPLC system connected to a Quattro Micro tandem MS configured with an electrospray source (Waters Corporation, Milford MA, USA). The LC was equipped with a Zorbax Rx-Sil 2.1 X 150 mm column and a precolumn 2.1 3 X 12.5 mm, both with 5 um particle size (Agilent Technologies, Santa Clara CA, USA). The MS was operated in positive ion multiple reaction monitoring (MRM) mode using the transitions of m/z   30 103.9 / 59.9 (choline), m/z 113 / 68.9 choline-d9, Pchol 184.1/125.0, pchol-d9 193.1/125.0, GPC 258/124.9 GPC d9 289.0/221.0 (sodium adduct), betaine 117.9/58.9 and betaine-d9 126.9/67.9. For analysis, aliquots of 20 μl of human milk were transferred to eppendorf tubes containing 10 μl of internal standards and vortexed. Protein was precipitated by adding 30 μl of methanol. containing 0.1% formic acid. The supernatant was recovered after centrifugation at 18,000g x 4°C for 10 min, transferred to an auto sampler vial then mixed with acetonitrile with 0.1% formic acid in dilutions of 1:5. The chromatographic separation was carried out at a flow of 0.5 ml/min with a mobile phase consisting of A= acetonitrile with 0.1% formic acid and 0.1% trifluoroacetic acid, B= 15 mmol/L ammonium formate with 0.1% formic acid and 0.2% trifluoroacetic acid in H20. The gradient separation started at 95% A plus 5% B for 2 min, then a linear gradient to 55% A plus 45% B at 4 min, then held for 5.8 minutes after which the column was flushed with 100% B and re-equilibrated to starting conditions. The autosampler and column were maintained at 5°C and 25°C, respectively. A sample volume of 4 μl was used for analysis. The total analytical time was 9.0 min. The inter-assay and intra-assay coefficient of variation (CV) based on 5 replicates each were as follows: for FC, 5.5% and 4.1%, respectively; for Pchol, 6.4% and 5.2%, respectively; and for GPC 9.5% and 2.3%, respectively. The LC-MS/MS analysis was done with the guidance and assistance of the lab technician expert in this method, Mr. Roger Dyer. Milk caloric and macronutrient content The macronutrient composition and calorie content of the human milks were analyzed using a Human Milk Analyzer (HMA, Miris), which uses the technique of mid-infrared transmission spectroscopy based on the absorption of energy of the functional groups in the molecules. The HMA consists of an emitter, cuvette with capacity of 3μl, and a detector. A 50 um fluid film of milk   31 sample is exposed to infrared light and transmits infrared wave lengths, detected at specific wavelengths within the infrared spectrum, using different wave lengths for determination of concentrations of fat, protein and carbohydrates using different calibration models. The HMA uses different filters for the milk components, with 4 wavebands specific for the functional carbonyl groups (5.7μm) and carbon-hydrogen groups (3.5 μm) for fat determination, the waveband for amine groups (6.5μm) for protein, and for carbohydrate determination the waveband is for hydroxyl groups (9.6 μm). For analysis, 1.5 ml frozen milk samples were thawed and warmed to 40 ºC in a water bath, then homogenized using the Human Milk Sonicator (Miris, SE) for 2.25 sec with amplitude of 70% full scale using a micro-tip. For analysis, the milk samples were transferred to a syringe then injected into the HMA, with analysis of each sample separated by analysis of an equal volume of “standard” milk from a large volume of pooled human milk. The standard sample injection between each study milk sample allows us to check any analytical variance or problems with the instrument flow tubes i.e. repeated accuracy of analysis of samples. Cleaning of the HMA was done before starting the analysis, then after every 10 samples, then at the end of the set using 15 ml of the cleaning solution provided (Miris CLEAN solution). Calibration was also checked after every cleaning with 5 ml of the solution provided (Miris ZERO CHECK solution) to ensure consistent accuracy of results. The HMA uses the operative system Windows Compacts 7. Values for each human milk sample are displayed as g/dl for macronutrients and kcal/dl for energy. The analytical accuracy is <0.1 g/dl, and repeatability <0.05 g/dl, based on the manufacturer. In our analysis, inter-assay and intra-assay CVs for fat were 0% and 2.8% respectively; and for kcal/dl were 5.0% and 0.8%, respectively, based on 10 injections of the same sample on the same day, to calculate intra-assay CV, and 8 injections of the same sample within a day.   32 Maternal diet analysis The dietary information provided by the mothers was analyzed for energy, total fat, saturated fat, cholesterol, and choline intake using the dietary records. Dietary intakes were determined using the ESHA Food Processor SQL (Version 10.1.0, ESHA research, Salem, OR, 2012) and the Canadian Nutrient File for total energy, fats and cholesterol (2010), and the USDA database of the choline content of common foods (Release Two, 2008). For analysis and entry into the nutrient analysis software, home-prepared foods or restaurant foods, as needed, were separated into ingredients for entry into the database. All foods were checked for complete and the most possible accurate information for all choline compounds. FC, Pchol, GPC, PC, SM and total choline dietary intakes were calculated using the USDA database on choline, as mg/day, with the estimated amount of each choline compound calculated from different food sources. To do this, foods were classified into different groups, as follows: • Cereals: including bread, bagel, English muffin, breakfast cereals, granola, rice, couscous, pasta, quinoa, tortilla, pita bread • Pastry: including cake, cupcakes, pies, doughnuts, pancakes, cookies and other pastries • Dairy: including cow’s milk, yogurt, cheese, cottage cheese, sour cream, ice cream • Eggs: including all types, including standard, free range, omega 3 enriched, and egg beaters (cartons of whole eggs or egg whites) • Fruit, vegetable & beans: including all fruits, vegetables, beans and legumes • Meat & fish: including all sources of meat, including beef, pork, veal, bison, poultry and all fish and shellfish • Mixed dishes: including all dishes prepared with a combination of carbohydrate, meat, and dairy such as lasagna, pizza and pastas.   33 • Oils and fats: including all cooking oils, margarine, butter, and salad dressings • Nuts & seeds: including all ground and tree nuts and seeds • Other: including chocolate bars, candy, condiments, and sauces 2.5.2.3 Statistical Analysis Participant characteristics were analyzed using descriptive statistics. Normality of data distribution was assessed using Kolmogorov-Smirnov test, with a significant level <0.05 to indicate when data was not normally distributed. Values for all analytical data are expressed as means ± SD, or medians and interquartile ranges (IQR), and 2.5th to 97.5th percentile ranges, as appropriate. Statistical tests to address the research question were then based on the normality or non-normality of the data. To summarize, two-way repeated measures ANOVA was used for the stability of WSC in human milk using temperature and storage time as factors. Within women, variability of water-soluble milk cholines in 5 samples was analyzed with calculations of the percent of change from the lowest to highest value. ANOVA was used to compare the average milk WSC across woman, and the minimum and maximum amount of choline per ml and per kcal in the milk from the 20 different women. The relationship between dietary choline intake, as well as the intakes of each choline compound, energy, total fat, saturated fat, polyunsaturated fat and cholesterol intake was analyzed using Pearson’s correlation. The relationship between egg intake and whether the current dietary AI for choline was or was not met was assessed using chi-square test. The relationship between dietary choline intakes and WSC secretion in milk was analyzed using Pearson’s Correlation. All statistical analyses were performed using the IBM SPSS statistics software (IBM SPSS Statistics for Windows, Version 22.0. Chicago, IL, USA: SPSS Inc), using a two-sided model and P values <0.05 as statistically significant.   34 Chapter 3: Results 3.1 Demographics The milk choline stability and variability studies involved 6 and 20 healthy lactating women, respectively. All recruited women met the inclusion criteria and finished the study, none were excluded or withdrew. The mean ± SD age of the mothers was 34.5 ± 1.87 and 32.3 ± 4.07 y for the stability and variability study, respectively. The median age of the infants was 4 mo. (IQR = 3 mo.) in both studies. Table 4. Maternal and infant characteristics a Mean ± SD; bmedian (range). Infant age is equivalent to time from delivery, months of milk production. Supplements intake includes prenatal vitamin and mineral supplements or vitamin D. Mothers included in the stability and variability study had a Caucasian ethnic background in a proportion 2/6 and 11/20, respectively, Table 4. All participants in the stability study were breastfeeding for the first time, while 50% of participants in the variability study had breastfed either one or two infants previously. Characteristic Stability study Variability study n = 6 n = 20 Maternal age (y) 35 ± 1.9a 32 ± 4.1 Infant age (mo.) 4.0 (3.0) b 4.5 (3.0) Ethnic background (n) Caucasian 2 (33%) 11(55%) Latin American 2 (33%) 4(20%) Arabic 0 3(15%) First nation 1 (17%) 1(5%) Chinese 1 (17%) 1(5%) First time breastfeeding 6 (100%) 10 (50%) Supplement use 4 (67%) 7 (35%)   35 3.2 Stability Study Milk choline analysis and storage The WSC were analyzed at different times after collection and with different milk storage conditions to address if the amount could change. Baseline analysis reflected a mean WSC content of 1231 ± 305 µmol/L. Major changes compared to baseline were noted at some storage conditions, such as milk storage at 4°C for 48h with a non-statistically difference of 26% loss and 19% gain after 6 months storage at -80°C, Table 5. Table 5. Change in milk water-soluble choline concentration after different temperatures and storage conditions. Time RT 4°C -20°C -80°C Baseline 1231 ± 305 - - - 30 min 1037 ± 258 (5 , 47%) - - - 60 min 1104 ± 260 (3 , 16%) - - - 90 min 1057 ± 222 (2 , 27%) - - - 120 min 1059 ± 291 (4 , 40%) - - - 3h 1116 ± 200 (2 , 13%) - - - 4h 1217 ± 194 (1 , 26%) - - - 6h - 1208 ± 282 (3 , 25%) - - 24h - 1087 ± 155 (3 , 20%) 1179 ± 154 (4 , 34%) 1144 ± 223 (4 , 30%) 48h - 905 ± 340 (13 , 54%) 1051 ± 283 (4 , 34%) 1228 ± 488 (4 , 36%) 1w - - 1455 ± 504 (24 , 125%) 1236 ± 354 (33 , 121%) 2w - - 1166 ± 301 (15 , 29%) 1166 ± 301 (21 , 47%) 6 mo. - - - 1467 ± 304 (31 , 190%) Data is presented as µmol/L, Mean ± SD (minimum , maximum change compared to baseline values) of water-soluble cholines content in a complete human milk expression from n=6 mothers. -, No analysis done. RT, room temperature.   36 It was also important to address if the WSC FC, Pchol and GPC would also be stable, or whether breakdown of the phosphorylated cholines may occur and alter the content of FC in milk. GPC had minor changes from baseline throughout storage at RT for 4 h, with mean levels of 547 ± 221 µmol/L to 559 ± 209 µmol/L, respectively, Table 6. Table 6. Change in milk glycerophosphocholine concentration after different temperatures and storage conditions. Time RT 4°C -20°C -80°C Baseline 547 ± 221 - - - 30 min 482 ± 170 (5 , 47%) - - - 60 min 539 ± 245 (3 , 16%) - - - 90 min 524 ± 248 (2 , 27%) - - - 120 min 516 ± 252 (4 , 40%) - - - 3h 540 ± 206 (2 , 13%) - - - 4h 559 ± 209 (1 , 26%) - - - 6h - 570 ± 275 (3 , 25%) - - 24h - 585 ± 221 (3 , 20%) 569 ± 210 (4 , 34%) 601 ± 241 (4 , 30%) 48h - 512 ± 311 (13 , 54%) 569 ± 309 (4 , 34%) 606 ± 336 (4 , 36%) 1w - - 691 ± 290 (24 , 125%) 616 ± 223 (33 , 121%) 2w - - 634 ± 348 (15 , 29%) 592 ± 364 (21 , 47%) 6 mo - - - 853 ± 267 (31 , 190%) Data is presented as µmol/L, Mean ± SD (minimum , maximum change compared to baseline values)of glycerophosphocholine content in a complete human milk expression from n=6 mothers. -, No analysis done. RT, room temperature. In contrast, on milk analyzed within the same day of collection, held at room temperature showed a gradual non-statistically significant increase in free choline from 122 ± 38 µmol/L at baseline to 190 ± 75 µmol/L after 4 h. Simultaneously, Pchol decreased in similar amounts, from 561 ± 224 at baseline to 468 ± 132 µmol/L after 4 h, suggesting breakdown of Pchol to FC occurs when milk is   37 left at room temperature, Figure 5. Over this time, the total WSC showed little change, with 1231 ± 305 µmol/L to 1217 ± 194 µmol/L at baseline and 48 hr., respectively. Figure 5. Change in free choline and phosphocholine in milk left at room temperature for up to 4 h after expression. Data is presented as µmol/l. A, free choline; B, phosphocholine. Bars represent the mean and 2.5th – 97.5th percentile. Differently from room temperature results, storage at refrigeration temperatures (4 °C) showed gradual decrease in mean concentration of milk WSC compared to baseline. The highest change on the mean concentration of milk WSC was seen after 48h of storage. Total WSC had a mean decrease of 326 µmol/L for WSC, with 305 µmol/L decrease in milk Pchol concentration and 35 µmol/L decrease in milk GPC, but an increase of FC of only 16 µmol/L. After 24hr of expression, WSC also had the highest change when milk was stored at 4°C with more stable concentrations of WSC in milk stored at -80 °C. Mean FC concentrations, changed Baseline 30 60 90 120 180 240 Time after expression (min) Phosphocholine µmol/L A B Baseline 30 60 90 120 180 240 Time after expression (min) 800 600 400 200 100 0 Free Choline µmol/L   38 from 122 ± 38 µmol/L at baseline to 130 ± 24 µmol/L after 2 wks. of storage at -80 °C and Pchol concentration decreased from 561 ± 225 µmol/L at baseline to 443 ± 63 µmol/L after 2 weeks, with minor changes seen for GPC. In summary, milk stored at -80 °C for 24h, 48h, 1wk and 2 wk. had the lowest changes in the concentrations of total WSC, FC and Pchol compared to the other temperatures of storage. Interestingly, after 6 months of expression, the mean milk WSC concentration changed from 1231 ± 305 µmol/L at baseline to 1467 ± 304 µmol/L, when milk was stored at -80°C, with a increase of 19%. Mean concentrations of milk FC and GPC increased after 6 months in concentration of 104 µmol/L and 306 µmol/L, respectively, Table 7, 8. Of interest, the mean GPC was higher after 6 months of storage, when compared to baseline, to -80°C (P< 0.05), or after storage for 1 wk. at -20°C, then 5 mo. 3 weeks at -80°C when compared to baseline (P<0.05). However, the change of milk GPC seen after 6 months of expression could be the result of phospholipids breakdown after long periods of storage, which remains to be confirmed in future research. Additionally, after 6 mo. of storage at -80°C, we found no significant difference in the mean milk total WSC or individual compounds, when compared to samples stored for 1 week at -20°C then transferred to -80°C for the next 5 months and 3 weeks.   39 Table 7. Change in milk free choline concentration after different temperatures and storage conditions. Time RT 4°C -20°C -80°C Baseline 122 ± 38 - - - 90 min 138 ± 49 (7 , 27%) - - - 120 min 158 ± 55 (0 , 51%) - - - 3 h 177 ± 64 (8 , 66%) - - - 4 h 190 ± 75 (31 , 79%) - - - 6 h - 129 ± 49 (2 , 28%) - - 24 h - 182 ± 115 (20 , 95%) 133 ± 66 (4 , 48%) 128 ± 46 (2 , 38%) 48 h - 138 ± 62 (8 , 54%) 164 ± 22 (6 , 67%) 127 ± 38 (16 , 115%) 1 wk. - - 175 ± 48 (9 , 126%) 131 ± 31 (7 , 37%) 2 wk. - - 253 ± 52 (55 , 82%) 130 ± 24 (11 , 21%) 6 mo. - - - 226 ± 64 (15 , 147%) Data presented as µmol/L, mean free choline concentration ± SD (minimum , maximum change compared to baseline values) from n = 6 participants. -, no analysis done. RT, room temperature. Regardless the temperature, FC tended to increase in concentration with time, opposite to Pchol, which showed a tendency to decrease in concentration in the milk, Table 8. This might explain the low variability of the total WSC concentration in the milk and also be suggestive of a breakdown of Pchol into FC secondary mainly to the time of storage.   40 Table 8. Change in milk phosphocholine concentration after different temperatures and storage conditions. Time RT 4°C -20°C -80°C baseline 561 ± 225 - - - 90 min 395 ± 148 (1 , 48%) - - - 120 min 386 ± 219 (6 , 47%) - - - 3h 399 ± 169 (2 , 50%) - - - 4h 468 ± 132 (8 , 45%) - - - 6h - 510 ± 263 (8 , 45%) - - 24h - 320 ± 107 (8 , 55%) 476 ± 99 (1 , 12%) 415 ± 150 (10 , 70%) 48h - 256 ± 42 (10 , 51%) 318 ± 52 (10 , 51%) 365 ± 68 (8 , 37%) 1w - - 471 ± 200 (10 , 43%) 489 ± 228 (8 , 53%) 2w - - 364 ± 9 (2 , 29%) 443 ± 63 (3 , 55%) 6 mo. - - - 582 ± 260 (8 , 32%) Data presented as µmol/L, mean phosphocholine concentration ± SD (minimum , maximum change compared to baseline values) from n = 6 participants. -, no analysis done. RT, room temperature. Based on these results on variability on milk WSC and individual WSC, all subsequent work on intra and inter-individual variability in milk cholines was done such that all participants were requested to store their collected milk at -20°C or -10°C in their home immediately after collection. The milk samples were then collected the day following sample collection, transferred to -80°C to the laboratory within 20 min of pick up on ice, and analyzed for WSC in less than 2 wks. after expression. 3.3 Variability study The intra- and inter-individual variability of the total and individual WSC in milk were determined for 20 healthy lactating women, each providing 5 different milk samples collected within one day, at specific times.   41 Composition of water-soluble cholines in human milk First, to compare variability in milk water-soluble cholines across women, the average of the five milk samples was determined for each woman, then the 20 women compared, Table 9. Table 9. Concentration of water-soluble cholines (µmol/L) in milk samples of healthy women. Concentrations of water-soluble cholines were determined as the average of the 5 samples for each woman, then the average for the group milk cholines analyzed across women. a Skewed due to one upper extreme outlier, mean ± SD of 138 ± 55 µmol/L for 19 women without outlier (value for n=20 women: 149 ± 73). Range of mean presented is the minimum and maximum mean for the 20 women included. All other data shown are normally distributed. The total FC concentration in milk produced on one day was not normally distributed across the women (n=20), explained by one upper extreme concentration, with the WSC normally distributed for the other 19 women. The median (IQR) of FC in milk was 131 (105) for n=20 participants, with a mean ± SD of 138 ± 55 µmol/l for the 19 women after excluding the woman with very high FC. The mean ± SD Pchol, GPC, and total WSC were present at 514 ± 244, 617 ± 255 and 1279 ± 277 µmol/l, respectively, Table 9, n=20. According to the data, GPC was on average the most abundant WSC compound in the mid-feed mature milk samples from this set of women; using mean values, GPC contributed to 48% of total WSC, followed by Pchol at 40% and 12% as FC. However, the data on the human milk cholines showed a large SD and wide IQR, which in turn are the evidence of large variability in the amount Compound Mean ± SD Range of means Median (IQR) Free Choline a 138 ± 55 62 , 362 131 (105) Phosphocholine 513 ± 244 106 , 1022 482 (335) Glycerophosphocholine 617 ± 255 131 , 1074 623 (311) Total Water-Soluble Cholines 1279 ± 277 830 , 1774 1254 (421)   42 of the different WSC in milk from different women. The mean amounts of choline compounds in milk varied among women as much as 6 fold for FC, 9.6 fold for Pchol, 8.2 fold for GPC, but with only a 2.1 fold difference for the total-WSC from the lowest to highest. The concentration of Pchol was inversely correlated with the milk FC (P=0.019), with no significant relationship of Pchol to GPC. Maternal age and ethnicity showed no relationship to the concentration of the individual WSC compounds in the milk. However, the milk FC was positively associated with length of breastfeeding (P=0.047) i.e. how long the post-partum time was. The inter-individual variability in water-soluble cholines in milk was next addressed as the percent change in FC, Pchol, GPC and total WSC in the 20 participants, using the mean concentration of choline for the 5 samples in a day for each subject. The mean variability across woman for the concentrations of FC, Pchol, GPC and total WSC in human milk were 50%, 46%, 85% and 42%, respectively, suggesting the variability is higher in the individual forms of WSC and not the total of WSC in milk. The individual variability in the 5 cholines in the different milk samples showed the mothers differed in the extent of variability among samples. As consequence, the determination of the inter-individual variability in WSC is complicated by the inconsistent and sometimes high variability of WSC within a mother. As such, using the mean concentration for WSC for a mother is not representative of the inter-individual variability. Figure 6 shows the median 2.5-97.5 percentile for the 5 milk samples for each of the 20 women.   43 Figure 6. Intra-individual variability in the milk water-soluble cholines. Boxplots show median, 25-75th and 2.5 to 97.5 percentile, with outliers *, O. The x-axis shows individual participants, labeled A-T; Data for each participant is 5 different milk samples collected at separate feeds on a single day. Note, the Y-axis scale is different for the different compounds. Participant Participant Free choline (µmol/L)  Phosphocholine (µmol/L)  Glycerophosphocholine (µmol/L)  Total water-soluble choline (µmol/L)  A B C D   44 The concentration of each compound was determined in milk samples from 5 different times of the day, for which the ranges and percent of change, as well as concentration of FC, Pchol, GPC and total WSC in each mother’s milk are in Tables 10, 11, 12, 13 and 14, respectively. Table 10. Water-soluble cholines in 5 milk samples from different women, minimum and maximum concentration and the percent difference in the highest and lowest. Subject Free choline Phosphocholine Glycerophosphocholine Total water-soluble cholines Range (µmol/L) Difference (%) Range (µmol/L) Difference (%) Range (µmol/L) Difference (%) Range (µmol/L) Difference (%) A 78 , 114 45 234 , 433 85 675 , 1268 88 994 , 1786   80 B 176 , 226 29 212 , 332 57 693 , 927 34 1127 , 1480   31 C 236 , 474 100 441 , 534 21 213 , 1363 539 1113 , 2117 90 D 128 , 272 113 350 , 487 39 670 , 882 32 1207 , 1518 26 E 122 , 224 83 225 , 362 61 460 , 566 23 855 , 1056 23 F 112 , 169 51 235 , 505 115 558 , 694 24 963 , 1294 34 G 127 , 172 36 440 , 533 21 910 , 1176 29 1485 , 1859 25 H 107 , 142 32 213 , 338 59 535 , 819 53 867 , 1242 43 I 78 , 118 51 906 , 1188 31 538 , 756 41 1700 , 1896 12 J 91 , 118 29 632 , 747 18 435 , 567 30 1216 , 1413 16 K 72 , 116 62 596 , 808 36 420 , 547 30 1169 , 1352 16 L 176 , 282 60 382 , 559 47 99 , 187 89 752 , 955 27 M 50 , 86 69 873 , 988 13 123 , 514 316 1091 , 1469 35 N 99 , 130 32 548 , 713 30 509, 1955 284 1239 , 2729 120 O 68 , 81 19 815 , 1113 36 379 , 900 138 1263 , 1894 50 P 100 , 149 49 541 , 703 30 279 , 429 54 938, 1281 37 Q 209 , 254 21 288 , 493 71 154 , 576 273 652 , 1308 100 R 143 , 221 54 308 , 402 31 603 , 772 28 1120 , 1341 20 S 52 , 69 33 481 , 662 37 769 , 1261 64 1365, 1794 31 T 201 , 258 28 79 , 138 75 494 , 669 36 786 , 1041 33 Data is presented is the minimum and maximum and range as µmol/L and the % difference (calculated as highest from lowest) for the 5 samples.   45 Table 11. Free choline (µmol/L) in human milk in samples collected at 5 time points on a single day. Subject Collection time Mean (µmol/L) 1 2 3 4 5 A 84 103 78 104 114 97 B 178 219 176 226 220 204 C 314 236 474 453 332 362 D 149 130 128 272 152 166 E 122 154 161 164 224 165 F 115 137 169 164 112 139 G 127 148 135 162 172 149 H 114 107 127 120 142 122 I 78 89 118 87 82 91 J 91 102 110 118 108 106 K 72 80 116 80 87 87 L 208 179 282 271 176 223 M 56 60 79 86 51 66 N 112 119 99 130 116 115 O 69 76 81 68 69 72 P 149 107 137 100 122 123 Q 238 231 217 254 210 230 R 143 157 221 209 152 176 S 52 68 69 60 59 62 T 258 253 201 213 214 228 Each letter denotes a participant. Data is free choline as µmol/L. Collection times are as follows: 1, early morning; 2, before lunch; 3, after lunch; 4, after dinner; 5, last feed of the day.   46 Table 12. Phosphocholine (µmol/L) in human milk samples collected at 5 time points on a single day. Subject Collection time Mean (µmol/L) 1 2 3 4 5 A 234 330 263 342 433 320 B 283 272 258 212 332 271 C 441 466 491 534 506 488 D 487 364 384 350 390 395 E 333 270 235 362 225 285 F 505 358 235 247 412 351 G 485 509 440 521 533 498 H 308 338 262 213 283 281 I 906 936 1120 1188 958 1022 J 701 632 690 747 728 700 K 687 596 608 808 674 675 L 559 559 423 382 457 476 M 899 879 988 881 873 904 N 548 713 582 654 658 631 O 815 1113 913 919 942 940 P 703 541 602 592 686 625 Q 493 335 441 393 288 390 R 373 308 402 369 345 359 S 481 530 662 537 550 552 T 138 82 114 79 89 100 Each letter denotes a participant. Data is presented as concentration of phosphocholine as µmol/L. Collection times are as follows: 1, early morning; 2, before lunch; 3, after lunch; 4, after dinner; 5, last feed of the day.   47 Table 13. Glycerophosphocholine (µmol/L) in human milk samples collected at 5 time points on a single day. Subject Collection time Mean (µmol/L) 1 2 3 4 5 A 675 998 1188 1268 1240 1074 B 780 758 693 848 927 801 C 1363 1115 221 213 276 637 D 882 712 707 670 790 752 E 566 479 460 530 461 499 F 674 558 559 562 694 610 G 1086 1038 910 1177 1102 1063 H 676 651 820 535 818 700 I 720 675 538 621 756 662 J 551 482 435 548 567 517 K 477 547 445 464 420 471 L 187 107 104 99 155 131 M 514 151 124 162 167 223 N 579 548 650 510 1955 848 O 379 489 900 514 432 543 P 429 290 354 279 388 348 Q 576 202 259 270 155 292 R 604 772 603 763 703 689 S 1261 842 865 769 828 913 T 646 559 520 494 669 578 Each letter denotes a participant. Data is presented as concentration of glycerophosphocholine as µmol/L. Collection times are as follows: 1, early morning; 2, before lunch; 3, after lunch; 4, after dinner; 5, last feed of the day.   48 Table 14. Total water-soluble cholines (µmol/L) in human milk samples collected at 5 time points on a single day. Subject Collection time Mean (µmol/L) 1 2 3 4 5 A 994 1432 1529 1714 1786 1491 B 1242 1248 1127 1286 1480 1277 C 2117 1818 1186 1200 1113 1487 D 1518 1207 1219 1292 1331 1313 E 1021 904 855 1056 911 949 F 1294 1053 963 973 1218 1100 G 1698 1695 1485 1859 1808 1709 H 1098 1096 1209 867 1242 1102 I 1705 1700 1775 1896 1796 1774 J 1343 1216 1235 1413 1403 1322 K 1235 1222 1169 1352 1182 1232 L 955 846 808 752 788 830 M 1469 1091 1190 1128 1091 1194 N 1239 1380 1330 1293 2729 1594 O 1263 1678 1894 1500 1443 1556 P 1281 938 1093 972 1196 1096 Q 1308 768 917 916 652 912 R 1120 1237 1226 1341 1199 1225 S 1794 1441 1597 1365 1437 1527 T 1041 895 835 786 972 906 Each letter denotes a participant. Data is presented as concentration of total water-soluble choline as µmol/L. Collection times are as follows: 1, early morning; 2, before lunch; 3, after lunch; 4, after dinner; 5, last feed of the day. It is important to note that, even within a mother, the concentration of WSC differed in milk collected at different times. In some women, FC was in similar amounts and equal or higher to the amount of Pchol or GPC. Of importance, variability in milk cholines showed no apparent pattern of change across women at different times during the day, i.e. there was no definitive evidence of a diurnal pattern.   49 The within women variability was assessed with the percent of difference of the concentration of the 5 samples retrieved on the same day. The ranges and percent of change for each woman are mainly lower than 100%. No set of 5 milk samples from a woman showed differences of less than 10% in any of the choline compounds. The same day changes in milk FC, Pchol and GPC within a mother also did not show any consistent diurnal pattern, with the WSC concentrations, peaking at different times during the day among the women.  As understood from the previous tables, it might not be correct to show milk WSC concentration for a mother based on a single milk sample, particularly when assessing the relation of milk choline to infant outcome using data from an individual sample. This is because a single milk sample may not represent the amount of total or individual choline compounds in the human milk. Milk caloric content as source of choline variability It is well known that milk volume (ml/day) and caloric density (kcal/dl) vary widely among women. For that reason, we investigated whether or not the intra-individual and inter-individual variability found for the water-soluble cholines, when expressed per volume of milk (/ml) was explained by variability in the milk caloric density. The kcal/dl of milk per participant is summarized on Table 15. With the data on caloric content/dl, the choline /kcal was determined for each milk sample. The ranges and % change of milk choline/kcal is shown in Table 16.   50 Table 15. Caloric content (kcal/dL) of human milk samples collected at 5 time points on a single day. Subject Collection time Mean ± SD 1 2 3 4 5 A 115 58 78 78 68 79 ± 22 B 45 68 73 91 73 70 ± 16 C 51 59 70 56 58 59 ± 7 D 80 71 68 73 51 69 ± 11 E 87 89 92 63 92 85 ± 12 F 39 64 53 69 62 57 ± 12 G 71 96 85 49 75 75 ± 18 H 44 49 65 67 70 59 + 12 I 72 74 58 51 52 61 ± 11 J 69 102 84 75 51 76 ± 19 K 64 72 99 67 100 80 ± 18 L 55 75 76 60 47 63 ± 13 M 51 53 58 65 57 57 ± 5 N 62 52 106 83 72 75 ± 21 O 76 80 90 58 61 73 ± 13 P 56 62 61 51 67 59 ± 6 Q 78 72 61 85 42 68 ± 17 R 78 84 81 74 55 74 ± 11 S 51 78 88 60 53 66 ± 16 T 60 57 63 57 69 61 ± 5 Each letter denotes one healthy breastfeeding woman. Sampling times are: 1, early morning; 2, before lunch; 3, after lunch; 4, after dinner; 5, last feed of the day.   51 Table 16. Range of water soluble-cholines (µmol/kcal) in human milk collected at 5 time points in a day. Subj. FC Pchol GPC WSC Range (µmol/kcal) Difference(%) Range (µmol/kcal) Difference(%) Range (µmol/kcal) Difference (%) Range (µmol/kcal) Difference (%) A 0.07 , 0.18 143 0.20 , 0.64 213 0.59 , 1.82 210 0.86 , 2.63 204 B 0.24 , 0.40 64 0.23 , 0.63 170 0.93 , 1.73 86 1.41 , 2.76 95 C 0.40 , 0.81 102 0.70 , 0.95 36 0.32 , 2.67 748 1.69 , 4.15 145 D 0.18 , 0.37 103 0.48 , 0.76 60 0.92 , 1.55 69 1.70 , 2.61 54 E 0.14 , 0.26 85 0.24 , 0.57 135 0.50 , 0.84 68 0.93 , 1.68 80 F 0.18 , 0.32 77 0.36 , 1.29 261 0.81 , 1.73 112 1.41 , 3.32 135 G 0.15 , 0.33 11% 0.52 , 1.06 105 1.07 , 2.40 124 1.75 , 3.79 117 H 0.18 , 0.26 46 0.32 , 0.70 120 0.80 , 1.54 92 1.29 , 2.50 93 I 0.11 , 0.20 87 1.26 , 2.33 85 0.91 , 1.45 59 2.30 , 3.72 62 J 0.10 , 0.21 112 0.62 , 1.43 130 0.47 , 1.11 136 1.19 , 2.75 131 K 0.09 , 0.12 36 0.61 , 1.21 96 0.42 , 0.76 81 1.18 , 2.02 71 L 0.24 , 0.45 89 0.56 , 1.02 83 0.14 , 0.34 149 1.06 , 1.74 63 M 0.09 , 0.14 53 1.36 , 1.76 30 0.21 , 1.01 373 1.74 , 2.88 66 N 0.09 . 0.23 145 0.55 , 1.37 150 0.61 , 2.71 343 1.25 , 3.79 202 O 0.09 , 0.12 30 1.01 , 1.58 56 0.50 , 1.00 101 1.66 , 2.59 56 P 0.17 , 0.27 54 0.87 , 1.26 44 0.47 , 0.77 64 1.51 , 2.29 51 Q 0.30 , 0.50 67 0.46 , 0.72 57 0.28 , 0.74 164 1.07 , 1.68 57 R 0.18 , 0.28 54 0.37 , 0.63 71 0.74 , 1.28 72 1.44 , 2.18 52 S 0.08 , 0.11 41 0.68 , 1.04 53 0.98 , 2.47 151 1.81 , 3.52 94 T 0.31 , 0.44 43 0.13 , 0.23 79 0.83 , 1.08 30 1.33 , 1.74 31 Data presented are the lowest and highest µmol/kcal, % difference in 5 samples of milk collected from n=20 participants. FC= free choline; Pchol= phosphocholine; GPC= glycerophosphocholine; WSC= total water-soluble cholines. When choline is expressed per kcal milk, variability in the milk choline was still present. The milk cholines content per kcal milk showed variability on the choline content for the majority of the participants with no clear evidence that variability in the WSC is due to variability in milk volume produced. Variability was lower in the FC in just 3 subjects, in which variability decreased in 10%,   52 26% and 16%, in 1 subject for Pchol with 14% of variability reduction and in 3 subjects for GPC, where the variability was lower in 37%, 109% and 6%, compared to choline content in the milk per ml (Appendix 3). For total WSC, variability was increased for all participants when determined /kcal, except for 1 subject who had a minor reduction of 2% in the variability, compared to the values obtained as WSC/ml of milk. All the changes were seen in different subjects. Milk caloric content also showed no relation to the concentration of individual or total WSC of the milk, in the milk collection times in one day. Grams of fat content were also analyzed to determine if a relation existed to the WSC content of the milk, with no association found and similar results to the ones for energy content of the milk. Diet as source of the variability The mothers diet showed a wide variety of foods, also with a high range of energy and macronutrient intake as seen in Table 17. The mean caloric intake of the 20 study participants was 2400 ± 770 kcal/day, with a mean content of protein of 117 g, carbohydrate of 279 g, and a fat of 99 g, representing on average of 18%, 46% and 36% of total energy, respectively.   53 Table 17. Dietary intakes of energy, macronutrient, fatty acid and cholesterol among lactating women. Dietary intake Energy intake (kcal) 2400 ± 770 Protein (g) 112 ± 52 Carbohydrate (g) 279 ± 99 Fat (g) 99 ± 39 Saturated, % total fat 41 % Monounsaturated, % total fat 41 % Polyunsaturated, % total fat 18 % Trans fats (g) 1.04 (1.55) Cholesterol (mg) 395 (503) Results shown as mean ± SD; or median (IQR), n=20 women. Table 18. Dietary choline intakes (mg) among lactating women, n=20. Dietary choline Dietary Intake, mg Mean ± SD Median (IQR) Range Total choline, mg 471+ 251 423 (264) 198 , 1310 Water-soluble cholines, mg 151 ± 57 144 (54) 71 , 312 Free Choline (mg) 80 ± 30 76.4 (36) 42 , 160 Phosphocholine (mg) 14 ± 7 13 (10) 5 , 29 Glycerophosphocholine (mg) 57 ± 30 52 (37) 18 , 123 Lipid-bound cholines, mg 321 ± 240 238 (256) 95 , 1097 Phosphatidylcholine (mg) 293 ± 218 212 (240) 91 , 980 Sphingomyelin (mg) 27 ± 24 23.9 (18) 4.3 , 117 Data is presented as mean ± SD, median (IQR) and range intake of each choline compound in the maternal diet from n=20 women, as mg.   54 As expected, the lipid bound cholines were the important sources of choline in the maternal diet, attributing to at least 1/3 of the total choline intake, with about 92% of the lipid-bound choline as PC. The major dietary WSC was FC, followed by GPC then Pchol representing 53%, 37.7%, and 9.2%, respectively, Table 18, 19. Notably, the choline composition of the diet is distinctly different from the choline composition of human milk. Table 19. Contribution of different forms of cholines to the maternal diet Total choline Water-soluble cholines Lipid-bound cholines Water soluble, % 32.5 Free Choline, % 53.2 Phosphocholine, % 9.3 Glycerophosphocholine, % 37.5 Lipid-bound, % 67.5 Phosphatidylcholine, % 91.6 Sphingomyelin, % 8.4 Data is mean % of each choline form in the diet of 20 breastfeeding women who collected their milk 5 times in one day. The main food groups of choline in the diet of the women (n=20) are summarized in Table 20. The food groups giving the highest amount of water-soluble cholines were mixed dishes, 22% from the group containing sushi, hamburgers, casseroles, lasagna, pizza; 20% from dairy products; and 15% from fruit and vegetables. The lipid-bound cholines were mainly from eggs, mixed dishes, and meat and fish that gave an average of 37%, 29% and 15% of their intake, respectively. The major dietary source of choline was the lipid-bound choline, with eggs and mixed dishes alone   55 contributing to more than 50% of total choline intake, Table 20. Notably, one single food eaten by an individual (eggs) could provide the same or more choline in the diet of subjects. Table 20. Food sources of choline in the maternal diet Food group % Water-soluble choline % Lipid-bound choline % Total choline Cereals 10 3 6 Dairy 20 2 8 Vegetable oils 0 0 0 Cake and pastry 8 6 7 Egg 1 37 25 Fruit, vegetables and legumes 15 4 7 Meat & fish 8 15 13 Mixed dishes 22 29 27 Nuts & seeds 4 2 3 Other 11 2 5 Data is presented as percent of total choline, water-soluble choline and lipid-bound choline provided by each one of the food groups consumed. WSC, water-soluble cholines; LBC, lipid bound cholines; TC, total choline. The total choline intake, mean ± SD and median (IQR) for study participants who consumed each food group is in Table 21, as well as the average intake data for all study participants. Intakes of total choline were found to be highly variable across women. It is important to note that not all women were consuming foods from all food groups; for example, the consumption of choline from eggs is underestimated if analysis is based in all study participants, while only 11 participants were consuming eggs.   56 Table 21. Mean concentration of total choline in the maternal diet, from different food sources in all subjects and consumers. Food group Food consumers All study participants n=20 n Mean ± SD Median (IQR) Mean ± SD Median (IQR) Cereals 18 27 ± 20 22 (23) 24 ± 20 19 (28) Dairy 18 38 ± 39 27 (49) 34 ± 39 19 (49) Vegetable oils 11 1 ± 2 0 (2) 1 ± 2 0 (1) Cake & Pastry 16 41 ± 33 36 (42) 31 ± 33 21 (58) Egg 11 174 ± 113 126 (133) 96 ± 121 16 (158) Fruit, vegetables & legumes 18 37 ± 35 33 (28) 33 ± 35 28 (30) Meat & fish 15 77 ± 64 48 (65) 57 ± 64 41 (54) Mixed dishes 18 137 ± 157 123 (129) 118 ± 152 75 (144) Nuts & seeds 9 28 ± 38 16 (15) 13 ± 29 0 (16) Other 19 22 ± 21 18 (16) 21 ± 21 18 (23) Data is presented as Mean ± SD and Median (IQR) intake of each food group, in mg. Food consumers: study participants who reported consumption of foods in each food group. All study participants: all study participants included (n=20). Total choline intake was skewed with a higher mean than median. The median total choline intake was 426 mg/day (5th- 95th percentile: 248 - 807). The dietary AI recommendation for choline during lactation is 550 mg/d (DRI Choline, 1998). According to this AI, 75% of the mothers were not meeting the AI recommendation. The AI for non-lactating and non-pregnant women is 450 mg/day. According to the median total choline intake in this set of women, more than half of the participants are not meeting the AI.   57 When expressing the choline intake per 2000 kcal, the distribution of choline intake is closer to a normal distribution, meaning that the choline intake is related to the energy consumption of the mother, Figure 7.  Figure 7. Distribution of total choline intake for lactating mothers. A, total dietary choline intake as mg/d; B. choline as mg/2,000 kcal. n=20; mean ± SD choline dietary intake is 478 ± 254; median (IQR) is 426 (248); range, 198 to 1310 mg/d. As egg as a sole food was responsible for a major amount of choline in some women, an association was explored between egg intake and meeting the AI. This provided a positive value of X2=5.455, P= 0.020, showing that eating eggs helps to meet the dietary recommendations for choline. Also the amount of eggs consumed was important, since it is positively related with meeting the recommended intake of choline (X2=6.857, P=0.032), suggesting 2 eggs/day might make mothers more able to consume adequate levels of choline. None of the mothers that were not consuming eggs met the AI, with 57% of the mothers who consumed 2 or more eggs met the AI and 25% who consumed 1/d egg met the AI, Table 22. Some other variables were explored as Number of Participants 8 6 4 2 0 Total choline in diet (mg/d) 0 250 500 750 1000 1250 Number of Participants 6 5 4 3 2 1 0 100 200 300 400 500 600 700 Choline intake (mg/2000 kcal) A B   58 being potentially related with meeting the dietary recommendations, such as milk, yogurt, and total dairy intake, but significant association was not found. Table 22. Choline intake of lactating women with daily intakes of ≥2, 1 or no egg, and the participants who met current AI for choline intake. To better understand how important it may be to meet the AI for choline, we categorized mothers according to meeting or not meeting the current AI. All choline compounds, except FC, were higher in the diet of mothers who met the AI, compared with those who did not, although this difference in choline intake to milk choline was not significant. WSC, Pchol, GPC and FC content of milk had no statistically significant differences in milk in mothers who met the AI and those who did not meet the AI, probably due the small sample size of the study (FC P=0.993; Pchol P=0.329; GPC P=0.364; WSC P= 0.132). When consuming AI, participants showed similar WSC concentrations compared to some mothers not consuming the current choline AI, however, none of the participants consuming the AI showed lower concentration as seen in some participants not consuming the AI, Figure 8. Total choline intake (mg/d)at different egg intake n Mean Median Range % Meeting AI ≥ 2 eggs2 7 694 562 425 - 1310 57% 1 egg 4 436 423 250 - 649 25% No eggs 9 321 281 197 - 461 0% AI, adequate Intake for choline in lactating women is 550 mg/d. 2One woman consumed 3 eggs with a daily total choline intake of 720 mg   59 Figure 8. Milk total water-soluble cholines (µmol/L) from women meeting (n=5) and not meeting (n=15) the dietary choline AI of 550 mg/d. Boxplots show median, 25-75th and 2.5 to 97.5 percentile. The x-axis shows AI choline consumption status (meeting vs. not meeting AI).The Y axis shows the total milk water-soluble choline concentration. We addressed if the WSC content in milk was related to intake of total choline in the maternal diet, finding that dietary PC had a trend to a positive association r=0.413, P=0.070, although non significant maybe due to the small sample size, Figure 9. 1800 1600 1400 1200 1000 800 Water-soluble cholines (µmol/L) Not meeting AI Meeting AI  AI consumption status   60 Figure 9. Relation between dietary phosphatidylcholine (mg) and milk water-soluble cholines (µmol/L). n = 20, r = 0.413, P = 0.070. In human milk, the lipid-bound cholines PC and SM are found in the phospholipids forming the milk fat globule membrane. The phospholipids in human milk represent around 0.5 to 1% of total milk fat (Gallier, 2010). The amount of PC and SM is reported in papers as being highly variable, with 8 to 45% PC and 4 to 29% SM in human milk phospholipids (Graves, 2007; Contarini, 2013). An estimation of the PC and SM content in human milk was set as 0.5% of the total fats, 30% as PC and 20% as SM. It is important to keep in mind these are proxy concentrations for a theoretical PC and SM content for a theoretical calculation of the PC and SM provided in the 5 milk samples in a day for each mother, as seen in Table 23. 400 500 600 700 800 900 1000 0 100 200 300 400 500 600 700 800 900 1000 Water-soluble choline in human milk (µmol/L) Phosphatidylcholine in diet (mg)   61 Table 23. Estimation of total phospholipids, phosphatidylcholine and sphingomyelin in milk samples of healthy breastfeeding women. Total fat, g/dl Phospholipids, mg/dl Phosphatidylcholine, mg/dl Sphingomyelin, mg/dl A 5.36 26.8 8.0 5.4 B 4.22 21.1 6.3 4.2 C 2.82 14.1 4.2 2.8 D 4.06 20.3 6.1 4.1 E 5.82 29.1 8.7 5.8 F 2.84 14.2 4.3 2.8 G 4.20 21.0 6.3 4.2 H 3.06 15.3 4.6 3.1 I 3.22 16.1 4.8 3.2 J 5.02 25.1 7.5 5.0 K 5.42 27.1 8.1 5.4 L 3.48 17.4 5.2 3.5 M 2.66 13.3 4.0 2.7 N 4.78 23.9 7.2 4.8 O 4.36 21.8 6.5 4.4 P 3.26 16.3 4.9 3.3 Q 4.28 21.4 6.4 4.3 R 4.74 23.7 7.1 4.7 S 3.60 18.0 5.4 3.6 T 3.08 15.4 4.6 3.1 Total fat is the mean fat content of 5 milk samples in one day, g/dL. Phospholipids, phosphatidylcholine and sphingomyelin are estimated at 0.5% with 30% of this from phosphatidylcholine and 20% from sphingomyelin, as mg/dl. With the estimation of PC and SM, a proxy of the total WSC and lipid-bound choline in human milk was calculated and used to determine whether or not the total amount of milk choline in woman meeting their AI (550 mg/d) would provide to the infant with milk with 1500-2000 µmol/L. The milk of only 7/20 women (35%) had total choline to deliver to their infant with 125 mg/d, equivalent to the 1500-2000 µmol/L, Figure 10.   62 Figure 10. Water-soluble choline (colored bars), as mean and estimated lipid-bound choline (open bars, **) for each subject. The milk total choline of 1500-2000 µmol/L is needed to meet the AI for infant of 125 mg/d. We determined if the mothers meeting the choline AI had higher WSC in their milk than mothers who did not. With this dataset, only five women met the AI for choline, and 4 of these women (80%) had milk WSC that met the infant AI of 125 mg/d, equivalent to 1200 µmol/day, assuming a mean milk output and intake of 0.78 L/d. For the mothers who did not met the AI (n=15), only 30% produced milk with choline of 1500-2000 µmol/L. It is important to remember that most of the choline in human milk is in the water-soluble form (about 90% of the total choline), thus the milk choline achieving 1500-2000 µmol/L of total choline is driven mainly by the WSC. 0 500 1000 1500 2000 A B C D E F G H I J K L M N O P Q R S T Lipid-Bound Choline** Free Choline Phosphocholine Glycerophosphocholine Water-soluble cholines + Estimation of lipid-bound cholines in human milk, µmol/L Participant   63 In summary, we found mean concentrations of milk total WSC to have minor changes in almost all storage conditions, with GPC having low changes in all situations, except after 6 months of storage. In contrast, FC and Pchol showed increase and decrease in human milk with time of storage, respectively, with faster changes seen when milk was stored at room temperature, with a proposed breakdown of Pchol into FC. Changes in all WSC were lower when milk samples were stored at -80°C for 24hr, 48hr, 1wk and 2 wk compared to milk samples stored at 4°C and -20°C. Milk samples stored for 6 months at -80°C however, showed high increase in the concentration of GPC compared to baseline, for a reason still unknown. This change also led to a large increase in the total concentration of milk WSC after 6 months of milk storage. With the variability study, we could find that using mean concentrations of individual and total WSC was not representative of milk samples from each woman, since no clear patterns for the concentration of any of the milk WSC across women were found. Additionally, variability was large in milk samples from the same woman at different time in a day were analyzed, and with large differences in the degree of variability across women. The major variability seen was in individual components of WSC, rather than the total amount of milk WSC, with no clear patterns of change seen for the different women. Analyzing the milk choline in relation to the milk kcal content did not decrease the variability in any of the WSC. Maternal dietary choline intake for this group was also unrelated to the variability of the milk concentration of FC, Pchol, GPC or total WSC. However, dietary data was based on one day and including only 20 women. Our data suggests that mothers consuming the AI of 550 mg/d or more produce milk with higher content of WSC compared to the mothers consuming less than the AI, with maternal dietary PC intake showing a trend in the relation to the WSC content in the milk, which stills needs to be confirmed in a study including a larger number of participants, with dietary data from more than one day.   64 Chapter 4: Discussion, Limitations and Future Directions 4.1 Discussion In this study, the storage effect was addressed on milk WSC from samples of 6 women, to better understand the changes in WSC resulting from different storage conditions. Storage length and temperature were found to have an overall minor influence on the amount of WSC across milk samples, with similar results for GPC at all storage conditions, except after 6 months of expression. In contrast, milk levels of FC and Pchol were inversely altered when left at room temperature, but also during refrigeration conditions. After assessing storage conditions, the variability in milk choline from 20 women was identified. Milk samples for the variability study were stored overnight at -20°C in the participants homes, and a maximum of 2 weeks at -80°C after transfer to the laboratory. Milk WSC average content across women showed GPC as the most abundant WSC compound, contributing to 48% of total WSC, then Pchol at 40% and FC at 12%. However, some milk samples showed high Pchol or GPC concentrations, with mean amounts of the different WSC varying as much as 6 fold for FC, 9.6 fold for Pchol, 8.2 fold for GPC. The total WSC however, only varied 2.1 fold. In some participants FC was consistently equal to Pchol or GPC in their milk, with some also showing FC concentrations higher than GPC. The wide variability in choline composition showed that the mean concentrations of the WSC were not representative of the WSC composition across women. Analysis of individual variability showed no consistent amounts or degree of variability of individual WSC in milk, with patterns of change being inconsistent across women. The variability in choline content was not related to the caloric density of the milk, with determination of milk WSC/kcal, not decreasing the within or among individual variability in milk choline. Dietary choline intake showed no relation to any of the individual or total milk WSC   65 concentrations. Dietary PC, however, showed a positive trend to significant positive association with the amount of total milk WSC. Participants who consumed the choline AI showed milk WSC to be present in higher amounts compared to mothers not consuming the choline AI. The role of egg intake on milk choline still needs to be confirmed, but our data suggest it is the major food source of milk total choline and PC, therefore helping mothers to achieve AI for lactation choline more easily. To date, only one published study included an experiment related to the stability of choline. Zeisel et al., (1986) compared fresh human milk to samples incubated for 15 min at 37°C, and to samples frozen for 72 h at -10°C, then incubated 37°C x 15 min (n = 3/group). They labeled PC and SM in the fresh milk and determined the amount of these lipid compounds after the storage and incubation to address if milk phospholipases would break down the phospholipids into free choline. They found no differences on the concentration of any of the choline compounds in milk, and concluded phospholipase activity did not form free choline from PC or SM. One probable reason for the results is that PC and SM are mostly present in human milk as part of the milk fat globule membrane with high stability. These experiments also lasted for short periods of time, and may not have been long enough to observe changes. Of importance, Zeisel et al., (1986) did not determine the amount or breakdown of water-soluble cholines, which are ready available and in major amounts in the aqueous phase of the milk. For short lengths of time, their findings of FC are consistent with ours, suggesting also that if frozen for 3 days, FC, which was found in our study to be the most variable compound, would be stable. One of the breakdowns we found was not investigated in the Zeisel et al., study, since in our study Pchol breaks down into FC at room temperature. Zeisel et al., (1986) and our findings combined   66 suggests that for a short period of storage, both water- and lipid-soluble cholines would remain in similar concentrations when compared to fresh milk, although more research is needed in this area. In regards to GPC, we saw an increase in the mean concentration in the milk after 6 months of storage, which could be due to a breakdown of the phospholipids in the milk fat globule membrane (i.e. PC or SM). Phospholipids showed to be stable in the study performed by Zeisel et al., (1986) at short periods of storage, but were not addressed after longer periods. No further information about choline stability is available, however, there are many studies that have determined the storage effect on other related nutrients in the human milk. In a study performed by Lev et al., (2014), fat, energy and carbohydrates concentration in preterm milk had a statistically significant lower content in milk associated with storage at -80°C (mean time = 44 d, range: 8-83d; P<0.0001). Similar results were found by Garcia-Lara et al., (2012), term milk stored for up to 3 months at -20°C, which is the temperature used in common household freezers. They found a significant decrease of fat and energy in human milk, even after controlling for some confounding variables, such as homogenization (P < 0.001). Slutzah el al., (2010) found an increase in free fatty acid concentration as storage at 4°C increased in time, from 0.35 at baseline to 1.28 g/L after 96 hours of refrigeration (P < 0.001), and also reported a decrease in total protein concentration. These findings open the door for more investigation on stability for all types of choline and other nutrients, which is of high relevance for the maintenance of the nutrient levels of milk for the optimal development of infants consuming it. Data obtained would help to guide recommendations on proper storage for expressed human milk. In the present study, after 4 hours at room temperature, the balance of FC and Pchol in the human milk was altered with an increase in concentration of FC. It is still unknown how the balance of the   67 choline compounds would affect the infants, but it is known that infants fed human milk have higher plasma free choline than their peers fed with infant formula. However, the formula has a different balance of the choline-containing compounds, compared to human milk, as found by Ilcol et al., (2005). What we don’t know yet is whether breastfed infants have a higher serum free choline than infants fed with expressed human milk because of a shift in FC/Pchol balance, or the difference in the milk matrix. In addition to cold storage of the expressed milk, thawing and heating of milk may also be common practice. Before analysis, we thawed the human milk samples in ice baths, but did not consider other thermic practices, this needs to be addressed because it is unknown if choline in milk would be affected by heat. In summary, storage temperatures and duration are important aspects that could alter the balance of the WSC in milk and should be addressed. After seeing the variability of WSC in fresh milk, it is clear that under the conditions tested, stability was not the main determinant of the wide variability in the milk WSC, Figure 11.   68 Figure 11. Baseline concentration of milk free choline, phosphocholine, glycerophosphocholine and total water-soluble cholines from 6 lactating women (S1 to S6), as µmol/L. A, free choline; B, phosphocholine; C, glycerophosphocholine; D, total water-soluble cholines. In our study, mean WSC content in milk samples across woman (n=20), had a mean GPC that contributed 48% of total WSC, 40% Pchol and 12% FC. In 1996, Holmes-McNary et al., determined the mean concentration of each choline compound, finding Pchol being the higher choline-containing compound, followed by GPC with FC representing just a small proportion of the total choline in milk. The mean ± SE of FC from Holmes Mc-Nary et al., (1996) was 116 ± 22 µmol/L, similar to our results of a median concentration of choline equal to 131 (IQR=105) µmol/L; their milk concentration of Pchol 570 ± 136 µmol/L was similar to our findings of 513.5 ± 244 µmol/L. However, the amount of GPC found by them was almost half that which we found, with 362 ± 70 0 200 400 600 800 1000 1200 S-1 S-2 S-3 S-4 S-5 S-6 Phosphocholine (umol/L) 0 30 60 90 120 150 180 210 S-1 S-2 S-3 S-4 S-5 S-6 Free Choline (umol/L) 0 150 300 450 600 750 900 S-1 S-2 S-3 S-4 S-5 S-6 Glycerophosphocholine (umol/L) Participant 0 300 600 900 1200 1500 1800 S-1 S-2 S-3 S-4 S-5 S-6 Water-soluble cholines (umol/L) Participant A B C D   69 µmol/L, compared to a mean GPC of 617.1 ± 255.2 µmol/L. Holmes et al., (2000) found mean concentrations of Pchol and GPC were higher compared with the rest of the choline-containing compounds, but similar in concentration when compared between them, which is consistent to our results. However, their total WSC was different compared to Holmes Mc-Nary et al., (1996) results and ours, since their FC concentration results were mean ± SE of 210 ± 5, GPC of 410 ± 8 and PC of 480 ± 7 µmol/L. One possibility for the lower phosphorylated cholines and higher FC is the storage of the milk samples, since the study does not provide any information on how long milk samples were stored before analysis. An alternate possibility for these differences between studies is the difference in infant age and the duration of milk production. The milk samples in our study correspond to a mean of 4 months post-partum, whereas milk used by Holmes et al., (2000) was between 1 and 1.5 months post-partum. Although it is well known that after the first few days the maturity of human milk is reached, it has been also proposed that choline content in the human milk is related to the length of milk production. We could identify a correlation between the free choline content in milk and time post-natal (P = 0.047), as WSC gradually decrease in human milk from 2 to 5 months of age, perhaps reflecting the brain and body growth requirements. Interestingly, when infants are 6 months of age, mean milk WSC concentration and variability increases, by a still unknown cause, Figure 12.   70 Figure 12. Change in the concentration of the water-soluble cholines concentration in the milk, based on infant age (2-7 mo.), as µmol/L. Sample size is as follows: 2 mo., 4; 3 mo., 2; 4mo., 4; 5 mo., 3; 6 mo., 5; 7mo, 2. Boxplots show median, 25-75th and 2.5 to 97.5 percentile. Sakurai et al., also showed that human milk choline changes from 821 ± 160 µmol/L (n = 62) at 3 wks-3 mo., to a maximum peak from 3-6 mo of 935 ± 141 µmol/L (n = 36), and decreasing at 7-12 mo to 888 ± 141 µmol/L (n = 35), however, they did not specify which forms of choline were included in their analysis. Zeisel et al., (1986) did not identify a correlation between the infant’s age and the amount of FC in the milk once the milk reached maturity, but this remains to be clarified. Bitman et al., 1984 showed phospholipid concentrations, including PC and SM, decrease in the first 3 months of lactation. Their turnover may also affect the amount of the WSC across time if they breakdown into free fatty acids and choline compounds, but more research is needed to confirm it. In our study, not all women showed the same degree of variability in WSC within a day. The variability in the concentration of WSC in the 5 milk samples collected at different times on the Time of breast milk production (mo.) Milk water-soluble cholines (µmol/L)   71 same day was high in many woman, with change of up to 113%, whereas others showed only 19% change. No participant had less than 10% of variability in the WSC in milk. Similarly to our results showing no consistent changes in a day, Zeisel et al., (1986) could not find a consistent pattern of changes in choline compounds after collecting mature milk samples (n = 5, sample = 1 ml) at each feed during 72 hr. They showed graphically that some mothers were producing milk with FC concentrations similar across their feeds, while others had higher variability in FC. Thus, some mothers are more variable than others; however, they did not report how much the milk choline varied for each woman across 72hr. Notably, the choline forms that they analyzed were only FC and the lipid-bound forms, while they didn’t include the phosphorylated WSC, which represents the bigger proportion of the total cholines in human milk. Interestingly, in our samples the total amount of WSC is less variable across and within mothers, compared to the variability of the individual WSC compounds. Notably, milk Pchol is negatively correlated to FC concentration (P = 0.019). This brings the possibility to have a compensatory or balancing mechanism, ensuring milk with adequate amounts of choline. This idea is supported by choline metabolism being highly dynamic, with several pathways of how choline compounds get in the milk, including uptake from maternal circulation, to some extent the de novo synthesis in the mammary gland, and the breakdown of phospholipids from the milk fat globule membrane. However, it is still not clear what proportion of each choline compound is delivered to the milk through which pathway. We found that women consuming an adequate amount of dietary choline, as determined by the AI of 550 mg/d, had similarly high milk values of WSC concentration than mothers not meeting the choline AI. The AI is set when not enough scientific evidence is available to determine an   72 Estimated Average Requirement (EAR), with the AI level covering the needs of almost all the population, meaning most people could be exceeding their needs when consuming the AI. Conversely, mothers not reaching the AI, might still be meeting their personal needs for choline, and producing milk with adequate amounts of choline. However, more studies are needed to confirm this idea. Importantly, in our study none of the mothers who met the AI for choline had low milk WSC concentration, as seen in some participants not meeting the AI. To this point, it remains unclear if these results can be seen in a larger sample of women, where it would be also important to address where each compound is coming from and if inadequate consumption of choline can alter (i.e. increase) the degree of choline biosynthesis. In synthesis, it is important to address if one source is not providing enough choline, how the maternal body acts in order to provide the required milk WSC to the infant; further investigation is required. We explored the caloric content of the milk as a means to explain the variability, since it is well known that many nutrients are related to it; however, this was not the case with the individual or the sum of WSC. Contrary to the expected results, the variability did not decrease when determining the WSC content per kcal or per grams of fat in the milk, compared to choline per ml of milk. One explanation is that we did not include lipid-bound choline in our analysis; although lipid-bound cholines are found in minor concentration in human milk, adding their content to get the total milk choline may alter the association between milk caloric density and total choline. The analysis of the maternal dietary choline intake was analyzed as source of the milk choline variability. Our milk samples were taken at specific times related both to the meals of the mother and the feeding of the infants. Our first result showed no relation between the amount of milk WSC   73 and the samples following the maternal meals, taken around 2 hr. after the meals. Chiao et al., (1988) found FC concentrations were 15-fold higher in rat milk than in maternal rat serum obtained simultaneously. The rat milk choline showed high peak at 6 hr. after choline administration, when choline in milk was found 60-fold higher in concentration compared to maternal serum. It is possible that we did not measure the major peak of choline content in the milk after ingestion, as the times of sampling were closer to the meals compared to the study in rats. Zeisel et al., (1986) studies in rats showed that adequate levels of maternal dietary choline and choline supplementation increased milk FC and Pchol compared to diets deficient in choline. However we did not see any relation between different levels of choline intake and the increase of the milk WSC, which may be explained by our small sample size. When determining the total choline intake per day for each woman, we saw a trend association between dietary PC and the milk concentration of WSC. In turn, PC was the compound found in higher amounts in the maternal diet with mean intake ± SD of 293 ± 218 mg/d. In all populations, except during infancy, dietary choline intake is mainly in the form of PC. In a recent study in Alberta, Canada, Lewis et al., (2014) reported that the major choline compound in the diet of healthy lactating women (n = 448) was PC with an intake of 170 ± 114 mg/d, which are lower than our intake values, perhaps due to the differences on the typical maternal diets in Vancouver or the small number of 20 women in the present study. They found the main contributors to maternal dietary choline were dairy products and meat, while we found that on average, eggs were a higher contributor to choline in the maternal diet. Similar results have shown PC, as being a major part of the cell membranes foods, is also the main contributor to the choline dietary intake in populations consuming omnivorous diets, with eggs being the main average PC food source across countries and age groups. Chu et al., (2012), analyzed diets from healthy individuals in Taiwan aged 13 to 64   74 years and showed that eggs contributed to 25% of the total choline intake. West et al., (2014) found that healthy non-pregnant lacto-ovo vegetarian women of childbearing age (n = 15) were not meeting the AI for choline (425 mg/d), with a mean intake of 170 mg/d with no egg consumption, and a mean of 265 mg/d when eggs were consumed. Consistent with our findings, many other studies have reported inadequacy of choline consumption in women of childbearing age, and in pregnant and lactating women. Fischer et al., (2010) reported a mean choline intake of 364 mg/d with only 8.3% of the participants meeting or exceeding the AI and 6.25% with a choline intake below the AI. Interestingly, we also found that mothers who consumed 2 eggs/day had higher consumption of choline and were more likely to meet the AI than mothers consuming only 1 or none. Although regular egg consumption has not been recommended by health professionals in the past few decades due to their cholesterol content, studies have shown that the LDL cholesterol levels do not increase in healthy adults after egg consumption. Miller et al., (2014) found that after the consumption of 2 or more eggs, in some, but not all participants had an increase in plasma trimethylamine oxidase (TMAO). TMAO has been suggested to increase the risk of cardiovascular disease, and it is thought that the different responses are due to gut and intestinal bacteria capable to convert choline into TMAO. West et al., (2014) investigated egg consumption in lacto-ovo-vegetarian women in relation to the plasma free choline with a 3x8 week crossover study. After comparing no egg consumption or the intake of 6 omega-3 enriched eggs/week, or 6 regular large eggs/week, they found a statistically significant increase in plasma FC (P = 0.02) in the omega-3 enriched eggs compared to no egg consumption. They also found higher plasma phosphocholine (P = 0.05) when consuming any of the egg types, compared to no egg consumption and found TMAO was not affected by any of the manipulations. It is, therefore, possible that women with   75 increased choline requirements, such as during lactation, may benefit from regular egg consumption. Other studies have addressed the relationship between the maternal dietary choline intake and the milk choline content of both animals and humans. Zeisel et al., (1986), fed rats with different levels of choline and showed a clear relationship between the rat choline intake administered as PC and the amount of total choline in the milk. Fischer et al., (2010) supplemented healthy pregnant women (n = 103) with PC (5400 mg/d of PC, equivalent to 750 mg of net choline), or a placebo from the 18th week of pregnancy until day 90 of lactation, and reported higher concentration of milk FC and Pchol in the supplemented PC group than the placebo group (P<0.001). Similarly, we found that women not consuming the AI were more likely to have lower milk concentrations of FC. However, the differences in the number of woman in our study was low, and we did not use a supplementation as done by Fischer et al., (2010). Of importance, we found that even when milk WSC concentration was not significantly related to the mother meeting the choline AI, it was more probable mothers consuming the AI had concentrations of milk sufficient to provide the AI = 125 mg/d to infants. Future work on dietary choline is needed to better understand the relation between the dietary choline and its sources, and milk cholines, as well as the choline requirements during infancy. Although de novo synthesis of choline as PC from PE does occur in the mammary gland, it is likely low due to a normal decrease in the production of estrogen during lactation (Fischer et al., 2007). Women with a single-nucleotide polymorphism (SNP) in the methylenetetrahydrofolate reductase (MTHFR) gene, estimated to be present in around 20% or the humans (Kohlmeier et al., 2005), may have higher requirements for dietary choline as proposed by Fischer et al., (2010) based on   76 the results from mice. It would be important to investigate more in depth the role of the maternal hepatic stores of choline. In our study, the importance of this relationship is indicated by the trend of having lower WSC in human milk of mothers with their second infant, when compared with milk of women with their first infant (r = -0.432 P = 0.057). If choline stores are depleted during the first pregnancy and lactation, and the mother does not consume an adequate amount of choline prior to the second pregnancy or if the time between pregnancies is short, it is possible that the amount of choline provided to the second infant in milk may be inadequate, which could impact the infant and the mother’s health. However, this needs to be better investigated. Chiao et al., (1988) showed that after oral administration of free choline to rats, an accumulation of labeled choline took place in the mammary epithelial cells, in the form of free choline or other choline compounds, although which ones and what amounts was not specified. This leads to many other possibilities that should be considered for future studies to better understand where choline in the milk comes from, what is making the amounts of each compound and total choline variable across and within women or, since choline metabolism is highly dynamic, the pathway where each choline compound comes from is not important. The dietary recommendation for choline during lactation and infancy assume that all choline forms are equivalent and only the total amount of dietary choline is important. Our study has shown the WSC levels in milk across women can differ and contain a different balance of compounds. The reasons and implications are yet to be determined, but it is clear that not all cholines are metabolized equally, especially when comparing the water-soluble forms with the lipid-bound forms (DRI Choline, 1998). The importance of the water-soluble forms in the human milk is not well understood, but according to available research, infancy is the only stage in which WSC forms represents the majority of the dietary choline intake. This may be explained by their lower production of bile salts and pancreatic phospholipases   77 required to digest phospholipids such as PC and SM and potentially, the different choline compounds might have different fates in the infant body, which is yet to be determined. Our findings also stress the importance of adequate storage of milk after expression to maintain the balance of milk WSC, as well as the need for controlling the amounts, type of milk (colostrum, fore-milk, hind-milk), and the need of pooling different milk samples from each one of the participants in the same day, in contrast to the use of random samples. Ideally, future research should include a larger number of women consuming the AI, in order to address if variability is lower on those women, and to confirm findings from maternal diet in relation to the milk choline content. 4.2 Limitations • We don’t have intermediate storage times between 2 weeks and 6 months for the stability study. It is clear that at 6 months the water-soluble cholines are not stable anymore, but we don’t know at which point between those times, the water-soluble cholines changed to be unstable. • We tried to recreate normal conditions of what a mother would do to feed her infant with expressed human milk, including storage in a refrigerator. Heating treatment to warm the milk before giving it to the infant, which may also potentially influence the amount of choline on the milk, was not done. • We did not determine the concentration of the lipid-bound cholines. Thus, the amount of total choline provided is only an estimation based on a mean proportion of phosphatidylcholine and sphingomyelin in human milk. The total milk choline remains unknown for this set of women.   78 • Our sample size is too small for conclusions on the relationship between the maternal dietary choline intake and the amount of water-soluble cholines in human milk, and also may not reflect Vancouver or Canada’s population. As a result, the data from this thesis cannot be generalized. • The inter-individual variability for the 20 women was determined. However, the results are presented as mean and ranges for one day and may not represent the real degree of variation of the milk cholines. • Recall and food intake bias might have impacted since participants when completing their dietary diary. Instructions and examples were explained and provided in written format, however, it is possible that they could over- or under- estimate consumption, forget or prefer not to report certain foods. It is also possible that the obtained data from one single day might be different to the typical diet. • Although we had 5 samples per mother in the variability study, only one day was assessed which may or may not represent a “normal day” for the women. 4.3 Future directions 1. The cause of the breakdown of phosphocholine and increase in free choline, whether is an enzymatic, microorganism or other, needs to be determined. With this information, further actions can be taken on how to prevent changes in the balance of the cholines. 2. Research is needed to determine the cause of the increase of glycerophosphocholine after 6 months of storage at -80°C, analyzing also lipid-bound cholines and their possible breakdown as sources of glycerophosphocholine.   79 3. Stability of choline in regards heating should be addressed, as well as determining the times and temperatures where choline starts being unstable. 4. Future studies should include a larger sample size, with different study days. Information on dietary intake of choline and parallel samples of maternal blood and milk are also needed to better understand the role of maternal choline intake and sources of choline in the amount of milk choline. 5. It is well known that the mammary gland can produce PC de novo through PE methylation, and that there is an active uptake of choline from the maternal blood with the CDP-choline pathway being also present in the mammary gland. There is no information on how much and what types of choline in milk come from different sources. 6. As the choline compounds are different in structure, it is important to determine whether or not they are equivalent in milk. Whether or not only the amount of total choline is important or the balance of the different compounds have health and development implications for the healthy infant is not known. 7. Further studies should address the TMAO production secondary to egg consumption in woman during lactation. This is important for developing recommendations for intake of eggs during lactation. 8. A closer look into the fate of each choline compound after being consumed by the infant and lactating women is needed, with close look in the intake-plasma-hepatic stores, uptake by the mammary gland and de novo synthesis of choline in the mammary gland. This could be addressed the use of animal models.   80 REFERENCES  Academy of Breastfeeding Medicine Protocol Committee, Eglash A. ABM clinical protocol #8: human milk storage information for home use for full-term infants. Breastfeed Med. 2010; 5(3):127-130.  Albright CD, Friedrich CB, Brown EC, Mar MH, Zeisel SH. Maternal dietary choline availability alters mitosis, apoptosis and the localization of TOAD-64 protein in the developing fetal rat septum. Brain Res Dev Brain Res. 1999; 115(2):123-129. Bitman J, Wood DL, Mehta NR, Hamosh P, Hamosh M. Comparison of the phospholipid composition of breast milk from mothers of term and preterm infants during lactation. Am J Clin Nutr. 1984; 40(5):1103-1119. Blanton CA, Moshfegh AJ, Baer DJ, Kretsch MJ. The USDA Automated Multiple-Pass Method accurately estimates group total energy and nutrient intake. J Nutri 2006; 136(10):2594-2599. Blesso CN, Andersen CJ, Barona J, Volek JS, Fernandez ML. Whole egg consumption improves lipoprotein profiles and insulin sensitivity to a greater extent than yolk-free egg substitute in individuals with metabolic syndrome. Metabolism. 2013; 62(3):400-410. Blusztajn JK. Choline, a vital amine. Science. 1998; 281(5378):794-795.   81 Braun LD, Cornford EM, Oldendorf WH. Newborn rabbit blood-brain barrier is selectively permeable and differs substantially from the adult. J Neurochem. 1980; 34(1):147-152. Buchman A, Dubin M, Moukarzel A, et al. Choline deficiency: a cause of hepatic steatosis during parenteral nutrition that can be reversed with intravenous choline supplementation. Hepatology 1995; 22(5):13991403. Buchman AL, Sohel M, Moukarzel A, Bryant D, Schanler R, Awal M, Burns P, Dorman K, Belfort M, Jenden DJ, Killip D, Roch M. Plasma choline in normal newborns, infants, toddlers, and in very-low-birth-weight neonates requiring total parenteral nutrition. Nutrition. 2001; 17(1):18-21. Buchman AL, Ament ME, Sohel M, Dubin M, Jenden DJ, Roch M, Pownall H, Farley W, Awal M, Ahn C. Choline deficiency causes reversible hepatic abnormalities in patients receiving parenteral nutrition: proof of a human choline requirement: a placebo-controlled trial. JPEN J Parenter Enteral Nutr. 2001; 25(5):260-268. Chaio CK, Pomfret EA, Zeisel SH. Uptake of choline by rat mammary-gland epithelial cells. Biochem J. 1988; 254(1):33-38. Cheng RK, MacDonald CJ, Williams CL, Meck WH (2008) Prenatal choline supplementation alters the timing, emotion, and memory performance (TEMP) of adult male and female rats as indexed by differential reinforcement of low-rate schedule behavior. Learn Mem 2008; 15(3):153–162.   82 Chu DM, Wahlqvist ML, Chang HY, Yeh NH, Lee MS. Choline and betaine food sources and intakes in Taiwanese. Asia Pac J Clin Nutr. 2012; 21(4):547-557. Cole LK, Vance JE, Vance DE. Phosphatidylcholine biosynthesis and lipoprotein metabolism. Biochim Biophys Acta. 2012; 1821(5):754-761 Conway JM, Ingersen LA, Vinyard BT, Moshfegh AJ. Effectiveness of the USDA 5-Step Multiple Pass Method in Assessing Food Intake in Obese and Non-obese Women. Am J Clin Nutr. 2003; 77:1171-1178. Fisher MC, Zeisel SH, Mar MH, Sadler TW. Inhibitors of choline uptake and metabolism cause developmental abnormalities in neurulating mouse embryos. Teratology. 2001; 64(2): 114-122. Fischer LM, daCosta KA, Kwock L, Stewart PW, Lu TS, Stabler SP, Allen RH, Zeisel SH. Sex and menopausal status influence human dietary requirements for the nutrient choline. Am J Clin Nutr. 2007; 85(5): 1275-1285. Fischer LM, da Costa KA, Galanko J, Sha W, Stephenson B, Vick J, Zeisel SH. Choline intake and genetic polymorphisms influence choline metabolite concentrations in human breast milk and plasma. Am J Clin Nutr. 2010; 92(2):336-346.   83 Gallier S, Gragson D, Jiménez-Flores R, Everett D. Using confocal laser scanning microscopy to probe the milk fat globule membrane and associated proteins. Food Chem. 2010; 58(7):4250-4257 García-Lara NR, Escuder-Vieco D, García-Algar O, De la Cruz J, Lora D, Pallás-Alonso C. Effect of freezing time on macronutrients and energy content of breastmilk. Breastfeed Med. 2012; 7:295-301. Giuffrida F, Cruz-Hernandez C, Flück B, Tavazzi I, Thakkar SK, Destaillats F, Braun M. Quantification of phospholipids classes in human milk. Lipids. 2013; 48(10):1051-1058. Health Canada, Canadian Nutrient File, 2010. Retrieved from: www.healthcanada.gc.ca/cnf Heiman H, Schanler RJ. Enteral nutrition for premature infants: the role of human milk. Semin Fetal Neonatal Med. 2007; 12(1):26-34. Holmes-McNary MQ, Cheng WL, Mar MH, Fussell S, Zeisel SH. Choline and choline esters in human and rat milk and in infant formulas. Am J Clin Nutr. 1996; 64(4):572-576. Holmes-Mcnary MQ, Loy R, Mar M-H, Albright CD, Zeisel SH. Apoptosis is induced by choline deficiency in fetal brain and in PC12 cells. Brain Res Dev Brain Res. 1997; 101(1-2):9-16.   84 Holmes HC, Snodgrass GJ, Iles RA. Changes in the choline content of human breast milk in the first 3 weeks after birth. Eur J Pediatr. 2000; 159(3):198-204. Ilcol YO, Ozbek R, Hamurtekin E, Ulus IH. Choline status in newborns, infants, children, breast-feeding women, breast-fed infants and human breast milk. J Nutr Biochem. 2005; 16(8):489-499. Infante JP, Kinsella JE. Phospholipid synthesis in mammary tissue. Choline and ethanolamine kinases: kinetic evidence for two discrete active sites. Lipids. 1976;11(10):727-735. Institute of Medicine (US) Standing Committee on the Scientific Evaluation of Dietary Reference Intakes and its Panel on Folate, Other B Vitamins, and Choline. Dietary Reference Intakes for Thiamin, Riboflavin, Niacin, Vitamin B6, Folate, Vitamin B12, Pantothenic Acid, Biotin and Choline. Washington, DC: National Academy Press, 1998. Kohlmeier M, da Costa KA, Fischer LM, Zeisel SH. Genetic variation of folate-mediated one carbon transfer pathway predicts susceptibility to choline deficiency in humans. Proc Natl Acad Sci USA. 2005; 102(44):16025-16030. Konstantinova SV, Tell GS, Vollset SE, Ulvik A, Drevon CA, Ueland PM. Dietary patterns, food groups, and nutrients as predictors of plasma choline and betaine in middle-aged and elderly men and women. Am J Clin Nutr. 2008; 88(6):1663-1669.   85 Kular J, Tickner JC, Pavlos NJ, Viola HM, Abel T, Lim BS, Yang X, Chen H, Cook R, Hool LC, Zheng MH, Xu J. Choline Kinase Beta Mutant Mice Exhibit Reduced Phosphocholine, Elevated Osteoclast Activity and Low Bone Mass. J Biol Chem. 2015; 290(3):1729-1742. Lawrence RM, Pane CA. Human breast milk: current concepts of immunology and infectious diseases. Curr Probl Pediatr Adolesc Health Care. 2007; 37(1):7-36. Lev HM, Ovental A, Mandel D, Mimouni FB, Marom R, Lubetzky R. Major losses of fat, carbohydrates and energy content of preterm human milk frozen at -80°C. J Perinatol. 2014; 34(5):396-398. Lewis ED, Subhan FB, Bell RC, McCargar LJ, Curtis JM, Jacobs RL, Field CJ; APrON team. Estimation of choline intake from 24 h dietary intake recalls and contribution of egg and milk consumption to intake among pregnant and lactating women in Alberta. Br J Nutr. 2014; 112(1):112-121. Meck WH, Smith RA, Williams CL. Organizational changes in cholinergic activity and enhanced visuospatial memory as a function of choline administered prenatally or postnatally or both. Behav Neurosci. 1989; 103(6):1234-1241. Michel V, Yuan Z, Ramsubir S, Bakovic M. Choline transport for phospholipid synthesis. Exp Biol Med. 2006; 231(5):490-504.   86 Miller CA, Corbin KD, da Costa KA, Zhang S, Zhao X, Galanko JA, Blevins T, Bennett BJ, O'Connor A, Zeisel SH. Effect of egg ingestion on trimethylamine-N-oxide production in humans: a randomized, controlled, dose-response study. Am J Clin Nutr. 2014; 100(3):778-786. Newton ER. Breastmilk: the gold standard. Clin Obstet Gynecol. 2004; 47(3):632-642. Niculescu MD, Zeisel SH. Diet, methyl donors and DNA methylation: interactions between dietary folate, methionine and choline. J Nutr. 2002; 132(8):2333S-2335S. Oshida K, Shimizu T, Takase M, Tamura Y, Shimizu T, Yamashiro Y. Effects of dietary sphingomyelin on central nervous system myelination in developing rats. Pediatr Res. 2003; 53(4):589-593. Otten JJ, Hellwig JP, Meyers LD. Dietary Reference Intakes: The Essential Guide to Nutrient Requirements. Choline. Institute of Medicine. Washington, DC: The National Academies Press, 2006. Patterson KY, Bhagwat SA, Williams JR, Howe JC, Holden JM. USDA Database for the Choline Content of Common Foods Release Two. Nutrient Data Laboratory. Agricultural Research Service U.S. Department of Agriculture. 2008. Rillema JA. Hormone regulation of choline uptake and incorporation in mouse mammary gland explants. Exp Biol Med. 2004; 229(4):323-326.   87 Sakurai T, Furukawa M, Asoh M, Kanno T, Kojima T, Yonekubo A. Fat-soluble and water-soluble vitamin contents of breast milk from Japanese women. J Nutr Sci Vitaminol. 2005; 51(4):239-247. Shin OH, Mar MH, Albright CD, Citarella MT, da Costa KA, Zeisel SH.Methyl-group donors cannot prevent apoptotic death of rat hepatocytes induced by choline-deficiency. J Cell Biochem. 1997; 64(2):196-208. Siân E.C. Davies, David A. Woolf, Ronald A. Chalmers, Joan E.M. Rafter, Richard A. Iles. Proton nmr studies of betaine excretion in the human neonate: consequences for choline and methyl group supply. J Nut Bioch. 1992; 10(3):523–530 Slutzah M, Codipilly CN, Potak D, Clark RM, Schanler RJ. Refrigerator storage of expressed human milk in the neonatal intensive care unit. J Pediatr. 2010; 156(1):26-28. Sousa SG, Delgadillo I, Saraiva JA.Human Milk Composition and Preservation: Evaluation of High-Pressure Processing as a Non-Thermal Pasteurisation Technology. Crit Rev Food Sci Nutr. 2014: 14:0. Tayebati SK, Amenta F. Choline-containing phospholipids: relevance to brain functional pathways. Clin Chem Lab Med. 2013 Mar 1;51(3):513-21.   88 Walker A. Breast milk as the gold standard for protective nutrients.J Pediatr. 2010; 156(2):3-7. West AA, Shih Y, Wang W, Oda K, Jaceldo-Siegl K, Sabaté J, Haddad E, Rajaram S, Caudill MA, Burns-Whitmore B. Egg n-3 fatty acid composition modulates biomarkers of choline metabolism in free-living lacto-ovo-vegetarian women of reproductive age. J Acad Nutr Diet. 2014; 114(10):1594-1600. World Health Organization. Exclusive breastfeeding for six months best for babies everywhere. 2011. Retrieved from: http://www.who.int/mediacentre/news/statements/2011/breastfeeding_20110115/en/ Wu BT, Dyer RA, King DJ, Richardson KJ, Innis SM. Early second trimester maternal plasma choline and betaine are related to measures of early cognitive development in term infants. Plos One. 2012;7(8): e43448. Yang, E. K. Blusztajn, J. K., Pomfret, E. A., and Zeisel, S. H. (1988) Rat and human mammary tissue can synthesize choline moiety via the methylation of phosphatidylethanolamine. Biochem J. 1988; 256(3): 821-828 Zeisel SH. Dietary choline: biochemistry, physiology, and pharmacology. Annu Rev Nutr. 1981;1:95-121.   89 Zeisel SH, Char D, Sheard NF. Choline, phosphatidylcholine and sphingomyelin in human and bovine milk and infant formulas. J Nutr. 1986; 116(1):50-58. Zeisel SH, Da Costa KA, Franklin PD, Alexander EA, Lamont JT, Sheard NF, Beiser A. Choline, an essential nutrient for humans. FASEB J. 1991; 5(7):2093-2098. Zeisel SH, Blusztajn JK: Choline and human nutrition. Ann Rev Nutr 1994; 14:269–296. Zeisel SH, Mar M-H, Zhou Z-W, da Costa K-A. Pregnancy and lactation are associated with diminished concentrations of choline and its metabolites in rat liver. J. Nutr. 1995; 125:3049–3054. Zeisel SH, Albright CD, Shin OH, Mar MH, Salganik RI, da Costa KA. Choline deficiency selects for resistance to p53-independent apoptosis and causes tumorigenic transformation of rat hepatocytes. Carcinogenesis. 1997;18(4):731-738. Zeisel SH. Choline: needed for normal development of memory. J Am Coll Nutr. 2000;19(5): 528-531. Zeisel SH. Choline: an essential nutrient for humans. Nutrition. 2000; 16(7-8): 669-671. Zeisel SH. Nutritional importance of choline for brain development. J Am Coll Nutr. 2004; 23(6): 621S-626S.   90 Zeisel SH, Niculescu MD. Perinatal choline influences brain structure and function. Nutr Rev. 2006; 64(4):197-203. Zeisel SH. Choline: critical role during fetal development and dietary requirements in adults. Annu Rev Nutr. 2006;26: 229–250. Zeghari N, Younsi M, Meyer L, Donner M, Drouin P, Ziegler O. Adipocyte and erythrocyte plasma membrane phospholipid composition and hyperinsulinemia: a study in nondiabetic and diabetic obese women. Int J Obes Relat Metab Disord. 2000; 24(12):1600-1607.   91 APPENDIX 1. Instructions for milk sample collection. Instructions for Collecting Milk Samples In this study you will be asked to collect 5 mid-feed milk samples in a day (15-20 ml (~ 1-1 ½ Tablespoon). You will freeze and store the milk samples for later pick-up. € Wash your hands and gently clean your breast before feeding your baby or expressing milk. € Try not to touch the inside surfaces of the collection bottles. € Drink plenty of water during the day and plan to drink a beverage every time you breastfeed or pump. € Milk sample collection will be on the time of your baby feeding, close to the following times: o As soon as you wake up in the morning o Before your lunch o Around 2 hours after your lunch (early afternoon) o Around 2 hours after your dinner (evening) o Before going to bed at night € If you have extra milk to donate, then we can give you more collection bottles. € If you have any questions about collecting the samples then do not hesitate to contact the Nutrition & Metabolism Research Program. Expressing milk means obtaining milk from your breast without your baby feeding. Some women find it easy to express milk by hand or you may choose to use a pump to remove milk quickly from your breasts, which works well if you have a lot of milk. The milk which appears early in the feed is called ‘foremilk’. For this study we are interested in the milk appearing later in the feed, after the baby has nursed for a few (~ 3-4) minutes, called mid-feed milk. € The collection bottles have been labeled with your unique study number. € Use the permanent felt-pen “Sharpie” marker provided to clearly write the date and time of the sample on the bottle. You may find it easier to label the bottle before you collect the sample. € Collect the milk into a clean sanitized container and then pour into the container or use the funnel provided to express directly into the container. € Screw the cap onto the bottle and place the sample into your freezer. € When you first start sampling just put as much milk in the bottle as you can comfortably provide. After a day or so you will notice that your milk supply is increased and it will be easier to obtain the sample. € Contact us to arrange the collection of the frozen samples the day after you collected the samples, at a convenient time. You can call us or contact us via email. D e p a r t m e n t o f P e d i a t r i c s 950 West 28th Avenue, Room 171 Vancouver, BC, V5Z 4H4   92 APPENDIX 2. Dietary diary form Date____________________ Participant #_____________________ FOOD INTAKE: Dietary dairy Instructions: Please record every food or drink that you consume, with the time, place and amount of consumption. You can use grams/ml or tsp., tbsp., cups in the estimation of the amount of food/drink you consumed. If you prepared and ate a dish, please record all the ingredients you used in the preparation and what type of food you are consuming (example: whole wheat vs. white bread; skim, 1%, 2%, homo milk; margarine vs. butter, type of oil, etc.) Time Place Food Amount   93 APPENDIX 3. Milk caloric (Kcal/dL) and water-soluble choline content (µmol/L) in 20 healthy breastfeeding women across one day. Each letter denotes a participant. Sampling times are as follows: 1, early morning; 2, before lunch; 3, after lunch; 4, after dinner; 5, last feed of the day. Calories are expressed as kcal/dL of milk, water-soluble cholines expressed as µmol/L. FC, free choline; Pchol, phosphocholine; GPC, glycerophosphocholine; WSC, total water-soluble choline. Collection time Collection time Collection time 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 A B         C        Calories 115 58 78 78 68 45 68 73 91 73 51 59 70 56 58 FC 84 103 78 104 114 178 219 176 226 220 314 236 474 453 332 Pchol 234 330 263 342 433 283 272 258 212 332 441 466 491 534 506 GPC 675 998 1188 1268 1239 780 758 693 848 927 1363 1115 221 213 275 WSC 994 1432 1529 1714 1786 1242 1249 1127 1286 1480 2117 1818 1186 1200 1113         D                 E                 F        Calories 80 71 68 73 51 87 89 92 63 92 39 64 53 69 62 FC 149 130 128 272 152 122 154 161 164 224 115 137 169 163 112 Pchol 487 364 384 350 390 333 270 235 362 225 505 358 235 247 412 GPC 882 712 706 670 790 566 479 460 530 461 674 558 559 562 694 WSC 1518 1207 1218 1292 1331 1021 904 855 1056 911 1294 1053 963 973 1218             G                 H                 I        Calories 71 96 85 49 75 44 49 65 67 70 72 74 58 51 52 FC 127 148 135 162 172 114 107 127 120 142 78 89 118 87 82 Pchol 485 509 440 521 533 308 338 262 213 283 906 934 1120 1188 958 GPC 1086 1038 910 1176 1102 676 651 819 535 818 720 675 538 621 756 WSC 1698 1695 1485 1859 1808 1098 1096 1209 867 1242 1705 1700 1775 1896 1796             J                 K                 L        Calories 69 102 84 75 51 64 72 99 67 100 55 75 76 60 47 FC 91 102 109 118 108 72 80 116 79 87 208 179 282 271 176 Pchol 701 632 690 747 728 687 596 608 808 674 559 559 423 387 457 GPC 551 482 435 548 567 477 547 445 464 420 187 107 104 99 155 WSC 1343 1216 1235 1413 1403 1235 1222 1169 1352 1182 955 846 808 752 788             M                 N                 O        Calories 51 53 58 65 57 62 52 106 83 72 76 80 90 58 61 FC 56 60 79 85 50 112 119 99 130 116 69 76 81 68 69 Pchol 899 879 988 881 873 548 713 582 654 658 815 1113 913 919 942 GPC 5134 151 123 162 167 579 547 650 509 1955 379 489 900 513 432 WSC 1469 1091 1190 1128 1091 1239 1380 1330 1293 2729 1263 1678 1894 1500 1443             P                 Q                 R        Calories 56 62 61 51 67 78 72 61 85 42 78 84 81 74 55 FC 149 106 137 100 122 238 231 217 254 209 143 157 221 209 152 Pchol 703 541 602 592 686 493 335 441 393 288 373 308 402 369 345 GPC 429 289 354 279 388 576 202 259 269 154 604 772 603 763 703 WSC 1281 938 1093 972 1196 1308 768 917 916 652 1120 1237 1226 1341 1199             S                 T         Calories 51 78 88 60 53 60 57 63 57 69 FC 52 68 69 60 59 258 253 201 213 214 Pchol 481 530 662 537 550 138 82 114 79 89 GPC 1261 842 865 769 828 646 559 520 494 669 WSC 1794 1441 1597 1365 1437 1041 895 835 786 972