Sterols in Infant Formulas: A Bioaccessibility Study - American

Article Journal of Agricultural Sterols in infant and Food Chemistry is published by the American formulas: Chemical A Society. 1155 Sixteenth Street N.W., bioaccessibility Washington,study DC 20036

Subscriber access provided by READING UNIV Published by American Chemical Society. Copyright © American Chemical Society.

ISLAM J.A. HAMDAN, Luis Manuel Sanchez-Siles, Journal of Agricultural Guadalupe Garcia and Food Llatas, Chemistry is published by the American and María Jesús Lagarda Chemical Society. 1155

J. Agric. FoodSixteenth Chem.,Street JustN.W., Washington, DC 20036 Accepted Manuscript • DOI:


10.1021/acs.jafc.7b04635 • Publication Date (Web): 25 Jan 2018 Agricultural DownloadedJournal from of http:// and Food Chemistry is pubs.acs.org on January published by26, the 2018 American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Subscriber access provided by READING UNIV Published by American Chemical Society. Copyright © American Chemical Society.

Just Accepted

Journal of Agricultural “Just Accepted” manuscripts have been peer and Food Chemistry is online prior to technical formatting for publishedediting, by the American 1155 as a fre Society provides Chemical “Just Society. Accepted” Sixteenth Street N.W., dissemination of scientific Washington,material DC 20036 as soon a


appear in full in PDF format accompanied by a fully peer reviewed, but should not be conside Journal of Agricultural readers and citable by the Digital Object Ident and Food Chemistry is to authors. Therefore, theby “Just Accepted” W published the American Chemical Society. 1155is technical in the journal. After a manuscript Sixteenth Street N.W., Accepted” Web site and published Washington, DC 20036 as an ASA Subscriber access provided by READING UNIV Published by American Chemical Society. Copyright © American Chemical Society.

changes to the manuscript text and/or graph and ethical guidelines that apply to the jou of Agricultural or consequences Journal arising from the use of info

and Food Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Subscriber access provided by READING UNIV Published by American Chemical Society. Copyright © American Chemical Society.

In vitro gastrointestinal digestion (infant Page 1Journal of 34conditions) of Agricultural Salivaryand phase

3 diffetent IFs

Gastric phase: pH=4

IntestinalACS phaseParagon micelle formation

Food Chemistry LOD and Linearity

LOQ

Validation (GC-FID method for sterols determination)

Precision

Mixed

Accuracy

micelles:

bile fatty acids, phospholipids, cholesterol and other sterols.

Plus Environment salts,monoglyceride, BF

Journal of Agricultural and Food Chemistry

Sterols in infant formulas: A bioaccessibility study

Islam J.A. Hamdan1, Luis Manuel Sanchez-Siles2, Guadalupe Garcia-Llatas1*, María Jesús Lagarda1 1

Nutrition and Food Science Area, Faculty of Pharmacy, University of Valencia, Avda. Vicente Andrés Estellés s/n, 46100 – Burjassot (Valencia), Spain.

2

R&D Department, Institute of Infant Nutrition, Hero Group. Avda. Murcia 1, 30820 – Alcantarilla (Murcia), Spain. *To whom correspondence should be addressed (Telephone +34-963543766; Fax +34-963544909; E-mail: [email protected])

Author’s contribution: Islam J.A. Hamdan contributed with the analytical assays and to data acquisition and interpretation. Guadalupe Garcia-Llatas and María Jesús Lagarda contributed with the conception and design of the study. The last two authors and Luis Manuel SánchezSiles contributed with the interpretation of data. All authors have collaborated in the drafting and revision of the article, and have approved the final version of the manuscript for submission.

1 ACS Paragon Plus Environment

Page 2 of 34

Page 3 of 34


1

Abstract

2

The design of infant formulas (IFs) seeks to resemble human milk (HM) composition

3

and functionality. The fat sources used usually comprise vegetable oil blends to mimic

4

the fatty acid composition of HM, and introduce changes in the animal/plant sterol ratio.

5

In contrast, the use of milk fat globule membrane (MFGM)-rich ingredients could

6

improve this aspect by increasing the ratio. The present study evaluates the

7

bioaccessibility (BA) of sterols (cholesterol, desmosterol, brassicasterol, campesterol,

8

stigmasterol, and β-sitosterol) in three IFs (with or without MFGM) using an in vitro

9

digestion method simulating infant conditions. Analytical parameters confirmed the

10

suitability of the method for all these sterols. Results showed the presence of MFGM to

11

increase cholesterol content (6-7 vs 2 mg/100 mL), this being the most bioaccessible

12

sterol in the IFs. Although the BA of cholesterol was reduced in MFGM-enriched IF

13

(65.6-80.4% vs 99.7%), the intake of bioaccessible cholesterol from these IFs was

14

higher.

15

Keywords

16

Bioaccessibility, cholesterol, infant formulas, in vitro digestion, milk fat globule

17

membrane (MFGM), plant sterols.



18

INTRODUCTION

19

Human milk (HM) from healthy mothers is universally recognized as the ideal food

20

for newborn infants, since it perfectly suits their needs. With regard to macro- and

21

micronutrients, HM contains several bioactive compounds that are linked to the health

22

benefits associated with breastfeeding1. In some cases, when mothers are not able to

23

breastfeed or HM is not sufficient, infant formulas (IFs) are used as an alternative.

24

Animal and vegetable sterols are bioactive lipid compounds present in both HM

25

and IFs in different quantities and with variable profiles. Cholesterol is the main animal

26

sterol in HM. It is an essential component of cell membranes, and is the main

27

component of nerve cell membrane lipids2. Apart from its role as a structural

28

component, cholesterol is a precursor of bile acids, steroid hormones and vitamin D3,

29

and is necessary for the development of tissues and organs, particularly the brain2.

30

During the past decades, a beneficial relationship between breastfeeding and lower

31

serum cholesterol in later life and the possible prevention of cardiovascular diseases in

32

adulthood has been hypothesized by different authors4-6. Some research has shown that

33

higher cholesterol intake from HM in infancy contributes to cholesterol homeostasis and

34

lower serum cholesterol levels versus feeding with IFs in adulthood7-8. However, more

35

research is needed to explain the suggested programming effect of early exposure to

36

cholesterol9-10 and the functional benefits of the other sterols in development and

37

growth.

38

Over the last decades, advances in food science, technology and nutrition have

39

contributed to the improvement of IFs11. Infant formula manufacturers continuously

40

seek to develop IFs that are more similar to HM in terms of composition and functional

41

outcomes. Specifically, in relation to the lipid portion, most IFs are formulated with oil

42

blends in order to mimic the fatty acid composition of HM11. The sterol quantity and 3 ACS Paragon Plus Environment

Page 4 of 34

Page 5 of 34


43

profile of IFs is highly dependent upon the type and quantity of oils and fats used in

44

their formulation12. The main sources of plant sterols are vegetable oils (sunflower,

45

rapeseed, palm and coconut, among others), while the main sources of animal sterols

46

come from dairy ingredients where bovine or caprine milk fat is present. Since most

47

commercially available IFs are based on vegetable oils, the sterol profile between HM

48

and IF differs significantly: the cholesterol and other animal sterols (such as

49

desmosterol, and lathosterol) contents are higher in HM (12.0-16.6 mg/100 mL13 vs 0.4-

50

5.47 mg/100 mL in IFs12-14), while the opposite applies in relation to the plant sterol

51

contents (β-sitosterol and campesterol, among others)(0.02 mg/100 mL in HM15 vs

52

2.45-5.07 mg/100 mL in IFs12,14).

53

In this regard, milk fat globule membrane (MFGM)-rich ingredients have been used

54

in the formulation of IFs to mimic both HM composition12,16 and HM functional

55

outcomes17. Currently, some infant food manufacturers are commercializing MFGM-

56

enriched infant formulas in several European, Asian and American markets. In a double

57

blind randomized clinical trial in Sweden, formula-fed healthy infants supplemented

58

with a protein-rich MFGM fraction (Lacprodan MFGM-10) showed a higher cognitive

59

score on Bayley-III at 12 months18 and a lower incidence of acute otitis media19 than the

60

control group. Moreover, due to its composition (rich in phospholipids (~ 30%) and

61

cholesterol, and with lesser amounts of other sterols and also proteins and

62

glycoproteins17,20, MFGM acts as a natural emulsifier21.

63

From a functional perspective, it is currently seen to be of interest to have

64

information on the bioavailability of sterols from IFs, and not only on their contents, for

65

optimizing IFs. For the evaluation of bioavailability in vivo or in vitro methods can be

66

used. In vitro methodologies allow the screening of various ingredients or foods, and are

67

less time-consuming, laborious and expensive, although they are unable to totally



68

simulate in vivo human conditions22-24. In this regard, bioaccessibility (BA) is defined as

69

the fraction of a compound released from the food matrix within the gastrointestinal

70

tract, thus becoming available for intestinal absorption25. The BA or intestinal

71

absorption of highly lipophilic micro-constituents (such as cholesterol, phytosterols, fat-

72

soluble vitamins, triglycerides and carotenoids) can be affected by several factors, such

73

as their chemical structure26, the food matrix27,28 to which they are incorporated, the

74

presence of other dietary lipids (triglycerides27 and fatty acids26,27,29), interfacial

75

molecules27, protein composition27, the phospholipids26,27 and/or fiber contents26, the

76

size27 and surface area27 of the lipid droplets, and effectiveness of the enzymes

77

(cholesterol esterase29 and pancreatic lipase27,29). On the other hand, several studies

78

have assessed the solubility or BA of cholesterol and/or phytosterols from standard

79

solutions or food model systems30-37, confirming an important effect of the matrix,

80

chemical structure and concentration of sterols, and the digestion conditions.

81

To the best of our knowledge, only four studies38-41 have evaluated the BA of

82

sterols from food products, and none have assessed the BA of cholesterol or other

83

sterols from IFs. Due to the potential effect which sterols such as cholesterol may have

84

in terms of correct infant growth and development, and considering the lack of studies

85

in this field, the main objective of the present study was to evaluate the BA of sterols

86

(cholesterol, desmosterol, brassicasterol, campesterol, stigmasterol, and β-sitosterol)

87

present in three IFs, two of them including MFGM in their formulation, using an in

88

vitro digestion method simulating infant conditions. In addition, we validated gas

89

chromatography-flame ionization detection (GC-FID) determination of the sterol

90

contents present in the bioaccessible fraction (BF) obtained by this gastrointestinal

91

procedure. The study could help to better understand the fate of sterols from IFs, and

92

this might yield useful information for the design and development of new IFs. 5 ACS Paragon Plus Environment

Page 6 of 34

Page 7 of 34


93 94

MATERIALS AND METHODS

95

Samples

96

Three different powdered IFs marketed in Europe were used in this study (IFA, IFB

97

and IFC) (see Table 1). Two of them (IFB and IFC) contained whey protein enriched

98

with MFGM, bovine milk fat and a vegetable fat blend, while IFA only contained a

99

vegetable fat blend as the main source of fat.

100

Chemicals and standards

101

The enzymes and reagents used for in vitro digestion were: human salivary α-

102

amylase (EC 3.2.1.1), colipase, cholesterol esterase (EC 3.1.1.13), lipase (EC 3.1.1.3)

103

and phospholipase A2 (EC 3.1.1.4) from porcine pancreas, bile and serum albumin

104

(BSA) from bovine, mucin (type II) and pepsin (EC 3.4.23.1) from porcine stomach,

105

and porcine pancreatin from Sigma Chemical Co. (St. Louis, MO, USA). Calcium

106

chloride dihydrate, glucose, glucosamine hydrochloride, glucuronic acid, magnesium

107

chloride, potassium hydroxide, potassium thiocyanate, sodium dihydrogen phosphate,

108

sodium taurocholate and tris (hydroxymethyl) aminomethane were from Sigma

109

Chemical Co. (St. Louis, MO, USA). Ammonium chloride, anhydrous sodium sulfate,

110

hydrochloric acid (purity 37%), methanol, potassium chloride (KCl), potassium

111

dihydrogen phosphate, sodium bicarbonate, sodium chloride and urea were from Merck

112

(Whitehouse Station, NJ, USA). Chloroform, diethyl ether, n-hexane, and isopropanol

113

were from Scharlau (Barcelona, Spain). Ethanol and sodium hydroxide were from

114

Panreac (Barcelona, Spain), and uric acid was purchased from Prolabo (Sacramento,

115

CA, USA).

116 117

The

derivatization

reagent

was

a

mixture

of

N,O-

bis(trimethylsilyl)trifluoroacetamide (BSTFA) + 1% trimethylchlorosilane (TMCS)



118

from Fluka and anhydrous pyridine from Acros Organics (Geel, Belgium) (10:3 v/v)

119

prepared in an amber glass flask and kept at 4ºC to be used in one week.

120

Sterol standards: epicoprostanol as internal standard (IS) (5β-cholestan-3α-ol),

121

cholesterol (5-cholesten-3β-ol, 99%), stigmasterol (24S-ethylcholest-5,22-dien-3β-ol,

122

97%) and β-sitosterol (24R-ethylcholest-5-en-3β-ol, 97%) were from Sigma Chemical

123

Co. (St. Louis, MO, USA). Brassicasterol (24-methylcholest-5,22-dien-3β-ol, 98%) and

124

desmosterol (5, 24-cholesta-dien-3β-ol, 98%) were from Steraloids (Newport, RI,

125

USA). Campesterol (24α-methyl-5-cholesten-3β-ol, 99%) was purchased from Chengdu

126

Biopurify Phytochemicals Ltd. (Sichuan, China). Standard solutions of sterols were

127

prepared by dissolving the standards in n-hexane: 2-propanol (3:2, v/v) mixture except

128

the IS that was prepared in ethanol.

129

Millipore MilliQ deionized water (Millipore Ibérica S.A., Barcelona, Spain) was

130

used.

131

Simulated gastrointestinal digestion

132

The in vitro digestion applied to the IFs was similar to that employed by

133

Claumarchirant et al.42, which consists of three sequential phases (salivary, gastric and

134

intestinal with micelle formation). Two independent digestions for each IF were

135

performed.

136

Twenty grams of IF reconstituted according to instructions of the manufacturer (see

137

Table 1) were weighed in an Erlenmeyer flask, and 9 mL of saliva solution containing

138

organic and inorganic components at pH 6.5 ± 0.2 with 0.017 mg (1.8 U) of α-amylase

139

was then added. The Erlenmeyer flasks were placed in a shaking water bath (Stuart

140

SBS30) for 5 min at 37°C and 95 opm. Subsequently, gastric juice (13.5 mL, containing

141

organic and inorganic solutions) with BSA (54.7 mg), pepsin (6102 U) and mucin from

142

porcine stomach (pH 4.0 ± 0.07) were added. The pH of the mixture was adjusted to 4.0


Page 8 of 34

Page 9 of 34


143

± 0.07 in order to resemble infant conditions43 and incubated for 1 h (37°C, 95 opm in a

144

shaking water bath). After that, the duodenal juice was added (25 mL at pH 7.8 ± 0.2)

145

along with the bile solution (9 mL with 11.76 mM of bovine bile salt, at pH 8.0 ± 0.2),

146

and the mixture was then neutralized (pH 6.8-7.2). Subsequently, 5 U of cholesterol

147

esterase, 12.5 µg of colipase, 59.2 U of porcine pancreatic lipase, 501.2 U of

148

phospholipase A2 and 0.48 mM of sodium taurocholate were added. This mixture was

149

then incubated for 2 h (37°C, 95 opm). Finally, the digested mixture (70-80 g) was

150

centrifuged (Eppendorf 5810R) for 90 min at 3050 g and 4° to obtain the supernatant or

151

BF of the samples (aqueous-micellar fraction), which is the soluble fraction available

152

for absorption in the gastrointestinal tract. The BA of the studied sterols from IFs was

153

calculated as follows: (sterol content in BF (µg/g of undigested IF)/sterol content in IF

154

(µg/g of undigested IF)) x 100.

155

Digestions of blanks (by duplicate) composed of 20 g of water were performed with

156

the aim of eliminating any possible presence of interferences, since it has been reported

157

that crude bile extract used in the simulated digestion process can contain traces of

158

cholesterol and other sterols44.

159

Sterol determination

160

Infant formulas

161

Sterol determination in IFs was validated by our research group in a previous

162

work14. Briefly, fat was extracted from 2 mL of reconstituted IF (containing

163

approximately 0.25 g of IF), in triplicate, and saponified with KOH 2N in ethanol (90%)

164

(at 65ºC during 1 h) after IS addition (20 µg of epicoprostanol). Then, the

165

unsaponifiable fraction was extracted with diethyl ether and subjected to derivatization

166

with a mixture of BSTFA+1%TMCS: pyridine (10:3, v/v). The trimethylsilyl ether



167

(TMSE) derivatives obtained were analyzed by GC-FID (1 µL) under the

168

chromatographic conditions described elsewhere14.

169

The identification of sterols (cholesterol, desmosterol, brassicasterol, campesterol,

170

stigmasterol, β-sitosterol) was performed by comparing their relative retention times

171

from GC-FID with those of commercial sterol standards and by GC-MS analysis14. The

172

sterols were quantified by calibration curves performed with added standards of IS

173

derivatized the same way as the samples (see Table 2).

174

Bioaccessible fractions

175

The sterol content in the BFs was performed in quadruplicate (two aliquots from

176

each digestion). Ten grams of the BF (containing approximately 0.33 g of IF) was

177

directly saponified; the unsaponifiable fraction was extracted, derivatized and analyzed

178

by GC-FID using the same procedure as for the IFs.

179

Analytical parameters for sterol determination in BF

180

The validation of sterol determination in the BF was performed by evaluating the

181

following analytical parameters: linearity, limit of detection (LOD), limit of

182

quantification (LOQ), precision and accuracy, in order to check the suitability and

183

goodness of the method. A pool of the BF obtained from two digestions of IFC

184

performed on the same day was used for this purpose.

185

Linearity was evaluated in duplicate by the analysis of 5 concentration levels of

186

sterol standards added with 20 µg of epicoprostanol as IS for all studied sterols (see

187

Table 2). These calibration curves were employed for sterol quantification in IFs and

188

BFs, since the assayed concentrations included the corresponding contents in 2 mL of

189

reconstituted IFs or 10 g of BF. The LOD and LOQ for each sterol were calculated from

190

the digestion of four blanks (where BF was substituted by 2 mL of water), and were

191

calculated as follows according to the American Chemical Society guidelines45: LOD =


Page 10 of 34

Page 11 of 34


192

3SD/S and LOQ = 10SD/S (where SD is the standard deviation of the digestion blanks

193

and S is the slope of the calibration curve). The within-assay and between-assay

194

precisions were assessed by calculating the relative standard deviation (RSD), as a

195

percentage, of the sterol contents in four aliquots of BF from IFC analyzed on the same

196

day (within-assay) and 8 aliquots analyzed on two non-consecutively different days

197

(between-assay). Accuracy was evaluated by recovery assays by adding a known

198

amount of the studied sterol standards to 10 g of BF (or 5 g in the case of cholesterol

199

and desmosterol). Percentage recovery was calculated as: (sterol content in spiked BF

200

aliquots (µg) - sterol content in non-spiked BF aliquots (µg)) x 100/amount of spiked

201

sterol standard (µg).

202

Statistical analysis

203

Distribution normality of data and homogeneity of variance were tested by Shapiro-

204

Wilk and Levene test, respectively (p value ≥ 0.05). In order to assess statistically

205

significant differences in the same compound (individual or total sterol content) and in

206

the same kind of sample (IF, BF, or BA) one-way analysis of variance (ANOVA)

207

followed by Tukey’s post hoc test or Kruskal-Wallis (with Bonferroni correction) were

208

carried out for parametric and non-parametric data, respectively. The significance level

209

was p < 0.05 for all comparisons. All statistical analyses were performed using the

210

Statgraphics Centurion XV statistical package (Statpoint Technologies Inc., VA, USA).

211 212

RESULTS AND DISCUSSION

213

Sterol contents in IFs

214

Cholesterol, desmosterol, brassicasterol, campesterol, stigmasterol and β-sitosterol

215

were identified and quantified in the three studied IFs (see Figure 1). The sterol contents

216

(mg/100 mL of reconstituted IF) are shown in Table 4.



217

The amount of total sterols ranged from 9.2-15.2 mg/100 mL of reconstituted IF, in

218

the case of IFA being around two-thirds that of the other IFs. Regarding plant sterols,

219

the contents were significantly higher in IFB (8.6 mg/100 mL) than in the other IFs

220

(mean value 6.5 mg/100 mL). As expected, cholesterol was the most abundant sterol in

221

the IFs containing milk fat and whey protein enriched with MFGM (42% of total sterols

222

for IFB and 53% for IFC), being two-fold higher than in IFA (25%). However, although

223

the addition of MFGM increased the desmosterol content to double in IFC versus IFA,

224

it did not result in any significant increase in IFB - probably because of the different

225

proportions of algae, fungal and fish oils used as ingredients, which make a major

226

contribution to desmosterol content12.

227

In IFA the most abundant sterol was β-sitosterol (44%), with 38% and 29% in IFB

228

and IFC, respectively. The abundance of the rest of sterols, in general, was the same:

229

campesterol (12-23%) > stigmasterol (3-5%) ≈ brassicasterol (2-4%) ≈ desmosterol (1-

230

2%).

231

Our cholesterol (2.3-7.1 mg/100 mL of reconstituted IF), desmosterol (0.2-0.3

232

mg/100 mL) and total plant sterol contents (6.1-8.6 mg/100 mL) were comparable to

233

those reported by other authors (1.7-12.812-13, 46-47, 0.3-0.412 and 3.1-9.2 mg/100 mL12,13,

234

respectively) in cow’s milk based IFs.

235

As can be seen in Table 4, the IFs with dairy lipids and MFGM-rich ingredient

236

showed an animal/plant sterol ratio two to three times higher than the sample without

237

them, as has also been reported by Claumarchirant et al.12. Although the physiological

238

significance of this difference has not yet been established, it must be pointed out that

239

the more this ratio increases, the closer it comes to the situation found in HM. It

240

therefore could be taken into account when developing IFs.


Page 12 of 34

Page 13 of 34


241

Analytical parameters for sterol determination in BF Tables 2 and 3 show the analytical parameters obtained in the determination of

242 243

sterols in the BF.

244

Good correlation coefficients (all > 0.99) (Table 2) were obtained for all sterols at

245

the tested concentrations. Brassicasterol was the sterol which showed the greatest

246

sensitivity, while β-sitosterol showed the lowest sensitivity.

247

The LODs of plant and animal sterols were at concentrations below 5 and 11 µg/g

248

IF, respectively, and their LOQs were below 15 and 35 µg/g IF, respectively. The

249

highest LOD and LOQ values corresponded to cholesterol, and the lowest to β-sitosterol

250

and brassicasterol.

251

Regarding the precision values (Table 3), within-assay precision was lower than 6%

252

for all the sterols (values ranging from 1.4% for cholesterol to 5.5% for stigmasterol),

253

whereas between-assay precision was lower than 13% for all the sterols (values ranging

254

from 4.8-12.7%). Precision values obtained for stigmasterol were the highest, probably

255

due to low contents found in the studied IFs. At the concentrations of sterols present in

256

the BF, the Association of Official Agricultural Chemists (AOAC)48 considers precision

257

values of under 4-6% within-assay and 8-11% between-assay as being acceptable. Our

258

precisions were thus comprised within these ranges, except in the case of the between-

259

assay values obtained for cholesterol and stigmasterol, which were borderline.

260

According to the AOAC48, the acceptable recovery rate for the concentration of our

261

sterols should range between 80-115%. The greatest recovery corresponded to β-

262

sitosterol (106.2%) and the lowest to desmosterol (86.9%). We therefore can accept the

263

accuracy of our methods as satisfactory for the determination of sterols in BFs (Table

264

3).



265

Page 14 of 34

Bioaccessibility of sterols

266

The sterol contents in the BF of the IFs (mg sterol/100 mL of reconstituted IF) and

267

the corresponding BA values (%) are also shown in Table 4. Chromatograms of the BF

268

of the IFC and a blank are shown in Figure 2.

269

The highest total and animal sterol contents in the BF corresponded to IFC (9.7 and

270

5.9 mg sterol/100 mL of reconstituted IF, respectively), followed by IFB (8.4 and 4.3

271

mg sterol/100 mL, respectively) and IFA (7 and 2.4 mg sterol/100 mL, respectively).

272

The individual sterol contents in the BF of the three samples generally followed the

273

same order of abundance as in the undigested samples. In this sense, the cholesterol

274

contents in samples B and C were around two-fold higher than in the IF without

275

MFGM-rich ingredient (sample A). The abundance of the plant sterols in the three BFs

276

was as follows: β-sitosterol (24-41% of total sterols) > campesterol (11-19%) >

277

brassicasterol (2-4%) ≈ stigmasterol (≈2%). The BA of total sterols was similar in IFA and IFC (76% and 72%, respectively)

278 279

and

higher

than

in

IFB

(55%).

In

this

latter

280

galactooligosaccharides (soluble fiber) may have had an effect in the sterols BA. As it

281

has been previously described, dietary fibers have cholesterol adsorption capacity, may

282

reduce the BA of fatty acids and cholesterol through a depletion flocculation mechanism

283

antagonizing the emulsification by bile salts, and can decrease lipolytic activity49,50.

284

Regarding total animal or total plant sterols, the BA order was IFA > IFC > IFB, though

285

the ratio between animal sterols BA and plant sterols BA remained relatively constant.

286

In contrast, different trends in BA were observed on considering individual sterols: in

287

general, cholesterol showed higher BA than plant sterols, which is in concordance with

288

the lesser solubility and higher hydrophobicity exhibited by the latter (the animal/plant

289

sterol ratio increased in BF compared with the corresponding IF). Accordingly, in the


case,

the

presence

of

Page 15 of 34


290

mixed micellar phase, cholesterol would show the highest solubility. In this respect,

291

Alemany et al.38 and Alvarez-Sala et al.39 found cholesterol generally to be the most

292

bioaccessible sterol versus plant sterols in dairy products.

293

On the other hand, the different behavior of BA observed for cholesterol among the

294

IFs presumably could be explained by two factors related to differences in emulsion

295

interface composition and structure. Specifically: i) Competition among sterols: taking

296

into account that IFB has the highest total sterol content (with a greater proportion of

297

plant sterols versus animal sterols) (Table 4), this could produce competition between

298

cholesterol and plant sterols for incorporation to the mixed micelles - thereby reducing

299

cholesterol BA compared to the other IFs. Moreover, we consider that the amount of

300

plant sterols with saturated lateral chains (β-sitosterol and campesterol) was higher in

301

IFB (7.6 mg/100 mL of reconstituted IF) versus IFA and IFC (6.2 and 5.5 mg/100 mL,

302

respectively)(Table 4); the greater hydrophobicity exhibited by these sterols could

303

produce greater transference to the micelles, which in turn may displace cholesterol and

304

reduce BA, as evidenced in prior research39,51; ii) Type and quantity of emulsifiers: the

305

contents of phospholipids, proteins and other emulsifiers can affect the digestion rate of

306

lipid compounds52. The content and type of phospholipids of these formulas have been

307

published by our group16. The formula in which cholesterol exhibits least BA (IFB) was

308

that with the highest sphingomyelin content (13.73 mg/100 mL), while IFA (with

309

greater cholesterol BA) was the formula with least sphingomyelin (9.9 mg/100 mL) and

310

phospholipid contents (41.35 mg/100 mL vs 54.79 for IFB and 56.18 for IFC). This

311

could have an effect, as phospholipids and milk sphingomyelin have the ability to

312

decrease cholesterol intestinal absorption26,27, 53,54.

313

Although it has been described that the BA of sterols can be improved with the

314

increase in fat content in food composition/formulation by promoting the formation of



315

mixed micelles during gastrointestinal digestion39, this aspect cannot be discussed in our

316

study, since the IFs showed similar fat contents (27-31%, Table 1). In our study, we did

317

not control for the size of the fat globules; however, previous research has found that the

318

effect of fat globule size on digestibility is not clear55. Consequently, it does not seem to

319

be a key parameter in homogenized milks52.

320

With regard to other sterols, the order of BA was campesterol > β-sitosterol >

321

stigmasterol in IFB and IFC (Table 4). Matsuoka et al.34 observed the same order of

322

solubility for sterol standards in model systems, and furthermore, the presence of an

323

ethyl group in the side chain of the last two sterols can strongly diminish their solubility

324

- as evidenced in earlier research56,57. Similarly, a higher BA exhibited by campesterol

325

versus stigmasterol and β-sitosterol has also been reported by Alemany et al.38 in fruit

326

and/or skimmed milk-based beverages.

327

In conclusion, the present study offers a validated method for sterol determination

328

by GC-FID in BF obtained after in vitro digestion of IFs, affording good accuracy and

329

precision. Overall, the order of sterols BA was as follows: cholesterol > desmosterol =

330

brassicasterol = campesterol > β-sitosterol > stigmasterol. The important variability in

331

sterols BA - with special emphasis on cholesterol - evidenced in this work reveals an

332

important effect of the selection of fat-source/providing ingredients employed in the

333

manufacture of IFs. Although the BA of sterols may be influenced by several factors

334

(some of them not controlled in this study), the addition of MFGM increases the

335

cholesterol content and makes the formula more similar to HM from the sterols

336

composition perspective. The BA of animal sterols in formulas supplemented with

337

MFGM (IFB and IFC) was lower. However, given that the amount of cholesterol was

338

higher, there may be a higher intake of bioaccessible cholesterol from these formulas.


Page 16 of 34

Page 17 of 34


339

For all above mentioned reasons, this preliminary study reports for the first time the

340

BA of sterols from IF which is of importance to increase the knowledge on the role that

341

these sterols can exert in the infant’s development and growth. However, more future

342

research is needed on the BA of sterols from HM and IF in order to define a formulation

343

with a more similar behavior to that of HM. In this respect, the use of in vitro

344

gastrointestinal digestion simulating infant conditions and validated methodologies for

345

sterol determination – such as that proposed in this study can help academic circles and

346

the food industry to obtain better understanding of the behavior of bioactive

347

components, and thus comprehend the beneficial effects observed in clinical trials with

348

IF enriched with MFGM.

349

ABBREVIATIONS USED

350

ARA: arachidonic acid

351

AOAC: Association of Official Agricultural Chemists

352

BA: bioaccessibility

353

BF: bioaccessible fraction

354

BHT: butylated hydroxytoluene

355

BSA: bovine serum albumin

356

BSTFA: N,O-Bis (trimethylsilyl)trifluoroacetamide

357

DHA: docosahexaenoic acid

358

GC-FID: gas chromatography-flame ionization detector

359

GOS: galactooligosaccharide

360

HIV: human immunodeficiency virus

361

HM: human milk

362

IFA: infant formula A

363

IFB: infant formula B



364

IFC: infant formula C

365

IF: infant formula

366

IS: internal standard

367

KCl: potassium chloride

368

KOH: potassium hydroxide

369

LOD: limit of detection

370

LOQ: limit of quantification

371

MFGM: milk fat globule membrane

372

NaTDC: sodium taurodeoxycholate

373

Ps: plant sterols

374

TMCS: trimethylchlorosilane

375

TMSE: trimethylsilyl ether

376

ACKNOWLEDGMENT

377

Thanks are due to Prof. José D. Bermúdez from Department of Statistics and O.R. from

378

University of Valencia for his valuable advice in the statistical analysis.

379

FUNDING SOURCES

380 381

This work belongs to a CDTI (Centro para el Desarrollo Tecnológico Industrial) project granted to Hero Spain, S.A.


Page 18 of 34

Page 19 of 34


382

REFERENCES

383

1. Andreas, N.J.; Kampmann, B.; Le-Doare, K.M. Human breast milk: A review on its

384

composition and bioactivity. Early hum. dev. 2015, 91, 629-635.

385

2. Dietschy, J.M.; Turley, S.D. Thematic review series: brain Lipids. Cholesterol

386

metabolism in the central nervous system during early development and in the

387

mature animal. J. Lipid Res. 2004, 45, 1375-1397.

388 389

3. Cook, R. P. (Ed.). (2015). Cholesterol: chemistry, biochemistry, and pathology. New York, United States: Elsevier.

390

4. Fall, C.H.; Barker, D.J.; Osmond, C.; Winter, P.D.; Clark, P.M.; Hales, C.N. Relation

391

of infant feeding to adult serum cholesterol concentration and death from ischaemic

392

heart disease. Brit. Med. J. 1992, 304, 801-805.

393

5. Ravelli, A.C.; van der Meulen, J.H.; Osmond, C.; Barker, D.J.; Bleker, O.P. Infant

394

feeding and adult glucose tolerance, lipid profile, blood pressure, and obesity. Arch.

395

Dis. Child. 2000, 82, 248–252.

396

6. Singhal, A.; Cole, T.J.; Fewtrell, M.; Lucas, A. Breastmilk feeding and lipoprotein

397

profile in adolescents born preterm: follow-up of a prospective randomised study.

398

Lancet. 2004, 363, 1571–1578.

399

7. Owen, C.G.; Whincup, P.H.; Odoki, K.; Gilg, J.A.; Cook, D.G. Infant feeding and

400

blood cholesterol: a study in adolescents and a systematic review. Pediatrics. 2002,

401

110, 597-608.

402

8. Owen, C.G., Whincup, P.H., Kaye, S.J.; Martin, R.M.; Smith, G.D.; Cook, D.G.;

403

Bergstrom, E.; Black, S.; Wadsworth, M.E.J.; Fall, C.H.; Freudenheim, J.L.; Nie, J.;

404

Huxley, R.R., Kolacek, S., Leeson, C.P., Pearce, M.S., Raitakari, O.T., Lisinen, I.,

405

Viikari, J.S., Ravelli, A.C., Rudnicka, A.R., Strachan, D.P.; Williams, S.M. Does



406

initial breastfeeding lead to lower blood cholesterol in adult life? A quantitative

407

review of the evidence. Am. J. Clin. Nutr. 2008, 88, 305-314.

408

9. Woollett, L.; Heubi, J.E. Fetal and Neonatal Cholesterol Metabolism. [Updated 2016

409

Jun 22]. In: De Groot, L.J.; Chrousos, G.; Dungan, K. et al., editors. Endotext

410

[Internet]. South Dartmouth (MA): MDText.com, Inc.; 2000-. Available from:

411

https://www.ncbi.nlm.nih.gov/books/NBK395580/ (Access date: 21/09/2017).

412

10.

Koletzko, B. Human Milk Lipids. Ann. Nutr. Metab. 2016, 69, 27-40.

413

11. Zou, L., Pande, G., Akoh, C.C. Infant formula fat analogs and human milk fat: new

414

focus on infant developmental needs. Annu. Rev. Food Sci. T. 2016, 7, 139-165.

415

12. Claumarchirant, L.; Matencio, E.; Sanchez-Siles, L.M.; Alegria, A.; Lagarda, M.J.

416

Sterol composition in infant formulas and estimated intake. J. Agric. Food Chem.

417

2015, 63, 7245-7251.

418

13. Huisman, M.; van Beusekom, C.M.; Lanting, C.I.; Nijeboer,H.J.; Muskiet, F.A.J.;

419

Boersma, E.R. Triglycerides, fatty acids, sterols, mono- and disaccharides and

420

sugar alcohols in human milk and current types of infant formula milk. Eur. J. Clin.

421

Nutr. 1996, 50, 255-260.

422

14. Hamdan, J.A. I.; Claumarchirant, L.; Garcia-Llatas, G.; Alegría, A.; Lagarda, M.J.

423

Sterols in infant formulas: validation of a gas chromatographic method. Int. J. Food

424

Sci. Nutr. 2017, 68, 695-703.

425

15. Benoit, B.; Fauquant, C.; Daira, P.; Peretti, N.; Guichardant, M.; Michalski, MC.

426

Phospholipid species and minor sterols in French human milks. Food Chem. 2010,

427

120, 684-691.

428

16. Claumarchirant, L.; Cilla, A.; Matencio, E.; Sanchez-Siles, L.M.; Castro-Gomez,

429

P.; Fontecha, J.; Alegría, A.; Lagarda, M.J. Addition of milk fat globule membrane


Page 20 of 34

Page 21 of 34


430

as an ingredient of infant formulas for resembling the polar lipids of human milk.

431

Int. Dairy J. 2016, 61, 228-238.

432 433 434

17. Hernell,O.; Timby, N.; Domellöf, M.; Lönnerdal, B. Clinical benefits of milk fat globule membranes for infants and children. J. Pediatr. 2016, 173S, S60-5. 18. Timby,

N.;

Domellöf,

E.; Hernell,

O.; Lönnerdal,

B.; Domellöf,

M.

435

Neurodevelopment, nutrition, and growth until 12 mo of age in infants fed a low-

436

energy, low-protein formula supplemented with bovine milk fat globule

437

membranes: a randomized controlled trial. Am. J. Clin. Nutr. 2014, 99, 860-868.

438

19. Timby, N.; Hernell, O.; Vaarala, O.; Melin, M.; Lönnerdal, B.; Domellöf, M.

439

Infections in infants fed formula supplemented with bovine milk fat globule

440

membranes. J. Pediatr. Gastroenterol. Nutr. 2015, 60, 384-389.

441

20. Dewettinck, K.; Rombaut, R.; Thienpont, N.; Le, T.T.; Messens, K.; Van Camp J.

442

Nutritional and technological aspects of milk fat globule membrane material. Int.

443

Dairy J. 2008, 18, 436-457.

444 445 446 447

21. Singh, H.; Gallier, S. Nature's complex emulsion: The fat globules of milk. Food Hydrocolloid. 2017, 68, 81-89. 22. Hur, S.J.; Lim, B.O.; Decker, E.A.; McClements, D.J. In vitro human digestion models for food applications. Food Chem. 2011, 125, 1–12.

448

23. Minekus, M.; Alminger, M.; Alvito, P.; Ballance, S.; Bohn, T.; Bourlieu, C.;

449

Carrière, F.; Boutrou, R.; Corredig, M.; Dupont, D.; Dufour, C.; Egger, L.;

450

Golding, M.; Karakaya, S.; Kirkhus, B.; Le Feunteun, S.; Lesmes, U.; Macierzanka,

451

A.; Mackie, A.; Marze, S.; McClements, D.J.; Ménard, O.; Recio, I.; Santos, C.N.;

452

Singh, R.P.; Vegarud, G.E.; Wickham, M.S.J.; Weitschies,W.; Brodkorb, A. A

453

standardised static in vitro digestion method suitable for food – An international

454

consensus. Food Func. 2014, 5, 1113–1124.



455

24. Ménard, O.; Bourlieu, C.; De Oliveira, S.C.; Dellarosa, N.; Laghi, L.; Carrière, F.;

456

Capozzi, F.; Dupont, D.; Deglaire, A. A first step towards a consensus static in

457

vitro model for simulating full-term infant digestion. Food Chem. 2018, 240, 338-

458

345.

459

25. Alegria, A; Garcia-Llatas, G; Cilla, A. Static Digestion Models: General

460

Introduction. In: The impact of food bioactives on health; Verhoeckx, K., Cotter, P.,

461

López-Expósito, I., Kleiveland, C., Lea, T., Mackie, A., Requena, T., Swiatecka,

462

D., Wichers, H., Eds; Springer International Publishing, 2015.

463

26. Borel, P. Factors affecting intestinal absorption of highly lipophilic food

464

microconstituents (fat-soluble vitamins, carotenoids and phytosterols). Clin. Chem.

465

Lab. Med. 2003, 41, 979-994.

466

27. Gleize, B.; Nowicki, M.; Daval, C.; Koutnikova, H.; Borel, P. Form of phytosterols

467

and food matrix in which they are incorporated modulate their incorporation into

468

mixed micelles and impact cholesterol micellarization. Mol. Nutr. Food Res. 2016,

469

60, 749-759.

470

28. Hernández-Alvarez, E.; Blanco-Navarro, I.; Pérez-Sacristán, B.; Sánchez-Siles,

471

L.M.; Granado-Lorencio, F. In vitro digestion-assisted development of a β-

472

cryptoxanthin-rich functional beverage; in vivo validation using systemic response

473

and faecal content. Food chem. 2016, 208, 18-25.

474 475 476 477

29. Marze, S. Bioaccessibility of lipophilic micro-constituents from lipid emulsion. Food Funct. 2015, 6, 3218-3227. 30. Armstrong, M.J.; Carey, M.C. Thermodynamic and molecular determinants of sterol solubilities in bile salt micelles. J. Lipid Res. 1987, 28, 1144-1155.


Page 22 of 34

Page 23 of 34


478

31. Mel’nikov, S.M.; ten Hoorn, J.W.S.; Eijkelenboom, A.P. Effect of phytosterols and

479

phytostanols on the solubilization of cholesterol by dietary mixed micelles: an in

480

vitro study. Chem. Phys. Lipids. 2004, 127, 121-141.

481 482

32. Jesch, E.D.; Carr, T.P. Sitosterol reduces micellar cholesterol solubility in model bile. Nutr. Res. 2006, 26, 579-584.

483

33. Matsuoka, K.; Hirosawa, T.; Honda, C.; Endo, K.; Moroi, Y.; Shibata, O.

484

Thermodynamic study on competitive solubilization of cholesterol and β-sitosterol

485

in bile salt micelles. Chem. Phys. Lipids. 2007, 148, 51-60.

486

34. Matsuoka, K.; Kajimoto, E.; Horiuchi, M.; Honda, C.; Endo, K. Competitive

487

solubilization of cholesterol and six species of sterol/stanol in bile salt micelles.

488

Chem. Phys. Lipids. 2010, 163, 397-402.

489 490

35. Brown, A.W.; Hang, J.; Dussault, P.H.; Carr, T.P. Phytosterol ester constituents affect micellar cholesterol solubility in model bile. Lipids. 2010, 45, 855-862.

491

36. Nik, A.M.; Corredig, M.; Wright, A.J. Release of lipophilic molecules during in

492

vitro digestion of soy protein-stabilized emulsions. Mol. Nutr. Food Res. 2011,

493

55(S2).

494

37. Gleize, B.; Nowicki, M.; Daval, C.; Koutnikova, H.; Borel, P. Form of phytosterols

495

and food matrix in which they are incorporated modulate their incorporation into

496

mixed micelles and impact cholesterol micellarization. Mol. Nutr. Food Res. 2016,

497

60, 749-759.

498

38. Alemany, L.; Cilla, A.; Garcia-Llatas, G.; Rodriguez-Estrada, M.T.; Cardenia, V.;

499

Alegría, A. Effect of simulated gastrointestinal digestion on plant sterols and their

500

oxides in enriched beverages. Food Res. Int. 2013, 52, 1-7.

501

39. Alvarez-Sala, A.; Garcia-Llatas, G.; Cilla, A.; Barberá, R.; Sánchez-Siles, L.M.;

502

Lagarda, M.J. Impact of lipid components and emulsifiers on plant sterols



503

bioaccessibility from milk-based fruit beverages. J. Agric. Food Chem. 2016, 64,

504

5686-5691.

505 506

40. Vaghini, S.; Cilla, A.; Garcia-Llatas, G.; Lagarda, M.J. Bioaccessibility study of plant sterol-enriched fermented milks. Food Funct. 2016, 7, 110-117.

507

41. Bohn, T.; Tian, Q.; Chitchumroonchokchai, C.; Failla, M.L.; Schwartz, S.J.; Cotter,

508

R.; Waksman, J.A. Supplementation of test meals with fat-free phytosterol products

509

can reduce cholesterol micellarization during simulated digestion and cholesterol

510

accumulation by caco-2 cells. J. Agric. Food Chem. 2007, 55, 267-272.

511

42. Claumarchirant, L.; Sanchez-Siles, L.M.; Matencio, E.; Alegría, A.; Lagarda, M.J.

512

Evaluation of sialic acid in infant feeding: contents and bioavailability. J. Agric.

513

Food Chem. 2016, 64, 8333-8342.

514 515

43. Nguyen, T. T.; Bhandari, B.; Cichero, J.; Prakash, S. A comprehensive review on in vitro digestion of infant formula. Food Res. Int. 2015, 76, 373-386.

516

44. Alvarez-Sala, A.; Blanco-Morales, V.; Cilla, A.; Garcia-Llatas, G.; Sánchez-Siles,

517

L.M.; Barberá, R.; Lagarda, M.J. Safe intake of a plant sterol-enriched beverage

518

with milk fat globule membrane: Bioaccessibility of sterol oxides during storage. J.

519

Food Comp. Anal. 2017 https://doi.org/10.1016/j.jfca.2017.03.011.

520 521

45. MacDougall D, Crummett WB. Guidelines for data acquisition and data quality evaluation in environmental chemistry. Anal Chem. 1980, 52, 2242–2249.

522

46. Ramalho, H.M.M.; Casal, S.; Oliveira, M.B.P.P. Total cholesterol and desmosterol

523

contents in raw, UHT, IF powder and human milks determined by a new fast micro-

524

HPLC method. Food Anal. Met. 2011, 4, 424-430.

525

47. Kamelska, A.M.; Pietrzak-Fiecko, R.; Bryl, K. Determination of cholesterol

526

concentration in human milk samples using attenuated total reflectance Fourier

527

transform infrared spectroscopy. J. Appl. Spectrosc. 2013, 80, 148-152.


Page 24 of 34

Page 25 of 34


528

48. AOAC. Guidelines for single-laboratory validation of chemical methods for dietary

529

supplements and botanicals. Official methods of analysis of AOAC International

530

(19th ed.). Appendix K, Gaithersburg, MD, USA: AOAC International. 2012.

531

49. Minekus, M.; Jelier, M.; Jin-zhong, X.; Kondo, S.; Iwatsuki, K.; Kokubo, S.; Bos,

532

M.; Dunnewind, B.; Havenaar, R. Effect of partially hydrolyzed guar gum (PHGG)

533

on the bioaccessibility of fat and cholesterol. Biosci. Biotechnol. Biochem. 2005,

534

69,932-938.

535

50. Lopez-Marcos, M.C.; Bailina, C.; Viuda-Martos, M.; Perez-Alvarez, J.A.;

536

Fernandez-Lopez, J. Effects of various fibre-rich extracts on cholesterol binding

537

capacity during in vitro digestion of pork patties. Food Funct. 2015, 6, 3473-3478.

538

51. Moran-Valero, M.I.; Martin, D.; Torrelo, G.; Reglero, G.; Torres, C.F. Phytosterols

539

esterified with conjugated linoleic acid. In vitro intestinal digestion and interaction

540

on cholesterol bioaccessibility. J. Agric. Food Chem. 2012, 60, 11323-11330.

541

52. Garcia, C.; Antona, C.; Robert, B.; Lopez, C.; Armand, M. The size and interfacial

542

composition of milk fat globules are key factors controlling triglycerides

543

bioavailability in simulated human gastro-duodenal digestion. Food Hydrocolloid.

544

2014, 35, 494-504.

545

53. Noh, S.K.; Koo, S.I. Milk sphingomyelin is more effective than egg sphingomyelin

546

in inhibiting intestinal absorption of cholesterol and fat in rats. J. Nutr. 2004, 134,

547

2611-2616.

548

54. Norris, G.H.; Jiang, C.; Ryan, J.; Porter, C.M., Blesso, C.N. Milk sphingomyelin

549

improves lipid metabolism and alters gut microbiota in high fat diet-fed mice. J.

550

Nutr. Biochem. 2016, 30, 93-101.

551

55. Bourlieu, C.; Ménard, O.; De La Chevasnerie, A.; Sams, L.; Rousseau, F.; Madec,

552

M.N.; Robert, B.; Deglaire, A.; Pezennec, S.; Bouhallab, S.; Carrière, F.; Dupont,



553

D. The structure of infant formulas impacts their lipolysis, proteolysis and

554

desintegration during in vitro gastric digestion. Food Chem. 2015, 182, 224-235.

555

56. Armstrong, M.J.; Carey, M.C. Thermodynamic and molecular determinants of

556

sterol solubilities in bile salt micelles. J. Lipid Res. 1987, 28, 1144-1155.

557

57. Stevens, M.M.; Honerkamp-Smith, A.R.; Keller, S.L. Solubility limits of

558

cholesterol, lanosterol, ergosterol, stigmasterol and β-sitosterol in electroformed

559

lipid vesicles. Soft Matter, 2010, 6, 5885-5890.


Page 26 of 34

Page 27 of 34


560

FIGURE CAPTIONS

561

Figure 1. Chromatogram of IFC (~ 2 ml) obtained by GC-FID showing the studied

562

sterols. Epicoprostanol (IS), cholesterol, desmosterol, brassicasterol, campesterol,

563

stigmasterol and β-sitosterol.

564

Figure 2. Chromatograms of bioaccessible fraction of IFC (~10g) (grey line) and blank

565

of in vitro digestion (~10g) (black line) obtained by GC-FID showing the studied

566

sterols. Epicoprostanol (IS), cholesterol, desmosterol, brassicasterol, campesterol,

567

stigmasterol and β-sitosterol.



Page 28 of 34

Table 1. Energy, macronutrients composition (g/100g), main ingredients and reconstitution factors (RF) of studied infant formulas. Sample

Energy (Kcal)

Fat

Carbohydrates

Protein

Casein

Whey

Ingredientsa

%RF (w/v)

Demineralized whey protein, skimmed milk, vegetable oils (palm, rapessed, coconut, sunflower and Mortierella alpina oil (ARA)), A 519 27.7 57.8 9.6 30 70 lactose, minerals, fish oil, emulsifier (soy 12.9 lecithin), L-phenylalanine, vitamins, taurine, inositol, nucleotides, L-carnitine, milk proteins and bifidobacteria Lactose, vegetable oils (sunflower, rapeseed, palm and coconut), skimmed milk, GOS, milk fat, whey protein rich in MFGM, whey protein rich in α-lactalbumin, minerals, emulsifier (soy 13.0 B 509 27 56 10 38 62 lecithin), fish oils, fungal oil, vitamins, taurine, nucleotides, choline chloride, inositol and carnitine Lactose, vegetable oils (sunflower, rapeseed, palm and coconut), skimmed milk, milk fat, whey protein rich in MFGM, whey protein rich in αC 533 31 53 11 29 71 11.4 lactalbumin, milk protein, vitamins, fish oil, algae and fungi oil, emulsifier (soy lecithin), minerals, choline chloride, taurine, inositol and L-carnitine a ARA: arachidonic acid; GOS: galactooligosaccharide; MFGM: milk fat globule membrane; the fungal oil from Mortierella alpina is a source of ARA.

27

ACS Paragon Plus Environment

Page 29 of 34


Table 2. Linearity, and limits of detection and quantification for the determination of sterols in the bioaccessible fractions. Linearitya (n=2) Sterols

Range µg (in assay)

LODb (n=4) µg

r

Calibration curve

(in assay)

µg/g IF

LOQc (n=4) µg (in assay)

µg/g IF

Cholesterol

39.60-247.50

0.9994

y=0.0528x+0.0314

3.37d

10.36

11.23d

34.54

Desmosterol

0.89-7.12

0.9977

y=0.0575x+0.0045

0.67d

2.06

2.23d

6.85

Brassicasterol

1.97-19.72

0.9999

y=0.0691x-0.0021

0.33

1.01

1.10

3.37

Campesterol

24.91-84.69

0.9995

y=0.0538x-0.0710

1.44

4.44

4.82

14.81

Stigmasterol

0.24-19.40

0.9980

y=0.0616x+0.0130

1.13

3.47

3.76

11.57

β-Sitosterol

44.15-147.15

0.9993

y=0.0492x+0.1671

0.25

0.76

0.82

2.53

a

r=linear correlation coefficient; y=sterol area/ IS area; x=µg sterol; bLOD=limit of detection; cLOQ=limit of quantification; dµg of sterols in 5 g of BF.

28



Page 30 of 34

Table 3. Precision and accuracy for the determination of sterols in the bioaccessible fractions.

Precision (%RSDa) Sterols Cholesterol

Recovery assaysb (n=4)

Within

Betweenc

Original content

Spiked

Found

(n=4)

(n=8)

(µg in assay)

(µg in assay)

(µg in assay)

1.35

8.58

86.65 ± 2.12c

114.16 ± 0.16d

191.71 ± 2.08d

c

d

d

Recovery (%) 92.03 ± 1.82

Desmosterol

2.26

7.33

2.91 ± 0.06

4.58 ± 0.07

6.80 ± 0.18

86.89 ± 0.86

Brassicasterol

1.79

8.15

5.80 ± 0.19

10.61 ± 0.21

15.53 ± 0.25

91.70 ± 2.37

Campesterol

2.02

4.79

29.65 ± 0.59

52.12 ± 0.83

80.61 ± 1.56

97.77 ± 2.99

Stigmasterol

5.49

12.69

5.48 ± 0.51

11.32 ± 0.15

16.01 ± 0.45

93.08 ± 3.94

β-Sitosterol

2.85

6.42

58.67 ± 1.54

79.22 ± 1.52

144.40 ± 3.26

106.24 ± 1.42

a

RSD=relative standard deviation; bvalues are expressed as mean ± standard deviation; canalysis performed by quadruplicate in two different days; dµg of sterols in 5g of BF.

29


Page 31 of 34


Table 4. Sterol contents in infant formulas and bioaccessible fractions (mg/100 mL of reconstituted IF) and their bioaccessibilitya. A Sterols

Cholesterol

Desmosterol

Brassicasterol

Campesterol

IF

BF

BA

IF

BF

BA

IF

BF

BA

2.28±0.01 a,x

2.27±0.10 a,x

99.68±4.37 a,x

6.39±0.09 a,y

4.19±0.22 a,y

65.61±3.48 a,y

7.11±0.39 a,z

5.72±0.15 a,z

80.38±2.15 a,z

0.20±0.005 b,y

73.74±1.67 a,b,x

0.21±0.01 b,y

72.73±2.51 b,x,y

1.08±0.02 c,y

69.01±1.39 b,x

0.16±0.02 b,x

47.13±6.00 c,z

2.29±0.07 d,y

58.35±1.66 d,z

(24.73) 0.15±0.01 b,x

β-Sitosterol

(42.17) 0.10±0.01 b,x

68.43±3.65 b,x

0.39±0.01 c,x

66.19±2.54 b,x

0.30±0.01 b,y

1.35±0.06 c,x

65.02±2.66 b,x

1.90±0.05 c,y

0.11±1.0.02 b,x

65.28±11.65 a,x

0.25±0.02 b,x

81.25±8.01 b,y

0.29±0.02 b,y

1.23±0.09 c,x

64.89±4.98 a,x

1.57±0.07 c,z

(2.16)

(12.53)

0.16±0.02 b,x

64.66±6.43 b,x

(2.71)

(11.63)

0.68±0.10 b,y

0.16±0.04 b,x

23.54±5.53 c,y

(4.48) 2.85±0.21 d,x

69.81±5.25 b,x

5.72±0.48 d,y

0.34±0.06 b,x (2.53)

2.43±0.18 d,y

42.47±3.23 d,y

3.92±0.23 d,x

(37.74)

(44.25)

0.27±0.02 b,y (2.01)

(1.97)

(22.45)

4.08±0.03 f,x

0.17±0.01 b,x

0.26±0.01 b,x

(4.23) 2.07±0.02 d,x

(52.64)

(1.11)

(1.63)

0.25±0.01 e,x Stigmasterol

C

B

(29.03)

Σ Animal sterols

2.43±0.01 x

2.37±0.10 x

97.77±4.32 x

6.56±0.09 y

4.30±0.23 y

65.60±3.48 y

7.39±0.40 z

5.92±0.16 z

80.13±2.12 z

Σ Plant sterols

6.79±0.001 x

4.61±0.27 x

67.95±3.92 x

8.59±0.60 y

4.06±0.31 y

47.28±3.61 y

6.12±0.30 x

3.74±0.10 y

61.14±1.68 z

Σ Total sterols

9.22±0.01 x

6.99±0.34 x

75.81±3.71 x

15.15±0.55 y

8.37±0.29 y

55.21±1.92 y

13.51±0.40 z

9.66±0.22 z

71.52±1.64 x

0.4

0.5

1.4

0.8

1

1.4

1.2

1.6

1.3

Ratio animal/plant sterols a

Values are expressed as mean ± standard deviation of three replicates for IF analysis and four replicates for BF analysis. Relative percentage to total sterols content is indicated in parentheses. IF: infant formula. BF: bioaccessible fraction. BA: bioaccessibility, calculated as [sterol content in BF (µg/g of undigested IF)/sterol content in IF (µg/g of undigested IF)] x 100. Different superscript letters denote significant differences (p < 0.05) in the same column (a-f) and in the same line (x-z) for each sample (IF, BF, or BA). Σ Animal sterols: sum of cholesterol and desmosterol. Σ Plant sterols: sum of brassicasterol, campesterol, stigmasterol, and β-sitosterol.

30



Page 32 of 34

FIGURE GRAPHICS

mV

β-Sitosterol Stigmasterol

Campesterol

Desmosterol Brassicasterol

Epicoprostanol (IS)

Cholesterol

min

Figure 1.


Page 33 of 34


Cholesterol

β-Sitosterol Stigmasterol

Campesterol

Desmosterol Brassicasterol

Epicoprostanol (IS)

mV

min

Figure 2.



GRAPHIC FOR TABLE OF CONTENTS In vitro gastrointestinal digestion (infant conditions)

Salivary phase

Gastric phase: pH=4

3 IFs

Intestinal phasemicelle formation BF

Linearity

LOD and LOQ

Validation (GC-FID method for sterols determination)

Precision

Accuracy

Mixed micelles: bile salts,monoglyceride, fatty acids, phospholipids, cholesterol and other sterols.


Page 34 of 34

Sterols in Infant Formulas: A Bioaccessibility Study - American

Recommend Documents