Inulin increases Bifidobacterium animalis Bb-12 in vitro gastrointestinal resistance in margarine

Accepted Manuscript Inulin increases Bifidobacterium animalis Bb-12 in vitro gastrointestinal resistance in margarine Cínthia Hoch Batista de Souza, Luiz Antonio Gioielli, Susana Marta Isay Saad PII:

S0023-6438(17)30032-4

DOI:

10.1016/j.lwt.2017.01.032

Reference:

YFSTL 5984

To appear in:

LWT - Food Science and Technology

Received Date: 20 July 2016 Revised Date:

6 January 2017

Accepted Date: 12 January 2017

Please cite this article as: de Souza, C.H.B., Gioielli, L.A., Saad, S.M.I., Inulin increases Bifidobacterium animalis Bb-12 in vitro gastrointestinal resistance in margarine, LWT - Food Science and Technology (2017), doi: 10.1016/j.lwt.2017.01.032. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

1

ACCEPTED MANUSCRIPT

2 3

Inulin increases Bifidobacterium animalis Bb-12 in vitro gastrointestinal resistance in

4

margarine

5

RI PT

6 Cínthia Hoch Batista de Souza a,1, Luiz Antonio Gioielli a, Susana Marta Isay Saad a,*

7 8 9

SC

10 11 12

a

13

Universidade de São Paulo. Av. Prof. Lineu Prestes, 580, 05508-000 - São Paulo, SP, Brazil.

14

1

15

Rua Marselha, 591, 86041-140 - Londrina, PR, Brazil.

M AN U

Departamento de Tecnologia Bioquímico-Farmacêutica, Faculdade de Ciências Farmacêuticas,

Present address: Mestrado em Ciência e Tecnologia de Leite e Derivados, Universidade Norte do Paraná.

16 17

*

18

Tel: +55-11-30912378; Fax: +55-11-38156386

19

E-mail address: [email protected] (S.M.I. Saad)

22 23 24 25 26

TE D

EP

21

AC C

20

Corresponding author:

Shortened running title: In vitro GI resistance of Bifidobacterium Bb-12 in margarine.

27 28

1

29

Abstract

30

This study aimed to investigate the effect of inulin, whey protein concentrate (WPC), and/or

31

caseinomacropeptide (CMP), in different proportions up to 3%, on the viability and resistance to simulated

32

gastric and enteric conditions of Bifidobacterium animalis Bb-12 added in probiotic and synbiotic margarine,

33

during 35 days of storage at 5°C. Supplementation was important, since the control margarine presented very

34

low Bb-12 populations. Inulin at 3% resulted in higher Bb-12 counts (8 log cfu g-1). CMP contributed for

35

higher Bb-12 counts, compared to WPC. Inulin increased Bb-12 in vitro survival significantly after 6h

36

(P<0.05). In margarines with WPC at 3%, Bb-12 populations decreased drastically during the in vitro assays

37

for all storage periods. For the other formulations, Bb-12 populations decreased 2 log cfu g-1 after 2h. Taken

38

together, results on viability of B. animalis Bb-12 and on in vitro GI resistance of the strain incorporated in

39

margarine produced with inulin were unique, but the addition of inulin was necessary.

SC

RI PT

ACCEPTED MANUSCRIPT

40

Keywords: Synbiotic; Probiotic; Prebiotic; Spread; Gastrointestinal resistance.

M AN U

41 42 43

1. Introduction

Probiotics are “live microorganisms that, when administered in adequate amounts, confer a health

45

benefit on the host” (Hill, Guarner, Reid, Gibson, Merenstein, & Pot, 2014). One of the well-investigated

46

probiotics is Bifidobacterium animalis subsp. lactis Bb-12, with demonstrated efficacy through clinical trials

47

(Haschke et al., 1998). In order to exert their functional properties, probiotics need to be delivered to human

48

intestine in an active and viable form (Vandenplas, Huys, & Daube, 2015). Therefore, it is important to

49

develop probiotic food with suitable probiotic populations throughout shelf life and that are ingested as part

50

of a normal diet for maintaining a regular probiotic intake. Nevertheless, ingested microorganisms are

51

exposed to several stress factors that influence their viability through the human gastrointestinal tract (GIT).

53 54 55

EP

AC C

52

TE D

44

Among these factors, probiotics must tolerate the stomach acidic environment and also the reduced water activity and the presence of bile in the upper small intestine (Vandenplas, Huys, & Daube, 2015). Once probiotic bacteria must preserve viability and arrive at high populations in the colon after

surviving the food processing steps and the digestive process, a careful selection of the food matrix is an

56

important factor that should be considered in developing a probiotic product. The food matrix is considered

57

as one of the major factors in regulating colonization of microorganisms in the GIT. Food helps to buffer the

58

bacteria through the stomach and may contain other functional ingredients that could interact with probiotics

59

to alter their functionality. Fat content, concentration and type of proteins, sugars, and pH of foods are some

60

factors that could affect probiotic growth and survival in food products (Ranadheera, Baines, & Adams,

2

61

2010). These features might help probiotic survival during passage through the GIT, where survival is

62

dependent on both the strain and the food matrix involved. Likewise, the supplementation of food with some

63

ingredients like the prebiotic fibre inulin and milk proteins may stimulate probiotic bifidobacteria growth and

64

also protect them during GIT passage (Janer, Peláez, & Requena, 2004; Kos, Šušković, Goreta, & Matošić,

65

2000).

ACCEPTED MANUSCRIPT

Limitations of dairy products such as the presence of allergens and the requirement for cold storage

67

facilities, as well as an increasing demand for new foods and tastes have initiated a trend in non-dairy

68

probiotic product development (Martins, Ramos, Vanzela, Stringheta, Pinto, & Martins, 2013; Rößle, Auty,

69

Brunton, Gormley, & Butler, 2010; Tripathi & Giri, 2014). Among these new food matrices, margarine has

70

excellent perspectives as a matrix for probiotic incorporation. Margarine has several advantages over

71

yoghurt-type products in terms of delivery of viable probiotics, including its higher pH. Moreover, desirable

72

characteristics for probiotic foods as higher fat content and more solid consistency (Ong, Henriksson, &

73

Shah, 2006) are observed in a margarine food matrix and may offer protection to probiotics through the GIT

74

transit. Also, spreads, like margarine, are a particularly interesting vehicles for functional ingredients,

75

because they are eaten daily. Therefore, the aim of this study was to investigate the effect of inulin, WPC,

76

and/or CMP, in different proportions up to 3%, on the viability and resistance to simulated gastric and enteric

77

conditions of a commercial probiotic strain of Bifidobacterium animalis Bb-12 added in probiotic and

78

synbiotic margarine.

SC

M AN U

TE D

79

RI PT

66

2. Materials and methods

81

2.1. Experimental design and margarine manufacture

EP

80

Seven pilot scale margarine-making trials, named as M1-M7, were produced according to Table 1,

83

using a simplex-centroid design, changing the inulin, whey protein concentrate (WPC), and

84 85 86 87

AC C

82

caseinomacropeptide (CMP) proportions in the margarines (three repetitions of each trial were produced on different days). For each formulation, 3 kg of margarine were obtained. Combinations of the ingredients inulin (Beneo ST-Gel, Orafti, Oreye, Belgium), whey protein concentrate (Lacprodan 80, Arla Foods Ingredients, Sønderhøj, Denmark), and caseinomacropeptide (Lacprodan CGMP 10, Arla Foods Ingredients)

88

were used. A control trial, without supplementation with inulin, WPC and/or CMP (M8) was also prepared.

89

All trials were produced using the freeze-dried probiotic culture of Bifidobacterium animalis subsp. lactis

90

Bb-12 (Christian Hansen, Hørsholm, Denmark).

3

91

Margarine production was carried out in five steps: 1) fat phase production, 2) water phase

92

production, 3) emulsification, 4) crystallization process, and 5) packaging and storage. Different ingredients

93

were employed for preparing the fat and water margarine phases (Table 2).

ACCEPTED MANUSCRIPT

The following commercial ingredients were employed for the preparation of the fat phase: palm oil

95

(Agropalma, Tailândia, Brazil), canola oil (Liza, Mairinque, Brazil), emulsifiers monoacylglycerol (Myvatex

96

Smooth 5Z10729 and Myverol 18-92K) and polyglycerol esters of ricinoleic acid (Admul Wol 1408K)

97

(Kerry Bio-Science, Campinas, Brazil), colouring agent β-carotene (DSM, São Paulo, Brazil), butter flavour

98

(Danisco, Cotia, Brazil), and vitamin A (Fortitech, Campinas, Brazil). For the water phase, the following

99

ingredients were employed: water, NaCl (Cisne, Cabo Frio, Brazil), food-grade lactic acid 85% solution

100

(Purac Sínteses, Rio de Janeiro, Brazil), skimmed powdered milk (Nestlé, Araçatuba, Brazil). Additionally,

101

according to experimental design (Table 1), the following ingredients were also used in the water phase:

102

inulin (Beneo ST-Gel), whey protein concentrate (Lacprodan 80), and caseinomacropeptide (Lacprodan

103

CGMP 10). Water phase ingredients were manually mixed, after which the probiotic inoculum was added.

M AN U

SC

RI PT

94

104

The probiotic inoculum of Bb-12 was grown in 40 ml of reconstituted skimmed powdered milk for 2

105

h at 37°C, prior to its addition to the water phase, in order to obtain Bb-12 populations of a minimum

106

between 8 and 9 log cfu g-1 after margarine production (day 1). For the preparation of the fat phase, palm oil

107

was heated at 45°C and manually mixed with the additional fat phase ingredients mentioned above. During the emulsification step (step 3), the water phase was gently added to the fat phase under

109

agitation using a mixer (Stand Mixer 300, Brastemp, São Paulo, Brazil) until the emulsion formation took

110

place. For the crystallization step (step 4), the emulsion obtained on step 3 was transferred to an associated

111

equipment consisting of a 10 l jacketed emulsion tank, fitted with a refrigerated scraped-surface (Skymsen,

112

Brusque, Brazil), similarly to that described by Goli, Sahri, and Keramat (2009). The emulsion (35°C) was

113

submitted to a crystallization process until reaching the temperature of 17°C. Subsequently, the margarine

115 116 117 118

EP

AC C

114

TE D

108

was transferred to a manual dispenser (Delgo, Cotia, Brazil) and packaged in individual polypropylene plastic pots for food products (Tries Aditivos Plásticos, São Paulo, Brazil) in portions of 60 g of product. The margarines were then stored at 5±1°C for up to 35 days.

2.2. Determination of B. animalis Bb-12 viability during refrigerated storage

119

Margarines from each batch were used for microbiological analysis. Counts of B. animalis Bb-12

120

were monitored during margarine production (day 0), and after 1, 7, 14, 21, 28, and 35 days of refrigerated

121

storage at 5±1°C, for all trials. For this purpose, 25 g portions of duplicate margarine samples were collected

122

aseptically, blended in a Bag Mixer 400 (Interscience, St. Nom, France) with 225 ml of 0.1% peptone water

4

123

preheated to 40°C, according to Charteris, Kelly, Morelli, and Collins (2002), during 3.5 min, to obtain a

124

suitable dispersion, and submitted to serial dilutions with the same diluent.

ACCEPTED MANUSCRIPT

B. animalis Bb-12 was counted by pour-plating 1 ml of each dilution in DeMan-Rogosa-Sharpe agar

126

(MRS agar, Oxoid, Basingstoke, UK) to which sodium propionate (0.3 g 100 ml-1) and lithium chloride (0.2

127

g 100 ml-1) were added (Lithium Propionate MRS, LP-MRS). Plates were incubated during 3 days under

128

anaerobic condition (Anaerobic System Anaerogen, Oxoid) at 37°C (Vinderola & Reinheimer, 2000). Counts

129

of Bb-12 populations were carried out in quadruplicates.

RI PT

125

130 131

2.3. Survival of B. animalis Bb-12 under in vitro simulated gastrointestinal conditions

The in vitro evaluation of probiotic B. animalis Bb-12 survival in refrigerated margarines submitted

133

to gastric and enteric simulated conditions was carried out according to the methods described by Buriti,

134

Castro, and Saad (2010), with modifications, as described below. The in vitro assays were carried out after 7,

135

14, 21, and 28 days of refrigerated storage, since some trials did not present sufficient counts for a probiotic

136

food product on day 35 (end of storage period). In order to enumerate Bb-12 before and after the in vitro

137

assays, samples were collected initially (time 0) and after 2h (first step – after the gastric phase), 4h (after the

138

first enteric phase), and 6h after the beginning of the assay (after the second enteric phase).

M AN U

SC

132

For the in vitro procedure, in each sample collection, 10 ml from each triplicate dilution of

140

margarine in 0.5% NaCl solution (blended in a Bag Mixer 400) was transferred to 3 sterile flasks, totalizing 9

141

flasks containing the samples, and the pH of flasks was adjusted to 1.6-2.0 with 0.2 ml of 1 N HCl. Pepsin

142

(from porcine stomach mucosa, Sigma-Aldrich, St. Louis, USA) and lipase (Amano lipase from Penicillium

143

camemberti, Aldrich Chemical Company Inc., Milwaukee, USA) solutions were added to samples to reach a

144

concentration of 3 g l-1 and of 1.3 mg l-1, respectively. Flasks were incubated at 37°C, with agitation of

145

approximately 150 r.p.m in a metabolic water bath (Dubnoff MA-095, Marconi, Piracicaba, Brazil), during

147 148 149

EP

AC C

146

TE D

139

2h (first step - gastric phase). In the next step (first enteric phase), the pH of samples was increased to 4.05.2, using an alkaline solution pH 12 (1N NaOH, PO4H2Na.2H2O and distilled water). Bile (bovine bile, Sigma-Aldrich) and pancreatin (from porcine pancreas, Sigma-Aldrich) were added to reach a concentration of 5 g l-1 and of 1.6 g l-1, respectively. Samples were incubated again at 37°C for 2h under agitation. In the

150

last step, the pH was increased to 6.77-7.21, using the same alkaline solution. Bile and pancreatin

151

concentrations were adjusted (7.95 g l-1 and 0.79 g l-1, respectively), and samples were incubated again at

152

37°C for 2h under agitation (second enteric phase), achieving 6h of the in vitro assay.

5

153

Enumeration of B. animalis Bb-12 was carried out as described above in aliquots collected from

154

triplicate samples after 2h, 4h, and 6h (three different flasks of the same trial for each time) of the in vitro

155

assays. All results were presented as log cfu g-1 of margarine.

ACCEPTED MANUSCRIPT

156 157

2.4. Statistical analysis Statistical analysis of results obtained was carried out using STATISTICA v.8.0 software (Statsoft,

159

Tulsa, OK, USA). Analysis of variance (ANOVA) was used to determine significant differences (P<0.05)

160

for every parameter analysed among different margarine trials and different days of storage of each

161

margarine, using repeated measures. Differences between means were detected using post hoc Tukey test.

162

Data were checked for the homogeneity of variances before ANOVA evaluation, using the Brown-Forsythe

163

test. When homogeneity of variances was not verified, equivalent non-parametric tests were applied: Kruskal

164

Wallis test followed by post hoc Mann Whitney U (different margarines in the same period of storage) or

165

Friedman test followed by post hoc LSD rank (different storage period for a same margarine) (Bower,

166

1998a,b).

M AN U

SC

RI PT

158

167

The experimental results obtained both for B. animalis Bb-12 viability in margarines and for the in

168

vitro resistance tests after 21 days of refrigerated storage were applied to obtain the linear, quadratic, and

169

special cubic regression models, as a function of the proportions of each ingredient (x1: inulin, x2: WPC, and

170

x 3:

171

(

172

coefficients estimated by the least square method, and xi = dependent variables (x1: inulin, x2: WPC, and x3:

173

CMP) with

174

was evaluated by ANOVA and F test. Also, the lack of fit and adjusted coefficient of determination (R2) were

176 177 178

in

margarine

TE D

present

yˆ ( i ) = b1 x1 + b2 x 2 + b3 x 3 + b2 b3 x 2 x 3 ),

EP

n

formulations. where

yˆ ( i )

A

quadratic

model

= estimated response, bi =

∑ i = 1 x =1, 1> x > 0 was used to describe data obtained. The quality of the models obtained i

i

AC C

175

CMP)

determined and used to evaluate the model.

3. Results and Discussion 3.1. Probiotic viability during refrigerated storage of margarines

179

Table 3 shows the viability of B. animalis Bb-12 added to margarines M1-M8 during the

180

refrigerated storage at 5±1°C for up to 35 days. Brazilian present legislation states that the minimum viable

181

quantity of probiotic culture should be between 8.00 and 9.00 log cfu per daily serving portion and 10 g is the

182

portion recommended for margarine, butter, and spreads (Anvisa, 2003, 2008). Therefore, if a portion of

6

183

around 10 g of margarine is consumed daily, each gram of the product would necessarily contain, at least,

184

7.00 log cfu of probiotic bacteria.

ACCEPTED MANUSCRIPT

The addition of inulin, WPC, and CMP was necessary for the survival of Bb-12 in margarine, since

186

M8 (control) presented mean counts of 5.12±0.10 and 3.10±0.15 log cfu g-1, respectively, after the 1st and the

187

7th day of storage. After 7 days of storage, Bb-12 was not detected in M8. Except for M2, M3, and M7 after

188

14, 35, and 28 days of storage, respectively, and M8, all other margarines presented populations of Bb-12

189

above the minimum required for a probiotic food during the whole storage period.

RI PT

185

Among the different margarines studied, the formulation M1, containing higher amounts of inulin

191

(3%), presented populations of Bb-12 above 8.00 log cfu g-1 throughout the whole storage period. On days 28

192

and 35, populations in M1 were significantly higher (P<0.05). Margarines M3, M5, and M6 revealed

193

populations of Bb-12 above 8.00 log cfu g-1 from day 0 up to day 21 of storage. After this period, significant

194

reductions were observed (P<0.05). In the same way, M4 and M7 Bb-12 populations were above 8.00 log

195

cfu g-1 until day 14. In contrast, M2 was not considered a probiotic margarine after 14 days of storage, since

196

it presented Bb-12 populations below 7.00 log cfu g-1. Regarding the margarines supplemented with milk

197

proteins (WPC and CMP), Bb-12 populations in M3 (3% of CMP) were significantly higher than in M2 (3%

198

of WPC) (P<0.05). On day 35, they were as low as 4.64 log cfu g-1 (M2) and 6.87 log cfu g-1 (M3).

M AN U

SC

190

The lipid food matrix is in fact a great technological challenge for probiotic incorporation, since the

200

water-in-oil emulsion which makes up the margarine structure is a relatively inhospitable environment for

201

their growth. The size of the water phase droplets, the absence of nutrients inside the droplets, the addition of

202

some ingredients like salt, and the lack of ability of microorganisms to move between droplets also reduce

203

the ability of margarine to support microbial growth (Charteris, Kelly, Morelli, & Collins, 2002). Preliminary

204

tests carried out by our research group revealed that Lactobacillus acidophilus La-5 was not able to survive

205

in margarine matrix (data not shown). Therefore, margarine features reinforce the importance of the aqueous

207 208 209

EP

AC C

206

TE D

199

phase supplementation in order to obtain suitable populations of a specific probiotic microorganism. The presence of ingredients that could be fermented by beneficial microorganisms added during margarine manufacture may ensure their growth and/or survival during shelf life. The results obtained for margarine M1 probably are related to the fact that B. animalis Bb-12 is able

210

to metabolize prebiotic ingredients. The inulin employed in the presented study has a low polymerization

211

degree (DP ~ 10) (Beneo®

212

M1 containing 3% inulin presented lower pH values, varying from 4.89 (day 1) to 4.50 (day 35) when

213

compared to M2 (5.42-5.10), M3 (5.28-5.12), and M6 (5.36-5.10). Inulin has a stimulant effect on the

214

bifidobacteria growth, acting as a bifidogenic factor. Fermentation of inulin by bifidobacteria results in the

HP), therefore easily metabolized by Bifidobacterium spp. In fact, margarine

7

215

production of organic acids (Bosscher, Loo, & Franck, 2006). These compounds, especially lactic acid, might

216

also be formed in synbiotic food, and probably contributed for the decrease in margarine pH values.

ACCEPTED MANUSCRIPT

217

Some studies revealed Bifidobacterium animalis subsp. lactis populations of 107 cfu g-1 , when dairy

218

foods were manufactured with inulin (Bruno, Lankaputhra, & Shah, 2002; Cardarelli, Buriti, Castro, & Saad,

219

2008), whereas, in the present study, M1 presented even higher Bb-12 populations - of 108 cfu g-1. Several factors, including the presence of oxygen and acid, are responsible for the loss in viability of

221

probiotic bacteria in foods (Tripathi & Giri, 2014). However, in margarine M1, the presence of 3% of inulin

222

probably contributed for Bb-12 increased survival, particularly conferring protection against the presence of

223

organic acids and salt. Regarding supplementation of margarine with WPC and/or CMP, probably CMP was

224

more easily metabolized by Bb-12. Caseinomacropeptide (CMP) contains not only available nitrogen for

225

bacterial growth but also amino-sugars, such as sialic acid and N-acetylgalactosamine, which could be

226

fermented by bifidobacteria (Janer, Peláez, & Requena, 2004).

SC

RI PT

220

A noticeable difference in the composition of WPC that may influence upon growth of probiotic

228

microorganisms is the content of mainly β-lactoglobulin and α-lactoalbumin, which were selectively

229

removed during the isolation of the CMP fraction (Martín-Diana, Fraga, & Fontecha, 2014). WPC also

230

includes whey proteins with excessively high molecular weight to be rendered available for direct bacterial

231

uptake, but could be enzymatically cleaved, leading to formation of bifidobacterial growth factors (Janer,

232

Peláez, & Requena, 2004). In the present study, according to the manufacturer (Arla Foods Ingredients,

233

Sønderhøj, Denmark), CMP added in margarines M3, M5, and M7 had higher glutamine, isoleucine, proline,

234

serine, and threonine levels than those of WPC. The presence of these amino acids contributed for the high

235

Bb-12 populations observed in M3 (above 8.00 log cfu g-1 up to the 21st day of storage). Probably, the

236

probiotic culture here employed was not capable of using the proteins present in M2, since several

237

bifidobacteria strains were reported as being only weakly proteolytic (Bergamini, Hynes, Palma, Sabbag, &

239 240 241

TE D

EP

AC C

238

M AN U

227

Zalazar, 2009).

Only a few studies were reported on probiotic lipid-based emulsions. Charteris, Kelly, Morelli, and

Collins (2002) studied a table biospread containing Lactobacillus casei and Bifidobacterium infantis and reported populations of 4.5×108 cfu ml-1 (L. casei) and 1.0×106 cfu ml-1 (B. infantis), when different

242

ingredients like L-cysteine and sodium alginate were added to the water phase. Similarly, Dommels,

243

Kemperman, Zebregs, Draaisma, and Jol (2009) evaluated the viability of Lactobacillus reuteri DSM 17938

244

and Lactobacillus rhamnosus GG added to a low fat spread with 28% of lipids. Populations of, respectively,

245

2.85×108 and of 1.65×109 cfu g-1 were observed after 6 weeks of refrigerated storage.

8

246

Considering the results obtained for probiotic viability in margarine formulations on day 21 (all

247

margarines presented suitable Bb-12 populations, homogeneity of variances, and significant differences

248

between formulations), a quadratic model was generated for this variable response, in order to verify the

249

influence of inulin, WPC, and CMP on Bb-12 viability. The quadratic model and its coefficients and quality

250

analysis are presented in Table 4. The model obtained was statistically significant (P < 0.01) and able to

251

explain 86% of results (variation). Also, lack of fit was not statistically significant (P=0.065). The model

252

obtained revealed that B. animalis Bb-12 viability on day 21 was dependent on the individual presence inulin,

253

WPC, and CMP. The simultaneous supplementation with WPC and CMP was also important. The contour

254

plot generated from the quadratic model for viability revealed that inulin was more important than WPC and

255

CMP to maintain Bb-12 viability in margarine (Fig. 1). Higher populations of Bb-12 were observed with

256

inulin concentrations between 3% and 0.75% (1.00 and 0.25 in the diagram). The same behaviour was

257

observed with CMP between 2.25 and 3%.

259

M AN U

258

SC

RI PT

ACCEPTED MANUSCRIPT

3.2. Probiotic survival in margarines under in vitro simulated human gastrointestinal conditions The survival of B. animalis Bb-12 in margarines M1-M7 submitted to simulated in vitro human

261

gastrointestinal conditions after 7, 14, 21, and 28 days of storage at 5±1°C is presented in Fig. 2. The control

262

trial (M8) was not submitted to in vitro evaluation, since Bb-12 populations observed after 7 days of storage

263

were below 6.00 log cfu g-1).

TE D

260

During the whole in vitro assays - simulated gastric (2 h) and enteric (4 h and 6 h) conditions,

265

considering the storage period evaluated (28 days), reductions in Bb-12 populations varied from 0.95 - 2.98

266

log cfu g-1; 1.0 - 3.26 log cfu g-1, and 0.71 - 3.47 log cfu g-1 after 2, 4, and 6h, respectively, for most of the

267

margarines. M2 presented higher reductions, varying from 4.3 to 5.5 log cfu g-1 during the same period of the

268

in vitro assays.

270 271 272

AC C

269

EP

264

Significant decreases in Bb-12 populations were observed after 2 h of the in vitro assays for all

margarines tested during storage for up to 28 days (P<0.05). On day 7, after 2h of in vitro assays, M2 presented a 5.00 log cfu g-1 decrease (P<0.05). In contrast, others margarines presented a reduction of 2.00 log cfu g-1 after the gastric phase. M1, M3, and M7 presented the lowest reductions, probably due to the

273

presence of inulin and CMP. At the end of the in vitro assays (6h) significant reductions in Bb-12

274

populations were observed only for M5 after 7 days, M3 after 21 days, and M4 and M6 after 28 days. M1

275

and M6 on day 7, M1, M2 and M7 on day 14, M2 on day 21, and M5 on day 28 presented significant

276

increase in probiotic populations (P<0.05). Probably these results reflect a recovery of Bb-12 when it was

277

submitted to enteric phase 1 and 2 (pH 5 and 7, respectively). The hydrochloric acid present in the gastric

9

278

phase resulted in an injury of Bb-12 cells, and their recovery through the microbiological analysis was

279

difficult. After 7 days, all margarines studied, except M5, presented an increase or maintenance in the Bb-12

280

populations until the end of the in vitro assays (6h). From day 7 to 14, a decrease in probiotic populations

281

was observed during the whole in vitro assay, but with counts always above 6.00 log cfu g-1, except for M2

282

(1.20 cfu g-1 and of 1.98 log cfu g-1 after 2 and 6 hours).

ACCEPTED MANUSCRIPT

After 21 days of storage, most margarines presented probiotic populations below 6.00 log cfu g-1,

284

varying from 1.14 (M2) to 5.85 log cfu g-1 (M7) after 6h of the in vitro assay. M1 and M2 differed

285

significantly from the other formulations (P<0.05), since populations of, respectively, 7.15 cfu g-1 and 1.14

286

cfu g-1 were observed after 6h of the in vitro assays. At 28 days of storage, after 6h of in vitro assays, Bb-12

287

populations in M3 were significantly higher than in M4 (P<0.05). In margarine supplemented with inulin +

288

WPC (M4), a significantly increase in Bb-12 populations was observed between 4 and 6h of the in vitro

289

assay on day 7 (P<0.05).

SC

RI PT

283

During the whole in vitro assays period, margarines supplemented with inulin presented lower

291

reductions in Bb-12 populations when compared to initial counts. As mentioned earlier, inulin protected

292

probiotic microorganism during the in vitro assays, since when mixed with water inulin forms a particle gel

293

network. This gel probably avoided the action of HCl and enzymes over Bb-12 during the in vitro assays.

294

Inulin assured probiotic viability in suitable concentrations for margarines M4, M5, M7, and especially for

295

M1 during storage. On day 28, by the end of in vitro assays (6h), inulin assured populations of 6.32 log cfu g-

296

1

in M1.

TE D

M AN U

290

High reductions in probiotic viability after simulated gastric phase are expected results, since to

298

reach the intestine, strains must first pass through the stomach, which provides a powerful barrier against the

299

entrance in the gut (Morelli, 2007). Similarly to what was observed in the present study, Hansen, Allan-

300

Wojtas, Jin, and Paulson (2002) reported a 1.3 log cfu g-1 decrease in B. animalis Bb-12 populations after

302 303 304

AC C

301

EP

297

exposure to pH 2.0 during 2 hours. Charteris, Kelly, Morelli, and Collins (1998) reported that milk proteins protected probiotic

Lactobacillus casei and Bifidobacterium lactis during simulated gastric transit. This was not observed here when whey protein concentrate was employed, since margarines supplemented with WPC solely did not

305

present suitable Bb-12 populations after 2, 4, and 6h of the in vitro assays during all the sampling periods.

306

Probiotic counts in M2 during shelf life were the lowest (P<0.05) (Table 4). This fact contributed towards

307

the unsatisfactory results observed after the in vitro assays of margarine M2. Different results were reported

308

by Kos, Šušković, Goreta, & Matošić (2000), regarding the addition of WPC. The authors observed that

309

WPC was an important protector for Lactobacillus acidophilus M92 exposed to simulated gastric and small

10

310

intestinal juices, whereas, in the present study, the addition of WPC did not protect Bb-12 and M2 presented

311

a progressive decrease in Bb-12 populations during the in vitro assays along storage. After 28 days, Bb-12

312

counts observed in all sampling periods of the in vitro assay of M2 were below the detection limits of the

313

method (<1.14 log cfu g-1).

ACCEPTED MANUSCRIPT

Guo, Wang, Yan, Chen, Liu, and Zhang (2009) observed that the incubation of B. animalis Bb-12 in

315

simulated gastric juice containing pepsin at pH 2.5 resulted in significant decreases in its survival rate. In the

316

present study, reductions observed in probiotic populations were lower, due to the presence of the food

317

matrix and its constituents, mainly its fat content, a factor that could affect probiotic growth and survival in

318

food during the digestive process (Ranadheera, Baines, & Adams, 2010).

RI PT

314

As observed in the present study, Buriti, Castro, and Saad (2010) reported that the supplementation of

320

synbiotic guava musses with inulin and WPC simultaneously did not influence L. acidophilus La-5

321

protection positively during in vitro resistance assays. However, in refrigerated synbiotic guava musses the

322

authors observed counts below 2.00 log cfu g-1 when products supplement with different percentages of

323

inulin (1.33%, 2% or 4%) where submitted to in vitro resistance assays after 28 days. Similarly, Bedani,

324

Rossi, and Saad (2013) observed that inulin did not protect La-5 in fermented soy product during in vitro

325

resistance assays, whereas B. animalis Bb-12 presented mean populations above 7.00 log cfu g-1 up to the

326

end of in vitro assays.

M AN U

SC

319

The presence of inulin was important to protect Bb-12, since margarine with 3% of inulin (M1)

328

presented Bb-12 populations of 6.32 log cfu g-1 after 6h of in vitro assays. The survival rate of Bb-12 in M1

329

after in vitro assays was 91.33% and 78.51% after 7 and 28 days of storage, respectively. Similar results

330

regarding protection given by the prebiotic ingredient inulin, when combined with fructooligosacharide, to

331

Bb-12 during GI in vitro assays were reported by Padilha, Villarreal-Morales, Vieira, Costa, and Saad (2016)

332

in petit-suisse cheese. The authors observed mean Bb-12 survival rates of 79% in the synbiotic cheese,

334 335 336

EP

AC C

333

TE D

327

whereas this rate was reduced to 60% in the probiotic product. In order to verify the influence of inulin, WPC, and CMP on Bb-12 resistance, results obtained after

6h of the in vitro assays on day 21 were used, since the parameters mentioned previously regarding Bb-12 viability were observed. A quadratic model was generated for this variable response and it was statistically

337

significant (P < 0.01) and able to explain 98% of results (variation). Also, lack of fit was not statistically

338

significant (P=0.665). The model and its coefficients and quality analysis are presented in Table 5. The

339

contour plot generated from the quadratic model obtained for Bb-12 survival during in vitro assays is

340

presented in Fig. 3.

11

341

The quadratic model obtained revealed that probiotic Bb-12 survival was dependent on isolated

342

ingredients and some ingredients interactions. The higher viability of Bb-12 (>7.00 log cfu g-1) after the in

343

vitro tests was observed when inulin was added in concentrations varying between 3 and 2.25%. Similarly, a

344

satisfactory result was observed when margarine was supplemented with CMP, since concentrations between

345

3 and 2.25% resulted in Bb-12 populations between 5.25 and 6.25 log cfu g-1 after the entire in vitro assay

346

(6h). In contrast, WPC in the same concentrations resulted in Bb-12 populations below 4.25 log cfu g-1 after

347

6h of the in vitro assay.

RI PT

ACCEPTED MANUSCRIPT

The conditions found in human gastrointestinal tract are hostile for probiotic microorganisms.

349

Nevertheless, the food matrix protects the cells, and avoids high reductions in probiotic populations (El-

350

Shafei, Tawfik, Dabiza, Sharaf, & Effat, 2010). Therefore, the food matrix and its composition are very

351

important for protecting probiotic strains. The fat content, the pH and the solid matrix, may protect probiotic

352

bacteria more efficiently than a fluid environment during storage of the food product and its transit through

353

the human gastrointestinal tract (Ranadheera, Baines, & Adams, 2010). In the present study, the solid matrix

354

and the fat present in margarine assured the survival of the probiotic strain Bb-12 during the in vitro

355

resistance assays. However, it is important to emphasize that high counts during the shelf life of a specific

356

product are necessary to obtain appropriate populations of probiotic viable cells by the end of the digestive

357

process.

M AN U

SC

348

359 360

4. Conclusion

TE D

358

The results obtained showed that the production of probiotic margarine with suitable populations of

362

B. animalis Bb-12, and without hydrogenation process, with palm and canola oil, is possible. The

363

supplementation of the water phase was very important, since the control margarine (without the addition of

365 366 367

AC C

364

EP

361

inulin, WPC, and/or CMP) presented very low and insufficient Bb-12 populations. The presence of inulin, especially in the proportion of 3% resulted in significant higher Bb-12 populations (P<0.05) and assured Bb12 resistance through the in vitro assays. Populations observed for margarines (except M2) by the end of in vitro assays (after 6h) revealed that recovery of Bb-12 viable cells in satisfactory populations was possible,

368

even after the digestive process. As far as we know, taken together, results on viability of B. animalis Bb-12

369

and on in vitro GI resistance of the strain incorporated in margarine produced with inulin were unique, when

370

compared to other probiotic products described in the scientific literature. Therefore, functional margarine

371

supplemented with a probiotic strain like B. animalis Bb-12 ought to be produced commercially, though the

372

addition of inulin is necessary to lead to promising probiotic potential.

12

373 374

ACCEPTED MANUSCRIPT Acknowledgments

375 The authors would like to thank Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

377

(Projects 07/59260-0 and 06/54843-5), and Conselho Nacional de Desenvolvimento Científico e Tecnológico

378

(CNPq) for financial support and fellowships. The authors would also like to thank Agropalma, Kerry Bio-

379

Science, DSM, Danisco, Fortitech, Purac Sínteses, Orafti, and Arla Foods, for providing part of the material

380

resources employed in this study. The authors gratefully acknowledge Regina Célia Cabral Estrotra for her

381

technical assistance.

RI PT

376

382 References

384

Anvisa. (2003). Aprova regulamento técnico de porções de alimentos embalados para fins de rotulagem

SC

383

385

nutricional.

386

http://bvsms.saude.gov.br/bvs/saudelegis/anvisa/2003/res0359_23_12_2003.html.

M AN U

Available

from

387

Anvisa. (2008). Comissões de assessoramento tecnocientífico em alimentos funcionais e novos alimentos.

388

Aprova alimentos com alegações de propriedades funcionais ou de saúde, novos alimentos/ingredientes,

389

substâncias

390

http://www.anvisa.gov.br/alimentos/comissoes/tecno_lista_alega.htm.

e

probióticos.

Available

from

TE D

bioativas

Bedani, R., Rossi, E. A., & Saad, S. M. I. (2013). Impact of inulin and okara on Lactobacillus acidophilus

392

La-5 and Bifidobacterium animalis Bb-12 viability in a fermented soy product and probiotic survival

393

under in vitro simulated gastrointestinal conditions. Food Microbiology, 34(2), 382-389.

395 396 397 398 399 400 401 402 403 404

Beneo® HP. Product Sheet Beneo® HP, Orafti, DOC.A4-05* 01/02-B. Mannheim, Germany, 2012. Available from http://www.orafti.com.

AC C

394

EP

391

Bergamini, C. V., Hynes, E. R., Palma, S. B., Sabbag, N. G., & Zalazar, C. A. (2009). Proteolytic activity of three probiotic strains in semi-hard cheese as single and mixed cultures: Lactobacillus acidophilus, Lactobacillus paracasei and Bifidobacterium lactis. International Dairy Journal, 19(8), 467-475.

Bosscher, D., Loo, J. V., & Franck, A. (2006). Inulin and oligofructose as prebiotics in the prevention of intestinal infections and diseases. Nutrition Research Reviews, 19(2), 216-226. Bower, J. A. (1998a). Statistics for food science – V: ANOVA and multiple comparisons (part B). Nutrition and Food Science, 28(1), 41-48. Bower, J. A. (1998b). Statistics for food science – V part C: non-parametric ANOVA. Nutrition and Food Science, 28(2), 102-108.

13

405 406

Bruno, F. A., Lankaputhra, W. E. V., & Shah, N. P. (2002). Growth, viability and activity of Bifidobacterium

ACCEPTED MANUSCRIPT

spp. in skim milk containing prebiotics. Journal of Food Science, 67(7), 2740-2744.

407

Buriti, F. C. A., Castro, I. A., & Saad, S. M. I. (2010). Viability of Lactobacillus acidophilus in synbiotic

408

guava mousses and its survival under in vitro simulated gastrointestinal conditions. International

409

Journal of Food Microbiology, 137(2-3), 121-129. Cardarelli, H. R., Buriti, F. C. A., Castro, I. A., & Saad, S. M. I. (2008). Inulin and oligofructose improve

411

sensory quality and increase the probiotic viable count in potentially synbiotic petit-suisse cheese. LWT

412

- Food Science and Technology, 41(6), 1037-1046.

RI PT

410

Charteris, W. P., Kelly, P. M., Morelli, L., & Collins, J. K. (2002). Edible table (bio)spread containing

414

potentially probiotic Lactobacillus and Bifidobacterium species. International Journal of Dairy

415

Technology, 55(1), 44-56.

SC

413

Charteris, W. P., Kelly, P. M., Morelli, L., & Collins, J. K. (1998). Development and application of an in

417

vitro methodology to determine the transit tolerance of potentially probiotic Lactobacillus and

418

Bifidobacterium species in the upper human gastrointestinal tract. Journal of Applied Microbiology,

419

84(5), 759-768.

M AN U

416

Dommels, Y. E. M., Kemperman, R. A., Zebregs, Y. E. M. P., Draaisma, R. B., Jol, A., Wolvers, D. A. W.,

421

et al. (2009). Survival of Lactobacillus reuteri DSM 17938 and Lactobacillus rhamnosus GG in the

422

human gastrointestinal tract with daily consumption of a low-fat probiotic spread. Applied and

423

Environmental Microbiology, 75(19), 6198-6204.

TE D

420

El-Shafei, K., Tawfik, N. F., Dabiza, N. M. A., Sharaf, O. M., & Effat, B. A. (2010). In vitro assessment of

425

gastrointestinal viability of potentially probiotic Lactobacilli. Journal of American Science, 6(11), 357-

426

367.

428 429 430 431 432

Goli, S. A. H., Sahri, M. M., Kadivar, M., & Keramat, J. (2009). The production of an experimental table

AC C

427

EP

424

margarine enriched with conjugated linoleic acid (CLA): physical properties. Journal of American Oil Chemistry Society, 86(5), 453-458.

Guo, Z., Wang, J., Yan, L., Chen, W., Liu, X. M., & Zhang, H. (2009). In vitro comparison of probiotic properties of Lactobacillus casei Zhang, a potential new probiotic, with selected probiotic strains. LWT - Food Science and Technology, 42(10), 1640-1646.

433

Hansen, L. T., Allan-Wojtas, P. M., Jin, Y. L., & Paulson, A. T. (2002). Survival of Ca-alginate

434

microencapsulated Bifidobacterium spp. in milk and simulated gastrointestinal conditions. Food

435

Microbiology, 19(1), 35-45.

14

436

Haschke, F., Wang, W., Ping, G., Varavithya, W., Podhipak, A., Rochat, F., et al. (1998). Clinical trials

437

prove the safety and efficacy of the probiotic strain Bifidobacterium Bb12 in follow-up formula and

438

growing-up milks. Monatsschr Kinderheilkd, 146(1), 26-30.

ACCEPTED MANUSCRIPT

439

Hill, C., Guarner, F., Reid, G., Gibson, G. R., Merenstein, D. J., Pot, B., et al. (2014). The international

440

scientific association for probiotics and prebiotics consensus statement on the scope and appropriate use

441

of the term probiotic. Nature Reviews Gastroenterology & Hepatology, 11(8), 506-514.

443

Janer, C., Peláez, C., & Requena, T. (2004). Caseinomacropeptide and whey protein concentrate enhance

RI PT

442

Bifidobacterium lactis growth in milk. Food Chemistry, 86(2), 263-267.

Kos, B., Šušković, J., Goreta, J., & Matošić, S. (2000). Effect of protectors on the viability of Lactobacillus

445

acidophilus M92 in simulated gastrointestinal conditions. Food Technology and Biotechnology, 38(2),

446

121-127.

SC

444

Martín-Diana, A. B., Fraga, M. J., & Fontecha, J. (2002). Isolation and characterisation of

448

caseinomacropeptide from bovine, ovine and caprine cheese whey. European Food Research

449

International, 214(4), 282–286.

M AN U

447

450

Martins, E. M. F., Ramos, A. M., Vanzela, E. S. L., Stringheta, P. C., Pinto, C. L. O., & Martins, J. M.

451

(2013). Products of vegetable origin: A new alternative for the consumption of probiotic bacteria. Food

452

Research International, 51(2), 764-770.

454

Morelli, L. (2007). In vitro assessment of probiotic bacteria: From survival to functionality. International

TE D

453

Dairy Journal, 17(11), 1278-1283.

Ong, L., Henriksson, A., & Shah, N. P. (2006). Development of probiotic Cheddar cheese containing

456

Lactobacillus acidophilus, Lb. casei, Lb. paracasei and Bifidobacterium spp. and the influence of these

457

bacteria on proteolytic patterns and production of organic acid. International Dairy Journal, 16(5), 446-

458

456.

460 461 462 463

AC C

459

EP

455

Padilha, M., Villarreal-Morales, M. L., Vieira, A. D. S., Costa, M. G. M., & Saad, S. M. I. (2016). Prebiotic mixture improved Lactobacillus acidophilus and Bifidobacterium animalis gastrointestinal in vitro resistance in petit-suisse. Food & Function, 7(5), 2312-2319.

Ranadheera, R. D. C. S., Baines, S. K., & Adams, M. C. (2010). Importance of food in probiotic efficacy. Food Research International, 43(1), 1-7.

464

Rößle, C., Auty, M. A. E., Brunton, N., Gormley, R. T., & Butler, F. (2010). Evaluation of fresh-cut apple

465

slices enriched with probiotic bacteria. Innovative Food Science and Emerging Technologies, 11(1),

466

203-209.

467

Statsoft Inc. Statistica for Windows, Version 8.0, 2300 East 14th Street, Tulsa, OK, 74104, USA.

15

468 469

Tripathi, M. K., & Giri, S. K. (2014). Probiotic functional foods: survival of probiotics during processing and

ACCEPTED MANUSCRIPT

storage. Journal of Functional Foods, 9, 225-241. Vandenplas, Y., Huys, G., & Daube, G. (2015). Probiotics: an update. Journal of Pediatrics, 91(1), 6-21.

471

Vinderola, C. G., & Reinheimer, J. A. (1999). Culture media for the enumeration of Bifidobacterium bifidum

472

and Lactobacillus acidophilus in the presence of yoghurt bacteria. International Dairy Journal, 9(8),

473

497-505.

AC C

EP

TE D

M AN U

SC

RI PT

470

16

ACCEPTED MANUSCRIPT 1

Table 1

2

Simplex-centroid experimental design employed in the present study. Quantities of each ingredient (%) CMP3 (x3)

0 3.0 0 1.5 0 1.5 1.0 -

0 0 3.0 0 1.5 1.5 1.0 -

RI PT

WPC2 (x2)

2

Whey protein concentrate (Arla Foods Ingredients, Sønderhøj, Denmark; 82% protein).

3

Caseinomacropeptide (Arla Foods Ingredients, 85% protein).

4

M8: control trial.

– = without inulin, WPC and/or CMP addition.

AC C

EP

TE D

8

M AN U

3 4 5 6 7

Proportion of ingredients in the mixture Inulin1 (x1) (x1, x2, x3) M1 (1, 0, 0) 3.0 M2 (0, 1, 0) 0 M3 (0, 0, 1) 0 M4 (½, ½, 0) 1.5 M5 (½, 0, ½) 1.5 M6 (0, ½, ½) 0 (⅓, ⅓, ⅓) 1.0 M7 M84 (0, 0, 0) 1 Inulin (Orafti, 92% inulin + 8% fructose, glucose, and sucrose).

SC

Margarine Trials

1

ACCEPTED MANUSCRIPT 9

Table 2

10

Ingredients used for the production of the margarine trials studied, according to the experimental design

11

described in Table 1. Margarine formulations Ingredients (%)

M AN U

60.000 3.400 0.001 0.050 0.880 3.000

Control trial.

**

33.999 0.760 0.010 0.400 0.500

SC

30.999 0.760 0.010 0.400 0.500

60.000 3.400 0.001 0.050 0.880 -

60% of palm oil (Agropalma, Tailândia, Brazil) + 40% of canola oil (Liza, Mairinque, Brazil).

***

Margarines M1 - M7 were supplemented with a mixture containing inulin, WPC, and/or CMP, according to

Table 1. 1

TE D

– = without supplementation with inulin, WPC, and/or CMP. : (Cisne, Cabo Frio, Brazil); 2: (food-grade lactic acid 85% solution, Purac Sínteses, Rio de Janeiro, Brazil); 3:

(Nestlé, Araçatuba, Brazil); 4: (Christian Hansen, Hørsholm, Denmark);

5

(Kerry Bio-Science, Campinas,

EP

Brazil); 6: (β-carotene, DSM, São Paulo, Brazil); 7: (Danisco, Cotia, Brazil), 8: (Fortitech, Campinas, Brazil).

AC C

20

*

RI PT

Water phase Water Salt1 Lactic acid2 Skimmed powdered milk3 Probiotic culture4 Fat phase Fat** Emulsifiers5 Colouring agent6 Butter flavour7 Vitamin A8 Mixture***

12 13 14 15 16 17 18 19

M8*

M1-M7

2

ACCEPTED MANUSCRIPT

Table 3

22

Populations (mean values1 ± standard deviation) of B. animalis Bb-12 (log cfu g-1) in margarines M1-M8 after 0, 1, 7, 14, 21, 28, and 35 days of storage

23

at 5±1°C. Margarines M2

M3

M4

M5

M6

M7

03

8.39±0.33BCab

8.24±0.14Ca

8.71±0.10Aa

8.49±0.11Ba

8.77±0.10Aa

8.42±0.14Ba

8.14±0.12Ca

1

8.77±0.24Aa

7.82±0.19Bb

8.60±0.16Aa

8.43±0.05Aa

8.42±0.09Aab

8.63±0.22Aa

8.44±0.14Aa

5.12±0.10Cb

7

8.19±0.23Bb

7.55±0.17Cb

8.36±0.15ABab

8.18±0.07Bb

8.55±0.18Aa

8.26±0.13ABb

8.29±0.12Ba

3.10±0.15Dc

14

8.54±0.01Aa

6.74±0.17Cc

8.19±0.08ABb

8.03±0.06Bb

8.16±0.10ABbc

8.27±0.13ABb

8.40±0.16Aa

-

21

8.23±0.20Aab

6.54±0.14Dc

8.07±0.21ABbc

7.62±0.08Cc

8.10±0.04ABc

8.03±0.15ABc

7.81±0.29BCb

-

28

8.05±0.04Ab

5.40±0.16Dd

7.64±0.03Bc

7.56±0.15Bcd

7.52±0.09Bd

7.62±0.12Bd

6.81±0.01Cc

-

35

8.01±0.22Ab

4.64±0.26De

6.87±0.52Bd

7.32±0.04Bd

7.29±0.23Bd

7.27±0.09Be

6.26±0.16Cd

-

TE D

See Table 1 for description of trials M1-M8. n=6.

2

Control trial.

3

Day 0: counts observed in the same day of margarine production.

AC C

EP

1

A,B,C,D

8.52±0.07Ba

M AN U

(Days)

24 25 26 27 28 29 30

M82

M1

SC

Storage

RI PT

21

: For each day of storage, different capital letters in a row denote significant differences (P<0.05) between different trials.

a,b,c,d,e

: For each trial, different lowercase superscripts in a column denote significant differences (P<0.05) between different days of storage.

– = not detected.

31

3

ACCEPTED MANUSCRIPT 32

Table 4

33

Coefficients (b1, b2, b3) with their respective standard deviations, P values (model and lack of fit), and adjusted

34

R2 of the quadratic model obtained for B. animalis Bb-12 viability in margarines at 21 days of storage at 5±1°C.

35

Coefficients Inulin (b1) WPC (b2) CMP (b3) Inulin + WPC (b12) Inulin + CMP (b13) WPC + CMP (b23) P value P lack of fit Adjusted R2 n.s: non-significant.

SC

RI PT

Values 8.26 (0.08) 6.61 (0.09) 8.03 (0.09) n.s n.s 2.72 (0.34) <0.01 0.065230 0.86

M AN U

36 37

AC C

EP

TE D

38

4

ACCEPTED MANUSCRIPT 39

Table 5

40

Coefficients (b1, b2, b3) with their respective standard deviations, P values (model and lack of fit), and adjusted

41

R2 of the quadratic model obtained for B. animalis Bb-12 survival during in vitro tests at 21 days of storage at

42

5±1°C. Values 7.14 (0.12) 1.14 (0.07) 5.48 (0.08) 6.16 (0.48) -3.31 (0.39) 8.28 (0.34) <0.01 0.665 0.98

SC

RI PT

Coefficients Inulin (b1) WPC (b2) CMP (b3) Inulin + WPC (b12) Inulin + CMP (b13) WPC + CMP (b23) P value P lack of fit Adjusted R2

AC C

EP

TE D

M AN U

43

5

ACCEPTED MANUSCRIPT

Figure caption

Figure 1. Contour plot showing the effect of inulin, whey protein concentrate (WPC), and

RI PT

caseinomacropeptide (CMP) on B. animalis Bb-12 viability (log cfu g-1) at 21 days of refrigerated storage.

Figure 2. Survival of B. animalis Bb-12 (log cfu g-1) in margarines M1-M7 submitted to in

at 5±1°C (a, b, c, and d, respectively) before ( )], and enteric [4h (

) and 6h (

) and during exposition to simulated gastric

)] conditions. A,B,C,D,E: different superscript capital

M AN U

[2h (

SC

vitro simulated gastrointestinal conditions after 7, 14, 21, and 28 days of refrigerated storage

letters denote significant differences between margarines for the same sampling period of the in vitro assay.

a,b,c

: different lowercase superscript letters denote significant differences

between the sampling periods of the in vitro assays for the same margarine. See Table 1 for

TE D

description of margarines M1-M7.

Figure 3. Contour plot showing the effect of inulin, WPC, and CMP on B. animalis Bb-12

AC C

EP

resistance (log cfu g-1) during in vitro assays at 21 days of refrigerated storage.

AC C

EP

Figure 1

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

B. animalis Bb-12 (log cfu g-1)

10

Ba

8

Aa

ABa

Ab Ab Ac

Bb

Bb BCb

6 4

ABa

Ba

Ca

Ba ABb Bb Ab

Ab ABb

BCb Cb c Cc

Bb Bb Cc

Cc

M4

M5

M6

Db Db Db

2 0 M1

M2

M3

M7

Aa

8

Ac

Ab

Ab

ABa Ab

Bb

Ab ABbCb

Db

Ab Bb Db

Cb Eb Bc

2 0 M1

M2

M3

M4

Ab Bb Db

SC

6 4

Aa

ABa

ABa

Ba

Ca

M5

M6

M AN U

B. animalis Bb-12 (log cfu g-1)

10

RI PT

Margarines

(a)

ABbc Bb Ac

M7

Margarines

10

Aa

8

ABa

Ab Ab Ab

4

M2

EP

(c)

Bb Cc Bb

Ba

Cc Cb Bbc

Bb Bb

M6

M7

Bb

M3

M4

M5

Margarines

10 8

Aa

Ba

Ba Bb

Da

6

Cb

Ba

Ba

Ab Ab Bb

AC C

B. animalis Bb-12 (log cfu g-1)

BCb BCb Bb

Dc Db Cb

2 M1

Figure 2

ABa

ABa

Ca

Bb Bc CBc

6

0

(d)

Da

TE D

B. animalis Bb-12 (log cfu g-1)

(b)

Bb Bb

Cc

Dc

Bc

Bc

Ca

Ab

Bb Cb

BCb Bb

Dc Ec

4 2 0

M1

M2

M3

M4

Margarines

M5

M6

M7

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

Figure 3

ACCEPTED MANUSCRIPT Original manuscript “Margarine produced with inulin increases Bifidobacterium animalis Bb-12 gastrointestinal survival in vitro”, authors Cínthia Hoch Batista de Souza, Luiz Antonio Gioielli, and Susana Marta Isay Saad.

SC

Research Highlights

RI PT

Submission – LWT - Food Science and Technology

M AN U

Margarine with inulin, whey protein concentrate, and/or caseinomacropeptide at 3% was studied.

Viability and GI in vitro resistance of Bifidobacterium Bb-12 was evaluated Inulin improved Bb-12 survival in probiotic margarine during storage. Margarine with inulin improved Bb-12 resistance to in vitro GI assays.

AC C

EP

TE D

Inulin in the proportion of 3% resulted in significant higher Bb-12 survival.