Quantification of Cholesterol Solubilized in Dietary Micelles

Subscriber access provided by UNIV OF CALIFORNIA SAN DIEGO LIBRARIES

Article

Quantification of cholesterol solubilized in dietary micelles: dependence on human bile salt variability and presence of dietary food ingredients Filipe Manuel Coreta-Gomes, Winchil L. C. Vaz, Emeric Wasielewski, Carlos F.G.C. Geraldes, and Maria João Moreno Langmuir, Just Accepted Manuscript • DOI: 10.1021/acs.langmuir.6b00723 • Publication Date (Web): 15 Apr 2016 Downloaded from http://pubs.acs.org on April 18, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Langmuir is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Quantification of cholesterol solubilized in dietary micelles: dependence on human bile salt variability and presence of dietary food ingredients Filipe M. Coreta-Gomes1,3, Winchil L. C. Vaz 1,4, Emeric Wasielewski 3, Carlos F. G. Geraldes 2,3* and Maria João Moreno 1,3*

1.

Chemistry Department, University of Coimbra Rua Larga, Largo D. Dinis, 3004-535 Coimbra, Portugal

2.

Department of Life Sciences, Faculty of Science and Technology, University of Coimbra 3000-393 Coimbra, Portugal

3.

Coimbra Chemistry Center, CQC, Rua Larga, University of Coimbra, 3004-535 Coimbra, Portugal

4.

CEDOC, NOVA Medical School, Faculdade de Ciências Médicas, Universidade Nova de Lisboa, Lisbon, Portugal

1 ACS Paragon Plus Environment

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 32

Abstract Solubility of cholesterol in bile salt (BS) micelles is important to understand the availability of cholesterol for absorption in the intestinal epithelium and to develop strategies to decrease cholesterol intake from the intestinal lumen. This has been the subject of intense investigation, due to the established relation between the development of diseases such as atherosclerosis and high levels of cholesterol in the blood. In this work we quantify the effect of BS variability on the amount of cholesterol solubilized. The effect of some known hypocholesterolemic agents usually found in the diet is also evaluated, as well as some insight regarding the mechanisms involved. The results show that, depending on the bile salt composition, the average value of sterol per micelle is equal to or lower than one. The amount of cholesterol solubilized in the BS micelles is essentially equal to its total concentration until the solubility limit is reached. Altogether, this indicates that the maximum cholesterol solubility in the BS micellar solution is the result of saturation of the aqueous phase and depends on the partition coefficient of cholesterol between the aqueous phase and the micellar pseudophase. The effect on cholesterol maximum solubility for several food ingredients usually encountered in the diet was characterized using methodology developed recently by us. This method allows the simultaneous quantification of both cholesterol and food ingredient solubilized in the BS micelles even in the presence of larger aggregates, therefore avoiding their physical separation with possible impacts on the overall equilibrium. The phytosterols stigmasterol and stigmastanol decreased significantly cholesterol solubility with a concomitant reduction in the total amount of sterol solubilized, most pronounced for stigmasterol. Those results point towards coprecipitation being the major cause for the decrease in cholesterol solubilization by the BS micelles. The presence of tocopherol and oleic acid leads to a small decrease in the amount of cholesterol solubilized while palmitic acid slightly increases the solubility of cholesterol. Those dietary food ingredients are completely solubilized by the BS micelles indicating that the effects on cholesterol solubility are due to changes in the properties of the mixed micelles.


Page 3 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Introduction Of the 1800 mg of cholesterol absorbed per day by an average human, 600 mg are supplied by the diet and the remaining is due to endogenous production.1 Although sterols are vital molecules for eukaryotic life forms, including humans, cholesterol is also known for its potentially lethal properties. Indeed, high cholesterol levels in blood are supposed to be responsible for atherosclerosis, one of the most deadly and incapacitating diseases in humans.2 It is therefore crucial to develop methods that may reduce cholesterol blood level. The solubilization of cholesterol in the intestine, which is strongly dependent on the diet, can be a key research target for this task. Given its low solubility in water,3 cholesterol needs to be solubilized prior to absorption. This is achieved with the help of several molecules present in the duodenum content, namely bile salts (BS), the most representative of which are the ones conjugated with glycine, phospholipids (PL) and fatty acids (FA). This mixture of components self-aggregates and forms either intermixed micelles or vesicles depending on the BS/lipid ratio.4 These structures solubilize cholesterol and other hydrophobic nutrients and carry them to the apical membrane of the brush border located in the intestinal epithelium.5-7 BS are crucial for cholesterol absorption.4-6,8 Cholesterol solubilization by BS has been extensively studied in the past, focusing mainly on the therapeutic potentialities for gallstone disease.9-11 A substantial mapping of ternary and binary phase diagrams of PL, BS and cholesterol was determined.4,11-15 The size of the micelles, mixed micelles and vesicles formed with the different proportions of these components as well as cholesterol solubility has been addressed.9,12-29 One important observation was that BS with distinct hydrophobicities have different capacities to solubilize and to deliver cholesterol to membranes. However, the delivery efficiency is not directly dependent on the amount of cholesterol solubilized by the micelles. This has been interpreted as delivery being through aqueous cholesterol monomer and not by the fusion of cholesterol loaded micelles and the membranes.6 Another important factor in the relation between solubilization of cholesterol in BS micelles and its delivery to membranes may be related with the interaction of BS with the lipid bilayers and the effects on the properties of the latter.6,30-33 Most work has been done with taurine conjugates, although glycine conjugated BS have a higher prevalence in humans.34 This apparent contradiction is due to the lower pKa value of the taurine conjugates,35 with the ionized species being the only relevant one 3 ACS Paragon Plus Environment

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 32

in the whole range of pH values found in the duodenum (6 to 8)18. Another limitation associated with most previous work is the need to physically separate the distinct components in the system (micelles, vesicles and/or crystals). We have developed a method to quantitatively follow the solubilization of cholesterol in BS micelles by 13C NMR spectroscopy using [4-13C]-cholesterol.19 The main advantage of this method is that it permits monitoring cholesterol solubilization in the micelle in a non-invasive way, i.e., without using any technique that could disrupt the equilibrium, such as filtration or chromatography. In this work we extended the application of the NMR method developed to quantitatively evaluate the dependence of cholesterol solubilization in BS micelles upon the variation of some conditions present in the duodenum. Our approach was to reduce the complexity of the system (duodenum content) to two major class components only (BS and cholesterol). First we addressed the possibility of synergistic effects between the most representative BS present in the duodenum and how they affect cholesterol solubilization. Different compositions of BS, namely single, binary and a physiologically relevant ternary composition were studied at the concentrations usually observed in the duodenum. These studies provide valuable information regarding the effect of BS variability on cholesterol solubilization in the intestinal lumen, and therefore on its absorption through the intestinal epithelium. One of the most used strategies to reduce cholesterol absorption in the intestine is to include sterols of vegetable origin in the diet.36-37 Phytosterols are known to reduce cholesterol solubilization in BS micelles and in this way decrease cholesterol absorption.22,26,37-42 The mechanism involved in the reduction of cholesterol solubilization by the BS micelles is not clear, and may be the result of a decrease in cholesterol solubility in the micellar phase (due to co-solubilization) and/or a reduction of cholesterol solubility in the aqueous phase (due to co-precipitation). The distinction between the two mechanisms is relevant because a reduction in the capacity of BS micelles to solubilize hydrophobic solutes may compromise the absorption of important dietary compounds such as lipophilic vitamins. Important insight regarding the mechanism involved may be obtained from the quantification of both cholesterol and phytosterol solubilized in the same BS micellar solution. The method used in this work permits the direct quantification of solubilized cholesterol (due to enrichment in

13

C at carbon 4) as well as phytosterol

(from the chemical shift induced in the BS resonances)19. The phytosterols studied in this


Page 5 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

work are stigmasterol and stigmastanol, which are representative of the two most abundant classes known to affect cholesterol solubility. Fatty acids are another important class of dietary compounds known to affect cholesterol absorption.43 In this regard we addressed also the quantitative effect of the saturated FA palmitic acid (PA) and the unsaturated FA oleic acid (OA) upon cholesterol solubilization by BS micelles. Finally we also studied the effect of α-tocopherol (the component of vitamin E) on the solubilization of cholesterol as a model of a lipophilic vitamin. Figure 1 shows the structures of the different BS and dietary phytochemicals used in this work.

Figure 1. Molecular structures of BS (GDCA, GCDCA and GCA), cholesterol (enriched in 13C in position 4), phytosterols (stigmastanol and stigmasterol), vitamin (α-tocopherol) and FA (saturated – palmitic acid, unsaturated - oleic acid).


Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 32

Experimental Section N-Phenyl-1-naphthylamine (NPN) was obtained from Merck (Hohenbrunn, Germany), [4-13C]cholesterol and deuterium oxide (99.8%) for NMR experiments were obtained from CortecNet (Paris, France), and the BS sodium glycochenodeoxycholate (GCDCA), sodium glycocholate (GCA), sodium glycodeoxycholate (GDCA), αtocopherol, stigmastanol and stigmasterol were obtained from Sigma (Steinheim, Germany), while oleic acid and palmitic acid were provided by Avanti. The non-aqueous solvents used for sample preparation (chloroform, methanol, and acetone) were of spectroscopy grade, and the aqueous buffer components (Tris-HCl, NaCl, and NaN3) were of high purity and purchased from Sigma; water was first distilled and further purified by activated charcoal and deionization. The critical micellar concentration (CMC) of single, binary and ternary BS mixtures used were determined using a method described in the literature,44 based on the increase in fluorescence quantum yield and blue shift of the fluorescence emission spectra of N-phenyl-1-naphthylamine (NPN) upon association with preformed micelles. The total concentration of NPN was 1 µM, and the concentrations of simple, binary and ternary BS mixtures in the aqueous buffer were changed between 0 and 50 mM at 37 ºC. The fluorescent emission of NPN was followed using an excitation wavelength of 330 nm on a Cary Eclipse fluorescence spectrophotometer from Varian (Victoria, Australia) equipped with a thermostated multi-cell holder accessory. Tocopherol solubility was followed by absorption spectroscopy using its molar absorptivity coefficient (3.2×103cm-1M-1 at λ=292 nm). The concentration of tocopherol solubilized in GDCA micelles was obtained from its absorption at 292 nm, using a pathlength 0.1 cm. The size of BS micelles at a total concentration of 50 mM, with and without cholesterol solubilized, was also characterized by dynamic light scattering (DLS) using a Beckman Coulter N4 Plus apparatus operating with a He neon laser at 632.8 nm as light source and a scattering angle of 90o. The autocorrelation curves were analyzed with the PSC control software provided with the equipment. The refraction index considered for the samples was 1.33, and the viscosity was 1.127 cp.45 The experiments were done at 37 ºC, up to six measurements were run for each sample, and two independent samples were analyzed. Proton-decoupled

13

C NMR spectra acquisition was performed on a Varian

VNMRS 600 NMR spectrometer equipped with a high field ‘‘switchable’’ broadband 5mm probe with z-gradient operating at a frequency for 1H spectra (599.72 MHz) and for 6 ACS Paragon Plus Environment

Page 7 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

13

C spectra (150.8 MHz). 13C NMR spectra were acquired at 37 ºC with a 90º pulse angle

sequence, a spectral width of 31,250 Hz with an acquisition time of 2.3 s, a relaxation delay of 7 s, and 1024 acquisition scans. Proton decoupling was achieved using a WALTZ-16 decoupling sequence.

13

C-{1H} nuclear Overhauser enhancement (NOE)

was obtained by comparing 13C spectra with full proton decoupling and 13C spectra with proton decoupling only during acquisition (spectra acquired with 3000 scans). The values used for correcting the NOE effect upon signal intensity are shown in Table S1 and Table S2 (supplementary material). Spectra were processed with MestreNova 6.1.1 (Mestrelab Research, Santiago de Compostela, Spain). The relaxation delay used ensured that the 13C signals are quantitative for carbons with a relaxation time equal to 1.4 s or shorter. This is the case for C4 of cholesterol and for all BS carbons except carbonyl and quaternary carbon atoms.19 1H NMR spectra of stigmasterol and oleic acid were obtained at 37 ºC with a 90º pulse angle sequence, a spectral width of 7225 Hz, acquisition time of 1.5 s, a relaxation delay of 2.5 s, and 512 acquisition scans. Aqueous suspensions of [4-13C]cholesterol with or without different competitors (phytosterols, FA, tocopherol) were prepared by evaporating the required volume of a solution in chloroform/methanol (87:13, v/v), blowing dry nitrogen over the heated solution, and removing traces of organic solvent in a vacuum desiccator for 30 min at 23 ºC. The glass tube with the dry residue was then transferred to a thermostatic bath at 37 ºC and hydrated with a preheated solution of BS in the aqueous buffer (10 mM Tris–HCl (pH 7.4), 0.15 M NaCl, 1 mM ethylenediaminetetraacetic acid (EDTA), and 0.02% NaN3 in D2O) and was continuously stirred at 100 rpm for 24 to 48 h. In the case of FA another assay was done by solubilizing first the FA in the BS solution, waiting 24 h and then adding to a film of cholesterol. The total concentration of BS plus sterol was maintained at 50 mM in all solubilization experiments. In the competition experiments using phytosterols, FA and tocopherol the total concentration of GDCA used was 50 mM. The samples were left in the bath with continuous stirring during 24 to 48 h (experiments on the maximum solubilization) and then characterized by 13C NMR to obtain the amount of solubilized cholesterol. BS CH2 resonances (same chemical group followed in cholesterol) or TSP (trimethylsilyl propionate) were used as internal standards. NMR assignments were done based on information available in the literature.46 For the studies of cholesterol solubilization in the bile salt micelles the total concentration was kept constant (BS + Cholesterol = 50 mM) while changing the proportion between cholesterol and BS. However, in the competition experiments this 7 ACS Paragon Plus Environment

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 32

approach leads to a significant decrease in the bile salt concentration that would complicate the analysis of the results. We have therefore opted to maintain the BS concentration at 50 mM in those studies.


Page 9 of 32

Results and Discussion Determination of the CMC of glycine conjugated BS The CMC of GDCA was characterized previously using two independent methods, isothermal calorimetric titration and dye solubilization (NPN).19 The results obtained agreed within the experimental error. In this work we have determined the CMC of some additional BS common in human bile, both simple (GCA and GCDCA), binary mixtures (GCA/GDCA, GCA/GCDCA and GDCA/GCDCA, at equimolar concentrations) and a physiologically relevant ternary mixture (GCA/GCDCA/GDCA, at the molar ratio 37.5:37.5:25), using the NPN solubilization method. The results obtained are shown in Figure 2 and the calculated CMC values are presented in Table 1.

A

B

1.0

GCA GDCA GCDCA GDCA/GCA GDCA/GCDCA GCA/GCDCA GCA/GCDCA/ GDCA

0.8 0.6 0.4

0.8 0.6 0.4

0.2

0.2

0.0

0.0 360

400

440

λ (nm)

480

520 -1.5

-0.8

0.0

0.8

IFNormalized (a.u)

1.0

IFNormalized (a.u)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

1.5

Log[BS(mM)]

Figure 2. A) Fluorescent emission spectra of 1 µM NPN at increasing concentrations (ranging from 0.25 to 50 mM) of a ternary GCA/GCDCA/GDCA (37.5:37.5:25) BS solution in aqueous buffer pH=7.4 and I=150 mM at 37 ºC (excitation wavelength of 330 nm). B) Fluorescence intensity area from 340 to 530 nm for the different concentrations of BS mixtures tested. The CMC is calculated from the intersection of the extrapolated fit of the two different regimes, as shown for the case of GCA. The values obtained for the CMC and their standard deviations are shown in Table 1.

The results obtained show that GCA, a BS with three hydroxyl groups, has a much higher CMC than the other two BS studied (GCDCA and GDCA) with two hydroxyl groups (Figure 1), whose CMCs are identical within the experimental error. Thus, the


Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 32

trihydroxy BS is soluble in water in the monomeric state up to a higher concentration than the others, in qualitative agreement with data available in the literature.18,26 The CMC values obtained in this work for the isolated BS are systematically smaller than those reported in the literature using ITC for similar conditions (GDCA, 2.0 mM, GCDCA, 2.2 mM, and GCA, 8.5 mM)26. The observed differences between the two datasets are a consequence of the different approaches used to calculate the CMC value from the experimental results. The CMC is, by definition, the concentration at which micelles start to form.1 Accordingly, the CMC values reported in table 1 were calculated from the intercept between the linear best fits to the measured property at low concentrations and when the property starts changing abruptly due to the formation of micelles (figure 2B). The CMC values obtained for the binary mixtures, GDCA/GCA (50:50), GCA/GCDCA (50:50) and GDCA/GCDCA (50:50), were equal to the values predicted from ideal miscibility of surfactants in the micelles, i.e.: the free energy associated to the mixing is equal to the sum of free energies of each surfactant.47 The CMC obtained for the ternary mixture GCA/GCDCA/GDCA (37.5:37.5:25) is somewhat smaller than the value predicted considering an ideal mixture. This indicates that there are interactions of the components in this BS mixture, which cannot be described by the average behavior of each BS.

Dependence of Cholesterol Solubilization Index on the composition of bile salt micelles The determination of sterol solubilized in the different BS micelles was obtained using the NMR method developed by us.19 Briefly, labelled [4-13C]cholesterol solubilized in the micelles was followed by 13

13

C NMR spectroscopy directly through the [4-

C]cholesterol signal and also by the chemical shift induced in the BS resonances that

were most sensitive to the presence of cholesterol. In the chemical shift analysis, the concentration of cholesterol in the micelles was quantified from the chemical shift observed in some selected BS resonances using the slope obtained as a function of cholesterol concentration for the concentration region where this was known (total solute solubilization). The BS resonances were selected based on the sensitivity to the presence of cholesterol and on the linearity observed between the chemical shift and cholesterol

1

http://goldbook.iupac.org/C01395.html, accessed on 2016 February 4


Page 11 of 32

concentration (R2>0.98). Figure 3 shows the results obtained for the cholesterol saturation index (CSI) in the distinct mixtures, while the average values and standard deviations are shown in Table 1.

A

B

3.0

3.0

2.0

2.0 GDCA GCA GCDCA GDCA/GCA GDCA/GCDCA GCA/GCDCA GCA/GCDCA/GDCA

1.0 0.0 0

1

2

3

4

5 0

[Chol]T(mM)

1

2

3

4

1.0

[Chol]M (mM)

[Chol]M (mM)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

0.0 5

[Chol]T(mM)

Figure 3. Concentration of [4-13C]cholesterol solubilized by BS micelles ( [ Chol]M ) as a function of the total concentrations of cholesterol ( [ Chol]T ) for a total concentration of lipid (BS+ Chol) equal to 50 mM. The experiments were performed at 37 oC. The amount of solubilized cholesterol

was obtained from the area of its resonance for [4-13C]cholesterol (plot A) or from the chemical shift observed for the bile salt resonances (plot B). The CSI value can be obtained from the maximum [ Chol]M , or from the intersection of the best fit to both regimes in cholesterol solubilisation. The average and standard errors associated with the CSI obtained from at least three independent experiments are shown in Table 1.

The same values for CSI were obtained from both methods, showing that the chemical shift of BS resonances is a reliable method to quantify solubilization of cholesterol even for more complex mixtures (binary and ternary) of BS. GDCA, which was the BS solubilizing more cholesterol, is the one with lowest water solubility (lowest CMC) and the highest hydrophobicity. The most hydrophilic BS (GCA) showed the lowest solubilizing capacity for cholesterol. Surprisingly, although GCDCA has a similar CMC to GDCA, its cholesterol solubilizing capacity is closer to that of GCA. Similar qualitative, but not quantitative, results were previously obtained by other authors for the same BS.10,15,23 This emphasizes that the change in the hydroxyl position between the structures of GDCA and GCDCA is not affecting their solubility in water as monomer but considerably modifies the capacity of the hydrophobic core of the micelles to 11 ACS Paragon Plus Environment

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 32

accommodate cholesterol. The studied BS mixtures showed a CSI equal to the one obtained using the average of CSI of each BS weighted by their molar fraction. An exception is observed for the GDCA/GCDCA mixture, which showed a lower capacity to solubilize cholesterol than that expected from the average of each BS. The capacity of the mixture GCA/GCDA to solubilize cholesterol is also slightly higher than predicted from the average of the CSI for the single component BS micelles although this is in agreement with the strong decrease observed in the CMC of the mixture. The significant decrease in the CSI observed for the GDCA/GCDCA mixture is not in concordance with the effects observed on the CMC (only slightly larger than predicted from ideal mixing). This suggests an increased perturbation of the micelles’ core and reinforces the conclusion that the presence of the hydroxyl group (see structure of BS at Figure 1) at C7 has a particularly high destabilizing effect regarding cholesterol solubility. As expected from the results obtained with single component micelles, GCA/GCDCA mixtures show the lower cholesterol solubilizing capacity. The addition of GDCA to this mixture decreases the CMC and increases cholesterol solubility. The surprising result obtained from this work is that the addition of GCDCA to GDCA leads almost to the same effectivity in reducing the cholesterol solubilizing capacity. To the best of our knowledge, this is the first time that this effect among different BS compositions has been evaluated. To gain insight regarding changes in the structure of the mixed micelles, their size has been measured using DLS. For single component micelles, the size obtained correlates well with the micelle size and aggregation number obtained from literature (Table 1), with smaller micelles for the three hydroxyl BS and micelles from DCA derivatives being slightly smaller than for CDCA.48 The results obtained in this work for the micelles in the binary and ternary BS mixtures clearly show that the accommodation of the distinct BS in the same micelle leads to smaller micelles with a consequent increase in the concentration of micelles for the same total BS concentration. The results obtained for the biologically relevant ternary mixture shows that the system is not optimized for CSI (neither maximal nor minimal). It does suggest a preference for a large quantity of very small micelles maintaining a relatively high CSI. To better compare the results of cholesterol solubilization obtained by us with the results of others groups, we will use the ratio of the concentration of cholesterol solubilized in the micelles to the concentration of micelles ([Chol]M/[Micelles]). The aggregation number used for GDCA, GCA and GCDCA micelles is shown in Table 1. 12 ACS Paragon Plus Environment

Page 13 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Table 1. Values of CSI and CMC determined for single, binary and ternary mixtures of BS. The predicted values for CSI and CMC are also shown assuming that the mixtures of BS behave as an ideal solution. The values of aggregation number obtained by others for micelles in similar conditions to our studies are also given.

BS compositions

a)

CSI (mM)

CSIb) (mM)

Radius (nm)

Α

c)

[Micelles] [Chol ]M (mM) [ Micelles]

GDCA

1.7 ± 0.1

3.0 ± 0.1

2.6 ± 0.2

16

3.0

1.0

GCA

6.5 ± 0.1

1.0 ± 0.1

2.4 ± 0.4

6

7.8

0.1

GCDCA

1.6 ± 0.1

1.3 ± 0.1

2.9 ± 0.1

18

2.7

0.5

GDCA/GCA

2.7 ± 0.1

2.7

1.9 ± 0.1

2

2.4 ± 0.2

GDCA/GCDCA

1.8 ± 0.1

1.6

1.6 ± 0.2

2.2

2.4 ± 0.2

GCA/GCDCA GCDCA/GCA/ GDCA

2.7 ± 0.2

2.6

1.4 ± 0.1

1.2

2.3 ± 0.1

1.9 ± 0.2

2.3

1.8 ± 0.2

1.6

2.4 ± 0.2

The CMC of the mixtures is calculated from molar fraction of each component in solution (xi) and its

CMC when pure (CMCi), b)

CMCa) (mM)

CMC (mM)

=∑

.47

CSI predicted for the complex BS mixtures using the average value of cholesterol solubilization obtained

for each single BS weighted by their molar fraction. c)

Aggregation number per micelle, values obtained from literature.17,23-24

These results allow us to calculate the quantity of sterol per micelle (mean occupation number) which are also given in Table 1 for the simple BS micelles. The results obtained show that GDCA micelles contain an average of one molecule of cholesterol per micelle, these micelles being those that best accommodate cholesterol. The other dihydroxy BS, GCDCA, shows half of the solubilizing capacity of GDCA. GCA shows even less affinity for cholesterol with a mean occupancy of 0.1. The results are consistent with the available literature on cholesterol solubilization, where GDCA shows an average cholesterol solubilization per micelle of 0.9, GCDCA 0.4 and GCA 0.2.10,15,23 The solubilization profiles obtained show an essentially complete solubilization of cholesterol up to the saturation limit for all BS and BS mixtures tested, Figure 3. This, together with the small mean occupancy numbers, clearly indicates that the association between cholesterol and the BS micelles cannot be treated as binding to well defined sites 13 ACS Paragon Plus Environment

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 32

but rater by a partition between the aqueous intermicellar space and the micellar pseudophase.49 At the CSI, the BS aqueous solution is saturated with cholesterol. Whether saturation of the solution is due to the achievement of the limiting solubility in the aqueous or micellar phases cannot be judged easily. At a given concentration of total cholesterol, the concentration in the aqueous and micellar phases depends on the relative affinity for the two phases (partition coefficient). If the partition coefficient was known, the concentrations in each phase could be calculated and the comparison between the values obtained at the CSI and those at saturation when each single phase is present would lead to the unequivocal attribution of saturation in the aqueous or micellar phase. This is not however possible due to the low solubility of cholesterol in the aqueous phase and the impossibility of have micelles as a single phase solution. This discussion is nevertheless relevant because saturation of the micellar pseudo-phase could compromise the solubilization of other non-polar solutes and reduce their intestinal absorption. The observation that the maximal solubility of non-polar compounds (not prone to form micelles by their one) in micellar solutions decreases as their hydrophobicity increases,49 is strong evidence for solubility being limited by saturation of the aqueous phase. The values obtained for the CSI in the distinct micellar solutions, Table 1, show that the affinity of cholesterol for the single component BS micelles is highest for GDCA and lowest for GCA. The affinity of cholesterol for the physiologically relevant ternary mixture of BS is intermediate, showing that the system is not optimized for a maximal cholesterol solubility. In addition to the eventual preference for smaller size micelles (as discussed above), the physiological composition of BS micelles in the intestinal lumen may reflect an additional step required in the process of cholesterol absorption – association with the apical membrane of the brush border cells in the intestinal epithelium. To be absorbed the solutes must associate with the membranes of the epithelium (either with the lipid bilayer for passive absorption or with specific transporters in the membrane). This association is not mediated by dissolution of the BS micelles with the membrane but rather it is mediated by the solute in the aqueous phase. Therefore, maximal absorption is a trade-off between efficient solubilization in the intestinal lumen and a relatively large concentration in the aqueous phase to allow partition to the gastrointestinal membrane.


Page 15 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Effect of phytosterols on cholesterol solubilization in GDCA micelles Several compounds are known to affect cholesterol absorption in the intestine and have been implicated in changes in the solubilization capacity of the dietary mixed micelles (DMM) in the intestinal lumen. Phytosterols are probably one of the most studied families of molecules for this purpose.23,41,50 Fatty Acids (FA), namely the unsaturated ones, and some vitamins (namely α-tocopherol) are also involved in the decrease of cholesterol absorption and other lipophilic molecules.51 To understand the mechanism of action of these compounds and, most importantly, to determine quantitatively their effect, we used our previously established NMR method 19 to follow the solubilization of both cholesterol and dietary compound in the BS micelles. Figure 4 shows an expansion of a typical 13C NMR spectrum of GDCA micelles acquired in the presence of labelled [4-13C]cholesterol and of various dietary phytochemicals. The resonance at 44 ppm, corresponding to carbon four (C4) of labelled [4-13C]cholesterol, allows the quantitative determination of cholesterol solubilized in the micelles.

Figure 4. 13C NMR expanded spectra of 50 mM GDCA with 3.5 mM [4-13C]cholesterol (A) and

in the presence of 3.5 mM of dietary phytochemicals: α- tocopherol (B), Palmitic acid (C), Oleic acid (D), Stigmastanol (E), Stigmasterol (F). The spectra were acquired with 1H decoupling and NOE, in D2O aqueous buffer at 37 ºC. The peak of 13C4 enriched cholesterol appears at 44 ppm (black dotted box). Note the relative height of the peak at 44 ppm (cholesterol) showing an increase in the presence of palmitic acid and a decrease in the presence of all other phytochemicals.


Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 32

The results clearly showed that stigmasterol and stigmastanol were most effective at reducing cholesterol dissolution by BS micelles. They belong to two classes of phytosterols designated as sterols and stanols, respectively, known as inhibitors of cholesterol absorption. Two mechanisms can explain the decrease of cholesterol solubility in the presence of phytosterols.26,37,50 One possibility is the competition of both molecules for solubilization in the hydrophobic core of BS micelles in a process referred to as co-solubilization. Alternatively, the decrease in cholesterol solubilization may be due to a reduction of its aqueous solubility due to the formation of mixed crystals with the phytosterols, a mechanism of co-precipitation. The two processes are not mutually exclusive and therefore a combination of both processes may occur. The distinction between the two mechanisms is of practical relevance because if the co-precipitation is the major one, the capacity of BS micelles to solubilize other non-polar solutes is expected to be unaffected. Otherwise, the use of high concentrations of phytosterols in the diet could lead to a deficient absorption of other important non-polar solutes such as vitamins and/or pharmaceuticals. Insight regarding the mechanism of the decrease in cholesterol solubilization may be obtained from the simultaneous quantification of cholesterol and dietary food ingredients solubilized by the BS micelles. Stigmasterol structure is very similar to that of cholesterol except for an additional double bond in the side chain (see figure 1). This allowed the direct quantification of both sterols by NMR. The proton spectrum of cholesterol in BS micelles (Figure 5) has a peak at 5.3 ppm which is characteristic of alkene protons. Stigmasterol also has a proton at C4 position which resonates at this same frequency (black arrow in Figure 5). The two additional protons of the double bond in the side chain resonate closer to the water frequency at 4.6 ppm (gray arrows in Figure 5). Although these two latter signals could not be used to quantify stigmasterol due to baseline distortions caused by the proximity to the water resonance, the 5.3 ppm signal of the protons located in the cholesterol and stigmasterol backbone could be informative of the solubilization of both sterols.


Page 17 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Figure 5. 1H NMR spectra of 50 mM GDCA micelles in D2O buffer Tris-HCL pH=7.4 with 3.5

mM [4-13C]cholesterol. The inset show an expansion from 5.5 to 3.0 ppm with the peak of cholesterol solubilized in micelles (black line) and the peaks from cholesterol and stigmasterol (gray line). The arrows indicate the resonance from cholesterol (black) and from stigmasterol (gray) additional two alkene protons.

While in the proton spectrum the area of the signal at 5.3 ppm corresponds to both sterols solubilized in the micelle, in the 13C NMR spectrum only cholesterol is quantified because only it, and not stigmasterol, is enriched in 13C. Thus using a known quantity of TSP as internal reference, the total concentration of solubilized sterols was quantitatively determined from the 5.3 ppm proton peak area. The concentration of cholesterol in the same solution was obtained from quantitative

13

C NMR spectra.19 Therefore the

difference between both concentrations gave the solubility of stigmasterol in GDCA micelles. Figure 6 shows the results obtained for the effect of stigmasterol on cholesterol solubility in GDCA micelles, as well as the solubility of stigmasterol and the total sterol concentration.


Langmuir

4.0 3.0

Chol Stigmasterol Stigmasterol+Chol

B

Chol Stigmastanol Stigmastanol+Chol

3.0

2.0

2.0

1.0

1.0

0.0

[Sterol]M (mM)

4.0

A [Sterol]M (mM)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 32

0.0 0.00

1.75

3.50

[Stigmasterol] (mM)

0.00

1.75

3.50

[Stigmastanol] (mM)

Figure 6. Effect of stigmasterol (A) and stigmastanol (B) on cholesterol (3.5 mM) solubilization

by 50 mM GDCA micelles. The average and standard errors associated with values of solubilization at equilibrium for both cholesterol and stigmasterol in the GDCA micelles (

[Sterol]M ) obtained are from at least three independent experiments. In the absence of dietary phytochemicals the solubility of cholesterol in a 50 mM GDCA solution is 3.2 mM. This value is higher than the CSI reported in Table 1 for GDCA micelles because in the competition study with co-solutes a 50 mM total concentration of BS was used, while in the previous measurements the total concentration of cholesterol plus BS was 50 mM. As expected, the amount of cholesterol solubilized by the BS micelles decreases in the presence stigmasterol. This phytosterol is also solubilized by GDCA micelles, although the total amount of sterols solubilized (Cholesterol + Stigmasterol) decreases. The fraction of each sterol solubilized by the GDCA micelles is essentially equal to the total fraction in solution suggesting no preferential solubilization of any of the sterols. To the best of our knowledge this is the first time that solubilized cholesterol and phytosterol are measured directly in the same micellar solution using non-invasive techniques. Previous reports required the physical separation between micelles and precipitates, such as filtration or chromatography, and significant perturbations in the equilibrium could occur. The observation that the total sterol solubilized decreases as the proportion of stigmasterol is increased points towards the involvement of co-precipitation with a lower aqueous solubility for the mixture cholesterol plus stigmasterol than that of cholesterol alone. Sterols solubility in aqueous solutions are not known with high accuracy but available data in the literature indicates a smaller solubility of stigmasterol 18 ACS Paragon Plus Environment

Page 19 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

as compared with cholesterol.52 This is in fact expected due to the additional two carbons in the side chain which are not compensated by the extra double bond. Another evidence for co-precipitation was obtained from the analysis of formed crystals by polarized microscopy (data not shown), which show crystal morphologies different from those of cholesterol alone, and are in agreement with data shown in the literature50. The solubilization of cholesterol and stigmastanol in GDCA micelles was also characterized and is shown in Figure 6, Panel B. As observed for stigmasterol, the presence of stigmastanol leads to a decrease in cholesterol solubility. However, the results suggest a relatively higher effect at a cholesterol:stigmastanol ratio of 0.5. To elucidate the mechanism involved it would be important to quantify the solubility of stigmastanol in GDCA micelles. The method used above for stigmasterol cannot be used due to the absence of a specific proton or carbon resonance from stigmastanol that could be separately observed and used for the quantification of its micellar solubilization. The quantification of stigmastanol in the GDCA micelles was quantitatively followed through its effects in the resonances of GDCA, using quantitative chemical shift analysis (qCSA).19 This methodology may be used to quantify the total sterol solubilized in the micelles, considering that the effect induced by the solutes is known. Due to their similar chemical structures, this effect is expected to be the same for both cholesterol and the phytosterols. This approach was first validated using stigmasterol by comparison with the results obtained from its direct quantification by 1H NMR (see above). Table S3 in the supplementary material shows the chemical shift variation and the slopes of such plots obtained for the micelles in the presence of cholesterol or cholesterol plus stigmasterol. The slopes relating chemical shift with solubilized sterol show similar values in the two situations. This allows determination of the total solubilized sterol by following the chemical shift of the most sensitive BS resonances. The determined value of solubility was equal to the value of solubilized sterol in the micelle determined by quantitative 1H NMR (Table 2). This demonstrates the robustness of the method and the validity of the initial assumption regarding similar locations and effects in the BS micelles. This same strategy was followed for stigmastanol allowing the quantification of stigmastanol solubility in the GDCA micelle in the presence of cholesterol, Table S4 of the supplementary material. The results obtained for the amount of stigmastanol and total sterol solubilized in the GDCA micelles is shown in Figure 6B and Table 2. The absence of double bonds in stigmastanol would predictably decrease its solubility in aqueous solutions even below 19 ACS Paragon Plus Environment

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 32

that of stigmasterol. Data available in the literature however indicates otherwise, with the solubility of stigmastanol being between that of cholesterol and stigmasterol.52 The observation that the total amount of sterol solubilized in the GDCA micelles does not change significantly as the concentration of stigmastanol is increased, suggests that the overall sterol solubility is not significantly affected. The literature regarding solubilization of phytosterols and their quantitative effects on decreasing cholesterol solubility in BS micelles is not consensual. Cholesterol and stigmasterol solubility in non-conjugated BS was studied as a function of BS concentration. The results showed that the presence of stigmasterol in deoxycholate micelles, at a similar concentration to the one used in our work, induced a mild decreasing effect on cholesterol solubility.27 In the same work, cholestanol (a stanol similar to stigmastanol) showed a strong decrease in the amount of solubilized cholesterol. Also, the total sterol solubilized at the BS micelle as compared with the cholesterol solubilized in absence of phytosterols was lower for the case of cholestanol and similar for the case stigmasterol. Other authors have encountered a significant increase in the total amount of sterol solubilized by taurine conjugated DCA when both cholesterol and sitosterol or sitostanol are present (the amount of cholesterol solubilized being reduced to similar extent by both phytosterols).52 The same authors have characterized the effects of several phytosterols and the results encountered were most frequently an increase in the total amount of sterol solubilized in the BS micelles despite the smaller solubility of the phytosterols alone and a reduction in the amount of cholesterol solubilized in the presence of the phytosterol.40 Those results are unexpected both when interpreted as competition for solubilization in the BS micelle or co-precipitation in mixed crystals. The methodologies used to quantify the amount of sterols solubilized in the micelles, involving the physical separation between the micelles, may be the basis for those discrepancies. In fact, with the approach followed in previous work it is not possible to distinguish between sterol in small crystals (smaller than 0.2 µm) and solubilized in BS micelles. Another source of discrepancy may be due to different methodologies followed in the sterol solubilization by the BS micelles. The distinct sterols were added to the BS solution as a mixture of preformed single crystals. In this work a film of the distinct sterols was prepared by solvent evaporation from a homogeneous solution, thus favoring the formation of mixed crystals when the aqueous solution containing the BS is added. This methodology is physiologically more relevant as sterol precipitation is expected to occur during digestion, as the emulsifying lipids from diet are progressively digested. 20 ACS Paragon Plus Environment

Page 21 of 32

Table 2. Comparison between total sterol (cholesterol+ stigmasterol) solubilized in GDCA micelles determined by quantitative 13C and 1H NMR spectroscopy

and the total sterol determined from quantitative chemical shift analysis (qCSA). The determination of stigmastanol solubilized in the micelle was done using the same chemical shift analysis as for stigmasterol. The total cholesterol present in the solutions was 3.5 mM and the concentration of GDCA was always 50

Quantitative NMR

mM.

qCSA NMR

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47

Langmuir

mM

Cholesterol

Chol/Stigmasterol 2:1

Chol/Stigmasterol 1:1

Chol/Stigmastanol 2:1

Chol/Stigmastanol 1:1

[Chol ]M

3.2 ± 0.1

1.6 ± 0.2

1.0 ± 0.2

1.8 ± 0.1

1.6 ± 0.1

0.6 ± 0.2

1.0 ± 0.1

2.2 ± 0.3

2.0 ± 0.1

2.1 ± 0.3

2.0 ± 0.1

2.5 ± 0.2

2.9 ± 0.2

0.7 ± 0.2

1.3 ± 0.1

[ Stigmasterol]M [Total Sterol]M [Total Sterol]M [ Stigmastanol]M [ Stigmasterol]M

a)

a)

a)

0.5 ± 0.1

1.0 ± 0.1

The concentration of stigmasterol and stigmastanol was obtained by the difference between the total sterol obtained by chemical shift

analysis and the concentration of cholesterol determined by 13C NMR quantification for at least three independent experiments.


Langmuir

Effect of saturated and unsaturated fatty acids on cholesterol solubilization in GDCA micelles To study the effect of saturated and unsaturated FA on cholesterol solubilization, we used palmitic (PA) and oleic acid (OA). Those two fatty acids were chosen because they have a similar solubility in aqueous solutions 53. Therefore, the effect on cholesterol solubility should be due only to the presence of the cis double bond (and not also due to a different aqueous solubility as would be the case for oleic and stearic acid). The results obtained are shown in

4.0

A

Chol PA

B

Chol OA

4.0

3.0

3.0

2.0

2.0

1.0

1.0

[Fatty acid]M (mM)

Figure 7.

[Cholesterol]M (mM)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 32

0.0

0.0 0.00

1.75

[PA] (mM)

3.50

0.00

1.75

3.50

[OA] (mM)

Figure 7. Effect of saturated (palmitic acid-left panel) and unsaturated fatty acids (oleic acid-right

panel) on cholesterol solubilization. Values shown are for fatty acids/cholesterol ratios equal to 0, 0.5 and 1. BS composition is 50 mM GDCA and the cholesterol is 3.5 mM in all experiments. The values indicated for the solubility are the average and standard deviation of at least three independent experiments. The solubility of the FA in the BS micelles in the presence of cholesterol is also shown being obtained by 1H NMR (for OA) or by qCSA (for PA).

The presence of the saturated palmitic acid leads to a small increase in the solubility of cholesterol. For these experiments two different assays were performed. In the first one a film of cholesterol and FA was prepared and allowed to equilibrate with the BS aqueous solution for 24 h. In the other assay, a pre-equilibrated solution of BS with FA was added to the cholesterol film. The results obtained from the two experiments were similar and are included in the results shown in Figure 7. In the absence of FA, GDCA micelles solubilize 3.2 mM out of a total cholesterol concentration of 3.5 mM. Addition of 1.75 mM PA to the BS micelles leads to a slight increase of cholesterol solubility. Doubling the concentration of PA in the system leads to the total solubilization of the cholesterol present in solution. A higher 22 ACS Paragon Plus Environment

Page 23 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

cholesterol concentration (5 mM) was used for both PA concentrations (1.75 and 3.5 mM) but no further solubilization was achieved (results not shown). The results showed that the presence of PA in the BS micelles leads to a modest increase on the solubility of cholesterol in small micelles. The analysis of the changes in the 13C chemical shifts of the different resonances of BS upon cholesterol and FA addition showed that several resonances were sensitive to the increase of FA in the system and could be used to quantify solubilized FA. Table S5 (Supplementary Material) shows a typical example of the quantitative determination of the solubilized PA in the BS micelles followed by qCSA. This result demonstrates that PA was completely solubilized by BS micelles at both concentrations used. The results obtained for cholesterol solubility in the presence of OA (Figure 7, right panel) showed a different trend when compared with PA. The unsaturated OA leads to a slight decrease in cholesterol solubility, which is very small when compared with the effect of phytosterols in the same proportions. The solubility of OA in the micelles was also followed by quantitative 1H NMR and by chemical shift analysis (Table S6 Supplementary Material). Complete solubilization of OA was obtained, at both concentrations used (1.75 and 3.5 mM). This result is different from that observed for phytosterols, where the dietary compound was not completely solubilized by the micelles and showed a defined solubility limit. This was consistent with observations done with DMM where oleic acid showed to decrease cholesterol solubility when high OA/cholesterol ratios were used.41 Evidence for the solubilization of both FA and cholesterol in the BS micelles was obtained from the chemical shift variations of BS induced by the presence of the solutes and by the absence of large FA micelles as measured by DLS. No significant increase in the size of the mixed micelles was observed, as expected from the small number of solute molecules solubilized in each BS micelle.

Effect of α-tocopherol on cholesterol solubilization in GDCA micelles Vitamin E is one of the most abundant lipid-soluble antioxidants in higher mammals.51 It can exist in two forms: tocopherols and tocotrienols. α-tocopherol (molecular structure in Figure 1) is one of the most representative forms of tocopherols. Its effect on cholesterol solubilization by BS micelles was studied and the results obtained are shown in Figure 8.


Cholesterol Tocopherol

4

[Chol]M (mM)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 32

4

3

3

2

2

1

1

0

0

[Tocopherol]M (mM)

Langmuir

a

0.00

1.75

3.50

[Tocopherol] mM Figure 8. Effect of tocopherol on cholesterol solubility by the BS micelles. The values shown are for

tocopherol/cholesterol ratios of 0, 0.5 and 1. The BS composition is 50 mM GDCA and the total cholesterol content used in each experiment is 3.5 mM. The solubility values are the average and standard deviation of at least three independent experiments. The solubilization of tocopherol in the BS micelles in the presence of sterol was measured by UV absorption spectroscopy and qCSA and is also shown.

The solubilization profile shows a decrease of cholesterol solubility in the presence of α-tocopherol. This effect is clearly lower than that of phytosterols, but is higher than observed for oleic acid. Tocopherol solubility in GDCA micelles was followed by UV absorption spectroscopy and by qCSA (Table S7 Supplementary Material) and it was complete for both concentrations studied.


Page 25 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Conclusions The results presented in this work clearly show that the BS micelles composition present in vivo is not optimized for the highest solubilizing capacity, instead they show a compromise between the concentration of micelles, their size and the ability to solubilize cholesterol. This reflects the dual physiological role of BS micelles in the intestinal lumen which not only help in the solubilization of hydrophobic molecules but also work as a vehicles for their delivery to the intestinal membrane. In order to understand the effect of diet changes on cholesterol solubility by BS micelles in the intestinal lumen, we studied the effect of several hydrophobic molecules, such as phytosterols (stigmasterol and stigmastanol), fatty acids (saturated palmitic acid and mono-unsaturated oleic acid) and vitamin E (α-tocopherol). These molecules, namely phytosterols, unsaturated FA and tocopherol, are known to decrease cholesterol absorption in the intestine,1,54 although the mechanism is not fully understood 22,26,37-38,41. The methodology used in this work permits the direct quantification of both cholesterol and dietary food ingredient, allowing in this way a better elucidation of the mechanism involved in the effect on cholesterol solubilization. The results obtained show that all dietary food ingredients tested are co-solubilized with cholesterol in the BS micelles. This co-solubilization leads to a decrease in the maximal solubility of cholesterol for most dietary food ingredients tested (phytosterols, α-tocopherol and OA) and to an increase for the case of the saturated fatty acid (PA). The total amount of solute solubilized by the BS micelles (cholesterol plus dietary food ingredient) decreases in the case of phytosterols (more pronounced for stigmasterol) and increases for all other food ingredients added with its complete solubilization in the BS micelles. Those results clearly show that saturation of the BS micelles capacity to accommodate solutes is not the mechanism involved in the reduction of cholesterol solubilization by the BS micelles. In contrast, effects in the maximal cholesterol solubility in the micellar solution should reflect variations in its tendency to precipitate (co-precipitation with the added dietary food ingredient, as observed in the case of the phytosterols) and/or variations in the properties of the mixed micelles formed by the BS and the added food ingredient (as observed for the case of FA and α-tocopherol). In the former mechanism, the efficiency of the decrease in cholesterol solubilization depends on the aqueous solubility of the dietary food ingredient added and on the relative stability of mixed crystals. The efficiency observed in the later mechanism depends on the relative affinity of cholesterol for BS micelles when pure and mixed with the dietary food ingredient. In this 25 ACS Paragon Plus Environment

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 26 of 32

respect, the shape and rigidity of the added food ingredient plays a crucial role due to the interactions with the rigid steroid moiety of cholesterol. The understanding of those mechanisms and the molecular constraints involved is the first step towards the design better strategies to decrease cholesterol absorption and/or in the optimization of BS based models for drug delivery to the gastrointestinal membrane.


Page 27 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Acknowledgements

The Coimbra Chemistry Centre (CQC) is supported by the Portuguese “Fundação para a Ciência e a Tecnologia” (FCT) through the Project Nº 007630 UID/QUI/00313/2013, co-funded by COMPETE2020-UE. This work was also supported by project RECI/QEQ-QFI/0168/2012. Filipe Coreta-.Gomes acknowledges FCT for the PhD fellowship SFRH/BD/40778/2007. We acknowledge the Rede Nacional de RMN for access to the facilities. The Varian VNMRS 600 MHz spectrometer is part of the National NMR Network (PTNMR) and was purchased in the framework of the National Program for Scientific Re-equipment, contract REDE/1517/RMN/2005, with funds from POCI 2010 (FEDER) and FCT. We also acknowledge the Centre for Neuroscience and Cell Biology for access to the DLS.


Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 28 of 32

References

(1) Wilson, M. D.; Rudel, L. L., Review of Cholesterol Absorption with Emphasis on Dietary and Biliary Cholesterol. J. Lipid Res. 1994, 35 (6), 943-955. (2) Glass, C. K.; Witztum, J. L., Atherosclerosis: The road ahead. Cell 2001, 104 (4), 503-516. (3) Haberlan.Me; Reynolds, J. A., Self-Association of Cholesterol in Aqueous-Solution. Proc. Natl. Acad. Sci. USA 1973, 70 (8), 2313-2316. (4) Staggers, J. E.; Hernell, O.; Stafford, R. J.; Carey, M. C., Physical-Chemical Behavior of Dietary and Biliary Lipids During Intestinal Digestion and Absorption .1. Phase-Behavior and Aggregation States of Model Lipid Systems Patterned After Aqueous Duodenal Contents of Healthy Adult Human-Beings. Biochem. 1990, 29 (8), 2028-2040. (5) Watt, S. M.; Simmonds, W. J., Specificity of Bile-Salts in Intestinal-Absorption of Micellar Cholesterol in Rats. Clinical and Experimental Pharmacology and Physiology 1976, 3 (4), 305-322. (6) Westergaard, H.; Dietschy, J. M., Mechanism Whereby Bile-Acid Micelles Increase Rate of Fatty-Acid and Cholesterol Uptake Into Intestinal Mucosal Cell. J. Clin. Invest. 1976, 58 (1), 97-108. (7) Woollett, L. A.; Wang, Y.; Buckley, D. D.; Yao, L.; Chin, S.; Granholm, N.; Jones, P. J. H.; Setchell, K. D. R.; Tso, P.; Heubi, J. E., Micellar solubilisation of cholesterol is essential for absorption in humans. Gut 2006, 55 (2), 197-204. (8) Siperstein, M. D.; Chaikoff, I. L.; Reinhardt, W. O., C-14-Cholesterol .5. Obligatory Function of Bile in Intestinal Absorption of Cholesterol. J. Biol. Chem. 1952, 198 (1), 111114. (9) Carey, M. C.; Small, D. M., Physical-Chemistry of Cholesterol Solubility in Bile Relationship to Gallstone Formation and Dissolution in Man. J. Clin. Invest. 1978, 61 (4), 998-1026. (10) Igimi, H.; Carey, M. C., Cholesterol Gallstone Dissolution in Bile - Dissolution Kinetics of Crystalline (Anhydrate and Monohydrate) Cholesterol with Chenodeoxycholate, Ursodeoxycholate, and Their Glycine and Taurine Conjugates. J. Lipid Res. 1981, 22 (2), 254-270. (11) Wang, D. Q. H.; Cohen, D. E.; Carey, M. C., Biliary lipids and cholesterol gallstone disease. J. Lipid Res. 2009, 50, S406-S411. (12) Hofmann, A. F., Molecular Association in Fat Digestion. Interaction in Bulk of Monoolein, Oleic Acid, and Sodium Oleate with Dilute, Micellar Bile Salt Solutions. In Molecular Association in Biological and Related Systems, Goddard, E. D., Ed. American Chemical Society Publications: USA, 1968; Vol. 84, pp 53-66. (13) Donovan, J. M.; Timofeyeva, N.; Carey, M. C., Influence of Total LipidConcentration, Bile-Salt Lecithin Ratio, and Cholesterol Content on Inter-Mixed Micellar/Vesicular (Non-Lecithin-Associated) Bile-Salt Concentrations in Model Bile. J. Lipid Res. 1991, 32 (9), 1501-1512. (14) Small, D. M.; Bourges, M.; Dervichi.Dg, Ternary and Quaternary Aqueous Systems Containing Bile Salt Lecithin and Cholesterol. Nature 1966, 211 (5051), 816-818. (15) Neiderhiser, D. H.; Roth, H. P., Cholesterol Solubilization by Solutions of Bile Salts and Bile Salts Plus Lecithin. Proceedings of the Society for Experimental Biology and Medicine 1968, 128 (1), 221-&. (16) Armstrong, M. J.; Carey, M. C., Thermodynamic and Molecular Determinants of Sterol Solubilities in Bile-Salt Micelles. J. Lipid Res. 1987, 28 (10), 1144-1155.


Page 29 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

(17) Carey, M. C.; Montet, J. C.; Phillips, M. C.; Armstrong, M. J.; Mazer, N. A., Thermodynamic and Molecular-Basis for Dissimilar Cholesterol-Solubilizing Capacities by Micellar Solutions of Bile-Salts - Cases of Sodium Chenodeoxycholate and Sodium Ursodeoxycholate and Their Glycine and Taurine Conjugates. Biochem. 1981, 20 (12), 36373648. (18) Carey, M. C.; Small, D. M., Micelle Formation by Bile-Salts - Physical-Chemical and Thermodynamic Considerations. Arch. Intern. Med. 1972, 130 (4), 506-527. (19) Coreta-Gomes, F. M.; Vaz, W. L. C.; Wasielewski, E.; Geraldes, C.; Moreno, M. J., Quantification of cholesterol solubilized in bile salt micellar aqueous solutions using C-13 nuclear magnetic resonance. Anal. Biochem. 2012, 427 (1), 41-48. (20) Garidel, P.; Hildebrand, A.; Neubert, R.; Blume, A., Thermodynamic characterization of bile salt aggregation as a function of temperature and ionic strength using isothermal titration calorimetry. Langmuir 2000, 16 (12), 5267-5275. (21) Janich, M.; Lange, J.; Graener, H.; Neubert, R., Extended light scattering investigations on dihydroxy bile salt micelles in low-salt aqueous solutions. J. Phys. Chem. B 1998, 102 (31), 5957-5962. (22) Jesch, E. D.; Carr, T. P., Sitosterol reduces micellar cholesterol solubility in model bile. Nutrition Research 2006, 26 (11), 579-584. (23) Matsuoka, K.; Maeda, M.; Moroi, Y., Micelle formation of sodium glyco- and taurocholates and sodium glyco- and taurodeoxycholates and solubilization of cholesterol into their micelles. Colloids Surf., B 2003, 32 (2), 87-95. (24) Matsuoka, K.; Suzuki, M.; Honda, C.; Endo, K.; Moroi, Y., Micellization of conjugated chenodeoxy- and ursodeoxycholates and solubilization of cholesterol into their micelles: comparison with other four conjugated bile salts species. Chem. Phys. Lipids 2006, 139 (1), 1-10. (25) Mazer, N. A.; Carey, M. C., Quasi-Elastic Light-Scattering-Studies of Aqueous Biliary Lipid Systems - Cholesterol Solubilization and Precipitation in Model Bile Solutions. Biochem. 1983, 22 (2), 426-442. (26) Mel'nikov, S. M.; ten Hoorn, J. W. M. S.; Eijkelenboom, A. P. A. M., Effect of phytosterols and phytostanols on the solubilization of cholesterol by dietary mixed micelles: an in vitro study. Chem. Phys. Lipids 2004, 127 (2), 121-141. (27) Nagadome, S.; Okazaki, Y.; Lee, S.; Sasaki, Y.; Sugihara, G., Selective solubilization of sterols by bile salt micelles in water: A thermodynamic study. Langmuir 2001, 17 (14), 4405-4412. (28) Neiderhiser, D. H.; Roth, H. P., Effect of Modifications of Lecithin and Cholesterol on Micellar Solubilization of Cholesterol. Biochim. Biophys. Acta 1972, 270 (3), 407-&. (29) Small, D. M., Size and Structure of Bile Salt Micelles. Influence of Structure, Concentration, Counterion Concentration, pH, and Temperature. In Molecular Association in Biological and Related Systems, Goddard, E. D., Ed. American Chemical Society Publications: USA, 1968; Vol. 84, pp 31-52. (30) Coreta-Gomes, F. M.; Martins, P. A. T.; Velazquez-Campoy, A.; Vaz, W. L. C.; Geraldes, C. F. G.; Moreno, M. J., Interaction of Bile Salts with Model Membranes Mimicking the Gastrointestinal Epithelium: A Study by Isothermal Titration Calorimetry. Langmuir 2015, 31 (33), 9097-9104. (31) Chen, Y.; Lu, Y.; Chen, J.; Lai, J.; Sun, J.; Hu, F.; Wu, W., Enhanced bioavailability of the poorly water-soluble drug fenofibrate by using liposomes containing a bile salt. Int. J. Pharmaceut. 2009, 376 (1-2), 153-160. (32) Mello-Vieira, J.; Sousa, T.; Coutinho, A.; Fedorov, A.; Lucas, S. D.; Moreira, R.; Castro, R. E.; Rodrigues, C. M. P.; Prieto, M.; Fernandes, F., Cytotoxic bile acids, but not cytoprotective species, inhibit the ordering effect of cholesterol in model membranes at 29 ACS Paragon Plus Environment

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 30 of 32

physiologically active concentrations. Biochim. Biophys. Acta-Biomembr. 2013, 1828 (9), 2152-2163. (33) Schubert, R.; Schmidt, K. H., Structural-Changes in Vesicle Membranes and Mixed Micelles of Various Lipid Compositions After Binding of Different Bile-Salts. Biochem. 1988, 27 (24), 8787-8794. (34) Gowda, G. A. N.; Somashekar, B. S.; Ijare, O. B.; Sharma, A.; Kapoor, V. K.; Khetrapal, C. L., One-step analysis of major bile components in human bile using H-1 NMR spectroscopy. Lipids 2006, 41 (6), 577-589. (35) Fini, A.; Feroci, G.; Roda, A., Acidity in bile acid systems. Polyhedron 2002, 21 (1415), 1421-1427. (36) De Smet, E.; Mensink, R. P.; Plat, J., Effects of plant sterols and stanols on intestinal cholesterol metabolism: Suggested mechanisms from past to present. Molecular Nutrition & Food Research 2012, 56 (7), 1058-1072. (37) Trautwein, E. A.; Duchateau, G.; Lin, Y. G.; Mel'nikov, S. M.; Molhuizen, H. O. F.; Ntanios, F. Y., Proposed mechanisms of cholesterol-lowering action of plant sterols. European Journal of Lipid Science and Technology 2003, 105 (3-4), 171-185. (38) Brown, A. W.; Hang, J.; Dussault, P. H.; Carr, T. P., Phytosterol Ester Constituents Affect Micellar Cholesterol Solubility in Model Bile. Lipids 2010, 45 (9), 855-862. (39) Matsuoka, K.; Hirosawa, T.; Honda, C.; Endo, K.; Moroi, Y.; Shibata, O., Thermodynamic study on competitive solubilization of cholesterol and P-sitosterol in bile salt micelles. Chem. Phys. Lipids 2007, 148 (1), 51-60. (40) Matsuoka, K.; Kajimoto, E.; Horiuchi, M.; Honda, C.; Endo, K., Competitive solubilization of cholesterol and six species of sterol/stanol in bile salt micelles. Chem. Phys. Lipids 2010, 163 (4-5), 397-402. (41) Matsuoka, K.; Rie, E.; Yui, S.; Honda, C.; Endo, K., Competitive solubilization of cholesterol and beta-sitosterol with changing biliary lipid compositions in model intestinal solution. Chem. Phys. Lipids 2012, 165 (1), 7-14. (42) Romer, S.; Garti, N., The activity and absorption relationship of cholesterol and phytosterols. Colloids Surf., A 2006, 282, 435-456. (43) Hunter, J. E.; Zhang, J.; Kris-Etherton, P. M., Cardiovascular disease risk of dietary stearic acid compared with trans, other saturated, and unsaturated fatty acids: a systematic review. American Journal of Clinical Nutrition 2010, 91 (1), 46-63. (44) Brito, R. M. M.; Vaz, W. L. C., Determination of the Critical Micelle Concentration of Surfactants Using the Fluorescent-Probe N-Phenyl-1-Naphthylamine. Anal. Biochem. 1986, 152 (2), 250-255. (45) Fontell, K., Micellar behaviour in solutions of bile-acid salts. Colloid & Polymer Science 1971, 246 (1), 614-625. (46) Ijare, O. B.; Somashekar, B. S.; Jadegoud, Y.; Gowda, G. A. N., H-1 and C-13 NMR characterization and stereochemical assignments of bile acids in aqueous media. Lipids 2005, 40 (10), 1031-1041. (47) Jonsson, B.; Lindman, B.; Holmberg, K.; Kronberg, B., Surfactants and Polymers in Aqueous Solution. John Wiley & Sons Ltd.: England, 1998. (48) Mazer, N. A.; Carey, M. C.; Kwasnick, R. F.; Benedek, G. B., Quasi-Elastic LightScattering Studies of Aqueous Biliary Lipid Systems - Size, Shape, and Thermodynamics of Bile-Salt Micelles. Biochem. 1979, 18 (14), 3064-3075. (49) Melo, E.; Freitas, A. A.; Chang, Y. W.; Quina, F. H., On the significance of the solubilization power of detergents. Langmuir 2001, 17 (26), 7980-7981. (50) Rozner, S.; Popov, I.; Uvarov, V.; Aserin, A.; Garti, N., Templated cocrystallization of cholesterol and phytosterols from microemulsions. Journal of Crystal Growth 2009, 311 (16), 4022-4033. 30 ACS Paragon Plus Environment

Page 31 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

(51) Nielsen, P. B.; Mullertz, A.; Norling, T.; Kristensen, H. G., The effect of alphatocopherol on the in vitro solubilisation of lipophilic drugs. Int. J. Pharmaceut. 2001, 222 (2), 217-224. (52) Matsuoka, K.; Nakazawa, T.; Nakamura, A.; Honda, C.; Endo, K.; Tsukada, M., Study of thermodynamic parameters for solubilization of plant sterol and stanol in bile salt micelles. Chem. Phys. Lipids 2008, 154 (2), 87-93. (53) Richieri, G. V.; Ogata, R. T.; Kleinfeld, A. M., A Fluorescently Labeled Intestinal Fatty-Acid Binding-Protein - Interactions with Fatty-Acids and its use in Monitoring Free Fatty-Acids. J. Biol. Chem. 1992, 267 (33), 23495-23501. (54) Ros, E., Intestinal absorption of triglyceride and cholesterol. Dietary and pharmacological inhibition to reduce cardiovascular risk. Atherosclerosis 2000, 151 (2), 357379.


Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 32 of 32

TOC Graphics


Quantification of Cholesterol Solubilized in Dietary Micelles

Recommend Documents