Protein Matrix Involved in the Lipid Retention of Foie Gras during

Article pubs.acs.org/JAFC

Protein Matrix Involved in the Lipid Retention of Foie Gras during Cooking: A Multimodal Hyperspectral Imaging Study Laeẗ itia Théron,†,‡,§ Annie Vénien,∥ Frédéric Jamme,⊥,# Xavier Fernandez,†,‡,§ Frédéric Peyrin,∥ Caroline Molette,‡,†,§ Paul Dumas,⊥ Matthieu Réfrégiers,⊥ and Thierry Astruc*,∥ †

INRA, UMR 1289 Tissus Animaux Nutrition Digestion Ecosystème Métabolisme, F-31326 Castanet-Tolosan, France Université de Toulouse, INPT ENSAT, UMR 1289 Tissus Animaux Nutrition Digestion Ecosystème Métabolisme, ENSAT, F-31326 Castanet-Tolosan, France § INPT ENVT, UMR 1289 Tissus Animaux Nutrition Digestion Ecosystème Métabolisme, F-31076 Toulouse, France ∥ INRA Clermont-Ferrand Theix, UR370 QuaPA, F-63122 Saint-Genès-Champanelle, France ⊥ Synchrotron SOLEIL, BP48, L’Orme des Merisiers, F-91120 Gif-sur-Yvette, France # INRA, UAR1008 CEPIA, rue de la Géraudière, F-44316 Nantes, France ‡

ABSTRACT: Denaturation of the protein matrix during heat treatment of duck foie gras was studied in relationship to the amount of fat loss during cooking. A low fat loss group was compared with a high fat loss group by histochemistry, FT-IR, and synchrotron UV microspectroscopy combination to characterize their protein matrix at different scales. After cooking, the high fat loss group showed higher densification of its matrix, higher ultraviolet tyrosine autofluorescence, and an infrared shift of the amide I band. These results revealed a higher level of protein denaturation and aggregation during cooking in high fat loss than in low fat loss foie gras. In addition, the fluorescence and infrared responses of the raw tissue revealed differences according to the level of fat losses after cooking. These findings highlight the importance of the supramolecular state of the protein matrix in determining the fat loss of foie gras. KEYWORDS: UV microspectroscopy, infrared microspectroscopy, histochemistry, foie gras, heating, fat loss, quality

■

INTRODUCTION In some birds such as palmipeds in which the liver is the main site of de novo lipogenesis, nonpathological steatosis can be triggered by dietary carbohydrates. This ability is used to produce foie gras, a coveted product highly appreciated for its flavor and sensorial qualities. Usually, ducks or geese are reared and overfed in standardized conditions. Immediately after slaughter, the foie gras is removed, trimmed, and seasoned before cooking. However, even under these standardized conditions, quality is uneven. Despite the controlled cooking conditions, fat loss ranges from 2% to 50%, with a marked impact on technological yield and organoleptic qualities. A study suggested that lipid and fatty acid characteristics are not closely related to fat loss during cooking.1 However, a recent ultrastructural analysis showed that lipid staining intensity was linked to fat loss during cooking, thus suggesting a role of the composition of the lipid droplets in this effect.2 The protein component, though minor in amount, seems to play a major role in the molecular mechanisms underlying fat loss during cooking. Enzymes involved in redox processes are over-represented in livers with high fat loss during cooking;3 this finding was interpreted as precocity in the steatosis stage. An extracellular matrix of hepatocytes forms the support tissue of cells. The integrity of this matrix is probably important for the structural support of lipid-filled hepatocytes. A proteomic study performed on duck livers showed that during postmortem storage, the proteolytic pattern was linked to fat loss during cooking.4 A variation in the molecular composition of this matrix could affect its mechanical resistance during cooking © 2014 American Chemical Society

and thereby modulate lipid retention. Cooking causes thermal denaturation of proteins5 with consequent changes in the morphology of the protein components.2 Using electron microscopy, these last authors found a coagulation of protein matrix that was difficult to quantify. New imaging tools allow a quantitative approach to the local chemical changes in composition and molecular structure, offering an opportunity to establish links between such changes and fat loss during cooking. This is the case of microspectroscopy tools consisting of an optical microscope coupled to a spectrometer. The microscope allows one to select an area of interest in the histological tissue section, and the spectrometer allows spectral acquisition in the same area. Results from different spectral acquisition modalities provide additional informations. From a series of spectral acquisition, it is possible to edit spectral maps of the region of interest. The image is multimodal when spectra come from different wavelength ranges such as infrared, visible, or UV. Analysis of infrared spectra yields information on the molecular structure with a spatial resolution of 10 μm.6 Synchrotron deep ultraviolet (DUV) fluorescence spectra provide knowledge of chemical composition with a spatial resolution that may reach 300 nm.7,8 The simultaneous approaches of located microspectroscopy and histochemistry are therefore well-suited to characterize cellular morphology and molecular Received: Revised: Accepted: Published: 5954

March 5, 2014 May 22, 2014 May 25, 2014 May 26, 2014 dx.doi.org/10.1021/jf5009605 | J. Agric. Food Chem. 2014, 62, 5954−5962

Journal of Agricultural and Food Chemistry

Article

Figure 1. Multimodal and multiscale imaging (visible, IR, and UV) of the same raw foie gras tissue section. (A) Histological section of foie gras examined in bright field microscopy. (B) Infrared spectral distribution of the 1658 cm−1 band, assigned to the α-helix structure of proteins. (C) Magnification of the square delimited area of the infrared spectral image in B. (D,E,F,G) Magnification of the square delimited area depicted in C, characterized by different stains and UV spectroscopy imaging. (D) Bright field without stain. (E) Toluidine Blue staining highlighting the proteins in blue. (F) Histofluorescence of the Nile Red probe. Proteins in black; lipids in red. (G) Deep UV spectral image of the 290−375 nm emission band, assigned to the protein. The difference between the spatial resolution of FT-IR microspectroscopy (25 μm; B,C) and UV fluorescence microspectroscopy (1 μm; G) does not allow for a comparison of images at the same intensity. The UV spectrofluorescence imaging confirms protein histological identification without staining.

structures of targeted areas on the same tissue section. A recent study conducted by FT-IR microspectroscopy characterized the effect of heating on the secondary structure of muscle proteins previously identified by immunohistofluorescence.9 The spatial resolution was reduced from about 25 to 6 μm by the use of a synchrotron source. In addition, the combination of FT-IR and UV fluorescence microspectroscopies allowed a fine in situ characterization of liver steatosis in human subjects.10 Considering the fat loss of foie gras, we suspected that slight differences in the molecular structure of the protein matrix could affect its mechanical resistance during cooking and thereby modulate lipid retention. So, our objectives were to characterize the in situ protein molecular structure changes of foie gras during cooking in order to find lipid loss markers and to acquire knowledge on the molecular mechanism involved in fat loss variability. Therefore, we conducted a multimodal imaging study combined with histological and spectroscopy analyses to characterize the thermal denaturation of foie gras

proteins and to elucidate the molecular mechanisms underlying fat loss during cooking.

■

MATERIALS AND METHODS

Animals, Foie Gras Processing, and Sample Collection. Animals and samples were from an experiment that has been described in previous articles.2,4 Briefly, 120 male mule ducks (Cairina moschata × Anas platyrhynchos) were reared with access to free range until age 13 weeks in a poultry house under natural conditions of light and temperature at the Agricultural College of Périgueux (EPLEFPA, 24, France). The ducks were then overfed in individual cages for 12 days by dispensing a soaked corn mixture (42% grain, 58% flour) twice a day. Ducks were slaughtered at the experimental slaughterhouse of the Agricultural College of Périgueux, 10 h after their last meal. The birds were electrically stunned head-only using scissor tongs and bled by ventral cutting of neck vessels within 10 s after the stun. After 5 min of bleeding, the carcasses were scalded and plucked. The experiments described here fully comply with the legislation on research involving animal subjects in the European Communities Council Directive of November 24, 1986 (86/609/EEC). The investigators were 5955

dx.doi.org/10.1021/jf5009605 | J. Agric. Food Chem. 2014, 62, 5954−5962


Article

Figure 2. Cooking effect on foie gras tissue structure. Protein matrix (arrows) revealed by Toluidine Blue staining appeared thin before cooking (A) and thick after cooking (C) owing to coagulated proteins (cp). Lipids were stained with Nile Red fluorescent dye and were surrounded by the matrix, which appeared black before cooking (B) and fluorescent after cooking (D). Before cooking, the lipid droplets (stars) were small and spherical (B); after cooking, they were large and had lost their spherical shape (D). After cooking, gaps (g) are seen on the tissue sections (C and D). were dried and stored for several hours to several days at −20 °C until use. Histochemistry. Slides were immersed in a 1% Toluidine Blue (EMS Company, Hatfield, PA 19440, USA) solution maintained at 4 °C for 10 min to contrast the nonlipidic matrix. Two washes of 1 min and one wash of 2 min were then performed in ultrapure water. After 1 h of drying, the same sections were stained with Nile Red fluorescent dye (Sigma-Aldrich corporation, St Louis, Missouri, USA) to reveal the lipids. Slides were incubated in a Nile Red solution (90 μg/mL in 75% glycerol) for 30 min in the dark at 4 °C. After three washes of 5 min in PBS buffer and one in ultrapure water, the slides were dried and stored at 4 °C in the dark until acquisition. Image Acquisition and Image Analysis. Histological examination and image acquisitions were performed using an Olympus BX61 transmission bright field/fluorescence microscope coupled to a highresolution digital acquisition kit (Olympus DP 71 digital camera and Olympus CellF̂ software). Digital images were successively acquired on the same optical field in bright field and fluorescence mode (550 nm excitation and 570 nm emission). Proteins appeared in blue and lipids fluoresced in orange. Twenty-five images of each acquisition condition were acquired at 200× magnification on the same histological section. For image analysis, the ×200-magnified Toluidine Blue stained section images were quantitatively analyzed with the open source ImageJ image processing software (http://rsbweb.nih.gov/ij) using exactly the same acquisition settings for all images. Each RGB-colored image was separated into its three monochromatic components (i.e., red, green, and blue) to keep only the green component of the image, which provided the best contrast between protein and lipid components. Protein matrix was extracted after thresholding segmentation of the gray levels, and its relative area was estimated by pixel counting of the region relative to the whole field-of-view area. The number of pixels stained was counted for each region of interest, in bright field acquisition mode, before and after cooking, and was expressed as a percentage of stained matrix per field. The protein matrix densification index, defined as the ratio of the percentage of stained matrix per field after to that before cooking, was calculated.

certified by the French government authority to conduct these experiments (agreement no. 31-11.43.501). At the end of the slaughter process, 20 min after bleeding, livers were removed from the carcasses and weighed. About 1 cm3 was excised from each liver and frozen in cooled isopentane (−160 °C) chilled with liquid nitrogen (−196 °C). Samples were stored at −80 °C until use. Livers were chilled on ice for 6 h and trimmed of their main blood vessels. Each foie gras was then transversally divided into three parts including its two lobes. In the middle part of each lobe, a slice of approximately 200 g was excised and placed in a closed glass jar. Salt (12 g/kg) and pepper (2 g/kg) were added, and the jars were left for 1 h in water in an autoclave (“Brouillon Process”, Sainte-Bazeille, France) at 85 °C under a pressure of 0.8 bar. Temperature was controlled in the water and in two control jars equipped with temperature sensors. After 30 min of chilling (circulating cool water), the jars were stored at 4 °C for two months until opened for estimation of technological yield. The jars were opened, and the surface fat exuded during cooking was carefully removed from the liver. The technological yield was evaluated by the expression of fat loss during cooking as a percentage of initial liver weight: technological yield = (cooked liver weight trimmed of all visible fat/raw liver weight) × 100. Among the 120 livers collected, 25 with similar lipid contents following Folch analysis11 were selected for histological analysis, 13 with low cooking fat loss or high technological yield (89.9% ± 0.8%, mean ± SEM) and 12 with high cooking fat loss or low technological yield (68.3% ± 1.3%). Given the 96 h synchrotron beam time allocation, microspectroscopy analyses were performed on eight livers from the 25 selected, four livers with high technological yield (90.8% ± 1.6%) and four livers with low technological yield (67.0% ± 1.1%). About 1 cm3 of cooked foie gras was cryofixed in cooled isopentane chilled with liquid nitrogen in the same way as that for raw foie gras and stored at −80 °C until use. Tissue Section Preparation. Sections (10 μm thick) cut using a cryostat (Microm, HM 560) at −20 °C were collected on glass slides for histology and on infrared transparent BaF2 windows for UV and IR microspectroscopy. Sections for both histology and microspectroscopy 5956



Article

Figure 3. Representative UV fluorescence deconvoluted spectra of duck foie gras. Deconvolution of a UV spectrum of duck foie gras showing the autofluorescence bands assigned to Tyrosine (305 nm), tryptophan (333 nm), and to a lesser degree collagen (405 nm). mid-IR region from 4000 to 800 cm−1. The spectra were saved in absorbance format using OMNIC software (Thermo Scientific). Processing of Spectra. To study the chemical variations of interest, all of the spectra underwent data preprocessing before multivariate analysis. Autofluorescence spectra were spike- and noise-filtered using an in-house program written in MATLAB, version 7.3 (The MathWorks, Natick, MA), and images of each individual fluorescent component were produced. The Unscrambler software (v9.8, Camo Software AS, Norway) was used to perform a baseline adjustment to zero and to apply unit vector normalization. Data processing of IR spectra was performed using the Unscrambler software (v9.8, Camo Software AS, Norway). An extended multiplicative signal correction (EMSC) was used as a flexible method for correcting artifacts in spectra by removing physical light scattering effects from chemical light absorbance effects.12 The reference spectrum representative of the data set was calculated as the average of all spectra. The EMSC model was calculated over the 1700− 1560 cm−1 range. On the basis of the 9-point Savitzky-Golay smoothing and using a third polynomial degree, the second derivative was first applied, and the calculated EMSC model was used to obtain the EMSC corrected spectra. Statistical Analysis. Numerical variables of histology (protein matrix area) were expressed as the mean ± standard error of the mean (SEM). Variance analysis and mean comparisons were performed using one-way variance analysis and the Student−Newman−Keuls test under the general linear model procedure of the Statistical Analysis System software (Statistical Analysis Systems Institute Inc. SAS/STAT User’s Guide, release 6.03, SAS Institute Inc., Cary, NC, 1988).

Lipid proportion was determined on the black and white images of Nile Red fluorescence. The same approach was used to extract lipid droplets. In each field of view, the total area of lipid droplets was estimated by its pixel count. A conversion factor was applied to express the measured surface areas in μm2. The densification index was calculated as the ratio of percentage of matrix stained after/before cooking within each fat loss group. Microspectroscopies Spectra Acquisitions. Synchrotron DUV microspectroscopy was performed on the DISCO beamline of the SOLEIL synchrotron radiation facility (Saint-Aubin, France).7,8 DUV monochromatized light (typically between 270 and 330 nm) was used to excite tissue sections through a 40× ultrafluar immersion objective (Zeiss, Germany). The emission spectra were acquired from 290 to 539 nm and recorded with a 0.1 nm spectral resolution. The fluorescence spectrum from each excited pixel was recorded. Rastering of the sample allows the recording of a chosen map of interest. Mapping of 50 × 50 μm2 was performed with 1 μm spatial resolution and a 3 s acquisition time per spectrum. Among all of the spectra acquired, 100 were selected for their high 335 nm peak intensity, characteristic of the protein target (tryptophan fluorescence). FT-IR spectra acquisitions were performed on a Nicolet iN10 MX microscope (Thermo Scientific), using the internal (globar) infrared source, available at the SMIS beamline (Synchrotron SOLEIL, SaintAubin, France). Spectra were acquired on whole sections of each sample with a spectral resolution of 8 cm−1. All spectra were recorded in transmission mode. The microscope comprises a motorized sample stage and a liquid nitrogen cooled array detector. A 15× Schwarzschild objective (NA = 0.5) was used, and all spectra were recorded in the 5957



Article

Figure 4. Synchrotron deep UV microspectroscopy comparison between raw and cooked foie gras within low and high fat loss groups. Dark blue, raw high fat loss; light blue, raw low fat loss; red, cooked high fat loss; and orange, cooked low fat loss. (A) Score plot of principal component analysis from the 290-510 nm fluorescence spectra. (B) PC2 loadings. (C) Autofluorescence intensity of 305, 355, and 405 nm bands. PCA score plots show that the groups separate along PC2. Compared to raw foie gras, cooked ones show higher fluorescence intensity at 305 and 405 nm wavelengths (assigned to tyrosine and collagen, respectively; p < 0.001) and lower fluorescence at 355 nm (assigned to the hydrophilic shift of tryptophan; p < 0.001). As shown in C, both the high and low fat loss groups are significantly different for raw foie gras at 305 and 405 nm and for cooked ones at 305 and 355 nm. NS: not significant; *, p < 0.05; **, p < 0.01; ***, p < 0.001. PC: principal component. Processed spectra were analyzed by principal component analysis (PCA) using the Unscrambler software (v9.8, Camo Software AS, Norway). PCA was applied as an unsupervised approach to handle this new set of data and reveal variances or combinations of variables among this large multivariate data set. For this study, the infrared and fluorescence spectral domains were in the ranges 1700−1560 cm−1 and 290−530 nm, respectively. The number of possible components was always left sufficiently high (7). After analysis, the family label of each spectrum was revealed, and the first two components were plotted. The mean characteristic spectrum of each group was also plotted to relate the separation to spectral features. Score plots were used to show similarity maps allowing comparison of spectra regardless of sample categories. Loading plots derived from lists and second principal component X-loading plots were used to reveal and identify characteristic vibrational absorption bands.

A deep UV spectral image of the 290−375 nm emission band, which is assigned to proteins, evidences the protein matrix according to histology. Foie Gras Ultrastructure. The histological sample preparation performed in this study was used here on foie gras for the first time. Cryofixation followed by double staining with Toluidine Blue and the Nile Red fluorescent probe allowed the colocation of the lipid droplets and nonlipid matrix, composed mainly of protein. Previous histological studies of foie gras were carried out after fixation with formalin followed by organic solvent dehydration and inclusion in paraffin.13,14 However, in this type of sample preparation, lipid esters are not fixed by formalin but are eluted by an organic solvent during dehydration.15 In our study, cryo-fixation allowed lipid preservation in the tissue section, and the high contrast of the protein matrix facilitated its density measurement by image analysis. The foie gras ultrastructure study included characterizing the nonlipid structure (taken as protein) around the lipid droplets. Before cooking, the protein matrix located around the lipid droplets (Figure 2A) was thin, and lipid droplets appeared

■

RESULTS AND DISCUSSION Complementarities of Imaging Modalities. The complementarity of the different imaging modalities used in this study is illustrated in Figure 1. The infrared maps of the spectral region 1600−1700 cm−1 allowed one to identify the protein matrix on the whole tissue section but with much lower spatial resolution than with fluorescence spectra imaging and histochemistry. 5958



Article

small (3.8 ± 0.7 μm2) and spherical (Figure 2B) as already described for duck foie gras from a semithin section embedded in epoxy resin.2 In fluorescence mode, the protein matrix was not visible, underlining the specificity of Nile Red fluorescent dye for lipids in raw foie gras. After cooking, lipid droplets were larger (9.0 ± 3.6 μm2) and had lost their spherical shape (Figure 2 D), probably because of the merging of droplets under heat treatment, as previously shown by Théron et al.2 The protein matrix appeared thicker than that before cooking (Figure 2C). These findings were consistent with results obtained on cooked meat.16 Heating leads to protein aggregation through protein thermal denaturation5,9,16 and protein oxidation17 that can form disulfide and bityrosine bonds. Such protein aggregations probably explain the protein matrix structural changes. Also, in contrast to the raw foie gras, the protein matrix of cooked foie gras expressed fluorescence because of the affinity of Nile Red dye for protein aggregates.17,18 These authors highlighted in situ protein aggregation on histological sections of cooked meat by detecting the Nile Red fluorescence dye fixed on protein aggregates formed during heating. The protein matrix densification index, defined as the ratio of the percentage of stained matrix per field after to before cooking, was 1.55 ± 0.04 and 1.35 ± 0.02 (p < 0.05) in the high and low fat loss groups, respectively. These results evidence the link between the level of fat loss during cooking and the densification of the protein matrix, i.e., protein denaturation and aggregation. Cooking Effects on the Protein Matrix Fluorescence Spectroscopy Response. The high brightness and high spatial resolution provided by the DISCO synchrotron beamline allowed the protein matrix to be characterized on an ultrastructural scale on raw and cooked foie gras. The UV spectra of duck foie gras (Figure 3) showed a marked peak at 333 nm assigned to the autofluorescence of tryptophan,10,19 with a shoulder at 305 nm characteristic of tyrosine.19 The band at 405 nm that could be assigned to collagen was not as welldefined as in human pathologic fibrosis studies.10 This result is not surprising since foie gras is not the result of a disease, unlike human steatosis, which causes liver fibrosis. Figure 4A,B presents the score and loadings plots of the principal component analysis performed on UV spectra from the four groups: before and after cooking in low and high fat loss groups. The loadings show that cooking was characterized by peaks at 305, 355, and 405 nm. Cooking resulted in an increase in fluorescence intensity (p < 0.001) at 305 nm (tyrosine) and 405 nm (collagen), while fluorescence intensity decreased at 355 nm (p < 0.001); regardless of the fat loss during cooking. The discriminatory effect of cooking at 355 nm could be due to a shift in the tryptophan peak as a consequence of changes in protein hydrophobicity20 or to its partial degradation. Indeed, a decrease in tryptophan was still observed after meat cooking.21 This amino acid can be oxidized in 5- and 7-hydroxytryptophan22 or reacted with other molecules to form heterocyclic aromatic amines (HAAs).23 In Figure 4C, compared to raw foie gras, we observed an increase of fluorescence intensity of tyrosine in cooked foie gras. Thermal denaturation tends to move the hydrophobic amino acids to the external parts of proteins,17,24 and tyrosine can be considered predominantly hydrophobic as its hydroxyl residue is not ionized at the pH of foie gras (about 6).25 The hydrogen bond environment is known to affect the fluorescence intensity of a molecule.26 In the present case, heating led to the cleavage

of many hydrogen bonds, which probably helped to increase the fluorescence yield of tyrosine. The greater accessibility of the excitation radiation to well-exposed tyrosine residues, together with their lower molecular interactions, may explain this increase in tyrosine autofluorescence intensity. In addition, heating causes melting and peroxidation of lipids, which are particularly abundant in foie gras. Aldehyde compounds from lipid peroxidation react with proteins and increase the fluorescence intensity of cooked meat.27 This effect could be transposed to foie gras, where lipid movement during cooking could influence autofluorescence parameters of molecules in their vicinity. Before cooking, compared with high fat loss foie gras, we observed in low fat loss foie gras, a slight but significant increase in the fluorescence intensity at 305 nm assigned to tyrosine (p < 0.05) and at 405 nm that could be assigned to collagen (p < 0.001) (Figure 4 C). These results suggest that tyrosine and collagen autofluorescence of raw foie gras could be used to predict fat loss during cooking. Significant differences were observed after cooking on emission fluorescence between low and high fat loss groups. The low fat loss group showed a decrease in fluorescence at 305 nm (p < 0.001) and an increase at 355 nm (p < 0.001). This finding reflects different changes in the protein matrix structure and/or composition during cooking, related to fat loss level. Cooking Effects on Protein Matrix Infrared Spectroscopy Response. FT-IR microspectroscopy was applied on the tissue sections to characterize the secondary structure of proteins in situ. The global profile obtained from FT-IR spectra (Figure 5) was similar to that observed on human nonalcoholic

Figure 5. Mean infrared spectra of raw and cooked foie gras. The 16001700 cm−1 spectral range, which corresponds to the amide I band of protein, was used to characterize molecular structure. A close-up of this part of the spectrum shows that the cooked samples have a shoulder at 1622 cm−1, assigned to aggregated β-sheet structures.

steatosis liver samples.28 The spectrum shows major peaks at 3000−3060, 2800−3000, 1710−1780, and 1475−1710 cm−1 assigned to C = C, CH3-CH2, C = 0, and amide (I and II) bands of proteins, respectively.28 The 1600−1700 cm−1 spectral range, was used to characterize the molecular structure of proteins.29 A close-up of this part of the spectrum shows a peak at 1658 cm−1 reflecting α-helix structure.29 Cooked samples 5959



Article

Figure 6. Infrared microspectroscopy comparison between raw and cooked foie gras within low and high fat loss groups. Dark blue, raw high fat loss; light blue, raw low fat loss; red, cooked high fat loss; orange, cooked low fat loss. (A) Score plot of principal component analysis from the 1558− 1700 cm−1 band infrared spectra. (B) PC1 loadings. (C) Infrared response of 1658, 1643, and 1622 cm−1 bands. PCA score plots show that the groups separate along PC1. Compared to raw foie gras, cooked ones show lower infrared absorbance intensities at 1658 and 1643 cm−1 (attributed to α-helix and antiparallel β sheets structures, respectively; p < 0.001) and higher response at the 1622 cm−1 band assigned to aggregated β sheet structure (p < 0.001). From the ANOVA analysis, shown in C, the two fat loss groups (high and low) are significantly different for raw foie gras at 1658 cm−1 and for cooked ones at 1658 and 1622 cm−1. NS, not significant; *, p < 0.05; **, p < 0.01; ***, p < 0.001. PC: principal component.

have a shoulder at 1622 cm−1, assigned to aggregated β-sheet structures.29 Multivariate statistical analysis (Figure 6A,B) separated cooked samples from raw samples on the amide 1 band of proteins. This finding results in the changes in the secondary structure of proteins during heating characterized by changes in spectral bands at 1658, 1643, and 1622 cm−1, as shown by PC1 loadings (Figure 6B) and the close-up of the amide 1 band (Figure 5). These bands are assigned to the α-helical structures, pleated (parallel/antiparallel) β-sheet structure, and aggregated β-sheet structure, respectively.29−31 Compared with raw foie gras, after cooking, the peak heights at 1658 and 1643 cm−1 decreased by about 25% and 75%, respectively, while the peak height at 1622 cm−1 was increased by seven times (Figure 6). These results agree with those of Kirschner et al.32 and Astruc et al.9 showing a decrease in α-helix and pleated β-sheet structures and an increase in aggregated β-sheet structures of skeletal muscle proteins subjected to heating. The decrease in native secondary structure of protein (α-helix and pleated β-sheet structures) associated with an increase of aggregated β-sheet structure is an illustration of the thermal denaturation of proteins.

The level of α-helix structure was slightly lower in the raw low fat loss group than in the raw high fat loss group, but the difference was highly significant (p < 0.001; Figure 6C). This finding could be linked to the lower content of heat shock proteins (HSP) in low fat loss foie gras,3 which protects proteins against denaturation. After cooking, the level of α-helix structures was higher, and the aggregated β-sheet structures were lower in the low fat loss group. Thus, the foie gras that was most sensitive to thermal denaturation of proteins also displayed lower lipid retention after cooking. It seems that the thermostability of proteins that maintain the structural integrity of hepatocytes is involved in the retention of lipid droplets during cooking. The variability in protein thermal denaturation between high and low fat loss groups could also come from differences in the proportion of thermostable protein in the foie gras protein matrix. Indeed, FT-IR microspectroscopy does not allow for characterizing a specific single protein but does allow all proteins probed by the beam, i.e., a set of colocalized proteins. However, each protein has a specific thermal stability.5 We have recently shown that myofibrillar proteins of muscle begin to denature at 40 °C, whereas elastin is thermostable up to at least 5960



Article

80 °C.9 In the case of foie gras, the lower thermal denaturation of overall matrix proteins from low fat loss foie gras could be explained by a higher proportion of thermostable proteins. Coupling morphological, multimodal imaging methods and spectroscopies enabled us to characterize the structure of foie gras at mesoscopic, ultrastructural, and supramolecular scales. The coagulation of heated proteins observed by histology is the result of their aggregation at the molecular scale after their thermal denaturation, which was characterized by microspectroscopy. Our results link the evolution of the protein matrix to the level of fat loss during cooking. Each of the three methods (histology, infrared microspectroscopy, and fluorescence microspectroscopy) showed that a reduced thermal change in foie gras proteins promotes better lipid retention and thereby enhances the technological quality of foie gras.

■

(9) Astruc, T.; Peyrin, F.; Venien, A.; Labas, R.; Abrantes, M.; Dumas, P.; Jamme, F. In situ thermal denaturation of myofibre subtype proteins studied by immunohistofluorescence and synchrotron radiation FT-IR microspectroscopy. Food Chem. 2012, 134, 1044− 1051. (10) Petit, V. W.; Refregiers, M.; Guettier, C.; Jamme, F.; Sebanayakam, K.; Brunelle, A.; Laprevote, O.; Dumas, P.; Le Naour, F. Multimodal spectroscopy combining time-of-flight-secondary ion mass spectrometry, synchrotron-FT-IR, and synchrotron-UV microspectroscopies on the same tissue section. Anal. Chem. 2010, 82, 3963−3968. (11) Folch, J.; Lees, M.; Sloane Stanley, G. H. A simple method for the isolation and purification of total lipides from animal tissues. J. Biol. Chem. 1957, 226, 497−509. (12) Kohler, A.; Sule-Suso, J.; Sockalingum, G. D.; Tobin, M.; Bahrami, F.; Yang, Y.; Pijanka, J.; Dumas, P.; Cotte, M.; van Pittius, D. G.; Parkes, G.; Martens, H. Estimating and correcting Mie scattering in synchrotron-based microscopic Fourier transform infrared spectra by extended multiplicative signal correction. Appl. Spectrosc. 2008, 62, 259−266. (13) Bénard, G.; Labie, C. In Evolution histologique du foie des palmipèdes au cours du gavage; 3èmeJournées de la Recherche sur les Palmipèdes à Foie Gras, Bordeaux, France, Oct 27−28, 1998; Bordeaux, ITAVI: Paris, France 1998; pp 31−35. (14) Locsmandi, L.; Hegedus, G.; Andrassy-Baka, G.; Bogenfurst, F.; Romvari, R. Following the goose liver development by means of crosssectional digital imaging, liver histology and blood biochemical parameters. Acta Biol. Hung. 2007, 58, 35−48. (15) Fujimoto, T.; Ohsaki, Y.; Cheng, J.; Suzuki, M.; Shinohara, Y. Lipid droplets: a classic organelle with new outfits. Histochem. Cell Biol. 2008, 130, 263−279. (16) Astruc, T.; Gatellier, P.; Labas, R.; Lhoutellier, V. S.; Marinova, P. Microstructural changes in m. rectus abdominis bovine muscle after heating. Meat Sci. 2010, 85, 743−751. (17) Sante-Lhoutellier, V.; Astruc, T.; Marinova, P.; Greve, E.; Gatellier, P. Effect of meat cooking on physicochemical state and in vitro digestibility of myofibrillar proteins. J. Agric. Food Chem. 2008, 56, 1488−1494. (18) Demeule, B.; Gurny, R.; Arvinte, T. Detection and characterization of protein aggregates by fluorescence microscopy. Int. J. Pharm. 2007, 329, 37−45. (19) Pallu, S.; Rochefort, G. Y.; Jaffre, C.; Refregiers, M.; Maurel, D. B.; Benaitreau, D.; Lespessailles, E.; Jamme, F.; Chappard, C.; Benhamou, C. L. Synchrotron ultraviolet microspectroscopy on rat cortical bone: involvement of tyrosine and tryptophan in the osteocyte and its environment. PLoS One 2012, 7, 1−11. (20) Lakowicz, J. Principles of Fluorescence Spectroscopy; Kluwer Academic and Plenum Publishers: New York, 1999. (21) Gatellier, P.; Kondjoyan, A.; Portanguen, S.; Greve, E.; Yoon, K.; Sante-Lhoutellier, V. Determination of aromatic amino acid content in cooked meat by derivative spectrophotometry: Implications for nutritional quality of meat. Food Chem. 2009, 114, 1074−1078. (22) Josimovic, L.; Jankovic, I.; Jovanovic, S. V. Radiation-induced decomposition of tryptophan in the presence of oxygen. Radiat. Phys. Chem. 1993, 41, 835−841. (23) Skog, K. I.; Johansson, M. A. E.; Jagerstad, M. I. Carcinogenic heterocyclic amines in model systems and cooked foods: A review on formation, occurrence and intake. Food Chem. Toxicol. 1998, 36, 879− 896. (24) Morita, J. I.; Yasui, T. Involvement of hydrophobic residues in heat-induced gelation of myosin tail subfragments from rabbit skeletalmuscle. Agric. Biol. Chem. Tokyo 1991, 55, 597−599. (25) Petsko, G. A.; Ringe, D. Protein Structure and Function; Lawrence, E., Robertson, M., Eds.; New Science Press Ltd.: London, U.K., 2004. (26) Valeur, B. Molecular Fluorescence: Principles and Applications; Wiley-VCH: Weinheim, Germany, 2001.

AUTHOR INFORMATION

Corresponding Author

*Tel: 33 4 73 62 41 56. Fax: 33 4 73 62 42 68. E-mail: Thierry. [email protected]. Funding

This work was supported by funding from INRA, CIFOG, and Région Midi-Pyrénées. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS Ducks were slaughtered and processed at the facilities of the Lycée Agricole de Périgueux (EPLEFPA, 24, France). We thank François Hérault (EPLEFPA) for technical supervision of the breeding and slaughter process. Microspectroscopy spectra acquisitions were performed at the SOLEIL synchrotron on the DISCO beamline in the framework of Project No. 20100044.

■

REFERENCES

(1) Hermier, D.; Salichon, M. R.; Guy, G.; Peresson, R. Differential channelling of liver lipids in relation to susceptibility to hepatic steatosis in the goose. Poultry Sci. 1999, 78, 1398−1406. (2) Theron, L.; Astruc, T.; Bouillier-Oudot, M.; Molette, C.; Venien, A.; Peyrin, F.; Vitezica, Z. G.; Fernandez, X. The fusion of lipid droplets is involved in fat loss during cooking of duck “foie gras”. Meat Sci. 2011, 89, 377−383. (3) Theron, L.; Fernandez, X.; Marty-Gasset, N.; Pichereaux, C.; Rossignol, M.; Chambon, C.; Viala, D.; Astruc, T.; Molette, C. Identification by proteomic analysis of early post-mortem markers involved in the variability in fat loss during cooking of mule duck “foie gras”. J. Agric. Food Chem. 2011, 59, 12617−12628. (4) Theron, L.; Fernandez, X.; Marty-Gasset, N.; Chambon, C.; Viala, D.; Pichereaux, C.; Rossignol, M.; Astruc, T.; Molette, C. Proteomic analysis of duck fatty liver during post-mortem storage related to the variability of fat loss during cooking of “foie gras”. J. Agric. Food Chem. 2013, 61, 920−930. (5) Tornberg, E. Effects of heat on meat proteins - Implications on structure and quality of meat products. Meat Sci. 2005, 70, 493−508. (6) Lasch, P.; Naumann, D. Spatial resolution in infrared micro spectroscopic imaging of tissues. Biochim. Biophys. Acta, Biomembr. 2006, 1758, 814−829. (7) Giuliani, A.; Jamme, F.; Rouam, V.; Wien, F.; Giorgetta, J. L.; Lagarde, B.; Chubar, O.; Bac, S.; Yao, I.; Rey, S.; Herbeaux, C.; Marlats, J. L.; Zerbib, D.; Polack, F.; Refregiers, M. DISCO: a lowenergy multipurpose beamline at synchrotron SOLEIL. J. Synchrotron Radiat. 2009, 16, 835−841. (8) Jamme, F.; Villette, S.; Giuliani, A.; Rouam, V.; Wien, F.; Lagarde, B.; Refregiers, M. Synchrotron UV fluorescence microscopy uncovers new probes in cells and tissues. Microsc. Microanal. 2010, 16, 507−514. 5961



Article

(27) Gatellier, P.; Sante-Lhoutellier, V.; Portanguen, S.; Kondjoyan, A. Use of meat fluorescence emission as a marker of oxidation promoted by cooking. Meat Sci. 2009, 83, 651−656. (28) Le Naour, F.; Bralet, M. P.; Debois, D.; Sandt, C.; Guettier, C.; Dumas, P.; Brunelle, A.; Laprevote, O. Chemical imaging on liver steatosis using synchrotron infrared and ToF-SIMS microspectroscopies. PLoS One 2009, 4, 1−10. (29) Barth, A. Infrared spectroscopy of proteins. Biochim. Biophys, Acta, Bioenerg. 2007, 1767, 1073−1101. (30) Jackson, M.; Mantsch, H. H. The use and misuse of ftir spectroscopy in the determination of protein-structure. Crit. Rev. Biochem. Mol. 1995, 30, 95−120. (31) Bocker, U.; Ofstad, R.; Wu, Z. Y.; Bertram, H. C.; Sockalingum, G. D.; Manfait, M.; Egelandsdal, B.; Kohler, A. Revealing covariance structures in Fourier transform infrared and Raman microspectroscopy spectra: A study on pork muscle fiber tissue subjected to different processing parameters. Appl. Spectrosc. 2007, 61, 1032−1039. (32) Kirschner, C.; Ofstad, R.; Skarpeid, H. J.; Host, V.; Kohler, A. Monitoring of denaturation processes in aged beef loin by Fourier transform infrared microspectroscopy. J. Agric. Food Chem. 2004, 52, 3920−3929.

5962


Protein Matrix Involved in the Lipid Retention of Foie Gras during

Recommend Documents