HAL Id: hal-00929674https://hal.archives-ouvertes.fr/hal-00929674

Submitted on 1 Jan 1999

HAL is a multi-disciplinary open accessarchive for the deposit and dissemination of sci-entific research documents, whether they are pub-lished or not. The documents may come fromteaching and research institutions in France orabroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire HAL, estdestinée au dépôt et à la diffusion de documentsscientifiques de niveau recherche, publiés ou non,émanant des établissements d’enseignement et derecherche français ou étrangers, des laboratoirespublics ou privés.

Multiple fluorescence labelling of proteins, lipids andwhey in dairy products using confocal microscopy

Sophie Herbert, Brigitte Bouchet, Alain Riaublanc, Eric Dufour, Daniel J.Gallant

To cite this version:Sophie Herbert, Brigitte Bouchet, Alain Riaublanc, Eric Dufour, Daniel J. Gallant. Multiple fluo-rescence labelling of proteins, lipids and whey in dairy products using confocal microscopy. Le Lait,INRA Editions, 1999, 79 (6), pp.567-575. �hal-00929674�

Lait (1999) 79, 567-575© InralElsevier, Paris

567

Original article

Multiple fluorescence labelling of proteins, lipidsand whey in dairy products using confocal microscopy

Sophie Herbert-", Brigitte Bouchet", Alain Riaublanc'',Éric Dufour-v, Daniel J. Gallant'"

a Unité d'étude des interactions des molécules alimentaires, Inra,BP 71627, 44316 Nantes cedex 3, France

b Unité de recherches sur les polysaccharides: leurs organisations et interactions, Inra,BP 71627, 44316 Nantes cedex 3, France

c Département qualité et économie alimentaires, Enita de Clermont-Ferrand,63370 Lempdes, France

(Received 24 March; accepted 22 July 1999)

Abstract - Texture optimisation of dairy products is a major aim for manufacturers. A betterknowledge of the structure and spatial organisation of, their main components would allow the opti-misation of their texture. In this study, using confocal scanning laser microscopy, a multiple fluorescentlabelling of proteins, lipids and whey was developed to visualise these main components simultaneouslyin dairy products. Different extrinsic fluorescent probes were tested by confocal microscopy andfluorescence spectroscopy. Fuchsin acid, Bodipy" 665/676 and DM-NERF were selected to label pro-teins, lipids and whey, respectively. Methods for selecting stable and specifie fluorescent probesand for obtaining the multiple fluorescent labelling are presented. An application example on a dairygel is also shown. © Inra/Elsevier, Paris.

microscopy / fluorescence / protein /lipid / curd

Résumé - Colocalisation des protéines, des lipides et du lactosérum dans les produits laitierspar microscopie confocale. L'optimisation de la texture des produits laitiers est un enjeu majeur pourles industries. Une meilleure connaissance de la structure et de l'organisation spatiale des princi-paux constituants de ces produits permettrait une optimisation de leur texture. Dans cette étude, réa-lisée à l'aide d'un microscope confocal à balayage laser, un marquage fluorescent spécifique desprotéines, des lipides et du lactosérum a été développé pour localiser simultanément ces principauxcomposés dans des produits laitiers. Différentes sondes extrinsèques ont été testées par microscopieconfocale et spectroscopie de fluorescence. L'acide fuchsin,le Bodipy" 665/676 et le DM-NERF ontété sélectionnés pour marquer respectivement les protéines, les lipides et le lactosérum. La démarcheutilisée pour sélectionner les différentes sondes stables et spécifiques et pour réaliser la colocalisa-tion sont détaillées. Un exemple d'application sur un gel laitier est également présenté. © InralElsevier, Paris.

microscopie / fluorescence / protéine / lipide / caillé

* Correspondence and reprints. gallant@nantes.inra.fr

568 S. Herbert et al.

1. INTRODUCTION

Texture optimisation of dairy productsis a major aim for dairy manufacturers. Thisoptimisation implies a better knowledge ofthe structure of the se products. Lightmicroscopy is a well-developed techniquefor studying the microstructure of food sys-tems [8, 21]. However, this technique can-not be very weIl applied to the study of thicksamples. Indeed, the large field depth of thelight microscope produces images contain-ing superimposed information from differ-ent focal planes and diffracted light fromareas which are out of focus. Thus, it pre-vents the recording of images with high res-olution in thick specimens [15,16,20]. Onthe other hand, confocal microscopy, mainlyused in the biomedical fields, presents animportant potentialfor the microstructurecharacterization of food products [1, 3, 11].Compared to conventional fluorescencemicroscopy, confocal microscopy allowsthe recording of optical sections preventingthe out-of-focus radiation. It provides a bet-ter signal-to-noise ratio and consequentlyan improved resolution although the confo-cal microscope is also limited by both laserexcitation wavelengths, and by transparencyof the sample [5, 15, 16,20]. In addition tothe possibility of 3-D reconstruction, mul-tiple images of the same field can beobtained showing the spatial distribution ofdifferent chemical components, which canthen be superimposed in order to visualisetheir relative distributions.

Although the multiple labelling for thestudy of the microstructure and the spatialorganisation of different components in aproduct had some interest, this method ishowever seldom applied in the food field.In many studies on the structure of dairyproducts using fluorescence microscopy,only one of the main components, eitherproteins or lipids, is labelled [3, II, 19]. Itappears interesting to localise simultane-ously proteins, lipids and whey. Indeed,when only proteins are labelled, it is not

possible to understand if the pores are filledby lipids, whey or air. A double labellingof proteins and lipids using fluorescein isoth-iocyanate (FITC) and Nile Red, respectively,was reported in the literature [1]. However,fluorescence emissions of Nile Red andFITC cannot be completely separated and,consequently, an artefactuallabelling of pro-teins is observed in the lipid fluorescenceimage. Moreover, FITC rapidly undergoes aphotobleaching, consisting of an irreversibledestruction of the excited fluorophore underhigh-intensity illumination conditions [10].This phenomenon prevents the study of pro-tein network structure for long illuminationexposures, such as kinetic studies. As a con-sequence, the realisation of a multiplelabelling and the recording of good imagesrequire that each of the fluorescent probesused presents a strict specificity for one corn-ponent, does not undergo photobleachingand presents emission fluorescence spectrawith a minimum of overlapping in order torecover only the fluorescence emission ofeach probe.

Thus, the main purpose of this study wasto develop a multiple fluorescent labelling ofproteins, lipids and whey in order to visu-alise simultaneously the different compo-nents of dairy products and to follow theirevolution during milk coagulation kinetics.In the present paper, the methodology forobtaining the multiple labelling of proteins,lipids and whey on a model dairy gel is pre-sented.

2. MATERIALS AND METRODS

2.1. Coagulation system

Raw bovine milk was purchased from a localdairy plant. Before the coagulation kinetics, themilk pH was adjusted to 6.7 using Hel 1N. Milkcoagulation was performed by progressive acid-ification with glucono-ô-lactonc (Roquette,Lestrem, France) added at the concentration of1.75 g-L-1 The coagulation kinetics were realisedat 30 oc.

Confocal microscopy of dairy gels

Coagulations were performed inside cavityslides (SOOum). The slides were submitted toslow rotary movement during the coagulationkinetics to avoid the creaming of the fat glob-ules. For ail the experiments, the extrinsic fluo-rescent probes were added in the milk prior tothe coagulation.

2.2. Fluorochromes and stocksolutions preparation

BODIPY® 665/676 (=(E,E)-3,5-bis-(4-phenyl-l,3-butadienyl)-4,4-difluoro-4-bora-3a-diaza-s-indacene), DM-NERF and 9, lO-dipheny-lanthracene were from Molecular probes, Inc(Eugene, OR, USA). 1-anilino-S-naphthalenesulfonic acid (ANS), and diphenylhexatriene(DPH) were provided by Sigma (Saint-Quentin-Fallavier, France). Fuchsin acid was from Merck(Chelles, France).

The solubilization conditions and the con-centrations of the stock solutions are reported intable J.

569

2.3. Confocallaser scanningmicroscopy

The multiple fluorescent labelling methodwas developed using a Zeiss LSM 410 Confo-cal microscope (Zeiss, Le Pecq, France) used inepi mode and fitted with the following four lasers:UV (363.S nrn), Blue (488 nrn), Green (543 nm)and Red (633 nm) and mounted with 9 differentfilters, either Long Pass filter (LP) or Band Passfilter (BP) (figure J). The BP filters allowed torecover fluorescence emission between 2 wave-lengths and were used to ensure spectral dis-crimination.

2.3.1. Utilisation conditionsof jluorophores for confocalmicroscopy

Ali the extrinsic fluorescent probes were sep-arately screened by confocal microscopy to elim-inate the non specifie fluorochromes and thosewhich were sensitive to photobleaching.

The final concentrations of fluorophores inmilk, excitation lasers and filters were tested. In

Table I. Solubilisation conditions and concentrations of stock solutions of each tested probe andtheir utilisation conditions for confocal microscopy.Tableau I. Conditions de solubilisation et concentrations des solutions mères pour chaque sondetestée et leurs conditions d'utilisation pour la microscopie confocale.

Tested Probes Stock solutions Final concentration Lasers Filtersinmilk

Protein probesANS 10 mrnol-L:' in ethanol Iûumol-L" UV BP 450--490 nmFuchsin acid 100 mrnol-L"! in water 0.4 mmol-L:' Green BP 575-640 nm

Lipidic Probes9, 10-diphenylanthracene 5 mmol-L:' in acetonitrile 5/lmol-L-1 UV BP 400--435 nmDiphenylhexatriene 5 mmol-L:' in acetonitrile 5/lmol·L-1 UV BP 450--490 nmBODIPY® 665/676 65Ilmol·L-1 in chloroform, 6.5Ilmol·L-1 Red LP665 nm

methanol and ethanol(1 VII VII V)

Whey probeDM-NERF 1 mmol·L-1 in phosphate 5Ilmol·L-1 Blue BP 515-565 nm

buffer (5 mrnol-L -1, pH = 8)

570 S. Herbert et al.

LP397

1

BP 400·435--BP 450-490--uv

lP515360.8

1

BP51~525-B~

BkI. lP570488

1 Be 575-640

Green lP66554.'l ::1

350 400 450 550 600 650500

Longueurs d'onde (nm)

table l, are presented the optimal utilisation con-ditions of each tested probe for confocalmicroscopy. It is noticed that the fluorescenceintensity of Fuchsin acid is sensitive to pH, con-sequently the optimal concentration of this probein milk depends on sample pH.

2.3.2. Realisation of multiplefluorescent labelling

To perform the multiple fluorescent labelling,each probe is successively excited by a laser, inthe same field, and their fluorescence emission isdetected with specifie filters allowing to localiseeach of the three main components on separateimages. To minimise photobleaching by highenergy radiation, the sample is first excited bythe red laser, then by the green laser and finallyby the blue laser. To display simultaneously thethree components on the same image, the fluo-rescence images of each component (grey lev-els) are transferred in different channels calledRed, Green and Blue channel s, attributing foreach component a false colour (red, green andblue, respectively). These images are superim-posed in a red/green/blue (RGB) channel yield-ing a single image with each labelled compounddisplayed by the false col our previously deter-mined.

2.4. Recording of excitationand emission spectra by frontface fluorescence spectroscopy

Fluorescence spectra were recorded using aSLM 4800 spectrofluorimeter (Bioriteeh, Chama-rande, France) provided with a thermostatedfront-surface accessory. Considering the cuvette

700 750 800

Figure 1. Lasers and filtersfitted on ZEISS LSM 410confocal microscope. BandPass filter (BP) and LongPass filter (LP).Figure 1. Lasers et filtresdisponibles sur le micro-scope confocal ZEISS LSM410. Filtre à bande passante(BP) et filtre passe haut(LP).

holder, the incidence angle of the excitation radi-ation was 60°. Ali spectra were corrected forinstrumental distortions in excitation using a rho-damine cell in reference channel.

It is weil known that the fluorescence prop-erties of molecules depend on their environment.Consequently, for each selected probe by con-focal microscopy, the excitation and emissionspectra were recorded in the milk in order toselect the appropriate filters to specifically detectthe fluorescence emission of these probes.DM-NERF, Fuchsin acid and Bodipy" 665/676were added to the milk at a final concentration of2.5Ilmol·L-l, 0.5 mmol-L:' and 3.25Ilmol·L-1,respectively. The excitation spectra of DM-NERF(420-530 nm), Fuchsin acid (445-555 nm) andBodipy" 665/676 (600-685 nm) were recordedwith emission wavelengths set at 534, 620 and705 nm, respectively and the emission spectraof DM-NERF (515-650 nm), Fuchsin acid(565-650 nm) and Bodipy®665/676 (670-720 nm)were recorded with excitation wavelengths setat 509, 510 and 633 nm, respectively.

3. RESULTS AND DISCUSSION

To perform a multiple labelling, intrin-sic or extrinsic fluorescent probes must beused to discriminate the different compo-nents. In milk, the tryptophan residues andvitamin Amay be considered as the mostappropriate intrinsic fluorescent probes,which are specifie of proteins and lipids,respectively [6]. However, theses' probescannot be excited by lasers installed on theconfocal microscope because the lowestexcitation ray of lasers is located at

Confoeal microseopy of dairy gels

363.8 nm, while tryptophan excitation spec-trum spreads between 250 and 310 nm andvitamin A between 270 and 350 nm. Con-sequently, extrinsic fluorescent probes mustbe used to observe proteins, lipids and wheyin milk.

3.1. Selection of fluorescent extrinsicprobes using confocal microscopy

As described previously, to perform amultiple fluorescent labelling, fluorophoresmust specifically label the different com-ponents and, as far as possible, without pho-tobleaching. However, most fluorophoresare used in a simple medium (for spectro-scopie studies) and under illumination con-ditions of low intensity. Thus, a first selec-tion of probes was made using confocalmicroscopy to eliminate the non specifieprobes and those which were sensitive tophotobleaching.

Six fluorophores known to label eitherproteins or lipids or being used as pH indi-cators and commonly used in spectroscopyand microscopy studies were investigated(table Il). To label protein network, ANS

571

was first tested. This probe is frequentlyused in fluorescence spectroscopy studiesto label proteins [12,13,17,18]. However,ANS photobleached upon the UV lightexposition. Moreover, this probe alsolabelled lipid compounds. Indeed, on thefluorescence image obtained with ANS, thediffuse structures were labelled as weil asthe round structures. In the literature, lipidsare described as round structures (globules),while proteins are reported as diffuse struc-tures. Consequently, this probe is not spe-cific to proteins and it is impossible to dis-criminate the protein network and the fatglobules. Another probe known to label pro-teins [7], Fuchsin acid, was then tested.Fuchsin acid was revealed to be specifie toproteins and underwent no photobleaching.Fuchsin acid was selected for the labelling ofproteins. Although mc is a dye frequentlyused to label proteins in microscopy [l, 3],this probe was not been screened in thisstudy. Indeed, this fluorophore is weIlknown to photobleach, to be excited by theblue and green lasers and to present a largefluorescence emission spectrum yieldingdifficult the specifie detection of fluores-cence emission of another probe [1, 10].

Table II. Fluorescent extrinsic tested probes for labelling proteins, Iipids and whey.Tableau II.Sondes fluorescentes extrinsèques testées pour le marquage des protéines, des lipides etdu lactosérum.

Tested Probes Lasers Observations

Protein probesANSFuehsin acid

UVBlue and

green

Lipidie probes9,IO-diphenylanthraeeneDipheny IhexatrieneBODlPY® 6651676

uvUVRed

WheyprobeDM-NERF Blue Specifie of whey and specifie deteetion

of fluorescence ernission • Seleeted

Not specifie of proteins and photobleaehingSpecifie of proteins and specifie deteetionof fluorescence emission • Seleeted

Specifie of lipids but photobleaehingSpecifie of Iipids but photobleaehingSpecifie of Iipids and specifie deteetionof fluorescence emission • Selected

572 S. Herbert et al.

To label the lipids, very apolar probes,as 9,1O-diphenylanthracene [10] and DPH[4, 14, 18] were tested. These probes, spe-cific for lipids, presented severe photo-bleaching upon UV light excitation. In fact,it appeared that most of the fluorescentprobes photobleached upon exposure to highenergy of UV laser. Consequently, fluo-rophores excited by the UV laser were dis-carded from the screening process. Bodipy'"665/676 that offers an unusual combinationof apolar structure and long-wavelength flu-orescence was then tested [10]. This probewas excited by the red laser, revealed to bea specifie dye for lipids and stable upon illu-mination. Considering these characteristics,Bodipy'" 665/676 was selected for labellingof lipids. In the literature, it appears thatNile Red [l, 3] and Nile Blue [11, 21] arethe fluorophores the most frequently usedin microscopy to stain lipids. But, theseprobes, as well as PITC, can be excited bythe blue and green lasers and present a largeemission spectrum. These fluorescent prop-erties did not match well with the multiplefluorescence labelling approach developedin this study.

To reveal whey, DM-NERF was tested.This probe is usually used for monitoringpH by fluorescence spectroscopy [10]. Itwas excited with the blue laser, labelled onlythe aqueous phase and presented no photo-bleaching. DM-NERF was selected tolocalise the whey.

The three selected probes to specificallylabel proteins, lipids and whey are Fuchsinacid, Bodipyv 665/676 and DM-NERF,respectively.

3.2. Selection of filters for the specifiefluorescence emission detectionof each probe using front facefluorescence spectroscopy

In order to confirm the selection of fluo-rophores by confocal microscopy, the fluo-rescence emission of these probes must bespecifically detected by appropriate filters to

simultaneously display proteins, lipids andwhey. It is well known that the excitationand emission wavelengths of the fluo-rophores as well as their quantum yield,depend on their environment [9]. For thesereasons, it appeared essential to record thefluorescence spectra of the selected probesin raw milk, which was achieved by frontface fluorescence spectroscopy. These datamade it possible to choose the appropriatelasers and filters for confocal microscopy.

The excitation and emission spectra ofDM-NERF, Fuchsin acid and Bodipy'"665/676 are presented infigure 2. For DM-NERF, the excitation and emission spectraspread from 420 to 530 nm and from 520to 650 nm, respectively. In the case ofFuchsin acid, excitation and emission spec-tra spread from 445 to 555 nm and from 565to 650 nm, respectively. Fuchsin acid dis-plays a maximum excitation around 540 nm,in agreement with literature [7], but thisprobe may also be excited by the blue laserat 488 nm (figure 2). Thus, the blue laserexcites both DM-NERF and Fuchsin acid. Itis therefore necessary to specifically recoverthe fluorescence emission of the DM-NERFby a Band Pass filter between 515 and565 nm. Since DM-NERF and Fuchsin aciddisplay distinct emission spectra, it was pos-sible to specifically detect their fluorescenceemission. Fuchsin acid was excited at543 nm by the green laser and fluorescenceemission was detected between 575 and640 nm with a Band Pass fil ter. In the caseof Bodipy'" 665/676, the excitation spec-trum spreads from 600 to 685 nm and theemission spectrum from 670 to 720 nm.Bodipy'" 665/676 was excited at 633 nmwith the red laser and fluorescence emis-sion was detected above 665 nm with aLong Pass filter. Since the fluorescenceemission of the three probes selected forconfocal microscopy can be specificallyrecovered, DM-NERF, Fuchsin acid andBodipy" 665/676 were used to visualisewhey, proteins and lipids, respectively, ina dairy gel.

Confocal microscopy of dairy gels 573

BP515·565 BP575·640 1.1'665- .. ~ ~1.0

:::i-5 3a 3b

.t''<;jc~.5.,.-c 0.5.,.-'".,..0:::1

i:i:

----0.0400 450 500 550 600 650 700 750

Blue Green Red

Wavclcngth (nm)

Figure 2. Excitation and emission spectra for DM-NERF. Fuchsin acid and Bodipyf 665/676 probesand their utilisation conditions for confocal microscopy (lasers and filters). Spectra obtained in themilk using front face fluorescence. (- - -): excitation spectra of DM-NERF (la). Fuchsin acid (2a)and Bodipy" 665/676 (3a); (-) : emission spectra of DM-NERF (lb). Fuchsin acid (2b) andBodipy'P 665/676 (3b).Figure 2. Spectres d'excitation et d'émission du DM-NERF, acide fuchsine et Bodipy" 665/676 etleur condition d'utilisation pour la microscopie confocale (lasers et filtres). Spectres obtenus dans lelait utilisant la spectroscopie de fluorescence frontale. (- - -) : spectres d'excitation du DM-NERF(la), acide Fuchsine (2a) et Bodipy" 665/676 (3a) ; (-) : spectres d'émission du DM-NERF(lb), acide fuchsine (2b) et Bodipyê 665/676 (3b).

3.3. Multiple fluorescent labellingof proteins, Iipids and wheyin a dairy gel

In order to validate the developedmethod, a multiple labelling of proteins,lipids and whey was performed on a dairygel. Images of the fat globules stained withBodipy" 665/676, protein network labelledwith Fuchsin acid and whey stained withDM-NERF were visualised, respectively,in the red channel (figure 3a), in the greenchannel (figure 3b) and in the blue channel(figure 3e). In figure 3d is presented thetriple-Iabelling image allowing to localisesimultaneously the main components of thegel in the RGB channel. On this image, thefat globules appeared in red, the protein net-work appeared in cyan (not in green asfig-ure 3b) and the whey in dark blue. The

labelling of the protein network in cyanresults from casein micelle hydration [2]involving the superposition of green andblue colours.

4. CONCLUSIONS

The multiple labelling of proteins, lipidsand whey allows the study of the microstruc-ture and the simultaneous localisation ofthese components. Moreover, since theimages of each component are recorded sep-arately, a quantification of their texture maybe performed by different texture imageanalysis techniques. This method will beused in further study to monitor the porosityof the protein network during different milkcoagulation processes.

574 S. Herbert et al.

Figure 3. Fluorescence image of fat globules (Red channel) (a), protein network (Green channel) (b),and whey (Blue channel) (c). The three images were superimposed in the red/green/blue channel(RGB) giving the triple-labelling image (d).Figure 3. Image de fluorescence des globules gras (canal rouge) (a), du réseau protéique (canalvert) (b), et du lactosérum (canal bleu) (c). Superposition de ces trois images dans le canalrouge/vert/bleu (RVB) donnant l'image du triple marquage (d).

ACKNOWLEDGEMENTS REFERENCES

This research was financed as part of FAIREuropean program n° CT96-1056. The authorswould like to acknowledge Dr Soulié (Soredab,La-Boissière-École, France) for valuable dis-cussions.

[1] Blonk l.C.G., van Aalst H., Confocal scanninglight microscopy in food research, Food Res.Int. 26 (1993) 297-311.

[2] Bloomfield VA., MOITC.V., Structure of caseinmicelles: physical methods, Neth. Milk Dairyr. 27 (1973) 103-120.

Confocal microscopy of dairy gels

[3] Brooker B.E., Imaging food systems by confo-cal laser scanning microscopy, in: Dickinson E.(Ed.), New physico-chemical techniques for thecharacterizalion of complex food systems,Blackie Academie and Professional, London,1995, pp. 53-68.

[4] Chen L.A., Dale R.E., Roth S., Brands L.,Nanosecond time-dependent fluorescence depo-larizalion of diphenylhexatriene in dimyris-toyllecithin vesicles and the determination ofmicroviscosity, J. Biol. Chem. 252 (1977)2163-2169.

[5] Cheng P.c., Summers R.G., Image contrast inconfocallight microscopy, in: Pawley J.B. (Ed),Handbook of biological confocal microscopy,Plenum Press, New York, USA, 1990,pp. 179-195.

[6] Dufour E., Riaublanc A., Potentiality of spec-troscopic methods for the characterisation ofdairy products. 1. Front-face fluorescence studyof raw, heated and homogenised milks, Lait 77(1997) 657-670.

[7] Fulcher R.G., Irving D.W., de Francisco A, Flu-orescence microscopy: applications in food anal-ysis, in: Munck L. (Ed.), Fluorescence Analysisin Foods, Longman Scientific and Technical,Harlow, UK, 1989, pp. 59-109.

[8] Fulcher R.G., Wong S.L, Inside cereals- a fluo-rescence microchemical view, in: Inglett G.E.,Munck L. (Eds.), Cereals for food and bever-ages, Academie Press, New York, 1980,pp. 1-26.

[9] Greenspan P., Mayer E.P., Flowler S.D., NileRed: a selective fluorescent stain for intracel-luIar Iipid droplets, J. Cell. Biol. III (1985)965-973.

[10] Haugland R.P., Handbook of fluorescent probesand research chemicals, 6th ed., MolecularProbes, Inc., Eugene, USA, 1996.

[II] Heertje L, van der Vlist P., B10nk J.C.G., Hen-drickx H.A.C.M., Brakenhoff G.I., Confocalscanning laser microscopy in food research:sorne observations, Food Microstruct. 6 (1987)115-\20.

575

[12] Iametti S., Negri E., Bonomi F., Giangiacomo R.,A spectrofluorimetric approach to the estima-tion of changes in protein surface hydrophobie-ity during cheese ripening, Neth. Milk DairyJ. 45 (1991) 183-191.

[I3] Laligant A, Dumay E., Valencia C.C., Cuq J.L.,Cheftel J.c., Surface hydrophobicity and aggre-gation of ~-Iactoglobulin heated near neutralpH, 1. Agric. Food Chem. 39 (1991) 2147-2155.

[14] Lakowicz J.R., Thompson R.B., Differentiaipolarized phase fluorometric studies of phos-pholipid bilayers under high hydrostatic pres-sure, Biochim. Biophys. Acta 732 (1983)359-371.

[15] Laurent M., Johannin G., Le Guyader H.,Fleury A, Confocal scanning optical microscopyand three-dimensional imaging, Biol. Cell 76(1992) 113-124.

[16] Laurent M., Johannin G., Gilbert N., Lucas L.,Cassio D., Petit PX., Fleury A., Power and lim-its of laser scanning confocal microscopy, Biol.Cell80 (1994) 229-240.

[17] Matarella N.L., Richarson T., Physico-chemi-cal and functional properties of positivelycharged derivates of ~-Iactoglobulin, J. Agric.Food Chem. 31 (1983) 972-978.

[18] Strasburg G.M., Ludescher R.D., Theory andapplications of fluorescence spectroscopy infood research, Trends Food Sei, Technol. 6(1995) 69-75.

[19] Sutheerawattananonda M., Fulcher R.G., Mar-tin F.B., Bastian E.D., Fluorescence image anal-ysis of process cheese manufactured withtrisodium citrate and sodium chloride, J. DairySei, 80 (1997) 620-627.

[20] Vodovotz Y., Vittadini E., Coupland J.,McClements D.J., Chinachoti P., Bridging thegap: use of confocal microscopy in foodresearch, Food Technol. 50 (6) (1996) 74-82.

[21] Yiu S.H., A fluorescence microscopie study ofcheese, Food Microstruct. 4 (1985) 99-106.