Comparative X-Ray Studies on the Interaction of Carotenoids with a Model Phosphatidylcholine Membrane Mario Suwalskya,*, Paulina Hidalgoa, Kazimierz Strzalkab, and Anna Kostecka-Gugalab a Faculty of Chemical Sciences, University of Concepcio´ n, Casilla 160-C, Concepcion, Chile. Fax: + 56-41 245974. E-mail: [email protected] b Institute of Molecular Biology, Jagiellonian University, Krakow, Poland * Author for correspondence and reprint requests Z. Naturforsch. 57c, 129Ð134 (2002); received September 3/October 5, 2001 Carotenoids, Xanthophylls, Membrane The interaction of structurally different carotenoids with a membrane molecular model was examined by X-ray diffraction. The selected compounds were -carotene, lycopene, lutein, violaxanthin, zeaxanthin, and additionally carotane, a fully saturated derivative of -carotene. They present similarities and differences in their rigidity, the presence of terminal ionone rings and hydroxy and epoxy groups bound to the rings. The membrane models were multibi- layers of dipalmitoylphosphatidylcholine (DPPC), chosen for this investigation because the 3 nm thickness of the hydrophobic core of its bilayer coincides with the thickness of the hydrophobic core of thylakoid membranes and the length of the carotenoid molecules. Re- sults indicate that the six compounds induced different types and degrees of structural per- turbations to DPPC bilayers in aqueous media. They were interpreted in terms of the molecu- lar characteristics of DPPC and the carotenoids. Lycopene and violaxanthin induced the highest structural damage to the acyl chain and polar headgroup regions of DPPC bilayers, respectively. Introduction of dipalmitoylphosphatidylcholine (DPPC). The According to the current view, in photosynthetic carotenoids were i) -carotene; ii) its fully satu- membranes physiologically active carotenoids are rated form carotane; iii) lycopene, which lack the terminal rings; iv) lutein, with one hydroxy group functionally attached to proteins. On the other ε hand, certain carotenoids are found in the lipidic in each of the - and -rings; v) violaxanthin, con- moiety of membranes such as that of the retina, taining one hydroxy and one epoxy group in each terminal -ionone ring, and vi) zeaxanthin with bacteria and also in photosynthetic membranes (Gruszecki, 1999). Among the most commonly oc- one hydroxy group in each -ionone ring. Their curring carotenoids three main groups differing in structural formulae are presented in Figures 1 and structure can be distinguished: cyclic, linear and 2. These studies were performed by X-ray diffrac- xanthophylls containing oxygen in the form of hy- tion using a range of concentrations of each carot- droxy or epoxy groups. These structurally different enoid. groups have one feature in common: the conju- gated double bond system that gives rise to rigid, Materials and Methods rod-like structures. This property is lost upon hy- Synthetic dipalmitoylphosphatidylcholine drogenation; for instance, carotane, a fully satu- (DPPC) (lot 45H8383, 99+% purity, MW 734) and rated derivative of -carotene is structurally sim- -carotene (lot 72H0060, MW 537) were from ilar to -carotene but is a flexible molecule. Sigma (St. Louis, MO, USA); zeaxanthin (lot The aim of the present work was to obtain infor- 706007, MW 568) and carotane (MW 558) were a mation about the significance of the different gift from Hoffmann La Roche (Basle, Switzer- chemical elements present in these three carot- land); Violaxanthin (MW 600) and lutein (MW enoid groups when they interact with biomem- 568) were isolated from daffodil (Narcissus poet- branes by comparing the structural perturbations icus) petals by extraction with acetone, cold sapon- induced by different carotenoids to multibilayers ification followed by column chromatography on 0939Ð5075/2002/0100Ð0129 $ 06.00 ” 2002 Verlag der Zeitschrift für Naturforschung, Tübingen · www.znaturforsch.com · D 130 M. Suwalsky et al. · Carotenoids in Membranes alkalized silica gel F254 (Merck) in hexane:ace- tating devices. The blanks, constituted by pure tone 4:1 v/v. Lycopene was isolated from tomato samples of DPPC, were recrystallized as described paste by extraction with chloroform:methanol 1:1, above and X-ray diffracted in Debye-Scherrer v/v, then water was added in a stepwise manner (Philips, The Netherlands) and flat-plate cameras until a good phase separation was obtained; the (built by us). Specimen-to film distances were 8 tomato pigments from the chloroform fraction and 14 cm, standardized by sprinkling calcite pow- were separated by chromatography on MgO in der on the capillary surface. Ni-filtered CuKα radi- hexane:benzene 9:1 v/v. DPPC was used without ation from a Philips PW 1140 X-ray generator was further purification; the carotenoids, except caro- used. The relative reflection intensities on film tane were purified using HPLC performed on re- were measured by peak integration using a Bio- verse Apex Prepsil SIN phase column (Jones Rad GS-700 densitometer (Richmond, CA, USA) Chromatography, Mid Glamorgan, U. K.), using a and Molecular Analyst/PC image software; no cor- Jasco PU 980 isocratic pump and PU 970 UV/Vis rection factors were applied. The experiments detector (Jasco Corp., Tokyo, Japan) set on were performed in the dark at 17 ð 2 ∞C. 440 nm. HPLC solvents were from Lab-Scan (An- alytical Sciences, Dublin, Ireland); others were of Results the highest purity available. All carotenoids, ex- cept carotane, were purified using the following The molecular interactions of DPPC multibi- eluents: methanol:ethyl acetate 17:8 v/v for -caro- layers with -carotene, carotane, lycopene, lutein, tene, acetonitrile:hexane:methanol 5:2:2 v/v/v for violaxanthin and zeaxanthin were determined by lycopene, acetonitrile:methanol: deionized water X-ray diffraction in an aqueous media. Fig. 1 72:8:1 v/v/v for zeaxanthin and lutein, and acetoni- shows a comparison of the diffraction pattern of trile:methanol:deionized water 72:8:8 v/v/v for vio- DPPC and of its mixtures with -carotene (ex- laxanthin, To check the purity of the carotenoids pressed as moles of carotenoid per 100 moles of aliquots were separated in a Nucleosil 100 C18 an- DPPC), its fully saturated derivative carotane and alytical column (Teknokroma, Barcelona, Spain) the linear carotenoid lycopene. As expected, water with the detector set on 250 nm, and absorption altered the structure of DPPC: its bilayer width spectra in the range of 300Ð550 nm were recorded increased from about 5.8 nm in its dry form to using a SLM DW 2000 Aminco spectrophotometer 6.46 nm when immersed in water, and its reflec- (Urbana, IL, USA). Carotane purity, checked by tions were reduced to only the first three, which thin layer chromatography on silica gel (Merck, are related to the bilayer width. On the other Darmstadt, Germany) in acetone:benzene:water hand, a new and strong reflection of 0.42 nm 81:30:4 v/v/v), revealed no impurities. The final ca- showed up, whose appearance was indicative of rotenoid purity was estimated to be between 90 the fluid state reached by DPPC bilayers and cor- and 98%. The carotenoids were dark stored at responded to the average distance between DPPC Ð35 ∞C in a nitrogen atmosphere. fully extended acyl chains organized with rota- For the X-ray experiments about 3 mg of DPPC tional disorder in hexagonal packing. It can be ob- were mixed with the corresponding amount of served that 3% -carotene induced a drastic re- each carotenoid (except carotane), dissolved in duction of DPPC low angle reflection intensities, CHCl3 and left to evaporate the solvent for about which were further reduced with increasing carot- 24 h in the dark in a nitrogen atmosphere; each enoid concentrations. This result implies a struc- dry sample was placed in a 2 mm diameter special tural perturbation of the polarhead region of the glass capillary and 200 µl of bidistilled water were lipid. On the other hand, no significant changes added, into which nitrogen had previously been were noticed in the 0.42 nm reflection, which streamed. In the case of carotane, which is a vis- means that -carotene did not affect DPPC hy- cous liquid at room temperature, the adequate drocarbon chain arrangement. Carotane (5% and amounts were first dissolved in CHCl3 and then 10%) induced a marked decrease of the low angle added to DPPC. The samples thus prepared were reflection intensities, while the intensity of the X-ray diffracted in flat-plate cameras with 0.25- 0.42 nm reflection remained practically un- mm-diameter glass collimators provided with ro- changed. However, a 7.5% concentration pro- M. Suwalsky et al. · Carotenoids in Membranes 131 Fig. 1. Microdensitograms from X-ray diagrams of dipalmitoylphosphatidylcholine (DPPC) with water and the carot- enoids -carotene, carotane and lycopene; (a) low-angle and (b) high-angle reflections. Insets: structures of the corresponding carotenoids. duced an increase of all the reflection intensities, pounds, lycopene induced the highest disorder in particularly that of the 0.42 nm reflection, which the acyl chain region. The results obtained from can be interpreted as an ordering effect of DPPC the interaction of the three xanthophylls (lutein, acyl chains. The results of the interaction of lyco- zeaxanthin and violaxanthin) are presented in pene with DPPC are also exhibited in Fig. 1. It Fig. 2. It may be observed that exposure to lutein may be noticed that this carotenoid (2.5%) had a gives results quite different from those exhibited moderate effect on the low angle reflections but by the action of lycopene. 2.5% lutein induced a ensured complete disappearance of the 0.42 nm significant decrease of the low angle reflection in- reflection, a result that implies the total disorder tensities and a moderate increase of the intensity of DPPC hydrophobic phase. Increasing lycopene of the 0.42 nm reflection. A lutein increase to 5% concentration to 5% resulted in a further fainting produced a slight decrease of DPPC reflection in- of the low angle reflections arising from the lipid tensities.