ARTICLE pubs.acs.org/ac
Structural Changes of Carotenoid Astaxanthin in a Single Algal Cell Monitored in Situ by Raman Spectroscopy Agnieszka Kaczor* and Malgorzata Baranska Faculty of Chemistry, Jagiellonian University, Ingardena 3, 30-060 Krakow, Poland
bS Supporting Information ABSTRACT: The changes of structure of astaxanthin (AXT), a superpotent antioxidant, upon thermal stress were investigated in unicellular microalgae Haematococcus pluvialis by measuring Raman spectra in situ and analyzing obtained results with DFT calculations. Although no visual changes are observed in the Haematococcus cells upon heating, discernible changes in Raman spectra occur from 100 °C systematically up to 150 °C. The exponential increase of the Raman shift of the ν CdC band at ca. 1520 cm1 along with the change of the intensity ratio of bands at 1190 and 1160 cm1 is observed, that correlates with the changes predicted by calculations for astaxanthin conformers ordered by decreasing energy. It is assumed that AXT molecules, initially in the form of H-aggregates with the trans conformations of the endrings, interconvert toward more stable gauche forms upon thermal stress of the algae. The applied approach enables one to follow structural changes of the carotenoid upon temperature stress both in a single algal cell and in a multicellular sample in situ. Obtained information might be of use to improve the industrial process of extraction of AXT in its most bioavailable form.
D
ue to a growing market of carotenoid-related nutrients and changes occurring in carotenoids bioavailability upon temperature,1,2 studying of heat-promoted structural modifications of these compounds is of considerable importance. Extraction of carotenoids out of plants and further food processing (blanching, pasteurization, heating, canning) result frequently in their degradation or isomerization to cis isomers. In general, the latter have lower bioavailability and antioxidant capacity compared to the respective trans forms.29 Illustrative examples are the vitamin A precursors: 9-cis and 13-cis-β-carotene, with 38 and 53% activity, respectively, compared to the all trans isomers.1 The β-carotene conversion processes are of importance as kinetic and thermal accessibility of 13-cis and 15-cis forms of β-carotene at 37 °C next to the most stable all-trans isomer was previously indicated10 as well as cis-isomers were found in plants and fruits.2 Additionally, bioavailability and bioconversion of carotenoids is strongly affected by various factors including among others a form of carotenoid, which in turn depends on the nutrient source.1 For instance, Dunaliella, an algae containing approximately equal ratio of 9-cis and all-trans forms produced lower response in human than the synthetic carotene (predominantly in the all trans form).1 Moreover, bioavailability depends strongly on matrix in which carotenoids are incorporated. For instance, absorption from dark-green leafy vegetables (carotenoids in complex with proteins in chloroplasts and within cell structures) is much lower than from orange-yellow vegetables (carotenoids dissolved in oil droplets in chromoplasts).1 Apart from cistrans isomerism and the matrix effect, other factors, such as arrangement of the rings in respect to the chain and stereoisomerism may impact bioavailability in the case of astaxanthin (AXT). r 2011 American Chemical Society
One of the most powerful carotenoid antioxidants is AXT (3,30 -dihydroxy-β,β-carotene-4,40 -dione), the red pigment required in a diet of some crustaceans and salmonoids, synthesized de novo by simple plants, bacteria, fungi, and microalgae.1116 The most potent biomanufacture producing AXT is Haematococcus (Chlorophyceae), a green unicellular algae that can accumulate AXT up to 34% of total dry weight,11,14 presently gathering attention as an alternative, next to the synthetic compound, source of AXT in crustaceans and salmonoids farming. The synthetic AXT consists of 3R,30 R:3R,30 S:3S,30 S forms in the 1:2:1 ratio, while in Haematococcus, AXT, synthesized from lycopene upon stress conditions,11,14,17 exists primarily as a monoestrified 3S,30 S isomer.17 Carotenogenesis is associated with some distinct changes of the cell such as its significant growth and thickening of walls (encystment) and decrease in photosynthesis and cellular metabolism.11,18 AXT is most probably accommodated in lipid globules/vesicles in cytoplasm in line with the significant increase of lipid content11 and degradation of protein18 occurring during encystment, while the site of biosynthesis is not determined up to now.11 Recently, we employed Raman micromapping to study AXT distribution and its structure in a single Haematococcus cell in situ.19 It was shown that the structure of AXT in the cell is rather uniform and differs distinctly from the AXT structure in the synthetic standard due to considerable differences in Raman shifts of the marker bands. Moreover, it was hypothesized that the difference in structure is Received: May 23, 2011 Accepted: September 1, 2011 Published: September 01, 2011 7763
dx.doi.org/10.1021/ac201302f | Anal. Chem. 2011, 83, 7763–7770
Analytical Chemistry related predominantly to the change in the chemical environment of AXT chains due to the interaction with fatty acids.19 The Raman spectra of carotenoid molecule measured directly in a single cell of alga have been reported before.20 With the use of polarized resonance Raman spectroscopy, it was possible to investigate the orientation of carotenoid chains in the eyespot of Chlamydomonas and Euglena.20 However, detection of carotenoid at the subcellular level followed by the observation and interpretation of its structural changes is still an analytical challenge. Our approach allowed not only for obtaining information about the structure of AXT in a Haematococcus single cell in situ but also for monitoring of its modifications upon an external factor (temperature). Haematococcus farming is becoming a more and more popular method of obtaining AXT, and one of the methods of inducing astaxanthin production is temperature increase.21 Additionally, many methods of extraction of the pigment out of microalgae require elevated temperature.2224 Answering the question of how the thermal stress affects the structure of carotenoid in the algae in situ may result in optimization of the extraction process toward production of AXT of higher bioavailability. Previous works demonstrate that dimorphism is present in AXT crystal.25,26 Raman spectra of these two forms differ substantially,25,26 and some parameters influencing possible applications of the compound, e.g., solubility in solvents, depend on a polymorphic form.26 Nevertheless, according to our knowledge, the structure of both polymorphs is not described yet. Information about the structure of the pigment in Haematococcus also seems scarce, and there is lack of data about structural changes of AXT in microalgae upon changes of temperature. In this work, Raman spectroscopy is applied to monitor changes in the AXT structures in the standard as well as in Haematococcus cells in situ upon thermal stress. Unicellular organisms are good models for biological investigations in view of the fact that each of them consists of only one cell and each cell (organism) has the same function, in contrast to multicellular organisms where cells are almost always heterogeneous in function and fate.27 We chose a single-cell approach to avoid misinterpretation of the results due to possible interaction between cells when they occur as a colony. We wanted to be sure that observed structural changes of astaxanthin molecules are only due to thermal stress and are not caused by the contact with other cells and pressure they put. Heat-promoted changes in Raman spectra of AXT combined with quantum-chemical calculations of various AXT models allowed one to rationalize modifications of the structure of the carotenoid upon thermal stress both in the synthetic compound as well as in the studied biological material. According to our best knowledge, there are no reports on the consequences of thermal stress on the structure of a carotenoid studied in a place of its generation in situ. This work, performed inside a single microalgal cell, demonstrates that such study is possible with the application of Raman spectroscopy.
’ EXPERIMENTAL SECTION Astaxanthin (3,30 -dihydroxy-β-carotene-4,40 -dione, CAS: 47261-7) was obtained from Fluka (e98.5%, mp. 220225 °C). Haematococcus pluvialis Flotow, strain number CCALA 883, was provided by Culture Collection of Autotrophic Organisms (CCALA), Institute of Botany, Academy of Sciences of the Czech Republic. The algae were cultured on vertical laboratory shaker
ARTICLE
(speed 130 cpm, ampl 90). Bristol modified Bold (1949) medium was applied. The average air temperature was 20 °C, and light conditions were 12 h/12 h (day/night), at light condition 200 μmol m2 s1 PAR. Cells were measured after settlement in the metal rings without any preparation. The samples exposed on laser light/air were stable up to several days. Raman spectra were recorded on a MultiRAM FT-Raman spectrometer equipped with a 1064 nm laser line, a germanium detector, and the RamanscopeIII microscope module (Bruker). Olympus BX 10 (for AXT standard measurements) or 40 (for cells measurement) objectives were used in micro mode. We have not done any technical modification of the instrument used for measurement of algae; however, due to the use of fibers, the Raman signal from the microscope is very weak, and the measurement requires to be optimized every time, including vigilant focusing at the sample. All spectra were collected in a 504000 cm1 range with 4 cm1 resolution and laser power in the range of 100450 mW (higher laser powers were used for micromeasurements in order to amend for power loss in the optical fiber). A Linkam THMS600 stage, placed in micro- or macrochamber, was used to heat the samples in the temperature range maximally from 150 to 150 °C. Spectra were measured every 5 or 10 °C, respectively. The heating ratio was set on 10° per min. The starting temperatures were room temperature or 150 °C, and five measurement cycles were done both for the AXT standard and for the algae. The temperature stage was scavenged with gaseous nitrogen. This inert environment of the samples decreased the probability of AXT oxidation and other chemical reactions that potentially might occur in Haematococcus upon thermal treatment, although they might not be absolutely excluded. It is, however, necessary to note that no bands other than those related to astaxanthin were observed in the Raman spectra of the algae upon heating. Micromeasurements (Raman measurements done via microscope) were done to investigate a single cell with the parallel observation of a cell’s visual picture. Macromeasurements (done classically) were conducted in order to reassure that obtained single-cell results can be generalized for the statistical number of cells.
’ COMPUTATIONAL The quantum chemical calculations were performed with Gaussian09 (Gaussian, Inc., USA, RevisionA1)28 at the Density Functional Theory (DFT) level. The DFT calculations were carried out with the three-parameter density functional abbreviated as B3LYP, which includes Becke’s gradient exchange correction29 and the Lee, Yang, and Parr correlation functional.30 Two dimensional scan of the potential energy surface was performed by varying independently two dihedral angles defining orientations of the ionone rings in respect to the chain at the B3LYP/6-31G(d,p) level of theory for the AA 3S,30 S optical isomer of AXT. The minima found on the potential energy surface (PES) and additional points predicted by applying some assumptions (described in detail in the Results and Discussion section), were subsequently fully optimized at the B3LYP/6-31+G(d,p) level of theory; vibrational frequencies were calculated for them at the same level of theory, and the nature of the stationary points on the potential energy surface resulting from optimization was determined, confirming that the optimized structures of AXT 7764
dx.doi.org/10.1021/ac201302f |Anal. Chem. 2011, 83, 7763–7770
Analytical Chemistry Scheme 1. Conformations of the Ionone Ringa
ARTICLE
Table 1. Relative Zero-Point Corrected Energies (ΔEcorr, in kJ mol1), Symmetry Point Groups (S), and Values of Torsion Angles (in Degrees) Defining 3S,30 S AXT Conformers Calculated at the B3LYP/6-31+G(d,p) Level of Theorya conformer
a
The position of the chain denoted by the blue ball.
ΔEcorrb
S
C5dC6C7dC8
C23dC24C25dC26
GAGA
0.0
C2
38.0
38.0
GAGA
0.5
C1
39.0
37.9
GAGA
1.3
C2
39.0
39.0
GATA
3.9
C1
38.5
166.3
TAGA
4.5
C1
166.2
39.4
TATA
8.1
C2
166.2
166.2
GAGB GAGB
16.5 17.3
C1 C1
40.7 39.0
38.1 40.4
GBGA
18.0
C1
39.7
38.1
GAGB
18.7
C1
39.7
39.2
GATB
20.5
C1
38.3
155.4
GBTA
20.8
C1
40.6
166.7
TBGA
21.3
C1
155.4
38.9
TAGB
22.1
C1
166.6
39.9
TATB GBGB
24.6 34.2
C1 C2
155.1 40.7
166.0 40.7
GBGB
34.8
C1
39.8
40.2
GBGB
36.4
C2
39.7
39.7
GBTB
37.3
C1
40.4
155.5
TBGB
38.8
C1
155.3
39.7
TBTB
41.3
C2
155.6
155.6
a Subscripts denote conformation of the ring. b Ecorr = 1855.695800 Hartree for conformer GAGA.
Figure 1. B3LYP/6-31G(d,p) potential energy contour map of AXT 3S,30 S AA form as a function of C5dC6C7dC8 and C23dC24 C25dC26 dihedral angles (upper panel) and the numbering scheme of AXT nonhydrogen atoms (lower panel). Energies (in kJ mol1) are relative to the most stable conformer.
conformers correspond to true minimum energy conformations on the PES. Calculated harmonic frequencies were used in the analysis of the experimental spectra. They were scaled down by a factor of 0.97, to account mainly for anharmonicity effects and limitations of the basis set.
’ RESULTS AND DISCUSSION Geometries, Energies, and Theoretical Raman Spectra of Astaxanthin Models. In order to describe conformational space
for a given optical isomer (for instance 3S,30 S), while assuming all-trans conformation of the chain, possible conformations of the ionone rings relative to the chain and conformations of the rings themselves have to be taken into account. In order to include both factors, two-dimensional potential energy scan of the 3S,30 S stereoisomer was performed by varying independently two dihedral angles defining conformations of ionone rings in respect to the chain, namely C5dC6C7dC8 and C23dC24C25dC26
dihedral angles for the chosen, lowest-energy ring geometries, i.e., AA (for definition of ring geometries, refer to Scheme 1, the nomenclature adopted after Hashimoto et al.,31 atom numbering in Figure 1). The results of the scan as the contour map along with the numbering scheme of the nonhydrogen atoms are presented in Figure 1. Nine minima in total and six of them independent were predicted by calculations for the AA 3S,30 S AXT conformers, and the same number of forms exist for BB forms. Twelve “mixed” conformers, i.e., six AB and six BA might be formally expected, but in fact, this number reduces to nine as for the C2 symmetry structures the AB conformer is equal to the respective BA one. In total, 21 3S,30 S conformers were submitted to full geometry optimization and frequency calculations. Zero-point corrected and Gibbs free energies along with the key geometrical parameters characterizing the obtained structures are given in Table 1. The first and second index refer to orientations around C5dC6C7dC8 and C23dC24C25dC26 angles, respectively, and G, G, and T symbols denote gauche (ca. 40°), minus gauche (ca. 40°), and trans (ca. 160° or 160°) conformations, respectively. Subscripts denote conformations of the ionone rings. Calculations predict that the type of the ionone ring is a factor deciding the energy of AXT 3S,30 S isomer, as all AA forms are significantly stabilized in respect to both “mixed” (AB/BA) and BB conformers. It is, at least partially, caused by the orientation of the OH group in the plane of the carbonyl moiety for the A-type ring and, therefore, better conditions to create the intramolecular CdO 3 3 3 HO H-bond (compare Scheme 1). 7765
dx.doi.org/10.1021/ac201302f |Anal. Chem. 2011, 83, 7763–7770
Analytical Chemistry
ARTICLE
Figure 3. Theoretical Raman spectra of conformers of the AA forms of the AXT 3S,30 S stereoisomer (B3LYP/6-31G+(d,p)). Spectra have been broadened by Lorentzian profiles having a fwhm (full width at half maximum) of 5 cm1 and centered at the calculated (scaled) frequencies.
Figure 2. Optimized geometries of conformers of the AXT AA 3S,30 S stereoisomer (B3LYP/6-31G+(d,p)). Numbers in brackets denote zeropoint corrected energy relative to the most stable GG form in kJ mol1. The naming system is given in the text.
For a given type of structure (AA, “mixed”, or BB), (1) gauche conformations are overall stabilized compared to the trans ones and (2) the energy differences between GG, GG, and GG conformers do not differ considerably (the maximal difference between GBGB and GBGB forms is equal to 2.2 kJ mol1). The structures of fully optimized AA conformers are presented in Figure 2. Interestingly, differences in the conformations of the rings practically do not translate to differences in the predicted bond
lengths neither in the chain nor in the ring (maximal changes of the respective rings equal to 0.005 Å) apart from the C5dC6 and C25dC26 bonds that slightly differ in the gauche forms compared to the trans ones (compare Tables S1S2, Supporting Information). Nevertheless, these subtle changes influence the theoretical Raman spectra of conformers by decreasing Raman shift of the bands due to the stretching vibration of the CdC bonds with the change of conformation from gauche to trans (for low-energy AA forms, compare Figure 3; for other 3S,30 S conformers, see Figures S1 and S2 in Supporting Information). Additionally, with the increase of energy of conformers in the AA, BB, or “mixed” group, a nearly linear increase of the intensity of the bands at ca. 1520 cm1 relative to the bands at ca. 1510/ 1502 cm1 (Table S3, Supporting Information) and decrease of the intensity of bands at ca. 1190 cm1 relative to the bands at ca. 1160 cm1 (Table S4, Supporting Information) are predicted. The differences in the theoretical spectra of the 3S,30 S conformers were used to rationalize changes observed in the experimental spectra of the synthetic AXT and Haematococcus pluvialis cells upon temperature stress conditions. Raman Spectra of Synthetic Astaxanthin and Haematococcus pluvialis upon Stress Conditions. Typical spectra obtained for the synthetic AXT and Haematococcus pluvialis samples at room temperature and upon thermal stress are presented in Figures 4 and 5, respectively. It should be mentioned that the only compound possible to observe experimentally in Haematococcus Raman spectra was astaxanthin and no knowledge about other metabolites can be gained from our experiment. Presented spectra show that Raman shift and relative intensities of the marker bands change distinctly upon thermal stress, both for the synthetic AXT and microalgae, although the 7766
dx.doi.org/10.1021/ac201302f |Anal. Chem. 2011, 83, 7763–7770
Analytical Chemistry
ARTICLE
Figure 4. FT-Raman spectra of the synthetic AXT (black line) and Haematococcus pluvialis (red line) at room temperature.
directions of these changes are not uniform for these two cases. As the values of integral intensities are not relevant due to slight changes of the laser focus with the change of temperature, only changes of relative intensities were considered. Additionally, parameters such as width and center of marker bands were investigated (Figures 6 and 7). The general trends in the spectra are as follows. For the synthetic AXT, both marker bands, observed at 1514 cm1 (ν CdC) and 1158 cm1 (ν1 CC) at 150 °C, shift toward lower wavenumbers with the increase of the temperature (Figures 5, 6, and 7). For the microalgae, the marker bands, observed at 1520 and 1159 cm1 at 100 °C, shift in the opposite directions; namely, the band at ca. 1520 cm1 shifts toward higher wavenumbers, and the band at ca. 1157 cm1 shifts toward lower wavenumbers with the increase of the temperature. Additionally, quite distinct and approximately exponential, increase of the 1190(ν2 CC):1160(ν1 CC) cm1 bands’ ratio is observed upon heating of Haematococcus (Figure 8). The proposed rationalization of these findings is presented below. In the crystalline structure of AXT, obtained from pyridine/water, the molecules are connected in the chains by weak CH 3 3 3 O hydrogen bonds. 32 Molecules adopt minus gauche conformations of both end rings in respect to the chain (dihedral angle of 50.4°) and the intramolecular H-bond is formed between the hydroxyl and keto oxygen atoms.32 Previous works demonstrate that dimorphism is present in AXT crystal.25,26 Raman spectra of these two forms differ substantially.25,26 For instance, Raman shift of the most intense bands due to ν CdC and ν1 CC vibrations are 1510/ 1513 cm1 and 1155/1157 cm1 for I/II polymorphs, respectively (1064 nm excitation).25 Guo et al.25 suggest that the structure described by Bartalucci et al.32 refers to the metastable form II. On the basis of the Raman shifts of these marker bands of synthetic AXT, at room temperature, the mixture of forms I and II coexists (it is necessary to note that the same spectrum at room temperature was obtained while heating the sample from 150 °C and if nonthermally treated AXT was measured),
Figure 5. FT-Raman spectra of the synthetic AXT (upper panel) and Haematococcus pluvialis (middle panel) obtained during heating of the sample from 150 to 150 °C: black, red, green and blue line refer to 150, 50, 50, and 150 °C, respectively, compared with theoretical Raman spectra of AXT conformers (color coded, lower panel). All spectra were normalized by the height of the band at ca. 1157 cm1.
while the Raman spectrum at 150 °C corresponds to the metastable form II.25 It means that quick cooling of the AXT sample to 150 °C results in the formation of the metastable GG form II from the stable polymorph I. On the basis of obtained calculations and taking into account assumptions that (1) cooling of the sample should not result in drastic change of the structure of AXT due to steric constrains and (2) the same trend as for 3S,0 3S form is valid also for 3R,0 3R and 3S,0 3R forms, one can state that cooling of the AXT produces the change of conformations of the end rings from GG (predicted Raman shifts of marker bands: 1508/1160 cm1, observed shift at room temperature: 1512/1157 cm1, Raman shifts of form I: 1510/1155 cm125) to GG (predicted Raman shifts of marker bands: 1510/1161 cm1, observed shift 7767
dx.doi.org/10.1021/ac201302f |Anal. Chem. 2011, 83, 7763–7770
Analytical Chemistry
ARTICLE
Figure 6. Change of parameters (FWHM and center) of the marker band at ca. 1510/1520 cm1 for synthetic AXT (a: micro and b: macro) and Haematococcus pluvialis (c: micro and d: macro) samples upon temperature. For micro measurements (a and c), two different measurement series were presented.
Figure 7. Change of parameters (FWHM and center) of the marker band at ca. 1157 cm1 for synthetic AXT (a: micro and b: macro) and Haematococcus pluvialis (c: micro and d: macro) samples upon temperature. For micro measurements (a and c), two different measurement series were presented.
at 150 °C: 1514/1158 cm1, Raman shifts of form II: 1513/ 1157 cm125). Heating of the sample from 150 °C produces reverse changes, i.e., conformational interconversion from GG toward the most stable GG form that dominates at higher temperatures as the observed shifts of marker bands at 150 °C: 1510/1155 cm1 correspond to form I according to Guo’s et al. results.25 The computed vibrational frequencies are a bit overestimated as the Raman shift of the ν CdC band is observed at 1527 cm1 in the AXT monomer.33 The presence of this band at ca. 1512 cm1 in the spectra of crystalline AXT is due to intermolecular interactions in the crystal structure of AXT in agreement with Bartalucci et al.32 The AXT ν CdC band is observed at 1520 cm1 in Raman spectra of Haematococcus obtained at room temperature. The
value of the red-shift of this band if compared to the monomer (ca. 7 cm1) is characteristic for the AXT linked in H-type aggregates.33 Additionally, it is necessary to mention that AXT in the algae, deposited in lipid globules in the cytoplasm,11 is mostly in its ester form. In principal, esterification does not change vibrational frequencies in the chain as predicted by calculations (frequencies of two most intense bands change up to 1 cm1 in monoesters of two most stable conformers, data not shown here). Hence, it might be assumed that AXT esters form H-aggregates in microalgae lipid globules. Cooling of the microalgae to 150 °C produces a slight shift of the ν1 CC band (to 1522 cm1) and further heating of the sample, from ca. 100 °C, produces the exponential increase of the Raman shift of this band (Figure 6) along with the decrease of the wavenumber of the ν1 CC band at ca. 1160 cm1 (Figure 7) and change of 7768
dx.doi.org/10.1021/ac201302f |Anal. Chem. 2011, 83, 7763–7770
Analytical Chemistry
Figure 8. Ratio of Raman intensities of the band at ca. 1190 cm1 to the band at ca. 1160 cm1 as a function of temperature for Haematococcus pluvialis (the red line shows the exponential fit of points: I1190/I1160 = 0.059 exp(T/119.218) + 0.204; N = 31, adj R2 = 0.824).
1190:1160 cm1 bands’ ratio (Figure 8). The changes upon heating in the Raman spectra of Haematococcus correlate with the changes predicted by calculations upon decrease of the energy of conformers (excluding the very subtle ν1 CC shift). One may state that the increased rotational freedom upon thermal treatment allows for loosening bonding in the H-aggregates (note that the Raman shift of the ν CdC band at 150 °C is 1527 cm1, i.e., as in the monomer33) and change of conformations of the end-rings toward the gauche forms, i.e., forms that are most stable at this temperature. It strongly suggests that initial conformation of the AXT molecules in lipid globules is close to TT and that interconversion toward more stable gauche forms occurs in the microalgae upon thermal stress, i.e., while trans conformations are no more stabilized by binding in aggregates. No visual changes are observed in the microalgal cell upon temperature change between 25 and 150 °C (Figure S3, Supporting Information) that shows usefulness of Raman spectroscopy in in situ studies of Haematococcus. Moreover, the trend obtained for single cell micromeasurements is valid for a statistical number of cells as verified by macromeasurements, in which the scattered signal is averaged over several cells.
’ CONCLUSIONS Thermal stress has an important consequence on bioavailability of carotenoids due to structural changes of the molecules occurring upon the increase of temperature.1,2 One of the most promising carotenoids is at the moment astaxanthin, a superpotent antioxidant, whose disuccinate sodium salt derivative under the brand name Cardax is already after preclinical evaluation as a radical scavenger and myocardial salvage agent.34 Besides the potential applications in humans, AXT is necessary in a diet in some crustaceans and salmonoids and naturally produced by some simple organisms, of which particularly important is a unicellular algae Haematococcus pluvialis.1116 This organism is the astaxanthin manufacture on a industrial scale upon stress conditions, for instance, thermal stress.12,14,21 Additionally, although the details of industrial methods of extraction of the pigment out of algae
ARTICLE
are not known, laboratory methods of extraction are quite often associated with elevated temperatures.2224 This study, showing that there is a change of structure occurring in situ upon heating, may have application in designing and modifications of farming conditions and the extraction process toward ones resulting in a more bioavaliable compound. Calculations predicted that, in the considered population of 3S,3 0 S conformers (the isomer that is present in Haematococcus17 ), the conformation of the ionone ring is a factor deciding of the energy with one type of the ring (socalled A) significantly stabilized over the other (B, Scheme 1). Energy is also dependent on the orientation of the ring in respect to the chain with gauche conformations stabilized over the trans ones. For theoretically predicted Raman spectra of 3S,3 0 S conformers, (1) a decrease of the Raman shift of the bands due to the stretching vibration of the CdC bonds, (2) a decrease of the intensity of bands at ca. 1190 cm1 relative to the bands at ca. 1160 cm1 , and (3) nearly linear increase of the intensity of bands at ca. 1520 cm1 relative to the bands at ca. 1502/1510 cm1 with the increase of energy (and change from gauche to trans) are predicted. Effects opposite to changes 1 and 2 were observed upon heating of Haematococcus from 100 °C systematically up to 150 °C. The direction of these changes (exponential increase of the Raman shift of the ν CdC band at ca. 1520 cm1 and increase of the ratio of 1190:1160 cm1 band) demonstrates that trans to gauche conversion of end-ring conformations occurs in AXT molecules upon thermal treatment. It is concluded that AXT in microalgae, initially in the form of H-aggregates with both trans conformations of the end-rings, are released from H-aggregates and interconvert toward more stable gauche forms upon thermal stress. It was confirmed that these results, obtained for a single algal cell, can be generalized for their statistical number. The changes in the Raman spectra upon thermal treatment of synthetic AXT are related to the slight redshift of both marker bands (at ca. 1512 and 1157 cm1) upon increase of the temperature. On the basis of calculated spectra and knowledge of the crystal structure of the metastable polymorph II,32 it is proposed that cooling of the sample to 150 °C causes interconversions of the end-rings from gauche toward minus gauche conformations corresponding to the metastable polymorph II. The effect is reversible, and further increase of the temperature up to 150 °C results in inverse conversion to the stable polymorph I of GG type. It is worth stressing that the obtained data not only resulted in rationalization of the effect of thermal stress on the studied samples but also allowed one to reach conclusions about the molecular structure of AXT in the initial (nonthermally treated) system.
’ ASSOCIATED CONTENT
bS
Supporting Information. Additional information as noted in text. This material is available free of charge via the Internet at http://pubs.acs.org.
’ AUTHOR INFORMATION Corresponding Author
*Tel: +48 12 6632064. Fax: +48 12 6340515. E-mail: kaczor@ chemia.uj.edu.pl. 7769
dx.doi.org/10.1021/ac201302f |Anal. Chem. 2011, 83, 7763–7770
Analytical Chemistry
’ ACKNOWLEDGMENT The research was supported by the Polish Ministry of Science and Higher Education (Grant N204311037, 2009-2012). Academic Computer Centre CYFRONET is acknowledged for CPU time. Prof. Katarzyna Turnau is acknowledged for Haematococcus pluvialis samples. ’ REFERENCES (1) Castenmiller, J. J. M.; West, C. E. Annu. Rev. Nutr. 1998, 18, 19. (2) Schieber, A.; Carle, R. Trends Food Sci. Technol. 2005, 16, 416. (3) Bushway, R. J.; Wilson, A. M. Can. Inst. Food. Sci. Technol. 1982, 15, 165. (4) Chen, B. H.; Chen, T. M.; Chien, J. T. J. Agric. Food Chem. 1994, 42, 2391. (5) Chen, B. H.; Peng, H. Y.; Chen, T. M. J. Agric. Food Chem. 1995, 43, 1912. (6) Khachik, F.; Beecher, G. R.; Whittaker, N. F. J. Agric. Food Chem. 1986, 34, 603. (7) Noguchi, T.; Hayashi, H.; Tasumi, M.; Atkinson, G. H. J. Phys. Chem. 1991, 95, 3167. (8) Saguy, I.; Goldman, M.; Karel, J. J. Food Sci. 1985, 50, 526. (9) Yeum, K. J.; Russell, R. M. Annu. Rev. Nutr. 2002, 22, 483. (10) Doering, W. E.; Kitagawa, T. J. Am. Chem. Soc. 1991, 113, 4288. (11) Boussiba, S. Physiol. Plant 2000, 108, 111. (12) Boussiba, S.; Bing, W.; Yuan, J. P.; Zarka, A.; Chen, F. Biotechnol. Lett. 1999, 21, 601. (13) Fjerdingstad, E.; Fjerdingstad, K. K. E.; Vanggaard, L. Arch. Hydrobiol. 1974, 73, 70. (14) Higuera-Ciapara, I.; Felix-Valenzuela, L.; Goycoolea, F. M. Crit. Rev. Food Sci. Nutr. 2006, 46, 185. (15) Tsubokura, A.; Yoneda, H.; Mizuta, H. Int. J. Syst. Bacteriol. 1999, 49, 277. (16) Yokoyama, A.; Miki, W. FEMS Lett. 1995, 128, 139. (17) Hussein, G.; Sankawa, U.; Goto, H.; Matsumoto, K.; Watanabe, H. J. Nat. Prod. 2006, 69, 443. (18) Kobayashi, M.; Kurimura, Y.; Kakizono, T.; Nishio, N.; Tsuji, Y. J. Ferment. Bioeng. 1997, 84, 94. (19) Kaczor, A.; Turnau, K.; Baranska, M. Analyst 2011, 136, 1109. (20) Kubo, Y.; Ikeda, T.; Yang, S.; M., T. Appl. Spectrosc. 2000, 54, 1114. (21) Lorenz, R. T.; Cysewski, G. R. Trends Biotechnol. 2000, 18, 160. (22) Machmudah, S.; Shotipruk, A.; Goto, M.; Sasaki, M.; Hirose, T. Ind. Eng. Chem. Res. 2006, 45, 3652. (23) Mendez-Pinto, M. M.; Raposo, M. F. J.; Bowen, J.; Young, A. J.; Morais, R. J. Appl. Phycol. 2001, 13, 19. (24) Sarada, R.; Vidhyavathi, R.; Usha, D.; Ravishankar, G. A. J. Agric. Food. Chem. 2006, 54, 7585. (25) Guo, J.; Ulrich, J. Cryst. Res. Technol. 2010, 45, 267. (26) Leigh, S.; Leigh, M. S. L.; Van Hoogevest, P. Crystal forms of astaxanhin, U.S. Patent 7,563,935 B2, July 21, 2009. (27) Spiller, D. G.; Wood, C. D.; Rand, D. A.; White, M. R. H. Nature 2010, 465, 736. (28) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H. P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, J. A., Jr.; Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin, K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.; Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega, N.; Millam, N. J.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.; Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Dapprich, € Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; S.; Daniels, A. D.; Farkas, O.;
ARTICLE
Fox, D. J. GAUSSIAN 09, Revision A1; Gaussian, Inc.: Wallingford CT, 2009. (29) Becke, A. D. J. Chem. Phys. 1993, 98, 5648. (30) Lee, C.; Yang, W.; Parr, R. G. Phys. Rev. B 1988, 37, 785. (31) Hashimoto, H.; Yoda, T.; Kobayashi, T.; Young, A. J. J. Mol. Struct. 2002, 604, 125. (32) Bartalucci, G.; Coppin, J.; Fisher, S.; Hall, G.; Helliwell, J. R.; Helliwell, M.; Liaaen-Jensen, S. Acta Crystallogr., Sect. B: Struct. Sci. 2007, 63, 328. (33) Salares, V. R.; Young, N. M.; Carey, P. R.; Bernstein, H. J. J. Raman Spectrosc. 1977, 6, 282. (34) Sliwka, H.-R.; Partali, V.; Lockwood, S. F. Hydrophilic Carotenoids: Carotenoid Aggregates. In Carotenoids. Physical, Chemical, and Biological Functions and Properties; Landrum, J. T., Ed.; CRC Press Taylor & Francis Group: Boca Raton, FL, 2010; pp 51.
7770
dx.doi.org/10.1021/ac201302f |Anal. Chem. 2011, 83, 7763–7770
