Boletín de la Sociedad Chilena de Química
versión impresa ISSN 0366-1644
Bol. Soc. Chil. Quím. v.45 n.3 Concepción set. 2000
SPECTROSCOPIC PROPERTIES OF HYDROPHOBIC
A ONE AND TWO-DIMENSIONAL 1H-NMR AND 13C-NMR STUDY
*A.M. Edwards, A. Saldaño, C. Bueno, E. Silva and S. Alegría
*Depto. de Química Física, Facultad de Química, Pontificia Universidad Católica de Chile,
Casilla 306, Santiago 22, Chile. Teléfono 56-2-6864394, Fax 56-2-6864744,
(Received: January 27, 2000 - Accepted: June 7, 2000)
In memorian of Dr. Guido S. Canessa C.
Spectroscopic characterization of hydrophobic riboflavin esters was performed using IR, UV-visible absorbance and fluorescence emission spectroscopy. An accurately One and Two dimensional 1H-NMR and 13C-NMR study with H-HCOSY and C-HCOSY and double irradiation applications were necessary for a good resolution of complex spectra.
KEYWORDS: flavins, flavin-esters, 1H-NMR, 13C-NMR, HHCOSY, CHCOSY
Se caracterizó espectroscópicamente cuatro ésteres hidrofóbicos de riboflavina, utilizando espectroscopía IR, de absorción UV-visible y de emisión fluorescente. Fue necesario efectuar un estudio RMN-1H y RMN-13C en una y dos dimensiones, utilizando aplicaciones de H-HCOSY y C-HCOSY y doble irradiación para lograr una buena resolución, debido a la complejidad de los espectros.
PALABRAS CLAVES: flavinas, esteres flavínicos, 1H-NMR, 13C-NMR, HHCOSY, CHCOSY
Flavins are important in chemistry and in biology, because of their properties as photosensitizers and/or as redox enzyme cofactors. Riboflavin is one of the components of the B2 vitaminic complex, being the precursor of the flavinic redox cofactors, flavin mononucleotide (FMN) and flavin adenine dinucleotide (FAD). Both are biological electron carriers and can adopt radical forms in vivo1,2.
Many biomolecules do not absorb visible light and need the presence of a photosensitizer to be altered by this radiation. When studying the sensitized photooxidation of aminoacid residues in proteins we found an exceptional efficiency for photooxidation when riboflavin (RF) was used as a sensitizer, comparing with that observed using photosensitizers such as methylene blue (MB) or rose bengal (RB)3. This result was not expected because RF is photolabile4. Studying the photooxidation of the aromatic photooxidable aminoacids, we found a photoinduced binding between RF and Trp. This binding was observed with the free aminoacid and also with different proteins containing Trp, such as Lysozyme, human and bovine serum albumin, a-lactalbumin, etc.5,6,7,8
We have investigated the effect of visible light, sensitized by RF on tumoral human and mouse cells in culture9,10. We found a massive deletereous effect which was higher when Trp was added in adittion to RF, prior to the irradiation. We investigated the effect of the photoproducts of the Trp-RF irradiation, working in the presence and in the absence of molecular oxygen. The effect of the anaerobic photoproducts was higher than that of the oxygenated ones. By transmission electronic microscopy we observed morphological alterations similar to those described for apoptotic cells9,10, cells suffering a kind of programmed death. These results made interesting to study the possibility to apply RF type photosensitizers in cancer therapy, such as the recently developed photodynamic therapy (PDT), based in the accumulation of a sensitizer in the tumoral tissue with a later localized visible light irradiation11.
RF is a hydrophilic vitamin and its normal concentration in tissues is too low for an appreciable photochemical activity, because its biological function requieres only catalytic amounts of it. As a consequence, it is necessary the presence of exogenous introduced flavins. The increase in the hydrophobicity of the sensitizer molecule is known to enhance its affinity to cells and tissues12, therefore, we synthetized a serie of 2,3,45-esters of RF. We have recently reported13, the photochemical and pharmacokinetic properties of these RF esters, and we have demonstrated that the RF esters retain the photosensitizer properties of RF, but they are considerably more photostable than RF. The pharmacokinetic studies demonstrated13, that the more hydrophobic esters, RTB and RTPa showed a high incorporation to mouse tissues, an important condition for their use in different types of therapy.
The aim of the present work is to achieve an accurate characterization of the RF esters, using spectroscopic techniques, with special relevance on One and Two dimensional 1H-NMR and 13C-NMR study.
MATERIALS AND METHODS
RF, acetic, propionic, butyric and palmitic acids and anhydrides were obtained from Sigma Chem.Co. All other reagents were analytical grade.
Preparation of the RF esters
RF 2',3',45'-tetraacetate (RTA), RF tetrapropionate (RTP), RF tetrabutyrate (RTB) and RF tetrapalmitate (RTPa) were synthesized according to the procedure described by Ogasawara et al.14, modified increasing the reaction time and the temperature with the length of the alkyl chain, from 30 min and 40ºC in the original work to 3 hours for RTA up to 2 weeks at 65ºC for RTPa. Briefly, for each ester, RF was added to a beaker containing a 1:1 mixture of the corresponding (acetic, propionic, butyric or palmitic) acid and anhydride. After a dropwise addition of 70% perchloric acid, the mixture was stirred during the adequate time and temperature (see above). The mixture was cooled in an ice bath and diluted with an equal volume of water, and the solution was then extracted three times with chloroform. The combined chloroform extracts were washed four times with deionized water, followed with an extraction with a saturated solution of NaCl. The solution was evaporated to dryness and the product was then recrystallized from 95% ethanol. The reaction yield was 75% for RTA (Calc. for C25H28N4O10: C, 55.15; H, 5.15; N, 10.29. Found: C, 54.88; H, 5.16; N, 10.06 %) ; 70 % for RTP (Calc. for C29H36N4O10: C, 58.00; H, 6.00; N, 9.33. Found: C, 57.49; H, 5.88; N, 9.00 %); 65% for RTB (Calc. for C33H44N4O10: C, 60.37; H, 6.71; N, 8.54. Found: C, 60.32; H, 6.71; N, 8.34%) and for RTPa (Calc. for C81H140N4O10: C, 73.19; H, 10.54; N, 4.22. Found: C, 73.10; H, 10.94; N, 4.27%) the yield was only 1% after 3 fold recristallization. Yagi et al.14 reported a 8% yield with one recristallization step.
Octanol-water distribution ratio
The distribution between n-octanol and water was determined by the procedure described by Kessel12. The compounds (5mM) were partitioned between 1-octanol and 50 mM phosphate buffer pH 7.0. After shaking for 2 min, the phases were separated by centrifugation and the flavin concentration in each phase was determined by absorbance measurements.
Absorption and fluorescence spectra were recorded on a 8453 Hewlett Packard spectrophotometer and on a Perkin-Elmer 650-10S fluorescence spectrometer, respectively. IR spectra were obtained using a Bruker VECTOR 22 IR spectrophotometer. 1H-NMR and 13C-NMR spectra (CDCl3, 200 MHz and 50 MHz respectively) were obtained and processed using a Bruker AC-200P NMR spectrometer.
RESULTS AND DISCUSSION
Four RF esters; RF-2',3',45'-tetraacetate (RTA), tetrapropionate (RTP), tetrabutyrate (RTB) and tetrapalmitate (RTPa) were synthesized and characterized. The structures of the esters are shown in Fig. 1. The procedure described by Ogasawara et al.14 must be modified increasing the reaction time and the temperature with the length of the alkyl chain, from 30 min, and 40ºC in the original work to 3 hours for RTA up to 2 weeks at 65×C for RTPa, and the reaction yield decrease with the increase in the length of the alkyl chain.
|Fig. 1 Chemical structures of riboflavin (RF), riboflavin tetraacetate (RTA), rivoflavin tetrapropionate (RTP), rivoflavin tetrabutyrate (RTB) and rivoflavin tetrapalmitate (RTPa).|
The physical properties of the RF esters are shown in Table I. The melting points of the esters are very similar to those described by Miyamoto et al.15 for RTA, RTP and RTB and by Yagi et al.16 for RTPa.
The values of the octanol-water partition ratio clearly indicate the great change in hydrophobicity12 experimented by RF after esterification, and it also indicates that the hydrophobicity increase with the length of the alkyl chain is important between acetate and butyrate, however the addition of more than three CH2 groups to the chain seems to have a low effect.
The photophysical properties of the flavin esters are shown in Table II. The UV-visible absorption spectrum of RF in methanol consists of four structureless peaks; however, for the esters in chloroform, some fine structure is observed for the long wavelength absorption band (So ® S1 electronic transition), with the partial resolution of three vibrionic transitions as a result of a reduction in solvent-solute interactions. All four absorption maxima possess high molar extintion coefficients ( > 104 M-1cm-1), indicative of p ® p* type transitions.
The esters exhibit the bright yellow fluorescence characteristic of isoalloxazines, assigned to the So ® S1 transition4. The wavelength independence of the fluorescence quantum yield indicates that excitation to upper excited states ( Sn+1) is followed by extremely rapid internal conversion to the S1 level. The spectrum of RF in methanol is similar to that in aqueous solution4, however, in the fluorescence emission spectra of the esters in chloroform, a blue shift is observed together with a partial resolution of some vibration structure , as a consequence of reduction in solvent-solute interactions.
The more characteristics IR bands are shown in Table III. The band assignment was carried out on the basis of data reported for RF in a previous work of our group17. The results clearly confirm the ester formation, with the dissappearence of the OH bands and the formation of the ester C=O bands, and also that the characteristics of the isoalloxazine ring of RF were unaffected.
The assignment of the signals in the 1H-NMR spectra is shown in Table IV. A number of them are coincident with those previously reported by other authors18-21. Grande et al.18,19 have reported the 1H- and 13C-NMR characteristics of isoalloxazines; however, all the compounds had a methyl group attached to N-10 instead of the ribityl group in RF, therefore the spectra were less complex than that of RF. Van Schagen and Müller21 described the natural abundance of reduced isoalloxazines (13C-NMR) using the 2',3',4',5'-tetra-acetylated ester of reduced N-3-methyl-RF. Keller et al.21 analyzed biosynthetically 13C-labeled riboflavin by double quantum and two dimensional NMR. They also studied the tetra-acetylated ester of RF, however, none of these reports made a clear assignment of the H and C signals of either the ribityl chain or the alkyl chain of the esters, and it was not possible to find this information in literature.
Fig. 2 1H-NMR spectra of rivoflavin tetrapropionate (RTP).
Methylene protons on C-5 were clearly identified as the ABX spin system located at 4.2 and 4.4 ppm.
HH Correlation spectroscopys, (HHCOSY) spectra were performed for all compounds to assign protons by observing spin coupling through 2 (2J) and 3 (3J) bonds. Geminal and vicinal interaction between C-1 methylene protons among themselves and that with the C-2methine respectively were not observed. See Fig. 3.
Fig. 3 H-H correlation Spectroscopy (H-HCOSY) of rivoflavin tetrapropianate (RTP).
It was necessary to make a selective homonuclear decoupling to assign these protons correctly in the 1H-NMR spectra. This experiment is shown in Fig. 4. There is a great broadening of the C-1 methylene protons due to the quadrupolar coupling to Nitrogen-10, but it was possible to observe a significative narrowing of the signal in 5.7 ppm when the broad signal in 5.15 ppm was irradiated.
Fig. 4 Homonuclear decoupling experiment in the region of the ribityl protons of rivoflavin tetrapropionate (RTP).
Methine protons on C-3 and C-4 are overlapped, but C4H was easily differenciated by its complex spin pattern due to coupling with its neighbouring protons.
These assignments of proton signals together with the CHCOSY allowed the unequivocal identification of the carbon 13 signals in the spectra of C-2, C-3 and C-4. An expansion of the two dimensional CHCOSY is shown in Fig. 5.
Fig. 5 C-H correlation Spectrocopy (C-HCOSY) of rivoflavin tetrapropionate (RTP).
The C-1 was identified by exclusion, because there was no correlation between this carbon and the attached protons, but it was clasified as a methylene by a DEPT 135 experiment (Fig. 6).
Fig. 6 13C DEPT 135 experiment of rivoflavin tetrapropionate (RTP).
The chemical shifts of the 13C-NMR spectra of RTA and RTP are shown in Table 6. The low solubility of RTB and RTPa avoided to obtain 13C-NMR spectra.
CH Correlation spectroscopy also helped to assigned many of the carbons in the alkyl groups because the proton spectra were already identified by homonuclear correlation spectroscopy.
Its important to observe that the variation in the length of the alkyl group of the ester produce no significant variation in the proton and carbon NMR spectra.
The almost complete assignment of the 1H-NMR and 13C-NMR of these riboflavin esters can be useful to determine the purity and grade of esterification of the ribityl moiety. If a partial esterification occurs, the non esterified location can be identified by simple comparation of their 1H-NMR and 13C-NMR spectra.
This work received financial support from Fondecyt, Grant Nº 2970081 and Grant Nº 1000310 (for the preparation of the samples for elemental analysis).
1. S.T. Kim, P.F. Heelis, A. Sancar. Biochemistry, 31, 11244-11248 (1992). [ Links ]
2. E. Silva, R. Ugarte, A. Andrade, A.M. Edwards. J. Photochem.Photobiol.B: biol. 23, 43-48 (1994). [ Links ]
3. E. Silva, Sensitized photooxidation of aminoacids in proteins, in chemical. Modification of enzimes active site studies (J. Eyzaguirre ed.), John Wiley and Sons, New York, 1987, pp 63-73. [ Links ]
4. P.A. Heelis, Chem. Soc. Rev. 11, 15-39 (1982). [ Links ]
5. E. Silva, M. Salim-Hanna, M.I. Becker, A. De Ioannes. Internat. J. Vit. Nutr. Res., 58, 394-401 (1988). [ Links ]
6. A.M. Edwards, E. Silva. Radiat. Environ. Biophysic, 24, 141-148 (1985). [ Links ]
7. A.M. Edwards, E. Silva. Radiat. Environ. Biophysic., 25, 113-122 (1986). [ Links ]
8. G. Tapia, E. Silva. Radiat. Environ. Biophysic., 30, 131 (1991). [ Links ]
9. A.M. Edwards, E. Silva, B. Jofré, M.I. Becker, A. De Ioannes. J. Photochem. Photobiol. B: biol., 24, 179-186 (1994). [ Links ]
10. A.M. Edwards, F. Barredo, E. Silva, A. De Ioannes, M.I. Becker. Photochem. Photobiol., 70, 645- 649 (1999). [ Links ]
11. G. Jori. J. Photochem. Photobiol. B: biol., 36, 87-93 (1996). [ Links ]
12. D. Kessel. Photochem. Photobiol., 49, 447-452 (1989). [ Links ]
13. A.M. Edwards, C. Bueno, A. Saldaño, E. Silva, K. Kassab, L. Polo, G. Jori. J. Photochem. Photobiol. B: biol., 48, 36-41 (1999). [ Links ]
14. F. Ogasawara, Y. Wang, D. Bobbett. Anal. Chem., 64, 1637-1642 (1992). [ Links ]
15. K. Miyamoto, Y. Takahashi, Tokio Tanabe Pharmaceutical Co., Ltda. Japan 3750, (66) (C.I. 16E62) March 3, Appl. Oct 10, 1963, 2pp. [ Links ]
16. K. Yagi, J. Okuda, A. Dimitrovskii, R. Honda, T. Matsubara. J. Vitaminol, (Kioto) 7, 276-280 (1961). [ Links ]
17. M. Campos, R. Díaz, A.M. Edwards, S. Kennedy, E. Silva, J. Derouault, M. Rey-Lafon. Vibrational Spectroscopy, 6, 173-183 (1994). [ Links ]
18. H.J. Grande, C.G van Schagen, T. Jarbandhan, F. Müller, Helv. Chim. Acta, 60, 348 (1977). [ Links ]
19. H.J. Grande, R. Gast, C.G. van Schagen, W.J.H van Berkel, F. Müller, Helv. Chim. Acta, 60, 367 (1977). [ Links ]
20. P. Keller, Q. Le Van, A. Bacher, J. Am. Chem. Soc., 105, 2505-2507 (1983). [ Links ]
21. C.G. van Schagen, F. Müller, Helv. Chim. Acta, 61, 3139-3142 (1978). [ Links ]