CryoLetters 24, 253-259 (2003) Ó CryoLetters, c/o Royal Veterinary College, London NW1 0TU, UK
CRYO-BIOORGANIC CHEMISTRY: FREEZING EFFECT ON STEREOSELECTION OF DL-ALANINE-N-CARBOXYANHYDRIDE OLIGOMERIZATION IN DIOXANE SOLUTION a
b
Tamás Vajda *, Marianna Mák and Miklós Hollósi
a
a
Department of Organic Chemistry, Eötvös Loránd University, P.O.Box 32, Budapest 112, H-1518 (Hungary), Fax +36-1-372-2620. E-mail:
[email protected] b G. Richter Ltd., Spectroscopic Res. Lab., Mass Spectrometry Group, Budapest Abstract The possibility of stereoselection through the DL–alanine–N–carboxyanhydride (NCA–DL– Ala) oligomerization, and the effect of freezing on it have been investigated. To this end, the chirality of peptides obtained by oligomerization for 1 and 3 days, respectively, in liquid (+22o C) and frozen (–18ºC) dioxane solutions, was analyzed. These water-soluble samples were fractionated by gel filtration, aliquots of the fractions were completely hydrolyzed and then derivatized with Marfey reagent (1–fluoro–2,4–dinitrophenyl–5–L–alanine amide). These derivatives were traced and evaluated by RP–HPLC analysis. The relatively best effects appeared in a given fraction, where after 1 day of oligomerization the L–alanine enantiomeric excess (ee) was 3.8% in liquid and 8.6% in frozen conditions. After 3 days, however, the ee contents decreased to 2.0% and 4.1%. The mass spectrometric data of the peptides pointed to the formation of open chain and cyclic peptide mixtures, where the residue numbers of 8-11 and 4-5 dominated. The formation of some chiral peptides from a racemic amino acid suggests the possibility of preferential incorporation of the L-enantiomer into the growing chain, beside the achiral statistical succession of residues. Here we provide the first example of the role of freezing in the increased formation of chirality from racemic amino-acid through oligomerization, together with some speculations about the implications of our model in the events of prebiotic chemistry. Keywords: freezing, oligomerization, enantiomeric excess, prebiotic chirality generation. INTRODUCTION The elucidation of how the mirror symmetry of the bio-organic world could be broken may help us to understand the dynamics of the prebiotic evolution. The puzzle of the prebiotic chemistry of the amino acid/peptide/protein chirality represents still a high-level of research challenge. These abiotic events comprise two basic questions of a sequential scenario, i.e. the asymmetry generation from a racemic amino acid mixture and then the chiral amplification of the few optically active molecules (1, 5, 7, 15, 16).The first aspect of these two questions will be addressed here. Many abiotic mechanisms have been proposed to explain both questions; however, the laboratory experiments for chirality generation from racemic biomolecules by
chemical means are quite rare. Although, it was shown in 1952 by IR investigations (6) that films cast from aqueous solution of poly-DL-alanine appeared in a or b conformations, which may point to some ee domains among the achiral succession of alanine residues of the polypeptide chain, the presumed configurational consequences were not proven. The first direct evidence for asymmetry generation was given by polymerization of NCA-DL-alanine, where the polymer showed some small optical rotation indicating a slight accumulation of the L-amino-acid in excess to the D-amino-acid (17). The experiments were carried out very carefully to avoid any systematic error. Another proof was obtained by Kricheldorf and Hull (11), who polymerized different 15N-enriched DL-amino acid N-carboxyanhydrides and traced the stereochemistry of the polymers by 15N-NMR spectroscopy. Depending on the conditions, the formation of some isotactic and syndiotactic sequences were found; however, the stereospecificity was low as the average lengths of the stereo blocks never exceeded 4 monomer units. Later, some other similar examples were given for stereoselection of 4-6 unit oligopeptides tested with different methods (9, 23). On the other hand, there is a growing demand to study molecular interactions in frozen aqueous or organic solutions because of their great importance from biochemical, biophysical, biotechnological but also theoretical points of view, since reactions in such phases proceed in a different manner from the homogeneous ones. Although, quite different bio-organic chemical reactions (18, 19) were carried out in frozen systems (see also a survey in 20), to the best of our knowledge, the freezing effect on stereoselection is investigated only in our laboratory. Namely, we have detected some freezing effect on the L-enantiomer amplification in the L-/DL-leucine co-oligomerization (21). This previous result prompted us to study the freezing effect also on chirality generation, and the racemic DL-alanine has been chosen for this purpose. This amino acid is a good a-helix-forming molecule and it may have been also a stock molecule of the prebiotic world. The NCA-DL-Ala oligomerization was carried out in liquid and frozen dioxane solutions, the peptide mixture obtained was fractionated by gel filtration, and then the samples of the hydrolyzed and derivatized fractions were tested by RPHPLC chromatography. Concerning the gel filtration, it seemed worthwhile to fractionate the mixture obtained, to find some enantiomeric excess containing peptides, and to eliminate the unreacted alanine molecules. Because of our poor preliminary experience with the 1,1’carbonyl-diimidazole activation in aqueous solution, we have chosen the oligomerization of NCA-derivative in dioxane. MATERIALS AND METHODS Chemicals All chemicals purchased from Aldrich-Sigma were of reagent grade. The RP-HPLC eluents originated from Carlo-Erba. NCA-DL-alanine The preparation was similar to that of Day and Poche (3). DL-alanine (1.5 g) suspended in 30 ml dioxane was treated with 2 g of triphosgene for 4 h at 50 ± 1ºC and stored at ambient temperature overnight. The insoluble material (0.50 g) was removed by filtration, the filtrate evaporated in vacuo and kept under petrol ether (30 ml) at -20ºC for several hours. The separated pulver was recrystallized from ethyl acetate/petrol ether, kept at -20ºC for 48 h and after drying in a vacuum desiccator over P2O5, it resulted a faint yellow crystalline pulver (0.96 g , 51%; subtracting the insoluble DL-alanine: 78% yield).
Oligomerization Stock solutions of 0.14 M NCA–DL–alanine in dioxane were prepared before the experiments and then distributed into samples. The oligomerization was started with triethylamine initiator (NCA/TEA = 38). After rigorous shaking (30s), half of the samples were left at ambient temperature (22 + 1ºC), while the other half were frozen and kept in the freezer (-18 ± 1ºC). The samples were treated for 1 or 3 days, and then those frozen were thawed in a bath (ca 30ºC). After acidification with hydrochloric acid to pH 3-4, the solutions were evaporated in vacuo (bath 30-35ºC), dissolved in water and freeze-dried. Gel filtration and hydrolysis The crude oligomers obtained were gel filtered on Sephadex G25 Medium in a syringe (20 ml, 80x20 mm) with an eluent of 5% acetic acid and flow rate of 0.5-0.6 ml/min at ambient temperature. In one run a 5-6 mg/ml solution of oligomer in 5% acetic acid was gel filtered and fractionated (à 1ml) into fifteen polypropylene centrifuge tubes. This procedure was repeated 6-8 times, filtering the given fractions into the same tubes. The fractions were freeze-dried and used for hydrolysis and mass spectrometry. A weighed amount (3-4 mmol) of each fraction was completely hydrolyzed with 200 ml of 6 M hydrochloric acid at 110±1ºC for 24 h. The hydrolyzates were evaporated in a vacuum desiccator over P2O5 / granular KOH and, after dissolution the remaining parts in 200 ml of water, then this evaporation was repeated. For control, the DL-alanine amino acid was treated identically to the samples under investigation. Derivatization The method was as published by Marfey (14) with some modifications. The solid hydrolyzate was dissolved in water (2 mmol/100 ml) and 50 ml (1 mmol) of this was mixed with 16 ml (8 mmol) of 0.5 M sodium bicarbonate solution (pH 8.0) and 78 ml (1.4 mmol) of 0.5% solution of Marfey’s reagent in acetone. This solution was heated at 40ºC for 60 min and after cooling, 10 ml of 1 M hydrochloric acid was added. After 100-fold dilution with methanol, the mixture was kept in the refrigerator until the RP-HPLC analysis. RP-HPLC analysis A vertex column (250 x 4 mm) was used with C18-silica as the stationary phase and 0.02 M sodium acetate buffer (pH 4)-methanol-acetonitrile (7:2:1, v/v/v) as the mobile phase, with a flow rate of 1 ml/min at ambient temperature. A volume of 80 ml (16 nmol) of the 100fold diluted sample was injected, the peaks were monitored at 340 nm and identified with authentic samples of L- and D-alanine derivatives. The integrated areas of the identified peaks served for evaluation of the L-alanine yield. All chromatography experiments were repeated at least three times. Mass spectrometry The fast atom bombardment mass spectrometric (FAB-MS/LSIMS) experiments were carried out using a Finningan MAT 95SQ hybrid tandem mass spectrometer equipped with a liquid secondary ion source. The Cs+ ion gun was operated at 20 keV. The peptide samples dissolved in methyl alcohol were mixed with glycerol as liquid matrix.
RESULTS AND DISCUSSION Optical rotation The racemity of DL-alanine was examined by optical rotation and RP-HPLC analysis (see next paragraph). The specific optical rotation data of Table 1 represent well, that the slight rotation values of racemic alanine are equal to those of the 6M hydrochloric acid solvent. However, the small difference found at 546 nm stays around the error limit of the method (measured ao = +0.020; +0.020 and + 0.010; +0.015 for DL-Ala and 6N HCl, respectively). The rotation values of the L-enantiomer are also given to see the large difference between the active and racemic amino acids. Shorter wavelengths were chosen instead of l = 589.3 nm, because of the higher rotations at shorter wavelength values. Table 1. Control of DL-alanine racemity by optical rotation o
[+a ] l (nm) 366 406 436 546 DL-Ala 5.37 2.00 1.00 0.50 6M HCl 5.37 2.05 1.00 0.31 L-Ala 62.2 40.9 31.7 16.7 All measurements at c = 4 g 6 M HCl and 25 ± 0.2ºC. The values are means with a deviation of 0.06-0.12º for DL-Ala, 6 M HCl and 0.06-0.25º for L-Ala.
RP-HPLC analysis The peaks appeared with the following mean retention times (min): Marfey’s reagent: tR = 2.2 0.0; L-Ala-derivative tR = 7.3 0.1; D-Ala-derivative tR = 13.8 0.4 and hydrolyzed reagent (1-hydroxy-2,4-dinitrophenyl-5-L-alanine amide) tR = 18.7 0.4. Table 2. Stereoselection of NCA-DL-alanine oligomerization in liquid (22ºC) and frozen (-18ºC) dioxane solutions. RP-HPLC analysis of hydrolyzed fractions Time (days) 1
3
liquid frozen
Conditions
L-Ala (%) 53.8 ± 0.7 58.6 ± 0.8
liquid frozen
52.0 ± 0.8 54.1 ± 2.1
Control monomer 49.6 ± 1.3 Oligomerization of DL-alanine-N-carboxyanhydride (NCA) with triethylamine (TEA) initiation in dioxane solution (NCA/TEA = 38). All data relate to the 9th fractions of gel filtration. L-Ala(%) = 100 x L/(L+D); the values are means ± sem; ee = [L-Ala(%)]–50 (%).
The identically treated DL-alanine monomer control resulted in peaks with the same retention times.The integrated areas under the L- and D-alanine derivative peaks served for quantitative evaluation of the hydrolyzates (Table 2). Data of a repeated experiment are included in means
± sem. The hydrolyzates of some fractions of the oligomerized racemate showed a slight enantiomeric excess of L-alanine, however, the significant ee yields were found only in the 9th fractions. As can be seen from Table 2, the over-representation of L-alanine was larger after a one-day treatment than those after three days. This indicates that with elapsing time an increasing amount of the D-antipode may also incorporate into some peptides. Furthermore, in both cases the production of L-enantiomer excess was approximately twice as large in frozen than in liquid conditions. Because some excess of the D-enantiomer did not appear in any fraction, it can be concluded that a part of the D-alanine remained unoligomerized, beside the achiral peptides of a random sequence of left- and right-handed units. Nevertheless, it is to be considered that the gel used for fractionation was a cross-linked dextran derivative and thus a chiral material which could even discriminate between the enantiomers. However, from the above results, since some ee of the D-alanine was never shown in any fraction and because we know of no paper showing a definite positive influence of the Sephadex gel towards the stereoselection of peptides/oligopeptides, therefore this argument can be eliminated. Mass spectrometry The data of the 9th fractions indicate the mixtures of open chain and cyclic peptides of different size, where the residue numbers of 8-11 and 4-5, respectively, dominated (Figure 1) and the peptides of frozen solutions appeared mostly in larger amounts.
35
A
30 25 20 15 10 5 0 6-7 8-9 10 - 11 12 - 13 range of residue number
relative amount (%)
relative amount (%)
40
90 80 70 60 50 40 30 20 10 0
B
4-5 6-7 8-9 range of residue number +
Figure 1. FAB-MS spectrometry. Relative amount of peptides to the M fractions of open (A) and cyclic (B) chains, respectively, vs number of residues in the given range is depicted. Columns: 1 day treatment at 22ºC (empty); -18ºC (black) and 3 days treatment at 22ºC (pointed); -18ºC (hatched). All data relate to the 9th fractions of gel filtration.
The obtained stereoselections among the racemic oligomers were small but significant and unidirectional, i.e. albeit it could be considered the L-alanine dominance as a chance, its enantiomeric counterpart has not appeared also in the repeated experiment. The obtained results may allow us to suggest some hypotheses, as they indicate that the mirror symmetry can be broken spontaneously in far-from-equilibrium microheterogeneous systems. The unstirred liquid and more the frozen solutions of the oligomerization satisfy this criterion. Structured clusters can be formed even within the bulk of solutions and then the oligomerization may occur within these clusters. The far-from-equilibrium condition is valid to a greater extent for the frozen solutions. On the other hand, some a-helical turns at the N-terminal part of the
peptide chains or the formations of cyclic peptides favour the homochirality, i.e. the preferential incorporation of L-lanine residues into the chain. An occasional incorporation of D-amino acid is disfavoured by steric hindrance and destabilizes the given structures (13, 22). Although the ice and the frozen dioxane (melting point: 11.8ºC) are quite different solvents in several relations, their low permittivity values (dielectric constants) are similar to each other (Table 3) and therefore these data suggest the possibility of some similar effects on their solutes. Table 3. Relative permittivity ( e ) data of water, ice and dioxane (12, 24) Temperature (ºC) 20 0 -10 20 0 (frozen)
Solvent
e
water
80.3 88.0 3.3 2.2 2.1
ice 1,4-dioxane
The effects of ice (8) are the following: (i) the frozen surface of low polarity can help the interactions between molecules positioned onto the surface, i.e., incorporation of the NCA-L-derivative into the chain by ligating both of them; (ii) the bimolecular reaction rate between the NCA-L-enantiomer and aminoterminal of the peptide chain may be significantly enhanced by freezeconcentration of the solutes in the diminished liquid phase of the cavities, and (iii) freezing as partial drying means some desolvation of the solvation shells, which can cause the activation (destabilization) of reactants. Deming and her co-workers have shown that life can exist also in ice eutectic phases as a medium at extremely low temperature (-20ºC), i.e. organisms (psychrophiles) living at the critical interface inherent to the phase change of water to ice. They recorded exciting results of an Arctic wintertime study of sea-ice bacteria within the brine inclusions of the ice (4, 10). On the other hand, if the similarity of the ice surface and that of frozen dioxane can be supposed, then our results have implications for life-favouring processes at the eutectic point, i.e., the above experiments provide a simple model for generation of optically active oligopeptides in frozen solutions. Therefore, it seems worthwhile to speculate on the role of ice in the prebiotic chemistry of chirality. This relates to the fundamental question of the origin of molecular chirality on the Earth at the most extensive glaciation in the past billion years (2). Acknowledgements: This work was supported by a Hungarian grant (FKFP) No 0075/2000 to M. Hollósi. REFERENCES 1. 2. 3. 4.
Bonner WA (1994) Orig Life and Evol Biosphere 24, 63-78. Calder N (1983) Timescale - An Atlas of the Fourth Dimension, Viking Press, NY. Daly WH & Poche D (1988) Tetrahedron Lett 29, 5859-5862. Deming JW (2002) Curr Opinion in Microbiol 5, 301-309.
5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24.
Eliel EL, Wilen SH & Mander LN (1994) Stereochem Org Comp, Wiley, NY, pp 209214. Elliott A (1952) Nature, Lond 170, 1066-1067. Frank FC (1953) Biochim Biophys Acta 11, 459-463. Franks F (1985) Biophysics and Biochemistry at Low Temperatures, Camb Univ Press. Hitz T, Blocher M, Walde P & Luisi PL (2001) Macromolecules 34, 2443-2449. Junge K, Krembs C, Deming JW, Stierle A & Eicken H (2001) Ann Glaciol 33, 304310. Kricheldorf HR and Hull WE (1982) Biopolymers 21, 1635-1655. Landolt-Börnstein (1959) Elektrische Eigenschaften I, vol 2, Part 6, Springer, Berlin. Majer ZS, Láng E, Vass E, Szabó S, Hallgas B and Hollósi M (1999) in Advances in BioChirality, (eds) G Pályi, C Zucchi and L Caglioti, Elsevier, Amsterdam, pp. 285295. Marfey P (1984) Carlsberg Res Commun 49, 591-596. Mason SF (1991) Chemical Evolution, Clarendon Press, Oxford. Mislow K (2003) Collect Czech Chem Commun 68, 849-864. Thiemann W & Darge W (1974) Orig Life 5, 263-283. Vajda T (1998) CryoLetters 19, 361-366. Vajda T, Szókán Gy and Hollósi M (1998) J Peptide Sci 4, 300-304. Vajda T (1999) Cell Mol Life Sci 56, 398-414. Vajda T & Hollósi M (2001) Cell Mol Life Sci 58, 343-346. Wada A (1961) J Mol Biol 3, 507-519. Weissbuch I, Bolbach G, Zepik H, Shavit E, Tang M, Frey J, Jensen TR, Kjaer K, Leiserowitz L & Lahav M (2002) J Am Chem Soc 124, 9093-9104. Wohlfarth C (1994) in CRC Handbook of Chem and Phys, (ed) DR Lide, CRC Press, Boca Raton, pp 155-188.
Accepted for publication 11/7/03
