Int. J. PeptideProtein Res. 10,107-112 Published by Munksgaard, Copenhagen, Denmark No part may be reproduced by any process without written permission from the author(s)

ASYMMETRIC HYDROGENATION OF UNSATURATED PEPTIDES 0. PIERONI, D. BACCIOLA, A. FISSI, R. A. FELICIOLI* and E. BALESTRERI

C.N.R., Laboratorio per lo Studio delle Propnet; Fisiche d i Biomolecole e Cellule, Pisa. and *Cattedra di Chimica, Facolth di Medicina, Universith di Cagliari, Italy

Received 18 September 1976 Eleven unsaturated peptides, containing one or two dehydro-phenylalanyl (dehydro-Phe) residues and a C-terminal L-amino acid have been hydrogenated in the presence o f palladium-on-charcoal. Hydrolysis o f the saturated peptides thus obtained gave optically active phenylalanine showing the occurrence of asymmetric induction during the hydrogenation. Both monoansaturated dipeptides and doubly unsaturated tripeptides with L-Glu, L-Leu and L-Val as chiral end-group afforded L-Phe in 40-50% optical yield. In the case of the trip ep t ide N-ace t y I-(dehy dro -Phe )-(dehydro -Tyr)-L -Glu the asymmetric induction was higher (70%)f o r the unsaturated residue which is farther from the chiral endgroup along the peptide chain. The results are discussed o n the basesofPrelog rule and the rigid dissymmetric conformation of the dehydropeptides in solution.

Dehydro-amino acid units are present in many naturally occurring peptides having antibiotic (Bycroft, 1969) and phytotoxic (Koncewicz et d., 1973; Meyer er al., 1974) activity. In this connection several studies have been carried out in order to clarify their role in the biosynthesis and mechanism of action of microbial peptides. A growing interest has recently been shown in the synthetic methods and stereochemistry of these dehydropeptides (Shin et al., 1974; Morin & Gordon, 1973; Breitholle & Stammer, 1975; Bycroft, 1972; Rich & Mathiaparanam, 1974; Meyer e t al., 1975). We have reported (Pieroni er al., 1973; 1975) a detailed investigation of a series of N-acylated dehydropeptides containing one or two dehydro-phenylalanyl (dehydro-Phe) residues and a C-terminal optically active Lamino acid or m i n e residue. The determination of the structure allowed us

to conclude that the preparation method of unsaturated peptides originally proposed by Doherty et al. (1943) is stereospecific, as it gives double bonds with “Z” configuration both in mono- and doubly unsaturated derivatives. Crystallography data as well as circular dichroism measurements provided evidence for a certain degree of rigidity of the dehydro-Phe residue owing to the partial conjugation of the unsaturated bonds. In the case of doubly unsaturated peptides, steric hindrance forces the two unsaturated groups to stay in two distinct skewed planes, thus originating a rigid conformation probably stabilized by hydrogen bonding (Pieroni et al., 1975). Considering the conformational rigidity of the above dehydro-Phe peptide, we have carried out their catalytic hydrogenation in order to investigate whether the conversion of the 107

0. PIERONI, D. BACCIOLA, A. FISSI, R.A. FELICIOLI AND E. BALESTRERI

dehydro-Phe into phenylalanyl residue occurs a Perkin-Elmer model 141 Polarimeter; infrared with appreciable asymmetric induction spectra were recorded with a Unicam SP1200 spectrophotometer, and ultraviolet absorption (Equation 1). spectra were obtained by means of a Unicam *(L) SP700 spectrophotometer. CH, -CO-

NH-C-CO [ : H I

-NH-CH-COOH RI

H, ,Pd

Hydrogenation. One gramme of dehydropep-

tide in 100ml of methanol was hydrogenated in the presence of 0.1 g of 10% palladium-oncharcoal and 0.5 mi acetic acid at 25°C and 760 * *(L) mm Hg. At the end of the reaction (2 h) the -NH-CH-COOH intense absorption maximum, centred at 280 nm, of the unsaturated chromophore (Pieroni et al., 1975) disappeared. Catalyst was filtered off and the filtrate was evaporated to dryness. The obtained saturated peptide was not recrysRecent investigations have shown that a tallized in order to avoid possible diastereomeric great number of microbial antibiotic peptides enrichment. contain both dehydro and D-amino acid residues, and Bycroft (1969) has suggested that dehydro- Hydrolysis. The residue containing the saturesidues may be biosynthetic intermediates in rated peptide was hydrolysed by refluxing with the conversion of L- to D-amino acids by a 10 ml concentrated HCl under nitrogen atmosphere (Nakayama etal., 1971), then the reaction dehydrogenation-hydrogenation sequence. The biosynthesis certainly operates under mixture was extracted with ethyl ether to enzymatic conditions with a high degree of remove any organic oil. The aqueous layer was stereospecificity. However, it could be interest- evaporated to dryness under reduced pressure ing to determine the degree of stereospecificity to a mixture containing the hydrochlorides of possible to obtain under nonenzymatic con- Phe and the amino acid originally present as ditions, such as catalytic hydrogenation. end-group in the peptide. The mixture was At present, two sets of data are available in redissolved with water and the procedure the literature concerning catalytic hydrogen- repeated until free HCl was completely reation of peptides containing one dehydro-Phe moved. In the case of La-phenylethylamine residue and a C-terminal optically active group: derivatives the residue was dissolved in a the resulting Phe showed, in one case the minimum amount of water, alkalinized with opposite absolute configuration (Sheehan & 1 N sodium hydroxide and extracted with Chendler, 1961) and, in the second case, the portions of ether to remove the amine. same configuration (Nakayama et al., 1971) of the asymmetric carbon atom already present in Chromatographic separation. After acid hydrolysis, amino acid hydrochlorides were the molecule. separated by means of ionexchange column chromatography according to Moore e t al. MATERIAL AND METHODS (1958). The standard procedure was modified to a preparative scale: in order to shorten the Material The synthesis as well as i.r., U.V. and n.m.r. operative time, after complete elution of the characterization of the unsaturated peptides chiral amino acid with citrate buffer, the Phe have already been reported (Pieroni etal., 1975). was eluted with sodium hydroxide. The detailed The resin Amino Q 1506,expressly designed procedure was as follows: aliquots (0.200g) of for preparative work, was obtained from Bio- the mixture of amino acid hydrochlorides were dissolved in lOml of citrate buffer 100mM, Rad . pH4.25 and applied to an Aminex Q 150-S column (5.6 x 11 cm) equilibrated with citrate Methods General. Optical rotations were measured using buffer previously deoxygenated. The column

1 c,;, J

1.2

1

108

'

HYDROGENATION OF DEHYDROPEPTIDES

was eluted with citrate buffer and the nonaromatic amino acid was eluted after 140ml. Phe was then eluted with 1 NNaOH. The elution of both amino acids was followed by ninhydrin reaction. The identity and purity of the Phe was confirmed in each instance by t.1.c. plates cellulose-developed with pyridine: HzO (4: 1) and subsequently stained with ninhydrin reagent using authentic Phe as reference standard. As optically active Phe is more soluble in water than the racemic one, Phe was not crystallized. The eluted solution was neutralized to pH7 and concentrated to a definite volume (10 ml). The optical rotation of this solution was measured and the Phe content spectrophotometrically determined (Amax = 260; emax= 220). In run 11, Phe and Tyr were separated according to the method of Partridge (1949). RESULTS

All the unsaturated peptides have been hydrogenated at 760 mm Hg and at room temperature in the presence of 10%palladium-oncharcoal. In order to evaluate the asymmetric induction, the obtained saturated peptides were hydrolysed with hydrochloric acid. The mixture of the hydrochlorides was separated by chromatography on an ionexchange resin. The amino acids originally present as end-group in the peptides (L-Asp, L-Glu, L-Ala, LLeu, L-Val, L-Pro) were eluted with citrate buffer, Phe was then obtained by elution with 1 NNaOH in a practically quantitative yield. Identity and purity of the recovered Phe were confirmed in each instance by t.1.c. and in some cases by comparison of its infrared spectra with that of an authentic sample. Finally, measurements of optical rotation of the recovered Phe allowed us to determine the minimum optical yield of the hydrogenation reaction. In order to evaluate the eventual racemization during the hydrolysis of hydrogenated peptides, N-acetyl-LPhea-phenylethylamidewas hydrolysed under the same conditions. La-phenylethylamine was chosen as terminal chiral residue of the molecule to facilitate separation of the products after hydrolysis. In this case the recovered Lphenylalanine and Lar-phenyethylamine had the same optical rotatory powers as

the starting compounds, thus ruling out any racemization during the HCI treatment, to an appreciable extent. Even if this case cannot be applied to the other runs, as the racemization may be dependent on the sequence, the actual racemization amounts are probably small (Smart et al., 1960). In any case the determined optical yield of the recovered Phe must be considered the minimum optical yield of the hydrogenation reaction. Mono-unsaturated dipeptides

Hydrogenation of unsaturated peptides containing one dehydro-Phe residue occurred with optical yield between 1.5 and 47.5% depending on the nature of the C-terminal chiral residue (Table 1). For the peptides containing L-Asp, L-Glu, L-Ala, L-Val and L-Leu as C-terminal residue (runs 1-5), the prevalent configuration of the produced Phe was the same as that of the chiral residue originally present in the unsaturated compound. When L-Pro or La-phenylethylamine were present in the unsaturated peptide (runs 6 and 7) a prevalence of D-Phe was obtained. The asymmetric induction was very low in the hydrogenation of N-acetyl-(dehydro-Phe)-LAsp (run I), but the hydrogenations of the other peptides gave considerably greater prevalence of one enantiomer of Phe over the other. The magnitude of the induced asymmetry was dependent on the nature of the alkyl portion of the originally present optically active amino acid residue; so the optical purity of the produced Phe increased in the series of the LAla, L-Val, L-Leu derivatives, depending on the bulkiness of the lateral alkyl group. Doubly unsaturated tripeptides

Asymmetric induction occurs also in the hydrogenation of some tripeptides containing two dehydro-Phe residues and .an optically active end-group (Table 2 runs 8 , 9 and 10). With L-Glu and LVal as C-terminal residues the asymmetric induction is similar to that of the mono-unsaturated compounds while the tripeptide with terminal a-phenylethylamine gave apparently no asymmetric induction. However, it should be noted that in this case the induced asymmetry cannot be directly based on the optical rotatory power of the 109

0. PIERONI, D. BACCIOLA, A. FISSI, R.A. FELICIOLI AND E. BALESTRERI TABLE 1 Asymmetric hydrogenation of unsaturated peptides containing one dehydro-Phe residue Run

Unsaturated peptide

Recovered Phea [a1 % (H, 0,PH 7)

1 2 3 4 5 6 7

Optical yieldb %

-0.5 -12.8 -5.5 -12.8 -16.8 +2.1 +4.2

acetyl-(dehydro-Phe)-L-Asp acetyl-(dehydro-Phe)-LGlu acetyl-(dehydro-Phe)-L-Ala acetyl-(dehydro-Phe)-L-Val' acetyl-(dehydro-Phe)-L-Leu acetyl-(dehydro-Phe)-L-Pro acetyl-(dehydro-Phe)-La-phenylethylamided

Induced configuration

L L L L L D D

1.5 35.5 15.5 36.5 47.5 6 .O 12.0

Specific rotation for optically pure L-Phe: [a]i;, -35 (H,O; pH 7) b -~ [a]of recovered Phe x 100 [a]of optically pure Phe In the literature it is reported (Nakayama, 197 1) that the catalytic hydrogenation of N-acetyl-(dehydro-Phe)D-Val afforded D-Phe having [a]5 8 9 + 12 + 14 (34-40% optical yield) Hydrogenation in the presence of Raney-Ni gave 6% optical purity for D-Phe (Sheehan & Chendler, 1961)

recovered Phe. In fact, the specific rotatory power observed for the recovered Phe corresponds to an average value deriving from the asymmetric reductions of the two unsaturated residues enclosed in the tripeptide. Indeed, the two dehydro-Phe residues are quite unlikely to afford Phe with the same optical yield. The asymmetric induction might be higher for the residue which first undergoes reduction, and lower for the residue which is reduced later on. Another hypothesis might be that the asymmetric induction is higher for the unsaturated residue next to the chiral end-group and lower for the residue which is farther from the chiral end-group in the sequence of the peptide chain. Moreover, it is also possible that

the hydrogenation of the dehydro-Phe residues affords two Phe residues having opposite absolute configuration. In order to preliminarly clarify the above hypotheses and the steric course of the reaction, the hydrogenation of the tripeptide N-acetyl(dehydro-Phe)(dehydro-Tyr)-L-Glu, containing two different unsaturated residues, was carried out (run 11). The data reported in Table 2 show that the dehydro-Phe residue, which is farther from the asymmetric centre, affords GPhe in high optical yield. The asymmetric induction of the dehydro-Tyr residue is of the same order of magnitude as that observed for mono-unsaturated peptides (Table 1).

TABLE 2 A symme rnk hydrogenation of dou bly unsaturated tripcptides Asymmetric induction Run

8

9 10 11

Unsaturated peptide

Ac-(dehydro-Phe), -LGlu Ac-(dehydro-Phe), -L-Val Ac-(dehydro-Phe), -La-phenylethylamide

Ac-(dehydro-Phe)-(dehydro-Tyr)-LGlu

Recovered amino acid Phe Phe Phe Phe 5 r

Observed rotation

Optical purity

Configuration

blm I[.

40 46 racemic 70 30

L L

- 14 - 16 [a],,, 0 [a15 8 9 - 25 [a],,,- 22 58,

-

For optically pure L-Phe: [a],,,- 35 (H,O, pH 7); for optically pure L-Tyr: [a]436 - 73 (H,O)

110

L L

HYDROGENATION OF DEHY DROPEPTTIDES

DISCUSSION

Several literature reports (Sheehan & Chendler, 1961; Nakayama et d., 1971; Hiskey & Northfop, 1965) have utilized the Prelog rule to predict the enantiomer which will be predomiaantly formed in asymmetric hydrogenation under the induction of a chiral centre already present in the molecule. According to the Prelog rules ( F’relog, 1956) hydrogen is absorbed on the surface of the catalyst and the substrate approaches the catalyst surface from the least sterically hindered side. The peptides under consideration have an unsaturated residue which tends to be mostly planar, owing to conjugation of the unsaturated bonds: indeed they may be regarded as derivatives of trans-cinnamic acid. Furthermore, the conformation is supposed to be stabilized by intramolecular hydrogen bonding with the formation of a ring-like structure (Cann, 1972). The ‘Huggins type’ structure I (Scheme l), proposed for N-acylated amino acids (Cann, 1972), seems to be sterically impossjble in dehydro-amino acids. In fact, the N-H . . . 0 angle is very far from th; normal values for ‘hydrogen bonding (0-30 ) (Pieroni et d., 1975). SCHEME

N-C

The correct geometry for hydrogen bonding is obtained only between the C = 0 of the acyl group and the carboxyl terminal group (Scheme 1, structure 11). Structure I1 seems to be the most probable conformation during the reaction. In fact, when the amino acid residue present as end-group in the peptide has L-absolute configuration, the alkyl group is placed on the front of the paper. Assuming that the substrate approaches the catalyst surface from the less bulky side of the molecule, hydrogen should attack the double bond from the back of the paper. Accordingly the cis hydrogen addition should result in the formation of L-Phe, as experimentally observed. Steric effects due to the differences of methyl, isopropyl and isobutyl groups respectively in LAla, L-Val and L-Leu residues, were also observed. In run 6 it can be seen from the analogous conformer 111 that, by the use of D-amine, L-Phe should predominate. Conversely, the use of L-amine should lead to a predominance of D - f i e , as actually obtained. In the case of L-Pro derivative (run 7), a more detailed investigation is necessary in order to find an explanation of the stereochemical course of the reaction.

1

H

I

coon

I

H

111

H

IV

111

0. PIERONI, D. BACCIOLA, A. FISSI, R.A. FELICIOLI AND E. BALESTRERI

For doubly unsaturated peptides, we have already hypothesized (F'ieroni, 1975) on the bases of circular dichroism measurements that they have a rigidly fixed conformation in solution (Scheme 1, structure IV). Significative values of asymmetric induction are usually obtained in cyclic structures or rigid conformations. Therefore the rather high values of stereospecificity observed in the catalytic hydrogenations of doubly unsaturated peptides (Table 2) are in agreement with the circular dichroisrn data (Reroni, 1975). Moreover, in the case of Ac-(dehydro-F'he)-(dehydro-Tyr)-LGlu (run l l ) , the hydrogenation of dehydroPhe residue occurred with a higher degree of stereospecificity than the one observed for dehydro-Tyr residue. The dehydro-Phe is more distant than the dehydro-Tyr residue from the chiral centre along the primary structure of the peptide chain. Thus the obtained data confirm the substantially ordered secondary structure of the tripeptides. The stereochemical course of the reaction might consist firstly in the attack of hydrogen on the dehydro-Phe residue from the front of the molecule, so affording a high yield of L-Phe , then the dehydro-Tyr residue should be hydrogenated as described for mono-unsaturated peptides. Finally, the hydrogenation of N-acetyl(dehydro-Phe), a-L-phenylethylamide (run 10) gave practically racemic Phe. No definitive explanation is at present available for the absence of any asymmetric induction in this one case. A reasonable explanation could involve opposite asymmetric inductions in the reduction of the two dehydro-Phe residues. A more detailed investigation of this system is currently in progress. REFERENCES Breitholle, E. G. & Stamme1,C.H. (1975) Tetrahedron Letters 2381 -2384

112

Bycroft, B. W. (1969)Nature (Lond.) 224,595-598 Bycroft, B. W. (1972) J. Chem. SOC.Chem. Comm.

660-661 Cann, J. R. (1972)Biochemistry 11,2654-2659 Doherty, D. G., Tietzman, J. E. & Bergmann, M. (1943)J. Biol. Chem. 147,617-637 Hiskey, R. G. & Northrop, R. C. (1965)J. Am. Chem. SOC.87,1753-1757

Koncewicz, M., Mathiaparanam, P., Uchytil, T. F., Sparapano, L., Tam, J., Rich, D. H. & Durbin, R. D. (1973) Biochem. Biophys. Res. Commun. 53,

653-658 Meyer, W. L., Kuyper, L. F., Lewis, R. B., Templeton, G. E. & Wo0dhead.S. H. (1974)Biochem. Biophys. Res. Commun. 56,234-239 Meyer, W. L., Templeton, D. E., Grable, C. I., Jones, R., Kuyper, L. F., Lewis, R. B., Sigel, C. W. & Woodhead, S. H. (1975) J. Am. Chem. SOC. 97,

3802-3809 Moore, S., Spackman, D. H. & Stein, W. H. (1958) Anal. Chem. 30,1185-1206 Morin, R. B. & Gordon, E. M. (1973) Tetrahedron Letters 2163-2166 Nakayama, M., Maeda, G., Kaneko, T. & Katsura, H. (1971)Bull. Chem. SOC.Japan 44,1150-1152 Partridge, S . M. (1949)Biochem. J. 44,521-527 Pieroni, O., Fissi, A. & Montagnoli, G. (1973) Biopolymers 12,1445-1449 Pieroni, O., Montagnoli, G., Fissi, A., Merlino, S. & Ciardelli, F. (1975)J. Am. Chem. SOC. 97,6820-

6826 Prelog, V. (1956)Bull. SOC.Chim. France, 987-995 Rich, D. H. & Mathiaparanam, P. (1974)Tetrahedron Letters 4037-4040 Sheehan, J. C. & Chendler, R. E. (1961)J. Am. Chem.

SOC.83,4795-4191 Shin, C., Nanjo, K., Ando, E. & Yoshimura, J. (1974) Bull. Chem. SOC.Japan 47,3109-3113 Smart, N. A., Young, G. T. & Williams, M. W. (1960) J. Chem. SOC.3902-3912 Address: 0.Pieroni C.N.R.-Laboratorio per lo Studio delle Proprieti Fisiche di Biomolecole e Cellule Via F. Buonarroti, 9 56100 Pisa Italy

Asymmetric hydrogenation of unsaturated peptides.

Int. J. PeptideProtein Res. 10,107-112 Published by Munksgaard, Copenhagen, Denmark No part may be reproduced by any process without written permissio...
412KB Sizes 0 Downloads 0 Views