This work has been digitalized and published in 2013 by Verlag Zeitschrift für Naturforschung in cooperation with the Max Planck Society for the Advancement of Science under a Creative Commons Attribution4.0 International License.

Dieses Werk wurde im Jahr 2013 vom Verlag Zeitschrift für Naturforschungin Zusammenarbeit mit der Max-Planck-Gesellschaft zur Förderung derWissenschaften e.V. digitalisiert und unter folgender Lizenz veröffentlicht:Creative Commons Namensnennung 4.0 Lizenz.

Conformations of Chlorophylls a and a' and their Magnesium-Free Derivatives as Revealed by Circular Dichroism and Proton Magnetic Resonance Paavo H . Hynninen*

Department of Biochemistry, University of Kuopio, P.O.B. 138, SF-70101 Kuopio (Finland)

Gunnel Sievers Department of Biochemistry, University of Helsinki, Unionink. 35, SF-00170 Helsinki (Finland)

Z. Naturforsch. 36b, 1000-1009 (1981); received May 19, 1981

Stereochemistry, Conformational Analysis, Chlorophyll and Pheophytin Epimers, XH NMR Spectra, Circular Dichroism

The electronic absorption (UV/VIS), circular dichroism (CD) and proton magnetic resonance (*H NMR) spectra have been recorded for C-l0 epimeric chlorophylls a and a', pheophytins a and a' as well as pheophorbides a and a'. Although the epimers in each pair showed virtually identical UV/VIS spectra, their CD spectra were profoundly different and exhibited opposite signs at most wavelengths in the UV region. The differences were interpreted as arising, in part, from different C-10 configurations, and, in part, from conformational alterations induced by the steric strain in the crowded periphery of the macrocycle. The conformational alterations were also clearly indicated by the XH NMR zJö-values observed for the a,ß and <5 methine protons, the C-10 protons and most methyl group protons of the epimers in each pair. This was considered to imply changes in the geometry of the whole macrocycle. The A 6-values were larger for the Mg-free epimers than for the chlorophyll epimers, which shows that the central Mg-atom makes the macrocycle more rigid. Correlations between the signs of the CD bands and configurations are discussed.

Introduction

The stereochemistry of chlorophylls (Chi) and related compounds has been an impor tan t aspect of chlorophyll research for about f if ty years [1, 2]. Chlorophylls a ( l a ) und b ( lb) have been shown to be subst i tuted 7,8-dihydroporphyrins (chlorins) containing asymmetric centers a t C-7, C-8, C-10, C-7' and C-l 1' with the absolute configurations 7(S), 8(S), 10(R), 7'(R), l l ' ( R ) . The relative and absolute configurations of these asymmetric carbons have been determined through extensive investigations by chemical and physical methods including CD/ORD and i l l NMR (see recent reviews by Brockmann [3, 4], Scheer [5] and Wolf and Scheer [6]. The hydrogens a t C-7 and C-8 have been shown to be in trans configuration and the relative configu-rat ion of C-7 and C-10 has also been demonstrated to be trans. The above stereochemistry has been conclusively established through Woodward 's total synthesis of Chi a [7] and through recent X-ray investigations [8, 9]. The lat ter have also provided direct information of the conformations of Chi a and its Mg-free derivatives (3, 5) in the solid state.

In recent years, it has become more evident tha t a detailed knowledge of the relationship between the stereochemistry and chemical reactivi ty in chlorophylls is essential for an understanding of the organization and function of these compounds in photosynthetic membranes. I n a t tempt ing to obtain knowledge of the correlation between the stereo-chemical and the electron donor-acceptor (EDA) properties in the chlorophylls [10], we have explored the conformation and E D A properties of Chi a ' (2) which is stereoisomeric with Chi a a t C-10, i.e. Chi a and Chi a ' possess the 10(R) and 10(S) configura-tions, respectively. Previous work [11] had con-firmed the epimeric relationship between the two chlorophylls and had led to the conclusion t h a t the unexpected properties of Chi a ' are best interpreted in terms of the conformational al terations arising from the single stereochemical change a t C-10.

The altered steric position of the C-10 methoxy carbonyl group and its conformational consequences prevent Chi a ' from forming water linked chlorophyll oligomers ("crystalline chlorophyll"). This differ-ence as compared to chlorophyll a, was utilized for the first t ime in 1975 for the separation of chloro-phylls a and a ' [12]. Also the fact t h a t the epimeric chlorophylls are so easily separable on sucrose [13] is probably based on the stereochemically induced

differences in the EDA properties of the two compounds.

In the present work, we compare the CD and *H NMR spectra of chlorophylls a and a' ( la , 2) and a t tempt to draw some detailed conclusions regarding the conformations of these epimeric compounds and their magnesium-free derivatives. The CD spectra previously measured for chlorophylls a and a ' [14, 15] are compared with the present results. We also report, for the first time, CD and XH NMR spectra for pheophytin a ' (4) and pheophorbide a' (6).

Results Electronic absorption spectra

The electronic absorption spectrum of Chi a ( la) in T H F is presented in the lower part of Fig. 1. Consistently with previous measurements in diethyl ether [11, 13] the absorption spectrum of Chi a ' (2) in T H F exhibits only slight differences from the spectrum of Chi a (Table I). Both spectra show, however, some essential differences when compared with the spectra measured in diethyl ether. Band I (QVo) is redshifted by 4 to 5 nm. The second band a t 626 nm, is not completely resolved from the third

one at 615 nm. The fourth band at about 593 nm is of low intensity and shows some fine structure probably implying that there is another overlapping band in this region. The fifth band at 540 nm is of very low intensity. The Soret band (B) is redshifted by 7 nm, and the bands in the UV region show more fine structure (bands rj3 and L) as compared to the spectrum in diethyl ether.

The absorption spectrum in Fig. 1 resembles that reported by Katz and Evans [16] for Chi a in dry pyridine. This spectrum has been interpreted as arising from chlorophyll species where the central Mg-atoms is hexacoordinated. According to Shipman et al. [17], the conversion from penta- to hexa-coordination results in a strong red shift of the Qx transition relative to the Qv transition, and, consequently, the Qxo-o band appears located be-tween the QV-o and bands. We therefore assign the band at 626 nm in Fig. 1 to the Qzo-o transition. The bands of very low oscillator strength ( I I I -V in Fig. 1) are assigned to higher vibronic components of the Q* and Qv transitions. In the assignment of the bands in the blue region (rji, r/2, rj3, N, L, M) we have followed the notation by Gouter-

Table I. Electronic absorption spectra and chiroptical properties of chlorophylls a and a' and their magnesium-free derivatives in tetrahydrofuran.

Absorption Molar Molar Circular Compound maximum absorptivity Assignment ellipticity dichroism

Amax e X 10-3 [0] X 10-3 (A) Ae (A) (nm) (M-icm-i) (degree M -1cm_1) (M-icm-i)

Chlorophyll a 664.1 86.6a Qyo-o l a 626.3 15.1 Q*o-o

615.0 13.4 Q^o-i 593.0 8.10 Q^o-2 540.0 3.00 Q*o-i 505.0 1.66 Q*0-2 435.5 116.2 B 18.4 (436) 5.58(436) 413.3 67.2 Vi 14.7 (402) 4.46(402) 386.3 43.1 V2 — 7.36(359) — 2.23(359) 333.5 26.8 N 24.5 (329) 7.43(329) 300.0 20.4 L 9.20(295) 2.79(295) 249.0 23.3 M 12.0 (268) 3.64(268)

— 20.2 (231) — 6.12(231) Chlorophyll a' 664.7 86.4b Qyo-o 2 626.7 15.9 Qxo-o

615.0 13.9 Q^o-i 593.0 8.38 2 537.0 3.25 Qxo-i 505.0 2.05 Q*0-2 435.9 114.7 B 68.1 (436) 20.6 (436) 414.1 69.1 m 9.20(419) 2.79(419) 386.7 44.5 T) 2 — 53.3 (384) — 16.2 (384) 333.2 26.6 N 15.3 (359) 4.64(359) 300.0 20.2 L — 15.6 (310) — 4.73(310) 252.0 24.9 M — 21.2 (280) — 6.43(280)

— 38.6 (232) — 11.7 (232) Pheophytin a 668.0 56.1° Q^o-o 3 609.0 9.52 Q^o-i

559.5 3.82 QV0-2 534.4 11.1 Qxo-o 3.68(542) 1.12(542) 505.0 13.1 Q*o-i 470.0 4.36 Qz0—2 411.4 120.7 B 34.9 (411) 10.6 (411)

~ 395.0 ~ 100 m 26.7 (395) 8.10(395) - 3 7 0 . 0 ~ 60 r] 2 — 4.60(355) — 1.39(355)

322.0 23.6 N 25.7 (322) 7.79(322) 276.5 13.6 L 14.7 (277) 4.46(277)

29.4 (245) 8.91(245) 227.5 25.4 M — 11.0 (223) — 3.34(223)

Pheophytin a' 668.3 56.ld Q^o-o 4 609.5 9.45 Qvo-i

559.5 3.56 0^0-2 534.7 12.1 Qzo-o 73.0 (533) 22.1 (533) 505.4 13.0 Q^o-i 10.1 (496) 3.06(496) 470.0 4.95 QZ0-2 6.44(465) 1.95(465) 411.4 120.8 B 159.5 (411) 48.4 (411)

~ 370.0 60 V 2 — 63.4 (371) — 19.2 (371) 321.4 24.0 N 19.3 (323) 5.85(323)

— 18.4 (301) — 5.58(301) 276.5 17.4 L —27.6 (274) — 8.37(274) 225.0 29.9 M — 58.9 (230) — 17.9 (230)

Pheophorbide a 668.2 52.7 Qfo-o 5 609.4 9.15 Qyo-i

560.0 3.94 Qv0-2 534.9 11.4 Qxo-o 4.52(545) 1.37(545) 505.7 12.8 Q*o-i 471.0 4.67 QZ0-2 411.6 113.5 B 62.1 (411) 18.8 (411)

Table I (continued)

Compound Absorption maximum Amax (nm)

Molar absorptivity e X lO"3

(M-icm-i)

Assignment Molar ellipticity [0] x lO-3 (A) (degree M -1cm_1)

Circular dichroism Ae (X) (M-icm-i)

- 3 9 5 . 0 ~ 94 Vi 43.3 (395) 13.1 (411) ~ 370.0 ~ 56 V2 — 8.28(356) — 2.51(356)

322.0 23.0 N 32.0 (324) 9.70(324) — 3.76(297) — 1.14(297)

276.5 16.8 L 9.03(276) 2.74(276) 34.6 (244) 10.5 (244)

225.0 28.9 M — 18.8 (223) — 5.70(223) Pheophorbide a' 668.8 52.7e Qyo-o 6 610.0 8.83 Qfo-i

560.0 3.33 Q"o-2 534.9 11.0 Q*o-o 38.7 (534) 11.7 (534) 505.9 12.3 Q*o-i 3.76(495) 1.14(495) 471.0 3.86 Qxo—2 1.88(465) 0.57(465) 411.6 111.4 B 135.5 (412) 41.1 (412)

~ 370.0 ~ 55 m —41.4 (370) — 12.6 (370) 322.1 21.8 N 20.6 (324) 6.25(324)

— 16.9 (298) — 5.12(298) 276.0 15.5 L — 15.0 (278) — 4.55(278) 228.0 25.4 M — 37.6 (228) — 11.4 (228)

a This value is based on the assumption that the chlorophyll preparation contained molecular species Chi a • H2O (MW 911.5). 1.79 mg of solid Chla • H 2 0 was dissolved in 10 ml of THF; c = 1.964 x lO"4 M. The measured A664 of this solution was 1.700.

b This value is based on the assumption that the chlorophyll preparation contained molecular species Chi a' • H 2 0 . 1.60 mg of solid Chi a' • H 2 0 was dissolved in 10 ml of THF; c = 1.755 x IO-4. The measured A665 of this solution was 1.516.

c The molar absorptivity was assumed to be identical with that of 4. d This value was obtained by weighing 1.60 mg of solid Chi a' • H 2 0 which was converted quantitatively to

pheophytin (see Experimental). The pheophytin a' obtained was dissolved in 10 ml of THF; c = 1.755 X lO^M. The measured Aees of this solution was 0.985.

e The molar absorptivity was assumed to be identical with that of 5.

m a n [18] and Weiss [19]. According to Weiss, these bands probably represent separate electronic t ran-sitions, in spite of their incomplete resolution and ra ther low intensity. This concept is supported by recent quan tum mechanical calculations of Petke et al. [20] suggesting a t least ten separate electronic transit ions for t he Soret band region of Chi a.

The electronic absorption spectra of the Mg-free derivatives (3-6) in T H F are all virtually identical (Fig. 2 and Table I). The differences between the spectra measured in T H F and those measured in diethyl ether are also small. The bands of low oscillator s t rength ( I I -V) are more resolved from one another t h a n the corresponding bands in the chlorophyll spectrum. The assignment of the bands to specific transit ions is therefore easier in the case of the Mg-free derivatives. The band corresponding to the Qxo-o transi t ion is now located a t 535 nm. In the assignment of the absorpt ion bands in Fig. 2 to

specific transitions, we have followed the interpreta-tions by Gouterman [18] and Weiss [19].

Circular dichroism spectra

Chlorophylls a and a ' exhibit profoundly different CD spectra (Fig. 1) in spite of the similarity of their electronic absorption spectra. The intensities of the CD bands a t the Soret band region are quite different for the two chlorophylls. The signs of these bands, however, are both positive. I n the region f rom 420 to 250 nm the CD spectra of the two chlorophylls show opposite signs a t almost every position (Fig. 1, Table I), which suggests t h a t this region should be particularly suitable for the resolution and analysis of the chlorophyll C-10 epimers. The ellipticities in the region f rom 600 to 450 nm are very close to zero for both chlorophylls. This is in accordance with the assignment of the absorption band a t 626 nm to the Q*o-o transition.

150

- 100

200 300 400 500 600 Alnm)

700

Fig. 2. Circular dichroism spec-tra (upper curves) of pheophytin a (- - -) and a' (—) in tetrahydro-furan. Electronic absorption spectra (lover curve) of pheo-phytin a and a' in tetrahydro -fur an.

The CD spectra obtained by us for Chi a and a ' are quite similar to those reported by Prokhorenko et al. [15]. The differences exist mainly in the intensities and fine structures of t he CD bands. A closer examinat ion of the CD spectrum of Chi a in Fig. 1 reveals, however, t h a t the intensity of the band corresponding to the B transi t ion is only half of t h a t reported for t he same band by Prokhorenko et al. [15] and by Houssier and Sauer [14]. This discrepancy is probably due to different pur i ty grades of t he chlorophyll samples used in the measurements. I t is difficult to prepare Chi a com-pletely free f rom Chi a ' and vice versa. Another probable reason for the discrepancy derives f rom the fact t h a t the interconversion between Chi a and a ' is more rapid in diethyl ether t h a n in T H F [11] (this was the reason why we selected T H F for our measurements).

The CD spectra of pheophyt in a and a ' (Fig. 2) show even more pronounced differences than those of Chi a and a ' . Form Fig. 2, it can be observed t h a t CD is positive for both pheophytins a t 533 nm (Q*o-o transition), 411 nm (B transition) and 325 nm (N transi t ion). The CD intensities, however, are

remarkably higher for pheophytin a ' a t 533 nm and 411 nm (Table I). Below 380 nm, the two pheo-phytins exhibit CD bands with opposite signs a t several positions (Table I).

Comparison of the CD spectrum obtained by us for pheophytin a with t h a t measured by Houssier and Sauer [14] for the same derivative in diethyl ether, shows tha t the main difference between the two results exists again a t the Soret band wave-length. The explanation for this discrepancy is similar to tha t presented in the case concerning the differences in the CD spectra of Chi a.

The CD spectra of pheophorbides a and a ' (Table I) are closely similar to those of pheophytins a and a ' . This result is expectable since the asym-metric carbons of the phyty l group (C-7' and C - l l ' ) are too far from the TI electron system of the macro-cycle to induce optical act ivi ty in the macrocycle electronic transitions [14]. The molar ellipticities of pheophorbide a ' are, however, somewhat lower t han the corresponding values obtained for pheophyt in a ' (Table I). In particular, the 0 (or As) values cor-responding to the Qxo-o and B transit ions are clearly lower t h a n those for pheophyt in a ' . This is obviously

due to t h e fac t t ha t the puri ty of the preparat ion used in t h e measurement probably did no t exceed 60%. A pu r i t y of ca. 50% was est imated on the basis of t h e 1 H NMR spectrum measured in acetone-d6. As the inter con version between pheophorbide a and a ' in this solvent is more rapid than in T H F , we es t imate a value of ca. 60% for the pur i ty of the pheophorbide a ' used in the CD measurements. Unfor tuna te ly , there is no good method available a t t he moment for splitting off the phy ty l group f rom chlorophyll a ' . The enolization and epimeriza-t ion a t C-10 is inevitable in the 30% HCl/diethyl ether solvent system used for this purpose, no m a t t e r how low the temperature is kept during the procedure. The other alternatives such as trans-esterification with chlorophyllase (E.C. 3.1.1.14) or by t h e methods of organic chemistry would probably not be a much better choice, as polar organic solvents have to be used also in these methods.

Proton magnetic resonance spectra

Table I I presents a comparison of the *H NMR chemical shifts for chlorophyll a and a', pheophytin a and a ' and pheophorbide a and a ' in acetone-d6. The A <5 values obtained for Chi a and a ' are close to

Table II. Comparison of the *H NMR chemical shifts (<5 [ppm] from internal TMS) of chlorophylls a and a', pheophytins a and a' and pheophorbides a and a' in acetone-dg.

Compounds Assignment «a 68L' AS

Chlorophylls ß-H 9.57 9.56 0.01 l a , 2 a-H 9.21 9.19 0.02

(5-H 8.42 8.37 0.05 10-H 6.11 5.99 0.12 lOb-CHg 3.73 3.70 0.03 5a-CH3 3.49 3.47 0.02 8a-CH3 1.68 1.58 0.10

Pheophytins ß-B. 9.66 9.59 0.07 3 , 4 a-H 9.35 9.29 0.06 3 , 4

<5-H 8.86 8.83 0.03 10-H 6.32 6.22 0.10 lOb-CHs 3.87 3.81 0.06 5a-CH3 3.63 3.58 0.05 3a-CH3 3.12 3.07 0.05 8a-CH3 1.83 1.67 0.16

Pheophorbides ß-B. 9.70 9.65 0.05 5 , 6 a- H 9.39 9.35 0.04

(5-H 8.88 8.85 0.03 10-H 6.35 6.23 0.12 10b-CH3 3.87 3.81 0.06 5a-CH3 3.64 3.60 0.04 3a-CH3 3.15 3.12 0.02 8a-CH3 1.83 1.68 0.15

the values reported previously [11] for these com-pounds, with the exception for the 8a-CH3 group (A <5 = 0.10 ppm). No chemical shift difference could previously be observed for this last-mentioned group owing to the incomplete resolution in the high field region of the measured spectra. The informa-t ion concerning the chemical shift of the 8a-CH3 group is highly desirable in a t tempt ing to disclose conformational alterations very likely to occur in ring IV due the sp3 hybridizations a t C-7 and C-8. The relatively high A <5-value (0.10 ppm) obtained for the 8a-CH3 group shows t h a t the largest con-formational alterations indeed occur where expected.

The A (5-values obtained for the epimeric Mg-free derivatives are clearly higher than the values ob-tained for the epimeric chlorophylls. The highest values are now observed for the 8a-CH3 group. The chemical shifts of the 3a-CH3 group are also different for the Mg-free derivatives while no difference in this respect was observed for the epimeric chloro-phylls. Consistently with the differences observed in t h e CD-spectra, the XH NMR A (5-values indicate t h a t the conformational alterations are more pro-nounced in the Mg-fiee derivatives. Accordingly, the central Mg-atom seems to make the macrocycle more rigid as a whole.

Discussion The great differences observed above for the CD

spectra of the epimeric chlorophylls or their Mg-free derivatives, arise, in par t , f rom the change of configuration a t C-10 and, in par t , f rom conforma-tional al terations which induce chirality in the chlorin macrocycle. The large increase in the in-tensi ty of the CD band corresponding to the B transi t ion in y- subst i tuted chlorins has been inter-preted as arising f rom the steric repulsion between the bulky subst i tuents a t C-y and C-7 [14, 21]. Our results support this conclusion showing the enhance-ment effect very clearly in the case of the C-10 epimeric Mg-free derivatives.

The *H NMR chemical shift differences given in Table I I provide evidence for the fact t ha t con-formational al terations (puckering) take place in the whole macrocycle when the epimerization a t C-10 occurs (compare to Ref. [6]). The changes in the geometry are, however, most prominent in ring IV.

There are two principal factors determining the conformation of the chlorin macrocycle: (1) the

TT-orbital overlap and resonance energy should be kept to a maximum. This means actually a trend to planarity; (2) the steric strain arising from the peripheral bulky substituents, should be kept to a minimum. Steric strain is relieved by conformational alterations (twists, distortions, puckering) in the macrocycle and, consequently, deviations from planarity occur. These alterations take place more easily in chlorins than in porphyrins, as in the former group flexibility is considerably enhanced by the sp3 hybridization of C-7 and C-8.

The results from the X-ray diffraction analysis for methylpheophorbide a [9] show tha t C-7 is bent up by 0.29 Ä and C-10 is pushed down by 0.14 Ä from the least square plane (see Fig. 4 in Ref. [9]) resulting in a dihedral angle Cio-Cy-Ci7-C- of 16° [6]. The corresponding values for chlorophylls a and b can be estimated to be much lower on the basis of the X-ray analysis by Strouse et al. [8]. The latter values seem to be approximately of the same order of magnitude as those reported for vanadyl desoxo-erythroetioporphyrin (0.11 Ä, 0.05 Ä and 5°, respec-tively) [22]. Accordingly, the metal atom coordinated at the center of the molecule, makes the macrocycle more rigid. The results from the above X-ray analysis are analogous to our results obtained by CD and XH NMR for Mg-free and Mg-coordinated chlorophyll derivatives.

On the other hand, however, the introduction of a metal atom to the center of the molecule, brings about a possibility for a new type of isomerism: the metal atom can be in the plane (e.g. hexacoordinated chlorophyll molecules) or out of the plane (e.g. pentacoordinated chlorophyll species). This new kind of isomerism has gained attention only recently [5]. I t should be explored in more detail, because it certainly effects the conformation of the macrocycle in an essential way.

The observation tha t the methine proton chemical shifts of the prime derivatives are located at higher field as compared to the corresponding values for the derivatives with the "natural" configuration at C-10, can be interpreted as implying slightly higher electron densities at the methine bridges of the prime derivatives. This fact, if it proves to be correct, could be very important, since it would mean a higher susceptibility to electrophilic sub-stitution [23] or possibly a change in some other chemical property of the macrocycle. These ideas are not new. Woodward [7] first focused attention

on the important correlation between stereochem-istry and chemical reactivity in asymmetrically substituted porphyrins and chlorins. Woodward's concept of an "overcrowded periphery" was basic in his stragedy for the total synthesis of Chi a and, subsequently, it has been supported by several investigators [5].

The comparison of our CD spectra with those obtained by Wolf et al. [24, 26] for stereoisomer^ 10-methoxy-methylpheophorbides and 10-methoxy-pyromethylpheophorbides, reveals some interesting aspects in respect to the correlation of the signs of the CD bands with the configurations at the asym-metric carbons. For these pheophorbides, Wolf and Scheer [6] have presented the following two argu-ments: (1) The more space-filling substituent deter-mines the sign of the CD bands, i.e. the propionic methyl ester side chain at C-7 and not the C-8 methyl group; (2) the sign of the CD band at 285 nm is clearly determined by the configuration of the 10-methoxy group (negative in the 10(R) and posi-tive in 10(S) series, irrespective of the presence of the carbomethoxy substituent). The examination of the CD spectra in Figs. 1 and 2 clearly shows tha t the argument (2) is not valid for the C-10 epimeric chlorophylls and Mg-free derivatives used in our measurements. The sign of the CD band at ea. 280 nm is positive for 10(R)-chlorophyll a ( la), 10(R)-pheophytin a (3) and 10(R)-pheophorbide a (5), whereas it is negative for the 10(S) configurations of these derivatives (2, 4 and 6). The argument (1) concerning the dominant effect of the space-filling propionic ester side chain at C-7 is valid for the CD band corresponding to the B transition, i.e. this band is positive for all 7(S) derivatives irrespective of the configurations at C-8 and C-10. However, the extension of this argument also to the CD bands in the UV region (210-400 nm) leads to difficulties in the interpretation. According to our results, the signs and intensities of these CD bands are clearly effected also by the nature and configuration of the substituents at C-10 relative to those at C-7, i.e. the space-filling effect of the methoxy carbonyl group at C-10 should also be taken into account. When the propionic ester group at C-7 and the methoxy carbonyl group at C-10 become positioned on the same side of the ring plane, a considerable amount of steric strain results. This leads to con-formational chirality which enhances the intensities of most CD bands. The more possibilities there are

R3 _ C20H39 = phytyl

l a : R i = H, R2 = CO2CH3, R 4 = CH3 Chlorophyll a = 10(R) Chi a

l b : R1 = H, R2 = C02CH3, R4 = CHO Chlorophyll b = 10(R) Chi b

2: R 1 = CO2CH3, R 2 = H, R 4 = CH3 Chlorophyll a' = 10(S) Chi a

3: R1 = H, R2 = CO2CH3, R 3 = phytyl 10(R) Pheophytin a;

4: R1 = CO2CH3, R2 = H, R 3 = phytyl 10(S) Pheophytin a = pheophytin a'

0: R 1 = H , R 2 = CO2CH3, R 3 - H 10(R) Pheophorbide a

6: Ri = CO2CH3, R 2 = H, R 3 = H 10(S) Pheophorbide a = pheophorbide a'

for t he relief of the steric strain through conforma-t ional alterations, the more pronounced is this enhancement (compare Figs. 1 and 2). Nevertheless, t h e conformational alterations also have a distinct effect on the signs and positions of the CD bands in t h e UV region. For the above reasons, the addit ivi ty of group effects [6] is no longer valid for the CD spectra of the 10(S) derivatives 2, 4 and 6.

Experimental Electronic absorption spectra

The electronic absorption spectra were recorded on a Cary 15 spectrophotometer a t 25 °C. The spectra were measured using the spectropolarimeter cells with fused quartz windows (b = 0.1 cm).

Circular dichroism spectra The CD spectra were recorded on a Cary 61

spectropolarimeter. Cells with fused quartz windows and a pathlength of 0.1 cm were employed. The ins t rument was calibrated with an aqueous solution of D-(+)-camphorsulphonic acid ( c = l mg/ml) a t 290 nm (6 = 0.31 degrees). The full scale of the recorder paper corresponded to an ellipticity of 0.05 degrees. The spectra were recorded a t 17 °C

using a scan speed of 0.1 nm/s. Baselines were also recorded and substracted from the sample spectra. Due to ins t rumental limitations (lack of a red-sensitive photomultiplier), the spectra could not be recorded beyond 600 nm.

Molar ellipticities, [0], were calculated from the equat ion

where 0 is the measured ellipticity in degrees, b is t he optical pa th length in cm and M is the molar concentration, mol • 1_1. Circular dichroism values, ZU = £L—£R, were obtained from the relationship

[0] = 3298-Zle.

Proton magnetic resonance spectra The XH NMR spectra were measured a t ambient

tempera ture with a Jeol FX-60 P F T instrument . F ive m m sample tubes were employed. The solvent was acetone-de containing te t ramethyl silane (TMS) as an internal reference. The sample concentration was in the range 1.5 x 10~2 to 1.7 X 10~2 mol • H .

Purity of solvents and reagents The t e t r ahydrofuran (THF) used in the measure-

ment of the electronic absorption spectra and CD spectra, was of spectroscopic grade pur i ty (Merck,

Darmstadt). I t was used without further purifica-tion. Other solvents and reagents were of reagent grade purity and were also used without further purification.

Chlorophyll a (la) Chi a was isolated from frozen clover leaves by

the method described previously [12]. Repeated precipitation of the chlorophyll as its water adduct after the separation on a sucrose column, yielded a preparation free from colourless galacto- and other lipids as well as from chlorophyll a ' and b' . TLC on sucrose [27] showed only one spot. UV/VIS and CD: Table I. *H NMR (60 MHz, acetone-de/TMSmt) <5 [ppm]: 9.57 (s, 1H, ß-H); 9.21 (s, 1H, a-H); 8.42 (s, 1H, <5-H); 8.03 (dd, 1H, «7= 12Hz, 18Hz, 2 a - H x ) ; 6.10 (dd, 1H, «7 = 2 Hz, 18 Hz, 2b-H B ) ; 5.89 (dd, 1H, J = 2 Hz, 12 Hz, 2b-HA) ; 6.11 (s, 1H, 10-H); 4.84 (t, 1H, 2'-H); 4.40 (q, 1H, l ' -H) ; 4.21 (m, 1H, 8-H); 4.03 (m, 1H, 7-H); 3.73 (s, 3 H , 10b-CH3); 3.66 (q, 2H, «7 = 7 Hz, 4a-CH2); 3.49 (s, 3H, 5a-CH3); 3.25 (s, 3H, 1 a-CH3); 3.18 (s, 3H, 3a-CH3); 2.35 (m, 4H, 7a, 7b-CH2); 1.76 (t, 2H, «7 = 7 Hz, 4'-CH2); 1.68 (d, 3H, «7 = 7 Hz, 8a-CH3); 1.60 (t, 3H, «7 = 7 Hz, 4b-CH3); 1.44 (s, 3H, 3a'-CH3); 1.21 (s, 3H, 7'-15'-CH); 1.17 (s, broad, 18H, 4'-14'-CH2); 0.785, 0.756, 0.723,0.694 (s, s, s, s, 12H, 7a'-16'-CH3).

Chlorophyll a' (2) Chi a ' was prepared from purified 1 a as described

elsewhere [11, 27]. TLC on sucrose revealed the presence of only one component. UV/VIS and CD: Table I. XH NMR: Table II . In other respects the XH NMR spectrum was identical with tha t of 1 a.

Pheophytin a (3) Pheophytin a was prepared from purified l a by

shaking it for 10 min with cold (0 °C) 12% aqueous HCl-diethyl ether in a separatory funnel (equal phase volumes). The ether solution was washed three times with cold distilled water and evaporated to dryness. TLC on cellulose [28] yielded only one spot. UV/VIS and CD: Table I. *H NMR (60 MHz, acetone-de/TMSmt) ö [ppm]: 9.66 (s, 1H, ß-H); 9.35 (s, 1H, a-H); 8.86 (s, 1H, <5-H); 8.05 (dd, 1H, J = 12 Hz, 18 Hz, 2a -H x ) ; 6.29 (dd, 1H, «7 = 2 Hz, 18 Hz, 2b-HB ) ; 6.14 (dd, 1H, «7 = 2 Hz, 12 Hz, 2b-HA) ; 6.32 (s, 1H, 10-H); 5.21 (t, 1H, «7 = 7 Hz, 2'-H); 4.45 (m, 3H, l ' -H, 8H, 7H) ; 3.87 (s, 3H, 10b-CH3); 3.63 (s, 3H, 5a-CH3); 3.48 (q, 2H, 4a-CH2); 3.40 (s, 3H, 1 a-CH3); 3.12 (s, 3H, 3a-CH3); 2.38 (m, 4H, 7a, 7b-CH2); 1.83 (d, 3H, «7 = 7 Hz, 8a-CH3); 1.60 (t, 3H, «7 = 7 Hz, 4b-CH3); 1.48 (s, 3H, 3a'-CH3); 1.27 (s, 3H, 7'-15'-CH); 1.09 (s, broad, 18H, 4'-14'-CH2); 0.849, 0.754, 0.649 (s, s, s, 12H, 7a'-16'-CH3).

Pheophytin a' (4) Pheophytin a ' was prepared from 2 analogously

to the preparation of 3 from l a . The aqueous

HCl-diethyl ether solvent system was cooled below 0 °C prioi to the treatment. The ether solution was rapidly washed three times with cold distilled water and evaporated to dryness. TLC on cellulose re-vealed only one component. The purity of the pheophytin a ' prepared by this means was estimated to be ca. 70% on the basis of the XH NMR spectrum measured in acetone-dß at room temperature. The purity was probably higher than this, as the inter-conversion between 3 and 4 is quite rapid in acetone-dö at RT. We therefore estimate tha t the purity of the preparation used in the CD measure-ment was 80 to 90%. UV/VIS and CD: Table I. m NMR: Table I I . In other respects the m NMR spectrum was identical with that of 3.

Pheophorbide a (5) Pheophorbide a was prepared from purified l a

by shaking it for 30 min with 30% aqueous HCl-diethyl ether in a separatory funnel. The solvent system (equal phase volumes) was cooled below 0 °C prior to the treatment. The small amount of pig-ments remaining in the ether phase after 30 min was discarded. The pigments in the acid phase were transferred to fresh ether by dilution and neutraliza-tion of the lower phase. Pheophorbide a was extracted from the ether solution with 16% (w/w) hydrochloric acid and transferred again to fresh ether by dilution and neutralization of the lower phase. After washing, the ether extract was evap-orated to dryness. The product yielded only one spot in TLC on cellulose. UV/VIS and CD: Table I. m NMR (60 MHz, acetone-d6/TMSi0t) 5 [ppm]: 9.70 (s, 1H, ß-H); 9.39 (s, 1H, a-H); 8.88 (s, 1H, <5-H); 8.10 (dd, 1H, «7= 12 Hz, 18 Hz, 2a -H x ) ; 6.31 (dd, 1H, J = 2 Hz, 18 Hz, 2b-H B ) ; 6.16 (dd, 1H, «7 = 2 Hz, 12 Hz, 2b-HA); 6.35 (s, 1H, 10-H); 4.64 (m, 1H, «7 = 7 Hz, 7-H); 4.18 (m, 1H, <7 = 7 Hz, 8-H); 3.87 (s, 3H, 10b-CH3); 3.64 (s, 3H, 5a-CH3) ; 3.60 (q, 2H, «7=7 Hz, 4a-CH2); 3.42 (s, 3H, la-CH3) ; 3.15 (s, 3H, 3a-CH3); 2.35 (m, 4H, 7a, 7b-CH2); 1.83 (d, 3H, «7 = 7 Hz, 8a-CH3); 1.62 (t, 3H, «7 = 8 Hz, 4b-CH3); 0.904 (s, broad, NH); —1.815 (s, broad, NH).

Pheophorbide a' (6) Pheophorbide a ' was prepared from 2 analogously

to the preparation of 5 from 1 a with the exception tha t the shaking time was only 10 min. TLC on cellulose yielded one spot. UV/VIS and CD: Table I. m NMR: Table I I . In other respects the *H NMR spectrum was identical with that of 5. On the basis of the XH NMR spectrum measured in acetone-de, the purity achieved was ca. 50%. The principal impurity was 5.

This work was supported by the Research Council for the Natural Sciences of the Academy of Finland. We wish to thank Dr. S. Lötjönen for the measure-ment of the *H NMR spectra.

[1] H. Ficher, O. Süs, and G. Klebs, Liebigs Ann. Chem. 490, 38 (1931).

[2] H. Fischer and H. Siebel, Liebigs Ann. Chem. 499, 84 (1932).

[3] H. Brockmann (Jr.) and J. Bode, Liebigs Ann. Chem. 1974, 1017.

[4] H. Brockmann (Jr.), in D. Dolphin (ed.): The Porphyrins, Vol. H, p. 288, Academic Press, New York 1978.

[5] H. Scheer, in D. Dolphin (ed.): The Porphyrins, Vol. II, p. 1, Academic Press, New York 1978.

[6] H. Wolf and H. Scheer, Ann. N. Y. Acad. Sei. 206, 549 (1973).

[7] R. B. Woodward, Pure Appl. Chem. 2, 383 (1961). [8] H.-C. Chow, R. Serlin, and C. E. Strouse, J. Am.

Chem. Soc. 97, 7230 (1975). [9] M. S. Fischer, D. H. Templeton, A. Zalkin, and

M. Calvin, J. Am. Chem. Soc. 94, 3613 (1972). [10] J. J. Katz and J. R. Norris, Curr. Top. Bioenerg.

5, 41 (1973). [11] P. H. Hynninen, M. R. Wasielewski, and J. J.

Katz, Acta Chem. Scand. B 33, 637 (1979). [12] P. H. Hynninen, Acta Chem. Scand. B 31, 829

(1977). [13] P. H. Hynninen, Acta Chem. Scand. 27, 1487

(1973). [14] C. Houssier and K. Sauer, J. Am. Chem. Soc. 92,

773 (1970). [15] I. R. Prokhorenko, V. M. Lobachev, and V. M.

Kutyurin, Z. Obsc. Chim. 46, 2147 (1976).

[16] T. A. Evans and J. J. Katz, Biochim. Biophys. Acta 396, 414 (1975).

[17] L. L. Shipman, T. M. Cotton, J. R. Norris, and J. J. Katz, J. Am. Chem. Soc. 98, 8222 (1976).

[18] M. Gouterman, in D. Dolphin (ed.): The Por-phyrins, Vol. III, p. 1, Academic Press, New York 1978.

[19] C. Weiss, in D. Dolphin (ed.): The Porphyrins, Vol. I l l , p. 211, Academic Press, New York 1978.

[20] J. D. Petke, G. M. Maggiora, L. L. Shipman, and R. E. Christoffersen, Photochem. Photobiol. 30, 203 (1979).

[21] B. Briat, D. A. Schooley, R. Records, E. Bunnen-berg, and C. Djerassi, J. Am. Chem. Soc. 89, 6170 (1967).

[22] R. C. Pettersen and L. E. Alexander, J. Am. Chem. Soc. 90, 3873 (1968).

[23] P. H. Hynninen and S. Lötjönen, Tetrahedr. Lett. 22, 1845 (1981).

[24] H. Wolf, H. Brockmann (Jr.), H. Biere, and H. H. Inhoffen, Liebigs Ann. Chem. 704, 208 (1967).

[25] H. Wolf, H. Brockmann (Jr.), I. Richter, C.-D. Mengler, and H. H. Inhoffen, Liebigs Ann. Chem. 718, 162 (1968).

[26] H. Wolf, I. Richter, and H. H. Inhoffen, Liebigs Ann. Chem. 725, 177 (1969).

[27] P. H. Hynninen, to be published. [28] G. Sievers and P. H. Hynninen, J. Chromatogr.

134, 359 (1977).