BIoINoRGANK’
CHEMSTR Y 6,77-82
(1976)
77
Magnetic Circular Dichroism of Protoporphyrin Derivatives in the Ultraviolet Region TORU SHIMIZU, TSUNENORI
Chemical Research Sendai 980, Japan.
Institute
NOZAWA and MASAHIRO
of Non-Aqueous
Solutions,
HATANO
Tohoku
University,
ABSTRACT The MCD spectra of ferri- and ferro-heme in high and low spin states and zinc porphyrin for reference have been measured in the ultraviolet region. Zinc porphyrin offered an A term at 325 nm attributable to the N transition. A ferrous lowspin complex with CO or CN exhibited MCD spectra composed of B terms around 325 nm This striking difference from that of zinc porphyrin in the uItravioIet region is noted from the fact that their Soret and vistbte MCD’s gave A terms similar to those of zinc porphyrin A ferric complex with high spin had MCD spectrum different from that of low spin, the former showing a negative C term at 287 nm and a B term at 339 run, but the fatter positive C terms at 275 nm and 320 nm Some of these Faraday parameters were as-bed to charge-transfer transitions.
INTRODUCTION It has been reported that the MCD spectra in the Soret and visible region of hemoproteins are very sensitive to the spin state of the heme iron [l-S] _ However, no MCD of the hemoproteiu in the ultraviolet region with respect to its redox and spin states has been reported so far, possibly because of the presence of some aromatic amino acids such as tyrosiue and tryptophane etc. which have MCD and absorption bands in this region [6,7] _The purposes of this work are to investigate the effect of the spin and oxidation states on the MCD spectra of heme in the ultraviolet region and to try to assign the bands by determining the Faraday parameters of the transitions from the temperature variation experiments on MCD. We used zinc porphyrin for reference. As become clear in many reports 18,91 the MCD has three components, conventionally called A, B and C terms. Each term has a direct relation with the ground and excited electronic states of the transition. The A term which is nonzero if the ground or excited state is degenerate can be distinguished from B and C terms from the characteristic S shaped band of the A term. B and C terms which are both bell shaped can be separated by their different temperature dependence_ Thus, the C term, which originates from the population differences in the ground state sublevels separated by a magnetic fieId, is temperature dependent, while the B term is not. @American
Elsevier Publishing Company,
Inc., 1976.
78
MASAHIRO
HATANO
et al_
EXPERIMENTAL Hemin was supplied from Daiichi-Pharmaceutical CO., Ltd. The other compounds used in this investigation were all commercially available guaranteed grade reagents and were used without further purification. Zinc hematoporphyrin dimethylester was synthesized by the method of Falk [I 01. Heme iron complex was dissolved in 50% aqueous ethanol solution in order to prevent it from aggregation and dimerization [ 1 1 ] _ Low temperature MCD and absorption spectra were measured in 50% aqueous glycerol solution. Absorption spectra were obtained on a Hitachi Model EPS-3T spectrophotometer using a cuvette of 2 or 1 mm optical path length. MCD spectra were recorded on a JASCO J-20A spectropolarimeter equipped with a JASCO electromagnet to produce a longitudinal magnetic field at the sample up to 11.4 kG. A cell of 1 or 2 mm path equipped with a Cu-(Au-Co) thermocouple was employed for the measurements below O” down to - 120°; the cell was placed in a quartz Dewar and the temperature was regulated by a stream of cold nitrogen gas-
RESULTS
AND
DISCUSSION
Zinc hematoporphyrin dimethylester would reflect the electronic state of porphyrin per se with Dab symmetry since zinc (II) has no vacant 3d electron orbital which exerts crucial influences on the electronic state of porphyrin_ Thus its MCD would provide a valuable comparison for that of iron porphyrin_ Zinc porphyrin exhibited peaks at 397 run and 310 nm and troughs at 355 nm, 342 nmand244nmasshowninFig l.Thepeakat3lOnmandthetroughat342 nm seem to compose one A term around 325 nm. On the basis of SCFMO-PPP calculation, Weiss et al [ 123 pointed out that porphyrin with D4h symmetry has the L and N bands due to e&rr) f a’ ,JT) and e,(n) + bau(rr) transitions, respectively, which are expected to give A terms in this spectral region because of the degeneracy of the excited state. Hence the S shaped band will be assigned to the A term associated with the N band from the observed position of the band [ 12, I3] _ The trough at 355 nm may possibly be a vibrational component of the Soret transition_ A ferrous lowspin complex such as ferrous protoporphyrin-CO or CN complex has MCD spectrum similar to that of zinc porphyrin in the Soret and visible region (not shown) and was expected to exhibit an A term for the N transItion of porphyrin in the ultraviolet region. However, MCD’s of the CO and CN compIexes are strikingly different from that of zinc porphyrin_ Fig_ 2(A) shows the MCD (upper) and absorption (below) spectra for ferroheme-CO as a ferrous complex with Iow spin at room and lower temperatures. The reduction from Fe3* to Fe” was affected by sodium dithionite. Though the absorption peak at about 330 nm is due to the sodium dithionite present in the solution, it has no MCD band at this sample concentration in this spectral region Since all troughs and peaks showed little temperature dependence, the C term contri-
MCD OF HEME
79
FIG. 1. MCD (upper) and absorption (lower) spectra of zinc-hematoporphyrin dimethylester in dioxane solution at room temperature.
of the MCD bands was estimated to be almost negligible at room temperature and can be assigned to mainly B and/or A terms The lack of A term around 325 nm suggests the absence of the transition with a degenerate excited state in this spectral region. Axial ligands and ferrous low-spin iron will exert some influences to the electronic states of porphyrin, so that they might remove the degeneracy of the excited state or change the transition energies of the L and N bands to lower energies than 300 nm_ In fact, it may be possible to assign the trough at 295 nm and the peak at 270 nm in the MCD of ferroheme-CO to a Faraday A term for the shifted L or N band. Figure 2(B) shows the MCD and absorption spectra for ferroheme in a 50% aqueous ethanol solution, in which a ferrous high-spin complex forms. The trough around 380 nm may be assigned to a component of the Soret MCD, and thus will be a C term C3,Sl. Another possible assignment may be a porphyrin-to-iron charge-transfer transition_ bution
. 80
MASAHIRO
HATANO et al_
FIG_ 2. MCD (upper) and-absorption (lower) spectra of Fe* - protoporphyrin complex with (A) and without (El) CO in 50% aqueous ethanol solution at room temperature (solid line). Broken line in (A) indicates MCD spectrum in 50% aqueous glycerol solution at 158 OK, Reduction was performed by adding a few mg of sodium dithionite, and the CO complex by gently bubbiing CO gas through the reduced heme solution for 20 sec. The absorption spectra were drawn with a broken line because they did not show true spectra from the presence of dithionite band around 330 nm.
Ferr5heme cyanide, a ferric low-spin complex, showed three troughs at 275 run, 320 run and 37’C nm in the waveiength region from 250 nm to 380 nm (Fig. 3(A)). The magnitude of troughs at 275 mn and 320 nm had fairly distinct temperature dependences with linear relations to I/kT (k: Boltzmann constant, T: absohrte temperature) as shown in the inset in Fig_ 3(A)_ The absorption spectrum, on the other hand, remained unchanged even at low temperatures_ Therefore, they are assigned to the Faraday C terms. On the other hand, the trough at 370 nm was found to have a B term contribution to some extent, since its magnitude does not increase with decreasing temperature as much as the other two bands. One plausibIe candidate for the bands at 275 nm and 320 nm is the charge-transfer band such as blg(dxz_Yz) fbzu(n) (at higher energy) or ar,(dz2> +~‘~~(?r) (at lower energy). They are transitions from ‘Eg to 2EU under the D4h symmetry_ Even if ferriheme cyanide would have a symmetry Iower than D4h as evidenced by the anisotropy in EPR g values for ferriheme azide 114 1, they at Ieast have spin degeneracy 15, 14, 15 1. Hence, these charge-transfer bands should exhibit a paramagnetic C term due to a spin degeneracy, which is in agreement with the- experimental results. Another band at 370 MI may have possibIy been assigned to a component for a vibrational band of the Soret transition_
MCD OF HEME
81
cer, O-
$ -2 \ -4 -
l....l....f..
250
IiAacEa,
390
FIG. 3. MCD (upper) and absorption (lower) spectra of Fe”*- protoporphyrin compIex with (A) and without (B) CN in 50% aqueous ethanol solution at room temperature (solid line). The CN complex was formed by the addition of excess
solid KCN. Broken lines in (A) and (B) show MCD spectra at 155 “K and 77 “K, respectively, in 50% aqueous glycerol solutioa The insets show the dependence of
the extrema indicated on l/kT_
A ferric high-spin complex made of feniheme in a 50% aqueous ethanol solution showed a peak at 275 nm, a peak at 340 nm, a small trough at 3 15 nm and a trough at 367 nm at ro‘om temperature (Fig. SB)). Attention should be
paid to the fact that the M-CD spectrum for the ferric high-spin complex is significan!ly different from that for the ferric low-spin complex. This high MCD sensitivity to spin states should be compared with the relatively small dependence of the electronic absorption spectra on the spin states [16, 171. Decrease of the sample temperature intensified the peak at 285 nm and the trough at 3 15 nm, while the peak at 340 run being attenuated. At liquid-nitrogen temperature, an MCD peak at 355 nm appeared together with the disappearance of the 367~nm trough. The drastic change of the MCD spectra with temperature variation might suggest the spin state change from high to low with decreasing temperature, as reported for some high-spin ferric hemoproteins (18, 13]_ However, the fact that the resultant MCD spectrum does not Tesemble that of low-spin excludes this possibility, The peak around 287 nm is estimated to be mainly C term, while the peak around 355 nm, hidden at room temperature, appeared and made the trough at 367 nm disappear at liquid-nitrogen temperature. Although it is difficult to definitely determine numbers and
MASAHIRO
82
HATANO
et al.
of the transition in this region even with the aid of the MCD, it can be argued, at least, that there are more than five transitions in this region and three transitions among them have strong C term characterIn conclusion, despite lack of rigid interpretation of the magneio-opticai effects, striking differences between ferric and ferrous with low and high spin complexes may serve as a useful working hypothesis in further pursuit of the electronic structure of heme complexes.
positions
REFERENCES 1. J. Bolard and A_ Gamier, Biochim Biophys. Acfa 263,535 (1972). 2 J. C_ Sutherland and M. P. Klein, J. Chem Phys 57,76 (1972). 3_ J_ 1.Treu and J. J. HopfieId,J. Chem Whys 63,613 (1975). 4. T_ Shimizu, T_ Nozawa, M. Hatano, Y. Imai and R_ Sato, Biochemk?y
14, 4172
(1975). 5. L. Vickery, T_ Nozawa and K. Sauer,J. Am Chem Sot. 98,343,351 (1976). 6. G. Barth, W. Voelter, E. Bunnenberg and C. Djerassi, J. Am Chem Sot. 94, 1293 (1972). 7_ B. Hohuquist and B. L_ VaBee, Biochemistry 22,909 (1973). 8. A’. B. Buckingham and P. J. Stephens, Ann. Rev. Phys_ Chem. 17,399 (1966). 9. P. N. Schatz and A. J. McCaffery, Quart. Rev. Chem Sot. 23,552 (1969). Elsevier,Amsterdam (1964). 10. J. E. Falk,PorphytinsandMetalloporphynizs, ll_ T_ H. Davies,Biochim. Biophyz Acta 329,108 (1973). 12_ C. Weiss, H_ Kobayashi and M_ Gouterman, J. MoL Spect_ 16,415 (1965). 13. M. Zemer, hf. Gouterman and H. Kobayasbi, Theoret Chim. Acta 6,363 (1966). 14_ M_ Weissbluth, Srruct. Bond. (Berifne) 2,1 (1967). 15. B. Briat, D. Berger and hf. Lehboux,J. Chem. Whys 57,5606 (1972). 16. P_ George, J. Beetlestone and J. S. Griffith, Rev. Mod. Phys. 36,441 (1964). 17_ G. I_ Hanania, A_ Yeghiayan and B. F_ Cameron, Biochem J. 98,189 (1966). 18. T. IizuIcaand M. Kotani, Biochim Biophys Acta 181,275 (1969). 19. T. Iizuka and M. Kotani, Biochim Biophys. Acta 194,351 (1969). Received I9 November
I975
CHEMSTR Y 6,77-82
(1976)
77
Magnetic Circular Dichroism of Protoporphyrin Derivatives in the Ultraviolet Region TORU SHIMIZU, TSUNENORI
Chemical Research Sendai 980, Japan.
Institute
NOZAWA and MASAHIRO
of Non-Aqueous
Solutions,
HATANO
Tohoku
University,
ABSTRACT The MCD spectra of ferri- and ferro-heme in high and low spin states and zinc porphyrin for reference have been measured in the ultraviolet region. Zinc porphyrin offered an A term at 325 nm attributable to the N transition. A ferrous lowspin complex with CO or CN exhibited MCD spectra composed of B terms around 325 nm This striking difference from that of zinc porphyrin in the uItravioIet region is noted from the fact that their Soret and vistbte MCD’s gave A terms similar to those of zinc porphyrin A ferric complex with high spin had MCD spectrum different from that of low spin, the former showing a negative C term at 287 nm and a B term at 339 run, but the fatter positive C terms at 275 nm and 320 nm Some of these Faraday parameters were as-bed to charge-transfer transitions.
INTRODUCTION It has been reported that the MCD spectra in the Soret and visible region of hemoproteins are very sensitive to the spin state of the heme iron [l-S] _ However, no MCD of the hemoproteiu in the ultraviolet region with respect to its redox and spin states has been reported so far, possibly because of the presence of some aromatic amino acids such as tyrosiue and tryptophane etc. which have MCD and absorption bands in this region [6,7] _The purposes of this work are to investigate the effect of the spin and oxidation states on the MCD spectra of heme in the ultraviolet region and to try to assign the bands by determining the Faraday parameters of the transitions from the temperature variation experiments on MCD. We used zinc porphyrin for reference. As become clear in many reports 18,91 the MCD has three components, conventionally called A, B and C terms. Each term has a direct relation with the ground and excited electronic states of the transition. The A term which is nonzero if the ground or excited state is degenerate can be distinguished from B and C terms from the characteristic S shaped band of the A term. B and C terms which are both bell shaped can be separated by their different temperature dependence_ Thus, the C term, which originates from the population differences in the ground state sublevels separated by a magnetic fieId, is temperature dependent, while the B term is not. @American
Elsevier Publishing Company,
Inc., 1976.
78
MASAHIRO
HATANO
et al_
EXPERIMENTAL Hemin was supplied from Daiichi-Pharmaceutical CO., Ltd. The other compounds used in this investigation were all commercially available guaranteed grade reagents and were used without further purification. Zinc hematoporphyrin dimethylester was synthesized by the method of Falk [I 01. Heme iron complex was dissolved in 50% aqueous ethanol solution in order to prevent it from aggregation and dimerization [ 1 1 ] _ Low temperature MCD and absorption spectra were measured in 50% aqueous glycerol solution. Absorption spectra were obtained on a Hitachi Model EPS-3T spectrophotometer using a cuvette of 2 or 1 mm optical path length. MCD spectra were recorded on a JASCO J-20A spectropolarimeter equipped with a JASCO electromagnet to produce a longitudinal magnetic field at the sample up to 11.4 kG. A cell of 1 or 2 mm path equipped with a Cu-(Au-Co) thermocouple was employed for the measurements below O” down to - 120°; the cell was placed in a quartz Dewar and the temperature was regulated by a stream of cold nitrogen gas-
RESULTS
AND
DISCUSSION
Zinc hematoporphyrin dimethylester would reflect the electronic state of porphyrin per se with Dab symmetry since zinc (II) has no vacant 3d electron orbital which exerts crucial influences on the electronic state of porphyrin_ Thus its MCD would provide a valuable comparison for that of iron porphyrin_ Zinc porphyrin exhibited peaks at 397 run and 310 nm and troughs at 355 nm, 342 nmand244nmasshowninFig l.Thepeakat3lOnmandthetroughat342 nm seem to compose one A term around 325 nm. On the basis of SCFMO-PPP calculation, Weiss et al [ 123 pointed out that porphyrin with D4h symmetry has the L and N bands due to e&rr) f a’ ,JT) and e,(n) + bau(rr) transitions, respectively, which are expected to give A terms in this spectral region because of the degeneracy of the excited state. Hence the S shaped band will be assigned to the A term associated with the N band from the observed position of the band [ 12, I3] _ The trough at 355 nm may possibly be a vibrational component of the Soret transition_ A ferrous lowspin complex such as ferrous protoporphyrin-CO or CN complex has MCD spectrum similar to that of zinc porphyrin in the Soret and visible region (not shown) and was expected to exhibit an A term for the N transItion of porphyrin in the ultraviolet region. However, MCD’s of the CO and CN compIexes are strikingly different from that of zinc porphyrin_ Fig_ 2(A) shows the MCD (upper) and absorption (below) spectra for ferroheme-CO as a ferrous complex with Iow spin at room and lower temperatures. The reduction from Fe3* to Fe” was affected by sodium dithionite. Though the absorption peak at about 330 nm is due to the sodium dithionite present in the solution, it has no MCD band at this sample concentration in this spectral region Since all troughs and peaks showed little temperature dependence, the C term contri-
MCD OF HEME
79
FIG. 1. MCD (upper) and absorption (lower) spectra of zinc-hematoporphyrin dimethylester in dioxane solution at room temperature.
of the MCD bands was estimated to be almost negligible at room temperature and can be assigned to mainly B and/or A terms The lack of A term around 325 nm suggests the absence of the transition with a degenerate excited state in this spectral region. Axial ligands and ferrous low-spin iron will exert some influences to the electronic states of porphyrin, so that they might remove the degeneracy of the excited state or change the transition energies of the L and N bands to lower energies than 300 nm_ In fact, it may be possible to assign the trough at 295 nm and the peak at 270 nm in the MCD of ferroheme-CO to a Faraday A term for the shifted L or N band. Figure 2(B) shows the MCD and absorption spectra for ferroheme in a 50% aqueous ethanol solution, in which a ferrous high-spin complex forms. The trough around 380 nm may be assigned to a component of the Soret MCD, and thus will be a C term C3,Sl. Another possible assignment may be a porphyrin-to-iron charge-transfer transition_ bution
. 80
MASAHIRO
HATANO et al_
FIG_ 2. MCD (upper) and-absorption (lower) spectra of Fe* - protoporphyrin complex with (A) and without (El) CO in 50% aqueous ethanol solution at room temperature (solid line). Broken line in (A) indicates MCD spectrum in 50% aqueous glycerol solution at 158 OK, Reduction was performed by adding a few mg of sodium dithionite, and the CO complex by gently bubbiing CO gas through the reduced heme solution for 20 sec. The absorption spectra were drawn with a broken line because they did not show true spectra from the presence of dithionite band around 330 nm.
Ferr5heme cyanide, a ferric low-spin complex, showed three troughs at 275 run, 320 run and 37’C nm in the waveiength region from 250 nm to 380 nm (Fig. 3(A)). The magnitude of troughs at 275 mn and 320 nm had fairly distinct temperature dependences with linear relations to I/kT (k: Boltzmann constant, T: absohrte temperature) as shown in the inset in Fig_ 3(A)_ The absorption spectrum, on the other hand, remained unchanged even at low temperatures_ Therefore, they are assigned to the Faraday C terms. On the other hand, the trough at 370 nm was found to have a B term contribution to some extent, since its magnitude does not increase with decreasing temperature as much as the other two bands. One plausibIe candidate for the bands at 275 nm and 320 nm is the charge-transfer band such as blg(dxz_Yz) fbzu(n) (at higher energy) or ar,(dz2> +~‘~~(?r) (at lower energy). They are transitions from ‘Eg to 2EU under the D4h symmetry_ Even if ferriheme cyanide would have a symmetry Iower than D4h as evidenced by the anisotropy in EPR g values for ferriheme azide 114 1, they at Ieast have spin degeneracy 15, 14, 15 1. Hence, these charge-transfer bands should exhibit a paramagnetic C term due to a spin degeneracy, which is in agreement with the- experimental results. Another band at 370 MI may have possibIy been assigned to a component for a vibrational band of the Soret transition_
MCD OF HEME
81
cer, O-
$ -2 \ -4 -
l....l....f..
250
IiAacEa,
390
FIG. 3. MCD (upper) and absorption (lower) spectra of Fe”*- protoporphyrin compIex with (A) and without (B) CN in 50% aqueous ethanol solution at room temperature (solid line). The CN complex was formed by the addition of excess
solid KCN. Broken lines in (A) and (B) show MCD spectra at 155 “K and 77 “K, respectively, in 50% aqueous glycerol solutioa The insets show the dependence of
the extrema indicated on l/kT_
A ferric high-spin complex made of feniheme in a 50% aqueous ethanol solution showed a peak at 275 nm, a peak at 340 nm, a small trough at 3 15 nm and a trough at 367 nm at ro‘om temperature (Fig. SB)). Attention should be
paid to the fact that the M-CD spectrum for the ferric high-spin complex is significan!ly different from that for the ferric low-spin complex. This high MCD sensitivity to spin states should be compared with the relatively small dependence of the electronic absorption spectra on the spin states [16, 171. Decrease of the sample temperature intensified the peak at 285 nm and the trough at 3 15 nm, while the peak at 340 run being attenuated. At liquid-nitrogen temperature, an MCD peak at 355 nm appeared together with the disappearance of the 367~nm trough. The drastic change of the MCD spectra with temperature variation might suggest the spin state change from high to low with decreasing temperature, as reported for some high-spin ferric hemoproteins (18, 13]_ However, the fact that the resultant MCD spectrum does not Tesemble that of low-spin excludes this possibility, The peak around 287 nm is estimated to be mainly C term, while the peak around 355 nm, hidden at room temperature, appeared and made the trough at 367 nm disappear at liquid-nitrogen temperature. Although it is difficult to definitely determine numbers and
MASAHIRO
82
HATANO
et al.
of the transition in this region even with the aid of the MCD, it can be argued, at least, that there are more than five transitions in this region and three transitions among them have strong C term characterIn conclusion, despite lack of rigid interpretation of the magneio-opticai effects, striking differences between ferric and ferrous with low and high spin complexes may serve as a useful working hypothesis in further pursuit of the electronic structure of heme complexes.
positions
REFERENCES 1. J. Bolard and A_ Gamier, Biochim Biophys. Acfa 263,535 (1972). 2 J. C_ Sutherland and M. P. Klein, J. Chem Phys 57,76 (1972). 3_ J_ 1.Treu and J. J. HopfieId,J. Chem Whys 63,613 (1975). 4. T_ Shimizu, T_ Nozawa, M. Hatano, Y. Imai and R_ Sato, Biochemk?y
14, 4172
(1975). 5. L. Vickery, T_ Nozawa and K. Sauer,J. Am Chem Sot. 98,343,351 (1976). 6. G. Barth, W. Voelter, E. Bunnenberg and C. Djerassi, J. Am Chem Sot. 94, 1293 (1972). 7_ B. Hohuquist and B. L_ VaBee, Biochemistry 22,909 (1973). 8. A’. B. Buckingham and P. J. Stephens, Ann. Rev. Phys_ Chem. 17,399 (1966). 9. P. N. Schatz and A. J. McCaffery, Quart. Rev. Chem Sot. 23,552 (1969). Elsevier,Amsterdam (1964). 10. J. E. Falk,PorphytinsandMetalloporphynizs, ll_ T_ H. Davies,Biochim. Biophyz Acta 329,108 (1973). 12_ C. Weiss, H_ Kobayashi and M_ Gouterman, J. MoL Spect_ 16,415 (1965). 13. M. Zemer, hf. Gouterman and H. Kobayasbi, Theoret Chim. Acta 6,363 (1966). 14_ M_ Weissbluth, Srruct. Bond. (Berifne) 2,1 (1967). 15. B. Briat, D. Berger and hf. Lehboux,J. Chem. Whys 57,5606 (1972). 16. P_ George, J. Beetlestone and J. S. Griffith, Rev. Mod. Phys. 36,441 (1964). 17_ G. I_ Hanania, A_ Yeghiayan and B. F_ Cameron, Biochem J. 98,189 (1966). 18. T. IizuIcaand M. Kotani, Biochim Biophys Acta 181,275 (1969). 19. T. Iizuka and M. Kotani, Biochim Biophys. Acta 194,351 (1969). Received I9 November
I975
