ON THE QUESTION OF ENERGY MIGRATION IN BIOLOGICAL PROCESSES

A. N. Terenin and A. A. Krasnovskii

The physicist N. Riehl, known for his work on the luminescence of crystals, develops in an article recently published in the present journal[^1] the idea of energy migration in biological systems, based on likening these systems to inorganic semiconductors—phosphors—in which, upon excitation by a quantum of light, an electron can pass into the conduction band. An analogous conception was also put forward by the well-known Hungarian biologist A. Szent-Györgyi[^2]. In a more vague formulation, the notion of electronic conductivity had also earlier been invoked by some biochemists to explain the action of enzymes[^3]. In his article, N. Riehl adduces a number of experimental justifications for this point of view and gives a developed physical picture embracing a wide range of biocatalytic processes. However, the experimental evidence presented by the aforementioned authors can hardly be considered convincing, and the theoretical interpretation proposed in N. Riehl’s article of protein as a semiconductor, in our opinion, encounters serious objections.

One of us (A. N. Terenin), in his address at the All-Union Conference on Photosynthesis[^4], already pointed out that the transfer to protein of ideas from the theory of semiconductors is so far without sufficient foundation; and he also briefly outlined those conclusions for biochemistry which follow from the mobility of the electron and proton in organic compounds*).

Let us proceed to a more detailed consideration of the question.

ON THE EXPERIMENTAL EVIDENCE FOR ENERGY MIGRATION IN BIOLOGICAL PROCESSES

As an experimentum crucis proving the fact of energy transfer in a protein molecule over considerable distances, Riehl cites Bucher’s experiments on the photochemical decomposition of the complex CO + myoglobin, without giving a bibliographic reference; apparently,

*) A more detailed exposition of this point of view will be given in an article intended for publication in the present journal.

the question concerns the work of Bücher and Kasper, published last year[^5]. In this work the quantum yield of the reaction of photochemical decomposition of the compound CO + myoglobin under the action of wavelengths 280, 313, 334, 366, and 546 mµ was measured. The value of the quantum yield proved to be equal to \(1 \pm 0.1\) in all regions of the spectrum. In the same spectral region the value of the molar absorption coefficient was also measured, both for CO + myoglobin and separately for the hemin component of this compound, which directly binds CO. The absorption spectra proved to coincide completely from the visible region down to a wavelength of 320 mµ. This indicated that in this part of the spectrum light is absorbed exclusively by the hemin component. The difference between the spectra is noticeable only in the shorter-wavelength part of the spectrum: CO + myoglobin has a maximum at 280 mµ, CO + hemin at 265 mµ1. The absorption of the hemin component at 280 mµ is 40% lower in comparison with the absorption spectrum of the myoglobin complex.

From the fact that the values of the quantum yield are practically identical and equal to unity in all regions of the spectrum, the authors conclude that light quanta with wavelength 280 mµ, absorbed not only by the hemin but also by the protein component—namely, by the aromatic rings of tyrosine and tryptophan distributed throughout the mass of the myoglobin protein—are used for the detachment of CO from the hemin part of the system.

In Riehl’s article this fact is interpreted as decisive for his conception of the migration of electronic energy; he writes: “The quantum energy is transported without any losses through the entire large protein molecule to the hemin group.”

Meanwhile the fundamental question is unclear: in what manner does the detachment of CO from hemin occur when the latter absorbs light? The circumstance that the separation of CO from myoglobin proceeds at a measurable rate also without illumination, even at \(0^\circ\)C, indicates a weak bond of CO with hemin, for the breaking of which it proves sufficient to supply a few kcal/mole of thermal energy.

Upon absorption of a quantum already in the first maximum, located in the visible region (about 570 mµ), energy of 50 kcal/mole is supplied to the hemin molecule; this energy is completely converted into intramolecular vibrations, as is evidenced by the absence of fluorescence in hemin. A rather small part of this energy is sufficient to cause detachment of CO[^6]. Vibrational energy will be supplied to the hemin–CO bond with an excess upon absorption of quanta of still greater magnitude in the maxima of the ultraviolet spectrum of hemin.

This explains the unusual constancy, for photochemical reactions of decomposition, of the quantum yield, which retains a value equal to unity over a broad range of the absorption spectrum of hemin, beginning with the visible region. From this point of view, the enhancement of CO liberation under the action of light absorbed by hemin should be interpreted as a kind of internal heating of the molecule, supplying to the vibration of the hemin–CO bond the comparatively small dissociation energy required. As is known, the cleavage of a valence bond even in complex molecules does not occur directly upon the input of electronic excitation energy, but only in the subsequent process of conversion of the excitation energy into motion of the nuclei.

When myoglobin containing hemin is illuminated at the maximum of 280 mμ, there occurs, along with the process described above, absorption by the tyrosine and tryptophan structural units of the protein of quanta with an energy of 100 kcal. With the usual degradation of absorbed energy into vibrational energy, it is quite conceivable that significant portions of the vibration energy propagate along the rigidly connected chain of the principal valences of the protein to the mobile hydrogen atoms, more precisely, protons, which form “hydrogen-bond” bridges between the structural units of the protein. As a result of such activation, transfer of a proton from one unit to another may occur, with the corresponding valence rearrangement of a very extended system, like the known tautomeric reaction of organic compounds^4. Proton transfer under the action of light between conjugated molecules was established by us by direct experiment^7. The energy released in such a reversible reaction of “macrotautomerism” is transferred along the valence bonds participating in it over large distances in the form of potential energy, subsequently being expended on vibrations of large amplitude. Apparently it is precisely in this way that the small portion of vibrational energy required for cleavage of CO is delivered to hemin*). In the protein it is not the electronic energy of the excited level (exciton) that migrates, and by no means a free electron in the conduction band, but the potential energy of vibrations created by a sudden tautomeric rearrangement of the structural skeleton of myoglobin as a single whole. One can speak only of a rearrangement of valence electrons affecting simultaneously a large number of atoms. Migration of a quantum of electronic energy, as such, in a protein is improbable because of the absence of conjugated double bonds and of the regular alternation of molecules of one type. But precisely this circumstance, on the contrary, promotes the unimpeded propagation of vibrations along a chain of ordinary valences, having—

*) This transfer is favored by the circumstance that the Fe atom of hemin is coordinatively bound to the protein—globin—through the imidazole group of histidine^8.

which have approximately identical elastic constants and reduced masses of the vibrating particles.

The transfer of a quantum of electronic excitation energy without any losses would be proved if it were possible to demonstrate sensitized fluorescence fixed on a protein and capable of making molecules fluoresce under the action of light absorbed only by the protein.

Leaving aside the question of the methodological difficulty of measuring light absorption and quantum yield in microheterogeneous colloidal structures, the interpretation of such an experiment will encounter the difficulty that, in the spectral region of protein absorption, the adsorbed component usually also absorbs light.

Under the conditions of a living leaf, sensitized fluorescence of chlorophyll under the action of light absorbed by carotenoids has been demonstrated^9. However, the latter, in contrast to protein, are long chains of conjugated bonds, with deep coloration; moreover, it is unknown whether the components were fixed, and, if so, on one and the same protein.

In a study of energy transfer in biochemical systems carried out in our laboratory, an attempt was also made at one time to obtain data on the possibility of transfer of electronic energy in a protein–dye system by the method of studying sensitized fluorescence.

Two kinds of dyes were adsorbed on a protein (silk fibroin), the absorption maxima of which were considerably shifted relative to one another. It was assumed that, upon illumination of the system in the absorption region of the dye with the maximum located in the short-wavelength region of the spectrum, one would be able to observe fluorescence of the other dye, which has a more long-wavelength maximum, i.e. luminescence should have occurred at the lower energy level of the system. The experiments were carried out under high-vacuum conditions at the temperature of liquid air, in order to prevent rapid degradation of the excitation energy into heat. Under these conditions we were unable to observe the phenomena of sensitized fluorescence.

In the light of the considerations presented above, it is difficult to regard the material cited above as direct proof of the existence of migration of electronic energy in the protein molecule.

Szent-Györgyi, in his book^2, describes experiments on the phosphorescence and photoconductivity of proteins stained with dyes of the gelatinous-phosphor type. In protein without dyes, no photoconductivity was established. The author believes that the conductivity of the protein–dye system indicates the presence of common electronic levels of the type found in inorganic phosphors.

However, studies of recent years have shown above all that the nature of the phosphorescence of organic compounds is fundamentally different

from the phosphorescence of inorganic crystals and is explained by the transition of an organic molecule into a long-lived metastable biradical state (Terenin^6, Lewis). As for the photoconductivity of “dry” protein–dye systems, the works of Soviet researchers first showed that not only dye microcrystals^10, but also dyes in colloidal films and benzene solution, i.e., in non-protein media, possess considerable photoconductivity in themselves^11.

From recent experiments^12 it follows that dye molecules adsorbed on semiconductors can, upon absorption of visible light, cause the appearance of electrons in the conduction band, but such an effect is completely absent on crystalline amino acids.

On the basis of these experimental facts it must be acknowledged that the hypothesis of electron migration in a protein, as in a semiconductor, is without foundation.

Let us now consider the theoretical ideas about biocatalytic processes set forth in N. Riehl’s article.

1. Model of the action of oxidizing and reducing enzymes

The model proposed by N. Riehl for the action of various oxidation–reduction systems is based on the transfer (over a distance) through a protein molecule of an electron, with compensating release or uptake of a proton from the aqueous phase. The currently accepted scheme for the action of dehydrogenases is based on the known facts of the dissociation of the enzyme into a prosthetic group and a specific protein, and of the reversible oxidation–reduction transformation of the prosthetic group, occurring on different types of specific protein.

Thus, the transfer of an electron from one system to another occurs by diffusion of the dissociating prosthetic group of the dehydrogenase enzyme. The mechanism of the reaction of hydrogen transfer from AH to B may be expressed by the following simplified scheme, where P denotes the prosthetic group, and HP the hydrogenated prosthetic group (see, for example,^14):

\[ \begin{aligned} \mathrm{AH} + \mathrm{P} &\to \mathrm{AH}^{+} + \mathrm{P}^{-} \to \mathrm{A} + \mathrm{HP} \quad \text{(on type I protein)}\\ \mathrm{HP} + \mathrm{B} &\to \mathrm{BH} + \mathrm{P} \quad \text{(on type II protein)} \end{aligned} \]

We shall not dwell here on a detailed analysis of the schemes of oxidation–reduction reactions presented in N. Riehl’s article; we shall merely point out that the elementary process

\[ 2\mathrm{Fe}^{++} + \frac{1}{2}\mathrm{O}_{2} \to \mathrm{O}^{--} + 2\mathrm{Fe}^{+++}, \]

accompanied by the appearance of the ion O—, is extremely unlikely, while the addition of a second electron to the oxygen molecule is a strongly endothermic process[^13].

2. Model of the Action of Hydrolases

The author considers hydrolases, including enzymes that hydrolyze the peptide bond, as two-component systems containing a prosthetic group and a protein carrier. It is further stated that the prosthetic group must exist in two different energy states. The addition of water to the peptide bond is considered from the electronic point of view as a “discharge” of H\(^+\) with addition of H to the NH group of the peptide bond and as a compensating discharge of OH\(^-\) with addition of OH to the CO group.

Against this scheme two principal objections should be raised.

a) Enzymes hydrolyzing the peptide bond are single-component systems: they are proteins not associated with a prosthetic group existing in different energy states, as is the case with oxidation–reduction enzymes[^14].

b) From the energetic point of view it is difficult to imagine the possibility of a mechanism of hydration proceeding through the intermediate formation of free radicals H and OH.

The dissociation of water into radicals requires 110 kcal; preliminary dissociation into ions reduces this value to approximately 95 kcal. At the expense of what energy resources are such elementary processes possible on a protein molecule?

3. On the Migration of Energy in the Process of Assimilation of Carbon Dioxide

In experiments on the investigation of photosynthesis in intermittent light it was found that the ratio of the number of chlorophyll molecules absorbing light to the number of reduced carbon dioxide molecules is, for a light-flash duration of \(10^{-5}\) sec, a value from 2,000 to 14,000.

Proceeding from these data, and also in order to explain a number of inconsistencies arising in calculating the duration of the dark reaction of photosynthesis, Gaffron and Wohl proposed a hypothesis according to which the number of reduction centers is several thousand times smaller than the number of available chlorophyll molecules; in this connection two possible models were proposed:

1) the “optical model” of the photosynthetic unit, according to which a large number of chlorophyll molecules is bound in a single complex—the “crystal” (this conception is developed by N. Riehl in his article);

2) the “kinetic model,” according to which energy from a large number of individual chlorophyll molecules is transferred to a small number of reduction centers; these ideas were also developed in a number of works (for references see\({}^{15}\)).

Subsequently other schemes were proposed: the most mathematically developed was the Franck–Gaffron scheme\({}^{*}\), based on the assumption that the limiting dark reaction is controlled by a catalyst whose working period is longer than the lifetime of the unstable photoproduct. Another possible explanation is based on the idea that the period of the reversible reactions in which chlorophyll is involved is several thousand times longer than the duration of the light interval in experiments with intermittent illumination. Thus, we see that the experimental data obtained can be interpreted by means of a number of schemes differing from one another\({}^{**}\).

With his conception, H. Ril revives Reincke’s views, which regarded chlorophyll as a “physical” sensitizer, transferring only energy to the reacting molecules, without supporting this view with new experimental evidence.

In contrast to this, K. A. Timiryazev, in his classical works, regarded chlorophyll as a sensitizer both “optical” and “chemical,” undergoing a reversible chemical transformation in the process of photosynthesis.

The direction put forward by K. A. Timiryazev has the greatest amount of experimental evidence and best explains the factual material of photosynthesis (see\({}^{15,17,18}\)).

Recently it has also proved possible to show the possibility of a reversible photochemical reduction of chlorophyll, proceeding with an increase in the free energy of the system, and to note the significance of this process in sensitized reactions\({}^{17}\).

\({}^{*}\) The general scheme of photosynthesis of Franck and Gaffron, proposed in 1941, is at the present time unacceptable, since it does not correspond to contemporary experimental results. In addition to the earlier criticism of this scheme\({}^{15}\), the following objections should be given here:

1) The assumption of a photochemical act of carbon dioxide reduction does not correspond to reality; studies in which isotopes of carbon C\({}^{11}\) and C\({}^{14}\) were used showed that the fixation and reduction of carbon dioxide occur as a result of a dark reaction.

2) The idea of the existence of oxidized and reduced forms of chlorophyll, preserving similar absorption spectra, has not received experimental confirmation; it has recently been possible to show that the reversible reduction and oxidation of chlorophyll is accompanied by the opening of the system of conjugated bonds, the disappearance of the red absorption maximum, and the appearance of a new maximum in the region of 500–550 mµ\({}^{16,17}\).

\({}^{**}\) Here we leave aside discussion of the correctness of using the available experimental data (studies of photosynthesis in intermittent light, etc.) in calculations made by various authors; in particular, the question of whether, under the experimental conditions, all the chlorophyll molecules contained in the chloroplasts of the plant studied were accessible.

It is not our task here to discuss the question of the form in which chlorophyll is bound in the plastid. We should merely like to note that the Gubert scheme cited in H. Riehl’s article is, to a considerable degree, speculative, since there is no direct evidence that all the chlorophyll in the plastid exists in the form of a monolayer with the porphyrin ring adjacent to the protein molecule.

It is more probable that chlorophyll in the plastid is not in a single state, and that the photochemically active part consists of chlorophyll molecules bound to the lipoid portion of the lipoprotein complex.

In conclusion, we shall allow ourselves to express the opinion that, despite its fundamental interest, the physical conception of electron migration as applied to biological systems requires more convincing experimental proof. The broad use of this hypothesis to interpret known facts of biocatalysis has no advantage over the views now generally accepted. H. Riehl’s considerations are more appropriate as applied to catalytic reactions occurring on the surface of crystalline solids, i.e. true semiconductors, for which electronic ideas were developed long ago in our country by L. V. Pisanzhevsky^19 and which are again becoming topical at the present time^20. There is no need artificially to confine the energetics of biochemical processes within the narrow framework of a detailed physical picture taken from the field of very interesting phenomena that, however, have no direct relation to organic substances. Undoubtedly, in the latter as well there is electron displacement accompanied by energy transfer (see the examples in^6), but the mechanism of this migration is of a different nature than in inorganic semiconductor crystals. We shall return to a detailed substantiation of this point of view in the pages of the present journal.

CITED LITERATURE

1. H. V. Riehl, Uspekhi Fiz. Nauk 35, 186 (1948), No. 2.
2. A. Szent-Györgyi, Nature 9, 157 (1941); “Chemistry of Muscular Contraction,” N. Y. 1947.
3. H. Borsook, Ergebn. Enzymforsch. 4, 19 (1935).
4. A. N. Terenin, Izv. Akad. Nauk, Ser. Biol., No. 3, 369 (1947).
5. T. Bücher, J. Kaspers, Biochimica et biophysica acta 1, 21 (1947).
6. A. N. Terenin, “Photochemistry of Dyes,” Izd. AN SSSR (1947).
7. A. N. Terenin and A. V. Karyakin, DAN 88, 425 (1947).
8. H. Theorell, Ann. Rev. Biochem. 7, 265 (1947).
9. H. Dutton, W. Manning, R. Dugar, J. Phys. Chem. 47, 308 (1943). E. Wassink, I. Kersten, Enzymologia 12, 3 (1946).
10. A. T. Vartanian, Zh. fiz. khim. 20, 1065 (1946); 22, 769 (1948). E. K. Putseiko, Dokl. Akad. Nauk SSSR 59, 471 (1948); Zh. fiz. khim. 22, 1172 (1948).

1. J. Coudreau, Izv. Ros. Akad. Nauk 6, 727 (1919); Izv. fiz. in-ta 1, 571, 132, J. chimie physique 26, 59, 117, 178 (1929).
2. A. Terenin and E. Klyueiko, Zh. fiz. khim. (1949).
3. I. A. Kazarnovskii, Zh. fiz. khim. 14, 320 (1940).
4. Enzymes, Publ. House of the Academy of Sciences (1940) (collection).
5. A. A. Krasnovskii, Usp. sovr. biolog. 21, 153 (1946); Izv. Akad. Nauk, ser. biol., No. 3, 377 (1947).
6. A. A. Krasnovskii, Dokl. Akad. Nauk SSSR 58, 617 (1947); 58, 835 (1947).
7. A. A. Krasnovskii, Dokl. Akad. Nauk SSSR 60, 421 (1948); 61, No. 1 (1948).
8. E. Rabinowitch, “Photosynthesis” No. 4 (1945).
9. L. V. Pisarzhevskii and M. Rozenberg, “Inorganic Chemistry” (1928), DBNTU.
10. F. F. Vol’kenshtein, Zh. fiz. khim. 22, 311 (1948).

1. It should be emphasized that the authors measured the absorption spectrum of the hemin component separated from the protein carrier; meanwhile the possibility is not excluded that, in the bound state, the maximum of the spectrum of the hemin component will be still closer to the maximum of the spectrum of the CO + myoglobin complex. ↩