You are on page 1of 22

Chern. Rev.

1995, 95, 3-24

3

Magnetic Field Effects in Biology: A Survey of Possible Mechanisms with Emphasis on Radical-Pair Recombination
Charles B. Grissom
Department of Chernisw, Universily of Utah, SaH Lake Civ, Utah 84112 Received Juiy 27, 1994 (Revised Manuscript Received October 9, 1994 )

Contents
I. Introduction II.
A. Producing and Measuring Laboratory Magnetic Fields Magnetic Field Effects on Radical Pair Recombination in Biological Systems A. Theory of Magnetic Spin Effects on RP Recombination B. Hyperfine Induced Intersystem Crossing C. Spin Rephasing (AgMechanism) Induced Intersystem Crossing D. Singlet-Triplet Energy Level Crossing E. Spin-Orbit Coupling F. Enhanced Recombination by Compartmentalization G. Random Radical Pair Recombination H. Radical Chain Reactions I. Static vs Modulated Magnetic Field Effects on RP Recombination J. Oscillating Magnetic Fields To Induce Changes in RP Recombination K. Radical Pair Recombination in Biological Systems L. Magnetic Field Effects on Thermal (Nonphotochemical) RP Reactions M. Maanetic Field Effects in Catalvtic Reactions N. Magnetic Field Effects in Enzymatic Reactions 0. Coenzvme BWPhotochemistrv P. Magneiic Fieid Effects on B12:Dependent Enzymes ( . Biological and Health Relevance of B12 1 Maanetic Field Effects R. Other Enzymes for Which RP Mechanisms Have Been Considered Non-Radical-Pair Magnetic Field Effects in Biological Systems: What Is the Mechanism? A. Magnetoreception B. Melatonin C. Ca2+Binding and Ion Cyclotron (Parametric) Resonance D. Miscellaneous Magnetic Field Effects E. Electric Field Effects Acknowledgments References

3
4
4

5
6

6
7 7 7

0 0 0
9 9 10 10 11 14 17 19 20 20 20 21 21 22 22 22

Charles 8. 'Chuck" Grissom received his B.A. in Biochemistry and Chemistry from the University of California, Riverside in 1981. He completed his Ph.D. in Biochemistry under the direction of W. W. Cleland at the University of Wisconsin-Madison in 1985. He was an NIH Postdoctoral Fellow at Wisconsin in the laboratory of John L. Markley from 1985 to 1987 and at the University of Califomia, Berkeley in the laboratory of Paul A. Bartlett from 1987 to 1989. In 1989, he assumed his position in the Chemistry Department at the University of Utah where his laboratoryfocuses on magnetic spin effects in enzymatic and chemical reactions. His passion in lile is understanding the mechanism of biological catalysts. Outside of the lab, his passions are mountain biking, exploring old uranium mines, and neon art. research began in earnest in t h e 1960s with t h e w o r k o f Barnothy.'S2 These early experiments were phenomenological observations: I s a magnetic f i e l d beneficial, inconsequential, o r deleterious to biological organisms? M o r e than 30 years later, t h i s fundam e n t a l question h a s y e t to b e answered unambiguously.3.4 I f t h e effect was profound and an exogenous magnetic f i e l d resulted in t h e immediate death, mutation, or change in morphology o f an organism, n o question would r e m a i n as to t h e safety o f magnetic fields. Instead, t h e effects a r e more subtle. A continuing series o f phenomenological studies h a s provided equivocal evidence for magnetic field effects in biological ~ y s t e m s . 3 , In 1979,intense scientific ~ curiosity and social activism was sparked by t h e epidemiological study of Werthheimer and Leeper that suggested an increase in childhood cancer for individuals living near electric power Beyond t h i s citation classic, n o f u r t h e r discussion o f epidemiological data will b e included in t h i s review. Instead, t h e purpose of t h i s review i s to summarize t h e evidence f o r possible biological effects o f static and time-varying magnetic fields within the context of proposed mechanisms. In t h e t r a d i t i o n a l practice o f science, a n e w f i e l d begins with a series o f phenomenological observations

111.

IV. V.

1. Introduction
F e w areas o f research are as controversial, b o t h scientifically and politically, as research i n t o t h e biological effects o f magnetic fields. This area o f

0009-2665/95/0795~015.50/0 0 1995 American Chemical Society

4 Chemical Reviews, 1995, Vol. 95, No. 1

Grissom

ing static magnetic fields up to 1.8 T. The high that cannot be explained within existing paradigms. permeability metal core allows the lines of magnetic New hypotheses that describe a fundamental prinflux t o be “concentrated” and spatially confined to ciple are developed and they are adjudicated by experimentation. Further iteration of hypothesis and produce a substantially higher field than would be achieved with an air gap alone. Furthermore, the experiment refine the concept until a new paradigm is accepted by the scientific community. In the study high reluctance provided by the high permeability metal attenuates the residual ac ripple that was not of magnetic field effects on biological systems, phefiltered out by the constant current dc power supply. nomenological-based observations that cannot be This affords adjustable static fields with an imexplained within existing theory have only recently measurably low electric field gradient and an altergiven way to hypothesis-based experimentation. nating magnetic field component of less than 0.1 This article is organized around the principle of ppm. Magnetic flux densities in the range 1.8-16.5 hypothesis-based research into biological magnetic T are most easily obtained by superconducting magfield effects. It is intended to serve as an introduction to current issues and to review recent progress nets. The common lH NMR spectrometer frequencies of 200,300,400,500,600,and 750 MHz provide easy toward answering the fundamental question of how magnetic fields interact with biological systems. access t o high magnetic flux densities of 4.4,6.6,8.8, This review will focus on those magnetic field 11.0, 13.2, and 16.5 T. In addition to field strength effects that can be addressed within the existing and stability, the homogeneity of the required field paradigms of chemistry and physics in order to must be considered. Typically, the questions related to biological systems are not concerned with exencourage further research within this community. tremely narrow field ranges as is required in magIn the last section of this review, those biological magnetic field effects that are understood at only a netic resonance spectroscopy. However, magnetic field effects do occur in specific “windows” of field minimal level will be summarized, but not evaluated critically. A broader introduction to the subject and strength and frequency, so a more modest magnetic flux homogeneity of approximately 1%within the an historical perspective of biological magnetic field effects can be found in several collective b o o k ~ . ~ - l ~ area of the experiment is probably sufficient. This criterion of minimal homogeneity may have t o be A. Producing and Measuring Laboratory Magnetic revised if fine features and resonances in biological Fields systems are discovered. Well-defined alternating magnetic fields can be I . Measuring Magnetic Fields more problematic to produce and characterize. Paired Magnetic flux density can be measured by a variety coils of wire can be oriented orthogonally around the of techniques. In common practice, magnetic flux sample area. Helmholtz coils are often employed to densities greater than 0.05 mT are monitored with achieve magnetic fields of calculatable intensity. An a teslameter (gaussmeter) based on a Hall effect alternating current of the desired frequency and transducer, whereas magnetic flux densities less than waveform will produce an alternating magnetic field about 0.2 mT are commonly measured with a fluxat right angles to the direction of current flow. If gate magnetometer. Both instruments can be calitiming synchronization is important, it must be brated by reference to NMR frequencies. Nearby remembered that the magnetic field leads the electric high-energy RF sources can occasionally lead to field by d 2 . Laminated-core electromagnets (conincorrect measurements due to instrumental artifacts struction similar to a transformer) provide access t o and proper shielding must be used. small volumes (3-5 cm3)with flux densities of 0-80 Magnetic field strength (H) measured in Oeris mT at frequencies less than 750 Hz. steds, and it decreases rapidly as the distance from Magnetic shielding can be achieved with lowthe source is increased. Magnetic flux density ( B )is carbon, high-iron steel or special high-permeability measured in Tesla (or Gauss) and it gives the density mumetal. Laminated shielding typically proves more of lines of magnetic flux per unit area (eq 1). In air, efficient than thicker, unlaminated material. High magnetic field strength is closely approximated by efficiency shielding material for sensitive devices magnetic flux density, such that magnetic field such as photomultiplier tubes, ion traps, and light strength is commonly specified in units of Tesla (or sources are commercially available. Gauss). 1 T = 10000 G (1) 11. Magnetic Field Effects on Radical Pair 2. The Geomagnetic Field Recombination in Biological Systems The geomagnetic field is near 0.05 mT, and it does Magnetic field effects on the rate of radical pair not vary on a time scale relevant to living organisms. (RP) recombination is the best understood mechaIt is directional and is best represented as a vector nism by which magnetic fields interact with biological quantity. When required, the geomagnetic field can systems. However, the health relevance of this be subtracted by orienting an experimental apmechanism of magnetic field sensitivity is uncertain paratus along the geomagnetic north-south axis and because the best-known effects only become signifiapplying a suitable bucking current t o a small cant at moderate magnetic flux densities above 1-10 electromagnet or coil. mT. Significant discussion of this mechanism is 3. Producing Laboratory Magnetic Fields provided because recent calculations suggest magnetic field effects through changes in RP recombinaElectromagnets with metal cores of high magnetic tion may even occur at magnetic field strengths near permeability are the most suitable method of produc-

The size of the Zeeman splitting is given by eq 2.^^. AG becomes more tibility. Figure 2.35 In contrast. This includes the influence of an exogenous magnetic field. In the absence of a magnetic field. we use the term “magnetic isotope effect (MIE)” to refer to the influence of nuclear spin on the reaction rate. TO. and T+I.’Lwd Crossing Magnetlc Field. The term B + + AE =gpB (2) is the magnetic flux density and /3 is the Bohr . as reflected in the distribution of isotopes in products.2334 However. T+Iand T-I are split to higher and lower energy from SOand To.^^-^^ The MFE neophyte is directed to several reviews that are especially readable and provide a good introduction to the theory of spin-dependent c h e m i ~ t r y . and the singlet RP. of favorable by the trivial amount of 0.are energy degenerate. In contrast. the three triplet states. ~ ~ nisms that lead to magnetic spin-dependent chemistry are summarized in the following sections. TO.0002 kcal/mol in a magnetic field of 1T. Prior t o dissociation. an enzyme with radical pair intermediates. t I I I 8 l.22 In the following section. T A.Magnetic Field Effects in Biology Chemical Reviews. T+I. consider the case of a RP that is produced by homolysis of the A-B bond in the ground state.lgV2l one other only biological system. if the ground-state starting material absorbs a photon to create the first excited singlet state. SI. At B = 0. General scheme for generating a triplet radical pair by photolysis. physical mechaThe ~ ~ . the two electrons in the a-bond were spin-paired by virtue of the Pauli exclusion principle. 95.and T+1are split by the electronic Zeeman interaction energy into three distinct states that are no longer degenerate (Figure 2). the theory of magnetic field effects on radical pair recombination will be summarized and the results of studies on the photosynthetic reaction center and enzymes with RP intermediates will be discussed. At B 0.The ~~ theory behind magnetic spin effects in chemical reactions. ~ ~ . The reactivity of a doublet state radical will not be altered by a magnetic field. A n isolated molecule with a single unpaired electron can adopt two orientations relative to an external magnetic field. 1995. a so-called “magnetic field effect (MFE)” as well as the effect of an endogenous magnetic field that originates from a nonzero nuclear spin (If 0) that has an intrinsic magnetic moment. According to the Wigner Spin Conservation Rule. will place the molecule on a dissociative energy surface and the separated triplet RP {At tB} with parallel electron spins will result (Figure 1). - Throughout this article. At B > 0. The importance of magnetic spin effects in chemical reactions presents a conceptual problem if only the energy imparted by the field is considered. electron “spin” refers to the total angular momentum of the unpaired electron. {}I will retain their original orientation immediately after homolysis. Thermal reactions that undergo a change from diamagnetic substrates to paramagnetic intermediates. has been described with both classical and quantum mechanical formalism^. products. the elements of the RP (denoted by brackets. TI. Theory of Magnetic Spin Effects on RP Recombination The most general term that covers the role of magnetic interactions in chemical reactivity is “magnetic spin effect”. especially photochemically produced RP’s. will prevail. We call this a “doublet”state by virtue of its having two possible orientations in the magnetic field. No.and T-1 are energy degenerate. with electron spins paired in an antiparallel fashion {At JB}. rapid decay t o the (usually) lower energy triplet state. These orientations are with the spin vector aligned parallel or antiparallel to the external field. Ax. In a reaction that undergoes an increase in magnetic suscepcm3/mol.16-18 Since the yield of excited triplet electronic states in the photosynthetic reaction center was first shown to be magnetic field sensitive in 1977. Because different isotopes have different nuclear magnetic moments. or transition states. has been shown to exhibit any magnetic field-dependent parameters. the three triplet spin states. T-1. might be expected to be accelerated by a magnetic field that imparts a stabilizing interaction to the paramagnetic species. Vol. T-1.. TO. 1 5 1 Figure 1. even a magnetic field in the range of 1-10 mT can split the Zeeman energy levels of a radical pair and provide an alternate (nonadiabatic) reaction pathway that can change the observed reaction rate or alter the product d i s t r i b u t i ~ n . Each electron in the RP has a spin quantum number of &2. Splitting of triplet energy levels. the intensity of the geomagnetic field.

The value of g varies slightly as the surrounding nuclear and electronic environment is varied. If an external magnetic field is applied.5 T is applied. ~ ~ . ~ ~ Most hyperfine coupling constants for organic radicals are 1-10 mT.g. there are other mechanisms that can lead t o magnetic field-induced ISC. EST. Since RP recombination usually occurs from the spinpaired singlet state. described by a vector quantity. it can lead to changes i n the rate of the reaction or the product distribution. Radical pairs that contain 13C at either of the radical centers will undergo more ISC than radical pairs with 12C at the radical center. which is close t o 2. if the photolysis is carried out in aqueous detergent micelles. l - . but with opposite relative signs (Figure 3). the latter of which can undergo decarbonylation) and cage recombination (to produce DBK). The magnetic field dependence of the isotope effect provides a characteristic determinant of the importance of magnetic spin and RP chemistry in any reaction with an unusually large isotope effect. as well as the mass of the reacting i s o t ~ p e . In solution. so--HFI T O . promote SO To.37 Because the majority of photochemically produced radical pairs are borne in the triplet state. The net result is that HFI induced ISC between S Oand T-l and T-l is decreased. (e. 1 A 1 Grissom S o . a net decrease in RP recombination will occur.05. so they are populated to an equal extent (Figure 3 ) .37-42 Absorption of a photon by DBK produces the first excited singlet state that undergoes rapid ISC to the lowest energy triplet state (dissociative surface). I n reactions that are sensitive to the spin angular momentum of the unpaired electron. C..00 for the free electron and most organic radicals. Vol. all orientations are averaged by molecular motion such that g becomes isotropic and we can regard it as a scalar quantity. the net magnetic moment. k1dk13. Hyperfine Induced Intersystem Crossing Hyperfine interactions (HFI) between the magnetic moment of an unpaired electron and the magnetic moment of a nearby (bonded) nucleus results in a “torque” that promotes flipping of the electron spin. Hyperfine interactions promote mixing of the singlet (attractive) and triplet (dissociative) reaction manifolds of the free energy surface by interconverting the SOand T-I. magneton (9. This leads t o the NMR-observable phenomenon of chemically-induced dynamic nuclear polarization (CIDNP) and the ESR-observable phenomenon of chemically-induced dynamic electron polarization (CIDEP).47 and is beyond the range of reasonable mass-dependent isotope effects. k&13 increases t o 1. the decrease in ISC will decrease population of the singlet state. but this yields a high-energy triplet molecule that lies on a dissociative energy surface. ~ * .1 B= 0 Magnetic Field B> 0 Figure 3. Bond homolysis occurs to produce the triplet RP..~ ~ T o . 95.40 This enormous kinetic isotope effect decreases to 1. This interaction between the electron and a magnetically active nucleus provides a mechanism to interconvert both electron and nuclear spin quantum numbers in the absence of an external magnetic field. this leads to a “magnetic” isotope effect that depends upon the nuclear magnetic moment of the isotope. thus producing a small enrichment of 13Cin starting materia1. Given sufficient time. TO. B.-1 and T-1 spin states by the Zeeman interaction energy and decreases the HFI induced ISC. The value g is the Lande g factor.6 Chemical Reviews. A net increase in “escape” (nonrecombination) products derived from the triplet RP will be observed. At B > 0. the directionality of g may be very important. If the RP was produced in the triplet spin state. g is a tensor. HFI can only promote SO To intersystem crossing and population of the T+l sublevels is diminished. The geminate RP partitions between cage escape products (benzyl and benzylcarbonyl radical. Because the amount of HFI-induced ISC will depend upon the nuclear spin (isotope). As an example. No. Spin Rephasing ( A s Mechanism) Induced Intersystem Crossing In addition to electron-nuclear hyperfine interactions. Hyperfine interactions. the observed isotope effect. and a physical mechanism by which to interconvert the spin states. In a rigorous analysis. In compartmentalized biological systems or at interfaces. 1995. If photolysis is carried out in benzene. an increase in ISC will lead to an increase in the singlet RP population and a corresponding increase in RP recombination. In the presence of an external B-field.42 However. The spin angular momentum of an electron is quantized as either +l/z or -l/Z relative to an external frame of reference. The spinning electron has a net magnetic moment that is similarly quantized and can occupy only one of two relative orientations.HFI.T.41. the degeneracy between the three triplet spin states is removed by the Zeeman interaction energy that splits T+1and T-1 from To equally.and T+1 states.37 The external field now splits the T. Since we are considering the interac- . The substrate will be enriched in 13C relative to product. This decrease in HFI-promoted ISC can remove up to 2/3 of the enhanced reactivity (throughput) of the magnetically active nucleus. The rate of intersystem crossing (ISC)between the singlet and triplet spin states is dependent upon having an accessible energy gap. Recombination from the triplet RP is possible. .12 when an external magnetic flux density of 1. is 1. an equilibrium distribution between singlet and triplet states (assuming A E s T is so small as t o be approximately zero) will result in a 25% singlet and 75% triplet spin population.274 x J T-l).T.*l == intersystem crossing at B = 0. The three triplet spin states are energy degenerate at B = 0. consider the photolysis of dibenzyl ketone (DBK). RP recombination). will precess at its Larmor frequency given by eq 3 .

at B > 0.00 and g = 2. To a first approximation. If the RP is held close for at least this long.69 CPat 37 “C and 1 atmL51 The cell itself offers a further degree of compartmentalization. For two unpaired electrons that are on different atoms and do not “see” each other. Spin-Orbit Coupling A radical on an atom of high electron density (sulfur) will tend t o have a more anisotropic g value than a radical on an atom of lower electron density (carbon).04 to 1. a biradical in which the two unpaired electron spins reside on the same molecule). This allows primary geminate recombination to be characterized by the solvent independence of the process. it is convenient to express eq 3 as the difference in precession rates for the two unpaired electrons (eq 4). either through space. In the DBK example above.00255 and 2. producing a net torque on the magnetic moment of the unpaired electron. no conversion between the singlet and triplet states will occur at B = 0. (the case where W > EHFI). A similar increase in the isotope effect is observed when the rate of diffusion is decreased by an increase in solvent viscosity.6 cP) to dodecane (7 = 1. however. and ISC can out compete RP separation or geminate recombination. exchange-interaction-promoted ISC disappears. 95. -545 cm-l. -2194 cm-l. On going from benzene (7 = 0. No.e. Ag = 0 and both unpaired electrons will precess at exactly the same frequency. Vol.07. Although controversial and poorly defined.05 and to 1.32946SOC can lead to the relaxation (randomization) of electron spin intermolecularly. it may be regarded as producing a basal degree of ISC that is not influenced by external factors. and there will be a preference for either the triplet or singlet state.of less than -10 A.32 SOC is caused by coupling between the electron spin angular momentum and the orbital angular momentum. Photochemists have used “internal heavy atom effects” and “solvent heavy atom effects” to alter the reactivity of spin-correlated RP’s by increasing ISC.gl = 2. { adenosyl-5’-CHz cob(II)alamin}. A spincorrelated RP that is produced by bond homolysis will experience only a small MFE or MIE if the RP elements are free to diffuse in solution. w = 4. SOC will result in ISC regardless of magnetic field strength.For ~ ~ ~RP. r e s p e ~ t i v e l y . whereas photolysis in the restricted space of a micelle brought the isotope effect up to 47%. (i. J. If the exchange interaction is so large (definite preference for S or T) that HFI cannot promote ISC.43the rate of rephasing (ISC) 2 can be very fast.54 x s at 1 T. Diffusion and viscosity effects are now important and solvatioddesolvation of the RP elements may limit the rate of recombination. this may be regarded as the spinning bonded electrons . Primary geminate recombination is limited to to s and occurs without intervention of solvent and little or no diffusion. I. ~ ~ ~ ~ ~ Enhanced RP recombination by compartmentalization may be important t o observing significant RP magnetic field effects in biological systems. the cytosolic absolute viscosity may be similar to the viscosity of H2O (0. Fe2+. 1995. typically 10-8-10-4 s.0031. The distance between the radical centers is still small enough to define a singlet and triplet surface. S.-114 cm-l. only a small isotopic enrichment of 5% was observed in benzene. Only systems that do not undergo signXcant conformational changes or atomic reorganization can undergo primary cage recombination. In this sense. Equation 4 gives the time required for spin evolution to interconvert the SO and TOstates by rephasing of the electron spins. In the case of a heteronuclear RP. then the precession rates for the two unpaired electrons will be different. As an example. rsep. This is known as spin-orbit coupling (SOC) and it increases in importance as the atomic number increases.is a meqsure of this interaction. escape products are favored over cage recombination products. The relative importance of SOC in a given RP is proportional to the absolute value of the SOC constant for the atom with unpaired electron density: C.35 cP) and cyclohexanol (7 = 30 cP). the S and T surfaces can cross in a very narrow region of B and rsepto promote ISC (Figure 2). J > 0.00055 ~ ~ this and Am = 2. E. In the physical organic nomenclature. Under these circumstances. C1. as well as intramolecularly. 5060 cm-1.25. 1 7 tion between two unpaired electrons. Singlet-Triplet Energy Level Crossing In the case of a RP that exists at a well-defined separation distance. EST > 0.Magnetic Field Effects in Biology Chemical Reviews. However. consider the organic radical pair (CH3 CH3CH OEt} with g values of 2. or through a n-bonding interaction. Enhanced Recombination by Compartmentalization Another important aspect of magnetic spin-dependent chemistry is enhanced recombination by compartmentalization of the RP. RP separation decreases the likelihood of reencounter through collisions that would otherwise lead to a discrimination based on the spin state of the RP. ISC by spin-rephasing can compete with RP separation. 13 cm-l. In a homo- nuclear RP. the Coulombic repulsion between the two electrons will be important in determining how they interact.07 x s at B = 1T. The net effect is to produce a narrow window at which ISC is increased. but D. The exchange integral (or exchange interaction). Br. Cage recombination processes can be characterized further as either primary or secondary cage recombination. such as the photolysis product of adenosylcob(II1)alamin (or in the active site of a B12 enzyme) where one element of the RP is a paramagnetic metal ion. Recombination in bulk solvent occurs on a much slower time scale. Co2+.-189 cm-l. r e s p e c t i ~ e l y . In a freely diffusing environment. 0. -184 cm-l. J = 0 and their ESR signatures will each be doublets. At fields higher than B . If the two electrons “communicate”. hg = 0.29. the isotope effect increases from 1. However. no magnetic spin-dependent chemistry is observed. if Ag f 0. -79 cm-’.46-48 F.

a RP that exists at a large separation distance. Since only one unpaired electron spin is present. Any magnetic field-induced change in ISC would provide only a small change in the concentration of radical propagators.^^. the RP (of random origin) will stay close enough for ISC t o alter its spin state. In the short RP lifetime the reactivity of the 25% of radical pairs that are singlet will be nearly unaffected by magnetic field-dependent ISC. the probability of radicalradical recombination is low. R’ Under steady-state conditions. ROO’ + RH ?L ROOH + R’ R’ + 0. then a 60 Hz magnetic field exerts the same instantaneous effect (within 10-8-10-6 s) as a static ~~ ~~ field of equal m a g n i t ~ d e . but this would have a large effect on the rate because each radical can cause multiple chain events.]ldt = (k.instantaneous field is described by eq 14. The . k ~ k4. ~ ~ As an example. 1 the volume of a cell is large when considering the mean free path of small molecule diffusion./2k liZk5 [RHI[R-RI ’’. Interfacial (boundary) effects and organelle compartmentalization are probably the most important restrictions to diffusion in cells. A RP that is held extremely close will usually have a large AEST that is greater than the interaction energy provided by the magnetic field or HFI. In a more viscous medium or a compartmentalized environment. However. and k8). and k5 only .56 -d[O. and termination are shown in eqs 5-12. whereas encounter between triplet radical pairs typically will not result in bond formation. 1995.j7J8 R-R . AESTis undefined and there will be no effect of a magnetic field (endogenous HFI or external MFE). consider the effect of a magnetic field produced by 60 Hz alternating current. k2. The magnetic vector component of a time-varying magnetic field that is modulated slowly.2R” kl k2 (5) R” + 0. This prevents chain propagation in autoxidation and peroxidation reactions of biological interest from being affected by a magnetic field. k7. involve one unpaired electron species and will not exhibit a magnetic spin dependence. Static vs Modulated Magnetic Field Effects on RP Recombination Magnetic field effects on RP recombination originate from the splitting of Zeeman energy levels by static fields. Vol. These random radical pairs of non-geminate origin are sometimes called “F-pairs” to distinguish them from “G-pairs”(geminate pairs). -ROO’ (6) The slow oscillation of a weak magnetic field in the presence of a larger static magnetic field has also been used to increase the signal-to-noise ratio in spectrophotometrically observable reaction^.j2 In contrast. has the same net e f f e ~ t . The sign of the B field is unimportant and the absolute value of the field is changing at 120 Hz. and no increase in ISC will be observed. rSep. with AEST % 0. Radical-radical encounter of these species demands a high concentration of these species to produce a significant rate of recombination. -. A lock-in amplifier allows . Although Izl and ks appear in the rate expression. In a solution of low viscosity. propagation. the derivative of the spectrophotometric signal is followed in phase with a weakly modulated magnetic field. A magnetic field-dependent radical chain reaction involving the peroxyl radical has been demonstrated e ~ p e r i m e n t a l l y . d[R’Ydt = 0 and the net rate of lipid peroxidation is described by eq 13. The series of reactions that describe initiation. Note that the propagation rate constants. the concentration of R and ROO’ is very low. Radical Chain Reactions At an infinite separation distance. consider the autoxidation of a fatty acid. No. compared to radicalmolecule reaction that is more likely.j3. 95.ROO’ k4 ROO’ + RH 5ROOH + R’ 2R’ RR G. H.j4 + ROO’ 5 ROOR 2R00’ k.^^ In this application. If the radical concentration is low. when compared t o the lifetime of the spin-correlated RP and the rate of ISC. The lifetime of the collision complex is too short for ISC after the RP is brought together and before subsequent dissociation. ~ ~ .56 I. If RP recombination is occurring in the time window s. AEST = 0 and the elements of the RP are characterized as “free radicals” that react independently.8 Chemical Reviews. the rate of a radical chain reaction is directly dependent on the concentration of radical species and any decrease in the radical concentration (through decreased initiation or increased termination events) will directly affect the rate. Random Radical Pair Rec~mbination~~s~~ At equilibrium. ( 13) The combination of two radical species occurs only in recombination of the initiator radicals (reverse of k d and chain termination (k6. random pairs will exhibit a 3:l distribution between the triplet and singlet spin states. ~ ~ As an example. whereas reactivity of the 75% of radical pairs that are triplet will exhibit the greatest dependence on ISC (and magnetic field). encounter events between singlet radical pairs almost always result in reaction. ROOR + 0. will generally have a small AESTas required for magnetic spin-induced ISC.

Magnetic Field Effects on the Photosynthetic Reaction Center The first biological system to be probed by magnetic field effects was the light-harvesting reaction center from the photosynthetic bacterium Rhodopseudomonus sphueroides.4 0. Magnetic Flux Donrlty. of magnetic field effects on RP recombination rates was developed. this theory is extended to biological systems with RP intermediates. a net increase in Qr is observed. This method of data collection will not work with coupled enzyme assays. I is an intermediate electron acceptor.. J. 3P. accumulates.) 1.68969 even At higher magnetic fields.03 Magnotic Flux Donrlty.e. ~ ~ At high magnetic fields greater than 500 mT. HFI mixes So-Til. T Figure 4. Oscillating Magnetic Fields To Induce Changes in RP Recombination16--'8 The application of perturbation theory treatment to oscillating magnetic fields and RP recombination allows a theoretical consideration of weak oscillating magnetic fields (on the order of 0. Vol.62 In close analogy to the photochemical reaction of DBK described above.16-18 0. Qr. along with its nine prosthetic groups. the l{P+* X} radical pair transfers an electron to '-I The quantum yield.0 0 1 2 3 4 5 K. this technique may offer a superior signalto-noise ratio. (A: Reprinted from Moehl. X. the lifetime of the l{P+* X-*} RP is extended and can undergo recom'-1 bination (disfavored)or ISC to the triplet RP 3{P+ *-I} that can recombine (by virtue of its prereduced state) to yield 3P*M-*. A. the general theorj.16J7 perturbation theory approach to the Schrodinger equation suggests that considerable alterations in ISC can be realized at surprisingly low fields that may be relevant to magnetic field bioeffects via changes in RP recombination. of the triplet chromophore.8 ps. only the So-To states can i n t e r ~ o n v e r t . fluorescence or luminescence yield).decreases by about 50% at 50 mT (Figure 4A).3 0. This technique is best suited to reactions that produce a short-lived chemical species that can be probed on a rapid time scale (i. 1985. 1 9 a high-precision measurement of the magnetic field effect on the spectrophotometrically observable species. The lower free energy of the l{P+* '-X} RP makes this I an irreversible process. the integrated rate expression that describes [PI as a function of time only weakly reflects the small modulation of d[PYdt. 1995.8 0" f p: 3 - 0.63164 electron donor (chromophore).~ O In Figure 4B.003 mT) in the This presence of the geomagnetic field of 0. Phys. At B = 0 T.121. but at magnetic fields greater than zero. 95.63964 In this scheme. When a stable end-product. Recent reviews by Scaiano et al. Copyright 1982 American Chemical Society.01 0. ~ ~ .63 Absorption of a photon by P produces the excited P*M complex that decomposes to the metastable P+* radical cation and '-1 radical anion in about 2. E. Hoff. converts visible light energy into a transmembrane proton gradient that drives ATP synthesis. This case is described by eq 16. P is the primary quinone acceptor X to produce l{P+*I*-X}in a process that requires a minimum of 200 ps.P. is chemically reduced (X-*) prevent it from accepting an electron. Radical Pair Recombination in Biological Systems In sections A-J.Magnetic Field Effects in Biology Chemical Reviews.58and Walleczek et al. This 100000 M membrane-bound W protein (from R. W. a .o states. K. Nevertheless. In the next sections. Lous. This linear increase in ISC at high fields is ascribed to the interconversion of the SO-TO spin states via the Ag m e ~ h a n i s m . and X is the first stable electron acceptor.V. 22.61also consider the possibility of biological magnetic field effects via changes in radical-pair recombination. Chem. the absorption of a photon produces the first excited singlet state that irreversibly donates an electron to an acceptor and leads to formation of a compartmentalized radical pair (eq 15). No. Instead. In the normal biological reaction. uiridis). ~ ~ . Lett. Copyright 1985 Elsevier Science Publishers B.5 0.. 0 0. if the electron to acceptor. in sensitive assays with highly fluorescent products. J. sphaeroides reaction center at high magnetic fields.05 mT. (A) Relative triplet quantum yield of Rhodoabsolute pseudomonas sphaeroides reaction center and (B) triplet quantum yield of R . T T=100 K J.6556 As the applied magnetic field increases. B: Reprinted from ref 68. Qr begins to increase slowly and attains parity with the B = 0 T value at B 3 T (Figure 4B).02 0. the Zeeman interaction energy between the T*I levels increases and this causes a decrease in HFI.

the rate of product formation decreases and goes below the control rate at 7 T. so a distinct singlet and triplet reaction product manifold does not exist. "Any process that alters the concentration of the active catalyst (steps a and f. If the R P stays together for -1O-l0 s. Figure 6) will affect the reaction rate. geminate recombination is very favorable because it liberates substantial free energy when the covalent bond is formed once again. Figure 6). Ma netic Field Effects on Thermal (Nonp otochemical) RP Reactions R .ROOH + R " cases of unusual stability. No. a magnetic + n-BuLi - '{F.PhCl ROO' + R'H . Magnetic Field Effects in Catalytic Reactions A remarkable magnetic field dependence has been observed for the cobalt and manganese catalyzed The oxidation of 2.10 Chemical Reviews.Ph' n-Bu'} . Magnetic field effects on thermal reactions have received very little attention.74 This is a landmark result in magnetic field effect studies of catalytic reactions.ROH + R " HO' + R H . the peroxide formed by the reaction of molecular oxygen with an unsaturated lipid is a hydroperoxide (see eqs 5-13). a. The normal reaction pathway is composed of a series of coupled vectorial processes that create an irreversible free energy cascade and drives H' pumping.71 This magnetic field-dependent increase in cage recombination originates from a decrease in HFI . or there is both a triplet and a singlet product manifold. but in fact. 95. Thermally driven homolysis reactions produce a RP that is born in the singlet electronic state. Hence. When RP recombination from the singlet state is being monitored.ROO' ROOH (20) (21) (22) (23) (24) (25) Magnetic field effects are best described for photochemical reactions where the RP is prepared in the triplet spin state. the significant outcome of these experiments is the absolute proof of the formation of a RP in the early events of photosynthesis.68 The biphasic dependence of q vs magnetic field is characteristic h of many magnetic spin-dependent photochemical RP reactions in which HFI decreases ISC at low field. Perito and Corden point out. Rather. Vol. 1995. The thrust of these experiments (vide infra) was not to test the harmfulness of magnetic fields toward photosynthetic bacteria and plants. M. The possibility of observing magnetic spin-dependent chemistry with *OH is uncertain because of the unquenched orbital angular momentum and the ensuing SOC. Furthermore.e. and the competing Ag mechanism increases ISC at higher fields. n-butyllithium will react with pentafluorobenzyl chloride t o form LiCl and n-butylpentafluorobenzene (eq 17)? At magnetic fields greater than 20 mT.ROOH (minor) RO' + R H . or the phenoxy radical concentration (steps c and f. the cage recombination of the spin-correlated lauroyl radicals is increased by 3-6%.73374 rate of quinone product formation exhibits a biphasic increase to a maximum 50%enhancement at 100 mT (relative t o the control rate at 0 T). the oxygen-centered radicals are 10-15 kcaVmol less stable than the possible alkyl radicals. a limiting value of h will be reached when w 2 kT (where kT is the rate constant for formation of 3PIX-*).72 This reaction might appear to be relevant t o the chemistry of biological lipid autoxidation. The fate of the R radical is varied and this radical can participate in free-radical chain chemistry.68 The limit of & = kT/(ks kT) and the rate of reaction of the singlet RP. although they are the most relevant to possible magnetic field effects on biological reactions through changes in RP recombination.n-Bu.'2 At CllH23COOOCOCllH23 - 1{C11H23' 'CllH23) + 2CO2 (18) magnetic fields greater than 150 mT.H 2 0 + R ' R + 0. no electronic spin conversion is required for recombination and only limited atomic motion may have occurred (i. A magnetic field effect has also been observed on the thermal decomposition of dilauroyl peroxide. no magnetic field effect on the photosynthetic quantum yield would be expected in intact and unadulterated photosynthetic systems. . The reaction has only one product. The thermal decomposition of a hydroperoxide will lead to the alkoxy1 and hydroxyl radicals as a RP. A radical pair mechanism is proposed in which the magnetic field-dependent (partially) rate-determining step is regeneration of the active Co(I1) catalyst and production of the phenolic radical in a spin-correlated step (Figure 6). it is easy to understand how changing the rate of ISC will alter the course of the reaction.6-di-tert-butylphenol(Figure 5). incomplete atomic reorganization t o the optimal ground-state geometry). k s . will be independent of magnetic field. Several thermal reactions do exhibit a marked magnetic field dependence.'{ RO' 'OH} '{ROO 'OH} . A RP in which ISC and recombination could compete with the forward step had t o be created in order t o observe a magnetic field-dependent process. This will give the alcohol and HzO as the stable products of the starting hydroperoxide. Except in L. At higher fields. Biological Relevance. In refluxing hexane. Although the w = BPAg term (where w zx kIsc) continues t o increase. the F." Thus. F. 1 Grissom plateau in 4~ becomes apparent above B x 5 T. These factors make it more difficult to understand how a magnetic field can affect a thermally generated RP.Ph + LiCl (17) amount of n-butylpentafluorobenzene (cage product) relative t o (F5Ph)z (escape product) is increased by 50%. + that would otherwise populate the triplet spin states and prevent recombination.

95. (Reprinted from ref 74.H-O-O-C~L~ d L&o[II] + HO2 *e + 0 (e) OH %$ (+ L&o(III)OH % 0 6 Figure 6 Mechanism of 2. other negative reports of magnetic field effects on enzymatic reactions followed (Table l).77 At that time. and (3) ex- OH O.) field can alter singlet RP recombination rates in catalytic reactions with only one product.6-di-tert-butylphenol oxidation catalyzed by cobalt(I1) bis(3-(salicylideneamino)propyl).8 T. T Figure 5. 1 11 1.6 l.6-di-tert-butylbenzoquinone formation vs magnetic field. In 1984. Magnetic Field Effects in Enzymatic Reactions + HzO 6 1. Literature Reports Magnetic field effects on enzymatic reactions have long been proposed. This statement is supported by a theoretical treatment of MFE on the steady-state rate of product formation in a catalytic reaction. methylamine (CoSMDPT). In 1986. Copyright 1988 American Chemical Society.77-ss 1 I I Magnetic Flux DenrHY.Magnetic Field Effects in Biology Chemical Reviews. 0. can be broken down into a series of discrete unimolecular steps that can be described as a series of microscopic rate constants. 1995. S.o : -.8 a - e_ .) . to product. they considered the possibility of MFE on enzymatic reactions through changes in RP chemistry.4 1 0. Relative rate of 2.. no significant magnetic field effects on enzymatic reactions had been observed. and they were carried out before the theory of MFE on RP recombination was developed.76 With a few noteworthy exceptions.2 I I / / 1 I 1.75 Only a transient pair of paramagnetic particles involving either the catalyst or the substrate and catalyst complex is required to invoke a possible magnetic field effect. Vol. Copyright 1988 American Chemical Society. (Reprinted from ref 74. Vanag and Kuznetsov examined the evidence for magnetic spin effects in biological reactions other than the bacterial photosyntheticreaction center. P. (2) the overall conversion of substrate. No.87 Their analysis is based on the central paradigms of enzymatic catalysis: (1)product formation can only occur from the enzyme-substrate (ES) complex. The earliest studies of enzymatic reaction rate vs magnetic field were phenomenological in nature.**- * e *. N. The rate of ribonuclease and cytochrome c reductase was independent of applied magnetic field from 0 to 4.

req. P: is binding of E and S t o form ES. 1 Grissom Table 1.. 17. so the reaction mechanism can be expanded to include the enzyme catalyst (eq 27). the interconversion of these enzyme-substrate complexes occurs with much faster rates. This includes how tightly S is . trapolation t o infinitely low substrate ([SI < K.0 1.F. P. Unpublished results.4 1. T succinate cytochrome c reduction of cytochrome c 5. With these assumptions. supported by NIH GM 32634 to R. Grissom. e Farmer. B. Entries 1-8 are from ref 77. oxidation by H202 horseradish peroxidase linoleate oxidation by 0 2 lipoxygenase Tyr.05 1. [PI = 0 and the reverse reaction.65 1. staphylococcal nuclease. C. However. if the microscopic steps of catalysis are considered. No information as to the rate of ES formation is contained in V.) hexokinase glucose glucose-6-P (coupled assay with GGPDH/T\JADP) guaicol. t o product. The requirements for a magnetic field-dependent enzymatic reaction.0 reductase oxidation of a-dianidizine with hydrogen peroxide oxidation of L-tyrosine by tyrosinase molecular oxygen alcohol dehydrogenase oxidation of CzHbOH. C.0 0.20 0-0. conversion of P to S. since the overall turnover rate of an enzyme may be as slow as a few molecules of product formed per second. No radical pair intermediate is proposed for these enzymes. such that when E exists only as ES. this describes the behavior of the ES complex. the first irreversible step. The first step E +SzES-+E + P kl k3 (27) Saturation kinetics are observed (Figure 71. C. G.84 1985 82 1990 b 1990 b 1990 b 1991 c 1994 85 1993 d 1993 e 1994 86 1994 86 17.contains information about each step prior to. The latter two points are important. We define the maximum catalytic rate as V m a . 92 less than experimental error 8.) will < change the overall kinetics from zero order in S (at saturating substrate where ES predominates) to pseudo-first-order in S (where free E and S predominate). on an intuitive level.B.F. the tangent to the initial slope of the hyperbola (extrapolated to infinitely low [SI).0 same same 0 0 experimental error 92 110 -12 f3-7 12 52 year ref 1965 76 1967 78 1967 78 1971 79 1971 79 1967 80 1967 80 1978 81 1989 84 1989 83. Fitzpatrick. Lee. The asymptote to the saturation curve is V. B. Silverman. B. ~ restated below using the more familiar biochemical formalisms of enzyme kinetics. C. S.28 0-0. In the conditions of a typical in vitro assay. reduction of NAD degradation of H202 catalase liberation of 0 2 during degradation of H202 catalase degradation of HZOZ oxidation of L-ascorbate by 0 2 ascorbate oxidase FMNHz decanal oxidation by bacterial luciferase 0 2 . F. the kinetic rate expression that describes the rate of product formation takes the form of a hyperbola.05 0-1. K. Grissom. and including.. Since P can only be formed from ES.12 Chemical Reviews.. does not occur.4 6. the maximum rate of catalysis is obtained. No. J. B. Chymotrypsin.27 0-0. If the first irreversible step in the conversion of S to P were slower than about lo3 s-l. J. supported by NIH GM 47291 t o P.0 -(5-10) +(5-9) +20 0 0 0 0 0 0 0 0 0 0 no statistical treatment same 15 12 12 15 rtl 13 1<l - - 0-0.25 0-0. 1995.. Under initial velocity conditions. Unpublished results.- 510 +5 f20 13 15 15 -25 -60 a Table adapted from ref 77 (with permission). where Vmax = k3LEItotal.. S==P (26) Product formation can only occur from the ES complex. and..2 0-0. B. Consider a unimolecular enzymatic reaction that converts substrate. as suggested by Vanag and K u z n e t ~ o vare~ . P..5. reduction of NAD glutamate dehydrogenase oxidation of 2-oxoglutarate.25 0-0. Grissom. G6PDH = glucose-6-phosphate dehydrogenase. C..27 0-0. Vol. Hillas. 0 2 L-Dopa tyrosine hydroxylase benzylamine benzalmonoamine oxidase B dehyde A NH3 ethanolamine ammonia lyase ethanolamine acetaldehyde + NH3 (coenzyme BIZ requirement) (unlabeled EA) (deuterated EA) peroxidase change in reaction rate. 95. R. Grissom.27 0-0. Woo. At the other extreme of [SI. random molecular interactions would cause a loss of spin coherence and no magnetic spin-dependent chemistry would be observed. and hexokinase were used in the author's laboratory as controls for spectrophotometer/electromagnet evaluation. This is the MichaelisMenten equation that describes simple enzymatic catalysis (eq 28). [El < [SI.25 . D. and the steady< state assumption can be employed t o describe [ES]: d[ESlldt 0. Unpublished results. Magnetic Field Effects on Enzyme Reaction Rates In Vitro" induction enzyme catalyzed reaction of MF. Unpublished results. d[Plldt = kdES1..0 6. light emission chymotrypsin p-nitrophenyl ester staphylococcal nuclease p-nitrophenyl-dTp p-Nphenol (Ca2.S. reduction of NAD lactate dehydrogenase oxidation of lactate.

125. + 3. in the general case) present at the same time during the reaction. The rate of E S association is. the spin-correlated nature of the RP will be lost. Hence.Magnetic Field Effects in Biology Chemical Reviews. This is tantamount to one round of free radical chain chemistry with an initiation and termination step. 1995. a. all substrate molecules will go forward (given sufficient time for spin randomization).. Generally. 1 13 Figure 7. When 122 7 k3. as all substrate molecules. where K m = (122 k3)/k1. Requirements for MFE on Enzymatic Reactions In light of the theoretical treatment of MFE on chemical and enzymatic reactions (vide supra) and .88 Under this condition. The stable Tyr-122 radical either abstracts a hydrogen atom from substrate d i r e ~ t l y or ~ ~ ~ ~ .e.. Vanag and Kuznetsov suggest that assay conditions in which [SI < K m will produce the largest < dependence of observed rate on magnetic field. the fidelity of product formation is maintained by high barriers to alternate reaction pathways so that only one product is formed. If electron spin relaxation in the {E’‘ S } radical pair is fast compared to ISC. In these cases./Km. consider the case of a MFE on 123 rather than a kinetic isotope effect.The full isotopic discrimination (based on a kinetic isotope effect on k3) for a sticky substrate will not be observed. Saturation kinetics observed in enzyme. ~ Subsequent homolysis of this ~#~l two-electron reduced species or another group on the enzyme generates the active radical enzyme form that abstracts a hydrogen atom from substrate to initiate catalysis. It is largely + through 124 (recombination of {E’‘S} to {ES})that the observed rate of the reaction will be magnetic field dependent. No. In contrast. S is called a “sticky substrate” because once it binds to E. Also. the value of VmJKm will yield the same information but with greater accuracy and precision. where the rate of catalysis (k3) is either much slower or much faster than the rate of substrate dissociation (122). KmRZ kdkl and a true equilibrium of free S 7 and bound S (as ES) is established. will tend to go forward through catalysis. the activation step may still involve a RP and exhibit magnetic field dependent recombination. determined by the rate of diffusion in solution. is the catalyzed reactions. S’ undergoes subsequent conversion to P. Vmax and Vmax/Km. Although catalytic turnover involves only one radical species. Following reductive activation. 95. k3. consider the mechanism of ribonucleotide reductase from Escherichia coli (Figure 8). enzymes do not allow release of a reactive intermediate.and reverse hydrogen atom abstraction produces P and the regenerated tyrosyl radical (Figure 8). Consider the two extreme cases. If the precatalytic step(s) prior to the magnetic field sensitive step are irreversible. if k2 k3.. metal ions) that exert a large magnetic field at the molecular level. since it also contains k3. halogen. At no time during the reaction (after initial activation) does more than one radical species exist in the active site. The parameter K m is often called the Michaelis constant. the tendency is to go forward through catalysis ~~ rather than to d i s s o ~ i a t e . a radical exists at tyrosine-122 of the B2 subunit. Because only one radical species exists at a time. may or may not have a requirement for spin correlation. It is not always a true dissociation constant. cysSH. The kinetic parameter V maximal observed rate when all enzyme exists as the ES complex. etc.~ produces another transient radical speciesg3 that abstracts a hydrogen atom from substrate. 4. it is the ratio of the fundamental parameters. rapid spin randomization that is independent of an applied external magnetic field will occur. metal ion. or nearby paramagnetic centers (i. to a first approximation. Fast relaxation can be promoted by nearby atoms with large SOC (Le. The kinetic parameter K m is the substrate concentration that corresponds to half-maximal V“. As an example. The importance of reversibility in the {ES} complex is obvious when eq 27 is expanded to eq 29 to show the ES complex as a radical pair. no MFE on turnover of ribonucleotide reductase is apected. Now. This leads to the full expression of any kinetic isotope effect on the chemical step. rather than follow the rate of single assays at ultralow [SI. Ribonucleotide Reductase. Many enzymes with radical intermediates must be activated by two-electron reduction prior to catalysis (there are many excellent reviews on the mechanism of enzymes with radical i n t e r m e d i a t e ~ ) . A MFE on catalysis will be masked if the magnetic field sensitive step is preceded by an irreversible step. The tangent to the initial slope is defined as VmJKm.87 In practice. The single radical species exists as a doublet. The tangent to the observed rate at infinitely low substrate concentration is V. regardless of isotopic composition. so the {E’ *S} complex will not undergo diffusive separation. This corresponds to the condition when [SI K m (typically when [SI= ‘ / l a m or less). Vol. rather than either a singlet or triplet RP. Rather. any kinetic isotope effect on k3 will be minimized by the propensity of all S to go forward through catalysis (123) rather than t o dissociate (122). Requirement for Two Unpaired Electrons During Catalysis Perhaps the most significant caveat to observing MFE in enzymatic reactions is the requirement of having two unpaired electrons (or two paramagnetic species. then { E ‘ S } can only react via 125 and the chemical transformation step.). If 124 = 0. the “tightness of binding“ in the ES complex will be largely determined by k2. bound to ES and how fast ES is converted to product (if the conversion of ES to product is the first irreversible step).

^^ The common structural element is the macrocyclic corrin ring that holds Co3+in a square-planar coordination geometry (Figure 9). A strongly coupled RP (with AESTx large) will not undergo ISC. substrate is "non-sticky"). (Reprinted from ref 90. The next two sections describe these results. Furthermore. ( 5 ) The radical pair E' ' complex must exist long S enough for ISC to compete with other modes of reaction. RP recombination). (3) A physical mechanism must exist to promote magnetic field-dependent ISC. no spin-dependent chemistry is defined for a noncoupled RP (AEsT = 0 and rsep= infinite). This may be HFI.) R I n CHZOH Figure 9. Vol. ( 6 )Binding steps and conformational changes that precede formation of the enzyme-substrate RP must be reversible (Le. Mechanism of iron-dependent mammalian ribonucleotide reductase. long- lived). a set of rules can be formulated that describes the situation that must exist in order t o observe a MFE on an enzymatic reaction: (1)There must be at least one step in the reaction that generates a pair of spin-correlated radicals or paramagnetic particles. The form found in nutrition supplements .^^. These stringent requirements suggest that many enzymes with radical intermediates will not satisfy all of the conditions necessary (especially 1and 4)t o produce magnetic field dependent reaction kinetics. Similarly.14 Chemical Reviews. 1 Grissom I I F HO H HO H s-s s-s Figure 8. 95. No. or similar. (2) The RP must be weakly coupled. 1995. Coenzyme Bt2 Photochemistry In its various forms. Following the initial enzyme activation step (not shown). In the author's laboratory. level crossing. Copyright 1988 American Chemical Society. If the ES complex is too stable (Le. vitamin BIZis a cofactor for over a dozen enzymatic reaction^.only one radical species is present in the enzyme active site. An examination of Table 1 will reveal the limited observation of magnetic field dependent reaction kinetics in enzymatic reactions. electron spin relaxation by interaction with the enzymeholvent lattice will remove spin correlation. Copyright 1993 American Chemical Society. along with the magnetic field dependent photochemistry of the Blz cofactor. (Reprinted from ref 103. this active form must require spin correlation. the unique spin-correlated RP chemistry of one coenzyme BIZ-dependent enzymatic reaction has been explored. (4)The observed rate of the enzymatic reaction must be sensitive to the fraction of ES complex in the active form. Structure of cob(II1)alamincofactor. Ag mechanism. 0.) the example of ribonucleotide reductase. No radical pair exists during turnover of the enzyme. o r be directly convertible to a catalytically inactive complex via a reaction pathway that requires spin correlation (Le.

1 0.lo3J08 tolysis was carried out by irradiation at 514 nm in the presence of 50 mM buffer and the indicated viscosigen. no 0 0. but ~ J ~ ~ ~~ the analogous photoproducts are the same: {CH$ Cb111}.15 0. 1 15 hv I JI' I L s' 1 Escape I 1 S T L ISC Escape (Diffusion) I Figure 10. Both the singlet and triplet RP can escape the solvent cage.lo8 In contrast. T +CW decreases by up to 50% in the presence of 75% Figure 11.05 0. the continuous-wave quantum yield (q&) for AdoCbl"' and MetCbl"' was examined as a function of magnetic field in solvents of varying Anaerobic phoviscosity (Figures 11 and 12).~ ~ Visible light below 610 nm will induce homolysis of the C-Co bond to produce a spin-correlated geminate RP consisting of the 5'-deoxyadenosyl radical and cob(1I)alamin {AdoCHz' CbP1}.5 1. is vitamin BIZ. Only the singlet RP can recombine.(c) buffered HzO ( ~ l q = 1). Proposed scheme for the photolysis of alkylcob(II1)alaminto produce a spin-correlated radical pair. q/qo = 30). Magnetic field dependence of &J for anaerobic glycerol (a microviscosigen. 20 "C. q/qo = 30) and 20% ficollAdoCbl"' photolysis at 514 nm. 200 pM AdoCblIII. 4. pH 7.05 0. In buffered water with a relative viscosity of 1.0 and (a) 75% glycerol (q/qo= 30). 95.4 cally produced RP. 1995. curves represent best-fit empirical lines through the data.) viscosity (commonly measured with an Ostwald viscosimeter) and decrease the rate of small-molecule diffusion in parallel (Figure 13). The singlet and triplet radical pairs can be interconverted by intersystem crossing.15 0.1 0. Both viscosigens 50 mM Hepes. adenosylcob(II1)alamin (AdoCblI'I). Magnetic Flux Denrity.Wave Photolysis As a probe of spin correlation in the photochemi4. copolymer of sucrose and epichlorohydrin) increase macroviscosigens such as ficoll-400 (400000 kDa the bulk viscosity. Microviscosigens such as glycerol increase the bulk (Reprinted from ref 103.cyanocob(II1)alamin. No. and the form that is a cofactor for methyl transferase reactions is methylcob(II1)alamin (MetCbl"').0s 0.a-migrations is coenzyme BIZ.4 1 - - A 3. 400 (a macroviscosigen. Vol. ~ ~ . 2. but 20% Ficoll-400( q / q= 30).(b) make RP recombination magnetic field sensitive. but do not significantly change the 't c - .99J06J07 Photolysis of the alkylcob(II1)alamins and subsequent partitioning between cage recombi0 0. In contrast.Copyright 1993 American Chemical Society.Magnetic Field Effects in Biology c Chemical Reviews. Whether bond homolysis occurs from the excited singlet or triplet state has not been demonstrated unambiguously.15 0.the form that is a cofactor for about a dozen enzymes that catalyze 1. The Carbon-Cobalt Bond I All biologically active forms of BIZhave an unusu0 0. Continuous.The ~ o in different ways.1 0.2 nation and escape is described in Figure 10.99-103 C-Co bond in The MetCbl"' is slightly stronger (37 k c a l / m ~ l ) .2 magnetic field dependence is observed.2 ally labile C-Co bond that has a bond dissociation energy as low as 31 kcaVmol (for A ~ O C ~ P I ' ) .

Magnetic field dependence of methylcob(II1)alamin anaerobic photolysis in ( 0 ) buffered water.15 0. 200 M AdoCbl"'.05 0.) rate of small-molecule diffusion (Figure 13). Clearly.) 0 0 0.) Figure 14. Relative macroviscosity and microviscosity of buffered water.2 F" E 6 9.4 1. ( 0 )glycerol microviscosity.04 0. 95. (Reprinted from ref 108. 1 Grissom 1. 200 pM AdoCblIII. and glycerol solutions a t 25 "C. the dependence of kre. The curves represent best-fit empirical lines through the data. Copyright 1993 Oldenbourg Verlag GmbH.16 Figure 13. Ficoll400 can still potentiate the magnetic field dependence of RP recombination.. True geminate RP recombination is being monitored. 50 mM Hepes.1 0. In remarkthen KISC able agreement. pH 7.8 0. Ficoll-400.C) and krec % k ~ s c . since kre. if Ag for the RP is % 2.16 Chemical Reviews. LegFicoll-400 macroviscosity.08 0.. p 1. the Ag mechanism for spin rephasing must be contributing to the observed magnetic field dependence! . Copyright 1993 American Chemical Society. following photolysis of AdoCblIII at 532 nm.( glycerol macroviscosity. 1995. Figure 16.~~~J~~ 3 Laser Flash Photolysis . and (v)75% glycerol. 75% glycerol. a 4-fold increase in krec is observed at 50 mT.08 0. (Reprinted from ref 103.2 U 0. The initial radical pair is formed in the triplet spin state and ISC limits the rate of recombination.) Magnetlc flux Denrlty.does not vary as the microviscosity increases from V/VO = 1 t o V / ~ O 30 (Figure 15). Copyright 1993 Oldenbourg Verlag GmbH..0. 50 mM Hepes. T Figure 12. The curves represent best-fit empirical lines through the data. pH 7. Adenosylcob(II1)alamin photolysis. The rate of decomposition of methylcob(II1)alaminwas determined by monitoring the decrease in absorbance a t 520 nm. The line is the result of fitting the data to the first-order rate equation. In buffered = HzO. the ~ magnetic spin dependence of geminate primary recombination (10-10-10-9 s) can be dissected away from recombination in the bulk solvent lop4 s). with a maximum 3-fold increase in krec in the range 60-120 mT (Figure 15). (Reprinted from ref 108.'03 The geminate RP {AdoCHz' Cbl"} exhibits extraordinarily fast recombination with a first-order rate constant of k. but probably through t h e formation of hydrogen-bonded cage structures produced by the interdigitation of linear polymer ~trands. In 75% glycerol.. Copyright 1993 American Chemical Society. = 1.2 x lo9 s-l (see section 1.0 and (A) buffered HzO (q/qo = 1)or (B) 75% glycerol ( ~ $ 7 0= 30).2 Magnetlc Flux Denstty. Kinetic trace of [cob(II)alamin] after the 30 ps 532 nm pulse as a function of time as determined by the integrated transient absorbance centered a t 470 nm. The absolute macroviscosity was measured with an ostwald viscosimeter and the relative microviscosity was determined as t.12 0. 20 "C. (Reprinted from ref 103.16 Concentration (%) 0 ' o I I 0.0 x lo9 s. on magnetic field is biphasic. (v) 20% (w/v) Ficoll-400.0 { 9: TIME (ns) 0.04 0.o 2.6 0. (in ns) for 4-(2-iodoacetamido)-TEMPO. ( 0 )Ficoll-400 microend: (0) viscosity. No. Vol. Conditions are 20 "C.0 3. Magnetic field dependence of K. The magnetic field dependence of AdoCblI" and MetCbl"' photolysis was also probed by picosecond laser flash p h o t o l y ~ i s .12 0.). Figure 14 shows the time-dependent disappearance of Cbl" following photodissociation of AdoCbl"' by a 30 ps laser pulse. ' In~these time regimes.0 0.

However. AdoCblIII serves a dual role as the initiator. very little force is required to homolyze the bond in the active site of an enzyme.Magnetic Field Effects in Biology Chemical Reviews. to generate enzyme-bound l(AdoCH2' Cbl"} RP in the singlet spin state. the enzyme uses the excess free energy of multiple hydrogen bonding interactions to weaken the C-Co bond such that. In the following section.0. This hydrolytically unstable . The sum of multiple hydrogen bonds to the adenosine moiety and the - CH. In effect. followed by homolysis of the C-Co bond. The enzyme catalyzes the conversion of ethanolamine to acetaldehyde and ammonia (eq 32) in bacteria (the best described sources are Clostridium and Salmo- nella). the RP recombination of {AdoCHz' 'Cbl'I} observed in picosecond laser flash photolysis experiments must be fast enough such that a 4-fold increase in k.ll0Jl1 However.. the net MFE on &W must be the result of RP recombination that occurs in bulk solvent.2-migrations are thought to occur via a transient radical intermediate that is produced by hydrogen atom abstraction by AdoCHZ'. and perhaps as propagator.2-groupmigration 7 in adenosylcob(1II)alamin-dependent enzymes.o 2. 1. The RP will be AdoCblI'I '{AdoCH. Because the rate of an enzyme is proportional to the active form of the enzyme. 200 pM MetCblrrr. Conditions are 20 "C. of radical chain chemistry (Figure 17).NH.CHO + NH4+ (32) The catalytic mechanism is illustrated in Figure 18. refs 110 and 111. as would be encountered in BIZ-dependent enzymatic systems.OHCH. Vol.0 3. Methylcob(II1)alamin 6 photolysis. No.0 TIME (ne) Figure 1 . Many of these 1. From the magnetic field dependence of {AdoCHz' Cbl"} recombination in the photolytically produced RP. the continuous-wave quantum yield for MetCbl"' in 20% Ficoll-400 and 75% glycerol exhibits a biphasic magnetic field dependence that is similar to the corresponding data for AdoCblI". on a time-averaged basis. CH..a-Migration (rearrangement) occurs to form the product radical that is trapped by reverse H donation from 5'deoxymethyladenosine. Because the C-Co bond in AdoCb1"I is very weak.5 ns (Figure 16). Therefore. a decrease in nonproductive recombination will produce an increase in the rate of catalysis. 75% glycerol.' - '{AdoCH.) +Y Analogous picosecond laser flash photolysis experiments with MetCbl'II show that no recombination occurs in less than 3. Because the continuous-wave quantum yield of neither AdoCblII' nor MetCbl"' exhibits a significant magnetic field dependence in buffered HzO of low viscosity. Generalized mechanism of 1.+ P. pH 7. corrinhenzimidazole moiety could easily overcome a bond dissociation energy of 31 kcdmol. recombination from a photolytically produced RP is markedly different than recombination from a thermally produced singlet RP. and (2) secondary recombination that occurs in the bulk solvent at longer times. Kinetic trace of [cob(II)alaminl after the 30 ps 532 nm pulse as a function of time as determined by the integrated transient absorbance centered at 470 nm.' Cbl"} - Cbl"} (30) (31) AdoCbl"' produced on the lowest energy (singlet) surface. 1995. This suggests two different magnetic field-dependent time regimes for RP recombination: (1)primary geminate recombination that occurs in less than 3. it is clear the system can exhibit magnetic spin dependent chemistry. (Unpublished result. Figure 1 . a dynamic equilibrium is established between AdoCbl"' and l{AdoCHz*CblTT} (eqs 30 and 31). By analogy to radical reactions in organic chemistry. 812 Ethanolamine Ammonia Lyase One of the most intensely studied BIZ-dependent enzymes is ethanolamine ammonia lyase.50 mM Hepes. Nonproductive RP recombination of l{AdoCHZ*Cbl'I} competes with forward catalytic throughput that occurs by 5'-deoxyadenosyl radical abstraction of a hydrogen atom from ethanolamine to generate the substrate radical.5 ns. 95. the opportunity to observe magnetic spin-dependentchemistry in BIZ-dependent enzymes will be considered. 1. Magnetic Field Effects on Blz=Dependent Enzymes All of the reactions that require AdoCbl'I' involve a 1.22J12J13 The catalytic cycle begins with binding of substrate to the enzyme-cofactor complex.a-migration as the key step in catalysis. by virtue of the paired electron interaction that existed in the covalent C-Co bond. does not significantly affect the overall fraction of {AdoCHz' Cbl"} in the primary cage that undergoes recombination. 1 17 1.

whereas VmaxIKm exhibits a decrease of 25% at 100 mT.18 Chemical Reviews. ethanolamine ammonia lyase is placed in one syringe and ethanolamine and AdoCblII' is placed in the other syringe..22 If step k13 is circumvented. In this experiment. If HFI that normally populates the triplet RP is disfavored by the application of a magnetic field. product dissociation is followed immediately by the binding of another substrate molecule that is poised for the next round of catalysis. the Ag mechanism becomes significant and increases ISC. kz. 95: No. conditions. a greater magnetic field effect on kz is expressed. k l l and k g should out compete k2 and hq. Enzyme-induced homolysis of the C-Co bond produces the 5'-deoxyadenosyl radical and cob(I1)alamin in the singlet spin state. product - H O I Hq-CHz YH2 X H HCHzAdO HCHzAdO k11 1 product substrate Figure 18. Vol.22 Under conditions of saturating substrate (expressed by the kinetic parameter Vmax). Under Vmax conditions. A decrease in k g will increase the fraction of enzyme that exists as the E-S-(AdoCHz' Cbl'I} complex prior to hydrogen atom abstraction from substrate. This would begin the catalytic cycle with the transient {AdoCHz' Cbl"} radical pair in the singlet spin state. The solutions are . under conditions of less-than-saturating substrate (expressed by the kinetic parameter Vmax/Km). Further experiments at higher magnetic flux density are required to decide which explanation is correct.. Under V.) carbinolamine can be released into solution or decompose t o the ultimate products of acetaldehyde and ammonia. recombination of the {AdoCHz' Cbl"} radical pair ( k g ) is more likely. 1995. Only the singlet l{AdoCHz' Cbl"} RP can undergo nonproductive recombination via k2.22 In contrast.22 The magnetic field dependence of enzyme-bound {AdoCHs' Cbl"} recombination has been verified by stopped-flowkinetic s t u d i e ~ . The net effect is no dependence of Vmaxon magnetic field. The recombination rate constant. This will be expressed as a decrease in Vmax/Km. (Reprinted from ref 22. thus producing the biphasic magnetic field dependence that is observed./K. Copyright 1994 American Association for the Advancement of Science.22 Subsequent turnover must start with homolysis of the C-Co bond t o produce the spincorrelated (AdoCH2' Cbl'I} RP in the singlet state. then kz will increase and cause a net decrease of the forward flux through the first irreversible step. Alternatively. 1 Grissom H O pHpL-CH2 CI H -X.22 The kinetic parameter Vmax is independent of applied magnetic field up to 250 mT. The 5'-deoxyadenosyl radical abstracts H' from C-2 of ethanolamine to generate the initial substrate radical. Deuteration of ethanolamine produces a larger 60% decrease in Vmax/Km 150 mT.22 At higher magnetic fields. The amine group migrates to form the carbinolamine radical that abstracts H from 5'-methyladenosine to produce the hydrolytically unstable carbinolamine product and regenerate the 5'-deoxyadenosyl radical. Reaction mechanism for ethanolamine ammonia lyase. Thus. a specific So-T-l-level crossing might be responsible for the dip in Vmax/Km.22 Replacement of H at with D introduces a primary isotope effect on kj. product dissociation occurs via k13 and the "resting" state of the enzyme-cofactor complex is restored. The magnetic field dependence of Vmax and V m a d Km for ethanolamine ammonia lyase is shown in Figure 19. the enzyme always has ethanolamine bound and the {AdoCHa' CbP} radical pair does not have to recombine between turnover (kll includes product dissociation and substrate binding before {AdoCHz' Cbl"} recombination occurs. will remain unchanged and the net flux via hz[E-S-{AdoCHn* Cbl"}] will increase. ' l ~The~ J ~ apparent firstorder rate constant for Cbl" formation can be determined by monitoring CblII formation on the enzyme.

Acute megaloblastic anemia and other diseases associated with B 2 deficiency can 1 be induced with nitrous oxide.50 0. No deuterium isotope effect on the rate of CblII formation is observed and the magnetic field dependence of Cbl'I formation is independent of isotopic composition of ethanolamine. Without sufficient B12.00 0.. No. Each assay contained 100 mM N-(2-hydroxyethyl)piperazine-N '-2-ethane sulfonic acid (Hepes) pH 7. Magnetic field dependence of (A) Vmax with unlabeled ethanolamine. and (C) Vm.25 i o o h h 0:1 0. The enzyme methylmalonyl-CoA mutase is noteworthy as an AdoCblIII-dependent enzyme that is found in mammals. Tissue becomes increasingly transparent to light above 600 nm (Figure 21)'19 and the possibility of photochemical production of CH3*in vivo and the magnetic field-induced alteration of { CH3' Cbl'I} recombination cannot be summarily discounted. Figure 20 shows the rate of CbPI formation vs magnetic field.4 2. l l ~Nitrous oxide.ll* Considering the high photochemical quantum yield for MetCbP photodissociation.05 0. Copyright 1994 American Association for the Advancement of Science.48.) + rapidly mixed and the rate of Cbl" formation on the enzyme is monitored as a function of magnetic field. is ~ ~ teine to methionine in the cycling of the potent methylating agent. M i 1.0 1 0.4 i z 0. In order to keep the measured rates with deuterated and unlabeled substrates similar.10 0. methyl radical appears to be an unwieldy and uncontrollable molecule that is too dangerous to generate in a biological setting.10 0.20 0.85. n. although mammals have an absolute requirement for B12 to thrive.2-D~ethanolamine. Direct methylation of DNA by MetCbl"' has been reported.0s 0.2. Because enzymatic processes involving Co(I1)and Co(II1)are unaffected. Coenzyme BIZ-dependent1. A caveat to minimizing the biological importance of magnetic field effects on B12 reactions in animals lies in the unusual photoreactivity of methylcob(II1)alamin. In view of the absence of geminate recombination in the photochemically produced (CH3' Cbl"} RP (vide supra).2 .l14 Methionine synthase is a MetCbP-dependent methyl transferase that is found in bacteria and mamm a l ~ It ~ important in the conversion of homocys. Each data point represents the kinetic parameter derived by fitting observed d[PYdt vs [ethanolamine] data to d[Pl/dt = Vmm[Sln/Km [SIn by nonlinear methods.25 3 0. and ethanolamine ammonia lyase a t 25 "C.5 pM adenosylcob(III)alamin. radical chemistry via Co(I1) is not implicated in the mechanism of BIZ-dependentmethionine ~ y n t h a s e .05 0. at least some of the BE-dependent processes that are important for human health will be insensitive to magnetic field on the basis of RP recombination considerations.l12s113 This result unambiguously identifies {AdoCHz' Cbl"} recombination as the magnetic field sensitive step in ethanolamine ammonia lyase.2-migrations appear to be most important in bacterial metabolism.12 0.15 0. Biological and Health Relevance of Magnetic Field Effects 812 Other BIZenzymes that catalyze a 1. rather than the radical. Standard error bars may be smaller than the plotted symbol. varied only slightly between 0. (B) Vm. this may be an adventitious process that is light dependent.2 and DV-/Km = 5.59-fold more EAL enzyme was used in assays with deuterated ethanolamine than in assays with unlabeled ethanolamine.) 1.A 0 .8 f 0.00 0.20 Figure 20. 8. 1995. 6 1 b 0. The Hill number./Km with 1. (Reprinted from ref 22.00 0.16' Magnetic Flux Density. 1 19 *I 0.00 0. Ethanolamine ammonia lyase. The magnetic field dependence of the firstorder rate of appearance of Cbl" with unlabeled ethanolamine is shown. S-adenosylmethionine (SAM). is a potent inactivator of Co(1) processes in vitro and is known t o deplete B12 stores in humans who are chronically exposed. Q.4a t 0 T. It is safer to transfer a one carbon unit as the cation. 95.Magnetic Field Effects in Biology Chemical Reviews. l l ~Methyl transfer in methionine synJ~~ thase appears to occur from the 2-electron reduced Co(1) form of methylcobalamin.15 TT BI 0.50 I 0. or laughing J~~ gas. The magnetic flux density that is currdtly known to alter BIZ-dependent processes is far greater than the .25 Magnetic Flux Density. T Figure 19.14 0. Identical rates are observed with unlabeled and deuterated ethanolamine.2-migration via initial hydrogen atom abstraction might be expected to exhibit a similar magnetic field effect. T 0. However. humans develop pernicious anemia.15 0. (Reprinted from refs 112 and 113. Vol./Km with unlabeled ethanolamine.10 0. The evidence for the importance of Co(1)alamin in vivo is ~ o m p e l l i n g .1. This yields a n observed kinetic isotope effect of DVmax= 6.75and 0. Ethanolamine ammonia lyase stopped-flow kinetic study.20 0.

Thickness = 22-32 mm. R P recombination has been estimated to exceed lo9 s-1. Magnetoreception The evidence that some organisms-from bacteria t o vertebrates-can sense and respond to the geomagnetic field. an enzyme with a nonheme iron center for which a similar R P mechanism has been proposed. salmon.05 mT (depending on latitude) and the sign of the field.143 130 yes yes yes 131-135 137 138 139 140 141 400 500 600 700 800 900 wrvelength. Magnetoreception is the only biological magnetic field effect that is demonstrably of consequence to the organism. the gross magnitude of the field is always near 0. It has been isolated from most of the organisms listed in Table 2. magnetoreception is generally thought of as being distinct from the magnetic field-induced changes in chemistry discussed previously. No. In an end-to-end alignment. the cytochrome P-450 family of enzymes is an obvious place to look for spin-correlated R P intermediates that can partition between nonproductive recombination and forward catalytic throughput. their poles will repel each other and the size of the composite iron granules will be enlarged. nm Figure 21. The sum of the torque exerted on the individual microsomes by an external magnetic field is sufficient t o rotate the cell and impart a signal t o the organism. Honeybees. Biogenic Magnetite Biogenic magnetite (crystals of FesOa) may be responsible for magnetoreception.6 pm in diameter. north vs south. A.) The light dependence is unspecified if unknown. Organisms with Magnetite behavior or light ornanism DroDosed function deDendent" ref(s) directional 123 algae flagellar propulsion 124. thereby suggesting these granules are the primary site for magnetic field signal transduction. A complete treatment of all proposed mechanisms of biological magnetic field effects is beyond the scope of this review and would require several volumes of this journal. environmental magnetic f u densities that are of lx concern to human health.g0 Of these. Non-Radical-Pair Magnetic Field Effects in Biological Systems: What Is the Mechanism? In this section.lZ8 Electron microscopy has revealed innervation of the cells containing the iron granules. Besides enzymes that use AdoCbl"' t o initiate radical chemistry. physics. and mollusks can use the geomagnetic field for systematic food acquisition.130 In a side-by-side orientation. 95. and whales can sense the . hornets. Copyright 1981 American Society for Photobiology. sea turtles. This change in size could be transmitted to the rudimentary nervous system. magnetite is sequestered in iron granules that are about 0.123-141 Birds. and magnetotactic bacteria and salamanders alter their swimming patterns in response to changes in the environmental magnetic field. the magnetosomes (the general term for subcellular organelles containing particles of magnetite) may be placed in a rigid cellular matrix (cytoskeleton) with their magnetic moments aligned. chemistry. 1995. systematic feeding marine migration pigeons. their poles will be aligned and the iron granules will collapse t o a more compact size.85 geomagnetic field and use it for migratory navigation.128 Each iron granule is further composed of magnetite crystals. is important. honey food location hornets food location humans unknown mollusks. homing salamander directional swimming salmon migration directional swimming tuna turtles migration migration whales 126-128 129 142. R. (Reprinted from ref 119. 125 bacteria. that scientists at the interface between mathematics. I. Spirillium bees. some of the more common reports of biological magnetic field effects that occur by mechanisms other than changes in R P recombination will be surveyed. and the clinical sciences can have their greatest impact. tuna. does not exhibit magnetic spin-dependent chemistry. Lipoxygenase. It is in these research areas where theory and chemical intuition does not yet exist. Vol. 1 Grissom I 1o-2 10-3 Table 2. Furthermore. Spectral transmittance through human abdominal wall.120-122 this reaction. Ill. Alternatively.20 Chemical Reviews. is undeniable (Table 2). an In activated (Fe-O)3+ species abstracts a hydrogen atom from substrate to produce the {(Fe-0)2f:'R) radical pair. only a small subset can be considered radical pair mechanisms that may exhibit magnetic spin-dependent ISC. In honeybees. One hypothesis envisions these magnetite particles as behaving like bar magnets.122The symmetrical ligand environment of the heme may offer an additional advantage in minimizing the otherwise highly anisotropic electronic environment of the iron that would strongly couple t o the R P and promote ISC through magnetically insensitive SOC. biology. with the largest having a diameter of about 10 nm. In magnetoreception. Other Enzymes for Which RP Mechanisms Have Been Considered Enzymes with radical mechanisms are numerous. Either mechanism of magnetoreception and signal transduction is sufficient to explain the behavioral response of honeybees and many other magnetoreceptive organisms.

A ~ ~ ~ candidate for the avian magnetoreceptor is the retinal pigment rh0d0psin. The reason for bacterial magnetotaxis is unclear.4)in the pituitary.149 This negative result may be caused by the unusually high magnetic field. trace metal ions may be the desired substance.05 mT. the In ~ ~ excited state of the pigment is the ultimate magnet0recept0r. Vol. Because much of the concern of biological MFE is related to electric powerline and electrical device exposure. sleep. Melatonin controls circadian rhythms. but rather optical pumping of a retinal pigment.05 mT. In single magnetosomes. 1 21 a. rats that were acutely blinded were insensitive to magnetic fields. Melatonin There is a large body of evidence that serum levels of the hormone melatonin can be decreased by exposure to slowly pulsed static magnetic f i e l d ~ .150 Considerable evidence indicates this nocturnal increase in melatonin can be suppressed by exposure to weak magnetic fields. No.1. generally expressed as kT.05 mT magnitude. other research groups have failed to observe a magnetic fieldinduced decrease in m e l a t ~ n i n .l@J@ Remarkably. such that decreased serum melatonin levels are considered undesirable and may be oncopotentiative. and physical and mental performance. B. The signal-to-noise ratio of a signal that is sampled by M detectors will increase by the square root of M . ~ ~ ~ J ~small 60 Hz powerline Thus. i . l ~ ~ MRI relevant At J ~ ~ magnetic fields near 1 5 mT (with the application of . there is an obligate role for long-wavelength (red) light in setting the organism’s biological magnetic compass. it is useful to consider the effect of a small alternating field in the presence of the geomagnetic field. Light-Dependent Magnetoreception. mood. has been put forth to account for magnetoreception in the Australian ~ i l v e r e y e . In some organisms.l~~ net torque experienced by The the magnetosome complex must be in excess of the thermal noise limit. the effect of a 60 Hz alternating magnetic field of 0. A combination of photoreception and magnetoreception has long been proposed for migratory navigation in birds. Without these assumptions. no perturbation in the normal nocturnal increase in melatonin was observed in humans. the direction of light they first see “sets” their magnetoreceptive device to allow them to navigate successfully to feeding grounds. Such a system would have no difficulty responding to the geomagnetic field vector of about 0.005 mT RMS.148J52J54J55 Exposure of the same subjects to light during the same nocturnal hours had the same effect as exposure to pulsed magnetic fields. l @ .140When the initial light comes from the east (geomagnetic and geographical east).l~~ Only wavelengths below 616 nm are effective in allowing m a g n e t ~ r e c e p t i o n . When the initial light comes from the west (geomagnetic and geographical west).147 Adjudication of these different conclusions is beyond the scope of this review. the net magnetic moment may be as high as 5000kT. the magnetic moment may be as large as 25kT. Magnetotactic Bacteria Spirillium is a magnetotactic bacterium that also contains m a g r ~ e t i t e .001 mT by biogenic magnetite.l~~ Although many research groups have observed the magnetosensitivity of melatonin levels. At this level of organization. serum melatonin levels increase during darkness (especially during sleep). Ion Cyclotron Resonance One of the more controversial theories to account for MFE at the cellular level involves the specific .154 This observation indicates the magnetic field-induced decrease in melatonin levels may require light and suggests that retinal magnetosensitivity may be inv01ved. just as multiple sampling of a weak signal is used in laboratory measurements. where k is Boltzmann’s constant and T is the absolute temperature. in the presence of the static geomagnetic field of 0. and salamanders. ~ magnetic fields probably have little effect on magnetite-based magnetoreception that is designed to sense the direction of static magnetic fields near 0. or it may be caused by exposure of the subjects in the dark.lthis hypothesis. This precludes a large fixed-array from sensing an alternating magnetic field as an intact e n t i t ~ . l ~ ~ J ~ ~ The direction and intensity of flagellar motion changes in response to the application of a magnetic field. the same type of calculation allows for the sensing of a magnetic field as low as 0. or even better. the turtles always begin swimming toward the west. further increases the signal-to-noise ratio.Magnetic Field Effects in Biology Chemical Reviews. even in the absence of continued irradiation. Ca2+ Binding and Ion Cyclotron (Parametric) Resonance 1. In this case.1. 1995. There is evidence that induced electric currents may be more important than the magnetic field component in inducing changes in pineal gland metab01ism. an array of cells containing magnetosomes. Migratory Navigation. most notably. sea turtles.la In animals and humans. N-acetylserotonin is converted to melatonin by acetylserotonin methyltransferase (EC 2. a proposal not involving magnetite. we must consider the effect of an alternating magnetic field on individual magnetosomes with magnetic moments not much larger than kT.134J37J40 When loggerhead sea turtles hatch. the turtles swim toward the east. l ~ ~ . birds.l~~ 3.l ~ ~ In what appears to be the rate-limiting step of melatonin biosynthesis. a magnetite- C. gradient pulses for imaging).145~146 To transmit the time-dependent information from an alternating magnetic field.145 Coherent sensing by an array of magnetosomes. but it may be analogous to bacterial chemotaxis that is aimed at food acquisition.144 Recently.l~~ 2.145 In large assemblies with aligned magnetic moments. Beneficial oncostatic properties have been ascribed to melatonin. but the magnetic moment in an average magnetosome is probably on the order of kT. 95. will be i n c o n ~ e q u e n t i a l . l ~ ~ J ~ ~ Thus. based magnetic field sensor must be free to rotate. Thermal Noise Constraints on Magnetoreception A magnetosome that is held loosely enough to sense and transmit changes in the geomagnetic field vector will also experience the small random effects of Brownian m0ti0n.145J46 The underlying assumptions of absolutely free rotation of the microsome and the consequential lack of coherence is not canonical.

chemical. The ion cyclotron resonance model was refined by Lednev within the broader and more general concept of ion parametric r e s ~ n a n c e . Eds.. 0 .173 induction of cellular t r a n ~ c r i p t i o n . M. M. 1989. R. In addition to the studies summarized herein. Biological Effects ofElectropollution. (7) Blank.02 mT (RMS). ’ ~ ~ . (14) Tenford. K. Prentice Hall: Englewood Cliffs.increase ’~~ in c-myc t r a n ~ c r i p t i o nchange in transcription of .April. 1990. C.. dropped over a distance of 1 pm. Prentice Hall: Englewood Cliffs. and this lead to a proposed “dampening”of the oscillation by solvent molecules or active site binding. Errors in the original application of ion parametric resonance to biological systems have been c ~ r r e c t e d .Li+. W. even a modest potential of 1V. 1 Grissom interaction of an oscillating magnetic field and a static magnetic field at a frequency that corresponds to a specific energy level of a metal ion cofactor.158-160 these frequencies. D. Ed.170 Under nonresonance conditions. P.170These data support specific predictions of the ion parametric resonance model. NJ. J. 1993. Noteworthy among these are reports of a magnetic field-induced change in cell surfaces. (10) Carstensen. Mechanistic Approaches to Interactions of Electric and EZectromagnetic Fields with Living Systems. E. no change in neurite outgrowth was observed. F..183-188 Electroconformational coupling has been demonstrated for the transport of ions by the membrane-spanning ( N ~ . T. 1986. 1979. W. (4) Bennett. Schulten. L. As with any global theory that tries to explain disparate observations. Miscellaneous Magnetic Field Effects The focus of this review has been biological magnetic field effects for which plausible chemical mechanisms have been put forth. Sept 25-30. Chicago.170 The percent of neurite outgrowth (NO) following stimulation of PC-12 cells with nerve growth factor was determined under 45 Hz resonance conditions for the following ions: Ca2+. L. Reports of experimental data that support and refute the concept of ion cyclotron resonance effects on ion binding and transport have appeared. E.. S. K. 1994. In 1985. Gandhi. is electroconformational coupling of the dipole moment of membrane proteins and enzymes to the alternating electric field vector. J . DC. Ed. Acknowledgments A portion of this work was supported by a grant from the National Institute of Environmental Health Sciences (ES 05728) to C. Biochem. Phys. 23. Millis. Biophys... W. ~ ~ ~ J ~ ~to the theory of ion According parametric resonance. (15) U S . and H+. Information Ventures. Eds. Biological Effects of Power Frequency Electric and Magnetic FieldsBackground Paper.. J. is amplified to a voltage drop (electric field strength) of 1 x lo6 V/m! There are demonstrated biological effects of oscillating electric fields that appear t o depend upon voltage amplification across the cell membrane. (13) Dutta. Electric Field Effects In this review. Ed. H. L.l~~-l’* increase in ornithine decarboxylase gene e x p r e s s i ~ n . Brain Tumors and Experimental Models. Ed. References (1) 1.. 1969. Debrunner. Electromagnetic Interaction with Biological Systems. (6) Silverman. Biological Effects of Magnetic Fields. F. i l l ) Blank. Belford.172 enhanced biocide activity against biofilm-sequestered bacte- . Elsevier: New York. N. 297.. 1978. Because the cell membrane is an efficient dielectric that is only a few micrometers thick. G. Government Printing Office: Washington. Submitted for publication.. Third International Converence on Magnetic Spin Effects in Chemistry and Related Phenomena. Ed.. Barnothy.. S. E. Biological Effects of Transmission Line Fields.. K ) .005 mT (RMS) t o 0. Up t o a 70% decrease in neurite outgrowth was observed as the magnitude of the ac field was changed from 0. 161-164 a. Findl. 272.Mg2+. Vol. U.. Ed. Office of Technology Assessment. Epidemiol. M. Postow.. the resonant path At of oscillation was greater than 1 m. Belford. Schulten. l and -similarity t o heat shock~~ ~~~ induced changes in protein distribution. Biological Effects and Medical Applications ofElectromagnetic Energy. (17) Canfield. Electricity and Magnetism in Biology and Medicine.V4+. Plenum Press: New York.l ~ ~ Recent data on changes in neurite outgrowth from PC-12 (nervous system) cells supports the window effects and resonance conditions predicted by ion parametric resonance theory. Schulten. (16) Canfield. 1990. P.. In Biological Effects and Medical Applications of Electromagnetic Energy. including the binding of an ion to a macromolecular ligand. M. 2.. 273.G. 1995.~~~ other gene~. CRC Handbook of Biological Effects of Electromagnetic Fields. Phys. Debrunner. The author thanks the generous efforts of reviewers in the physical. 1989.. 1964.182. M. G. R.Mn4+. Ed. (5) Werthheimer. C. there are numerous phenomenological observations of magnetic field effects for which a satisfactory mechanistic explanation remains elusive.lg2 The reader is directed to the primary literature cited above (vide infra) for lead references. Leeper.109. K. Liboff applied the equations of ion cyclotron resonance to the binding of Ca2+by biological molecule~. Chem. Debrunner.. Ion Parametric Resonance. CRC Press: Boca Raton. OTA-BP-E-53. 0. Today 1994. K. T. 1987. Congress. M. NJ.171 alteration of bacterial growth rates. (2) Barnothy. R. J. P. 119) Blankenship. As with any measurement involving the visual quantification of cell morphology. J . these results await duplication and verification in other laboratories. 1.7.B. Plenum Press: New York. Fe3+. Inc: Philadelphia. and biological sciences. C. more experimental results are needed to evaluate its relevance to biological magnetic field effects.A T P ~ S ~ . J. (3) Frey. J. ~ ~ ~ - IV. M. 1987.. ( 8 ) Gandhi. P. P. R. Schaafsma.. M. E. Abstracts Book. This theory was attractive because it predicted a “window effect” by which only certain frequencies (near the 50-60 Hz powerline frequency)would lead t o the resonant absorption of energy by an unhydrated metal ion such as Ca2+.22 Chemical Reviews. FL. many physical parameters that describe the interaction of an ion with its environment will change. E. 1993. (18) Canfield. E.S. 1986. No. 1994. L. A m . V. Acta 1977. 95. ria. San Francisco Press: San Francisco. Plenum Press: New York. Vol. (12) Polk. p 414. Plenum Press: New York. the effect of the electric field vector that accompanies an alternating magnetic field has been ignored. R.461. Ion parametric resonance offers a mechanism that specifically predicts frequency window effects near 50-60 Hz. IL.. A. R. Magnetic Field Effects on Biological Systems. FASEB J . G.l~~J~~ Ion cyclotron resonance theory describes the frequency-specific absorption of electromagnetic energy by ions in a weak magnetic field and the path circumscribed by these ions. Plenum Press: New York. (9) Lin. Parson. Biological Effects of Magnetic Fields. Among these. Belford.

.. (70)Chidsey. 25..115. CRC Crit. (41)Buchachenko. Focesi. (99)Endicott. B. A. Shein. T. J. R. M. J. R.124. White. B.. Naturwissenschaften 1992... Biol.1073. B. Bicudo.27. J. Aug 23. A.5. I. E. 106. DC. 1995. Phys. Salikhov. (59)McLauchlan. Elsevier: Amsterdam. Martl. Schaafsma. Lindauer. (Usp. 1994. Chem. (117)Drummond. Biochem 1982..33... 1982.33. Am. CA. J. N. (43)Schepler. P. G. (120)Groves. Mulay. 1445. l . (100)Krautler. 958. (29)McLauchlan.. (97)Koenig. SOC. Wolfe. B. . U. T. E. M. Dunham. Znd. Phys. Buncel. SOC.101. Ed. Russ. J. Weller. Ed.. C.7. 169. (89)Cleland.885. van Grondelle. N. Nickel.. J . (36)Turro.51. J. Acta 1977. D. 7. 1965.51.-J. R. SOC.. A. F.. Chem. Photobiol.2040..1978. 547... J . Chem. 1990. Wiley: New York. Biophys.. Matthews. C. Kim. Zzu. SOC. (121) Guengerich. Biochem. Abeles. W. A. (24)Chen.385.551. Whitear. A. N. F. Molin.15. Clarendon Press: Oxford. N.Biol. Pritze. A. J . 5071. M. J . (95)Schneider. S. 1987. SOC.. Biochemistry 1993. K. J. Nishino. Chem. (54)Margulis. D.2674. A. Hsu. Photobiol. 3249.162.110.. Stearns. 0. I. 1965.. A. Chagovetz. Khim. Damle.. Eklund. 1978. Chem. 1993. Acta 1977. L.116. 1973. Roelfs. Redemarker. S. In Isotopes i n Organic Chemistry.278. T.73. B I Z .Vol.. E. Waggoner. Turro.. E.) 1977.34. N. J. J . (21)Parson. T. J . Blakemore.. T.. (61)Grundler.113. J.101. J. Washington. Photochem. J. K. P. U. B.. 1986. 58. B. R. C. Pritchett. M. SOC.547.. F. (63)van den Brink. Hoff. J . Parrish. 8045. Hewitt.Wiley: New York. R. J. Molin. (110)Lott. 24. Natl. Acta 19711. J . 1970..5204. SOC. Chem. Y. Chem. (56)Huyser. T. 1981. (86)Harkins. Biofizika 1978. Coon. R. 1990. H.. Biochim. 1986. Biophys.. Magn. E.. Duysens. 1967. F. R. A.3732. B. 1986. L. Salikhov.203. 1988. Lee. (101)Chen.. L. A. McLauchlan. J. (73)Perito. S. C. Acta 1977. Biophys. Acta 1994.. K. Am.11. P. V. B. (82)Petushkov. 1991. W. Turro. Chem. Kuznetsov.. Finke. (105) Martin. H. Fedorov.537. P..89.. . E.. Presented at the American Chemical Society Fall National Meeting. Van Grondelle. Mumumbar.8317. J.265. B.263.. Washington. (34)Salikhov. C. 1981. McLean. Bioelectromagnetics 1994. Ed. Biol. Acad. R.1994.. C. Chem. W. (107)Pratt. Yasina. Biochemistry 1994. Chem. 1965. M. J. S. H. (22)Harkins. Duysens.252. B. Matthews..p 191. K. 1983.. E. E. (102)Chen.-J. 1993. Sjoberg. R. Photobiol. In Photochemistry and Photophysics. R. T. Soc. A. 1985. 41..Vol.1469. M. Biol.V. A. 1990. J. G. Chem. Res. Chem. T. T. 1993..117. Grissom.January.318. Chidsey.. Finke. 371. J. W.. T. (103)Chagovetz. C. Mol. R. J. (112)Harkins. M... Am. W. J . P. (52)Zimmt. A. V. Commun. Kraeutler. Buchachenko. 297. Chem. A..241.213. A n Introduction to Free Radicals. A m . McDonald. Kuznetsov. 1988. Biochemistry 1988. (Usp. 1984. (25)Hayashi.. J . 1992. H. Int. K. (55) Buchachenko. C.103. (62)Diesenhofer. Sou. K. 767. Am. L. Mol. Rev.618.679. J .Jpn.-W. 1977. C. Acc. Acc. S. B. G. 1980.. Kuznetsov. Chem. Podoplelov. (27)Salikhov. E.. Rev. Leshina. Keilmann. R. in press. G. T. Chem. J. University Science Books: Mill Valley. 1985. G. M.54.. Chem. (109)Lavrenko. R.99. S. 1984. B. Sagdeev.569. P. V.-K.. 334. (46)Richards. (23)Steiner. 517..80. H.. R. A m . K. R. A. Polyhedron 1988. K. 1977. Am...3365. A. C.. 3415. E. B. Jpn. P. H.795. S S S R Ser.45.-. Chem. Am. C. (35)Buchachenko.T. 1978. H.. B. M. SOC. W. I.74. 1982. Nature 1967.S. N. Grissom. (126)Martin. V.. (87)Vanag. M. L.. G. (123)Torres de Araujo. FL. W. L. M. R. S. Am. T.107. Photochem. I.. Sterna. Madden.29..23.Phys. 1983. J . B. R. W.. W.Phys.. Annu.82. Kobayashi Kirschvink. (49)Levin.405.. Hay.. Kraeutler. (88)Cook. (98)Halpern. Chem. (119)Wan. R. P. Biochem.A... V. Sci.Naturforsch.Magnetic Field Effects in Biology (20)Hoff. S. (39)Leffler. Soedin. A.265.46.-Y. Grissom.de Gruyter: Berlin. Polyhedron 1988.. 1480.. D.. Chem.. Sagdeev. (83)Ghole. Abstract MA 1. 2419. K. Trivedi. L.. A.. J . Lawrence Berkeley Laboratory. Maling. Chem. N. Weller. Comp. Phys. Ulrich.. W. P.. Leung.7432.. (125)Frankel.. N. Molin. R... G. A. 1986. R. 163. J.. A. B. Chagovetz..369. Mokryukova. Biophys. 1971. Res.4 1450..31. (116)Drummond.145. N. Y. Boxer. J. (113)Harkins..460. J. Acta 1975. F. 1988. Phys. H.2498. Rev. T. Chem. SOC. W. Chem. 1978. Mol. (28)Schulten. Chem.Phys. (66)Blankenship. J. R. (67)Werner. H. Rev. P.. N. Kuzmin. 234. A. G..122.. P.13.. G. B. Bull.1452. DC. V.2. 1991. B. (31)Turro. FASEB J . Res.-. J. J.182S. D.. SOC.. J. Chem. 1984. Z. 1985. Comprehensive B I Z . CRC Press: Boca Raton. P. Didenko. 1991. A.13. (74)Perito. Michl. R... Galbraith. J. M. Proc. Zool. J. Eds. 958.Miki. (122)Ortiz de Montellano. B. R. Warner. A.. (114)Cannata. (106)Endicott. M. H. Am. W. Borden. L. Chem. Y. Weissbluth.. A.62. N. Vysokomol. In Photochemistry and Photophysics.3. 1-2. 1989. H.33. Chem.. A. J. Am. 5329. Chem. C. M. Nature 1986.187. L. T. J. Heinsohn. B. J . (85)Hwang. R.31.. M. K. 28A. (77)Vanag. Jacobs. F. Steiner. W..4418. (68)Boxer. J .236. M. R.585. M. (60)Hamilton. Lett. J...30. D. Phys. N. Brussels. CRC Press: Boca Raton. 181... 1979. Biochem. Russ. Biophys. 95. (57)Scaiano. Mazumda. K.. J. G. (111) Lott.. (69)Boxer. Biochemistry 1994. Kaiser. Am. 6726. Presented at the American Chemical Society Fall National Meeting... G. J. 215.. Experientia 1977. M. G. K... G. Stroinski. (76)Maling. (94)Dolphin. (38)Buchachenko. Staerk. J.593. R. H. M. No. A. C. 1985. J. 1993. (58)Scaiano. 1986. M. Steiner.5. McLean. Ishay. 6. A. (48)Koziar. C. L. E. B. 510. L. G. World 1992... N. Wolff.93. G.. R. J. T. 1977. Epp.N. L. Enzyme Mechanism from Zsotope Effects.1990. SOC. Russ. Pires. (42)Turro. J. Walleczek. E. 1987.p 259. T. R. C.. R.. P.) 1976... Staerk.1185..17 9. B. R.-J. R.240. Cowan. M.. R. A. D.4000.265. 1993. J. 375.114.. F. Science 1994. 1963. C. Serebrennikov. 2. (26)Steiner. B. J .. SOC. 2. J . Biophys. V. (124)Blakemore. J . Reu. J. J.. Phys.25. Rademaker. M. L. H. Science 1994. A. (118)Pfohl-Leszkowicz. V... (127)Kirschvink. F. Conference on the Primary Electron Transport and Energy Transduction in Photosynthetic Bacteria. Biochem. J. ACC. Phys. J . Am. Mujumdar. K. S. P. Manikowski.. A. Chem. Galimov. Cozens. (32)Khudyakov..243. Leshina. J . J . Sands. J . 1990. Spin Polarization and Magnetic Effects in Radical Reactions.'T. 1 23 (71)Sagdeev. 1993. Am. R. Z. D. H. Chem. Grissom. V Chapter 1. J . . M. 2. Y. Spin Orbit Coupling in Atoms. N. B. (92)Stubbe. Chem. A. Ferraudi. Itoh. J . C. K. P.397. 1994. E. Y. (51) Luby-Phelps. Scaiano. B. Nauk. J . Phys. D. A.. Catal.. Schuler. J. Chem. Grissom. C... E. McClusky.. A. J. SOC. (33)Sagdeev. CRC Press: Boca Raton. Chemical Reviews. Grissom. C.377. A m . Neue Folge 1976. Mohtat. T.. (78)Rabinovitch.. L. V. Gast. Wiley: New York. C.3742. A. (93)Nordlund. 2419. R. Chem.. A. R.. Akad. Doubleday. Biochemistry 1991. A. Finke. (84)Hummel. Chem. (37)Turro. L. M. 95. Biofizika 1985. (75)Vanag. Chem. Biochem. (80)Haberditzl. J .. Cooper. A. E. C. Weissbluth..43 727.. J.. H. Chance. (44)Fessenden. (72)Tanimoto. 1993. Khim. .. Res. 1977. Rabek.62. V. Angew.. 1987. S. Reson. (79)Muller. Chem. N. (129)Kisliuk. Biophys. V. 23.. M. Anderson. L. B.375. A. U. (30)Werner. Biochem.263. Werner. Phys. W. H. R. Rev. Z. K. Ed.112.483. C. T. 1355. 1982. Corden. Khudyakov. Grissom. B. R. M. D... B. Lett. N. F. Grissom. Stepanek. N. Cozens. L. (128)Li. Russ. Chem. M.. W. Free Radical Chain Reactions. 1980. (115)Banergee. 3893. W. Chem. Elsevier: Amsterdam.603. 12987. (64)Hoff. 1989. K. A. J . (90)Stubbe. A. Modern Molecular Photochemistry. F. B. A. 1977.461. Ernst.. K. SOC..460. B..p 17. Phys.1984. (47)Turro.Vols. (104)Martin.177.502.58.p 54. P. Annual Report. M. Am... C.. Parson. E. M. Frankel. Netzel. 0.. J. 1994. G. SOC.68. Chem. N. Biochem. In Diradicals. L..79. J . E. 349. 185. Physiol. A. Kamha. Photochem. Phys.7. M.. Chem. (91)Stubbe. Chem. B. (40)Turro. A. J .585. Keith...100. Soc. Aug 23. SOC. R.. (108)Grissom. J .81 154. 0. P. J . C. A m .' S.. B.32. Fee. M. H. S. Kuzmin. Rev. C. Chapter 3.-J. Rabek. (45)Norman. Vol. R. R. (96)Hay.549. Roelfs. Thiemann. G. Martl. 12152. Org. W. Vol.2147. F.. Huber..79. Science 1994.. Lett.. 1986. 257. J. Dirheimer.190. B. (81)Vainer. Ochoa. A. 0. W. W.345. V. U. (65) Hoff. A. Y. Phys.. Science 1979. Schulten. R. Chem. G. V... Chance. R. 319. L.. Molin.Haberditzl.566567. S.1499. F. Finke. Nature 1990. 1995.50. M. (53)Cozens. 1968. 2. C. Kraeutler.115.5. Phys. Chapter 2. Corden. Biophys. Engl. Matthews. J . E.109..65. E.1970.104. R.-P. 1991. B. G. Weller.. M.41c. 1984. 1989. A m .. C. SOC. L. L.-C. Kraeutler. R.1992.65 423..109. Chem. Roelfs. SOC.39. Chidsey. Biophys. Biophys. Wiley: New York. R. L1.. M. 355.59.3472. S. R. N. J. L. S. 609. Chem. Chagovetz. Thansandote. Chem. J . Science 1975.. U. Biophys. Biofizika 1984. E. (50) Pines.104. 1991.

(157) McLeod. J. Dixon. R. 1156) Liboff A. D. 1984. Comp. (161) Liboff.140. Blank. L.. Plenum Press. B. Science 1990. Janata. L. B. (153) Lerchl. 1990. 1177) Phillips... Sci. (134) Wiltschko. E. Janata. J . B. Greenebaum.. R. J . J. 144.. H. 1990.. J .. B.275. Painter. L. Proc.142-144. A..6 . C. 381. C. (154) Olcese. Science 1987. J. (132) Walcott. D.54. Mild. A. C. (144) Leask. Biol. P. K.. Tsong.138.7683. Radiat. (165) Lednev.A. R.. Exp. 459. J. Blackman. W. A. M.: Goodman. SOC.. J. Ed. J . (178) Goodman. R. B.: Chen. Anderson.49. . Physiol. J. J.: Blanchard. Ann. 1984. Sharpe. J . 547. (136) Phillips. J.382. Mohammad.Transl.6. W. Phys. T. Schwan. Vol.: House.. G.. R.: San Frakisco Press: San"Francisc0.267. Microbiol. H.1132. C. S. J . F. 1993. (190) Blank. San Francisco Press: San Francisco. J. A. Res. Burns. (142) Kirschvink. (131) Walcott. C. In Mechanistic Approaches to Interactions of Electric and Electromagnetic Fields with Living Systems. (164) Bruckner-Lea. (151) Lee.Ed.: Soo.89. Science 1983. Blank.5062. Parkinson. Tsong. B. p 398. Flanders.: Sokolic.1 (147) Polk. (160) Sandweiss. L. 1982. Derpapas. R. B. 1992. 1174) Goodman.857. K.: Ford. (172) Moore.333.. Nature 1977. ~. 0. S.. P. R. W. Artificial Internal Organs 1994 40. M. A. Phys. p 528. 217. 77.. L.: Benane. Rushforth. T. 0.. D. (145) Adair.. 625 .92. K. T.359. 1993.. 1990. H. F. Bruckner-Lea. Plenum Press. S. Bioelectromagnetics 1994. 0. Green. Behau.91.. V. In Electricity and Magnetism in Biology and Medicine: Blank. B. R. Pineal Res. Biochem.48. FASEB J . V. A. A. (182) Blank.. B. 1994. Robertson. Vollrath. (148) Reiter. (169) Blanchard.. Astumian. T.. 1992.. R. 4222. P. Brannon.. R. Bioelectromagnetics 1993.... Physiol.24. Biol. Phys. 95.: Adey.. Ecol. Yaswen...89. Brain Res. Bioenerg.-0. J. p 281.. Carcinogenesis 1987.S. Nature 1993.Reuss.155.199. S. S. Ed. Pieper.281. M.. E. D. Neuroscience 1992. K. (133) Leask. L. L. C. Bioelectromagnetics 1994. F. A. Acad.4. C.91. 969. (168) Lednev.51.71.--- Grissom (163) Durney. J. R.673. R.10.259. Conf IEEE Eng. J.. E.: Shirley-Henderson.239: (171) Goodman..267. Sokolic. T. A.19. L.. Biophys. W. Nature 1977. C. 25. R.12.13.3189.1089. J. H. (175) Lin. (158) Durney. 1986. 1993. L.: Kobavashi-Kirschvink. Willows. Thompson. L. (185) Robertson.. R. R. L. SOC. Biol.Rappaport. Blank. Biochem. E. Kartun.58.. R.. A. J. Y. C.. M. B. A..-Usp. A. Noore.V. 1992. (167) Lednev. P. Bioelectromagnetics 1988. J . 1994. Kaminski. T.331. J. C. Ed. S.. V. 1985. 7155. J . B. 313. No.25.8.. M. 2283. Thomas. L. A.. In Electricity and Magnetism in Biology and Medicine. Sloma..184. P..4891.S.. Acad. Lasch. Bioelectromagnetics 1992.510.. Durney.24 Chemical Reviews. Cell.?64.. Imag. D. Cancer Res.180. C. (170) Blackman. Exp. (187) Weaver. (155) Liburdy. M.. Comp. Eds. J . (138) Quinn.5. Chem. J. 1993. Nature 1992.147.: Ishida-Jones. 1989. D. 1993. 22.. J. H. 1. D. E. (166) Podgoretskii. S. M. F. W. (189) Serpersu. V. (179) Byus. M. G. D. D. Pineal Res.. 128. Pieper. J. 11431 Kirschvink. 109. C.. C. C.-D. C. R. J . N. (137) Phillips. Henderson.. Joines. C. Astumian. Science 1974.A. 1991.)1964. V. V. San Francisco Press: San Francisco. A. Nicolini. Anderson. T. 1990. J.. E. J. In Electricity and Magnetism in Biology and Medicine. Y. S.12. L.65. (141) Kirschvink.: Ishida-Jones. D.. R.. T. M. Kinney. J .1. Yaswen.. M. F. 1995. M. K. W. Sociobiol. J.. Wiltschko.: Munro. Haggren. R. A.. Benane.2925. Anninos... J. R. Goodman. NIH Res.. Int. C. Bioelectromagnetics 1992. A.147. A. Reson. W. (159) Halle. 1027. R. S. Abstr. W. 144. Biol.682.. M. C. 1993. Eds. H. C. (149) Schiffman. M. A . (140) Lohmann. 1993. Astumian. M.. M. D. Natl. Chem. Bioenerg. In Electricity and Magnetism in Biology and Medicine.. R... 315. M.. B. 1982. Biochem. K.9. CR940164W .:Reiter. Bioelectromagnetics 1988. p 500. Y. P. R.: Woodford.12. Chock. (191) Blank.. M.. W. Grissom. FEBS 1986.. Gould. V. P. Bassett. 1994.Fittinghoff Lohmann. H. 1993. H.. Adey. C. 1988. M. (162) Blackman. 1994. 1385. Burk. J.. W. A. (186) Markin. Natl. J . Reprod. S. Science 1979.. 1986. Bioelectromagnetics 1991. Khorkova. Borland. U.55. 1989. I. J .M371. Acad.. J. W. 190. L. (152) Liburdy. U. Astumian. Robertson. R. Stormshak.335. B. M. Liboff. Marron. E. M. A m . Sci. R.15.7 . R. Kaminski. J . Ed.14. A. Sci. Proc.: Adey.9. H. U. In Electricity and Magnetism i n Biology and Medicine. J.93. E. T. (146) Adair. (176) Phillips. (183) Westerhoff. Personal communication. 1987: p 97. Biophys. J . M. 261. M. P. J. R.14.. New York. Kirschvink. K. 4734. F. Bioelectromagnetics 1991.: Soo. Acta 1992. 1184) Astumian.: Nonaka. P...31. Rehnholm. Can. Kirschvink.. Adey. B. Thomas.A.. G.. Haggren.11.247.. Bioenerg.."S. Sou. Blank. R. C. T. Rollag. Bioelectromagnetics 1994. V. Henderson. Y. U. J. (Engl.15. R. U. Magn. 1979. F. K. M. K. J . J. Tsong. C. (139) Walker. Med. M. R. San Francisco Press: San Francisco. G. Brainard. Bioelectromagnetics 1990.61-8.1145 (173) Benson. D. J . E.. A.. Blank. R. p 497. A. (188)Robertson..220. Rappaport.. (135) Nowak. 120. (180) Byus.. M. Proc.: Westphal. H.. (181) Mattsson.205. 1985. M. Tsagas.. 203. J . Biochem..83. C. Natl.235.13. Chem. S. 1 (130) Lohmann.. Richardson. Woodford. Bioelectrochem.. J.. L.. Biochemistry 1992. Findl. B i d . Thinesen..: Astumian. P. E. S. Intl. L. R. 1991. R. San Francisco Press: San Francisco.: House. (150) Sandyak. C. V.. New York. P. E.15. T. p 550. R. In Interactions Between Electromaenetic Fields and Cells: Chiabrera.