Free Radicals and Lipid Peroxidation: What They Are and How They Got That Way

William A. Pryor

I Introduction

The aim of this review is to outline what radicals are and how they react. We will start with a discussion of the lifetimes and reactivities of different types of radicals. Then we will examine the mechanistic types of reactions that radicals undergo. The peroxidation of lipids, one of the oldest examples of a radical chain reaction that plays a role in biological changes, will be discussed. Finally, we will consider some “side reactions” that very often accompany the main processes that occur in lipid peroxidation and that have biological consequences.

The term “free radical” was first used in the debates that took place among chemists in the period 1750–1890 about the possibility that parts of molecules could have an independent existence. In that period, analytical methods in general and molecular weight methods in particular were so crude that it was not clear whether a part or a whole molecule was under observation. For example, if butane were to split to form two ethyl radicals, as shown in reaction (1), the methods of the time could not distinguish between the two “free” ethyl radicals on the right and the two “bound” ethyl radicals in the molecule butane, shown on the left.

H3−CH2−CH2−CH3→2CH3−CH⋅2

Some of the problems that chemists of the 1800s had with radicals can be seen from the quite acrimonious quotes from leading chemists of the era, taken from the review by Pryor (1968). For example, Laurent in 1842 stated that “chemistry has now become the science of [radicals] that do not exist!” In contrast, Frankland in 1850 argued that “the isolation of these four radicals [methyl, ethyl, valeryl, and amyl] eliminates any doubt of their actual existence and is a complete proof of the correctness of the theories of Berzelius and Liebig.” The murky status of radicals becomes clear from the statement of Wurtz in 1856: “Gerhardt’s assumptions that free radicals do not exist … cannot be maintained. Ethylene, carbon monoxide and sulfur dioxide may be regarded as free radicals since they form compounds by direct addition of two atoms of chlorine.” Clearly, Wurtz did not use the term “radical” in the same way that we do now, because neither carbon monoxide nor sulfur dioxide is a radical (although nitric oxide, another common gas, is). Mechanisms as well as atomic structures were unknown in those days, and chemists did not recognize that compounds that contain a carbon–carbon double bond (olefins) can add “two atoms of chlorine,” i.e., Cl2, either in a radical chain process or in an ionic process that does not involve radicals, depending on conditions (Pryor, 1966).

Thanks to pioneering work by Gomberg in 1900, Nernst in 1918, Herzfeld and Polanyi in the 1920s, and Paneth in 1929, the existence of “free” radicals both in the gas phase and in solution was unequivocally established in the period 1900–1930 (Pryor, 1968). Today, we use the terms “radical” and “free radical” interchangeably.

II Putting the Proper Spin on Radicals

We should start by defining the equivalent terms “radical” and “free radical”; the formal definition states that a radical is a chemical species with an unpaired electron (Pryor, 1966; Walling, 1957). Radicals can be neutral or negatively or positively charged. Most of the organic radicals we will be discussing are neutral (3., RO·, etc.), but many radical anions and radical cations are known and are important in biology. For example, superoxide, ⋅2−, is a radical anion. The oxidation of polycyclic aromatic hydrocarbons (PAHs) by a one-electron step is relatively easy, because the odd electron can be delocalized over the (generally) large number of benzene rings in these “bathroom tile” types of molecules; when a PAH loses an electron, it becomes a PAH cation radical, AH⋅+, and these species can be involved in the oxidative processes that convert PAHs to carcinogens (Todorovic et al., 1993; RamaKrishna et al., 1993; Cavalieri and Rogan, 1984; Rogan et al., 1993; Devanesan et al., 1993).

Chemical bonds are made up of a pair of electrons that have opposed spins, symbolized in this way: . In a two-electron bond, these two electrons occupy the same orbital (space), located between the two atomic nuclei that are bonded together by this electron pair. When a bond breaks, either both electrons can attach to one partner, as shown in reaction (2), or one electron can attach to each partner, as shown in reaction (3). In the former case, called heterolysis, ions are formed. In the latter case, called homolysis or homolytic bond scission, two fragments, each with one unpaired electron, are formed; these species with an odd number of electrons are radicals.

−Y→X++:Y−

−Y→X⋅+⋅Y

Because virtually all stable molecules have an even number of electrons in “closed” (complete) electronic orbitals, the scission shown in reaction (3) gives species with an odd number of electrons. These free radicals are called “open shell” species, because the orbital holding the single electron could hold another electron. If another electron were to be placed in this orbital, it must have a spin opposed in direction to the electron already there. This gives two electrons with opposed, or “paired,” spins, as shown in the representation . The spins of the two electrons must be different because the Pauli exclusion principle states that each electron in a molecule must have at least one quantum number (of the four total) that differs from those of all the other electrons. These quantum numbers can be thought of as the “address” of the electrons in a molecule. If the two electrons are in the same orbital, then the three quantum numbers that specify the location in space of the electrons are the same; therefore, to satisfy the Pauli principle, the spin quantum numbers of the two electrons must differ. One must be polarized “up” and one must be “down.” Incidently, it is this small energy difference that is utilized in electron spin resonance (ESR), a spectrographic technique that measures only those species with an odd electron. [ESR can also be referred to as electron paramagnetic resonance (EPR).] When the molecule is put in a magnetic field (so the electrons have an external reference to allow them to distinguish up from down, i.e., with or against the field), and the energy is scanned, species with pairs of electrons undergo transitions and no net energy is absorbed. However, radicals have one odd electron that “flips” its orientation from being with to against the field; the absorption of the small energy required by this electron flip is detected in ESR.

This business of spin can produce the unusual situation of a diradical, a species with two unpaired electrons that have the same spin. Because the two electrons have the same spin, they cannot go into the same orbital (Pauli principle) and are forced to stay apart. (Of course electrostatics also tells us that the two electrons would repel each other.) Dioxygen has two electrons with equal energy that occupy similarly shaped orbitals with identical energies (called “degenerate”), one on each oxygen atom. Hund’s rule states that when two electrons fill degenerate orbitals, they have the same spin, which forces them to stay apart. Thus, ground-state dioxygen (O2) is a diradical. If this ground-state “triplet” dioxygen is excited by 23 kcal/mol, it can pair the spins of these two electrons and put them both in the same orbital, forming the reactive species called singlet oxygen. (Singlet and triplet are terms that describe the number of equivalent spectroscopic states the species has.) The conversion of ground-state dioxygen (O2) to the excited singlet state (represented as 1O2) is shown in reaction (4).

O—O⋅triplet→O=Osinglet

III Radical Lifetimes

The biological lifetimes of a number of types of radicals can be approximated by calculations in which their rate constants for reaction with their principal targets are combined with the estimated concentration of the targets in the vicinity where the radical is formed (Pryor, 1986). These calculated lifetimes are shown in Table I. Note that radical lifetimes vary from extremely short to infinitely long. The hydroxyl radical is so short-lived that it can only diffuse about 50 molecular diameters before it reacts. Thus, it is extremely reactive and can pull off a hydrogen atom from even the least likely molecules. However, this very reactivity makes the hydroxyl radical quite unselective in the damage it produces. For example, in an attack on the protein α-1-proteinase inhibitor, hydrogen peroxide specifically oxidizes a methionine residue and produces an inactive protein. In contrast, hydroxyl radicals appear to do a random,...