Oxygen
Page 17
Oxygen Poisoning and X-Irradiation: A Mechanism in Common • 117
conjoin in blissful chemical union; or when the free radical product is so feebly reactive that the chain reactions fizzle out, like handbag thieves overcome with remorse. Some well-known antioxidants, such as vitamin C and vitamin E, act in this way. Although the products of their reactions with a free radical are themselves free radicals, they are so poorly reactive that the chain reactions squib out before too much damage is done.
If radiation strips a second electron from water, the next fleeting intermediate is hydrogen peroxide (H2O2) — whose bleaching properties gave its name to the peroxide blonde. Bleaching is caused by the oxidation of organic pigments as the hydrogen peroxide strips electrons from them. The oxidizing properties of hydrogen peroxide can kill bacteria, and are in part responsible for the mildly antiseptic properties of honey, which has been used to treat wounds since ancient times. Most industrial uses of hydrogen peroxide also draw on its power as an oxidizing agent.
For example, hydrogen peroxide is used to oxidize pollutants in water and industrial waste, to bleach textiles and paper products, and to process foods, minerals, petrochemicals and detergents.
Despite its widespread use as an oxidizing agent, hydrogen peroxide is unusual in that it lies chemically exactly half way between oxygen and water. This gives the molecule something of a split personality. Like a would-be reformed mugger, whose instinct is pitted against his judgement, it can go either way in its reactions (losing or gaining electrons) depending on the chemical company it keeps. It can even go both ways at once, when reacting with another hydrogen peroxide molecule. In this case, one of the molecules gains two electrons to become water, while the other loses two electrons to become oxygen. The decomposition of hydrogen peroxide in this way is partly responsible for the generation of oxygen from water by radiation:
2H
씮
2O2
2H2O + O2
A far more dangerous and significant reaction, however, takes place in the presence of iron, which can pass electrons one at a time to hydrogen peroxide to generate hydroxyl radicals. If dissolved iron is present, hydrogen peroxide is a real hazard. Organisms go to great lengths to avoid contamination with dissolved iron. The reaction between hydrogen peroxide and iron is called the Fenton reaction, after the Cambridge chemist Henry Fenton, who first discovered it in 1894:
H2O2 + Fe2+ 씮 OH– + •OH + Fe3+
118 • TREACHERY IN THE AIR
He later showed that the reaction could damage almost any organic molecule. Thus, the main reason that hydrogen peroxide is toxic is because it produces hydroxyl radicals in the presence of dissolved iron.
Ironically, the greatest danger lies in its slow reactivity in the absence of iron. It has time to diffuse throughout the cell. Hydrogen peroxide may diffuse into the cell nucleus, for example, and there mix with the DNA, before it encounters iron, which transforms it into a brutish hydroxyl radical.2 The insidious infiltration of hydrogen peroxide means that it is more dangerous than the hydroxyl radicals produced outside the nucleus.
Some proteins, such as haemoglobin, also contain iron. If they happen to run into hydrogen peroxide they can be mutilated on the spot. Hydrogen peroxide is like a gangland thug. Normally quiet, posing little danger to casual passers-by, it turns violent on meeting a rival gang member.
Damage to proteins containing embedded iron can be as swift and specific as a kneecapping operation.
We have now met two out of the three intermediates between water and oxygen. The first, the hydroxyl radical, is one of the most reactive substances known. It reacts with all biological molecules within billiseconds, initiating chain reactions that spread damage and devastation.
The second of the intermediates, hydrogen peroxide, is much less reactive, almost inert, until it meets iron (regardless of whether the iron is in solution or embedded in a protein). Hydrogen peroxide reacts quickly with iron to generate hydroxyl radicals, taking us back to square one.
What, then, of the third of our intermediates, the superoxide radical (O •–
2
)? Like hydrogen peroxide, the superoxide radical is not terribly reactive.3 However, it too has an affinity for iron, dissolving it from proteins and storage depots. To understand why this is harmful, we need to think again about the Fenton reaction.
The Fenton reaction is dangerous because it produces hydroxyl radicals, but it grinds to a halt when all the accessible iron is used up. Any chemical that regenerates dissolved iron is capable of re-starting the reaction. Because the superoxide radical is one electron away from molecular 2 Iron is normally tightly bound within proteins inside the cell, but some pathological conditions may bring about its release, allowing iron to be found even in the nucleus. Here, the positively charged iron binds to negatively charged DNA, exacerbating the damage to DNA.
3 Despite the heroic-sounding name, superoxide radicals are not very reactive with lipids, proteins or DNA. Superoxide does react vigorously with other radicals such as nitric oxide, however, and this may cause cellular damage. It is also reactive in slightly acidic conditions, which occur in the vicinity of cell membranes, so superoxide may damage membranes directly.
Oxygen Poisoning and X-Irradiation: A Mechanism in Common • 119
oxygen, it is more likely to lose that electron to form oxygen than it is to gain three electrons to form water. Only a few molecules are able to accept a single electron, however. One of the best places for the superoxide radical to jettison its spare electron is iron. This converts iron back into the form that can participate in the Fenton reaction:
O •–
2
+ Fe3+ 씮 O2 + Fe2+
In summary, then, the three intermediates between water and oxygen operate as an insidious catalytic system that damages biological molecules in the presence of iron. Superoxide radicals release iron from storage depots and convert it into the soluble form. Hydrogen peroxide reacts with soluble iron to generate hydroxyl radicals. Hydroxyl radicals attack all proteins, lipids and DNA indiscriminately, initiating destructive free-radical chain reactions that spread damage and destruction.
These same intermediates are also formed from the oxygen that we breathe. The parallel between radiation damage and oxygen toxicity was first described by Rebeca Gerschman in the early 1950s, while she was working at the University of Rochester in New York State — the centre selected to study the biological effects of radiation for the Manhattan Project. In a 1953 seminar, she caught the imagination of a young doctoral student with a background in muscle physiology named Daniel Gilbert, and together the pair pioneered the theory that oxygen free radicals are responsible for the lethal damage caused by both oxygen poisoning and radiation. Their findings were published in a seminal 1954
paper in Science, with the splendidly unambiguous title Oxygen Poisoning and X-irradiation: A Mechanism in Common, which I’ve taken as the subtitle to this chapter. Since the 1950s, research has confirmed that radiation damage and oxygen toxicity amount to much the same thing.
Oxygen is a paradox. From a theoretical point of view, it ought to be easier to add electrons to oxygen than it is to remove them from water.
Water is chemically stable. Taking electrons from water requires a large input of energy, which can be provided by ionizing radiation, ultraviolet rays, or by sunlight in the case of photosynthesis. Oxygen, on the other hand, releases energy when it reacts — a sure sign of favourable energetics. Burning is the reaction of oxygen with carbon compounds; and the
120 • TREACHERY IN THE AIR
massive amount of heat released suggests that burning should proceed almost spontaneously. In terms of energetics, it does not matter if fuel is burnt rapidly, as in combustion, or slowly, as in respiration. Regardless of whether we metabolize it or burn it, 125 grams [4 oz] of sugar (the amount required to make a sponge cake) will produce 1790 kilojoules [428 kilo-calories] of energy: enough to boil 3 litres [6.3 US pints] of wa
ter or light a 100-watt bulb for 5 hours.
With such favourable energetics, and with oxygen all around us, the fact that living things do not burst spontaneously into flame betrays an odd reluctance on the part of oxygen to react. The reason for its reticence is buried within the bonds of the oxygen molecule itself. Although slightly abstruse, the chemistry of oxygen explains not only why oxygen free radicals are formed inside us all the time, but also why we do not spontaneously combust. We will therefore consider it briefly.
One of the first signs that oxygen is a little odd was reported in 1891
by the great Scottish chemist Sir James Dewar, who found that oxygen is magnetic. His discovery came after a competitive and acrimonious race to liquefy oxygen, in which the Frenchman Louis Cailletet narrowly defeated his Swiss rival Raoul Pictet, when he succeeded in producing a few droplets of liquid oxygen just before Christmas in 1877. The following year, Dewar liquefied oxygen in a demonstration before the hawkish audience of the Royal Institution, in one of their celebrated Friday Evening Discourses. Dewar was a star turn on these occasions, which were tradi-tionally held in an auditorium known, to the terror of many invited to perform there, as ‘that semi-circular fountain of eloquence’. But Dewar was far more than a gifted performer: he was also one of the most brilliant practical scientists of his day. By the mid 1880s, Dewar had improved his methods and was able to produce liquid oxygen in large enough quantities to study its properties in detail. He soon found that liquid oxygen (and indeed ozone, O3) was attracted to the poles of a magnet. In 1891, he demonstrated his findings with characteristic flamboyance at one of the Discourses, using a strong magnet and his newly devised vacuum flask, still known in laboratories across the world as the Dewar flask. His classic demonstration is repeated today in many university foundation courses (and video demonstrations can be found on the Internet). Liquid oxygen is poured from a Dewar flask between the poles of a powerful magnet. The cascading liquid halts in mid air and sticks to the magnet, forming a plug that hangs majestically between the magnetic poles, until evaporating away.
Oxygen Poisoning and X-Irradiation: A Mechanism in Common • 121
What is going on? In 1925, Robert Mulliken finally explained why oxygen is magnetic, using recently developed quantum theory. Magnetism results from the spin of unpaired electrons, and Mulliken showed that molecular oxygen normally has two unpaired electrons.4 These electrons dominate the chemistry of oxygen and make it hard for the gas to receive a bonding pair of electrons; hence the reluctance of oxygen to react by forming new chemical bonds (Figure 8). There are only two ways out of this chemical cul-de-sac. First, oxygen can absorb energy from other molecules that have been excited by heat or light, and this can cause one of its unpaired electrons to flip its spin. Some excited pigment molecules have this effect, and are being put to medical purposes, as in photodynamic therapy in which a pigment is activated by light to destroy a tumour or other pathological tissue. Flipping the spin of an electron leaves one electron pair and one vacant bonding orbital, and so frees up oxygen to react. It is said to remove the ‘spin-restriction’. This form of oxygen is called singlet oxygen. Unlike its spin-restricted sister, singlet oxygen is fast to react with organic molecules. If, by a chemical quirk of fate, singlet oxygen was the only form that existed, we could never have accumulated oxygen in the atmosphere, or crawled out of the oceans.
The second way of coaxing oxygen to react is to feed it with electrons one at a time, so that each of the unpaired electrons receives a suitable partner independently. Iron can do this, as it has its own unpaired electrons (which makes it magnetic too). Iron loses these electrons without becoming unstable because it has several different ‘oxidation states’, all of which are energetically stable under relatively normal conditions. (This is partly because the iron atom is large, and the electrons furthest from the nucleus are loosely bound to the atom.) The ability of iron to feed electrons one at a time explains its affinity for oxygen, and the tendency 4 Rotating electrical charges generate magnetic fields. This applies to electrical current in a coil of wire or to a single spinning electron. In theory, all chemical (covalent) bonds are formed from pairs of electrons. The electrons within a bond usually spin in opposite direc-tions, and their spin is said to be antiparallel. The opposition of spins in typical bonds cancels out, leaving the molecule as a whole with no net spin. Most compounds are therefore not magnetic. Atoms or molecules that are magnetic, such as iron and oxygen, must have at least one unpaired electron; and this, as we have seen, is not usually a stable electronic arrangement. In the case of oxygen, though, quantum mechanical considerations mean that unpaired electrons are actually more stable than the regular double bond structure that most of us were taught at school (see Figure 8). On the basis of bond strengths and magnetism, oxygen has three bonds rather than two: one bond has two electrons, while the other two bonds have three electrons, one of which is unpaired in each bond.
122 • TREACHERY IN THE AIR
of iron minerals and cars to rust. It also explains our own need to lock iron away in molecular safe-houses in the body. Other metals that exist in two or more stable oxidation states, such as copper, can feed electrons to oxygen just as efficiently, and are equally dangerous unless well caged.
Life is not free from the restrictions imposed by the odd chemistry of oxygen, and we too must supply electrons one at a time in order to tap into its reactivity. Cells contrive to break down the oxidation of food into a series of tiny steps, each of which releases a manageable quantity of energy that can be stored in a chemical form as ATP (see Chapter 3).
Unfortunately, at each of these steps there is the risk of single electrons Ground-state
Singlet (excited state)
molecular oxygen
molecular oxygen
O
O
*
2
O
O
O2
O
Figure 8: Electron orbitals for (a) ground-state molecular oxygen (normal O2) and (b) singlet oxygen (the excited form of molecular oxygen). Each cross represents a pair of electrons, while a single diagonal line represents a single electron. The circles represent available electron orbitals; empty circles represent empty orbitals. According to Hund’s rule, electron orbitals of the same energy (shown on the same level) must be filled one at a time, before pairing can occur. This is why atomic oxygen (shown to the side of the box in each case) contains two unpaired electrons. When the electron orbitals are merged in molecular oxygen, the same rule applies, leaving ground-state oxygen with two unpaired electrons, each with parallel spin. Oxygen can therefore only receive single electrons with anti-parallel spin to complete the electron pairings. In singlet oxygen, one of the parallel-spin electrons has flipped spin, enabling a pair to form, but vacating a lower energy orbital in violation of Hund’s rule. This is why singlet oxygen is so reactive.
Oxygen Poisoning and X-Irradiation: A Mechanism in Common • 123
escaping from their shackles and joining with oxygen to form superoxide radicals. The continual production of superoxide radicals by cells means that, despite the emotive associations of radiation sickness, breathing oxygen carries a qualitatively similar risk.
Estimates suggest that, at rest, about 1 or 2 per cent of the total oxygen consumed by cells escapes as superoxide radicals, while during vigorous exercise this total might rise to as much as 10 per cent. Lest these figures sound trivial, we should remember that we consume a large volume of oxygen with each breath. An average adult, weighing 70 kilograms [154 pounds], gets through nearly a quarter of a litre of oxygen every minute. If only 1 per cent of this leaks away to form superoxide radicals, we would still produce 1.7 kilograms of superoxide each year.
From superoxide we can go on to produce hydrogen peroxide and hydroxyl radicals by way of the reactions discussed earlier.
We can produce hydrogen peroxide and hydroxyl radicals, but do we really? Our bodies have evolved efficient mechanisms for
eliminating both superoxide radicals and hydrogen peroxide before they can react with iron to produce hydroxyl radicals (we will look at these mechanisms in more detail in Chapter 10). Is there any way of working out how many hydroxyl radicals are actually formed in the body despite these mechanisms?
There are two possible approaches. First, from a theoretical point of view, we can calculate the rate of production of hydroxyl radicals on the basis of an estimated steady-state concentration of hydrogen peroxide and iron, and the know n reaction kinetics. In the human body, it is likely that both iron and hydrogen peroxide are present at a steady-state concentration of about a millionth of a gram per kilogram body weight. If so, we would generate less than a million millionth of a gram of hydroxyl radicals per second per kilogram. Such small numbers are virtually incomprehensible. However, if we convert this figure from the number of grams to the number of molecules produced each second, using Avogadro’s number, then we have a far more intelligible number. According to this calculation, we should produce around 50 hydroxyl radicals in each cell every second.5 In a full day, each cell would generate 4 million hydroxyl 5 Avogadro’s number states that there are 6.023 ⫻ 1023 molecules in one mole of any substance. One mole of a substance is the molecular weight in grams. The molecular weight of a hydroxyl radical is 17, so one mole is 17 grams. This means that 17 grams of hydroxyl radicals would contain 6.023 ⫻ 1023 hydroxyl radicals. The volume of an average mammalian cell is approximately 10–9–10–8 ml. The calculation is adapted from Halliwell and Gutteridge, Free Radicals in Biology and Medicine (see Further Reading).
124 • TREACHERY IN THE AIR
radicals! Many of these would be neutralized in one way or another, and damaged proteins or DNA replaced, but over a lifetime, in a body composed of 15 million million cells, it adds up to a very great deal of wear and tear: more than enough to underpin a process such as ageing.
This is all very well, but somewhat theoretical. If so much damage is taking place, we ought to be able to measure it. This brings us to the second way of estimating the production of hydroxyl radicals: analysing the damage they cause. One possible marker for their effects, which was first proposed in the late 1980s by Bruce Ames and his team at Berkeley, is the rate of excretion of oxidized DNA building blocks in the urine. There are some obstacles to this approach — DNA is being constantly tinkered with by various enzymes in the normal processes of DNA replication and repair, and only some of the types of oxidized fragments produced as a result are diagnostic of attack by hydroxyl radicals. Other types are sometimes produced by hydroxyl radicals but can also be generated normally.