Antioxidant Vitamins and Prevention of Lung Disease DANIEL B. MENZEL Department of Community and Environmental Medicine FRF North Campus University of California Iruine, California 9271 7-1825

INTRODUCTION Air pollution is a ubiquitous curse of all civilizations on the earth. We depend on fossil fuels or vegetable matter to supply our needs for energy. In North and South America, Europe, and Japan, fossil fuels supply, by far, most of the energy we use. Transportation vehicles are the largest source of outdoor air pollution, and cars contribute more pollution than any other form of transportation. Despite efforts to reduce the emissions of nitrogen oxides (NO,) by cars and trucks, levels of NO, in most major cities exceed a safe level.' NO, produced by cars also leads to a number of other toxic air pollutants. NO2 reacts with atmospheric oxygen in the presence of unburned hydrocarbons from fuel and sunlight to produce ozone (0,).0, is even more toxic than NO,. Power production and space heating contribute sulfur oxides to the air. Of these, sulfur dioxide and sulfuric acid are important to human health. Sulfuric and nitric acid combine with water vapor to produce acid fogs. When NO, and 0, are also present, smog occurs. Compounding the effects of outdoor air pollution is the problem of indoor air pollution. Tobacco smoke is an important source of indoor air pollution that resembles outdoor air pollution in many regards. Tobacco smoke contains nitrogen oxides, heavy metals, polycyclic hydrocarbons, and a variety of chemicals similar to the combustion of fossil fuels. Gas-fired appliances contribute to indoor pollution of nitrogen oxides. Building materials and human habitation itself also contribute to the carbon monoxide, carbon dioxide, aldehyde, and ammonia levels. It is this complex mixture with which our lungs must now deal. In this paper, some concepts of how antioxidants are important to the defense of the lung against air pollution toxicity are defined. The evidence for a protective role of antioxidants is best illustrated for 0, and NOz, and secondarily for tobacco smoke. I will consequently confine my remarks to these areas.

MANY PEOPLE ARE EXPOSED TO TOXIC LEVELS OF AIR POLLUTION In a recent review of the health effects of air pollution, van Bree et a1.l pointed out that more than 100 million people in the United States as well as the majority of the population in the Netherlands were exposed to levels of O3 in excess of the maximum safe level currently set by the U.S. Environmental Protection Agency and Dutch authorities. Ozone is of particular concern because the available evidence for its adverse human health effects on the respiratory tract is reported 141

ANNALS NEW YORK ACADEMY OF SCIENCES

142

to occur within the current range of exposure levels of large segments of the population. Worldwide reporting of air pollution, particularly O3and NO2, is unreliable for Eastern Europe, Russia, the Middle East, China, and South America. It is very likely that the troposphere over all of the populated areas of the globe is polluted with these two toxic chemicals. As tropospheric O3is increasing, the stratospheric 0, appears to be decreasing. The decrease in stratospheric O3 may result in an increase in tropospheric O3 because tropospheric O3 is produced by ultraviolet radiation reaching the troposphere and catalyzing the reaction of NO2with oxygen. O3is an effective “green house” gas that exacerbates the problem of air pollution by increasing temperatures and decreasing mixing on a global scale.’ Air pollution is truly a global problem that has major health consequences and for which there is no solution in sight. THE TOXICITY OF OZONE AND NITROGEN DIOXIDE Animal studies have shown that O3and NO2 are the two most toxic of the air pollutants.2 In some areas of the world the combustion of high sulfur coal has lead

TABLE 1. Time Course of NO,- or O1-Induced Lung Disease in Rodents Time after Exposure Effect 1-10 milliseconds 7-10 hours 10- 18 hours 3 days 6 months

12-18 months

Lipid peroxidation Increased lung permeability to proteins Inflammation and increased susceptibility to infection; cell death Maximum increase in antioxidant enzymes; cell replication to replace dead and/or dying cells Increase in number and shape of type-2 cell and loss of type1 cells; thickening of alveolar septa; changes in mucus glands; denuding of cilia Bronchitis (NO?) or emphysema (0,)

to an elevation of sulfur dioxide content as well. Eventually, NO, and SO, are converted to nitric and sulfuric acids, which form acid smogs. Acid smogs have devastating effects on the mortality and morbidity of patients with existing lung disease. In experimental animals exposed to controlled levels of either NO2, 03,or SO2 for almost their lifetimes (18 months out of 24 to 26 months) major lung disease results. Sulfur dioxide exposure at high concentrations results in asthma in rats with even a single exposure.2 Nitrogen dioxide exposure over a lifetime in rats and mice results in bronchitis, whereas lifetime O3 exposure produces e m p h y ~ e m a . ~ The events leading to such major lung diseases are complex, starting first with the appearance of inflammatory cells and then progressing slowly to remodeling of the architecture of the lung (TABLE1). The production of pulmonary disease by air pollutants, especially O3and NO2, is a slow process. It is difficult to relate “rodent months” to “human years” in any more than qualitative terms. The time required to produce lung disease in rodents is probably close to mid-life, or in equivalent

MENZEL: ANTIOXIDANTS AND LUNG DISEASE

143

human terms, close to 40-50 years. Chronic lung disease in humans is also mostly diagnosed at mid-life, and the disease continues on through to death. As with human lung disease, once the disease process of either bronchitis or emphysema is evident on an anatomical basis, recovery to normal lung architecture even in the absence of continued exposure to either NOz or 0, does not occur. Once changes in the lower respiratory tract occur they are permanent. The toxicity of 0, and NO, in experimental animals can be linked directly to the toxic effects observed in humans through such parameters as increased lung permeability or inflammation, which occurs in a dose-response manner in both humans and experimental animal^.^.' A broad range of studies with human volunteers has shown that both NO, and O3 produce transient reductions in respiratory function as well as the recruitment of inflammatory cells to the lung.* Patients with preexisting asthma are particularly sensitive and probably represent the most sensitive population to NOZ.’ Although epidemiological studies have suggested lung injury from ambient 0, and NO2, the statistical power of these studies has not been sufficient to support direct associations between ambient levels and lung d i ~ e a s e Short-term .~ correlations between air pollution exposure and a decrement in respiratory function has, however, been found with children living in summer camps or in areas of high air p ~ l l u t i o n . These ’ ~ ~ children are important subjects because their lungs are relatively free of disease. These studies clearly demonstrate that pulmonary injury occurs in children on exposure to current levels of air pollution. Trends show decreased lung development in children living year round in polluted air.6 Whereas most studies have focused on outdoor sources of NO,, gas-fired cooking stoves and space heaters are also sources of indoor NOz pollution.’ One of the most sensitive indices of exposure to either NO, or 0, is the inhibition of the bacterial and viral defense mechanisms of the lung.’When rodents are exposed to NO2 or 0, and then to an infectious agent, either bacteria or viruses, the mortality from a standard infection is increased in proportion to the concentration and duration of the air pollutant exposure. This assay, known as the infectivity assay, has shown that rodents are more susceptible to infection when exposed to levels of NO2or 0, below the current standards. Recent studies of the respiratory infection rate in children living in homes with gas cooking stoves have shown a greater incidence of upper respiratory infectiom8 These results are controversial inasmuch as tobacco smoking in the home also increases the upper respiratory infection rate, especially in children around two years of The controversy may be resolved when one considers that tobacco smoke contains large amounts of nitrogen oxides, which are converted to NOz.

CHEMICAL REACTIONS BETWEEN NITROGEN DIOXIDE AND OZONE, AND LUNG BIOMOLECULES Both NO, and O3 are strong oxidizing agents. Ozone reacts with the ethylene groups of unsaturated fatty acids to form an initial ozonide (FIG. 1). The initial ozonide can rearrange by way of the Criegee mechanism to a secondary ozonide. During the rearrangement, the initial ozonide can react with water to form peroxyl radical and peroxides. The secondary ozonide can also decompose into a peroxide and an aldehyde. Peroxides decompose thermally into peroxyl and hydroxyl free radicals. The initial ozonide can also decompose by way of the /3 scission to a

ANNALS NEW YORK ACADEMY OF SCIENCES

Criegec Ozorride

I

R

x

ll

FIGURE 1. Reaction of ozone with unsaturated fatty acids of cell membranes. O3 adds directly on the ethylene groups of unsaturated fatty acids esterified to phospholipids. In the relatively anhydrous environment of cell membranes and the alveolar lung lining fluid, the initial omnide rearranges to the Criegee ozonide. which can decompose to a hydroperoxide and an aldehyde. Hydroperoxides can imitate lipid peroxidation, whereas aldehydes may be oxidant. stress inflammatory signal molecules.

peroxide and an aldehyde (FIG.2). Trace amounts of iron catalyze the p scission of the oxygen-to-oxygen bond. Trace amounts of iron are always present in tissues. The decomposition of the initial ozonide by either mechanism shown in FIGURES 1 or 2 leads to fatty acid peroxyl radicals. The peroxyl radical is known to be the intermediate in propagations of lipid peroxidation. Vitamin E as a-tocopherol preferentially reacts with fatty acid peroxyl radicals to terminate or block lipid peroxidation and cell death. 0, reacts directly with vitamin E, reducing tissue stores o f the vitamin and increasing the susceptibility of the tissue to lipid peroxidation. The mechanism by which cells die is not clear but may involve the loss of membrane integrity, regulation of ion fluxes, or of intracellular calcium ion. Nitrogen dioxide abstracts a methyleneic hydrogen-forming nitrous acid and a carbon free radical (FIG.3). The carbon free radical reacts rapidly with molecular oxygen forming a peroxyl radical. The latter can initiate a chain reaction, exactly the same as in other forms of fatty acid peroxidation, to form fatty acid peroxides. Vitamin E as a-tocopherol terminates the peroxyl radical-supported chain reaction

MENZEL: ANTIOXIDANTS AND LUNG DISEASE

145

by reacting with the peroxyl radical to form relatively stable hydroperoxides and the vitamin E radical. NO2reacts directly with vitamin C, ascorbic acid, oxidizing ascorbic acid. This reaction is faster than the abstraction of a hydrogen from unsaturated fatty acids. The oxidation of ascorbic acid by NO, may be responsible for the low levels of vitamin C in smokers and the apparently increased requirement of smokers for vitamin C. Although not demonstrated in uiuo, vitamin C may reduce the tocopherol free radical back to tocopherol. Vitamins E and C may work together in this complex relationship to prevent oxidative damage to cell membranes. Oxidation of unsaturated fatty acids by either O3or NOz produces fatty aldehydes, which may be unsaturated depending on the original unsaturated fatty acid oxidized in the membrane. Fatty aldehydes have intriguing properties as “inflammatory signals.” The constant elaboration of such an inflammatory signal could be the reason why animals and human subjects exposed to NO, and O3have increased inflammatory cells in their lungs. These reactions have been demonstrated in a variety of systems and cells. Several laboratories have duplicated our original work to show that vitamin E is particularly important in preventing 0,toxicity, whereas ascorbic acid or vitamin C is more important in preventing NO2 toxicity (see refs. 2 and 9.) Inasmuch as tobacco smoke contains NOZ, it is not surprising that tobacco smoke also initiates oxidative damage in isolated human plasma lipoproteins. lo

FIGURE 2. Ozone reaction with unsaturated fatty acids and decomposition by p-scission. The initial ozonide can decompose by scission of the 0-0 bond to a peroxyl radical and an aldehyde. The peroxyl radical can initiate lipid peroxidation, which is terminated by vitamin E. Oxidant stress inflammatory signal aldehydes are also released.

O N 0 0’

I

t

H

H

1 t

0 R

A

H

R H

B scission

146

ANNALS NEW YORK ACADEMY OF SCIENCES

The organic components of cigarette smoke may also participate in this reaction or affect biological defense systems.

PEROXIDATION AS THE MECHANISM OF ACTION OF OXIDIZING AIR POLLUTANTS AND TOBACCO SMOKE

When and mice13 are made deficient in vitamin E, the cumulative mortality, as expressed as the time to death of 50% of the continuously exposed population, or LTSO, was decreased by 10-20-fold over a similar vitamin E supplemented population. Other studies have subsequently confirmed this original

1

I,IPIIl PEROXIDATION

ROOF1 +

KCHO

FIGURE 3. Reaction of nitrogen dioxide with unsaturated fatty acids in cell membranes. Nitrogen dioxide, a radical, abstracts a methylenic hydrogen from unsaturated fatty acids and produces a peroxyl radical. Vitamin E as a-tocopherol (TocOH) terminates lipid peroxidation by forming the relatively stable TocO. radical which may be reduced by vitamin C.

datal4.ls(see ref. 2 for a review). Direct evidence of peroxidative damage to the lungs of rats or mice exposed to NO2 and 0, has not been reported because of the technical difficulties of isolation of the original reaction products. Indirect evidence has been reported from increased levels of diene conjugation (as in the mechanism in FIG. 3 ) or 2-thiobarbituric acid (TBA) -reacting substances (see ref. 16 for a review of the complex nature of lipid peroxidation products from NO2 and 03). TBA-reactive substances are mixtures of oxidative products, including cyclic peroxides.” Recently, Dr. Finlayson-Pitts (personal communication, 1992) reported the presence of secondary ozonides in lung lavage fluid by use of infrared spectra. Because rats and mice biosynthesize vitamin C, direct experimentation on the role of vitamin C in either NOz or O3toxicity has not been possible in these species. Guinea pigs, as well as humans, do not biosynthesize ascorbic acid. Guinea pigs

MENZEL: ANTIOXIDANTS AND LUNG DISEASE

147

deficient in vitamin C are more susceptible to O3 and NO,. Vitamin C provides protection in proportion to the dietary content.z.18 Using a model system of the cell membrane consisting of lipid bilayers of phosphatidylcholine, Shoaf et al.17-19 demonstrated the rapid oxidation of the unsaturated fatty acids esterified to the phosphatidylcholine. The oxidation was directly proportional to the NO, concentration. Incorporation of vitamin E into the lipid bilayer prior to NOz exposure prevented fatty acid oxidation until the vitamin E was exhausted. Ascorbic acid sealed inside the lipid bilayer liposomes before exposure to NO, inhibited the oxidation of the unsaturated fatty acids. The ascorbic acid within the liposomes was rapidly oxidized. Only limited amounts of vitamin E could be incorporated into the lipid that would form a stable liposome, whereas the amount of ascorbate sealed within the liposomes was essentially limited by the solubility of ascorbic acid in water. Very similar effects occur in type-2 lung cells treated in uitro with a-tocopherol succinate or ascorbic acid.z0 Ascorbic acid prevented the formation of reactive oxygen species on treatment with either 0, or organic peroxides. Preincubation with increasing concentrations of a-tocopherol succinate, on the other hand, led to a concentration of a-tocopherol succinate beyond which additional protective effects were not detected.

WHAT IS THE DOSE-RESPONSE RELATIONSHIP BETWEEN ANTIOXIDANT VITAMIN PROTECTION AND AIR POLLUTION PEROXIDATION? Unfortunately, little quantitative data is available on the dose of antioxidant vitamins and their protective effects against either NO2 or 0,. Generally, only a single dose of one vitamin or a fixed combination of the two has been used. In rats and mice, increasing the dietary content of vitamin E beyond about 100 mg/kg of diet fails to provide additional protection (see ref. 2 for a more detailed analysis of this problem). Only one study of the dose-response between vitamin E intake and 0, effects in humans has been In this very limited study, volunteers ate a normal diet and were supplemented with increasing daily supplements of vitamin E as a tablet form of d,I-a-tocopherol acetate. Red cells were isolated from the volunteers after one week of no supplementation, or after one week of supplementation with 100 IU/day, then 200 mg/day for up to two weeks. Supplementation with 200 IU/day did not significantly increase the protection afforded their red cells to oxidative stress damage by fatty acid ozonides beyond supplementation with 100 IU/day. A clear protective effect was, however, seen with supplementation with either level of vitamin E beyond that afforded by a normal diet alone. Attempts have been made to provide super supplementation with vitamin E and selenium antioxidant^.^, In these studies, rats were fed graded amounts of vitamin E, sulfur amino acids, and selenium and then exposed to NO2 or O3 for seven days. Toxicity was measured by mortality and biochemically by the activity of antioxidant enzymes in the lung. Mortality occurred in the highest dietary supplementation group, and complete protection against NOz or 0, was not accomplished. In unexposed rats, the level of antioxidant enzymes remained constant when fed increasing levels of antioxidants. When unsupplemented o r supplemented rats were exposed to either NO, or O,, the antioxidant enzymes of

ANNALS NEW YORK ACADEMY OF SClENCES

148

the glutathione cycle shown in TABLE2 were induced. Theoretically, glutathione peroxidase acts to decompose lipid peroxides to their corresponding alcohols, preventing cellular damage. By providing vitamin E in the diet, less oxidation should occur on exposure to 0, or NO,. The failure to induce antioxidant enzymes was taken as prevention of oxidation by NOz or 0,. Using this very indirect measure of oxidation, saturation occurred in the protection afforded by vitamin E and other antioxidants, so that super amounts did not confer absolute protection. These studies are of limited value because the shape of the doseresponse curve for 0, and NO, toxicity is poorly described. If the dose-response curve follows a sigmoidal shape, an infinitely large dose of antioxidants would be needed to provide complete protection. More importantly, these studies were only for a short-level acute toxic dose of NOz or 03.As shown in TABLE1, the effects of O3 and NO, are chronic and extend over the lifetime of the exposed individual.

TABLE 2.

Antioxidant Enzymes Induced by O? or NO, Exposure in Rats

Glucose 6-Phosphate GSH

GSSG

+ NADP+

+ 6-Phosphogluconate Glucose 6-Phosphate Dehydrogenase

+

NADPH

+ ROOHGlutathione Peroxidase> ROH + GSSG

+ NADPH Glutathione Reductase 2 GSH + NADP'

It is clear that the current recommended daily allowance based on other symptoms of vitamin E deficiency is inadequate for protection against current levels of air pollution found globally in the air over our cities.

GLUTATHIONE S-TRANSFERASES AS AN IMPORTANT CLASS OF ANTIOXIDANT ENZYMES The experiments of Fletcher and Tappelz3did not take into account the multifunctional activity of glutathione S-transferases. Glutathione peroxidase, as detected in these experiments, might have been glutathione S-transferase activity. Glutathione S-transferase catalyzes the reduction of peroxides as well as the . ~ ~ glutathione S-transferases transfer of glutathione to reactive n ~ c l e o p h i l e s The are a broad class of multifunctional enzymes that exist as either homo- or heterodirners. Three isoenzymic forms predominate as alpha, mu, and pi in humans and rats. Because of their dimeric form and multiple subunits, the literature has become confused over the activity of the glutathione S-transferases. Vos i>ta1.2s.2h found that the glutathione S-transferases decomposed the secondary ozonides of fatty acids very efficiently. They also found that the mu form was the most effective catalyst. This observation is particularly interesting because the glutathione S-transferase mu subunit gene is deleted in about 40-50% of the Caucasian population.*' The Iack of the mu glutathione S-transferase has been associated with the incidence of lung cancer among cigarette smokers.28The alpha and pi glutathione S-transferases are absent in a variety of cancers.29

MENZEL: ANTIOXIDANTS AND LUNG DISEASE

149

It is fair to say that the question of the role of induction of glutathione S-transferase and air pollutant toxicity is still open. The induction of glutathione S-transferase might be a protective response. Despite the intake of large amounts of vitamin E, exposure to NO, or O3 could provide adequate signals to induce glutathione S-transferase as a protective system. The in uitro data clearly shows that human lung cells are protected against peroxidation by 0, by both vitamin E and ascorbic acid. The experiments on graded amounts of vitamin E were not carried out for a lifetime, so that the real protective effects of vitamin E supplementation were not demonstrated. Menzel and Wolpert3' developed several statistical models of morbidity from ambient levels of 0, and vitamin E protection. These models are based on two general statistical models of toxicity both of which fit the cumulative 0, toxicity data in mice equally well. If these relations also apply to human populations and to a lifetime continuous exposure to O,, then supplementation with vitamin E at about 10 times the current RDA should afford a reduction in morbidity approximately equal to that for mice or a reduction in toxicity of about 10-fold for the worst case. In other words, the probability that death would occur solely from 0, exposure would be decreased by 10-fold, if a person breathed a constant lifetime concentration of the current ambient air quality standard for 0,. One must bear in mind that lung disease is not the leading cause of death in the United States and that the relative reduction in mortality must be adjusted for the mortality due to bronchitis and emphysema or perhaps broadly to chronic obstructive lung disease. Fortunately, although the human population is exposed to 0, levels at or above the current standard, the exposures are transient and periods of much lower levels of O3 usually occur between peak concentrations. Nonetheless these statistical models suggest that major reductions in morbidity from lung disease could occur by a reduction in ambient 0, levels and supplementation above the current RDA with vitamin E.

REGIONAL DISTRIBUTION OF OZONE AND NITROGEN DIOXIDE CORRELATES WITH THE CELLULAR TOXICITY IN THE LUNG Most animal experiments measuring the toxicity of 0, and NO, have been relatively short, with exposures of 7-14 days (see ref. 2 ) . A few more recent studies have tried to mimic the lifetime exposure pattern of air pollution in urban areas. These studies in rats and subhuman primates are remarkably similar in result and clearly demonstrate that lifetime exposure to these gases or to mixtures of the gases as might occur in urban areas results in chronic and irreversible changes in the morphology of the distal airways. The region most affected by O3 and to a large extent by NO2 is the region of the respiratory tract where the conducting airways form a transition with the central acinar region or transitional zone region. Miller et developed a mathematical model for the deposition of 0, and other reactive gases in the airways. These models predict the deposition of 0, and NO, in the transitional zone region on the basis of the chemical and physical properties of 0, and NOz and the anatomical properties of the lung. The transitional zone region is not particularly sensitive but receives 10-1,000 times the dose of other regions of the lung. 0, and NO2 damage occurs at all levels of the respiratory tract. Therefore, antioxidants do not necessary have to be targeted to a small part of the lung to be effective. Dietary supplementation will lead to a distribution through-

150

ANNALS NEW YORK ACADEMY OF SCIENCES

out the lung, including those regions receiving the greatest dose of 0, and

NO?. In cigarette smokers patterns of damage very similar to that of animals exposed to NO, occur. Again the high content of NO, in cigarette smoke suggests that NO, in cigarette smoke is responsible for this damage. Although regions of the lung may receive more damage from cigarette smoke, there are no data to suggest that cigarette smoke “targets” specific cells in the lung. Dietary supplementation will reach all of the sites in the lung affected by inhaled cigarette smoke.

DIRECT EVIDENCE FOR ANTIOXIDANTS AS PROTECTIVE FACTORS IN THE HUMAN LUNG EXPOSED TO AIR POLLUTION OR TOBACCO SMOKE Studies of the protective effects of vitamins E and C from cigarette smoke and air pollution in humans are scanty, but more studies are in progress. Most investigations of air pollution effects have relied on changes in respiratory function. It has been recognized that these measurements provide important but limited assessment of the toxicity of air pollutants and cigarette smoke to the lung. Originally, pulmonary function measurements were designed as clinical tests to detect and diagnose pulmonary disease. More recent studies have used biochemical techniques to evaluate samples collected by the use of flexible fiber optic bronchoscopes. Bronchoalveolar lavage (BAL) provides a rapid and safe means of sampling the lungs of adult volunteers. Saline solution is lavaged into one lung and then aspirated for collection, which is the BAL fluid. BAL samples the lung surfactant layer and free cells present at the respiratory bronchiole and alveolar region of the lung, the area thought to receive the greatest dose of both NOz and 0,. Subjects exposed to either NO2 or 0, and undergoing mild to heavy intermittent exercise experience a decrease in the ventilatory function of their lungs (ref. 5 presents a comprehensive review). When BAL is performed at the same time as exposure to 4 ppm NOz for 3 h, an increase in inflammatory cells, an increase in diene conjugation (indiciative of lipid peroxidation), and ~ a decrease in elastase inhibitor concentration was d e t e ~ t e d . ,Supplementation of the volunteers with 1,500 mg vitamin C/day and 1,200 IU vitamin Eiday prevented the loss of elastase inhibitory capacity and the increase in diene conjugation found in volunteers receiving placebo. These studies provide direct evidence that vitamin E and C supplementation prevents lipid peroxidation in the lung and protects the elastase inhibitory capacity of the lung lining fluid. The elastase inhibitory capacity is critical because it is generally thought that the loss of elastase inhibitory capacity (a specific protein) leads to emphy~ema.,~ According to the theory of Janoff, emphysema occurs because of an imbalance among digestive enzymes, such as elastase, elaborated by inflammatory cells in the lung and the elastase inhibitory factor elaborated into the lung. The Janoff theory holds that inflammation is an essential part of the generation of emphysema. Inasmuch as cigarette smoking, NO,, and O3 all stimulate inflammation in the lung, they all may be contributors to the exacerbation or initiation of emphysema in humans as occurs in experimental animals. No systematic study has been reported of the effects of cigarette smoking and vitamins E and C. Oxidative stress has been suggested as a mechanism of cigarette smoke toxicity, leading to several studies of vitamins E and C and

MENZEL: ANTIOXIDANTS AND LUNG DISEASE

151

smoking. Pacht et found little vitamin E in BAL from cigarette smokers compared to non-smokers and concluded that the lung lining fluid was deficient in smokers. Chow and Bridges3s and Smith and H o d g e ~ contended ,~ that chronic cigarette smoking lowered the serum vitamin C content of smokers. Bui et a1.” found no differences between male smokers and non-smokers in their serum vitamin C content but did find a marginal increase in the BAL vitamin C content of smokers compared to non-smokers. Bui et al. studied a group of volunteers with a high vitamin C intake of greater than 100 mg vitamin C per day. They suggested that the serum vitamin C content may saturate at a daily intake of greater than 100 rng and make the serum vitamin C content a relatively insensitive index of vitamin C stores in the lung. Their conclusion that smokers may have higher vitamin C content of their lung lining fluid as an adaptive response to the oxidant stress of cigarette smoking is an interesting point. These results are similar to the effects of O3 on mice, which increased the lung vitamin E level at the expense of the spleen vitamin E stores.3s To date only a very small number of volunteers has been studied, and the variation in lung vitamin C content is great. In all of the studies, chronic cigarette smoking increases the inflammatory cells in the lung and reinforces the idea that oxidant stress in the lung leads to chronic inflammation. Elsewhere in this symposium, Bucca et al.39report the protective effect of vitamin C on respiratory challenge by histamine in police working in the city. These subjects were exposed to the ambient mixture of air pollutants, but NO, most likely was the dominant toxicant. The results from these subjects indicate that their lungs could be protected against histamine challenge by ascorbic acid in the face of exposure to air pollution. Bhalla et a/.& found that exposure to O,, NOz, or to mixtures such as occur in urban air increase the permeability of the airways. They hypothesized that allergens o r vasoactive substances, such as histamine, could reach the underlying mast cells and smooth muscle cells more easily when the lung is exposed to air pollutants. Vitamin C was particularly effective for the histamine challenge experiments of Bucca et al. because of the ease with which NO2 destroys vitamin C.17*19

SUMMARY AND CONCLUSIONS Although the evidence for oxidative stress for air pollution in the human lung is fragmentary, the hypothesis that oxidative stress is an important, if not the sole, mechanism of toxicity of oxidizing air pollutants and tobacco smoke is compelling and growing. First, biochemical mechanisms have been worked out for oxidation of lung lipids by the gas phase of cigarette smoke, NO, and 0,.The oxidation of lung lipids can be prevented by both vitamins C and E. Vitamin C is more effective in preventing oxidation by NO2, and vitamin E is more effective against 0,.Second, multiple species of experimental animals develop lung disease similar to human bronchitis and emphysema from exposure to NO2 and O,, respectively. The development of these diseases occurs over a near lifetime exposure when the levels of NOz or O3 are at near ambient air pollution values. Third, isolated human cells are protected against oxidative damage from NO2 and O3 by both vitamins C and E. Fourth, the vitamin C level in the lung either declines on exposure to NO, for short-term exposures or increases on chronic cigarette smoke exposure. The effects of cigarette smoking on serum vitamin C is apparently complex and may be related to the

152

ANNALS NEW YORK ACADEMY OF SCIENCES

daily intake of vitamin C as well as smoking. Serum vitamin C levels may be poor indicators of lung demands when daily vitamin C intakes are above 100 mglday . Fifth, vitamin C supplementation protects against the effects of ambient levels of air pollution in adults as measured by histamine challenge. An augmented response to histamine challenge may represent increased lung permeability brought about by air pollution. In experimental animal and human experiments, the amount of vitamin C or E that afforded protection was in excess of the current recommended dietary allowance. Although animal studies do not provide evidence for complete protection against NO2 or 03,they do illustrate that current recommended daily allowances are inadequate for maximum protection against air pollution levels to which over 100 million Americans are exposed. The problem of air pollution and its effects on humans is truly of global concern. Air pollution is not restricted to North America or Japan where it was first recognized, but is a major public health problem in Europe as well. When data are available, air pollution probably will be shown to be a major public health problem in all urban areas of the world. The chemical similarities between urban air pollution and tobacco smoke, and the apparent increased demand for vitamin C in smokers, correlates well with the general idea of oxidant stress as a mechanism of toxicity. The decline in pulmonary function of children living in polluted areas is alarming. The lungs of these children fail to keep up with the growth of their bodies. The reserve respiratory function of these children is less than that of their counterparts living in less polluted areas.5.hThe progress of air pollution-induced disease (TABLEI ) is slow, and thus intervention to protect children must be carried out early to avoid adult respiratory disease. The biological basis for protection by the antioxidant vitamins may be the prevention of inflammation by blocking the production of “oxidative stress signal” molecules. Oxidative stress signal molecules may be aldehydes or complexes with proteins that recruit inflammatory cells into the lung. Damage to the antiproteases by oxidants tilts the balance toward proteolysis over repair leading to chronic lung disease. By reducing the amount of oxidative damage done to the cells of the airways, antioxidant vitamins could prevent allergens or vasoactive substances from reaching the smooth muscle cells regulating airway caliber and protect asthmatics from further disease. Antioxidant vitamins could also enhance the ability of macrophages to scavenge infectious agents and prevent the more frequent infections known to occur on air pollutant exposure. Antioxidant vitamins could also decrease the amount of growth substances elaborated in response to oxidative injury and thereby favor repair of the epithelial cells over endothelial cells. Although plausible, these hypotheses need to be investigated. A timely intervention trial with children, and especially with those likely to be at risk in areas of high air pollution, is needed. These data do not, however, support any reduction in the current recommended daily allowance of either vitamin C or E.

ACKNOWLEDGMENTS Dr. Lawrence Machlin of Hoffmann-La Roche, Inc. has provided continuing encouragement through the years of study of vitamins E and C; Dr. W. A. Pryor has provided stimulating discussions and suggestions on the chemistry of free

MENZEL: ANTIOXIDANTS AND LUNG DISEASE

153

radical reactions. Dr. Bart Ziegler, Mr. Steven Oddo, and Mr. Steven Owh were essential colleagues in our studies of human lung cells. Mrs. Sally Menzel and Mrs. Nancy Clarke provided invaluable assistance in the preparation of the manuscript and collection of the literature. REFERENCES

I. 2. 3. 4.

5.

6. 7. 8. 9. 10. 11.

12. 13. 14. 15.

16. 17. 18.

19.

BREE,L., P. J. LIOY,P. J. A. ROUMBOUT & M. LIPPMAN.1990. A more stringent and longer-term standard for atmospheric ozone. Toxicol. Appl. Pharmacol. 103: 377-382. SHOAF,S. R. & D. B. MENZEL.1988. Oxidative damage and toxicity of air pollutants. In Cellular Antioxidants and Defense Mechanims. C. K. Chow, Ed. Vol. 1: p 197-213. CRC Press. Boca Raton, FL. American Thoracic Society. 1978. Health effects of air pollution. American Lung Association. New York. F. GROSE, CHANG,L., F. J. MILLER,J. ULTMAN,Y. HUANG,B. L. STOCKSTILL, J. A. GRAHAM,J. J. OSPITAL& J. D. CRAPO.1991. Alveolar epithelial cell injuries by subchronic exposure to low concentrations of ozone correlate with cummulative exposure. Toxicol. Appl. Pharmacol. 109: 219-34. SAMMET,J. M. & M. J. UTELL. 1990. The risk of nitrogen dioxide: What have we learned from epidemiological and clinical studies. Toxicol. Ind. Health 6 247262. BERKEY,C. S., J. H. WARE,D. W. DOCKERY & 8 . G. FERRIS JR. 1986. Indoor air pollution and pulmonary function growth in preadolescent children. Am. J. Epidemol. 123: 250-260. GARDNER, D. E. 1984. Oxidant-induced enhanced sensitivity to infection in animal models and their extrapolations to man. J. Toxicol. Environ. Health 13: 423439. D. G. ALTMAN& A . V. SWAN.1977. Association between MELIA,R., C. V. FLOREY, gas cooking and respiratory disease in children. Br. Med. J. 2: 149-152. Toxicol. MENZEL,D. B. 1984. Ozone: An overview of its toxicity in man and animals. .I. Environ. Health 13: 183-204. FREI, B., T. M. FORTE, B. N. AMES& C. E. CROSS. 1991. Gas phase oxidants of cigarette smoke induce lipid peroxidation and changes in lipoprotein properties in human blood plasma. Biochem. J. 277: 133-138. & D. B. MENZEL.1971a. Antioxidants versuslungdisease. ROEHM,J. N., J. G. HADLEY Arch. Intern. Med. 128: 88-93. & D. B. MENZEL.1971b. Oxidation of unsaturated fatty ROENM,J. N., J. G. HADLEY acids by ozone and nitrogen dioxide: A common mechanism of action. Arch. Environ. Health 23: 142-148. DONOVAN, D. H., S . J. WILLIAMS,J. M. CHARLES& D. B. MENZEL.1977. Ozone toxicity: Effect of dietary vitamin E and polyunsaturated fatty acids. Toxicol. Lett. 1: 135-139. CHOW,C. K.,C. G. PLOPPER,M. CHIU& D. DUNGWORTH. 1961. Dietary vitamin E and pulmonary biochemical and morphological alterations in rats exposed to 0.1 ppm ozone. Environ. Res. 24: 315-324. CHOW,C. K. & A. L. TAPPEL.1972. An enzymatic protective mechanism against lipid peroxidation damage in lungs of ozone-exposed rats. Lipids 7: 518-523. PRYOR, W. A. 1991. Can vitamin E protect humans against the pathological effects of ozone in smog? Am. J. Clin. Nutr. 53: 702-722. SHOAF,C. R., R. L. WOLPERT& D. B. MENZEL.1989a. Nitrogen dioxide-initiated peroxidation of liposomal membrane systems. Inhalation Toxicol. 1: 301-3 14. HATCH,G. E., H. KOREN& M. AISSA.1989. A method for comparison of animal and human alveolar dose and toxic effect of inhaled ozone. Health Phys. 1: 37-40. SHOAF,C. R., R. L. WOLPERT& D. B. MENZEL.1989b. Antioxidant effects of Qtocopherol and ascorbate in liposomes exposed to nitrogen dioxide. Inhalation Toxicol. 1: 315-329. VAN

154

ANNALS NEW YORK ACADEMY OF SCIENCES

& B. ZtEciLER. 1991. Imaging of reactive oxygen 20. M ~ N Z E LD., B., R. VANDAGRIFF species in living lung cells exposed to oxidizing air pollutants. I n Oxidative Damage and Repair. J. F. K. Davies, Ed.: 533-538. Pergamon Press. New York. A. M. BRYANT& H. 0. JAUREGUI. 1975a. Heinz 21. MFNZEI.,D. B., R. J . SLAUGHTER, bodies formed in erythrocytes by fatty acid ozonides and ozone. Arch. Environ. Health 30: 296-301. D. B., R. J . SLAUGHTER, A. M. BRYANT & H. 0. JAUREGUI. 197%. Preven22. MENZEL, tion of ozonoide-induced Heinz bodies in human erythrocytes by vitamin E. Arch. Environ. Health 30: 234-238. B. L. & A. L. TAPPEL.1973. Protective effects of dietary a-tocopherol in 23. FLETCHER, rats exposed to toxic levels of ozone and nitrogen dioxide. Environ. Res. 6 165175. 1988. Glutathione S-trasferases-Structure and B. & U. H. DANIELSON. 24. M.ANNERVIK, catalytic activity. CRC. Crit. Rev. Biochem. 23: 283-339. 1987. Methyl 25. Vos, R. M., I. M. REITJENS,G . M. ALINK& P. J. VAN BLADEREN. linoleate ozonide: A substrate for rat glutathione S-transferase. Eur. J. Drug Metab. Pharmacokinet. 12: 245-277. L. H. STEVENS & P. J. VAN BLADEREN. 1989. Methyl 26. Vos, R. M., 1. M. REITJENS, linoleate ozonide a s a substrate for rat glutathione S-transferases: Reaction pathway and isoenzyme selectivity. Chem.-Biol. Interactions 69: 268-278. J., W. R. VORACHEK, R. W. PERO& W. R. PEARSON.1988. Hereditary 27. SEIDEGARD, differences in the expression of the human glutathione transferase active on transstilbene oxide are due to a gene deletion. Proc. Natl. Acad. Sci. USA 85: 72947297. 28. SEIDEGARD, J., R. W. PERO,D. G. MILLER& E. J. BEATTIE.1986. A glutathione transferase in the human leukocytes as a marker for the susceptibility to lung cancer. C'arcinogenesis 7: 751-753. R. HUSSNATTER, J . HARRIS,D. J. MEYER,S . PETER,B. , U. WEIDNER, 29. K L O N EA,, & H . SIES.1990. Decreased expression of the glutathione S-transferase KETTERER alpha and pi genes in human renal cell carcinoma. Carcinogenesis 11: 2179-2183. 30. MI.NZEL,D. B. & R. L . WOLPERT.1989. Is There A Threshold for Ozone? I n Ozone Research and Its Policy Implications. T. Schneider & L. D. Grant, Eds. Elsevier Biomedical Press. Amsterdam. 31. MII.LER, F. J., J. H. OVERTON JR., R. H. JASKOT& D. B. MENZEL.1985. A model of the regional uptake of gaseous pollutants in the lung. I. The sensitivity of the uptake of ozone in the human lung to lower respiratory tract secretions and exercise. Toxicol. Appl. Pharmacol. 79: 11-27, V. 1991. Lipid peroxidation and antielastase activity in the lung under 32. MOHSEININ, antioxidant stress. J. Appl. Physiol. 70: 1456-1462. 33. JANOFF,A. 1985. Elastase and emphysema: Current assessment ofthe protease-antiprotease hypothesis. Am. Rev. Respir. Dis. 132 417-433. & W. B. DAVIS.1986. D. G. CONNWELL 34. PACHT,E. R., H. KASEKI,J. R. MOHAMMED, Deficiency of vitamin E in the alveolar fluid of cigarette smokers. Influence of alveolar macrophages. J. Clin. Invest. 77: 789-796. 35. CHOW,C. K. & R. B. BRIDGES. 1989. Chronic cigarette smoking and plasma micronutrients. Fed. Proc. Fed. Am. SOC.Exp. Biol. 43: 860. 36. S M I T HJ., L . & R. E. HODGES.1987. Serum levels of vitamin C in relation to dietary and supplemental intake of vitamin C in smokers and nonsmokers. Ann. N.Y. Acad. Sci. 498: 144-152. 37. Bui, M. H.. A. SAUTY,F. COLLET& P. LEUENBERGER. 1992. Dietary vitamin C intake and concentrations in the body fluids and cells of male smokers and nonsmokers. J. Nutr. 122: 312-316. 38. MORGAN,D. L., T. L. FURLOW& D. B. MENZEL.1988. Ozone initiated changes in erythrocyte membrane and loss of deformability. Environ. Res. 45: 108-1 17. 39. BUCCA,C., G . ROLLA& J. C. FARINA. 1992. Effect of vitamin C on transient increase of bronchial responsiveness in conditions affecting the airways. Ann. N.Y. Acad. Sci. This volume. 40. BHALLA,D. K., R. E. RASMUSSEN & S. TJIEN.1990. Interactive effects of 0,. cyto-

MENZEL ANTIOXIDANTS AND LUNG DISEASE

155

chlasin D, and vinblastine on transepithelialtransport and cytoskeletonin rat airways. Am. J . Respir. Cell Mol. Biol. 3: 119-129.

DISCUSSION UNIDENTIFIED SPEAKER: In my studies, I have found that ozone can deplete vitamin A. Have you ever seen anything similar? D. B. MENZEL(University of Calqornia at Irvine): Beta-carotene, a precursor of vitamin A, is a better antioxidant for singlet oxygen than is vitamin E. A lot of the products of decomposition of ozone do involve singlet oxygen and not the hydroxyl radical, so it is likely that ozone could deplete beta-carotene and vitamin A. F. COLBY(New York, N Y ) : Do you agree that indoors microorganisms are at least as important as “chemicals.” MENZEL:The question of microorganisms indoors is part of the sick building syndrome. A lot of ozone is produced within buildings. For instance, the photocopy machine is an abundant source of ozone. You can smell it, above one part per million. But the issue here is that in the infectivity model, the animal is given a nonlethal dose of infectious agent too. Exposure to those chemicals that I discussed, either simultaneously or ahead of time, will increase the lethality. It is very clear that the respiratory infection rate is much higher for children in homes that have cigarette smokers or that have gas cooking stoves or space-heating units. L. J. MACHLIN(Hoffmann-Lu Roche, Nutley, N J ) : Dr. Mohseinin at Yale showed that vitamin C would protect against the destruction of alpha 1 trypsin inhibitor, and that certainly fits into the idea ofa protective role by the antioxidative vitamins. Dr. Wolf in Germany also found a protective role of vitamin E in respiratory distress syndrome, and Dr. Pacht has reported a decrease in vitamin E in alveolar fluid of the lung. Are there any other human studies pertaining to a possible protective role of antioxidant vitamins in the lung? MENZEL:There has been a very interesting series of experiments by Koran and his co-workers at the University of North Carolina and the EPA, in which they lavaged the lungs of healthy young volunteers before and after exposure to ozone and nitrogen dioxide. They were able to show that these volunteers had depressed levels of both vitamins C and E after exposure to ozone and nitrogen dioxide. Paradoxically, the effect is far more obvious with ozone, because the chemical rates of reaction of nitrogen dioxide with ascorbic acid are about 1000fold greater than the rate of reaction with vitamin E. Therefore, I would expect NOz to have a much stronger effect on vitamin C, which is in line with what you said.