Studies in adaption to ambient oxidant air pollution: effects of ozone exposure in Los Angeles residents vs. new arrivals.

To test the hypothesis that adaptation protecting against acute effects of ambient ozone (O3) exposures develops in Los Angeles residents, human volunteers were exposed to 0.4 ppm O3 under conditions simulating ambient pollution exposures. Blood biochemical, pulmonary physiological, and clinical responses were assessed. Los Angeles residents (N = 6) showed only minimal clinical or physiological response to O3, while new arrivals (N = 9) showed significant losses in pulmonary function and a tendency toward increased symptoms. Most biochemical responses did not differ significantly between residents and new arrivals. These results agree with others in suggesting that exposures to elevated ambient concentrations of O3 produce adaptation in a least some residents of photochemical pollution areas. The underlying mechanisms and long-term consequences of such adaptation are unknown.


Introduction
Development of tolerance to ozone (03) and other irritant gases in experimental animals was first described by Stokinger and co-workers approximately 20 years ago (1) and has been studied extensively since. The subject has been reviewed by Fairchild (2) and Morrow (3). Salient features of animal tolerance include the following. Pretreatment with a relatively low 03 dose will prevent death or severe lung injury which would otherwise occur with a higher dose. This tolerance gradually disappears after cessation of 03 exposure. Cross tolerance exists among 03 and other irritant gases, including some which, like *Environmental Health Service, Rancho Los Amigos Hospital, 7601 East Imperial Highway, Downey, Calif. 90242.
tUSC School of Medicine, Physical Therapy Department, 12933 Erickson Avenue, Downey, Calif. 90242. 03, are powerful oxidizing agents and others which are not. Tolerance does not prevent the development of chronic lung lesions following repeated exposures. Tolerance results in decreased edema formation in response to 03 challenge, but no diminution of cytotoxic effects of 03 iS observable (4). The biological mechanisms responsible for tolerance are largely unknown.
The observation that animals can respond to a toxic inhalation challenge in a manner which prevents some of the short-term adverse effects of further exposures suggests the possibility that an analogous response might occur in humans exposed to community air pollution. We use the term "adaptation" to describe this hypothetical response in humans, since the doses of toxicants being considered are much less than in animal "tolerance" studies, and since responses are less severe and perhaps depend on different biological phenomena. Metropolitan Los Angeles experi-ences uncommonly high ambient levels of 03 and other oxidants during photochemical smog episodes; thus residents of this area constitute an attractive group in which to investigate the possibility of adaptation. That Los Angeles residents suff6ri'-l-ess deleterious~effects-ofambient exposures than visitors to the area has been previously suggested (5), but the hypothesis has never been tested extensively.
Previous work in our laboratory (6,7) showed that some healthy Los Angeles residents develop respiratory symptoms and function changes when exposed to 03 concentrations of 0.37-0.50 ppm-less than maximum ambient concentrations in the area. Similar studies in Canadians not frequently exposed to ambient oxidants (8,9) appeared to show a greater mean effect of a given dose, suggesting that responses in Los Angeles residents might have been reduced by adaptation. Methodological differences between studies might have explained the apparently different response, however. To test this possibility, a cooperative investigation was undertaken to compare experimental methods and responses of a small sample of subjects to 0.4 ppm 03 (10). The results reproduced to a great extent the previous finding of less reactivity in Los Angeles residents as compared to Canadians and failed to reveal any methodological factors which could account for this difference. The hypothesis of adaptation was thus supported. To test the hypothesis more rigorously, the present study was undertaken in order to compare the effects of 0.4 ppm 03 in somewhat larger and more carefully matched groups of Los Angeles residents and nonresidents.

Methods
The null hypothesis tested was as follows. Healthy Los Angeles residents (three years or more in area) and new arrivals (five days or less in area) will not differ in mean clinical, physiological or biochemical response to 0.4 ppm 03 exposure under conditions simulating ambient pollution episodes. Rejection of the null hypothesis with new arrivals showing significantly greater mean response would be the necessary result if the hypothesis of 03 adaptation in residents were to be supported.
The exposure facility and basic experimental design have been described in detail previously (11). Volunteer subjects were studied on two successive days. The first day's exposure was to purified air only; the second day's exposure was to 0.40 ppm 03 in purified air. Exposures lasted 2 hr 15 min. During the first 2 hr, each subject exercised at a workload sufficient to increase minute volume to approximately twice the resting level (150-200 kg-m/min) for 15 min in every half-hour. During the last 15 min pulmonary function tests were performed; these included forced vital capacity (FVC), one-second forced expiratory volume (FEVy), maximum midexpiratory flow rate (MMF), total respiratory resistance by forced oscillation (R), and indices of the singlebreath nitrogen test: closing volume as a percent of vital capacity (CV/VC), and slope of the alveolar plateau (AN2). Each subject's test results were expressed as control values (those obtained after purified-air exposure) and as O3 responses (differences between post-03 exposure and control values). Subjects' symptoms during and following exposure were recorded and scored semiquantitatively according to severity and duration using a standard interview questionnaire administered by the project medical officer. The symptom response to 03 was expressed as the difference in symptom score between 03 exposure and control days. Venous blood samples were drawn immediately following exposure, and erythrocyte (RBC) and serum analyses were performed to detect changes expected to result from an oxidant challenge, as described previously (12).
Paired statistical tests with each subject serving as his own control were applied to detect differences between control and 03 conditions for the resident group and for the new-arrival group. Unpaired tests were applied to compare between groups. For physiological measures, only 03 responses were compared between groups, as control values were expected to depend mostly on body size and not on adaptation. For biochemical measures, control values could have differed between groups as a consequence of adaptation, therefore both control values and 03 responses were compared statistically. In addition to the commonly employed t tests, analogous nonparametric tests-the Wilcoxon signed-rank test for paired analyses and the Mann-Whitney U test for between-group analyses-were applied to the pulmonary function data. The nonparametric tests were expected to be possibly more powerful in analyzing these data since the data were expected to be skewed, whereas t tests require a normal distribution for greatest reliability. Skewness is inherent in data of this nature since there is considerable variability between individuals in reactivity to exposure, and since Environmental Health Perspectives function measures remain similar to control values in relatively unreactive subjects but deviate from control values in only one direction in more reactive subjects. Symptom data, which were not rigorously quantitative and not necessarily expected to show a normal distribution even under control conditions, were analyzed only with the nonparametric tests.
Subjects were recruited within the incoming 1975 class of the USC School of Physical Therapy. Fifteen of a possible 44 individuals volunteered to be studied; six of these were residents of metropolitan Los Angeles and nine were nonresidents. Studies were conducted during September, i.e., late in the summer smog season when residents should have had ample time to develop adaptation. Nonresidents were studied within five days of their arrival in Los Angeles; they were instructed to minimize intercurrent ambient oxidant exposures by remaining in coastal areas of metropolitan Los Angeles and/or remaining indoors and at rest during peak oxidant hours.
Individual subject characteristics are given in Table 1. Since the nonresidents included two males, while the residents were all female, the possible effect of sex differences on the overall results was examined. The males' data were compared individually with the female nonresidents' for the three measures which showed significant (p <0.05) group differences. Both males' values fell within the females' range, except that one male had the largest control and post-exposure FEV,. When statistical analyses were repeated excluding the males' data, mean group responses were actually larger than when the males were included; however, due to the reduction in sample size the level of significance of the group differences decreased-0.05 <p < 0.10 with males excluded. Overall, no evidence was found that sex differences affected the results; this was also the case in the previous study (10).

Results
Individual physiological and clinical responses are given qualitatively in Table 1, and group mean physiological and symptom measures are summarized in Table 2. The residents as a group showed no significant 03 responses except for slight decrease in AN2. Increases in AN2 are normally expected in chronic pulmonary dysfunction and in acute responses to 03 exposure (6). Decreased values represent increased uniformity of ventilation distribution and thus could be considered an improvement in function. On the other hand, more uniform distribution could be the result of adverse physiological changes, such as complete "closure" of a few small airways previously only partially obstructed. Nonresidents "P = physiological response-significant (p<0.05) loss in FVC and/or FEV, with 03 exposure relative to control, determined by t test, three measurements under each condition. S = symptom response-increase in symptom score of > 4 units (arbitrary definition of "clinically significant" response).
' Spent previous summer in area. showed a smaller, nonsignificant decrease in AN2, but showed significant 03 responses in FVC, FEV1, and MMF. The MMF response was significantly more severe than in the residents according to the intergroup comparison, but the FVC responses did not differ significantly between the groups. The FEV, loss was significantly more severe in nonresidents than in residents according to the U test (p = 0.03), but not according to the t test (p = 0.06). Since the distributions of FEV, responses appear skewed (Fig. 1), the results of the U test may be more reliable.
Neither group showed significant responses of CV/VC, R,, or symptom score, but the non-  residents showed a trend toward increased symptom score with 03 exposure.
Group mean biochemical measurements and significant changes related to exposure are summarized in Table 3. None of the analyses showed significant differences in control values between residents and new arrivals, although residents showed trends toward less fragility of RBCs as determined by hydrogen peroxide challenge, and higher serum concentrations of Vitamin E. Both groups showed 03 responses generally similar to those seen previously (12): increased RBC fragility, reduced RBC acetylcholinesterase activity, and tendencies toward increased activity of pentose pathway enzymes (which would tend to protect against excessive oxidation of cellular components). Lactate dehydrogenase (LDH) activity was the only biochemical measure to show a significant difference between groups in response to 03. New arrivals showed the expected increase in LDH activity, while residents showed a decrease, in contrast to previous findings (12). The biological significance of this observation, if indeed it represents other than a chance occurrence, is unclear.

Discussion
These results support the hypothesis of adaptation to 03 in Los Angeles residents. Statistical differences found between residents and new arrivals are relatively small, as should be expected given the unavoidably small sample sizes and the Environmental Health Perspectives I typically large individual variability in 03 responses. Controlled-exposure studies cannot be done on a large enough scale to conclusively establish differenes in response between populations, but the essetial agreement of present and previous results in small-scale studies considerably strengthens the case for the existence of such differences. Various factors unrelated to inherent adaptive biological reponses could explain these results-selective migration or diet, for example (10). No such factor has yet been identified, leaving adaptation as the most plausible explanation for the experimental observations. No biochemical index of the adapted state has yet been found in animals or in man, nor are the physiological and biochemical mechanisms of 03 toxicity well understood. Further investigations in these areas will be necessary before the biological mechanisms of the adaptive response (if it exists) can be elucidated. Of particular interest is the possibility that adaptive mechanisms may be inoperative in certain individuals, who might then be at increased risk of developing chronic respiratory disease.
The phenomenon of adaptation may ultimately, but should not presently, be taken into account in setting ambient or occupational air-quality standards. By analogy with animal studies, it appears that human adaptation to acute 03 effects might not protect against the possible development of chronic lung damage after many exposures. Unless this possibility and the possibility of failure of adaptation are conclusively ruled out, air quality standards should continue to be set to protect the susceptible, least well-adapted individuals in the exposed population.