- Email: [email protected]
Nitrate formation in rats exposed to nitrogen dioxide
67,284-291(1983)
TOXlCOLOCYANDAPPLIEDPHARMACOLOGY
Nitrate Formation
in Rats Exposed to Nitrogen
Dioxide
ROBERT L. SAUL AND MICHAEL C. ARCHER’ Department
of Medical Biophysics, 500 Sherbourne Street,
Received
University of Toronto, Toronto, Ontario M4X
July 16, 1982; accepted
October
Ontario Cancer IK9, Canada
Institute,
25, 1982
Nitrate Formation in Rats Exposed to Nitrogen Dioxide. SAUL, R. L., AND ARCHER, M. C. (1983). Toxicol. Appl. Pharmacol. 67, 284-291. Sprague-Dawley rats exposed to atmospheres containing low levels of nitrogen dioxide (Nq) for 24 hr had increased levels of nitrate in their mine on the day of exposure and on the 3 subsequent days. The total increase in urinary nitrate was linearly related to the nitrogen dioxide concentration administered. We recovered in urine 8.4 + 1. I pmol nitrate/ppm N02/24-hr exposure (slope + 95% confidence limits) for 185-g rata. Both the linearity and magnitude of this effect imply that reaction with respiratory tract water is not a major pathway of NOr absorption in the lung. Instead, our observations support the hypothesis that the major interaction of NOz in the lung is with readily oxidizable tissue components to form nitrite. We estimate that 9.6 pmol of nitrite is formed in the respiratory tract ofthe rat per ppm NO1 per 24-hr exposure. We also estimate that humans breathing air containing 0.1 ppm NO2 have about 3.6 mg of nitrite formed in their respiratory tract per day.
Nitrogen dioxide (NOz) is a common atmospheric pollutant that arises from a wide range of high-temperature combustion processes. It is a lung irritant, producing a pathological condition similar to human emphysema in experimental animals (Freeman et al., 1968). The chemical mechanisms of NO2 toxicity are not understood, but are probably related to the oxidative nature of this agent (Roehm et al., 197 1). Recently, NO2 has also been studied as a potential precursor of carcinogenic N-nitroso compounds which may be formed in vivo (Iqbal et al., 1980; Mirvish et al., 1981). The fate of inhaled NO2 was studied by Svorcova and Kaut ( 197 1) who exposed rabbits to a single high concentration of NO* for 4 hr and observed elevated levels of both nitrate and nitrite in the blood and urine. More recently, Oda et al. (198 1) found that the nitrate and nitrite levels in the blood of mice rose quickly to constant values during expo-
sure to N02, and that afer exposure, the blood levels declined with half lives of about 1 hr for nitrate, and a few minutes for nitrite. The short half-life of nitrite was attributed to its reaction with oxyhemoglobin to form methemoglobin and nitrate. Goldstein et al. (1977) studied the distribution of inhaled NO2 in rhesus monkeys with radioactive i3N02. They demonstrated that 50 to 60% of inspired NO2 was retained during quiet respiration and confirmed that N02, or its chemical reaction products, is transported to distant sites via the bloodstream. These authors suggested that NO2 reacts with water in the lung to form nitric and nitrous acids (conjugate acids of the nitrate and nitrite ions) and that the toxicity of NO2 is due to the reaction of these acids with pulmonary and extrapulmonary tissues. Postlethwait and Mustafa (198 1) studied the fate of NO2 in isolated perfused rat lungs. Surprisingly, for perfusate containing no erythrocytes they found that nitrite, but not nitrate, ’ To whom requests for reprints should be addressed. was produced in lungs exposed to NO*. The
0041-008X/83/020284-08$03.00/0 Copyright 0 1983 by Academic Press. Inc. All rights of reproduction in any form reserved.
284
URINARY
NITRATE
OF RATS EXPOSED
lack of nitrate indicated that the primary reaction of NO2 in the lung is not with water, since this reaction is well known to yield both nitrate and nitrite as products (Denbigh and Prince, 1947). Postlethwait and Mustafa (198 1) have suggested that instead of water, NO2 reacts with readily oxidizable tissue components to yield only nitrite. Intratracheal administration of [“Nlnitrate and [ i3N]nitrite to rodents has shown that both these ions are rapidly absorbed from the respiratory tract into the blood (Parks et al., 1981). Nitrate in the blood is known to be excreted rapidly and in high yield as urinary nitrate (Greene and Hiatt, 1954; Hawksworth and Hill, 197 1). Nitrite in the blood is quantitatively oxidized by oxyhemoglobin to form nitrate (Kosaka et al., 1979) and so should also be recovered as urinary nitrate. Thus, the formation of either nitrate or nitrite from NO2 in the lung could be detected by observing the nitrate levels in the urine during and after exposure to NO*. In the present study, we have used this approach to determine the extent of nitrate and nitrite formation in rats exposed to various concentrations of NO2 and have used our data to suggest the mechanism of this process.
METHODS An airtight glove box (Labconco Corp., Kansas City, MO.; internal volume approximately 350 liters), equipped with an air lock, was modified for use as a controlled atmosphere chamber by disconnecting its air circulation tubing and using the two connection ports for the inflow and outflow of gases. The NO* levels inside the chamber were checked periodically by sampling the gas inside the glove box or in the exit tube. The two sampling sites yielded equivalent results. The NO2 concentrations were determined by the Saltzmann method with either the syringe technique or the fritted bubbler technique (Katz, 1976). Three metabolic cages (Nalgene Co., Rochester, N.Y.) were placed inside the chamber, and for each experiment, three male Sprague-Dawley rats, (Charles River Canada, Inc., La Prairie, Quebec) initially 36 days old and weighing 162 f 8 g (i & SD), were individually housed in the cages for a period of 7 days. With he exception of the fourth day, the inlet gas was synthetic air (Liquid Carbonic of Canada, Ltd., Scarborough, Ontario) at a
TO NO,
285
flow rate of 2 liters/min. We determined that this gas contained no NO* (detection limit 0.01 ppm). On the fourth day of each experiment the animals were exposed to atmospheres containing NO*. A mixture of 3 14 ppm (by volume) NO, in air from a gas stock tank (Liquid Carbonic of Canada Ltd., concentration self-determined) was dynamically combined with synthetic air to give the desired dilution at an overall flow rate of 5 liters/min. During the first 2 hr of the 24-hr exposure period, the NOr concentration of the inlet gas was set 60% higher than the desired level to achieve a rapid increase in the chamber NO, concentration. Once the desired chamber concentration was reached, the inlet gas concentration was reduced and periodically adjusted so that the chamber NO2 concentration remained constant within +lO% over the next 20 hr. During the last 2 hr of the exposure period, the chamber NO* concentration was returned to zero by purging the chamber with synthetic air. The average NO2 concentration was calculated by integrating the measured NO2 concentrations over the entire 24-hr exposure period. Six treatment groups (three rats per group) were exposed to 24hr average NOz concentrations of zero, 1.2, 3.4, 4.1, 5.6, and 8.8 ppm. On the 3 days before and 3 days after the exposure day, the rats were allowed unlimited amounts of deionized water and a casein and corn starch-based, low-nitrate, low-nitrite diet (Biomix 1435, Bioserve Inc., Frenchtown, N.J.). Using the method of Sen and Donaldson (1978), we measured the nitrate content in this diet as 0.085 pmol/ g, and no detectable nitrite with a detection limit of 0.005 pmol/g. Since we found that NOr reacts with this food to form nitrate and nitrite, the animals were given only water on the day of NOr exposure. Water bottles and food containers were changed daily, and food consump tion was determined. Urine was collected for each 24-hr period. Ten milliliters of 0.6 M NaOH solution was added before use to each urine collection container to prevent nitrate or nitrite destruction or synthesis due to contaminating fecal microorganisms (Saul et al., 198 1). To minimize the accumulation of feces on the collection funnel, separating cone, and urine ring, the metabolic cages were disassembled, washed inside the glove box with deionized water, and reassembled each day. The alkaline, preserved urine samples were transferred to lOO-ml volumetric flasks, diluted to volume with deionized water, stored for no longer than 7 days at 4”C, and analyzed as described below. It was shown by repeated analyses that the nitrate and nitrite levels in these samples were stable for several months when stored under these conditions. Urinary nitrate and nitrite were analyzed by a method based on that of Sen and Donaldson ( 1978). Twenty milliliters of the diluted alkaline urine sample was heated at 60°C for 10 min. Two milliliters of 0.42 M ZnSO, solution was added to precipitate protein, and the mixture was heated a further 10 min with occasional swirling. After cooling to room temperature, the mixture was suc-
286
SAUL AND ARCHER
tion filtered through a disposable tilter unit (0.45 pm, Nalgene Co.; unit was prewashed with 300 ml of deionized water to remove contaminating nitrate), and the filtrate was transferred to a 50-ml volumetric Bask. The residue on the ftlter was washed with 20 ml water, and the wash was combined with the filtrate and diluted to volume. Two lO-ml portions of this solution were assayed for nitrate and nitrite as described by Sen and Donaldson (1978) by cadmium reduction of nitrate to nitrite and the Griess reaction for nitrite detection. The Greiss reaction was always performed at 20°C since the azo dye yield is temperature dependent. Some urine samples were spiked with standard solutions of nitrate or nitrite, and a standard curve was constructed to relate the resulting azo dye absorbances with the amounts of these ions in the urine. Recoveries of both nitrate and nitrite were greater than 95%. Based on multiple analyses, we estimate that our individual measurements of nitrate and nitrite in urine were accurate to within 0.5 pmo1/24-hr urine sample. The detection limit was 0.5 pmol/sample for both nitrate and nitrite.
RESULTS Average daily food consumption for the six groups of rats treated with NO2 (not including Day 4 when all animals were starved) was 18.7 + 3.5 g (2 + SD). This value represents an average dietary nitrate intake of 1.6 pmol/ day. Average weight gain over the 7&y experimental period for these rats was 45 f 6 g (2 + SD). There were no significant NOzdependent differences in food consumption or weight gain between treatment groups. All rats weighed approximately 185 g on the day of NOz exposure. Figure la shows the average daily urinary nitrate output of the three rats which were exposed to zero ppm NOz (synthetic air) on Day 4. No nitrite (detection limit 0.5 gmol/ day) was found in any of these urines. The reduced nitrate output on Day 4 was a result of their fast. On all 7 days of this experiment, daily urinary nitrate output exceeded dietary nitrate intake by 3.2 * 0.2 firno1 (J? +_ SE, n = 2 1). This observation is consistent with those ofGreenetal.(1981)andWitteretal.(1981) who also reported excess urinary nitrate for rats consuming low-nitrate diets. Figure lb shows the average daily urinary nitrate output for a typical experiment in-
+
,
lzJzlz 2
3
L 4
I-
5
Ill 6
7
FIG. 1. Average daily urinary nitrate levels for rata exposed to NO, on Day 4. Bars represent X + SE for three rats. Average 24-hr NO, concentrations for these two experiments were (a) zero ppm and (b) 4.1 ppm.
volving three rats which were exposed to NOz on Day 4, in this case at an average concentration of 4.1 ppm. The levels shown in Fig. lb are significantly higher than those in Figure la for Days 4 through 7. Figure 2 shows the 4-day total nitrate excretion for Days 4 through 7 as a function of NO* exposure concentration. Regression analysis indicates that these data are best fit with a straight line. The least-squares fit shown in Fig. 2 has a slope of 8.4 f 1.1 pmol nitrate/ppm NOz/day (slope + 95% confidence limits, correlation coefficient 0.970). Some nitrite was also found in the urine of animals exposed to NOz, but only in samples collected from the day of exposure. As will be discussed below, this nitrite was not due to urinary nitrite excretion but was a result of the in vitro reaction of urine with NO*. In addition to interactions within the respiratory tract, NO2 could also react with the rats’ food, drinking water, and body hair to form nitrate or nitrite. These reaction products could then be ingested by the animal
URINARY
90
NITRATE
OF RATS EXPOSED
TO NO*
2
4 dioxide
0
287
-
00 -
0
I
3 Nitrogen
5 6 7 concentrotion(ppm)
9
IO
FIG. 2. Total urinary nitrate output for Days 4 through 7 for rats exposed to NO1 for 24 hr on Day 4. Each symbol represents the total for one rat.
through the processes of eating, drinking, or licking and excreted as nitrate in the urine. Nitrogen dioxide could also react with the plastic walls of the urine collection apparatus or with the urine itself, either in the collection container, or while dripping along the walls of the collection funnel. All of these reactions could potentially lead to an overestimation of the contribution of inhaled NO2 to urinary nitrate. Nose- or head-only exposure to NOz might have circumvented this problem. Such exposure systems, however, suffer from the disadvantage that the exposure time must normally be less than 1 hr (Menzel and McClellan, 1980). Since much longer exposure times were necessary to produce measurable increases in urinary nitrate at low NOz concentrations, the nose- or head-only approach was impractical. Also, as will be seen in the discussion, we were interested in the effect of NO2 on the whole animal for the purpose of studying nitrate balance. To assess the potential problems of our whole-body exposure method, several experiments were performed which are described below.
The powdered food used in our experiments reacted readily with NO1 to form both nitrate and nitrite. Therefore, all animals were fasted on the day of NO2 exposure to prevent ingestion of this nitrate and nitrite which would ultimately appear in urine as nitrate. Deionized water reacts extremely slowly with NOz at the concentrations used in this study to form small amounts of nitrate and nitrite. Since water was administered by an inverted drinking bottle, contact of the water with chamber air was minimal. The animals were therefore allowed water during the NOa exposure period. Analysis of water from the drinking bottles after the higest NO2 concentration was used (8.8 ppm over 24 hr) showed no detectable nitrate or nitrite (detection limit 0.0 1 pmol/ml). Although water reacts slowly with low concentrations of NOz, urine was found to react more rapidly to form nitrite. In a number of experiments designed to examine this reaction, neutral or alkaline urine exposed to NO* yielded increased levels of nitrite, but no detectable increase in nitrate. Apparently it is
288
SAUL AND ARCHER
not the water, but rather, oxidizable chemicals present in urine which reduce NOz to nitrite. Since this in vitro reaction of NO2 with urine does not produce nitrate, it did not lead to an overestimation of the levels of nitrate in our rat urines for Day 4. Although our nitrate determinations were based on the subtraction of the nitrite concentration from the nitrate plus nitrite total (Sen and Donaldson, 1978), the levels of nitrite in urines from Day 4 were small enough that they did not reduce the accuracy of this subtraction. A clean metabolic cage, without a rat inside, was exposed to 6.7 ppm NO* for 24 hr and then rinsed with deionized water. The rinse water was collected and analyzed for nitrate and nitrite. Trace quantities (less than 1 pmol) of both ions were found but these were too small to contribute significantly to our observations of urinary nitrate. Several experiments were performed to test the extent of NO;? reaction with the animals’ hair to produce nitrate or nitrite which could be ingested as a result of licking. On the day of NO2 exposure (Day 4), rats were placed in a restraining device which prevented them from licking themselves but still allowed access to their drinking water. Unrestrained rats exposed to the same atmosphere were treated as described under Methods and served as controls. After the NOz exposure period, the restrained rats were released and immediately rinsed with deionized water to remove nitrate or nitrite from their body exteriors. The differences in total urinary nitrate output for Days 4 through 7 between restrained and unrestrained animals were then used to assess the contribution of licking to the total effect. We estimate that about 20% of the observed slope in Fig. 2 is a result of licking, while about 80% of this slope is a result of interactions of NO* within the respiratory tract. Additional data used to support this estimation came from our analysis of the water used to rinse NOz-exposed animals which showed the presence of nitrite, but not nitrate. On a molar basis, the amount of nitrite found on the animals’ hair was approximately equal to
60
0
IO
20 Injected
d 30 40 nitrite(pmol)
I 50
J 60
FIG. 3. Total urinary nitrate output for Days 4 through 7 for rats injected with sodium nitrite solution on Day 4. Each symbol represents the total for one rat.
the difference in urinary nitrate between the restrained and unrestrained rats. As will be seen in the Discussion, the linear relationship between urinary nitrate and NO2 concentration suggests that the major interaction of NO2 in the respiratory tract is the formation of nitrite which is absorbed into the blood and excreted in the urine as nitrate. To determine the amount of nitrite in the blood which is recovered as nitrate in the urine, we injected rats iv with nitrite solutions and subsequently observed the urinary nitrate levels. The experimental protocol used for this purpose was identical to that used in the NO2 experiments except that on Day 4, the animals were exposed to synthetic air and were given an injection of nitrite solution at the midpoint of this day. Zero, 15, 30, and 45 pm01 of sodium nitrite in 0.3 ml of isotonic saline were administered via a tail vein. The nitrate excretion profile for the rats injected with nitrite was very similar to that of the rats exposed to NO*, with most of the excess nitrate appearing on Days 4 and 5. No nitrite was found in any of the urines of these animals. Figure 3 shows the linear relationship between total nitrate excreted on Days 4 through 7 and the injected nitrite dose. The slope of the linear regression line is 0.70 rf: 0.10 pm01 nitrate/pm01 nitrite (slope + 95%
URINARY
confidence 0.994).
limits,
correlation
DISCUSSION
NITRATE
OF RATS EXPOSED
coefficient
TO NO2
289
the concentration of the dimer is proportional to the square of the NO2 concentration, the rate of absorption NOz into water is given by dN02 = kA[N0212, dt
Our finding of a urinary nitrate increase of 8.4 pmol/ppm NO*/day is the total increase where A is the area of the gas-water interface, and k is the absorption rate constant equal to which results from all interactions of NO2 with 1.3 X lo3 lite3 rnp2 mol-’ set-’ (Denbigh and the rat. This total effect is the quantity that is needed to assess the contribution of at- Prince, 1947, for a temperature of 40°C). We mospheric NOz to nitrate balance in the rat. have used this equation to show that the rate Green et af. (1981) and Witter et al. (1981) of reaction of NO2 with water in the respiobserved that rats excrete more urinary ni- ratory tract of the rat is too slow to account for our observations. Our calculation assumes trate than they ingest from food. We originally believed that the observations of these that the conditions of constant gas flow and authors might be explained by the exposure uniform NO2 concentration used in the model of their rats to ambient levels of NOz. How- system of Denbigh and Prince are applicable. These are simplifying assumptions since there ever, our observations have shown a urinary is in fact an NO2 concentration gradient along nitrate excess even for rats exposed to synthetic air on all 7 days (Fig. la). Thus our the length of the respiratory tract, and the gas flow is nonuniform. The actual NO2 concendata confirm the observations of these authors and support their conclusion that the source tration at any point within the respiratory tract of the observed nitrate excess is an endoge- will be lower than that assumed by our calnous mammalian process. culation, and so the true rate of reaction of For the purpose of using our data to suggest NO* with water will be even lower than that the mechanism of NO2 absorption and to es- derived from the equation. If we assume that timate the contribution of atmospheric NO2 the entire surface area of the respiratory tract to human nitrate/nitrite exposure, only the (0.44 m2; Tenney and Remmers, 1963) is covered with water, then for a rat inhaling NO2 effect of NO1 which results from respiratory tract interactions will be considered. We have at a concentration of 2.23 X lo-’ mol/liter (5 ppm), the equation predicts an absorption rate estimated that this effect alone contributes of 2.8 X lo-” mol/sec or 2.5 pmol/day. This about 6.7 pmol nitrate to rat urine per ppm NO* per 24-hr exposure (about 80% of the absorption rate would lead to a nitrate plus nitrite formation rate in the respiratory tract total effect of NO* on urinary nitrate). The linear relationship between NO2 con- of no more than 2.5 pmolfday which is far centration and urinary nitrate shows that the too small to account for our observations. rate-limiting step of NO2 absorption in the Also, since the slowness of this step would respiratory tract is first order with respect to make it rate limiting in our concentration NOz concentration. Both the linearity and the range, the amount of NO2 absorbed by this magnitude of this effect may be used as in- process and detected as urinary nitrate should direct evidence to show that reaction with re- be a parabolic function of the NO2 concentration. The observed linear function is furspiratory tract water could not be an important step in NO2 absorption. Denbigh and ther evidence that reaction with respiratory Prince ( 1947) showed that the rate of absorp- tract water is not an important interaction of NO2 at low NOz concentrations. tion of NO2 into aqueous solution to yield nitrate and nitrite is proportional to the gas Postlethwait and Mustafa ( 198 1) proposed phase concentration of its dimer, N204. Since that the major interaction of NO2 in the lung
290
SAUL AND ARCHER
is not with water, but with readily oxidizable tissue components such as proteins, lipids, glutathione, or amines to form nitrite. Their hypothesis, based on experiments with the isolated perfused rat lung, is indirectly sup ported by our results for whole animals. Since we have shown that 70% of the nitrite in blood is recovered as urinary nitrate under our conditions (Fig. 3), we may expect that 1.43 gmol of nitrite must enter the blood from the respiratory tract to account for each pmol of nitrate observed in the urine. Assuming that all nitrite which is formed in this way is absorbed into the blood, we estimate that 9.6 pmol of nitrite is formed in the respiratory tract of the rat per ppm NO* per day. We compared our observed respiratory tract nitrite formation rate with a theoretical maximum rate based on the assumption that 100% of the NO* in each breath is converted to nitrite. We used the resting minute volume measured by Guyton (1947) of 73 ml of air/ min for 113-g rats, and extrapolated this value on the basis of the 3/4 power of body weight (as suggested by Guyton) to obtain 106 ml of air/min for our 185-g rats. This value leads to a theoretical maximum nitrite formation rate of 8.4 pmol/ppm Nor/day. Our observed rate of 9.6 pmol nitrite/ppm NO&lay is slightly higher than this theoretical rate, but the discrepancy is not surprising, considering that we have not allowed for factors such as variations in activity on respiratory parameters. We conclude that the observed rate is similar to the theoretical maximum rate, indicating that most inhaled NOz is absorbed. This conclusion is consistent with the observation of Haagen-Smit et al. ( 1959) who found that if cigarette smoke containing NO and NOz was inhaled into the lungs, the exhaled smoke contained no detectable oxides of nitrogen. To estimate the contribution of atmospheric NO2 to the nitrite exposure of humans, we have extrapolated our rat data. Our calculation, based on an interspecies extrapolation, should not be considered to be precise, but may be useful as an order of mag-
nitude estimation. If we assume that humans absorb NOz with the same efficiency as the rat per milliliter of inhaled air, we may use the relative minute volumes as our extrapolation factor. A 68.5-kg man, with a resting minute volume of 8.73 liters/min (Guyton, 1947) will therefore breathe 82.4 times as much air as a 185-g rat. From our calculated rate of nitrite formation of 9.6 pmol/ppm NOJday for the rat, we estimate this effect in the human to be about 790 pmol/ppm NOz/ day. For a typical urban level of 0.1 ppm N02, this calculation predicts that 79 pmol or about 3.6 mg of nitrite is formed in the human respiratory tract per day. This level is greater than the average daily dietary intake of nitrite, which has been estimated by Hartman (1982) to be 0.6 mg/day. Atmospheric NOz , and its related oxides of nitrogen N203 and N204, may react with amines through a variety of mechanisms to form carcinogenic N-nitroso compounds (Challis and Kyrtopoulos, 1977). Nitrosation reactions which involve these oxides of nitrogen can take place at neutral pH and so could occur in the respiratory tract. Alternatively, NOz could interact with oxidizable tissue components in the respiratory tract to form nitrite, and this nitrite could act as a precursor for N-nitroso compounds. However, nitrosation reactions involving nitrite generally require acidic conditions (Archer, 1982). Smith et al. ( 1978) found that low concentrations of nitrite in the blood of pigs could survive for periods in excess of the time required for circulation within the vascular system. Thus, nitrite formed in the lung from NO2 could be transported to sites where conditions for nitrosation reactions involving this ion are more favorable. Also, some nitrite which is oxidized in the blood to nitrate could eventually be reduced back to nitrite at sites such as the oral cavity where endogenous microorganisms possess a nitrate-reductase activity (Tannenbaum et al., 1976). In these two ways, nitrite produced from inhaled NO* could act as a precursor for nitrosation reactions in viva at sites other than the lung.
URINARY
NITRATE
OF RATS EXPOSED
ACKNOWLEDGMENTS The authors thank Peter F. Zucker for his assistance. This research was supported by the Ontario Cancer Treatment and Research Foundation and the National Cancer Institute of Canada. Studentships from the Province of Ontario are gratefully acknowledged by R.L.S.
REFERENCES ARCHER, M. C. (1982). Hazards of nitrate, nitrite, and nitrosamines in human nutrition. In Nutritional Toxicology (J. N. Hathcock, ed.), Vol. 1, pp. 327-381. Academic Press, New York. CHALLIS, B. C., AND KYRTOPOULOS, S. A. (1977). Rapid formation of carcinogenic N-nitrosamines in aqueous alkaline solutions. Brit. J. Cancer 35, 693-696. DENBIGH, K. G., AND PRINCE, A. J. ( 1947). Kinetics of nitrous gas absorption in aqueous nitric acid. J. Chem. Sot. 790-80 1. FREEMAN, G., CRAN, S. C., STEPHENS, R. J., AND FuRIO%, N. J. (1968). Pathogenesis of the nitrogen dioxide-induced lesion in the rat lung. A review and presentation of new observations. Amer. Rev. Resp. Dis. 98,429-443. GOLDSTEIN, E., PEEK, N. F., PARKS, N. J., HINES, H. H., STEFFEY,E. P., AND TARKINGTON, B. (1977). Fate and distribution of inhaled nitrogen dioxide in rhesus monkeys. Amer. Rev. Resp. Dis. 115,403-412. GREENE, I., AND HIAIT, E. P. (1954). Behavior of the nitrate ion in the dog. Amer. J. Physiol. 176,463-467. GREEN, L. C., TANNENBAUM, S. R., AND GOLDMAN, P. (1981). Nitrate synthesis in the germfree and conventional rat. Science 212, 56-58. GUYTON, A. C. (1947). Measurement of the respiratory volumes of laboratory animals. Amer. J. Physiol. 150, 70-77. HAAGEN-SMIT, A. J., BRUNELLE, M. F., AND HARA, J. (1959). Nitrogen oxide content of smokes from different types of tobacco. J. Arch. Ind. Health 20, 399-W. HARTMAN, P. E. (1982). Nitrates and nitrites: Ingestion, pharmacodynamics, and toxicology. In Chemical Mutagens (F. J. deSerres and A. Hollaender, eds.), Vol. 7, pp. 21 l-294. Plenum, New York. HAWKSWORTH, G. M., AND HILL, M. J. (197 1). Bacteria and the N-nitrosation of secondary amines. Brit. J. Cancer 25, 520-526. IQBAL, Z. M., DAHL, K., AND EPSTEIN, S. S. (1980). Role of nitrogen dioxide in the biosynthesis of nitrosamines in mice. Science 208, 1475-1477. KATZ, M. ( 1976). Nitrogen compounds and oxidants. In
TO NO*
291
Air Pollution (A. C. Stem, ed.), 3rd ed Vol. 3, pp. 259305, Academic Press, New York. KOSAKA, H., IMAIZUMI, K., INAI, K., AND TWMA, I. ( 1979). Stoichiometry of the reaction of oxyhemoglobin with nitrite. B&him. Biophys. Acta 581, 184-l 88. MENZEL, D. B., AND MCCLELLAN, R. 0. (1980). Toxic responses of the respiratory system. In Casarett and DoullS Toxicology: The Basic Science of Poisons (J. Doull, C. D. Klaassen, and M. 0. Amdur, eds.), 2nd ed., MacMillan, New York. MIRWSH, S. S., ISSENBERG,P., AND SAMS, J. P. (1981). N-Nitrosomorpholine synthesis in rodents exposed to nitrogen dioxide and morpholine. In N-Nitroso Compounds (R. A. Scanlan and S. R. Tannenbaum, eds.), Vol. 174, pp. 18 1- 19 1. ACS Symposium Series, Amer. Chem. Sot., Washington, D.C. ODA, H., TSUBONE, H., SUZUKI, A., ICHINOSE,T., AND KUBOTA, K. (1981). Alterations of nitrite and nitrate concentrations in the blood of mice exposed to nitrogen dioxide. Environ. Res. 25, 294-301. PARKS, N. J., KROHN, K. A., MA~IS, C. A., CHASKO, J. H., GEIGER, K. R., GREGOR, M. E., AND PEEK, N. F. (198 1). Nitrogen-13-labeled nitrite and nitrate: Distribution and metabolism after intratracheal administration. Science 212, 58-6 1. POSTLETHWAIT, E. M., AND MUSTAFA, M. G. (1981). Fate of inhaled nitrogen dioxide in isolated perfused rat lung. J. Toxicol. Environ. Health 7, 86 l-872. ROEHM, J. N., HADLEY, J. C., AND MENZEL, D. B. (1971). Oxidation of unsaturated fatty acids by ozone and nitrogen dioxide. Arch. Environ. Health 23, 142-148. SAUL, R. L., KABIR, S. H., COHEN, Z., BRUCE, W. R., AND ARCHER, M. C. (1981). Reevaluation of nitrate and nitrite levels in the human intestine. Cancer Res. 41,2280-2283.
SEN, N. P., AND DONALDSON, B. (1978). Improved colorimetric method for determining nitrate and nitrite in foods. J. Assoc. Off Anal. Chem. 61, 1389-1394. Sm, P. L. R., WALTERS, C. L., AND WALKER, R. (1978). The transport of nitrite in blood. Biochem. Sot. Transact. 6, 665. SVORCOVA, S., AND TAUT, V. (1971). Arteriovenous differences in the nitrite and nitrate ion concentrations in rabbits after inhalation of nitrogen oxides. Cesk. Hyg. 16, 71-76. (English translation in Chem. Abstr. 75, 33253m) TANNENBAUM, S. R., WEISMAN, M., AND FRT, D. (1976). The effect of nitrate intake on nitrite formation in human saliva. Food Cosmet. Toxicol. 14, 549-552. TENNEY, S. M., AND REMMERS, J. E. ( 1963). Comparative quantitative morphology of the mammalian lung: Diffusing area. Nature @ondun) 197, 54-56. WITTER, J. P., GATLEY, S. J., AND BALISH, E. (198 1). Evaluation of nitrate synthesis by intestinal microorganisms in vivo. Science 213, 449-450.
TOXlCOLOCYANDAPPLIEDPHARMACOLOGY
Nitrate Formation
in Rats Exposed to Nitrogen
Dioxide
ROBERT L. SAUL AND MICHAEL C. ARCHER’ Department
of Medical Biophysics, 500 Sherbourne Street,
Received
University of Toronto, Toronto, Ontario M4X
July 16, 1982; accepted
October
Ontario Cancer IK9, Canada
Institute,
25, 1982
Nitrate Formation in Rats Exposed to Nitrogen Dioxide. SAUL, R. L., AND ARCHER, M. C. (1983). Toxicol. Appl. Pharmacol. 67, 284-291. Sprague-Dawley rats exposed to atmospheres containing low levels of nitrogen dioxide (Nq) for 24 hr had increased levels of nitrate in their mine on the day of exposure and on the 3 subsequent days. The total increase in urinary nitrate was linearly related to the nitrogen dioxide concentration administered. We recovered in urine 8.4 + 1. I pmol nitrate/ppm N02/24-hr exposure (slope + 95% confidence limits) for 185-g rata. Both the linearity and magnitude of this effect imply that reaction with respiratory tract water is not a major pathway of NOr absorption in the lung. Instead, our observations support the hypothesis that the major interaction of NOz in the lung is with readily oxidizable tissue components to form nitrite. We estimate that 9.6 pmol of nitrite is formed in the respiratory tract ofthe rat per ppm NO1 per 24-hr exposure. We also estimate that humans breathing air containing 0.1 ppm NO2 have about 3.6 mg of nitrite formed in their respiratory tract per day.
Nitrogen dioxide (NOz) is a common atmospheric pollutant that arises from a wide range of high-temperature combustion processes. It is a lung irritant, producing a pathological condition similar to human emphysema in experimental animals (Freeman et al., 1968). The chemical mechanisms of NO2 toxicity are not understood, but are probably related to the oxidative nature of this agent (Roehm et al., 197 1). Recently, NO2 has also been studied as a potential precursor of carcinogenic N-nitroso compounds which may be formed in vivo (Iqbal et al., 1980; Mirvish et al., 1981). The fate of inhaled NO2 was studied by Svorcova and Kaut ( 197 1) who exposed rabbits to a single high concentration of NO* for 4 hr and observed elevated levels of both nitrate and nitrite in the blood and urine. More recently, Oda et al. (198 1) found that the nitrate and nitrite levels in the blood of mice rose quickly to constant values during expo-
sure to N02, and that afer exposure, the blood levels declined with half lives of about 1 hr for nitrate, and a few minutes for nitrite. The short half-life of nitrite was attributed to its reaction with oxyhemoglobin to form methemoglobin and nitrate. Goldstein et al. (1977) studied the distribution of inhaled NO2 in rhesus monkeys with radioactive i3N02. They demonstrated that 50 to 60% of inspired NO2 was retained during quiet respiration and confirmed that N02, or its chemical reaction products, is transported to distant sites via the bloodstream. These authors suggested that NO2 reacts with water in the lung to form nitric and nitrous acids (conjugate acids of the nitrate and nitrite ions) and that the toxicity of NO2 is due to the reaction of these acids with pulmonary and extrapulmonary tissues. Postlethwait and Mustafa (198 1) studied the fate of NO2 in isolated perfused rat lungs. Surprisingly, for perfusate containing no erythrocytes they found that nitrite, but not nitrate, ’ To whom requests for reprints should be addressed. was produced in lungs exposed to NO*. The
0041-008X/83/020284-08$03.00/0 Copyright 0 1983 by Academic Press. Inc. All rights of reproduction in any form reserved.
284
URINARY
NITRATE
OF RATS EXPOSED
lack of nitrate indicated that the primary reaction of NO2 in the lung is not with water, since this reaction is well known to yield both nitrate and nitrite as products (Denbigh and Prince, 1947). Postlethwait and Mustafa (198 1) have suggested that instead of water, NO2 reacts with readily oxidizable tissue components to yield only nitrite. Intratracheal administration of [“Nlnitrate and [ i3N]nitrite to rodents has shown that both these ions are rapidly absorbed from the respiratory tract into the blood (Parks et al., 1981). Nitrate in the blood is known to be excreted rapidly and in high yield as urinary nitrate (Greene and Hiatt, 1954; Hawksworth and Hill, 197 1). Nitrite in the blood is quantitatively oxidized by oxyhemoglobin to form nitrate (Kosaka et al., 1979) and so should also be recovered as urinary nitrate. Thus, the formation of either nitrate or nitrite from NO2 in the lung could be detected by observing the nitrate levels in the urine during and after exposure to NO*. In the present study, we have used this approach to determine the extent of nitrate and nitrite formation in rats exposed to various concentrations of NO2 and have used our data to suggest the mechanism of this process.
METHODS An airtight glove box (Labconco Corp., Kansas City, MO.; internal volume approximately 350 liters), equipped with an air lock, was modified for use as a controlled atmosphere chamber by disconnecting its air circulation tubing and using the two connection ports for the inflow and outflow of gases. The NO* levels inside the chamber were checked periodically by sampling the gas inside the glove box or in the exit tube. The two sampling sites yielded equivalent results. The NO2 concentrations were determined by the Saltzmann method with either the syringe technique or the fritted bubbler technique (Katz, 1976). Three metabolic cages (Nalgene Co., Rochester, N.Y.) were placed inside the chamber, and for each experiment, three male Sprague-Dawley rats, (Charles River Canada, Inc., La Prairie, Quebec) initially 36 days old and weighing 162 f 8 g (i & SD), were individually housed in the cages for a period of 7 days. With he exception of the fourth day, the inlet gas was synthetic air (Liquid Carbonic of Canada, Ltd., Scarborough, Ontario) at a
TO NO,
285
flow rate of 2 liters/min. We determined that this gas contained no NO* (detection limit 0.01 ppm). On the fourth day of each experiment the animals were exposed to atmospheres containing NO*. A mixture of 3 14 ppm (by volume) NO, in air from a gas stock tank (Liquid Carbonic of Canada Ltd., concentration self-determined) was dynamically combined with synthetic air to give the desired dilution at an overall flow rate of 5 liters/min. During the first 2 hr of the 24-hr exposure period, the NOr concentration of the inlet gas was set 60% higher than the desired level to achieve a rapid increase in the chamber NO, concentration. Once the desired chamber concentration was reached, the inlet gas concentration was reduced and periodically adjusted so that the chamber NO2 concentration remained constant within +lO% over the next 20 hr. During the last 2 hr of the exposure period, the chamber NO* concentration was returned to zero by purging the chamber with synthetic air. The average NO2 concentration was calculated by integrating the measured NO2 concentrations over the entire 24-hr exposure period. Six treatment groups (three rats per group) were exposed to 24hr average NOz concentrations of zero, 1.2, 3.4, 4.1, 5.6, and 8.8 ppm. On the 3 days before and 3 days after the exposure day, the rats were allowed unlimited amounts of deionized water and a casein and corn starch-based, low-nitrate, low-nitrite diet (Biomix 1435, Bioserve Inc., Frenchtown, N.J.). Using the method of Sen and Donaldson (1978), we measured the nitrate content in this diet as 0.085 pmol/ g, and no detectable nitrite with a detection limit of 0.005 pmol/g. Since we found that NOr reacts with this food to form nitrate and nitrite, the animals were given only water on the day of NOr exposure. Water bottles and food containers were changed daily, and food consump tion was determined. Urine was collected for each 24-hr period. Ten milliliters of 0.6 M NaOH solution was added before use to each urine collection container to prevent nitrate or nitrite destruction or synthesis due to contaminating fecal microorganisms (Saul et al., 198 1). To minimize the accumulation of feces on the collection funnel, separating cone, and urine ring, the metabolic cages were disassembled, washed inside the glove box with deionized water, and reassembled each day. The alkaline, preserved urine samples were transferred to lOO-ml volumetric flasks, diluted to volume with deionized water, stored for no longer than 7 days at 4”C, and analyzed as described below. It was shown by repeated analyses that the nitrate and nitrite levels in these samples were stable for several months when stored under these conditions. Urinary nitrate and nitrite were analyzed by a method based on that of Sen and Donaldson ( 1978). Twenty milliliters of the diluted alkaline urine sample was heated at 60°C for 10 min. Two milliliters of 0.42 M ZnSO, solution was added to precipitate protein, and the mixture was heated a further 10 min with occasional swirling. After cooling to room temperature, the mixture was suc-
286
SAUL AND ARCHER
tion filtered through a disposable tilter unit (0.45 pm, Nalgene Co.; unit was prewashed with 300 ml of deionized water to remove contaminating nitrate), and the filtrate was transferred to a 50-ml volumetric Bask. The residue on the ftlter was washed with 20 ml water, and the wash was combined with the filtrate and diluted to volume. Two lO-ml portions of this solution were assayed for nitrate and nitrite as described by Sen and Donaldson (1978) by cadmium reduction of nitrate to nitrite and the Griess reaction for nitrite detection. The Greiss reaction was always performed at 20°C since the azo dye yield is temperature dependent. Some urine samples were spiked with standard solutions of nitrate or nitrite, and a standard curve was constructed to relate the resulting azo dye absorbances with the amounts of these ions in the urine. Recoveries of both nitrate and nitrite were greater than 95%. Based on multiple analyses, we estimate that our individual measurements of nitrate and nitrite in urine were accurate to within 0.5 pmo1/24-hr urine sample. The detection limit was 0.5 pmol/sample for both nitrate and nitrite.
RESULTS Average daily food consumption for the six groups of rats treated with NO2 (not including Day 4 when all animals were starved) was 18.7 + 3.5 g (2 + SD). This value represents an average dietary nitrate intake of 1.6 pmol/ day. Average weight gain over the 7&y experimental period for these rats was 45 f 6 g (2 + SD). There were no significant NOzdependent differences in food consumption or weight gain between treatment groups. All rats weighed approximately 185 g on the day of NOz exposure. Figure la shows the average daily urinary nitrate output of the three rats which were exposed to zero ppm NOz (synthetic air) on Day 4. No nitrite (detection limit 0.5 gmol/ day) was found in any of these urines. The reduced nitrate output on Day 4 was a result of their fast. On all 7 days of this experiment, daily urinary nitrate output exceeded dietary nitrate intake by 3.2 * 0.2 firno1 (J? +_ SE, n = 2 1). This observation is consistent with those ofGreenetal.(1981)andWitteretal.(1981) who also reported excess urinary nitrate for rats consuming low-nitrate diets. Figure lb shows the average daily urinary nitrate output for a typical experiment in-
+
,
lzJzlz 2
3
L 4
I-
5
Ill 6
7
FIG. 1. Average daily urinary nitrate levels for rata exposed to NO, on Day 4. Bars represent X + SE for three rats. Average 24-hr NO, concentrations for these two experiments were (a) zero ppm and (b) 4.1 ppm.
volving three rats which were exposed to NOz on Day 4, in this case at an average concentration of 4.1 ppm. The levels shown in Fig. lb are significantly higher than those in Figure la for Days 4 through 7. Figure 2 shows the 4-day total nitrate excretion for Days 4 through 7 as a function of NO* exposure concentration. Regression analysis indicates that these data are best fit with a straight line. The least-squares fit shown in Fig. 2 has a slope of 8.4 f 1.1 pmol nitrate/ppm NOz/day (slope + 95% confidence limits, correlation coefficient 0.970). Some nitrite was also found in the urine of animals exposed to NOz, but only in samples collected from the day of exposure. As will be discussed below, this nitrite was not due to urinary nitrite excretion but was a result of the in vitro reaction of urine with NO*. In addition to interactions within the respiratory tract, NO2 could also react with the rats’ food, drinking water, and body hair to form nitrate or nitrite. These reaction products could then be ingested by the animal
URINARY
90
NITRATE
OF RATS EXPOSED
TO NO*
2
4 dioxide
0
287
-
00 -
0
I
3 Nitrogen
5 6 7 concentrotion(ppm)
9
IO
FIG. 2. Total urinary nitrate output for Days 4 through 7 for rats exposed to NO1 for 24 hr on Day 4. Each symbol represents the total for one rat.
through the processes of eating, drinking, or licking and excreted as nitrate in the urine. Nitrogen dioxide could also react with the plastic walls of the urine collection apparatus or with the urine itself, either in the collection container, or while dripping along the walls of the collection funnel. All of these reactions could potentially lead to an overestimation of the contribution of inhaled NO2 to urinary nitrate. Nose- or head-only exposure to NOz might have circumvented this problem. Such exposure systems, however, suffer from the disadvantage that the exposure time must normally be less than 1 hr (Menzel and McClellan, 1980). Since much longer exposure times were necessary to produce measurable increases in urinary nitrate at low NOz concentrations, the nose- or head-only approach was impractical. Also, as will be seen in the discussion, we were interested in the effect of NO2 on the whole animal for the purpose of studying nitrate balance. To assess the potential problems of our whole-body exposure method, several experiments were performed which are described below.
The powdered food used in our experiments reacted readily with NO1 to form both nitrate and nitrite. Therefore, all animals were fasted on the day of NO2 exposure to prevent ingestion of this nitrate and nitrite which would ultimately appear in urine as nitrate. Deionized water reacts extremely slowly with NOz at the concentrations used in this study to form small amounts of nitrate and nitrite. Since water was administered by an inverted drinking bottle, contact of the water with chamber air was minimal. The animals were therefore allowed water during the NOa exposure period. Analysis of water from the drinking bottles after the higest NO2 concentration was used (8.8 ppm over 24 hr) showed no detectable nitrate or nitrite (detection limit 0.0 1 pmol/ml). Although water reacts slowly with low concentrations of NOz, urine was found to react more rapidly to form nitrite. In a number of experiments designed to examine this reaction, neutral or alkaline urine exposed to NO* yielded increased levels of nitrite, but no detectable increase in nitrate. Apparently it is
288
SAUL AND ARCHER
not the water, but rather, oxidizable chemicals present in urine which reduce NOz to nitrite. Since this in vitro reaction of NO2 with urine does not produce nitrate, it did not lead to an overestimation of the levels of nitrate in our rat urines for Day 4. Although our nitrate determinations were based on the subtraction of the nitrite concentration from the nitrate plus nitrite total (Sen and Donaldson, 1978), the levels of nitrite in urines from Day 4 were small enough that they did not reduce the accuracy of this subtraction. A clean metabolic cage, without a rat inside, was exposed to 6.7 ppm NO* for 24 hr and then rinsed with deionized water. The rinse water was collected and analyzed for nitrate and nitrite. Trace quantities (less than 1 pmol) of both ions were found but these were too small to contribute significantly to our observations of urinary nitrate. Several experiments were performed to test the extent of NO;? reaction with the animals’ hair to produce nitrate or nitrite which could be ingested as a result of licking. On the day of NO2 exposure (Day 4), rats were placed in a restraining device which prevented them from licking themselves but still allowed access to their drinking water. Unrestrained rats exposed to the same atmosphere were treated as described under Methods and served as controls. After the NOz exposure period, the restrained rats were released and immediately rinsed with deionized water to remove nitrate or nitrite from their body exteriors. The differences in total urinary nitrate output for Days 4 through 7 between restrained and unrestrained animals were then used to assess the contribution of licking to the total effect. We estimate that about 20% of the observed slope in Fig. 2 is a result of licking, while about 80% of this slope is a result of interactions of NO* within the respiratory tract. Additional data used to support this estimation came from our analysis of the water used to rinse NOz-exposed animals which showed the presence of nitrite, but not nitrate. On a molar basis, the amount of nitrite found on the animals’ hair was approximately equal to
60
0
IO
20 Injected
d 30 40 nitrite(pmol)
I 50
J 60
FIG. 3. Total urinary nitrate output for Days 4 through 7 for rats injected with sodium nitrite solution on Day 4. Each symbol represents the total for one rat.
the difference in urinary nitrate between the restrained and unrestrained rats. As will be seen in the Discussion, the linear relationship between urinary nitrate and NO2 concentration suggests that the major interaction of NO2 in the respiratory tract is the formation of nitrite which is absorbed into the blood and excreted in the urine as nitrate. To determine the amount of nitrite in the blood which is recovered as nitrate in the urine, we injected rats iv with nitrite solutions and subsequently observed the urinary nitrate levels. The experimental protocol used for this purpose was identical to that used in the NO2 experiments except that on Day 4, the animals were exposed to synthetic air and were given an injection of nitrite solution at the midpoint of this day. Zero, 15, 30, and 45 pm01 of sodium nitrite in 0.3 ml of isotonic saline were administered via a tail vein. The nitrate excretion profile for the rats injected with nitrite was very similar to that of the rats exposed to NO*, with most of the excess nitrate appearing on Days 4 and 5. No nitrite was found in any of the urines of these animals. Figure 3 shows the linear relationship between total nitrate excreted on Days 4 through 7 and the injected nitrite dose. The slope of the linear regression line is 0.70 rf: 0.10 pm01 nitrate/pm01 nitrite (slope + 95%
URINARY
confidence 0.994).
limits,
correlation
DISCUSSION
NITRATE
OF RATS EXPOSED
coefficient
TO NO2
289
the concentration of the dimer is proportional to the square of the NO2 concentration, the rate of absorption NOz into water is given by dN02 = kA[N0212, dt
Our finding of a urinary nitrate increase of 8.4 pmol/ppm NO*/day is the total increase where A is the area of the gas-water interface, and k is the absorption rate constant equal to which results from all interactions of NO2 with 1.3 X lo3 lite3 rnp2 mol-’ set-’ (Denbigh and the rat. This total effect is the quantity that is needed to assess the contribution of at- Prince, 1947, for a temperature of 40°C). We mospheric NOz to nitrate balance in the rat. have used this equation to show that the rate Green et af. (1981) and Witter et al. (1981) of reaction of NO2 with water in the respiobserved that rats excrete more urinary ni- ratory tract of the rat is too slow to account for our observations. Our calculation assumes trate than they ingest from food. We originally believed that the observations of these that the conditions of constant gas flow and authors might be explained by the exposure uniform NO2 concentration used in the model of their rats to ambient levels of NOz. How- system of Denbigh and Prince are applicable. These are simplifying assumptions since there ever, our observations have shown a urinary is in fact an NO2 concentration gradient along nitrate excess even for rats exposed to synthetic air on all 7 days (Fig. la). Thus our the length of the respiratory tract, and the gas flow is nonuniform. The actual NO2 concendata confirm the observations of these authors and support their conclusion that the source tration at any point within the respiratory tract of the observed nitrate excess is an endoge- will be lower than that assumed by our calnous mammalian process. culation, and so the true rate of reaction of For the purpose of using our data to suggest NO* with water will be even lower than that the mechanism of NO2 absorption and to es- derived from the equation. If we assume that timate the contribution of atmospheric NO2 the entire surface area of the respiratory tract to human nitrate/nitrite exposure, only the (0.44 m2; Tenney and Remmers, 1963) is covered with water, then for a rat inhaling NO2 effect of NO1 which results from respiratory tract interactions will be considered. We have at a concentration of 2.23 X lo-’ mol/liter (5 ppm), the equation predicts an absorption rate estimated that this effect alone contributes of 2.8 X lo-” mol/sec or 2.5 pmol/day. This about 6.7 pmol nitrate to rat urine per ppm NO* per 24-hr exposure (about 80% of the absorption rate would lead to a nitrate plus nitrite formation rate in the respiratory tract total effect of NO* on urinary nitrate). The linear relationship between NO2 con- of no more than 2.5 pmolfday which is far centration and urinary nitrate shows that the too small to account for our observations. rate-limiting step of NO2 absorption in the Also, since the slowness of this step would respiratory tract is first order with respect to make it rate limiting in our concentration NOz concentration. Both the linearity and the range, the amount of NO2 absorbed by this magnitude of this effect may be used as in- process and detected as urinary nitrate should direct evidence to show that reaction with re- be a parabolic function of the NO2 concentration. The observed linear function is furspiratory tract water could not be an important step in NO2 absorption. Denbigh and ther evidence that reaction with respiratory Prince ( 1947) showed that the rate of absorp- tract water is not an important interaction of NO2 at low NOz concentrations. tion of NO2 into aqueous solution to yield nitrate and nitrite is proportional to the gas Postlethwait and Mustafa ( 198 1) proposed phase concentration of its dimer, N204. Since that the major interaction of NO2 in the lung
290
SAUL AND ARCHER
is not with water, but with readily oxidizable tissue components such as proteins, lipids, glutathione, or amines to form nitrite. Their hypothesis, based on experiments with the isolated perfused rat lung, is indirectly sup ported by our results for whole animals. Since we have shown that 70% of the nitrite in blood is recovered as urinary nitrate under our conditions (Fig. 3), we may expect that 1.43 gmol of nitrite must enter the blood from the respiratory tract to account for each pmol of nitrate observed in the urine. Assuming that all nitrite which is formed in this way is absorbed into the blood, we estimate that 9.6 pmol of nitrite is formed in the respiratory tract of the rat per ppm NO* per day. We compared our observed respiratory tract nitrite formation rate with a theoretical maximum rate based on the assumption that 100% of the NO* in each breath is converted to nitrite. We used the resting minute volume measured by Guyton (1947) of 73 ml of air/ min for 113-g rats, and extrapolated this value on the basis of the 3/4 power of body weight (as suggested by Guyton) to obtain 106 ml of air/min for our 185-g rats. This value leads to a theoretical maximum nitrite formation rate of 8.4 pmol/ppm Nor/day. Our observed rate of 9.6 pmol nitrite/ppm NO&lay is slightly higher than this theoretical rate, but the discrepancy is not surprising, considering that we have not allowed for factors such as variations in activity on respiratory parameters. We conclude that the observed rate is similar to the theoretical maximum rate, indicating that most inhaled NOz is absorbed. This conclusion is consistent with the observation of Haagen-Smit et al. ( 1959) who found that if cigarette smoke containing NO and NOz was inhaled into the lungs, the exhaled smoke contained no detectable oxides of nitrogen. To estimate the contribution of atmospheric NO2 to the nitrite exposure of humans, we have extrapolated our rat data. Our calculation, based on an interspecies extrapolation, should not be considered to be precise, but may be useful as an order of mag-
nitude estimation. If we assume that humans absorb NOz with the same efficiency as the rat per milliliter of inhaled air, we may use the relative minute volumes as our extrapolation factor. A 68.5-kg man, with a resting minute volume of 8.73 liters/min (Guyton, 1947) will therefore breathe 82.4 times as much air as a 185-g rat. From our calculated rate of nitrite formation of 9.6 pmol/ppm NOJday for the rat, we estimate this effect in the human to be about 790 pmol/ppm NOz/ day. For a typical urban level of 0.1 ppm N02, this calculation predicts that 79 pmol or about 3.6 mg of nitrite is formed in the human respiratory tract per day. This level is greater than the average daily dietary intake of nitrite, which has been estimated by Hartman (1982) to be 0.6 mg/day. Atmospheric NOz , and its related oxides of nitrogen N203 and N204, may react with amines through a variety of mechanisms to form carcinogenic N-nitroso compounds (Challis and Kyrtopoulos, 1977). Nitrosation reactions which involve these oxides of nitrogen can take place at neutral pH and so could occur in the respiratory tract. Alternatively, NOz could interact with oxidizable tissue components in the respiratory tract to form nitrite, and this nitrite could act as a precursor for N-nitroso compounds. However, nitrosation reactions involving nitrite generally require acidic conditions (Archer, 1982). Smith et al. ( 1978) found that low concentrations of nitrite in the blood of pigs could survive for periods in excess of the time required for circulation within the vascular system. Thus, nitrite formed in the lung from NO2 could be transported to sites where conditions for nitrosation reactions involving this ion are more favorable. Also, some nitrite which is oxidized in the blood to nitrate could eventually be reduced back to nitrite at sites such as the oral cavity where endogenous microorganisms possess a nitrate-reductase activity (Tannenbaum et al., 1976). In these two ways, nitrite produced from inhaled NO* could act as a precursor for nitrosation reactions in viva at sites other than the lung.
URINARY
NITRATE
OF RATS EXPOSED
ACKNOWLEDGMENTS The authors thank Peter F. Zucker for his assistance. This research was supported by the Ontario Cancer Treatment and Research Foundation and the National Cancer Institute of Canada. Studentships from the Province of Ontario are gratefully acknowledged by R.L.S.
REFERENCES ARCHER, M. C. (1982). Hazards of nitrate, nitrite, and nitrosamines in human nutrition. In Nutritional Toxicology (J. N. Hathcock, ed.), Vol. 1, pp. 327-381. Academic Press, New York. CHALLIS, B. C., AND KYRTOPOULOS, S. A. (1977). Rapid formation of carcinogenic N-nitrosamines in aqueous alkaline solutions. Brit. J. Cancer 35, 693-696. DENBIGH, K. G., AND PRINCE, A. J. ( 1947). Kinetics of nitrous gas absorption in aqueous nitric acid. J. Chem. Sot. 790-80 1. FREEMAN, G., CRAN, S. C., STEPHENS, R. J., AND FuRIO%, N. J. (1968). Pathogenesis of the nitrogen dioxide-induced lesion in the rat lung. A review and presentation of new observations. Amer. Rev. Resp. Dis. 98,429-443. GOLDSTEIN, E., PEEK, N. F., PARKS, N. J., HINES, H. H., STEFFEY,E. P., AND TARKINGTON, B. (1977). Fate and distribution of inhaled nitrogen dioxide in rhesus monkeys. Amer. Rev. Resp. Dis. 115,403-412. GREENE, I., AND HIAIT, E. P. (1954). Behavior of the nitrate ion in the dog. Amer. J. Physiol. 176,463-467. GREEN, L. C., TANNENBAUM, S. R., AND GOLDMAN, P. (1981). Nitrate synthesis in the germfree and conventional rat. Science 212, 56-58. GUYTON, A. C. (1947). Measurement of the respiratory volumes of laboratory animals. Amer. J. Physiol. 150, 70-77. HAAGEN-SMIT, A. J., BRUNELLE, M. F., AND HARA, J. (1959). Nitrogen oxide content of smokes from different types of tobacco. J. Arch. Ind. Health 20, 399-W. HARTMAN, P. E. (1982). Nitrates and nitrites: Ingestion, pharmacodynamics, and toxicology. In Chemical Mutagens (F. J. deSerres and A. Hollaender, eds.), Vol. 7, pp. 21 l-294. Plenum, New York. HAWKSWORTH, G. M., AND HILL, M. J. (197 1). Bacteria and the N-nitrosation of secondary amines. Brit. J. Cancer 25, 520-526. IQBAL, Z. M., DAHL, K., AND EPSTEIN, S. S. (1980). Role of nitrogen dioxide in the biosynthesis of nitrosamines in mice. Science 208, 1475-1477. KATZ, M. ( 1976). Nitrogen compounds and oxidants. In
TO NO*
291
Air Pollution (A. C. Stem, ed.), 3rd ed Vol. 3, pp. 259305, Academic Press, New York. KOSAKA, H., IMAIZUMI, K., INAI, K., AND TWMA, I. ( 1979). Stoichiometry of the reaction of oxyhemoglobin with nitrite. B&him. Biophys. Acta 581, 184-l 88. MENZEL, D. B., AND MCCLELLAN, R. 0. (1980). Toxic responses of the respiratory system. In Casarett and DoullS Toxicology: The Basic Science of Poisons (J. Doull, C. D. Klaassen, and M. 0. Amdur, eds.), 2nd ed., MacMillan, New York. MIRWSH, S. S., ISSENBERG,P., AND SAMS, J. P. (1981). N-Nitrosomorpholine synthesis in rodents exposed to nitrogen dioxide and morpholine. In N-Nitroso Compounds (R. A. Scanlan and S. R. Tannenbaum, eds.), Vol. 174, pp. 18 1- 19 1. ACS Symposium Series, Amer. Chem. Sot., Washington, D.C. ODA, H., TSUBONE, H., SUZUKI, A., ICHINOSE,T., AND KUBOTA, K. (1981). Alterations of nitrite and nitrate concentrations in the blood of mice exposed to nitrogen dioxide. Environ. Res. 25, 294-301. PARKS, N. J., KROHN, K. A., MA~IS, C. A., CHASKO, J. H., GEIGER, K. R., GREGOR, M. E., AND PEEK, N. F. (198 1). Nitrogen-13-labeled nitrite and nitrate: Distribution and metabolism after intratracheal administration. Science 212, 58-6 1. POSTLETHWAIT, E. M., AND MUSTAFA, M. G. (1981). Fate of inhaled nitrogen dioxide in isolated perfused rat lung. J. Toxicol. Environ. Health 7, 86 l-872. ROEHM, J. N., HADLEY, J. C., AND MENZEL, D. B. (1971). Oxidation of unsaturated fatty acids by ozone and nitrogen dioxide. Arch. Environ. Health 23, 142-148. SAUL, R. L., KABIR, S. H., COHEN, Z., BRUCE, W. R., AND ARCHER, M. C. (1981). Reevaluation of nitrate and nitrite levels in the human intestine. Cancer Res. 41,2280-2283.
SEN, N. P., AND DONALDSON, B. (1978). Improved colorimetric method for determining nitrate and nitrite in foods. J. Assoc. Off Anal. Chem. 61, 1389-1394. Sm, P. L. R., WALTERS, C. L., AND WALKER, R. (1978). The transport of nitrite in blood. Biochem. Sot. Transact. 6, 665. SVORCOVA, S., AND TAUT, V. (1971). Arteriovenous differences in the nitrite and nitrate ion concentrations in rabbits after inhalation of nitrogen oxides. Cesk. Hyg. 16, 71-76. (English translation in Chem. Abstr. 75, 33253m) TANNENBAUM, S. R., WEISMAN, M., AND FRT, D. (1976). The effect of nitrate intake on nitrite formation in human saliva. Food Cosmet. Toxicol. 14, 549-552. TENNEY, S. M., AND REMMERS, J. E. ( 1963). Comparative quantitative morphology of the mammalian lung: Diffusing area. Nature @ondun) 197, 54-56. WITTER, J. P., GATLEY, S. J., AND BALISH, E. (198 1). Evaluation of nitrate synthesis by intestinal microorganisms in vivo. Science 213, 449-450.