Vol. 16, No. 1 Printed in U.S.A.
INFECTION AND IMMUNITY, Apr. 1977, p. 240-248 Copyright ©D 1977 American Society for Microbiology
Effects of Cigarette Smoke Components on In Vitro Chemotaxis of Human Polymorphonuclear Leukocytes RAYMOND B. BRIDGES,* J. H. KRAAL, L. J. T. HUANG, AND M. B. CHANCELLOR Department of Oral Biology and Periodontology, College of Dentistry, University of Kentucky, Lexington, Kentucky 40506 Received for publication 4 November 1976
Some ciliostatic components of cigarette smoke were studied as inhibitors of in vitro chemotaxis of human polymorphonuclear leukocytes (PMNs). In comparison to their concentration in an inhibitory level of cigarette smoke, the unsaturated aldehydes acrolein and crotonaldehyde were the most potent inhibitors, whereas nicotine, cyanide, acetaldehyde, and furfural were the next strongest inhibitors. In contrast, sulfide, propionaldehyde, butyraldehyde, and the phenols (phenol and o-, m-, and p-cresol) were relatively weak inhibitors of PMN chemotaxis. Acrolein and crotonaldehyde mimicked whole cigarette smoke in their effects on PMNs by not causing loss of PMN viability, yet their effects were prevented by the addition of cysteine. On the other hand, addition of nicotine, cyanide, acetaldehyde, and furfural to PMN suspensions resulted in a limited loss of cellular viabilities, and their effects on PMNs were not prevented by cysteine. Of the tested components, only cyanide significantly altered PMN glucose metabolism by increasing carbon flow via the glycolytic and hexose monophosphate pathways in a manner similar to that observed with whole cigarette smoke. The results of this study suggest that the unsaturated aldehydes, including acrolein and crotonaldehyde, are major contributors to the inhibitory properties of cigarette smoke. The inhibitory effects of these unsaturated aldehydes are probably due to a direct interaction of these oxidants and/or thiol-alkylating agents with PMNs, yet the glucose metabolism of these cells is unaffected. One interpretation of these data is that PMN chemotaxis is dependent upon particular cellular proteins containing one or more essential thiol group(s) but that these proteins are unrelated to glucose metabolism.
exclusion or enzyme (i.e., lactic acid dehydrogenase [LDH] or ,8-glucuronidase) release. Furthermore, these tobacco smoke products did not inhibit glucose metabolism in PMNs, but rather increased (twofold) glucose catabolism via both the glycolytic and hexose monophosphate pathways. Finally, cysteine was shown to afford complete protection against the inhibition of PMN chemotaxis by the water-soluble fraction but only partial protection against the inhibitory effects of whole smoke and the gas phase of smoke. In the present study, we have tested components of cigarette smoke previously shown to cause cilia toxicity (26) for their ability to inhibit PMN chemotaxis. The aim of this study was to determine which of these components was most deleterious to PMN function.
The function of human polymorphonuclear leukocytes (PMNs) is severely affected by cigarette smoke both in vivo and in vitro. The motility and oxygen consumption of human oral PMNs from smokers were inhibited as compared to oral PMNs from nonsmokers (14). Also, the chemotaxis of peripheral blood leukocytes has been shown to be depressed in subjects who smoke as compared either to the same subjets abstaining from smoking or to nonsmokers (23). In a more recent study (10), the water-soluble fraction of cigarette smoke, whole tobacco smoke, and the gas phase of smoke were shown to be potent inhibitors of in vitro PMN chemotaxis. Whole smoke was more inhibitory than gas phase, suggesting the possibility that the inhibitory components of smoke were associated with both the particulate and gas phase fractions. However, none of the tested tobacco smoke products at concentrations inhibitory to chemotaxis were found to have a cytotoxic effect on PMNs, as measured by trypan blue
MATERIALS AND METHODS Isolation of PMNs. PMNs were isolated from peripheral blood of nonsmoking, healthy, male volunteers who were fasted at least 12 h before venipunc240
VOL. 16, 1977
EFFECTS OF CIGARETTE SMOKE COMPONENTS
ture. PMNs were separated from lym-iphocytes and erythrocytes by dextran sedimentation of heparinized whole blood (11), Hypaque-Ficoll gradient centrifugation (9), and hypotonic saline lysis of residual erythrocytes (11). PMNs were resuspended in Gey medium containing 2% bovine serum albumin and bicarbonate to give 3 x 106 PMNs per ml. This procedure produced PMN populations of greater than 98% purity with greater than 98% viability as measured by trypan blue exclusion (25). Pure substances of cigarette smoke. The compound to be tested was dissolved in Gey medium (100 AD) or dimethyl sulfoxide (10 ,lI) and added to PMN suspensions (1 ml), each in amounts to give the final concentrations indicated. An equivalent amount of solvent was added to control PMN suspensions. These solutions were freshly prepared on the day of the experiment, and the pH of the compounds prepared in Gey medium was adjusted to 7.4, if necessary. Dimethyl sulfoxide was an acceptable solvent for some compounds, since it did not affect in vitro PMN chemotaxis, confirming the previous observation of Borel (6). Cigarette smoke components were added to PMN suspensions in final concentrations that were previously shown to result in cilia toxicity (26). If this concentration of agent was found either not to affect or to inhibit PMN chemotaxis, the concentration of agent was then increased or decreased in 10-fold increments, respectively. In this manner, the range of agent concentration necessary to produce PMN chemotaxis inhibition was approximated. From these data, components for further study were selected on the basis of their relative concentration in tobacco smoke as compared to the concentration inhibitory to chemotaxis and their limited cytotoxic effects on PMNs. With the demonstration of minimal cytotoxic effects, the effects of each component on PMN glucose metabolism were then studied. Finally, cysteine was tested as a protective agent against the inhibitory action of each of these select components, since cysteine should provide at least some protection against the inhibitory effects of these components, if indeed they mimic the effects of whole tobacco smoke (10). Chemotaxis assay. Chemotaxis experiments were done using a modified Boyden chamber (8). To avoid errors introduced by diminished adherence of PMNs, two membrane filters (mesh sizes, 0.45 and 3 ,tm; Millipore Corp., Bedford, Mass.) were used to separate the two compartments of the modified Boyden chamber (18). Fresh, human, autologous serum activated with endotoxin was prepared as described previously (10) and layered in the lower compartment of the Boyden chamber adjacent to the smaller mesh size (0.45 ,m) Millipore filter. The upper compartment was filled with cell suspension (3 x 106 PMNs per ml) with or without the addition of test agent. The chambers were incubated at 37°C for 2 h in an atmosphere of humidified air containing 5% carbon dioxide. After incubation, the Millipore filters were removed, stained with Ehrlich hematoxylin, cleared, and mounted on microscopic slides (8). Cells were counted on the attractant side of the 3,m filter in 10 consecutive evenly spaced fields as
defined by an ocular grid. Chemotaxis assays were done in duplicate or triplicate. The chemotaxis results are expressed as a ratio of experimental to control. A numerical value for this ratio of 1 represented stimulation of chemotaxis. In experiments testing the protection of cysteine against inhibition of PMN chemotaxis, cysteine (final concentration, 10 mM) was added prior to the addition of the cigarette smoke constituent. Effects of cigarette smoke components on PMN integrity. The test agents, at concentrations inhibitory to in vitro PMN chemotaxis, were incubated with PMN suspension (3 x 10" cells per ml) in Gey medium for 2 h at 37°C. The viability of the treated PMN was measured by the exclusion of trypan blue dye, as previously described (25). The number of PMNs per milliliter of suspension was also quantitated in the treated cellular suspensions to insure that there were no cellular losses. The PMNs were removed from the remaining suspension by centrifugation, and the cell-free supernatants, collected by decantation, were analyzed for LDH (EC 22.214.171.124) and /3-glucuronidase (EC 126.96.36.199) activities. As suitable controls, PMNs in an equivalent suspension (3 x 106 PMNs per ml) in Gey medium were disrupted by treatment with a Branson Sonifier (W140) at maximal voltage output for two 30-s bursts. After centrifugation, the supernatants from the disrupted cell suspension were collected by decantation and incubated in the presence and absence of test agents for 2 h at 37°C. The treated and untreated supernatants were analyzed for LDH and 1glucuronidase activities. LDH activity of the supernatant in the absence of test agent was taken as an index of complete cellular lysis. Since the f8-glucuronidase activity of these supernatants represented a constant proportion (65%) of the total 83-glucuronidase activity of these cells, as obtained by sonic treatment in the presence of Triton X-100 (Packard Instrument Co., Inc., Downers Grove, Ill.), this enzyme activity was used as an index of lysosomal enzyme release. Furthermore, these data indicated the effects of the test agents on these enzymatic activities. The assay for LDH activity was done as previously described (5). The oxidation of reduced nicotinamide adenine dinucleotide was followed continuously at 340 nm using a Unicam SP 1700 ultraviolet spectrophotometer equipped with a Unicam AR 25 linear recorder. A unit of LDH activity is defined as the amount of enzyme required to catalyze the oxidation of 1 gmol of nicotinamide adenine dinucleotide per min. ,B-Glucuronidase activity was measured as previously described (15). Phenolphthalein glucuronide was incubated with cell supernatants or cell-free sonic extracts for 18 h at 37°C. A unit of /8-glucuronidase activity is defined as the amount of enzyme required to liberate 1 ,umol of phenolphthalein. Glucose metabolic studies. Glucose metabolism of PMNs in the presence and absence of cigarette smoke constituents was studied by the previously described methods (10). Cigarette smoke constituents in 0.3 ml of Gey medium were added to cellular
BRIDGES ET AL.
suspensions (3 ml, 3 x 106 PMNs per ml in Gey medium containing 1 g of glucose per liter). To start the reactions, 5 ,ul of [1_14C]_ or [6-'4C]glucose (0.5 ,uCi, 50 ,Ci/,umol) was added. After incubation for 90 min at 37°C, the reactions were stopped by acidification, and the "4CO2 released was trapped and quantitated. The 14CO2 released from incubation mixtures containing either [1-_4C]- or [6-"4C]glucose was taken as an index of glucose metabolism via either the hexose monophosphate pathway or the glycolysis/oxidative metabolism, respectively. PMNs were collected from the acidified reaction mixture by centrifugation and analyzed for glycogen content by the method of Anderson and Stowring (1). The deproteinized incubation mixtures were then analyzed for glucose and lactic acid by using glucose oxidase (Worthington Biochemicals Corp., Freehold, N.J.) and the method of Barker and Summerson (3), respectively. Glucose consumed by PMNs was calculated by the difference between the nanomoles present in incubation mixtures without PMNs and the nanomoles present in test incubation mixtures. Some test agents (acetaldehyde, propionaldehyde, butyraldehyde, acrolein, and crotonaldehyde) produced an interfering colorimetric reaction in the lactic acid determination. In these instances, blanks containing an equivalent concentration of test agent were used. Glucose utilization or lactic acid production were expressed as nanomoles of glucose utilized or lactic acid produced per 90 min per 9 x 10V PMNs, respectively. The expression of metabolic results in terms of PMN number was previously shown to be valid (10), since equivalent PMN suspensions had a constant protein concentration. Although the metabolic results were highly reproducible within a given experiment, some variability in results was observed with different leukocyte preparations. Therefore, metabolic results were expressed as a ratio of the experimental values to the control values. A numerical value for the ratio of >1 represented metabolic stimulation, whereas a numerical value for the ratio of < 1 represented metabolic depression. Statistical analysis of data. Chemotactic and glucose metabolic data were subjected to statistical analysis by using Student's t test, testing the hypothesis that the ratio of treated to control was equal to 1. RESULTS
Chemotaxis and glucose metabolism of untreated PMNs. The mean values for chemotaxis and glucose metabolism of untreated PMNs are shown in Table 1. Since the glycogen content of PMNs was negligible (data not shown), the Gey medium was the sole source of glucose. The extent of PMN glucose metabolism via glycolysis (712 nmol) was approximately 130-fold greater than via the hexose monophosphate pathway (5.6 nmol). The glucose catabolic rate via oxidative metabolism as measured by 14CO0, released from [6-'4C]glucose
was negligible and unaffected by the test agents reported in this study (data not shown). Effect of cigarette smoke constituents on PMN chemotaxis. Figure 1 shows the effects of various concentrations of cyanide, sulfide, nicotine, crotonaldehyde (2-butenal), and acrolein (2-propenal) on PMN chemotaxis. The addition of increasing concentrations of these components caused a progressive suppression of PMN chemotaxis. As indicated by the standard errors, a high degree of variability in the results was generally obtained with concentrations of agent yielding less than 50% suppression of chemotaxis. The concentrations of cyanide, sulfide, nicotine, crotonaldehyde, and acrolein required to reduce the chemotactic responsiveness to 50% of control levels was approximately 3.5 mM, 6.5 mM, 3.5 mM, 40 uM, and 15 AM, respectively. Thus, among these components, the unsaturated aldehydes (i.e., acrolein and crotonaldehyde) were the most potent inhibitors of PMN chemotaxis. For comparative purposes, the effects of other aldehydes of cigarette smoke are given in Table 2. These saturated aldehydes (i.e., acetaldehyde, propionaldehyde, and butyraldehyde) were shown to be much less effective inhibitors of in vitro PMN chemotaxis than the corresponding unsaturated analogues. Comparison of the concentrations of these compounds necessary to inhibit chemotaxis with their respective concentrations in an inhibitory level of whole tobacco smoke was not favorable, making complete dose-response curves for these components unnecessary. The concentrations of the saturated aldehydes (i.e., propionaldehyde and butyraldehyde) necessary to inhibit PMN chemotaxis were at least 200-fold greater than the concentrations of their respective unsaturated analogues (i.e., acrolein and crotonaldehyde). Furfural (2-furaldehyde) was also shown TABLE 1.
Chemotaxis and glucose metabolism of untreated PMNs
Measured parameter Chemotaxis (PMNs per 10 fields)
Mean + SEa 1,528 + 142 (38)b
Glucose utilization (nmol of glucose utilized)"
Lactic acid produced (nmol of lactic acid produced)"
Hexose monophosphate activity ("'CO2 from [1-"4C]glucose,
± 64 (24)
"SE, Standard error.
Number of experiments.
'"Metabolic results are expressed in the units indicated
per 9 x
10' PMNs per 90 min of incubation.
VOL. 16, 1977
EFFECTS OF CIGARETTE SMOKE COMPONENTS
1 2 4 6880
-20 ~20 E..T 0
COOSU LFDE HYE(m,UM1)
C>0 °0 I-
FIG. 1. Effects of tobacco smoke components on in vitro chemotaxis of PMNs isolated from human peripheral blood. The results represent the mean percentages of control (+,- standard error) of three to six separate e-xperiments. * Mean percentages significantly (P < 0.05) different from control values.
to be a rather weak inhibitor of in vitro PMN
chemotaxis.Phenol and o-, m-, and p-cresol were also weak inhibitors of PMN chemotaxis. Concen-
trations of these phenolic compounds in excess of 5 mM were required to cause a significant (P < 0.05) inhibition of PMN chemotaxis (data not shown).
BRIDGES ET AL.
TABLE 2. Effects of the saturated aldehydes of cigarette smoke on PMN chemotaxis Test agent Concn (mM) Ratio to controls (± SE) pa Acetaldehyde 0.67 0.965 ± 0.167 (6)b NS 6.7 0.422 ± 0.029 (3)