INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Nitric oxide and other free radicals contained in cigarette smoke may trigger oxidation reactions and modify biologic membranes and proteins (1). Through the formation of peroxynitrite after the reaction between nitric oxide and superoxide in inflamed tissues, nitration of tyrosine residues of proteins occurs (2, 3), either on albumin or on lipoproteins (4, 5). Peroxynitrite may also interact with nucleic acids, causing DNA strand breakage (6), and react with guanine residues with the formation of 8-nitroguanine (7), thus contributing to the multistage process of carcinogenesis. Therefore, the formation of nitro-aromatic compounds may represent a useful marker of peroxynitrite-mediated reactions in biologic systems. In this context, the nitration of the ortho position of tyrosine leads to the formation of 3-nitrotyrosine (NTYR), which has been found in animal and human acute lung injury (8, 9) and in serum and synovial fluid of patients with rheumatoid arthritis as a marker of oxidative damage in chronic inflammation (10). Other protein residues, such as phenylalanine and tryptophan, may be nitrated as well by peroxynitrite or nitrogen dioxide (11, 12). NTYR formation after cigarette smoke exposure may thus result either from direct oxidant and free radical activity or from the activation of inflammatory cells in the respiratory tract (13). On the other side, the effects of cigarette smoking on antioxidant defenses are still controversial (14, 15), but it has been shown that the interaction of peroxynitrite with plasma constituents reduces plasma antioxidants (16). We developed a highly sensitive method to detect NTYR in human plasma and tested it in cigarette smokers and in healthy control subjects, along with the measurement of the total plasma antioxidant capacity.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Twenty healthy, asymptomatic cigarette smokers (13 males, 7 females; mean age: 49 ± 11 yr; average cigarette consumption: 26 ± 8 per day) and 18 healthy nonsmokers, taken as controls (10 males and 8 females; mean age: 36 ± 6 yr), were studied. FVC, FEV_1.0, and FEV_1.0/ FVC ratio in smokers were proved to be in the normal range (99.4 ± 14.1%, 99.5 ± 15.2%, and 100.3 ± 8.0% of predicted values, respectively). In smokers, expired CO levels were also determined before blood sampling. Peripheral venous blood (10 ml) was obtained in each subject. Plasma samples were prepared soon after blood withdrawal and stored at 20° C for nicotine and cotinine determination (17) or immediately processed for measurement of NTYR and total antioxidant capacity.

Measurement of Plasma NTYR

A modified version of the gas chromatography/thermal energy analysis (GC-TEA) method of Ohshima and associates was developed (18). Proteins were extracted from plasma by centrifugation at 500 × g for 15 min, precipitated with ammonium sulfate (80% saturation), and dialyzed (Dialysis Tubing; Medicell International Ltd, London, UK) against distilled water overnight. Dialyzed proteins were then dried under nitrogen, and NTYR was subsequently released by acid hydrolysis in vacuo with 6 N HCl (1 ml) at 110° C for 24 h. Hydrolysates were dried down, resuspended with 1 ml of distilled water, and then applied to C₁₈ end-capped columns (ISOLUTE^TM; International Sorbent Technology, UK) preconditioned with methanol (6 ml) and 0.05 N HCl (2 ml). Solid-phase extraction of NTYR was performed by applying the sample (1 ml) to the column which was eluted with 7 ml of distilled water. The first 1-ml fraction of the eluate was discarded, and the remaining 6 ml were collected, pooled, and dried down. Fractions and authentic NTYR (2 to 500 ng; Sigma-Aldrich, Milan, Italy) were derivatized with N-methyl-N-(tert-butyldimethylsilyl)trifluoroacetamide (MTBSFA, 80 µl) and pyridine (20 µl) at 70° C for 20 min. Aliquots (1 µl) of derivatized samples were injected by an automatic sampler into a Hewlett-Packard 5890 gas chromatograph equipped with a Hewlett-Packard capillary column (25 m × 0.32 mm, 0.52-µm film thickness). The injection port was at 295° C, and the column oven was increased from 80° C to 280° C, 70° C/min, by a gradient controller. Helium was used as carrier at a flow rate of 3 ml/min. A thermal energy analyzer (ThermedeTec 543; Analytical Instruments, Woburn, MA) was used as detector with a pyrolyzer temperature of 700° C. Calibration curves were generated with known amounts of NTYR standards, spiked plasma samples, and in vitro nitrosated bovine serum albumin (BSA). To this purpose, formation of NTYR in proteins by nitrating agents was followed by incubating 20 ml of 0.5% BSA with 1.0 mM NaNO₂ adjusted to pH 3.5 with acetic acid. The reaction was stopped by adding 5 ml of 25% trichloroacetic acid solution containing 10% (NH₄)₂SO₄. Proteins were precipitated, washed with 10 ml ethanol:ether mixture (1:1), and hydrolyzed as above.

Micropreparative HPLC for Detection of Plasma NTYR

With the above procedure, the presence of a chromatographic peak with authentic NTYR standards could be observed, but the detection of very low amounts of NTYR in human plasma samples was hampered by a high background signal in the neighbor retention times (Figure 1). We therefore decided to add a micropreparative step to the hydrolysates before the injection into the GC-TEA system. Dried hydrolysates were resuspended in distilled water (1 ml) and HPLC separated through a µ-Bondapak C₁₈ column 10 µm, 300 × 3.9 mm; Waters, Milford, MA) fitted with a guard column (4.6 × 45 mm). A quaternary pump (Millipore Co., Waters Division, Milford, MA) coupled with a photodiode array ultraviolet detector was used to control the gradient. Trifluoroacetic acid (1:1,000 vol/vol; BDH, UK) was released 99% for 5 min and then linearly mixed to 90% over 20 min with acetonitrile, which was thereafter increased to 95% over an additional 15 min. The flow rate was 1 ml/min. Given the possibility of nitration of amino acids other than L-tyrosine by nitrating agents, fractions corresponding to L-tyrosine, L-phenylalanine, L-tryptophan, and the corresponding 3-nitro-L-tyrosine, 4-nitro-L-phenylalaine, and 5-nitroindole were identified by ultraviolet spectrophotometry of authentic standards, collected, and dried down for GC-TEA processing.

View larger version (13K): [in this window] [in a new window]   Figure 1.   (Left panel  ) GC-TEA chromatogram of NTYR authentic standard (0.2 ng, arrow). (Right panel ) GC-TEA chromatogram of human plasma sample spiked with 0.5 ng of authentic NTYR (arrow).

Measurement of Plasma Total Antioxidant Capacity

The total Trolox^®-equivalent antioxidant capacity (TEAC) of plasma was measured according to Miller and coworkers (19). Metmyoglobin was obtained by adding myoglobin type III stock solution (400 µM) to an equal volume of freshly prepared 740 µM potassium ferricyanide. The solution was passed through a G15-120 Sephadex column, and the metmyoglobin fraction was collected. Visible spectra were analyzed with a spectrophotometer (Beckman, Fullerton, CA), and absorbance at 490, 560, and 580 nm was measured. After subtracting for background absorbance at 700 nm, the purity of metmyoglobin was estimated by applying the Whitburn equations (20). 2,2'-azinobis-(3-ethylbenzothiazoline-6-sulfonic acid) (ABTS; 500 µM, 300 µl), metmyoglobin (70 µM, 36 µl), and 497 µl of buffer (of which 4 µl was replaced by plasma in the human experiments) were mixed, and the reaction was initiated by the addition of 167 µl of hydrogen peroxide (450 µM). Visible spectra (450 to 900 nm) were recorded at 90-s intervals for 12 min. The absorbance at 734 nm after 6 min was measured, and the absorbance value of the buffer blank minus the value of the sample divided by the blank value was the fractional inhibition of the reaction. The same procedure was followed with 0.5, 1.0, 1.5, 2.0, and 2.5 mM solutions of Trolox (6-hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid) in place of plasma, and the inhibition value of plasma was then expressed as Trolox-equivalent from the Trolox dose-inhibition curve (Figure 2).

View larger version (12K): [in this window] [in a new window]   Figure 2.   Dose-response inhibition curve of the absorbance at 734 nm of 2,2'-azinobis-(3-ethylbenzothiazoline-6-sulfonic acid) (ABTS) visible spectrum, in the presence of metmyoglobin and hydrogen peroxide, with known amounts of 6-hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid ( Trolox). Circles represent mean values ± SD of 18 experiments.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The normal, non-nitrated, amino acids L-lyrosine, L-phenylalanine, and L-tryptophan can be singled out and identified by HPLC but they did not generate, as expected, GC-TEA signals (data not shown). On the other hand, NTYR and 4-nitro- L-phenylalanine almost coeluted in the HPLC runs with 18 to 19-min retention times, about 15 min earlier than 5-nitroindole (Figure 3, left panels). However, the three nitroamino acids could be singled out by GC-TEA, with the NTYR signal occurring 2 to 3 min later than the other two nitroamino acids (Figure 3, right panels). With this HPLC/GC-TEA technique, NTYR standard was detected down to the limit of 0.02 ng/injection, which represents a 25-fold increase in sensitivity compared with the original method (18). A linear dose-response relationship was observed in the range of 0.02 to 5.00 ng/µl concentration (Figure 4). BSA nitration with different amounts of NaNO₂ was linear in the range of 0.037 to 10.540 nmol/mg protein (data not shown).

View larger version (17K): [in this window] [in a new window]   Figure 3.   (Left panels) HPLC separation of authentic 3-nitrotyrosine (3-NTYR, 1 µg, top panel  ), 4-nitro-L-phenylalanine (N-PHE, 1 µg, middle panel ), and 5-nitroindole (5-N-IND, 1 µg, bottom panel ). (Right panels) GC-TEA separation of HPLC fractions (1:1,000 dilutions) corresponding to authentic 3-nitrotyrosine (3-NTYR, top panel ), 4-nitro-L-phenylalanine (N-PHE, middle panel ), and 5-nitroindole (5-N-IND, bottom panel ).

View larger version (15K): [in this window] [in a new window]   Figure 4.   Dose-response relationship of GC-TEA signal intensity with known amounts of authentic 3-nitrotyrosine. Circles represent mean values ± SD of 15 experiments.

NTYR was detectable in plasma of 11 of 20 (55%) healthy cigarette smokers (mean ± SD: 1.60 ± 1.24 ng/mg protein; range: 0.59 to 4.17) and in two of 18 (11%) healthy control subjects (1.10 and 1.20 ng/mg protein, respectively). In smokers, the amount of plasma NTYR was not associated with the number of cigarettes smoked per day or of pack-years, nor was it associated with indicators of active smoking such as blood nicotine (mean: 13.3 ± 6.9 ng/ml; range: 5 to 28) and cotinine (mean: 271.3 ± 101.1 ng/ml; range: 124 to 489) levels and expired CO (mean: 23.0 ± 13.7 ppm; range: 5 to 52) (Table 1). No subjects with GC-TEA peaks at the retention time corresponding to that of authentic 4-nitro-L-phenylalanine were observed in our series. Collecting HPLC fractions at the typical retention time of NTYR (18 to 19 min) prevented the potential noise of GC-TEA tracings with other nitroamino acids, such as 5-nitroindole. A representative GC-TEA chromatogram from a smoker with detectable plasma NTYR is shown in Figure 5.

View larger version (7K): [in this window] [in a new window]   Figure 5.   Representative GC-TEA chromatogram of HPLC-separated hydrolysate of human plasma sample. A visible peak at the retention time of authentic NTYR standard is indicated (arrow).

Plasma TEAC in smokers was significantly lower (0.43 ± 0.38 mM) than in healthy control subjects (1.42 ± 0.3 mM; p < 0.001), whose value corresponded with the normal reference value (1.46 ± 0.14 mM) reported from a much greater population sample (19). As shown for NTYR, no relationships between TEAC and the indicators of active and past smoking were found. Five subjects had extremely low TEAC, corresponding to < 2% inhibition of A₇₃₄ signal intensity. When plasma TEAC in smokers was plotted against the corresponding plasma NTYR, a biphasic, negative relationship was observed (r = 0.96, p < 0.001), which was fitted by the expression of TEAC as an exponential function of NTYR (Figure 6). The two healthy control subjects with detectable plasma NTYR had TEAC values just around the mean value of that category group (1.38 and 1.44 mM, respectively).

View larger version (13K): [in this window] [in a new window]   Figure 6.   Relationship between plasma NTYR (x-axis) and corresponding plasma TEAC (y-axis) in 11 healthy cigarette smokers.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

This highly sensitive HPLC/GC-TEA method for detection and quantitation of plasma NTYR revealed the presence of this nitroamino acid in 55% of healthy cigarette smokers, whereas only 11% of control subjects had detectable levels of NTYR (² = 6.27, p = 0.012). This is in contrast with the results of Ohshima and associates (18), who, on the other hand, sought to identify and quantitate urinary metabolites of NTYR in humans (actually they measured the major metabolite NHPA). Interindividual variations in protein metabolism and the contribution of sources of NHPA other than cigarette smoke in human urine may account for the lack of difference between smokers and nonsmokers in their study. Direct nitrating agents in cigarette smoke such as nitric oxide and/or peroxynitrite formation as a result of cigarette smoke-induced inflammatory reactions may account for the higher prevalence of NTYR detectability in cigarette smoke-exposed subjects compared with control subjects (1, 4, 8, 9). None of the study subjects had known exposure to other pollutants or particular sources of oxidants. However, other exogenous (e.g., nitrogen dioxide) and endogenous nitrating agents may have contributed to the measured level of NTYR in our subjects, in particular in the healthy nonsmokers (12, 18). Because tyrosine nitration is an irreversible reaction, NTYR might accumulate into proteins (21) and thus serve as a stable marker for noninvasive monitoring of cumulative oxidative damage in cigarette smokers, as proposed for circulating F₂-isoprostanes and for hydrogen peroxide in the exhaled air (22, 23). The double-step HPLC/GC-TEA method we used for plasma analysis was specifically focused on NTYR, although the nitration of other amino acid residues (such as tryptophan and phenylalanine) in proteins could have occurred (11, 12). However, none of the samples showed GC-TEA peaks corresponding to 4-nitro-L-phenylalanine, whereas oxidation of tryptophan to 5-nitroindole, whose detectability would have been missed in our system, was not observed after incubation of BSA with 1 mM peroxynitrite (16). In smokers, no relationships were found between NTYR and estimators of current and cumulative tobacco smoke exposure (Table 1).

In the same subjects, the presence of oxidative damage was associated with a reduction of the total plasma antioxidant capacity (TEAC), as compared with normal, healthy nonsmokers. The observed level of TEAC in our group of smokers (0.43 ± 0.38 mM) was in the same range of that reported for active and chronic smokers (24) and it corresponded to about one third of TEAC in normal subjects. This confirms that peroxynitrite-mediated reactions may cause depletion of plasma antioxidants (16). For instance, urate, which is estimated to contribute as much as 33% to the total antioxidant activity of plasma (19), has been shown to inhibit tyrosine nitration by peroxynitrite (1, 11). It is seen in Table 1 that three subjects without measurable levels of NTYR had very low TEAC values. This might be due to the direct consumption of antioxidants by tobacco smoke constituents without the priming of peroxynitrite-mediated reactions or to the presence of interfering substances in the assay (such as heme proteins or drugs) which may cause a marked underestimation of TEAC (19). However, in cigarette smokers, oxidative reactions characterized by the formation of NTYR are paralleled by a concomitant consumption of plasma antioxidants that counteract plasma constituent nitration. When we tested for the relationship between plasma NTYR and corresponding TEAC in the 11 smokers with detectable levels of NTYR, a negative, nonlinear relationship was found (Figure 6): for NTYR levels > 0.9 ng/mg protein plasma TEAC was severely reduced (< 0.2 mM in five of six smokers), whereas NTYR levels < 0.9 ng/mg protein were associated with TEAC values in the range of 0.5 to 1.1 mM. This figure is similar to that reported for the relationship between peroxynitrite concentration and thiol content of dialyzed plasma (16). Since thiols in plasma are mainly represented by the cysteine residues of albumin and albumin accounts for ~ 50% of normal TEAC (19), the observed relationship might indicate the reduction of the peroxyl radical-scavenging capacity of plasma proteins while the nitration of tyrosine residues occurs. The nonlinear shape of the above relationship may reflect, at least in part, the requirement for a double oxidation reaction on tyrosine to form NTYR, as reported for nitrogen dioxide (9).

Whatever the route of formation of NTYR might be (auto-oxidation of nitric oxide, peroxynitrite formation), the use of NTYR as a marker of free radical oxidative damage appears promising. We herein suggest a highly sensitive, noninvasive method for NTYR detection and quantitation in plasma, which seems suitable to monitor oxidation reaction in cigarette smokers. However, besides being a dosimeter of oxidative damage, NTYR formation may also have pathophysiologic relevance, since nitration of tyrosine residues on functional proteins such as ₁-proteinase inhibitor (25) may directly contribute to the development of emphysema in cigarette smokers. Therefore, the application of the above techniques in symptomatic smokers and/or patients with chronic obstructive pulmonary disease will be our goal in the near future. However, in light of the reported genotoxic effect of peroxynitrite (26), the use of NTYR as a biomarker in molecular epidemiology studies of human malignancies in which the exposure to nitrosating and alkylating agents is a relevant risk factor may have even broader applications (27). Its use and that of other biologic markers of exposure to tobacco smoke in easily accessible fluids and cells (28) may represent new tools for preventing tobacco smoke-associated lung diseases.