Intraphagosomal Chlorination Dynamics and Yields Determined

1080

Chem. Res. Toxicol. 1997, 10, 1080-1089

Intraphagosomal Chlorination Dynamics and Yields Determined Using Unique Fluorescent Bacterial Mimics Qing Jiang,† Donald A. Griffin,‡ Douglas F. Barofsky,‡ and James K. Hurst*,§ Departments of Biochemistry and Biophysics and of Chemistry, Washington State University, Pullman, Washington 99164-4630, and Department of Agricultural Chemistry and Environmental Health Sciences Center, Oregon State University, Corvallis, Oregon 97331-7301 Received June 5, 1997X

Fluorescein was covalently attached through a cystamine linker group to carboxy-derivatized polyacrylamide microspheres to generate phagocytosable particles containing fluorescent reporter groups. A unique feature of these beads is that the dye was recoverable in nearquantitative yield from intracellular environments by thiol reduction of the cystamine disulfide bond. Fluorescence microscopy indicated that individual neutrophils could bind as many as ∼20 serum-opsonized beads, although no appreciable cellular association was observed for unopsonized beads. By using methyl viologen to quench external fluorescence, it was demonstrated that 70-90% of the neutrophil-associated fluorescein on opsonized beads was inaccessible to the medium. The particle-bound fluorescein underwent near-stoichiometric conversion to chlorinated derivatives when reacted with HOCl or the cell-free myeloperoxidase (MPO)-H2O2-Cl- system; products were identified by HPLC separation and electrospray ionization mass spectrometry of the recovered dye. Fluorescence changes accompanying phagocytosis were consistent with chlorination of the dye; fluorescence spectrometric and chemical trapping measurements indicated that intraphagosomal chlorination was far more extensive than extracellular chlorination. Yields of recovered chlorofluoresceins determined by HPLC indicated that sufficient HOCl had been produced intracellularly to kill entrapped bacteria. Fluorescein chlorination coincided approximately with phagocytosis and stimulated uptake of O2 by the cells. Demonstration that HOCl is produced within phagosomes in sufficient concentrations to kill bacteria on a time scale associated with death constitutes strong evidence in support of a primary role for HOCl in the microbicidal action of neutrophils.

Introduction A wide variety of reactive oxygen and nitrogen species have been proposed as phagocyte-generated toxins and pathogenic agents in human disease (1-4). Prominent among these are hydrogen peroxide (5) or hypervalent metals and metal-peroxo complexes derived from H2O2 (6, 7), hydroxyl (6), and carbonate (8) radicals, hypochlorous acid (1, 3) and other chlorinating or oxidizing agents formed by its secondary reactions with nitrogen-based (9-11) or oxygen-based (12, 13) reactants, nitrogen oxides (14), and peroxynitrite ion (15, 16) and its carbon dioxide adduct (17). In principle, these oxidants can all be formed in biological reactions involving superoxide ion and/or nitric oxide, which are ubiquitous components of mammalian tissues. Although several of the proposed deleterious reactions may be inconsequential because the oxidant is either too unreactive or nonselective or is formed too slowly under physiological conditions to be biologically significant (18-23), most of the reactions are chemically plausible and probably occur to some extent in respiring cells. The critical issue in assessing the potential importance of a particular oxidant as an antimicrobial or pathogenic agent is therefore not whether it can be detected, but whether it is generated in sufficient amounts to effect significant damage within the time frame associated with development of cellular dysfunction or destruction. †

Department of Biochemistry and Biophysics. Department of Agricultural Chemistry and Environmental Health Sciences Center. § Department of Chemistry. X Abstract published in Advance ACS Abstracts, September 15, 1997. ‡

S0893-228x(97)00098-2 CCC: $14.00

To address this issue in phagocyte biochemistry, we have developed some unique fluorescein-labeled particles which are capable of efficiently trapping HOCl, nitrating agents, and other oxidants in intraphagosomal environments, giving distinct fluorescein reaction products. Quantitation of product yields is possible because nearly all of the dye can be recovered from the cells. Hypochlorous acid has long been thought to be an important microbicide generated by leukocyte peroxidases (1-4). Earlier studies had clearly demonstrated that stimulated neutrophils generate HOCl in myeloperoxidase-catalyzed reactions, although the subcellular localization and oxidant yields could not be determined (24-28). In this study, we use the fluorescein-conjugated beads to show that the extent of chlorination occurring within phagosomes is sufficient to account for the cytocidal activity of neutrophils toward prototypical bacteria.

Experimental Procedures Caution: Several chemicals used in these studies, including valinomycin, phorbol myristate acetate, and methyl viologen, are highly toxic and possibly carcinogenic. Proper precautions should be taken to avoid their inhalation, ingestion, or adsorption through the skin. Materials. Human neutrophils were isolated from the blood of individual donors by Hypaque-Ficoll density gradient centrifugation (29) and suspended in cold Hanks balanced salt solution without Ca2+ or Mg2+ [137 mM NaCl, 5.4 mM KCl, 0.34 mM Na2HPO4, 0.44 mM NaH2PO4, 4.2 mM NaHCO3, 5.5 mM glucose (HBSS1)]; the cell suspensions were maintained on ice and used within several hours of isolation. The number of cells in suspension was determined by counting on a hemocytometer in a solution containing 23 µg/mL methylene blue. Myeloper-

© 1997 American Chemical Society

HOCl Production within Neutrophil Phagosomes Scheme 1

oxidase (MPO) was purified from bovine spleens by column chromatography (30). The purified enzyme had an A430/A280 ) 0.85 and a specific activity (31) of 284 guaicol units/mg of protein; MPO concentrations were determined spectrophotometrically using 430 ) 91 mM-1 cm-1/heme (32). Hypochlorous acid (HOCl) solutions were prepared by distilling under vacuum commercial bleach solutions whose acidities had been adjusted to pH 7.0-7.5 with sulfuric acid; the concentration of HOCl in the distillate was determined spectrophotometrically using 236 ) 100 M-1 cm-1 (33). Reagent solutions of methyl viologen (N,N′-dimethyl-4,4′-bipyridinium, MV2+) were prepared from its dichloride salt; concentrations were determined spectrophotometrically using 258 ) 2.09 × 104 M-1 cm-1 (34). Other chemicals and biochemicals were the highest grade commercially available and were used as supplied. All solutions were prepared from deionized water that was purified by passage through a laboratory reverse osmosis/ion exchange unit. Synthesis of Fluorescein-Conjugated Beads. Custommade polyacrylamide microspheres with ∼1 µm diameters were generously provided by Bio-Rad Laboratories (Hercules, CA). The beads were digested in alkaline media to hydrolyze amide bonds, generating carboxyl end groups (35). Typically, heating to 80-90 °C for 3-4 h in 0.5 M Na2CO3 (pH 10.5) generated 0.2 mmol of carboxylate groups on 0.1 g dry weight of beads, as determined by titration with standard HCl (35). This corresponds to ∼8 × 108 carboxyl groups/particle. As outlined in Scheme 1, fluorescein was covalently attached to the carboxylmodified beads using cystamine as a spacer group. N-Hydroxysuccinimidyl-5-carboxyfluorescein (I) was first reacted at room temperature with a 20-fold excess of cystamine (II) in 0.05 M phosphate, pH 7.6, for 1-2 h to give the amide-linked cystamine derivative III. The reaction was monitored by thin layer chromatography on silica gel using methanol as eluent. Upon completion of the reaction, excess cystamine and a small amount (∼5%) of unidentified byproduct were removed by column chromatography on silica gel with methanol as eluent. The purified fluorescein derivative III was then amide-linked to the carboxyl-modified beads using 1-ethyl-3-[3-(dimethylamino)propyl]carbodiimide (EDC) as a coupling agent to give the 1 Abbreviations: CGD, chronic granulomatous disease; DTNB, 5,5′dithiobis(2-nitrobenzoic acid); DTT, dithiothreitol; EDC, 1-ethyl-3-[3(dimethylamino)propyl]carbodiimide; ESI, electrospray ionization; HBSS, Hanks’ balanced salt solution; MPO, myeloperoxidase; MV2+, N,N′dimethyl-4,4′-bipyridinium; PBS, phosphate-buffered saline; PMA, phorbol 12-myristate 13-acetate; PMN, polymorphonuclear leukocytes (neutrophils); TNB, 5-thio-2-nitrobenzoic acid.

Chem. Res. Toxicol., Vol. 10, No. 10, 1997 1081 desired fluorescein-conjugated particles IV. This reaction was carried out overnight at 4 °C. The labeled beads were washed repeatedly with 0.2 M NaCl and stored in pure water at 4 °C. Treatment of the particles with ∼0.25 M dithiothreitol (DTT) to cleave the linker group disulfide bond (Scheme 1) caused nearquantitative (g95%) release of the bound fluorescein, as determined by comparing the fluorescence intensities and absorbances of the beads before and after addition. Chemically synthesized compound V had an absorbance maximum at 494 nm in alkaline solutions, for which 494 ) 7.5 × 104 M-1 s-1 at pH 9. From the absorbance of supernatant solutions containing the DTTreleased fluorescein derivative, the amount of originally bound dye was calculated to be 0.7-4 × 107 fluorescein molecules/bead for the various preparations. The mean particle size of the fluorescein-conjugated beads measured by scanning electron microscopy was 0.9 µm, with a size distribution of 0.5-2.0 µm. Phagocytosis and Stimulation of Neutrophils. For most experiments, the beads were opsonized by suspension to a particle density of 3 × 109 beads/mL in 25% human serum75% HBSS, incubated at 37 °C for 20-25 min, then washed once with cold HBSS, resuspended in HBSS, and kept on ice until used. Neutrophils were prewarmed to 37 °C for 2 min, and phagocytosis was initiated by adding 1 mM CaCl2 and 1 mM MgCl2 and the opsonized beads. The mixture was then gently rotated in an incubator at 37 °C. Phagocytosis was terminated at 30 or 60 min by placing the mixture on ice; for some experiments, 1% formaldehyde was also added to inhibit respiration (29). The two procedures gave equivalent results. Unopsonized beads were used to minimize phagocytosis in experiments for which phorbol 12-myristate 13-acetate (PMA) was used to stimulate respiration and extracellular degranulation. Unopsonized beads were also used in experiments utilizing 100% serum as the medium. In some studies using neutrophils from individuals with chronic granulomatous disease (CGD), glucose and glucose oxidase were added to provide an external source of hydrogen peroxide. In other experiments, 20 mM taurine or 1 mM methionine was added to the reaction medium to act as an extracellular scavenger of HOCl. The extent of phagocytosis under the various experimental conditions was estimated qualitatively using a Nikon Microphot-FX fluorescent microscope. Rate Measurements. At periodic intervals following initiation of phagocytosis, small aliquots from the reaction mixture were diluted into cold HBSS and fluorescence changes were measured using a PTI Model A1010 dual monochromator fluorescence spectrophotometer. In some experiments, extracellular fluorescence was quenched by adding 35 mM MV2+ to the medium immediately prior to recording the fluorescence spectrum. Control experiments were also done in which 1 mM sodium azide was added to the medium to inhibit MPO activity. The order of addition in these experiments was neutrophils, azide, and beads. These experiments were required to distinguish effects due to encapsulation of the beads within sealed phagosomes and probe chlorination. Specifically, phagocytosis leads to enhancement of the measured fluorescence intensity in MV2+-containing solutions, and chlorination leads to partial quenching of fluorescence and a bathochromic shift of the fluorescence band (26). Inhibition of peroxidase activity allows the phagocytic process to be independently monitored. Oxygen consumption by neutrophils was determined using a Rank Brothers O2 polarographic cell thermostatted at 37 °C. Typically, 2 × 106 neutrophils were mixed with 1 × 108 beads in 1.5 mL of buffer to initiate respiratory activation. Oxygen uptake was monitored using a Linear stripchart recorder; the recorder was calibrated by measuring the pen displacement following addition of excess sodium dithionite to air-saturated buffer in the electrode chamber. The rate of stimulated respiration was determined from the maximal rate of O2 uptake, which occurred after a brief lag period following addition of the beads. At this point, the rate was linear over a 5-10 min period. Product Identification. Following phagocytosis, the neutrophils were broken by homogenization in 8.6% sucrose, and the dye was recovered by addition of ∼0.25 M DTT followed by

1082 Chem. Res. Toxicol., Vol. 10, No. 10, 1997 centrifugation to remove cell debris. Fluorescein and reaction products in the supernatant were separated by C18 reverse phase HPLC utilizing a Gilson 305/306 system equipped with an LDL Analytical Spherisorb ODS2 column (5 µm, 4.6 × 250 mm). Isocratic elution using 29% methanol-71% 20 mM sodium phosphate (pH 7.4) gave three major peaks in addition to fluorescein as detected by absorption spectroscopy at 498 nm. Reference standards were prepared by reacting fluoresceinated beads in 50 mM sodium phosphate (pH 7.4) containing 0.1 M NaCl with either HOCl or MPO plus H2O2. In the former case, ∼1.5 × 109 beads containing 60 nmol of fluorescein were either flow-mixed with 60-120 nmol of HOCl or an equivalent amount of HOCl was added in small aliquots to a vigorously stirred suspension of the beads. For the enzymatic reaction, the bead suspension was prewarmed to 37 °C and then 70 nM MPO and 50-100 µM H2O2 were added to initiate the reaction. The mixture was vortexed several times over the course of the reaction, which was terminated after 30 min. Following release of the dye by reaction with DTT, the products were separated by HPLC. Fractions containing the major peaks were desalted by evaporating to dryness using a rotary evaporator and extracting the residue with absolute methanol. This process was repeated until negligible residue remained in the extraction flask. Rechromatography of the purified products established that the extraction process did not elicit any changes in chemical composition. The products were identified by pneumatically assisted electrospray (ionspray) mass spectrometry on a Perkin-Elmer SCIEX API III+ triple-quadrupole mass spectrometer. The structures were confirmed by tandem mass spectrometry (MS/ MS) of the major ions. The potential of the ionspray needle was placed at 4700 V to produce positive ions, and the potential of the orifice leading into the mass analyzer was set to 80 V. The flow rates for the nebulizer gas (air) and the curtain gas (nitrogen) were both set at 0.6 L/min. The instrument was mass-calibrated in the positive ionization mode with a mixture of poly(propylene glycol)s. The samples were chromatographed on an LC column as follows: The chromatography system consisted of a Perkin-Elmer ABI 140 B syringe pump, a Rheodyne 8125 injector, and a Phase Separations Spherisorb ODS2 column (5 µm, 1 × 100 mm). The solvents, which were water (A) and acetonitrile (B), each containing 0.1% trifluoroacetic acid, were delivered at 40 µL/min and programmed from 5% B to 95% B in 15 min. Tandem mass spectrometry was performed using argon with 10% nitrogen as the collision gas. The collision gas thickness was set at 230, and the collision energy was adjusted to the optimum value for each compound. The extent of taurine chloramine formation was determined by reacting product solutions with 5-thio-2-nitrobenzoic acid (TNB) and measuring spectrophotometrically the extent of colorimetric loss at 412 nm [412(TNB) ) 1.41 × 104 M-1 cm-1] associated with its oxidation to colorless 5,5′-dithiobis(2-nitrobenzoic acid) (DTNB) (9). Reagent solutions of TNB were prepared by dissolving 2 mM DTNB in 50 mM phosphate, pH 7.4, titrating to pH 12 with NaOH to promote its hydrolysis, and then back-titrating with 1 M HCl to pH 7.4. The TNB solutions were stored in amber bottles at 4 °C. Following phagocytosis experiments in media containing taurine, the cells were pelleted by centrifugation at 4 °C and 10000g for 10 min, and the supernatant was immediately removed. Bovine catalase (940 U/mL) was added to scavenge any H2O2 present in the medium, followed by TNB, and the spectral change at 412 nm was measured. Measurement of Intraphagosomal Acidities. Opsonized beads at 3 × 108 beads/mL ([fluorescein] ) 5 µM) were incubated in a slowly rotating test tube with 2 × 107 neutrophils/mL at 37 °C in HBSS containing 1 mM NaN3. At timed intervals, 50 µL aliquots were diluted into 1.4 mL of cold phosphate-buffered saline solution [137 mM NaCl, 2.7 mM KCl, 8.1 mM Na2HPO4, 1.5 mM KH2PO4 (PBS)] containing 1 mg/mL glucose, then 35 mM methyl viologen was added to quench extracellular fluorescence, and the fluorescence excitation spectrum for emission at 523 nm was immediately recorded. The intracellular pH was

Jiang et al.

Figure 1. Liquid chromatograms of fluorescein recovered from oxidized beads: (trace A) 50 µM HOCl reacted with beads containing 30 µM total fluorescein; (trace B) 5 × 108 beads/mL containing 25 µM total fluorescein incubated with 107 PMN/ mL in HBSS, pH 7.4, for 30 min at 37 °C. Peaks identified by mass spectrometry were fl, unreacted fluorescein; fl-Cl, chlorofluorescein; fl-Cl2, dichlorofluorescein; fl-Cl3, trichlorofluorescein. determined by comparing the ratios of experimentally determined excitation intensities at 493 and 433 nm (I493/I433) to calibration curves constructed from I493/I433 ratios measured for the fluoresceinated beads at various pH values (36, 37). Two separate calibration protocols were used to generate the curves. In one, the fluoresceinated beads were merely suspended in K+Ringers buffer (90 mM KCl, 2 mM KH2PO4, 1 mM MgCl2, 1 mM MgCl2) containing 25 mM potassium acetate, 25 mM MES, or 25 mM HEPES at various pH values and the excitation spectrum recorded. In the other, aliquots were taken at 15 min following initiation of phagocytosis as described above and diluted into Ringers buffers to which 30 µM nigericin and 20 µM valinomycin had been added. After incubation for 5 min to allow collapse of transmembrane electrochemical gradients (38), methyl viologen was added and the excitation spectra were recorded. The spectra were corrected for background scattering and intrinsic fluorescence of the cells by subtracting the excitation spectra of solutions containing the same cell density of neutrophils without added beads.

Results Product Identification and Yields. The fluoresceinated beads were reacted with HOCl, the cell-free MPO-H2O2-Cl- enzymatic system, and stimulated neutrophils; products from each of these reactions were isolated by HPLC following reduction of the disulfide bond in the spacer group with DTT as described in Experimental Procedures. The chromatograms gave qualitatively identical patterns indicating formation of three major products (Figure 1); the corresponding products from each of the reactions coeluted when mixed and gave identical excitation and emission spectra. The product spectra were different from the unreacted fluorescein derivative and showed a progressive red shift in optical bands with increasing retention time. Specifically, for the peak eluting at 7.3 min, excitation (λex) and emission (λem) peak maxima were λex ) 501 nm and λem ) 530 nm; for the 12 min peak, λex ) 508 nm and λem ) 536 nm; for the 22 min peak, λex ) 514 nm and λem ) 542 nm. These optical changes are consistent with

HOCl Production within Neutrophil Phagosomes

Chem. Res. Toxicol., Vol. 10, No. 10, 1997 1083 Table 1. Chlorination of Fluorescein-Conjugated Beads by Myeloperoxidase-H2O2-Cl- a conditions complete system: MPO (70 nM) + H2O2 (100 µM) + NaCl (100 mM) + beads (2.0 × 109, [fl] ) 100 µM) complete system minus: MPO H2O2 Clcomplete system plus: NaN3 (1 mM) taurine (20 mM) methionine (1 mM) catalase (940 U/mL)

Figure 2. Ionspray mass spectrum of chlorofluorescein. A 5 µL portion of the HPLC fraction from HOCl-reacted beads that was nominally identified as the monochlorofluorescein derivative was injected onto the LC-ionspray system. The inset shows an expanded view of the mass scale between m/z 468 and 476.

progressive chlorination of the fluorescein chromophore (26). The products of the reaction with HOCl were identified by ionspray ionization (ESI) and tandem (MS/MS) mass spectrometry. Three parent ion peaks were often observed for a single purified product, one of which was attributable to the [M + H]+ ion of the sulfhydryl compound and the other two to the [M + H]+ and [M + 2H]2+ ions of the corresponding disulfide, which apparently formed by aerobic sulfhydryl oxidation during the chromatographic desalting step immediately prior to introduction into the mass spectrometer. A representative ESI spectrum of the sulfhydryl compound eluting at 7.3 min is reproduced in Figure 2. The parent ion peak at m/z 470 is the value expected for addition of a single Cl atom to the fluorescein derivative (Scheme 1). At higher mass resolution, this peak is seen to be due to a cluster of ions (Figure 2, inset), whose relative signal strengths correspond to the isotopic distribution of a monochlorinated species. Specifically, ions were detected at nominal m/z values of 470 and 472 with ion signal ratios of about 3:1, which is the 35Cl:37Cl isotope ratio. The second set of peaks at 471 and 473 amu, which appeared in a 3:1 ratio and were ∼25% as intense as the first set of peaks, is attributable to the corresponding ions containing a single 13C atom, as expected from the natural abundance of this isotope in an ion containing 23 carbons. The sulfhydryl compound eluting at 12 min gave peaks at 504 ([M + H]+ for 35Cl2), 506 ([M + H]+ for 35 Cl37Cl), and 508 ([M + H]+ for 37Cl2) amu in the ion current ratio of 9:6:1, identifying this compound as a dichlorofluorescein. The mass spectrum of the compound eluting at 22 min could be obtained only as the disulfide, which showed a multiplet of peaks separated by 2 amu beginning at 1073 amu with intensity ratios 0.52:1.0:0.80: 0.34:0.08, indicative of trichlorination of each of the fluorescein rings. As with the monochloro derivative, an identical set with ∼25% the ion currents that was displaced 1 amu to higher masses was also observed, corresponding to the analogous ions containing a single 13C atom. The structures of the molecular ions were confirmed by tandem mass spectrometry. The fragmentation patterns for the nonchlorinated, monochlorinated, and dichlorinated sulfhydryl compounds were very similar. For example, the 35Cl2-[M + H]+ parent at m/z 503.6 dissociated primarily into fragments at m/z 469.6 ([M + H]+ - [H2S]), 443.6 ([M + H]+ - [CH2dCHSH]), 414.0

HOCl trapped (nmol)b,c 65 ( 4 (3)

0 0 0 0 8.3 ( 1.8 (2) 3.9 ( 0.5 (2) 2.6

a HOCl trapped by fluorescein-conjugated beads was quantified by reverse phase HPLC following release of the dye from the beads as described under Experimental Procedures. b In 1 mL of 50 mM sodium phosphate, pH 7.4, at 37 °C, after 30 min. c Error limits are average deviations from the mean value determined for the number of measurements given in parentheses.

([M + H]+ - [CH3CH2SH + CO]), 400.0 ([M + H]+ [CH2dCHSH and CO]), and 354.8 ([M + H]+ [HCONHCH2CH2SH + COOH]). Unlike the electron impact spectra obtained for similar fluorescein compounds (26), no evidence was found for cleavage of the carbon-chlorine bond. Myeloperoxidase was found to adhere strongly to both opsonized and unopsonized fluoresceinated beads. When MPO was incubated for 10 min with 1.5 × 109 beads/mL suspended in PBS, and then beads and supernatant were separated by centrifugation, >99.8% of the MPO activity was found associated with the particulate fraction. Chlorination of the microsphere-bound dye by MPOH2O2-Cl- required the presence of all reaction components and, despite strong adsorption of the enzyme, was blocked by the MPO inhibitor sodium azide and extensively inhibited by addition of the HOCl scavengers taurine and methionine (Table 1). Catalase, which competes with the peroxidase for H2O2, was also inhibitory. This pattern of behavior is commonly observed for MPO-catalyzed chlorination reactions and suggests that the enzymatic system functions as a source of HOCl, which subsequently reacts nonenzymatically with fluorescein. Chlorination yields were determined spectrophotometrically following HPLC separation of the chlorinated products using molar extinction coefficients that were measured for authentic samples of the monochloro (500 ) 7.5 × 104 M-1 cm-1) and dichloro (508 ) 8.0 × 104 M-1 cm-1) derivatives and estimated from values for other halogenated fluoresceins (39) (515 = 9.0 × 104 M-1 cm-1) for the trichloro derivative. The total amount of HOCl trapped was calculated as the sum of the amounts of monochlorofluoresceins, 2 times the dichlorofluoresceins, and three times the trichlorofluoresceins recovered. In the enzymatic reaction, the HOCl trapping yield exhibited a bell-shaped pH profile with an optimum at pH ∼6.5 (Figure 3A), approximately the same as the maximal activity of MPO under these reaction conditions (40). At the optimal pH, about 80% of the added H2O2 was trapped as chlorofluorescein compounds. This nearstoichiometric conversion was observed over a relatively wide range of H2O2 concentrations below about 100 µM; above this concentration, the yield of chlorinated products rapidly declined (Figure 3B). The reduced yields are

1084 Chem. Res. Toxicol., Vol. 10, No. 10, 1997

Figure 3. pH and H2O2 dependencies of chlorination yields from reaction of fluorescein-conjugated beads with MPO-H2O2Cl-. Conditions: (panel A) 2 × 109 beads/mL containing 100 µM total fluorescein reacted with 100 µM H2O2 for 30 min at 37 °C in 50 mM sodium phosphate at various acidities containing 35 nM MPO and 0.1 M NaCl; (panel B) identical conditions at pH 7.4 with varying [H2O2]. Table 2. Bead-Induced Phagocytosis, Respiratory Stimulation, and HOCl Production by Neutrophils beads/ PMNa

mediumb

% entrappedc-e

HOCl trappedc,f (nmol)

O2 uptake ratec,f (nmol/min)

0 50/1 (U) 15/1 (O) 15/1 (U) 50/1 (O) 50/1 (U) 50/1 (O)

buffer buffer buffer serum buffer serum buffer + NaN3g

0 50 ( 5 (4) 80 ( 4 (3) 23 ( 4 (4) 28 ( 2 (3) 23 ( 4 (4)

0 9.5 ( 0.5 (3) 15 ( 0.5 (3) 34 ( 3 (4) 51 ( 3 (3) 0.3

3(2 3(2 16 ( 2 (3) 30 ( 2 (3) 37 ( 4 (4) 55 ( 3 (3) 45 ( 2 (2)

a The notation U and O refers to unopsonized beads and beads opsonized in 25% serum, respectively. b Beads were incubated with neutrophils (PMN) for 30 min at 37 °C in either HBSS or PBS, pH 7.4, or 100% human serum; equivalent results were obtained in the two buffers. Respiration rates were measured in PBS containing 1 mg/mL glucose at 37 °C. c Error limits are average deviations from the mean value determined for the number of measurements given in parentheses. d The extent of entrapment was equated with the fraction of fluorescence that could not be quenched by adding 35 mM MV2+ to cellular suspensions containing 1 mM NaN3. e At 15/1 beads/PMN, >90% of the beads were attached to the cells, and at 50/1 beads/PMN, ∼40% of the beads were attached. The fraction of attached beads that were entrapped is given by % entrapped/% attached. f Yields of HOCl trapped and maximal rates of O2 uptake are scaled to 107 neutrophils (PMN)/ mL. g 1 mM.

potentially attributable to inactivation of MPO and/or its catalytic disproportionation of H2O2, reactions that are known to occur at relatively high concentrations of this oxidant (40). Addition of HOCl to the fluoresceinated beads in 1:1 or 2:1 HOCl:fluorescein ratios led to nearly 100% trapping as chlorofluoresceins. Phagocytosis and Chlorination of the FluoresceinConjugated Beads. Opsonized beads or unopsonized beads mixed with neutrophils in serum markedly stimulated neutrophil respiration (Table 2) and were avidly phagocytosed. In contrast, unopsonized beads in aqueous buffers did not stimulate respiration in the cells and underwent negligible phagocytosis. When examined by fluorescence microscopy, the number of opsonized beads associated with the cells rapidly increased within the first few minutes following initiation of phagocytosis; by 15 min, less than 10% were unattached at bead/neutrophil ratios below 15/1. At higher ratios, as many as 20 beads were attached to a single neutrophil. In similar experiments, only a few percent of the unopsonized beads in

Jiang et al.

aqueous buffers were found adherent to the neutrophils following incubation. Phagocytosis of opsonized beads led to extensive chlorination, whereas incubation of unopsonized beads gave no evidence of chemical reaction (Table 3). The amount of HOCl trapped per opsonized bead increased proportionately with the amount of fluorescein that was bound under otherwise identical conditions (Figure 4). Addition of taurine or methionine to the reaction medium did not diminish the extent of fluorescein chlorination of the opsonized beads, although these HOCl scavengers effectively inhibited chlorination in reactions between unopsonized beads and PMAstimulated neutrophils (Table 3). Formation of HOCl by the PMA-stimulated cells was extensive, as is evident from the high yields of chlorofluoresceins and N-chlorotaurine (Table 3). The less efficient inhibition of chlorofluorescein formation by taurine than methionine may be attributable to secondary reactions of N-chlorotaurine with the beads. We have found that N-chlorotaurine and other chloramines also react slowly with fluoresceins to give chloro derivatives (data not given). Sodium azide nearly completely inhibited chlorination of the opsonized beads by neutrophils, although phagocytosis and respiration (Table 2) were normal. Neutrophils from an individual with CGD were capable of phagocytosing the beads, but, as expected (26), fluorescein chlorination occurred only when an exogenous source of H2O2 (glucose plus glucose oxidase) was provided (Table 3). The chlorination yield with respect to O2 consumption was determined by measuring the amount of chlorofluoresceins recovered from reaction between neutrophils and particles in a closed reaction chamber containing no air space. Oxygen consumption was monitored with an O2 polarographic electrode, and the reactions were allowed to proceed until the solutions became anaerobic. When 1.8 × 107 neutrophils were challenged with 2.3 × 108 unopsonized beads (bead/neutrophil ) 13/1) in 1 mL of serum, 22 nmol of HOCl was recovered as chlorofluoresceins from the beads. Since the solution originally contained ∼200 nmol of O2, the efficiency of conversion (HOCl trapped/O2 consumed) was ∼11%. Very similar results were obtained for opsonized beads with neutrophils in PBS, for which 19 nmol of HOCl was trapped at identical particle densities, giving an efficiency of 9.5%. At bead/neutrophil ratios of 50/1 (3.5 × 108 beads, 7.0 × 106 neutrophils), the corresponding efficiencies were 11% for unopsonized beads in serum and 8% for opsonized beads in PBS. All of these measurements utilized a preparation containing 3.6 × 107 fluorescein molecules/ bead to optimize trapping efficiencies (Figure 4). Fluorescence Quenching Experiments. The extent of phagocytosis and intraphagosomal chlorination of opsonized beads was further probed by fluorescence spectroscopy using methyl viologen to quench the fluorescence of extracellular dye. The MV2+concentration dependence of fluorescence intensity of beads suspended in PBS exhibited simple Stern-Volmer behavior, i.e., Io/I ) 1 + KQ[MV2+] (Figure 5), where Io and I are the measured intensities in the absence and presence of MV2+, respectively. The quenching constant (KQ), measured over the range of 0.2-3.0 × 108 beads/mL, containing 0.23-18 µM total fluorescein, was KQ ) 806 ( 24 M-1. With [MV2+] ) 35 mM, the conditions of our experiments with neutrophils, the extent of fluorescence quenching of solvent-exposed beads was 95-97%.2


Chem. Res. Toxicol., Vol. 10, No. 10, 1997 1085

Table 3. Chlorination of Fluorescein-Conjugated Beads by Neutrophils (PMN) in the Presence or Absence of HOCl Scavengers HOCl (nmol)b trapped by conditionsa particle-stimulated normal neutrophils: PMN + beadsU PMN + beadsO PMN + beadsO + methionine (1 mM) PMN + beadsO + taurine (20 mM) PMN + beadsO + NaN3 (1 mM) PMA-stimulated normal neutrophils: PMN + PMA (50 nM) + taurine (20 mM) PMN + beadsU + PMA PMN + beadsU + PMA + taurine PMN + beadsU + PMA + methionine (1 mM) particle-stimulated CGD neutrophils: PMN + beadsO PMN + beadsO + glucose (10 mM) PMN + beadsO + glucose + glucose oxidase (70 mU/mL)

beads 0 21 ( 2 (5) 20 ( 1 (3) 21 ( 2 (3) 0.3 (2)

taurine

11 ( 1 (3) 53 ( 5 (2)

19 ( 1 (2) 4.0 ( 0.7 (2) 2.1 ( 0.2 (2)

42 ( 4 (2)

0 0 12 ( 2 (2)

a Reactions, carried out by incubating 1.0 × 109 beads/mL containing 50 µM total fluorescein with 8 × 106 PMN/mL in 0.5 mL of HBSS, pH 7.4, for 30 min at 37 °C, were terminated by adding 1% formaldehyde. The notation beadsO and beadsU refers to beads that were opsonized in 25% human serum and unopsonized beads, respectively. b Error limits are average deviations from the mean value determined for the number of measurements given in parentheses.

Figure 4. Dependence of HOCl trapping yields upon extent of fluorescein loading. A total of 5 × 108 opsonized beads/mL containing various amounts of fluorescein were incubated with 1 × 107 PMN/mL for 30 min in HBSS, pH 7.4, at 37 °C.

In general, phagocytosis of the fluorescein-conjugated beads led to progressive loss of fluorescence intensity and a red shift of the fluorescence bands, as illustrated in Figure 6A. These effects are consistent with the reduction in the fluorescence quantum yield and optical shifts that accompany chlorination of the dye (26). Periodic addition of MV2+ caused an instantaneous decrease of only a portion of the fluorescence; the residual fluorescence underwent no further changes in intensity for periods extending to 30 min following addition of the quencher. Thus, the residual fluorescence is attributable to dye that is inaccessible to MV2+. The intensity of the 2 Analogous viologens containing other N-alkyl substituent groups were more effective quenching agents, i.e., N,N′-dioctyl-4,4′-bipyridinium, for which KQ ) 2.6 × 103 M-1, and N,N′-didecyl-4,4′bipyridinium, for which KQ ) 2.0 × 104 M-1. However, these more lipophilic ions were apparently capable of disrupting or permeating the neutrophil membrane and gaining access to the occluded fluorescein, based upon the observation that a slow continuous decline in fluorescence intensity occurred over a period of ∼20 min following the initial quenching phase.

Figure 5. Fluorescence quenching of fluorescein-conjugated beads by methyl viologen. Conditions: 2 × 107 beads/mL containing 0.23 µM total fluorescein in HBSS, pH 7.4 (λex ) 494 nm, λem ) 525 nm, 3 nm slit width). The line corresponds to the equation: Io/I ) a + b[MV2+], with constants a ) 1.0 and b ) 830 M-1.

residual fluorescence initially increased with time in a manner that paralleled visual observation of phagocytosis but then decreased (Figure 6B). When NaN3 was added to the reaction medium to inhibit MPO, the overall loss of intensity was markedly attenuated (Figure 6C) and the residual intensity maximized after 15 min at somewhat higher values than when the enzyme inhibitor was absent and underwent no appreciable red shift in the fluorescence maximum (Figure 6D). Qualitatively, these data implicate MPO in the processes that lead to reduction of intensity and optical shifts of the residual fluorescence. The relative magnitude of residual fluorescence can be used to estimate the apparent fraction of beads that were entrapped. These measurements were made in the presence of NaN3 to minimize interference from MPO-catalyzed reactions. This fraction was found to vary widely, depending primarily upon the bead/neutrophil ratio (Table 2). The data indicate that up to about


Jiang et al.

Figure 7. Kinetics of phagocytosis and fluorescein chlorination by neutrophils. A total of 3 × 108 beads/mL containing 17.5 µM total fluorescein were incubated with 1.9 × 107 PMN/mL at 37 °C. At timed intervals, phagocytosis was terminated, and both the fraction of fluorescence that was unquenched by MV2+ and the extent of chlorination of the isolated fluorescein were determined: (panel A) opsonized beads in HBSS, pH 7.4; (panel B) unopsonized beads in 100% serum. Symbols: percent unquenched fluorescence in the presence (O) and absence (4) of 1 mM NaN3; total chlorination yield of recovered dye (9).

Figure 6. Fluorescence changes accompanying phagocytosis. A total of 3 × 108 opsonized beads/mL containing 7.5 µM total fluorescein were incubated with 2 × 107 PMN/mL in HBSS, pH 7.4, at 37 °C. At timed intervals, 50 µL aliquots were removed and diluted into 1.4 mL of cold HBSS: (panel A) total emission spectra at λex ) 494 nm; (panel B) emission spectra following addition of 35 mM MV2+ to the dilution buffer; (panel C) emission spectra when the incubation buffer contained 1 mM NaN3; (panel D) emission spectra in the presence of 35 mM MV2+ for samples from incubation buffers containing 1 mM NaN3. The numbers accompanying the individual traces indicate the time interval in minutes between initiation of phagocytosis and sampling.

10 beads can be readily assimilated by a single neutrophil and, at higher ratios, the remainder of the beads are primarily solvent-exposed. Phagocytosis of unopsonized beads by neutrophils in serum was more efficient than phagocytosis of opsonized beads in aqueous buffer under otherwise identical conditions (Table 2), which presumably also accounts for the greater respiratory stimulation (Table 2) and more extensive chlorination (Figure 7) of the beads in serum. Kinetics of Phagocytosis and Chlorination by Neutrophils. Changes in the magnitude of the unquenched fluorescence following mixing of opsonized beads with neutrophils were used to monitor the rate of bead entrapment, which was compared to the extent of chlorination of the recovered dye at the corresponding times. In all cases, chlorination was extensive and closely followed phagocytosis. A typical result is given in Figure 7A; the apparent halftime for phagocytosis was ∼4 min, based upon the results with N3--inhibited cells, while t1/2 for chlorination was ∼7 min. Very similar results were obtained when unopsonized beads were mixed with neutrophils in serum (Figure 7B), quantitative differences being that the apparent halftimes for both phagocytosis (t1/2 = 3.5 min) and chlorination (t1/2 = 4.5 min) and the extent of entrapment and chlorination were somewhat greater (50% greater under the conditions of Figure 7) at comparable bead and neutrophil densities. The onset and extent of dye chlorination also coincided with respiratory stimulation of the neutrophils (Figure 8). For the experiments described in Figures 6 and 7,

Figure 8. Kinetics of O2 consumption and fluorescein chlorination by neutrophils. A total of 6.7 × 107 opsonized beads were mixed with 1.3 × 106 PMN/mL in 10 mM PBS, pH 7.4, to stimulate respiration. O2 uptake is compared to the extent of chlorination of recovered fluorescein in experiments where 5 × 108 opsonized beads/mL containing 29 µM total fluorescein were incubated with 1 × 107 PMN/mL. Symbols: total O2 consumed (O); total chlorination yield of recovered dye (9).

the O2 concentration was maintained near atmospheric saturation levels by equilibration with a large volume of air in the reaction vessel; similarly, the O2 uptake experiments described in Figure 8 were made with relatively dilute suspensions of neutrophils to ensure that O2 was not depleted over the course of the reaction. Intraphagosomal Acidities. Examination of the excitation spectral band shapes for the unquenched fluorescence allowed determination of the intraphagosomal pH at various times following phagocytosis. Calibration curves constructed for the unopsonized fluoresceinconjugated beads in buffers were nearly identical to the curves obtained for opsonized beads that had been phagocytosed (Figure 9A). In both cases, an apparent


Figure 9. Intracellular pH changes during phagocytosis. A total of 3 × 108 opsonized beads/mL containing 5.0 µM total fluorescein were incubated at 37 °C with 2 × 107 PMN/mL in HBSS, pH 7.4, containing 1 mM NaN3. Aliquots taken at timed intervals were diluted into cold HBSS containing 35 mM MV2+ and excitation spectra immediately recorded (λem ) 523 nm): (panel A) calibration curves of the 493/433 nm absorbance ratio for fluorescein-conjugated beads in various buffers (0) and within buffer-equilibrated neutrophils (b); (panel B) pH determined from the 493/433 absorbance ratio of the excitation band for unquenched fluorescence (ordinate scaled according to data from the calibration curves). Data points are averages of three determinations; error bars are the range of values obtained.

pKa = 6.7 for the bound fluorescein was measured from the curves. The intracellular pH was found to rise slightly from an initial value of 7.4 to 7.5-7.7 over the first 15 min following phagocytosis and then slowly declined to about pH 7.0 after 60 min (Figure 9B).

Discussion Chlorination of Fluorescein-Conjugated Beads. Stopped-flow kinetic studies have established that the rate constants for formation of mono- and dichlorofluorescein from HOCl and fluorescein are k g 105 M-1 s-1 and k ) 4 × 103 M-1 s-1, respectively, at 23 °C in 25 mM phosphate, pH 7.4, containing 0.1 M NaCl (26). These constants are comparable to values obtained for the most reactive biological compounds under physiological conditions (33). Consistent with this high reactivity for fluorescein, reaction of HOCl with the dye-conjugated beads led to near-stoichiometric conversion to chlorinated derivatives. Similarly, enzymatic chlorination by the cellfree MPO-H2O2-Cl- system occurred with ∼80% efficiency with respect to the amount of added H2O2. MPO is a cationic protein that binds negatively charged surfaces, including bacterial cell walls (41). The beads used for these studies are also anionic, each containing 7-8 × 108 free carboxyl groups; presumably, this accounts for the extensive adsorption of MPO by the beads, which resembles adsorption on bacteria (41) and may be significant in directing the chlorinating agent to reactive sites on the particles. Phagocytosis of Fluorescein-Conjugated Beads. The fluoresceinated beads appear capable of eliciting the normal phagocytic response, as indicated by fluorescence microscopic examination of adhesion, 5-20-fold stimulation of cellular respiration under various conditions (Table 2), and probe chlorination, detected by both fluorescence spectral shifts (Figures 6 and 7) and HPLC of the recovered dye (Table 3). Each neutrophil was capable of binding up to ∼20 opsonized beads, although binding of unopsonized beads did not occur to any appreciable extent. Methyl viologen was found to be an

Chem. Res. Toxicol., Vol. 10, No. 10, 1997 1087

effective quencher of fluorescein fluorescence. In general, phospholipid membranes are impermeable to MV2+ (17), suggesting its possible use to determine the extent to which the neutrophil-associated beads were exposed to the extracellular environment. From microscopic observations and the data in Table 2, one estimates that about two-thirds of the fluorescein on the bound beads was inaccessible to added MV2+ in experiments where opsonized beads were mixed with neutrophils in buffer, and nearly all of the fluorescein was protected when phagocytosis was carried out in serum. In their study of intraphagosomal acidification, Cech and Lehrer made the distinction between sealed and unsealed phagosomes to account for similar quenching phenomena (37); for the experiments we conducted in buffer, this interpretation would suggest that two-thirds of the bound beads were ultimately encapsulated in sealed phagosomes. Alternatively, the partial quenching might simply reflect the condition that, on average, two-thirds of the total bead surface was in intimate contact with the neutrophil plasma membrane, from which externally added reagents were excluded (42). This distinction is irrelevant for experiments in serum because here nearly all of the dye was protected from MV2+; hence, by definition, the beads were in sealed phagosomes. Intraphagosomal Chlorination Yields. The following observations indicate that chlorination within phagosomes was much more extensive than in the external medium when beads were present in both environments: (i) at all bead/neutrophil ratios examined, the magnitude of the bathochromic shift in the emission band was greater for the unquenched fraction of the emission than the overall emission (cf. Figure 5, where λem ) 538 nm for overall emission and λem ) 542 nm for the unquenched emission after 30 min incubation); (ii) at bead/neutrophil ratios of 50/1, conditions for which maximal attachment was achieved, yet ∼60% of the beads were free in suspension (Table 2), the average amount of HOCl trapped corresponded to ∼1 HOCl/ fluorescein (Figure 8), as compared to ∼1.5 HOCl/ fluorescein when most of the particles were in sealed phagosomes; (iii) HOCl trapping efficiencies (i.e., HOCl trapped/O2 consumed) did not increase when excess beads were added to the medium; (iv) both methionine and taurine were ineffective in preventing fluorescein chlorination of opsonized beads, even when the bead/neutrophil ratio was 120/1, whereas extracellular chlorination of unopsonized beads by phorbol myristate acetatestimulated neutrophils was inhibited by ∼90% under comparable conditions (Table 3); (v) fluorescein analyses of unattached and neutrophil-bound beads separated by differential centrifugation (150 G) at various times during phagocytosis indicated that >95% of the chlorination occurred within the neutrophil-containing particulate fraction (data not shown). Points iv and v, in particular, indicate that chlorination was far more efficient on beads that were neutrophil-bound. This conclusion is consistent with cytochemical studies indicating that the MPO substrate, H2O2, is formed primarily at the sites of attachment (including the phagosomal membrane) in particle-stimulated neutrophils (42, 43). The data in Figure 7B can be used to estimate the extent of chlorination of beads when phagocytosed in human serum. In this experiment, ∼80% of the beads (or 2.4 × 108 beads/mL) were localized within sealed phagosomes. The total amount of HOCl trapped was (15 nmol/107 PMN) × (1.9 × 107 PMN/mL) ) 28 µM;


assuming that ∼90% of this total arose from reactions within the phagosome, the average number of HOCl molecules reacting with each phagocytosed bead is (0.9 × 28 nmol/mL) × (6 × 1014 HOCl/nmol)/(2.4 × 108 beads/ mL) ) 6 × 107 HOCl/bead. This number is the same as the amount of HOCl required to kill prototypical bacteria, which for Escherichia coli 25922, Pseudomonas aeruginosa 27853, and Streptococcus lactis 7962 have all been determined from titrimetric experiments to lie in the range of (4-8) × 107 HOCl/cell (44, 45). Thus, it appears that bactericidal levels of HOCl can be generated within sealed phagosomes under conditions mimicking the physiological reaction environment. In this context, it should be noted that trapping of neutrophil-generated HOCl by the fluoresceinated beads was not quantitative. This is evident from comparisons of product yields from PMA-stimulated cells, where under otherwise identical conditions the beads trapped only about 40% as much of the extracellularly generated HOCl as taurine (Table 3), and from the approximately linear dependence of chlorofluorescein yields upon the extent of fluorescein labeling (Figure 4).3 Potential alternative targets include soluble serum-based biological antioxidants (3, 33) and amines (9), as well as oxidizable amino groups within the neutrophil membrane proteins (3). Consequently, the ∼11% efficiency of conversion of O2 to HOCl measured by trapping yields is probably an unrealistic lower limit to the actual amount of bleach that is generated by stimulated neutrophils. For example, Weiss and co-workers have estimated that as much as 40% of the H2O2 formed during the respiratory burst can be trapped by taurine as the chloramine (24). From these data, one infers that intraphagosomal production of HOCl may be severalfold greater than the amount trapped by the fluorescein-conjugated beads. Temporal Correlation of Events. Under our mixing conditions, phagocytosis was complete within 10-15 min, as indicated by the progressive increase in the unquenched fraction of fluorescein fluorescence over this time period (Figures 6 and 7). After a very brief lag, comprising no more than ∼1-2 min, fluorescein chlorination commenced and paralleled phagocytosis (Figure 7); the extent of chlorination also coincided with the extent of respiratory uptake of O2 (Figure 8) and occurred on the same time scale as previously reported by Winterbourn and associates for killing of E. coli by neutrophils (46). It is often assumed that the neutrophil phagosome is acidic despite publication of several recent studies (26, 37, 47) indicating that the phagosomal environment of neutrophils remains slightly alkaline over the temporal range associated with the respiratory “burst” and bacterial killing. Confusion might arise, in part, because earlier studies (reviewed in ref 26) had erroneously interpreted MPO-catalyzed decolorization of pH indicator dyes as acidification. The present studies using N3- to inhibit MPO activity (Figure 9) indicate a slight alkanization for the first 15 min following initiation of 3 If all of the HOCl were trapped, the product yields would have “saturated”, i.e., become independent of the extent of fluorescein loading at the higher dye/bead ratios. 4 It has recently been suggested that secondary oxidants formed from HOCl and NO2-, which are capable of aromatic nitration and chlorination, may be significant neutrophil-generated antimicrobial agents (11). Nitration of dye-bound fluorescein molecules by, e.g., peroxynitrite, is readily distinguishable from chlorination (Q. Jiang, unpublished observations) and may be useful in probing for these reactions in physiological environments.

Jiang et al.

phagocytosis followed by a slow acidification, reaching only pH 6.8 after incubating for 60 min; this behavior is very similar to the other recent studies utilizing various particles (26, 37, 47). Myeloperoxidase inhibition, if anything, is expected to increase the overall rate of acidification because the sequence of events comprising glucose oxidation by O2 forming H2O2 followed by peroxide disproportionation occurs with a release of 1.0 H+/O2 consumed, but any sequences involving MPO-catalyzed formation of HOCl followed by its reaction with biological substrates and/or fluorescein have a lower overall stoichiometric release of protons (or could even involve net H+ uptake) (3). Thus, these studies clearly indicate that the intraphagosomal environment remains neutral, even when chlorination is inhibited. Role of HOCl in Bactericidal Action. Winterbourn and associates have recently concluded from measured killing rates that MPO-catalyzed reactions constitute the predominant bactericidal mechanism expressed by normal neutrophils toward Staphylococcus aureus (48). HOCl is a potent MPO-generated bactericide (1, 21) that appears to act by selective oxidative inactivation of plasma membrane-localized proteins associated with cellular energy transduction (3, 45). We have shown in these studies that the neutrophil is capable of intraphagosomal generation of HOCl in sufficient quantities to kill entrapped bacteria on a time scale that is associated with bacterial death. Others have demonstrated that HOCl can react with •O2- to give hydroxyl radical (13) and that this reaction apparently occurs in PMAstimulated neutrophils (12). However, our data give no indication of formation of hydroxylated fluoresceins, as might be expected if •OH were a major product of the respiratory burst. Thus, unless HOCl is diverted in some as-yet unidentified reaction in physiological systems,4 one must conclude that HOCl is the principal oxidative toxin produced by neutrophils.

Acknowledgment. We (Q.J. and J.K.H.) are grateful to Jian-shen Qi and Joanne Beck for preliminary investigations establishing the feasibility of these studies, to Dr. Henry Rosen (University of Washington Medical School, Seattle) for guidance in aspects involving phagocytosis and N3- poisoning of MPO function in neutrophils and for providing CGD neutrophils, and to Drs. Eberhardt Kuhn and Mohammed Aboulez (Bio-Rad Laboratories) for preparing the polyacrylamide microspheres. Financial support was provided by the National Institutes of Health under Grant AI-15834 (to J.K.H.) and Grant ES-00210 (to D.H.B.).

References (1) Klebanoff, S. J. (1988) Phagocytic cells: products of oxygen metabolism. In Inflammation: Basic Principles and Chemical Correlates (Gallin, J. I., Goldstein, I. M., and Snyderman, R., Eds.) pp 391-444, Raven Press, New York. (2) Weiss, S. J. (1989) Tissue destruction by neutrophils. N. Engl. J. Med. 320, 365-376. (3) Hurst, J. K., and Barrette, W. C., Jr. (1989) Leukocyte oxygen activation and microbial oxidative toxins. CRC Crit. Rev. Biochem. Mol. Biol. 24, 271-328. (4) Jesaitis, A. J., and Dratz, E. A., Eds. (1992) The Molecular Basis of Oxidative Damage by Leukocytes, CRC Press, Boca Raton, FL. (5) Elzanowska, H., Wolcott, R. G., Hannum, D. M., and Hurst, J. K. (1995) Bactericidal properties of hydrogen peroxide and copper or iron-containing complex ions in relation to leukocyte function. Free Radical Biol. Med. 18, 437-449. (6) Halliwell, B., and Gutteridge, J. M. (1990) Free radicals and catalytic metal ions in human disease: an overview. Methods Enzymol. 186, 1-85.

HOCl Production within Neutrophil Phagosomes (7) Chevion, M. (1988) A site-specific mechanism for free radical induced biological damage: the essential role of redox-active transition metals. Free Radical Biol. Med. 5, 27-37. (8) Wolcott, R. G., Franks, B. S., Hannum, D. M., and Hurst, J. K. (1994) Bactericidal potency of hydroxyl radical in physiological environments. J. Biol. Chem. 269, 9721-9728. (9) Thomas, E. L., Grisham, M. B., and Jefferson, M. M. (1986) Preparation and characterization of chloramines. Methods Enzymol. 132, 569-593. (10) Koppenol, W. H. (1994) Thermodynamic considerations on the formation of reactive species from hypochlorite, superoxide and nitrogen monoxide. Could nitrosyl chloride be produced by neutrophils and macrophages? FEBS Lett. 347, 3-8. (11) Eiserich, J. P., Cross, C. E., Jones, A. D., Halliwell, B., and van der Vliet, A. (1996) Formation of nitrating and chlorinating species by reaction of nitrite with hypochlorous acid. J. Biol. Chem. 271, 19199-19208. (12) Ramos, C. L., Pou, S., Britigan, B. E., Cohen, M. S., and Rosen, G. M. (1992) Spin trapping evidence for myeloperoxidase-dependent hydroxyl radical formation by human neutrophils and monocytes. J. Biol. Chem. 267, 8307-8312. (13) Candeias, L. P., Patel, K. B., Stratford, M. R. L., and Wardman, P. (1993) Free hydroxyl radicals are formed on reaction between the neutrophil-derived species superoxide anion and hypochlorous acid. FEBS Lett. 333, 151-153. (14) Tamir, S., Burney, S., and Tannenbaum, S. R. (1996) DNA damage by nitric oxide. Chem. Res. Toxicol. 9, 821-827. (15) Beckman, J. S. (1996) Oxidative damage and tyrosine nitration from peroxynitrite. Chem. Res. Toxicol. 9, 836-844. (16) Goldstein, S., and Czapski, G. (1996) Formation of peroxynitrite from the nitrosation of hydrogen peroxide by an oxygenated nitric oxide solution. Inorg. Chem. 35, 5935-5940. (17) Lymar, S. V., and Hurst, J. K. (1995) Rapid reaction between peroxonitrite ion and carbon dioxide: implications for biological activity. J. Am. Chem. Soc. 117, 8867-8868. (18) Held, A. M., and Hurst, J. K. (1978) Ambiguity associated with use of singlet oxygen trapping agents in myeloperoxidasecatalyzed reactions. Biochem. Biophys. Res. Commun. 81, 878885. (19) Zhu, L., Gunn, C., and Beckman, J. S. (1992) Bactericidal activity of peroxynitrite. Arch. Biochem. Biophys. 298, 452-457. (20) Koppenol, W. H., Moreno, J. J., Pryor, W. A., Ischiropoulos, H., and Beckman, J. S. (1992) Peroxynitrite, a cloaked oxidant formed by nitric oxide and superoxide. Chem. Res. Toxicol. 5, 834-842. (21) Lymar, S. V., and Hurst, J. K. (1995) Role of compartmentation in promoting toxicity of leukocyte-generated strong oxidants. Chem. Res. Toxicol. 8, 833-840. (22) Lymar, S. V., and Hurst, J. K. (1996) Carbon dioxide: physiological catalyst for peroxynitrite-mediated cellular damage or cellular protectant? Chem. Res. Toxicol. 9, 845-850. (23) Hurst, J. K., and Lymar, S. V. (1997) Toxicity of peroxynitrite and related reactive nitrogen species toward Escherichia coli. Chem Res. Toxicol. 10, 802-810. (24) Weiss, S. J., Klein, R., Slivka, R., and Wei, M. (1982) Chlorination of taurine by human neutrophils. Evidence for hypochlorous acid generation. J. Clin. Invest. 70, 598-607. (25) Foote, C. S., Goyne, T. E., and Lehrer, R. I. (1983) Assessment of chlorination by human neutrophils. Nature 301, 715-716. (26) Hurst, J. K., Albrich, J. M., Green, T. R., Rosen, H., and Klebanoff, S. J. (1984) Myeloperoxidase-dependent fluorescein chlorination by stimulated neutrophils. J. Biol. Chem. 259, 4812-4821. (27) Domigan, N. M., Charlton, T. S., Duncan, M. W., Winterbourn, C. C., and Kettle, A. J. (1995) Chlorination of tyrosyl residues in peptides by myeloperoxidase and human neutrophils. J. Biol. Chem. 270, 16542-16548. (28) Kettle, A. J. (1996) Neutrophils covert tyrosyl residues in albumin to chlorotyrosine. FEBS Lett. 379, 103-106.

Chem. Res. Toxicol., Vol. 10, No. 10, 1997 1089 (29) Rosen, H., Michel, B. R., and Chait, A. (1991) Phagocytosis of opsonized oil droplets by neutrophils. J. Immunol. Methods 144, 117-125. (30) Ikeda-Saito, M. (1985) Spectroscopic, ligand binding, and enzymatic properties of the spleen green hemeprotein. A comparison with myeloperoxidase. J. Biol. Chem. 260, 11688-11696. (31) Klebanoff, S. J., Waltersdorph, A. M., and Rosen, H. (1984) Antimicrobial activity of myeloperoxidase. Methods Enzymol. 105, 399-403. (32) Kettle, A. J., and Winterbourn, C. C. (1988) Superoxide modulates the activity of myeloperoxidase and optimizes the production of hypochlorous acid. Biochem. J. 252, 529-536. (33) Albrich, J. M., McCarthy, C. A., and Hurst, J. K. (1981) Biological reactivity of hypochlorous acid: implications for microbicidal mechanisms of leukocyte myeloperoxidase. Proc. Natl. Acad. Sci. U.S.A. 78, 210-214. (34) Thompson, D. H. P., Barrette, W. C., Jr., and Hurst, J. K. (1987) One-electron reduction of dihexadecyl phosphate vesicle bound viologens by dithionite ion. J. Am. Chem. Soc. 109, 2003-2009. (35) Inman, J. K. (1974) Covalent linkage of functional groups, ligands and proteins to polyacrylamide beads. Methods Enzymol. 34, 3158. (36) Ohkuma, S., and Poole, B. (1978) Fluorescence probe measurements of the intralysosomal pH in living cells and the perturbation of pH by various agents. Proc. Natl. Acad. Sci. U.S.A. 75, 3327-3331. (37) Cech, P., and Lehrer, R. I. (1984) Phagolysosomal pH of human neutrophils. Blood 63, 88-95. (38) Slavik, J. (1993) Fluorescent Probes in Cellular and Molecular Biology, CRC Press, Boca Raton, FL. (39) Haugland, R. P. (1992) Handbook of Fluorescent Probes and Research Chemicals, 5th ed., Molecular Probes, Inc., Eugene, OR. (40) Hurst, J. K. (1990) Myeloperoxidase: active site structure and catalytic mechanisms. In Peroxidases in Chemistry and Biology (Everse, J., Everse, K. E., and Grisham, M. B., Eds.) pp 37-62, CRC Press, Boca Raton, FL. (41) Klebanoff, S. J., and Clark, R. A. (1978) The Neutrophil: Function and Clinical Disorders, Elsevier/North-Holland Biomedical Press, Amsterdam. (42) Ohno, Y.-I., Hirai, K.-I., Kanoh, T., Uchino, H., and Ogawa, K. (1982) Subcellular localization of H2O2 production in human neutrophils stimulated with particles and an effect of cytochalasin-B on the cells. Blood 60, 253-260. (43) Vissers, M. C., Day, W. A., and Winterbourn, C. C. (1985) Neutrophils adherent to a nonphagocytosable surface (glomerular basement membrane) produce oxidants only at the site of attachment. Blood 66, 161-166. (44) Albrich, J. M., and Hurst, J. K. (1982) Oxidative inactivation of Escherichia coli by hypochlorous acid. Rates and differentiation of respiratory from other sites. FEBS Lett. 144, 157-161. (45) Barrette, W. C., Jr., Hannum, D. M., Wheeler, W. D., and Hurst, J. K. (1989) General mechanism for the bacterial toxicity of hypochlorous acid: abolition of ATP production. Biochemistry 28, 9172-9178. (46) Hampton, M. B., Vissers, M. C. M., and Winterbourn, C. C. (1994) A single assay for measuring the rates of phagocytosis and bacterial killing by neutrophils. J. Leuk. Biol. 55, 147-152. (47) Segal, A. W., Geisow, M., Garcia, R., Harper, A., and Miller, R. (1981) The respiratory burst of phagocytic cells is associated with a rise in vacuolar pH. Nature (London) 290, 406-409. (48) Hampton, M. B., Kettle, A. J., and Winterbourn, C. C. (1996) Involvement of superoxide and myeloperoxidase in oxygendependent killing of Staphylococcus aureus by neutrophils. Infect. Immun. 64, 3512-3517.

TX9700984

Intraphagosomal Chlorination Dynamics and Yields Determined

Recommend Documents