Fluorescence-Based Measurement of Nitric Oxide Synthase Activity in Activated Rat Macrophages Using Dichlorofluorescin

NITRIC OXIDE: Biology and Chemistry Vol. 1, No. 4, August, pp. 359–369 (1997) Article No. NO970135

ANALYTICAL METHODS Fluorescence-Based Measurement of Nitric Oxide Synthase Activity in Activated Rat Macrophages Using Dichlorofluorescin1 Amy Imrich*,2 and Lester Kobzik*,† *Physiology Program, Harvard School of Public Health, and †Department of Pathology, Brigham and Women’s Hospital, Boston, Massachusetts 02115

Received July 29, 1997

We investigated the utility of the oxidant-reactive probe dichlorofluorescin (DCFH) for measurement of NO synthase (NOS) activity in rat alveolar macrophages (AMs) activated by culture for 18 h with interferon-g (IFN-g, 25 U/ml). Using both microplatebased fluorometry and flow cytometric analysis, AMs treated with l (but not d)-arginine showed a dose- and time-dependent increase in DCFH oxidation above buffer control (e.g., DCF production (nM): control, d-arginine 100 uM, l-arginine 100 mM respectively; 91 { 9, 76 { 11, 396 { 45, mean { SE, n Å 6, 120 min). Furthermore, the NOS inhibitor (nitro-larginine) showed complete inhibition of the l-arginine-dependent DCF production. Parallel assays showed a strong correlation in DCFH oxidation with nitrite production in the same samples (e.g., DCF production (nM): 143, 222, 409; nitrite (mM): 2.2, 3.6, 7.1; for 5, 10, 50 mM l-arginine, respectively). In contrast, the alternate oxidant-reactive probes hydroethidine (HE) and dihydrorhodamine (DHR) did not report increased oxidant production in activated AMs incubated with l-arginine, despite their ability to easily detect intracellular superoxide anion production in cells treated with menadione (100 mM). We conclude that DCFH is a useful probe

1

This work was supported by NIH ES-00002 and HL 56725. To whom correspondence should be addressed at Physiology Program, Harvard School of Public Health, 665 Huntington Ave., Boston, MA 02115. Fax: 617-432-3468. 2

for quantitation of NOS-2 activity in activated rat lung macrophages. q 1997 Academic Press

Dichlorofluorescin–diacetate (DCFH-DA)3 is a sensitive and widely used probe for the detection of intracellular oxidant production (1–3). DCFH-DA freely enters the cell and becomes modified by intracellular esterases into the hydrophilic (and hence ‘‘trapped’’), nonfluorescent reporter molecule dichlorofluorescin (DCFH). Oxidation of DCFH creates the highly fluorescent dichlorofluorescein (DCF), which can be detected by flow cytometry or other fluorescence detection methods. A number of reactive oxygen species and enzymes are capable of oxidizing DCFH, including H2O2 –Fe2/-derived oxidants, and peroxidases (4, 5). Moreover, Rao et al. reported that NO can directly oxidize DCFH (3). These investigators were able to detect NO production in human neutrophils treated with W-13 to inhibit superoxide production (3). Similarly, use of DCFH to monitor 3

Abbreviations used: DCFH-DA, dichlorofluorescin–diacetate; DCF, dichlorofluorescein; NOS, nitric oxide synthase; IFN-g, interferon-g; AMs, alveolar macrophages; PBS-E, phosphate-buffered saline containing EDTA; BSS, balanced salt solution; BSA, bovine serum albumin; HE, dihydroethidium; DHR, dihydrorhodamine 123; NLA, nitro-l-arginine; NDA, nitro-d-arginine; PI, propidium iodide; GFL, green fluorescence. 359

1089-8603/97 $25.00 Copyright q 1997 by Academic Press All rights of reproduction in any form reserved.

AID

NO 0135

/

am05$$$161

11-04-97 22:08:04

noal

AP: NO

360

IMRICH AND KOBZIK

intracellular NO production in neuronal cells has been reported (6). Activated rodent macrophages produce substantial amounts of NO, often measured by detection of nitrate/nitrite products using the Griess reaction (7). Fluorescence-based assays of intracellular oxidant production offer a number of advantages, including sensitivity and potential to assay multiple fluorescent probes simultaneously (e.g., multiparameter flow cytometry) and repeatedly over time (cuvette or microplate fluorometry). In the experiments reported here, we tested the utility of DCFH-DA as a probe of nitric oxide synthase (NOS) activity in lung macrophages. To induce the high output isoform of NOS (iNOS or NOS-2), lavaged rat AMs were cultured in vitro with interferon-g (IFN-g). Initial experiments used arginine-free media to allow synchronous production of NO upon the addition of l-arginine. After 18–20 h of IFN-g treatment, stimulated cells were suspended in buffer containing DCFH-DA and treated with NOS substrate (l-arginine) and/or NOS inhibitor (nitro-l-arginine). Green fluorescence (oxidation of DCFH into fluorescent DCF) within treated cell suspensions was quantified using both plate-based fluorometry and flow cytometric analysis. We found that activated AMs caused substantial increases in DCFH oxidation and nitrite production when given l (but not d)-arginine, and that these increases were prevented by the NOS inhibitor, nitro-l-arginine. We also compared microplate-based and flow cytometric methods of fluorescence analysis.

MATERIALS AND METHODS

Buffers and Reagents Phosphate-buffered saline containing EDTA (PBSE) was prepared as 137 mM NaCl, 8 mM Na2PO4 (H2O)7 , 2.6 mM KCl, 1.5 mM KH2PO4 and 0.6 mM EDTA(Na)2 (H2O)2 . Balanced salt solution (BSS) consisted of 124 mM NaCl, 5.8 mM KCl, 10 mM dextrose, 20 mM Hepes, 0.3 mM CaCl2 (H2O)2 . Cell culture medium RPMI 1640 without l-arginine or phenol red (R0/0) was prepared from the RPMI Select Amine kit (Gibco BRL Grand Island, NY) and supplemented with antibiotics and 0.1% bovine serum albumin (BSA). DCFH-DA (Molecular Probes, Eugene, OR) was used to detect oxidants. A 2 mM

solution was prepared by dissolving 31 mg of DCFHDA powder in 1.5 ml 100% ethanol which was then stirred into 30.5 ml BSS. Aliquots were stored at 0807C. Dihydroethidium (HE) and dihydrorhodamine 123 (DHR) were additional oxidant probes used in this study (Molecular Probes). Concentrated stock solutions of l- and d-arginine were prepared at 100 mM in PBS. Nitro-l-arginine (NLA) and nitro-d-arginine (NDA, Alexis, San Diego, CA) were made at 10 mM and propidium iodide (PI) at 2 mg per milliliter, both in PBS. DCF was used for standards and was prepared as a 1 mM stock solution in 100% ethanol. Unless otherwise specified, all reagents were obtained from Sigma (St.Louis, MO).

AMs Isolation and Culture Female CD rats (Harlan Sprague Dawley Inc., Indianapolis, IN) were euthanized with a lethal intraperitoneal injection of sodium pentobarbital (75 mg/ kg). To harvest AMs, bronchoalveolar lavage was performed using PBS-E as lavage buffer (8). Lavaged cells were centrifuged at 250g for 7 min, resuspended, and adjusted to 106 cells/ml R0/0. Cytocentrifuge preparations of the lavaged cells were counterstained with Quik-Diff, a modified Wright– Giemsa stain (VWR, Boston MA), to allow differential analysis of the lavaged cells. Only samples containing greater than 95% AMs were used. Isolated AMs were stimulated in culture for 18– 20 h with recombinant rat interferon-g (Gibco BRL, No. 3283SB) used at 25 units per 106 AMs/ml R0/0 in a 6-well low-binding plate (CoStar, Cambridge, MA). For experiments which tested the effect of stimulating cells under more normal culture conditions, 10% heat-inactivated (30 min at 567C) fetal bovine serum (FBS) and 1 mM l-arginine were added to R0/ 0. Following overnight incubation, stimulated AMs appeared as large clumps of cells when viewed by light microscopy. The cells were cooled by placing the culture plates on ice for 5 min. Cell suspensions were then transferred into 40 ml cold PBS-E and held on ice for an additional 15 min to promote a single-cell suspension. AMs were centrifuged for 7 min at 250g and resuspended in cold BSS. Aliquots were diluted in trypan blue (to assess viability) and counted in a hemocytometer. Cell suspensions were adjusted to 106 live cells/ml BSS.

Copyright q 1997 by Academic Press. All rights of reproduction in any form reserved.

AID

NO 0135

/

am05$$$161

11-04-97 22:08:04

noal

AP: NO

FLUORESCENCE-BASED MEASUREMENT OF NOS ACTIVITY

Oxidation of DCFH in Cell Suspensions (96-Well Plate-Based Assay) Interferon-g-stimulated AMs were tested for their ability to metabolize l-arginine in the presence of the oxidant-reactive probe DCFH. Cell suspensions were assayed in a 96-well low binding plate and green DCF fluorescence was measured using a Cytoflour fluorescence plate reader (Millipore system 2300, excitation 485 nm/emission 530 nm). First, 101 concentrated solutions of NOS substrate (l-arginine), NOS inhibitor (NLA), isomer controls or PBS were added to replicate wells in 10 ml of volume. Next, freshly thawed 2 mM DCFH-DA was added to the cell suspension (106 cells/ml BSS kept on ice) to achieve a final concentration of 20 mM. The cold, DCFH-DA-containing cell solution was then dispensed into the 96-well low binding plate (containing treatments) at 100 ml per well. The plates were incubated at 377C and the green fluorescence in each well was measured at 30, 60, 90, and 120 min. After 2 h of incubation, all plates were frozen at 0807C for subsequent detection of nitrite levels in each well (see below). A standard curve of DCF ranging from 0 to 1 mM was generated in BSS and read at the same sensitivity settings as the plate assay. The corresponding mean fluorescence values from the known DCF standards were used to convert the cell suspension fluorescence readings into units of DCF per well. Nitrite Production The total nitrite produced by activated rat AMs in the DCFH plate assay described above (2 h time point) was measured using the Greiss method. The plates containing cell suspensions were thawed (see above) and 100 ml of Griess solution ( 1% sulfanilamide, 0.1% naphthyl(ethylene)diamine dihydrochloride, 2.5% phosphoric acid, in distilled water) was made. Sodium nitrite was diluted in BSS at concentrations ranging from 0 to 100 mM and a standard curve was generated in the empty wells of each plate (110 ml in replicate wells). After adding standards, an equal volume of Griess solution was added to all wells containing either cell suspensions or standards. Absorption at 550 nM was read after 10 min incubation at room temperature (Vmax microplate reader, Molecular Devices, Menlo Park, CA).

Oxidation of DCFH in Cell Suspensions (Flow Cytometric Assay) Interferon-g-stimulated AM suspensions were treated with arginine and/or NOS inhibitors in a microcentrifuge tube format and intracellular DCFH oxidation was determined using flow cytometry. In most experiments, cell suspensions were adjusted to 106/ml BSS containing 20 mM DCFH-DA. Cells were dispensed into 1.7-ml polypropylene microcentrifuge tubes (0.5 ml per tube) containing 50 ml of 101 concentrated l- or d-arginine, NLA, NDA, or control buffer. Microtubes containing cell suspensions and treatments were incubated at 377C for 30 min while rotating (10 rpm) to prevent adherence during the assay. To stop the assay, samples were placed on ice for 10 min in the dark. Propidium iodide (PI) was added to each tube at a final concentration of 20 mg/ ml and the relative green DCF fluorescence within live cells was measured using flow cytometry. In several experiments, stimulated AM suspensions (0.5 1 106 cells/ml) were pretreated with or without 1007 M antimycin A (a mitochondrial respiration inhibitor) for 5 min at room temperature. After pretreatment, 20 mM DCFH-DA was added and cells were aliquoted (0.5 ml) into microtubes containing 101 treatments as described above. Finally, for one set of experiments, DHR (5 mM final) and HE (1 mg/ml final) were used in place of DCFH to determine if these other fluorescent probes of intracellular oxidant production are sensitive to intracellular NOS activity. We used an Ortho 2150 Cytoflourograf equipped with a 15-mW 488-nm emitting air-cooled argon laser (Cyonics/Uniphase, Sunnyvale CA), and a Cyclops data acquisition and analysis system (Cyclops software, Cytomation, Fort Collins, CO). DCF and DHR green fluorescence was collected through a 530 bandpass filter, while PI fluorescence was measured through a 630 long pass filter. Live (PI excluding) AMs were identified by their forward- and rightangle light-scattering properties (related to cell size and granularity) and the green fluorescence (GFL) of 5000 AMs was displayed as a GFL univariate histogram, providing a measurment of relative mean fluorescence for each sample. All data were collected using a linear scale which ranged from 0 to 1024 relative fluorescent units. When analyzing cells containing HE, a red fluorescing probe, the signal was

Copyright q 1997 by Academic Press. All rights of reproduction in any form reserved.

AID

NO 0135

/

am05$$$161

11-04-97 22:08:04

361

noal

AP: NO

362

IMRICH AND KOBZIK

collected using a 630 long pass filter and PI was not added to the samples. To compare flow cytometric- and plate-based DCF fluorescence measurements, the sample volume in the flow cytometric assay described above was increased from 0.5 to 0.75 ml and cell suspensions (at 106/ml BSS) were analyzed both by flow cytometry and with a fluorescence plate reader. After 30 min of incubation (in 20 mM DCFH-DA-containing treatments), microtubes were cooled on ice for 10 min. From each sample tube we removed 250 ml of cell suspension which was centrifuged at 300g for 1 min and placed back on ice and the supernatant was collected for analysis. Aliquots (110 ml) of both the centrifuged supernatant and of the original cell suspension (cells / sup) were dispensed into a 96-well plate (plate sitting on ice) and the green DCF fluorescence was measured by a fluorescence plate reader. PI was added to the remaining sample volume still on ice and intracellular fluorescence of live AMs was determined using flow cytometry.

RESULTS

l-Arginine Dependent DCFH Oxidation by IFN-gTreated AM Suspensions We treated rat AMs with IFN-g to induce expression of the high output isoform of NOS (iNOS or NOS-2 (7)). For initial experiments, cells were cultured in arginine-free and serum-free media to optimize the detection of NOS metabolism upon addition of l-arginine. Recovery of rat AMs after culture ranged from 70 to 80% with 94–97% viability (determined by trypan blue exclusion). Stimulated cells were washed into BSS, 20 mM DCFH-DA was added, and cells were dispensed into 96-well low binding plates containing various arginine treatments or buffer control. Cell suspensions were incubated at 377C and the green fluorescence (DCFH oxidation to DCF) within each well was measured at 30-min intervals for a period of 2 h. Figure 1 illustrates the dose response and time course of DCF production by AM suspensions provided with varying amounts of l-arginine substrate. Cells treated with l-arginine showed a dose- and time-dependent increase in DCFH oxidation above control buffer. Furthermore, suspensions treated with the isomer d-arginine at 100 mM showed no

FIG. 1. Oxidation of DCFH by IFN-g-stimulated rat AMs using a microplate-based fluorometric assay. Cells treated with 5 mM (s), 10 mM (n), and 100 mM (l) l-arginine (isomer-specific NOS substrate) for 2 h showed a dose- and time-dependent increase in DCF production per well, while the isomer control d-arginine used at 100 mM (m) was equal to buffer control (L). Green fluorescence (GFL) measurements were collected at 30-min intervals for 2 h (X axis) and were converted to units of DCF per well (Y axis) using linear regression analysis from a standard curve created from wells containing known DCF concentrations in buffer. (Data points are means { SE, n Å 3–6 experiments performed using duplicate wells.)

increase, indicating that the l-arginine response is stereospecific. The DCF produced in wells treated with 50 and 500 mM l-arginine (data not shown) followed the same pattern as wells given 100 mM larginine. An expanded dose–response experiment is presented in Fig. 2. Using the plate based assay, oxidation of DCFH after 2 h (Fig. 2A) was dose dependent in the range of 5 to 50 mM l-arginine treatment. Higher concentrations of l-arginine (100 or 500 mM) did not result in correspondingly higher amounts of DCFH oxidation. Cells treated with 500 mM d-arginine showed only a minimal DCFH oxidation equal to buffer control. Also shown in Fig. 2A, the NOS inhibitor NLA showed potent inhibition of the l-arginine-dependent DCF production (e.g., 86% at 50 mM, 100% at 500 mM). In contrast, adding the isomer control NDA (500 mM) caused no significant inhibition of the DCF produced in response to 100 mM l-arginine, indicating that the DCFH oxidation

Copyright q 1997 by Academic Press. All rights of reproduction in any form reserved.

AID

NO 0135

/

am05$$$161

11-04-97 22:08:04

noal

AP: NO

FLUORESCENCE-BASED MEASUREMENT OF NOS ACTIVITY

363

FIG. 2. Comparison of DCFH oxidation and nitrite production by stimulated AMs incubated with DCFH using the microplate-based assay (one representative experiment of four is shown). The amount of DCF (2A) produced per well after 2 h increases as the dose of l-arginine increases up to 50 mM. The NOS inhibitor NLA (500 mM) was able to completely reduce the 100 mM l-arginine response back to control values, while its isomer control NDA had no inhibitory effect on l-arginine-dependent DCFH oxidation. In addition to GFL measurments, nitrite production (B), a separate indicator of NOS activity, was determined for each well after 2 h incubation using the Greiss method and shows a strong correlation with the DCF produced in the same wells.

Copyright q 1997 by Academic Press. All rights of reproduction in any form reserved.

AID

NO 0135

/

am05$$0135

11-04-97 22:08:04

noal

AP: NO

364

IMRICH AND KOBZIK

caused by the metabolism of l-arginine is NOS specific. Correspondence of DCFH Oxidation with Nitrite Production To further examine the contribution of NO in the oxidation of DCFH described above, we measured nitrite production (a separate indicator of NOS activity) in the wells of the DCFH plate assay shown in Fig. 2A. After measuring the green fluorescence at 2 h (the last time point), the plates were immediately frozen (to lyse cells) and later thawed for nitrite detection using the Greiss method. The results are shown in Fig. 2B. Similar to DCFH oxidation, nitrite levels were dependent on the dose and stereoisomer of arginine given. When comparing DCFH oxidation with nitrite production, a strong correlation in the magnitude of response was observed with 5, 10, and 50 mM l-arginine treatment. However, with concentrations higher than 50 mM, DCF production remained constant, while nitrite production continued to increase. In wells treated with NLA (but not NDA) and 100 mM l-arginine together, nitrite production was greatly diminished, corresponding with the inhibitory effect NLA had on DCFH oxidation in the same wells (shown in Fig. 2A). Flow Cytometric Analysis of DCF Fluorescence Activated rat AMs suspensions (in BSS containing 20 mM DCFH-DA) were given arginine treatments in a microcentrifuge tube format. After 30 min of rotation at 377C, cell-associated GFL of live cells was measured using flow cytometry. Cell viability after the assay (determined as the percentage of PI-negative cellular events) ranged from 88 to 93%. Figure 3 is a set of GFL histograms from one representative experiment. As increasing concentrations of l-arginine were given, the relative fluorescence intensity curve shifted to the right (fluorescence increased) and remained bell-shaped, suggesting a uniformly responsive cell population. As in the plate-based assay, intracellular DCF fluorescence was maximal with 50 mM l-arginine treatment, while cells given isomer control (100 mM d-arginine) were equal to buffer control. When cells were incubated with both 100 mM l-arginine and 500 mM NLA, DCF fluorescence was completely inhibited, while 500 mM NDA

had no effect on l-arginine-dependent DCFH oxidation. These data demonstrate that flow cytometric analysis can also be used to measure relative NOS activity in activated AM populations. Comparison of Flow Cytometric- and Plate-Based Analysis We have presented data obtained using a fluorescence plate reader and a flow cytometer, two different methods of cell suspension analysis. In order to better understand and compare these methods of analysis, we used the microcentrifuge tube format and performed both flow cytometric analysis and microplate-based analysis on the same cell suspensions. We gave stimulated AMs (in 20 mM DCFHDA) 100 mM l-arginine, or control buffer for 30 min (with rotation, 10 rpm) and measured GFL of these samples both by flow cytometry and by fluorometry in a 96-well plate. We also removed the cells from suspension by centrifugation and measured the fluorescence contained in the buffer supernatant. Table I shows the results from one representative experiment. Using fluorometry, we found 70–90% of the total fluorescence measured (105 cells in 110 ml) to actually be in the extracellular buffer. Furthermore, the GFL fold increase (over control) caused by l-arginine treatment was about 35% higher in cells plus supernatant (4.7-fold increase) than that measured by flow cytometry (3.5-fold increase). Since flow cytometry measures only cell-associated fluorescence, it is noteworthy that the fold increase of GFL in cells alone (Table I) determined by simple subtraction (3.6-fold increase) was very close to the flow cytometric result (3.5-fold increase). Culture Conditions during IFN-g Treatment Affect l-Arginine Responses The medium used in the above experiments (R0/ 0) is not commonly used in routine cell culture. Hence, we investigated the effect of more typical culture conditions on the ability of DCFH to detect NOS activity using the plate-based assay with multiple readings, including short (30 min) and long (2 h) incubation periods. We pretreated AMs with IFN-g in R10 (R0/0 to which 1 mM l-arginine, the usual concentration in RPMI, and 10% FBS were added, or R0/0 alone. After 18–20 h of treatment, cultured

Copyright q 1997 by Academic Press. All rights of reproduction in any form reserved.

AID

NO 0135

/

am05$$$161

11-04-97 22:08:04

noal

AP: NO

FLUORESCENCE-BASED MEASUREMENT OF NOS ACTIVITY

365

FIG. 3. Flow cytometric analysis of activated rat AMs. AMs in suspension (containing 20 mM DCFH-DA) were given arginine treatments for 30 min in a microcentrifuge tube format and relative cell-associated green fluorescence (GFL) was measured using flow cytometry. This figure shows eight univariate histograms (cell number (Y axis) versus GFL intensity channel (X axis)) from one representative experiment showing the dose response and inhibition of l-arginine-dependent intracellular DCFH oxidation.

AMs (in BSS with 20 mM DCFH-DA) were treated with 100 mM l-arginine or control buffer. At both time points, cells cultured in R10 showed increased baseline (control) DCF production and oxidized less DCFH when treated with l-arginine (Fig. 4). DCF production by cells in R10, as found in R0/0, was inhibited by NLA (not by NDA), both at baseline and upon addition of l-arginine (data not shown). Nitrite production after 2 h (data not shown) also showed diminished NOS activity of cells cultured in R10. TABLE I

Comparison of Microplate-Based and Flow Cytometric Green Fluorescence Measurements Fluorescence plate reader

Flow cytometer: Treatment

Cells

C L100 D100

92 328 106

Cells / Supernatant Cells alone supernatant only (subtraction) 430 2031 524

322 1635 390

108 396 134

Note. Activated AM suspensions (in 20 mM DCFH-DA) were treated with 100 mM arginine or buffer control in microcentrifuge tubes for 30 min. Fluorescence measurements of the complete suspension (cells / supernatant) and the cell-free supernatant show that a substantial amount of fluorescence residues in the buffer (110 ml per well). Cell-associated fluorescence per well was calculated by subtraction of the supernatant only from the cells / supernatant and was very similar to the cell-associated GFL measured using flow cytometry.

The Source of l-Arginine-Dependent DCFH Oxidation Is Partially Mitochondrial NO produced within activated macrophages could potentially oxidize DCFH either directly (3) or indirectly (e.g., after combining with superoxide anion to form peroxynitrite (9), see Discussion). Because mitochondrial respiration is one source of superoxide anion within the cell, we tested the effect of antimycin A, a mitochondrial respiration inhibitor, on l-arginine-dependent DCFH oxidation. In addition, since we had seen a decrease in the detection of NOS activity in cells stimulated in R10 (Fig. 4A), we compared the effects of antimycin A treatment within AMs cultured in R0/0 versus R10. Using the flow cytometric assay, we measured intracellular DCFH oxidation during the first 30 min after cells were removed from culture (Fig. 5). When cells cultured in R0/0 were pretreated with antimycin A, there was a significant albeit incomplete reduction of l-arginine-dependant DCFH oxidation (33 { 7%, n Å 9). It is worth noting that we observed some unexplained variability from animal to animal in the proportion of DCFH oxidation that was antimycin-sensitive (i.e., 2 of the 9 animals studied in Fig. 5 showed õ10% inhibition, while the values for the remaining 7 were substantially higher). This result suggests that at least some of the NO-mediated DCFH oxidation observed within AMs occurs indirectly, possibly via peroxynitrite derived from mitochondrial 020 and

Copyright q 1997 by Academic Press. All rights of reproduction in any form reserved.

AID

NO 0135

/

am05$$$161

11-04-97 22:08:04

noal

AP: NO

366

IMRICH AND KOBZIK

FIG. 4. Presence of serum and l-arginine during IFN-g activation of AMs alters subsequent DCFH oxidation. DCF production was compared in AMs cultured in R0/0 (RPMI without l-arginine or serum) vs R10 (RPMI containing the standard 1 mM l-arginine and 10% serum). While NOS activity is still detected with DCFH, AMs cultured in R10 show increased basal oxidation (open bars) and somewhat diminished l-arginine-dependent oxidation (shaded bars) at both 30 (A) and 120 min (B) of observation (n Å 3–6, mean { SE).

NO. When cells were IFN-g-stimulated in R10, both basal and l-arginine-dependent intracellular DCFH oxidation was diminished (Fig. 5). In contrast, microplate-based measurement of cell suspensions (and their surrounding buffer) showed increased DCFH oxidation at both 30 and 120 min (Fig. 4, see

also Discussion). Addition of antimycin had no effect on the increase in nitrite production seen upon addition of l-arginine (data not shown). We also measured nitrite levels in the 18-h culture supernatants and found that R0/0 contained 1.4 mM nitrite ({ 0.1 mM) while cells incubated in R10 (containing ample NOS substrate) showed evidence of high NOS activity (23.8 mM nitrite { 3.1 mM). Comparison of Fluorescent Probes for Detecting NOS Activity in AMs

FIG. 5. Effect of antimycin A on intracellular DCFH oxidation in activated AMs. Flow cytometric analysis of AMs activated with IFN-g in R0/0 after 30 min incubation with 10 mM l-arginine (open bars) shows that 33 { 7% of DCF production was inhibited by antimycin treatment. In contrast to results shown in Fig. 4, AMs activated with IFN-g in R10 (containing the standard 1 mM l-arginine and 10% serum) showed diminished basal (shaded bars) and l-arginine stimulated intracellular DCFH oxidation, both of which were less sensitive to the mitochondrial respiration inhibitor antimycin A (see Discussion). n Å 3–9, mean { SE.

After exploring the utility of DCFH in the detection of NOS activity, we compared two additional fluorescent probes, DHR and HE. Both probes are thought to be more reactive with superoxide anion than DCFH (10, 11). We used the flow cytometric assay (cells at 500K/ml BSS) and compared the fluorescence of DCFH, HE, and DHR when stimulated rat AMs were treated with control buffer, 100 mM larginine, or 100 mM menadione (which causes intracellular production of superoxide anion). In Fig. 6 we show that, unlike DCFH, both DHR and HE are not able to detect any oxidation above control levels when cells are given l-arginine. In contrast, both DHR and HE are very reactive with the oxidants (020) supplied by menadione, while DCFH was unable to detect this source of oxidant production within AMs. DISCUSSION

We sought to investigate the utility of DCFH for detection of NOS activity in macrophages. This fluo-

Copyright q 1997 by Academic Press. All rights of reproduction in any form reserved.

AID

NO 0135

/

am05$$$161

11-04-97 22:08:04

noal

AP: NO

FLUORESCENCE-BASED MEASUREMENT OF NOS ACTIVITY

FIG. 6. Flow cytometric analysis of intracellular oxidant production using dichlorofluorescin (DCFH), hydroethidine (HE), and dihydrorhodamine (DHR). Rat AMs activated with IFN-g in R0/0 were treated with control buffer, 100 mM l-arginine, or 100 mM menadione (source of superoxide) in the presence of the three oxidant-sensitive probes for 30 min and then analyzed. Menadione treatment caused substantial increases in intracellular fluorescence in AMs containing HE and DHR, but not DCFH. In contrast, DCFH showed l-arginine-dependent oxidation, while HE and DHR did not (n Å 3, mean { SE).

rescent probe of cellular oxidant production offers the potential advantages of high sensitivity and use in conjunction with other fluorescent markers or functional assays. The microplate-based format allows for multiple, real-time measurements in cell suspensions, while flow cytometric analysis is most useful for determining total intracellular DCF production at one time point. Our experiments were stimulated by previous investigations demonstrating the effective use of DCFH for assay of macrophage oxidative metabolism (2) and for assay of NOS activity in neutrophils (3) and neuronal cells (6). Using rat AMs that were activated with IFN-g, a potent inducer of the high output, calcium-independent isoform of NOS (iNOS, NOS2), we have observed a dose-dependent oxidation of DCFH (increased fluoresence) in response to l-arginine treatment (figure 1). This response required stereospecific l-arginine, not d-arginine, and was completely reversible with the NOS inhibitor NLA. Furthermore, nitrite production corresponded very closely with DCF production when cells were treated with 5, 10, or 50 mM l-arginine (Fig. 2), supporting the validity of using DCFH to measure NOS activity in stimulated AMs. We observed an increase in DCFH oxidation over time with 50 mM l-arginine treatment (Fig. 1); yet, for each time point measured, DCF production had reached a plateau at this concentration of substrate. In contrast, nitrite production (measured at 2 h) continued to increase with 100 and 500 mM l-arginine treatment (Fig. 2). The basis for this plateau in DCF production is unclear. One possibility is NO-dependent oxidation of DCFH requires a constantly produced and yet limited supply of another component within the cell which enables NO to react with DCFH. One such candidate is suggested by the results showing inhibition of NO-dependent DCFH oxidation by the mitochondrial respiration inhibitor antimycin A (Fig. 5). We postulate that mitochondrial respiration within the cell provides superoxide anion (020) which can react with NO to form peroxynitrite, which in turn, directly or through other derivatives, oxidizes DCFH. Since DHR reacts with peroxynitrite to form the fluorescent rhodamine reporter (12), this postulate is called into question by our observation of no increased fluorescence in activated AMs provided larginine, despite demonstration of adequate loading

Copyright q 1997 by Academic Press. All rights of reproduction in any form reserved.

AID

NO 0135

/

am05$$$161

11-04-97 22:08:04

367

noal

AP: NO

368

IMRICH AND KOBZIK

and sensitivity by use of the positive control menadione. One possible explanation is suggested by the observation that under some conditions (concentration-dependent), NOx can reduce rhodamine back to DHR and confound the fluorescence-based results using this probe (13). Alternatively, the absence of larginine-dependent HE and DHR oxidation suggests that the NO-derived oxidant species which can react with DCFH inside the cell are either chemically or spatially unable to react with the former probes. Further experiments are needed to more precisely identify the oxidant species that mediate DCFH oxidation in NO-producing cells. There is substantial evidence that prolonged NO exposure can downregulate mitochondrial respiration (14, 15). We have shown that mitochondrial activity contibutes to l-arginine-dependent DCFH oxidation (Fig. 5) and at the same time found a dramatic decrease in intracellular DCF fluorescence after cells have been subjected to culture conditions (R10, containing 1 mM l-arginine and 10% serum) where they are stimulated to produce NO for a period of time (Fig. 5). An additional possible explanation for this decreased responsiveness is the finding that NO can itself inhibit NOS (16). Finally, on a technical note, when using DCFH to compare different levels of NOS activity, conditions must be such that the amount of NO produced remains within a sensitive range of DCFH detection (preliminary l-arginine dose–response experiments) and, second, this range of sensitivity may be partially determined by the level of mitochondrial respiration within the cell. We found that both plate-based fluorometry and flow cytometric analysis can be used to detect NO production within activated AMs cultured in R0/0. Both assays showed a plateau of DCFH oxidation at 50 mM l-arginine treatment, stereospecificity of responses, and complete inhibition with NLA. However, we should also consider the inherent differences between these methods. The most significant is that flow cytometry measures only cell associated fluorescence, while the plate format measures cells and their surrounding buffer. Analysis of the fluorescence signal in cell suspension samples with or without cells (Table I) shows that the majority of fluorescence measured in cell suspensions is actually in the buffer, similar to results reported for studies of human monocytes and DCFH (17). The basis for the presence of extracellular DCF is unclear. One

possibility is the diffusion of intracellular DCF out of the cell and into the buffer. We consider this unlikely in view of experiments where incubation of rat AMs in solutions containing high concentrations of extracellular DCF resulted in virtually no intracellular fluorescence (data not shown). More likely, oxidants produced at the cell surface by plasma membrane oxidases or released from the cell could potentially react indirectly with extracellular probe. We urge investigators to be aware that the results obtained using this method on cell suspensions in a microplate format may potentially be different from what the strictly intracellular results obtained with flow cytometry. This is especially important for situations wherein cells are studied after exposure to conditions (serum, arginine in the media) where NOS activity may result in impaired mitochondrial respiration (and diminished intracellular oxidant production), while leaving the ability of cells to oxidize extracellular DCFH relatively intact (Figs. 4 and 5). This concern is illustrated by the results in Fig. 5 where cells cultured in R10 show a lower baseline oxidation (intracellular fluorescence) than cells cultured in R0/0, which we speculate is due at least in part to impaired mitochondrial metabolism or other alterations in cell phenotype during culture in R10. Although buffer fluorescence is the main component of plate based measurements, this method still has many advantages. Plate-based assays allow repeated fluorescence readings over long incubation periods and allow quick analysis of numerous samples and replicates. Also, by using a DCF standard curve, mean GFL measurements read at multiple GFL sensitivity settings can be effectively compared. Finally, the subtraction method of determining cell associated fluorescence in a plate-based assay (Table I) allows reasonable estimation of oxidant production within the cell as measured by flow cytometry. This approach can also be useful in adherent cell systems where creating a cell suspension for flow cytometric analysis could potentially affect the intracellular oxidative state and subsequent results. When cells are naturally living in suspension, flow cytometric analysis is the preferred way to measure cell-associated fluorescence. The multiparameter, single-cell analysis afforded by flow cytometry provides a powerful tool which will enable fluorescence

Copyright q 1997 by Academic Press. All rights of reproduction in any form reserved.

AID

NO 0135

/

am05$$$161

11-04-97 22:08:04

noal

AP: NO

FLUORESCENCE-BASED MEASUREMENT OF NOS ACTIVITY

measurements to be combined with other functional analysis.

REFERENCES

9.

10.

1. Szejda, P., Parce, J. W., Seeds, M. S., and Bass, D. A. (1984). Flow cytometric quantitation of oxidative product formation by polymorphonuclear leukocytes during phagocytosis. J. Immunol. 133, 3303–3307. 2. Kobzik, L., Godleski, J., and Brain, J. (1990). Oxidative metabolism in the alveolar macrophage: Analysis by flow cytometry. J. Leukocyte Biol. 47, 295–303. 3. Rao, K., Padmanabhan, J., Kilby, D., Cohen, H., Currie, M., and Weinberg, J. (1992). Flow cytometric analysis of nitric oxide production in human neutrophils using dichlorofluorescein diacetate in the presence of a calmodulin inhibitor. J. Leukocyte Biol. 51, 496–500.

11.

12.

13.

4. LeBel, C., Ischiropoulos, H., and Bondy, S. (1992). Evaluation of the probe 2*,7*-dichlorofluorescin as an indicator of reactive oxygen species formation and oxidative stress. Chem. Res. Toxicol. 5, 227–231.

14.

5. Zhu, H., Bannenberg, G., Moldeus, P., and Shertzer, H. (1994). Oxidation pathways for the intracellular probe 2*,7*dichlorofluorescin. Arch. Toxicol. 68, 582–587.

15.

6. Gunasekar, P., Kanthasamy, A., Borowitz, J., and Isom, G. (1995). Monitoring intracellular nitric oxide formation by dichlorofluorescin in neuronal cells. J. Neurosci. Methods 61, 15–21. 7. Nathan, C. (1992). Nitric oxide as a secretory product of mammalian cells. FASEB J. 6, 3051–3064. 8. Kobzik, L. (1997). Methods to isolate and study lung macrophages. In Lung Macrophages and Dendritic Cells in Health

16.

17.

and Disease (Lipscomb, M., and Russell, S., Eds.), pp. 111– 129, Dekker, New York. Pryor, W., and Squadrito, G. (1995). The chemistry of peroxynitrite: A product from the reaction of nitric oxide with superoxide. Am. J. Physiol. 268, L699–L722. Rothe, G., and Valet, G. (1987). Use of hydroethidine (HE) and 2,7-dichlorofluorescin (DCFH) for the flow-cytometric measurement of NADPH-oxidase and mitochondrial oxygen radical formation in phagocytes. Cytometry (Suppl.) 1, 77. Rothe, G., Emmendorffer, A., Oser, A., Roesler, J., and Valet, G. (1991). Flow cytometric measurement of the respiratory burst activity of phagocytes using dihydrorhodamine 123. J. Immunol. Methods 138, 133–135. Kooy, N., Royall, J., Ischiropoulos, H., and Beckman, J. (1994). Peroxynitrite-mediated oxidation of dihydrorhodamine 123. Free Radical Biol. Med. 16, 149–156. Szabo, C., Salzman, A., and Ischiropoulos, H. (1995). Peroxynitrite-mediated oxidation of dihydrorhodamine 123 occurs in early stages of endotoxic and hemorrhagic shock and ischemia–perfusion injury. FEBS Lett. 372, 229–232. Brown, G., and Cooper, G. (1994). Nanomolar concentrations of nitric oxide reversibly inhibit synaptosomal respiration by competing with oxygen at cytochrome oxidase. FEBS Lett. 356, 295–298. Stadler, J., Billiar, T., Curran, R., Stuehr, D., Ochoa, J., and Simmons, R. (1991). Effect of exogenous and endogenous nitric oxide on mitochondrial respiration of rat hepatocytes. Am. J. Physiol. 260, C910–C916. Griscavage, J., Rogers, N., Sherman, M., and Ignarro, L. (1993). Inducible nitric oxide synthase from a rat alveolar macrophage cell line is inhibited by nitric oxide. J. Immunol. 151, 6329–6337. Robinson, J. P., Bruner, L. H., Bassoe, C. F., Hudson, J. L., Ward, P. A., and Phan, S. H. (1988). Measurement of intracellular fluorescence of human monocytes relative to oxidative metabolism. J. Leukocyte Biol. 43, 304–310.

Copyright q 1997 by Academic Press. All rights of reproduction in any form reserved.

AID

NO 0135

/

am05$$$161

11-04-97 22:08:04

369

noal

AP: NO