- Email: [email protected]
Detection of phagocyte-derived free radicals with spin trapping techniques: Effect of temperature and cellular metabolism
236
Biochimica etBiophysicaActa 969 (1988) 236-241
Elsevier BBA12233
Detection of phagoctye-derived free radicals with spin trapping techniques: effect of temperature and cellular metabolism G e r a l d M. R o s e n a,b, B r a d l e y E. B r i t i g a n c,., M y r o n S. C o h e n S h a r o n P. E l l i n g t o n a,b a n d M i c h a e l J. B a r b e r e
c,d
" Department of Pharmacology, Duke University Medical Center, and b GRECC, Veterans Administration Medical Center, Durham, NC, Departments of " Medicine and a Microbiology and Immunology, University of North Carolina at Chapel Hill, Chapel Hill, NC and e Department of Biochemistry, University of South Florida College of Medicine, Tampa, FL (U.S.A.)
(Received 16 October 1987) (Revised manuscript received 1l January 1988)
Key words: Free radical; Superoxide generation; Phagocyte stimulation; ESR; Temperature dependence; (Human)
Human neutrophils activated with either particulate or soluble stimuli generate oxygen-centered free radicals which are detected by spin trapping in conjunction with electron spin resonance (ESR) spectroscopy. We investigated the effect of temperature on ESR spectra resulting from stimulation of human neutrophils with phorbol myristate acetate (PMA) or opsonized zymosan in the presence of the spin trap, 5,5-dimethyl-l-pyrroline 1-oxide (DMPO). At 20°C with either stimuli, neutrophil superoxide production was manifested predominantely as the superoxide spin-trapped adduct, 5,5-dimethyi-5-hydroperoxy-l-pyrrolidinyloxy (DMPO-OOH). In contrast, at 37°C, the hydroxyl spin-trapped adduct, 2,2-dimethyl-5-hydroxy-l-pyrrolidinyloxy (DMPO-OH) was dominant. No evidence of hydroxyl radical (defined as the methyl spin-trapped adduct, 2,2,5-trimethyl-l-pyrrolidinyloxy, DMPO-CH 3) was observed, suggesting that elevated temperatures increased the rate of DMPO-OOH conversion to DMPO-OH. In addition, the elevated temperature activated a neutrophil reductase which accelerated the rate of D M P O - O H reduction to its corresponding hydroxylamine, 2,2-dimethyl-5-hydroxy-l-hydroxypyrrolidine. This bioreduction was dependent upon the presence of both superoxide and a phagocyte-derived factor (possibly a thiol) released into the surrounding media.
* Present address: Department of Medicine, University of Iowa, Iowa City, IA 52242, U.S.A. Abbreviations: PMA, 4-phorbol 12-myristate 13-acetate; DMPO, 5,5-dimethyl-l-pyrroline 1-oxide; DMPO-OOH, 5,5dimethyl-5-hydroperoxy-l-pyrrolidinyloxy; DMPO-OH, 2,2dimethyl-5-hydroxy-l-pyrrolidinyloxy; DMPO-CH 3, 2,2,5-trimethyl-l-pyrrolidinyloxy; DTPA, diethylenetriaminepentaacetic acid; TEMPO, 2,2,6,6-tetramethylpiperidinoxy; DMSO, dimethyl sulfoxide; HBSS, Hanks' balanced salt solution. Correspondence: G.M. Rosen, Department of Pharmacology, Duke University Medical Center, Durham, NC 27710, U.S.A.
Introduction In response to soluble or particulate stimuli, human neutrophils generate superoxide as the result of the activation of an NADPH-oxidoreductase [1]. Subsequent dismutation yields hydrogen peroxide. In the presence of ferric ions, superoxide and hydrogen peroxide produce hydroxyl radical via a metal-catalyzed Haber-Weiss reaction [2]. Although formation of hydroxyl radical by activated neutrophils has been previously reported
0167-4889/88/$03.50 © 1988 Elsevier Science Publishers B.V. (Biomedical Division)
237 [3-8], the techniques used are not without problems and limitations. Consequently, we recently undertook an extensive investigation of neutrophil free radical production using spin trapping techniques [9]. When either PMA or opsonized zymosan were used to stimulate neutrophils at room temperature (25 ° C), we only spin trapped super° oxide, even although, under some experimental conditions, we observed the hydroxyl radical spin-trapped adduct, DMPO-OH [9,10]. Our findings suggested that for previous investigations [6-8], the purported spin trapping of hydroxyl radical may actually be incorrect. However, several earlier studies [6-8] differed from ours in that they were performed with neutrophils incubated at 37°C. This suggested two possibilities: (a) ESR spectra resulting from spin trapping of neutrophil-derived superoxide vary with temperature or (b) hydroxyl radical formation by neutrophils occurs at 37°C, but not at 25°C. In this present study, we monitored the spin trapping of neutrophil-generated free radicals at 20 °C and 37 ° C. Our data demonstrate that at 37°C, decomposition of DMPO-OOH is rapid, however, we do not find any evidence for the formation of hydroxyl radical at this elevated temperature. Materials and Methods
Reagents. Diethylenetriaminepentaacetic acid (DTPA), hypoxanthine, superoxide dismutase and zymosan A were purchased from Sigma Chemical Company (St. Louis, MO). PMA and xanthine oxidase were obtained from Midland Chemical Company (Brewster, NY) and Calbiochem, a division of Behring Diagnostics, Inc. (La Jolla, CA). Succinylated cytochrome c was prepared as described earlier [11]. 2,2,6,6-Tetramethylpiperidinoxy (TEMPO) was synthesized according to procedures outlined by Rozantsev [12]. DMPO was prepared by methods described in Bonnett et al. [13]. PMA was dissolved in DMSO at a concentration of 10 ~tg/ml. Zymosan A was opsonized with normal human serum prior to use. Neutrophil isolation. Neutrophils were isolated from heparinized blood derived from normal human volunteers as previously described [14]. Leukocytes were separated from erythrocytes by
the addition of Plasmagel (Roger Bellon, Neuilley, France). The resulting leukocyte suspension was centrifuged through a Ficoll-Hypaque gradient (Pharmacia Chemicals, Piscataway, N J). Erythrocytes contaminating the neutrophil layer were removed by hypotonic lysis. Neutrophils were then suspended in Hanks' balanced salt solution (HBSS) and the concentration was adjusted as determined by a Coulter Model D2N automated blood cell counter (Coulter Electronics, Hialeah, FL). This procedure yielded cells which were more than 98% neutrophils (Giemsa stain) and more than 95% viable (Trypan blue exclusion). Cell-free supernatant. Neutrophils were suspended in HBSS to 4. 107/ml and stimulated with PMA, if required, for 30 rain at room temperature, pelleted (300 × g, 10 rain) and the resulting supernatant was removed for additional experiments. Glutathione detection. Concentration of reduced and oxidized glutathione in neutrophil-derived cell-free supernatants was determined by HPLC analysis. Samples were prepared for analysis by the method of Livesey and Reed [15], and analyzed as described by Chaney and Spector [16] except that the sample was eluted with a 14 min gradient (Gradient Curve 8, Watts Model 660 Solvent Programmer) from 75% A, 25% B to 25% A, 75% B. Neutrophil superoxide release. Neutrophil superoxide release as a consequence of PMA stimulation was determined by measuring the rate of superoxide dismutase-inhibitable reduction of succinylated cytochrome c [11]. Spin trapping/ESR spectroscopy studies. Spin trapping of free radicals was undertaken by mixing neutrophils (1. 107/ml), DMPO (0.1 M), DMSO (0.14 M), DTPA (0.1 mM) and sufficient buffer (HBSS) to reach a final volume of 0.5 ml. Stimulation of neutrophil oxygen reduction was mediated by the addition of either PMA (100 ng/ml) or opsonized zymosan A (3 or 6 mg/ml). Reaction mixtures were transferred to an ESR quartz cell and placed into the cavity of an ESR spectrometer (Varian Associates, Palo Alto, CA, model E-9), equipped with a variable temperature accessory. Sample temperature was determined by the inclusion of a thermistor in the spectrometer cavity. In some experiments, a superoxide-generating system consisting of hypoxanthine (0.1 mM) and xanthine oxidase (0.01 U/ml) such that the
238
rate of superoxide dismutase-inhibitable reduction of succinylated cytochrome c was 10 /~M/min at p H 7.8, was substituted for stimulated neutrophils. The rate of T E M P O reduction was determined by monitoring the change in amplitude of the lowfield peak of the nitroxide triplet with time. ESR spectra were routinely obtained using a microwave power of 20 mW, a modulation amplitude of 1 G, a field sweep rate of 12.5 G / m i n , a time constant of I s, and a receiver gain of 5 • 103, except where noted. In a series of experiments, we examined the effect of D M P O (0.1 M), D M S O (0.14 M) and D T P A (0.1 mM) on superoxide production at 2 0 ° C and 37°C. At these concentrations, we observed no inhibition in the generation of superoxide, as monitored by superoxide dismutase-inhibitable reduction of succinylated cytochrome c. These results are identical to results from our earlier studies [9]. Results
and Discussion
When human neutrophils were stimulated with P M A in D M S O (0.14 M) or opsonized zymosan at 20 ° C, the spectra shown in Fig. 1 were recorded. For PMA, the ESR spectrum was primarily D M P O - O O H and to a lesser extent D M P O - O H (Fig. 1A). "The ESR spectrum for D M P O - O O H results from the spin trapping of superoxide by DMPO, whereas D M P O - O H may arise as the result of either the spin trapping of hydroxyl radical by D M P O or the degradation of D M P O O O H to D M P O - O H [17,18]. However, the detection of D M P O - C H 3 is more specific than D M P O O H for the formation of hydroxyl radical, since this free radical reacts with D M S O at diffusioncontrolled rates to generate methyl radical ("CH 3) [19], which can then be spin trapped. A small quantity (approx. 3%) of D M P O - O O H decomposes to give hydroxyl radical, which can then be spin trapped as D M P O - C H 3 [17]. Thus, detection of a concentration of D M P O - C H 3, in the range described above, is not necessarily the result of cell-mediated hydroxyl radical production, but rather arises as an artifact of superoxide spin trapping by DMPO. In the case of opsinized zymosan (Fig. 1B and C), two different ESR spectra were recorded at 20 ° C , depending on the concentration of
lOG
I 2
I I
A.
B.
C•
I 2
i 2
2
2~
i
Fig. 1. ESR spectra resulting from neutrophil (107/ml) stimulation at 2 0 ° C in the presence of D M P O (0.1 M), D T P A (0.1 mM) and D M S O (0.14 M). Stimufi were P M A (100 n g / m l , scan A), opsinized zymosan (3 m g / m l , scan B) and opsinized zymosan (6 m g / m l , scan C). Low- and high-field peaks corresponding to D M P O - O H and D M P O - O O H are designated 1 and 2, respectively. See Materials and Methods for instrumentation settings: scans were, however, initiated 1 min after the addition of the stimulus to the neutrophils.
opsinized zymosan added to the cell suspension. At low concentrations of opsinized zymosan (3 m g / m l ) (Fig. 1B), D M P O - O H was the dominant spin-trapped adduct detected; at higher concentrations (6 m g / m l ) (Fig. 1C), the ESR spectrum corresponding to D M P O - O O H was most prevalent. With the inclusion of D M S O (0.14 M), no change in ESR spectra was observed, indicating the absence of hydroxyl radical. These data with PMA and opsinized zymosan were identical both in regard to the nature of the free radicals spin trapped and the effects of stimulus concentration on the ESR spectrum recorded to that previously observed at 25 ° C [9]. When neutrophils were stimulated with PMA at 3 7 ° C in the presence of DMPO, the resulting spectrum was that characteristic of D M P O - O H (Fig. 2). Since D M S O (0.14 M) was included in the reaction mixture as the solvent for PMA at a concentration considerably greater than D M P O (0.1 M), and the reaction kinetics of the hydroxyl
239 lOG
I I 2
2
I
I
Fig. 2. ESR spectrum resulting from neutrophil (107/ml) stimulation at 37 o C with PMA (100 n g / m l ) in the presence of DMPO (0.1 M), DTPA (0.1 raM) and DMSO (0.14 M). Lowand high-field peaks corresponding to DMPO-OH and DMPO-OOH are designated 1 and 2, respectively. See Materials and Methods for instrumentation settings, however, the scan was started 1 min after the addition of PMA to the neutrophils.
radical with DMSO and DMPO are similar [20], if hydroxyl radical had arisen, the ESR spectrum corresponding to D M P O - C H 3 should have been recorded. Since this did not occur, as noted in Fig. 2, it seems likely that the elevated temperature led to an acceleration of D M P O - O O H decomposition rather than hydroxyl radical formation. When neutrophils were stimulated with the low dose of opsinized zymosan (3 m g / m l ) at 37 ° C in the presence of DMPO (0.1 M) and DMSO (0.14 M), we obtained only the ESR spectrum characteristic of DMPO-OH. However, when these experiments were repeated with the higher concentration of opsinized zymosan (6 mg/ml), the ESR signals for both D M P O - O O H and DMPOO H were observed, although D M P O - O H was the more dominant species, as noted in the high-field portion of the scan in Fig. 3. This conversion from 10G
I 2
2
I
Fig. 3. ESR spectrum resulting from neutrophil (107/ml) stimulation at 37 ° C with opsinized zymosan (6 m g / m l ) in the presence of DMPO (0.1 M), DTPA (0.1 mM) and DMSO (0.14 M). Low- and high-field peaks corresponding to D M P O - O H and DMPO-OOH are designated 1 and 2, respectively. See Materials and Methods for instrumentation settings; the scan was, however, initiated 1 min after the addition of opsinized zymosan to the cells.
D M P O - O O H to D M P O - O H has recently been described by Kleinhans and Barefoot [21], although they left open the question of the origin of these spin-trapped adducts. These results suggest that at 37 ° C even with enhanced rates of superoxide formation, as measured by the increased rate of superoxide dismutase-inhibitable reduction of succinylated cytochrome c, in response to either stimuli, the only (for PMA and low dose opsinized zymosan) or prevalent (for high dose opsinized zymosan) ESR spectrum observed is that of DMPO-OH. To elaborate on these findings, we incubated D M P O with neutrophils stimulated with PMA at 2 0 ° C for 5 min, which resulted in an observable ESR spectrum characteristic of only D M P O - O O H (Fig. 1A) and then rapidly raised the temperature to 3 7 ° C while monitoring the ESR signal. Almost immediately, the spectrum began to shift to that of DMPO-OH. We next monitored the early time points in the conversion of D M P O - O O H to D M P O - O H at 37 ° C. As shown in Table I, rapid degradation ensued. This transformation could have resulted from TABLE I EFFECT OF T E M P E R A T U R E ON THE CONVERSION OF DMPO-OOH TO DMPO-OH (a) Neutrophils (107/ml), PMA (100 ng/ml), DTPA (0.1 mM) and DMPO (0.1 M) were incubated for 5 min at 2 0 ° C , and then transferred to the ESR spectrometer set at 37° C. Peak heights (and thus ratios) are measured at designated times. Each point is the average of two independent experiments. (b) Hypoxanthine (0.1 mM), xanthine oxidase (0.01 U / m l ) , DTPA (0.1 mM) and DMPO (0.1 M) were mixed at 20 ° C for 5 min, and then transferred to the ESR spectrometer set at 3 7 ° C . Peak heights (and thus ratios) are measured at the designated times. Each point is the average of two independent experiments. (c) Hypoxanthine (0.1 raM), xanthine oxidase (0.01 U / m l ) , DTPA (0.1 mM), unstimulated neutrophils (107Jml) and DMPO (0.1 M) were mixed at 20 o C, and then transferred to the ESR spectrometer set at 37 o C. Peak heights (and thus ratios) are measured at designated times. Each point is the average of two independent experiments. Time (s)
0 30 60
D M P O - O O H / D M P O - O H ratio peak height (ram) a
b
c
3.60 0.29 0.11
7.30 1.86 1.10
1.1 0.7 0.50
240
either chemical (cell-independent) or enzymic (cell-dependent) conversion of D M P O - O O H to DMPO-OH. Accordingly, we monitored the effect of temperature on the transformation of DMPOOOH to D M P O - O H using the model superoxidegenerating system of h y p o x a n t h i n e / x a n t h i n e oxidase at pH 7.8. As shown in Table I, as the temperature was increased from 20 to 37 °C, the ESR signal changed from D M P O - O O H to DMPO-OH. However, even at 37 ° C, the rate was substantially slower than that observed with stimulated neutrophils (Table I). Finally, when unstimulated neutrophils were incubated with our model superoxide-generating system at 37 ° C, the rate of D M P O - O O H conversion was markedly increased, as compared to similar studies in the absence of neutrophils (Table I). These data indicate that elevated temperatures enhance the rate of DMPO-OOH degradation to DMPO-OH, despite the increased production of superoxide. If the degradative process discussed above was indeed mediated by neutrophils, an enhancement of this phenomenon might be expected by increasing cell concentration, independent of the temperature used. When neutrophils (1.10V/ml) were stimulated with PMA at 20 ° C, D M P O - O O H was the primary spin-trapped species observed. However, when-cell concentration was increased to 4. 107/ml, DMPO-OH became the only spintrapped adduct noted. At intermediate cell concentrations, both species were present. At no time was DMPO-CH 3 detected, excluding the possibility of hydroxyl radical formation. Thus, as the concentration of neutrophils increases, in spite of enhanced superoxide production, there is a shift in the ESR spectrum from D M P O - O O H to DMPOOH. One unusual finding was that the overall peak height of D M P O - O H decreased with increasing neutrophil concentration, despite the fact that superoxide production was also enhanced. This suggested that these cells, at 4- 10V/ml, are reducing the nitroxide moiety of D M P O - O H to 2,2-dimethyl-5-hydroxy-l-hydroxypyrrolidine, which would not be observable by ESR spectroscopy. Several years ago, we reported that nitroxides could be reduced to their corresponding hydroxylamines in the presence of superoxide and thiols. The absence of either component would prevent
10
v i
2
5
4
5
6
7
8
TIME (minutes)
Fig. 4. Effect of stimulated neutrophils on the stability of TEMPO. Neutrophils (4.10V/ml) were stimulated with PMA (100 n g / m l ) in the presence of TEMPO (10/zM) at 20 ° C. The ESR magnetic field was set at the top of the low-field peak of the nitroxide triplet and scanned with time. Time zero is the time at which recording began (1 min after the reaction commenced). (A) A marked decrease in TEMPO peak amplitude was recorded over the 8 min scan. (B) In the absence of either neutrophils or in the presence of superoxide dismutase (10 /~g/ml) ESR instrumentation settings were the same as those described in the Materials and Methods, except that the modulation amplitude was raised to 1.6 G and the gain was decreased to 3.2.103. Scan time was 8 min.
this reduction [22]. To determine whether this reaction was taking place in our cell system, we incubated T E M P O (10 ffM) with neutrophils (4107/ml) stimulated with PMA at 20 ° C. Following an initial lag phase, which is similar to the time necessary for stimulated neutrophils to generate superoxide, a reduction of T E M P O was observed (Fig. 4A), which was inhibited by the addition of superoxide dismutase (10 ffg/ml, Fig. 4B) to the mixture. To further explore this reaction, we added cell-free supernatant, obtained following neutrophil stimulation with PMA, to our model superoxide-generating system, h y p o x a n t h i n e / xanthine oxidase at pH 7.8. Again, rapid reduction of T E M P O was observed, which was inhibited in the presence of superoxide dismutase (10 ffg/ml). Yet, neither the superoxide-generating system nor the cell-free supernatant alone was able to manifest this reduction. Since glutathione is known to be important in neutrophil physiology [23], we examined the cell-free supernatant following neutrophil stimulation by PMA for the presence of this thiol. However, using HPLC techniques, we were unable to detect any glutathione at concentrations above i nM. It seems possible that other thiol-containing compounds derived from neutrophils may be responsible for this reaction, and not detected by our HPLC methods. We are currently investigating this possibility. In conclusion, this study further supports our
241 c o n t e n t i o n that following n e u t r o p h i l s t i m u l a t i o n with either P M A or opsinized zymosan, hydroxyl radical is n o t generated unless exogenous iron salts are added, at least within the limits of detection b y spin t r a p p i n g techniques. A l t h o u g h neutrophil s t i m u l a t i o n at 3 7 ° C results in a n E S R spectrum in which D M P O - O H dominates, this is the result of accelerated conversion of D M P O O O H to D M P O - O H a n d n o t the spin t r a p p i n g of hydroxyl radical at the elevated temperature. T h e m e c h a n i s m for this process is complicated, involving a n increase in the rate of d e g r a d a t i o n c o n c o m i t a n t with a temperature-sensitive e n z y m a t i c transformation. Finally, investigators a p p l y i n g spin t r a p p i n g procedures to the s t u d y of free radical g e n e r a t i o n by stimulated n e u t r o p h i l s should be cognizant of the potential for biochemical reactions which m a y m a r k e d l y alter the E S R s p e c t r u m observed as a result of n e u t r o p h i l p r o d u c t i o n of superoxide.
Acknowledgements This work was supported b y grants from the N a t i o n a l Institutes of Health, AM-3322, AI-15036, HL-33550 a n d AI-107001; the K n i g h t s T e m p l e r Eye F o u n d a t i o n a n d the A m e r i c a n C a n c e r Society, BC-453. B.E. Britigan is a recipient of the Burroughs W e l c o m e Fellowship of the Infectious Disease Society of America.
References 1 Babior, B.M., Kipnes, R.S. and Curnette, J.T. (1973) J. Clin. Invest. 52, 741-744. 2 Haber, F. and Weiss, J. (1934) Proc. R. Soc. Lond. A. Math. Phys. Sci. 147, 332-351. 3 Weiss, S.J., Rustagi, P.K. and LeBuglio, A.F. (1978) J. Exp. Med. 147, 316-323.
4 Tauber, A.I. and Babior, B.M. (1977) J. Clin. Invest. 60, 374-379. 5 Repine, J.E., Eaton, J.W., Anders, M.W., Hoidal, J.R. and Fox, R.B. (1979) J. Clin. Invest. 64, 1642-1651. 6 Rosen, H. and Klebanoff, S.J. (1979) J. Clin. Invest. 64, 1725-1729. 7 Green, M.R., Hill, H.A.O., Okolow-Zubkowska, M.J. and Segal, A.W. (1979) FEBS Lett. 100, 23-26. 8 Hawley, D.A., Kleinhans, F.W. and Biesecker, J.L. (1983) Am. J. Clin. Pathol. 79, 673-677. 9 Britigan, B.E., Rosen, G.M. Chai, Y. and Cohen, M.S. (1986) J. Biol. Chem. 261, 4426-4431. 10 Britigan, B.E., Rosen, G.M., Thompson, B.Y., Chai, Y. and Cohen, M.S. (1986) J. Biol. Chem. 261, 17026-17032. 11 Finkelstein, E., Rosen, G.M., Patton, S.E., Cohen, M.S. and Rauckman, E.J. (1981) Biochem. Biophys. Res. Commun. 102, 1008-1015. 12 Rozantsev, E.G. (1970) in Free Nitroxyl Radicals, pp. 217, Plenum Press, NY. 13 Bonnett, R., Brown, R.F.C., Clark, V.M., Sutherland, I.O. and Todd, A. (1959) J. Chem. Soc. 2094-2101. 14 Cohen, M.S. and Cooney, M.H. (1984) J. Infect. Dis. 150, 49-56. 15 Livesey,J.C. and Reed, D.J. (1984) Int. J. Rad. Oncol. Biol. Phys. 10, 1507-1510. 16 Chaney, W.G. and Spector, A. (1984) Curr. Eye Res. 3, 345-350. 17 Finkelstein, E., Rosen, G.M. and Rauckman, E.J. (1982) Mol. Pharmacol. 21,262-265. 18 Finkelstein, E., Rosen, G.M. and Rauckman, E.J. (1980) Arch. Biochem. Biophys. 200, 1-16. 19 Cohen, M.S., Metcalf, J.A. and Root, R.K. (1980) Blood 55, 1003-1009. 20 Finkelstein, E., Rosen, G.M. and Rauckman, E.J. (1980) J. Am. Chem. Soc. 102, 4994-4999. 21 Kleinhans, F.W. and Barefoot, S.T. (1987) J. Biol. Chem. 262, 12452-12457. 22 Finkelstein, E., Rosen, G.M. and Rauckman, E.J. (1984) Biochim. Biophys. Acta 802, 90-98. 23 Root, R.K. and Cohen, M.S. (1981) Rev. Infect. Dis. 3, 565-598.
Biochimica etBiophysicaActa 969 (1988) 236-241
Elsevier BBA12233
Detection of phagoctye-derived free radicals with spin trapping techniques: effect of temperature and cellular metabolism G e r a l d M. R o s e n a,b, B r a d l e y E. B r i t i g a n c,., M y r o n S. C o h e n S h a r o n P. E l l i n g t o n a,b a n d M i c h a e l J. B a r b e r e
c,d
" Department of Pharmacology, Duke University Medical Center, and b GRECC, Veterans Administration Medical Center, Durham, NC, Departments of " Medicine and a Microbiology and Immunology, University of North Carolina at Chapel Hill, Chapel Hill, NC and e Department of Biochemistry, University of South Florida College of Medicine, Tampa, FL (U.S.A.)
(Received 16 October 1987) (Revised manuscript received 1l January 1988)
Key words: Free radical; Superoxide generation; Phagocyte stimulation; ESR; Temperature dependence; (Human)
Human neutrophils activated with either particulate or soluble stimuli generate oxygen-centered free radicals which are detected by spin trapping in conjunction with electron spin resonance (ESR) spectroscopy. We investigated the effect of temperature on ESR spectra resulting from stimulation of human neutrophils with phorbol myristate acetate (PMA) or opsonized zymosan in the presence of the spin trap, 5,5-dimethyl-l-pyrroline 1-oxide (DMPO). At 20°C with either stimuli, neutrophil superoxide production was manifested predominantely as the superoxide spin-trapped adduct, 5,5-dimethyi-5-hydroperoxy-l-pyrrolidinyloxy (DMPO-OOH). In contrast, at 37°C, the hydroxyl spin-trapped adduct, 2,2-dimethyl-5-hydroxy-l-pyrrolidinyloxy (DMPO-OH) was dominant. No evidence of hydroxyl radical (defined as the methyl spin-trapped adduct, 2,2,5-trimethyl-l-pyrrolidinyloxy, DMPO-CH 3) was observed, suggesting that elevated temperatures increased the rate of DMPO-OOH conversion to DMPO-OH. In addition, the elevated temperature activated a neutrophil reductase which accelerated the rate of D M P O - O H reduction to its corresponding hydroxylamine, 2,2-dimethyl-5-hydroxy-l-hydroxypyrrolidine. This bioreduction was dependent upon the presence of both superoxide and a phagocyte-derived factor (possibly a thiol) released into the surrounding media.
* Present address: Department of Medicine, University of Iowa, Iowa City, IA 52242, U.S.A. Abbreviations: PMA, 4-phorbol 12-myristate 13-acetate; DMPO, 5,5-dimethyl-l-pyrroline 1-oxide; DMPO-OOH, 5,5dimethyl-5-hydroperoxy-l-pyrrolidinyloxy; DMPO-OH, 2,2dimethyl-5-hydroxy-l-pyrrolidinyloxy; DMPO-CH 3, 2,2,5-trimethyl-l-pyrrolidinyloxy; DTPA, diethylenetriaminepentaacetic acid; TEMPO, 2,2,6,6-tetramethylpiperidinoxy; DMSO, dimethyl sulfoxide; HBSS, Hanks' balanced salt solution. Correspondence: G.M. Rosen, Department of Pharmacology, Duke University Medical Center, Durham, NC 27710, U.S.A.
Introduction In response to soluble or particulate stimuli, human neutrophils generate superoxide as the result of the activation of an NADPH-oxidoreductase [1]. Subsequent dismutation yields hydrogen peroxide. In the presence of ferric ions, superoxide and hydrogen peroxide produce hydroxyl radical via a metal-catalyzed Haber-Weiss reaction [2]. Although formation of hydroxyl radical by activated neutrophils has been previously reported
0167-4889/88/$03.50 © 1988 Elsevier Science Publishers B.V. (Biomedical Division)
237 [3-8], the techniques used are not without problems and limitations. Consequently, we recently undertook an extensive investigation of neutrophil free radical production using spin trapping techniques [9]. When either PMA or opsonized zymosan were used to stimulate neutrophils at room temperature (25 ° C), we only spin trapped super° oxide, even although, under some experimental conditions, we observed the hydroxyl radical spin-trapped adduct, DMPO-OH [9,10]. Our findings suggested that for previous investigations [6-8], the purported spin trapping of hydroxyl radical may actually be incorrect. However, several earlier studies [6-8] differed from ours in that they were performed with neutrophils incubated at 37°C. This suggested two possibilities: (a) ESR spectra resulting from spin trapping of neutrophil-derived superoxide vary with temperature or (b) hydroxyl radical formation by neutrophils occurs at 37°C, but not at 25°C. In this present study, we monitored the spin trapping of neutrophil-generated free radicals at 20 °C and 37 ° C. Our data demonstrate that at 37°C, decomposition of DMPO-OOH is rapid, however, we do not find any evidence for the formation of hydroxyl radical at this elevated temperature. Materials and Methods
Reagents. Diethylenetriaminepentaacetic acid (DTPA), hypoxanthine, superoxide dismutase and zymosan A were purchased from Sigma Chemical Company (St. Louis, MO). PMA and xanthine oxidase were obtained from Midland Chemical Company (Brewster, NY) and Calbiochem, a division of Behring Diagnostics, Inc. (La Jolla, CA). Succinylated cytochrome c was prepared as described earlier [11]. 2,2,6,6-Tetramethylpiperidinoxy (TEMPO) was synthesized according to procedures outlined by Rozantsev [12]. DMPO was prepared by methods described in Bonnett et al. [13]. PMA was dissolved in DMSO at a concentration of 10 ~tg/ml. Zymosan A was opsonized with normal human serum prior to use. Neutrophil isolation. Neutrophils were isolated from heparinized blood derived from normal human volunteers as previously described [14]. Leukocytes were separated from erythrocytes by
the addition of Plasmagel (Roger Bellon, Neuilley, France). The resulting leukocyte suspension was centrifuged through a Ficoll-Hypaque gradient (Pharmacia Chemicals, Piscataway, N J). Erythrocytes contaminating the neutrophil layer were removed by hypotonic lysis. Neutrophils were then suspended in Hanks' balanced salt solution (HBSS) and the concentration was adjusted as determined by a Coulter Model D2N automated blood cell counter (Coulter Electronics, Hialeah, FL). This procedure yielded cells which were more than 98% neutrophils (Giemsa stain) and more than 95% viable (Trypan blue exclusion). Cell-free supernatant. Neutrophils were suspended in HBSS to 4. 107/ml and stimulated with PMA, if required, for 30 rain at room temperature, pelleted (300 × g, 10 rain) and the resulting supernatant was removed for additional experiments. Glutathione detection. Concentration of reduced and oxidized glutathione in neutrophil-derived cell-free supernatants was determined by HPLC analysis. Samples were prepared for analysis by the method of Livesey and Reed [15], and analyzed as described by Chaney and Spector [16] except that the sample was eluted with a 14 min gradient (Gradient Curve 8, Watts Model 660 Solvent Programmer) from 75% A, 25% B to 25% A, 75% B. Neutrophil superoxide release. Neutrophil superoxide release as a consequence of PMA stimulation was determined by measuring the rate of superoxide dismutase-inhibitable reduction of succinylated cytochrome c [11]. Spin trapping/ESR spectroscopy studies. Spin trapping of free radicals was undertaken by mixing neutrophils (1. 107/ml), DMPO (0.1 M), DMSO (0.14 M), DTPA (0.1 mM) and sufficient buffer (HBSS) to reach a final volume of 0.5 ml. Stimulation of neutrophil oxygen reduction was mediated by the addition of either PMA (100 ng/ml) or opsonized zymosan A (3 or 6 mg/ml). Reaction mixtures were transferred to an ESR quartz cell and placed into the cavity of an ESR spectrometer (Varian Associates, Palo Alto, CA, model E-9), equipped with a variable temperature accessory. Sample temperature was determined by the inclusion of a thermistor in the spectrometer cavity. In some experiments, a superoxide-generating system consisting of hypoxanthine (0.1 mM) and xanthine oxidase (0.01 U/ml) such that the
238
rate of superoxide dismutase-inhibitable reduction of succinylated cytochrome c was 10 /~M/min at p H 7.8, was substituted for stimulated neutrophils. The rate of T E M P O reduction was determined by monitoring the change in amplitude of the lowfield peak of the nitroxide triplet with time. ESR spectra were routinely obtained using a microwave power of 20 mW, a modulation amplitude of 1 G, a field sweep rate of 12.5 G / m i n , a time constant of I s, and a receiver gain of 5 • 103, except where noted. In a series of experiments, we examined the effect of D M P O (0.1 M), D M S O (0.14 M) and D T P A (0.1 mM) on superoxide production at 2 0 ° C and 37°C. At these concentrations, we observed no inhibition in the generation of superoxide, as monitored by superoxide dismutase-inhibitable reduction of succinylated cytochrome c. These results are identical to results from our earlier studies [9]. Results
and Discussion
When human neutrophils were stimulated with P M A in D M S O (0.14 M) or opsonized zymosan at 20 ° C, the spectra shown in Fig. 1 were recorded. For PMA, the ESR spectrum was primarily D M P O - O O H and to a lesser extent D M P O - O H (Fig. 1A). "The ESR spectrum for D M P O - O O H results from the spin trapping of superoxide by DMPO, whereas D M P O - O H may arise as the result of either the spin trapping of hydroxyl radical by D M P O or the degradation of D M P O O O H to D M P O - O H [17,18]. However, the detection of D M P O - C H 3 is more specific than D M P O O H for the formation of hydroxyl radical, since this free radical reacts with D M S O at diffusioncontrolled rates to generate methyl radical ("CH 3) [19], which can then be spin trapped. A small quantity (approx. 3%) of D M P O - O O H decomposes to give hydroxyl radical, which can then be spin trapped as D M P O - C H 3 [17]. Thus, detection of a concentration of D M P O - C H 3, in the range described above, is not necessarily the result of cell-mediated hydroxyl radical production, but rather arises as an artifact of superoxide spin trapping by DMPO. In the case of opsinized zymosan (Fig. 1B and C), two different ESR spectra were recorded at 20 ° C , depending on the concentration of
lOG
I 2
I I
A.
B.
C•
I 2
i 2
2
2~
i
Fig. 1. ESR spectra resulting from neutrophil (107/ml) stimulation at 2 0 ° C in the presence of D M P O (0.1 M), D T P A (0.1 mM) and D M S O (0.14 M). Stimufi were P M A (100 n g / m l , scan A), opsinized zymosan (3 m g / m l , scan B) and opsinized zymosan (6 m g / m l , scan C). Low- and high-field peaks corresponding to D M P O - O H and D M P O - O O H are designated 1 and 2, respectively. See Materials and Methods for instrumentation settings: scans were, however, initiated 1 min after the addition of the stimulus to the neutrophils.
opsinized zymosan added to the cell suspension. At low concentrations of opsinized zymosan (3 m g / m l ) (Fig. 1B), D M P O - O H was the dominant spin-trapped adduct detected; at higher concentrations (6 m g / m l ) (Fig. 1C), the ESR spectrum corresponding to D M P O - O O H was most prevalent. With the inclusion of D M S O (0.14 M), no change in ESR spectra was observed, indicating the absence of hydroxyl radical. These data with PMA and opsinized zymosan were identical both in regard to the nature of the free radicals spin trapped and the effects of stimulus concentration on the ESR spectrum recorded to that previously observed at 25 ° C [9]. When neutrophils were stimulated with PMA at 3 7 ° C in the presence of DMPO, the resulting spectrum was that characteristic of D M P O - O H (Fig. 2). Since D M S O (0.14 M) was included in the reaction mixture as the solvent for PMA at a concentration considerably greater than D M P O (0.1 M), and the reaction kinetics of the hydroxyl
239 lOG
I I 2
2
I
I
Fig. 2. ESR spectrum resulting from neutrophil (107/ml) stimulation at 37 o C with PMA (100 n g / m l ) in the presence of DMPO (0.1 M), DTPA (0.1 raM) and DMSO (0.14 M). Lowand high-field peaks corresponding to DMPO-OH and DMPO-OOH are designated 1 and 2, respectively. See Materials and Methods for instrumentation settings, however, the scan was started 1 min after the addition of PMA to the neutrophils.
radical with DMSO and DMPO are similar [20], if hydroxyl radical had arisen, the ESR spectrum corresponding to D M P O - C H 3 should have been recorded. Since this did not occur, as noted in Fig. 2, it seems likely that the elevated temperature led to an acceleration of D M P O - O O H decomposition rather than hydroxyl radical formation. When neutrophils were stimulated with the low dose of opsinized zymosan (3 m g / m l ) at 37 ° C in the presence of DMPO (0.1 M) and DMSO (0.14 M), we obtained only the ESR spectrum characteristic of DMPO-OH. However, when these experiments were repeated with the higher concentration of opsinized zymosan (6 mg/ml), the ESR signals for both D M P O - O O H and DMPOO H were observed, although D M P O - O H was the more dominant species, as noted in the high-field portion of the scan in Fig. 3. This conversion from 10G
I 2
2
I
Fig. 3. ESR spectrum resulting from neutrophil (107/ml) stimulation at 37 ° C with opsinized zymosan (6 m g / m l ) in the presence of DMPO (0.1 M), DTPA (0.1 mM) and DMSO (0.14 M). Low- and high-field peaks corresponding to D M P O - O H and DMPO-OOH are designated 1 and 2, respectively. See Materials and Methods for instrumentation settings; the scan was, however, initiated 1 min after the addition of opsinized zymosan to the cells.
D M P O - O O H to D M P O - O H has recently been described by Kleinhans and Barefoot [21], although they left open the question of the origin of these spin-trapped adducts. These results suggest that at 37 ° C even with enhanced rates of superoxide formation, as measured by the increased rate of superoxide dismutase-inhibitable reduction of succinylated cytochrome c, in response to either stimuli, the only (for PMA and low dose opsinized zymosan) or prevalent (for high dose opsinized zymosan) ESR spectrum observed is that of DMPO-OH. To elaborate on these findings, we incubated D M P O with neutrophils stimulated with PMA at 2 0 ° C for 5 min, which resulted in an observable ESR spectrum characteristic of only D M P O - O O H (Fig. 1A) and then rapidly raised the temperature to 3 7 ° C while monitoring the ESR signal. Almost immediately, the spectrum began to shift to that of DMPO-OH. We next monitored the early time points in the conversion of D M P O - O O H to D M P O - O H at 37 ° C. As shown in Table I, rapid degradation ensued. This transformation could have resulted from TABLE I EFFECT OF T E M P E R A T U R E ON THE CONVERSION OF DMPO-OOH TO DMPO-OH (a) Neutrophils (107/ml), PMA (100 ng/ml), DTPA (0.1 mM) and DMPO (0.1 M) were incubated for 5 min at 2 0 ° C , and then transferred to the ESR spectrometer set at 37° C. Peak heights (and thus ratios) are measured at designated times. Each point is the average of two independent experiments. (b) Hypoxanthine (0.1 mM), xanthine oxidase (0.01 U / m l ) , DTPA (0.1 mM) and DMPO (0.1 M) were mixed at 20 ° C for 5 min, and then transferred to the ESR spectrometer set at 3 7 ° C . Peak heights (and thus ratios) are measured at the designated times. Each point is the average of two independent experiments. (c) Hypoxanthine (0.1 raM), xanthine oxidase (0.01 U / m l ) , DTPA (0.1 mM), unstimulated neutrophils (107Jml) and DMPO (0.1 M) were mixed at 20 o C, and then transferred to the ESR spectrometer set at 37 o C. Peak heights (and thus ratios) are measured at designated times. Each point is the average of two independent experiments. Time (s)
0 30 60
D M P O - O O H / D M P O - O H ratio peak height (ram) a
b
c
3.60 0.29 0.11
7.30 1.86 1.10
1.1 0.7 0.50
240
either chemical (cell-independent) or enzymic (cell-dependent) conversion of D M P O - O O H to DMPO-OH. Accordingly, we monitored the effect of temperature on the transformation of DMPOOOH to D M P O - O H using the model superoxidegenerating system of h y p o x a n t h i n e / x a n t h i n e oxidase at pH 7.8. As shown in Table I, as the temperature was increased from 20 to 37 °C, the ESR signal changed from D M P O - O O H to DMPO-OH. However, even at 37 ° C, the rate was substantially slower than that observed with stimulated neutrophils (Table I). Finally, when unstimulated neutrophils were incubated with our model superoxide-generating system at 37 ° C, the rate of D M P O - O O H conversion was markedly increased, as compared to similar studies in the absence of neutrophils (Table I). These data indicate that elevated temperatures enhance the rate of DMPO-OOH degradation to DMPO-OH, despite the increased production of superoxide. If the degradative process discussed above was indeed mediated by neutrophils, an enhancement of this phenomenon might be expected by increasing cell concentration, independent of the temperature used. When neutrophils (1.10V/ml) were stimulated with PMA at 20 ° C, D M P O - O O H was the primary spin-trapped species observed. However, when-cell concentration was increased to 4. 107/ml, DMPO-OH became the only spintrapped adduct noted. At intermediate cell concentrations, both species were present. At no time was DMPO-CH 3 detected, excluding the possibility of hydroxyl radical formation. Thus, as the concentration of neutrophils increases, in spite of enhanced superoxide production, there is a shift in the ESR spectrum from D M P O - O O H to DMPOOH. One unusual finding was that the overall peak height of D M P O - O H decreased with increasing neutrophil concentration, despite the fact that superoxide production was also enhanced. This suggested that these cells, at 4- 10V/ml, are reducing the nitroxide moiety of D M P O - O H to 2,2-dimethyl-5-hydroxy-l-hydroxypyrrolidine, which would not be observable by ESR spectroscopy. Several years ago, we reported that nitroxides could be reduced to their corresponding hydroxylamines in the presence of superoxide and thiols. The absence of either component would prevent
10
v i
2
5
4
5
6
7
8
TIME (minutes)
Fig. 4. Effect of stimulated neutrophils on the stability of TEMPO. Neutrophils (4.10V/ml) were stimulated with PMA (100 n g / m l ) in the presence of TEMPO (10/zM) at 20 ° C. The ESR magnetic field was set at the top of the low-field peak of the nitroxide triplet and scanned with time. Time zero is the time at which recording began (1 min after the reaction commenced). (A) A marked decrease in TEMPO peak amplitude was recorded over the 8 min scan. (B) In the absence of either neutrophils or in the presence of superoxide dismutase (10 /~g/ml) ESR instrumentation settings were the same as those described in the Materials and Methods, except that the modulation amplitude was raised to 1.6 G and the gain was decreased to 3.2.103. Scan time was 8 min.
this reduction [22]. To determine whether this reaction was taking place in our cell system, we incubated T E M P O (10 ffM) with neutrophils (4107/ml) stimulated with PMA at 20 ° C. Following an initial lag phase, which is similar to the time necessary for stimulated neutrophils to generate superoxide, a reduction of T E M P O was observed (Fig. 4A), which was inhibited by the addition of superoxide dismutase (10 ffg/ml, Fig. 4B) to the mixture. To further explore this reaction, we added cell-free supernatant, obtained following neutrophil stimulation with PMA, to our model superoxide-generating system, h y p o x a n t h i n e / xanthine oxidase at pH 7.8. Again, rapid reduction of T E M P O was observed, which was inhibited in the presence of superoxide dismutase (10 ffg/ml). Yet, neither the superoxide-generating system nor the cell-free supernatant alone was able to manifest this reduction. Since glutathione is known to be important in neutrophil physiology [23], we examined the cell-free supernatant following neutrophil stimulation by PMA for the presence of this thiol. However, using HPLC techniques, we were unable to detect any glutathione at concentrations above i nM. It seems possible that other thiol-containing compounds derived from neutrophils may be responsible for this reaction, and not detected by our HPLC methods. We are currently investigating this possibility. In conclusion, this study further supports our
241 c o n t e n t i o n that following n e u t r o p h i l s t i m u l a t i o n with either P M A or opsinized zymosan, hydroxyl radical is n o t generated unless exogenous iron salts are added, at least within the limits of detection b y spin t r a p p i n g techniques. A l t h o u g h neutrophil s t i m u l a t i o n at 3 7 ° C results in a n E S R spectrum in which D M P O - O H dominates, this is the result of accelerated conversion of D M P O O O H to D M P O - O H a n d n o t the spin t r a p p i n g of hydroxyl radical at the elevated temperature. T h e m e c h a n i s m for this process is complicated, involving a n increase in the rate of d e g r a d a t i o n c o n c o m i t a n t with a temperature-sensitive e n z y m a t i c transformation. Finally, investigators a p p l y i n g spin t r a p p i n g procedures to the s t u d y of free radical g e n e r a t i o n by stimulated n e u t r o p h i l s should be cognizant of the potential for biochemical reactions which m a y m a r k e d l y alter the E S R s p e c t r u m observed as a result of n e u t r o p h i l p r o d u c t i o n of superoxide.
Acknowledgements This work was supported b y grants from the N a t i o n a l Institutes of Health, AM-3322, AI-15036, HL-33550 a n d AI-107001; the K n i g h t s T e m p l e r Eye F o u n d a t i o n a n d the A m e r i c a n C a n c e r Society, BC-453. B.E. Britigan is a recipient of the Burroughs W e l c o m e Fellowship of the Infectious Disease Society of America.
References 1 Babior, B.M., Kipnes, R.S. and Curnette, J.T. (1973) J. Clin. Invest. 52, 741-744. 2 Haber, F. and Weiss, J. (1934) Proc. R. Soc. Lond. A. Math. Phys. Sci. 147, 332-351. 3 Weiss, S.J., Rustagi, P.K. and LeBuglio, A.F. (1978) J. Exp. Med. 147, 316-323.
4 Tauber, A.I. and Babior, B.M. (1977) J. Clin. Invest. 60, 374-379. 5 Repine, J.E., Eaton, J.W., Anders, M.W., Hoidal, J.R. and Fox, R.B. (1979) J. Clin. Invest. 64, 1642-1651. 6 Rosen, H. and Klebanoff, S.J. (1979) J. Clin. Invest. 64, 1725-1729. 7 Green, M.R., Hill, H.A.O., Okolow-Zubkowska, M.J. and Segal, A.W. (1979) FEBS Lett. 100, 23-26. 8 Hawley, D.A., Kleinhans, F.W. and Biesecker, J.L. (1983) Am. J. Clin. Pathol. 79, 673-677. 9 Britigan, B.E., Rosen, G.M. Chai, Y. and Cohen, M.S. (1986) J. Biol. Chem. 261, 4426-4431. 10 Britigan, B.E., Rosen, G.M., Thompson, B.Y., Chai, Y. and Cohen, M.S. (1986) J. Biol. Chem. 261, 17026-17032. 11 Finkelstein, E., Rosen, G.M., Patton, S.E., Cohen, M.S. and Rauckman, E.J. (1981) Biochem. Biophys. Res. Commun. 102, 1008-1015. 12 Rozantsev, E.G. (1970) in Free Nitroxyl Radicals, pp. 217, Plenum Press, NY. 13 Bonnett, R., Brown, R.F.C., Clark, V.M., Sutherland, I.O. and Todd, A. (1959) J. Chem. Soc. 2094-2101. 14 Cohen, M.S. and Cooney, M.H. (1984) J. Infect. Dis. 150, 49-56. 15 Livesey,J.C. and Reed, D.J. (1984) Int. J. Rad. Oncol. Biol. Phys. 10, 1507-1510. 16 Chaney, W.G. and Spector, A. (1984) Curr. Eye Res. 3, 345-350. 17 Finkelstein, E., Rosen, G.M. and Rauckman, E.J. (1982) Mol. Pharmacol. 21,262-265. 18 Finkelstein, E., Rosen, G.M. and Rauckman, E.J. (1980) Arch. Biochem. Biophys. 200, 1-16. 19 Cohen, M.S., Metcalf, J.A. and Root, R.K. (1980) Blood 55, 1003-1009. 20 Finkelstein, E., Rosen, G.M. and Rauckman, E.J. (1980) J. Am. Chem. Soc. 102, 4994-4999. 21 Kleinhans, F.W. and Barefoot, S.T. (1987) J. Biol. Chem. 262, 12452-12457. 22 Finkelstein, E., Rosen, G.M. and Rauckman, E.J. (1984) Biochim. Biophys. Acta 802, 90-98. 23 Root, R.K. and Cohen, M.S. (1981) Rev. Infect. Dis. 3, 565-598.