Inactivation of NADPH oxidase from human neutrophils by affinity labelling with pyridoxal 5′-diphospho-5′-adenosine

Vol.

181. No.

December

BIOCHEMICAL

3, 1991

AND

BIOPHYSICAL

RESEARCH

31, 1991

Inactivation

of NADPH oxidase from human neutrophils with pyridoxal S-diphospho-5’-adenosine

COMMUNICATIUNS Pages 1259-1265

by affinity

labeling

Pascale Ravel and Florence Meter CNRS URA 1461, Hopital Necker, 161 Rue de S&es, 75743 Paris Cedex 15, France Received

October

3, 1991

Summary: When a particulate NADPH oxidase prepared from phorbol ester-activated human neutrophils was treated with pyridoxal S-diphospho-S-adenosine (PLP-AMP), the superoxide anion-producing activity was inhibited according to affinity labeling kinetics. NADPH afforded a protection against inactivation which was competitive with respect to PLP-AMP, 2.5’~ADP and 2’-phospho-5’ diphosphoadenosine (ATP ribose) appeared to be as potent as NADPH as protecting agents. NADP+ and ATP were less effective, while ADP and GTP-Y-S did not protect significantly. These results suggest that PLP-AMP can be used, in conjunction with tritiated cyanoborohydride, to identify the elusive NADPH-dependent flavoprotein which is part 0 1991Academic PESS, Inc. of the electron transfer chain of NADPH ox&se.

NADPH oxidase (EC 1.6.99.6) is a membrane-associated enzyme found essentially in “professional” phagocytes : neutrophils, eosinophils and macrophages. These cells have a common function : host defense against infections. Gpsonized microorganisms, as well as a variety of particulate and soluble stimuli, are responsible for cell activation; this in turn leads to activation of the resting NADPH oxidase. The enzyme catalyses, at the expense of NADPH, the one-electron reduction of molecular oxygen to superoxide anion (a-) (1). The latter is then transformed into an array of microbicidal oxidants (2,3). It is nowadays generally accepted that the electron transfer chain from NADPH to oxygen is comprised of two components : an FAD-linked flavoprotein and a low-potential b type cytochrome (cytochrome b55g)(4-6). The evidence concerning the participation of cytochrome bss8is incontrovertible. The protein has been purified (7-g), the genes of its two subunits are cloned (10-13). Genetic alterations of either of these genes are responsible for certain forms of chronic granulomatous disease (10, 14-16), a condition in which the respiratory burst is absent or defective. The participation of a flavoprotein also appears certain. Flavins are the redox cofactors usually linked to nicotinamide and they are capable of coupling a two-electron donor like NADPH to a monoelectronic one, which in this case would be the cytochrome. An FAD requirement for NADPH oxidase activity of particulate oxidase in Triton X-100 was indeed observed (17). This result was strengthened by the finding that NADPH oxidase activity could

1259

0006-291X/91 $1.50 Copyright 0 1991 by Academic Press, Inc. All rights of reproduction in any form reserved.

Vol. 181, No. 3, 1991

BIOCHEMICAL

AND BIOPHYSICAL

RESEARCH COMMUNICATIONS

not be supported by a flavin analogwhich is incapableof one-electronchemistry (18). Further evidencein favour of the importanceof a flavoprotein for superoxideproduction is summarized in (19). On the other hand, the cellular localization of the flavoprotein in resting cells (it is membrane-boundin activated cells), as well as its molecular composition have remained elusive. Attempts to purify the superoxide-producingenzyme led to variable resultsasto FAD and hemecontent as well as molecular weight (20-26). Chemical labeling concomitant with inactivation of the superoxideproducing activity led to a molecularweight estimateof 47 kDa with diphenylene iodonium (27,28), but this was later challenged (29). The use of affinity labelsbasedon the structureof NADP(H) led to proposefor the flavoprotein molecularweights of 65 kDa (in beef) (30), 77 kDa (in rabbit) (31), 66 kDa (in man) (32). The most popular of thesereagentsis the periodatecleaved, dialdehyde form of NADPH (NADPI-I& tirst proposed by Umei et al. (32). It is however now known not to provide a highly specific labeling (33-36); furthermore, using this samereagent with membranesfrom activated cells instead of the fractions of the cell-free reconstitution assay,Umei et al. (36) recently suggesteda MW of 32 kDa for the “NADPH binding component”of the oxidase. This brief review makesit clear that no consensushasemergedconcerningeven at leastthe molecular weight of NADPH oxidase. We decided to investigate the useof affinity labeling reagentsother than NADPI-Idx, which might provide either new or confirmatory evidence. We chosethe nucleotide pyridoxal S-diphospho-5’-adenosine(PALP-AMP). The synthesisof this reagent was reported by two groupsin 1986 (37,38). It hasbeen usedfor labeling nucleotide binding sites in a number of proteins (37-40), among them an NADH (38) and an NADPH binding site (39). Although such a reagent might appearas lessspecific than NADPI&, the idea was to use as a criterion the protection afforded by NADPH against inactivation by covalent labeling, aswasdone in the studieswith other inhibitors.

Materials

and Methods

Materials. PLP-AMP was synthesizedaccordingto a publishedprocedure (37). It was stored at -8O’C in 5 n&I KC1 in 50% aqueousethanol.Aliquots weredried asrequiredjust beforeeach experiment. Preparation of particulate NADPH oxidase. Human neutrophilswere obtainedfrom the blood of normal volunteers accordingto (40). The cells (lO%nl) were suspendedin Ca2+-and Mg2+free Hanks balancedsalt solution (HBSS) pH 7.5 and activated using 10 @ml phorbol-12my&ate-13-acetate (PMA) at 37°C for 4 min. The suspensionwas then diluted in cold HBSS; the cells were washedfour times in the samebuffer then resuspended in homogenizationbuffer (20 mM phosphatepH 7.4 supplementedwith 0.34 M sucrose,20 mM NaN3,7 mM MgS04, 1 mM CaC12,l mM phenylmethanesulfonylfluoride (PMSF) and 200 pM leupeptin) at 5 x 10’ cells/ml. They were then lysed with a Branson250 sonicator (6 x 10-spulses);the homogenate wascentrifuged at 1OOOg for 10min at 4OCand the supematantkept until useat - 8O’C for up to several months. For each experiment, the suspensionwas centrifuged at 48,ooOg for 1 h at 4’C. The pellet was suspendedin half the original volume of solubilization buffer (20 mM phosphatepH 7.4 containing 0.34 M sucrose,20 it NaN3. 1 mM PMSF, 200 pM leupeptin. 0.25% (w/u) deoxycholate to which 15% (u/u) dimethylsulfoxide wasadded.The suspension was homogenized by sonication (3 x 10-s bursts); it is called the particulate fraction. Superoxide production was measuredwith the continuous superoxide dismutase-inhibitable 1260

Vol. 181, No. 3, 1991

BIOCHEMICAL

AND BIOPHYSICAL RESEARCH COMMUNICATIONS

ferricytochrome c reduction assay, using a Uvikon 930 spectrophotometer (42). Assay mixtures contained 100 PM ferricytcchrome c, 200 p,M NADPH in 100 mM phosphate buffer pH 7.4 at 2S’C. The reaction was started by adding the enzyme aliquot to both cuvettes. Inactivation experiments. PLP-AMP concentration was determined in 50 mM sodium phosphate buffer pH 7 with & = 4900 M-1 cm -1 at 388 nm. The particulate NADPH oxidase preparation described above was incubated in the presence of various concentrations of PLPAMP at 10°C in the dark. For each concentration, the extent of inactivation was monitored as a function of time by withdrawing an aliquot and diluting it 25fold into the superoxide assay mixture at 2S’C. For protection experiments the ligand was added to the incubation mixture before the inactivator. Results The preparation used for the experiments is inspired from that described by Tauber & Goetzl(43), in that the particulate fraction from activated cells was treated with deoxycholate and dimethylsulfoxide, as described in Methods. Its specific activity was about 25% that of whole cells; it showed no FAD requirement. Upon centrifugation at 100 OOOg,the activity was recovered in the particulate fraction in 95% yield. Inactivation with PLP-AMP Fig.1 shows the time- and concentrationdependence of inactivation of our particulate fraction with various PLP-AMP concentrations. The reaction was found to follow a fist-order kinetic course (Fig.lA). At each inactivator concentration a kobs was thus obtained from the slope of the line. When l/kobs was plotted as a function of l/I, a straight line not intersecting

-1 Time (min)

0

1 2 l/PLP-AMP (mW)

&& Kinetics of the inactivation by PLP-AMP and protection by NADPH. A. Time dependenceof the inactivation. Experimental details are given in Methods. Additions: O.none; A, 0.6mM; 0,1.2mM; +, 2.2mMPLF’AMF’. B. Concentration dependenceof the inactivation. The experimental points (k& are derived from plots similar to those of Fig IA after correction for the usually small activity loss of the contm1. The concentration of NADPH was0.5 mM. In all cases,straight lines werefitted to experimental points using a linear regressionanalysiscomputer program. 1261

Vol.

181,

No.

3, 1991

BIOCHEMICAL

AND

BIOPHYSICAL

RESEARCH

COMMUNICATIONS

the origin was obtained (Fig.lB). This is indicative of a saturable phenomenon, which can be described by the scheme: k +l E+I =EI

kinact -E-I

k-1

where E and I arc the enzyme and inactivator, EI is a non covalent inactivator-enzyme complex and E-I the covalendy modified inactivated enzyme (44). The scheme yields

From the plot of Fig. 1B values of 1.6 + 0.6 mM and 2.2 f 0.5 x 10s2min -l were obtained for Kinact and kinact respectively, where Kinact = (k-l+ kh&k+l). Protection exueriments NADPH at 5 mM afforded complete protection against inactivation (not shown). At lower concentrations, it provided partial protection. The effect obtained with 0.5 mM NADPH is graphically represented in Fig. 1B. The fact that the lines with and without NADPH intersect on the ordinate indicates that the inactivator behaves competitively with the cofactor. The slope of the line obtained in the presence of NADPH is

(&act / kinact)(1 + PKp) where P is the protector, i.e. NADPH and Kp its dissociation constant (44). The plot of Fig.2B yielded Kp = 304* 20 PM. Other nucleotides were also tested for a possible protective action, at a fixed concentration but variable PLP-AMP concentrations. Their effects, determined at 50% inactivation in the absence of protection, are compared in Table I. It is clear that ADP and GTP-Y-S were inactive, while NADP+, NADH and ATP afforded a less marked protection. The most effective agents were NADPH itself, and interestingly enough, 2’,5’-ADP and ATP-ribose (Z-monophosphoadenosine S-diphosphoribose), agents which mimic a part of the NADPH structure since they carry a %‘-phosphate group. .

.

.

enc stoichiometrv &cttvauon Pending the use of a radioactive reducing agent, a kinetic stoichiometry of inactivation can be determined as shown in Fig.2. The double log plot of k&s versus inactivator concentration obeys the equation log k&S = dog I+ log k where n is the number of moles of reagent required to inactivate the system according to k E + n1 + E -I,, and kchs = k [IJ” at each inactivator concentration. From Fig.2, values of n = 0.6 + 0.1 and n = 0.75 f 0.15 were found in the absence and the presence of NADPH respectively. These figures 1262

Vol. 181, No. 3, 1991

BIOCHEMICAL

AND BIOPHYSICAL RESEARCH COMMUNICATIONS

Table 1. Protecting effect of various nucleotide analogs Addition (1 nw

PLP.AMP 0.77 mM

PLP.AMP 1.35 mM

PLP.AMP 1.5 mM

None

50%

50%

50%

81.5 * 4.5 (n = 2)

(n’=” 1)

NADPH NADP+ NADH ATP

77 zk 3.4 (n = 3)

65 (n = 1)

70 (n = 1)

62+4 (I-l =2)

(n z’l)

63.5 f 1.5 (n = 2)

(n6z 1)

GTP-?/-S (n5i1 )

(n”=” 1)

(try 1)

(n5z 1)

56.5 31 5.5 (n = 2)

ADP 2’,5’-ADP

80f 1 (n = 2)

ATP-ribose

74.5 * 3.5 (n = 2)

At each PLP-AMP concentration and for each nucleotide, the activity loss was monitored function of time as described in Methods. Semilog plots similar to those of Fig.lA constructed. The time point at which the unprotected control reached 50% inactivation determined graphically and the residual activity in the presence of protecting nucleotide read on the graph at this time point ; n is the number of independent experiments.

as a were was was

strongly suggestthat inactivation of the particulate NADPH oxidase is caused by one mole of

reagent reacting with one component in the preparation. This result does not preclude other chemical modification events which would be without consequence for the activity.

-1.6

-

-NADPH

-1.8

-

+NAOPH

-2.6 -0.4

I

-0.2

I

I

I

0 0.2 0.4 Log (PLP-AMP)

I

0.6

&,& Kinetic determination of the stoichiometry of NADPH oxidase inactivation by PLPAMP. The kobs values ate the same as those that were used in Fig.lB. 0 : no NADPH present during inactivation; 0 : 0.5 mM NADPH present.

1263

Vol. 181, No. 3, 1991

BIOCHEMICAL

AND BIOPHYSICAL

RESEARCH COMMUNICATIONS

Discussion The results reported above suggest that PLP-AMP is a promising reagent for identifying the flavoprotein of NADPH oxidase by labeling its NADPH binding site. The nucleoude analogue behaves as an affinity label with respect to the superoxide producing activity, and its inactivating effect is efficiently prevented by NADPH and those nucleotide analogues which carry a 2’-phosphate group on the ribose ring. Inactivation appears to result from a single chemical modification event, as kinetically determined. The protection constant against inactivation, Kp, was found to be 300 I.~M for NADPH. Measuring superoxide anion production, we found a K, for the cofactor of 47 pill under our experimental conditions for inactivation (lO”C>, a value which lies well within the range of published ones (16,17,19,43). Thus the values for K, and K, differ by a factor of about 6. This difference however does not contradict the idea that PLP-AMP labels the NADPH binding site of the oxidase. Indeed, while K, is expected to be the dissociation constant of NADPH for the binding site of PLP-AMP, the K, for superoxide production is not necessarily identical to a K, especially in such a complex reaction sequence as that involving an electron donor, an electron acceptor and two redox enzymes. Further work should lead to the identification of the labeled target, since the reversible Schiff base linkage formed between an aldehyde and a lysine can be rendered irreversible by reduction with borohydride or cyanoborohydride. The use of tritiated reductants allows incorporation of a radioactive label into the modified protein. Even if the chemical modification by PLP-AMP turns out not to be entirely specific, as might be expected from previous work with NADPHo, (33-36), it can be hoped that the use of protecting agents will allow an identification by difference. It should be noted that the experiments described in this work were carried out in the absence of reductant. The residual NADPH oxidase activity was measured after a 2%fold dilution of the inhibited sample. Preliminary experiments in the presence of cyanoborohydride showed that this dilution factor is insufficient to significantly reverse the equilibrium for the putative Schiff base formation during the few minutes required for the assay. This suggests that the Schiff base must be rather stable.

References Babior, B.M., Curnutte, J.T., and MC Munich, B.J. (1976) J. Clin. Invest. 58, 989-996. :. Gabig, T.G., and Babior, B.M. (1981) Annu. Rev. Med. 32, 313-326.

3: Babior, B.M. (1984) Blood 64,959~966. 4. Cross, A.R., Jones, O.T.G., Harper, A.M., and Segal, A.W. (1981) B&hem. J. 194, 599-606. Babior, B-M., and Kipnes. R.S. (1977) Blood 50, 517-524. 2 Rossi, F. (1986) Biochim. Biophys. Acta 853,65-89. 7: Harper, A.M., Dunne, M.J., and Segal, A.W. (1984) Biochem. J. 219, 519-527. 8. Parkas, C.A., Allen, R.A., Cochrane, C.G. and Jesaitis, A.J. (1987) J. Clin. Invest. 80, 732-742. 9. Segal, A.W. (1987) Nature 326,88-91. 10. Royer-Pokora, B., Kunkel, L.M., Monaco, A.P., Goff, S.C., Newburger, P.E., Bachner, R.L., Cole, F.S., Cumutte, J.T., and Grkin, S.H. (1986) Nature (London) 322, 32-38. 1264

Vol.

181, No. 3, 1991

BIOCHEMICAL

AN0 BIOPHYSICAL

RESEARCH COMMUNICATIONS

11. Parkos, C.A., Dinauer, M.C., Walker, L.E., Allen, R.A., Jesaitis, A.J., and G&in, S.H. (1988) Proc. Natl. Acad. Sci. USA 85,3319-3323. 12. Dinauer, M.C., G&in, S.H., Brown, R., Jesaitis, A.J., and Parkos, C.A. (1987) Nature (London) 327.717-720. 13. Teahan, C., Rowe, P., Parker, P., Totty, N., and Segal, A.W. (1987) Nature (London) 327, 720-721. 14. Hamers, M.N., de Boer, M., Meerhof, L.J., Weening, R.S., and Roos, D. (1984)

Nature, 307, 553-555. 15. Dinauer, M.C., Cumutte, J.T., Rosen, H., and Grkin, S.H. (1989) J. Clin. Invest. 64, 2012-2016. 16. Dinauer, M.C., Pierce, E.A., Bruns, G.A.P., Curnutte, J.T., and Grkin, S.H. (1990) J. Clin. Invest. 86, 1279-1737. 17. Gabig, T.G., and Babior, B.M. (1979) J. Biol. Chem. 254, 90709074. 18. Light, D.R., Walsh, C., O’Callaghan, A.M., Goetzl, E.J., and Tauber, A.I. (1981) Biochemistry 20, 1468-1476. 19. Bellavite, P. (1988) Free Rad. Biol. Med. 4.225-261. 20. Doussitre, J., Vignais, P.V. (1985) Biochemistry 24,7231-7239. 21. Ma&en, M., Glass, G.A., and Babior, B.M. (1985) Proc. Natl. Acad. Sci. USA 82, 3144-3148. 22. Glass, G.A., Delisle, D.M., DeTogni, P., Gabig, T.G., Magee, G.H., Markert, M., and Babior, B.M. (1986) J. Biol. Chem. 261, 13247-13251. 23. Bellavite, P., Cassatella, M.A., Papini, E., Magyeri, P., and Rossi, F. (1986) FEBS Lett. 199, 159-163. 24. Kakinuma, K., Fukuhara, Y., and Kaneda, M. (1987) J. Biol. Chem. 262, 12316-12322. 25. Laporte, F., Doussitre, J., Mechin, V., and Vignais, P.V. (1991) Eur. J. B&hem. 196, 59-66. 26. Yea, C.M., Cross, A.R., and Jones, O.T.G. (1990) B&hem. J. 265, 95-100. 27. Cross, A.R., and Jones, O.T.G. (1986) B&hem. J. 237, 111-116. 28. Ellis, J.A., Mayer, S.J., and Jones, O.T.G. (1988) B&hem. J. 251, 887-891. 29. Doussiere, J., and Vignais, P.V. (1991) B&hem. Biophys. Res. Commun. 175, 143151. 30. Doussiere, J., Laporte, F., and Vignais, P.V. (1986) Biochem. Biophys. Res. Commun. 139, 85-93.

31. Doussiere, J., Laporte, F., and Vignais, P.V. (1990) B&hem.

Biophys. Res. Commun.

168, 78-84. 32. Umei, T., Takeshige, K., and Minakami,

S. (1986) J. Biol. Chem. 261, 5229-5232. 33. Smith, R.M., Curnutte, J.T., and Babior, B.M. (1989) J. Biol. Chem. 264, 1958-1962. 34. Smith, R.M., Curnutte, J.T., Mayo, L.A., and Babior, B.M. (1989) J. Biol. them. 264, 12243-12248. 35. Takasugi, S-i., Ishida, K., Takeshige, K., and Minakami, S. (1989) J. B&hem. (Tokyo) 105, 155-157. 36. Umei, T., Babior, B.M., Curnutte, J.T., and Smith, R.M. (1991) J. Biol. Chem. 266, 60196022. 37. Tamura, J.K., Rakov, R.D., and Cross, R.L. (1986) J. Biol. Chem. 261, 4126-4133. 38. Tagaya, M., and Fukui, T. (1986) Biochemistry 25,2958-2964. 39. Mojana, N.A., Lyons, C., and Flynn, T.G. (1989) J. Biol. Chem. 264, 2912-2919. 40. Tamura, J.K., and Gellert, M. (1990) J. Biol. Chem. 265, 2 1342-21349. 41. BByum, A. (1968) Stand J. Clin. Lab. Invest. 21 (suppl. 97), 77-89. 42. Marker& M., Andrews, P.C., and Babior, B.M. (1984) Methods Enzymol. 105, 358-365. 43. Tauber, A.I., and Goetzl, E. (1978) Biochemistry 18.5576-5584. 44. Meloche, P. (1967) Biochemistry 6,2273-2280.

1265