NADPH oxidase 4 function as a hydrogen peroxide sensor.

Nox4, a member of the NADPH- and oxygen-dependent oxidoreductases that generate reactive oxygen species (ROS), is widely expressed and constitutively active. To understand better its function and regulation, specific mutations in the Nox4 dehydrogenase (DH) domain were examined for effects on Nox4 oxidase activity. Transfection of His6-tagged Nox4 increased the amount of p22phox subunit in HEK293 cells, and a higher level of the heterodimer was observed in the nucleus-enriched fraction (NEF). NEF from Nox4-expressing HEK293 cells exhibited oxygen and H2O2 concentration-dependent NADPH oxidation rate. In Nox4-expressing cells, NEF and its partially purified form, the Nox4(P437H) mutant almost completely lost its oxidase activity, while Nox4(C546S), Nox4(C546L) or/and (C547L) had a significantly decreased rate of ROS production. The NADPH-dependent reduction of cytochrome c or cytochrome b5 by purified Nox4 DH domain was found regulated by the H2O2 concentration, and C546L and C547L mutants showed lower rates of the hemeprotein reduction. These conserved Cys residues in the DH domain respond to the cytosolic H2O2 concentration to regulate Nox4 activity.

Nox4: a hydrogen peroxide-generating oxygen sensor.

Nox4 is an oddity among members of the Nox family of NADPH oxidases [seven isoenzymes that generate reactive oxygen species (ROS) from molecular oxygen] in that it is constitutively active. All other Nox enzymes except for Nox4 require upstream activators, either calcium or organizer/activator subunits (p47(phox), NOXO1/p67(phox), and NOXA1). Nox4 may also be unusual as it reportedly releases hydrogen peroxide (H2O2) in contrast to Nox1-Nox3 and Nox5, which release superoxide, although this result is controversial in part because of possible membrane compartmentalization of superoxide, which may prevent detection. Our studies were undertaken (1) to identify the Nox4 ROS product using a membrane-free, partially purified preparation of Nox4 and (2) to test the hypothesis that Nox4 activity is acutely regulated not by activator proteins or calcium, but by cellular pO2, allowing it to function as an O2 sensor, the output of which is signaling H2O2. We find that approximately 90% of the electron flux through isolated Nox4 produces H2O2 and 10% forms superoxide. The kinetic mechanism of H2O2 formation is consistent with a mechanism involving binding of one oxygen molecule, which is then sequentially reduced by the heme in two one-electron reduction steps first to form a bound superoxide intermediate and then H2O2; kinetics are not consistent with a previously proposed internal superoxide dismutation mechanism involving two oxygen binding/reduction steps for each H2O2 formed. Critically, Nox4 has an unusually high Km for oxygen (∼18%), similar to the values of known oxygen-sensing enzymes, compared with a Km of 2-3% for Nox2, the phagocyte NADPH oxidase. This allows Nox4 to generate H2O2 as a function of oxygen concentration throughout a physiological range of pO2 values and to respond rapidly to changes in pO2.

Nox4 B-loop creates an interface between the transmembrane and dehydrogenase domains.

By targeting redox-sensitive amino acids in signaling proteins, the NADPH oxidase (Nox) family of enzymes link reactive oxygen species to physiological processes. We previously analyzed the sequences of 107 Nox enzymes and identified conserved regions that are predicted to have important functions in Nox structure or activation. One such region is the cytosolic B-loop, which in Nox1-4 contains a conserved polybasic region. Previous studies of Nox2 showed that certain basic residues in the B-loop are important for activity and translocation of p47(phox)/p67(phox), suggesting this region participates in subunit assembly. However, conservation of this region in Nox4, which does not require p47(phox)/p67(phox), suggested an additional role for the B-loop in Nox function. Here, we show by mutation of Nox4 B-loop residues that this region is important for Nox4 activity. Fluorescence polarization detected binding between Nox4 B-loop peptide and dehydrogenase domain (K(d) = 58 +/- 12 nm). This interaction was weakened with Nox4 R96E B-loop corresponding to a mutation that also markedly decreases the activity of holo-Nox4. Truncations of the dehydrogenase domain localize the B-loop-binding site to the N-terminal half of the NADPH-binding subdomain. Similarly, the Nox2 B-loop bound to the Nox2 dehydrogenase domain, and both the Nox2 and Nox4 interactions were dependent on the polybasic region of the B-loop. These data indicate that the B-loop is critical for Nox4 function; we propose that the B-loop, by binding to the dehydrogenase domain, provides the interface between the transmembrane and dehydrogenase domains of Nox enzymes.

Constitutive NADPH-dependent electron transferase activity of the Nox4 dehydrogenase domain.

NADPH oxidase 4 (Nox4) is constitutively active, while Nox2 requires the cytosolic regulatory subunits p47(phox) and p67(phox) and activated Rac with activation by phorbol 12-myristate 13-acetate (PMA). This study was undertaken to identify the domain on Nox4 that confers constitutive activity. Lysates from Nox4-expressing cells exhibited constitutive NADPH- but not NADH-dependent hydrogen peroxide production with a K(m) for NADPH of 55 +/- 10 microM. The concentration of Nox4 in cell lysates was estimated using Western blotting and allowed calculation of a turnover of approximately 200 mol of H(2)O(2) min(-1) (mol of Nox4)(-1). A chimeric protein (Nox2/4) consisting of the Nox2 transmembrane (TM) domain and the Nox4 dehydrogenase (DH) domain showed H(2)O(2) production in the absence of cytosolic regulatory subunits. In contrast, chimera Nox4/2, consisting of the Nox4 TM and Nox2 DH domains, exhibited PMA-dependent activation that required coexpression of regulatory subunits. Nox DH domains from several Nox isoforms were purified and evaluated for their electron transferase activities. Nox1 DH, Nox2 DH, and Nox5 DH domains exhibited barely detectable activities toward artificial electron acceptors, while the Nox4 DH domain exhibited significant rates of reduction of cytochrome c (160 min(-1), largely superoxide dismutase-independent), ferricyanide (470 min(-1)), and other electron acceptors (artificial dyes and cytochrome b(5)). Rates were similar to those observed for H(2)O(2) production by the Nox4 holoenzyme in cell lysates. The activity required added FAD and was seen with NADPH but not NADH. These results indicate that the Nox4 DH domain exists in an intrinsically activated state and that electron transfer from NADPH to FAD is likely to be rate-limiting in the NADPH-dependent reduction of oxygen by holo-Nox4.

Activation of NADPH oxidase 1 in tumour colon epithelial cells.

In the plasma membrane fraction from Caco-2 human colon carcinoma cells, active Nox1 (NADPH oxidase 1) endogenously co-localizes with its regulatory components p22(phox), NOXO1, NOXA1 and Rac1. NADPH-specific superoxide generating activity was reduced by 80% in the presence of either a flavoenzyme inhibitor DPI (diphenyleneiodonium) or NADP(+). The plasma membranes from PMA-stimulated cells showed an increased amount of Rac1 (19.6 pmol/mg), as compared with the membranes from unstimulated Caco-2 cells (15.1 pmol/mg), but other components did not change before and after the stimulation by PMA. Spectrophotometric analysis found approx. 36 pmol of FAD and 43 pmol of haem per mg of membrane and the turnover of superoxide generation in a cell-free system consisting of the membrane and FAD was 10 mol/s per mol of haem. When the constitutively active form of Rac, Rac1(Q61L) or GTP-bound Rac1 was added exogenously to the membrane, O(2)(-)-producing activity was enhanced up to 1.5-fold above the basal level, but GDP-loaded Rac1 did not affect superoxide-generating kinetics. A fusion protein [NOXA1N-Rac1(Q61L)] between truncated NOXA1(1-211) and Rac1-(Q61L) exhibited a 6-fold increase of the basal Nox1 activity, but NOXO1N(1-292) [C-terminal truncated NOXO1(1-292)] alone showed little effect on the activity. The activated forms of Rac1 and NOXA1 are essentially involved in Nox1 activation and their interactions might be responsible for regulating the O(2)(-)-producing activity in Caco-2 cells.

Rosmarinic acid inhibits the formation of reactive oxygen and nitrogen species in RAW264.7 macrophages.

Antioxidant action of Rosmarinic acid (Ros A), a natural phenolic ingredient in many Lamiaceae herbs such as Perilla frutescens, sage, basil and mint, was analyzed in relation to the Ikappa-B activation in RAW264.7 macrophages. Ros A inhibited nitric oxide (NO) production and inducible nitric oxide synthase (iNOS) protein synthesis induced by lipopolysaccharide (LPS), and also effectively suppressed phorbol 12-myristate 13-acetate (PMA)-induced superoxide production in RAW264.7 macrophages in a dose-dependent manner. Peroxynitrite-induced formation of 3-nitrotyrosine in bovine serum albumin and RAW264.7 macrophages were also inhibited by Ros A. Moreover, Western blot analysis demonstrated that LPS-induced phosphorylation of Ikappa-Balpha was abolished by Ros A. Ros A can act as an effective protector against peroxynitrite-mediated damage, and as a potent inhibitor of superoxide and NO synthesis; the inhibition of the formation of reactive oxygen and nitrogen species are partly based on its ability to inhibit the serine phosphorylation of Ikappa-Balpha.

Activation of the flavoprotein domain of gp91phox upon interaction with N-terminal p67phox (1-210) and the Rac complex.

A series of truncated forms of His(6)-tagged gp91phox were expressed, solubilized, and purified in the presence of 30 microM FAD. The truncated gp91phox with the longest sequence in the C-terminal region (221-570) (gp91C) showed the highest activity (turnover rate, 0.92) for NADPH diaphorase in the presence of either 0.3% Triton X-100 or 0.5% Genapol X-80. Activity was not inhibited by superoxide dismutase but was blocked by an inhibitor of the respiratory burst oxidase, diphenylene iodonium. The flavinated gp91C contained approximately 0.9 mol of FAD/mol of protein (MW 46 kDa) and 12% alpha-helix content. In the absence of p47phox, p67phox showed considerable activation of gp91C in the presence of Rac. Carboxyl-terminal truncated p67phox (1-210) (p67N), which is the minimal active fragment, was fused with Rac or Q61LRac. The fusion protein p67N-Rac (or p67N-Q61LRac) showed a 2-fold higher stimulatory effect on NBT reductase activity of gp91C than the combination of the individual cytosolic p67N and Rac proteins. In contrast, Rac-p67N, a fusion with the opposite orientation, showed a smaller significant effect on the enzyme activity. The EC(50) values for p67phox, p67N, p67N-Rac, and Rac-p67N were 8.00. 4.35, 2.56, and 15.2 microM, respectively, while the K(m) value for NADPH in the presence and absence of the cytosolic components was almost the same (40-55 microM). In the presence of Rac, p67N or p67phox bound to gp91C with a molar ratio of approximately 1:1 but neither p67N nor Rac alone showed significant binding.

Interaction between p21-activated protein kinase and Rac during differentiation of HL-60 human promyelocytic leukemia cell induced by all-trans-retinoic acid.

Undifferentiated human promyelocytic leukemia HL-60 cells show little or no superoxide production, but generate a very low O(2)(-) concentration upon incubation with all-trans-retinoic acid (ATRA). Its production reaches a maximum within 20 h, and thereafter is maintained at an almost constant level. The differentiated cells show phorbol 12-myristate 13-acetate (PMA)-stimulated NADPH oxidase activity consistent with the amount of gp91phox (phagocytic oxidase) expressed in the plasma membrane. Three isoforms of p21-activated serine/threonine kinases, PAK68, PAK65 and PAK62, were found in both cytosolic and membrane fractions, and their contents were significantly increased during induced differentiation. The amount of Rac identified in the two fractions was also markedly enhanced by ATRA- induced differentiation. In contrast, neither PAK nor Rac was seen in the plasma membrane of undifferentiated HL-60 or human neutrophil, but they were abundant in the cytoplasmic fraction. Binding of Rac with PAK isoforms was shown in the membrane upon induced differentiation of HL-60 cells. Direct binding of purified Rac1 to PAK68 was quantified using a fluorescent analog of GTP (methylanthraniloyl guanosine-5'-[beta,gamma-imido]triphosphate) bound to Rac as a reporter group. Rac1 bound to PAK68 with a 1 : 1 stoichiometry and with a K(d) value of 6.7 nm.