| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Departments of Medicine (H.G., S.R.) and Pediatrics (S.R.), and Committee on Genetics (S.R.), The University of Chicago, Chicago, Illinois 60637; Institut de Recherche Interdisciplinaire en Biologie Humaine et Moléculaire (X.D.D., F.M.), Université Libre de Bruxelles, B-1070 Brussels, Belgium; and Children’s Hospital of the Johannes Gutenberg University (J.P.), D-55101 Mainz, Germany
Address all correspondence and requests for reprints to: Helmut Grasberger, The University of Chicago, MC3090, 5841 South Maryland Avenue, Chicago, Illinois 60637. E-mail: hgrasber{at}uchicago.edu.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Dual oxidases (DUOX1 and DUOX2; formerly known as thyroid oxidases) have been identified as candidates for the thyroid H2O2 generator by purification of a thyroidal NADPH oxidase (NOX) flavoprotein (7) and by screening of a thyroid cDNA library for homologs of gp91phox/NOX2 (8), the catalytic core of the phagocyte NOX involved in transfer of electrons across the membrane. The gp91phox homolog region is composed of six transmembrane helices harboring conserved coordination sites for two heme prosthetic groups and a C-terminal domain comprising NADPH and flavin adenine dinucleotide binding sites (Fig. 1
). DUOXs are long members of the NOX/DUOX family, with a unique N-terminal peroxidase-like domain, followed by a membrane-spanning segment and an additional cytosolic domain containing two EF-hand calcium-binding motifs.
|
Previously, molecular studies of DUOX proteins had been hampered by the failure to reconstitute active DUOX enzymes in heterologous systems, because recombinant DUOX was completely retained inside the endoplasmic reticulum (ER) in an immature form (13, 14, 15, 16, 17). We recently showed that reconstitution of a DUOX-based H2O2 generator requires activation by specific maturation factors (DUOXA1 and DUOXA2) that are genetically linked and coexpressed with the respective DUOX genes (18). The two DUOXA genes encode novel five transmembrane proteins with a conserved, N-glycosylated domain between TM2 and -3 (Fig. 1
).
We have now used a DUOXA2 reconstituted system to characterize three natural DUOX2 missense variants (Q36H, R376W, D506N) (10, 11, 12) at the molecular level, analyzing the effects of these mutations on DUOX2 function, trafficking, stability, folding, and DUOXA2 interaction. Our results indicate that trafficking defects are the common feature of these congenital hypothyroidism-linked DUOX2 mutations. DUOXA2 is not involved in the oxidative folding of DUOX2 in the ER, but rather appears to interact with the already folded DUOX2 conformation resulting in ER exit of wild-type (WT) DUOX2 but enhanced degradation of mutant DUOX2 not competent for ER exit.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
). In contrast, cotransfections of the Q36H and R376W DUOX2 vectors with DUOXA2 did not reconstitute DUOX2 activity. Significant H2O2 production was observed with the D506N mutant, albeit at approximately 50% of WT level (0.80 ± 0.18 nmol/well/h). The H2O2 release triggered by WT and D506N mutant DUOX2 was completely blocked by including in the incubation medium 10 µM diphenyleneiodonium, a flavoprotein inhibitor (data not shown). In cotransfection experiments, the Q36H and R376W mutant DUOX2 did not affect the activity of WT DUOX2, and the activities of D506N and WT DUOX2 were additive, indicating absence of dominant-negative effects of these mutant DUOX2 proteins in vitro (data not shown). It should be mentioned that neither WT nor mutant DUOX2 released detectable amounts of superoxide anion (tested with the sensitive chemiluminescence substrate Diogenes), and they were also devoid of significant peroxidase activity (tested with 10-acetyl-3,7-dihydroxyphenoxazine and tetramethyl benzidine) (data not shown). Taken together, we conclude that all three DUOX2 missense mutations indeed cause either a partial or complete defect of H2O2 generation in the DUOXA2 reconstituted system.
|
). We found that Ca2+ stimulated the activity of WT DUOX2 (5.1 ± 0.2 vs. 2.7 ± 0.3 nmol H2O2/100 µg protein/h) and D506N DUOX2 (2.7 ± 0.2 vs. 0.7 ± 0.1 nmol H2O2/100 µg protein/h) in the heterologous system, but had no stimulatory effect on the Q36H and R376W mutants lacking basal activity.
To explore the functional characteristics of WT DUOX2 and the partial deficiency mutant D506N in greater detail, we determined the effect of DUOX2 expression level and DUOX2-to-DUOXA2 ratio on H2O2 release. Within the tested range, the amount of transfected DUOX2 cDNA was proportional to the amount of DUOX2 protein and did not interfere with the expression of DUOXA2 driven from a constant amount of cotransfected DUOXA2 vector (Fig. 2D). Protein expression and H2O2 generation by D506N DUOX2 were equivalently reduced to 35–60% of WT protein (Fig. 2, C and D). The plateau in H2O2 generation with increasing DUOX2 expression level occurred within the linear range of the fluorometric assay (Fig. 2A, inset) and was apparently not due to heme or NADPH depletion because addition of iron porphyrins to the culture medium (1 µg/ml hemin or hematin) or D-glucose (5 mM) to the incubation buffer had no effect on DUOX2 activity (data not shown). The amount of coexpressed DUOXA2 was also not limiting in our system and DUOXA2 overexpression did not restore the activity of the D506N mutant to the level of WT DUOX2 (Fig. 2, E and F). These dose-response characteristics were congruent with the concept that DUOXA2 is required for ER exit of DUOX2, but not integral part of the active enzyme complex at the plasma membrane (18).
Impaired Trafficking of Mutant DUOX2 to the Cell Surface
To further explore the mechanisms leading to absent or reduced H2O2 generation by the mutant proteins, we assessed the targeting of the mutant proteins to the plasma membrane by indirect immunofluorescence microscopy and flow cytometry. We found that the D506N DUOX2 protein could be detected at the cell surface of nonpermeabilized cells, albeit at clearly reduced level compared with WT DUOX2 (Fig. 3A). In contrast, no specific surface signal was obtained for the Q36H and R376W mutants. Analysis of permeabilized cells by confocal laser-scanning microscopy revealed that this was not due to increased intracellular accumulation or aggrosome formation of the mutant proteins (Fig. 3B). WT and mutant proteins were detected in the same compartment as the ER marker calnexin (CANX) (Ref. 18 and data not shown) and found associated with individual ER tubules as well as the nuclear envelope (Fig. 3B). Colocalization of DUOX2 with a Golgi marker, golgin 97, was not detectable at steady state (data not shown), suggesting that trafficking through the Golgi apparatus is a rapid process, compared with the residence time of immature DUOX2 in the ER or the half-life of mature DUOX2 at the cell surface. D506N cell surface expression was quantitated by immunofluorescence microscopy (Fig. 3C) and flow cytometry (Fig. 3, D and E), with both methods showing a reduction to approximately 35% of WT DUOX2 level. Plotting the activity of WT and mutant DUOX2 transfected cells as function of their corresponding surface expression confirmed that the D506N mutation does not diminish the intrinsic activity of DUOX2 in generating H2O2 (Fig. 3F). Thus, impaired trafficking to the cell surface appeared to be the common molecular mechanism underlying the functional deficiency of these DUOX2 missense mutations linked to congenital hypothyroidism.
|
, lanes 3 and 5). Importantly, the ratio of mature to immature DUOX2 remained unaltered irrespective of the attained DUOX2 expression level (Fig. 4B). Such low intrinsic maturation efficiency of DUOX2 was not peculiar to the heterologous expression system but rather recapitulated the findings for endogenous DUOX2 protein in various tissues, for which both glycosylated forms had been observed in similar proportion (13, 21, 22). In marked contrast to WT DUOX2, the three DUOX2 missense mutations remained in a completely Endo H-sensitive form despite coexpression of DUOXA2 (Fig. 4, A, lanes 9, 13 and 17, and B).
|
-mannosidase II, on trafficking and activity of WT DUOX2. The hybrid N-glycosylation of WT DUOX2, synthesized in the presence of swainsonine, was completely sensitive to Endo H (Fig. 5A, compare lanes 6 and 7). Yet, there was no effect of swainsonine on the surface targeting (Fig. 5B) or activity of WT DUOX2 (Fig. 5C). These findings imply that maturation of the N-glycans in the Golgi was indeed dispensable for trafficking and activity of DUOX2, compatible with the cell surface expression and full intrinsic activity of Endo H-sensitive D506N DUOX2. On the contrary, cotranslational core N-glycosylation of DUOX2 appeared to be crucial for stabilizing the nascent protein, because unglycosylated WT DUOX2 did not accumulate to a detectable level when synthesized in the presence of tunicamycin (Fig. 5A, lanes 8 and 9).
|
, lanes 7, 9, 11, 13, 15, and 17). Because neither WT nor mutant DUOX2 proteins remained in the insoluble fraction after protein extraction (data not shown), we hypothesized that DUOXA2 increased the turnover of DUOX2 proteins not competent for ER exit. To test this concept, we analyzed the stability of WT and mutant DUOX2 proteins in cycloheximide (CHX) chase experiments. We found that DUOXA2 indeed accelerated the turnover of the Q36H and R376W DUOX2 proteins (Fig. 6A). DUOXA2 had no apparent effect on the turnover of WT DUOX2, the estimated half-life of which was comparable to that reported for endogenous DUOX proteins in CHX-treated dog thyrocytes (12–16 h) (13). CHX seemed not to prevent maturation and Golgi transition of WT DUOX2 as indicated by the increase in the level of Endo H-resistant form within the first 2 h of the chase (Fig. 6B). As expected from the constant steady-state expression level in the absence of CHX, the stability of DUOXA2 was not affected by the coexpression of WT or mutant DUOX2 constructs. Overall, these results suggested that DUOXA2 is not only crucial for ER export of DUOX2, but can also drive DUOX2 proteins, that are incompetent for ER exit, along a degradation pathway.
|
). No bands were observed in the region corresponding to twice the apparent molecular weight of DUOX2, as would be expected with the formation of an intermolecular bond. Deglycosylation with peptide N-glycosidase F (PNGase F) demonstrated that both conformations were to the same extent N-glycosylated (Fig. 7B, lanes 2 and 4). In the presence of DUOXA2, the oxidized form had a slightly decreased electrophoretic mobility, consistent with complex N-glycosylation of DUOX2 (Fig. 7B, lane 6). However, the oxidized forms of DUOX2 synthesized in the absence or presence of DUOXA2 had identical mobility after PNGase F treatment (Fig. 7B, lanes 4 and 8). Because the latter comprised the mature protein at the cell surface, we can conclude that DUOX2 does not depend on DUOXA2 to acquire native disulfide bonds, because incomplete or nonnative disulfide bond formation would affect the electrophoretic mobility (24, 25). Overall, the high level of reduced folding intermediate indicated that the reduced-to-oxidized transition of DUOX2 is intrinsically inefficient. Furthermore, oxidative folding appeared to be the rate-limiting step in DUOX2 maturation, because, in the presence of DUOXA2, the vast majority of oxidized DUOX2 is Endo H resistant, indicating rapid ER exit of the oxidized form (Figs. 7C, lane 3, and 8).
|
|
, lanes 2 and 3), suggesting that the D506N mutant folded as inefficiently and/or slowly as WT DUOX2 in terms of disulfide bond formation. Coexpression of DUOXA2 did not cause a mobility shift of oxidized D506N DUOX2, a finding consistent with the lack of complex-type N-glycosylation. Nevertheless, the oxidized-to-reduced ratios of WT and D506N DUOX2 remained similar (Fig. 7B, lanes 6 and 7). There was no indication of nonnative disulfide bond formation in D506N, because the oxidized forms of D506N and WT had identical electrophoretic mobility after enzymatic deglycosylation (Fig. 7B, lanes 8 and 9). In contrast, the two complete deficiency mutants (Q36H, R376W) were almost exclusively present in the reduced folding intermediate (Fig. 7D). Using longer exposure times, we could detect trace amounts of the folded conformation for these mutants (Fig. 7E). Because coexpression of DUOXA2 caused a decrease of the oxidized-to-reduced ratio of these mutant DUOX2 proteins at steady state (Fig. 7E), DUOXA2-mediated turnover may primarily target the oxidized conformation.
Abnormal Association of DUOX2 Mutants with DUOXA2 and CANX
The effects of DUOXA2 on DUOX2 maturation and/or degradation suggested a direct physical interaction of the two proteins. To test whether WT and mutant DUOX2 interact with DUOXA2, we coimmunoprecipitated DUOX2-associated proteins after in situ chemical cross-linking. All DUOX2 proteins were immunoprecipitated with similar efficiency, with the relative amount of DUOX2 in the immunoprecipitates matching the expression levels in the cell lysates (data not shown). DUOXA2 was specifically precipitated from lysates containing DUOXA2 and either WT or mutant DUOX2 (Fig. 9A). Relative to the amount of immunoprecipitated DUOX2, the amount of bound DUOXA2 was, however, significantly reduced for the R376W mutant (19 ± 6%; P < 0.01; two-tailed t test; n = 3) (Fig. 9D, left panel). This was confirmed by analyzing the amount of DUOX2 proteins in DUOXA2 immunoprecipitates: the fraction of R376W DUOX2 complexed with DUOXA2 was lower compared with the fraction of WT DUOX2 in complex with DUOXA2 (Fig. 9, C and D). Although our results suggested a similarly reduced association of the Q36H mutant DUOX2 with DUOXA2, the level of precipitated proteins were at the detection limit precluding reliable quantitation.
|
) could reflect its inability to associate with CANX. CANX has also been implicated in the quality control of protein folding by retaining misfolded glycoproteins within the ER (26). We, therefore, asked whether DUOX2 interacts with CANX and whether such interaction would be altered by the DUOX2 mutations and/or the presence of DUOXA2 in the system. We found that the proportions of Q36H and R376W mutant DUOX2 proteins in complex with CANX were higher compared with WT DUOX2, both in the presence and absence of DUOXA2 (Fig. 9B). These results are compatible with prolonged association with CANX during futile folding cycles retaining these DUOX2 mutant proteins within the ER, eventually leading to their degradation. DUOXA2 coexpression did not substantially, if at all, modify the expression level of CANX (Fig. 9A, input CAXN) and the relative amount of CANX complexed with WT or mutant DUOX2 in the ER. Note that the apparent decrease of CANX association with WT and D506N DUOX2 in the presence of DUOXA2 is likely accounted for by DUOX2 protein in post-ER compartments. Again, these results supported the view that DUOXA2 does not play a role as molecular chaperone in the early folding of DUOX2 in the ER. |
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
The mutated residues reside in the N-terminal peroxidase-like domain (Fig. 1). Morand et al. (15) have studied a truncated porcine DUOX2 protein consisting of the peroxidase-like domain and the first transmembrane segment (residues 1–685). This protein, but not the human counterpart, was partially targeted to the cell surface and had Endo H-resistant N-glycosylation in a heterologous system lacking DUOXA. Four conserved cysteine residues (C351, C370, C568, C582 of human DUOX2) within this domain appeared to be crucial for maturation of this domain, because their mutation rendered the N-glycans completely Endo H sensitive. Conceivably, those cysteine residues participate in intramolecular disulfide bonds stabilizing the native tertiary structure. Protein conformations differing in their thiol redox state can typically be separated by nonreducing electrophoresis, which therefore provided a means to probe the effect of the mutants on the folding of the peroxidase-like domain.
Our results showed that WT DUOX2 is present in two distinct redox forms at steady state. For most proteins, the mobility of the oxidized form is higher than that of the reduced form, which is attributed to an assumed more compact structure of the disulfide bonded form. The oxidized conformation of DUOX2, however, had lower mobility than the reduced form. The migration of the oxidized form was incompatible with that expected for a DUOX2 homodimer. Furthermore, separation of the oxidized form in a second dimension under reducing conditions only revealed a single band with the molecular weight of reduced DUOX2 by silver staining (Grasberger, H., unpublished observation). Reduced mobility of the oxidized form has been described for a few other proteins (27, 28, 29) and could relate to the formation of extended loops upon disulfide bond formation resulting in a larger Stokes radius.
DUOXA2 does not affect the reduced-to-oxidized transition, indicating that DUOXA2 likely functions at a later stage in DUOX2 maturation, once the DUOX2 protein has acquired a stable disulfide-bonded conformation. Importantly, the rate-limiting step in the maturation of DUOX2 is not the ER export per se: in the presence of DUOXA2, the oxidized DUOX2 is almost completely Endo H resistant (Fig. 7C), indicating efficient DUOXA2-mediated export of folded DUOX2 (Fig. 8). The substantial amounts of DUOX2 detectable in the ER are due to unfolded DUOX2 lacking stable disulfide bonds and the oxidized-to-reduced ratio of DUOX2 consequently mirrors the ratio of Endo H-resistant to -sensitive protein. Inefficient folding appears to be an intrinsic feature of the DUOX2 protein, with the equilibrium between the disulfide-bonded and reduced conformation not affected by the DUOX2-to-DUOXA2 ratio or the overall DUOX2 expression level. The latter finding also implies that inefficient folding is not caused by specific chaperones being limiting in the experimental system. It is expected that even slight perturbations of the folding process can shift the equilibrium further to the unfolded state resulting in complete ER retention. This is indeed the mechanism underlying the Q36H and R376W DUOX2 mutants, but presumably a common theme for other mutations yet to be characterized. This would not be unprecedented given that other proteins with notably inefficient folding, such as, GnRH receptor (30), erythropoietin receptor (31), and opioid receptors (32), are known to be particularly prone to intracellular retention when affected by a wide spectrum of mutations. Because there was no evidence for inappropriate interchain disulfide bonds, e.g. in form of dimers or higher order multimers, the Q36H and R376W DUOX2 mutants are conceivably retained by the quality control in the ER. These mutants showed indeed increased association with CANX (Fig. 9B), which can target misfolded proteins to retrotranslocation and ultimately proteosomal degradation (33, 34, 35, 36, 37). The interaction with the membrane-bound lectin CANX is likely crucial for stabilizing the nascent DUOX2 proteins (Fig. 8), because unglycosylated DUOX2 did not accumulate to detectable amounts when core N-glycosylation was blocked (Fig. 5A).
The D506N DUOX2 mutation also produces a trafficking defect as indicated by the reduced expression at the plasma membrane. However, it showed the same ability to fold into the disulfide-bonded conformation as the WT protein, as indicated by a similar ratio of oxidized-to-reduced forms. Once successfully routed to the plasma membrane, it also had the same intrinsic H2O2-generating activity as WT DUOX2. Compared with WT DUOX2, the amounts of oxidized and reduced forms, and the amount of surface targeted protein all appeared to be diminished to a similar degree. We thus believe that the defect in D506N DUOX2 manifests at a very early point in the ER maturation, preceding the oxidative folding. Unexpectedly, the D506N DUOX2 mutant protein was exclusively expressed with Endo H-sensitive N-glycan moieties, indicating that the latter are not subject to the same Golgi modification as in WT DUOX2. Clearly, N-glycans of plasma membrane proteins are not always rendered Endo H resistant in the Golgi (38, 39, 40), probably because of physical inaccessibility for the modifying enzymes (41). The latter would imply that D506N DUOX2 is routed through the secretory pathway despite nonnative conformation of its extracellular domain. Because the surface-expressed D506N protein has normal intrinsic activity, our results indicate that the complex N-glycosylation in the Golgi, and maybe the peroxidase-like domain in general, are not essential for the H2O2-generating activity of DUOX2. It should be noted that our functional data potentially underestimate the actual effect of this mutation on thyroidal H2O2 generation because the surface expression of D506N may be higher in the heterologous system due to escape from the ER quality control systems at elevated expression levels. In fact, the functional defect of D506N appeared to be more pronounced at low expression levels (<20% of WT activity) and saturated at higher expression levels (at 
60% of WT DUOX2) (Fig. 2F).
An interesting observation in the present study is the accelerated turnover of DUOX2 when DUOXA2 is coexpressed (Fig. 6A). Our best explanation for this phenomenon is a model where DUOXA2 would be involved in a partitioning step, leading either to ER exit of DUOX2 or targeting for degradation if a defect in the protein renders it incompetent for ER exit. Such concept is supported by the new equilibrium with diminished oxidized-to-reduced ratio of the Q36H and R376W DUOX2 mutants when DUOXA2 is coexpressed (Fig. 7E), likely reflecting the primary turnover of the oxidized conformation. Thus, beyond enabling ER exit of DUOX2, DUOXA2 may be part of a secondary quality control mechanism (42).
Our findings that the Q36H and R376W DUOX2 mutants are completely inactive in vitro may seem surprising in view of the partial organification defects observed in patients harboring one of these missense mutations in a compound heterozygous state with DUOX2 truncation mutations (10, 11). The most likely explanation for residual iodide organification activity in the thyroid glands of these patients, and also of other patients with presumptive truncation mutations on both DUOX2 alleles (11), is partial compensation by DUOX1, which is able to release H2O2 when activated with DUOXA1 (our unpublished results). At the mRNA level, DUOX1 and DUOXA1 are expressed at about 5-fold lower level in the human thyroid gland compared with DUOX2 and DUOXA2, respectively.
In conclusion, we report the first functional study of natural DUOX2 missense mutations linked to congenital hypothyroidism in a reconstituted heterologous system. The three mutations investigated showed loss or marked reduction of plasma membrane expression. The effect of the mutations on oxidative folding and/or conformation of the peroxidase-like domain is indicated by their inability to readily accumulate in an oxidized conformation or the abnormal maturation of the N-glycosylation in the Golgi complex. These mutations thus highlight the critical role of posttranslational processing of this domain in ER exit of DUOX2. DUOXA2 mediates ER exit of folded DUOX2, without directly affecting the oxidative folding process, and is likely involved in the secondary quality control in the ER.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
C (Q36H), c.1126CT (R376W), and c.1516GA (D506N) mutants were introduced into an N-terminal hemagglutinin (HA)-tagged DUOX2 expression vector (HA-DUOX2) by site-directed mutagenesis (QuikChange; Stratagene, La Jolla, CA). The sense primers (mutated nucleotides are underlined) were 5'-ACTGCCCTGGGAAGTGCACCGCTATGACGGCTGGT-3' for Q36H, 5'-GGTCTGCAACAACTACTGGATTTGGGAGAACCCCAATCTGA-3' for R376W, and 5'-TGTTCAGTGCCATTGTCCTCAACCAGTTTGTACGGCTGC-3' for D506N. All constructs were verified by bidirectional DNA sequencing. Presence of the HA-tag at the N terminus of the mature protein does not affect the peroxide-generating activity compared with untagged recombinant DUOX2 (18) and readily allows assessment of surface expression of the recombinant protein. The DUOXA2-myc/His and DUOXA2-EGFP/myc expression vectors were prepared as described (18).
Cell Culture and Transfection
PC CL3 cells, a rat thyroid cell line (43), were grown in Coon’s modified Ham’s F12 medium supplemented with 5% decomplemented fetal calf serum, 10 mU/ml bovine thyrotropin (Sigma, St. Louis, MO), 1 µg/ml insulin, and 5 µg/ml transferrin under 5% CO2/95% air at 37 C. HeLa cells, derived from human cervical cancer, and COS-7 cells, derived from transformed simian fibroblasts, were maintained in DMEM (Invitrogen Life Technologies, Carlsbad, CA) supplemented with 2 mM L-glutamine, 4.5 g/l D-glucose, 50 µg/ml gentamycin, and 10% fetal bovine serum. Adherent cells were transfected at 70–80% confluence with the indicated amounts of plasmid DNA using FuGENE 6 reagent (Roche Applied Science, Indianapolis, IN) or Effectene (Qiagen, Hilden, Germany). Renilla luciferase activity from cotransfected pRL-Tk plasmid (Promega, Madison, WI) or green fluorescence protein expression from a cotransfected pEGFP vector (Clontech Laboratories, Palo Alto, CA) were used in some experiments to monitor transfection efficiency in cell lysates or whole cells, respectively. The total amount of DNA transfected per square centimeter of cell monolayer (130 ng) was kept constant by adjusting with empty pcDNA3.1 vector.
Cell Lysates and Immunoblot Analysis
For analysis under reducing conditions, cells extracts were prepared in lysis buffer [50 mM Tris/HCl (pH 8.0), 150 mM NaCl, 1 mM DTT, 1% Nonidet P-40 supplemented with a protease inhibitor cocktail (Complete; Roche)] by rocking for 45 min at 4 C and cleared by centrifugation (12,000 x g, 15 min, 4 C). Total protein concentrations were determined by the Bradford method (Bio-Rad, Hercules, CA). Samples containing equal amounts of protein were incubated for 30 min (20 C) with equal volume of 4x Laemmli sample buffer, separated on 5 or 12.5% sodium dodecyl sulfate (SDS)-polyacrylamide gels and transferred to polyvinylidine fluoride membranes. To check for DUOX2 protein in the insoluble fraction, the 12,000 x g pellets were washed twice in 25 mM Tris/HCl (pH 8.0), 150 mM NaCl, and resolubilized in SDS-PAGE loading buffer by sonication. For analysis under nonreducing conditions, cells were washed and reacted for 15 min with 20 mM N-ethylmaleimide (NEM; Sigma) in ice-cold PBS to alkylate-free sulfhydryl groups. Cell lysates (in lysis buffer adjusted to pH 7.4, containing 20 mM NEM but no DTT) were denatured with an equal volume of 4x Laemmli sample buffer without reducing agent. Blots were probed, as indicated, with anti-HA (clone12CA5), anti-c-Myc (clone 9E10; both from Roche Applied Science), rabbit anti-CANX (StressGen, Victoria, British Columbia, Canada), or rabbit anti-actin (Sigma), all diluted to 1:4000. Relative expression levels were determined by scanning densitometry using Epson4870 (Epson, Long Beach, CA) and NIH ImageJ.
Cross-linking and Coimmunoprecipitation
Cells were washed in PBS and incubated for 20 min at 37 C in 2 mM cell-permeable cross-linker dithiobis(succinimidylpropionate) (Pierce, Rockford, IL), freshly prepared in PBS from a stock solution in dry dimethylsulfoxide. Cells were subsequently washed in cold PBS, alkylated with 20 mM NEM as described above, and extracted in immunoprecipitation buffer [25 mM Tris/HCl (pH 7.4), 150 mM NaCl, 0.5% Triton X-100] supplemented with 50 mM glycine, 20 mM NEM, and protease inhibitors. Proteins were immunoprecipitated from the cleared lysates using anti-HA (clone HA-7; Sigma) or anti-c-Myc (clone 9E10) antibodies covalently conjugated to agarose beads. Precipitates were washed five times in immunoprecipitation buffer and captured proteins eluted at room temperature in 4x Laemmli sample buffer.
Deglycosylation and Glycosylation Inhibitor Studies
After denaturation in 0.5% SDS and 40 mM DTT, samples were adjusted to 50 mM sodium citrate (pH 5.5) and incubated with or without 2 U/ml Endo H (New England Biolabs, Beverly, MA) for 16 h at 20 C. For PNGase F treatment under nonreducing conditions, samples were denatured in 0.5% SDS without reducing agent. SDS was quenched with 1% Nonidet P-40, followed by incubation with 7 U/ml PNGase F for 16 h at 20 C. For the in vivo inhibition of Golgi -mannosidase II or N-acetylglucosamine phosphototransferase, cells were changed 6 h after transfection to complete medium containing 10 mg/ml swainsonine or tunicamycin, respectively.
H2O2 Production Assay
Release of H2O2 was routinely determined by reaction with cell-impermeable 10-acetyl-3,7-dihydroxyphenoxazine (Amplex Red reagent; Invitrogen Life Technologies) in the presence of excess peroxidase, producing fluorescent resorufin (44). Briefly, cell monolayers were incubated, with or without 10 µM diphenyleneiodonium, in 1 mM HEPES-buffered Dulbecco’s PBS (D-PBS) (pH 7.4) supplemented with 50 µM Amplex Red reagent and 0.1 U/ml horseradish peroxidase for 1 h at 37 C. Relative fluorescence units (excitation/emission, 535/595 nm) of the medium were measured within the linear range of the H2O2 concentration response curve and corrected for Amplex Red oxidation in wells containing nontransfected cells. Changes in fluorescence intensity were converted into absolute nanomoles of H2O2 using a calibration curve. As internal control for transfection efficiency, Renilla luciferase activity (Dual Luciferase Assay System; Promega) from cotransfected pRL-Tk plasmid (5 ng/well of 12-well plate) was determined in the remaining cells. In some experiments, H2O2 was determined using the fluorogenic substrate homovanillic acid (excitation/emission, 315/425) in Krebs-Ringer HEPES medium as described in detail previously (13), and results normalized for the total protein content in the corresponding cell lysates.
Immunofluorescence Studies
For surface staining of HA-tagged DUOX2, transfected cells grown on glass coverslips were incubated with rat anti-HA clone 3F10 at 1 µg/ml in Hanks’ buffered saline solution/10 mM HEPES (pH 7.4), 1% BSA at 4 C for 30 min. Cells were then washed in D-PBS and fixed in 4% paraformaldehyde/D-PBS. Nonspecific binding sites were blocked with 3% BSA/5% goat serum/D-PBS. Affinity-purified Alexa Fluor 568 goat anti-rat IgG (Molecular Probes, Eugene, OR) was used as secondary antibody. After staining with Hoechst 33342 fluorochrome, the slides were mounted with Prolong Gold antifade (Molecular Probes). For staining of intracellular antigens, fixed cells were permeabilized for 5 min in 0.2% Triton X-100/D-PBS, washed in D-PBS, and blocked as described above. Rabbit anti-CANX (StressGen) and anti-golgin-97 (clone CDF4; Invitrogen Life Technologies) were used as primary antibodies in colocalization experiments, and revealed with appropriate Alexa Fluor 488-labeled secondary antibodies. Epifluorescence and confocal laser-scanning microscopy images were captured on a Nikon Eclipse E800 microscope equipped with PCM-2000 (Nikon, New York, NY). For quantitation of surface expression by epifluorescence, nonsaturated images were captured at equal exposure time for randomly chosen fields. Background-subtracted cell surface fluorescence in the red channel was determined using ImageJ and normalized for the green fluorescence of coexpressed DUOXA2-EGFP.
Flow Immunocytofluorometry
Cells were detached from the plates with EDTA/EGTA (5 mM each). For anti-HA surface immunofluorescence staining, they were incubated sequentially with anti-HA (clone 3F10) and fluorescein-conjugated anti-rat IgG diluted in D-PBS/0.1% BSA. Propidium iodide (5 µg/ml) was included in the second incubation step to exclude damaged cells during the cytometric analysis. For detection of intracellular HA-DUOX2 or DUOXA2-myc, detached cells were fixed in 1% paraformaldehyde/D-PBS for 10 min at 4 C, washed in D-PBS and permeabilized for 30 min with 0.2% saponin in D-PBS/0.1% BSA. Binding of antibodies was done as above, but with 0.2% saponin in the incubation buffers. Fluorescence was assayed after a final washing step in D-PBS using a FACScan Flow Cytofluorometer (BD Biosciences, Mountain View, CA) counting 20,000 events per sample. Relative protein expression was determined by calculating differences in total fluorescence between the sample and an equal-sized population of empty pcDNA3.1 transfected cells.
|
FOOTNOTES |
|---|
The authors have nothing to disclose.
First Published Online March 20, 2007
Abbreviations: CANX, Calnexin; CHX, cycloheximide; D-PBS, Dulbecco’s PBS; DTT, dithiothreitol; DUOX, dual oxidase; Endo H, endoglycosidase H; ER, endoplasmic reticulum; HA, hemagglutinin; NADPH, reduced NAD phosphate; NEM, N-ethylmaleimide; NOX, NADPH oxidase; PNGase F, peptide N-glycosidase F; SDS, sodium dodecyl sulfate; WT, wild type.
Received for publication January 11, 2007. Accepted for publication March 15, 2007.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
opioid receptor. J Biol Chem 275:13727–13736
F508 CFTR pool in the endoplasmic reticulum is increased by calnexin overexpression. Mol Biol Cell 15:563–574This article has been cited by other articles:
 |
|
 |
|
  M. Tonacchera, G. De Marco, P. Agretti, L. Montanelli, C. Di Cosmo, A. C. Freitas Ferreira, A. Dimida, E. Ferrarini, H. E. Ramos, C. Ceccarelli, et al. Identification and Functional Studies of Two New Dual-Oxidase 2 (DUOX2) Mutations in a Child with Congenital Hypothyroidism and a Eutopic Normal-Size Thyroid Gland J. Clin. Endocrinol. Metab., November 1, 2009; 94(11): 4309 - 4314. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Luxen, D. Noack, M. Frausto, S. Davanture, B. E. Torbett, and U. G. Knaus Heterodimerization controls localization of Duox-DuoxA NADPH oxidases in airway cells J. Cell Sci., April 15, 2009; 122(8): 1238 - 1247. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Morand, T. Ueyama, S. Tsujibe, N. Saito, A. Korzeniowska, and T. L. Leto Duox maturation factors form cell surface complexes with Duox affecting the specificity of reactive oxygen species generation FASEB J, April 1, 2009; 23(4): 1205 - 1218. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Rigutto, C. Hoste, H. Grasberger, M. Milenkovic, D. Communi, J. E. Dumont, B. Corvilain, F. Miot, and X. De Deken Activation of Dual Oxidases Duox1 and Duox2: DIFFERENTIAL REGULATION MEDIATED BY cAMP-DEPENDENT PROTEIN KINASE AND PROTEIN KINASE C-DEPENDENT PHOSPHORYLATION J. Biol. Chem., March 13, 2009; 284(11): 6725 - 6734. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
I. Zamproni, H. Grasberger, F. Cortinovis, M. C. Vigone, G. Chiumello, S. Mora, K. Onigata, L. Fugazzola, S. Refetoff, L. Persani, et al. Biallelic Inactivation of the Dual Oxidase Maturation Factor 2 (DUOXA2) Gene as a Novel Cause of Congenital Hypothyroidism J. Clin. Endocrinol. Metab., February 1, 2008; 93(2): 605 - 610. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
