HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 275, Issue 30, 23227-23233, July 28, 2000

This Article Abstract Full Text (PDF) All Versions of this Article: 275/30/23227    most recent M000916200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by De Deken, X. Articles by Miot, F. Search for Related Content PubMed PubMed Citation Articles by De Deken, X. Articles by Miot, F. Social Bookmarking                      What's this?

Cloning of Two Human Thyroid cDNAs Encoding New Members of the NADPH Oxidase Family^*

Xavier De Deken^§, Dantong Wang, Marie-Christine Many^¶, Sabine Costagliola, Frédérick Libert, Gilbert Vassart^**, Jacques E. Dumont, and Françoise Miot

From the  Institut de Recherche Interdisciplinaire and ^** Service de Génétique Médicale, Hôpital Erasme, Université Libre De Bruxelles, Campus Erasme, 808, route de Lennik, 1070 Bruxelles and ^¶ Laboratoire d'Histologie, Université Catholique de Louvain, 1200 Bruxelles, Belgium

Received for publication, February 4, 2000, and in revised form, March 21, 2000

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Two cDNAs encoding NADPH oxidases and constituting the thyroid H₂O₂ generating system have been cloned. The strategy of cloning was based on the functional similarities between H₂O₂ generation in leukocytes and the thyroid, according to the hypothesis that one of the components of the thyroid system would belong to the gp91^Phox/Mox1 gene family and display sequence similarities with gp91^Phox. Screening at low stringency with a gp91^Phox probe of cDNA libraries from thyroid cells in primary culture yielded two distinct human cDNA clones harboring open reading frames of 1551 (ThOX1) and 1548 amino acids (ThOX2), respectively. The encoded polypeptides display 83% sequence similarity and are clearly related to gp91^Phox (53 and 47% similarity). The theoretical molecular mass of 177 kDa is close to the apparent molecular mass of 180 kDa of the native corresponding porcine flavoprotein and the protein(s) detected by Western blot in dog and human thyroid. ThOX1 and ThOX2 display sequence similarities of 53% and 61%, respectively, with a predicted protein of Caenorhabditis elegans over their entire length. They show along their first 500 amino acids a similarity of 43% with thyroperoxidase. The corresponding genes of ThOX1 and ThOX2 are closely linked on chromosome 15q15.3. The dog mRNA expression is thyroid-specific and up-regulated by agents activating the cAMP pathway as is the synthesis of the polypeptides they are coding for. In human thyroid the positive regulation by cAMP is less pronounced. The proteins ThOX1 and ThOX2 accumulate at the apical membrane of thyrocytes and are co-localized with thyroperoxidase.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Iodine enters the thyroid as iodide through an active transport process mediated by the Na⁺/I symporter NIS (1-3). Iodide is oxidized to a higher oxidation state before it acts as an effective agent capable of iodinating thyroglobulin, the precursor of thyroid hormones (4). Both the iodination of thyroglobulin and the oxidative coupling of the resulting iodotyrosines into iodothyronines take place in the follicular space in close contact with the apical pole of thyrocytes. According to current knowledge three different systems are co-localized within the apical plasma membrane and cooperate to the hormonogenic reaction: an iodide transporter, recently identified as the product of the pendrin gene (5, 6), thyroperoxidase (7, 8), and an H₂O₂ generating system. H₂O₂ is the limiting factor in the oxidation of iodide in the synthesis of thyroid hormones reaction (9).

H₂O₂ is formed from the oxidation of NADPH by an NADPH oxidase, a flavoprotein that has been partially purified from solubilized pig thyroid plasma membrane (10). It is known that the enzyme requires Ca²⁺ to be fully active, but the precise molecular structure of the H₂O₂ generating system operating in the thyroid remains ill defined. In polymorphonuclear neutrophils, an NADPH oxidase activity, with functional similarities to the thyroid system, is responsible for the burst of oxygen consumption ("respiratory burst"). It catalyzes the production of superoxide anions (O₂) in activated leukocytes. In these cells the molecular species implicated have been well defined; they comprise a glycoprotein, gp91^Phox, and a smaller 22-kDa protein, p22^Phox located at the membrane and at least three cytosolic components (p40^Phox, p47^Phox, and p67^Phox) migrating to the membrane after stimulation of the cell (11). In nonphagocytic mammal cells, Mox1 has been identified as another superoxide generating NADPH oxidase belonging to the same family as gp91^Phox. It has been implicated in mitogenesis and cancer (12). More recently NOH-1, generated from Mox1 transcripts by alternative splicing, has been described as a mammalian H⁺ channel (13).

From the functional similarities between H₂O₂ generation in leukocytes and the thyroid, we based a cloning strategy on the hypothesis that one of the components of the thyroid system would belong to the gp91^Phox/Mox1 gene family and, accordingly, display sequence similarities with gp91^Phox. Screening of cDNA libraries from thyroid cells in primary culture, at low stringency with a gp91^Phox probe, yielded two distinct cDNA clones harboring open reading frames of 1551 (ThOX1) and 1548 amino acids (ThOX2), respectively. The encoded polypeptides display 83% sequence similarity and are clearly related to gp91^Phox (53 and 47% similarity). The two corresponding genes are closely linked on chromosome 15q15.3. Their transcripts are thyroid-specific, subjected to positive regulation by cAMP agonists, and encode polypeptides accumulating at the apical membrane of thyrocytes.

During the completion of the present study, Dupuy et al. (14), using a strategy based on the purification of the enzymatic activity, reported cloning of a single cDNA encoding a 1210-residue thyroid NADPH oxidase. Comparison with the present results demonstrates that their p138^Tox corresponds to a carboxyl fragment of ThOX2, lacking the first 338 residues.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Screening of Thyroid cDNA Libraries-- A  ZapII cDNA library was constructed as described (15) with poly(A)⁺ RNA prepared from human thyroid cells maintained in primary culture for 5 days with TSH.¹ 18 × 10⁶ recombinant phages were screened at low stringency (hybridization at 42 °C: 30% formamide, 6× SSC (1× SSC is 0.15 M NaCl, 0.015 M sodium citrate), 5 mM EDTA, pH 8, 0.5% SDS, 0.25% milk powder; washings at 50 °C: 2× SSC, 0.1% SDS) with a 1.3-kilobase PstI-EcoRI fragment of gp91^Phox cDNA (a kind gift of Prof. S.Orkin). Two different 1.1-kilobase clones were obtained. Segments corresponding to their 5' portions were used to screen 4 × 10⁶ and 3 × 10⁶ clones of  gt11 cDNA libraries made from human (16) and dog (17) thyroids, respectively. Segments containing the human counterparts of the canine coding sequences were obtained by polymerase chain reaction (ThOX1: forward 5'-GACGGAATTCATATTCATCATGGGCTTCTGC, reverse 5'-ACTACTCGAGCTGGAGAAACTTGAGTTCCGA; ThOX2: forward 5'-GTGCAGCGATTTGATGGGTGG, reverse 5'-ACTACTCGAGACTCCAGTTCCTTGGGATGTC). The 5' noncoding sequence of each human cDNA was obtained by 5'-rapid amplification of cDNA ends, starting from 1 µg of poly(A)⁺ RNA of human thyrocytes in culture and three specific primers (ThOX1 5'-GGATGCGAATGTTGAGGAACT, 5'-CCACGCTCACCAGGTCTGAAA, 5'-CCCAACACTGTGCGGTTTCTC; ThOX2 5'-GTCCCAGCGGCTCCTCTGGAA, 5'-CGCGCTGGTCGGGGTCGAACA, 5'-GGATGCGGATGTTGAGGAACT) according to the manufacturer's protocol (Life Technologies, Inc.). cDNAs subcloned in pBluescript SK⁺ (pBS) or in pCR2.1 (Invitrogen) were sequenced with the BigDye Terminator cycle sequencing method on an automated ABI Prism 377 DNA Sequencer (Perkin-Elmer Applied Biosystems). Sequence data handling and analysis were performed using the BLAST sequence programs.

Northern Blot-- Total RNA was obtained from different dog tissues or from dog and human thyrocytes in primary culture by the method of Chirgwin et al. (18). 10 µg of total RNA were separated by electrophoresis, transferred to a positively charged nylon membrane (Amersham Pharmacia Biotech), and hybridized as described previously (19). cDNA probes, inserted in pBS, were chosen in the 3' noncoding regions of ThOX1 and ThOX2 to avoid any cross-hybridizations. The dog cDNA probes were a 842-bp EagI-EcoRI fragment of ThOX1 and a 1427-bp KpnI-EcoRI fragment of ThOX2. The human probes were a 687-bp SmaI-EcoRI fragment of ThOX1 and a 1628-bp HincII-XhoI fragment of ThOX2. Filters were washed four times for 30 min in 0.1× SSC, 0.1% SDS at 65 °C.

Immunodetection-- A rabbit polyclonal antibody was raised against the Arg⁶¹⁸-His¹⁰⁴⁴ fragment of ThOX1 produced in Escherichia coli by the pET-24d(+) vector system (Novagen). New Zealand White rabbits (Iffa Credo) were subcutaneously injected three times, at a 3-week interval, with 150 µg of recombinant ThOX1 fragment, in the presence of complete (first injection) or incomplete (subsequent injections) Freund's adjuvant (Sigma). Sera were collected 10 days after the third injection and were directly used at a dilution of 1:10,000 in Western blot analyses and 1:200 in immunohistochemistry experiments.

Western Blot-- Protein extracts were obtained by lysis in classical Laemmli buffer of thyrocytes or Cos-7 cells transfected with ThOX1 cDNA in pcDNA3 and Fugene (Roche Molecular Biochemicals). Proteins were separated by SDS/polyacrylamide gel electrophoresis on 6 or 7.5% polyacrylamide gels. 10-30 µg of protein extracts, measured by a paper dye binding assay (20), were loaded, and the resolved proteins were transferred to nitrocellulose. Immune complexes were detected with a horseradish peroxidase-coupled anti-rabbit IgG antibody (Amersham Pharmacia Biotech) according to the ECL method (NEN Life Science Products).

Protein Localization by Immunochemistry-- Human paranodular tissue was obtained at surgery and quickly frozen in isopentane cooled in liquid nitrogen. After inhibition of endogenous peroxidase activity by addition of 0.228% periodic acid for 45 s, the frozen sections were incubated for 1 h with the anti-ThOX1 antibody diluted at 1:200 and for 1 h with a second anti-rabbit immunoperoxidase-conjugated antibody. Antibody binding was revealed using diaminobenzidine tetrachloride. Controls were performed by using preimmune serum instead of the first antibody, by omission of the first antibody, or by omission of the first and second antibodies.

Thyroperoxidase was immunolocalized using the N-vision technique. The first mouse monoclonal antibody (gift from Dr De Mico) was added on the frozen sections for 1 h at dilution 1:500.

The secondary antibody is a goat anti-mouse immunoglobuline conjugated to peroxidase-labeled polymer (Dako En Vision +). The chromogen used for revelation was AEC giving a red staining (3-amino-9 ethylcarbazote).

Thyroid Cell Culture-- Follicles were isolated by collagenase digestion and differential centrifugation from fresh dog and human thyroid as described previously (21, 22). The follicles were seeded in 60- or 90-mm tissue culture-treated Petri dishes and cultured for 4 days in control medium. This procedure yields cultures containing thyrocytes pure at 99% (23). The cells were stimulated for different times with 10 µM forskolin (Calbiochem), 1 milliunit/ml TSH (Sigma), or 25 ng/ml EGF (Sigma).

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Molecular Cloning of Two cDNA Encoding Proteins of the Flavoprotein Family-- Screening by hybridization at low stringency of two human thyrocyte cDNA libraries with a gp91^Phox cDNA segment yielded systematically two types of clones displaying significant similarity with the probe. When sequenced and aligned, they contained an uninterrupted reading frame corresponding to the 3' region of gp91^Phox, and none of them presented an initiation codon in a favorable consensus. After several unsuccessful attempts to isolate full-length cDNAs from human libraries, a dog  gt11 cDNA library known to contain large inserts was screened with the most 5' portions of each type of clones. This yielded canine homologs of the two categories of human clones which, when aligned with gp91^Phox, displayed open reading frames extending upstream of gp91^Phox initiation codon. Reverse transcription-polymerase chain reaction with a human-specific reverse primer and a canine forward primer (see "Experimental Procedures") allowed extension of the 5' portions of the two human cDNA clones to a total length of 5,369 and 6,126 bp, respectively. Finally 5'-rapid amplification of cDNA ends polymerase chain reaction with primers specific of each sequence yielded two presumably complete cDNA sequences: the first was 5,693 bp long (GenBank^TM accession number AF230495) with the potential to encode a protein of 1551 amino acids (ThOX1 for thyroid oxidase 1); the second was 6,410 bp long (GenBank^TM accession number AF230496) and contained an open reading frame of 1548 amino acids (ThOX2).

The primary structures of ThOX1 and ThOX2 showed 83% similarity (Fig. 1A). The molecular weight of the predicted unglycosylated proteins was 177,000. According to von Heijne's algorithm (24), both contained a putative signal peptide with Asn²³ and Gln²⁶ being the preferred candidates for the N terminus of the mature proteins in ThOX1 and in ThOX2, respectively. Both proteins contained two putative EF-hand motifs, ⁸²⁸DKDGNGYLSFREF⁸⁴⁰ and ⁸⁶⁴DFDGNGLISKDEF⁸⁷⁶ in ThOX1 and ⁸³²DKDGNGYLSFREF⁸⁴⁴ and ⁸⁶⁸DLDENGFLSKDEF⁸⁸⁰ in ThOX2. The sequence of the 1210 last amino acids of ThOX2 was identical to that of p138^Tox, a recently reported cDNA claimed to encode a component of thyroid H₂O₂ generating system (14). When compared with gp91^Phox and Mox1, ThOX1 and ThOX2 displayed 53 and 47% similarity with the former and 54 and 47% identity with the latter, respectively. The similarity was, however, limited to their last 569 C-terminal residues because both of them contained an N-terminal extension with no counterpart in gp91^Phox (or Mox1) (982 and 979 residues in ThOX1 and ThOX2, respectively). The homology with gp91^Phox allowed, in both ThOX1 and ThOX2, identification of a conserved putative NADPH-binding site (¹³⁸⁹GIGVTPF¹³⁹⁵, ¹⁴²⁰IWVTR¹⁴²⁴, ¹⁴⁸⁶GLRSITHFGR¹⁴⁹⁵, and ¹⁵¹⁷VFSCGP¹⁵²² in ThOX1 and ¹³⁸⁶GIGVTPF¹³⁹², ¹⁴¹⁷IWVTR¹⁴²¹, ¹⁴⁸³GLRSITHFGR¹⁴⁹², and ¹⁵¹⁴VFSCGP¹⁵¹⁹ in ThOX2), a putative FAD-binding site ¹³¹⁹HPFTLTS¹³²⁵ in ThOX1 and ¹³¹⁶HPFTLTS¹³²² in ThOX2), and the four histidines (His¹¹³⁰, His¹¹⁴⁴, His¹²²⁵, and His¹²³⁸ in ThOX1 and His¹¹²⁷, His¹¹⁴¹, His¹²²², and His¹²³⁵ in ThOX2) and an arginine (Arg¹⁰⁸⁷ in ThOX1 and Arg¹⁰⁸⁴ in ThOX2) considered to be implicated in the binding of the heme moiety (25). The N-terminal region harbors five putative sites of N-glycosylation (Asn⁹⁴, Asn³⁴², Asn³⁵⁴, Asn⁴⁶¹, and Asn⁵³⁴ in ThOX1 and Asn¹⁰⁰, Asn³⁴⁸, Asn³⁸², Asn⁴⁵⁵, and Asn⁵³⁷ in ThOX2).

View larger version (63K): [in this window] [in a new window]   Fig. 1.   Primary structure of ThOX1 and ThOX2 and sequence comparison. A, deduced amino acid of ThOX1 (GenBankTM accession number AF230495) and ThOX2 (GenBankTM accession number AF230496) aligned with C. elegans predicted protein (GenBankTM accession number AF043697). Shaded boxes indicate identical residues, and + denotes the position of the four histidines and arginine conserved in gp91Phox. Putative FAD-binding and NADPH-binding domains and EF-hand regions are indicated with dotted lines. The seven putative transmembrane regions are shown with asterisks. B, presumed structure of ThOX protein; model of the transmembrane topology and putative functional domains.

The sequence organization of ThOX1 and ThOX2 is shared by a predicted protein of Caenorhabditis elegans (GenBank^TM accession number AF043697) found by sequencing of chromosome III of C. elegans (26), with which they display respectively 53 and 61% similarity over their entire length. Surprisingly, stretches of amino acids displaying significant sequence similarity with proteins belonging to the peroxidase gene family were found both in the C. elegans protein and in ThOX1,2 in the N-terminal extensions (43% similarity with thyroperoxidase over 500 residues). The hydropathy profiles (TMpred program (27)) of ThOX1 and ThOX2 include seven hydrophobic stretches that could be membrane spanning regions. Considering that Ca²⁺ binding to EF-hands motifs as well as NADPH and FAD binding presumably occur in the intracellular portions of the protein, we propose a common structure for ThOX1 and ThOX2 shown in Fig. 1B.

The genes corresponding to ThOX1 and ThOX2 have been co-localized on chromosome 15q15.3 using the radiation hybrid panel method of Genebridge 4 (Research Genetics). Comparison of their nucleotide sequence with the current human genomic data base (high throughput genomic sequences, NCBI, National Institutes of Health) revealed stretches of identity corresponding to exons in the partially sequenced 163_P_10 clone on chromosome 15 (AC 009700.3, release of October, 1999). This allowed positioning of at least 20 intron-exon junctions in each gene and confirmed their location in close proximity on chromosome 15.

Distribution and Regulation of mRNA Accumulation-- The presence of ThOX1 and ThOX2 mRNA has been explored in 12 different dog tissues by Northern blotting (liver, lymph node, heart, lung, brain, testis, kidney, cerebellum, pancreas, stomach, spleen, and thyroid). Except for a very weak ThOX2 signal in the stomach, it was exclusively found in the thyroid (Fig. 2).

View larger version (81K): [in this window] [in a new window]   Fig. 2.   ThOX1 and ThOX2 mRNA expression in dog tissues. Total RNA was extracted from different dog tissues. Lane 1, liver; lane 2, lymph node; lane 3, heart; lane 4, lung; lane 5, brain; lane 6, testis; lane 7, kidney; lane 8, cerebellum; lane 9, pancreas; lane 10, stomach; lane 11, spleen; lanes 12-14, thyroid. 10 µg of total RNA were subjected to Northern blot analysis and hybridized with a cDNA probe of ThOX1 or ThOX2. The bottom panel shows acridine orange staining of ribosomial RNA (rRNA).

Regulation of ThOX1 and ThOX2 gene expression was studied in dog and human thyroid cells in primary culture after stimulation for up to 72 h with TSH (1 milliunit/ml), forskolin (10⁵ M), or EGF (25 ng/ml). In dog and human thyrocytes, Northern blotting revealed transcripts estimated at 5.7 and 6.4 kilobases when hybridizations were performed with the 3' region of ThOX1- and ThOX2-specific probes, respectively. 10⁵ M forskolin stimulated accumulation of both ThOX1 and ThOX2 mRNAs of dog thyrocytes after 48 h with a further increase after 72 h (Fig. 3B). This stimulation was reproduced with TSH (Fig. 3C). EGF had no effect on the mRNA expression. In human thyrocytes the up-regulation by forskolin was less pronounced (Fig. 3A). The same signal was observed when the blots were hybridized with the 5'-extremity of the ThOX1 (700 bp) and ThOX2 (1200 bp) cDNAs (not shown).

View larger version (75K): [in this window] [in a new window]   Fig. 3.   ThOX1 and ThOX2 mRNA expression after chronic stimulation with forskolin (105 M), TSH (1 milliunits/ml), or EGF (25 ng/ml). Total RNA was extracted, and 10 µg were subjected to Northern blot analysis. Panel a shows the mRNA expression in the human thyrocytes cultured for 24 or 48 h with forskolin. mRNA expression is measured in dog thyrocytes cultured for 8, 24, 48, or 72 h with forskolin or EGF (panel b) and for 72 h with TSH (panel c). The bottom panels of the figure show acridine orange staining ascertaining that equal amount of RNA was spotted in each lane (45). C, control; Fsk, forskolin; kb, kilobases.

Localization and Regulation of ThOX1 Protein-- The polyclonal antibody raised against a fragment of ThOX1 made it possible to localize ThOX1 in the supranuclear apical pole of all thyroid cells. In some cells the labeling was more pronounced on the apical plasma membrane (Fig. 4A). No labeling was obtained when using preimmune serum instead of anti-ThOX1 (Fig. 4B). Localization of ThOX1 protein is similar to that of thyroperoxidase also detected in the apical pole of the thyroid cells (Fig. 4C).

View larger version (72K): [in this window] [in a new window]   Fig. 4.   Immunodetection of human ThOX and TPO. Frozen sections of human paranodular thyroid tissue (× 300). A, immunostaining of ThOX1; the protein is localized in the apical pole of all thyroid cells or on the apical plasma membrane. B, immunostaining with preimmune serum. There is no labeling. C, Immunostaining of thyroperoxidase. Thyroperoxidase is also localized in the apical poles of the cells.

Western blot analysis of lysates of Cos-7 cell transfected with an expression plasmid coding for human ThOX1 detected a major immunoreactive band corresponding to a protein with an apparent molecular mass of 180 kDa (Fig. 5). This mass was close to the theoretical value calculated from the predicted amino acid sequence, the difference being most probably accounted for by N-linked carbohydrate chains. The antibody recognized a protein of the same molecular mass in three rat thyroid cell lines: FRTL-5, WRT, and PCCL3. The protein was also detected in dog and human thyrocytes. Western blots made with thyroid tissues revealed two bands corresponding to proteins with molecular masses of 180 and 190 kDa (Fig. 5). Prior incubation of the antibody with 10 µg of the recombinant ThOX1 fragment used to immunize the rabbit suppressed completely and specifically the bands. Stimulation of dog thyrocytes in culture by forskolin for 48 or 72 h resulted in a steady increase of the accumulation of the immunoreactive material. EGF had no effect even at 72 h. In human thyrocytes a weak stimulation by forskolin was observed at 72 h.

View larger version (70K): [in this window] [in a new window]   Fig. 5.   Immunoblotting analysis of ThOX protein. Cell lysates of human or dog thyrocytes stimulated with forskolin (105 M) or EGF (25 ng/ml) for 48 h and 72 h, of COS-7 cells transfected with ThOX1 cDNA or pcDNA3 vector alone, and of rat thyroid cell lines were run on SDS/polyacrylamide gel electrophoresis and immunodetected with a polyclonal antibody raised against a fragment of ThOX1. C, control; Fsk, forskolin; Vect., pcDNA3 vector; Hum., human; Thyr., thyrocytes.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Using a strategy based on the functional homology existing between the NADPH oxidase systems in the thyroid and the polymorphonuclear neutrophil, we cloned from cDNA libraries of thyroid cells in primary culture, two different cDNAs displaying convincing sequence similarity with gp91^Phox over their entire C-terminal portion (Fig. 1A). It should be noted that previous attempts using cDNA libraries made from whole thyroid tissue has only yielded the cDNA of gp91^Phox. The available arguments to identify the two candidates as components of the thyroid H₂O₂ generating system include: 1) the level of sequence similarity with gp91^Phox (53 and 54% for ThOX1 and ThOX2, respectively) which, apart from indicating a common ancestor (see below), is strongly suggestive of a similar function; 2) the almost exclusive thyroid specificity of the two transcripts (Fig. 2) and the co-localization of ThOX1,2 polypeptides with TPO at the apical pole of thyrocytes (Fig. 4); and 3) the perfect sequence identity of the last 1210 residues of ThOX2 with p138^Tox, a thyroid NADPH oxidase reported recently by Dupuy et al. (14) as the catalytic moiety of the thyroid H₂O₂ generating system. Because this clone was obtained from the sequence of endoproteinase Lys-C peptides generated from a 180-kDa protein displaying NADPH oxidase activity (10), the most likely explanation is that the p138^Tox clone of Dupuy et al. (14) is an incomplete version of ThOX2 lacking 1014 bases of coding sequence at its 5' end. In contrast, ThOX2 would encode a full-size 180-kDa NADPH oxidase polypeptide (see Western blot on Fig. 5). The 83% sequence similarity between ThOX2 and ThOX1 strongly suggests that the latter would also encode a thyroid NADPH oxidase. The presence in the protein sequence of conserved FAD-, NADPH-, and heme-binding sites of the gp91^Phox fits with the known biochemistry and physiology of the thyroid (28). Finally the regulation of mRNA expression of ThOX1,2 in dog thyroid cells in culture is in total agreement with the previous findings on the functional generation of H₂O₂ by these cells (29). Despite these convincing arguments, the definite proof that ThOX1 and ThOX2 are components of the thyroid H₂O₂ generating system will await reconstitution of H₂O₂ production in transfected nonthyroid cells. This may require co-transfection of constructs encoding additional components that remain to be identified (10).

In their present state, sequence data bases contain five sequences displaying similarity with NADPH oxidases: gp91^Phox, ThOX1, ThOX2, Mox1, and a conceptual protein of C. elegans. Together, they constitute a subfamily of genes encoding flavoproteins. These flavoproteins contain NADPH- and FAD-binding domains, four specific histidines, and a conserved arginine involved in the binding of the heme prosthetic group. gp91^Phox possesses a NADPH oxidase activity, which catalyzes the production of superoxide anions (O₂) by the one-electron reduction of oxygen using NADPH as the electron donor. The cloning of members of the flavoprotein family, restricted to gp91^Phox two years ago, has been extended today to new members that are expressed in nonphagocytic cells; Mox1 presumably involved in cell proliferation and expressed in colon and vascular smooth muscle cells (12) and NOH, a mammalian H⁺ channel that is in fact a spliced form of Mox1 (13).

Alignment of the various members of the protein family allows to classify them in two categories on the basis of the presence of two different building blocks. All five contain amino acid residue homologous segments, the NADPH- and FAD-binding sites, with the potential to encode an NADPH oxidase. Only the C. elegans protein and both ThOX1 and ThOX2 contain, in addition, a 500-residue N-terminal extension with sequence similarity with peroxidases. The structure of the NADPH oxidase segment has been modelled in gp91^Phox (25), which allows proposal of a possible structure for the corresponding portion of the ThOX proteins. The presumed structure of this portion of the proteins (Fig. 1B) shows a large first intracellular loop containing two EF-hand domains probably involved in the direct activation of the enzyme by Ca²⁺ as demonstrated in porcine and human thyroid membrane fractions (30, 31), in intact follicles (32), and in thyroid slices (9, 33). The four histidines and the arginine involved in the heme binding in the gp91^Phox (25) are conserved in ThOX1 and ThOX2 as the FAD- and NADPH-binding sites. These sites are the presumed functional domains of the thyroid NADPH oxidase and should be located at the intracellular side of the apical pole of the thyrocyte.

The N-terminal portions of the ThOX polypeptides and the C. elegans predicted protein display homology with peroxidases and harbor three conserved potential sites for N-glycosylation. They are expected to be extracellular. Indeed, the EF-hand containing segment of the NADPH oxidase domain, in ThOX1 and ThOX2, is separated from the peroxidase homology domain by a hydrophobic segment presenting the characteristics of a transmembrane helix. This structural difference between the ThOX proteins and gp91^Phox could perhaps be related for an expected functional difference; the transfer of electrons from NADPH to oxygen results in the neutrophil in the formation of superoxide anions, whereas in the thyroid, H₂O₂ is formed directly (34).

In evolutionary terms, the presence of both the NADPH oxidase and the peroxidase homology domains in the ThOX and the C. elegans proteins indicates that a gene fusion event, at the origin of these proteins, predates the separation between vertebrates and invertebrates. The absence of the peroxidase domain in gp91^Phox is compatible with two scenarios; either this domain has been lost after the gene duplication event having generated the ThOX genes in the vertebrate lineage or the gp91^Phox gene evolved from an ancestral gene devoid of the peroxidase domain, which would have been lost in C. elegans. The relatively high sequence similarity between the ThOX genes in their NADPH oxidase domains and gp91^Phox strongly favors the first scenario. The co-localization of the two ThOX genes on chromosome 15q15.3, as determined by the radiation hybrid method, was confirmed by current information in the genomic data base; the two genes are very closely linked, perhaps in tandem, within a 153,400-bp bacterial artificial chromosome in the process of being fully sequenced.

Evolution of genes by fusion of segments encoding functional entities, which may remain separate in other genes, has been described (35, 36). The organification of iodide and thyroid hormonogenesis make use both of ThOX proteins, containing an NADPH oxidase and a peroxidase homology domain, and thyroperoxidase. Considering the toxicity of H₂O₂, association of two enzymes in a complex allowing channeling of H₂O₂ would make physiological sense (37-39). Future dissection of structure-function relationships of the ThOX proteins will be needed to identify the respective roles and possible interactions of thyroperoxidase and the N-terminal peroxidase homology domain of ThOX1 and ThOX2.

The existence of the two distinct ThOX genes raises the question of their respective roles. Both are expressed in a thyroid-specific manner and controlled in a similar way by TSH. Among other possibilities, they could constitute redundant entities or be implicated in a heteropolymeric structure, with each subunit playing a distinct role in the catalytic reaction. In the first case, with four copies of ThOX per diploid genome, the system should be particularly resistant to alteration by loss-of-function mutations. On the contrary, if a heteropolymeric structure is involved, depending on the stoechiometry of the complex, one could observe dominant negative effects in heterozygotes with mutations affecting the structure of the holoprotein. The rarity of cases with congenital defects of organification without mutations of the thyroperoxidase gene (40, 41) would rather argue in favor of the first hypothesis, but here also, clarification will require activity reconstitution.

Regulation of H₂O₂ generation is known to occur both at the gene expression and posttranscriptional levels. At the former level the results of Northern and Western blotting experiments with dog thyrocytes agree with biochemical data showing that the capacity of H₂O₂ formation is stimulated after a chronic elevation of cAMP (29). Although Northern blotting experiments clearly indicate that both ThOX1 and ThOX2 mRNAs are up-regulated by cAMP agonists, in the absence of antibodies specific for the individual proteins, it can only be concluded from Western blots that one or both proteins accumulate under the same conditions. In human but not in dog thyroid, TSH activates the phosphatidylinositol triphosphate-phospholipase C cascade. Acute stimulation of iodide organification and H₂O₂ production by the PIP2-phospholipase C cascade and by calcium has been observed in humans, dogs, and other species (9, 33, 42). The presence of the two EF-hands in the ThOX proteins provides a structural basis for these effects.

The successful cloning of human H₂O₂ generating system opens new perspectives in the physiopathology and diagnosis of thyroid diseases. It should allow to demonstrate the genetic lesion of this system in iodination defects observed in congenital hypothyroidism, thyroid cold adenomas, and thyroid cancer with normal thyroperoxidase (43, 44). The close proximity of thyroperoxidase and ThOX at the apical pole of thyrocytes suggests that autoimmune reactions leading to the generation of autoantibodies directed against the former might spread to the latter in Hashimoto's thyroiditis or Graves' diseases. By analogy with anti-thyroperoxidase autoantibodies, it is conceivable that some antibodies will have an inhibitory effect on the catalytic activity.

	ACKNOWLEDGEMENTS

We thank the technician team of the Service de Génétique moléculaire for expert manipulations of the sequencers, Christiane Christophe for providing us the oligonucleotide primers, and Marie-Jeanne Simons for excellent technical advice.

	FOOTNOTES

^* This work was supported by the Ministère de la Politique Scientifique (Pôle d'Attraction Interuniversitaire), Fonds National de la Recherche Scientifique, Fonds National de la Recherche Scientifique Médicale, Association contre le Cancer, and Association Sportive contre le Cancer.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

^§ Fellow of the Fonds pour la Formation à la Recherche dans l'Industrie et l'Agriculture (FRIA).

Qualified Researcher of the Belgian Fonds National de la Recherche Scientifique (FNRS).

To whom correspondence should be addressed: IRIBHN, Université Libre de Bruxelles, Campus Erasme, Bat. C.; 808, route de Lennik, B-1070 Bruxelles Belgium. Fax: 32-2-5554655; E-mail: fmiot@ulb.ac.be.

Published, JBC Papers in Press, May 9, 2000, DOI 10.1074/jbc.M000916200

	ABBREVIATIONS

The abbreviations used are: TSH, thyrotropin; EGF, epidermal growth factor; bp, base pair(s).

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				H. J. Kim, Y.-D. Park, U. Y. Moon, J.-H. Kim, J. H. Jeon, J.-G. Lee, Y. S. Bae, and J.-H. Yoon The Role of Nox4 in Oxidative Stress-Induced MUC5AC Overexpression in Human Airway Epithelial Cells Am. J. Respir. Cell Mol. Biol., November 1, 2008; 39(5): 598 - 609. [Abstract] [Full Text] [PDF]

				Y. Maruo, H. Takahashi, I. Soeda, N. Nishikura, K. Matsui, Y. Ota, Y. Mimura, A. Mori, H. Sato, and Y. Takeuchi Transient Congenital Hypothyroidism Caused by Biallelic Mutations of the Dual Oxidase 2 Gene in Japanese Patients Detected by a Neonatal Screening Program J. Clin. Endocrinol. Metab., November 1, 2008; 93(11): 4261 - 4267. [Abstract] [Full Text] [PDF]

				P-R. Burgel and J. A. Nadel Epidermal growth factor receptor-mediated innate immune responses and their roles in airway diseases Eur. Respir. J., October 1, 2008; 32(4): 1068 - 1081. [Abstract] [Full Text] [PDF]

				B. Rada, K. Lekstrom, S. Damian, C. Dupuy, and T. L. Leto The Pseudomonas Toxin Pyocyanin Inhibits the Dual Oxidase-Based Antimicrobial System as It Imposes Oxidative Stress on Airway Epithelial Cells J. Immunol., October 1, 2008; 181(7): 4883 - 4893. [Abstract] [Full Text] [PDF]

				S. Pacquelet, M. Lehmann, S. Luxen, K. Regazzoni, M. Frausto, D. Noack, and U. G. Knaus Inhibitory Action of NoxA1 on Dual Oxidase Activity in Airway Cells J. Biol. Chem., September 5, 2008; 283(36): 24649 - 24658. [Abstract] [Full Text] [PDF]

				C. Meischl, H. P. Buermans, T. Hazes, M. J. Zuidwijk, R. J. P. Musters, C. Boer, A. van Lingen, W. S. Simonides, M. A. Blankenstein, C. Dupuy, et al. H9c2 cardiomyoblasts produce thyroid hormone Am J Physiol Cell Physiol, May 1, 2008; 294(5): C1227 - C1233. [Abstract] [Full Text] [PDF]

				S. Luxen, S. A. Belinsky, and U. G. Knaus Silencing of DUOX NADPH Oxidases by Promoter Hypermethylation in Lung Cancer Cancer Res., February 15, 2008; 68(4): 1037 - 1045. [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]

				Y. Song, N. Driessens, M. Costa, X. De Deken, V. Detours, B. Corvilain, C. Maenhaut, F. Miot, J. Van Sande, M.-C. Many, et al. Roles of Hydrogen Peroxide in Thyroid Physiology and Disease J. Clin. Endocrinol. Metab., October 1, 2007; 92(10): 3764 - 3773. [Abstract] [Full Text] [PDF]

				S. Dremier, M. Milenkovic, S. Blancquaert, J. E. Dumont, S. O. Doskeland, C. Maenhaut, and P. P. Roger Cyclic Adenosine 3',5'-Monophosphate (cAMP)-Dependent Protein Kinases, But Not Exchange Proteins Directly Activated by cAMP (Epac), Mediate Thyrotropin/cAMP-Dependent Regulation of Thyroid Cells Endocrinology, October 1, 2007; 148(10): 4612 - 4622. [Abstract] [Full Text] [PDF]

				K. R. Johnson, C. C. Marden, P. Ward-Bailey, L. H. Gagnon, R. T. Bronson, and L. R. Donahue Congenital Hypothyroidism, Dwarfism, and Hearing Impairment Caused by a Missense Mutation in the Mouse Dual Oxidase 2 Gene, Duox2 Mol. Endocrinol., July 1, 2007; 21(7): 1593 - 1602. [Abstract] [Full Text] [PDF]

				H. Grasberger, X. De Deken, F. Miot, J. Pohlenz, and S. Refetoff Missense Mutations of Dual Oxidase 2 (DUOX2) Implicated in Congenital Hypothyroidism Have Impaired Trafficking in Cells Reconstituted with DUOX2 Maturation Factor Mol. Endocrinol., June 1, 2007; 21(6): 1408 - 1421. [Abstract] [Full Text] [PDF]

				H. S. Li and G. Carayanniotis Induction of Goitrous Hypothyroidism by Dietary Iodide in SJL Mice Endocrinology, June 1, 2007; 148(6): 2747 - 2752. [Abstract] [Full Text] [PDF]

				H. Fischer, L. K. Gonzales, V. Kolla, C. Schwarzer, F. Miot, B. Illek, and P. L. Ballard Developmental regulation of DUOX1 expression and function in human fetal lung epithelial cells Am J Physiol Lung Cell Mol Physiol, June 1, 2007; 292(6): L1506 - L1514. [Abstract] [Full Text] [PDF]

				R. Liu, J. L. Garvin, Y. Ren, P. J. Pagano, and O. A. Carretero Depolarization of the macula densa induces superoxide production via NAD(P)H oxidase Am J Physiol Renal Physiol, June 1, 2007; 292(6): F1867 - F1872. [Abstract] [Full Text] [PDF]

				A. Orient, A. Donko, A. Szabo, T. L. Leto, and M. Geiszt Novel sources of reactive oxygen species in the human body Nephrol. Dial. Transplant., May 1, 2007; 22(5): 1281 - 1288. [Full Text] [PDF]

T. Nakanaga, J. A. Nadel, I. F. Ueki, J. L. Koff, and M. X. G. Shao Regulation of interleukin-8 via an airway epithelial signaling cascade Am J Physiol Lung Cell Mol Physiol, May 1, 2007; 292(5): L1289 - L1296. [Abstract] [Full Text] [PDF]