Peroxidases are enzymes that catalyze the breakdown of hydrogen peroxide to water and reactive oxygen with resultant oxidation of a wide variety of substrates. They have functions related to the regulation of potentially dangerous reactive oxygen species and are involved in several synthetic and degradative cellular functions. There are several families of peroxidases that appear to have different evolutionary origins (Daiyasu and Toh, 2000). The myeloperoxidase family is characterized by the presence of a conserved 500–600 amino acid catalytic domain that contains a heme prosthetic group. Myeloperoxidase family members are found in diverse groups of organisms, including bacteria and all animal groups (Daiyasu and Toh, 2000; Zhao et al., 2001). The prototypical member of this family is myeloperoxidase, which is an abundant protein in phagocytic neutrophils, that produces oxidizing species involved in killing invading microorganisms. Other members of the family include eosinophil peroxidase, lactoperoxidase, and thyroid peroxidase (TPO; Daiyasu and Toh, 2000). TPO is a critical enzyme to the synthesis of thyroid hormone, being required for the oxidation of iodide and its addition to thyroglobulin.
Peroxidasin, which was originally identified in Drosophila, is a member of the myeloperoxidase family and is of note because it has a novel domain structure (Nelson et al., 1994). Peroxidasin is a secreted protein that combines a peroxidase activity with the features of an extracellular matrix protein, containing four Ig loops and a carboxy terminal von Willebrand type C protein–protein interaction domain.
The activity of the Drosophila peroxidasin gene is associated with the function of hemocytes, which are migratory cells present in the hemolymph. Hemocytes are involved in both the intracellular destruction of phagocytosed apoptotic cells and foreign material, as well as the deposition of extracellular matrix. It has been proposed that peroxidasin has an intracellular role in the elimination of engulfed material and an extracellular role in the stabilization of the extracellular matrix, cross-linking component proteins such as collagen and laminin (Nelson et al., 1994).
The human peroxidasin orthologue has been identified in several contexts, as a partial sequence of a p53 regulate gene (PRG2) and as a cDNA coding for a near full-length peptide of 1,496 amino acids from melanoma (MG50) and immature myeloid cell lines (KIAA0230; Weiler et al., 1994; Nagase et al., 1997; Horikoshi et al., 1999). Northern blot and reverse transcriptase-polymerase chain reaction (RT-PCR) analyses have shown that human peroxidasin is widely expressed in adult tissues and cancer cell lines (Horikoshi et al., 1999; Mitchell et al., 2000).
Here, we report the identification of the amphibian orthologue of peroxidasin. We show that the Xenopus tropicalis peroxidasin (Xtpxn) gene exhibits a dynamic pattern of expression during early development. Maternal expression of Xtpxn is localized to the animal hemisphere where it persists through early cleavage stages. Initial zygotic Xtpxn expression is detected in the developing neural tube. Later expression within the nervous system is localized to the midbrain and hindbrain. Xtpxn expression is prominent in the pronephros throughout its development, suggesting a role for this novel secreted enzyme in the morphogenesis and function of the kidney.
RESULTS AND DISCUSSION
Identification of the Xenopus tropicalis peroxidasin gene
A database search was undertaken for Xenopus sequences related to human TPO. An expressed sequence tag (EST, accession no. AL783545) from a neurula stage cDNA library was identified that showed significant sequence identity to TPO but which was more closely related to Drosophila and human peroxidasin. Another related EST (accession no. AL86514) from an egg cDNA library was also identified. We used the sequence derived from the ESTs and sequences available from the Department of Energy, JGI, Xenopus tropicalis genome project to compile a 68,592-bp genomic contig for the sequence of the X. tropicalis peroxidasin (Xtpxn) gene, which codes for a conceptual peptide of 1,456 amino acids. The full-length human and Drosophila pxn peptides consist of 1,506 and 1,527 amino acids, respectively (Nelson et al., 1994). Figure 1 shows an alignment of the Xtpxn peptide sequence with the human and Drosophila orthologues. Analysis of the Xenopus peptide sequence using the SignalP software predicts the presence of a cleavable amino terminal signal peptide, suggesting that Xtpxn is likely to be a secreted protein.
Prosite protein motif identification software indicates the presence of several protein domains in the Xtpxn sequence that are conserved between Drosophila, frog, and human. These domains include four extracellular Ig-like domains, an animal myeloperoxidase domain, and a signature carboxy terminal von Willebrand factor type C domain (see Table 1; Fig. 2A).
Table 1. The Domain Structure of the Xenopus tropicalis Peroxidasin Peptidea
VWFC, von Willebrand factor C.
Ig-like domain 1
Ig-like domain 2
Ig-like domain 3
Ig-like domain 4
myeloperoxidase catalytic domain
The Drosophila pxn gene (accession no. NP995975) contains 10 exons (Nelson et al., 1994). Analysis of the genomic locus of human pxn reveals a gene consisting of 23 exons, coding for a full-length peptide of 1,506 (accession no. XP056455). The Xenopus genomic fragment that we have assembled also contains 23 predicted exons (Fig. 2B,C). An analysis of the exon/intron boundaries of the human locus and the predicted boundaries from the assembled Xenopus genomic contig suggests that genomic structure is highly conserved between the two species (data not shown).
The human and Xenopus peptides show predicted sequence identity of 79% overall and share 43% overall identity with Drosophila pxn. Within their catalytic domains, sequence identity between Xenopus and human rises to 84%. This figure is much higher than the observed sequence identity of 40% between the Xenopus pxn peroxidase domain and the corresponding domain of human TPO and myeloperoxidase. A phylogenetic analysis, using the sequences of the peroxidase domains from several animal myeloperoxidases, reveals that the Xenopus sequence groups with the Drosophila and human pxn orthologues (Fig. 3).
Expression of Xenopus tropicalis peroxidasin Gene
The identification of ESTs containing partial pxn sequences in both egg and neurula stage cDNA libraries suggests that Xtpxn is expressed both maternally and zygotically. Figure 4 is an RT-PCR analysis showing that Xtpxn is expressed in the fertilized egg and several stages after the activation of zygotic transcription in the mid-blastula. This finding confirms that Xtpxn is expressed both maternally and zygotically.
In Drosophila, the pxn gene is expressed in hemocytes as they migrate in the early embryo. Immunostaining for peroxidasin shows that it is present in hemocytes and mature basement membranes. Peroxidasin immunoreactivity is also detected in the fat body and the gastric caecae, suggesting that these organs may also be sources of secreted peroxidasin (Nelson et al., 1994).
No detailed description of the expression pattern of peroxidasin during early vertebrate development has been reported. We have determined the expression pattern of peroxidasin during the early development of the frog Xenopus tropicalis by whole-mount in situ hybridization. Figure 5A shows that strong maternal expression of Xtpxn is detected exclusively in the animal hemisphere of the single-celled zygote. Figure 5B is the open face of a zygote bisected along the animal vegetal axis and shows that Xtpxn mRNA is evenly distributed throughout the animal cytoplasm. This maternal expression in the animal hemisphere continues through early cleavage stages (Fig. 5C,D) but is eliminated from the embryo during later cleavage stages and is not detectable by in situ hybridization in late blastula stage 9 embryos (Fig. 5E).
The zygotic expression of Xtpxn is first detected by in situ hybridization in neurula stage embryos. Figure 5F,G is dorsal and lateral views of a neurula stage 17 embryo showing the strongest expression of Xtpxn along the dorsal axis within the developing neural tube. Low-level expression is also detected within lateral and ventral non-neural ectoderm. After the closure of the neural tube, several distinct regions of Xtpxn expression can be detected. Figure 5H is a lateral view of a stage 23 embryo that shows expression within the branchial arches, the tailbud region and the developing eyes. A distinct domain of expression is also detected anterior to the otic vesicle. By this stage, clear expression is detectable in the primordium of the pronephros within the intermediate mesoderm, although faint expression can be detected in this region as early as stage 21 (data not shown). A similar pattern of expression persists through to early tailbud stage 27 (Fig. 5I). However, by this stage, faint expression is also detected in the pronephric duct as it begins to extend toward the posterior, and the hindbrain adjacent to the otic vesicle (Fig. 5I,J).
The expression of Xtpxn within the neural tube is dynamic, and by tailbud stage 31, the stripe of expression within the hindbrain has disappeared and a new domain of expression is apparent at the midbrain/hindbrain boundary (Fig. 5K). The expression in the eye-forming region has resolved to expression within the developing lens. Other domains of expression are also detected in the posterior tail fin and the region of the proctodeum. Xtpxn continues to be expressed in the developing pronephros and prominent expression is detected in the pronephric tubules and the pronephric duct (Fig. 5L). These major sites of expression persist through later tailbud stages.
Figure 5M shows the anterior of a stage 34 embryo showing prominent expression in the pronephric rudiments, the eye, midbrain/hindbrain junction, and high levels of expression in the branchial arches. Figure 5N is a dorsal view of the same embryo and shows that Xtpxn expression extends, in a somewhat graded manner, from the midbrain/hindbrain junction some distance into the midbrain. Figure 5O is the posterior of an embryo and shows expression within the tail bud, the fin, and a region of posterior endoderm close to the proctodeum. Figure 5P shows high levels of expression within the pronephric tubules, which are undergoing complex looping morphogenesis, and within the cells of the extending pronephric duct. Figure 5Q is a transverse section through a stage 34 embryo at the level of the pronephros and shows high-level expression in the nephric mesoderm and the overlying ectoderm. Figure 5R is a high-power image of a transverse section through the posterior trunk of stage 34 embryo, showing expression with the cells lining the pronephric duct. In post–tailbud embryos, the expression of Xtpxn becomes much more general in all tissues but persists at high levels in the pronephros (data not shown).
We present the first detailed expression pattern reported for a vertebrate orthologue of the Drosophila peroxidasin gene. We show that Xenopus tropicalis Xtpxn is expressed in several distinct tissues during early development, including the neural tube and the tail-forming region. The pronephros represents a prominent site of expression for Xtpxn, and it is interesting to speculate what its role might be during the development of this organ. It has been proposed that secreted peroxidasin functions to modify and cross-link proteins in the extracellular environment (Nelson et al., 1994). Given that Xtpxn is expressed in the early pronephric primordium, it is possible that peroxidasin might have a roles in modifying extracellular matrix components necessary for the complex looping morphogenesis that characterizes the formation of this organ (Vize et al., 1995). Other roles might be associated with the ability of peroxidases to scavenge reactive oxygen species (ROS). The adult kidney is a prominent site of expression for other peroxidases such as the selenoprotein glutathione peroxidase, and oxidative stress is associated with chronic renal failure in humans (Brigelius-Flohe, 1999; Massy and Nguyen-Khoa, 2002). With this in mind, it is possible that peroxidasin may also be involved in detoxifying ROS in the functional pronephros. It will be interesting to see if the peroxidasin orthologues have conserved expression and roles in development of the kidneys of higher vertebrates.
It also is important to note that hydrogen peroxide, the major substrate for peroxidase enzymes, has been shown to be involved in regulating the activity of several signal transduction pathways (Stone and Collins, 2002; Aslan and Ozben, 2003; Zhougang and Schnellmann, 2004). This finding raises the intriguing possibility that the spatially restricted expression of Xtpxn in the early embryo might reflect a role for ROS species in cellular signaling processes during early development.
Embryos were generated by in vitro fertilization using a sperm suspension and eggs derived from females induced to lay by injection of 100 units of human chorionic gonadotrophin. Staging of Xenopus tropicalis embryos was carried out according to the same criteria used for staging Xenopus laevis (Nieuwkoop and Faber, 1967). Embryos were cultured to stage 8 in MRS/9 and then transferred to MRS/20 for further culture.
Identification of Xenopus tropicalis peroxidasin
The peptide sequence of human thyroid peroxidase was used to carry out a translated BLAST search of the Sanger Centre Xenopus tropicalis EST database (www.sanger.ac.uk/Projects/X_tropicalis/). The best recorded hit was for an EST from a neurula stage library. Subsequent analysis identified the EST as corresponding to the putative X. tropicalis pxn orthologue (accession no. AL783545). The EST clone was obtained from the MRC Geneservice (www.hgmp.mrc.ac.uk/geneservice/index.shtml) and the 1,250-bp insert was fully sequenced. To obtain more Xtpxn sequence, the cDNA sequence was used in a nucleotide BLAST search of the Department of Energy, JGI, Xenopus tropicalis genome project database (http://genome.jgi-psf.org/xenopus0/xenopus0.home.html). The 68,592 genomic contig of the putative Xtpxn gene was assembled from shotgun sequencing runs using the Seqman programme of the DNAstar Lasergene suite. The Megalign program of the DNAstar Lasergene suite was used to carry out sequence comparisons and alignments. Signal peptide analysis was undertaken using the SignalP 3.0 server (http://www.cbs.dtu.dk/services/SignalP/). Protein motifs were identified within the analyzed peptide sequences using the ExPASy Prosite (http://ca.expasy.org/prosite/) web server. The catalytic domains of myeloperoxidase family members were aligned by the Clustal method, and an unrooted phylogenetic tree was produced using the neighbor-joining method with the Tree View program. The sequences of the EST and genomic contig have been deposited in the GenBank sequence data base with accession nos. AJ810446 and BK005589, respectively.
RNA was isolated from batches of embryos from different stages using the Qiagen RNeasy minikit. Genomic DNA contamination was removed using the Ambion DNA-free reagent system. A total of 300 ng of each RNA was used to make random primed first-strand cDNA by using the Promega reverse transcription system. The cDNA was divided and subjected to PCR for detection of Xtpxn and XtODC, which is a ubiquitously expressed gene used as an internal loading control. PCR products were run on either 2% agarose or 6% polyacrylamide gels and visualized by staining with ethidium bromide. Analyses were undertaken three times with similar results. The primer sets used produced amplicons of 466 for Xtpxn and 118 bp for XtODC.
Xtpxn forward: GGAGACAACATCACCAAAGT; Xtpxn reverse: GTCCGTTATAGCATTCACAC. The amplicon for Xtpxn spans exons 20 to 23. Contamination with genomic DNA would produce an amplicon considerably larger than the cDNA amplicon. XtODC forward: GAGAGATCTAGACTGGAGGAAGGTTTCTGTGTGC; XtODC reverse: GAGAGAGAATTCAGATCAGCAACATAAAAGGC.
For Xtpxn, the PCR regimen was 25 cycles of 94°C for 30 sec, 51°C for 30 sec, and 72°C for 30 sec. For XtODC, the PCR regimen was 35 cycles of 94°C for 30 sec, 60.5°C for 30 sec, and 72°C for 30 sec. All PCR regimens included an initial denaturation for 2 min at 94°C and a final extension of 7 min at 72°C.
In Situ Hybridization Analysis
A DIG-labeled in situ probe was synthesized by linearizing the original Xtpxn EST with PstI and transcribing with T3 polymerase to yield a probe of 1,250 bases. Wild-type pigmented embryos were cultured to appropriate stages, fixed and processed for in situ hybridization. Embryos were cultured to appropriate stages and then fixed in MEMFA (0.1 M MOPS, 2 mM ethylenediaminetetraacetic acid [EDTA], 1 mM MgSO4, 3.7% formaldehyde) for 1 hr at room temperature and stored in 100% ethanol at −20°C until further processing. Embryos were rehydrated through a graded series of ethanol and then rinsed in phosphate buffered saline (PBS) with 0.1% Tween. Embryos were treated with 10 μg/ml proteinase K for 4 to 17 min at room temperature. Hybridization was carried out overnight at 60°C in 50% formamide, 5× standard saline citrate [SSC], 1 mg/ml total yeast RNA, 100 μg/ml heparin, 1× Denhardt's, 0.1% Tween, 0.1% CHAPS, 10 mM EDTA. Extensive washes in 2× SSC, and 0.2× SSC at 60°C were followed by washes at room temperature with maleic acid buffer, MAB (0.1 M maleic acid, 0.15 M NaCl, 0.1% Tween, pH 7.8), and blocking in 2% Roche Blocking Reagent and 20% heat treated lamb serum for 2 hr at room temperature. Embryos were then incubated with anti-DIG antibody at a dilution of 1/2,000 in blocking solution at 4°C overnight. After extensive washes in MAB, expression was visualized using the BM purple alkaline phosphatase substrate (Roche) containing 2 mM levamisole as an inhibitor of endogenous alkaline phosphatase activity. After color development, embryos were fixed and bleached in PBS-A + 5% H2O2 for several hours.
After development of the color reaction, embryos were refixed overnight in 4% buffered paraformaldehyde. They were then embedded in paraffin wax through an ethanol/Histoclear series (National Diagnostics), and 10-μm sections were cut before mounting in Histomount (National Diagnostics).
Digital photography of specimens was carried out by using a Spot Junior CCD camera (Diagnostic Instruments). Image manipulation was carried out using Adobe Photoshop.
A.J.T. and I.D.M. were funded by a Biotechnology and Biological Sciences Research Council CASE research studentship, and H.V.I. funded by a University Award from the Wellcome Trust.