Printer Friendly

Aquaporin in Chondrosia reniformis Nardo, 1847 and its possible role in the interaction between cells and engulfed siliceous particles.

Abstract. The sponge Chondrosia reniformis selectively engulfs siliceous particles that, when in crystalline form, become quickly dissolved in its ectosome. The molecular mechanism, identity, and physiological significance of the cells involved in this process are not completely understood. In the present study, we applied light and electronic microscopic techniques to show how the quartz particles in C. reniformis are enveloped through collagen fibers and host cells near the surface of these organisms. As various aquaporins from bacteria, animals, and plants bidirectionally conduct met alloids--including silicon ions--through the cell membrane, the presence and potential involvement of aquaporins in quartz dissolution in C. reniformis have been investigated. An aquaporin-like transcript (CrAQP) was isolated according to the transcriptome sequencing results in C. reniformis. The full-length CrAQP cDNA is 907 nucleotides long, with a 795-base pair (bp), open reading frame encoding a protein of 265 amino acids, a 29-bp, 5'-non-coding region, and a 83-bp, 3'-untranslated region. The Bayesian phylogenetic inference suggests that CrAqp is closely related to the Aqp8L grade, which is also implicated in [H.sub.2][O.sub.2] transport. Quantification of CrAQP mRNA through qPCR indicated that the transcript level was higher in the ectosome than in the choanosome. Immunofluorescence of a mammalian AQP8 in C. reniformis showed positivity in some cells near the quartz particles, a finding that may support the initial hypothesis of the potential involvement of CrAQP in quartz erosion. However, the features of the primary structure of this protein offer a new viewpoint about the functional role of these molecules in this process: that CrAQP may be involved in the permeation of [H.sub.2][O.sub.2] released during silica erosion.

Introduction

Marine sponges incorporate foreign materials from the environment (Cerrano et al., 2007a), a peculiar behavior that Haeckel (1872) was the first to observe over 140 years ago. Since then, other authors have examined this phenomenon, describing different examples of inorganic particles embedded in the spongin skeletons of these animals. The interaction of dermal cells with foreign particles suggests a direct role for these same cells in the movement, internalization, and orientation of sand grains, as described in the keratose demosponge Dysidea etheriade Laubenfels, 1936 (Teragawa, 1986a, b).

The marine demosponge Chondrosia reniformis Nardo, 1847 incorporates a wide range of foreign materials, including silica, in its ectosome; showing unusual selectivity, C. reniformis can partially dissolve only the crystalline form of this compound. The molecular mediator of this unique process has been identified as ascorbic acid (Bavestrello et al., 1995), and no cellular intervention has been described to date. The incorporation--and partial erosion--of quartz particles in C. reniformis have been associated with sponge collagen biosynthesis (Giovine et al., 2013), although there is no molecular evidence to support this hypothesis. Recently, a full-length cDNA encoding for a non-fibrillar collagen type and the enzyme involved in its posttranslational modification were identified and cloned from C. reniformis. Analysis of the gene expression showed positive regulation through soluble silicates (Pozzolini et al., 2012, 2015). These results have led to a hypothesis regarding the significance of the etching of quartz particles in C. reniformis: quartz erosion through ascorbic acid locally releases soluble silicate, thereby promoting collagen gene expression (Giovine et al., 2013).

The major intrinsic proteins (MIPs) are a large superfamily of integral membrane proteins present in all biological systems from Archaea and bacteria to eukaryotes (Engel and Stahlberg, 2002; Zardoya, 2005). MIPs were initially grouped into two major divisions according to substrate specificity: the water-selective-type channels, called aquaporins (AQPs), and the glycerol-uptake facilitators, or aquaglyceroporins (GLPs), when these proteins permeated glycerol or small solutes (Park and Saier, 1996; Heymann and Engel, 1999). However, the most recent studies have revealed a more complex subdivision structure for MIPs.

Aquaporins are now considered to belong, phylogenetically and functionally, to one of four grades: the water-selective classical aquaporins; the Aqp8-type aquaammoniaporins, which transport water, ammonia, and urea; the unorthodox aquaporins, for which cell permation properties have yet to be understood; and the classical aquaglyceroporins, which facilitate the transport of water, arsenic, urea, and polyols (Finn et al., 2014; Finn and Cerda, 2015). All AQPs form tetramers, in which each monomer functions as an independent pore. A single monomer comprises six transmembrane helices connected by five loops that delimit a polar channel. Two highly conserved signatures, known as NPA (Asn-Pro-Ala), are located at the center of the pore, forming a constriction (Wu and Beitz, 2007). The primary structure of an aquaporin is characterized by the repetition of two similar halves of the protein, comprising two imperfectly repeated regions, each containing three transmembrane helices (Pao et al., 1991). In the demosponge Suberites domuncula, AQPs were recently characterized as water channels in the dehydration and hardening of siliceous spicules produced from this animal (Muller et al., 2011). Among the members of the phylum porifera, phylogenetic evidence supports the existence of an ancestral., Aqp8-like channel in addition to an aquaglyceroporin (Finn et al., 2014). Apart from water and glycerol, various channels permeate a number of other compounds, such as N[H.sub.3] and C[O.sub.2], through AQP1, AQP4, and AQP5 in mammals, or ammonium and/or methylammonium transport B (AmtB) in bacteria (Musa-Aziz et al., 2008; Finn and Cerda, 2015); and [H.sub.2][O.sub.2], through some members of plasma membrane intrinsic proteins (PIPs), Nodulin 26-like intrinsic proteins (NIPs), X-intrinsic proteins (XIPs), and tonoplast intrinsic proteins (TIPs) in plants (Bienert et al., 2007; Bienert and Chaumont, 2014). Various AQPs that bidirectionally conduct silicon also have been described, such as NIPs in plants (Bhattacharjee et al., 2008) and AQP3, AQP7, AQP8, AQP9, and AQP10 in mammals (Garneau et al., 2015).

Because AQPs are able to permeate silicon, we hypothesized that AQPs were involved in managing the silica-water interface around quartz grains incorporated in the C. reniformis sponge. Using histological staining for collagen and cells, we describe the organization of C. reniformis as it relates to the engulfed siliceous particles; taking a molecular approach, we identified and cloned the full-length cDNA that encodes the sponge aquaporin CrAQP, whose transcript level was assayed in the ectosome and choanosome of C. reniformis. Finally, we assessed immunohistochemically the presence of CrAQP-like immunoreactivity in cells near the siliceous particles.

Materials and Methods

Experimental animals

Specimens of Chondrosia reniformis were collected in the area of Portofino Promontory, Liguria, Italy, at depths of 10-20 m. During sampling and transport of the specimens, water temperature was maintained at 14-15 [degrees]C; in short-term storage, the sponges were kept at 14 [degrees]C in a 200-1 aquarium containing natural seawater (NSW; salinity 37 ppt) and equipped with an aeration system. The NSW was collected in the same area of Portofino Promontory. Samples of C. reniformis were frozen at -80 [degrees]C or fixed in 4% paraformaldehyde (PAF; Carlo Erba Reagenti, Rodano, Italy) in 0.1 mol [1.sup.-1] phosphate-buffered saline (PBS), pH 7.4, and 8% NaCl (PBS), at 4 [degrees]C overnight for 16 hours.

Histological and histochemical staining

Samples of Chondrosia reniformis were affixed overnight (about 16 h) in 4% PAF in PBS. Subsequently, the specimens were rinsed with PBS, dehydrated in an increasing series of ethyl alcohol (70%, 80%, 90%, 95%, and 100% I, and 100% II, for 1 min each), cleared in xylene and ethyl alcohol at a 1:1 ratio for 15 min, cleared in xylene for 15 min, and embedded in Paraplast (McCormick Scientific, Richmond, IL) at 56 [degrees]C, after three changes of paraffin, for 2 h each. The embedded specimens were cut into 6-[micro]m sections and mounted on slides. The slides for light microscopy were alternately stained using Masson's Trichrome or Picrosirius Red staining for collagen (Junqueira et al., 1979). When using Masson's Trichrome, collagen appears blue and cytoplasm appears red; with Picrosirius red stain, however, collagen appears red in bright-field microscopy; when examined through cross-polarized light microscopy, larger collagen fibers are bright yellow or orange, and thinner fibers, including reticular fibers, are green (Junqueira et al., 1979). All of the histologically stained sections were visualized through a Leica DMRB light and epifluorescence microscope (Leica, Wetzlar, Germany), equipped with cross-polar and differential-interference contrast filters (using Nomarski contrast); images were acquired using a Leica CCD camera (DFC420C; Leica). Image analysis was performed using the open-source software ImageJ (Rasband, 2014).

Environmental scanning electron microscopy

For observation with the environmental scanning electron microscope (ESEM), histological sections were deparaffinized, as described for the histological techniques, ethanol-washed, dried, and graphite-covered. Thus, it was possible to observe the surface of the sections, or the internal area of the ectosome, the surface of some internalized grains, and the surrounding sponge tissue. Moreover, we were able to analyze the chemical elements present in the grains, ultimately identifying the silica grains. These observations were made using an ESEM (VEGA 3-TESCAN, LMU-type; TESCAN Orsay Holding, Brno, Czech Republic), equipped with a microanalyzer system (EDS-Apollo_x, and EDS Texture And Elemental Analytical Microscopy software (TEAM); EDAX, Inc., Mahwah, NJ). Silicon dioxide (Si[O.sub.2]) was used as a standard in the form of Wollastonite from a mineral block (MAC; Micro-Analysis Consultants Ltd., Cambridge, UK).

CrAQP cDNA identification and cloning

The 5' region of the aquaporin transcript of Chondrosia reniformis was identified using 1) a transcriptome sequencing approach, and 2) subsequently, the 3'-end was experimentally completed using a polymerase chain reaction (PCR).

1) Transcriptome sequencing and analysis of Chondrosia reniformis. Total RNA from C. reniformis was extracted through Isol-RNA Lysis (5'-Prime) using a freshly collected specimen. Then the poly(A) fraction was isolated using the FastTrack MAG mRNA Isolation Kit (Life Sciences, Woburn, MA), according to the manufacturer's instructions. The obtained poly-A fraction was subjected to a second round of mRNA messenger purification to improve the purity of the sample. Total and purified mRNA samples were assessed for quality using the Bioanalyzer 2100 (Agilent, Inc., Santa Clara, CA).

A total of 7 [micro]g of purified mRNA of C. reniformis was used for cDNA synthesis and normalization (Evrogen JSC Co., Moscow, Russia). A total of 5 [micro]g of normalized cDNA was analyzed through pyrosequencing using an FLX Titanium 454 sequencer (F. Hoffmann La Roche Ltd., Basel, Switzerland) at the BMR Genomics Company (Padua, Italy).

The obtained short nucleotide reads were assembled using MIRA 4.0 software (Mirametrics, Tucson, AZ) to produce error-free, unique, contiguous sequences (contigs). BlastX was used to align the obtained 19,678 transcripts (isotig) to the protein database (Swiss-Prot Ref_Seq_prot (e-value < 0.00001); March 2013 release; UniProt Consortium, 2015). A candidate AQP transcript, called porifera_c 14748, was identified through a sequence homology search with BLAST suite (NCBI; http://blast.ncbi.nlm.nih.gov/Blast.cgi).

2) 3'-end completion of CrAQP and full-length cloning. The identified sequence was incomplete at the 3'-end; therefore, this sequence was experimentally obtained using the GeneRacer kit (Life Technologies Corp., Carlsbad, CA), according to the manufacturer's instructions, using an OdT-adapter (provided in the kit) as a primer for cDNA synthesis.

Sponge cDNA was synthesized using the Superscript III RT-PCR System (Life Technologies Corp.). In a final reaction volume of 20 [micro]l, 200 ng of purified mRNA was incubated in the appropriate buffer containing 2.5 [micro]mol [1.sup.-1] Odt-adapter, 0.5 mmol [1.sup.-1] dNTP mix, 5 mmol [1.sup.-1] DTT, 40 units of ribonuclease inhibitor (RNASEOUT; Invitrogen, Carlsbad, CA), and 15 units of Superscript. The reaction was performed at 60 [degrees]C for 50 min. To remove RNA complementary to the cDNA, 2 units of E. coli RNase H (Invitrogen) was added, and the reaction was further incubated at 37 [degrees]C for 20 min.

PCR amplification was subsequently performed using FwAQ1 as a forward primer (Table 1), coupled with the 3'-GeneRacer reverse anchor primer.

The PCR reaction was performed using Platinum High-Fidelity Taq Polymerase (Life Technologies Corp.), according to the manufacturer's instructions, using 1 [micro]l of synthesized odT-adapted-cDNA. The touchdown thermal profile was performed at 94 [degrees]C for 2 min, followed by 5 cycles at 94 [degrees]C for 30 seconds, at 65 [degrees]C for 30 s, and at 68 [degrees]C for 30 s; 5 cycles at 94 [degrees]C for 30 s, at 62 [degrees]C for 30 s, and at 68 [degrees]C for 30 s; and 25 cycles at 94 [degrees]C for 30 s, at 60 [degrees]C for 30 s, and at 68 [degrees]C for 30 s; and one cycle at 68 [degrees]C for 10 min.

The obtained 705 base pair (bp) PCR product was directly cloned into a pJET1.2/blunt vector using the Clone-JET PCR Cloning Kit (Thermo Fisher Scientific), according to the manufacturer's instructions, for sequencing. The obtained sequence was subsequently aligned and joined to that of the previously identified partial putative transcript, to obtain the full-length cDNA. To ensure that the identified sequence represented a true gene product, experimental confirmation was performed through PCR amplification of sponge cDNA. For this amplification, two primers: Fw-CrAQP and RevCrAQP (Table 1), specific for the full-length CrAQP sequence, were used. Subsequently, the identity of the amplified product with the transcriptome sequence was assessed through DNA sequencing.

Sequence analysis of CrAQP

The candidate CrAQP's nucleotide and conceptual translation sequences were at first analyzed using the BLAST suite at the NCBI (http://blast.ncbi.nlm.nih.gov/Blast.cgi). The presence of transmembrane helices was detected using TMHMM Server ver. 2.0. (Center for Biological Sequencing Analysis, Technical University of Denmark, Kongens Lyngby, Denmark). Pfam annotation of the C. reniformis transcriptome was performed using TransDecoder (trinityrnaseq r20130225; Trinity Software Distribution, Brandon, FL; Grabherr et al., 2011), and was used to verify the presence of the MIP domain in the CrAQP candidate sequence.

To infer the phylogenetic relationships of CrAQP with other AQP grades already recognized as occurring in parazoan, cnidarian, or basal deuterostome subclusters, we used the dataset and alignment computed by Finn and colleagues to build the tree presented in figure S4 of their 2014 paper (Finn et al., 2014). To overcome the anomalous clustering of the aqpH1-5 orthologs between the unorthodox aquaporins and aquaglyceroporins (glps), the zooanthellate orthologs were removed. The conceptual translations of complete cDNA containing the MIP domain (identified using the Pfam annotations) from the eight transcriptomes of marine sponges from Riesgo et al. (2014) and other sequences retrieved from the DDBJ/EMBL/GenBank were also added to the dataset. The sequences used to place CrAQP among AQP grades are listed in Supplementary Table 1 (SM_l.doc, http://www.biolbull.org/content/supplemental).

Porifera's putative AQPs were added manually, using MacVector (MacVector, Inc., Cambridge, UK), to the above-cited multiple-sequence alignments previously obtained with Mafft ver. 7.058b with the L-INS-i algorithm and default settings (Katoh and Standley, 2013).

Bayesian inference was performed on the codon/amino acid alignment, as reported in Finn et al. (2014), with three million Markov chain Monte Carlo (MCMC methods) generations. The phylogenetic tree was rendered in Geneious (Biomatters Ltd., Auckland, New Zealand).

Tissue distribution of CrAqp

CrAqp transcript analysis. The amount of CrAqp mRNA in the two main regions of Chondrosia reniformis, the ectosome and choanosome, was analyzed using a qPCR approach with Chromo 4 Real-Time PCR Detector (MJ Research PTC200; Labcare Service, Braintree, UK).

Glyceraldehyde 3-phosphate dehydrogenase (GAPDH; accession no. KM217385) was used as a reference gene for normalization. Used previously in the same system (Pozzolini et al., 2015), this gene presents a higher stability level than the other reference gene considered (i.e., Beta Tubulin) and, at the same time, a threshold cycle (Ct) not too different from the Ct value of the analyzed gene. Each PCR reaction was performed in a 20-[micro]l reaction containing 1 x master mix iQ SYBR Green (Bio-Rad Laboratories, Hercules, CA), 0.2 [micro]mol [1.sup.-1] of each primer, and 0.8 [micro]l of synthesized cDNA, obtained through reverse transcription using the iScript cDNA Synthesis Kit (Bio-Rad Laboratories), according to the manufacturer's instructions, using 200 ng of mRNA purified from the ectosome or choanosome region. All samples were analyzed in triplicate.

The following conditions were used: initial denaturation for 3 min, followed by 45 cycles of denaturation at 95 [degrees]C for 15 s, and annealing and elongation at 57.7 [degrees]C for 60 s. Fluorescence was measured at the end of each elongation step. The next step involved the slow heating (1 [degrees]C/s) of the amplified product from 55 to 92 [degrees]C to generate a melting temperature curve. All primers, Fgapd and Rgapd for the reference genes, and FCrAQP and RCrAQP (Table 1) for the target gene, were designed using Beacon Designer 7.0 software (Premier Biosoft Intl., Palo Alto, CA), obtained from Tib Molbiol Srl, Genoa, Italy. The oligonucleotide sequences are indicated in Table 1. Data analyses were performed using the DNA Engine Opticon 3 Real-Time Detection System ver. 3.03 software program (Bio-Rad Laboratories); to detect the relative gene expression of CrAQP, compared to the sample with the lowest Ct value, the comparative threshold Ct method (Aarskog et al., 2000) was used with Gene Expression Analysis for iCycler iQ Real-Time Detection System software (Bio-Rad Laboratories; Vandesompele et al., 2002).

Immunofluorescence for AQPs. Because molecular analyses revealed that CrAQP is related to the AQP8L grade (see Results below), indirect immunohistochemical reactions were performed using polyclonal rabbit anti-AQP8 antiserum (Cat. Sc-28624; Santa Cruz Biotechnology, Inc., Dallas, TX). This antiserum was raised against amino acids 1-85 in the N-terminus of human AQP8. The N-terminal sequence of human AQPS was aligned with the CrAQP predicted protein using ClustalW (European Molecular Biology Laboratory-European Bioinformatics Institute (EMBL-EBI), Hinxton, UK) to determine whether the antisera could recognize CrAQP in the tissues of C. reniformis. The secondary antibody was Alexa 488-conjugated chicken anti-rabbit antiserum (1:800 in PBS, Molecular Probes; Invitrogen). As a negative control, the primary antiserum was omitted (Saper and Sawchenko, 2003; Saper, 2009).

Statistical analysis

All parameters were examined using a paired t-test. P-values < 0.05 were considered significant.

Results

General anatomy of the sponge Chondrosia reniformis

The sponge body is permeated with a network of canals (C) and chambers that form the aquiferous system (Fig. 1, a-e). The outer surface is typically bound by flattened cells, known as pinacocytes (Pin), which constitute a layer called the pinacoderm. The inner zone is called the choanoderm or choanosome, comprising a chamber whose walls are formed from a single layer of flagellated cells, called choanocytes (Ch). These choanocyte chambers (ChC) are considered the morphofunctional units of the sponge. Between these two layers is the mesohyl, also known as the mesoglea, which forms the bulk of the sponge body. Alternatively, the pinacoderm and the mesohyl are called the cortex, and the choanoderm (CD) is known as the endosome. In the aquiferous system, endopinacocytes border inhalant and exhalant canals. On the basal surface, the basopinacocytes (Bas) make contact with the substrate, forming the epithelium competent for adhesion. Chondrosia reniformis belongs to a class of demosponges, whose organizational level is the so-called Leucon. The leuconoid condition results from the additional folding of the endosome and thickening of the cortex, reflecting the growth of the mesohyl. Typically, the skeleton is formed by spicules, calcareous or siliceous, or with an organic support. However, in C. reniformis, endogenous spicule and related sclerocytes are completely missing. In the mesohyl, there are many types of ameboid cells, some of which have been implicated in the secretion of skelet al elements, including collencytes, lophocytes, and spongocytes; however, in C. reniformisthese elements are missing (Gaino, 2011).

Histological observations

Siliceous particles of appreciable sizes were observed in the ectosome--but not in the choanosome--of Chondrosia reniformis. These particles were visible through a polarized light microscope and affected the surrounding organization of the ectosome involving both cellular elements and coll+agen fibers. DAPI staining facilitated the detection of cell nuclei around the particles and on the surface (Fig. 2a).

Bright-field microscopy with Masson 's Trichrome staining. Basic fuchsin has been used as a cytoplasmic stain to visualize the shape and size of the cells around siliceous particles. The cells are characterized by an egg-shaped cell body approximately 3-4 [micro]m in size (Fig. 2b). The cells were aligned along the particle surface, occasionally within surface irregularities, and bearing long, thin processes on the particle surface (Fig. 2c; Supplementary video 1, http://www.biolbull.org/content/supplemental). Collagen fibers closely wrapping the particles were visible through Nomarski contrast (Fig. 2, d-g; Supplementary video 2, http://www.biolbull.org/content/supplemental).

Cross-polarized microscopy with Picrosirius Red staining. In bright-field microscopy, collagen was red (or dark gray in black and white images; see areas labeled (**), Fig. 2, b-g), while the color of the collagen fibers, when viewed through crossed-polarized microscopy, did not remain uniform throughout the entire sponge. Generally, in histologic sections the mesohyl presented as pale yellow and green collagen fibers (labeled ( + ); Fig. 2h, i), while the ectosome was primarily reddish-orange (labeled ( = ); Fig. 2, h-o). However, intensity and hue of staining varied within the same tissue and among different specimens. The fibers enveloping the particles were uniformly red when viewed through a cross-polarized microscope (labeled ( = ); Fig. 2, h-o); thus, these fibers were considered to be large collagen fibers. The red staining was particularly evident when the particles were thicker and protruded from beyond the histological sections, presenting as an enveloped surface with a net of collagen fibers (labeled ( + ); Fig. 2, h, i).

Environmental scanning electron microscopy (ESEM) analysis

ESEM analysis revealed the peculiar erosion pattern of the siliceous grains detected within the ectosome and the fiber organization enveloping the particle (Fig. 2p). This pattern has been described for Chondrosia reniformis (Bavestrello et al., 1995), revealing a sharp decrease in grain size and a blurring of their pristine shape. Energy-dispersive x-ray (EDX) analysis confirmed the chemical composition of the engulfed grain as nearly pure silicon dioxide (Si[O.sub.2]) (Fig. 3).

Characterization of CrAQP cDNA

Initially, the 618-bp, 5'-end of the putative AQP transcript (CrAQP) of Chondrosia reniformis was identified through transcriptome sequencing; subsequently, a 705-bp, 3'-end region was experimentally obtained through PCR using the GeneRacer Kit (Life Technologies, Woburn, MA). The obtained sequence was aligned and joined to that which previously identified the 5'-end, obtaining the full-length CrAQP cDNA.

CrAQP cDNA is 907 nucleotides long, and it contains a 5' untranslated region (UTR) of 29 nucleotides, an open reading frame of 795 nucleotides; it encodes a protein of 265 amino acids, followed by a 3'-UTR of 83 nucleotides, including a stop codon (TAA), a putative polyadenylation consensus signal (AATAAA), and a poly(A) tail (Fig. 4).

The putative translation product has an estimated relative molecular mass (Mr) of 27.93 kDa, with an isoelectric point (pI). 8.82. Hydropathy analysis predicted six transmembrane regions with an N-terminus localized in the cytoplasm, similar to the sequence reported for other MIP superfamily members. Amino acid sequence analysis revealed the presence of two conserved NPA motifs identified in all MIP superfamily members (Fig. 4).

Both BlastX and BlastP search results (eval 1e-44 over 42% identity and 4e-44 over 42%, respectively) indicated that the closest entry in the DDBJ/EMBL/GenBank was an aquaporin from the Demospongiae Suberites domuncula (accession no. CBY89223), bearing the typical Asn-Pro-Ala signature motifs.

Multiple-sequence alignment analysis revealed the residues present in the aromatic/arginine (ar/R) constriction of C. reniformis AQP: His62, He184, Glyl92, and Argl99 (Fig. 4). These residues appear to be identical to the residues present in human AQP8 (data not shown), but the residues could have evolved equally via convergent evolution; only a robust phylogenetic analysis can reveal relatedness among these genes. The AQP of C. reniformis shows 37.7% amino acid sequence identity with the AQP of Suberites domuncula, and 28.5% and 32.3% amino acid sequence identity with human AQP1 and AQP8, respectively.

The sequence of full-length cDNA has been deposited in the DDBJ/EMBL/GenBank (accession no. KR780752).

Phylogenetic placement of CrAQP

Only three complete cDNAs containing the MIP domain were identified among the eight transcriptomes of marine sponges from Riesgo et al. (2014), showing a compatible sequence length and BLAST-based homology: the contig_524 from Chondrilla nucula (Demospongiae), the contig_4169 from Petrosia ficiformis (Demospongiae), and the comp7958_c0_seql from Aphrocallistes vastus (Hexactinellidae). Those sequences, along with the aquaporin sequence from Suberites domuncula (accession no. CBY89223), were added to the existing alignment comprising representatives of AQP grades occurring in basal metazoan organisms. In the original dataset, the only two available Aqp8-like proteins from sponges were obtained from Suberites domuncula and Oscarella carmela (EC374967), whereas several aquaglyceroporins (Glps) were selected from the sponges Amphimedon queenslandica, Ephydatia muelleri, Lubomirskia baicalensis, and Heterochone calyx.

The present Bayesian inference obtained with the updated multiple-sequence alignment, with three million MCMC generations, showed that new sponge sequences come out with two (C. reniformis and C. nucula) clustering as Aqp8L, one (P. ficiformis) as Aqp 12L, and the last (A. vastus) as a Glp (Fig. 5, summarized Bayesian Inference; Supplementary Fig. 1, extended phylogenetic tree, SM_l.doc, http://www.biolbull.org/content/supplemental).

The CrAQP of C. reniformis and the C. nucula transcript form a well-supported cluster (Posterior Probability value 1) with the two other sponges, S. domunculaand the O. carmela, the latter having been described in a previous analysis (Finn et al., 2014) as belonging to the Aqp8-like aquammoniaporins grade. The placement of the CrAQP among this well-supported cluster is a good hint to consider the corresponding transcript as coding for an Aqp8-like channel.

The P. ficiformis sequence clusters together with Aqp 12-like representatives. To date, this is the first record of an unorthodox Aqp12-like sequence found in Porifera. With the expansion of transcriptomic studies on Porifera, it was bound to exist, but when the PLoS One paper was published (Finn et al., 2014), the sequence was not available in the public databases (R. N. Finn, University of Bergen, Norway; pers. comm.).

The A. vastus sequence forms a cluster with the H. calyx representatives of Glps (Posterior Probability value 1), in turn, clustering with Glp 1 orthologs rather than with GlpF-like variants found in A. queenslandica.

Tissue distribution of CrAqp

CrAQP transcript analysis. The tissue CrAQP gene expression pattern was examined in C. reniformis through qPCR analysis in two main sponge tissues: the ectosome and the choanosome. Expression analysis revealed that the CrAQP transcript is significantly more expressed in the ectosome than in the choanosome (threefold higher, P < 0.0025) (Fig. 6). This finding indicates that this protein is more represented and active in the cortical part of the animal body, i.e., the tissue directly involved in sediment incorporation.

Immunofluorescence

The alignment of human AQP8 AA 1-85 and CrAQP is shown in Fig. 7, and suggests that the anti-hAQP8 antibody can recognize CrAQP. However, because of the low identity between human AQP8 and CrAQP, we cannot exclude the possibility that these antibodies do not cross-react with other AQPs. AQP8-like immunoreactivity was observed around siliceous particles (Fig. 8), but not all nuclei close to the grains were surrounded by immunoreactive cytoplasm--a finding that suggests that more than one cell type (or one cell type in different stages) could adhere to the grains.

Discussion

The general morphological organization of Chondrosa reniformis has been described previously (Bonasoro et al., 2001). In addition, several scientific publications have provided details of the physiological reactivity of this animal with inorganic particles, describing its startling ability to select its mineralogical composition (e.g., Bavestrello et al., 1998). This unusual., and not completely understood, mechanism has been correlated with the expression of collagen genes, confirming the relevance of this ancestral model to the study of the evolution of extracellular matrix proteins. The selectiveness of mineral particles is more intriguing, considering that sponges are characterized by the incorporation of calcareous, not siliceous, particles. In this case, the development of these animals is not strictly associated with collagen production, and the embedding of quartz negatively affects biological processes such as morphogenesis and growth (Cerrano et al., 2007b).

With these assumptions, we examined the interactions between sandy grains and the sponge extracellular matrix, showing, for the first time, the close involvement of cells on the surface of quartz particles during processing. As seen in Figure 2d and g, and in supplementary video 2 (www.biolbull.org/content/supplemental), specific collagen staining also showed rapid embedding of siliceous particles within the collagenous matrix of sponges. The superficial position of the siliceous grains within the ectosome suggests that the close connection to cells and collagen fibers is a rapid reaction of the C. reniformis tissue.

The silica exposure of mammalian models consistently leads to fibrosis (Green and Vallyathan, 1996). In a similar manner, the production of collagen in sponges could be assisted through the incorporation of quartz particles. Partial quartz dissolution through ascorbic acid in C. reniformis tissue releases soluble silicates (Bavestrello et al., 1995). which could permeate some MIP superfamily members (Bhattacharjee et al., 2008; Mukhopadhyay et al., 2014; Garneau et al., 2015) and upregulate sponge collagen gene expression (Pozzolini et al., 2012). AQPs also have been identified in sponges, and are associated with dehydration and biosilica hardening in species that synthesize spicules (Muller et al., 2011). These considerations suggested the potential involvement of AQPs in quartz dissolution in C. reniformis.

A transcript encoding the AQP-like protein, CrAQP, was isolated from C. reniformis, and the deduced amino acid sequence has made it possible for us to infer its phylogenetic placement. Even if prudence should be applied when considering the relatedness of the isolated sequence due to the deficiency in sponges of a diversified pool of genetic resources and the low amino acid identity, the results of the Bayesian phylogenetic inference indicated that (similar to the AQP of S. domuncula) CrAQP robustly clustered with other representatives of the Aqp8-like aquammoniaporins grade (Fig. 6). For the current phylogenic analysis, eight transcriptomes of marine sponges from Riesgo et al. (2014) were also considered, from which we were able to identify only three transcripts with MIP domain from the Demospongiae Petrosia ficiformis and Chondrilla nucula, and from the Hexactinellid Aphrocallistes vastus. In C. nucula, a transcript coding for a putative Aqp8-like channel has been identified and found to be clustering with the CrAQP of C. reniformis and AQPs of S. domuncula and O. carmela. In A. vastus, a transcript has been found coding for a putative aquaglyceroporin that groups with other Porifera Glpl orthologs. Interestingly, in P. ficiformis we identified a transcript putatively coding for an unorthodox aquaporin (Aqp12L): it is the first report of this kind of channel to be found in Porifera, but it is to be expected with the expansion of transcriptomic data from this animal lineage, and is surely worth further investigation.

The presence of AQP8-like aquaporins in sponges was revealed by an analysis of the phylogenetic relationships between various AQP orthologs (Fig. 6). Indeed, AQP8 is an early divergent metazoan AQP that seems to be present as a single copy in animals (Zardoya, 2005). Considering the endosymbiotic theory, these proteins are also the only ones to be expressed on internal., mitochondrial., and endoplasmic membranes.

After the coding sequence of C. reniformisAQP was identified, tissue localization of CrAqp transcripts through quantitative PCR was conducted to verify the involvement of these proteins in silica erosion. However, knowledge of the deduced amino acid sequence has facilitated the selection of the most appropriate AQP antibody for tissue immune detection.

Sponges lack organized tissues and true organs. However, a general idea of the mRNA location of the identified CrAqp was possible after noting that the transcript levels in the two main regions of the animal--the ectosome and the choanosome--can be easily isolated and separately processed in this species. The data obtained indicated that CrAqp gene expression is three times higher in the ectosome than in the choanosome (Fig. 7), which is consistent with the initial hypothesis that AQPs may play a role in quartz dissolution. Indeed, the outer region is most subject to contact with the external environment and is directly involved in quartz incorporation and/or dissolution.

AQP8-like immunoreactivity was observed in cells near quartz particles (Fig. 8), a finding that does not demonstrate unequivocally the presence of a AQP8-like protein, but does, however, highlight an area for further research. Not all of the nuclei close to these grains were surrounded by immunoreactive cytoplasm. This difference in immunoreactivity to anti-AQP8 may reflect the involvement of more than one cell type or the same cell in different stages.

Analysis of both CrAQP transcripts and the immunofluorescence supports the hypothesis that AQP is involved in quartz etching in Chondrosia reniformis. However, the molecular information obtained in this study provides a new viewpoint as to the potential function of these proteins. The amino acid composition of the aromatic/arginine (ar/R) constriction in CrAQP appears to be identical to that in mammalian AQP8. Although there is no experimental evidence confirming the correlation between AQP permeants and the residues in ar/R constriction, identification of the residues in CrAQP revealed the presence of a histidine residue in the p1 position, as in the ar/R constriction of AQPs of mammals. Nevertheless, in the AQPs involved in [Si(OH).sub.4], permeation of the glycine (AQP10) or phenylalanine (AQP7 or AQP9) of animals was present (Garneau et al., 2015).

These findings do not support our initial hypothesis, that CrAQP is involved in silicate passage. Because of the phylogenic and structural correlation with mammalian AQP8s, the AQPs that are also involved in [H.sub.2][O.sub.2] permeation (Bienert et al., 2007), CrAQP, could play a role in [H.sub.2][O.sub.2] passage through the membrane. Indeed, studies have demonstrated that in C. reniformis the partial quartz dissolution through ascorbic acid releases soluble silicates into the medium (Bavestrello et al., 1995), and the ascorbic acid treatment of quartz dust generates hydroxyl radicals (OH-radicals) (Fenoglio et al., 2000), which can easily be converted into [H.sub.2][O.sub.2] in water (Hamilton et al., 2008). Moreover, murine macrophages incubated with ascorbic acidpretreated quartz showed an increment of reactive oxygen species (ROS), including [H.sub.2][O.sub.2] (Scarfi et at., 2009). The ROS are not only considered to be cell-damaging and destructive compounds, they have also been identified as crucial signaling molecules regulating various important responses and metabolic pathways (Bienert and Chaumont, 2010, 2014). It has been speculated that the [H.sub.2][O.sub.2] formed during quartz etching is internalized through CrAQP on the cell space, where this molecule acts on various protein and nonprotein targets that regulate diverse signaling cascades, such as fibrotic cytokine synthesis, as reported in mammalian models (Scarfi et al., 2009).

In conclusion, the microscopic observations obtained in the present study showed that the quartz particles incorporated in the tissue of C. reniformis are enveloped by collagen fibers and host cells on the surface. C. reniformis also possesses a gene that encodes an AQP8-like protein whose level of expression is higher in the ectosome than in the choanosome, where some of the cells near the quartz particles showed AQP8-like immunoreactivity. Furthermore, based on the structural features of the identified AQP, we propose that the potential function of these proteins is in permeation of the [H.sub.2][O.sub.2] released during the quartz erosion by ascorbic acid, even if the passage of water or silicates cannot be completely ruled out.

Acknowledgments

We are grateful to Roderick Nigel Finn, Department of Biology, Bergen High Technology Centre, University of Bergen, Norway, for kindly provinding AQPs alignments and supplemental phylogenetic analyses. We are also grateful to Laura Negretti, DiSTAV University of Genoa, Italy, for her help in the ESEM analysis. This work was supported in part by grants from EU (FP7, grant agreement no. 266033-SPonge Enzymes and Cells for Innovative AppLications-SPECIAL) to MG.

Literature Cited

Aarskog, N. K., and C. A. Vedeler. 2000. Real-time quantitative polymerase chain reaction. A new method that detects both the peripheral myelin protein 22 duplication in Charcot-Marie-Tooth type 1A disease and the peripheral myelin protein 22 deletion in hereditary neuropathy with liability to pressure palsies. Hum. Genet. 107: 494-498.

Bavestrello, G., A. Arillo, U. Benatti, C. Cerrano, R. Cattaneo-Vietti, L. Cortesogno, L. Gaggero, M. Giovine, M. Tonetti, and M. Sara. 1995. Quartz dissolution by the sponge Chondrosia reniformis (Porifera, Demospongiae). Nature 378: 374-376.

Bavestrello, G., U. Benatti, B. Calcinal., R. Cattaneo-Vietti, C. Cerrano, A. Favre, M. Giovine, S. Lanza, R. Pronzato, and M. Sara. 1998. Body polarity and mineral selectivity in the demosponge Chondrosia reniformis. Biol. Bull. 195: 120-125.

Bhattacharjee, H., R. Mukhopadhyay, S. Thiyagarajan, and B. P. Rosen. 2008. Aquaglyceroporins: ancient channels for met alloids. J. Biol. 7: 33.

Bienert, G.P., and F. Chaumont. 2010. Plant aquaporins: roles in water homeostasis, nutrition, and signaling processes. Pp. 3-36 in Transporters and Pumps in Plant Signaling, Vol. 7, M. Geisler and K. Venema. eds. Spinger-Verlag. Berlin.

Bienert, G. P., and F. Chaumont. 2014. Aquaporin-facilitated transmembrane diffusion of hydrogen peroxide. Biochim. Biophys. Acta 1840: 1596-1604.

Bienert, G. P., A. L. Moller, K. A. Kristiansen, A. Schulz, I. M. Moller, J. K. Schjoerring, and T. P. Jahn. 2007. Specific aquaporins facilitate the diffusion of hydrogen peroxide across membranes. J. Biol. Chem. 282: 1183-1192.

Bonasoro, F., I. C. Wilkie, G. Bavestrello, C. Cerrano, and M. D. Candia Carnevali. 2001. Dynamic structure of the mesohyl in the sponge Chondrosia reniformis (Porifera Demospongiae). Zoomorphology 121: 109-121.

Cerrano, C, B. Calcinal., C. G. Di Camillo, L. Valisano, and G. Bavestrello. 2007a. How and why do sponges incorporate foreign material? Strategies in Porifera. Pp. 239-246 in Porifera Research: Biodiversity, Innovation & Sustainahility, Serie Livros 28, M. R. Custodio. E. Hajdu. G. Lobo-Hajdu. and G. Muricy, eds. Museu Nacional., Rio de Janeiro.

Cerrano, C, P. Sambolino, B. Calcinal., F. Azzini, and G. Bavestrello. 2007b. Growth of the massive morph of Cliona nigricans (Schmidt. 1862) (Porifera, Clionaidae). Ital. J. Zool. 74: 13-19.

Engel, A., and H. Stahlberg. 2002. Aquaglyceroporins: channel proteins with a conserved core, multiple functions, and variable surfaces. Int. Rev. Cytol. 215: 75-104.

Fenoglio, I., G. Martra, S. Coluccia, and B. Fubini. 2000. Possible role of ascorbic acid in the oxidative damage induced by inhaled crystalline silica particles. Chem. Res. Toxicol. 13:971-975.

Finn, R. N., and J. Cerda. 2015. Evolution and functional diversity of aquaporins. Biol. Bull. 229: 6-23.

Finn, R. N., F. Chauvigne, J. B. Hlidberg, C. P. Cutler, J. Cerda. 2014. The lineage-specific evolution of aquaporin gene clusters facilitated tetrapod terrestrial adaptation. PLoS One 9: el 136X6.

Gaino, E. 2011. Overview of porifera. Pp. 1-53 in: Fauna d'ltalia, Vol. XLVI: Porifera I., M. Pansini, R. Manconi, R. Pronzato, eds. Calderini. Bologna.

Garneau, A. P., G. A. Carpentier, A.-A. Marcoux, R. Frenette-Cotton, C. F. Simard, W. Remus-Borel, L. Caron, M. Jacob-Wagner, M. Noel, J. J. Powell et al. 2015. Aquaporins mediate silicon transport in humans. PLoS One 10: e0136149.

Giovine, M., S. Scarfi, M. Pozzolini, A. Penna, and C. Cerrano. 2013. Cell reactivity to different silica. Prog. Mol. Subcell. Biol. 54: 143-174.

Grabherr, M. G., B. J. Haas, M. Yassour, J. Z. Levin, D. A. Thompson, I. Amit, X. Adiconis, L. Fan, R. Raychovvdhury, Q. Zenger al. 2011. Full-length transcriptome assembly from RNA-Seq data without a reference genome. Nat. Biotechnol. 29: 644-652.

Green, F. H. Y., and V. Vallyathan. 1996. Pathologic responses to inhaled silica. Pp. 39-59 in Silica and Silica-Induced Lung Diseases, V. Castranova, V. Vallyathan, W. E. Wallace, eds. CRC Press, Boca Raton, FL.

Haeckel, E. 1872. Die Kalkschwamme: Eine Monographic G. Reimer. Berlin.

Hamilton, R. F., Jr, S. A. Thakur, and A. Holian. 2008. Silica binding and toxicity in alveolar macrophages. Free Radic. Biol. Med. 44: 1246-1258.

Heymann, J. B., and A. Engel. 1999. Aquaporins: phylogeny, structure, and physiology of water channels. Physiology 14: 187-193.

Junqueira, L. C. U., G. Bignolas, and R. R. Brentani. 1979. Picrosirius staining plus polarization microscopy, a specific method for collagen detection in tissue sections. Histochem. J. 11: 447-455.

Katoh, K., and D. M. Standley. 2013. MAFFT multiple sequence alignment software version 7: improvements in performance and usability. Mol. Biol.Evol. 30: 772-780.

Mukhopadhyay, R., H. Bhattacharjee, and B. P. Rosen. 2014. Aquaglyceroporins: generalized met alloid channels. Biochim. Biophys. Acta 1840: 1583-1591.

Muller, W. E., X. Wang, M. Wiens, U. Schlossmacher, K. P. Jochum, and H. C. Schroder. 2011. Hardening of bio-silica in sponge spicules involves an aging process after its enzymatic polycondensation: evidence for an aquaporin-mediated water absorption. Biochim. Biophys. Acta 1810: 713-726.

Musa-Aziz, R., L. M. Chen, M. F. Pelletier, and W. F. Boron. 2008. Relative CO,/N[H.sub.3] selectivities of AQP1. AQP4, AQP5, AmtB, and RhAG. Proc. Natl. Acad. Sci. U. S. A. 106: 5406-5411.

Pao, G. M., L.-F. Wu, K. D. Johnson, H. Hofte, M. J. Chrispeels, G. Sweet, N. N. Sandal., and M. H. Saier, Jr. 1991. Evolution of the MIP family of integral membrane transport proteins. Mol. Microbiol. 5: 33-37.

Park, J. H., and M. H. Saier, Jr. 1996. Phylogenetic characterization of the MIP family of transmembrane channel proteins. J. Membr. Biol. 153: 171-180.

Pozzolini, M., F. Bruzzone, V. Berilli, F. Mussino, C. Cerrano, U. Benatti, and M. Giovine. 2012. Molecular characterization of a nonfibrillar collagen from the marine sponge Chondrosia reniformis Nardo, 1847 and positive effects of soluble silicates on its expression. Mar. Biotechnol. 14: 281-293.

Pozzolini, M., S. Scarfi, F. Mussino, S. Ferrando, L. Gallus, and M. Giovine. 2015. Molecular cloning, characterization, and expression analysis of a Prolyl 4-Hydroxylase from the marine sponge Chondrosia reniformis. Mar. Biotechnol. 17: 393-407.

Price, K., and C. Linge. 1999. The presence of melanin in genomic DNA isolated from pigmented cell lines interferes with successful polymerase chain reaction: a solution. Melanoma Res. 9: 5-9.

Rasband, W. S. 1997-2014. ImageJ. U. S. National Institutes of Health, Bethesda, MD. Available: http://imagej.nih.gov/ij/ [3 June, 2015].

Riesgo, A., N. Farrar, P. J. Windsor, G. Giribet, and S. P. Leys. 2014. The analysis of eight transcriptomes from all Porifera classes reveals surprising genetic complexity in sponges. Mol. Biol. Evol. 31: 1102-1120.

Saper, C. B. 2009. A guide to the perplexed on the specificity of antibodies. J. Histochem. Cytochem. 57: 1-5.

Saper, C. B., and P. E. Sawchenko. 2003. Magic peptides, magic antibodies: guidelines for appropriate controls for immunohistochemistry. J. Comp. Neurol. 465: 161-163.

Scarfi, S., M. Magnone, C. Ferraris, M. Pozzolini, F. Benvenuto, U. Benatti, and M. Giovine. 2009. Ascorbic acid pre-treated quartz stimulates TNF-[alpha] release in RAW 264.7 murine macrophages through ROS production and membrane lipid peroxidation. Respir. Res. 10: 25.

Teragawa, C. K. 1986a. Sponge dermal membrane morphology: histology of cell-mediated particle transport during skelet al growth. J. Morphol. 190: 335-347.

Teragawa, C. K. 1986b. Particle transport and incorporation during skeleton formation in a keratose sponge Dysidea etheria. Biol. Bull. 170: 321-334.

UniProt Consortium. 2015. UniProt: a hub for protein information. Nucleic Acids Res. 43: D204-D212.

Vandesompele, J., K. De Preter, F. Pattyn, B. Poppe, N. Van Roy, A. De Paepe, and F. Speleman. 2002. Accurate normalization of real-time quantitative RT-PCR data by geometric averaging of multiple internal control genes. Genome Bio3: research0034.

Wu, B., and E. Beitz. 2007. Aquaporins with selectivity for unconventional permeants. Cell. Mol. Life Sci. 64: 2413-2421.

Zardoya, R. 2005. Phylogeny and evolution of the major intrinsic protein family. Biol. Cell. 97: 397-414.

MARINA POZZOLINI (1,*), SARA FERRANDO (1), LORENZO GALLUS (1), CHIARA GAMBARDELLA (2), STEFANO GHIGNONE (3), AND MARCO GIOVINE (1)

(1) Department of Earth, Environmental and Life Science (DISTAV), University of Genoa, Corso Europa 26, 16132, Genoa, Italy; (2) Institute of Marine Sciences (ISMAR) - CNR, Via De Marini 6, 1-16149, Genoa, Italy; and (3) Institute for Sustainable Plant Protection (IPSP, Turin Unit) - CNR, V. le P. A. Mattioli 25, 1-10125, Turin, Italy

Received 27 July 2015; accepted 12 May 2016.

(*) To whom correspondence should be addressed. E-mail: marina.pozzolini@unige.it

Abbreviations: AQP, aquaporin; Bas, basopinacocytes; Ch, choancythes; CrAQP, aquaporin isolated in Chondrosia reniformis; DAPI, nuclear fluorescent dye 4',6-diamidino-2-phenylindole; ESEM, environmental scanning electron microscope; GAPDH, glyceraldehyde 3-phosphate dehydrogenase; GLPs, glycerol uptake facilitators or aquaglyceroporins; MIPs, major intrinsic proteins; NIPs, Nodulin 26-like intrinsic proteins; NPA, highly conserved signature of Asn-Pro-Ala; Pin, pinacocytes; PIPs. plasma membrane intrinsic proteins; qPCR, real-time polymerase chain reaction; TIPs, tonoplast intrinsic proteins; XIPs, X-intrinsic proteins.

Table 1
Sequence of oligonucleotides used as primers for PCR reactions

Name                Sequence (5'-3')               Position

GeneRacer
sense primer
 FwAQ1 (forward)    5'-TTGGGA CTGA TGGGTA TGGT-3'   219-238 (*)
Full-length
amplification
primers
 FwAQch (forward)   5'-AGAGGATTTATGGGTCGCTAA-.V       9-29 (*)
 RevAQch (reverse)  5'-CTTCTCCCTGGGTGCGTTT-3'       792-810 (*)
Real-time primers
 Fgapd (forward)    5'-AAGCCACCATCAAGAAGG-3'        882-899 ([dagger])
 Rgapd (reverse)    5'-CCACCAGTTTCACAAAGC-3'       1023-1040 ([dagger])
 FAQch (forward)    5'-TTGTGTTTGTCAGTGTCTTG-3'      127-146 (*)
 RAQch (reverse)    5'-TGTCAGCCAGTTGATACG-3'        309-326 (*)

GeneRacer, Life Technologies Corp., Carlsbad, CA.
(*) with respect to the KR780752 sequence; [dagger] with respect to the
KM217385 sequence.
PCR. polymerase chain reaction.


----------

Please note: Some tables or figures were omitted from this article.
COPYRIGHT 2016 University of Chicago Press
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2016 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Pozzolini, Marina; Ferrando, Sara; Gallus, Lorenzo; Gambardella, Chiara; Ghignone, Stefano; Giovine,
Publication:The Biological Bulletin
Article Type:Report
Date:Jun 1, 2016
Words:7438
Previous Article:Complexity of yolk proteins and their dynamics in the sea star Patiria miniata.
Next Article:Localization of phosphoproteins within the barnacle adhesive interface.
Topics:

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters