Ci-AMBP : a highly conserved member of the microglobulin superfamily of proteinase inhibitors in grass carp , Ctenopharyngodon idellus

Jianming Su, Hongyu Lei, Tiaoyi Xiao and Shuliang Cui* College of Veterinary Medicine, Hunan Agricultural University, No.1 Nongda Road, Furong District, Changsha, Hunan 410128, China College of Animal Science and Technology, Hunan Agricultural University, No.1 Nongda Road, Furong District, Changsha, Hunan 410128, China Australian Phenomics Facility, John Curtin School of Medical Research, Australian National University, 131 Garran Road, Acton ACT 2601, Australia

α-1-Microglobulin (α1m, also designated protein HC) (Åkerström and Lögdberg, 1990) is a plasma glycoprotein (Escribano et al., 1991) that is present in plasma as a proteinase inhibitor of the macroglobulin superfamily, as described in rat (Falkenberg et al., 1990).αlm belongs to the lipocalin superfamily, a group of distantly related and similarly folded proteins that transport small, lipophilic molecules (Igarashi et al., 1992).Bikunin is a serine proteinase inhibitor possessing tandemly arranged inhibitory domains of the Kunitz type (Gebhard and Hochstrasser, 1986;Salier, 1990), and is found in plasma in both free form and complexed via glycosaminoglycan bonds to large polypeptides, the heavy chains of the inter-α-inhibitor (IαI) family (Enghild et al., 1989;Salier, 1990).Bikunin and IαI are thought to have a role in endothelial cell growth and in stabilizing the cumulus extracellular matrix of early ovulated oocytes (Salier, 1990;Chen et al., 1992), but this remains largely unexplored.α1m and bikunin are derived from the α1-microglobulin/bikunin precursor (AMBP) polypeptide and are encoded by the same gene, ambp; they are co-expressed and co-translated as a fusion protein in animals (Yoshida et al., 1999) and released by cleavage of a connecting dipeptide (Kaumeyer et al., 1986).The cloning and characterization of AMBP has been reported in cat, human, pig, mouse and rat (Kaumeyer et al., 1986;Gebhard et al., 1990;Lindqvist et al., 1992;Nakata et al., 2011).
In this study, the gene encoding AMBP was isolated from a suppression subtractive hybridization (SSH) cDNA library of grass carp, Ctenopharyngodon idellus.It was named ci-ambp, and thus encodes Ci-AMBP.Full-length cDNA encoding Ci-AMBP was isolated from liver extracts of grass carp by reverse transcription-polymerase chain reaction (RT-PCR) and rapid amplification of cDNA ends (RACE).The cDNA and its encoded amino acid sequence were analyzed to characterize its molecular property and conservation by comparison with homologs from other species, which suggested a conserved tandem two-domain genomic organization and molecular structure, and a common evolutionary origin of the protein.The analysis showed strong molecular evidence of conserved biological function of the protein, which was supported by gene expression in tissue/organs.
Juvenile carps were transferred from pools and raised in an aquarium for a week, and then one group were intraperitoneally injected with Aeromonas hydrophila (1.2 × 10 8 cells) while the other group of experimental fish were injected with ordinary liquid broth.Intestine and liver tissue samples were taken from both groups at 24 h after injection, rinsed with sterile DEPC water, and stored in liquid nitrogen for later use.Other tissues were collected and stored in the same manner.
Total RNA was extracted from animal tissue using TRIzol Reagent (Invitrogen).An aliquot of total RNA was applied to the PolyATtract mRNA Isolation System (Promega) for mRNA purification.The mRNA samples were quantified using a BioPhotometer (Eppendorf), and their quality was checked by gel electrophoresis.
The subtractive cDNA library was constructed from SSH mRNAs of intestine tissue using the PCR-Select cDNA Subtraction Kit (Clontech).The cDNA samples reverse-transcribed from mRNAs of infected and noninfected intestines were used to prepare double-stranded Tester cDNA and Driver cDNA, respectively.Two Tester cDNAs were prepared using Adaptor-1 and Adaptor-2R for two rounds of hybridization with Driver cDNA followed by two rounds of universal PCR amplification.Products from the second PCR were ligated into pUCmT vector (MBI), and cloned by transformation of competent Escherichia coli DH5α to establish the library.Single colonies were picked to prepare plasmid DNA for sequencing.The sequences were used to search the EST database in GenBank, revealing a 541bp cDNA fragment (GC-B6) coding for a 125-amino acid polypeptide that shares high sequence similarity with a conserved Kunitz domain of AMBP (GenBank EST accession number EW688152).
RACE was performed in both directions using the SMART cDNA amplification kit (Clontech) to amplify the full-length cDNA of the ci-ambp gene from liver mRNA.mRNA was converted into RACE-Ready cDNA by the BD PowerScript reverse transcriptase (Clontech) at 42 °C for 1.5 h.The synthesized first-strand cDNAs were used as templates to amplify 5'-end (UPM, long and P1) and 3'-end (UPM, short and P2) fragments of the ci-ambp gene according to the manufacturer's protocol (Clontech).The PCR reactions were catalyzed by a mixture of LA Taq (TaKaRa) and AmpliTaq (Clontech).The amplification conditions were 85 °C for 2 min, followed by 20 cycles at 94 °C for 10 s, 65 °C for 30 s and 72 °C for 3 min, with a final incubation at 72 °C for 7 min.The cloning and sequencing of the PCR products were conducted as described previously (Cui et al., 2005).
Nucleotide sequencing was performed by Huada Gene (BGI) using the Dye Primer Sequenase Kit (Perkin-Elmer), and electrophoresis was on an ABI 373 A DNA Sequencer (Perkin-Elmer).
The putative amino acid sequence of Ci-AMBP was converted from ci-ambp cDNA sequence using an online tool (http://biophp.org/minitools/dna_to_protein/demo.php).Both nucleotide sequence and amino acid sequence were used for homology searching of databases, such as GenBank and EMBL, using the computer programs BLASTN and BLASTP (http://www.ncbi.nlm.nih.gov/BLAST/).The Ci-AMBP protein sequence was submitted to the web-based computer program SignalP 3.0 to identify the signal peptide.
The brain, eyes, gills, heart, head kidney, kidney, hepatopancreas, spleen, intestine, muscle and blood were collected from juvenile grass carps.Three replicates of cDNA were synthesized from each sample of total RNA.cDNA fragments of the ci-ambp gene were amplified by PCR using first-strand cDNA mixed with 4 mM MgCl 2 , 0.2 mM each dNTP, and 0.8 mM of ci-ambpF and ci-ambpR, in a final volume of 50 µl in 1 × PCR buffer.PCR reactions were incubated at 94 °C for 3 min and then subjected to 35 cycles of 94 °C for 30 s for denaturation, 55 °C for 40 s for annealing and 72 °C for 45 s for extension.
The PCR amplicons, as bands in agarose gels, along with β-actin amplified from the same tissues, were analyzed by ImageJ (Abràmoff et al., 2004) to measure the abundances of amplified DNA as integrated density.The abundances of amplified DNA from tissues/ organs were analyzed using one-way analysis of variance (ANOVA).The analysis was carried out using GraphPad Prism software (GraphPad Software) and the results were plotted as mean density with standard error.
Using a subtractive cDNA library constructed from the intestinal tract of grass carp experimentally infected with A. hydrophila, a subtraction rate of ~210 -fold was obtained judged by subtracted PCR, using the internal housekeeping gene β-actin, with a size range of 250 to 1,500 bp.These selected cDNAs were successfully cloned into plasmid vector pUCm-T to establish a subtracted PCR fragment library, from which positive clones were randomly selected and sequenced, including a clone termed GC-B6 (GenBank EST accession number EW688152) that showed a specific hit with the serine protease inhibitor bikunin of the Kunitz (KU) superfamily by BLAST searching.Bikunin is typically encoded by the ambp gene and transcribed in tandem with the α1m coding sequence.The sequence information was used to design grass carp-specific primers (P1 and P2, Table 1) for full-AMPB in Ctenopharyngodon idellus length cDNA cloning by the SMART RACE technique.
A full-length ci-ambp cDNA (1,230 nucleotides) and its encoded Ci-AMBP protein sequence (348 amino acid residues) were identified and submitted to GenBank (accession number EU186151).Ci-AMBP includes α1m and two KU domains of bikunin.
SignalP revealed a signal peptide at the N terminus with a probable cleavage site between A18 and G19, meaning that Ci-AMBP is a secretory protein (Fig. 1).The region between M39 and G184 encodes the lipocalin motif of the α1-microglobulin followed by the two domains of bikunin, kunitz inhibitor 1 and kunitz inhibitor 2, with multiple binding sites for inter-alpha-trypsin inhibitor light chain and trypstatin.Five helixes are found, with one between the signal region and the first coiled lipocalin domain, and four in the C-terminal bikunin coils.
Ci-AMBP showed the highest similarity at both DNA (80% identity) and protein (93% identity) levels to zebrafish AMBP, followed by rainbow trout AMBP (61% and 73% identity at the DNA and protein levels, respectively) (Table 2).Although Ci-AMBP shared less than 50% nucleotide sequence identity with AMBPs from other species analyzed, the identities at the protein level were all at least 50% (Table 2).Sequence alignment showed highly conserved signaling cores for lipocalin and two trypsin interaction sites of KU domains even with mammals, with which Ci-AMBP shared less sequence similarity (data not shown).
ci-ambp gene expression in tissues/organs was assessed by RT-PCR using gene-specific primers that generated a 379-bp product, along with the internal housekeeping gene β-actin as a control.ci-ambp transcripts were detected in all tissues/organs investigated, including brain, eyes, gills, heart, head kidney, hepatopancreas, spleen, intestine, muscle and the blood.The RT-PCR products were normalized using the expression abundance of β-actin, and plotted as a density index, which showed that ciambp was most highly expressed in spleen followed by head kidney, kidney and hepatopancreas, with the lowest expression in the muscle and gills (Fig. 2).
The expression of AMBP in tissues/organs has been studied in various species at the transcript and protein levels.The proteins are considered to be synthesized mainly in the liver, secreted, and transported to other tissues across the endothelial cell membrane (Berggård et al., 1998).The mature α1m and bikunin proteins are co-expressed as a precursor protein, and then separated by cleavage before their secretion (Bratt et al., 1993;Bratt et al., 1994), which implies that the two proteins act in common biological processes.However, only α1m has been extensively studied as an immunoregulator in the blood, while the biological functions of these proteins together remain largely unknown.The expression profile will provide the primary information about where and when these proteins might act (Grubb et al., 1983;Pervaiz and Brew, 1985;Falkenberg et al., 1990;Lindqvist et al., 1992).
The restricted expression of ambp mRNA in a few organs contrasts to a widespread and unique distribution of each of the two proteins.ambp transcript was found in the rat liver and pancreas by PCR (Lindqvist et al., 1992;Berggård et al., 1998) as well as in liver parenchymal cells in X. laevis, where it is activated around embryo hatching and repressed at the metamorphic climax stage, but the mRNA level remained low during adult life (Kawahara et al., 1997).AMBP is down-regulated by thyroid hormone in X. laevis during early development and in larval hepatocytes in primary culture, but its repressed expression Fig. 2. Expression pattern of the gene coding for Ci-AMBP.Expression was assessed by RT-PCR using cDNA templates converted from mRNA samples isolated from various grass carp tissues.A pair of Ci-AMBP-specific primers (ci-ambpF and ci-ambpR) was designed to amplify a fragment of 379 bp (between nucleotides 533 and 911) of the gene (top panel), with a specific primer pair for β-actin used as an internal control (middle panel) and total RNA (bottom panel) and H 2 O (rightmost lanes) as negative controls.The amplified PCR products were analyzed by gel electrophoresis, and the same amount of cDNA template was used in both Ci-AMBP and β-actin amplifications (A).The Ci-AMBP band densities were analyzed and converted into density index using ImageJ, and the digitized indexes were normalized to those of β-actin and plotted using the software GraphPad Prism for one-way ANOVA (B). in adult hepatocytes in culture is not affected by the hormone, which implies a role in the progression of amphibian metamorphosis (Kawahara et al., 1997).Although ambp is transcribed mainly in the rat liver and pancreas, studies have shown an expanded tissue expression pattern in other species.Using in situ hybridization, ambp mRNA was found to be strongly expressed in liver, pancreas and intestine, and weakly expressed in the umbilical cord vessels and developing vertebral bodies and kidney, in day 8.5-15.5 mouse embryos (Sánchez et al., 2002).Furthermore, microarray analysis detected ambp transcripts in the skin of zebrafish infected with the bacterium Citrobacter freundii in a study of innate immunity-associated proteins and different acute-phase proteins, one of which is α1m, suggesting a role for α1m in innate immune responses to bacterial infection (Lü et al., 2012).On the other hand, AMBP and its mature products, α1m and bikunin, are widely distributed in tissues/organs.The translated proteins were found in the microsomal fraction of rat liver homogenates (Lindqvist et al., 1992) and in developing hepatocytes, pancreas, kidney and gut of embryos (Sánchez et al., 2002).AMBP was also found in myocytes while bikunin was detected in cardiac muscle, nervous system microvasculature and connective tissue.Both proteins were found in brain mesenchyme and meninges (Sánchez et al., 2002).In addition to blood, the rat α1m is present in most tissues, including liver, heart, eye, kidney, lung, pancreas and skeletal muscle.α1m was detected by Western blot using perfused and homogenized rat tissue supernatants in its free, monomeric form and in high molecular weight forms, which had previously been reported in plasma (Grubb et al., 1983;Berggård et al., 1997), and non-active isoforms found in the liver, heart and kidney are not present in plasma (Berggård et al., 1998).The presence of intracellular α1m in hepatocytes and in the proximal epithelial cells of the kidney was detected by immunohistochemistry, with immunoreactivity found in the heart, lung and blood vessels as well as on cell surfaces of cardiocytes (Berggård et al., 1998).Additional sites of protein distribution were found that do not correlate to mRNA localization, supporting the notion that the protein is transported to its sites of action.Semi-quantitative RT-PCR was used to determine the Ci-AMBP gene expression pattern in this study.ci-ambp transcripts were detected in all the grass carp tissues/ organs examined, including brain, eyes, gills, heart, head kidney, kidney, hepatopancreas, spleen, intestine, muscle and the blood.However, the statistical plotting of the normalized expression index from these tissues showed that the most abundant expression is in the spleen, head kidney, kidney and hapatopancreas, followed by the heart, eyes and brain.Expression is at a low level in the blood and hard tissues (Fig. 2).The high level of expression in the more active organs appears to support its role in the immune response to infection and stress.Detailed expression of the gene and its protein distribution are to be investigated.

Fig. 1 .
Fig. 1.Molecular properties of Ci-AMBP.The protein sequence converted from its coding cDNA was subjected to comparative and functional bioinformatic analysis.A signal peptide was predicted at the N terminus by SignalP 3.0 between position 1 and 18, and highly conserved domains and motifs of predicted biologically active sites containing tandem functional domains of lipocalin (position 39 to 184) and two KU domains (position 225 to 278; 281 to 334), separated by an RARR protease cleavage site (position 200 to 203), were identified by UniProtKB in the Eukaryotic Linear Motif resource.

Table 2 .
Ci-AMBP identity with other species