- Research article
- Open Access
Transcriptome and proteome analysis of Pinctada margaritifera calcifying mantle and shell: focus on biomineralization
© Joubert et al; licensee BioMed Central Ltd. 2010
- Received: 8 July 2010
- Accepted: 1 November 2010
- Published: 1 November 2010
The shell of the pearl-producing bivalve Pinctada margaritifera is composed of an organic cell-free matrix that plays a key role in the dynamic process of biologically-controlled biomineralization. In order to increase genomic resources and identify shell matrix proteins implicated in biomineralization in P. margaritifera, high-throughput Expressed Sequence Tag (EST) pyrosequencing was undertaken on the calcifying mantle, combined with a proteomic analysis of the shell.
We report the functional analysis of 276 738 sequences, leading to the constitution of an unprecedented catalog of 82 P. margaritifera biomineralization-related mantle protein sequences. Components of the current "chitin-silk fibroin gel-acidic macromolecule" model of biomineralization processes were found, in particular a homolog of a biomineralization protein (Pif-177) recently discovered in P. fucata. Among these sequences, we could show the localization of two other biomineralization protein transcripts, pmarg-aspein and pmarg-pearlin, in two distinct areas of the outer mantle epithelium, suggesting their implication in calcite and aragonite formation. Finally, by combining the EST approach with a proteomic mass spectrometry analysis of proteins isolated from the P. margaritifera shell organic matrix, we demonstrated the presence of 30 sequences containing almost all of the shell proteins that have been previously described from shell matrix protein analyses of the Pinctada genus. The integration of these two methods allowed the global composition of biomineralizing tissue and calcified structures to be examined in tandem for the first time.
This EST study made on the calcifying tissue of P. margaritifera is the first description of pyrosequencing on a pearl-producing bivalve species. Our results provide direct evidence that our EST data set covers most of the diversity of the matrix protein of P. margaritifera shell, but also that the mantle transcripts encode proteins present in P. margaritifera shell, hence demonstrating their implication in shell formation. Combining transcriptomic and proteomic approaches is therefore a powerful way to identify proteins involved in biomineralization. Data generated in this study supply the most comprehensive list of biomineralization-related sequences presently available among protostomian species, and represent a major breakthrough in the field of molluskan biomineralization.
- Silk Fibroin
- Pearl Oyster
- Whey Acidic Protein
- Nacreous Layer
- Biomineralization Process
Mollusk shell is a natural biomaterial made up of a mineral phase - calcium carbonate (CaCO3) - and an organic cell-free matrix (proteins, glycoproteins, lipids and polysaccharides) secreted by the external mantle epithelium, the tissue layer underlying the shell. Although this matrix represents less than 2% of the total composition of the shell by dry weight , it interacts with the crystal surface to orientate its nucleation and control crystal polymorphism, in the form of aragonite or calcite, in the different structural layers of the shell . The highly organized internal structure of the shell has led to a very interdisciplinary approach to the study of biomineralization. The secretion of shell by mollusks is one of the best examples of a matrix-mediated mineralization process achieved outside living tissues [3, 4]. Models of mollusk shell biomineralization have therefore been proposed based on histochemical studies and ultrastructural observations of the shell, combined with biochemical analysis of the extracellular organic matrix. The current "chitin-silk fibroin gel proteins-acidic macromolecules" model proposed by Levi-Kalisman et al. , updated by Addadi et al.  and recently reviewed by Furuhashi et al. , was established from mollusk nacre analysis and involves the major matrix components of the shell. According to this model, the major components of biomineralization are relatively hydrophobic silk proteins and a complex assemblage of hydrophilic proteins (many of which are unusually rich in aspartic acid), highly structured in a polysaccharide β-chitinous framework. These components of the organic matrix are thought to control various aspects of the biomineralization process: the CaCO3 crystal polymorphisms (calcite and aragonite) and the microstructures of shell layers . Since the publication of the first complete amino-acid sequence of a nacre-shell protein in 1996 , major advances in the field of molecular biology have led to the identification of an increasing number of shell matrix proteins . However, the molecular aspects of shell building are still far from being fully understood.
As marine bivalves are organisms of major economic interest, attention has been turned to the study of their genomics during the last decade . In particular, various sequence-based strategies have been developed for transcriptome studies. Among them, Expressed Sequence Tag (EST) sequencing programs have proven to be an effective method for gene discovery and have been widely used for initiating genomic research in non-model organisms . EST collections provide information on the part of the genome that is expressed, and can be valuable in a number of ways, e.g. gene fishing, genome annotation and analysis, discovery of single nucleotide polymorphisms (SNPs), and expression studies such as microarrays. An EST approach to biomineralization offers the opportunity to rapidly identify transcripts encoding secreted shell proteins, proteins specific to the pallial space and proteins implicated in calcium regulation in mantle cells, as well as transcription factors responsible for the regulation of the process. EST programs have recently been developed for aquaculture bivalve species, in particular the Eastern oyster (Crassostrea virginica) [12–15], the Pacific oyster (Crassostrea gigas) [14, 16], and the common blue mussel (Mytilus galloprovincialis) , but these have mainly been aimed at investigating the mollusk immune response in the context of environmental or genome evolution studies. To date, only five studies report the analysis of EST programs performed on calcifying tissues with the aim of providing more insight into the biomineralization process. Suppression subtractive hybridization (SSH) studies were performed on the bivalve pearl oysters Pinctada fucata and P. margaritifera. Two other studies, involving the vetigastropod Haliotis asinina[20, 21] and the bivalve pearl oyster Pinctada maxima, revealed the high complexity of the calcifying mantle transcriptome, suggesting extensive differences between Bivalvia and Gastropoda in the molecular composition of the organic matrix guiding the deposit of calcium carbonate polymorphs within the shell. The most recent study  described the transcriptome of the mantle tissue of Laturnela elliptica, focusing on the datamining of genes involved in calcium regulation and shell deposition. Despite these genomic approaches, there is still a small amount of genomic data available on bivalve species and this limits our understanding of the dynamic process of biomineralization.
With the aim of increasing the genomic resources for the pearl-producing bivalve P. margaritifera, we conducted a pyrosequencing program to analyze the first EST library produced from the calcifying mantle of this bivalve. Here we report the functional analysis of 276 738 EST sequences, leading to the constitution of a P. margaritifera mantle transcript catalog of 82 sequences potentially implicated in the biomineralization process. Further structural characterization of a set of proteins was undertaken in addition to transcript localization and proteomic mass spectrometry analysis of proteins isolated from the shell matrix. Our results show that protein repertoire of the biomineralization process is conserved within pearl oysters, but also provide direct evidence that our EST data set covers most of the diversity of the shell matrix protein in P. margaritifera shell.
1. Mantle RNA Extraction and ESTs library construction
P. margaritifera pearl oysters raised in the Vairao lagoon were brought to the Ifremer laboratory in Tahiti, French Polynesia. Total cellular RNA was extracted from 12 mantle samples taken from separate P. margaritifera individuals, using TRIZOL® Reagent (Life Technologies) according to manufacturer's recommendations. RNA integrity and purity were assessed in a Bioanalyzer 2100 (Agilent - Bonsai Technologies) and using agarose gel analysis. RNA was quantified using a NanoDrop® ND-1000 spectrophotometer (NanoDrop® Technologies Inc). A pool of 24 μg total RNA (2 μg per sample) was used to construct a cDNA library. Five μg of full-length double-stranded cDNA was processed by the standard Genome Sequencer library-preparation method using the GS DNA Library Preparation Kit to generate single-stranded DNA ready for emulsion PCR (emPCR™). The cDNA library was pyrosequenced using GS FLX technology (454/Roche, http://www.454.com/).
2. Contig assembly and functional annotation
EST sequence analysis and assembly were performed by the Skuldtech Company http://www.skuldtech.com. ESTs were assembled into clusters using TGICL (TIGR Gene Indices Clustering tools), freely available on the sourceforge website http://sourceforge.net/projects/tgicl/. Overlapping identity percentage and minimum overlapping length parameters was set to 98% and 60 bp, respectively, in order to obtain highly reliable consensus sequences. Data were archived at NCBI Sequence Read Archive (SRA) under accession SRP002635. ESTs that did not form contigs (singletons) and contigs resulting from the assembly of multiple sequences are referred to as unique sequences. These unique sequences were translated into six reading frames and used as a query to search the non-redundant protein databases available at the National Center for Biotechnology Information (NCBI) using the BlastX algorithm with an E-value ≤10-3 (version # 2.2.15, GenBank release number #166) http://www.ncbi.nlm.nih.gov. Sequences with BlastX hits were manually assigned to the following five sequence categories: known, uncharacterized, predicted, unknown or unnamed, and hypothetical proteins. This classification was based on the information definition lines in each homologous sequence provided by NCBI. All unique sequences with BlastX hits (E-value ≤10-3) were functionally annotated using Blast2GO http://www.blast2go.org/ by mapping against gene ontology (GO) resources.
3. Identification of biomineralization-related proteins in P. margaritifera mantle EST library
Candidate genes from the biomineralization process were locally identified in the P. margaritifera mantle ESTs library using BlastX, according to the following parameters: E-value ≤10-3, expect feature set to a default value of 10, and low-complexity filter determined by the SEG program . For this purpose, we collected all available sequences regarding biomineralization in mollusks (bivalvia and gastropoda) from the literature or from public databases. The Pmarg-Pif nucleotide sequence was obtained by assembling ESTs with an overlapping identity percentage and minimum overlapping length parameters set to 100% and 60 bp, respectively. Motifs and conserved domains of Pmarg-Pif protein sequence were used as a query to search the non-redundant protein databases available at the National Center for Biotechnology Information (NCBI) using the BlastP algorithm, according to the following parameters: expect feature set to a default value of 10, and low-complexity filter determined by the SEG program . Sequence alignments were performed using the ClustalW program setting parameters to default for the gap criterions (gap open, no gap end, gap extension, gap distance, pairgap), followed by manual correction with BioEdit software http://www.ebi.ac.uk/Tools/clustalw2/index.html. The presence of signal peptides was inferred using the SignalP 3.0 server http://www.cbs.dtu.dk/services/SignalP/. Conserved domains were identified using Prosite http://www.expasy.ch/prosite/. Percentage identity and biochemical similarity between sequences were calculated using ProtParam http://www.expasy.ch/tools/protparam.html. Repeat detection in protein sequences was performed using RADAR http://www.ebi.ac.uk/Tools/Radar/index.html.
4. In situ hybridization analyses
a) Tissue preparation
P. margaritifera mantle tissues were fixed for 24 h in Davidson fixative (22% formalin, 33% ethyl alcohol, 11.5% glacial acetic, 33% sterile sea water), embedded in paraffin wax, and serially sectioned at 7 μm. Sections were collected onto polylysine coated slides (Silane-prep™, Sigma- Aldrich), dried overnight at 60°C and treated with proteinase K (10 μg.mL-1) in TE buffer (Tris 50 mM, EDTA 10 mM) at 37°C for 25 min. Slides were then dehydrated by immersion in an ethanol series and air dried. The sections were prehybridized for 1 h at 42°C with 500 μL hybridization buffer (4 × SSC, 50% formamide, 1× Denhardt's solution, 250 μg.mL-1 yeast tRNA, 10% dextran sulfate). The solution was replaced with 120 μL of the same buffer, containing 6 μL of the digoxigenin-labeled sense or antisense probes. The slides were incubated overnight at 42°C for hybridization. The sections were washed twice for 5 min in 2× SSC at room temperature and once for 10 min in 0.4× SSC at 42°C. The detection steps were performed according to manufacturer's instructions (Dig nucleic acid detection kit, Roche Molecular Biomedicals). Slides were finally counter-stained with a solution of Bismark Brown Yellow and mounted in Eukitt. The slides were examined using a DM4000B Leica microscope.
b) Specific probe preparation
In order to synthesize probes for in situ hybridisation, we used the PeS4 (GACATAGAGAGAGACAGATATGA)/PeAS4 (ATTCACCATTTCCGTTACCGT) primer set, specific to the pmarg-pearlin ORF (265bp), and AspF1 (CTCTTACACCAAAATGAAGGGG)/AspR1 (TCCGTCATCATTATCTGC), specific to the pmarg-aspein transcript (253 bp). These primers (4 μM final volume) were used in PCR reactions with the iQ™ Supermix (BIO-RAD) and pmarg-pearlin full-length cDNA as template. After DNA denaturation at 94°C for 5 min, 35 cycles were run with an MJ-Research thermocycler as follows: 94°C for 30 s; 55°C for 30 s; 72°C for 45 s ended by a final elongation step at 72°C for 10 min. Probes (sense or antisense) were synthesized by asymmetric PCR (using the same amplification program) in the presence of Dig-dUTP (0.7 mM), in a PCR reaction mixture containing a unique primer (sense or antisense, 2 μM final volume), 2 μL of the previously purified PCR fragment (Mini Quick Spin Columns, Roche Diagnostics), a mix of dGTPs-dCTPs-dATPs (200 μM each final), dTTPs (130 μM final), and Taq polymerase (Promega, 2.5 u). Labelling efficiency was assayed using the DIG high prime DNA labelling kit (Roche Diagnostics).
5. Purification and identification of proteins from P. margaritifera shell
Organic matrix was extracted from fresh shells of P. margaritifera specimens aged 3-5 years, after acid acetic decalcification . The acido-insoluble matrix was digested with trypsin prior to reduction and alkylation . Samples were injected into a nano LC-nanoESI-MS/MS system for analysis. Mass spectrometry (MS) was performed using a nanoESI-qQ-TOF, and data acquired automatically using Analyst QS 1.1 software (Applied Biosystems). A 1 s TOF-MS survey scan was acquired over 400-1600 amu, followed by three 3 s product ion scans over a mass range of 65-2000 amu. The three most intense peptides, with a charge state of two to four above a 30 count threshold, were selected for fragmentation and dynamically excluded for 60 s with ± 50 mmu mass tolerance. The collision energy was set by the software according to the charge and mass of the precursor ion. The MS and MS/MS data were recalibrated using internal reference ions from a trypsin autolysis peptide at m/z 842.51 [M + H]+ and m/z 421.76 [M + 2H]2+. Protein identification was done using the Mascot database-searching software (Matrix Science, London, UK; version 2.2.04) using our database of the pyrosequencing-based EST mantle library from P. margaritifera. Carbamidomethylation and oxidation were set as fixed and variable modifications, respectively. The mass tolerance was set to 0.5 Da and the MS/MS tolerance to 0.2 Da.
During recent decades, high-throughput techniques have been used to examine a broad range of physiological processes and applications in diverse fields of biology [33, 34]. To examine the biomineralization process in pearl oyster P. margaritifera, we performed transcriptome pyrosequencing of its calcifying tissue combined with a proteome analysis of the shell.
1. Transcriptome analysis of P. margaritifera calcifying mantle
a) Generation of ESTs and contig assembly
Summary statistics for pyrosequencing and annotation of P. margaritifera mantle ESTs
Total number of ESTs sequenced
Average lenght of ESTs (bp)
Number of assembled EST
Number of contigs
Number of singletons
Number of unique sequences
Ratio of singletons per unique sequences
Number of contigs containing 2 ESTs
Number of contigs containing 3 ESTs
Number of contigs containing 4 ESTs
Number of contigs containing 5 ESTs
Number of contigs containing > 6 ESTs
Number of annotated unique sequences:
- Known protein
- Unknown, Unnamed
- Hypothetical protein
Number of annotated contigs
Number of annotated singletons
Of the 19 257 contigs, 8717 (45.3%) contained 2 ESTs, 3419 (17.8%) contained 3 ESTs, 1779 (9.2%) contained 4 ESTs, 1119 (5.8%) contained 5 ESTs, and 4223 (21.9%) contained more than 6 ESTs (Table 1). In our study, 79.2% of the 276 738 ESTs were successfully assembled and remaining singletons only represented 20.8% of the reads, and a large part (74.9%) of the 76 790 unique sequences was singletons. In other recent 454 transcriptome studies, results showed that the remaining singletons represented 10 to 40% of the reads [36, 37]. It has already been observed that many ESTs resulting from deep sequencing of transcriptomes with 454 sequencing technology fail to assemble . These unassembled singletons could result from sequencing errors, contaminants from other sources, or can even from technical difficulties in assembling with overlaps that are too short in length or which contain highly repeated sequences. Interestingly, however, these singletons can also represent rare transcripts of genes expressed at low levels , and therefore constitute an interesting source of genomic data.
b) Putative identities of ESTs
BlastX searches of the 76 790 unique sequences in the non-redundant protein databases available at the National Center for Biotechnology Information (NCBI) revealed 29 479 (38.4%) significant matches (E-value ≤10-3). Among these 29 479 matches, 13 064 (44.3%) are known proteins, but 6010 are uncharacterized (20.4%), 4795 are predicted (16.3%), 2880 are either unknown or unnamed (9.8%), and 2730 are hypothetical proteins (9.3%) (Table 1). This apparently low rate of identification is common among mollusk EST databases, with which rates usually range from 15 to 40% [15, 17, 22, 40], although this is lower than for vertebrates , or even EST collections from model plants .
Although the lack of annotation can result from the difficulty of annotating some short length sequences, it can largely be explained by the lack of sequences available for mollusk species, and by the fact that a vast majority of genes on public databases come from taxa (in particular vertebrates species) whose amino acid sequences show great divergence with those of protostomians.
c) Functional Gene Ontology annotation
The distribution of the sequences between specialized terms in the binding section of the molecular function category showed that the greatest numbers fell under protein-binding (35%) and nucleotide-binding (19%). Interestingly, the third greatest number of the binding section fell into ion-binding (17%) (Figure 1D). Biomineral crystal matrix macromolecules play a key role in biologically-controlled biomineralization processes. In vitro crystallization experiments, microscopic and analytical methods revealed stereochemical properties of matrix proteins, which allow them to bind calcium ions and calcium carbonate, and therefore perform framework building and crystal growth during the construction of the molluskan shell[45–48]. A significant proportion of sequences in our mantle EST collection are implicated in binding, and particularly in ion binding. This result is consistent with observations from a previous study performed on the calcifying mantle of the bivalve L. elliptica. We therefore hypothesize that this classification could be a pattern typical of tissues of a secretory nature implicated in biomineralization processes.
2. Identification of transcripts encoding proteins involved in the biomineralization process of P. margaritifera
a) Identification of a catalogue of 82 proteins potentially involved in the biomineralization process
To obtain an integrated view of the transcriptional events of the biomineralization process in P. margaritifera mantle, we made BlastX searches with our EST mantle library focusing on proteins known to be involved in these mechanisms. For this purpose, we first collected all available sequences regarding biomineralization in calcifying invertebrates from the literature or from public databases. In mollusks, we found 140 bivalve and 103 gastropod proteins potentially implicated in biomineralization processes. These 243 molluskan sequences were isolated from shell or mantle tissue in previous studies, using either biochemical or molecular biology approaches. BlastX searches of the 140 bivalves and 103 gastropods proteins in our EST database revealed 121 and 56 significant matches (E-value ≤10-3), respectively. Analyzing these 177 sequences together with sequences from our EST library, we identified 82 P. margaritifera non-redundant unique sequences potentially implicated in the biomineralization process. Among these, 69 and 13 sequences could be recovered by homology with sequences from bivalve and gastropod, respectively.
Among the 69 unique P. margaritifera transcripts that were recovered by homology with the bivalve sequences, 55 sequences were obtained by homology with sequences from the Pinctada genus (Additional file 1). The overall identity percentage between P. margaritifera protein sequences potentially implicated in the biomineralization process and protein sequences from the Pinctada genus is ranging from 24% (C-type lectin 2 from P. fucata) to 95% (Ferritin-like protein from P. fucata). This level of identity is similar to percentages already observed for homolog proteins from the N66/Nacrein and N14/N16 families [49–51]. The N66 sequence from P. maxima and Nacrein sequence from P. fucata (P. maxima N44 homolog sequence) displayed identity percentages of 82% and 69%, respectively, with P. margaritifera homolog sequence. Similarly, the N14 sequence from P. maxima and N16 sequence from P. fucata displayed an identity percentage of 93% and 71% respectively with P. margaritifera homolog sequence, Perline matrix protein. Considering all sequences from the Pinctada genus, the identity percentage seems to be higher between P. margaritifera and P. maxima sequences than between P. margaritifera and P. fucata sequences.
Extending our analysis to biomineralization proteins from other bivalves led us to the identification of the 14 remaining sequences out of the 69 unique P. margaritifera transcripts that were recovered by homology with the bivalve sequences (Additional file 1). The overall identity percentage between P. margaritifera protein sequences potentially implicated in the biomineralization process and protein sequences from the other bivalves ranges from 28% (EP protein precursor from Mytilus edulis) to 58% (bone morphogenic protein type 2 receptor from Crassostrea gigas). This level of identity is lower than that observed between proteins within the Pinctada genus, except for proteins implicated in calcium regulation or signal transduction. For example, Calmodulin sequences from Hyriopsis schlegelii (genbank accession number: ACI22622) displayed an identity percentage of 99% with the P. margaritifera homolog sequence.
Finally, we identified 13 P. margaritifera unique sequences by homology with sequences of gastropod (Additional file 1). The overall identity percentage between P. margaritifera protein sequences potentially implicated in the biomineralization process and protein sequences from gastropods ranged from 27% (Veliger mantle 1 from H. asinina) to 100% (Calmodulin from Conus cuneolus). Interestingly, some sequences homologous to abalone (H. laevigata) proteins could be found in our EST database, namely Perlucin [52, 53], Perlustrin [53, 54] and Perlawpin  from Haliotis laevigata. Perlucin, Perlustrin and Perlwapin sequences were obtained by direct protein sequencing of proteins purified from the nacreous layer of abalone shell. All of the P. margaritifera homolog sequences for each of these 3 proteins found in the P. margaritifera EST library display the same motif and numerous conserved cystein positions as in the sequences from H. laevigata. Perlucin is a 155-amino acid protein which exhibits similarities with calcium dependent lectins (C-type). The P. margaritifera homolog sequence for Perlucin (Pmarg- perlucin) is not a complete sequence. However, of the 6 cysteins present in the abalone sequence, 3 are conserved between Pmarg- perlucin and Perlucin sequences. Moreover, Pmarg- perlucin displays an E-value of 9.00E-9 and an identity percentage of 38% (27/71 a.a.) with Perlucin and also has a C-type lectin domain. Perlustrin is a small protein (84 a.a.) with similarities to vertebrate insulin-like growth factor-binding protein (IGF-BP) sequences. The P. margaritifera homolog sequence for Perlustrin (Pmarg-perlustrin) is a complete 142-amino acid sequence with an E-value of 7.00E-6, and 39% (25/64 a.a.) identity with Perlustrin; it also exhibits a insulin-like growth factor binding proteins (IGFBPs). On the 12 cysteins scattered across the Perlustrin sequence, 11 (of the 14 cysteins of Pmarg-perlustrin) are conserved between Pmarg-perlustrin and the Perlustrin sequences. Finally, the Perlwapin protein consists of 134 amino acids that contain 3 repeats of 40 amino acids very similar to the well-known whey acidic protein (WAP) domains. The P. margaritifera homolog sequence for Perlwapin (Pmarg-perlwapin) is a complete 139-amino acid (a.a.) sequence with an E-value of 2.00E-11, 37% identity (40/107 a.a.) with Perlwapin, and two WAP domains. Out of the 25 cysteins spread along the Perlwapin sequence, all 14 cysteins of Pmarg-perlwapin are conserved between the Pmarg-perlwapin and Perlwapin sequences. These results would suggest that Perlucin, Perlustrin and Perlwapin are present in P. margaritifera. Previous studies have shown that there are significant differences in the molecular mechanisms in different mineralizing species and, therefore, between the proteins they use. Such differences may even exist among species that are phylogenetically very close, like the Mollusca. The cause of this "evolvability" remains a controversy, and it is still uncertain whether the biomineralization "molecular tool box" required for shell construction is inherited from an ancestral function, or whether this ability is the result of an adaptive convergence. Recent studies have explicitly demonstrated that shell or skeletal proteins had evolved independently among metazoans [8, 21, 56]. However, the identification of homolog proteins between bivalvia and gastropoda could support the idea that at least some of the shell component could have appeared early in the evolution of the molluscan phylum.
Taken together, this candidate approach allowed us to isolate 82 unique sequences potentially implicated in the biomineralization process in P. margaritifera. This study considerably increases the amount of transcriptomic data available in this field, making P. margaritifera the best documented marine protostomian with regard to biomineralization.
b) Identification of proteins from the "chitin-silk fibroin gel-acidic macromolecule" model
Mollusk shell construction is the result of biologically-controlled mineralization, a highly dynamic process mediated by an extracellular organic matrix secreted by the mantle epithelium . Histochemical studies and ultrastructural observations of the shell, together with biochemical analysis of the extracellular organic matrix, provided a better understanding of shell structure and led to the identification of proteins composing it, thereby allowing mollusk shell biomineralization models to be developed. The currently accepted "chitin-silk fibroin gel-acidic macromolecule" model involves the major matrix components of the shell, i.e. relatively hydrophobic silk proteins plus a complex assemblage of hydrophilic proteins (many of which are unusually rich in aspartic acid), highly structured in a polysaccharide β-chitinous framework .
In our study, beyond the consideration of protein homologies between species, it is interesting to note that our P. margaritifera EST mantle library includes sequences coding for proteinaceous components of the matrix following this model. Firstly, a sequence showing 78% identity with MSI60 from the silk fibroin matrix component could be retrieved. MSI60 is an insoluble framework protein purified from the nacreous layer of the shell  and expressed in the more dorsal region of the mantle . Poly-Ala and poly-Gly blocks conferring MSI60 homologies with spider silk fibroins are present in the P. margaritifera homologous sequence. MSI31  and Shematrins , displaying silk/fibroin-like domains, could be also retrieved. Secondly, a sequence showing 87% identity with the unusually acidic protein Aspein from P. fucata could be recovered in the EST database from P. margaritifera. This sequence homologous to Aspein is the first extremely acidic shell protein identified in P. margaritifera. In P. fucata, Aspein is specifically expressed in the mantle region, which secretes the calcite prism matrix . The main body of this protein includes a high proportion of Asp (60.4%) punctuated with Ser-Gly dipeptides, which are conserved in the P. margaritifera homologous sequence. Finally, recent electron microscopy studies on nacre have detected the presence of chitin in the shell of P. margaritifera, and chitin synthase gene has been cloned from P. fucata, Atrina rigida and Mytilus galloprovincialis. A P. margaritifera homolog sequence of chitin synthase from this species could be retrieved, revealing that chitin synthase sequences are well conserved among bivalves. More precisely, the chitin synthase sequences from Atrina rigida and Mytillus galloprovincialis displayed identity percentages of 91% and 84%, respectively, with the homologous P. margaritifera sequence.
Taken together, searches realized on the EST mantle library allowed us to identify proteinaceous components of the calcifying matrix from P. margaritifera. These results demonstrate how EST-based studies are a powerful way of dramatically increasing knowledge about proteins implicated in the biomineralization process, which constitutes an important prerequisite for establishing relevant biomineralization models.
c) Pmarg-Pif encodes an homolog of Pif-177 from P. fucata, a protein involved in nacre formation
Taken together, the numerous conserved sequence motifs, conserved cystein residue positions, charged amino acid residue composition and common isoelectric properties between Pmarg- Pif-97 and Pif-97 support the hypothesis that Pmarg-Pif might have a similar activity to Pif-177, and regulate nacre formation in P. margaritifera. However, the presence of the repeated 18 amino acid residues sequence specific to Pmarg- Pif-80 and the distinct number of repeats of the four-amino-acid motif (DD-R/K-R/K) between Pmarg- Pif-80 and Pif-80 also suggest that Pmarg- Pif-80 might have a function specific to P. margaritifera. Considering these features, further research needs to be undertaken in order to investigate Pmarg- Pif function and its role in the biomineralization process.
3. Expression pattern of biomineralization-related protein transcripts
4. Mantle transcripts encode proteins identified in P. margaritifera shell
Using the P. margaritifera EST mantle library, identification of shell matrix proteins was attempted by a complementary proteomic approach. The shell matrix proteins, extracted from decalcified shell powder, were digested with trypsin and the resulting peptides were analysed by MS/MS mode mass spectrometry. The raw MS/MS data were directly interrogated against the EST data set using Mascot software. After careful observation of the MS/MS data on the 50 first most intensive peptides, we estimated that almost all the main peptides analysed led to contig identification. We only considered matching proteins that presented at least 2 unambiguously identified peptides, i.e. those presenting individual scores superior to the threshold (calculated value of 32).
Protein identification in the shell matrix of P. margaritifera by a proteomic approach.
0.0 1 e-123
1 e-30 7
Our proteomic analysis enabled us to retrieve in silico all the sequences from P. margaritifera involved in the biomineralization process already published on databases in our peptide library, and we were also able to find a match in our database for all proteins experimentally found from P. margaritifera shell in our EST library. These results demonstrate that our EST data set covers most of the diversity of the matrix protein of the P. margaritifera shell.
This global approach combining transcriptome and proteome analysis of P. margaritifera calcifying mantle and shell is the first description of a pyrosequencing program performed on a pearl-producing bivalve species. It led to the functional analysis of 276 738 EST sequences, with the constitution of a P. margaritifera mantle transcripts catalog of 82 sequences potentially implicated in the biomineralization process. Our results showed that the biomineralization protein repertoire is conserved within pearl oysters, but also provided direct evidence that our EST data set covered most of the diversity of P. margaritifera shell matrix protein. These observations clearly demonstrate the high efficiency of this pyrosequencing-based EST library in accurately identifying shell proteins, in combination with shotgun proteomic analysis and automated database searches. These data represent the most comprehensive list of biomineralization-related sequences available among protostomian species, and represent a major breakthrough in the field of molluskan biomineralization.
This study is part of a collaborative project (GDR ADEQUA) supported by the "Service de la perliculture" of French Polynesia. It is also supported by Ifremer, Skuldtech and University of French Polynesia. Authors are grateful to Frédéric Marin, Marcel Le Pennec, Alexandre Tayalé, Florentine Riquet, Cédrik Lo and Anne-Sandrine Talfer for helpful discussions and assistance.
- Weiner S: Organization of extracellularly mineralized tissues: a comparative study of biological crystal growth. CRC Crit Rev Biochem. 1986, 20 (4): 365-408. 10.3109/10409238609081998.PubMedView ArticleGoogle Scholar
- Falini G, et al: Control of Aragonite or Calcite Polymorphism by Mollusk Shell Macromolecules. Science. 1996, 271 (5245): 67-69. 10.1126/science.271.5245.67.View ArticleGoogle Scholar
- Mann S: Biomineralization: principles and concepts in bioinorganic materials chemistry. 2001, Oxford University Press, 198:Google Scholar
- Rousseau M, et al: Dynamics of sheet nacre formation in bivalves. J Struct Biol. 2008, 165 (3): 190-5. 10.1016/j.jsb.2008.11.011.PubMedView ArticleGoogle Scholar
- Levi-Kalisman Y, et al: Structure of the nacreous organic matrix of a bivalve mollusk shell examined in the hydrated state using cryo-TEM. J Struct Biol. 2001, 135 (1): 8-17. 10.1006/jsbi.2001.4372.PubMedView ArticleGoogle Scholar
- Addadi L, et al: Mollusk shell formation: a source of new concepts for understanding biomineralization processes. Chemistry. 2006, 12 (4): 980-7. 10.1002/chem.200500980.PubMedView ArticleGoogle Scholar
- Furuhashi T, et al: Molluscan shell evolution with review of shell calcification hypothesis. Comp Biochem Physiol B Biochem Mol Biol. 2009Google Scholar
- Marin F, et al: Molluscan shell proteins: primary structure, origin, and evolution. Curr Top Dev Biol. 2008, 80: 209-76. full_text.PubMedView ArticleGoogle Scholar
- Miyamoto H, et al: A carbonic anhydrase from the nacreous layer in oyster pearls. Proc Natl Acad Sci USA. 1996, 93 (18): 9657-60. 10.1073/pnas.93.18.9657.PubMed CentralPubMedView ArticleGoogle Scholar
- Saavedra C: Bivalve genomics. Aquaculture. 2006, 256 (1-4): 1-14. 10.1016/j.aquaculture.2006.02.023.View ArticleGoogle Scholar
- Pi C, et al: Analysis of expressed sequence tags from the venom ducts of Conus striatus: focusing on the expression profile of conotoxins. Biochimie. 2006, 88 (2): 131-40. 10.1016/j.biochi.2005.08.001.PubMedView ArticleGoogle Scholar
- Jenny MJ, et al: Potential indicators of stress response identified by expressed sequence tag analysis of hemocytes and embryos from the American oyster, Crassostrea virginica. Mar Biotechnol (NY). 2002, 4 (1): 81-93. 10.1007/s10126-001-0072-8.View ArticleGoogle Scholar
- Peatman E: Development of Expressed Sequence Tags from Eastern Oyster (Crassostrea virginica): Lessons learnd from previous efforts. Mar Biotechnol (NY). 2004, 491-496. 6Google Scholar
- Tanguy A, Guo X, Ford SE: Discovery of genes expressed in response to Perkinsus marinus challenge in Eastern (Crassostrea virginica) and Pacific (C. gigas) oysters. Gene. 2004, 338 (1): 121-31. 10.1016/j.gene.2004.05.019.PubMedView ArticleGoogle Scholar
- Tanguy A, et al: Increasing genomic information in bivalves through new EST collections in four species: development of new genetic markers for environmental studies and genome evolution. Gene. 2008, 408 (1-2): 27-36. 10.1016/j.gene.2007.10.021.PubMedView ArticleGoogle Scholar
- Gueguen Y, et al: Immune gene discovery by expressed sequence tags generated from hemocytes of the bacteria-challenged oyster, Crassostrea gigas. Gene. 2003, 303: 139-45. 10.1016/S0378-1119(02)01149-6.PubMedView ArticleGoogle Scholar
- Craft JA, et al: Pyrosequencing of Mytilus galloprovincialis cDNAs: tissue-specific expression patterns. PLoS One. 2010, 5 (1): e8875-10.1371/journal.pone.0008875.PubMed CentralPubMedView ArticleGoogle Scholar
- Liu HL, et al: Identification and characterization of a biomineralization related gene PFMG1 highly expressed in the mantle of Pinctada fucata. Biochemistry. 2007, 46 (3): 844-51. 10.1021/bi061881a.PubMedView ArticleGoogle Scholar
- Duplat D, et al: Identification of calconectin, a calcium-binding protein specifically expressed by the mantle of Pinctada margaritifera. FEBS Lett. 2006, 580 (10): 2435-41. 10.1016/j.febslet.2006.03.077.PubMedView ArticleGoogle Scholar
- Jackson DJ, et al: A rapidly evolving secretome builds and patterns a sea shell. BMC Biol. 2006, 4: 40-10.1186/1741-7007-4-40.PubMed CentralPubMedView ArticleGoogle Scholar
- Jackson DJ, et al: Parallel evolution of nacre building gene sets in molluscs. Mol Biol Evol. 2010, 27 (3): 591-608. 10.1093/molbev/msp278.PubMedView ArticleGoogle Scholar
- Clark MS, et al: Insights into shell deposition in the Antarctic bivalve Laternula elliptica: gene discovery in the mantle transcriptome using 454 pyrosequencing. BMC Genomics. 2010, 11: 362-10.1186/1471-2164-11-362.PubMed CentralPubMedView ArticleGoogle Scholar
- Pertea G: TIGR Gene Indices clustering tools (TGICL): a software system for fast clustering of large EST datasets. Bioinformatics. 2003, Oxford, England, 19 (5): 651-2. 10.1093/bioinformatics/btg034.Google Scholar
- Conesa A: Blast2GO: a universal tool for annotation, visualization and analysis in functional genomics research. Bioinformatics. 2005, Oxford, England, 21 (18): 3674-6. 10.1093/bioinformatics/bti610.Google Scholar
- Federhen JCWS: Statistics of local complexity in amino acid sequences and sequence databases. Computers in Chemistry. 1993, 17: 149-163. 10.1016/0097-8485(93)85006-X.View ArticleGoogle Scholar
- Hall TA: BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucleic acids symposium series. 1999, 41: 95-98.Google Scholar
- Bendtsen JD, et al: Improved prediction of signal peptides: SignalP 3.0. J Mol Biol. 2004, 340 (4): 783-95. 10.1016/j.jmb.2004.05.028.PubMedView ArticleGoogle Scholar
- de Castro E: ScanProsite: detection of PROSITE signature matches and ProRule-associated functional and structural residues in proteins. Nucleic Acids Res. 2006, W362-5. 10.1093/nar/gkl124. 34 Web ServerGoogle Scholar
- Gasteiger EHC, Gattiker A, Duvaud S, Wilkins MR, Appel RD, Bairoch A: Protein Identification and Analysis Tools on the ExPASy Server, in The Proteomics Protocols Handbook. 2005, J.M.W.H.P. Inc., Humana Press Totowa, NJ, 561-607.Google Scholar
- Heger A, Holm L: Rapid automatic detection and alignment of repeats in protein sequences. Proteins. 2000, 41 (2): 224-37. 10.1002/1097-0134(20001101)41:2<224::AID-PROT70>3.0.CO;2-Z.PubMedView ArticleGoogle Scholar
- Marie B, et al: The shell matrix of the freshwater mussel Unio pictorum (Paleoheterodonta, Unionoida). Involvement of acidic polysaccharides from glycoproteins in nacre mineralization. FEBS J. 2007, 274 (11): 2933-45. 10.1111/j.1742-4658.2007.05825.x.PubMedView ArticleGoogle Scholar
- Marie B, et al: Evolution of nacre: biochemistry and proteomics of the shell organic matrix of the cephalopod Nautilus macromphalus. Chembiochem. 2009, 10 (9): 1495-506. 10.1002/cbic.200900009.PubMedView ArticleGoogle Scholar
- Margulies M, et al: Genome sequencing in microfabricated high-density picolitre reactors. Nature. 2005, 437 (7057): 376-80.PubMed CentralPubMedGoogle Scholar
- Zagrobelny M, S -AK, Bjerg Jensen N, Lindberg Moller B, Gorodkin J, Bak S: 454 pyrosequencing based transcriptome analysis of Zygaena filipendulae with focus on genes involved in biosynthesis of cyanogenic glucosides. BMC Genomics. 2009, 10 (1): 574-10.1186/1471-2164-10-574.PubMed CentralPubMedView ArticleGoogle Scholar
- Cheung F, et al: Sequencing Medicago truncatula expressed sequenced tags using 454 Life Sciences technology. BMC Genomics. 2006, 7: 272-10.1186/1471-2164-7-272.PubMed CentralPubMedView ArticleGoogle Scholar
- Cheung F, et al: Analysis of the Pythium ultimum transcriptome using Sanger and Pyrosequencing approaches. BMC Genomics. 2008, 9: 542-10.1186/1471-2164-9-542.PubMed CentralPubMedView ArticleGoogle Scholar
- Meyer E, et al: Sequencing and de novo analysis of a coral larval transcriptome using 454 GSFlx. BMC Genomics. 2009, 10: 219-10.1186/1471-2164-10-219.PubMed CentralPubMedView ArticleGoogle Scholar
- Trombetti GA, et al: Data handling strategies for high throughput pyrosequencers. BMC Bioinformatics. 2007, 8 (Suppl 1): S22-10.1186/1471-2105-8-S1-S22.PubMed CentralPubMedView ArticleGoogle Scholar
- Vera JC, et al: Rapid transcriptome characterization for a nonmodel organism using 454 pyrosequencing. Mol Ecol. 2008, 17 (7): 1636-47. 10.1111/j.1365-294X.2008.03666.x.PubMedView ArticleGoogle Scholar
- Venier P, et al: MytiBase: a knowledgebase of mussel (M. galloprovincialis) transcribed sequences. BMC Genomics. 2009, 10: 72-10.1186/1471-2164-10-72.PubMed CentralPubMedView ArticleGoogle Scholar
- Patil DP, et al: Generation, annotation, and analysis of ESTs from midgut tissue of adult female Anopheles stephensi mosquitoes. BMC Genomics. 2009, 10: 386-10.1186/1471-2164-10-386.PubMed CentralPubMedView ArticleGoogle Scholar
- Weber AP, et al: Sampling the Arabidopsis transcriptome with massively parallel pyrosequencing. Plant Physiol. 2007, 144 (1): 32-42. 10.1104/pp.107.096677.PubMed CentralPubMedView ArticleGoogle Scholar
- Ashburner M, et al: Gene ontology: tool for the unification of biology. The Gene Ontology Consortium. Nat Genet. 2000, 25 (1): 25-9. 10.1038/75556.PubMed CentralPubMedView ArticleGoogle Scholar
- Quilang J, et al: Generation and analysis of ESTs from the eastern oyster, Crassostrea virginica Gmelin and identification of microsatellite and SNP markers. BMC Genomics. 2007, 8: 157-10.1186/1471-2164-8-157.PubMed CentralPubMedView ArticleGoogle Scholar
- Addadi L, Weiner S: Interactions between acidic proteins and crystals: stereochemical requirements in biomineralization. Proc Natl Acad Sci USA. 1985, 82 (12): 4110-4. 10.1073/pnas.82.12.4110.PubMed CentralPubMedView ArticleGoogle Scholar
- Yan Z, et al: Biomineralization: functions of calmodulin-like protein in the shell formation of pearl oyster. Biochim Biophys Acta. 2007, 1770 (9): 1338-44.PubMedView ArticleGoogle Scholar
- de Paula SM, Silveira M: Microstructural characacterization of shell components in the mollusc Physa sp. Scanning. 2005, 27 (3): 120-5. 10.1002/sca.4950270303.PubMedView ArticleGoogle Scholar
- Kong Y, et al: Cloning and characterization of Prisilkin-39, a novel matrix protein serving a dual role in the prismatic layer formation from the oyster Pinctada fucata. J Biol Chem. 2009, 284 (16): 10841-54. 10.1074/jbc.M808357200.PubMed CentralPubMedView ArticleGoogle Scholar
- Kono M, Hayashi N, Samata T: Molecular mechanism of the nacreous layer formation in Pinctada maxima. Biochem Biophys Res Commun. 2000, 269 (1): 213-8. 10.1006/bbrc.2000.2274.PubMedView ArticleGoogle Scholar
- Miyashita T, et al: Identical carbonic anhydrase contributes to nacreous or prismatic layer formation in Pinctada fucata (Mollusca Bivalvia). Veliger. 2002, 45 (3): 250-255.Google Scholar
- Miyamoto H, Yano M, Miyashita T: similarities in the structure of nacrein, the shell-matrix protein, in a bivalve and a gastropod. j Mollusc Stud. 2003, 69: 87-89. 10.1093/mollus/69.1.87.View ArticleGoogle Scholar
- Mann K, et al: The amino-acid sequence of the abalone (Haliotis laevigata) nacre protein perlucin. Detection of a functional C-type lectin domain with galactose/mannose specificity. Eur J Biochem. 2000, 267 (16): 5257-64. 10.1046/j.1432-1327.2000.01602.x.PubMedView ArticleGoogle Scholar
- Weiss IM, et al: Purification and characterization of perlucin and perlustrin, two new proteins from the shell of the mollusc Haliotis laevigata. Biochem Biophys Res Commun. 2000, 267 (1): 17-21. 10.1006/bbrc.1999.1907.PubMedView ArticleGoogle Scholar
- Weiss IM, et al: Perlustrin, a Haliotis laevigata (abalone) nacre protein, is homologous to the insulin-like growth factor binding protein N-terminal module of vertebrates. Biochem Biophys Res Commun. 2001, 285 (2): 244-9. 10.1006/bbrc.2001.5170.PubMedView ArticleGoogle Scholar
- Treccani L, et al: Perlwapin, an abalone nacre protein with three four-disulfide core (whey acidic protein) domains, inhibits the growth of calcium carbonate crystals. Biophys J. 2006, 91 (7): 2601-8. 10.1529/biophysj.106.086108.PubMed CentralPubMedView ArticleGoogle Scholar
- Livingston BT, et al: A genome-wide analysis of biomineralization-related proteins in the sea urchin Strongylocentrotus purpuratus. Dev Biol. 2006, 300 (1): 335-48. 10.1016/j.ydbio.2006.07.047.PubMedView ArticleGoogle Scholar
- Sudo S, et al: Structures of mollusc shell framework proteins. Nature. 1997, 387 (5 june): 563-564. 10.1038/42391.PubMedView ArticleGoogle Scholar
- Takeuchi T, Endo K: Biphasic and dually coordinated expression of the genes encoding major shell matrix proteins in the pearl oyster Pinctada fucata. Mar Biotechnol (NY). 2006, 8 (1): 52-61. 10.1007/s10126-005-5037-x.View ArticleGoogle Scholar
- Yano M, et al: Shematrin: a family of glycine-rich structural proteins in the shell of the pearl oyster Pinctada fucata. Comp Biochem Physiol B Biochem Mol Biol. 2006, 144 (2): 254-62. 10.1016/j.cbpb.2006.03.004.PubMedView ArticleGoogle Scholar
- Tsukamoto D, Sarashina I, Endo K: Structure and expression of an unusually acidic matrix protein of pearl oyster shells. Biochem Biophys Res Commun. 2004, 320 (4): 1175-80. 10.1016/j.bbrc.2004.06.072.PubMedView ArticleGoogle Scholar
- Nudelman F, et al: Forming nacreous layer of the shells of the bivalves Atrina rigida and Pinctada margaritifera: an environmental- and cryo-scanning electron microscopy study. J Struct Biol. 2008, 162 (2): 290-300. 10.1016/j.jsb.2008.01.008.PubMedView ArticleGoogle Scholar
- Suzuki M, Sakuda S, Nagasawa H: Identification of chitin in the prismatic layer of the shell and a chitin synthase gene from the Japanese pearl oyster, Pinctada fucata. Biosci Biotechnol Biochem. 2007, 71 (7): 1735-44. 10.1271/bbb.70140.PubMedView ArticleGoogle Scholar
- Weiss IM, et al: The chitin synthase involved in marine bivalve mollusk shell formation contains a myosin domain. FEBS Lett. 2006, 580 (7): 1846-52. 10.1016/j.febslet.2006.02.044.PubMedView ArticleGoogle Scholar
- Suzuki M, et al: An acidic matrix protein, Pif, is a key macromolecule for nacre formation. Science. 2009, 325 (5946): 1388-90. 10.1126/science.1173793.PubMedView ArticleGoogle Scholar
- Jolly C, et al: Zona localization of shell matrix proteins in mantle of Haliotis tuberculata (Mollusca, Gastropoda). Mar Biotechnol (NY). 2004, 6 (6): 541-51. 10.1007/s10126-004-3129-7.View ArticleGoogle Scholar
- Samata T, et al: A new matrix protein family related to the nacreous layer formation of Pinctada fucata. FEBS Lett. 1999, 462 (1-2): 225-9. 10.1016/S0014-5793(99)01387-3.PubMedView ArticleGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.