Skip Navigation


International Immunology Advance Access originally published online on October 11, 2006
International Immunology 2006 18(12):1671-1680; doi:10.1093/intimm/dxl101
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
18/12/1671    most recent
dxl101v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (5)
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Zhu, Y.
Right arrow Articles by Ding, J. L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Zhu, Y.
Right arrow Articles by Ding, J. L.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?


© The Japanese Society for Immunology. 2006. All rights reserved. For permissions, please e-mail: journals.permissions@oxfordjournals.org

Diversity in lectins enables immune recognition and differentiation of wide spectrum of pathogens

Yong Zhu1,3, Patricia M. L. Ng1, Lihui Wang1,4, Bow Ho2 and Jeak Ling Ding1

1 Department of Biological Sciences, National University of Singapore, 14 Science Drive 4, Singapore 117543
2 Department of Microbiology, National University of Singapore, 5 Science Drive 2, Singapore 117597
3 Present address: Institute of Molecular and Cell Biology, Proteos, 61 Biopolis Drive, Singapore 138673
4 Present address: Department of Biochemistry, Oxford University, South Parks Road, Oxford, OX1 3QU, UK

Correspondence to: J. L. Ding; E-mail: dbsdjl{at}nus.edu.sg


   Abstract
Top
 Abstract
Introduction
 Methods
 Results
 Discussion
 Supplementary data
 References
 
Carbohydrate-binding lectins play essential roles as pattern recognition receptors in innate immunity in both vertebrates and invertebrates. The carcinolectins 5 (CL5a and CL5b, the CL5 isoforms of horseshoe crab, Carcinoscorpius rotundicauda, with apparent sizes of 36 and 40 kDa, respectively) are prominent plasma lectins that bind all representative microbes and pathogen-associated molecular pattern molecules. Different cDNA isoforms of both CL5a and CL5b were isolated, leading to our speculation on their functional divergence. Characterization of CL5 isoforms bound to microbial cell surfaces demonstrates the diversity of these lectins. The resolution patterns of the isoforms that associate with fungus differ from those that associate with bacteria, suggesting the unique roles these lectins play in the recognition and differentiation of microbes. We postulate that different populations of plasma lectins act in collaboration in frontline innate immune defense against disparate pathogens. The functional diversity of lectins in invertebrates appears to evolutionarily compensate for the lack of acquired immunity.

Keywords: functional diversity, innate immunity, lectin isoforms, pathogen recognition


   Introduction
Top
 Abstract
 Introduction
Methods
 Results
 Discussion
 Supplementary data
 References
 
Central to the defense system in both vertebrates and invertebrates is the rapid self–non-self differentiation by innate immune molecules. Innate immunity relies on germline-encoded pattern recognition receptors (PRRs) to recognize conserved and common pathogen-associated molecular patterns (PAMPs) displayed on the pathogen cell surface (1, 2). Peptidoglycan of bacteria, LPS of Gram-negative bacteria, lipoteichoic acid (LTA) of Gram-positive bacteria and ß-glucan of fungi are widely accepted PAMP molecules. The carbohydrate moieties in the PAMP molecules, with distinctive composition and configuration, are binding targets for lectins, which are universally deployed in animal hosts.

In the vertebrates, collectins and ficolins bind to oligosaccharide structures of PAMP molecules, leading to the activation of complement pathways where pathogens are phagocytosed and lysed (3–5). This is representative of the innate immune defense prior to the mounting of acquired immunity. In contrast, the invertebrates rely entirely on innate immunity to discriminate and clear invading micro-organisms. To compensate for the lack of acquired immunity, the invertebrates are well equipped with an arsenal of PRRs, the functional counterparts of immunoglobulins (Igs). The best understood example of invertebrate PRRs is peptidoglycan recognition proteins (PGRPs) that have been intensively studied in the fruit fly and silkworm, using both biochemical and genetic approaches [see an excellent review by Dziarski (6) and references cited therein]. Another well-known example of invertebrate PRRs is the zymogen factor C in the coagulation cascade of horseshoe crab, which is an extremely sensitive LPS sensor (7). However, lectin-type PRRs in invertebrates are relatively less understood in comparison to PGRPs. A number of lectins that recognize foreign elicitors of the immune response have been biochemically characterized in the cockroach (8–11) and the horseshoe crab [extensively reviewed by Iwanaga (12) and Kawabata and Tsuda (13)]. In the fruit fly, a C-type lectin (14) and a member of the selectin-type lectin (15) have been characterized. In the invertebrates, recognition of pathogens by PRRs leads to the initiation of various kinds of defense actions such as phagocytosis, activation of prophenoloxidase cascade that generate melanin to contain the pathogens and the initiation of signal pathways, which boost synthesis and release of anti-microbial peptides and superoxide anions (7, 16–18).

As a living fossil with >500 million years of evolutionary success, the horseshoe crab is an ideal model species to study the evolution of innate immune mechanisms. While the fruit fly is ideally suited for genetic manipulations, the horseshoe crab is a good invertebrate model organism for biochemical analysis (7). A repertoire of lectins has been identified in the horseshoe crab Tachypleus tridentatus (12, 13). Tachylectins (TL) 1–4, which exhibit diverse in vitro ligand specificities, are confined to the hemocytes and their in vivo function remains unknown. Two closely related plasma lectins (TL5a and TL5b), which are homologous to fibrinogen-like lectin, ficolin, were purified from the plasma using N-acetyl group-conjugated resin (19). The crystal structure of TL5a complexed with N-acetyl glucosamine (GlcNAc) has been solved, demonstrating homology between TL5a and ficolin (20). Recently, we have shown that the carcinolectins 5, CL5a and CL5b (counterparts of TL5a and TL5b, respectively) in the horseshoe crab (Carcinoscorpius rotundicauda), are the major proteins that bind all representative microbes and initiate the activation of a novel complement-like system, leading to the phagocytosis of pathogens (21).

In this study, characterization of CL5 cDNA sequences led to our speculation of the functional divergence of CL5, which called for further investigation of their binding to various microbes. By two-dimensional (2D) gel electrophoresis, the microbe-bound CL5s were resolved into different isoforms. The resolution pattern of CL5s bound to a marine fungus, Kluyveromyces marxianus, is different from those bound to bacteria species (Gram-negative Escherichia coli and Gram-positive Staphylococcus aureus). Our results demonstrate that different pools of CL5 isoforms are invoked to recognize and differentiate bacteria and fungi, thus fulfilling the requirements of innate immunity against a wide spectrum of microbes.


   Methods
Top
 Abstract
 Introduction
 Methods
Results
 Discussion
 Supplementary data
 References
 
Cell-free plasma
Horseshoe crabs (C. rotundicauda) were collected from the Kranji estuary, Singapore. The animals were bled by cardiac puncture using an 18-G needle. Cell-free plasma was obtained by centrifugation (100 x g for 5 min at 4°C) to remove hemocytes. The animals were handled according to national and institutional guidelines stipulated by the National Advisory Committee for Laboratory Animals Research, Singapore.

Isolation of CL5a and CL5b cDNA isoforms
A pair of degenerate reverse transcription (RT)–PCR primers for CL5a were designed for the isolation of CL5a cDNAs, based on the peptide sequences identified by tandem mass spectroscopy (MS)/MS analysis (20): forward primer 5'-ATG GA(C/T)AA(C/T)GA(C/T)AA(C/T)GGIGG(A/C/G/T)TGGAC-3' for MDNDNGGWTLL and reverse primer 5'-(C/G)(A/T)(A/G)TA(A/G)AA(A/G)TAIGT (A/C/G/T)CC(C/T)TC-3' for EGTYFYSLPENR. However, degenerate RT–PCR using this pair of primers on the cardiac cDNA templates resulted in the non-specific amplification of a TL5b homolog sequence, CL5b-C1. Thereafter, 5'-rapid amplification of cDNA ends (RACE) primer CL5-R1 (5'-GCATTGTACCACCAACCTCCTTTGTA-3') and 3'-RACE primer CL5-F1 (5'-GGGTGGACGCTAATACAAAGACGTGGA-3', covering residues YKGGWW and GWTLIQRRG, respectively, that are most conserved across species in TL5s and CL5b-C1 (see Fig. 1), were used to isolate cDNA sequences of both CL5a and CL5b. However, the 3'-RACE using CL5-F1 primer failed to amplify any product, possibly due to mismatch at the 3' anchor site of the 3'-RACE primed with CL5a cDNA sequences. This was demonstrated by comparison of the 5'-RACE-isolated sequences with the 3'-RACE primer, CL5-F1. Thus, a CL5a-specific 3'-RACE primer, CL5a-F2 (5'-ATCGTCTTCATATTGGCAACTACAGTG-3'), was designed for the isolation of 3' cDNA sequences of CL5a, based on the 5'-RACE-identified CL5a cDNA sequences.


Figure 1
View larger version (36K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 1 The identification of a CL5b fragment sequence by degenerate RT–PCR and design of the RACE primers for the cDNA amplification of both CL5a and CL5b. Degenerate RT–PCR using a pair of primers designed (see Methods) based on CL5a peptides MDNDNGGWTLL and EGTYFYSLPENR (20), with the cardiac cDNAs as template, non-specifically amplified a TL5b homologous sequence, (A) CL5b-C1. Capital letters denote non-coding sequences. Small letters denote coding sequences. (B) The alignment of derived amino acid sequence with both TL5a and TL5b. Oligonucleotides located in the region most conserved across species (indicated by arrows) were employed as RACE primers to isolate the full-length cDNAs of both CL5a and CL5b at lower stringency.

 
TRIZOL® RNA extraction reagent from GIBCO BRL (Gaithersburg, MD, USA) and Thermoscript RT–PCR system (Invitrogen, Carlsbad, CA, USA) were used for RNA purification and cDNA synthesis, respectively. Using SMARTTM RACE kit (BD Biosciences, Mountain View, CA, USA), the 5' and 3' ends of CL5a were amplified from the hepatopancreas, and those of CL5b were amplified from the cardiac tissues. The amplified fragments were cloned into TA cloning vector, pGEM-T® easy (Promega, Madison, WI, USA) and sequenced.

Analysis of CL5s bound to various PAMP molecules
Five hundred micrograms of each kind of the PAMP molecules, E. coli LPS, S. aureus LTA or Saccharomyces cerevisiae zymosan (Sigma, St Louis, MO, USA), was incubated with 1.5 ml of horseshoe crab plasma supplemented with complete protease inhibitor cocktail (Roche, Basel, Switzerland) in an ice bath for 15 min. The complex of PAMP–lectins was pelleted at 20 000 x g for 15 min. The pellets were washed twice in saline by re-suspension and centrifugation. The pellets were extracted with 200 µl of 5% SDS and boiled for 5 min after re-suspension. After removal of insoluble particulates by centrifugation at 20 000 x g for 15 min, the protein extracts were precipitated with cold acetone at –30°C for 2 h. After centrifugation at 20 000 x g for 15 min, the protein pellets were air-dried and dissolved in SDS-PAGE loading buffer, boiled for 5 min and analyzed. The protein extract from 50 µg each of plasma treated with PAMP was loaded per lane.

Analysis of CL5s bound to various microbes
Staphylococcus aureus (ATCC25923) and E. coli (ATCC25922) were cultured in Mueller Hinton broth at 37°C. Kluyveromyces marxianus was cultured in YPD medium (2% glucose, 2% peptone and 1% yeast extract, adjusted to pH 5.6) at 30°C. Mid-log phase microbial cells were washed three times by repeated centrifugation at 10 000 x g for 1 min and re-suspended in saline (150 mM NaCl). Washed cells were re-suspended at 5 U of OD600 nm. Ten volumes of plasma, supplemented with complete protease inhibitor cocktail from Roche, were mixed with microbes in suspension and the mixtures were gently shaken at room temperature for 15 min. the cells were pelleted, washed three times and re-suspended in saline at 5 U of OD600 nm and directly extracted with 1 volume of 2 x SDS-PAGE loading buffer and boiled for 3 min. After removing the insoluble debris by centrifugation, the protein samples were analyzed by SDS-PAGE, or further treated to remove SDS prior to 2D gel analysis. Protein extracts derived from 0.25 ml each of microbial cells/plasma were loaded per lane for 1D gel analysis. To remove SDS from samples, the proteins were precipitated with 5 volumes of ice-cold acetone and incubated at –30°C for 2 h. After centrifugation at 20 000 x g for 15 min, the protein pellets were re-suspended in 2% 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonate (CHAPS) from Sigma and precipitated again. The protein preparations were then dissolved in 2D gel sample buffer and analyzed using 7-cm immobilised pH gradient (IPG) strips, pH 5-8 (Bio-Rad Laboratories, Hercules, USA). The protein extract from 0.5 ml each of microbial cells/plasma was loaded onto a 7-cm IPG strip (pH 5–8) and isofocused. Focused strips were laid on top of polyacrylamide gels (10%) for the second dimension SDS-PAGE. All gels were stained with Coomassie Blue R-250. The recovered CL5 spots were in-gel digested with trypsin as described (22) and the extracted peptide samples were analyzed by MALDI-tof-tof using 4700 Proteomics Analyzer (Applied Biosystems, Foster City, CA, USA). The uninterpreted tandem MS/MS ion spectra were subjected to MASCOT database search at http://www.matrixscience.com. Selected peaks that were common in CL5a or CL5b isoform spots were sequenced by manual interpretation.


   Results
Top
 Abstract
 Introduction
 Methods
 Results
Discussion
 Supplementary data
 References
 
CL5 cDNA isoforms
Degenerate primers were designed based on the peptide sequences derived from tandem MS/MS analysis of CL5a (21) to amplify CL5a cDNA by RT–PCR. However, this effort isolated a CL5b cDNA 3'-fragment sequence instead, which is named CL5b-C1 (Fig. 1A). Thereafter, based on the alignment of the derived amino acid sequence with those of TL5a and TL5b (Fig. 1B), a pair of RACE primers located at the most conserved regions was designed with a view to isolate both the CL5a and CL5b cDNA isoforms. By this approach, the polymorphism of CL5s was discovered.

Both 5'- and 3'-RACE using cardiac cDNA templates yielded CL5b fragment sequences. Figure 2 shows the derived amino acid sequences together with CL5a fragment sequences described later in a multiple alignment. The six different 5'-fragment sequences can be divided into two groups (the longer CL5b-N1, CL5b-N2 and CL5b-N3 as group 1 and the shorter CL5b-N4, CL5b-N5 and CL5b-N6 as group 2) with insertions/deletions among the two groups, while only point mutations/replacements are found among members within each group. Only two 3'-sequences (CL5b-C1 and CL5b-C2) with small differences were identified by 3'-RACE. Thus, it appears that the C-terminus of CL5b is much more conserved than the N-terminus. However, it could also be explained by the greater diversity at the C-terminus; thus, the primer used failed to amplify more diversified sequences. Except for CL5b-N4 and CL5b-C1, which exactly overlap with each other over ~240 bp (likely to be transcribed from the same gene), all other sequences are not identical in the region of overlapping sequences, and are thus transcribed from different genes. Altogether, these cDNAs infer up to a minimum of seven different CL5b genes, but there are probably more of other sequences not primed by either the 3'- or 5'-RACE primer used.


Figure 2
View larger version (87K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 2 Multiple alignment of derived CL5 partial protein sequences and their homologs, TL5a and TL5b. CL5b sequences were isolated by either 3'-RACE (CL5b-C1 and -C2) or 5'-RACE (CL5b-N1 to -N6) with cardiac tissue cDNA preparation as template; CL5a sequences were isolated by either 5'-RACE (CL5a-N1 to -N4) or 3'-RACE (CL5a-C1 and -C3) with hepatopancreas cDNA preparation as template (see Methods). The multiple alignments of these sequences together with their homologs from the Japanese horseshoe crab were produced using ClustalW (23) and manually checked (see supplementary table 1 for database accession numbers of these sequences, available at International Immunology Online). The secretion signal peptides (boxed) were predicted by SignalP 3.0 (24). Amino acid residues in TL5a that are involved in Ca2+ binding are marked with asterisks, and those residues involved in the GlcNAc binding are marked with diamonds. The conserved cysteine residues engaged in the formation of intra-chain disulfide bonds are also indicated. Residues identical in all sequences are shaded black, while conserved residues are shaded gray. Tandem MS/MS analysis (see Fig. 5 in the later section) showed peptide sequences that were located in the derived protein sequences. These sequences are underlined (red for CL5a sequences and blue for CL5b sequences).

 
Four 5'-fragment sequences (CL5a-N1 to -N4) and two 3'-fragment sequences (CL5a-C1 and -C3) were isolated by RACE using the hepatopancreas cDNA as template (see Fig. 2 for the derived amino acid sequences). No overlapping 3'-sequences were amplified for CL5a-N1 and -N2. However, CL5a-N3 and -N4 can fully overlap with the 3'-sequences for ~100 bp. Taken together, these CL5a cDNA sequences theoretically encode up to 4–6 protein isoforms. There could be more isoforms if CL5a-N1 and -N2 were not conserved at the uncharacterized C-terminal. In addition, considering the sequences that were not primed by both primers, it is likely that there should be even more isoforms. This possibility is corroborated by the 2D gel analysis of the CL5a isoforms bound to different pathogens (see section below).

Figure 2 shows the multiple alignments of the derived CL5 sequences with their homologs, TL5a and TL5b. In general, it appears that the N-termini are more diversified, especially among the CL5b isoforms. The group I isoforms of CL5b (N1, N2 and N3) show higher homology to TL5b than the group II sequences. Interestingly, no CL5b sequence could be aligned to TL5b without gaps, demonstrating that multiple insertions/deletion mutations have occurred in evolution. In contrast to CL5b, the CL5a isoforms appear more evolutionarily conserved. The amino acid residues involved in Ca2+ binding are conserved in all the isolated CL5a sequences and CL5b group II sequences (but not in CL5b group I isoforms). This is in agreement with a concurrent study in our laboratory, where two populations of CL5b were observed to bind to the LPS affinity column in Ca2+-dependent and -independent manners (P. M. L. Ng et al., unpublished data). In addition, the residues in TL5a, which are involved in the binding of GlcNAc are conserved in all CL5a isoforms, but are not fully conserved in CL5b isoforms and TL5b, suggesting the potential functional divergence in ligand binding.

A homology tree (Fig. 3A) based on the multiple alignment is constructed, revealing much greater divergence of CL5b isoforms and suggesting that CL5a is probably the ancestral ortholog of various CL5 isoforms. By SWISS-MODEL using the crystal structure of TL5a complexed with GlcNAc as template, the 3D structures of TL5b, CL5a-N3/C1 and CL5b-N4/C1 were predicted (Fig. 3B). Although the overall folding patterns appear similar, CL5a is closer to TL5a and CL5b is closer to TL5b, indicating the generation of polymorphism after the divergence of CL5a and CL5b. Although CL5a and CL5b sequences were isolated from the hepatopancreas and cardiac tissues, respectively, these lectins appear to be more abundantly transcribed in intestines (data not shown).


Figure 3
View larger version (45K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 3 Comparison of sequence and structure homology between CL5s and TL5s. (A) The homology tree was constructed based on the multiple sequence alignment (Fig. 2). (B) The 3D structures of CL5s and TL5s were predicted by SWISS-MODEL online at http://swissmodel.expasy.org (25, 26) using the crystal structure of TL5a (Protein Data Bank ID: 1JC9) as template. The CL5a sequence used for modeling was assembled from overlapping sequences of CL5a-N3 and CL5a-C1. CL5b was assembled from overlapping sequences of CL5b-N4 and CL5b-C1.

 
Previously, the existence of multiple gene copies for TL5s was suggested by Southern analysis (19). However, our present data demonstrate the existence of multiple isoforms of the lectins 5 rather than multiple gene copies. Thus, the functional divergence of these lectins was speculated and tested.

Different pools of CL5 isoforms differentiate bacteria and fungi
In our recent discovery of an invertebrate homolog of the complement system in the horseshoe crab (21), we showed that CL5a (~36 kDa) and CL5b (~40 kDa) bind various microbes and appear to trigger the complement activation. In this work, the identification of multiple CL5a and CL5b isoforms renders a speculation that different pools of CL5 isoforms exhibit various ligand affinities to recognize and differentiate a broad spectrum of pathogens. To this end, we further employed 2D gel analysis to compare the profiles of total protein extracts from microbes incubated with the horseshoe crab plasma, aiming to see differential 2D resolution patterns of CL5 isoforms participating in the recognition of different kinds of pathogens.

Figure 4(A) shows 1D SDS-PAGE profiles of pathogen-binding proteins, featuring CL5a and CL5b as the prominent pathogen-binding lectins. In this study, a protease inhibitor cocktail was added into the binding assay to prevent the activation of C3, thus the pathogen-bound proteins appeared to be exclusively CL5s. In addition, we also tested if individual PAMP molecules were able to capture CL5s. Figure 4(B) shows the profile of plasma proteins that bind to representative PAMP molecules (LPS from Gram-negative bacteria E. coli, LTA from Gram-positive bacteria S. aureus and zymosan, the cell wall extract from yeast S. cerevisiae that contains mainly ß-glucan). Very similar overall profiles of plasma proteins were observed with those bound to live microbes, suggesting the surface PAMP molecules are the ligands of CL5s. Subsequently, we developed a protocol using acetone precipitation and detergent replacement to efficiently remove SDS from the proteins extracted from microbes that were incubated with plasma. This resulted in the ideal condition for isofocusing of protein samples (see Methods). Figure 4(C) shows the array of CL5 isoforms of similar sizes, bound to the representative microbes. In addition, the resolution patterns of the fungus-associated CL5a and CL5b are distinctive from those associated with the two bacterial species. Interestingly, the resolution profiles with Gram-negative E. coli and Gram-positive S. aureus are very similar. The fungal-binding CL5s appear to be acidic while the pI of bacteria-binding CL5s is approximately neutral.


Figure 4
View larger version (75K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 4 Analysis of CL5s bound to representative microbes and PAMP molecules. The prominent proteins that bind different kinds of microbes (Staphylococcus aureus, Escherichia coli and Kluyveromyces marxianus) and PAMP molecules (LPS, LTA and zymosan) are CL5a and CL5b. (A) SDS-PAGE analysis of SDS extracts of the three microbes (untreated or treated with horseshoe crab plasma). Between panels (A) and (B), a lane of the total plasma proteins is appended, featuring hemocyanin which constitutes >90% of total proteins and CRP as the second most abundant protein in the plasma. This control lane demonstrates the specific adsorption of CL5s to the (A) pathogens or (B) PAMP ligands. (B) SDS-PAGE analysis of extracted proteins from PAMP molecules incubated with horseshoe crab plasma. (C) The 2D gel resolution of CL5s associated with the three microbes tested. Since the proteins extracted from S. aureus and K. marxianus showed no visible background, E. coli extract was supplemented to the S. aureus and K. marxianus samples in order to provide ‘landmark’ protein spots for subsequent profile alignment. The CL5 spots (Ec-A1 to -A6, Ec-B1 to -B3 for CL5s bound to E. coli; Sa-A1 to -A6, Sa-B1 to -B3 for CL5s bound to S. aureus; Km-A1 to -A8, Km-B1 to -B6 for CL5s bound to K. marxianus) were excised from the gels for MS analysis.

 
Well-separated isoform spots (Fig. 4C) in the 2D gel were excised and in-gel digested with trypsin, and analyzed by MALDI-tof-tof [see Fig. 5 for the original peptide mass fingerprint (PMF) of all the characterized CL5 spots]. The identity of these proteins as CL5 isoforms was confirmed by homologous MS/MS ion spectra database search, which revealed the matches with TL5s (CL5a spots matched TL5a and CL5b spots matched TL5b, respectively). In addition, several peptide sequences from CL5a spots (Fig. 5A) or CL5b spots (Fig. 5B) that were obtained by tandem MS/MS analysis were located in the cDNA-derived protein sequences (see Fig. 2). Comparison of the CL5a and CL5b PMF profiles (peaks >5% relative intensity) are shown in Fig. 5(A and B), respectively. The CL5 isoform protein spots that associate with E. coli, S. aureus and K. maxianus are henceforth referred to as Ec, Sa and Km, respectively. The resolution patterns of the CL5 spots bound to E. coli and S. aureus can be accurately aligned using background proteins from E. coli as landmarks, suggesting that they are the same set of CL5s. Indeed, the MS analysis showed almost complete identity of PMFs between the aligned CL5 spots bound to E. coli and S. aureus. In addition, the two rather weak protein spots bound to K. maxianus: Km-A7 and -A8 that were well aligned with Ec-/Sa-A5 and -A6, respectively, also show very similar PMFs to their aligned spots in bacteria. However, in general, the CL5a and CL5b spots bound to yeast (except for the faintly stained Km-A7 and -A8) showed different PMFs from those bound to bacterial species, while still sharing certain peaks, thus proving the same origin of the CL5 isoforms.


Figure 5
View larger version (19K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 5 Comparison of PMFs of CL5 isoforms resolved by 2D gel. The trypsin digests of CL5 spots were analyzed by tandem MS/MS using 4700 Proteomics Analyzer. The PMFs of CL5a (A) and CL5b spots (B) were separately compared to show the presence or absence of a peptide at a certain molecular weight. To eliminate minor peaks which might have resulted from contamination and/or miscleavage by trypsin, only those peptide peaks at relative intensity >5% were arbitrarily included in the comparison. The peptide sequences obtained by MS/MS analysis, which were located in the derived protein sequences from isolated CL5 cDNA sequences (see Fig. 2), are indicated.

 

   Discussion
Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
Supplementary data
 References
 
This study shows the existence of multiple variants of CL5a and CL5b, and that different pools of isoforms are involved in recognizing and differentiating bacteria and fungi. To survive infection in the absence of adaptive immunity, invertebrates rely completely on innate immune defense to eliminate invading pathogens. The production of multiple isoforms of a PRR by gene duplication resulting in ligand affinity divergence might offer an alternative and formidable strategy to compensate for the lack of highly specific Igs.

Using detailed biochemical approaches, Ratcliffe and co-workers (9–11) characterized multiple lectins in the cockroach hemolymph with variable specificity in opsonizing foreign pathogens leading to the phagocytosis and activation of the prophenoloxidase cascade. Genome sequencing of the fruit fly (Drosophila melanogaster), mosquito (Anopheles gambiae) and human revealed the existence of families of highly diversified PGRP homologs (27–29). The fruit fly, mosquito and mammals have families of 13, 7 and 4 PGRP genes, respectively. Some of these genes are alternatively spliced. PGRPs are differentially expressed in various cells and tissues, and the expression is often up-regulated during infection (6). Also in the fruit fly, a group of genes that were generated by duplication and possibly code for members of the C-type lectin family was revealed by sequence analysis (30). The C-reactive protein (CRP) was first discovered in the human plasma as a non-Ig that precipitated with C-polysaccharides derived from bacterial cell wall (31). Three different types of CRPs with different substrate specificities were purified from the horseshoe crab T. tridentatus, and 22 clones of CRP genes with different amino acid sequences have been identified (31). Multiple genes of CRP were also isolated from the horseshoe crab, C. rotundicauda (32). Multiple fucose-binding lectins have been characterized in the eel and at least seven members of their genes revealed by cDNA cloning (33). Evidence obtained in both the vertebrates and invertebrates thus far suggests the strategy of generating a variety of PRRs as an alternative to the Igs in vertebrates, to fulfill the need to recognize and defend various kinds of invading pathogens. The findings presented in this study corroborate the evolution of a single type of lectin by way of gene duplication and functional divergence, with the development of functional diversity in pathogen/PAMP recognition in innate immunity.

It is interesting to note that although CRP is abundant in the horseshoe crab plasma, and is a polysaccharide agglutinin, the major proteins bound to various live pathogens and PAMP molecules turn out to be CL5s. Thus, it appears that CL5s play a central role in the frontline innate immune defense in the horseshoe crab. The 2D gel analysis clearly showed that fungus-binding CL5s are acidic, while the bacteria-binding CL5s are neutral. This probably indicates an evolutionary adaptation to the different surface charge distributions of bacteria and fungi. The major bacterial surface components (LPS in Gram-negative bacteria and LTA in the Gram-positive bacteria) are highly negatively charged, while ß-glucan, the component of fungal surface, is neutral.

Previously, TL5s were directly purified from the horseshoe crab plasma by using N-acetyl group-immobilized resin (18). However, this method of affinity chromatography resulted in the isolation and characterization of only one each of TL5a and TL5b. TL5a was shown to bind acetylated ligand via its C-terminal fibrinogen-like domain. While the predicted overall 3D structures of TL5s and CL5s appeared similar, the amino acid residues of TL5a forming the GlcNAc-binding pocket are not fully conserved in TL5b and CL5b isoforms. Thus, the diversity among the CL5 DNA sequence may have resulted in an array of CL5 isoforms with ligand-binding variations to fit different polysaccharides displayed on different microbes. This is corroborated by our observation of the different pools of CL5a and CL5b isoforms involved in the recognition of bacteria and fungi. Screening a hepatopancreas cDNA library using CL5b-C1 as probe yielded another 11 distinct CL5b sequences (supplementary Fig. 1, available at International Immunology Online); each of these protein sequences was obtained from two or more independent cDNA clones that were sequenced. In general, the diversity found in these sequences was less than that found in the RACE products, possibly due to the nature of the library screening, which employed a specific probe. Determination of the full sequence for all the isoforms would require a complete, large scale and tedious screening of more cDNA libraries in future, using different probes in combination with RACE by individual specific primers (instead of the current primers that are located in the very conserved regions). Nevertheless, our library screening effort here has further authenticated the richness and diversity of CL5 isoforms. Structure determination of selected fungus- and bacteria-specific CL5 isoforms will, in future, clarify the structural basis of the differential recognition spectra of these lectins.

Hitherto, it is not clear how the binding of CL5 to microbial surfaces activates the complement system. Nevertheless, CL5a and CL5b appear to collaborate in the recognition and differentiation of various micro-organisms. By combinatorial interactions among various isoforms of CL5s, which have different carbohydrate affinities, we envisage that the CL5s would rearrange/re-organize in their ensembles to adopt diverse and flexible oligomeric structures to discriminate and agglutinate a wide range of pathogenic invaders. Other parallel studies in our laboratory have shown that CL5s, CRPs and galactose-binding protein form a dynamic pathogen recognition complex (P. M. L. Ng et al., unpublished data). Furthermore, we have observed the association of galactose-binding protein with CrC2/Bf, a serine protease that activates the horseshoe crab complement C3. Therefore, we propose an integrated role of the CL5s in recognizing invading pathogens and triggering complement activation. Earlier study suggested the existence of lectin-binding receptors on the surface of the horseshoe crab hemocytes that bind plant lectins (34). Thus, it would be pertinent to search for CL5 receptors, and to demonstrate the functional consequence of the receptor recognition to phagocytosis, encapsulation and signaling during host–pathogen interaction. Since we have isolated a cDNA sequence homologous to vertebrate complement C3 receptor (T. H. H. Bui et al., unpublished data), it can be envisaged that there is potential collaboration between the C3 receptor and the putative CL5 receptor in the recognition of C3- and CL5-opsonized pathogen, which would enhance the accuracy and efficiency of pathogen elimination, as in the vertebrate system.

The conserved nature of innate immunity throughout the animal kingdom is clearly evidenced in the mechanism of pathogen recognition and opsonization by various kinds of PRR-mediated destruction via anti-microbial factors, and the elimination of the pathogen by phagocytosis. Yet, the innate immunity in both invertebrates and vertebrates is far more elaborate and powerful than is known to date. Before the appearance of genetic polymorphism and gene rearrangement in ancestral Igs in the vertebrates, the core mechanism of acquired immunity, viz, gene duplication and diversification, had already occurred in ancient animals. This is clearly indicated by PGRPs in insects and mammals, lectins in eel, fruit fly and cockroach and now CL5s in the horseshoe crab, a ‘living fossil.’ The multiplicity of the pathogen recognition molecules in the host appears to be a favored means toward its defense against the ever-diversifying foreign pathogens in evolution.


   Supplementary data
Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 Supplementary data
References
 
Supplementary Fig. 1 and table 1 are available at International Immunology Online.


   Acknowledgements
 
We thank Michelle Mok (Proteins and Proteomics Center, National University of Singapore) for technical assistance in MS. This work was supported by a BMRC grant (03/1/21/17/227) from the Agency for Science Technology Research, Singapore (A*STAR).


   Abbreviations
 
CL, carcinolectin
CRP, C-reactive protein
2D, two dimensional
GlcNAc, N-acetyl glucosamine
IGS, immunoglobulins
IPG, immobilised pH gradient
LTA, lipoteichoic acid
MS, mass spectroscopy
PAMP, pathogen-associated molecular pattern
PGRP, peptidoglycan recognition protein
PMF, peptide mass fingerprint
PRR, pattern recognition receptor
RACE, rapid amplification of cDNA ends
RT, reverse transcription
TL, tachylectin

   Notes
 
Transmitting editor: R. Medzhitov

Received 22 August 2006, accepted 12 September 2006.


   References
Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 Supplementary data
 References
 

  1. Gordon S. (2002) Pattern recognition receptors: doubling up for the innate immune response. Cell 111:927.[CrossRef][ISI][Medline]
  2. Janeway CA Jr and Medzhitov R. (2002) Innate immune recognition. Annu. Rev. Immunol. 20:197.[CrossRef][ISI][Medline]
  3. van de Wetering JK, van Golde LM, Batenburg JJ. (2004) Collectins: players of the innate immune system. Eur. J. Biochem. 271:1229.[ISI][Medline]
  4. Fujita T, Matsushita M, Endo Y. (2004) The lectin-complement pathway-its role in innate immunity and evolution. Immunol. Rev. 198:185.[CrossRef][ISI][Medline]
  5. Nonaka M and Yoshizaki F. (2004) Evolution of the complement system. Mol. Immunol. 40:897.[CrossRef][ISI][Medline]
  6. Dziarski R. (2004) Peptidoglycan recognition proteins (PGRPs). Mol. Immunol. 40:877.[CrossRef][ISI][Medline]
  7. Kurata S, Ariki S, Kawabata S. (2006) Recognition of pathogens and activation of immune responses in Drosophila and horseshoe crab innate immunity. Immunobiology 211:237.[CrossRef][ISI][Medline]
  8. Kawasaki K, Kubo T, Natory S. (1996) Presence of the Periplaneta lectin-related protein family in the American cockroach Periplaneta americana. Insect Biochem. Mol. Biol. 26:355.[CrossRef][ISI][Medline]
  9. Chen C, Durrant HJ, Newton RP, Ratcliffe NA. (1995) A study of novel lectins and their involvement in the activation of the prophenoloxidase system in Blaberus discoidalis. Biochem. J. 310:23.[Medline]
  10. Chen C, Rowley AF, Ratcliffe NA. (1998) Detection, purification by immunoaffinity chromatography and properties of 1,3-glucan-specific lectins from the sera of several insect species. Insect Biochem. Mol. Biol. 28:721.[CrossRef]
  11. Wilson R, Chen C, Ratcliffe NA. (1999) Innate immunity in insects: the role of multiple, endogenous serum lectins in the recognition of foreign invaders in the cockroach, Blaberus discoidalis. J. Immunol. 162:1590.[Abstract/Free Full Text]
  12. Iwanaga S. (2002) The molecular basis of innate immunity in the horseshoe crab. Curr. Opin. Immunol. 14:87.[CrossRef][ISI][Medline]
  13. Kawabata S and Tsuda R. (2002) Molecular basis of non-self recognition by the horseshoe crab tachylectins. Biochim. Biophys. Acta 1572:414.[Medline]
  14. Haq S, Kubo T, Kurata S, Kobayashi A, Natory S. (1996) Purification, characterization, and cDNA cloning of a galactose-specific C-type lectin from Drosophila melanogaster. J. Biol. Chem. 271:20213.[Abstract/Free Full Text]
  15. Leshko-Lindsay L and Corces VG. (1997) The role of selectins in Drosophila eye and bristle development. Development 124:169.[Abstract]
  16. Hoffman JA and Reichhart JM. (2002) Drosophila innate immunity: an evolutionary perspective. Nat. Immunol. 3:121.[CrossRef][ISI][Medline]
  17. Hoffman JA. (2003) The immune response of Drosophila. Nature 426:33.[CrossRef][Medline]
  18. Iwanaga S and Lee BL. (2005) Recent advances in the innate immunity of invertebrate animals. J. Biochem. Mol. Biol. 38:128.[ISI][Medline]
  19. Gokudan S, Muta T, Tsuda R, et al. (1999) Horseshoe crab acetyl group-recognizing lectins involved in innate immunity are structurally related to fibrinogen. Proc. Natl Acad. Sci. USA 96:10086.[Abstract/Free Full Text]
  20. Kairies N, Beisel HG, Fuentes-Prior P, et al. (2001) The 2.0-A crystal structure of tachylectin 5A provides evidence for the common origin of the innate immunity and the blood coagulation systems. Proc. Natl Acad. Sci. USA 98:13519.[Abstract/Free Full Text]
  21. Zhu Y, Thangamani S, Ho B, Ding JL. (2005) The ancient origin of the complement system. EMBO J. 24:382.[CrossRef][ISI][Medline]
  22. Shevchenko A, Wilm M, Mann M. (1997) Peptide sequencing by mass spectrometry for homology searches and cloning of genes. J. Protein Chem. 16:481.[CrossRef][ISI][Medline]
  23. Thompson JD, Higgins DG, Gibson TJ. (1994) CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, positions-specific gap penalties and weight matrix choice. Nucleic Acids Res. 22:4673.[Abstract/Free Full Text]
  24. Bendtsen JD, Nielsen H, von Heijne G, Brunak S. (2004) Improved prediction of signal peptides: SignalP 3.0. J. Mol. Biol. 340:783.[CrossRef][ISI][Medline]
  25. Guex N and Peitsch MC. (1997) SWISS-MODEL and the Swiss-PdbViewer: an environment for comparative protein modelling. Electrophoresis 18:2714.[CrossRef][ISI][Medline]
  26. Schwede T, Kopp J, Guex N, Peitsch MC. (2003) SWISS-MODEL: an automated protein homology-modeling server. Nucleic Acids Res. 31:3381.[Abstract/Free Full Text]
  27. Werner T, Liu G, Kang D, Ekengren S, Steiner H, Hultmark D. (2000) A family of peptidoglycan recognition proteins in the fruit fly Drosophila melanogaster. Proc. Natl Acad. Sci. USA 97:13772.[Abstract/Free Full Text]
  28. Liu C, Xu Z, Gupta D, Dziarski R. (2001) Peptidoglycan recognition proteins: a novel family of four human innate immunity pattern recognition molecules. J. Biol. Chem. 276:34686.[Abstract/Free Full Text]
  29. Christophides GK, Zdobnov E, Barillas-Mury C, et al. (2002) Immunity-related genes and gene families in Anopheles gambiae. Science 298:159.[Abstract/Free Full Text]
  30. Theopold U, Rissler M, Fabbri M, Schmidt O, Natori S. (1999) Insect glycobiology: a lectin multigene family in Drosophila melanogaster. Biochem. Biophys. Res. Commun. 261:923.[CrossRef][ISI][Medline]
  31. Iwaki D, Osaki T, Mizunoe Y, Wai SN, Iwanaga S, Kawabata S. (1999) Functional and structural diversities of C-reactive proteins present in horseshoe crab hemolymph plasma. Eur. J. Biochem. 264:314.[ISI][Medline]
  32. Ng PML, Jin Z, Tan SS, Ho B, Ding JL. (2004) C-reactive protein: a predominant LPS-binding acute phase protein responsive to Pseudomonas infection. J. Endotoxin Res. 10:163.[CrossRef][Medline]
  33. Honda S, Kashiwagi M, Miyamoto K, Takei Y, Hirose S. (2000) Multiplicity, structures, and endocrine and exocrine natures of eel fucose-binding lectins. J. Biol. Chem. 275:33151.[Abstract/Free Full Text]
  34. Gupta AP, Orenberg SD, Das YT, Chattopadhyay SK. (1991) Lectin-binding receptors, Na+,K(+)-ATPase, and acetylcholinesterase on immunocyte plasma membrane of Limulus polyphemus. Exp. Cell Res. 194:83.[CrossRef][ISI][Medline]

Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 Infect. Immun.Home page
Z. H. Fan, X. W. Wang, J. Lu, B. Ho, and J. L. Ding
Elucidating the Function of an Ancient NF-{kappa}B p100 Homologue, CrRelish, in Antibacterial Defense
Infect. Immun., February 1, 2008; 76(2): 664 - 670.
[Abstract] [Full Text] [PDF]

Home page
 GlycobiologyHome page
J. R. Bishop and P. Gagneux
Evolution of carbohydrate antigens--microbial forces shaping host glycomes?
Glycobiology, May 1, 2007; 17(5): 23R - 34R.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
18/12/1671    most recent
dxl101v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (5)
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Zhu, Y.
Right arrow Articles by Ding, J. L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Zhu, Y.
Right arrow Articles by Ding, J. L.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?