Composition of the Lectin Pathway of Complement in Gallus gallus: Absence of Mannan-Binding Lectin-Associated Serine Protease-1 in Birds
http://www.100md.com
免疫学杂志 2005年第8期
Abstract
The lectin pathway of complement is activated by multimolecular complexes that recognize and bind to microbial polysaccharides. These complexes comprise a multimeric carbohydrate recognition subunit (either mannan-binding lectin (MBL) or a ficolin), three MBL-associated serine proteases (MASP-1, -2, and -3), and MAp19 (a truncated product of the MASP-2 gene). In this study we report the cloning of chicken MASP-2, MASP-3, and MAp19 and the organization of their genes and those for chicken MBL and a novel ficolin. Mammals usually possess two MBL genes and two or three ficolin genes, but chickens have only one of each, both of which represent the undiversified ancestors of the mammalian genes. The primary structure of chicken MASP-2 is 54% identical with those of the human and mouse MASP-2, and the organization of its gene is the same as in mammals. MASP-3 is even more conserved; chicken MASP-3 shares 75% of its residues with human and Xenopus MASP-3. It is more widely expressed than other lectin pathway components, suggesting a possible function of MASP-3 different from those of the other components. In mammals, MASP-1 and MASP-3 are alternatively spliced products of a single structural gene. We demonstrate the absence of MASP-1 in birds, possibly caused by the loss of MASP-1-specific exons during phylogeny. Despite the lack of MASP-1-like enzymatic activity in sera of chicken and other birds, avian lectin pathway complexes efficiently activate C4.
Introduction
The lectin pathway provides an Ab-independent route of complement activation. It is an ancient system that antedates the evolution of adaptive immunity and the classical complement pathway, and it is thought to be particularly important in lower animals such as the urochordates and cephalochordates, which lack an adaptive immune system. In vertebrates, the lectin pathway provides a first line of defense during the lag phase that precedes the onset of an adaptive response (1, 2, 3). Deficiencies of the lectin pathway are associated with susceptibility to infectious disease, particularly in infants, young children, and those with another acquired immunodeficiency, such as cancer patients undergoing chemotherapy (4, 5, 6, 7, 8).
Activation of the lectin pathway occurs when a multimolecular fluid phase activation complex, comprising a recognition subcomponent and associated serine proteases, binds to carbohydrate structures present on microbial surfaces. Two types of fluid phase recognition molecule have been described, the mannan-binding lectin(s) (MBL),4 and the ficolins (9, 10, 11, 12). Structurally they are very similar, being homotrimers of a single polypeptide chain, with an N-terminal collagen-like domain and a C-terminal carbohydrate recognition domain (CRD). In MBL, the CRD is a C-type lectin domain, whereas the CRD in ficolins is a fibrinogen-like domain. In plasma, the basic homotrimers form higher order oligomers, which, in turn, form complexes with the MBL-associated serine proteases, MASP-1, -2, and -3 (13, 14, 15, 16). Only MASP-2 cleaves C4 and C4b-bound C2, generating C4b2a, the same C3 convertase that results from activation of the classical pathway (17, 18). The roles of MASP-1 and -3 in these complexes are still unclear.
All three MASPs share a conserved domain structure. Five N-terminal domains (the C1r/C1s/Uegf/bone morphogenetic protein I (CUBI), epidermal growth factor (EGF)-like, CUBII, complement control protein I (CCPI), and CCPII domains) precede a C-terminal serine protease domain. When activated, the enzymes split into two disulfide-linked chains, the A chain, which contains the five N-terminal domains, and the B chain, which contains the serine protease domain (2, 19). The mRNAs encoding MASP-1 and -3 are alternatively spliced products of one single gene. The A chain is common to both proteins, the B chain of MASP-1 is encoded by five or six exons (depending upon the species), and the B chain of MASP-3 is encoded by a single exon, located between the exons encoding the shared A chain and those coding for the B chain of MASP-1 (16, 19, 20). This genomic arrangement is found in amphioxus (a cephalochordate) and in all vertebrates investigated to date, but not in the ascidians, indicating that it arose after the divergence of the urochordates, but before the divergence of the cephalochordates (21). The lectin pathway activation complexes also contain MAp19, a protein composed of the two N-terminal domains of MASP-2 encoded by a truncated mRNA transcript that arises through alternative splicing of the MASP2 gene (19, 22, 23, 24). MAp19 has no enzymatic activity, and its function within the lectin pathway activation complexes remains unclear.
Commercial poultry are vulnerable to bacterial, viral, and parasitic infections that cause considerable mortality and economic loss. They are also reservoirs for transmissible human pathogens, including Salmonella, Campylobacter, and potentially dangerous stains of influenza, such as the H5N1 (Guangdong) virus that caused the 1997 bird flu outbreak in Hong Kong (25, 26). Although the adaptive immune system of the chicken has been extensively studied (27), the complement system and, in particular, the lectin pathway, is less well understood. It is known that Ab-independent activation of the complement system plays an important role in the host response to fowlpox virus; fowlpox virus infection is aggravated in chickens treated with cobra venom factor (28).
The classical pathway components C1 and C3 through C9 have been found in chickens (29). Because a homologue for C2 was not found in these early studies, it was assumed that the role of C2 may be fulfilled by the chicken factor B-like protease, an evolutionary remnant of a common C2/factor B ancestor described previously (29, 30). However, subsequent studies identified and mapped the chicken homologues of both the C2 and factor B genes and showed that chickens are not deficient in C2 (K. Skjoedt and J. Kaufman, unpublished observations). MBL is the only lectin pathway component that has been described in the chicken to date. Although most mammals have two separate genes for MBL (31), only one gene was found in the chicken, and phylogenetic analysis of its amino acid sequence indicated that it may represent an undiversified ancestor of the mammalian MBLs (32). The concentration of MBL in chicken serum ranges from 0.4 to 37.8 μg/ml (mean, 5.8 μg/ml), with no strain-to-strain variation and no deficiencies detected (33). MBL levels increase 2-fold in chickens infected with infectious bronchitis virus, infectious laryngotracheitis virus, and infectious bursal disease virus, indicating that MBL is a mild acute phase protein (34).
This report describes the molecular cloning of chicken MASP-2, MASP-3, and MAp19 and the organization of their genes and those for chicken MBL and a novel ficolin. We demonstrate the apparent absence of MASP-1 and MASP-1-like enzymatic activity in chickens and other birds.
Materials and Methods
RNA extraction and cDNA synthesis
Total RNA was prepared using TRIzol reagent (Invitrogen Life Technologies), according to the manufacturer’s instructions. Poly(A)+ mRNA was purified from total RNA using the Oligotex mRNA Midi kit, from Qiagen.
For cDNA synthesis, 8 μg of total RNA was digested with RNase-free DNase I (Promega), extracted once with phenol/chloroform/isoamyl alcohol (25/24/1), precipitated, washed with ethanol, then redissolved in water. The concentrations of the RNA samples were determined by measuring the absorbance at 260 nm. Three micrograms of each RNA sample was primed with oligo(dT)12–18, and single-stranded cDNAs were synthesized using the SuperScript First-Strand Kit (Invitrogen Life Technologies) according to the supplier’s instructions.
cDNA cloning
A ZAP chicken liver cDNA library (Stratagene) was screened by filter hybridization using [-32P]dCTP-labeled cDNA probes prepared with the Random Primed Labeling kit (Roche). Isolated ZAP clones were converted into pBluescript SK– plasmid clones by in vivo excision, following the supplier’s protocol, and sequenced (Genterprise). The 5' and 3' ends of some cDNA sequences were extended by RACE, using chicken liver RNA and kits (Invitrogen Life Technologies). RACE products were cloned into pGEM-Teasy (Promega) and sequenced.
Northern blotting
For Northern blot analysis, 10 μg of total RNA or 100 ng of poly(A+) mRNA was separated on formaldehyde-containing 1.2% agarose gels, transferred to nylon membrane, and hybridized with [-32P]dCTP-labeled cDNA probes using standard methods (35). The cDNA probes used were: for the A chain of MASP-2, a 376-bp RT-PCR product amplified from chicken liver cDNA with primers M2A-ch-F and M2A-ch-R (Table I); for the B chain of MASP-2, a 430-bp RT-PCR product generated with primers M2dg-F2 and M2dg-R1; for the A chain of MASP-3, a 740-bp RT-PCR product amplified with primers M1&3dg-F and M1&3dg-R; and for the B chain of MASP-3, a 1200-bp NcoI fragment of clone CM3/5, encompassing the last 500 bp of the coding sequence and the first 700 bp of the 3'-untranslated (3'UT) region. For controls, rat liver RNA was hybridized with rat MASP-3 A and B chain-encoding cDNA probes, as previously described (20).
Table I. Oligonucleotides used in this study
In situ hybridization
mRNA was localized by in situ hybridization using 35S-labeled antisense RNA probes generated by in vitro transcription. The templates for in vitro transcription were RT-PCR products, amplified from liver cDNA and cloned into pGEM-Teasy. For the A chain of chicken MASP-3 and for the B chain of chicken MASP-2, the cloned RT-PCR products were those described above. For the B chain of chicken MASP-3, a 564-bp RT-PCR product was generated using primers M3SP-F1 and -R1 (Table I); for chicken MAp19, a 170-bp product was obtained using primers MAp19-F1 and -R1; for chicken MBL, a 330-bp product was generated with primers MBL-F1 and MBL-R1; for chicken ficolin, a 324-bp product was amplified using primers ChkFCN_F1 and ChkFCN_R1; for the B chain of rat MASP-1, a 400-bp product was generated using primers RM1sp-F1 and RM1sp-R1; and for the B chain of rat MASP-3, a 407-bp product was obtained with primers RM3-BCF1 and RM3-BCR1. Antisense RNA probes were transcribed from the cloned templates using the method described by Melton et al. (36). For controls, the templates were transcribed in the opposite direction, generating sense RNA probes. Twenty-micron-thick tissue sections were cut using a cryostat, mounted on poly-lysine-coated microscope slides (Merck), and hybridized with the 35S-labeled probes according to the method of Sch?fer et al. (37). After hybridization and washing, the signals were detected by exposing the sections to Kodak BioMax MR x-ray film (Sigma-Aldrich).
Analysis of gene expression by quantitative RT-PCR (qRT-PCR)
cDNA was analyzed by quantitative PCR using a LightCycler (Roche) to follow the incorporation of SYBR Green I into the PCR products in real-time. Each 15 μl of PCR contained 1/100th of the original cDNA synthesis reaction (corresponding to 30 ng of total RNA), 0.5 μM of each primer, and 7.5 μl of QuantiTect SYBR Green Master Mix (Qiagen). Forty-five cycles of amplification were performed; the annealing temperature was reduced from 70 to 58°C during the first 15 cycles and was kept constant at 58°C thereafter. The fluorescence signal was detected at the end of each cycle, and results were analyzed using the Fit Points option in the LDCA software supplied with the machine (38). Melting curve analysis was used to confirm the specificity of the products (39). The primers used to amplify the A chain of MASP-3 were M3Ach-F1 and -R1 (generating a 261-bp product); for the B chain of MASP-3, M3SP-F1 and M3SP-R1 (564 bp); for MAp19, MAp19-F1 and MAp19-R1 (170 bp); for MASP-2, M2A-ch-F and M2A-ch-R (376 bp); for MBL, MBL-F1 and MBL-R1 (330 bp); and for chicken ficolin, ChkFCN-F1 and ChkFCN-R1. Standard curves were produced for each analysis using serial dilutions of the corresponding cDNAs cloned in pBluescript SK– (for MASP-2 and -3) or pGEMTeasy (for MAp19 and MBL).
Enzyme assays
Mannan-coated microtiter plates were used to capture MBL-MASP complexes from pooled normal human serum, chicken serum (from a pool of five animals), and turkey serum (also pooled from five animals). C4 cleavage activity of the captured complexes was assayed using the method developed by Petersen et al. (40), with minor modifications (41). MASP-1-like activity was measured using the fluorescent substrate, butyloxycarbonyl-Val-Pro-Arg-7-amino-4-methylcoumarin (AMC), as previously described (42).
Sequence analysis
The chicken genome database is available at www.ensembl.org/Gallus_gallus/. Similarity searches with nucleotide and polypeptide query sequences were run on this server using BLASTN and TBLASTN, respectively. The MutliContigView option was used to create chicken/human synteny maps. Signal peptides and N-linked glycosylation sites were predicted using SignalP v3.0 and NetNGlyc v1.0, which can be found at www.cbs.dtu.dk/services/.
Results
cDNA cloning
Degenerate oligonucleotides were designed to match parts of the MASP-1/3 A chain and the B chain of MASP-2 that are conserved in the human, rat, mouse, and Xenopus laevis sequences. The MASP-1/3 primers (M1&3dg-F and M1&3dg-R; Table I) were used to amplify a 740-bp fragment of cDNA from chicken liver RNA by RT-PCR, and the MASP-2 specific primers (M2dg-F2 and M2dg-R1) were used to amplify a 430-bp fragment from the same source. Both products were cloned into pGEM-Teasy and then sequenced to confirm that they encode the anticipated regions of the chicken MASP cDNAs. The RT-PCR products were subsequently labeled with [-32P]dCTP and used to screen a proprietary chicken liver cDNA library.
Seven clones were obtained using the MASP-1/3 A chain probe. Sequencing showed that all seven were incomplete MASP-3 cDNAs, the longest of which (clone CM3/5) was 3343-bp long and was complete at its 3' end, but began with DNA encoding the middle of the CUBI domain, 300 bp downstream of the expected transcription start. Where they overlapped, the sequences of the other six clones were identical with that of CM3/5. The 5' end of the MASP-3 sequence was completed by RACE, using chicken liver RNA and the primers M3race-R1, -R2, and -R3. A 565-bp RACE product was obtained, extending the known sequence by 293 bp.
The complete chicken MASP-3 cDNA (GenBank accession no. AY567829) is 3636 bp long and comprises a 159-bp 5'UT region, followed by a 2190-bp open reading frame, a stop codon TGA, and a 3'UT region of 1284 bp. A putative polyadenylation signal is located 13 bp upstream of the poly(A)+ tail. Analysis of the cDNA-derived amino acid sequence revealed extensive similarity between chicken MASP-3 and MASP-3 from other species. The chicken sequence shares 75% of its residues with the human sequence, 76% with the mouse sequence, and 75% with the Xenopus laevis MASP-3a sequence (Table II). The isoleucine/arginine cleavage site that separates the A and B chains is conserved, as are three residues that are important for catalytic activity (His, Asp, and Ser, the so-called catalytic triad). All the N-glycosylation sites are also conserved, as are all the cysteine residues, including the two that form the methionine loop, a cystine bond around a methionine residue in the serine protease domain (Fig. 1).
Table II. Cross-species comparison of MASP-3, MASP-2, and MAp19 peptide sequences (percentage of identical residues).
FIGURE 1. Alignment of the cDNA-derived peptide sequences of chicken, human, and mouse MASP-3 and X. laevis MASP-3a. Predicted leader peptides are italicized, potential N-glycosylation sites are underlined, conserved residues are marked with an asterisk, conservative substitutions are marked with a colon, and conserved cysteines are shown in bold. The arginine/isoleucine cleavage site that separates the A and B chains is indicated by an arrow, and the three residues that form the catalytic triad are boxed.
No MASP-1-like clones were obtained, either directly from the cDNA library or by 3'RACE using primers located in the A chain of MASP-3 (which is shared with MASP-1 in other species).
The largest clone obtained using the MASP-2 B chain probe was 2082 bp long, which did not comprise the full-length coding sequence. The 5' end of this clone extended to the cDNA encoding the CUBII domain. 5'RACE using primers M2race-R1, -R3, and -R4 gave a 601-bp product, which extended the known sequence by 336 bp and included the initiation codon. The complete MASP-2 cDNA (GenBank accession no. AY567828) is 2449 bp long and includes a 52-bp 5'UT region, a 2058-bp open reading frame, the stop codon, and a 336-bp 3'UT region, with a polyadenylation signal 19 bp upstream of the poly(A)+ tail. All the important features found in the primary structure of human, mouse, and Xenopus MASP-2 are conserved in chicken MASP-2 (Fig. 2), although the degree of identity among the MASP-2 amino acid sequences is less than that among the MASP-3 sequences (Table II).
FIGURE 2. The cDNA-derived amino acid sequences of chicken MASP-2 and MAp19. Predicted leader peptides are italicized. Features that are conserved in other species include the two N-glycosylation sites (underlined), all the cysteine residues shown in bold, the three residues that form the catalytic triad in MASP-2 (boxed), and the arginine/isoleucine cleavage site that separates the A and B chains of MASP-2 (indicated by an arrow).
Chicken MAp19 cDNA was cloned by RT-PCR. First, the 3' sequence was obtained by 3'RACE using primers located within the DNA encoding the CUBI domain of MASP-2 (MAp-race-F1 and MAp-race-F2). A 440-bp RACE product was obtained, cloned into pGEM-Teasy, and sequenced. Next, primers representing the 5' end of MASP-2 (MAp19-F2) and the 3' end of the MAp19 RACE product (MAp19-R2) were used to amplify full-length MAp19 cDNA from chicken liver cDNA. The first 588 bp of chicken MAp19 cDNA are identical with those of the MASP-2 cDNA and encode the CUBI and EGF-like domains (Fig. 2). The remainder of the cDNA is unique to MAp19 and includes the third nucleotide of the codon for the last residue, the stop codon, and a 3'UT region of 177 bp (GenBank accession no. AY567830). Mammalian MAp19 proteins have a four-residue C-terminal extension (EQSL) that is absent in the corresponding MASP-2 sequences (23, 24). Chicken MAp19 lacks this feature; its entire amino acid sequence is contained within the MASP-2 sequence.
Organization of the lectin pathway genes
The genes for chicken MASP-2 and MASP-3 were characterized by comparing the cDNA sequences with the recently completed chicken genome sequence.
The MASP-2 gene occupies 14 kb of chromosome 21, and its intron/exon structure is identical with that reported for the human and mouse MASP-2 genes (23, 24). Exon 1 encodes the 5'UT region and the initiation codon, exons 2 and 3 encode the signal peptide and the CUBI domain, and exon 4 encodes the EGF-like domain. Splicing between exons 4 and 5 generates the MAp19 mRNA. In mammals, the fifth exon encodes the last four residues of MAp19 (EQSL), the stop codon, and the 3'UT region. In the chicken, exon 5 only contains the last nucleotide of the coding sequence, followed by the stop codon and the 3'UT region. Exons 6 and 7 encode the CUBII domain of MASP-2, exons 8–11 encode the two CCP domains, and exon 12 encodes the serine protease domain (Table III). The codon phases are conserved between the chicken and human genes (43). The region of chromosome 21 that contains the chicken MASP2 gene is syntenic with the MASP2 loci on human chromosome 1p36, mouse chromosome 4E1, and rat chromosome 5q36. All four loci contain the genes TARDBP, MASP2, SRM, PMSCL2, and FRAP1 (or their homologues), in that order (43, 44).
Table III. Structure of the chicken MASP2 gene
The gene for chicken MASP-3 is located on chromosome 9. It comprises 11 exons, the organization of which is identical with that of the first 11 exons in the human and mouse MASP1/3 genes. The first 10 exons encode the 5'UT region and the A chain of MASP-3, and the 11th exon encodes the serine protease domain, stop codon, and 3'UT region. In the mammalian genes, exon 11 is followed by six additional exons, which encode the (TCN-type) serine protease domain of MASP-1. We looked for the equivalent exons in the chicken using the peptide sequences from the serine protease domains of human, mouse, and X. laevis MASP-1 as query sequences for TBLASTN searches of the chicken genome database. There is no evidence for MASP-1-specific exons in the chicken genome, either 3' of the MASP-3 gene or elsewhere in the genome. (The closest hits were actually MASP-2 and anticoagulant protein C.)
Examination of the sequence surrounding the MASP-3 gene revealed a possible explanation for the absence of MASP-1-specific exons in the chicken genome. The region 5' of the chicken MASP-3 gene, which includes the chicken homologues of hsBCL6 and hsSST, is syntenic with the human MASP1/3 locus on chromosome 3q27, but the synteny ends immediately 3' of the MASP-3 gene (Fig. 3). TM7L_HUMAN is 20 kb downstream of the human MASP1/3 gene, followed by RPL39L and SIAT1. In the chicken, the homologues of TM7L_HUMAN and RPL39L are missing altogether, and the homologue of SIAT1 (SAI1_chick) is found 5.8 Mb 5' of the chicken MASP-3 gene. Likewise, ENSGALG00000007474 and TFR1_chick are found 3' of the chicken MASP-3 gene, but their human homologues (PCYTIA and TFRC) are located 9 Mb upstream of the human MASP1/3 gene. These findings suggest that an early translocation event separated the MASP-1-specfic exons from the 3' end of the original MASP-1/3 gene, and that the MASP-1-specific exons together with the homologues of TM7L_HUMAN and RPL39L were either lost during translocation or disappeared at a later stage.
FIGURE 3. Comparison of the chicken MASP-3 and human MASP1/3 loci. Homologous genes are joined by dashed lines, and arrowheads indicate the direction of transcription. This figure is not to scale.
We also identified genes encoding the two lectin pathway recognition molecules in the chicken, MBL and a ficolin. Most mammals have two functional MBL genes, located on separate chromosomes, although in humans, MBL1 is a pseudogene, and only MBL2 produces a protein (31). The chicken has just one MBL gene, which is located on chromosome 6 in a conserved cluster that includes the gene for pulmonary surfactant protein D. It is homologous to the human pseudogene (MBL1) and the mouse and rat genes (Mbl1 and MABA_RAT).
Humans have two ficolin genes, FCN1 and FCN2, arranged back-to-back on chromosome 9q34, and a third, FCN3, on chromosome 1p35. FCN1 encodes M-ficolin, which is expressed on the surface of immature macrophages, whereas FCN2 and FCN3 encode L-ficolin and H-ficolin, which are both plasma proteins. The mouse homologues of FCN1 and FCN2 (Fcna and Fcnb, respectively) are arranged back-to-back on chromosome 2. Chickens have a single ficolin gene located on chromosome 17, which appears to represent an undiversified ancestor of FCN1 and FCN2. Its putative protein sequence is 60% identical with that of L-ficolin (the product of FCN2) and 57% identical with that of M-ficolin (the product of FCN1).
Expression of the lectin pathway components
In situ hybridization was used to analyze mRNA expression in embryonic tissue (Fig. 4); Northern blotting and qRT-PCR were used to analyze adult tissue (Figs. 5–7).
FIGURE 4. Localization of lectin pathway mRNAs by in situ hybridization. Cryostat sections (20 μm) were hybridized with 35S-labeled cRNA probes and exposed to Kodak BioMax MR film for 48 h. A–D, Day 12 chicken embryos hybridized with antisense probes for: MASP-2 B chain, MAp19, MBL, MASP-3 A chain, and MASP-3 B chain. E, Control section hybridized with a sense probe for MBL. F and G, Day 19 rat embryos hybridized with antisense probes for the B chains of MASP-1 and MASP-3. H, Control section hybridized with a sense probe for MASP-1 B chain. I and J, In situ hybridization results for chicken embryo sections obtained with antisense probes specific for the chicken lectin pathway recognition molecules MBL (I) and ficolin (J).
FIGURE 5. Expression of MASP-2 and MAp19 mRNA in chicken liver. Northern blots, prepared from chicken liver RNA, were hybridized with cDNA probes corresponding to the A chain (A) and the B chain (B) of MASP-2.
FIGURE 6. The qRT-PCR analysis of MASP-3 expression in adult chicken tissues. cDNA was prepared from the tissues indicated and was analyzed by real-time PCR using a LightCycler instrument. Results shown are copies per microgram of RNA and are the means of four separate experiments (two cDNA syntheses per sample, two analyses per cDNA). Error bars represent the SD.
FIGURE 7. Expression of MASP-3 mRNA in the livers of different birds. Poly(A)+ mRNA was prepared from the species indicated and analyzed by Northern blotting. Blots were hybridized with -32P-labeled cDNA probes corresponding to the A chain (A) and the B chain (B; serine protease domain) of chicken MASP-3. As a control, rat liver RNA was hybridized with the corresponding rat-specific probes encoding either the A or B chain of rat MASP-3 (right lanes). Distinct signals for MASP-1 (5 kb) and MASP-3 (3.5 kb) were seen using the A chain probe.
In the chicken, as in other species (23, 24), MASP-2 and MAp19 mRNA expression is restricted to liver tissue (Fig. 4, A and B). Northern blotting using a cDNA probe comprising the 5'-coding sequence for the A chain of MASP-2 revealed two mRNA species in adult liver, the 2.5-kb MASP-2 message and the 0.8-kb MAp19 mRNA (Fig. 5A). Using a probe comprising the coding sequence for the B chain, only the 2.5-kb MASP-2 message was seen (Fig. 5B). The qRT-PCR analysis showed that MAp19 mRNA is 3.5 times more abundant than the MASP-2 message in adult liver (data not shown).
Chicken MBL mRNA is restricted to the liver in both embryos (Fig. 4I) and adult tissue. This is in contrast to murine and human MBL, which is expressed in hepatic and extrahepatic tissues (45). The putative chicken ficolin gene is expressed in embryonic and adult liver tissue only (see Fig. 4J).
In humans, rats, and mice, MASP-1 expression is liver specific, whereas MASP-3 is expressed in the liver and in nonhepatic tissues, including the spleen, lung, small intestine, thymus, and brain (Fig. 4, F and G) (N. J. Lynch, unpublished observations). In day 12 chicken embryos, MASP-3 mRNA was detected by in situ hybridization in the liver, spleen, thymus, gizzard, mesonephros, and neuronal areas of the brain, but not in the lung. The same distribution and intensity of expression were observed using separate cRNA probes for the A and B chains of MASP-3 (Fig. 4, C and D). This result underlines the findings of the Northern blot analysis, showing that no evidence for a MASP-1-like transcript can be found.
This is also confirmed by the results obtained by qRT-PCR analysis of adult chicken tissues: MASP-3 expression was most abundant in the liver, followed by spleen, kidney, tonsil, duodenum, and brain. Again, using a combination of oligonucleotides that amplifies mRNA sequences specific for either the A chain or the B chain shows that identical abundances for MASP-3 A chain- and MASP-3 B chain-encoding mRNAs were detected and provided no evidence for the presence of a MASP-1-like transcript. Moreover, the amount of A and B chain message was identical in all tissues tested (Fig. 6).
Northern blot analysis using separate probes for the A and B chains of MASP-3 revealed a single 3.7-kb message in chicken liver, reinforcing the evidence that the MASP-1 transcript is missing (Fig. 7, first lane). This is in contrast to what was seen in mammals, where both MASP-3 and MASP-1 mRNA transcripts were detected (Fig. 7, upper last lane) (20). Analyzing liver mRNA from other birds (including turkey, duck, goose, ostrich, and pigeon) on Northern blots suggested that the absence of MASP-1 may be a general phenomenon in birds (Fig. 7).
Enzymatic activity of MBL-MASP complexes in avian sera
We tested the enzymatic activity of MBL-MASP complexes captured from chicken and turkey sera. Serum samples were diluted in a high salt buffer, which dissociates C1, thus avoiding interference from the classical pathway serine proteases, C1r and C1s (40). The diluted sera were added to ELISA plates coated with mannan, a polysaccharide that binds MBL-MASP complexes, then the bound complexes were assayed for MASP-1 and MASP-2. MASP-1-like activity was measured using the fluorescent substrate, Val-Pro-Arg-7-AMC, which is cleaved by MASP-1, but not by MASP-2 (42). The activity of MASP-2 was measured by following the cleavage of human C4.
MASP-1-like activity could be detected in MBL-MASP complexes from human serum, which was included as a control, but not in the avian sera, whereas MASP-2-like activity was present in all sera tested (Fig. 8). Subsequent experiments showed that MBL-MASP complexes from goose and ostrich sera also lack MASP-1-like enzymatic activity (data not shown).
FIGURE 8. Enzymatic activity of MBL-MASP complexes from human and avian sera. Mannan-coated microtiter plates were used to capture MBL-MASP complexes from serum. The immobilized complexes were assayed for MASP-1-like activity using the fluorescent substrate Val-Pro-Arg-7-AMC. To assay MASP-2, human C4 was added to the plates, and the deposition of C4b was measured. Results shown are percentages of the activity measured in human serum and are the means of duplicate determinations.
Discussion
Most vertebrates have three or four different lectin pathway recognition molecules, with differing, but overlapping, carbohydrate specificities that allow them to recognize a broad spectrum of microbial polysaccharides. The recognition molecules form complexes with three MASPs and MAp19 (the truncated, nonenzymatic form of MASP-2). Only one of these molecules has previously been reported in the chicken, namely MBL. In this study we present the most complete description of the lectin pathway in the chicken to date.
Laursen et al. (31) cloned the cDNA for chicken MBL and (based on a phylogenetic comparison of its deduced amino acid sequence with those of its mammalian homologues) concluded that it represents an undiversified ancestor of the two different forms of MBL found in mammals. Our findings support their conclusion; the chicken has a single MBL gene located on chromosome 6. Genomic analysis showed that it is more closely related to MBL1, Mbl1, and MABA-RAT than to MBL2, Mbl2, and MABC_RAT, suggesting that the first group represents the common ancestor. Likewise, the chicken has a single ficolin gene that appears to represent an undiversified ancestor of FCN1 and FCN2. It is not particularly surprising that the chicken has so few lectin pathway recognition molecules; the gene duplications that lead to the diversification of MBL and the ficolins were relatively recent evolutionary events (1), and such chromosomal rearrangements are thought to occur much more slowly in birds (and reptiles) than in mammals or amphibians (46, 47). Interestingly, chicken C1q (the recognition molecule of the classical pathway) appears to be a homotrimer, encoded by a single gene on chromosome 21. In this respect, it is more closely related to the ancestral form found in the lamprey than to mammalian C1q, which is a heterotrimer of three different polypeptides encoded by three separate genes (48).
The serine proteases found in the classical and lectin pathway activation complexes can be divided into two evolutionary groups: the ancestral TCN type (which includes MASP-1 and the ascidian MASPs) has a TCN codon for the serine in its catalytic triad and split exons for the protease domain, whereas the AGY type (which includes MASP-2, MASP-3, and C1s) uses AGY to encode its active serine and has a single exon for the protease domain (49). The AGY type evolved after, and appears to be derived from, the gene rearrangement that gave rise to the alternatively spliced MASP-1/3 gene (1, 21).
As expected, chicken MASP-2 is an AGY-type serine protease. Its primary structure shows 54% identity with those of human and mouse MASP-2. The organization of its gene is highly conserved; the intron-exon structure and codon phases are identical with those of the human and mouse genes (Tables II and III). As in other species, chicken MAp19 is an alternatively spliced product of the MASP-2 gene. It is slightly unusual in that it lacks the four-residue C-terminal extension found in mammalian MAp19 (22, 23, 24). In mammals these four residues (EQSL) are encoded by exon 5, which is unique to MAp19; in the chicken, exon 5 encodes only the third nucleotide of the last residue (which is identical with that of MASP-2), the stop codon, and the 3'UT region.
MASP-3 is the most highly conserved protein in the activation complexes; the primary structure of the chicken MASP-3 is 75% identical with those of the human, mouse, and Xenopus proteins. Its expression is more widespread than that of the other lectin pathway components; significant amounts of MASP-3 mRNA are present in liver, spleen, thymus, neuronal tissue, and gastrointestinal tissues, not only in chickens, but also in mice, rats and humans (Figs. 4 and 6) (N. J. Lynch, unpublished observations). In contrast, the messages for MASP-1, MASP-2, and MAp19 mRNA are restricted to the liver in all species investigated to date, and those for MBL and the plasma ficolins are mainly expressed in the liver (2, 20, 24, 45). The conservation of MASP-3 and its unique expression pattern suggest that its function may be decidedly different from those of the other lectin pathway components.
The 5' end of the chicken MASP-3 gene is similar to those of the mammalian MASP-1/3 genes and the Xenopus MASP1/3a gene; 10 exons encoding the A chain precede a single (AGY-type) exon that encodes the serine protease domain. However, the mammalian genes have an additional six exons that encode the (TCN-type) MASP-1 serine protease domain. In the chicken these exons appear to have been lost as a result of translocation and simultaneous or subsequent deletion that removed not only the MASP-1-specific exons, but also the chicken homologues of TM7L_HUMAN and RPL39L (Fig. 3). We found four separate pieces of evidence indicating that MASP-1 is absent in the chicken. 1) There are no MASP-1-specific exons, either 3' of MASP-3 gene or elsewhere the in the chicken genome. 2) Northern blots hybridized with a probe for the A chain of MASP-3 show a single mRNA species; two messages would be expected if the chicken had an alternatively spliced MASP-1/3 gene. 3) Similarly, qRT-PCR showed that the amount of MASP-3 A chain-encoding message is approximately equal to that of the B chain-encoding message in all tissues examined. If a second (MASP-1-like) message were present, we would expect the relative abundance of the common A chain to be greater than that of the MASP-3 B chain, at least in the liver. 4) No MASP-1-like proteolytic activity could be detected in MBL-MASP complexes from chicken serum. Moreover, we extended the Northern blot analysis and the analysis of enzymatic activity of MBL-MASP complexes to include other birds and obtained similar results, suggesting that MASP-1 might be absent in all Aves. The loss of the MASP-1 exons must have occurred after diversification of Reptilia and Mammalia, but might well predate the common ancestor of Crocodilia and Aves. In this context, it would be interesting to investigate whether the absence of MASP-1 extends to Crocodilia and perhaps other Reptilia.
Despite the apparent absence of MASP-1, chickens have functional MBL-MASP complexes that cleave C4, a finding that reinforces the evidence that MASP-1 is not essential for downstream complement activation (17). Nevertheless, given that MASP-1 has been conserved throughout evolution, ranging from tunicates to mammals, it seems likely that MASP-1 fulfills a biological function of which we are presently not aware.
Disclosures
The authors have no financial conflict of interest.
Acknowledgments
We are most grateful to Dr. Gennadij Raivich (Center for Perinatal Brain Protection and Repair, University College London, U.K.) for providing chicken embryos, Dr. Karsten Skjoedt (Department of Medical Microbiology, University of Southern Denmark, Odense, Denmark) for providing a chicken liver cDNA library, and Drs. Robert B. Sim (Medical Research Council Immunochemistry Unit, University of Oxford, Oxford, U.K.) and Jim Kaufman (Institute for Animal Health, Compton, U.K.) for kind attention and most valuable comments on the manuscript.
Footnotes
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 This work was supported by Wellcome Trust Grants 060574 and 062696.
2 N.J.L. and S.-u.H.K. contributed equally to this work.
3 Address correspondence and reprint requests to Dr. Wilhelm J. Schwaeble, Department of Infection, Immunity, and Inflammation, University of Leicester, University Road, Leicester, U.K. LE1 9HN. E-mail address: ws5{at}le.ac.uk
4 Abbreviations used in this paper: MBL, mannan-binding lectin; AMC, 7-amino-4-methylcoumarin; CCP, complement control protein; CRD, carbohydrate recognition domain; CUB, C1r/C1s/Uegf/bone morphogenetic protein; EGF, epidermal growth factor; MASP, MBL-associated serine protease; qRT-PCR, quantitative RT-PCR; UT, untranslated.
Received for publication November 24, 2004. Accepted for publication January 31, 2005.
References
Fujita, T.. 2002. Evolution of the lectin-complement pathway and its role in innate immunity. Nat. Rev. Immunol. 2:346
Holmskov, U., S. Thiel, J. C. Jensenius. 2003. Collections and ficolins: humoral lectins of the innate immune defense. Annu. Rev. Immunol. 21:547
Fujita, T., M. Matsushita, Y. Endo. 2004. The lectin-complement pathway: its role in innate immunity and evolution. Immunol. Rev. 198:185.
Super, M., S. Thiel, J. Lu, R. J. Levinsky, M.W. Turner. 1989. Association of low levels of mannan-binding protein with a common defect of opsonisation. Lancet 2:1236.
Summerfield, J. A., S. Ryder, M. Sumiya, M. Thursz, A. Gorchein, M. A. Monteil, M. W. Turner. 1995. Mannose binding protein gene mutations associated with unusual and severe infections in adults. Lancet 345:886.
Summerfield, J. A., M. Sumiya, M. Levin, M. W. Turner. 1997. Association of mutations in mannose binding protein gene with childhood infection in consecutive hospital series. Br. Med, J. 314:1229.
Peterslund, N. A., C. Koch, J. C. Jensenius, S. Thiel. 2001. Association between deficiency of mannose-binding lectin and severe infections after chemotherapy. Lancet 358:63738.
Neth, O., I. Hann, M. W. Turner, N. J. Klein. 2002. Deficiency of mannose-binding lectin and burden of infection in children with malignancy: a prospective study. Lancet 358:61418.
Ikeda, K., T. Sannoh, N. Kawasaki, T. Kawasaki, I. Yamashina. 1987. Serum lectin with known structure activates complement through the classical pathway. J. Biol. Chem. 262:7451.
Matsushita, M., T. Fujita. 1992. Activation of the classical complement pathway by mannose-binding protein in association with a novel C1s-like serine protease. J. Exp. Med. 176:1497.
Matsushita, M., Y. Endo, T. Fujita. 2000. Cutting edge: complement-activating complex of ficolin and mannose-binding lectin-associated serine protease. J. Immunol. 164:2281.
Matsushita, M., M. Kuraya, N. Hamasaki, M. Tsujimura, H. Shiraki, T. Fujita. 2002. Activation of the lectin complement pathway by H-ficolin (Hakata antigen). J. Immunol. 168:3502.
Sato, T., Y. Endo, M. Matsushita, T. Fujita. 1994. Molecular characterization of a novel serine protease involved in activation of the complement system by mannose-binding protein. Int. Immunol. 6:665.
Takayama, Y., F. Takada, A. Takahashi, M. Kawakami. 1994. A 100-kDa protein in the C4-activating component of Ra-reactive factor is a new serine protease having module organization similar to C1r and C1s. J. Immunol. 152:2308.
Thiel, S., T. Vorup-Jensen, C. M. Stover, W. Schwaeble, S. B. Laursen, K. Poulsen, A. C. Willis, P. Eggleton, S. Hansen, U. Holmskov, et al 1997. A second serine protease associated with mannan-binding lectin that activates complement. Nature 386:506.
Dahl, M. R., S. Thiel, M. Matsushita, T. Fujita, A. C. Willis, T. Christensen, T. Vorup-Jensen, J. C. Jensenius. 2001. MASP-3 and its association with distinct complexes of the mannan-binding lectin complement activation pathway. Immunity 15:127.
Vorup-Jensen, T., S. V. Petersen, A. G. Hansen, K. Poulsen, W. Schwaeble, R. B. Sim, K. B. Reid, S. J. Davis, S. Thiel, J. C. Jensenius. 2000. Distinct pathways of mannan-binding lectin (MBL)- and C1-complex autoactivation revealed by reconstitution of MBL with recombinant MBL-associated serine protease-2. J. Immunol. 165:2093.
Rossi, V., S. Cseh, I. Bally, N. M. Thielens, J. C. Jensenius, G. J. Arlaud. 2002. Substrate specificities of recombinant mannan-binding lectin-associated serine proteases-1 and -2. J. Biol. Chem. 276:40880.
Schwaeble, W. J., M. R. Dahl, S. Thiel, C. M. Stover, J. C. Jensenius. 2002. The mannan-binding lectin-associated serine proteases (MASPs) and MAp19: four components of the lectin pathway activation complex encoded by two genes. Immunobiology 205:455
Stover, C. M., N. J. Lynch, M. R. Dahl, S. J. Hanson, M. Takahashi, M. Frankenberger, L. Ziegler-Heitbrock, I. Eperon, S. Thiel, W. J. Schwaeble. 2003. Murine serine proteases MASP-1 and MASP-3, components of the lectin pathway activation complex of complement, are encoded by a single structural gene. Genes Immun. 4:374
Endo, Y., M. Nonaka, H. Saiga, Y. Kakinuma, A. Matsushita, M. Takahashi, M. Matsushita, T. Fujita. 2003. Origin of mannose-binding lectin-associated serine protease (MASP)-1 and MASP-3 involved in the lectin complement pathway traced back to the invertebrate, amphioxus. J. Immunol. 170:4701.
Wagner, S., N. J. Lynch, W. Walter, W. J. Schwaeble, M. Loos. 2003. Differential expression of the murine mannose-binding lectins A and C in lymphoid and nonlymphoid organs and tissues. J. Immunol. 170:1462.
Burt, D. W., C. Bruley, I. C. Dunn, C. T. Jones, A. Ramage, A. S. Law, D. R. Morrice, I. R. Paton, J. Smith, D. Windsor, et al 1999. The dynamics of chromosome evolution in birds and mammals. Nature 402:411.
International Chicken Genome Sequencing Consortium. 2004. Sequence and comparative analysis of the chicken genome provide unique perspectives on vertebrate evolution. Nature 432:695.
Matsushita, M., A. Matsushita, Y. Endo, M. Nakata, N. Kojima, T. Mizuochi, T. Fujita. 2004. Origin of the classical complement pathway: lamprey orthologue of mammalian C1q acts as a lectin. Proc. Natl. Acad. Sci. USA 101:10127.
Endo, Y., M. Takahashi, M. Nakao, H. Saiga, H. Sekine, M. Matsushita, M. Nonaka, T Fujita. 1998. Two lineages of mannose-binding lectin-associated serine protease (MASP) in vertebrates. J. Immunol. 161:4924.(Nicholas J. Lynch, Saeed-)
The lectin pathway of complement is activated by multimolecular complexes that recognize and bind to microbial polysaccharides. These complexes comprise a multimeric carbohydrate recognition subunit (either mannan-binding lectin (MBL) or a ficolin), three MBL-associated serine proteases (MASP-1, -2, and -3), and MAp19 (a truncated product of the MASP-2 gene). In this study we report the cloning of chicken MASP-2, MASP-3, and MAp19 and the organization of their genes and those for chicken MBL and a novel ficolin. Mammals usually possess two MBL genes and two or three ficolin genes, but chickens have only one of each, both of which represent the undiversified ancestors of the mammalian genes. The primary structure of chicken MASP-2 is 54% identical with those of the human and mouse MASP-2, and the organization of its gene is the same as in mammals. MASP-3 is even more conserved; chicken MASP-3 shares 75% of its residues with human and Xenopus MASP-3. It is more widely expressed than other lectin pathway components, suggesting a possible function of MASP-3 different from those of the other components. In mammals, MASP-1 and MASP-3 are alternatively spliced products of a single structural gene. We demonstrate the absence of MASP-1 in birds, possibly caused by the loss of MASP-1-specific exons during phylogeny. Despite the lack of MASP-1-like enzymatic activity in sera of chicken and other birds, avian lectin pathway complexes efficiently activate C4.
Introduction
The lectin pathway provides an Ab-independent route of complement activation. It is an ancient system that antedates the evolution of adaptive immunity and the classical complement pathway, and it is thought to be particularly important in lower animals such as the urochordates and cephalochordates, which lack an adaptive immune system. In vertebrates, the lectin pathway provides a first line of defense during the lag phase that precedes the onset of an adaptive response (1, 2, 3). Deficiencies of the lectin pathway are associated with susceptibility to infectious disease, particularly in infants, young children, and those with another acquired immunodeficiency, such as cancer patients undergoing chemotherapy (4, 5, 6, 7, 8).
Activation of the lectin pathway occurs when a multimolecular fluid phase activation complex, comprising a recognition subcomponent and associated serine proteases, binds to carbohydrate structures present on microbial surfaces. Two types of fluid phase recognition molecule have been described, the mannan-binding lectin(s) (MBL),4 and the ficolins (9, 10, 11, 12). Structurally they are very similar, being homotrimers of a single polypeptide chain, with an N-terminal collagen-like domain and a C-terminal carbohydrate recognition domain (CRD). In MBL, the CRD is a C-type lectin domain, whereas the CRD in ficolins is a fibrinogen-like domain. In plasma, the basic homotrimers form higher order oligomers, which, in turn, form complexes with the MBL-associated serine proteases, MASP-1, -2, and -3 (13, 14, 15, 16). Only MASP-2 cleaves C4 and C4b-bound C2, generating C4b2a, the same C3 convertase that results from activation of the classical pathway (17, 18). The roles of MASP-1 and -3 in these complexes are still unclear.
All three MASPs share a conserved domain structure. Five N-terminal domains (the C1r/C1s/Uegf/bone morphogenetic protein I (CUBI), epidermal growth factor (EGF)-like, CUBII, complement control protein I (CCPI), and CCPII domains) precede a C-terminal serine protease domain. When activated, the enzymes split into two disulfide-linked chains, the A chain, which contains the five N-terminal domains, and the B chain, which contains the serine protease domain (2, 19). The mRNAs encoding MASP-1 and -3 are alternatively spliced products of one single gene. The A chain is common to both proteins, the B chain of MASP-1 is encoded by five or six exons (depending upon the species), and the B chain of MASP-3 is encoded by a single exon, located between the exons encoding the shared A chain and those coding for the B chain of MASP-1 (16, 19, 20). This genomic arrangement is found in amphioxus (a cephalochordate) and in all vertebrates investigated to date, but not in the ascidians, indicating that it arose after the divergence of the urochordates, but before the divergence of the cephalochordates (21). The lectin pathway activation complexes also contain MAp19, a protein composed of the two N-terminal domains of MASP-2 encoded by a truncated mRNA transcript that arises through alternative splicing of the MASP2 gene (19, 22, 23, 24). MAp19 has no enzymatic activity, and its function within the lectin pathway activation complexes remains unclear.
Commercial poultry are vulnerable to bacterial, viral, and parasitic infections that cause considerable mortality and economic loss. They are also reservoirs for transmissible human pathogens, including Salmonella, Campylobacter, and potentially dangerous stains of influenza, such as the H5N1 (Guangdong) virus that caused the 1997 bird flu outbreak in Hong Kong (25, 26). Although the adaptive immune system of the chicken has been extensively studied (27), the complement system and, in particular, the lectin pathway, is less well understood. It is known that Ab-independent activation of the complement system plays an important role in the host response to fowlpox virus; fowlpox virus infection is aggravated in chickens treated with cobra venom factor (28).
The classical pathway components C1 and C3 through C9 have been found in chickens (29). Because a homologue for C2 was not found in these early studies, it was assumed that the role of C2 may be fulfilled by the chicken factor B-like protease, an evolutionary remnant of a common C2/factor B ancestor described previously (29, 30). However, subsequent studies identified and mapped the chicken homologues of both the C2 and factor B genes and showed that chickens are not deficient in C2 (K. Skjoedt and J. Kaufman, unpublished observations). MBL is the only lectin pathway component that has been described in the chicken to date. Although most mammals have two separate genes for MBL (31), only one gene was found in the chicken, and phylogenetic analysis of its amino acid sequence indicated that it may represent an undiversified ancestor of the mammalian MBLs (32). The concentration of MBL in chicken serum ranges from 0.4 to 37.8 μg/ml (mean, 5.8 μg/ml), with no strain-to-strain variation and no deficiencies detected (33). MBL levels increase 2-fold in chickens infected with infectious bronchitis virus, infectious laryngotracheitis virus, and infectious bursal disease virus, indicating that MBL is a mild acute phase protein (34).
This report describes the molecular cloning of chicken MASP-2, MASP-3, and MAp19 and the organization of their genes and those for chicken MBL and a novel ficolin. We demonstrate the apparent absence of MASP-1 and MASP-1-like enzymatic activity in chickens and other birds.
Materials and Methods
RNA extraction and cDNA synthesis
Total RNA was prepared using TRIzol reagent (Invitrogen Life Technologies), according to the manufacturer’s instructions. Poly(A)+ mRNA was purified from total RNA using the Oligotex mRNA Midi kit, from Qiagen.
For cDNA synthesis, 8 μg of total RNA was digested with RNase-free DNase I (Promega), extracted once with phenol/chloroform/isoamyl alcohol (25/24/1), precipitated, washed with ethanol, then redissolved in water. The concentrations of the RNA samples were determined by measuring the absorbance at 260 nm. Three micrograms of each RNA sample was primed with oligo(dT)12–18, and single-stranded cDNAs were synthesized using the SuperScript First-Strand Kit (Invitrogen Life Technologies) according to the supplier’s instructions.
cDNA cloning
A ZAP chicken liver cDNA library (Stratagene) was screened by filter hybridization using [-32P]dCTP-labeled cDNA probes prepared with the Random Primed Labeling kit (Roche). Isolated ZAP clones were converted into pBluescript SK– plasmid clones by in vivo excision, following the supplier’s protocol, and sequenced (Genterprise). The 5' and 3' ends of some cDNA sequences were extended by RACE, using chicken liver RNA and kits (Invitrogen Life Technologies). RACE products were cloned into pGEM-Teasy (Promega) and sequenced.
Northern blotting
For Northern blot analysis, 10 μg of total RNA or 100 ng of poly(A+) mRNA was separated on formaldehyde-containing 1.2% agarose gels, transferred to nylon membrane, and hybridized with [-32P]dCTP-labeled cDNA probes using standard methods (35). The cDNA probes used were: for the A chain of MASP-2, a 376-bp RT-PCR product amplified from chicken liver cDNA with primers M2A-ch-F and M2A-ch-R (Table I); for the B chain of MASP-2, a 430-bp RT-PCR product generated with primers M2dg-F2 and M2dg-R1; for the A chain of MASP-3, a 740-bp RT-PCR product amplified with primers M1&3dg-F and M1&3dg-R; and for the B chain of MASP-3, a 1200-bp NcoI fragment of clone CM3/5, encompassing the last 500 bp of the coding sequence and the first 700 bp of the 3'-untranslated (3'UT) region. For controls, rat liver RNA was hybridized with rat MASP-3 A and B chain-encoding cDNA probes, as previously described (20).
Table I. Oligonucleotides used in this study
In situ hybridization
mRNA was localized by in situ hybridization using 35S-labeled antisense RNA probes generated by in vitro transcription. The templates for in vitro transcription were RT-PCR products, amplified from liver cDNA and cloned into pGEM-Teasy. For the A chain of chicken MASP-3 and for the B chain of chicken MASP-2, the cloned RT-PCR products were those described above. For the B chain of chicken MASP-3, a 564-bp RT-PCR product was generated using primers M3SP-F1 and -R1 (Table I); for chicken MAp19, a 170-bp product was obtained using primers MAp19-F1 and -R1; for chicken MBL, a 330-bp product was generated with primers MBL-F1 and MBL-R1; for chicken ficolin, a 324-bp product was amplified using primers ChkFCN_F1 and ChkFCN_R1; for the B chain of rat MASP-1, a 400-bp product was generated using primers RM1sp-F1 and RM1sp-R1; and for the B chain of rat MASP-3, a 407-bp product was obtained with primers RM3-BCF1 and RM3-BCR1. Antisense RNA probes were transcribed from the cloned templates using the method described by Melton et al. (36). For controls, the templates were transcribed in the opposite direction, generating sense RNA probes. Twenty-micron-thick tissue sections were cut using a cryostat, mounted on poly-lysine-coated microscope slides (Merck), and hybridized with the 35S-labeled probes according to the method of Sch?fer et al. (37). After hybridization and washing, the signals were detected by exposing the sections to Kodak BioMax MR x-ray film (Sigma-Aldrich).
Analysis of gene expression by quantitative RT-PCR (qRT-PCR)
cDNA was analyzed by quantitative PCR using a LightCycler (Roche) to follow the incorporation of SYBR Green I into the PCR products in real-time. Each 15 μl of PCR contained 1/100th of the original cDNA synthesis reaction (corresponding to 30 ng of total RNA), 0.5 μM of each primer, and 7.5 μl of QuantiTect SYBR Green Master Mix (Qiagen). Forty-five cycles of amplification were performed; the annealing temperature was reduced from 70 to 58°C during the first 15 cycles and was kept constant at 58°C thereafter. The fluorescence signal was detected at the end of each cycle, and results were analyzed using the Fit Points option in the LDCA software supplied with the machine (38). Melting curve analysis was used to confirm the specificity of the products (39). The primers used to amplify the A chain of MASP-3 were M3Ach-F1 and -R1 (generating a 261-bp product); for the B chain of MASP-3, M3SP-F1 and M3SP-R1 (564 bp); for MAp19, MAp19-F1 and MAp19-R1 (170 bp); for MASP-2, M2A-ch-F and M2A-ch-R (376 bp); for MBL, MBL-F1 and MBL-R1 (330 bp); and for chicken ficolin, ChkFCN-F1 and ChkFCN-R1. Standard curves were produced for each analysis using serial dilutions of the corresponding cDNAs cloned in pBluescript SK– (for MASP-2 and -3) or pGEMTeasy (for MAp19 and MBL).
Enzyme assays
Mannan-coated microtiter plates were used to capture MBL-MASP complexes from pooled normal human serum, chicken serum (from a pool of five animals), and turkey serum (also pooled from five animals). C4 cleavage activity of the captured complexes was assayed using the method developed by Petersen et al. (40), with minor modifications (41). MASP-1-like activity was measured using the fluorescent substrate, butyloxycarbonyl-Val-Pro-Arg-7-amino-4-methylcoumarin (AMC), as previously described (42).
Sequence analysis
The chicken genome database is available at www.ensembl.org/Gallus_gallus/. Similarity searches with nucleotide and polypeptide query sequences were run on this server using BLASTN and TBLASTN, respectively. The MutliContigView option was used to create chicken/human synteny maps. Signal peptides and N-linked glycosylation sites were predicted using SignalP v3.0 and NetNGlyc v1.0, which can be found at www.cbs.dtu.dk/services/.
Results
cDNA cloning
Degenerate oligonucleotides were designed to match parts of the MASP-1/3 A chain and the B chain of MASP-2 that are conserved in the human, rat, mouse, and Xenopus laevis sequences. The MASP-1/3 primers (M1&3dg-F and M1&3dg-R; Table I) were used to amplify a 740-bp fragment of cDNA from chicken liver RNA by RT-PCR, and the MASP-2 specific primers (M2dg-F2 and M2dg-R1) were used to amplify a 430-bp fragment from the same source. Both products were cloned into pGEM-Teasy and then sequenced to confirm that they encode the anticipated regions of the chicken MASP cDNAs. The RT-PCR products were subsequently labeled with [-32P]dCTP and used to screen a proprietary chicken liver cDNA library.
Seven clones were obtained using the MASP-1/3 A chain probe. Sequencing showed that all seven were incomplete MASP-3 cDNAs, the longest of which (clone CM3/5) was 3343-bp long and was complete at its 3' end, but began with DNA encoding the middle of the CUBI domain, 300 bp downstream of the expected transcription start. Where they overlapped, the sequences of the other six clones were identical with that of CM3/5. The 5' end of the MASP-3 sequence was completed by RACE, using chicken liver RNA and the primers M3race-R1, -R2, and -R3. A 565-bp RACE product was obtained, extending the known sequence by 293 bp.
The complete chicken MASP-3 cDNA (GenBank accession no. AY567829) is 3636 bp long and comprises a 159-bp 5'UT region, followed by a 2190-bp open reading frame, a stop codon TGA, and a 3'UT region of 1284 bp. A putative polyadenylation signal is located 13 bp upstream of the poly(A)+ tail. Analysis of the cDNA-derived amino acid sequence revealed extensive similarity between chicken MASP-3 and MASP-3 from other species. The chicken sequence shares 75% of its residues with the human sequence, 76% with the mouse sequence, and 75% with the Xenopus laevis MASP-3a sequence (Table II). The isoleucine/arginine cleavage site that separates the A and B chains is conserved, as are three residues that are important for catalytic activity (His, Asp, and Ser, the so-called catalytic triad). All the N-glycosylation sites are also conserved, as are all the cysteine residues, including the two that form the methionine loop, a cystine bond around a methionine residue in the serine protease domain (Fig. 1).
Table II. Cross-species comparison of MASP-3, MASP-2, and MAp19 peptide sequences (percentage of identical residues).
FIGURE 1. Alignment of the cDNA-derived peptide sequences of chicken, human, and mouse MASP-3 and X. laevis MASP-3a. Predicted leader peptides are italicized, potential N-glycosylation sites are underlined, conserved residues are marked with an asterisk, conservative substitutions are marked with a colon, and conserved cysteines are shown in bold. The arginine/isoleucine cleavage site that separates the A and B chains is indicated by an arrow, and the three residues that form the catalytic triad are boxed.
No MASP-1-like clones were obtained, either directly from the cDNA library or by 3'RACE using primers located in the A chain of MASP-3 (which is shared with MASP-1 in other species).
The largest clone obtained using the MASP-2 B chain probe was 2082 bp long, which did not comprise the full-length coding sequence. The 5' end of this clone extended to the cDNA encoding the CUBII domain. 5'RACE using primers M2race-R1, -R3, and -R4 gave a 601-bp product, which extended the known sequence by 336 bp and included the initiation codon. The complete MASP-2 cDNA (GenBank accession no. AY567828) is 2449 bp long and includes a 52-bp 5'UT region, a 2058-bp open reading frame, the stop codon, and a 336-bp 3'UT region, with a polyadenylation signal 19 bp upstream of the poly(A)+ tail. All the important features found in the primary structure of human, mouse, and Xenopus MASP-2 are conserved in chicken MASP-2 (Fig. 2), although the degree of identity among the MASP-2 amino acid sequences is less than that among the MASP-3 sequences (Table II).
FIGURE 2. The cDNA-derived amino acid sequences of chicken MASP-2 and MAp19. Predicted leader peptides are italicized. Features that are conserved in other species include the two N-glycosylation sites (underlined), all the cysteine residues shown in bold, the three residues that form the catalytic triad in MASP-2 (boxed), and the arginine/isoleucine cleavage site that separates the A and B chains of MASP-2 (indicated by an arrow).
Chicken MAp19 cDNA was cloned by RT-PCR. First, the 3' sequence was obtained by 3'RACE using primers located within the DNA encoding the CUBI domain of MASP-2 (MAp-race-F1 and MAp-race-F2). A 440-bp RACE product was obtained, cloned into pGEM-Teasy, and sequenced. Next, primers representing the 5' end of MASP-2 (MAp19-F2) and the 3' end of the MAp19 RACE product (MAp19-R2) were used to amplify full-length MAp19 cDNA from chicken liver cDNA. The first 588 bp of chicken MAp19 cDNA are identical with those of the MASP-2 cDNA and encode the CUBI and EGF-like domains (Fig. 2). The remainder of the cDNA is unique to MAp19 and includes the third nucleotide of the codon for the last residue, the stop codon, and a 3'UT region of 177 bp (GenBank accession no. AY567830). Mammalian MAp19 proteins have a four-residue C-terminal extension (EQSL) that is absent in the corresponding MASP-2 sequences (23, 24). Chicken MAp19 lacks this feature; its entire amino acid sequence is contained within the MASP-2 sequence.
Organization of the lectin pathway genes
The genes for chicken MASP-2 and MASP-3 were characterized by comparing the cDNA sequences with the recently completed chicken genome sequence.
The MASP-2 gene occupies 14 kb of chromosome 21, and its intron/exon structure is identical with that reported for the human and mouse MASP-2 genes (23, 24). Exon 1 encodes the 5'UT region and the initiation codon, exons 2 and 3 encode the signal peptide and the CUBI domain, and exon 4 encodes the EGF-like domain. Splicing between exons 4 and 5 generates the MAp19 mRNA. In mammals, the fifth exon encodes the last four residues of MAp19 (EQSL), the stop codon, and the 3'UT region. In the chicken, exon 5 only contains the last nucleotide of the coding sequence, followed by the stop codon and the 3'UT region. Exons 6 and 7 encode the CUBII domain of MASP-2, exons 8–11 encode the two CCP domains, and exon 12 encodes the serine protease domain (Table III). The codon phases are conserved between the chicken and human genes (43). The region of chromosome 21 that contains the chicken MASP2 gene is syntenic with the MASP2 loci on human chromosome 1p36, mouse chromosome 4E1, and rat chromosome 5q36. All four loci contain the genes TARDBP, MASP2, SRM, PMSCL2, and FRAP1 (or their homologues), in that order (43, 44).
Table III. Structure of the chicken MASP2 gene
The gene for chicken MASP-3 is located on chromosome 9. It comprises 11 exons, the organization of which is identical with that of the first 11 exons in the human and mouse MASP1/3 genes. The first 10 exons encode the 5'UT region and the A chain of MASP-3, and the 11th exon encodes the serine protease domain, stop codon, and 3'UT region. In the mammalian genes, exon 11 is followed by six additional exons, which encode the (TCN-type) serine protease domain of MASP-1. We looked for the equivalent exons in the chicken using the peptide sequences from the serine protease domains of human, mouse, and X. laevis MASP-1 as query sequences for TBLASTN searches of the chicken genome database. There is no evidence for MASP-1-specific exons in the chicken genome, either 3' of the MASP-3 gene or elsewhere in the genome. (The closest hits were actually MASP-2 and anticoagulant protein C.)
Examination of the sequence surrounding the MASP-3 gene revealed a possible explanation for the absence of MASP-1-specific exons in the chicken genome. The region 5' of the chicken MASP-3 gene, which includes the chicken homologues of hsBCL6 and hsSST, is syntenic with the human MASP1/3 locus on chromosome 3q27, but the synteny ends immediately 3' of the MASP-3 gene (Fig. 3). TM7L_HUMAN is 20 kb downstream of the human MASP1/3 gene, followed by RPL39L and SIAT1. In the chicken, the homologues of TM7L_HUMAN and RPL39L are missing altogether, and the homologue of SIAT1 (SAI1_chick) is found 5.8 Mb 5' of the chicken MASP-3 gene. Likewise, ENSGALG00000007474 and TFR1_chick are found 3' of the chicken MASP-3 gene, but their human homologues (PCYTIA and TFRC) are located 9 Mb upstream of the human MASP1/3 gene. These findings suggest that an early translocation event separated the MASP-1-specfic exons from the 3' end of the original MASP-1/3 gene, and that the MASP-1-specific exons together with the homologues of TM7L_HUMAN and RPL39L were either lost during translocation or disappeared at a later stage.
FIGURE 3. Comparison of the chicken MASP-3 and human MASP1/3 loci. Homologous genes are joined by dashed lines, and arrowheads indicate the direction of transcription. This figure is not to scale.
We also identified genes encoding the two lectin pathway recognition molecules in the chicken, MBL and a ficolin. Most mammals have two functional MBL genes, located on separate chromosomes, although in humans, MBL1 is a pseudogene, and only MBL2 produces a protein (31). The chicken has just one MBL gene, which is located on chromosome 6 in a conserved cluster that includes the gene for pulmonary surfactant protein D. It is homologous to the human pseudogene (MBL1) and the mouse and rat genes (Mbl1 and MABA_RAT).
Humans have two ficolin genes, FCN1 and FCN2, arranged back-to-back on chromosome 9q34, and a third, FCN3, on chromosome 1p35. FCN1 encodes M-ficolin, which is expressed on the surface of immature macrophages, whereas FCN2 and FCN3 encode L-ficolin and H-ficolin, which are both plasma proteins. The mouse homologues of FCN1 and FCN2 (Fcna and Fcnb, respectively) are arranged back-to-back on chromosome 2. Chickens have a single ficolin gene located on chromosome 17, which appears to represent an undiversified ancestor of FCN1 and FCN2. Its putative protein sequence is 60% identical with that of L-ficolin (the product of FCN2) and 57% identical with that of M-ficolin (the product of FCN1).
Expression of the lectin pathway components
In situ hybridization was used to analyze mRNA expression in embryonic tissue (Fig. 4); Northern blotting and qRT-PCR were used to analyze adult tissue (Figs. 5–7).
FIGURE 4. Localization of lectin pathway mRNAs by in situ hybridization. Cryostat sections (20 μm) were hybridized with 35S-labeled cRNA probes and exposed to Kodak BioMax MR film for 48 h. A–D, Day 12 chicken embryos hybridized with antisense probes for: MASP-2 B chain, MAp19, MBL, MASP-3 A chain, and MASP-3 B chain. E, Control section hybridized with a sense probe for MBL. F and G, Day 19 rat embryos hybridized with antisense probes for the B chains of MASP-1 and MASP-3. H, Control section hybridized with a sense probe for MASP-1 B chain. I and J, In situ hybridization results for chicken embryo sections obtained with antisense probes specific for the chicken lectin pathway recognition molecules MBL (I) and ficolin (J).
FIGURE 5. Expression of MASP-2 and MAp19 mRNA in chicken liver. Northern blots, prepared from chicken liver RNA, were hybridized with cDNA probes corresponding to the A chain (A) and the B chain (B) of MASP-2.
FIGURE 6. The qRT-PCR analysis of MASP-3 expression in adult chicken tissues. cDNA was prepared from the tissues indicated and was analyzed by real-time PCR using a LightCycler instrument. Results shown are copies per microgram of RNA and are the means of four separate experiments (two cDNA syntheses per sample, two analyses per cDNA). Error bars represent the SD.
FIGURE 7. Expression of MASP-3 mRNA in the livers of different birds. Poly(A)+ mRNA was prepared from the species indicated and analyzed by Northern blotting. Blots were hybridized with -32P-labeled cDNA probes corresponding to the A chain (A) and the B chain (B; serine protease domain) of chicken MASP-3. As a control, rat liver RNA was hybridized with the corresponding rat-specific probes encoding either the A or B chain of rat MASP-3 (right lanes). Distinct signals for MASP-1 (5 kb) and MASP-3 (3.5 kb) were seen using the A chain probe.
In the chicken, as in other species (23, 24), MASP-2 and MAp19 mRNA expression is restricted to liver tissue (Fig. 4, A and B). Northern blotting using a cDNA probe comprising the 5'-coding sequence for the A chain of MASP-2 revealed two mRNA species in adult liver, the 2.5-kb MASP-2 message and the 0.8-kb MAp19 mRNA (Fig. 5A). Using a probe comprising the coding sequence for the B chain, only the 2.5-kb MASP-2 message was seen (Fig. 5B). The qRT-PCR analysis showed that MAp19 mRNA is 3.5 times more abundant than the MASP-2 message in adult liver (data not shown).
Chicken MBL mRNA is restricted to the liver in both embryos (Fig. 4I) and adult tissue. This is in contrast to murine and human MBL, which is expressed in hepatic and extrahepatic tissues (45). The putative chicken ficolin gene is expressed in embryonic and adult liver tissue only (see Fig. 4J).
In humans, rats, and mice, MASP-1 expression is liver specific, whereas MASP-3 is expressed in the liver and in nonhepatic tissues, including the spleen, lung, small intestine, thymus, and brain (Fig. 4, F and G) (N. J. Lynch, unpublished observations). In day 12 chicken embryos, MASP-3 mRNA was detected by in situ hybridization in the liver, spleen, thymus, gizzard, mesonephros, and neuronal areas of the brain, but not in the lung. The same distribution and intensity of expression were observed using separate cRNA probes for the A and B chains of MASP-3 (Fig. 4, C and D). This result underlines the findings of the Northern blot analysis, showing that no evidence for a MASP-1-like transcript can be found.
This is also confirmed by the results obtained by qRT-PCR analysis of adult chicken tissues: MASP-3 expression was most abundant in the liver, followed by spleen, kidney, tonsil, duodenum, and brain. Again, using a combination of oligonucleotides that amplifies mRNA sequences specific for either the A chain or the B chain shows that identical abundances for MASP-3 A chain- and MASP-3 B chain-encoding mRNAs were detected and provided no evidence for the presence of a MASP-1-like transcript. Moreover, the amount of A and B chain message was identical in all tissues tested (Fig. 6).
Northern blot analysis using separate probes for the A and B chains of MASP-3 revealed a single 3.7-kb message in chicken liver, reinforcing the evidence that the MASP-1 transcript is missing (Fig. 7, first lane). This is in contrast to what was seen in mammals, where both MASP-3 and MASP-1 mRNA transcripts were detected (Fig. 7, upper last lane) (20). Analyzing liver mRNA from other birds (including turkey, duck, goose, ostrich, and pigeon) on Northern blots suggested that the absence of MASP-1 may be a general phenomenon in birds (Fig. 7).
Enzymatic activity of MBL-MASP complexes in avian sera
We tested the enzymatic activity of MBL-MASP complexes captured from chicken and turkey sera. Serum samples were diluted in a high salt buffer, which dissociates C1, thus avoiding interference from the classical pathway serine proteases, C1r and C1s (40). The diluted sera were added to ELISA plates coated with mannan, a polysaccharide that binds MBL-MASP complexes, then the bound complexes were assayed for MASP-1 and MASP-2. MASP-1-like activity was measured using the fluorescent substrate, Val-Pro-Arg-7-AMC, which is cleaved by MASP-1, but not by MASP-2 (42). The activity of MASP-2 was measured by following the cleavage of human C4.
MASP-1-like activity could be detected in MBL-MASP complexes from human serum, which was included as a control, but not in the avian sera, whereas MASP-2-like activity was present in all sera tested (Fig. 8). Subsequent experiments showed that MBL-MASP complexes from goose and ostrich sera also lack MASP-1-like enzymatic activity (data not shown).
FIGURE 8. Enzymatic activity of MBL-MASP complexes from human and avian sera. Mannan-coated microtiter plates were used to capture MBL-MASP complexes from serum. The immobilized complexes were assayed for MASP-1-like activity using the fluorescent substrate Val-Pro-Arg-7-AMC. To assay MASP-2, human C4 was added to the plates, and the deposition of C4b was measured. Results shown are percentages of the activity measured in human serum and are the means of duplicate determinations.
Discussion
Most vertebrates have three or four different lectin pathway recognition molecules, with differing, but overlapping, carbohydrate specificities that allow them to recognize a broad spectrum of microbial polysaccharides. The recognition molecules form complexes with three MASPs and MAp19 (the truncated, nonenzymatic form of MASP-2). Only one of these molecules has previously been reported in the chicken, namely MBL. In this study we present the most complete description of the lectin pathway in the chicken to date.
Laursen et al. (31) cloned the cDNA for chicken MBL and (based on a phylogenetic comparison of its deduced amino acid sequence with those of its mammalian homologues) concluded that it represents an undiversified ancestor of the two different forms of MBL found in mammals. Our findings support their conclusion; the chicken has a single MBL gene located on chromosome 6. Genomic analysis showed that it is more closely related to MBL1, Mbl1, and MABA-RAT than to MBL2, Mbl2, and MABC_RAT, suggesting that the first group represents the common ancestor. Likewise, the chicken has a single ficolin gene that appears to represent an undiversified ancestor of FCN1 and FCN2. It is not particularly surprising that the chicken has so few lectin pathway recognition molecules; the gene duplications that lead to the diversification of MBL and the ficolins were relatively recent evolutionary events (1), and such chromosomal rearrangements are thought to occur much more slowly in birds (and reptiles) than in mammals or amphibians (46, 47). Interestingly, chicken C1q (the recognition molecule of the classical pathway) appears to be a homotrimer, encoded by a single gene on chromosome 21. In this respect, it is more closely related to the ancestral form found in the lamprey than to mammalian C1q, which is a heterotrimer of three different polypeptides encoded by three separate genes (48).
The serine proteases found in the classical and lectin pathway activation complexes can be divided into two evolutionary groups: the ancestral TCN type (which includes MASP-1 and the ascidian MASPs) has a TCN codon for the serine in its catalytic triad and split exons for the protease domain, whereas the AGY type (which includes MASP-2, MASP-3, and C1s) uses AGY to encode its active serine and has a single exon for the protease domain (49). The AGY type evolved after, and appears to be derived from, the gene rearrangement that gave rise to the alternatively spliced MASP-1/3 gene (1, 21).
As expected, chicken MASP-2 is an AGY-type serine protease. Its primary structure shows 54% identity with those of human and mouse MASP-2. The organization of its gene is highly conserved; the intron-exon structure and codon phases are identical with those of the human and mouse genes (Tables II and III). As in other species, chicken MAp19 is an alternatively spliced product of the MASP-2 gene. It is slightly unusual in that it lacks the four-residue C-terminal extension found in mammalian MAp19 (22, 23, 24). In mammals these four residues (EQSL) are encoded by exon 5, which is unique to MAp19; in the chicken, exon 5 encodes only the third nucleotide of the last residue (which is identical with that of MASP-2), the stop codon, and the 3'UT region.
MASP-3 is the most highly conserved protein in the activation complexes; the primary structure of the chicken MASP-3 is 75% identical with those of the human, mouse, and Xenopus proteins. Its expression is more widespread than that of the other lectin pathway components; significant amounts of MASP-3 mRNA are present in liver, spleen, thymus, neuronal tissue, and gastrointestinal tissues, not only in chickens, but also in mice, rats and humans (Figs. 4 and 6) (N. J. Lynch, unpublished observations). In contrast, the messages for MASP-1, MASP-2, and MAp19 mRNA are restricted to the liver in all species investigated to date, and those for MBL and the plasma ficolins are mainly expressed in the liver (2, 20, 24, 45). The conservation of MASP-3 and its unique expression pattern suggest that its function may be decidedly different from those of the other lectin pathway components.
The 5' end of the chicken MASP-3 gene is similar to those of the mammalian MASP-1/3 genes and the Xenopus MASP1/3a gene; 10 exons encoding the A chain precede a single (AGY-type) exon that encodes the serine protease domain. However, the mammalian genes have an additional six exons that encode the (TCN-type) MASP-1 serine protease domain. In the chicken these exons appear to have been lost as a result of translocation and simultaneous or subsequent deletion that removed not only the MASP-1-specific exons, but also the chicken homologues of TM7L_HUMAN and RPL39L (Fig. 3). We found four separate pieces of evidence indicating that MASP-1 is absent in the chicken. 1) There are no MASP-1-specific exons, either 3' of MASP-3 gene or elsewhere the in the chicken genome. 2) Northern blots hybridized with a probe for the A chain of MASP-3 show a single mRNA species; two messages would be expected if the chicken had an alternatively spliced MASP-1/3 gene. 3) Similarly, qRT-PCR showed that the amount of MASP-3 A chain-encoding message is approximately equal to that of the B chain-encoding message in all tissues examined. If a second (MASP-1-like) message were present, we would expect the relative abundance of the common A chain to be greater than that of the MASP-3 B chain, at least in the liver. 4) No MASP-1-like proteolytic activity could be detected in MBL-MASP complexes from chicken serum. Moreover, we extended the Northern blot analysis and the analysis of enzymatic activity of MBL-MASP complexes to include other birds and obtained similar results, suggesting that MASP-1 might be absent in all Aves. The loss of the MASP-1 exons must have occurred after diversification of Reptilia and Mammalia, but might well predate the common ancestor of Crocodilia and Aves. In this context, it would be interesting to investigate whether the absence of MASP-1 extends to Crocodilia and perhaps other Reptilia.
Despite the apparent absence of MASP-1, chickens have functional MBL-MASP complexes that cleave C4, a finding that reinforces the evidence that MASP-1 is not essential for downstream complement activation (17). Nevertheless, given that MASP-1 has been conserved throughout evolution, ranging from tunicates to mammals, it seems likely that MASP-1 fulfills a biological function of which we are presently not aware.
Disclosures
The authors have no financial conflict of interest.
Acknowledgments
We are most grateful to Dr. Gennadij Raivich (Center for Perinatal Brain Protection and Repair, University College London, U.K.) for providing chicken embryos, Dr. Karsten Skjoedt (Department of Medical Microbiology, University of Southern Denmark, Odense, Denmark) for providing a chicken liver cDNA library, and Drs. Robert B. Sim (Medical Research Council Immunochemistry Unit, University of Oxford, Oxford, U.K.) and Jim Kaufman (Institute for Animal Health, Compton, U.K.) for kind attention and most valuable comments on the manuscript.
Footnotes
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 This work was supported by Wellcome Trust Grants 060574 and 062696.
2 N.J.L. and S.-u.H.K. contributed equally to this work.
3 Address correspondence and reprint requests to Dr. Wilhelm J. Schwaeble, Department of Infection, Immunity, and Inflammation, University of Leicester, University Road, Leicester, U.K. LE1 9HN. E-mail address: ws5{at}le.ac.uk
4 Abbreviations used in this paper: MBL, mannan-binding lectin; AMC, 7-amino-4-methylcoumarin; CCP, complement control protein; CRD, carbohydrate recognition domain; CUB, C1r/C1s/Uegf/bone morphogenetic protein; EGF, epidermal growth factor; MASP, MBL-associated serine protease; qRT-PCR, quantitative RT-PCR; UT, untranslated.
Received for publication November 24, 2004. Accepted for publication January 31, 2005.
References
Fujita, T.. 2002. Evolution of the lectin-complement pathway and its role in innate immunity. Nat. Rev. Immunol. 2:346
Holmskov, U., S. Thiel, J. C. Jensenius. 2003. Collections and ficolins: humoral lectins of the innate immune defense. Annu. Rev. Immunol. 21:547
Fujita, T., M. Matsushita, Y. Endo. 2004. The lectin-complement pathway: its role in innate immunity and evolution. Immunol. Rev. 198:185.
Super, M., S. Thiel, J. Lu, R. J. Levinsky, M.W. Turner. 1989. Association of low levels of mannan-binding protein with a common defect of opsonisation. Lancet 2:1236.
Summerfield, J. A., S. Ryder, M. Sumiya, M. Thursz, A. Gorchein, M. A. Monteil, M. W. Turner. 1995. Mannose binding protein gene mutations associated with unusual and severe infections in adults. Lancet 345:886.
Summerfield, J. A., M. Sumiya, M. Levin, M. W. Turner. 1997. Association of mutations in mannose binding protein gene with childhood infection in consecutive hospital series. Br. Med, J. 314:1229.
Peterslund, N. A., C. Koch, J. C. Jensenius, S. Thiel. 2001. Association between deficiency of mannose-binding lectin and severe infections after chemotherapy. Lancet 358:63738.
Neth, O., I. Hann, M. W. Turner, N. J. Klein. 2002. Deficiency of mannose-binding lectin and burden of infection in children with malignancy: a prospective study. Lancet 358:61418.
Ikeda, K., T. Sannoh, N. Kawasaki, T. Kawasaki, I. Yamashina. 1987. Serum lectin with known structure activates complement through the classical pathway. J. Biol. Chem. 262:7451.
Matsushita, M., T. Fujita. 1992. Activation of the classical complement pathway by mannose-binding protein in association with a novel C1s-like serine protease. J. Exp. Med. 176:1497.
Matsushita, M., Y. Endo, T. Fujita. 2000. Cutting edge: complement-activating complex of ficolin and mannose-binding lectin-associated serine protease. J. Immunol. 164:2281.
Matsushita, M., M. Kuraya, N. Hamasaki, M. Tsujimura, H. Shiraki, T. Fujita. 2002. Activation of the lectin complement pathway by H-ficolin (Hakata antigen). J. Immunol. 168:3502.
Sato, T., Y. Endo, M. Matsushita, T. Fujita. 1994. Molecular characterization of a novel serine protease involved in activation of the complement system by mannose-binding protein. Int. Immunol. 6:665.
Takayama, Y., F. Takada, A. Takahashi, M. Kawakami. 1994. A 100-kDa protein in the C4-activating component of Ra-reactive factor is a new serine protease having module organization similar to C1r and C1s. J. Immunol. 152:2308.
Thiel, S., T. Vorup-Jensen, C. M. Stover, W. Schwaeble, S. B. Laursen, K. Poulsen, A. C. Willis, P. Eggleton, S. Hansen, U. Holmskov, et al 1997. A second serine protease associated with mannan-binding lectin that activates complement. Nature 386:506.
Dahl, M. R., S. Thiel, M. Matsushita, T. Fujita, A. C. Willis, T. Christensen, T. Vorup-Jensen, J. C. Jensenius. 2001. MASP-3 and its association with distinct complexes of the mannan-binding lectin complement activation pathway. Immunity 15:127.
Vorup-Jensen, T., S. V. Petersen, A. G. Hansen, K. Poulsen, W. Schwaeble, R. B. Sim, K. B. Reid, S. J. Davis, S. Thiel, J. C. Jensenius. 2000. Distinct pathways of mannan-binding lectin (MBL)- and C1-complex autoactivation revealed by reconstitution of MBL with recombinant MBL-associated serine protease-2. J. Immunol. 165:2093.
Rossi, V., S. Cseh, I. Bally, N. M. Thielens, J. C. Jensenius, G. J. Arlaud. 2002. Substrate specificities of recombinant mannan-binding lectin-associated serine proteases-1 and -2. J. Biol. Chem. 276:40880.
Schwaeble, W. J., M. R. Dahl, S. Thiel, C. M. Stover, J. C. Jensenius. 2002. The mannan-binding lectin-associated serine proteases (MASPs) and MAp19: four components of the lectin pathway activation complex encoded by two genes. Immunobiology 205:455
Stover, C. M., N. J. Lynch, M. R. Dahl, S. J. Hanson, M. Takahashi, M. Frankenberger, L. Ziegler-Heitbrock, I. Eperon, S. Thiel, W. J. Schwaeble. 2003. Murine serine proteases MASP-1 and MASP-3, components of the lectin pathway activation complex of complement, are encoded by a single structural gene. Genes Immun. 4:374
Endo, Y., M. Nonaka, H. Saiga, Y. Kakinuma, A. Matsushita, M. Takahashi, M. Matsushita, T. Fujita. 2003. Origin of mannose-binding lectin-associated serine protease (MASP)-1 and MASP-3 involved in the lectin complement pathway traced back to the invertebrate, amphioxus. J. Immunol. 170:4701.
Wagner, S., N. J. Lynch, W. Walter, W. J. Schwaeble, M. Loos. 2003. Differential expression of the murine mannose-binding lectins A and C in lymphoid and nonlymphoid organs and tissues. J. Immunol. 170:1462.
Burt, D. W., C. Bruley, I. C. Dunn, C. T. Jones, A. Ramage, A. S. Law, D. R. Morrice, I. R. Paton, J. Smith, D. Windsor, et al 1999. The dynamics of chromosome evolution in birds and mammals. Nature 402:411.
International Chicken Genome Sequencing Consortium. 2004. Sequence and comparative analysis of the chicken genome provide unique perspectives on vertebrate evolution. Nature 432:695.
Matsushita, M., A. Matsushita, Y. Endo, M. Nakata, N. Kojima, T. Mizuochi, T. Fujita. 2004. Origin of the classical complement pathway: lamprey orthologue of mammalian C1q acts as a lectin. Proc. Natl. Acad. Sci. USA 101:10127.
Endo, Y., M. Takahashi, M. Nakao, H. Saiga, H. Sekine, M. Matsushita, M. Nonaka, T Fujita. 1998. Two lineages of mannose-binding lectin-associated serine protease (MASP) in vertebrates. J. Immunol. 161:4924.(Nicholas J. Lynch, Saeed-)