| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
*
Department of Pathology and Laboratory Medicine, University of Pennsylvania, Philadelphia, PA 19104;
Department of Biology, University of Patras, Patras, Greece; and
Basel Institute for Immunology, Basel, Switzerland
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Complement activation through any of the three pathways results in the proteolytic cleavage of C3 to C3b and C3a, a reaction that is mediated by the C3 convertase. In the alternative pathway, factor B serves as the catalytic subunit of C3 convertase (8, 9, 10); in the classical pathway, this role is played by C2 (11, 12). In mammals, factor B and C2 share extensive amino acid homology; they have the same exon and intron organization and are located in tandem on the same chromosome (13, 14) within the mammalian MHC class III region. For these reasons, the two proteins are thought to have originated by gene duplication from an ancestral molecule (15). It is at present unclear in which animal phyla the duplication event took place.
The factor B protein has been purified only from mammals and birds (chicken). A partial amino acid sequence of chicken factor B was roughly equal in similarity to human and mouse factor B and C2 (16). In addition, the protein seemed to participate in both classical and alternative pathways of complement activation; nevertheless, the studies on the classical pathway in chicken were not clearly defined. It was suggested that the factor B protein in chicken is derived from a presumed common ancestral form of mammalian factor B and C2. In contrast to this situation, two genes encoding factor B-like molecules have been cloned from the amphibian (Xenopus). The two Xenopus molecules, designated Bf A (17) and Bf B (18), showed more sequence similarity to factor B than to C2 (40 and 30% identities to mouse factor B and C2, respectively) and consequently were considered to be factor B molecules. It was therefore hypothesized that the Bf/C2 gene duplication from a common ancestor occurred before the mammalian/amphibian divergence (17, 18). One gene encoding for a factor B-like molecule has recently been cloned from two teleost species, medaka fish (19) and zebrafish (20). Each of these molecules showed equal sequence similarity to factor B and C2 from mammals, and no additional factor B/C2-like molecules were found. In both cases, the factor B-like molecules were thought to function as both C2 and factor B; however, (as with Xenopus factor B), no functional studies of the proteins encoded by those genes were conducted to test this hypothesis. Functional studies have indicated the presence of both the classical and alternative pathways in teleost fish such as trout, a finding that suggests that these fish have both factor B and C2 molecules (21, 22). In this case, the Bf/C2 gene duplication would have predated the appearance of these fish.
Teleost fish have been shown to contain multiple forms of the third component of complement (C3)4 (23, 24, 25), which have been suggested to play an important role in generating immune diversity in fish (26). In the present study we demonstrate that trout, in addition to having multiple C3 genes, also possess two factor Bf/C2 genes. We have characterized their protein products and demonstrated that teleost fish contain a unique molecule that is involved in the activation of both alternative and classical pathways. Furthermore, we have succeeded in characterizing and reconstituting the trout alternative pathway with purified trout complement components (Bf, Df, and C3).
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Rainbow trout (Onchorhynchus mykiss) were obtained from Landenberg trout farm (Philadelphia, PA) and Limestone Trout Farm (Reading, PA). Blood was collected with a syringe from the caudal artery, and serum was obtained by incubating the blood at 4°C for 4 h. The serum was separated by centrifugation at 2000 x g for 10 min. Serum was stored at -70°C.
Determination of trout Bf-1 and Bf-2 cDNA sequences and phylogenetic analysis
Primers based on a partial trout cDNA sequence (clone RT-L72,
accession no. T23101, NID: g505992) with similarity to factor B and C2
from other species were designed, and a clone with an identical
sequence to the RT-L72 sequence was obtained by RT-PCR from trout liver
RNA. This PCR product was subcloned into the PCRTMII vector using the
TA cloning kit (Invitrogen, San Diego, CA) and was used to screen a
gt11 trout liver cDNA library. The full-length cDNA clone obtained
was designated Bf-1.
To obtain a full-length cDNA clone encoding Bf-2, we first produced a
partial cDNA clone by RT-PCR using degenerate primers based on the
internal protein sequence of the Bf-2 protein. We used primers P5F
(5'-(C,T)T(A,G,C,T)GA(C,T)AA(C,T)TT(C,T)AA(C,T)-3') and P3R
(5'-AT(A,T,C,G)CC(A,G)TA(A,G)TT(A,T,C,G)GG(A,G)TG-3') and the PCR
conditions described previously (23). The product obtained had the
expected molecular size of 0.6 kb, and its deduced amino acid sequence
showed high similarity to the corresponding region of trout Bf-1 and of
factor B and C2 molecules from other species (data not shown). The PCR
product was used to screen a gt11 trout liver cDNA library, and a
full cDNA clone encoding Bf-2 was obtained. Nucleotide sequences were
determined by the Sanger method (27). The deduced amino acid sequences
of Bf-1 and Bf-2 as well as all available factor B and C2 were aligned
using the Clustal W program (28), and the resulting alignments were
manually corrected. The obtained alignment (Fig. 1
) was used to calculate
Poisson-corrected distance matrexes to construct trees by the
neighbor-joining method (29).
|
A portion of 300 bp was amplified from both Bf-1 and Bf-2 cDNAs via PCR using Bf-1- and Bf-2-specific primers (Bf-1 sense primer, 5'-TAGCCCTGAAAACACAATGG-3' (nucleotides 1086–1106); Bf-1 antisense primer, 5'-TCCGTGCCATCCAGGGGTAT-3' (nucleotides 1377–1397); Bf-2 sense primer, 5'-CATCCCTTGCACCAAGGTAA-3' (nucleotides 1759–1779); Bf-2 antisense primer, 5'-TAGTAGGTTGACGACCACCCG-3' (nucleotides 2040–2051). Numbers in parentheses refer to trout Bf-1 and trout Bf-2 with GenBank accession numbers AF089861 and AF08960, respectively. The amplified products were purified using Qiaquick spin columns (Qiagen, Basel, Switzerland) and randomly primed (BRL, Gaithersburg, MD) with [32P]dCTP (Amersham, Arlington Heights, IL). Nonincorporated nucleotides were removed using G-50 spin columns (BMB, Rotkrevz, Switzerland), and the labeled fragments were then used as homologous probes for Northern and Southern blot analyses as described previously (30, 31).
Peptide synthesis
An 18-amino acid peptide (TBf1725–743) corresponding to the C-terminal portion (KYLGNDTEDYQPLEFLEN) of the deduced amino acid sequence of the RT-L72 clone that aligned with the C-terminal part of Bf and C2 molecules from other species was synthesized using an Applied Biosystems 430A peptide synthesizer (Foster City, CA) as described previously (32). A 14-amino acid peptide (TBf12–15) derived from the deduced amino acid sequence of Bf-1 corresponding to the N-terminal part of the molecule (RREWAWEGGSYTLT) was also synthesized.
Ab production
The synthesized peptides were coupled to keyhole limpet hemocyanin by the glutaraldehyde method (33) and used to raise Abs in rabbits. The Abs were purified by affinity chromatography, using synthetic peptides coupled to cyanogen bromide-activated Sepharose (Pharmacia, Piscataway, NJ). Polyclonal antisera against trout Bf-2 and against two trout protein contaminants from a post Mono-P Bf-2 preparation (65 and 23 kDa) were generated in rabbits by immunization with the SDS-PAGE-purified molecules (described below).
Purification of trout C3 isoforms, Bf-1, Bf-2, and Df
The trout C3 isoforms were purified as described previously (23). To purify trout Bf-2, trout serum (40 ml) was precipitated with 16% polyethylene glycol (PEG) at 4°C for 30 min in the presence of 20 mM EDTA, 10 mM benzamidine, and 1 mM PMSF and then centrifuged (15,000 x g, 20 min). The resulting supernatant was brought to 25% (NH4)2SO4, and the liquid phase containing PEG was removed. The remaining liquid phase containing the trout proteins was precipitated sequentially with 45, 60, and 75% (NH4)2SO4 at 4°C for 30 min; all incubations were followed by centrifugation at 15,000 x g for 20 min. The pellet from the final precipitation was resuspended in 10 mM sodium phosphate buffer, pH 7.5, loaded onto a DEAE 40 HR (6.5 x 5.0 cm) anion exchange column (Millipore, Bedford, MA) equilibrated in the same buffer, and eluted with a linear salt gradient (0–500 mM NaCl). Bf-2 was identified by immunoblotting with the immunoaffinity-purified anti-TBf1725–743 Ab and with the polyclonal anti-trout Bf-2 obtained after the Bf-2 purification. The immunoreactive fractions were pooled and concentrated with Amicon filters (30-kDa cut-off; Amicon, Beverley, MA), then exchanged into 25 mM imidazole buffer, pH 6.2, by passage over a PD10 column (Pharmacia). The sample was applied to a Mono P 10/10 isoelectric focusing column (Pharmacia) equilibrated in imidazole buffer, and eluted with a pH gradient (4.2–6.2) with polyampholites. The Bf-2 preparation contained some contaminants that were removed by preparing polyclonal Abs against two contaminating proteins (65 and 23 kDa). This reagent was used in affinity chromatography to obtain a homogeneous preparation of Bf-2.
Trout Bf-1 was partially purified from trout serum by a single precipitation with 45% PEG and anion (Mono Q HR 5/5) and cation (Mono S HR 5/5) chromatography. Bf-1 was identified with the anti-TBf12–15 Ab.
Trout Df was purified by gel filtration chromatography on a Superose 12 column. Fractions were concentrated 10-fold in Centricon filters (10-kDa cut-off) and tested for their ability to mediate the cleavage of purified trout Bf-2 into Bb and Ba fragments in the presence of purified trout C3-1 and a buffer containing Mg2+EGTA. The fractions mediating cleavage were reapplied to the Superose 12 column. Purification of each of these trout proteins was monitored by SDS-PAGE and by immunoblotting using Abs specifically recognizing the individual proteins. The concentrations of trout Bf-1, Bf-2, and Df in serum were determined as described previously (23).
Protein sequencing
N-terminal sequences were obtained by subjecting the purified molecules to electrophoresis, followed by electroblotting onto ProBlott membranes (Applied Biosystems). A modified version of the method of Matsudaira (34) was used for sequencing, as described previously (35). The electroblotted proteins were subjected to Edman degradation, using an Applied Biosystem 473A Protein Sequencer. The internal protein sequence of trout Bf was obtained by digesting the protein with the endoproteinase Lys-C from Lysobacter enzymogenes (Boehringer Mannheim, Indianapolis, IN) (35).
Trout antiserum against sheep E
Abs against sheep E were generated by immunizing rainbow trout (200–300 g) i.p. with a suspension (0.5 ml) of washed sheep E (5 x 108) mixed (1/1) in CFA. Thereafter, trout were injected weekly for 4 wk with a suspension (0.5 ml) of washed sheep E (5 x 108) mixed (1:1) in IFA. Fish were bled 1 wk after the last injection.
Complement assays
Factor Bf-1/Bf-2 activities were measured by two different assays: 1) formation of fluid phase C3bBb convertase and 2) reconstitution of hemolytic activity of Bf-1/Bf-2-depleted trout serum with purified Bf-2.
To determine whether the Bf-1/Bf-2 and Df molecules mediate cleavages homologous to those mediated by their mammalian counterparts, we reconstituted the trout alternative pathway of complement activation with purified trout C3, Bf-2 (or partially purified Bf-1) and Df. For each experiment 2 µg of C3, 1 µg of Bf-2, and 0.02 µg of Df were used. The three proteins or reaction mixtures containing only two of the proteins were incubated at 20°C for 40 min in the presence of Mg2+EGTA or EDTA. Cleavage of the C3 and Bf-2 molecules was studied by resolving the reaction mixtures on 9% SDS-PAGE under nonreducing and reducing conditions and staining with Coomassie blue. In the case of Bf-1, its cleavage to Bb and Ba was detected by Western blotting using the anti-Bf-1 Abs.
The hemolytic activity of trout serum or Bf-depleted trout was measured using rabbit E (RaRBC) and sensitized SRBC; SRBC were sensitized with trout anti-SRBC as described previously (21). Bf-1- and Bf-2-depleted trout serum was generated using a polyclonal anti-Bf-2 Ab; this Ab cross-reacts with both Bf-1 and Bf-2. The Ig fraction of this antiserum was purified and covalently coupled to cyanogen bromide-activated Sepharose. Trout serum (0.2 ml) was made 10 mM in EDTA and passed three times over the anti-Bf column. Bound material was eluted with 0.2 M glycine buffer, pH 2.6; the column was then re-equilibrated in PBS-EDTA, and the partially depleted Bf trout serum was again passed over the column. As a control, a second sample of trout serum was fractionated on a similar Sepharose column coupled with an unrelated Ab. The eluted samples were concentrated to one-half of their original volume, reconstituted with either 20 mM Ca2+ and Mg2+ or 20 mM Mg2+EGTA, and tested for hemolytic activity mediated through the classical and alternative pathways, respectively. RaRBC and SRBC were washed four times in veronal-buffered saline (VBS) and resuspended in VBS containing 10 mM Mg2+EGTA VBS (Mg2+EGTA-VBS) or 5 mM Ca2+Mg2+ (Ca2+Mg2+-VBS) to give a concentration of 2.5 x 108 cells/ml. For the classical pathway assays, ShRBC were sensitized with trout anti-ShRBC as described previously (21). Sensitized ShRBC (7 µl) were then added to 25 µl of serially diluted trout serum in Ca2+Mg2+-VBS. For alternative pathway assays, 7 µl of RaRBC were added to 25 µl of serially diluted trout serum in Mg2+EGTA-VBS. The reaction mixtures were incubated at room temperature for 40 min with shaking, and the reaction was stopped by adding 50 µl of VBS containing 40 mM EDTA. The extent of hemolysis was estimated by measuring the OD of the supernatant at 414 nm. The reciprocal of the serum dilution causing 50% lysis of RBC was designated the ACH50 or CH50 titer; results were presented as ACH50 and CH50 units per ml for the alternative and classical pathways, respectively, and were calculated as described previously (21, 36). To determine whether trout Bf-2 could restore the hemolytic activity of the Bf-depleted serum, various concentrations of purified Bf-2 were mixed with the depleted serum and serially diluted; hemolytic activity was assessed as described previously.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The full-length Bf-1 and Bf-2 cDNAs that we isolated were 2509 and
2560 bp in length, respectively, and encoded proteins of 743 and 749
amino acids. Both molecules had a domain structure similar to those of
factor B and C2 molecules from other species. The Bf-1 and Bf-2
molecules consisted of three short consensus repeats at the N-terminus,
a von Willebrand domain, and a serine protease domain at the C-terminus
(Fig. 1
). The distribution of the cysteine residues was highly
conserved, except for one cysteine in the von Willebrand domain that is
present in human and mouse factor B, but is not present in any other
known factor B/C2 molecules. Residues His, Asp, and Ser, which are
located at the active center of the serine protease domain, were also
conserved (Fig. 1).
The deduced amino acid sequences of Bf-1 and Bf-2 showed more similarity to human and mouse factor B than to C2 molecules. Trout Bf-1/Bf-2 showed amino acid identities of 38/39% to human factor B, 30/31% to human C2, 37/38% to mouse factor B, 28/31% to mouse C2, 34/33% to Xenopus factor B-A, 33/32% to Xenopus factor B-B, 50/51% to medaka fish factor Bf/C2, 42/43% to zebrafish factor B, and 16/22% to lamprey factor B. Trout Bf-1 was 75% identical with Bf-2. Thus, the trout Bf sequences showed about 9% more similarity to mammalian factor B than to C2. In contrast, the factor B/C2 sequences from the medaka fish and zebrafish were equally similar to mammalian factor B and mammalian C2 (19, 20). The situation in the case of the trout factor B molecules was very similar to that in Xenopus, which also contains two factor B molecules that show about 10% more similarity to mammalian factor B than to C2 (17, 18). The amino acid differences between trout Bf-1 and Bf-2 were scattered throughout the entire sequence, indicating that the two molecules are not generated by differential processing of transcripts from a single gene, but are the products of two distinct genes.
A phylogenetic tree generated from all available factor B and C2
sequences showed that despite the higher sequence similarity of trout
Bf-1 and Bf-2 to mammalian factor B than to C2, both trout molecules
clustered with the mammalian C2 sequences, as was the case also for the
medaka fish and zebrafish factor B molecules (Fig. 2). The Xenopus factor B
sequences clustered instead with those of mammalian factor B. Lamprey
factor B appeared as an outgroup.
|
The tissue-specific mRNA expression of Bf-1 and Bf-2 in trout was
investigated by Northern blot hybridization. Expression of trout Bf-1
(3 kb) and Bf-2 (2.7 kb) is limited to the liver (Fig. 3, A and B),
although slight expression of Bf-1, but not Bf-2, was detected in the
intestine with prolonged exposure (data not shown). Interestingly, both
Bf-1 and Bf-2 express high levels of message, in contrast to the low
levels of Bf-1 protein expression (2–4 µg/ml) compared with that for
Bf-2 (300–400 µg/ml). These results suggest a possible
post-transcriptional or translational mechanism for the differential
protein expression observed in the serum. Relative equivalency of
loading was verified by reprobing the Northern blot with a trout
housekeeping gene (EfTu-1; Fig. 3C).
|
A) and Bf-2 (Fig. 4B) using three different restriction enzymes
(HindIII, EcoRI, and EcoRV),
suggesting that both are single-copy genes in trout. However, two of
the four siblings for the Bf-2 analysis showed slight polymorphism for
HindIII (Fig. 4B), which is most likely due to
allelic variants of this gene. In contrast, Bf-1 displayed no
polymorphism for the enzymes or siblings used in this analysis.
|
An immunoaffinity-purified Ab recognizing residues 725 to 743 of
Bf-1 (corresponding to the C-terminal part of the deduced amino acid
sequence of the RT-L72 clone) was used to identify Bf-2 in trout serum.
This Ab cross-reacted to a low degree with an 81-kDa protein in trout
serum (data not shown) that we designated trout Bf-2 because its
N-terminal amino acid sequence differed from the deduced N-terminal
sequence of trout Bf-1. The reactivity of this anti-peptide Ab with
Bf-2 reflected the high sequence similarity of Bf-1 and Bf-2 in the
region spanned by the peptide (Fig. 1).
The resulting preparation obtained from the precipitation of trout
serum with PEG and ammonium sulfate contained mainly fish albumin and
hemoglobin along with the trout Bf-2 protein. Half of the fish albumin
and hemoglobin could be separated from Bf-2 by anion exchange
chromatography. Thereafter, isoelectric focussing was very effective in
removing most of the albumin and hemoglobin from the Bf-2-containing
fractions (Fig. 5). The remaining
contaminants were removed by affinity chromatography, and the Bf-2
protein was purified to homogeneity (>95% pure as judged by SDS-PAGE
and Coomassie blue staining; Fig. 6,
lane 3). In addition, N-terminal sequencing of the purified
Bf-2 molecule gave a single sequence, suggesting that the Bf-2
preparation was homogeneous. Moreover, the
anti-TBf12–15 peptide Ab, which
specifically recognized Bf-1 (see below), did not react to any extent
with the purified Bf-2 (data not shown). Bf-2 was present in trout
serum at 300 to 400 µg/ml.
|
|
).
The anti-TBf12–15 peptide Ab was used to
isolate Bf-1 from trout serum. The sequence spanned by the peptide was
very different from the corresponding Bf-2 sequence (Fig. 1).
Consequently, the Ab we raised against
TBf12–15 was unable to recognize Bf-2, but did
recognize a molecule of a similar size (data not shown). This molecule
was partially purified and was shown to react much more strongly with
the anti-TBf1725–743 Ab than did the Bf-2
protein, suggesting that the partially purified molecule was indeed
Bf-1. A polyclonal Ab raised against trout Bf-2 also strongly reacted
with Bf-1, probably because of the high sequence similarity between
Bf-1 and Bf-2 (data not shown). All of the individual fish analyzed for
the presence of Bf-1 using the
anti-TBf12–15 and the
anti-TBf1725–743 Abs were positive,
suggesting that Bf-1 is present in all fish and that the putative Bf-1
that we have identified is not a polymorphic form of Bf-2. A rough
calculation indicated that the serum concentration of Bf-1 was very low
(
2–4 µg/ml).
Trout factor D (Df) was purified to >90% homogeneity as judged by
SDS-PAGE and Coomassie blue staining (Fig. 6, lane 4). The
molecular size of trout Df was 24 kDa, consistent with the size of
factor D molecules from all other species analyzed to date. Its
concentration in serum was 25 to 50 µg/ml; this value was
significantly higher than that in the teleost fish Cyprinus
carpio (common carp; 6 µg/ml) (37) or in humans (1 µg/ml)
(38), although it was roughly similar to the values in mice (50–100
µg/ml) (39). The N-terminal amino acid sequence of trout factor D was
highly similar to that of factor D from other species (15/3 and 13/5,
identical/different residues to the corresponding sequenced residues in
carp and mouse, respectively; Fig. 7).
The residues that are conserved in these three species were also
conserved in trout Df, indicating that the purified protein was indeed
factor D.
|
Reconstitution of the trout alternative pathway. The ability of
Bf-1 and Bf-2 to participate in the formation of fluid phase C3bBb
convertase was assessed using purified trout C3 and Df. To date,
reconstitution of the alternative pathway with purified components has
only been achieved in mammalian species. We were able to reconstitute
the trout alternative pathway with purified trout complement components
(C3-1, Bf-2, and Df; Fig. 6). As previously observed in mammals, trout
C3-1 and trout Bf-2 in the presence of trout Df and EDTA remained
uncleaved (Fig. 6, lane 5); however, in the presence of
Mg2+EGTA, C3-1 was cleaved to C3b, and Bf-2 was cleaved to
yield fragments homologous to mammalian Bb and Ba (Fig. 6, lane
6) (40). These data are consistent with the requirement for
Mg2+ that is seen when trout serum is used to lyse rabbit
RBC through the alternative pathway (21, 41). Furthermore, combinations
of C3-1 and Df (Fig. 6, lane 7), C3-1 and Bf-2 (Fig. 6, lane 8), and Bf-2 and Df (Fig. 6, lane 9) did not
lead to cleavage of Bf-2 or C3-1 in the presence of
Mg2+EGTA, indicating that Df, Bf-2, and C3-1 preparations
were not cross-contaminated. Trout C3-3 and C3-4 also formed the
alternative pathway C3 convertase with trout Bf-2 in the presence of Df
and Mg2+EGTA (data not shown), indicating that trout Bf-2
was capable of interacting with all trout C3s. In contrast to the
situation in humans (42), trout Df was not the limiting factor in the
fish system, since as little as the equivalent of 0.2 µg/ml of Df was
capable of cleaving the same amount of Bf-2 as a physiologic
concentration of Df (25–50 µg/ml) in trout serum (data not shown).
In similar experiments a partially purified preparation of Bf-1 was
cleaved in the presence of trout C3, factor D, and
Mg2+EGTA, implying that Bf-1 can also act as a factor B
molecule (data not shown).
Role of trout Bf-2 in complement activation
The involvement of the Bf-2 protein in the hemolytic activity of
trout serum via both classical and alternative pathways was assessed
using Bf-depleted serum. The polyclonal anti-Bf-2 Ab was coupled to
Sepharose and used to deplete Bf-2 as well as Bf-1 from trout serum. To
our surprise this immunodepletion abolished the hemolytic activity of
the serum through both alternative and classical pathways, suggesting
that trout Bf-1 and Bf-2 were involved in either the alternative or the
classical pathway or both. Addition of purified Bf-2 to the depleted
serum restored both classical and alternative pathway activities (Fig. 8), suggesting that trout Bf-2 may
represent an ancestral molecule that has both Bf and C2 functions.
Trout-mediated lysis of SRBC through the alternative pathway is
negligible, and sensitization of SRBC with trout Abs did not contribute
to any significant lysis through the alternative pathway (data not
shown), indicating that Abs do not play a role in the activation of the
alternative pathway. Reconstitution of the hemolytic activity was dose
dependent (Fig. 8), and the concentration of Bf-2 needed to fully
restore the hemolytic activity (40 and 160 µg/ml for the alternative
and classical pathways, respectively) was less than that of Bf-2 in
serum (300–400 µg/ml), indicating that Bf-2 is in excess.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, A and B). The two molecules are expressed mainly
in liver, as is the case in higher animal species, although Bf-1
appears to be expressed in intestine also (Fig. 3); the functional
significance of expressing Bf-1 in the intestine is unknown at the
moment. Interestingly, both Bf-1 and Bf-2 show high levels of mRNA
expression; this contrasts with the very low expression of Bf-1 (2–4
µg/ml) compared with Bf-2 (300–400 µg/ml) at the protein level.
The mRNA of both Bf-1 and Bf-2 are different in size and are within the
size range found for factor B of other animal species. The higher
similarity of trout Bf-2 to mammalian factor B molecules than to C2
might suggest that trout Bf-2 would function only as a factor B
molecule, in contrast to the dual role that it appears to play. This
higher similarity may reflect the faster rate of evolution of C2
compared with that of factor B (20). However, this situation differs
from that in the medaka fish and zebrafish, whose factor B/C2 molecules
show equal similarity to mammalian factor B and C2 molecules. Seeger et
al. have postulated that this equal similarity in medaka fish reflects
the fact that the substitutions in these molecules have reached a
saturation point (20). If this were the case, however, the trout Bf
molecules would also be expected to show equal similarity to mammalian
factor B and C2. As Seeger and co-workers have observed, these
considerations make it difficult to draw any significant conclusions
from the phylogenetic tree (20). For example, it was unexpected to find
that trout Bf molecules clustered with mammalian C2 molecules (Fig. 2)
when they show about 9% more similarity to mammalian factor B than C2,
whereas the Xenopus factor B molecules (which are 10%
more similar to mammalian factor B than C2) do cluster, as expected,
with mammalian factor B. Furthermore, trout Bf-1 and Bf-2 appear to be
more similar to mammalian factor B than C2, since they both have 11
charged residues in the area that aligns with the exon 15-encoded
region from medaka fish. The number of charged residues differs greatly
between mammalian factor B and C2 in the exon 15-encoded region (human
and mouse Bf contain 15 and 12, whereas C2 contains only 6); this
difference may be related to the functional differences between Bf and
C2 (43). Consequently, from the phylogenetic analysis of the primary
sequences of trout Bf-1 and Bf-2, it was not possible to deduce whether
these molecules were factor B or C2. Therefore, the only way to
determine whether these molecules represented Bf or C2 molecules was by
analyzing their functions.
Our results suggest that the purified trout Bf-2 molecule can function
in both alternative and classical pathways of complement activation.
Bf-2 was able to reconstitute the alternative pathway in the presence
of purified trout C3-1 and factor D (Fig. 6). In this experiment, trout
Bf-2 behaved like a mammalian factor B, in that it was able to form the
alternative pathway C3 convertase and be cleaved to Bb and Ba fragments
in the presence of a buffer containing Mg2+EGTA.
Furthermore, Bf-2 fully reconstituted the hemolytic activity of the
Bf-depleted trout serum through the alternative and classical pathways
(Fig. 8). It is interesting that the amount of Bf-2 required to restore
the classical pathway was about fourfold higher than that needed to
restore the alternative pathway; this difference might reflect the
higher titers of the trout classical pathways (three- to fourfold
higher) than those of the alternative pathway. The fact that both trout
Bf-2 and Df are present in significantly higher serum concentrations
than are human Bf and Df and that both trout proteins appear to be
functionally in excess could explain why the hemolytic titers of the
fish alternative pathway are 5 to 10 times higher than those in humans.
Whether Bf-1 also works through the classical pathway is unknown at
present, because the very low concentration of Bf-1 in serum did not
allow us to obtain a pure Bf-1 preparation. To confirm whether Bf-2
functions as both Bf and C2, trout C4 and C1 will have to be purified
to analyze the requirement of Bf-2 in the formation of the classical
pathway convertase. This work is currently in progress in our
laboratory.
Our data are in agreement with predictions of Kuroda et al. (19) and
Seeger et al. (20), who have suggested that teleost fish might contain
a molecule functioning as both factor B and C2, since they were unable
to assign their fish (medaka fish and zebrafish) sequences to either
factor B or C2. Nevertheless, Kuroda et al. (20) suggested that if
medaka fish had a molecule that played a dual role, then the split
between Bf and C2 would have had to happen after the divergence of
teleosts but before the divergence of amphibians from a common
vertebrate ancestor, since they had previously found that
Xenopus contains two factor B molecules that are identified
as Bf on the basis of their higher similarity (10% more similar) to
mammalian factor B than to C2 molecules (17, 18). In contrast, our
results closely resembled those obtained in Xenopus. As in
the case of Xenopus, trout appear to have two molecules that
show about 9% more similarity to mammalian factor B than to C2, yet
one of the two molecules (Bf-2) appears to function in both classical
and alternative pathways of complement activation. Therefore, it is
possible that the Xenopus Bfs could also assume the roles of
both factor B and C2. This situation would imply that the split between
factor B and C2 happened after the divergence of the amphibians from a
common vertebrate ancestor.
It is interesting that in addition to having multiple forms of C3 (23, 44), trout also contain (in contrast to medaka fish and zebrafish) at least two factor B molecules. Our results cannot exclude the possibility that additional Bf isoforms are present or that a C2-like molecule exists in trout serum. However, trout Bf-2 alone (the most abundant trout Bf) was sufficient to completely reconstitute the hemolytic activity of trout serum through the alternative or classical pathway, even at lower concentrations than those present in serum. This finding suggests that a C2-like molecule may not be required, and therefore it is probably not present in the trout. The significance of trout Bf-1 remains unknown, although we have shown that it is present in serum at very low concentrations, and it can be cleaved at least through the alternative pathway; its role in the classical pathway is currently under investigation.
Our findings suggest that before the divergence of C2 and factor B from a common ancestor, a molecule existed that was able to function in both alternative and classical pathways. The need for higher evolved animals to have two separate molecules is uncertain; however, one could speculate that the system could be better regulated if each pathway was dependent on a distinct molecule, instead of both pathways relying upon a common Bf/C2 molecule.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 The sequences described in this paper have been deposited in the GenBank database under accession numbers AF089860 to AF089861.
3 Address correspondence and reprint requests to Dr. John D. Lambris, Protein Chemistry Laboratory, Department of Pathology and Laboratory Medicine, University of Pennsylvania, Philadelphia, PA 19104-6079. E-mail address:
4 Abbreviations used in this paper: C3, third component of complement; PEG, polyethylene glycol; RaRBC, rabbit E; VBS, veronal-buffered saline.
Received for publication March 13, 1998. Accepted for publication June 22, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
2-macroglobulin family. J. Immunol. 154:2164.[Abstract]
This article has been cited by other articles:
|
|
 |
|
  Y. Ohta, W. Goetz, M. Z. Hossain, M. Nonaka, and M. F. Flajnik Ancestral Organization of the MHC Revealed in the Amphibian Xenopus J. Immunol., March 15, 2006; 176(6): 3674 - 3685. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Mutsuro, N. Tanaka, Y. Kato, A. W. Dodds, T. Yano, and M. Nakao Two Divergent Isotypes of the Fourth Complement Component from a Bony Fish, the Common Carp (Cyprinus carpio) J. Immunol., October 1, 2005; 175(7): 4508 - 4517. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Boshra, T. Wang, L. Hove-Madsen, J. Hansen, J. Li, A. Matlapudi, C. J. Secombes, L. Tort, and J. O. Sunyer Characterization of a C3a Receptor in Rainbow Trout and Xenopus: The First Identification of C3a Receptors in Nonmammalian Species J. Immunol., August 15, 2005; 175(4): 2427 - 2437. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Boshra, A. E. Gelman, and J. O. Sunyer Structural and Functional Characterization of Complement C4 and C1s-Like Molecules in Teleost Fish: Insights into the Evolution of Classical and Alternative Pathways J. Immunol., July 1, 2004; 173(1): 349 - 359. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Terado, K. Okamura, Y. Ohta, D.-H. Shin, S. L. Smith, K. Hashimoto, T. Takemoto, M. I. Nonaka, H. Kimura, M. F. Flajnik, et al. Molecular Cloning of C4 Gene and Identification of the Class III Complement Region in the Shark MHC J. Immunol., September 1, 2003; 171(5): 2461 - 2466. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Q. Pan, R. O. Ebanks, and D. E. Isenman Two Clusters of Acidic Amino Acids Near the NH2 Terminus of Complement Component C4 {alpha}'-Chain Are Important for C2 Binding J. Immunol., September 1, 2000; 165(5): 2518 - 2527. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. D. Hansen, P. Strassburger, G. H. Thorgaard, W. P. Young, and L. Du Pasquier Expression, Linkage, and Polymorphism of MHC-Related Genes in Rainbow Trout, Oncorhynchus mykiss J. Immunol., July 15, 1999; 163(2): 774 - 786. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
