1. Introduction
Inflammation is one of the first lines of defense in the innate immune system of metazoans. It involves the recruitment of immunocytes to the infection sites, their extravasation, and consequent degranulation with the induction of cytotoxicity [
1]. The inflammatory events operate in tight association with the complement system (CS), one of the most ancient immune modulators of metazoans [
2].
Once activated, the CS leads to the cleavage of the core protein C3 into component proteins C3a and C3b [
2,
3]. The anaphylatoxin C3a, which shares many functional behaviors with the closely related molecules C4a and C5a, plays a central role in inflammatory response. C3a recruits and activates immunocytes, triggers degranulation, induces phagocytosis, and associated respiratory burst and reactive oxygen species (ROS) production. C3a works primarily by interacting with its receptor, known as C3a receptor (C3aR) [
4,
5,
6].
C3aR is a member of the G protein-coupled receptor family; all members of this family have seven transmembrane (7TM) alpha helixes [
7]. This receptor has been widely described and characterized in mammals [
8], where it is expressed in various immunocytes (e.g., monocytes, dendritic cells, neutrophils, basophils) [
9]. Mammalian C3aRs feature a large extracellular loop (ECL2) between the fourth and fifth transmembrane regions (TM4 and TM5) [
7] that is required for the high-affinity binding of C3a [
10]. As for non-mammalian vertebrates, up to now, C3aR has been described in amphibians and teleost fish [
11], although avian and reptilian C3aR gene sequences are present in databases.
Molecular data indicate that vertebrate C3aRs and C5aRs (the receptors for C5a) derive from a duplication event from a common ancestor. Since teleost fish possess both C3aR and C5aR, this suggests that the duplication occurred before the emergence of bony fish [
8].
In invertebrate chordates, functional C3a anaphylatoxin with chemotactic properties has been described in the cephalochordate
Branchiostoma japonicum [
12] and the tunicate
Ciona intestinalis [
13]. In
Ciona, a chemotactic receptor was also identified and characterized [
14].
Botryllus schlosseri is a cosmopolitan compound ascidian, and a member of the tunicates, a group of animals representing the closest relatives to vertebrates. It is one of the most used organisms for the study of tunicate immunobiology and immune system evolution [
15]. A colony includes three blastogenetic generations: actively filter-feeding zooids, primary buds emerging from the zooid body wall, and secondary buds on primary ones. Through cyclical generation changes or take-overs (TOs), the colony periodically rejuvenates: during this period (lasting 24–36 h), old zooids lose their functionality, close their siphons, and are progressively resorbed. In the meantime, buds grow to adult size and replace old zooids whereas budlets become buds able to give rise to a new budlet generation. The period of time between two successive TOs is defined as a colony blastogenetic cycle [
16].
In this species, we already identified some complement components such as C3 (BsC3), FactorB (BsfB), ficolin (Bsficolin), MASP (BsMASP) [
17]. In particular, BsC3 shares with other metazoan C3s the presence of a conserved cleavage site that generates C3a and C3b [
17]. We already studied some of the interactions between the complement system and immunocytes, demonstrating the involvement of complement C3, likely through C3b, in phagocytosis, as suggested by its reduction when C3 activation is blocked [
17,
18]. However, no data on the possible involvement of C3a in
Botryllus immunity has been reported up to now. In order to fill this gap and acquire new information on ascidian complement, in the present work, we mined our
Botryllus transcriptome [
18] looking for sequences with similarity to human C3aR.
2. Material and Methods
2.1. Animals
Colonies of
B. schlosseri were collected in the Southern part of the Lagoon of Venice and left to adhere to glass slides in aquaria at 18 °C for 5 days. Animals were fed daily with Phyto Marine (Oceanlife, Bologna, Italy) and living
Tetraselmis chuii cells. Each
B. schlosseri colony derives from a single tadpole-like larva that metamorphoses into an oozoid, the colony founder. The colony increases its size through several cycles of asexual reproduction (blastogenetic cycles). The filtering adults are organized in star-shaped systems of 10–15 zooids, each with its own oral siphon; in the center of each system, a common cloacal siphon opens to the exterior. Each blastogenetic cycle lasts about one week at 20 °C and begins with the appearance of thickenings of the zooid body wall (bud primordia): we called this phase, lasting one day, early cycle (EC). The blastogenetic cycles end with the TO, lasting 24–36 h [
16]. The colonial developmental phases lying more than one day from the TO [
16,
19] are collectively called mid-cycle (MC). A common circulatory system, in the form of a vessel network within the common tunic, connects zooids, buds, and budlets [
20,
21]. Among the circulating hemocytes, immunocytes, i.e., the effectors of immune reactions, are represented by phagocytes and cytotoxic granular (morula) cells [
22] and constitute the majority of cells [
23]. Phagocytes collectively account for 20–40% of the total circulating cells and include spreading and round phagocytes, the former being actively moving cells that, upon the ingestion of foreign particles or cells, acquire the round morphology. Morula cells are the most abundant hemocyte type (40–60% of total circulating hemocytes [
22]) and are the first cells perceiving the presence of nonself [
23,
24].
2.2. Hemocyte Collection
To collect hemocytes, the marginal vessel of colonies (previously immersed in 0.38% Na-citrate in filtered seawater (FSW) to prevent cell clumping) was punctured with a fine tungsten needle, and cells were collected with a glass micropipette. They were then pelleted at 800× g for 10 min and re-suspended in FSW to a final concentration of 5 × 105 cells/mL.
2.3. BsC3a/C5aR Sequence Characterization and Phylogenetic Analysis
Mining our
B. schlosseri EST collection (
http://botryllus.cribi.unipd.it; last access: 12 March 2020) [
18] and the database of the
B. schlosseri genome (
http://botryllus.stanford.edu/botryllusgenome/last access: 23 July 2020), we identified through BLASTn analysis using the predicted
C. intestinalis C3aR sequence (Genbank CAI84650.1) the sequence of a transcript with similarity to the vertebrate transcripts for C3aR, referred to as BsC3a/C5aR. The 3’ rapid amplification of the cDNA ends (RACE) was performed using the 2nd Generation of the 5’/3’ RACE Kit (Roche Diagnostics, Basel, Switzerland). In order to obtain the 3’ sequences of BsC3a/C5aR cDNA, specific primers, reported in
Table S1, were designed for nested PCR with anchor reverse primer according to the manufacturer’s instruction.
For the phylogenetic analysis, alignments were performed with MUSCLE software [
28] and assessed using molecular evolutionary genetics analyses (MEGA) version 7 program [
29]. We evaluated different amino acid substitution models using MEGA7 and we found that the JTT+G+F was the best fit for our dataset with the lowest Akaike information criterion (corrected AIC scores) = 42,837.9963 and maximum likelihood value (lnL) = −21274.8895. The maximum likelihood (ML) [
30] method was used to build phylogenetic trees with the MEGA 7 software [
29]. The non-distance-based phylogeny reconstruction neighbor-joining (NJ) [
31] and the maximum parsimony (MP) methods were also used to build phylogenetic trees with the MEGA 7 software [
29]. The nonparametric bootstrap test [
32], with 10,000 replicates, was used to assess the robustness of the tree topologies. Sequences used for phylogenetic analysis included C3aRs and C5aRs from vertebrates and sequences ascribed to C3aR from invertebrate chordates, found in Genbank and Aniseed. Outgroup sequences were obtained from Boshra et al. [
11]. They are all reported in
Table S2.
2.4. Primer Design, RNA Extraction, cDNA Synthesis, Cloning and Sequencing
The RNA NucleoSpin RNA XS (Macherey–Nagel, Düren, Germany) kit was used to isolate total RNA from colonies of B. schlosseri and its quality was determined by the A260/280 ratio and the quality of RNA was determined by the visualization of rRNAs in Midori green (Nippon genetics)-stained 1.5% agarose gels. The first strand of cDNA was reverse transcribed from 1 µg of total RNA at 42 °C for 1 h in a 20 µL reaction mixture containing 1 µL of ImPromII Reverse Transcriptase (Promega, Madison, WI, USA) and 0.5 µg oligo(dT)-Anchor primer or random primers (Promega, Madison, WI, USA).
The primers reported in
Table S1 were used for PCR reactions in a 25-µL reaction volume containing 100 ng of cDNA from
B. schlosseri colonies, 2.5 µL of 10× incubation buffer (PCRBIO Classic Taq, PCR BIOSYSTEMS, London, UK) with 15 mM MgCl
2, 0.25 µM of each primer, 10 mM of each of the deoxynucleotide triphosphates, and 2 units of Taq polymerase. PCR was performed on a MyCycler (BioRad; Hercules, CA, USA) thermocycler with the following conditions: 94 °C for 2 min, then 40 cycles of 94 °C for 30 s, 55–60 °C for 30 s, 72 °C for 40 s, and 72 °C for 10 min. Amplicons were separated by electrophoresis on 1.5% agarose gel and the corresponding bands were purified with ULTRAPrep Agarose Gel Extraction Mini Prep kit (AHN Biotechnologie, Nordhausen, Germany), ligated in pGEM-T Easy Vector (Promega, Madison, WI, USA), and cloned in DH-5α
Escherichia coli cells. Positively screened clones were Sanger sequenced at Eurofins Genomics (Ebersberg, Germany) on an ABI 3730XL Applied Biosystems apparatus (Life Technologies Europe BV, Monza, Italy).
2.5. Quantitative Real-Time PCT (qRTPCR)
Three laboratory colonies were split into three subclones, 2–3 systems each, and, after 5 days of acclimation, their blastogenetic cycle was followed under the dissection microscope. When at EC, MC, and TO, colonies were collected and their mRNA was extracted as already reported. qRT-PCR was carried out according to the method reported in Franchi and Ballarin [
17] to estimate the total amount of mRNA for BsC3a/C5aR. In this case, EC was considered as a reference control. Forward and reverse specific primers for the above-reported transcripts and for the elongation factor 1α (EF1α) were designed and reported in
Table S1. All the designed primers contained parts of contiguous exons to exclude contamination by genomic DNA; a qualitative PCR was also carried out before qRT-PCR. In addition, the analysis of the qRT-PCR dissociation curve gave no indications of the presence of contaminating DNA.
The following cycling parameters were used: 3 min at 95 °C (denaturation), 20 s at 95 °C plus 1 min at 60 °C, 45 times (annealing), 15 s at 95 °C, 1 min at 60 °C, 15 s at 95 °C, 15 s at 60 °C (extension). Each set of samples was run three times on an Applied Biosystem 7900 HT Fast Real-Time PCR System (Life Technologies Europe BV, Monza, Italy) and each plate contained cDNA from three different biological samples (
n = 3) and negative controls. The 2
−ΔΔCT method [
33] was used to estimate the total amount of mRNA. The amounts of transcripts in different conditions were normalized to EF1α to compensate for variations in the amounts of cDNA.
2.6. In Situ Hybridization (ISH)
Using the primers reported in
Table S1, as previously described [
17], we produced the biotin-labeled antisense riboprobes for BsC3a/C5aR. Hemocytes were collected as previously described and left to adhere for 30 min on SuperFrost Plus (Menzel–Glaser, Braunschweig, Germany) glass slides. Cells were then incubated for 1 h in FSW in the presence or in the absence (control) of either zymosan (1 mg/mL) or
Bacillus clausii (4 × 10
5 cells/mL). They were then washed in FSW and fixed for 30 min in a solution of 4% paraformaldehyde plus 0.1% glutaraldehyde in 0.4 M cacodylate buffer, containing 1.7% NaCl and 1% sucrose, at 4 °C. After their permeabilization in a solution of 0.1% Triton X in phosphate-buffered saline (PBS: 1.37 M NaCl, 0.03 M KCl, 0.015 M KH
2PO
4, 0.065 M Na
2HPO
4, pH 7.2) for 5 min, hemocytes were washed in PBS and preincubated in Hybridisation Cocktail 50% formamide (Amresco, Solon, OH, USA) for 1 h at 55 °C, and hybridized in the same solution containing 1 µg/mL riboprobe, overnight, at the same temperature reported above. They were then washed in SSC (0.3 M NaCl, 40 mM sodium citrate, pH 4.5) for 5 min, and in a solution of 50% formamide in SSC at 55 °C for 30 min followed by an additional washing in PBS containing 0.1% Tween 20 (PBST) at room temperature, for 5 min. Hemocyte monolayers were then incubated in 1% powdered milk in PBST for 1 h (to reduce aspecific staining), in 5% methanol for 30 min (to block endogenous peroxidases), and in Vectastain ABC (Vector) in PBS for 30 min. Finally, cells were incubated in 0.025% 3,3′-diaminobenzidine and 0.004% H
2O
2 in PBS for 15 min. Slides were washed in distilled water and mounted in Eukitt (Electron Microscopy Sciences, Hatfield, PA, USA) before cell observation under the light microscope (LM).
2.7. Effects of C3aR Agonist
In another experimental series, to investigate the relationship between BsC3a/C5aR activation and BsC3 transcription, three colonies were divided into two subclones and, after an acclimation period of 5 days, the marginal vessels of the subclones, when at MC, were injected with 5 µL of C3aR agonist (Santa Cruz Biotechnology, Dallas, TX, USA) at the concentration of 0.3 µM in DMSO or with the same amount of DMSO in controls. After 24 h, mRNA was extracted from treated and control colonies according to the protocol described above, reverse transcribed to cDNA, and the expression of bsc3 and bsc3a/c5ar was followed by qRT PCR. EF1α was used as a housekeeping gene due to its stable expression.
2.8. Statistical Analysis
Data are expressed as mean ± SD. qPCR experiments were replicated three times (n = 3) with three independent samples. Statistical analyses were performed with the PRIMER statistical program. One-way ANOVA was followed by the Student–Newman–Keuls test to assess significant differences with respect to either EC, in the case of blastogenetic cycle analysis, or colonies injected with DMSO in FSW, in the case of injection experiments. In ISH experiments, at least 200 cells were counted under the LM, in 10 optical fields, at 1250×. Data were compared with the χ2 test.
4. Discussion
In the present work we identified and characterized a new receptor with similarity with both human C3aR and C5aR receptors, and then named BsC3a/C5aR, from the colonial ascidian
B. schlosseri. Like the other described C3a and C5a receptors it belongs to the G protein-coupled family of 7TM-domain receptors. Analogously to the vertebrate C3aR, BsC3a/C5aR has a large extracellular loop, between the TM4 and TM5 domains, that, according to Chao et al [
10], is important for ligand binding. Indeed, the amino acid sequence alignment with the human orthologue demonstrates that, in this loop, the Tyr corresponding to the position 174 of the human sequence is conserved also in
B. schlosseri sequence, but not in the sequences of
Ciona R1 and
Halocynthia. In humans, this residue plays a critical role in the binding of human C3a and the consequent triggering of signal transduction [
10]; the binding is further strengthened by the negatively-charged Asp and Glu residues [
34] that are present, also in our case, in the above-cited loop. Therefore, we can argue that, in
Botryllus as in humans, the second loop is important for ligand binding and the activation of the effector cells.
Furthermore, the cytosolic C-terminus bears two Thr residues, corresponding to those located at positions 463 and 466 of the human sequence, that represent putative phosphorylation sites and are required for C3aR internalization [
35]. The result of multi-alignment indicates that none of the two amino acids are present in the R1 sequence from
C. intestinalis, whereas, only the second Thr is found in the
Halocynthia sequence. Receptor internalization is a fast feedback mechanism set up by immunocytes to avoid the deleterious effects of sustained complement activation and the presence of the two key residues reported above suggests that this regulation can take place also in our invertebrate model organism.
In addition to this putative control mechanism, we also demonstrated, in Botryllus, the presence of a positive autocrine feedback loop as the stimulation of BsC3a/C5aR with the C3aR agonist increases the transcription of BsC3 by morula cells. Since the production of C3a depends on the activation of the complement system upon the recognition of nonself, this positive loop can be a mechanism useful to enhance the inflammatory reaction required to kill and clear the foreign material.
In vertebrates, C3a and C5a receptors derive from duplication events which occurred early in their radiation, presumably during the invertebrate-vertebrate transition within chordates [
8]. Accordingly, the ascidian C3a/C5aR sequences cluster together as the sister group of the vertebrate C3aR and C5aR clades.
As an invertebrate,
Botryllus relies only on innate immunity and circulating immunocytes are responsible for both the cellular immunity and the synthesis and secretion of most of the humoral factors involved in immune responses. It has only two immunocyte types: phagocytes and morula cells. The former can easily ingest foreign cells or particles, especially when opsonized by rhamnose-binding lectin [
36] or complement [
37], whereas the latter are granular cells involved in inflammatory reactions [
22]. Morula cells are the first cells to sense the presence of nonself [
24] and, as a consequence of this recognition, they trigger an inflammatory reaction including the selective recruitment of additional morula cells in the infection site, their leakage from the circulation, and their degranulation with the induction of cytotoxicity [
36]. Morula cells are also the only cells synthesizing BsC3 upon the recognition of nonself [
17,
37].
In vertebrates, the C3a and C5a receptors are mainly expressed in immunocytes involved in innate immune responses, such as monocytes, macrophages, DCs, neutrophils, basophils, mast cells, and eosinophils, but also in lymphocytes, such as T cells [
9]. Here we demonstrate, using ISH, that only morula cells, i.e., the cytotoxic immunocytes able to synthesize BsC3, actively transcribe
bsc3a/c5ar. This is in agreement with what reported in
C. intestinalis, where the cells involved in complement-mediated inflammatory reactions are able to synthesize C3aR as well as C3 [
14].
In addition, in both qRT PCR and ISH, we observed a significant decrease of
bsc3a/c5ar transcription during the TO with respect to EC. Since we already reported a significant increase of the amount of
bsc3 transcription at TO with respect to EC [
37], the obtained data are indicative of the presence of an additional control against the deleterious effects of prolonged complement activation. It occurs in the form of a negative autocrine feedback loop blocking the transcription of
bsc3a/c5ar in the presence of a high amount of BsC3 in the circulation so to prevent a diffuse inflammation. Indeed, at the TO, a diffuse apoptosis occurs in tissues of old zooids. Additionally, cells are cleared by circulating phagocytes infiltrating the zooidal tissues without the induction of any inflammatory reaction, which is usually marked by the recruitment and degranulation of morula cells and the induction of cytotoxicity [
17].
Further studies are ongoing in our laboratory to better elucidate the relationships between BsC3 and BsC3a/C5aR expression in B. schlosseri.