LPS Challenge Regulates Gene Expression and Tissue Localization of a Ciona intestinalis Gene through an Alternative Polyadenylation Mechanism

A subtractive hybridization strategy for the identification of differentially expressed genes was performed between LPS-challenged and naive Ciona intestinalis. This strategy allowed the characterization of two transcripts (Ci8short and Ci8long) generated by the use of two Alternative Polyadenylation sites. The Ci8long transcript contains a protein domain with relevant homology to several components of the Receptor Transporting Protein (RTP) family not present in the Ci8short mRNA. By means of Real Time PCR and Northern Blot, the Ci8short and Ci8long transcripts showed a different pattern of gene expression with the Ci8short mRNA being strongly activated after LPS injection in the pharynx. In situ hybridization analysis demonstrated that the activation of the APA site also influenced the tissue localization of the Ci8short transcript. This analysis showed that the Ci8long mRNA was expressed in hemocytes meanwhile the Ci8short mRNA was highly transcribed also in vessel endothelial cells and in the epithelium of pharynx. These findings demonstrated that regulation of gene expression based on different polyadenylation sites is an ancestral powerful strategy influencing both the level of expression and tissue distribution of alternative transcripts.


Introduction
Alternative Polyadenylation (APA) has recently gained attention as a major player influencing the dynamics of gene regulation [1].
Usually, mature 39 ends of almost all eukaryotic mRNAs are created by a two-step reaction that involves an endonucleolytic cleavage of the pre-mRNA, followed by synthesis of a polyadenylate tail onto the upstream cleavage product. Polyadenylation influences many aspects of mRNA metabolism: transcription termination by RNAP II, mRNA stability, mRNA export to the cytoplasm and the efficiency of translation are all dependent on 39 processing.
In recent years it has become increasingly evident that APA regulates gene expression [2]. In some cases the alternative poly(A) + sites are located in internal introns/exons regions (Coding Region Alternative PolyAdenylation CR-APA) leading to different protein isoforms. In other cases, APA sites are all located in the 39untranslated region (UTR-APA), resulting in transcripts with 39 UTRs of different length but encoding the same protein. In this way, CR-APA can qualitatively affect the gene expression by producing distinct protein isoforms, whereas UTR-APA quantitatively affects the expression.
Ascidians (subphylum Tunicata) are chordate invertebrates whose immune system relies only on innate immunity including inflammatory humoral and cellular responses [3][4][5]. Due to the knowledge of the Ciona intestinalis genome [6], this ascidian has become a model to study the evolution of immune related genes [7]. In particular, previous research has shown that pharynx and hemocytes responses can be challenged by LPS inoculation through the body wall, therefore this experimental setting represents a well-established model to examine innate immunity gene expression. Previously published papers have described the inflammatory response and the immune role of C. intestinalis pharynx. In this respect, pharynx epithelia and hemocytes (mainly compartment/morula cells) express immune related genes (coding type IX collagen-like [8], TNFa-like [9,10], CAP-like [11], MBLlike [12], and galectin-like proteins [13]) upregulated by lipopolysaccharide (LPS). The pharynx occupies an extensive part of the adult body. It consists of two epithelial monolayers perforated by dorso-ventrally aligned rows of elongated elliptical, ciliated stigmata [14] enclosed in a mesh of vessels (also called transversal and longitudinal bars), where the hemolymph, containing abundant mature and immature hemocytes, flows. Hemopoietic nodules are associated with the bar epithelia that can be stimulated by mitogens [15,16]. In addition, in this organ, a C3-like protein gene is upregulated by LPS [17] suggesting the activation of a lectin-dependent complement-like system [18], while the activation of the proPO-system and an increased release of lectins with opsonic property have been shown [19,20].
In the present paper, a subtractive hybridization strategy for selective amplification of differentially expressed sequences showed that LPS challenge can induce an alternative polyadenylation mechanism in C. intestinalis. The LPS induced model correlates with the up-regulation and differential tissue localization of a novel gene.

Tunicates and LPS injection
Ascidians were collected from Sciacca Harbour (Sicily, Italy), a non-protected area in the Mediterranean sea, maintained in tanks with aerated sea water at 15uC and fed every second day with a marine invertebrate diet coraliquid (Sera Heinsberg, Germany) according to local guidelines. The work described in this study did not involve endangered or protected species. No specific permits were required for the described field studies.

Total RNA extraction and poly(A) + purification
Ascidian pharynx fragments (200 mg), excised at various times (from 1 to 72 hours), were immediately soaked in RNA later Tissue collection (Ambion, Austin, TX), and stored at 280uC. Total RNA extraction was performed by using an RNAqueous TM -Midi Kit purification system (Ambion, Austin, TX).
Poly(A) + RNA was prepared from control and injected animals (1 hour) using Illustra TM mRNA Purification Kit (GE Healthcare, UK) according to the manufacturer's instructions.

Subtractive hybridization and screening of the cDNA library
Subtractive hybridization was performed using the PCR-Select TM cDNA Subtraction Kit (Clontech Laboratories, USA) according to the manufacturer's instructions. This strategy is based on a PCR-based method for selective amplification of differentially expressed sequences allowing the isolation of transcript from activated tissues. Briefly, 2 mg of poly(A) + RNA from non-injected (driver) and injected (tester) animals (1 h p.i. of LPS) were retrotranscribed. The tester and driver cDNAs were digested with the restriction enzyme Rsa I to yield blunt ends. The tester cDNA was then subdivided into two parts and each was ligated with a different cDNA adaptor (ADAPTOR1: 59-CTAATACGACTCACTA-TAGGGCTCGAGCGGCCGCCCGGGCAGGT-39;  ADAP-TOR  2: 59-CTAATACGACTCACTATAGGG-CAGCGTGGTCGCGGCCGAGGT-39).The ends of the adaptor do not contain a phosphate group, so only one strand of each adaptor attaches to the 59 ends of the cDNA. Then two hybridizations were performed. In the first run, an excess of driver was added to each sample of tester. The samples were then heat denatured and allowed to anneal. In the second run of hybridization, the two primary hybridization samples were mixed together without denaturing to allow the subtracted single strand tester cDNAs to re-associate. These new hybrids were molecules with different ends, which correspond to the sequences of the two adaptors. After filling in the ends by DNA polymerase, the differentially expressed sequences display different annealing sites for the nested primers on their 59 and 39 ends. The entire population of molecules is then subjected to PCR to amplify the desired differentially expressed sequences using the following primers (Nested PCR Primer 1 59-TCGAGCGGCCGCCCGGG-CAGGT-39; Nested PCR Primer 2 59-AGCGTGGTCGCGGCCGAGGT-39) and PCR conditions (94uC for 3099, 68uC for 3099, 72uC for 1,59; 12 cycles). Screening of the library was performed hybridizing the subtracted library with P 32 labeled probes synthesized as first-strand cDNA from tester and driver. Clones corresponding to differentially expressed mRNAs will hybridize only with the tester probe, and not with the driver probe. P 32 labeled colonies were grown and plasmid DNA extracted.

Cloning and sequences analysis
Differentially expressed cDNA was cloned in the pCR4-TOPO vector (Invitrogen, USA) and sequenced. Sequence analysis identified a cDNA fragment of 102 nucleotides. Similarity searches performed using the FASTA algorithm (http://www.ebi.ac.uk/ Tools/fasta/) showed a relevant homology to some EST clones from mature adult Ciona intestinalis animal (data not shown). The full length sequence of the cDNA clone was obtained by using the GeneRacer TM kit (Invitrogen, USA). The kit ensures the amplification of only full length transcript via elimination of truncated messages from the amplification process. 59 RACE was performed by PCR (94uC 1 min, 52uC 1 min, 72uC 1 min for 30 cycles) using the Ci8 59Race R specific oligonucleotide (59-CATCCACCACCAACAGGAA-39) (see Figure 1 for details) and the GeneRacer TM 59-oligonucleotide (59-CGACTGGAGCAC-GAGGACACTGA-39).The 59 RACE technology has identified only one fragment of 379 bp; The 39 RACE was performed using the Ci8 39Race F specific oligonucleotide (59-GTGCAAATGGGGTGAGCTAT-39) and the GeneRacer TM 39 oligonucleotide (39-GCAATGCATCGCA-TAGCAACTGTCG-59). PCR products were diluted 1:100 and re-amplified using the Ci839Race Nested F specific oligonucleotide (59-GTTCCCATTCAACTACCGGT-39) and the GeneRacer TM 39 nested oligonucleotide (59-GTTCCCATTCAAC-TACCGGTT-39) (see Figure 1 and 2 for details). By means of 39RACE we were able to identify two cDNA fragments: a first one of 1370 bp and a second one of 176 bp. DNA fragments were purified and cloned in the pCR4-TOPO vector (Invitrogen, USA) and sequenced. Sequence analysis showed that both fragments contains a common 50 bp region overlapping with the originally isolated 102 bp fragments (see figures 1 and 2 for details).
In order to uniquely identify the 59 sequences of the two 39 RACE cDNA fragments, a second step of 59RACE analysis was performed. The full length longer cDNA was isolated by RT PCR using the Ci8long 39 UTR R oligonucleotide and the GeneRacer TM 59-oligonucleotide (named Ci8long). The full length shorter cDNA was isolated by PCR using the Ci8short 39UTR R and the GeneRacer TM 59-oligonucleotide (named Ci8short). Two fragments of 1662 and 486 bp were isolated, purified and cloned in the pCR4-TOPO vector (Invitrogen, USA).

Sequence, structural and phylogenetic analysis
Similarity searches were performed using the FASTA program (http://www.ebi.ac.uk/Tools/fasta/). Multiple alignments were accomplished with the Clustal W program [21]. The final sequence alignment was done using CLUSTAL W v.1.81 [21] and the similarity shaded with CLC workbench 6.4. A phylogenetic tree was constructed by the Neighbor-Joining method (NJ) after 1000 bootstrap iterations by using CLC Structural prediction was performed using the PSIPRED Protein Structure Prediction Server (http://bioinf.cs.ucl.ac.uk/ psipred/) and the Predict Protein algorithm (http://www. predictprotein.org/).

Gene expression analysis
Tissue differential expression of the two mRNAs was studied by Real-Time PCR using the Sybr-Green method. To discriminate the two transcripts specific sets of primers were designed on 39UTR regions using Custom Primers OligoPerfect TM Designers software (https://tools.invitrogen.com/) and synthesized commercially (Eurofins MWG Operon, Ebersberg, Germany). Real-time PCR analysis was performed using the Applied Biosystems 7500 real-time PCR System. Ci8long isoform tissue expression was performed in a 25 ml PCR reaction containing 2 ml cDNA converted from 250 ng of total RNA, 300 nM Ci8long 39UTR forward primer (59-TTGCATTTTATTCCATCATTGC-39) and Ci8long 39 UTR Reverse primers (59-TTGCGCA-TAAGCTTGGTTTA-39), 300 nM actin forward (59-TGATGTTGCCGCACTCGTA-39) and reverse (59-TCGA-CAATGGATCCGGT-39) primers, and 12.5 ml of Power Sybr-Green PCR Master Mix (Life Technologies, Milan, Italy).
The 50 cycles of the two-step PCR program consisted of initial polymerase activation for 3 min at 95uC followed by denaturing step at 95uC for 15 sec, and then the annealing/extension was carried out at 60uC for 45 sec when the fluorescent signal was detected. Each set of samples was run three times and each plate contained quadruplicate cDNA samples and negative controls.
The specificity of amplification was tested with real time PCR melting analysis. To obtain sample quantification, the 2 2DDCt method was used and the relative changes in gene expression was analysed as described in the Applied Biosystems Use Bulletin N.2 (P/N 4303859). The amount of Ci8long and Ci8short transcripts from different tissues was normalized to actin in order to compensate for variations in input RNA amounts. Relative Ci8long and Ci8 short expression was determined by dividing the normalized value of the target gene in each tissue by the normalized value obtained from the untreated tissue.
Northern blot analysis was performed as previously described [22]. A nucleotide fragment corresponding to the coding region of the Ci8short cDNA (including the common region between the two mRNAs) was labelled with a-CTP32 and the Rediprime II DNA Labeling System (GE Healthcare Life Science, Milan, Italy). Membrane was exposed to a Kodak X-Omat AR X-ray film for 48 hours.
region were in upper case letters; the first ATG and the STOP codon were underlined; sequence in bold displays the 102 bp fragment identified by Subtractive Hybridization. Box shows the canonical C.intestinalis polyadenylation site. The arrows indicate the oligonucleotides used for cloning procedures, Real time PCR and ISH assay. doi:10.1371/journal.pone.0063235.g001  Pharynx explants preparation and histology The tunic surface was cleaned and sterilized with ethyl alcohol and pharynx fragments (200 mg) were excised from the injection site of sham and LPS challenge ascidians. For in situ hybridization studies, pharynx fragments were fixed in Bouin9s fluid (saturated picric acid:formaldehyde:acetic acid 15:5:1) for 24 hours, paraffin embedded, and serially cut at 6 mm (Leica RM2035 microtome, Solms, Germany).

In situ hybridization assay (ISH)
To examine tissue excised from the inflamed body wall, ISH was carried out with digoxigenin-11-UTP-labeled riboprobes (1 mg/ml final concentration). The Ci8long probe was generated by PCR amplifying a cDNA fragment of 165 bp covering the 39untranslated region from nucleotide 1496 to nucleotide 1662 of the isolated cDNA using the Ci8long 39UTR forward oligonucleotide (59-TTGCATTTTATTCCATCATTGC-39) and the Ci8long 39UTR reverse oligonucleotides (59-TTGCGCA-TAAGCTTGGTTTA-39) (see Figure 1).
The DNA fragment was cloned in the pCR4-TOPO vector (Invitrogen, USA). The Ci8short probe was generated by PCR amplifying a cDNA fragment of 138 bp covering the 39untranslated region from nucleotide 348 to nucleotide 486 of the isolated cDNA using the Ci8 short 39UTR forward primer (-59TACCGGTTGTTCCTGTTGGT-39)and the Ci8 short 59UTR Race Reverse specific oligonucleotide (59-GACGTCAT-CAGACTTCTAAATGCT-39) (see Figure 2).
The digoxigenin-11-UTP-labeled riboprobes was carried out according to manufacturer9s instructions (Roche Diagnostics). The re-hydrated histological sections were digested with proteinase K (10 mg/ml) in PBS for 5 min, washed with PBS-T, and treated for hybridization with 50% formamide, 5X SSC, 50 mg/ml heparin, 500 mg/ml yeast tRNA, and 0.1% Tween 20, at 37uC overnight. After exhaustive washing in PBS-T and 4XSSC (twice for 10 min), the sections were incubated for 1hr with anti-DIG-Fab-AP conjugate (Roche Diagnostics, Milan, Italy) diluted 1:500 and washed in PBS-T. Finally, the sections were incubated in the 5bromo-4-chloro-3-indolyl phosphate/nitro blue tetrazolium liquid substrate system (Sigma-Aldrich, Milan, Italy). Colour development was stopped after 30 min at room temperature.

Statistical methods
Student9s t-test was used to estimate statistical significance. Multiple comparisons were performed with one-way analysis of variance (ANOVA) and different groups were compared by using Tukey9s t-test. Standard deviations were calculated on four experiments. P,0.01 was considered statistically significant.  (Figure 3 panel A). Alignments between the Ci8long and the Ci8short deduced amino acid sequences showed that the Ci8short protein represents a shorter form of the Ci8long protein (Figure 3

panels A and B).
A search in Ensembl genome browser performed with the Ci8long nucleotide sequence identified a five exons and four introns gene (ENSCING00000009651) localized on Chromosome 5: 555,293-559,003. This analysis identified a unique transcript (ENSCINT00000019621) for this gene. Then, a more detailed analysis was performed aligning the nucleotide sequences of the Ci8long, the Ci8short and the sequence of the annotated transcript (ENSCING00000009621). The Ci8long matches with the entire coding sequence of the annotated transcript. A comparison between the Ci8short versus the annotated genomic sequence showed that it matches with the 59untranslated region, the first 218 nucleotides of the coding region (corresponding to the first exon sequence) plus 91 nucleotides lying within the first intron of the gene (Figure 3 Panel B). In this region, we identified a noncanonical polyadenylation site (AAUACA) between nucleotides 402-407. In addition, two conserved tetranucleotides elements (UGUA) were identified in the positions 433-436 and 442-445, respectively ( Figure 4).
On the other hand, the Ci8long 39 UTR was analysed for the presence of polyadenylation sites. The Ci8long cDNA displays an ATTAAA sequence located between nucleotides 1640 and 1646 which is considered the most frequent variant of the canonical polyadenylation site [23] (Figure 1). In conclusion, in silico analysis showed that the 39 untraslated regions of the two mRNAs differ in the length, sequence and polyadenylation signals.

In silico structural analysis
In silico structural analysis of the Ci8long protein showed two putative transmembrane regions between aa 72-90 and aa 173-192 (Figure 3 panel C). None of these regions were detected in the Ci8short deduced sequence. In addition, the residue composition analysis of the Ci8short deduced sequence revealed a high percentage of proline and glycine residues (22% and 13%, respectively).  (Figure 5).

Phylogenetic analysis of the Ci8long RTP-like domain
The phylogenetic tree, constructed by comparing vertebrate and invertebrate components of the RTP family, showed the following main clusters. The first one containing fish RTP2 (D.rerio) and RTP3 (S. salar, O. mykiss), the Ci8long and the arthropod RTP sequences (I. scapularis). The second one consists of mammal RTPs separated into the RTP3 and RTP4 subgroups (H. sapiens, M. musculus, B.taurus, S.scrofa) and the RTP1 and RTP2 subgroups. The third one consists of the amphibian X. tropicalis and the flatworm C.sinensis RTP 3-like ( Figure 6).

Differential expression of Ci8long and Ci8short transcripts disclosed by Real Time PCR and Northern blotting
To study the expression pattern of the Ci8long and Ci8short mRNAs, specific primers were designed within the 39 UTR of the two cDNAs (see Figure 1 and Figure 2). Quantitative mRNA expression of Ci8long and Ci8short in naive, sham and LPS challenged ascidians was examined by Real Time PCR analysis. Four naive, sham and LPS-treated ascidians in three distinct experiments were examined at different post-inoculation time points (1,4,8,12,24,48, 72 h). The LPS treated ascidians were compared to specimens inoculated with marine solution, and the latter compared to naive ascidians.
In the LPS-treated ascidians, Ci8short expression, compared to the Ci8long one, disclosed a significantly higher RNA level at all time points (P,0.01). In particular, the Ci8short expression was enhanced at 1 h and reached a maximum of expression 12 h p.i., then decreased at 72 h p.i. (Figure 7, panel A).
The Ci8long mRNA level was slightly enhanced at 1 and 12 h p.i. (Figure 7, panel B). The inoculation procedure (sham ascidians) slightly modulated the expression levels in comparison to the naive specimens (Figure 7, panel A e B).
In addition, Figure 7 panel C shows the comparison of the level of expression of the Ci8short mRNA versus the Ci8long one. This assay demonstrates that the number of molecules of the short mRNA is statistically higher in the LPS challenged ascidians at all the time points.
The data are in agreement with the Northern blot assay showed inside Figure 8. Total RNA, from pharynx from naive and LPSchallenged ascidians 1 hour and 12 hours p.i., were fractionated and hybridized with a P 32 labeled probe covering the coding region of the Ci8short cDNA. This analysis showed a faint band corresponding to the size of the Ci8long mRNA in all the lines and an increasing hybridization signal in LPS-challenged ascidians in a region of about 500 nucleotides corresponding to the size of the Ci8short mRNA.
In situ hybridization assay of pharynx Figure 9 shows histological sections of the pharynx containing hemocytes from sham (panels A and F) and LPS-treated ascidians (panels B,C,D,E,G and H) 12 h p.i.
The Ci8short localization in the pharynx from LPS-treated ascidians shows an enhanced gene expression (Figure 9 panels B,C,D and E) when compared to the sham (panel A). In particular, a large part of the vessels appeared to be densely populated with hemocytes expressing the Ci8short transcript (Figure 9 panels B,C). The Ci8short is manly expressed by endothelial cells (panel d) that appeared to be marked in various regions of the pharynx bars (panel E).
On the contrary, differences in Ci8long transcript expression could not be observed between sham (Figure 9, panel F) and LPS treated ascidians (Figure 9, panel G). Figure 9 panel H shows that the Ci8long transcript is mainly expressed in compartment/morula and signet ring cells located in tightly packed cluster within the vessel lumen (Figure 9 panel H, insets).
Histological sections treated with the sense strand (negative control) did not display any positive staining.

Discussion
In recent years it has become evident that APA is an important mechanism in vertebrate and invertebrate organisms to produce different protein isoforms (coding region-APA) or regulate gene expression (UTR-APA). Differential processing at multiple poly(A) sites can be influenced by physiological and pathological conditions such as cell growth, differentiation, development, cancer and stress condition such as inflammation [1].
In this paper, we used a subtractive hybridization strategy on the attempts to identify LPS differentially expressed sequences in the C.intestinalis pharynx tissue that has been retained to be the main protagonist of the innate immunity responses. This strategy allowed us the identification of two mRNAs (Ci8long and Ci8short) derived from the transcription of the (ENSCING00000009651) annotated gene. In particular, LPS was able to weakly modulate the expression of the Ci8long transcript and to induce the activation of a LPS-induced APA mechanism responsible for the generation of a shorter mRNA (Ci8short). In fact, in silico analysis identified a non-canonical polyadenylation site within the first intron of the annotated gene. This region was composed by the hexanucleotide AATACA followed by two short tetranucleotides (TGTA). The latter sequences have been shown to be involved in alternative polyadenylation events in vertebrate binding specific cleavage factors [24].
Sequence analysis showed that the Ci8long deduced amino acid sequence displays a protein domain with homology to components of the Receptor Transporting Protein (RTP) family [25]. The RTP family is composed of four members (RTP1-4) who were first identified as partners for mammalian odorant receptors, promoting cell surface expression and leading to functional responses in heterologous cell system. RTP1 and RTP2 are expressed in   olfactory neurons and vomeronasal neurons, RTP3 is expressed in liver, lung and testis and RTP4 is expressed in a wide variety of tissues including lymph nodes, peripheral blood leucocytes, spleen and thymus (reviewed in [25]). The mechanism of action of this family of proteins is poorly understood and the existence of several closely related family members with disparate phenotypes suggests a wide role of these proteins.
The phylogenetic analysis supports that the CiRTP-like domain identified in the Ci8long RNA is a component of the RTP family while discloses the close relationship of RTPs inside the chordate clade sharing a common ancestor. Moreover, the RTP-like sequence found in the arthropod I. scapularis suggests a more ancient progenitor. A similar consideration arises from the phylogenetic branch formed with the amphibian X. laevis and the flatworm C. sinensis RTP3-like sequences.
We do not know the functional role of the presumptive RTPlike protein as deduced from Ci8long cDNA sequence as well as of the presumptive protein encoded by the Ci8short sequence. In any case, the short isoform do not contain the RTP domain and do not display any other homolog in the data banks different from the C.intestinalis annotated transcript (data not shown).  Furthermore, in silico prediction demonstrated that the Ci8long derived protein contains two transmembrane regions which are not present in the Ci8short protein suggesting that the short isoform may represent an LPS induced secreted form of the constitutively expressed gene.
In fact, it is noteworthy that Ci8long is expressed both in naive and LPS-challenged ascidians while the Ci8short transcript is significantly enhanced during the inflammatory process.
In particular, as demonstrated by means of Real Time PCR, the Ci8short expression profile showed a peak of activation within 1 hour p.i. followed by a second wave of activation at the stage of 12 hours. Interestingly, a similar pattern of activation has been observed for the expression of other LPS-induced components of the C. intestinalis inflammatory response [8,9,[11][12][13]. These data are in agreement with the Northern blot analysis.
Furthermore, the tissue localization of the Ci8short and Ci8long transcripts showed that LPS inoculation also induced a differential tissue localization of the two mRNAs probably related to the APA mechanism. The Ci8long transcript was expressed in some hemocytes of pharynx vessels, whereas the Ci8short mRNA appears to be strongly up regulated in compartment/morula and signet ring cells as well as in vessel endothelial cells and epithelium. In this respect, compartment/morula cell types are known to populate inflamed tissues engaged in the expression of immune related genes [8,9,[11][12][13]. Although precise quantitative data were not derived from the histological observations, the possibility that an increased number of Ci8short positive hemocytes in the vessels as well as positive regions of the endothelium can be related to LPS inoculation is supported by previous published papers [8,9,[11][12][13].
Finally, the finding that pharynx tissues gene expression can be upregulated by LPS is consistent with evidence on C3-like gene expression [17], and supports the finding that ascidian pharynx is involved in immune-surveillance. This is in accordance with the role of this organ that comes in contact with a large variety of microbes exerting an early recognition of Pathogen-associated molecular patterns. Conclusions This paper reports on the first identification of a CiRTP-like protein and an LPS induced APA mechanism in the invertebrate chordate Ciona intestinalis. In this respect, we showed that the mechanism based on different polyadenylation sites is an ancestral powerful strategy of gene regulation interfering with the level of expression and tissue distribution of alternative transcripts.