Analysis of Soluble Protein Contents from the Nematocysts of a Model Sea Anemone Sheds Light on Venom Evolution
- First Online:
The nematocyst is one of the most complex intracellular structures found in nature and is the defining feature of the phylum Cnidaria (sea anemones, corals, jellyfish, and hydroids). This miniature stinging organelle contains and delivers venom into prey and foe yet little is known about its toxic components. In the present study, we identified by tandem mass spectrometry 20 proteins released upon discharge from the nematocyst of the model sea anemone Nematostella vectensis. The availability of genomic and transcriptomic data for this species enabled accurate identification and phylogenetic study of these components. Fourteen of these proteins could not be identified in other animals suggesting that they might be the products of taxonomically restricted genes, a finding which fits well their origin from a taxon-specific organelle. Further, we studied by in situ hybridization the localization of two of the transcripts encoding the putative nematocyst venom proteins: a metallopeptidase related to the Tolloid family and a cysteine-rich protein. Both transcripts were detected in nematocytes, which are the cells containing nematocysts, and the metallopeptidase was found also in pharyngeal gland cells. Our findings reveal for the first time the possible venom components of a sea anemone nematocyst and suggest their evolutionary origins.
KeywordsNematocyst Toxin Cnidaria Nematostella Venom
Cnidaria is a phylum that includes a wide variety of marine animals such as sea anemones, corals, jellyfish, and hydroids. All cnidarians are carnivores and utilize venom in order to catch their prey and defend themselves from predators. The defining hallmark of this animal group is the nematocyst, a highly complex proteinaceous structure made of a capsule containing an inverted tubule, capable of extremely fast and powerful discharge (David et al. 2008; Kass-Simon and Scappaticci 2002; Nuchter et al. 2006). Nematocysts are found inside cells called nematocytes, also known as stinging cells. These cells are considered to produce the toxins and their nematocysts are miniature injectors which deliver the venom into their prey or predator (Kass-Simon and Scappaticci 2002). However, only in few cases toxins were directly shown to reside in the nematocyst capsule (Hessinger et al. 1973; Honma et al. 2003; Lotan et al. 1995; Schlesinger et al. 2009).
Compared to other venomous animals such as snakes or scorpions, relatively little is known about cnidarian toxins and their biological activity. However, in contrast to the situation in other cnidarians, peptide toxins from sea anemones (Actiniaria) are relatively well studied. The vast majority of known toxins from sea anemones can be divided into three functional groups: (a) toxins modulating voltage-gated sodium channels (Moran et al. 2009; Wanke et al. 2009), (b) toxins which block or modulate voltage-gated potassium channels (Castaneda and Harvey 2009), and (c) cytolytic toxins disrupting membranes (Anderluh and Macek 2002).
The starlet anemone Nematostella vectensis has become a major model for the study of evolutionary developmental biology since unlike many other cnidarians it can be grown throughout its full life cycle in the lab and advanced molecular tools for its study are available (Darling et al. 2005; Technau and Steele 2011). These tools, including gene knockdown and transgenesis techniques (Nakanishi et al. 2012; Renfer et al. 2010; Technau and Steele 2011), put Nematostella in a unique position for the study of cnidarian toxin production and delivery. Moreover, in light of the recent emergence of sea anemone nematocysts as a potential drug delivery device (Ayalon et al. 2011), the ability to maintain large Nematostella cultures makes this species an attractive nematocyst source. However, despite its growing popularity as a lab model, until recently very little was known about its venom. A bioinformatic search of the N. vectensis genome sequence revealed that this species contains only one toxin homologous to previously described sea anemone toxins (Moran and Gurevitz 2006). The toxin, called Nv1, belongs to the type I sea anemone toxin group and like other members of this group it inhibits the activation of voltage-gated sodium channels, resulting in strong contractile paralysis and death of arthropods and fish (Moran et al. 2012b; Moran et al. 2008). Unexpectedly, Nv1 was localized to ectodermal gland cells in the tentacles rather than nematocysts and was shown to be released in massive amounts to the medium upon tentacle contact with prey (Moran et al. 2012b). These findings raise the question which venom proteins, if any, are produced by N. vectensis nematocytes. To answer this question, we have isolated nematocysts from N. vectensis, discharged them in vitro and analyzed the released protein mixture by tandem mass spectrometry (MS/MS). In addition to pointing out putative toxins and auxiliary venom proteins, the analysis uncovers intriguing evolutionary links between venom and non-venom proteins and reveals a collection of taxonomically restricted putative venom proteins.
Materials and Methods
Capsule Isolation and Discharge and Isolation of Secreted Proteins
Capsule isolation was carried out in a similar fashion to a previously published method (Zenkert et al. 2011). In brief, the whole 6-month-old N. vectensis polyps were frozen in liquid nitrogen and then homogenized in isolation solution (50 % Percoll, 10 % sucrose and 0.003 % Triton X-100). The mixture was then centrifuged at 2,000×g, 4 °C for 10 min.
Isolated Nematostella capsules were activated for 15 min by water. Discharge and capsule purity were verified under the microscope. The discharged suspension was centrifuged at 20,000×g, 4 °C for 10 min. The supernatant was lyophilized and sent to the Smoler Proteomics Center in the Technion (Israel Institute of Technology) for protein identification by tandem mass-spectrometry (MS/MS).
Tandem Mass-Spectrometry (MS/MS) and Protein Annotation
The proteins from the samples were trypsinized and the tryptic peptides were analyzed by LC-MS/MS on the Orbitrap mass spectrometer (Thermo). The MS data was analyzed using the Sequest 3.31 software (Thermo) vs. the cnidaria section of the non-redundant protein sequences dataset (nr) of the National Center for Biotechnology Information (NCBI; http://blast.ncbi.nlm.nih.gov/Blast.cgi) and scores for each hit were calculated (Table S1). Conserved domains were detected using the CDD tool (Marchler-Bauer et al. 2011).
RNA Isolation and Rapid Amplification of cDNA Ends (RACE)
Total RNA was isolated from 9-day-old primary polyps of N. vectensis using Trizol (Ambion, USA) according to the manufacturer’s instructions. The PolyA RNA was selected using the PolyATract mRNA isolation system III (Promega, USA). The isolated polyA RNA was used as template for all reverse transcription reactions performed. 5′ and 3′ RACE experiments were conducted using the RACE SMARTer kit (Clontech, USA) according to manufacturer’s instructions. Advantage2 DNA polymerase mix (Clontech) was used for PCR under the touchdown conditions suggested in the RACE SMARTer kit manual. The product of each initial PCR reaction in a final dilution of 1:1,000 served as template for a nested PCR. All primer sequences are available in Table S2. The PCR products were ligated into pGEM-T (Promega) and sequenced from both sides. The full transcripts encoding NEP-6 and NEP-16 were deposited in GenBank (Accession numbers JQ829079 and JQ829080).
The boundaries of CAP and Astacin domains were determined according to PFAM. The domains were aligned using MUSCLE and for the SCP domains low quality alignment regions were removed by TrimAl (Capella-Gutierrez et al. 2009; Edgar 2004). ProtTest was used to find the most suitable model for phylogeny reconstruction (Abascal et al. 2005). For both domain alignments, the maximum-likelihood (ML) phylogenetic tree was constructed using PhyML with the WAG Model (+I + G), which got the highest score in the ProtTest analysis. Support values were calculated using 100 bootstrap replicates. A Bayesian tree was constructed using MrBayes version 3.1.2 with the same model. The run lasted 5,000,000 generations and every 100th generation was sampled. We estimated that the Bayesian analysis reached convergence when the potential scale reduction factor reached 1.0.
Single and Double In Situ Hybridization (ISH)
For ISH experiments, N. vectensis larvae were fixed at 48–168 h post fertilization in ice-cold 3.7 % formaldehyde in 1/3 seawater with 0.2 % glutaraldehyde for 90 s and then in 3.7 % formaldehyde in 1/3 seawater without glutaraldehyde for additional 60 min. Transcript fragments were amplified by PCR and cloned into pGEM-T. Antisense RNA probes for ISH were generated and labeled by using the T7 or SP6 MEGAscript kits (Ambion) and an RNA labeling mix with either digoxygenin (DIG) or fluorescein (FITC; Roche, Germany). The ISH procedure for single probes was performed as described previously using DIG-labeled probes (Genikhovich and Technau 2009). For double in situ, a DIG-labeled and a FITC-labeled probe were hybridized simultaneously according to the single probe protocol (Genikhovich and Technau 2009). After the hybridization step and the following washes, sheep anti-FITC coupled to alkaline-phosphatase (Roche) was applied at a dilution of 1:2,000 in blocking reagent (Roche) and incubated overnight at 4 °C. The next morning, the samples were washed 10 times with phosphate buffer saline containing 0.1 % Tween-20 (PTw) and then were incubated in 0.1 M Tris–HCl (pH 8.2) for 5 min twice. Then FastRed reagent (Roche) was applied in the same buffer. After development of a strong red signal, the reaction was stopped by five quick washes followed by an inactivation of the enzyme by a single wash in 0.1 M glycine–HCl (pH 2.2) for 10 min at room temperature. After five additional washes in PTw, the samples were blocked for 1 h at room temperature with blocking reagent solution. Then sheep anti-DIG alkaline-phosphatase-coupled antibody (Roche) was applied at a concentration of 1:3,000 in blocking reagent and the samples were incubated overnight at 4 °C. The next morning the samples were washed ten times with PTw and nitroblue tetrazolium chloride (NBT)/5-bromo-4-chloro-3-indolyl-phosphate (BCIP) blue signal was developed as in the single in situ procedure (Genikhovich and Technau 2009). The staining was stopped by three washes in PTw and stained sampels were mounted either in SlowFade Gold medium (Invitrogen) or 85 % glycerol and photographed in a Nikon Eclipse 80i fluorescent microscope connected to a Nikon Digital Sight DS-U2 camera.
Results and Discussion
The Proteins Released from Nematostella Nematocysts Upon Discharge
Proteins detected by tandem mass spectrometry in the secretion of discharged nematocysts of Nematostella and their homologies
No. of measured peptides
Best Nemaostella hit
Best non-model BLAST hit
Conserved domains and protein models
2 Endoglucanase C-terminal domain(COG4305); Chitin binding Peritrophin-A domain (PF01607)
RNA polymerase sigma-B factor (TIGR02941)
BAJ22666; nematogalectin Hydra vulgaris
Galactose binding lectin domain (PF02140)
ABC88377; Tolloid Nematostella vectensis
BAJ22673 nematogalectin related Aurelia aurita
Collagen triple helix repeat (PF01391). Galactose binding lectin domain (PF02140).
AAH59560 Ppib protein Danio rerio
Cyclophilin type peptidyl-prolyl cis–trans isomerase (PF00160)
Inosine-uridine preferring nucleoside hydrolase (PF01156)
AAX98722 Tolkin Drosophila simulans
CAP, Cysteine-rich secretory protein family (PF00188)
CAP, Cysteine-rich secretory protein family (PF00188)
Inosine-uridine preferring nucleoside hydrolase (PF01156)
AAZ79235 Slit1b Danio rerio
2 EGF-like domains (PF00008)
CAP, Cysteine-rich secretory protein family (PF00188)
Putative Venom Metallopeptidases in Nematostella are Members of the Tolloid-Related Family
CAP Domains in the Putative Venom Proteins of Nematostella
Convergent Toxin Recruitment Versus Novel Taxonomically Restricted Genes
In recent years with accumulation of data regarding venom components in various animals, it became clear that the same non-toxic protein families are re-currently becoming toxins via gene-duplication, accompanied by adaptive evolution and differential expression (Fry et al. 2009a; Fry et al. 2009b). The fact that the same gene families are recruited numerous times independently suggests that the non-toxic activity of these proteins can easily become toxic upon expression in a different context. Proteases and peptidases of all kinds can cause immediate tissue damage when lacking restraining selectivity, making them prime candidates for toxins in the venoms of diverse animal groups (Fry et al. 2009a). NEP-6 and NEP-14 are Astacin domain proteins that were probably recruited to the nematocyst merely due to their proteolytic activity. However, their clustering with Tolloid and Tolloid-like proteins suggest that their ancestor genes were involved in development (Fig. 1). The recruitment of Astacin-like proteins to spider venom and of many other non-Astacin peptidases to the venom of snakes demonstrates that the exact source of the proteolytic activity of venom is of little importance. In the cnidarian Hydractinia echinata an Astacin of another subfamily is expressed in developing nematocyte precursors (Mohrlen et al. 2006). Further, Astacins from Hydra were also shown to be present in the nematocyst capsule (Balasubramanian et al. 2012). It would be interesting to test whether these hydrozoan Astacins are secreted from the capsule upon discharge. A positive result will be pointing to a role as toxins and to an intriguing scenario where convergent recruitments of Astacin to venom occurred several times within Cnidaria.
Out of 20 proteins identified by MS/MS, 6 (30 %) do not show any clear homology to other proteins, and do not contain any conserved domains. Additional 8 proteins have at least 1 conserved domain but still do not exhibit profound homology (>25 % similarity) to any other metazoan protein currently present in the non-redundant protein database of the NCBI (Table 1). This means that in total 14/20 (70 %) of the proteins have no known homologue in another metazoan and may be considered as taxonomically restricted genes (TRGs; also known as orphan genes; Khalturin et al. 2009). This is a strong enrichment compared to results from recent re-annotation of the Nematostella transcriptome that found 16 % of the protein models to lack metazoan homologues and 5 % to lack both homologues and conserved domains (Fredman D. and Technau U. unpublished results). This is consistent with genetic studies in Hydra, indicating that a substantial fraction of the genes exclusively expressed in nematocytes are TRGs (Hwang et al. 2007; Milde et al. 2009). As the nematocyst is a unique cnidarian structure conserved for more than 600 million years it is plausible that many of its component proteins including toxins will be encoded by TRGs. Some nematocyst structural components, which are considered as TRGs such as Nematogalectins and Minicollagens, seem to be conserved among a wide range of cnidarians (David et al. 2008; Hwang et al. 2010), but as toxins are facing very strong selection due to prey-predator “arms race” their evolutionary turnover is usually much higher (Barlow et al. 2009; Duda and Palumbi 1999). CAP proteins in the Nematostella nematocyst lumen might represent a putative toxin class that is conserved in stony corals (Fig. 3), a remarkable feat when considering that the divergence time of stony corals and anemones is estimated at 500 million years ago (Shinzato et al. 2011). Thus, it is possible that CAP containing proteins represent an ancient toxin class whereas the other proteins we detected are newer venom recruits in the Nematostella lineage.
Multiple Venom Sources in Sea Anemones
The localization of Nv1 and other type I toxins to ectodermal gland cells revealed a new venom producing cell population in sea anemones (Moran et al. 2012b). In the present work, we show evidence suggesting that nematocysts in Nematostella produce toxins of classes never described before for a sea anemone (Table 1). In the vast majority of past studies, toxins were purified from whole tentacles by harsh chemical extractions, which put in risk the structural integrity of many proteins. Thus, it is difficult to assess the completeness of the detected arsenal and to what degree nematocyst toxins were represented in the peptides studied in the last 37 years by activity-guided fractionation (Béress et al. 1975; Bruhn et al. 2001; Diochot et al. 2004; Peigneur et al. 2011; Schweitz et al. 1981; Turk and Kem 2009). It is possible that many more sea anemone toxins are waiting to be discovered by finer methods like water-induced discharge of isolated nematocysts. In a recent study, tentacles were treated by dipping in distilled water for discharging nematocysts, resulting in an unprecedented richness of toxin peptides from just two anemone species (Rodriguez et al. 2011). Nevertheless, we suggest this method is likely to release the content of toxin-producing gland cells as well, since these are ectodermal cells located on the very outer boundary of the anemone tentacle (Moran et al. 2012b).
It is currently unknown why certain toxins are secreted from nematocytes while others are produced in gland cells. However, it is clear that some toxins are produced by more than one cell population: the type I toxins of Anemonia viridis are produced in both ectodermal gland cells of the tentacle and nematocytes whereas NEP-6 of Nematostella is found in both body wall nematocysts and gland cells of the pharynx (Fig. 2; Moran et al. 2012b). The expression of the same toxin by both gland cells and nematocysts supports the theory that nematocysts evolved originally from toxin secreting gland cells (Balasubramanian et al. 2012; Moran et al. 2012b). The expression of toxins in Nematostella and probably in other cnidarian species seems to be complex as distinct cell types express peptide toxins in different parts of the animal in various life stages for diverse functions such as prey capture, prey disintegration, and defense (Moran et al. 2012b). How such a complex system is regulated and what are the unique features of each toxin producing cell type and its biochemical arsenal remains to be described.
Nematostella nematocysts contain at least 20 proteins that are released upon capsule discharge. Some, like the Astacin domain metallopeptidases have a clear potential to act as potent toxins causing damage in stung prey or predator. Many others of these proteins lack clear homology to known proteins and are therefore regarded as taxon specific traits. Their detection opens the door for follow-up studies regarding their venomous function and origin. Our findings are setting the foundation for the study of nematocyst venom and its evolution in a rising lab model species with established experimental tools rarely available for a venomous animal.
We thank David Fredman (University of Vienna) for sharing his data and Michael Gurevitz (Tel Aviv University) for his critical comments on the manuscript. This work was supported by a grant of the Austrian National Science Foundation FWF (P22618-B17) to UT. YM is a Marie Curie Intra-European postdoctoral fellow.
This article is distributed under the terms of the Creative Commons Attribution License which permits any use, distribution, and reproduction in any medium, provided the original author(s) and the source are credited.
Open AccessThis article is distributed under the terms of the Creative Commons Attribution License which permits any use, distribution, and reproduction in any medium, provided the original author(s) and the source are credited.