Apoptotic gene loss in Cnidaria is associated with transition to parasitism

doi:10.21203/rs.3.rs-2238991/v1

Download PDF

Research Article

Apoptotic gene loss in Cnidaria is associated with transition to parasitism

https://doi.org/10.21203/rs.3.rs-2238991/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background

The phylum Cnidaria consists of several morphologically diverse classes including Anthozoa, Cubozoa, Hydrozoa, Polypodiozoa, Scyphozoa, Staurozoa, and Myxozoa. Myxozoa comprises two subclasses of obligate parasites – Myxosporea and Malacosporea, which demonstrate various degrees of simplification. The subclass Myxosporea is especially simplified, lacking any normal multicellular body plan and embryonic development. Myxosporea lack the majority of core protein domains of apoptotic proteins including caspases, Bcl-2 and APAF-1 homologs. Other sequenced Cnidaria, including the parasite Polypodium hydriforme from Polypodiozoa do not share this genetic feature. Whether this loss of core apoptotic proteins is unique to Myxosporea or also present in its sister subclass Malacosporea was not previously investigated. It is also unclear whether Myxosporea lost their apoptotic proteins abruptly or gradually (starting with their common ancestors with Malacosporea) during their evolutionary history.

Results

Using comparative genomics and transcriptomics approaches we describe a detailed picture of apoptotic protein gene loss in parasitic cnidaria in relation to each other and free-living members of the phylum. We show that Polypodium hydriforme lost the main components of the extrinsic apoptotic pathway such as death receptor and adaptor proteins. For the intrinsic pathway it has one predicted initiator and one effector caspase, compared to 2 and 11 in Hydrarespectively. Malacosporea retain one predicted effector caspase while Myxosporea universally lack all main actors of apoptosis including caspases, Bcl-2 family proteins, calpains, inhibitors of apoptosis proteins (IAPs), APAF-1 and p53 homologs. As an exception some Myxosporea species retained potentially functional cytochrome C, whose gene is however absent in Myxobolus squamalis, Henneguya salminicola and is a pseudogene with multiple inner stop-codons in Kudoa iwatai, Sphaeromyxa zaharoni, and Enteromyxum leei.

Conclusions

We show that the presence of core apoptotic proteins (such as the number of different caspases) gradually diminishes from free-living Cnidaria to Polypodium to Malacosporea to Myxosporea. This observation does not favor the hypothesis of catastrophic simplification of Myxosporea at the genetic level, but rather supports a stepwise adaptation to parasitism that likely started from early parasitic ancestors that gave rise to Myxozoa.

Apoptosis

Cnidaria

Myxozoa

simplification

parasitism

comparative genomics

evolution

Multicellular organisms can get rid of old, damaged, unuseful or precancerous cells by apoptosis, a form of programmed cell death or cell suicide. Key apoptotic proteins include proteolytic enzymes called caspases, which trigger cell death by cleaving specific cellular proteins. Inactive precursors, or procaspases, are activated through cleavage by other caspases. Extracellular or intracellular death signals can initiate this proteolytic cascade under tight regulation of adaptor and regulatory proteins such as Fas-associated death domain protein (FADD) and Bcl-2 family proteins [1], [2]. Notably, apoptosis-like programmed cell death is evolutionary conserved: it can be found even in free-living and parasitic unicellular organisms [3]. Many elements of programmed cell death in yeast are similar to those of the mammalian apoptotic pathways [4] including the presence of proteins that are structurally similar to caspases [5]. Bcl-2 and caspase homologs are universal among Metazoa [11] with rare exceptions.

There are two principal apoptotic pathways: the extrinsic pathway and intrinsic pathway. In its simplified version, the extrinsic apoptotic pathway starts with external signals that activate Death domain harboring transmembrane receptors (TNFR/Fas)[6]. This signal is transduced via adaptor proteins such as TRADD/FADD with Death and Death-effector domains that transform initiator procaspases into active initiator caspases which induce the apoptosis caspase cascade.

The intrinsic apoptotic pathway involves DNA damage signaling via transcription factors such as p53 [7]. In response BH3-only Bcl-2 family proteins are expressed, leading to the inhibition of anti-apoptotic Bcl-2. This allows liberated Bak proteins to form channels within the mitochondrial membranes causing cytochrome C release. Apoptotic protease activating factor 1 (APAF-1) proteins bind cytochrome C with their WD40 repeats and then bind with each other via their CARD domains and form the apoptosome [8]. The apoptosome activates the initiator procaspases of the intrinsic apoptotic pathway causing the apoptotic caspase cascade [9]. Inhibitors of apoptosis proteins (IAPs) inhibit caspases [10]. More detailed descriptions of apoptosis are reviewed elsewhere [6].

Variation exists between the apoptotic pathways of invertebrates. For example, the Pacific oyster has mammalian-like apoptotic pathways [53], whereas, Caenorhabditis elegans does not use cytochrome C for the apoptotic caspase cascade activation [51]. Apoptosis has been described and well-studied in model free-living Cnidaria, such as Hydra [14] which have numerous caspases, Bcl-2 protein family members, an APAF-1 homolog, components of a putative death receptor pathway and inhibitors of apoptotic proteases. Like other animals, Hydra are capable of developing tumors [15] and the transcriptomes of these tumors reveal misregulation of genes related to mammalian apoptosis genes.

Surprisingly, the Myxosporea (see Fig. 1 for simplified tree) species Thelohanellus kitauei, Kudoa iwatai, Myxobolus pronini, Sphaeromyxa zaharoni, Enteromyxum leei were reported to completely lack a number of Pfam-domains belonging to key apoptotic proteins including caspases, Bcl-2, APAF-1 and p53 homologs [11]. This finding was interpreted in favor of the SCANDALs (Speciated by CANcer Development AnimaLS) hypothesis that suggested that Myxosporea evolved from transmissible cancers that could have occurred in the parasitic ancestors of Polypodium, and have undergone catastrophic simplification. The limitations of that study were the consideration of Pfam-domains only, but not complete genes and that genetic data for the sister subclass Malacosporea was unavailable.

Unlike Myxosporea, Malacosporea are unlikely candidates for the SCANDAL hypothesis because of their more sophisticated morphology and development stages similar to gastrulation [16]. Malacosporean species, Tetracapsuloides bryosalmonae, can form ~ 350 µm spherical spore sacs [17] while another malacosporean, Buddenbrockia plumatellae, forms 1 mm long worm-like sacs with muscle cells that are capable of active movement in their bryozoan hosts [18]. In the fish host these species form a pseudoplasmodial structure in renal tubules [17], [19]. Tetracapsuloides bryosalmonae cause a proliferative kidney disease in salmonids which affects important wild and aquacultural populations [20]. Two other described, but less studied species of Malacosporea demonstrate similar morphology and structure [16], [21].

Recently transcriptomic data for members of Malacosporea Buddenbrockia plumatellae and Tetracapsuloides bryosalmonae became available [22], [23], [24], [25]. Also genomic data became available for additional Myxosporea species: Henneguya salminicola, Myxobolus squamalis [26] and Myxobolus honghuensis [27].

Notably, the Henneguya salminicola has been shown to completely lack a mitochondrial genome which is especially interesting in the context of the apparent loss of apoptosis in Myxosporea species [26] We decided to complete this data with our own sequences of Polypodium hydriforme and Myxobolus pronini to comprehensively study the loss of apoptosis-related genes in the parasitic Cnidaria.

We show that the SCANDAL hypothesis for Myxosporea origin is weakened by the observation that Malacosporea also lost many genes involved in apoptosis regulation, although to a lesser extent. Our analysis reveals and describes in detail the gradual process of apoptosis-related gene loss in Myxozoa that appears to have started in the common ancestor of both Myxosporea and Malacosporea subclasses.

Presence of genes coding key apoptotic proteins in parasitic Cnidaria

Comparison of cnidarian species reveals that the transition to parasitism is associated with gradual loss of key apoptotic factors in the parasitic classes Polypodiozoa and Malacosporea, up to a complete reduction of almost all pathway components in Myxosporea (Fig. 2 and Table 1).

Free-living Cnidaria (Hydra vulgaris, Nematostella vectensis, Acropora digitifera, Clytia hemisphaerica, Dendronephthya gigantea, Aurelia aurita) share all key components of both the intrinsic and extrinsic apoptotic pathways (Table 1). The parasitic Cnidaria Polypodium hydriforme lost the main components of the extrinsic apoptotic pathway: it completely lacks the Death domain required for the extrinsic receptor and adaptor proteins as well as the Death-effector domain required for the extrinsic initiator caspase. However, it maintains most of the intrinsic apoptotic pathway components with the exception of BH3-only Bcl-2 family members. Also, we were able to detect only one predicted initiator (CARD domain containing) and one effector caspase of the intrinsic pathway compared to two and eleven found in Hydra. Malacosporea have undergone further loss of apoptosis-related proteins including p53, all Bcl-2 family members, APAF-1, CARD domain and DEATH domain containing proteins (see Supplementary tables, “apoptosis-related” sheet, for full list of domains). One caspase without a CARD domain was found in Buddenbrockia. Myxosporea universally lack all main actors of apoptosis except cytochrome C, whose gene is however absent in Myxobolus squamalis and Henneguya salminicola, and is a pseudogene with multiple inner stop-codons in Kudoa iwatai, Sphaeromyxa zaharoni, and Enteromyxum leei. Henneguya salminicola apparently entirely lacks a mitochondrial genome [26]. Detailed findings and a comparison with other animals are presented in Table 1.

Table 1

*Presence of main actors of apoptosis in Cnidaria and other animals. Data for H. vulgaris, C. elegans and H. sapiens was taken from* [14], [47], [48], [49], [50], [51], [6]. The table contains information of the species from which a protein was found (Mp = Myxobolus pronini, El = Enteromyxum leei, Tk = Thelohanellus kitauei, Sz = Sphaeromyxa zaharoni, Ki = Kudoa iwatai, Mh = Myxobolus honghuensis, Budd = Buddenbrockia plumatellae, Tetr = Tetracapsuloides bryosalmonae, Pol = Polypodium hydriforme). * - these genes contain stop-codons. Note that C. elegans have an apoptotic pathway that involve homologs of mammalian BH3-only, BCL-2-like [52], *Apaf-1 and caspases, but that cascade does not involve cytochrome C release* [51]
	Myxosporea	Malacosporea	Polypodium hydriforme	free-living Cnidaria (Hydra vulgaris)	Caenorhabditis elegans	Homo sapiens
Caspases	-	BuddCasp	PolCasp, PolCARDCasp	HyCaspA, B, C, D, E, F, G, H, I, L, M HyCARD 1, 2 HyDEDCasp HyDDCasp	Ced-3, Csp-1, Csp-2	Caspase-1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 12, 13, 14
Bcl-2 multidomain	-	-	PolBCL-2-1, 2, PolBOK, PolBAK-1, 2	HyBcl-2-like 1, 2, 3, 4, 5, 6, 7, HyBak-like 1, 2	Ced-9	Bcl-2, Bcl-xL, Bcl-W, Mcl-1, Bcl-B, Bcl-2A1, Bax, Bak, Bok
BH3-only	-	-	-	HyBH3-only 1, 2, 3, 4	Egl-1, Ced-13	Bid, Bmf, NOXA, PUMA, Bad, Bim, Bik, Hrk
Death receptor	-	-	-	HyTNFR-like receptor	-	TNF-R and others
Adaptor protein	-	-	-	HyFADD	-	FADD, TRADD
IAP	-	-	PolIAP	HyIAP	-	XIAP, NAIP, c-IAP1, 2
APAF-1	-	-	PolAPAF-1	HyAPAF-1	Ced-4	APAF-1
p53	-	-	PolP53	p53	Cep-1	p53
Cytochrome C	MpCytC, TkCytC, MhCytC, ElCytC, SzCytC, KiCytC* Absent in: H. salminicola M. squamalis	BuddCytC, TetrCytC	PolCytC	Cytochrome C	CPS-6, WAH-1	Cytochrome C
Calpain	-	-	PolCalp-5, 7, PolClassCalp	Calpain-5, 7, 9, and others	CLP-1, TRA-3	Calpain-5, 7, 9, and others

This pattern of apoptotic-protein loss among parasitic Cnidaria is consistent with gradual loss-of-function in contrast to previous findings suggesting catastrophic loss of apoptotic and cancer-related proteins in Myxosporea [11]. While this study focuses only on Cnidarian apoptosis, we observe a gradual decline in the number of cancer-related PFAM domains in order from free-living Cnidaria to Polypodium to Malacosporea to Myxosporea (see Supplementary tables, “cancer-related” sheet).

Number of caspases gradually diminishes in parasitic Cnidaria

Caspases are proteases that play a crucial role in both intrinsic and extrinsic apoptotic pathways. Our results suggest the complete absence of caspases not only in Myxosporea, but also in one of two analyzed members of Malacosporea (Tetracapsuloides bryosalmonae). The other Malacosporea used in our analysis (Buddenbrockia plumatellae) has only one caspase (Fig. 3) in contrast to the free-living hydra that have fifteen [14]. Based on domain composition the single Buddenbrockia caspase is an effector caspase and not an initiator (CARD-domain containing) caspase, thus insufficient to trigger either of the apoptotic pathways. In contrast Polypodium has both an initiator and an effector caspase, allowing for the classical intrinsic apoptotic pathway to work. The loss of extrinsic apoptotic pathways in Malacosporea, Myxosporea and Polypodium can be predicted from the lack of DEATH receptors, adaptor proteins and Death-effector domain containing initiator caspases.

In contrast to Polypodium, Myxozoa likely lack Bcl-2 family members

The Bcl-2 family plays an important role in regulating the intrinsic apoptotic pathway. Some Bcl-2 family members are proapoptotic (Bak and Bok proteins in humans) and others are anti-apoptotic (human Bcl-2). Our analysis suggests that Myxozoa lack any Bcl-2 protein members, consistent with the predicted lack of intrinsic apoptotic pathways. In Polypodium we predicted five Bcl-2 protein family members and attempted to infer their pro- or anti-apoptotic roles based on similarity with known human and Hydra Bcl-2 family protein members (Fig. 4, see Supplementary materials, Fig. S1 for phylogenetic tree). One Polypodium Bcl-2 protein was predicted as a pro-apoptotic Bok homologue by InterProScan. The separation between Bcl-2 and Bak proteins is based on phylogenetic analysis (see Supplementary materials, Fig. S1). Overall, the results favor the existence of an intrinsic apoptotic pathway in Polypodium and its lack in Myxozoa.

Cytochrome C and mitochondrial complex III are likely to be lost in some myxosporeans

Predicted genes for cytochrome C were found in Polypodium, Malacosporea (both Tetracapsuloides and Buddenbrockia), and Myxosporea. However, in Myxosporea the situation with cytochrome C is rather complex. In Myxobolus pronini, M. honghuensis, and Thelohanellus kitauei the predicted cytochrome C proteins share high similarity with each other. Although, cytochrome C could not be found in a published M. cerebralis transcriptome (GBKL00000000.1). Also, we were unable to identify any potential genes of cytochrome C in the available genomes and transcriptomes of Henneguya salminicola and Myxobolus squamalis. Note that Henneguya salminicola was previously shown to lack a mitochondrial genome so this is a special case [26].

While we were able to find cytochrome C homologs in Sphaeromyxa zaharoni, Enteromyxum leei, and Kudoa iwatai, we found numerous stop codons and insertions within their predicted structural Cytochrome C domains. Enteromyxum leei and Kudoa iwatai have stop codons only within the above-mentioned insertions, whereas Sphaeromyxa zaharoni also has stop codons outside of them, within a highly conserved 69 aa region (Fig. 5).

Annotated mitochondrial genomes are available for myxosporeans Enteromyxum leei [54] and Kudoa iwatai [55]. Despite the presence of stop-codons in their cytochrome C genes, these organisms have been shown to have genes of core subunits of complex IV and cytochrome b (related to complex III) in their mitochondrial genomes. They possess the Pfam domains of respiratory subunits of complex III: PF00355 (Rieske) and PF02167 (Cytochrome C1) (Table 2). The latter statement is also true for Sphaeromyxa zaharoni, the third myxosporean with stop-codons in the cytochrome C gene. This suggests that the truncated cytochromes C of myxosporeans, resulting from these nonsense mutations, might still maintain their function in the respiratory chain.

Table 2 contains information on complex III respiratory subunit domains identified in studied Cnidaria. Among the domains of complex III respiratory subunit (encoded by nuclear genes) only the transmembrane domain (PF02921) was missing in all studied Myxosporea. In contrast, Henneguya salminicola, which reportedly lacks mitochondrial genome, is also missing all mentioned complex III domains. In Myxobolus squamalis and Thelohanellus kitauei only PF02167 (Cytochrome C1 family) complex III domain was found, respectively. Myxobolus honghuensis lacks the PF00355 (Rieske protein) domain. And Myxobolus pronini has the same complex III domain set as S. zaharoni, K. iwatai and E. leei. Malacosporea and Polypodium have all above-mentioned domains, including the PF02921 transmembrane region that the studied Myxosporeans lack.

Table 2.

Number of Pfam-domains constituting main functional subunits of complex III of mitochondria presented in genomic data of considered Myxosporea and Malacosporea species, Polypodium and Hydra. (UCR_TM – ubiquinol cytochrome C reductase transmembrane domain)

Not only Myxosporea, but also Malacosporea lack calpains, APAF-1, IAPs, and p53

Calpains are a family of proteases, some members of which are involved in caspase activation in response to calcium release from the endoplasmic reticulum [56]. The caspase activation is performed via the Peptidase_C2 and Calpain III peptidase domains, while the EF-hand domain is capable of binding calcium. We were unable to find any calpain family members in Myxozoa, however, we found 4 putative calpains in the parasitic cnidarians Polypodium (Fig. 6). The domain structure of two of those calpains (presence of both peptidase domains and EF-hand domains) suggests their role in apoptosis. The EF-hand domain is important for calcium binding, whereas the Peptidase_C2 and Calpain III domains are homologous to peptidase domains capable of caspase activation.

APAF-1 or apoptotic protease activating factor 1 is essential for forming the apoptosome when bound to cytochrome C, released from the mitochondria. Among the studied parasitic cnidaria only Polypodium has an APAF-1 homologue that we could identify. This homologue has the WD40 domain necessary for cytochrome C binding and the CARD domain for apoptosome formation (Fig. 7). This further suggests that Polypodium has functional apoptosis, while Myxosporea and Malacosporea do not.

IAPs are inhibitors of apoptosis. Among studied parasitic cnidaria only Polypodium has an IAP. However, this predicted protein lacks the XIAP/BIRC8_UBA domain found in its human homolog and has a shorter BIR_rpt domain which binds Zn + + and then binds the active site of caspases inhibiting their catalytic activity (Fig. 7).

p53 is a transcription factor that is activated in response to DNA damage and increases the expression of a number of pro-apoptotic genes (including APAF-1, Bak, Bax) [7]. The only parasitic cnidaria in our analysis with p53 is Polypodium. However, the Polypodium p53 homolog is truncated and lacks some of the domains present in the human p53 gene (Fig. 7). It lacks the p53_TAD2 and p53_tetr domains which are involved in p53 transactivation and tetramerization respectively. The p53 DNA binding domain is present in Polypodium.

One important limitation of the comparative genomics approach used in our study is data incompleteness: when a gene is missing from the data, it might be due to insufficient read coverage. However, when this is the case we expect different genes to be lost in independently sequenced related animals. For each species of Myxosporea we observe roughly the same set of lost apoptosis-related genes. A similar picture was observed for cancer-related gene loss of Myxosporean species as reported previously [11]. Nevertheless, the loss of functional cytochrome C genes in some myxosporeans might be due to incomplete data given that we were not able to connect this feature with the loss of complex III respiratory subunit domains. Note that cytochrome C is encoded in the nuclear genome and the observed presence of stop codons in its predicted pseudogenes is unlikely due to alternative genetic code use.

The near-completeness of malacosporean transcriptomes is supported by the observation that almost all Pfam domains found in myxosporean genomes are also present in malacosporean transcriptomes (see Supplementary tables, “check_completeness” sheet).

We evaluated the presence of core apoptotic proteins in free-living and parasitic cnidaria. We observe a gradual loss of apoptosis-related proteins, starting with Polypodium who lost the major components of the extrinsic apoptotic pathway, followed by Myxozoa who lost the intrinsic apoptotic pathway as well. Compared to Myxosporea, Malacosporea retained one partial (large subunit) caspase-domain containing protein (in Buddenbrockia). While both studied Malacosporea retained cytochrome C, the latter is missing in myxosporea Myxobolus squamalis, Henneguya salminicola and its gene contains multiple stop-codons in Kudoa iwatai, M. honghuensis, and Sphaeromyxa zaharoni. However, the presence of other respiratory chain proteins suggests that the genes might be producing truncated, yet possibly functional cytochrome C proteins. Thus, Cnidaria present us with a unique opportunity to observe different stages of gradual loss of apoptosis. The observation that Malacosporea lost many genes involved in apoptosis regulation, weakens the previously proposed the SCANDAL hypothesis for Myxosporea catastrophic simplification [11].

Cnidaria genomes and transcriptomes

We used genomic data for Myxobolus squamalis GCA_010108815, Henneguya salminicola GCA_009887335.1, Enteromyxum leei GCA_001455295.2, Sphaeromyxa zaharoni GCA_001455285.1, Kudoa iwatai GCA_001407235.2, Thelohanellus kitauei GCA_000827895.1, Myxobolus honghuensis GWHBFXL00000000, Hydra vulgaris GCA_000004095.1, Clytia hemisphaerica GCA_902728285.1, Nematostella vectensis GCA_000209225, Dendronephthya gigantea GCA_004324835.1, Acropora digitifera GCF_000222465.1, and Aurelia aurita GCA_004194415.1. We additionally sequenced and assembled the genomes of Polypodium hydriforme and Myxobolus pronini (see Materials and methods, Sequence assembly). Transcriptomic data was available for Tetracapsuloides bryosalmonae, Buddenbrockia plumatellae, and Polypodium hydriforme. For Tetracapsuloides bryosalmonae we used a set of transcriptomic data, namely ready-made T. bryosalmonae assemblies from Kumar et al., 2021 (BioProject: PRJNA680464) and Faber et al., 2021 (BioProject: PRJEB19471) and T. bryosalmonae paired-end reads from Faber et al., 2021, Kumar et al., 2021 and Ahmad et al., 2021 (BioProject: PRJNA668017) which we assembled using SPAdes [28] with k-mer lengths of 33, 49 and 93.

One of the approaches to filter contaminations of transcriptomic data of a parasitic animal is to compare its sequences from two hosts (as was done in [24]) or compare sequences from an infected and non-infected host of the same species. We created three filtered datasets using these approaches (Table 3). In two cases we used the intersection of RNA reads of T. bryosalmonae from a fish host and T. bryosalmonae assembly from a bryozoan host. In one case we used RNA reads from infected by T. bryosalmonae kidney of Salmo trutta and filtered out reads mapping to S. trutta reference genome. Reads were mapped on assemblies using Bwa mem algorithm [29]. After filtering, predicted T. bryosalmonae reads were assembled using SPAdes. SPAdes tool was used with k-mer lengths of 33, 49 and 93. All .sam and .bam files were processed with samtools package [30].

Table 3

Data sets which were used to make intersection transcriptomes. Reads were mapped on the ready-made assembly using Bwa mem algorithm and then mapped reads were assembled using SPAdes.
Reads	Assembly
T. bryosalmonae, fish-derived [24]	T. bryosalmonae, bryozoan-derived [24]
T. bryosalmonae, fish-derived [24]	T. bryosalmonae, bryozoan-derived [25]
Kidney of S. trutta infected by T. bryosalmonae [23]	Genome of S. trutta (Genbank AC: GCA_901001165.2)

Exclusion of host contaminations in Myxozoan genomes and transcriptomes

In the eight available genomes of myxosporean species (Myxobolus squamalis, Henneguya salminicola GCA_009887335.1, Enteromyxum leei GCA_001455295.2, Sphaeromyxa zaharoni GCA_001455285.1, Kudoa iwatai GCA_001407235.2, Thelohanellus kitauei GCA_000827895.1, Myxobolus pronini (see Materials and methods, Sequence assembly), Myxobolus honghuensis GWHBFXL00000000) we searched for possible host contaminations using an original Python script (see supplementary script genome_filter.py). Each contig from the genomes was used as a query for blastn-searches [31] against a BLAST database built from the host genome. In the case of Myxosporea we used the publicly available genome of either the exact fish host or its closest available relative from Genbank (see Supplementary materials, List S1). If the best hit demonstrated > 85% identity and e-value < 1-e75, the query contig was excluded from the final assembly.

Another original python script (see supplementary script comparison.py) was used to remove host-derived contaminating sequences from the transcriptomes of malacosporeans B. plumatellae and T. bryosalmonae. First, we created two BLAST databases. The first consisted of several cnidarian transcriptomes, namely of Aurelia aurita, Polypodium hydriforme, Hydra vulgaris, Pocillopora acuta, Myxobolus squamalis, Myxidium lieberkuehni, and malacosporean transcriptomes that were obtained from hosts other than those which are being filtered. The second database consisted of a wide range of non-cnidarian organisms that may be regarded as additional probable contaminants. It was shown that possible contaminants include algae and cyanobacteria [25], so we included Tetraselmis striata and Nodularia spumigena as representatives of these taxa. We also included the genome of bryozoan Cristatella mucedo, the transciptome of bryozoan Fredericella sultana, a transcriptome of another lophophorate – Lingula anatina, and also transcriptomes of Drosophila melanogaster, Ruditapes philippinarium, Trichinella pseudospiralis, Salmo trutta, Danio rerio, Cricetulus griseus. We found contaminations likely derived from these taxa when we manually checked BLAST search results of Malacosporean contigs. Such contaminations may sometimes occur when the same lab equipment is used for various sequencing projects or because bryozoans are filtrators and are susceptible to environmental contaminations.

Databases were created using translated ORFs obtained from the assemblies with the TransDecoder tool [32], using a minimal ORF length of 70 amino acids. Each contig from malacosporean assemblies was searched against each of the two BLAST databases using blastp from the BLAST+ [31] tool package. Contigs were divided into three groups (“cnidaria”, “contamination” and “not clear”) based on e-value and bit score differences between the best hits against cnidarian and contaminant blast databases. If the e-value of the best cnidarian hit was at least 100x less and its bit score at least 1.5x more than that of the best hit from the contaminant dataset, this contig was considered as “cnidaria”. If the opposite was true, the contig was flagged as “contamination”. Otherwise, it was placed in the “not clear” group and subjected to further GC-content analysis (see below).

It was reported that Myxozoa have relatively lower GC-content compared to other Metazoa [33]. To visualize GC-content of each group of contigs we built a scatter plot with the length of contig and the GC-content on the Y- and X-axis respectively for “cnidaria”, “contamination” and “not clear” groups of contigs. In the “cnidaria” group of contigs we obtained a set bounded by a bell curve with the average GC-content at 27–30% depending on the particular organism. The “contamination” group of contigs were placed as a flat bell-bounded set with average GC-content of 55%. In the “not clear” group we observed a distribution of GC-content indicating the presence of both “cnidaria” and “contamination” contigs (see Supplementary materials, Fig. S2) We used the following GC-content cutoffs: 39.03% for B. plumatellae, 40.80% for T. bryoslamonae assembly from fish kidney, 36.92% for T. bryosalmonae from bryozoan host to include the “not clear” contigs into the final “cnidaria” set of contigs.

Manual checks of potential contamination sources were conducted using BLAST searches and phylogenetic tree construction using IQ-TREE [34], [35] with further visualization in iTOL [36].

Identification of PFAM domains

We used the hmmscan tool from HMMER [47] version 3.3.2 and the PFAM A database (version 34.0) [48] to identify PFAM domains in TransDecoder-predicted ORFs within the filtered contigs. This was done for all filtered Myxozoan genomes and transcriptomes, the Polypodium hydriforme genome and several genomes of free-living Cnidaria genomes from the NCBI Genbank, namely Acropora digitifera GCF_000222465.1, Aurelia aurita GCA_004194415.1, Clytia hemisphaerica GCA_902728285.1, Dendronephthya gigantea GCA_004324835.1, Hydra vulgaris GCA_000004095.1, Nematostella vectensis GCA_000209225.

Search for apoptosis-related domains and proteins

To elucidate the presence of core apoptotic proteins as functional multidomain units we conducted additional BLAST searches. Similar to [14] we adopted the list of proteins that are most important for apoptosis, namely caspase homologs, Bcl-2-family proteins, Apoptotic protease activating factor 1 (APAF-1), tumor necrosis factor receptor (TNF-R), inhibitor of apoptosis proteins (IAPs), Fas-associated protein with death domain (FADD), calpains, p53, and cytochrome C. For each protein we collected several reference cnidarian protein sequences from UniProt and NCBI Genbank, Hydra sequences from [14] were also included (see Supplementary materials, List S2, for list of accessions ).

We performed blastp searches for each of the reference proteins against malacosporean (filtered and unfiltered B. plumatellae, T. bryosalmonae from invertebrate host, T. bryosalmonae from fish host with further manual analysis), Polypodium hydriforme genome and transcriptome, and also against filtered myxosporean genomes. Domain structures of identified hits were analyzed using the InterProScan tool [49] with InterPro database [50]. Multiple alignments were built with the Muscle tool [51] and visualized in the Jalview application [52].

Sample collection, DNA isolation

Round cysts of Myxobolus pronini were isolated from the abdominal cavity Carassius auratus gibelio host. Total DNA was isolated from tissue samples by Diatom DNA Prep (IsoGene). One paired-end and two mate-pair libraries were prepared for Myxobolus pronini using the Nextera library preparation protocol (Illumina). The mate-pair libraries were prepared with the estimated mean insert lengths of 3500 bp and 7000 bp. The libraries were sequenced using an Illumina HiSeq 2000 instrument, generating 163M 100 bp read pairs for the paired-end library, 31M read pairs for the 3500 bp mate-pair library and 38M read pairs for the 7000 bp mate-pair library.

Stolons of Polypodium hydriforme were isolated from infected oocytes of Acipenser gueldenstaedtii host. Total DNA was isolated from tissue samples by Diatom DNA Prep (IsoGene). One paired-end library was prepared for Polypodium hydriforme using the TruSeq library preparation protocol (Illumina). The Polypodium library was sequenced using an Illumina HiSeq 2500 instrument, generating 260M read pairs with the read length of 250 bp.

Sequence assembly

The sequencing reads for Myxobolus pronini and Polypodium hydriforme were trimmed using Trimmomatic [53] to remove the adapters. Prior to assembly, the Myxobolus pronini reads were processed using the error correction utility of the ALLPATHS-LG program [54]. The assembly for Myxobolus pronini data was performed using Velvet [55] with a k-mer size of 55. Scaffolds obtained with Velvet were then processed using the Redundans pipeline [56] with an identity threshold of 0.85 to obtain the final assembly draft. The assembly of Polypodium data was performed using SPAdes [28] with three k-mer sizes: 77, 99, 127. The assemblies were evaluated using QUAST [57].

Ethics approval and consent to participate

Not applicable to this study

Consent for publication

Not applicable to this study

Availability of data and materials

The datasets supporting the conclusions of this article are included within the article (and its additional files).

Competing interests

The authors declare no competing interest.

Funding

This work was supported by Russian Science Foundation grant [19-74-20147] аnd project of the Russian Academy of Science (No. AAAA-A17-117011810039-4).

Authors' contributions

AMN, AYP and YVP conceived the study. MDB, VVA, YVP performed sample collection. KVM performed sequencing and sequence assembly. AMN and AYP prepared the draft manuscript. AMN, AYP, YVP and KVM analyzed the data. YVP and VVA supervised the project. All authors read, edited and approved the final manuscript.

Alberts B, Johnson A, Lewis J. Molecular biology of the cell. 4th ed. London, England: Garland Science; 2002.
Kim R. Unknotting the roles of Bcl-2 and Bcl-xL in cell death. Biochem Biophys Res Commun. Elsevier BV; 2005;333:336–43.
Kaczanowski S, Sajid M, Reece SE. Evolution of apoptosis-like programmed cell death in unicellular protozoan parasites. Parasit Vectors. Springer Science and Business Media LLC; 2011;4:44.
Sharon A, Finkelstein A, Shlezinger N, Hatam I. Fungal apoptosis: function, genes and gene function. FEMS Microbiol Rev. Oxford University Press (OUP); 2009;33:833–54.
Madeo F, Herker E, Maldener C, Wissing S, Lächelt S, Herlan M, et al. A caspase-related protease regulates apoptosis in yeast. Mol Cell. Elsevier BV; 2002;9:911–7.
Elmore S. Apoptosis: a review of programmed cell death. Toxicol Pathol. 2007;35:495–516.
Aubrey BJ, Kelly GL, Janic A, Herold MJ, Strasser A. How does p53 induce apoptosis and how does this relate to p53-mediated tumour suppression? Cell Death Differ. Springer Science and Business Media LLC; 2018;25:104–13.
Kim H-E, Du F, Fang M, Wang X. Formation of apoptosome is initiated by cytochrome c-induced dATP hydrolysis and subsequent nucleotide exchange on Apaf-1. Proc Natl Acad Sci U S A. Proceedings of the National Academy of Sciences; 2005;102:17545–50.
Logue SE, Martin SJ. Caspase activation cascades in apoptosis. Biochem Soc Trans. Portland Press Ltd.; 2008;36:1–9.
Silke J, Meier P. Inhibitor of apoptosis (IAP) proteins-modulators of cell death and inflammation. Cold Spring Harb Perspect Biol. Cold Spring Harbor Laboratory; 2013;5:a008730–a008730.
Panchin AY, Aleoshin VV, Panchin YV. From tumors to species: a SCANDAL hypothesis. Biol Direct. Springer Science and Business Media LLC; 2019;14:3.
Shakeri R, Kheirollahi A, Davoodi J. Apaf-1: Regulation and function in cell death. Biochimie. Elsevier BV; 2017;135:111–25.
Ow Y-LP, Green DR, Hao Z, Mak TW. Cytochrome c: functions beyond respiration. Nat Rev Mol Cell Biol. Springer Science and Business Media LLC; 2008;9:532–42.
Lasi M, David CN, Böttger A. Apoptosis in pre-Bilaterians: Hydra as a model. Apoptosis. Springer Science and Business Media LLC; 2010;15:269–78.
Domazet-Lošo T, Klimovich A, Anokhin B, Anton-Erxleben F, Hamm MJ, Lange C, et al. Naturally occurring tumours in the basal metazoan Hydra. Nat Commun. Springer Science and Business Media LLC; 2014;5:4222.
Canning EU, Curry A, Hill SLL, Okamura B. Ultrastructure of Buddenbrockia allmani n. sp. (Myxozoa, Malacosporea), a parasite of Lophopus crystallinus (Bryozoa, Phylactolaemata). J Eukaryot Microbiol. Wiley; 2007;54:247–62.
Sudhagar A, Kumar G, El-Matbouli M. The malacosporean myxozoan parasite Tetracapsuloides bryosalmonae: A threat to wild salmonids. Pathogens. MDPI AG; 2019;9:16.
Gruhl A, Okamura B. Development and myogenesis of the vermiform Buddenbrockia (Myxozoa) and implications for cnidarian body plan evolution. Evodevo. Springer Science and Business Media LLC; 2012;3:10.
Naldoni J, Adriano EA, Hartigan A, Sayer C, Okamura B. Malacosporean myxozoans exploit a diversity of fish hosts. Parasitology. Cambridge University Press (CUP); 2019;146:968–78.
Okamura B, Hartikainen H, Schmidt-Posthaus H, Wahli T. Life cycle complexity, environmental change and the emerging status of salmonid proliferative kidney disease. Freshw Biol. Wiley; 2011;56:735–53.
Patra S, Hartigan A, Morris DJ, Kodádková A, Holzer AS. Description and experimental transmission of Tetracapsuloides vermiformis n. sp. (Cnidaria: Myxozoa) and guidelines for describing malacosporean species including reinstatement of Buddenbrockia bryozoides n. comb. (syn. Tetracapsula bryozoides). Parasitology. Cambridge University Press (CUP); 2017;144:497–511.
Hartigan A, Jaimes-Becerra A, Okamura B, Doonan LB, Ward M, Marques AC, et al. Recruitment of toxin-like proteins with ancestral venom function supports endoparasitic lifestyles of Myxozoa. PeerJ. 2021;9:e11208.
Ahmad F, Debes PV, Pukk L, Kahar S, Hartikainen H, Gross R, et al. Know your enemy - transcriptome of myxozoan Tetracapsuloides bryosalmonae reveals potential drug targets against proliferative kidney disease in salmonids. Parasitology. Cambridge University Press (CUP); 2021;148:726–39.
Faber M, Shaw S, Yoon S, de Paiva Alves E, Wang B, Qi Z, et al. Comparative transcriptomics and host-specific parasite gene expression profiles inform on drivers of proliferative kidney disease. Sci Rep. Springer Science and Business Media LLC; 2021;11:2149.
Kumar G, Ertl R, Nilsen F, Bartholomew JL, El-Matbouli M. Data of de novo transcriptome assembly of the myxozoan parasite Tetracapsuloides bryosalmonae. Data Brief. Elsevier BV; 2021;35:106831.
Yahalomi D, Atkinson SD, Neuhof M, Chang ES, Philippe H, Cartwright P, et al. A cnidarian parasite of salmon (Myxozoa: Henneguya) lacks a mitochondrial genome. Proc Natl Acad Sci U S A. Proceedings of the National Academy of Sciences; 2020;117:5358–63.
Guo Q, Atkinson SD, Xiao B, Zhai Y, Bartholomew JL, Gu Z. A myxozoan genome reveals mosaic evolution in a parasitic cnidarian. BMC Biol. Springer Science and Business Media LLC; 2022;20:51.
Igaki T, Miura M. Role of Bcl-2 family members in invertebrates. Biochim Biophys Acta. Elsevier BV; 2004;1644:73–81.
Tan FJ, Fire AZ, Hill RB. Regulation of apoptosis by C. elegans CED-9 in the absence of the C-terminal transmembrane domain. Cell Death Differ. Springer Science and Business Media LLC; 2007;14:1925–35.
Denning DP, Hatch V, Horvitz HR. Both the caspase CSP-1 and a caspase-independent pathway promote programmed cell death in parallel to the canonical pathway for apoptosis in Caenorhabditis elegans. PLoS Genet. Public Library of Science (PLoS); 2013;9:e1003341.
Chen Y-Z, Mapes J, Lee E-S, Skeen-Gaar RR, Xue D. Caspase-mediated activation of Caenorhabditis elegans CED-8 promotes apoptosis and phosphatidylserine externalization. Nat Commun. Springer Science and Business Media LLC; 2013;4:2726.
Arvanitis M, Li D-D, Lee K, Mylonakis E. Apoptosis in C. elegans: lessons for cancer and immunity. Front Cell Infect Microbiol. Frontiers Media SA; 2013;3:67.
Strasser A, Vaux DL. Viewing BCL2 and cell death control from an evolutionary perspective. Cell Death Differ. Springer Science and Business Media LLC; 2018;25:13–20.
Li Y, Zhang L, Qu T, Tang X, Li L, Zhang G. Conservation and divergence of mitochondrial apoptosis pathway in the Pacific oyster, Crassostrea gigas. Cell Death Dis. Springer Science and Business Media LLC; 2017;8:e2915–e2915.
Yahalomi D, Haddas-Sasson M, Rubinstein ND, Feldstein T, Diamant A, Huchon D. The multipartite mitochondrial genome of Enteromyxum leei (Myxozoa): Eight fast-evolving megacircles. Mol Biol Evol. Oxford University Press (OUP); 2017;34:1551–6.
Takeuchi F, Sekizuka T, Ogasawara Y, Yokoyama H, Kamikawa R, Inagaki Y, et al. The mitochondrial genomes of a myxozoan genus Kudoa are extremely divergent in Metazoa. PLoS One. Public Library of Science (PLoS); 2015;10:e0132030.
Momeni HR. Role of calpain in apoptosis. Cell J. 2011;13:65–72.
Prjibelski A, Antipov D, Meleshko D, Lapidus A, Korobeynikov A. Using SPAdes DE Novo Assembler. Curr Protoc Bioinformatics. Wiley; 2020;70:e102.
Li H. Aligning sequence reads, clone sequences and assembly contigs with BWA-MEM. arXiv; 2013; Available from: http://dx.doi.org/10.48550/ARXIV.1303.3997
Danecek P, Bonfield JK, Liddle J, Marshall J, Ohan V, Pollard MO, et al. Twelve years of SAMtools and BCFtools. Gigascience [Internet]. Oxford University Press (OUP); 2021;10. Available from: http://dx.doi.org/10.1093/gigascience/giab008
Camacho C, Coulouris G, Avagyan V, Ma N, Papadopoulos J, Bealer K, et al. BLAST+: architecture and applications. BMC Bioinformatics. Springer Science and Business Media LLC; 2009;10:421.
Haas BJ, Papanicolaou A, Yassour M, Grabherr M, Blood PD, Bowden J, et al. De novo transcript sequence reconstruction from RNA-seq using the Trinity platform for reference generation and analysis. Nat Protoc. Springer Science and Business Media LLC; 2013;8:1494–512.
Alama-Bermejo G, Holzer AS. Advances and discoveries in myxozoan genomics. Trends Parasitol. Elsevier BV; 2021;37:552–68.
Kalyaanamoorthy S, Minh BQ, Wong TKF, von Haeseler A, Jermiin LS. ModelFinder: fast model selection for accurate phylogenetic estimates. Nat Methods. Springer Science and Business Media LLC; 2017;14:587–9.
Nguyen BT, Shuval K, Yaroch AL. Nguyen et al. Respond. Am. J. Public Health. American Public Health Association; 2015. p. e2.
Letunic I, Bork P. Interactive Tree Of Life (iTOL) v5: an online tool for phylogenetic tree display and annotation. Nucleic Acids Res. Oxford University Press (OUP); 2021;49:W293–6.
Eddy SR. Accelerated profile HMM searches. PLoS Comput Biol. Public Library of Science (PLoS); 2011;7:e1002195.
Mistry J, Chuguransky S, Williams L, Qureshi M, Salazar GA, Sonnhammer ELL, et al. Pfam: The protein families database in 2021. Nucleic Acids Res. Oxford University Press (OUP); 2021;49:D412–9.
Jones P, Binns D, Chang H-Y, Fraser M, Li W, McAnulla C, et al. InterProScan 5: genome-scale protein function classification. Bioinformatics. Oxford University Press (OUP); 2014;30:1236–40.
Blum M, Chang H-Y, Chuguransky S, Grego T, Kandasaamy S, Mitchell A, et al. The InterPro protein families and domains database: 20 years on. Nucleic Acids Res. Oxford University Press (OUP); 2021;49:D344–54.
Edgar RC. MUSCLE: multiple sequence alignment with high accuracy and high throughput. Nucleic Acids Res. Oxford University Press (OUP); 2004;32:1792–7.
Waterhouse AM, Procter JB, Martin DMA, Clamp M, Barton GJ. Jalview Version 2--a multiple sequence alignment editor and analysis workbench. Bioinformatics. Oxford University Press (OUP); 2009;25:1189–91.
Bolger AM, Lohse M, Usadel B. Trimmomatic: a flexible trimmer for Illumina sequence data. Bioinformatics. Oxford University Press (OUP); 2014;30:2114–20.
Gnerre S, Maccallum I, Przybylski D, Ribeiro FJ, Burton JN, Walker BJ, et al. High-quality draft assemblies of mammalian genomes from massively parallel sequence data. Proc Natl Acad Sci U S A. Proceedings of the National Academy of Sciences; 2011;108:1513–8.
Zerbino DR, Birney E. Velvet: algorithms for de novo short read assembly using de Bruijn graphs. Genome Res. Cold Spring Harbor Laboratory; 2008;18:821–9.
Pryszcz LP, Gabaldón T. Redundans: an assembly pipeline for highly heterozygous genomes. Nucleic Acids Res. 2016;44:e113.
Gurevich A, Saveliev V, Vyahhi N, Tesler G. QUAST: quality assessment tool for genome assemblies. Bioinformatics. Oxford University Press (OUP); 2013;29:1072–5.

No competing interests reported.

Supplementarymaterials.docx
Supplementary figures and lists are stored in the file “Supplementary_materials.docx”. Figure S1 shows a phylogenetic tree of Bcl-2 family proteins, figure S2 shows an example of transcriptome filtration based on GC-content (example for Buddenbrockia plumatellae). List S1 consists of accession numbers of fish species’ genomes used for filtration of fish-derived contaminations in myxosporeans’ genomes. List S2 consists of reference proteins used in comparative BLAST analysis to predict their homologs in parasitic Cnidaria.
Supplementarytables.xlsx
All supplementary tables are stored in the file “Supplementary_tables.xlsx”. “Cancer-related” sheet presents the number of cancer-related domains which were found in considered Cnidaria. List of domains was taken from article about SCANDALs [11]. “Apoptosis-related” sheet presents the number of apoptosis-related domains which were found in considered Cnidaria. List of domains is based on the search of “apoptosis” keyword in Pfam database. “Check_completeness” list suggests the completeness of transcriptomes of Malacosporea, based on their comparison with genomes of Myxosporea
Proteins.fasta
All predicted apoptosis-related proteins of parasitic Cnidaria are stored in the file “Proteins.fasta” in fasta-format.
genomefilter.py
Script “genome_filter.py” was utilized for filtration of contaminations in myxosporeans’ genomes,
comparison.py
script “comparison.py” was utilized for filtration of contaminations in malacosporeans’ transcriptomes.

Download PDF

Version 1

posted

You are reading this latest preprint version

Apoptotic gene loss in Cnidaria is associated with transition to parasitism

Status:

Version 1

Abstract

Figures

Background

Results

Presence of genes coding key apoptotic proteins in parasitic Cnidaria

Number of caspases gradually diminishes in parasitic Cnidaria

In contrast to Polypodium, Myxozoa likely lack Bcl-2 family members

Cytochrome C and mitochondrial complex III are likely to be lost in some myxosporeans

Not only Myxosporea, but also Malacosporea lack calpains, APAF-1, IAPs, and p53

Limitations

Conclusions

Materials And Methods

Cnidaria genomes and transcriptomes

Exclusion of host contaminations in Myxozoan genomes and transcriptomes

Identification of PFAM domains

Search for apoptosis-related domains and proteins

Sample collection, DNA isolation

Sequence assembly

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1