A genome-wide screening for RNAi pathway proteins in Acari

doi:10.21203/rs.3.rs-38229/v2

Download PDF

Research article

A genome-wide screening for RNAi pathway proteins in Acari

https://doi.org/10.21203/rs.3.rs-38229/v2

This work is licensed under a CC BY 4.0 License

Journal Publication

published 12 Nov, 2020

Read the published version in BMC Genomics →

You are reading this older preprint version

Read the latest preprint version →

Background RNA interference (RNAi) is a highly conserved, sequence-specific gene silencing mechanism present in Eukaryotes. Three RNAi pathways critical for organismal development and survival are known, namely micro-RNA (miRNA), Piwi-interacting RNA (piRNA) and short interfering RNA (siRNA) pathways. Little knowledge exist about the genes involved in these pathways in Acari. Moreover, variable successes has been obtained in gene knockdown via siRNA pathway in functional genomics and management of Acari species. We hypothesized that the clue may be in the variability in the composition and the efficacy of siRNAi machinery among Acari.

Results Both comparative genomic analyses and domain annotation suggest that all the analyzed species have orthologs of genes that mediate gene silencing via the three RNAi pathways though gene duplication and/or loss have occurred in the different species. We also identified orthologs of Caenorhabditis elegans RNA-dependent RNA polymerase (RdRP) gene in all Acari species though no secondary Argonaute homologs that operate with this gene in siRNA amplification mechanism were found. This finding suggests that the siRNA amplification mechanism present in Acari may be distinct from that described in C. elegans . Moreover, the genomes of these Acari species encode no ortholog of C. elegans systemic RNAi defective 1 (Sid-1) that mediate systemic RNAi, suggesting that the phenomena of systemic and parental RNAi that has been reported in some Acari species probably occur through a different mechanism. Orthologs of RNAi spreading defective-3 (Rsd-3) gene and scavenger receptors namely Eater and SR-CI that mediate endocytosis cellular update of dsRNA in C. elegans and Drosophila melanogaster were found in Acari genomes. This result suggests that cellular dsRNA uptake in Acari is endocytosis-dependent. Detailed phylogenetic analyses of core RNAi pathway genes in the studied Acari species revealed that their evolution is compatible with the proposed monophyletic evolution of this group.

Conclusions Taken together, our in silico comparative analyses have revealed the potential activity of all three pathways in Acari. However, much experimental work remains to be done in elucidating the operating mechanisms behind these pathways in particular those that govern systemic/parental RNAi and siRNA amplification in Acari.

Epigenetics & Genomics

RNAi

siRNA

Acari

With over 55,000 described species, Acari (mites and ticks) are the most speciose group of organisms within the subphylum Chelicerata, which is considered to be the second most diverse group of terrestrial organisms after the subphylum Hexapoda to which insects belong [1]. This apparently monophyletic group is subdivided into two lineages (super-orders) based on their morphological characteristics, feeding habits and ecological niches namely: the Parasitiformes and Acariformes [2, 3]. While many Acari species provide important ecosystem services, others are known to impose serious economic and health problems to humans, managed crops, wild and domestic animals [2, 4, 5]. Harmful Acari cause direct and indirect damages on their hosts through their feeding activities and transmission of deadly pathogens, respectively. Although, many strategies have been successfully implemented to manage these pests, chemical treatment with synthetic acaricides remains the topmost management option that is currently used [6–8]. The economic losses to global food and animal production due to these harmful Acari and expenditure on acaricide usage to manage them is amounts to billions of dollars [5, 9, 10]. Growing evidence indicates loss of efficacy of these chemicals as mites developed resistance to most of them [8, 11]. Moreover, synthetic acaricides are notorious to the environment and non-target organisms. Therefore, new management options for harmful Acari are required.

RNA interference (RNAi) has been widely explored in the management of harmful Acari as well as in functional genomics [12, 13]. The ability for sequence-specific gene silencing by this molecular tool is known to be highly conserved in eukaryotic organisms [14, 15]. So far, three pathways for RNAi have been described in Eukaryotes, namely: The micro-RNA (miRNA), the piwi-interacting RNA (piRNA) and the short interfering RNA (siRNA) [16–18]. Although these pathways differ in a number of ways, interactions between them are known to occur [17, 19]. Firstly, while siRNA pathway can be exo- and endogenously triggered, the other two pathways are strictly endogenously triggered [17]. Secondly, both siRNA and miRNA pathways are triggered by small interfering RNAs (siRNAs) and microRNAs (miRNAs), respectively, transcribed from double-stranded RNAs (dsRNAs) that depend on the Dicer proteins in the cytoplasm for their biogenesis [17, 19]. In contrast, the piRNA pathway is triggered by piwi-interacting RNAs (piRNAs), transcribed from single-stranded RNA (ssRNA) precursors that are independent of Dicer proteins for their biosynthesis [19, 20]. Thirdly, all the three pathways depend on the proteins of the Argonaute superfamily for silencing the target genes, but the active Argonaute protein family differ between the pathways. Both siRNA and miRNA pathways depend on the Ago family Argonautes whereas the piRNA pathway depend on the Piwi family Argonautes [20– 22]. Lastly, these pathways can cross-regulate each other as reported in a few eukaryotic organisms [17].

The experimental application of RNAi for gene silencing in Arthropods including Acari exploits the siRNA pathway, which is activated mainly via the exogenous administration of specific synthetic dsRNAs [12, 13, 23–25]. In principle, three molecular processes are expected to define the success of gene silencing: the uptake of dsRNA by the target cell, the amplification of the siRNAs and the transport of the latter to distant cells to induce gene silencing throughout the treated organism, a phenomenon referred to as systemic RNAi [23]. These processes are chains of reactions involving a number of essential proteins. However, information on the number of proteins involved, their structure and function is still scanty in Acari [13, 26].

The first step following the administration of the trigger dsRNA molecule is its cellular uptake, which occurs either via a transmembrane channel-mediated mechanism and/or an endocytosis-mediated mechanism [23]. In the nematode- Caenorhabditis elegans, dsRNA uptake within somatic and germline cells involves both mechanisms. The cellular uptake of dsRNAs via the transmembrane channel mechanism is facilitated by a conserved systemic RNAi defective 1 (Sid-1) protein [27, 28]. Meanwhile, the uptake via the endocytosis mechanism is facilitated by the conserved RNAi spreading defective-3 (Rsd-3) protein [29]. In addition, other proteins namely: Rsd-2 and Rsd-6 mediate cellular dsRNA uptake only in C. elegans germline cells [30]. In D. melanogaster, cellular uptake of dsRNA is mediated predominantly by scavenger receptor endocytosis proteins: Eater and SR-CI [31, 32] and Rsd-3 [33]. The siRNA amplification process has been described in plants, fungi, C. elegans and a few other organisms and requires the action of RNA-dependent RNA polymerase (RdRP) [18, 23, 34, 35]. In C. elegans, the secondary siRNAs produced by RdRPs from primary siRNAs following Dicer cleavage associate with a specific group of Argonaute proteins of the WAGO family to direct target gene cleavage [34, 36, 37]. In C. elegans, the Sid-1 protein already mentioned above is involved in systemic RNAi [27, 28]. Homologs of this protein has been reported in other organisms including insects [33], but not yet in Acari. However, apparently the systemic spread is not entirely dependent on this protein, as organisms lacking this gene are still able to show systemic RNAi [33].

Even though successful gene silencing and systemic RNAi have been reported in some Acari species using siRNA pathway, failures or variable success in gene knockdown has also been reported (reviewed in [13, 26]). The reasons for the variable success or failure in gene silencing could be partly due to difference among the species in the proteins that are involved in either of its processes as was suggested for similar problems in insects (reviewed in [38]). Given the complexity and the interactions between the RNAi cascades, siRNA pathway efficacy may be connected to the structure and function of the other two RNAi pathways, which are also rarely described in Acari. In order to shed some light on the possible pathways that may be involved in Acari, we searched for putative homologs of proteins involved in the three RNAi pathways as well as those involved in cellular uptake of dsRNAs, siRNA amplification and systemic RNAi in the genome assemblies of five species of Parasitiformes and four species of Acariformes annotated at the protein levels.

Small RNA biosynthetic proteins

Using the orthology detection tool called Proteinortho5 [39] and/or Blastp against the NCBI protein database, we found homologs of Drosha, Pasha, Exportin1 and Exportin2 in the genomes of ecto-parasitic mites: Varroa destructor, V. jacobsoni and Tropilaelaps mercedesae, the predatory mite: Metaseiulus occidentalis and the black-legged tick: Ixodes scapularis. All of these species belong to Parasitiformes. Proteins similar to Exportin5 were also found in the genomes of these species, except T. mercedesae. Using the same search approaches, we also found homologs of all these proteins in Acariformes species that include the dust mites: Dermatophagoides pteronyssinus, Euroglyphus maynei, the two-spotted spider mite: Tetranychus urticae and the scabies mite: Sarcoptes scabiei. As presented in Table 1, multiple copies of Drosha, Pasha, Exportin1 and 5 homologs were found only in the genomes of V. destructor, V. jacobsoni, T. urticae, D. pteronyssinus, E. maynei and I. scapularis. Using InterProScan [40], HmmScan [41] and the online motif search tool (https://www.genome.jp/tools/motif/) we further confirmed the presence or absence of known conserved domains in homologs of these proteins identified in all these Acari species (Table 1).

Table 1. Number of protein homologs involved in the biosynthesis of miRNAs and their characteristic conserved domains. Asterick (*) following species name abbreviations indicates that its genome is fully sequenced whereas no asterisk indicates that its genome is partially sequenced. (^#) Indicates that the protein contains no conserved domains.

Proteins	Domain description	Acari Super-order	Parasitiformes					Acariformes
		Acari orders	Ixodida	Mesostigmata				Trombidiformes	Sarcoptiformes
		Domain ID (IPR/Pfam)	Is*	Mo*	TM	Vd	Vj	Tu*	Dp	Em	Ss
Drosha	Ribonuclease domain	IPR000999/PF14622	1	1	1	6	3	1	1	1	1
	One DsRNA binding domain	IPR014720/PF00035
Pasha	DsRNA binding domain	IPR014720/PF00035	2	1	1	3	3	2	2	1	1
Exportin1	Importin-beta_N	IPR001494/PF03810	1	1	1	2	2	3	1	1	1
	Exportin-1/Importin-b-like	IPR013598/PF08389
	Exportin-1, C-domain	IPR014877/PF08767
Exportin2	Importin-beta_N	IPR001494/PF03810	1	1	1	1	1	1	1	1	1
	Exportin-2_C	IPR005043/PF03378
	Exportin-2_Central	IPR013713/PF08506
Exportin5	Exportin-1/Importin-b-like	IPR013598/PF08389	1	1	0	1	1	2	1^#	2^#	1^#

Proteins shown are from Ixodes scapularis (Is), Metaseiulus occidentalis (Mo), Tropilaelaps mercedesae (Tm), Tetranychus urticae (Tu), Varroa destructor (Vd), Varroa jacobsoni (Vj), Dermatophagoides pteronyssinus (Dp) Euroglyphus maynei (Em) and Sarcoptes scabiei (Ss).

With the same search approaches mentioned above, we identified homologs of Dicer1 (Dcr1) and Dicer2 (Dcr2) proteins in the genome assemblies of all the studied Acari species. Phylogenetic analysis based on the alignment of the two catalytic conserved Ribonuclease domains of all Dicer homologs identified in these species along with homologs of this protein in C. elegans (Ce), D. melanogaster (Dm) and an Acari species Dermatophagoides farina (Df) was carried out. Additionally, we searched for the presence of known conserved domains of Dicer proteins in each identified homolog as outlined in the method's section. The analysis revealed that homologs of Dicer1 and Dicer2 in Acariformes species clustered separately from those found in Parasitiformes species (Fig. 1A). Moreover, we found that Dicer1 homologs identified in V. destructor (Vd), V. jacobsoni (Vj), T. urticae (Tu), I. scapularis (Is) and D. pteronyssinus (Dp), clustered together with significant bootstrap support (Fig. 1A) and have similar domain architecture to Dicer1 of both D. melanogaster (Dm-Dcr1) and C. elegans (Ce-Dcr1) (Fig. 1B). In the same vein, Dicer2 homologs in these species clustered together and have similar domain architecture to Dm-Dcr2 with the exception of the homolog identified in T. urticae, which clustered with Ce-Dcr1 (Fig. 1A, B). A single Dicer1 homolog was found in all these species while several copies of Dicer2 were detected in the genome assemblies of V. destructor (2), V. jacobsoni (2) and D. pteronyssinus (2). As shown in Fig. 1A, duplicates of Dicer2 in V. destructor and V. jacobsoni clustered together while those identified in D. pteronyssinus did not. It is worthwhile to note that duplicates of Dicer2 in V. destructor, V. jacobsoni and D. pteronyssinus have similar domain architecture to Dm-Dcr2 (Fig. 1B).

Dicer1 homolog found in T. mercedesae (Tm) clustered with Vj-Dcr1 and Vd-Dcr1 with significant bootstrap values (Fig. 1A) though it did not have similar domain architecture with Dm-Dcr1 and Ce-Dcr1 (Fig. 1B). Homolog of Dicer2 in this species has a similar domain structure with Dm-Dcr2 and clustered closely with Vj-Dcr2 and Vd-Dcr2 (Fig. 1A, B). On the other hand, homologs of Dicer1 and Dicer2 identified in E. maynei and S. scabiei clustered closely with Dp-Dcr1, Df-Dcr1, Dp-Dcr2 and Df-Dcr2 (Fig. 1A) though they differed significantly in domain architecture from Dm-Dcr1/Ce-Dcr1 and Dm-Dcr2, respectively (Fig.1B). As shown in Fig. 1A and B, one homolog of Dicer1 found in M. occidentalis (Mo-Dcr1) clustered with Tm-Dcr1, Vj-Dcr1 and Vd-Dcr1 with significant bootstrap values and clearly shared similarities in domain architecture with Dm-Dcr1 and Ce-Dcr1. However, the three Dicer2 homologs identified in this species (Mo-Dcr2a,b,c) grouped significantly with homologs of this protein identified in I. scapularis, T. mercedesae, V. destructor and V. jacobsoni though they did not have similar domain structure with Dm-Dcr2 (Fig. 1A, B).

Fig. 1 Phylogenetic analysis and schematic domain architecture of Dicer proteins. The tree in (A) was built based on the alignment of the two catalytic conserved Ribonuclease domains (IPR000999/PF00636) of Dicer homologs identified in the studied Acari species using MAFFT [42]. The studied species include: Varroa destructor (Vd), Varroa jacobsoni (Vj), Tetranychus urticae (Tu), Ixodes scapularis (Is), Dermatophagoides pteronyssinus (Dp), Tropilaelaps mercedesae (Tm), Euroglyphus maynei (Em), Metaseiulus occidentalis (Mo) and Sarcoptes scabiei (Ss). Homologs of Dicer proteins in Caenorhabditis elegans (Ce), Drosophila melanogaster (Dm) and Dermatophagoides farinae (Dp) were also included in this tree. The species names were abbreviated for convenience and the letters (a, b and c) following the names of some species indicate the number of gene copies found in their genomes. The tree was constructed using PhyML [43], with the model recommended by Lefort et al. [44] under the Akaike information criterion (LG+G) with 500 bootstrap replicates. The domain architecture of Dicer proteins in (B) was generated by searching for known conserved domains in the Pfam database using InterProScan [40], HmmScan [41] and the online motif search tool (https://www.genome.jp/tools/motif/). The letters (a, b and c) indicate the number of protein copies found in their genomes.

We also identified homologs of Loquacious in the genomes of all Acari species investigated herein and several copies of this protein were found only in V. jacobsoni (6) and T. urticae (2) genomes (see Additional file 1). All homologs of this protein found in these Acari species have the characteristic conserved dsRNA binding domain (IPR014720/PF00035). However, we did not find either R2D2 protein nor its homolog C3PO protein previously identified in the genome of the red flour beetle Tribolium castaneum [33], in the genomes of any of the studied Acari species (see Additional file 1).

Argonautes and RISC components

Argonautes are the core effector proteins of the RNA-induced silencing complexes (RISCs) in the cytoplasm. They belong to the Argonaute superfamily, which is functionally divided into three families based on their interactions with specific RNA substrates: Ago, Wago and Piwi [19]. In this study, we found homologs of Ago family Argonautes in Parasitiformes and Acariformes species using the same search approaches mentioned above. A maximum likelihood tree based on the alignment of the PIWI domain showed that homologs of the Ago family identified in Acari belong to two distinct groups that appear to be specialized in either mi- or siRNA- directed gene silencing (Fig. 2). It was intriguing to find that some of the Acari homologs clustered together with Dm-Ago1, Ce-Alg1 and Ce-Alg2, whereas others clustered closely with Dm-Ago2 and Ce-Rde1. Therefore, we referred to these homologs as Ago1 and Ago2, respectively. Except in E. maynei whose genome encodes a single copy of Ago1 and Ago2, we found more copies of Ago2 than Ago1 in the genomes of the remaining Acari species. The tree further showed that homologs of Ago1 and Ago2 in Acariformes species clustered separately from those identified in Parasitiformes species, with significant bootstrap values (Fig. 2).

Fig. 2 Phylogenetic analysis of Agonaute proteins. The tree was constructed based on the alignments of the conserved PIWI domain of Argonaute homologs identified in the studied Acari species: Varroa destructor (Vd), Varroa jacobsoni (Vj), Tetranychus urticae (Tu), Ixodes scapularis (Is), Dermatophagoides pteronyssinus (Dp), Tropilaelaps mercedesae (Tm), Euroglyphus maynei (Em), Metaseiulus occidentalis (Mo) and Sarcoptes scabiei (Ss) using MAFFT [42]. Homologs of these proteins in Caenorhabditis elegans (Ce), Drosophila melanogaster (Dm), Tribolium castaneum (Tc), Psoroptes ovis (Po) and Dermatophagoides farinae (Dp) were also included in this tree. The species names were abbreviated for convenience and the letters (a, b, c, d, e, f, g and h) following the names of some species indicate the number of protein copies found in their genomes. The tree was constructed using PhyML [43], with the model recommended by Lefort et al. [44] under the Akaike information criterion (LG+G+I+F) with 500 bootstrap replicates.

The search for the presence of known conserved domains revealed that almost all the homologs of Ago1 in Acari have the same domain architecture with Dm-Ago1 and Ce-Alg1 and 2, with the exception of the single homolog identified in E. maynei, which lacks the N-terminal domain (Fig. 3). Contrary to Ago1 homologs identified in Acari, which mostly shared the same domain architecture with homologs of this protein in C. elegans and D. melanogaster, most of the Ago2 homologs in Acari had a domain architecture that was quiet distinct from those of both C. elegans and D. melanogaster (Fig. 3). The majority of them lack the Mid domain or both the N-terminal and the Mid domains. Only all the homologs of Ago-2 identified in I. scapularis (Ago-2a, b, c and d) and one homolog of this protein identified in M. occidentalis (Ago-2c) and T. urticae (Ago-2h) have similar domain structure with Dm-Ago2.

Fig. 3 Schematic domain architecture of Argonaute proteins of the Ago family in Acari. The domain architecture of the proteins was generated by searching for known conserved domains in the Pfam database using InterProScan [40], HmmScan [41] and the online motif search tool (https://www.genome.jp/tools/motif/). The letters (a, b, c, d, e, f, g and h) following the names of the protein (Ago1or Ago2) indicate the number of protein copies found in the individual Acari species.

Argonautes of the Piwi family such as Argonaute-3 (Ago3), Aubergine (Aub) and Piwi were found in the genomes of I. scapularis, T. urticae, V. destructor, V. jacobsoni, M. occidentalis and T. mercedesae, but not in genomes of D. pteronyssinus, E. maynei and Sarcoptes scabiei. We identified homologs of Ago3 in the genomes of I. scapularis, T. urticae, V. destructor, V. jacobsoni, M. occidentalis and T. mercedesae (see Additional file 1). Notably, several copies of this homolog were found only in the genomes of I. scapularis (2) and V. jacobsoni (2). We further identified putative homologs of D. melanogaster-Aub and Tribolium castaneum (Tc)-Aub/Piwi only in the genome assemblies of I. scapularis and T. urticae (see Additional file 1). It is important to note that one homolog of Drosophila Aub/Piwi and Ago-3 were previously detected in the genome of the red flour beetle, Tribolium castaneum (Tc) [33]. In our study, we identified one and two copies of Dm-Aub and Tc-Aub/Piwi, respectively, in I. scapularis and one and four copies of Tc-Aub/Piwi and Dm-Aub, respectively in T. urticae (see Additional file 1). All homologs of the Piwi family Argonautes identified in T. urticae and I. scapularis had the PIWI and PAZ domains, while those identified in V. destructor, V. jacobsoni, M. occidentalis and T. mercedesae had only the PIWI domain. Phylogenetic analysis placed homologs of Ago3, Aub and Piwi in Acari with homologs of these proteins in D. melanogaster and T. castaneum (Fig. 2).

We found homologs of the Wago family Argonautes (Ce-PPW1, -PPW2, -SAGO1 and -SAGO2) in some Acari species studied herein using Proteinortho5 [39] (see Additional file 1). One and two homologs of Sago2 and Sago1, respectively, were identified in I. scapularis's genome. Also, one homolog of PPW2 was found in T. mercedesae while one and three homologs of Sago1 were identified in V. jacobsoni and V. destructor, respectively. In T. urticae, homologs of Sago1 (1) and PPW2 (1) were also found. Interestingly, most of the homologs of the Wago family Argonautes were the same homologs, which were highly similar to Ago family Argonautes (Ago2) identified in these species (represented as Is-Ago2a and Ago2b, Tm- Ago2c, Vd-Ago2b, Vj-Ago2b and Tu-Ago2c) (see Additional file 1). Since all the homologs of the Wagofamily Argonautes clustered with those of the Ago family Argonautes, with significant bootstrap support (Fig. 2), we tentatively referred to them as members of the Ago family Argonautes that is Ago2 homologs. It is worth mentioning that all C. elegans WagoO family Argonautes used as query sequences in our study have both the PAZ and the PIWI domains.

Homologs of the Ago and Piwi families Argonautes identified in all these Acari species were further examined for the presence of the conserved residues, Aspartate-Aspartate-Histidine/Aspartate/Lysine (DDH/DDD/DDK). These residues that are present within the catalytic PIWI domain presumably enable some Argonaute proteins to cleave the target mRNAs (reviewed in [45], [46]). As shown on Table 2, the DDH catalytic residue was found in all the Argonautes identified in I. scapularis. Similarly, Argonautes of the Ago family identified in E. maynei’s genome contained this motif. The only member of the Piwi family that is Ago3 identified in the genomes of M. occidentalis, T. mercedesae and V. jacobsoni did not have any of these residues in the sequences of their PIWI domains, whereas all the Argonautes of the Ago family identified in these species have the DDH residue. Also, Ago3 and one homolog of Ago-2 (that is Ago2d) identified in V. destructor lacked these conserved residues, though the remaining proteins of the Ago family had the conserved DDH residue. Also, Ago1 of S. scabiei did not have these residues, whereas Ago2 homologs did have. In T. urticae, all the Argonautes of the Ago family had the DDH motif, while only four out of the six Argonautes of the Piwi family had the DDH motif. In D. pteronyssinus, Ago1 and Ago2a have the DDH motif, while the remaining Argonautes that were Ago2b to 2g had the DDD motif.

Table 2. Catalytic residues of Argonautes of the Ago, Piwi and Wago groups in Acari species, Caenorhabditis elegans, Drosophila melanogaster and Tribolium castaneum.

Ce-Alg-1

Ago

Ce-Alg-2

Ago

Ce-Rde-1

Ago

Ce-PPW-1

Wago

Ce-PPW-2

Wago

Ce-Sago-1

Wago

Ce-Sago-2

Wago

Dm-Ago-1

Ago

Dm-Ago-2

Ago

Dm-Ago-3

Piwi

Dm-Aub

Piwi

Dm-Piwi

Piwi

Tc-Ago-1

Ago

Tc-Ago-2

Ago

Tc-Ago-3

Piwi

Tc-Aub

Piwi

Tc-Piwi

Piwi

Is-Ago-1

Ago

Is-Ago-2a

Ago

Is-Ago-2b

Ago

Is-Ago-2c

Ago

Is-Ago-2d

Ago

Is-Ago-3a

Piwi

Is-Ago-3b

Piwi

Is-Aub

Piwi

Is-Aub/Piwia

Piwi

Is-Aub/Piwib

Piwi

Mo-Ago-1

Ago

Mo-Ago-2a

Ago

Mo-Ago-2b

Ago

Mo-Ago-2c

Ago

Mo-Ago-2d

Ago

Mo-Ago-2e

Ago

Mo-Ago-3

Piwi

Tm-Ago-1

Ago

Tm-Ago-2a

Ago

Tm-Ago-2b

Ago

Tm-Ago-2c

Ago

Tm-Ago-2d

Ago

Tm-Ago-3

Piwi

Vd-Ago-1

Ago

Vd-Ago-2a

Ago

Vd-Ago-2b

Ago

Vd-Ago-2c

Ago

Vd-Ago-2d

Ago

Vd-Ago-2e

Ago

Vd-Ago-2f

Ago

Vd-Ago-2g

Ago

Vd-Ago-2h

Ago

Vd-Ago-3

Piwi

Vj-Ago-1a

Ago

Vj-Ago-1b

Ago

Vj-Ago-2a

Ago

Vj-Ago-2b

Ago

Vj-Ago-2c

Ago

Vj-Ago-2d

Ago

Vj-Ago-2e

Ago

Vj-Ago-3a

Piwi

Vj-Ago-3b

Piwi

Tu-Ago-1a

Ago

Tu-Ago-1b

Ago

Tu-Ago-1c

Ago

Tu-Ago-2a

Ago

Tu-Ago-2b

Ago

Tu-Ago-2c

Ago

Tu-Ago-2d

Ago

Tu-Ago-2e

Ago

Tu-Ago-2f

Ago

Tu-Ago-2g

Ago

Tu-Ago-3

Piwi

Tu-Auba

Piwi

Tu-Aubb

Piwi

Tu-Aubc

Piwi

Tu-Aubd

Piwi

Tu-Aub/Piwi

Piwi

Df-Ago-1

Ago

Df-Ago-2a

Ago

Df-Ago-2b

Ago

Df-Ago-2c

Ago

Df-Ago-2d

Ago

Df-Ago-2e

Ago

Df-Ago-2f

Ago

Df-Ago-2g

Ago

Po-Ago-1

Ago

Po-Ago-2

Ago

Em-Ago-1

Ago

Em-Ago-2a

Ago

Dp-Ago-1c

Ago

Dp-Ago-2a

Ago

Dp-Ago-2b

Ago

Dp-Ago-2c

Ago

Dp-Ago-2d

Ago

Dp-Ago-2e

Ago

Dp-Ago-2f

Ago

Dp-Ago-2g

Ago

Ss-Ago-1c

Ago

Ss-Ago-2a

Ago

Ss-Ago-2b

Ago

Argonautes shown are from Caenorhabditis elegans (Ce), Drosophila melanogaster (Dm), Tribolium Castaneum (Tc), Ixodes Scapularis (Is), Metaseiulus occidentalis (Mo), Tropilaelaps mercedesae (Tm), Varroa destructor (Vd), Varroa jacobsoni (Vj), Tetranychus urticae (Tu), Dermatophagoides farinae (Dp), Psoroptes ovis (Po), Euroglyphus maynei (Em), Dermatophagoides pteronyssinus (Dp), and Sarcoptes scabiei (Ss). The amino acid sequences of the PIWI domains were aligned with MAFFT [42]. Substitutions for the Aspartate-Aspartate-Histidine (Asp-Asp-His) motif such as Aspartate or Histidine to either Serine (S), Arginine (R), Alanine (A), Asparagine (N), Glycine (G) or Tyrosine (Y), are colored in pink. Substitutions of the Histidine to Aspartate (D) or Lysine (K) are colored in light blue. (-) indicate the absence of amino acid residues from the sequence.

Other components of the RISC complex notably Vasa intronic gene (Vig-1) and Tudor-staphylococcal nuclease (Tsn-1), that also contribute to the degradation of the target mRNAs [47], [48], were also identified in the genome of the Acari species in this study, with the exception of E. maynei, which had Vig-1 but not Tsn-1 homolog (see Additional file 1).

DsRNA uptake and spread

We searched for Sid-1, Rsd-2, Rsd-3, Rsd-6, Eater and SR-CI proteins that are involved in cellular dsRNA uptake and systemic spread of the silencing dsRNA molecules as mentioned above. We were unable to identify proteins similar to Sid-1 nor its homologs SilA, SilB and SilC, previously found in the T. castaneum genome [33] in the studied Acari species. Similarly, homologs of Rsd-2 and Rsd-6 were not found in any of the studied Acari species, though homologs of Rsd-3 protein were present in all of them (see Additional file 1). Duplicates of this protein were only found in the genomes of I. scapularis (2), V. jacobsoni (5) and V. destructor (9) respectively. Interestingly, all the identified homologs had a single conserved domain, the epsin N-terminal homology (ENTH) domain that has been shown to be sufficient to mediate the transport of dsRNA molecules into both somatic and germ cells of C. elegans [29]. We also identified homologs of Dm-Eater in the genomes of some Acari species but not in those of D. pteronyssinus, E. maynei and S. scabiei. Again, duplicates of this protein were found in T. urticae (3), V. destructor (5) and V. jacobsoni (2). Homologs of this protein in the Acari species have several epidermal growth factor (EGF)-like modules. Furthermore, one homolog of Dm-SR-CI receptor was identified only in the genome of T. urticae. The Drosophila SR-CI receptor and its homolog had a characteristic conserved domains coding for Meprin, A5 protein, and tyrosine phosphatase Mu (MAM).

siRNA secondary amplification

Systemic and trans-generational gene interference rely on the amplification of the initial siRNA trigger for efficient gene knockdown throughout the treated organism. This mechanism for enhancing RNAi potency necessitates the action of a cellular RNA-dependent RNA polymerase (RdRP) protein, which has been extensively studied in C. elegans [49, 50]. Interestingly, we found that all Acari species from the two sister lineages, Acariformes and Parasitiformes studied here, poses such RdRP proteins. Duplicates of this protein were found in the genomes of all studied species. Moreover, all the Acari RdRP proteins identified in this study including those previously identified in C. elegans (Ego-1 and Rrf-1) [37] and other Acari species e. g. D. farinae [51] and Psoroptes ovis [52] had the characteristic conserved RdRP domain [34]. In order to infer the evolutionary history of Acari RdRP proteins, we conducted a phylogenetic analysis based on the amino acid alignment of RdRP domain using C. elegans as outgroup, as described in the methods section. The phylogeny revealed that proteins of Acariformes (indicated with “green” color on Fig. 4) clustered strongly together (494/500 bootstraps) and separately from those of Parasitiformes. The Parasitormes proteins (indicated with “red” color on Fig. 4) also clustered strongly together (408/500 bootstraps). However, three out of the total four proteins identified in I. scapularis were found to be evolutionary closer to Acariformes proteins than to those of Parasitiformes (479/500 bootstraps) (Fig. 4).

Fig 4. Phylogenetic distribution of RNA-dependent RNA polymerase (RdRP) proteins in Acari species: Varroa destructor (Vd), V. jacobsoni (Vj), Tropilaelaps mercedesae (Tm), Metaseiulus occidentalis (Mo), Ixodes scapularis (Is), Dermatophagoides pteronyssinus (Dp), Euroglyphus maynei (Em), Tetranychus urticae (Tu), Sarcoptes scabiei (Sc) and Psoroptes ovis (Po). The nematode species, Caenorhabditis elegans (Ce) was used as outgroup. The species names were abbreviated for convenience and the numbers (1, 2, 3, 4 and 5) following their names indicate the protein copies found in each species. The tree was constructed using PhyML [43], with the model recommended by Lefort et al. [44] under the Akaike information criterion (LG+G+I+F) with 500 bootstrap replicates and is based on the amino acid alignment of the conserved RdRP domain (IPR007855/PF05183) using MAFFT [42].

Proteins associated with the three RNAi pathways in Acari

Our findings demonstrate that the genomes of all the nine Acari species that belong to diverse Acari orders of agricultural, veterinary and medical importance encode homologs of the proteins that mediate the silencing process in the three RNAi pathways. In particularly, we demonstrate that the genomes of all these species encode homologs of the RNase III enzyme Drosha, its co-factor Pasha and nuclear export receptors, Exportin1, 2 and/or 5 (Table 1 and 3) thereby suggesting that these proteins, which play vital roles in the initial step of miRNA biogenesis that takes place inside the nucleus are conserved in Acari as in other animal species [53–57].

Functional specialization of Dicer and Argonaute proteins was apparent in Acari. This assumption is supported by phylogenetic analysis of Dicer proteins which showed two Dicer paralogs with distinct functions in Acari: Dicer1 and Dicer2 involved in miRNA and siRNA biosynthesis, respectively (Table 3, Fig. 1A, B) as was previously shown in other two Acari species: P. ovis [52] and D. farina [51]. Two Dicer paralogs were also found in insect species [33], [58–60] and crustaceans [57, 61] analyzed so far, but only one (Dicer1) was so far detected in nematodes and mammals [19]. In case of Argonautes, our phylogenetic analysis revealed representatives of two families (Ago and Piwi) in Acari (Table 3, Fig. 2) out of three families known in eukaryotes [17, 19]. Within the Ago family, we further detected representatives of two distinct functional groups: Ago1 and Ago2 specialized in miRNA- and siRNA-directed gene silencing, respectively. The fact that these two gene families clustered separately from Wago family Argonaute specific to C. elegans with significant bootstrap value indicates that Acari, as insects [33] and crustaceans [24, 55], lack clear orthologs of Wago family Argonautes in their genomes. The match that occurred between C. elegans Wago family Argonautes and some orthologs of Ago family Argonautes (Ago2) in Acari, supports the idea of common ancestor for both protein families as was suggest by Swarts et al. [62]. The finding that functional separation of core effector proteins involved in the exogenous siRNA and endogenous miRNA/piRNA machineries have occurred in Acari as in insects [33, [60] and crustaceans [24, 55, 57], supports the previous hypothesis that duplication and specialization of Dicers and Ago family Argonautes have occurred already in ancestral Arthropoda [59], early before its split into two monophyletic groups: Chelicerata and Mandibulata, back in the Precambrian era [63]. However, further analysis of homologs of these proteins in the other Arthropod lineages is needed to ascertain whether these phenomena are widely conserved or merely lineage-specific.

Table 3. Summary of numbers of protein orthologs associated with the three RNAi pathways identified in the nine investigated Acari genomes. Asterick (*) following species name abbreviation indicates that its genome is fully sequenced whereas no asterisk indicates that its genome is partially sequenced.

Proteins	Acari Super-order	Parasitiformes					Acariformes
	Acari orders	Ixodida	Mesostigmata				Trombidiformes	Sarcoptiformes
	Species/Pathway	Is*	Mo*	TM	Vd	Vj	Tu*	Dp	Em	Ss
Drosha	miRNA	1	1	1	6	3	1	1	1	1
Dicer1	miRNA	1	1	1	1	1	1	1	1	1
Dicer2	siRNA	1	3	1	2	2	1	2	1	0
Argonaute1	miRNA	1	1	1	1	2	3	1	1	1
Argonaute 2	siRNA	4	5	4	8	5	8	7	1	2
Argonaute 3	piRNA	2	1	1	1	2	1	0	0	0
Aubergine/Piwi	piRNA	3	0	0	0	0	5	0	0	0
Rsd-3	Cellular dsRNA uptake	2	1	1	9	5	1	1	1	1
Eater		1	1	1	5	2	3	0	0	0
SR-CI		0	0	0	0	0	1	1	1	1
RdRP	siRNA amplification	4	3	3	4	5	5	5	4	1
Sid-1/SilA/SilB/SilC	Systemic RNAi	0	0	0

A genome-wide screening for RNAi pathway proteins in Acari

Status:

Journal Publication

Version 2

Abstract

Figures

Background

Results

Discussion