(C) PLOS One [1]. This unaltered content originally appeared in journals.plosone.org.
Licensed under Creative Commons Attribution (CC BY) license.
url:https://journals.plos.org/plosone/s/licenses-and-copyright

------------



Functional analysis of ADARs in planarians supports a bilaterian ancestral role in suppressing double-stranded RNA-response

['Dan Bar Yaacov', 'The Shraga Segal Department Of Microbiology', 'Immunology', 'Genetics', 'Faculty Of Health Sciences', 'Ben-Gurion University Of The Negev', 'Beer-Sheva', 'Department Of Integrative Biology', 'University Of Wisconsin-Madison', 'Madison']

Date: 2022-02 

ADARs (adenosine deaminases acting on RNA) are known for their adenosine-to-inosine RNA editing activity, and most recently, for their role in preventing aberrant dsRNA-response by activation of dsRNA sensors (i.e., RIG-I-like receptor homologs). However, it is still unclear whether suppressing spurious dsRNA-response represents the ancestral role of ADARs in bilaterians. As a first step to address this question, we identified ADAR1 and ADAR2 homologs in the planarian Schmidtea mediterranea, which is evolutionarily distant from canonical lab models (e.g., flies and nematodes). Our results indicate that knockdown of either planarian adar1 or adar2 by RNA interference (RNAi) resulted in upregulation of dsRNA-response genes, including three planarian rig-I-like receptor (prlr) homologs. Furthermore, independent knockdown of adar1 and adar2 reduced the number of infected cells with a dsRNA virus, suggesting they suppress a bona fide anti-viral dsRNA-response activity. Knockdown of adar1 also resulted in lesion formation and animal lethality, thus attesting to its essentiality. Simultaneous knockdown of adar1 and prlr1 rescued adar1(RNAi)-dependent animal lethality and rescued the dsRNA-response, suggesting that it contributes to the deleterious effect of adar1 knockdown. Finally, we found that ADAR2, but not ADAR1, mediates mRNA editing in planarians, suggesting at least in part non-redundant activities for planarians ADARs. Our results underline the essential role of ADARs in suppressing activation of harmful dsRNA-response in planarians, thus supporting it as their ancestral role in bilaterians. Our work also set the stage to study further and better understand the regulatory mechanisms governing anti-viral dsRNA-responses from an evolutionary standpoint using planarians as a model.

Today, more than ever, it is crucial to gain a deep understating of our anti-viral defenses. One of the ways to accomplish it is to study the principles governing anti-viral responses across various organisms. ADARs are a group of proteins that act on RNA molecules and alter their sequence compared to the genes that encode them (a process termed RNA editing). In recent years, ADARs have been shown to suppress abnormal anti-viral responses triggered by self-components of the cell (RNA encoded by the cell). Here, we show that the involvement of ADARs in anti-viral response regulation is conserved in planarians (free-living flatworms). We identified two ADAR proteins in planarians and showed that eliminating one (ADAR1) results in animal death and that an anti-viral response commenced in the absence of either ADAR1 or ADAR2. We further identified one of the proteins (PRLR1) that participate in initiating this anti-viral response in planarians, which its mammalian homolog (MDA5) serves a similar role. Thus, our work suggests that ADARs involvement in suppressing aberrant anti-viral response is an ancient evolutionary invention and is likely shared across multicellular organisms with bilateral symmetry.

Funding: DBY was awarded the Gruss Lipper Post Doctoral Fellowship ( http://www.eglcf.org/ ). DBY was also supported by Ben-Gurion University of the Negev startup grant. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Here, we describe planarian homologs of ADAR1 and ADAR2 and demonstrate roles for these proteins in the planarian dsRNA-response. RNA interference (RNAi) knockdown of adar1, but not adar2, resulted in lesions’ development and, ultimately, animal death. RNA-Seq analysis of ADAR-knockdown animals demonstrated increased expression of several genes that play roles in anti-viral immunity via RNAi and IFN-like pathways. Significantly, ADAR knockdowns led to a decreased load of SmedTV, an endogenous dsRNA virus of S. mediterranea [ 33 ]. Finally, simultaneous knockdown of prlr1, a planarian RIG-I-like receptor, and adar1 rescued lethality and delayed the dsRNA-response. Collectively, our findings demonstrate the essential immunomodulatory role of the ADAR1 homolog in invertebrates and suggest that this role is evolutionarily conserved across bilaterians.

Therefore, to broaden our perspective on the functional importance of ADARs in dsRNA-response and the evolutionary conservation of this role, we characterized and analyzed ADAR homologs in the planarian Schmidtea mediterranea. Along with mollusks, annelids, and several other animal phyla, planarians (free-living platyhelminths) belong to the superphylum Spiralia [ 27 – 29 ]. Planarians are best known for their remarkable ability to regenerate, mediated by a population of pluripotent stem cells (neoblasts) [ 30 ]. Interest in their remarkable biology has driven the development of a suite of functional-genetic tools [ 30 – 32 ]. As such, planarians make an attractive, tractable model for molecular-genetic studies from an evolutionary perspective [ 30 ].

Given these conserved functions between vertebrates and invertebrates, it has been postulated that one of the ancestral roles of ADARs is to prevent aberrant dsRNA-response [ 2 , 19 , 22 ]. However, the importance of the interaction between ADARs and RLR-mediated or RNAi pathways has only been described in the above invertebrate species. Nematodes and flies represent a limited segment of the animal evolutionary tree–both are members of the superphylum Ecdysozoa–and may lack essential characteristics to inform such evolutionary inferences [ 23 , 24 ]. For example, neither species has an apparent homolog of ADAR1 [ 17 , 18 ], whereas such homologs exist in other invertebrates such as octopuses and oysters (superphylum Spiralia; [ 25 , 26 ]). On the other hand, functional studies of ADARs’ role in the dsRNA-response have not yet been conducted in Spiralians.

Recent studies of the interaction between ADARs and dsRNA-responses in invertebrates have demonstrated intriguing parallels to vertebrates. Caenorhabditis elegans encodes two ADARs (ADR-1 and ADR-2) [ 17 , 18 ]. In adr-1; adr-2 mutant worms, components of the RNA interference (RNAi) pathway–the DICER and ARGONAUTE proteins DCR-1 and RDE-1 –have been shown to process ADAR targets [ 19 ]. Additionally, a loss-of-function mutation in drh1, which encodes an RLR homolog, suppresses the phenotype of ADAR-deficient worms, an interaction analogous to the observed interaction between ADAR1 and MDA5 in mammals [ 19 , 20 ].

While dsRNA molecules are inevitable products of normal cellular function, they are also commonly generated as intermediates of viral replication [ 10 ]. As such, they serve as molecular patterns that activate innate immune responses. Organisms must therefore balance between vigilance against foreign dsRNAs without overreacting to innocuous self dsRNA. Emerging evidence suggests a vital role for ADARs in this balancing act. In mammals, for example, ADAR1 is essential to life due to its role in suppressing an interferon (IFN) innate immune response activated by MDA5 (melanoma differentiation-associated protein 5), a dsRNA sensor in the RIG-I (retinoic acid-inducible gene I) like receptor (RLR) family, which binds to long, near-perfectly base-paired structures. [ 11 – 13 ]. Loss of ADAR1 function in mice triggers an embryonically lethal interferon response, which was rescued in Mda5 knockout mice [ 11 , 12 ]. Similarly, in humans, mutations in both ADAR1 and MDA5 (also known as IFIH1) are known to cause Aicardi-Goutières syndrome, a devastating inflammatory autoimmune disease [ 14 – 16 ].

ADARs are found across all multicellular animal lineages (including corals) [ 4 ] and play several essential roles. For example, vertebrates possess three ADARs: ADAR1 and ADAR2 are catalytically active and known to be essential for viability in mammals [ 1 , 2 , 5 – 8 ], while ADAR3 appears catalytically inactive [ 9 ]. Mammalian ADAR1 is responsible for most identified editing events, most of which occur in non-coding sequences. For example, in humans, ADAR1 targets mostly inverted Alu elements in introns and untranslated regions of mRNA, which form dsRNA structures post-transcriptionally [ 2 ]. ADAR2, on the other hand, is thought to mediate its effects primarily through protein recoding [ 2 , 8 ].

Results

Knockdown of adar1 is lethal To examine the function of ADARs in planarians, we used RNAi knockdown of gene expression. RNAi reduced adar1 and adar2 transcripts to 22% and 41%-58%, respectively, compared to their levels in control(RNAi) animals (S3 Fig and S2 Table). All adar1(RNAi) animals were smaller than control (RNAi) animals, developed lesions, and 73% (44/60) died (Fig 1B and 1C). In contrast, adar2(RNAi) animals did not display any gross morphological phenotype changes, and were similar to control(RNAi) animals (Fig 1B and 1C). Notably, the observed phenotype in adar1(RNAi) animals did not correspond to the canonical phenotype of neoblast loss (i.e., head regression and ventral curling) [35]. Indeed, adar1(RNAi) (and adar2(RNAi)) animals were able to regenerate upon the head or tail amputation, performed no more than five days before lesions formed in adar1(RNAi) animals (S4 Fig). Furthermore, neither WISH for the pan-neoblast marker piwi-1 nor flow cytometric analysis of cellular fractions revealed depletion of neoblasts after knockdown of adar1 or adar2 (S5 Fig). Therefore, our data collectively suggest that ADAR1 is essential in planarians but that its function is not critical for neoblast maintenance.

ADAR1 and ADAR2 suppress a bona fide anti-viral dsRNA-mediated response We next tested whether increased expression of the dsRNA-response genes following knockdown of either adar1 or adar2 constituted a bona fide dsRNA-response in planarians. If this were the case, one would expect a negative effect of adar1 or adar2 knockdown on RNA viruses in the treated animals (e.g., less infected cells / viral RNA due to upregulation of anti-viral factors). A recent report described a dsRNA virus, S. mediterranea tricladivirus (SmedTV), in the planarian nervous system [33]. Therefore, we assessed the prevalence of SmedTV infected cells and RNA as indicators of the activity of the planarian dsRNA-response. Notably, it was reported that the level of infection (i.e., number of infected cells per worm) varied considerably between individual worms [33]. To overcome this obstacle and obtain sufficient statistical power, we sampled more than 20 worms (pooled from two independent experiments) and counted the number of infected cells in the head of the animals (Fig 3A), as we observed that the majority of SmedTV infected cells, across RNAi treatments, resides in the cephalic ganglion of our sampled animals. Following our prediction, the average number of infected cells was reduced in adar1(RNAi) and adar2(RNAi) animals (Fig 3A and 3B). The reduction was statistically significant in adar2(RNAi) animals (p < 0.05) and marginally significant in adar1(RNAi) animals (p = 0.06). This could be due to the observed large inter-individual variability or could be the result of having a technical outlier (Fig 3A and 3B). In addition to the observed reduction in the number of infected cells, SmedTV RNA abundance was also significantly reduced in both adar1(RNAi) and adar2(RNAi) animals (Fig 3C). Taken together, these results suggest that both ADAR1 and ADAR2 dampen the dsRNA-response in planarians. PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 3. Knocking down adar1 or adar2 reduces viral RNA in infected planarians. (A) Representative confocal images (FISH–maximum-intensity projection (MIP)) of cells harboring dsRNA of the S. mediterranea tricladivirus (SmedTV—magenta) in adar1(RNAi), adar2(RNAi), and control(RNAi) animals after 21 days. Scale bar = 200 μm. The red box on the cartoon indicates the imaged area. Contrast and brightness were adjusted equally across all three images for better visualization. (B) Quantification of SmedTV+ cells in A (mean ± SD; N = 2, n ≥ 20 (pooled animals from both experiments)). One-way ANOVA with Dunnett’s multiple comparisons test. Adjusted p-value ≤ 0.05 (*). We pooled the data from two independent experiments after 21 and 23 days of RNAi. Data points corresponding to Fig 3A are marked in red. (C) Relative expression levels (RNA-Seq; mean ± SD; N = 4 (with three animals that were pooled together in each experiment)) of SmedTV RNA in adar1(RNAi), adar2(RNAi), and control(RNAi) after 19 and 28 days of RNAi. FDR ≤ 0.0001 (****). No RNA-Seq data for adar2(RNAi) animals at 19 days of RNAi. FC = Fold Change. https://doi.org/10.1371/journal.ppat.1010250.g003

PRLR1 is involved in mediating dsRNA-response in adar1(RNAi) and adar2(RNAi) animals Next, we asked whether PRLR1 function is necessary for the increased dsRNA-response following knockdown of adar1. Indeed, adar1(RNAi); prlr1(RNAi) animals displayed a lower average expression level of the dsRNA-response genes, relative to both adar1(RNAi) and adar1(RNAi); control(RNAi) animals after ten days of RNAi (Fig 5A). adar1 levels did not differ between single and double knockdowns (Figs 5A and S8), further demonstrating that the reduction in expression of dsRNA genes does not result from disruption of adar1 knockdown but rather from the effect on PRLR1. However, the reduction in expression was transient. After 14 days of RNAi, the expression of all examined dsRNA-response genes was similar between single and double RNAi treatments involving adar1 (S8 Fig). Thus, it is likely that additional factors are involved in inducing the dsRNA-response in adar1(RNAi) animals or that residual amounts of the PRLR1 protein following knockdown of prlr1 were still able to initiate the dsRNA-response in the absence of ADAR1 (albeit at a lower rate). Next, we asked if PRLR1 plays a role in mediating the dsRNA-response in adar2(RNAi) animals. We observed lower average expression levels of all examined dsRNA genes in adar2(RNAi); prlr1(RNAi) animals relative to both adar2(RNAi) and adar2(RNAi); control(RNAi) animals after 14 days of RNAi (Fig 5B). However, the effect was not as strong as in the case of adar1 (i.e., only being statistically significant for stat5), raising the possibility of additional factors involved in the regulation of dsRNA-response upon adar2 knockdown. PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 5. PRLR1 is involved in mediating dsRNA-response in adar1(RNAi) and adar2(RNAi) animals. (A) Relative expression levels (qPCR; mean ± SD; N = 3 (with three animals that were pooled together in each experiment)) of seven dsRNA-response genes and adar1 after ten days of RNAi. (B) Relative expression levels (qPCR; mean ± SD; N = 3; n = 3) of seven dsRNA-response genes and adar2 after 14 days of RNAi. FC = Fold change. Statistical analysis—One-way ANOVA with Sidak’s multiple comparisons test. Adjusted p-value ≤ 0.05 (*), ≤ 0.01 (**), ≤ 0.001 (***) and ≤ 0.0001 (****). https://doi.org/10.1371/journal.ppat.1010250.g005 In mammals, it was observed that activation of the IFN response by MDA5 in mice with deficient ADAR1 activity leads to an increase in cell death [11,41]. We, therefore, asked whether programmed cell death can explain lesion formation in adar1(RNAi) animals. However, we could not detect an increase in programmed cell death (apoptosis) as assayed by TUNEL (S9 Fig) [42], suggesting a different mechanism underlying lesion formation and lysis in adar1(RNAi) animals. Combined, these results are consistent with PRLR1 mediating a dsRNA-response in planarians, which ADARs at least partly suppress in healthy planarians.

[END]

[1] Url: https://journals.plos.org/plospathogens/article?id=10.1371/journal.ppat.1010250

(C) Plos One. "Accelerating the publication of peer-reviewed science."
Licensed under Creative Commons Attribution (CC BY 4.0)
URL: https://creativecommons.org/licenses/by/4.0/


via Magical.Fish Gopher News Feeds:
gopher://magical.fish/1/feeds/news/plosone/