RNA-guided RNA silencing by an Asgard archaeal Argonaute

Abstract

Argonaute proteins are the central effectors of RNA-guided RNA silencing pathways in eukaryotes, playing crucial roles in gene repression and defense against viruses and transposons. Eukaryotic Argonautes are subdivided into two clades: AGOs generally facilitate miRNA- or siRNA-mediated silencing, while PIWIs generally facilitate piRNA-mediated silencing. It is currently unclear when and how Argonaute-based RNA silencing mechanisms arose and diverged during the emergence and early evolution of eukaryotes. Here, we show that in Asgard archaea, the closest prokaryotic relatives of eukaryotes, an evolutionary expansion of Argonaute proteins took place. In particular, a deep-branching PIWI protein (HrAgo1) encoded by the genome of the Lokiarchaeon 'Candidatus Harpocratesius repetitus' shares a common origin with eukaryotic PIWI proteins. Contrasting known prokaryotic Argonautes that use single-stranded DNA as guides and/or targets, HrAgo1 mediates RNA-guided RNA cleavage, and facilitates gene silencing when expressed in human cells and supplied with miRNA precursors. A cryo-EM structure of HrAgo1, combined with quantitative single-molecule experiments, reveals that the protein displays structural features and target-binding modes that are a mix of those of eukaryotic AGO and PIWI proteins. Thus, this deep-branching archaeal PIWI may have retained an ancestral molecular architecture that preceded the functional and mechanistic divergence of eukaryotic AGOs and PIWIs.

Fig. 1. Diversity of Asgard archaeal Argonautes and a deep-branching PIWI-clade HrAgo1.

a Maximum-likelihood phylogenetic analysis of the MID-PIWI domains of Argonaute proteins showing that asAgos are polyphyletic. HrAgo1 is phylogenetically related to Argonautes in the eukaryotic PIWI clade. 334 representative sequences and 572 sites were analyzed using IQ-TREE based on Q.pfam + C60 + F + G4 model. Color pallets indicate host classification, outer ring indicate pAgo and eAgo classifications. Ultrafast bootstrap 2 (UFBoot2) values above 95, calculated based on 1000 replicates, are shown in gray circles. HrAgo1, HrAgo2, and the UFBoot2 values at the branches are highlighted. Inset, Maximum-likelihood phylogenetic analysis based on the MID-PIWI domains with a broader sampling of AGO and PIWI indicates that HrAgo1 is sister to all eukaryotic PIWIs. UFBoot2 values calculated based on 1000 replicates are indicated. b Fraction of Argonaute-encoding genomes in different prokaryotic phyla. The tree topology was adapted from GTDB v207 by removing phyla with less than 40 representatives. Source data are provided in the Source data file. c Genomic depiction of Asgard archaeon ‘Ca. H. repetitus’, where the genes encoding 16S and 23S rRNA, origin of replication protein Cdc6, and putative immune systems are indicated. The synteny and predicted domain compositions of genes surrounding pAgo-encoding genes are highlighted. RM, restriction-modification system. Blue bars indicate two genome assembly gaps with undetermined sequences.

Fig. 2. HrAgo1 mediates RNA-guided RNA cleavage.

a HrAgo1 associates with 5’ phosphorylated (5’ P) small RNAs in vivo from E. coli. Nucleic acids that co-purified with HrAgo1 were [γ-³²P] labeled, treated with RNase A or DNase I, and resolved on a denaturing gel (15% polyacrylamide 7 M urea). nt: nucleotides. b Length distribution of small RNAs associated with HrAgo1 as determined by small RNA sequencing. c Small RNAs associated with HrAgo1 have a bias for uracil bases at the 5’ end. d Sequences of guide and target oligonucleotides used in in vitro cleavage assays. e HrAgo1 cleaves ssRNA (but not ssDNA) targets with ssRNA guides, and ssDNA guides at lower efficiency, in the presence of Mg²⁺. HrAgo1 was incubated with ssDNA or ssRNA guides and Cy5-labeled ssDNA or RNA targets. Cy5-labeled cleavage products were resolved through denaturing (7 M urea) polyacrylamide gel electrophoresis and visualized by fluorescence imaging. Both ssRNA and ssDNA targets are 45nt. The HrAgo1-bound RNA extraction and digestion was carried out once; the results of the cleavage assays were confirmed by at least three repetitions.

Fig. 3. Molecular architecture of HrAgo1 bound to a guide RNA.

a Schematic diagram of the domain organization of HrAgo1. N, N-terminal domain; L1 and L2, linker domains; PAZ, PIWI-ARGONAUTE-ZWILLE domain; MID, Middle domain; PIWI, P-element induces wimpy testis domain. b Schematic representation of the HrAgo1-bound guide RNA. Structurally ordered residues are colored red, while disordered residues are colored gray. c Cryo-electron microscopic (cryo-EM) density map of HrAgo1 bound to a guide RNA Cartoon, colored according to individual domains. The unmodeled 3’-end guide RNA density is represented as a transparent surface. d Cartoon representation of the overall structure of the HrAgo1-guide RNA complex. e All-against-all structure comparison of selected Argonaute proteins. Source data are provided in the Source data file. f Close-up view of the HrAgo1 catalytic site aligned to that of other representative Argonaute proteins. g Close-up view of the HrAgo1 guide RNA 5’-end binding site in the MID domain aligned to that of other representative Argonaute proteins. h Efficient HrAgo1-mediated RNA cleavage requires a guide RNA with a 5’ phosphate and an intact catalytic site. HrAgo1 was incubated with ssDNA or ssRNA guides and Cy5-labeled ssRNA targets. DM: HrAgo1 catalytic mutant with D585A and E623A substitutions. i HrAgo1 mediates RNA-guided RNA cleavage at temperatures ranging from 9 °C to 71 °C. HrAgo1 was incubated with ssRNA guides and Cy5-labeled ssRNA targets. For (h) and (i), Cy5-labeled cleavage products were resolved on a denaturing (7 M urea) polyacrylamide gel and visualized by fluorescence imaging. In (h) and (i), both uncleaved ssRNA targets are 45nt. The results of the cleavage assays were confirmed by at least three repetitions.

Fig. 4. HrAgo1 displays a unique hybrid mode of guide organization and target binding.

a Close-up view of guide RNA organization by HrAgo1. b Comparison of structural features involved in guide RNA seed segment organization in HrAgo1, EfPiwi, and hAgo2. c Schematic of the single-molecule binding assay. Only when the HrAgo1-guide complex binds to the target, FRET will occur. d Schematic representation of a guide and target used in the single-molecule binding assay. Complementary nucleotides are indicated in dark gray and mismatched nucleotides are shown in light gray. N6 indicates base pairing with nucleotides 2–7 of the guide. e A representative time trace with four binding events, of which the dwell time (Δτ) of one is indicated. f Dwell time distributions for N4, N6, and N15. N4 and N6 are best fit with a single and double exponential, respectively. N15 cannot be fit and shows stable binding. The distributions and fits for the other match lengths and representative time traces can be found in Fig. S6. g Bubble plots showing the increase of dwell times for increasing complementarity between the guide and target for HrAgo1, EfPiwi and hAgo2. The area of the bubbles corresponds to the percentage of the total population belonging to this sub-population. The dashed lines indicate the time resolution (0.1 s) and the observation time limit (200 s). In the bubbles, the mean ± SD are indicated, and the darker shaded area indicates the SD of the fractional values of at least three independent experiments. The dwell times for EfPiwi were obtained in a similar way as for HrAgo1. For hAgo2, previously published dwell times were used. Source data are provided in the Source data file.

Fig. 5. HrAgo1 mediates RNA silencing in human cells.

a A schematic diagram for stable transfection of miR-1-1. pLKO.1 puro-pri-mir-1-1 vector was transfected into AGO1/2/3 KO HCT116 cells, which were subsequently subjected to puromycin selection for 16 days to generate cells that stably express miR-1-1. b A schematic diagram for the RNAi rescue experiment. Puromycin-selected cells were co-transfected with a dual-luciferase expression vector containing two perfect target sites for miR-1-1 in the 3′ UTR of the firefly luciferase gene (Fluc) and a protein expression vector encoding sfGFP or hAgo2 or HrAgo1. Two days after the transfection, total RNA was isolated and subjected to RT-qPCR. c qPCR results for relative mRNA expression levels between firefly luciferase (Fluc) and Renilla luciferase (Rluc). Bars indicate mean ± SD (n = 3, biological replicates). ns, not significant; **p < 0.01; *p < 0.05 by independent two-sided t-test. The p values are: 0.725 (hAGO2 vs. HrAGO1), 0.014 (sfGFP vs. hAGO2), and 0.0064 (sfGFP vs. HrAGO1). Source data are provided in the Source data file.