Skip to main content

Dicer-Dependent Defense in Mammals

Periodic Reporting for period 4 - D-FENS (Dicer-Dependent Defense in Mammals)

Reporting period: 2020-01-01 to 2020-06-30

The project D-FENS focused on how mammalian cells deal with threats associated with double-stranded RNA (dsRNA) while maintaining viability and genome integrity. dsRNA from viruses and mobile elements can be intercepted by RNase III Dicer acting in the RNA interference (RNAi) pathway, an ancient eukaryotic antiviral mechanism providing invertebrate innate immunity. In mammals, RNAi has been superseded by a different immune system but was not entirely lost. The project examined whether the vestigial RNAi pathway could be revived in vivo and what would be the fitness cost. Understanding principles governing evolution of RNAi and co-existence of RNAi with more recent antiviral pathways could bring new strategies for antiviral therapies. Analysis of evolution of genome maintenance and impact of mobile elements can reveal fundamental principles of evolution of new gene regulations and functions. The project had and achieved three major objectives:

(I) Explore consequences of functional RNAi in vivo. We produced a genetically modified mouse where RNAi was “reinstalled”. While this model revealed negative effects of hyperactive RNAi in vivo, it still enabled testing potential of RNAi to suppress viral infections in mammals. We observed that, under some conditions, the modified mouse can exhibit improved antiviral immunity.

(II) - Define common and species-specific features of RNAi in the oocyte. We explored evolutionary history of RNAi and its co-existence with the miRNA pathway. We discovered a molecular mechanism of mammalian gene evolution driven by retrotransposon insertions and a long-sought mechanism inactivating miRNAs in mouse oocytes and showed that it is common for mammalian oocytes.

(II) - Uncover relationship between RNAi and piRNA pathways in suppression of retrotransposons. We showed the degree of redundancy of RNAi and piRNA pathways in mouse oocytes and divergence between piRNA pathways between mice and hamsters.
(I) Short Dicer variant & antiviral potential of RNAi

The objective was to explore consequences of hyperactive RNAi in vivo. We proposed to produce a “super RNAi” mouse model where the Dicer gene would be modified to express an equivalent of a shortened Dicer isoform driving RNAi in mouse oocytes. First, we used genetically modified cells to define constraints for a functional highly active RNAi. The cells with enhanced RNAi were also challenged with viruses, however, there was no robust inhibitory effect on several examined viruses. The main publication outcome, so far, is the summary of data from cultured cells:

Demeter et al. (2019) Main constraints for RNAi induced by expressed long dsRNA in mouse cells. Life Sci Alliance. doi: 10.26508/lsa.201800289

The main milestone, production of mice expressing the shorter Dicer enzyme, was achieved by the end of 2017. During the rest of the project, we cleaned their genetic background and analyzed their phenotype & response to viral infection. We discovered that homozygosity of the truncated Dicer isoform is perinataly lethal. Heterozygotes, however, were viable and appeared normal. In five infection experiments, we observed once lack of development of a systemic infection, otherwise heterozygous mice showed slightly lower viral loads than wild type controls. Causes of these diverse results will be studied in the future together with analysis of susceptibility to different viruses. Work on a manuscript summarizing mouse phenotypes is still in progress.

(II) Common and species-specific features of RNAi in the oocyte.

This part aimed at understanding the roles of RNAi and miRNA pathways during oocyte-to-embryo transition (OET) in mammals.

(i) We explored evolutionary and functional aspects of maternal mRNA degradation by RNAi. Analysis of maternal transcriptomes of different mammals revealed many adaptations of the molecular machinery of RNAi and sources of small RNAs. We annotated hundreds of lncRNAs serving as substrates for gene-regulating siRNAs. We identified massive contribution of long terminal repeat insertions to mouse oocyte transcriptome remodeling and evolution of RNA interference:

Franke et a. (2017) Long terminal repeats power evolution of genes and gene expression programs in mammalian oocytes and zygotes. Genome Res. doi: 10.1101/gr.216150.116.

Another manuscript summarizing evolution of RNAi in mice and functional regulations that underlie the sterile phenotype of mice lacking RNAi will be submitted during fall 2020.

(ii) We studied co-existence of RNAi and miRNA pathways, particularly whether lack of miRNA activity in mouse oocytes (reported 2010) is linked to RNAi or is an independent phenomenon. We discovered that the miRNA pathway is commonly disabled in mammalian oocytes due to diluting effects of oocyte growth. This phenomenon is thus independent of evolution of RNAi. This is an unexpected major discovery of the project as the diluting effect of oocyte growth is a common but so far largely ignored player governing molecular biology of oocytes:

Kataruka et al. (2020) MicroRNA dilution during oocyte growth disables the microRNA pathway in mammalian oocytes. Nucleic Acids Res. doi: 10.1093/nar/gkaa543.

(III) Relationship of RNAi & piRNA pathways in the germline.

This part studied maintenance of genome stability and control of mobile elements in the germline. It brought insights into functional redundancy between RNAi and piRNA pathways, allowing for defining general and species-specific features. Three objectives of the part III of the project were developed:

(i) Investigating whether RNAi compensates lack of piRNAs in mouse oocytes. This involved producing a genetically modified mouse model lacking both pathways in oocytes. We showed that there is indeed functional redundancy between RNAi and piRNA pathways in mouse oocytes but it concerns only a subset of retrotransposons.

Taborska et al. (2019) Restricted and non-essential redundancy of RNAi and piRNA pathways in mouse oocytes. PLoS Genet. doi: 10.1371/journal.pgen.1008261.

(ii) Testing whether the short highly active Dicer is sufficient to compensate the loss of piRNAs in the male germline. This hypothesis did not gain experimental support and this part of the project was suspended.

(iii) A hamster knock-out model was used for testing piRNA function in a mammal that lacks highly active RNAi. We discovered that the hamster piRNA pathway is essential for both sexes but with very different phenotypic manifestation in each sex. A manuscript summarizing the work is being currently finalized and is planned to be submitted in September 2020.

Taken together, the scientific aims of the project were completed despite the project faced a number of common unpredictable issues, which always appear in projects of this kind. The work on publication of the remaining data is in progress and will be completed in due time.
The project progressed beyond the state-of-the-art through its animal models – (1) genetic engineering that yielded an in vivo RNAi animal model, and (2) piRNA pathway knock- out in hamster are exceptional achievements of genetic engineering allowing to address research questions that could not be examined in cultured cells (mouse model) or had to go beyond the mouse model (hamster model).