Mouse B2 SINE elements function as IFN-inducible enhancers

  1. Isabella Horton
  2. Conor J Kelly
  3. Adam Dziulko
  4. David M Simpson
  5. Edward B Chuong

  • Curated by eLife

    eLife logo

    eLife assessment

    This paper will be of interest to scientists studying species-specific immune responses and those studying how transposable elements rewire transcriptional regulatory networks. The work describes a new class of TEs that may act as enhancers of immune genes in mice. A combination of computational and experimental data supports most but not all conclusions in the paper.

Reviewed by

Read the full article See related articles

Listed in

Log in to save this article

Abstract

Regulatory networks underlying innate immunity continually face selective pressures to adapt to new and evolving pathogens. Transposable elements (TEs) can affect immune gene expression as a source of inducible regulatory elements, but the significance of these elements in facilitating evolutionary diversification of innate immunity remains largely unexplored. Here, we investigated the mouse epigenomic response to type II interferon (IFN) signaling and discovered that elements from a subfamily of B2 SINE (B2_Mm2) contain STAT1 binding sites and function as IFN-inducible enhancers. CRISPR deletion experiments in mouse cells demonstrated that a B2_Mm2 element has been co-opted as an enhancer driving IFN-inducible expression of Dicer1 . The rodent-specific B2 SINE family is highly abundant in the mouse genome and elements have been previously characterized to exhibit promoter, insulator, and non-coding RNA activity. Our work establishes a new role for B2 elements as inducible enhancer elements that influence mouse immunity, and exemplifies how lineage-specific TEs can facilitate evolutionary turnover and divergence of innate immune regulatory networks.

Article activity feed

  1. Version published to 10.7554/elife.82617 on eLife
  2. eLife

    eLife assessment

    This paper will be of interest to scientists studying species-specific immune responses and those studying how transposable elements rewire transcriptional regulatory networks. The work describes a new class of TEs that may act as enhancers of immune genes in mice. A combination of computational and experimental data supports most but not all conclusions in the paper.

  3. eLife

    Reviewer #1 (Public Review):

    In this manuscript, Horton et al. seek to define the role of TEs in shaping the murine innate immune regulatory landscape. This work follows previous studies that identified enrichment of RLTR30 elements within STAT1 binding sites in IFN-induced genes. Here, the authors re-analyze previously published transcriptomic and epigenomic datasets to screen for TEs showing signatures of inducible regulatory activity upon IFNG stimulation in mouse macrophages. Data presented in this study provide evidence that a specific B2 SINE subfamily (B2_Mm2) is enriched among regions bound by inducible STAT1 and H3K27ac, which are associated with enhancer activity. Additionally, the authors identify a putative B2_Mm2 derived inducible enhancer for Dicer1 located within its first intron. Cell lines harboring deletions of this element no longer show IFNG-inducible expression of Dicer1 and show a repressive effect on the expression of Serpina genes.

    While the data and analyses presented here are of good quality and the authors present some interesting data (specifically that deletion of B2_Mm2.Dicer1 ablates inducible expression of Dicer1), several conclusions drawn by the authors are overstated and not fully supported by the data presented. Furthermore, additional controls are required to firmly establish that B2_Mm2.Dicer1 functions as an inducible enhancer that regulates genes within the Serpina-Dicer1 locus.

  4. eLife

    Reviewer #2 (Public Review):

    Horton et al combined computational and functional approaches to identify a role for a mouse transposable element (TE) family in the transcriptional response to interferon gamma (IFNG, also known as type II interferon). This paper builds on previous work, some of which was done by the corresponding author, in which TE families have been shown to contribute transcription factor binding sites to genes in a species-specific manner. In the current work, the authors analyzed datasets from mouse primary macrophages that had been stimulated by IFNG to identify TEs that might contribute to the transcriptional response to IFNG treatment. In addition to previously identified endogenous retrovirus subfamilies, the authors identified sites from another TE family, B2_Mm2, that they found contained STAT1 transcription factor binding sites and whose binding by STAT1 was induced following IFNG stimulation. To test the hypothesis that a B2_Mm2 element was providing IFNG-inducibility to an associated gene, the authors chose one of the 699 mouse genes that had nearby (<50 kb) B2_Mm2 elements and was upregulated upon IFNG treatment in previous datasets. The gene they chose was Dicer1, which also is upregulated by IFNG in mouse macrophages but not in human primary macrophages, furthering the hypothesis that the presence of B2_Mm2 in mouse cells may provide IFNG-inducibility to Dicer1. Following KO of a ~500 bp region in two separate clones of immortalized mouse macrophages, the authors show a decrease in basal as well as IFNG-induced expression of Dicer1, providing support for their conclusion that a B2_Mm2 is important for IFNG-inducibility. The authors further show that two nearby genes that are also upregulated by IFNG, Serpina3f and Serpina3g, are also reduced at basal conditions as well as when stimulated with IFNG. The authors use these data to suggest that additional elements in the B2_Mm2 element in the Dicer1 gene, possibly CTCF elements, are have long distance effects on transcription of nearby genes.

    Overall, this is an interesting and well written manuscript. The computational conclusions are supported by their data and add to the growing field of TEs and their role in transcription regulatory network evolution. While the authors do a good job of experimentally validating one example, inclusion of additional data, all of which they already have, as detailed below would substantially increase the applicability of their work and strengthen their conclusions about the broad role of TEs in the IFNG response in mice versus other species.

    1. Following their genome-wide comparisons, the authors hone in on Dicer1 as an interesting example in which they hypothesize that a B2_Mm2 element near the Dicer1 gene could be contributing to the fact that this gene is upregulated by IFNG in mouse cells but not human cells. What would be very useful to the readers of this paper is knowing how many other examples there might be like this one. Adding DEseq values from human RNAseq data the authors already use (current references 10 and/or 37) for identifiable human orthologs to Table S7 would thus strengthen their conclusions. If Dicer1 is unique in this aspect of having (a) a nearby B2_Mm2 element and (b) a binary difference between inducibility in mouse versus human cells, that is interesting. If Dicer1 is not unique, that strengthens the authors' assertion that B2_Mm2 insertions have altered the transcriptional network in a host-specific manner. Either way, the answer is interesting, but without including this analysis, the authors leave out an important aspect of their work and it remains unclear how generalizable their conclusions are.

    2. The results with Serpina3g and Serpina3F gene expression in the authors' knockout cells are very interesting. However, the authors focus almost exclusively on Serpina3g and Serpina3F, which makes it difficult to understand what is happening genome wide. Are other IFNG-induced genes (including those not on chromosome 12) similarly affected at the level of basal or induced transcription? How many genes are different in WT versus KO cells, both at basal and induced states? Does this correlate with their CUT&TAG data shown in Fig. 5? By focusing only on nearby genes (Serpina3g and Serpina3F), the authors hint that this may be a long range regulatory effect, "potentially mediated by the CTCF binding activity of the element" that they removed. But by only focusing on two nearby IFNG-induced genes, their data do not rule out the (also potentially quite interesting) possibility that there may be a more indirect role for this TE site or Dicer1 in basal transcription of IFNG-induced genes or IFNG-mediated gene expression. Providing more data on other genes throughout the genome in WT and KO cells, which the authors have generated but do not include in the manuscript, would help distinguish between these models. While a broader effect of these KOs on IFNG expression, or gene expression in general, would not fit as neatly with their model for local gene regulation, these analyses are needed to understand the effects of TE insertion on gene regulation.

  5. eLife

    Reviewer #3 (Public Review):

    First of all, I enjoyed the manuscript by Horton et al. In the manuscript, they first re-analyzed published ChIP-seq data for STAT1 binding in INF-activated macrophages and found that a fourth of the >20,000 STAT1 binding sites were in transposable elements. Especially, about 10% of the total STAT1 binding sites were in B2_Mm2, a murine-specific SINE. They showed that these B2 elements are associated with H3K27ac signal upon INF treatment, thus likely serve as an INF-inducible enhancer through STAT1 binding. The authors then focus on the STAT1-bound B2_Mm2 in the Dicer1 gene (designated as B2_Mm2.Dicer1), and demonstrated that deletion of this B2 in a macrophage-like murine cell line resulted in loss of STAT1 binding, H3K27ac, and Dicer1 upregulation upon INF treatment. Their findings suggest that B2 transposition events has altered the transcriptional regulatory network in the innate immune response in the mouse.

    The manuscript is well organized, and the findings are potentially interesting in terms of the evolution of species-specific regulatory networks of the innate immune response. But, I am not convinced with the enhancer role of the B2_Mm2.Dicer1 copy for the Dicer1 expression (see below).

    Major Comments:

    (1) In Fig. 4, the degree of Dicer1 induction by INF was small (1.2-fold or so), and accordingly the effect of the B2 deletion on the Dicer1 induction was also small. In addition, this B2 binds to CTCF, and its deletion should also eliminate CTCF binding. Therefore, it is difficult to conclude from the presented data that this B2 serve as an enhancer for Dicer1. The B2 may increase the frequency of transcription (as suggested by the authors), may serve as an obstacle for transcriptional elongation (via binding to CTCF), or may regulate the splicing efficiency. In Fig.5C, promoter acetylation level does not seem to be affected in KO1. Pol II either does not seem to be affected if the Pol II peak is compared to the background level. Taken together, the enhancer role is not supported by strong evidence.

    (2) On the other hand, the authors discovered that the B2 deletion resulted in the decrease of Serpina3h, Serpina3g, Serpina3i and Serpina3f by >100-fold, which are 500 kb apart from the B2 locus. This is also interesting, and could be evidence for the B2 enhancer. Given that this B2 binds to both STAT1 and CTCF, the locus could interact with the Serpina3 locus to act as an enhancer. Were there STAT1 CUT&TAG peaks around the Serpina3 genes? Did H3K27ac and Pol II ChIP peaks in the Serpina3 promoters disappear in the KO cells? It would be interesting to see the IGV snapshots for H3K27ac, POLR2A and STAT1 ChIP-seq data around Serpina3 genes. In addition, HiC data for activated macrophages, if available, could be supportive evidence for the interaction between B2_Mm2.Dicer1 and the Serpina3 locus.

    Minor Comments:

    (3) Regarding Fig.1C, the authors calculated the B2 expression levels by mRNA-seq and DESeq2 analysis. But it does not accurately give the B2 transcription level, because the method does not discriminate B2 RNAs and B2-containing mRNA (and lncRNA as well). I wonder that the apparent upregulation of STAT1-binding B2 loci is due to the increase of Pol II transcription around the loci, rather than Pol III-mediated B2 transcription. This possibility should be discussed in page 6 after "Taken together, these data indicate that thousands of B2_Mm2 elements show epigenetic and transcriptional evidence of IFNG-inducible regulatory activity in primary murine bone marrow derived macrophages."

    (4) Fig. 2B shows that about 70-80% of B2_Mm2 loci carry the STAT1 motif, whereas only a limited number (2-3%) of B2_Mm2 bind to STAT1. Is this because of differences in their motif sequences, in genomic locations, or in epigenomic environments? For example, do these STAT1-binding loci have a C-to-A mutation at the second last position in the GAS motif (TTCNNGGAA), like B2_Mm2.Dicer1 (shown in Fig. S4)? Can the authors discuss about it? In addition, although the consensus sequence of B2_mm2 has a GAS motif with only a single mismatch, the presence of the STAT1 motif in >70% of B2_Mm2 is surprising, given that their average divergence to the consensus sequence is about 10% (ref. 26 of the manuscript). Is the binding site significantly conserved in compare to the other regions of the B2 sequence?

  6. Version published to 10.1101/2022.08.02.502523v1 on bioRxiv