The Firre locus produces a trans-acting RNA molecule that functions in hematopoiesis

RNA has been classically known to play central roles in biology, including maintaining telomeres, protein synthesis, and in sex chromosome compensation. While thousands of long noncoding RNAs (lncRNAs) have been identified, attributing RNA-based roles to lncRNA loci requires assessing whether phenotype(s) could be due to DNA regulatory elements, transcription, or the lncRNA. Here, we use the conserved X chromosome lncRNA locus Firre, as a model to discriminate between DNA- and RNA-mediated effects in vivo. We demonstrate that (i) Firre mutant mice have cell-specific hematopoietic phenotypes, and (ii) upon exposure to lipopolysaccharide, mice overexpressing Firre exhibit increased levels of pro-inflammatory cytokines and impaired survival. (iii) Deletion of Firre does not result in changes in local gene expression, but rather in changes on autosomes that can be rescued by expression of transgenic Firre RNA. Together, our results provide genetic evidence that the Firre locus produces a trans-acting lncRNA that has physiological roles in hematopoiesis.


138
Since Firre is found on the X chromosome, we first sought to determine if deletion of the 139 locus had an effect on the expected sex ratio of the progeny. Matings between ∆Firre mice 140 5 produced viable progeny with a normal frequency of male and female pups that did not exhibit 141 overt morphological, skeletal, or weight defects (Extended Data Table 1 and Extended Data Fig.   142 2). Moreover, deletion of Firre did not impact expression levels of Xist RNA in embryonic tissues 143 or perturb Xist RNA localization during random X chromosome inactivation (XCI) in mouse 144 embryonic fibroblasts (MEFs) (Extended Data Fig. 3A-C).

145
Because the ∆Firre allele removes the entire gene body, this model does not allow us to 146 distinguish between DNA-and RNA-mediated effects. Therefore, in order to be able to investigate

167
In contrast to ∆Firre MEFs, we detected pronounced localization of Firre RNA in the nucleus of 168 WT MEFs (Fig. 1H). In dox-treated Firre rescue MEFs, which only produce Firre RNA from the 169 transgene, we detected Firre RNA in both the nucleus and cytoplasm (Fig. 1H), which 170 corresponded to approximately a 2.7-fold increase in Firre RNA relative to WT (Fig. 1I). Notably,  (Fig. 1A), we took an initial unbiased approach to 177 explore the potential biological roles for the Firre locus, and performed poly(A)+ RNA-seq on eight 178 E11.5 tissues from WT and ∆Firre embryos (forebrain, midbrain, heart, lung, liver, forelimb, 179 hindlimb, and pre-somitic mesoderm). As expected, Firre expression was not detected in any of

183
Across these eight tissues, we identified a total of 3,910 significantly differentially 184 expressed genes, of which 271 genes were differentially expressed in two or more tissues 185 (Extended Data Tables 2-9). Interestingly, gene ontology (GO) analysis of the commonly 186 dysregulated genes showed that deletion of the Firre locus affected genes involved in hemoglobin 187 regulation and general blood developmental processes (Fig. 2C). We therefore analyzed publicly 188 available mouse RNA-seq datasets and found that Firre is expressed across many blood cell 189 types and note that expression is found highest in hematopoietic stem cells (HSCs) 48 and then 190 decreases in conjunction with hematopoietic differentiation 49 (Fig. 2D). Based on this information 191 we narrowed our investigation to evaluate potential roles for Firre in the blood system, and 192 leveraged the genetic mouse models to test DNA-and RNA-mediated effects.

193
194 LPS exposure to mice overexpressing Firre RNA impacts the innate immune response.

195
Firre is expressed in many innate immune cell types (Fig. 2D) and has been shown to regulate 196 the levels of inflammatory genes in human intestinal epithelial and mouse macrophage cell lines 34 .

197
Thus, we hypothesized that dysregulation of Firre might alter the inflammatory response in vivo.

198
To test this, we employed a commonly used endotoxic shock model by administering

202
We administered two different LPS preparations, one which broadly stimulates the pattern 203 recognition receptors toll-like receptors (TLR) 2, 4 and nitric oxide synthase, and an ultrapure LPS 204 preparation that specifically stimulates TLR4 51-53 . At 5 hours post LPS injection we measured 205 serum cytokine levels. Notably, we observed that Firre OE dox-fed mice administered broad-acting 206 LPS had significantly higher levels of inflammatory cytokines, including TNFa, IL12-p40, and MIP-207 2 compared to WT (Fig. 2F). In contrast, we did not observe a significant difference for these 208 7 cytokines in LPS-treated ∆Firre mice (Fig. 2F). Consistent with the increased cytokine response 209 using broad-acting LPS, dox-fed Firre OE mice administered TLR4 specific-acting LPS also had 210 significantly higher levels of TNFa, IL12-p40, and MIP-2 compared to WT (Fig. 2G), albeit at lower 211 serum concentrations compared to the broad-acting LPS (Fig. 2F,G). In addition, we confirmed 212 that overexpressing Firre RNA alone (without LPS) does not result in increased serum levels of 213 TNFa, IL12-p40, and MIP-2 (Extended Data Fig. 4).

214
Because increased levels of TNFa is a hallmark of endotoxic shock 54-56 , we next tested 215 whether the levels of Firre RNA had an impact on survival following LPS treatment. We 216 administered 5mg/kg of TLR4 specific-acting LPS to WT (n=30), ∆Firre (n=18), Firre OE

248
We assessed the long-term reconstitution ability of WT and ∆Firre HSCs to repopulate blood cell

356
In this study, we developed three genetic models in mice for the syntenically conserved  with a previous study using cell culture models 29 . Together, these data are notable because cis-395 acting mechanisms are speculated to be common feature at lncRNA loci 65 . While we did not find 396 any evidence for cis-activity at the Firre locus in vivo, a previous study from our group found active 397 DNA elements within the Firre locus using a cell-based enhancer reporter assay in 3T3 cells 22 .

398
We speculate that these candidate DNA regulatory elements are likely to regulate the Firre locus 399 rather than neighboring genes, as we did not find evidence of dysregulation in gene expression 400 for neighboring genes when the locus was deleted in vivo.

401
In summary, we have examined the role of Firre in the context of hematopoiesis in order 402 to test DNA-and RNA-mediated effects. This study does not exclude that Firre could be 403 functioning elsewhere, and even by other molecular modalities. Indeed, we identified that Firre is 404 abundantly expressed in a number of tissues, therefore going forward it will be important to       Fig. 1A,B). For each tissue, we generated the following libraries: WT male

537
The adapter-trimmed reads were mapped to the mm10 mouse reference genome using

542
Genes with an FDR<0.1 were deemed significant in a comparison between wildtype and ∆Firre 543 (Extended Data