USP30-AS1 functions as a post-transcriptional amplifier of interferon signalling to drive autoimmune pathogenesis

Abstract

Interferon (IFN) signalling is essential for antiviral defence yet pathogenic in autoimmunity, however, the mechanisms orchestrating this duality remain poorly defined. Here, we identify USP30-AS1, a cytoplasmic long non-coding RNA (lncRNA) induced by type I IFN, as a pivotal post-transcriptional amplifier of innate immunity. We demonstrate that USP30-AS1 selectively enhances the mRNA stability of nucleic acid sensors. Moreover, USP30-AS1 preferentially stabilizes the majority of AU-rich element (ARE)-containing interferon-stimulated gene (ISG) mRNAs. USP30-AS1 executes this function independently of its antisense partner, USP30. The deletion of USP30-AS1 impaired IFN-β-mediated antiviral defence and suppressed pro-inflammatory cytokine production. Consistent with its role in amplifying immune responses, USP30-AS1 was markedly upregulated in human autoimmune diseases characterised by dysregulated IFN signalling, including systemic lupus erythematosus, rheumatoid arthritis, and dermatomyositis. Thus, our work unveils USP30-AS1 as a key regulator that fine-tunes the stability of nucleic acid sensors and ARE-containing immune transcripts, providing a direct mechanistic link between IFN signalling and post-transcriptional gene regulation. These findings establish USP30-AS1 as a critical rheostat for immune homeostasis and a promising therapeutic target for IFN-associated diseases.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13578-026-01548-5.

Background

Type I interferons (IFN-I) are critical cytokines in innate immunity and serve as the first line of defence against viral infections [1, 2]. Upon viral invasion, pattern recognition receptors, such as cyclic GMP-AMP synthase (cGAS) and retinoic acid-inducible gene I (RIG-I), detect viral nucleic acids and initiate signalling cascades that lead to IFN production [3–5]. Secreted IFNs activate the Janus kinase-signal transducer and activator of transcription (JAK-STAT) pathway, triggering the transcription of hundreds of interferon-stimulated genes (ISGs) and establishing an antiviral state [6, 7]. Although this rapid and potent response is essential for host defence, it requires tight regulation to prevent excessive inflammation and autoimmunity [8–10]. Dysregulated IFN-I signalling, characterized by chronic ISG activation and aberrant sensing of self-derived nucleic acids, is strongly associated with autoimmune disorders, including systemic lupus erythematosus (SLE), rheumatoid arthritis (RA), and dermatomyositis (DM), commonly referred to as type I interferonopathies [11–17]. This duality underscores the necessity for precise molecular control that balances effective antiviral immune responses and prevention of autoimmunity.

While transcription factors like interferon regulatory factors (IRFs) and STATs are well-established regulators of IFN-I signalling [18, 19], transcriptional control alone is insufficient to account for the dynamic and rapid nature of immune responses. Growing evidence highlights the importance of post-transcriptional mechanisms in modulating this process [20, 21]. Regulation of mRNA stability and translation for key immune sensors and effector molecules serves as a critical mechanism that orchestrates the magnitude, duration, and overall outcome of IFN-I-mediated inflammation [22, 23]. Thus, in addition to transcription, post-transcriptional regulation provides a critical dimension of regulation that fine-tunes innate immune responses with precision.

Long non-coding RNAs (lncRNAs) have recently emerged as important regulators of innate immune signalling [24, 25]. These molecules typically function through interactions with mRNAs or proteins, influencing gene expression at various regulatory levels. This regulation includes transcriptional modulation, chromatin remodelling, and post-transcriptional regulation, which are especially relevant to rapid immune responses [25–30]. In particular, cytoplasmic lncRNAs are increasingly recognized as key post-transcriptional regulators that affect mRNA stability, translation, and protein activity in immune pathways [27, 31]. However, the functional roles of lncRNAs in regulating antiviral and inflammatory responses, particularly those driven by IFN signalling, remain incompletely understood.

In this study, we identify USP30-AS1 as a cytoplasmic lncRNA that is rapidly induced by IFN-I. We demonstrate that it functions as a post-transcriptional amplifier of IFN-I signalling by preferentially stabilizing mRNAs encoding nucleic acid sensors and ISGs. This mechanism is essential for antiviral defence but also contributes to pathological inflammation. Notably, we observed significantly elevated USP30-AS1 expression in patients with type I interferonopathies, highlighting its clinical relevance and potential as a therapeutic target in IFN-I-driven autoimmune conditions.

Methods

Bioinformatics analysis

We utilized RNA-seq datasets to characterize the transcriptomic alterations, a publicly available dataset (accession number PRJEB21332) and an in-house dataset comparing △ Intron cells to control cells. For both datasets, raw sequencing reads were aligned to reference genome using HISAT2 (v2.2.1) with default parameters, and gene-level quantification was performed with featureCounts (v2.0.3). Differential expression analysis was conducted in R (v4.3.3) using DESeq2 (v1.30.1), with significant genes defined as those exhibiting an absolute fold change (Log2) greater than 1 and an adjusted P value less than 0.05. Functional enrichment analyses, including KEGG pathway enrichment analysis, were carried out using clusterProfiler (v4.0.1).

Human 3′UTR sequences were obtained from the UTRDB database (https://utrdb.cloud.ba.infn.it/). To quantify the presence of AREs within 3′UTR of each gene, a custom Python (v3.7.3) script was developed using reference motif sequences from the ARED-Plus database (http://brp.kfshrc.edu.sa/ared). For each gene, the corresponding 3′UTR sequence was scanned for ARE motifs using string matching and regular expressions, and the number of ARE sites was recorded.

To assess USP30-AS1 expression in various autoimmune disease cohorts, we retrieved multiple publicly available datasets from Gene Expression Omnibus (GEO, https://www.ncbi.nlm.nih.gov/geo/). These datasets were generated using either RNA-seq or microarray platforms. For RNA-seq datasets, we adopted the reported transcripts per million values as the measure of gene expression. For microarray-based datasets, we used the provided normalized intensity values corresponding to the USP30-AS1 probes.

Data visualization was performed in R (v4.3.3) using ggplot2 (v3.5.1), pheatmap (v1.0.12), and enrichplot (v1.22.0).

Cross-species conservation of the USP30-AS1 genomic locus and promoter ISRE-like motif mapping

Locus-level conservation of USP30-AS1 across species was summarized using the LncBook database, which provides cross-species conservation estimates and gene-annotation status for lncRNA loci. Promoter-level conservation was assessed by defining the USP30-AS1 promoter as ± 1 kb relative to the transcription start site (TSS) based on genome annotation, extracting the corresponding human promoter sequence from the reference genome, and projecting the human TSS coordinate to primate genome assemblies using UCSC liftOver to obtain orthologous promoter intervals. All promoter sequences were oriented to the transcription direction. Each 2-kb promoter sequence was scanned for an ISRE-like consensus motif (5’-GAAANNGAAA-3’), allowing up to one mismatch, and motif positions were visualized relative to the TSS across species.

Antibodies and reagents

Anti-EGFP (MA1-952) was from Thermo Fisher Scientific; anti-USP30 (sc-515235) and anti-β-Actin (sc-47778) were from Santa Cruz Biotechnology; anti-MDA5 (5321S) and anti-PKR (12297S) were from Cell Signalling Technology; anti-GAPDH (10494-1-AP) was from Proteintech Group; anti-RIG-I (28137) was from Immuno-Biological Laboratories; anti-α-Tubulin (T5168) was from Sigma-Aldrich. Anti-human cGAS was prepared in our laboratory, the cGAS antibody was generated by immunizing rabbits with human cGAS (amino acids 1-160). Phorbol 12-myristate 13-acetate (PMA; 524400), 5-BrU (850187) and HT-DNA (D6898) were from Sigma-Aldrich, cGAMP (tlrl-cga23) was from InvivoGen; Recombinant human IFN-β (300-02BC) was from PeproTech.

Plasmids construction

To assess the protein-coding ability of USP30-AS1, the sequences of EGFP, USP30-AS1-EGFP, USP30-AS1-T-EGFP, and USP30-AS1-TT-EGFP were cloned into the pcDNA3.1 vector. For rescue assay, sequence of USP30-AS1 was subcloned into the pCDH-CMV-MCS-EF1-Blast vector (P14038, MiaoLingPlasmid). All constructs were verified by Sanger sequencing. To assess the regulatory roles of full-length 3′UTRs, we utilized the BTV plasmid system (84771, Addgene), which contains a bidirectional tetracycline-regulated promoter driving EGFP expression. The EGFP reporter includes a multiple cloning site for insertion of the wild-type or ARE-deleted 3’UTRs, followed by a polyadenylation signal to ensure proper mRNA processing.

Peripheral blood mononuclear cells (PBMCs) isolation

PBMCs were isolated from volunteer donors at Jiaxing central blood station following informed consent. The isolation was performed using HISTOPAQUE (1077, Sigma-Aldrich) according to the manufacturer’s instructions, and all procedures were approved by the Ethics Committee of Jiaxing Central Blood Station. Human primary macrophages were then differentiated from PBMCs by treatment with 10 ng/mL recombinant human granulocyte-macrophage colony stimulating factor (215-GM, R&D Systems) for 7 days.

Cell culture, cell transfection and lentiviral infection

HEK293T cells were cultured in Dulbecco’s modified Eagle’s medium with 2 mM L-glutamine (CM10013, Macgene) containing 10% (vol/vol) fetal bovine serum (FBS), 100 U/mL penicillin and 100 mg/mL streptomycin (CC004, Macgene). U937, THP1 and primary human macrophage cells were cultured in RPMI-1640 medium with 2 mM L-glutamine (CM10040, Macgene) containing 10% FBS, 100 U/mL penicillin and 100 mg/mL streptomycin. U937 and THP1 cells were differentiated with 0.1 µM PMA for 36 h before transfection or other treatment. Transfection of HT-DNA were performed with Lipofectamine 2000 (11668019, Thermo Fisher Scientific) at indicated concentrations. Transfection of cGAMP was performed with Lipofectamine 3000 (L3000015, Thermo Fisher Scientific) at indicated concentrations. Constructed plasmids were co-transfected into HEK293T cells with the packaging plasmids pVSVg (8454, Addgene) and psPAX2 (12260, Addgene) for 48 h to generate lentivirus. U937 cells were infected with the lentivirus for 24 h, followed by selection with 2 µg/mL puromycin (AK058, GPC biotechnology) or Blasticidin (ST018, Beyotime Biotechnology). U937, THP1 and HEK293T cells were obtained from ATCC. All cell lines were regularly tested to be mycoplasma free by PCR.

Virus

HSV-1 and VSV were propagated and amplified by infection of a monolayer of Vero cells. Cells were infected with HSV-1 [multiplicity of infection (MOI) = 5] and VSV (MOI = 1) for the indicated hours.

5′- and 3′-RACE

The lncRNA USP30-AS1 full-length containing the transcriptional initiation and termination sites was confirmed by 5′- and 3′-RACE using the SMARTer RACE 5′/3′Kit (634858, Takara) by following the manufacturer′s protocol. The gene-specific primers were designed for 5′-RACE (5’-TCACCATTACCGCTACATAGTACG-3’) and 3′-RACE (5’-GAGACATGCCACCATATTTAGG-3’).

Isolation of cytoplasmic and nuclear RNA

Cytoplasmic and nuclear RNA purification kit (NGB-21000, Norgen) was used to isolate cytoplasmic and nuclear RNA of U937 cells following the manufacturer’s protocol. RT-PCR was used to determine the expression level of USP30-AS1 in the nucleus and cytoplasm. U2 was used as control for nucleus, and β-Actin was used as control for cytoplasm. Sequence information of primers is provided in Supplementary Table 1.

Western blot

Cells were lysed with M2 buffer (20 mM Tris-HCl pH 7.5, 0.5% Nonidet P-40, 250 mM NaCl, 3 mM EDTA, 3 mM EGTA, 2 mM dithiothreitol) containing protease inhibitor cocktail (7012 L, Cell Signalling Technology) and then added by 2 × SDS loading buffer (WB-0081, Dingguo Biotechnology) for 15 min at 95°C. Cell lysates were separated by SDS-PAGE and proteins were visualized by enhanced chemiluminescence regent (abs920, Absin) with Amersham ImageQuant 800 (GE Healthcare).

Quantitative PCR (qPCR)

Total RNAs were extracted from collected cells using TRIzol reagent according to the manufacturer’s protocol. Total RNAs (500 ng) were reversed-transcribed to cDNA using the PrimeScript RT Master Mix (RR036A, TAKARA). qPCR was performed with PowerUp SYBR Green Master Mix (A25778, Applied Biosystems) on a StepOnePlus Real-Time PCR System (4376600, Applied Biosystems) according to the manufacturer’s instructions. Data were analyzed with StepOnePlus software. Human β-Actin was used for normalization. Sequence information of primers is provided in Supplementary Table 1.

5-BrU chasing and ActD assay

U937 cells (1 × 10⁷) were seeded in 10-cm dishes. Cells were incubated in growing media containing 200 µg/mL 5-BrU (850187, Sigma) for 3 h to allow for metabolic labelling of newly synthesized transcripts. Then, BrU-containing media were withdrawn, and cells were chased in growing media for 0, 6 and 12 h. Total RNA was extracted from cells at each time point of the uridine chase using TRIzol reagent (932879, Sigma-Aldrich). Next, 20 µg total RNA for each time point was incubated with 2 µg anti-bromodeoxyuridine antibody (MI-11-3, MBL Life Science) in 500 µL IP buffer (5 mM Tris-HCl pH 7.5, 0.5 × PBS, 0.05% Triton X-100, 1 mM EDTA and 40 U/mL RNaseOUT) for 2 h at 4 °C with constant rotation. Each sample uses 30 µL Protein A/G magnetic beads (88802, Thermo Scientific) in 100 µL IP buffer were added to the RNA-antibody mix and incubated for an additional 1 h at 4 °C with constant rotation. Beads were then washed four times with 700 µL IP buffer. The immunopurified RNA was eluted from the beads via proteinase K treatment, and incubated for 45 min at 50 °C. Next, BrU-containing RNA was extracted using TRIzol reagent and detected by qPCR analysis. U937 cells (5 × 10⁵) were seeded in 12-well plates. Cells were treated with 5 µg/mL ActD (HY-17559, MCE) with 0, 2, 4 and 8 h. Next, total RNA was extracted using TRIzol reagent and detected by qPCR analysis.

RNA-FISH

RNA FISH experiments were performed to detect lncRNA USP30-AS1 in U937 cells using the Ribo Fluorescent In Situ Hybridization Kit (C10910, RiboBio) according to the protocol. The USP30-AS1 probes (lncCM001) were designed and synthesized by RiboBio company with Cy3 fluorescent dye. Fluorescence detection was performed with ZEISS LSM 900 Confocal Microscopy.

Dual-KO using CRISPR/Cas9 system

Two single-guide RNAs (sgRNAs) that target the specific regions of USP30-AS1 were fused via PCR to segment containing a guide RNA scaffold and an H1 promoter. The resulting fragments were ligated into BsmBI-digested lentiCRISPRv2 (52961, Addgene) plasmid by seamless cloning (CL117, Biomed). Control cells were used two sgRNAs targeting GFP sequence. Genomic DNA of selected single clones was extracted using GeneJET Genomic DNA Purification Kit (K0721, Thermo Fisher Scientific). Genotyping was validated with primers listed in Supplementary Material 1.

ELISA

PMA-differentiated U937 cells were seeded into 24-well plates at a density of 3 × 10⁵ cells per well and subjected to the indicated treatments. The secreted IFN-β was measured using ELISA kits (EHC026b.96.5, Neobioscience) based on manufacturer’s instructions.

Statistical analysis

No statistical method was used to predetermine sample size. Statistical significance was assessed using two-tailed unpaired Student’s t-test for continuous variables. Data are presented as mean ± SEM unless otherwise indicated. P < 0.05 was considered statistically significant. All analyses were performed using GraphPad Prism software.

Results

Type I IFN induces the expression of lncRNA USP30-AS1

To identify lncRNAs regulators involved in IFN-I responses, we utilized a previously established transcriptomic dataset that comprehensively characterizes IFN-induced gene expression across multiple species [32]. We focused our analysis on human transcriptomic data from primary dermal fibroblasts and revealed IFN-I-induced lncRNA candidates (Fig. 1A and B). We next examined the expression of these candidates in PMA-differentiated U937 cells, a well-established macrophage-like cell model for studying IFN-mediated antiviral and autoimmune responses. Among the top ten induced candidates, USP30-AS1 exhibited the most robust induction following IFN-β stimulation (Fig. 1C). This upregulation of USP30-AS1 was confirmed as IFN-I-inducible in a time-dependent manner, consistent with the induction pattern of the canonical IFN-I-induced gene ISG15 (Fig. 1D, Fig. S1A). Similar data were obtained across different human macrophage cell models, including PMA-differentiated THP1 cells and primary human macrophage (Fig. 1E, Fig. S1B–D). We further demonstrated that IFN-α similarly induced USP30-AS1 expression in U937 cells, a response that paralleled the induction of ISG15 (Fig. 1F and Fig. S1E). The IFN-I-induced expression of USP30-AS1 was completely abolished upon treatment with Ruxolitinib, a JAK1/2 inhibitor that blocks the downstream IFN-I signalling cascade (Fig. 1G). This result confirms that USP30-AS1 induction depends on the JAK-mediated IFN-I signalling pathway.

Fig. 1.

Type I IFN induces the expression of lncRNA USP30-AS1. A Volcano plot illustrating differentially expressed lncRNAs following IFN-β treatment. The x-axis shows the fold change of lncRNA expression (Log2). The y-axis shows P value (-Log10). Red dots indicate significantly upregulated lncRNAs. Blue dots indicate downregulated lncRNAs. The top 10 upregulated lncRNAs are highlighted in bold. B List of the top 10 upregulated lncRNAs (by fold change) after IFN-β treatment. C qPCR analysis of candidate lncRNAs in U937 cells stimulated with IFN-β (20 ng/mL) for 6 h. D–E qPCR analysis of USP30-AS1 in U937 D and human primary macrophage E stimulated with IFN-β (20 ng/mL) for the indicated times. F qPCR analysis of USP30-AS1 in U937 stimulated with IFN-α (100 ng/mL) for the indicated times. G qPCR analysis of USP30-AS1 in U937 cells stimulated with IFN-β (20 ng/mL) for 6 h in the presence or absence of Ruxolitinib (2 µM). H Agarose gel electrophoresis of 5′- and 3′-RACE products from U937 cells confirming the complete sequence of USP30-AS1. I Schematic representation of the EGFP-fusion assay used to assess the protein-coding potential of USP30-AS1. J Immunoblotting analysis of HEK293T cells transfected with EGFP or indicated EGFP-USP30-AS1 constructs. ND, indicates not detected. Data are presented as mean ± SEM of three independent experiments C–G. **P < 0.01, ***P < 0.001, calculated by two-tailed t-test C–G. U937 and THP1 cells were differentiated with PMA prior to treatment

To further characterize this IFN-I-induced transcript, we defined its full-length sequence. By performing 5’- and 3’-RACE assays, we determined the full-length sequence of USP30-AS1 as a 656-nt transcript (Fig. 1H, Fig. S1F). To assess its protein-coding potential, we fused the coding sequence of enhanced green fluorescent protein (EGFP) to the sequence of USP30-AS1 and expressed these constructs in cells (Fig. 1I). Western blot analysis revealed no obvious expression of the fusion constructs at the expected size, supporting USP30-AS1 as a non-coding RNA (Fig. 1J). Together, these data identify USP30-AS1 as an IFN-I-induced lncRNA.

USP30-AS1 is a cytoplasmic non-coding RNA functionally independent of USP30

In LncBook database transcript annotations, no formally annotated USP30-AS1 orthologous transcript has been reported in non-human species, indicating its potential role as a human-specific regulator of innate immunity. We aligned the genomic sequences spanning USP30-AS1 locus and observed measurable conservation among several primate species (Fig. 2A). Next, we performed a comparative analysis of the promoter region around the putative transcription start site (TSS; ±1 kb) across primates, scanning for interferon-stimulated response elements (ISREs). We detected ISRE-like motifs in multiple species, with several showing high sequence similarity and positional conservation relative to the human TSS (Fig. S2A). These data suggest that the interferon-responsive regulatory architecture at this locus is partially conserved. Thus, while a fully annotated USP30-AS1 ortholog may be human-specific, the regulatory potential for IFN-I-responsive expression at this genomic locus is conserved in primates. This pattern implies that species differences may arise from divergent transcriptional output or post-transcriptional RNA processing.

Fig. 2.

USP30-AS1 is a cytoplasmic non-coding RNA functionally independent of USP30. A Phylogenetic overview of sequence conservation at the USP30-AS1 genomic locus. Branch colors indicate the conservation category (Q90/Q75/Q50; grey, not conserved), as defined by the quantile-based conservation scoring in LncBook. Species lacking an annotated orthologous gene for USP30-AS1 are indicated as “no gene annotation”, consistent with current lncRNA catalogs. B Schematic representation of the CRISPR/Cas9-mediated deletion strategy targeting different regions of USP30-AS1. Three knockout designs were employed: Exon 1 deletion, Intron deletion, and combined deletion of Exon 1 and Intron. C qPCR analysis of USP30-AS1 in control and USP30-AS1 KO U937 cells stimulated with IFN-β (20 ng/mL) for indicated time. ND, not detected. D left, qPCR analysis of USP30 mRNA expression in U937 cells transfected with non-targeting control small interfering RNA (si-NC) or USP30-specific small interfering RNA (si-USP30). Right, qPCR analysis of ISG15 mRNA expression in U937 cells transfected with si-NC or si-USP30, followed by treated with IFN-β (20 ng/mL) or left untreated. E left, qPCR analysis of USP30 mRNA expression in control and USP30-AS1 KO (△Intron) U937 cells transfected with si-NC or si-USP30. Right, qPCR analysis of ISG15 mRNA expression in control and USP30-AS1 KO U937 cells transfected with si-NC or si-USP30, followed by treated with IFN-β (20 ng/mL) or left untreated. F U937 cells were treated with IFN-β (20 ng/mL) for indicated times. USP30-AS1 expression in cytoplasmic and nuclear fractions was assessed by RT-PCR. G RNA FISH showing USP30-AS1 localization (red) in U937 cells treated with or without IFN-β (20 ng/mL) for 6 h. Nuclei were counterstained with DAPI (blue). Scale bar: 10 μm. H Volcano plot of differentially expressed genes in USP30-AS1 KO (△Intron) versus control U937 cells. Significantly upregulated genes are shown in red, downregulated genes in blue, and non-significant changes in gray. I KEGG pathway enrichment analysis of downregulated genes in USP30-AS1 KO cells. Bubble size indicates the number of genes, and the color gradient reflects the statistical significance of the enrichment. U937 cells were differentiated with PMA prior to any treatment. Data are presented as mean ± SEM of three independent experiments C–E. NS, not significant, ***P < 0.001, calculated by two-tailed t-test D, E

As USP30-AS1 is transcribed antisense of USP30 gene, and exon 2 of USP30-AS1 partially overlaps with USP30, we next investigated whether USP30-AS1 functions through regulating USP30 expression. We generated USP30-AS1 knockout (KO) U937 cells via CRISPR/Cas9 by targeting its non-overlapping regions (Fig. 2B, C, Fig. S2B-E). Neither knockout nor overexpression of USP30-AS1 altered USP30 expression at mRNA or protein levels, even under IFN-β treatment (Fig. S2F-H). Then we performed USP30 knockdown in both WT and USP30-AS1 KO U937 cells and assessed IFN-β-induced ISG15 expressions. In WT cells, knockdown of USP30 did not affect IFN-β-induced ISG15 mRNA level (Fig. 2D). In contrast, knockout of USP30-AS1 significantly attenuated this induction (Fig. 2E). Importantly, additional knockdown of USP30 in the USP30-AS1 KO cells did not further alter ISG15 expression (Fig. 2E). Together, these data demonstrate that the regulation of IFN-β-induced gene expression by USP30-AS1 is independent of USP30.

Since the function of a lncRNA is often associated with its subcellular localization, we next determined its compartmentalization. Subcellular fractionation and RNA-FISH revealed that USP30-AS1 was predominantly enriched in the cytoplasm, and its abundance increased upon IFN-β treatment (Fig. 2F and G). This cytoplasmic localization suggested a potential role of USP30-AS1 in post-transcriptional gene regulation. In line with these observations, transcriptomic analysis of USP30-AS1 KO macrophages revealed remarkable alterations in gene expression profiles (Fig. 2H), with downregulated genes significantly enriched in innate immune signalling pathways, including cytokine-cytokine receptor interaction, NOD-like receptor signalling pathway and herpes simplex virus 1 (HSV-1) infection (Fig. 2I). These findings suggest that USP30-AS1 is a cytoplasmic lncRNA that regulates innate immune gene expression independently of USP30.

USP30-AS1 deletion impairs the expression of nucleic acid sensors

Given the severe dampening of innate immune signalling pathways observed in USP30-AS1 KO cells, we investigated whether USP30-AS1 regulates the expression of nucleic acid sensors, which are pivotal for initiating IFN-I responses. Differential gene expression analysis revealed significant downregulation of multiple nucleic acid sensors in USP30-AS1 KO cells (Fig. 3A and B). We validated these findings at both protein and mRNA levels for key sensors, including RIG-I (encoded by RIGI), Melanoma differentiation-associated protein 5 (MDA5, encoded by IFIH1), Protein kinase R (PKR, encoded by EIF2AK2), and cGAS (encoded by CGAS) (Fig. 3C and D). Re-expression of USP30-AS1 in KO cells rescued these effects, confirming its role in regulating the expression of these factors (Fig. 3C and D). To determine the mechanistic basis for this regulation, we assessed whether USP30-AS1 affects the mRNA stability or transcription of these sensors by performing the Actinomycin D (ActD) and Bromouridine (BrU) labeling assays, respectively (Fig. 3E). Results showed that USP30-AS1 deficiency markedly reduced sensor mRNA levels across the ActD chase time course, with no significant change in their nascent transcription (Fig. 3F and G). Thus, these results suggest that USP30-AS1 maintains the expression of key nucleic acid sensors primarily by enhancing their mRNA stability.

Fig. 3.

USP30-AS1 deletion impairs the expression of nucleic acid sensors. A Heatmap showing differentially expressed nucleic acid sensors in USP30-AS1 KO (△Intron) and control U937 cells. The color scale represents z-score, with red indicating higher expression and blue indicating lower expression. B Schematic of nucleic acid sensing pathways, illustrating recognition of exogenous or endogenous nucleic acids and subsequent immune activation. Illustration created using MedPeer. C Western blot analysis of key nucleic acid sensors in control, USP30-AS1 KO (△Intron), and USP30-AS1 rescue U937 cells. D qPCR analysis of RIGI, IFIH1, EIF2AK2, and CGAS mRNA expression in control, USP30-AS1 KO (△Intron), and rescue U937 cells. E Schematic of the ActD assay (up) and 5-BrU (down) chasing experiment in U937 cells. Cells were treated with ActD to inhibit transcription, and total RNA was collected at 0, 2, 4 and 8 h to assess mRNA decay. Cells were treated with 5-BrU for 3 h and then chased in fresh medium for 0, 6, and 12 h to evaluate nascent RNA synthesis. F ActD chase assay showing the time-course of RIGI, IFIH1, EIF2AK2, and CGAS mRNA levels in control and USP30-AS1 KO (△Intron) U937 cells following ActD (2 µg/mL) treatment for 0, 2, 4, and 8 h. Statistical analysis was performed using ANOVA. G qPCR analysis of 5-BrU-labeled mRNAs in control and USP30-AS1 KO (△Intron) U937 cells after treating cells with 5-BrU (200 µg/mL). Data are presented as mean ± SEM of three independent experiments (D, F, G). NS, not significant, *P < 0.05, **P < 0.01, ***P < 0.001. Statistical significance was determined using two-tailed t-test (D,G). U937 cells were differentiated with PMA prior to any treatment

USP30-AS1 stabilizes ARE-containing ISGs mRNAs

Since AREs in the 3′UTR are crucial regulators of mRNA stability [33–35], we hypothesized that USP30-AS1 might stabilize mRNAs through AREs. Bioinformatic analysis revealed that nearly half of the downregulated transcripts in USP30-AS1 KO cells contained AREs (Fig. 4A, Fig. S3A), suggesting a preferential stabilization of ARE-containing mRNAs by USP30-AS1. This preference was further supported by a strong correlation between the number of ARE sites within a transcript and its susceptibility to downregulation upon USP30-AS1 depletion (Fig. 4B). KEGG pathway analysis of the downregulated genes confirmed their enrichment in innate immune pathways, with a significant portion being ISGs (Fig. 4C, Fig. S3B). Similarly, ARE-containing ISGs were more strongly downregulated than non-ARE ISGs in knockout cells (Fig. 4D). Given that more than half of the ISGs contain AREs (Fig. 4E), we found that USP30-AS1 depletion led to the downregulation of the majority (63.6%) of ARE-containing ISGs (Fig. 4F). To experimentally validate this ARE-dependent stabilization in ISGs, we performed an EGFP reporter assay (Fig. 4G). We fused the EGFP-coding sequence to either the wild-type (WT) or ARE-deleted (△ARE) 3′UTRs of CCL7, CCL8, IL1A, and IL1B. Overexpression of USP30-AS1 significantly increased the mRNA expression of EGFP only when the WT 3′UTRs were present, indicating that the AREs are required for stabilization (Fig. 4H). Collectively, these results establish USP30-AS1 as a crucial post-transcriptional regulator that preferentially stabilizes ARE-containing ISGs transcripts, thereby potentiating innate immune signalling.

Fig. 4.

USP30-AS1 stabilizes ARE-containing ISGs mRNAs. A Scatter plot illustrating the fold changes of downregulated genes in USP30-AS1 KO U937 cells. Each dot represents an individual gene: gray dots denote genes without AREs in their 3′UTRs, whereas blue dots indicate genes containing at least one ARE. B ECDF analysis of gene expression fold changes in U937 cells (△Intron versus control). Genes were grouped according to the number of ARE sites present in U937 cells. Statistical significance was assessed using two-sided Wilcoxon rank-sum tests. C KEGG pathway enrichment analysis of down-regulated genes in USP30-AS1 KO (△Intron) U937 cells. The color gradient indicates the ISGs ratio, with darker shades indicating a higher ratio. The dotted vertical line indicates the average ISGs ratio across all genes. D ECDF analysis of ISGs expression fold changes in U937 cells (△Intron versus control) based on the presence of ARE motifs. The red curve represents ISGs containing at least one ARE, whereas the gray curve corresponds to ISGs lacking AREs. Statistical significance was assessed using two-sided Wilcoxon rank-sum tests. E Bar chart showing the proportion of genes containing AREs in ISGs gene sets in wild-type U937 cells. F Bar chart showing the proportion of genes containing AREs in downregulated ISGs gene sets in USP30-AS1 knockout (△Intron) U937 cells. G Schematic of the EGFP reporter assay used to assess mRNA stability. The EGFP coding sequence was fused to the 3′UTRs of selected ISGs. H qPCR analysis of EGFP mRNA expression in HEK293T cells with empty vector or USP30-AS1-expressing constructs. qPCR was performed to measure EGFP mRNA expression when fused to the WT (ARE) or ARE-deleted (△ARE) 3′UTRs of CCL7, CCL8, IL1A, and IL1B. Data are presented as mean ± SEM of three independent experiments H. NS, not significant, *P < 0.05, **P < 0.01, ***P < 0.001, calculated by two-tailed t-test H. U937 cells were differentiated with PMA prior to any treatment

USP30-AS1 is essential for amplifying IFN-I responses

We next evaluated whether USP30-AS1 could function as a broad-spectrum amplifier of the IFN-I response cascade. Our results showed that USP30-AS1 deficiency dampened the anti-viral IFN response against both the DNA virus, HSV-1, and the RNA virus, Vesicular Stomatitis Virus (VSV) (Fig. S4A and B). This effect was rescued by re-expression of USP30-AS1 (Fig. S4A and B). We also challenged the cells with stimuli modelling autoimmune-like conditions, including herring testis DNA (HT-DNA) for cytosolic DNA mimicking, 2′,3′-cyclic guanosine monophosphate–adenosine monophosphate (2′3′-cGAMP) for STING activation, and exogenous IFN-β. Consistently, cells lacking USP30-AS1 exhibited remarkably reduced induction of key ISGs and IFNB compared to control cells (Fig. 5A–C, Fig. S4C and D). Importantly, re-expression of USP30-AS1 restored the expression of ISG15, CXCL10, and IFIT1 mRNA following each stimulus (Fig. 5A–C). These data indicate that USP30-AS1 potently amplifies IFN-I signalling, and its dysregulation may contribute to pathological IFN-I signalling in autoimmune diseases.

Fig. 5.

USP30-AS1 is essential for amplifying IFN-I responses. A–C qPCR analysis of indicated ISGs mRNA expression in control, USP30-AS1 KO and rescued U937 cells following stimulation with HT-DNA (1 µg/mL) A, cGAMP (1 µg/mL) B, and IFN-β (20 ng/mL) C for 0, 6, and 12 h. Data are presented as mean ± SEM of three independent experiments. **P < 0.01, ***P < 0.001, calculated by two-tailed t-test. U937 cells were differentiated with PMA prior to any treatment

USP30-AS1 as a potential indicator of autoimmune disease activity and therapeutic target

To explore the clinical relevance of USP30-AS1 in IFN-I-driven autoimmune diseases, we analyzed its expression in multiple published datasets from independent cohorts of patients with SLE [36–38], RA [39–41] and DM [42, 43]. Compared to healthy controls, patients with autoimmune diseases exhibited significantly elevated expression of USP30-AS1 (Fig. 6A–C). Notably, its expression positively correlated with disease severity in SLE (Fig. 6D) and was higher in untreated RA patients than healthy controls or treated patients (Fig. 6E), suggesting its potential as a dynamic indicator of autoimmune disease activity. ISG15 expression showed a similar pattern as USP30-AS1, further supporting the link between USP30-AS1 expression and IFN-I pathway activation (Fig. S5).

Fig. 6.

USP30-AS1 as a potential indicator of autoimmune disease activity and therapeutic target. A Boxplots showing USP30-AS1 expression in SLE patients and healthy controls from PBMC-derived datasets (GSE65391, GSE72509 and GSE88884). B USP30-AS1 expression in RA patients and healthy controls in joint-derived datasets (GSE236924, GSE89408) and PBMC-derived datasets (GSE93777). C USP30-AS1 expression in DM patients and healthy controls from skin-derived datasets (GSE46239, GSE142807) and muscle-derived datasets (GSE143323). D USP30-AS1 expression in healthy controls and SLE patients categorized by disease severity (mild and moderate), using dataset GSE88884. E USP30-AS1 expression in RA patients classified by treatment status (untreated versus treated) and healthy controls (GSE89408). F Spearman correlation analysis between USP30-AS1 and CCL7, CCL8, IL1A and IL1B expression was performed in three independent bulk RNA-seq cohorts including both patients and healthy controls (SLE: GSE72509; RA: GSE89408; DM: GSE143323). Heatmap shows Spearman’s ρ values (color scale). Asterisks indicate statistical significance (**P < 0.01, ***P < 0.001). Data points correspond to individual samples, boxes indicate the interquartile range, and the horizontal line within each box denotes the median A–E. Whiskers extend to the minimum and maximum values (excluding outliers) A–E. p-values indicate statistical significance for differential expression between disease and control groups, with sample sizes (n) for each group displayed in the legends A–E. Statistical significance was determined using Mann–Whitney U or Student’s t-test, depending on data distribution A–E

Next, to evaluate the pathophysiological relevance of our proposed mechanism, we evaluated the correlation between USP30-AS1 expression and the transcriptional levels of CCL7, CCL8, IL1A, and IL1B in public RNA-seq datasets from patients with systemic SLE [38], RA [39], and DM [42]. Notably, USP30-AS1 levels showed consistent and significant positive correlations with the expression of these ISGs across all three diseases (Fig. 6F). This association supports our hypothesis that USP30-AS1 post-transcriptionally regulates key IFN-related transcripts via stabilizing mRNAs. Taken together, our findings establish USP30-AS1 not only as a clinically relevant indicator of IFN-I pathway activity and disease severity but also a promising therapeutic target in IFN-I-driven autoimmune diseases.

Discussion

IFN-I signalling comprises a crucial component of innate immune responses, and is tightly controlled by intricate regulatory strategies to maintain the balance between effective immune defence and prevention of autoimmune pathology [44–47]. While the transcriptional regulation of IFN signalling is well-documented, the post-transcriptional layer, especially lncRNA-mediated modulation, remains incompletely understood. In this study, we identify USP30-AS1, an IFN-I-inducible cytoplasmic lncRNA, as a broad-spectrum amplifier of IFN-I signalling through preferentially stabilizing nucleic acid sensors and ARE-containing ISGs mRNAs. This post-transcriptional amplification is critical for mounting effective antiviral responses, and may contribute to pathological IFN-I signalling in autoimmune diseases, thereby revealing a previously unrecognized regulatory layer within the IFN-I signalling network.

Our findings delineate a novel mechanism of lncRNA-mediated mRNA stabilization that enhances IFN-I responses. The preference for ARE-containing transcripts is particularly significant. AREs are well-established regulators of mRNA turnover, and play a critical role in rapid shutdown of transient inflammatory responses, such as IFN-I signalling [20, 48–52]. By antagonizing this decay pathway for a select set of immune transcripts, USP30-AS1 provides a potent means to amplify immune signalling. This mechanism establishes a positive feedback loop that ensures robust and timely immune activation, paralleling the function of certain protein-coding ISGs, such as the Interferon-induced protein with tetratricopeptide repeats (IFIT) family, which enhances antiviral signalling by stabilizing key components of the IFN regulatory network [53, 54]. This similarity highlights an emerging paradigm where non-coding RNAs act as critical nodes within immune regulatory networks. Our model is further supported by a recent study showing that USP30-AS1 exerts antiviral effects against Influenza A Virus infection [55], consistent with its role as a broad-spectrum amplifier of antiviral immunity. Interestingly, although USP30-AS1 has been implicated in cancer progression via microRNA sponging [56, 57], our study reveals a distinct role in ISGs mRNA stabilization, suggesting that its function may be context-dependent across different pathophysiological conditions.

The capacity of USP30-AS1 to enhance mRNA stability promotes potent antiviral immunity during infection. However, under conditions of chronic immune activation, such as in autoimmunity, the persistent expression of USP30-AS1 may sustain a pathological positive feedback loop by continuously stabilizing the mRNAs of nucleic acid sensors and ISGs. This molecular model is directly supported by our observation of elevated USP30-AS1 expression in patients with IFN-driven autoimmune diseases. Therefore, USP30-AS1 emerges not only as a novel biomarker for IFN pathway activation but also as a compelling therapeutic target. Its inhibition offers the potential to dampen pathological autoimmunity driven by IFN-I without completely ablating the entire pathway, thereby preserving essential antiviral immunity.

Building on this therapeutic potential, we briefly consider translational strategies. The cytoplasmic localization and defined sequence of USP30-AS1 make it amenable to targeting by oligonucleotide-based therapeutics, such as antisense oligonucleotides (ASOs) [58] or small interfering RNAs (siRNAs) [59]. These agents can be designed for sequence-specific silencing, potentially degrading the transcript or blocking its function. However, key challenges for clinical translation remain, including achieving efficient [60] and specific delivery to relevant immune cells in affected tissues [61], minimizing off-target effects [59], and ensuring long-term tolerability [62]. Advances in chemical modifications and delivery platforms are actively addressing these hurdles, bringing RNA-targeted therapies closer to clinical reality for autoimmune diseases [63, 64].

Notably, USP30-AS1 appears to be a human-specific lncRNA, with no identifiable homologs in other species. However, our evolutionary analysis reveals that while the full-length transcript may be unique to humans, the genomic locus and the ISREs in its promoter are conserved in primates. This suggests that the regulatory potential for IFN-I induction at this locus predates the emergence of the human transcript. Therefore, USP30-AS1 may represent a recently evolved layer of post-transcriptional regulation that fine-tunes the IFN-I response specifically in humans, with species differences likely arising from divergent transcriptional or RNA processing outputs. Thus, the identification of USP30-AS1 underscores the importance of studying human-specific regulatory elements to fully understand the complexity of human autoimmune diseases characterized by type I interferonopathies. Furthermore, its human-specific nature enhances its potential as a therapeutic target, as targeting USP30-AS1 may mitigate pathological IFN signalling in patients while minimizing off-target effects due to sequence divergence.

Conclusions

In summary, we demonstrate that IFN-I induces the expression of USP30-AS1, which in turn amplifies the IFN response by post-transcriptionally stabilizing a network of key immune transcripts, including nucleic acid sensors and ISGs. This discovery provides a mechanistic basis for dysregulated IFN signalling in autoimmunity and opens new avenues for developing therapeutic strategies. A key future direction will be to determine the precise molecular mechanism by which USP30-AS1 recognizes and stabilizes its target mRNAs. Further structural and interactome analyses will be essential to elucidate whether it directly binds AREs or recruits specific RNA-binding proteins. Collectively, our findings unveil USP30-AS1 as a central regulator of immune homeostasis and a promising new target for the treatment of IFN-driven diseases.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

ActD

Actinomycin D

ARE

AU-rich element

BrU

Bromouridine

cGAS

cyclic GMP-AMP synthase

DM

Dermatomyositis

FBS

Fetal bovine serum

HSV-1

Herpes simplex virus 1

HT-DNA

Herring testis DNA

EGFP

Enhanced green fluorescent protein

GEO

Gene Expression Omnibus

IFIT

Interferon-induced protein with tetratricopeptide repeats

IFN

Interferon

IFN-I

Type I interferons

lncRNA

long non-coding RNA

IRFs

Interferon regulatory factors

ISGs

Interferon-stimulated genes

JAK-STAT

Janus kinase-signal transducer and activator of transcription

KO

Knockout

MDA5

Melanoma differentiation-associated protein 5

MOI

Multiplicity of infection

PBMCs

Peripheral blood mononuclear cells

RA

Rheumatoid arthritis

RIG-I

Retinoic acid-inducible gene I

PKR

Protein kinase R

PMA

Phorbol 12-myristate 13-acetate

qPCR

quantitative PCR

sgRNAs

single-guide RNAs

SLE

Systemic lupus erythematosus

SRA

Sequence Read Archive

VSV

Vesicular Stomatitis Virus

2′3′-cGAMP

2′,3′-cyclic guanosine monophosphate-adenosine monophosphate

Author contributions

HC, LC and YY supervised the project. LC, HC, YY, WX and JP designed the experiments. JP, M-YF, SJ and J-QW performed the bioinformatics analysis. YY, MZ and P-PZ performed cell experiments and qPCR analysis. YY, YL and X-PY performed immunoblotting and FISH imaging. TX and MZ constructed plasmids and performed the ELISA. WX and Z-HS performed virus infection experiments and RACE. YY, WX and JP performed 5-BrU chasing and ActD assay. JP and MZ performed PBMCs isolation. TL, HC, YY, WX, M-YF and JP analyzed the data and wrote the manuscript. All the authors read and approved the final manuscript.

Funding

This work is supported by grants from China National Natural Science Foundation (No.82130052 to TL. No. 82302491 to MZ.).

Data availability

The raw and processed data used in this study were obtained from a publicly available dataset (PRJEB21332), several datasets from the NCBI Gene Expression Omnibus (GEO; accession numbers: GSE65391, GSE72509, GSE88884, GSE236924, GSE89408, GSE93777, GSE46239, GSE142807, and GSE143323), and an in-house dataset comparing control U937 cells to USP30-AS1 knockout cells, which has been deposited in the NCBI Sequence Read Archive (SRA) under the accession number PRJNA1251834.

Declarations

Ethics approval and consent to participate

All procedures of PBMCs isolation were approved by the Ethics Committee of Jiaxing Central Blood Station.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.