The interferon stimulated gene-encoded protein HELZ2 inhibits human LINE-1 retrotransposition and LINE-1 RNA-mediated type I interferon induction

Abstract

Some interferon stimulated genes (ISGs) encode proteins that inhibit LINE-1 (L1) retrotransposition. Here, we use immunoprecipitation followed by liquid chromatography-tandem mass spectrometry to identify proteins that associate with the L1 ORF1-encoded protein (ORF1p) in ribonucleoprotein particles. Three ISG proteins that interact with ORF1p inhibit retrotransposition: HECT and RLD domain containing E3 ubiquitin-protein ligase 5 (HERC5); 2'-5'-oligoadenylate synthetase-like (OASL); and helicase with zinc finger 2 (HELZ2). HERC5 destabilizes ORF1p, but does not affect its cellular localization. OASL impairs ORF1p cytoplasmic foci formation. HELZ2 recognizes sequences and/or structures within the L1 5'UTR to reduce L1 RNA, ORF1p, and ORF1p cytoplasmic foci levels. Overexpression of WT or reverse transcriptase-deficient L1s lead to a modest induction of IFN-α expression, which is abrogated upon HELZ2 overexpression. Notably, IFN-α expression is enhanced upon overexpression of an ORF1p RNA binding mutant, suggesting ORF1p binding might protect L1 RNA from "triggering" IFN-α induction. Thus, ISG proteins can inhibit retrotransposition by different mechanisms.

Fig. 1. Identification of an ORF1p RNA binding mutant critical for L1 retrotransposition and ORF1p cytoplasmic foci formation.

a Schematic of a full-length RC-L1 (L1.3: Genbank Accession #L19088). ORF1p functional domains are noted below the schematic and include the coiled-coil domain, the RNA recognition motif (RRM), and carboxyl-terminal domain (CTD). Green arrowhead, position of the in-frame FLAG epitope tag. Open triangle, relative position of a triple mutant (R206A/R210A/R211A) in the RRM domain. b WT ORF1p and the ORF1p-FLAG R206A/R210A/R211A mutant are stably expressed in HeLa-JVM cells. Western blot with an anti-FLAG antibody. A construct lacking the FLAG epitope tag (pJM101/L1.3 [no FLAG]) served as a negative control. GAPDH served as a sample processing control. c Schematics of the retrotransposition indicator cassettes used in this study. A retrotransposition indicator cassette (REP) was inserted into the 3′UTR of an L1 in the opposite orientation relative to sense strand L1 transcription. The REP gene contains its own promoter (upside down arrow) and polyadenylation signal (open lollipop). The REP gene is interrupted by intron in the same orientation relative to sense strand L1 transcription. This arrangement ensures that REP expression only will occur if the sense strand L1 transcript is spliced and successfully integrated into genomic DNA by retrotransposition (bottom schematic, open triangles, target site duplications that typically are generated upon L1 retrotransposition). Three retrotransposition indicator cassettes are shown at the right of the figure: mneoI, which confers resistance to G418; mblastI, which confers resistance to blasticidin; and mEGFPI, which leads to enhanced green fluorescent protein (EGFP) expression. d Results of a representative mneoI-based retrotransposition assay. HeLa-JVM cells were co-transfected with phrGFP-C (transfection control) and either pJM101/L1.3FLAG (WT) or pALAF008 (M8 [RBM]). X-axis, L1 construct names, and representative retrotransposition assay results. Y-axis, relative retrotransposition efficiency; the number of G418 resistant (retrotransposition-positive) foci was normalized to the transfection efficiency (i.e., the percentage of hrGFP-positive cells). Pairwise comparison relative to the WT control: p = 2.1 × 10⁻¹²***. e The ORF1p-FLAG R206A/R210A/R211A mutant (M8 [RBM]) reduces the number of ORF1p cytoplasmic foci. Representative immunofluorescence microscopy images of U-2 OS cells expressing either WT ORF1p-FLAG (pJM101/L1.3FLAG) or ORF1p-FLAG R206A/R210A/R211A mutant (pALAF008 [M8 (RBM)]). The U-2 OS cells also expressed a doxycycline-inducible (Tet-On) mCherry-G3BP1 protein. White scale bars, 5 µm. f Quantification of immunofluorescence assays in U-2 OS cells. X-axis, L1 construct names. Y-axis, percentage of transfected cells containing ORF1p cytoplasmic foci. The number (n) inside the green bars indicates the number of individual cells counted in the assay. Pairwise comparisons relative to the WT control: p = 7.5 × 10⁻¹¹***. g RNA-immunoprecipitation (RNA-IP) reveals an L1 RNA binding defect in the ORF1p-FLAG R206A/R210A/R211A mutant (M8 [RBM]). HeLa-JVM cells were transfected with either pJM101/L1.3 (no FLAG), WT ORF1p-FLAG (pJM101/L1.3FLAG), or the ORF1p-FLAG R206A/R210A/R211A mutant (pALAF008 [M8 (RBM)]). An anti-FLAG antibody was used to immunoprecipitate ORF1p-FLAG; reverse transcription-quantitative PCR (RT-qPCR) using a primer set (L1 [SV40]) that amplifies RNAs derived from the transfected L1 plasmid was used to quantify L1 RNA. X-axis, constructs name. Y-axis, the enrichment of L1 RNA levels between the IP and input fractions. Blue rectangles, relative levels of control GAPDH RNA (primer set: GAPDH). Gray rectangles, relative levels of L1 RNA. In panels (d), (f), and (g), values represent the mean ± the standard error of the mean (SEM) of three independent biological replicates. The p-values were calculated using a one-way ANOVA followed by Bonferroni–Holm post-hoc tests; *** p < 0.001.

Fig. 2. The proteins encoded by interferon-responsive genes are enriched in WT ORF1p-FLAG, but not ORF1p-FLAG (M8 [RBM]) mutant complexes.

a Experimental rationale for identifying host factors enriched in WT ORF1p-FLAG vs. ORF1p-FLAG (M8 [RBM]) immunoprecipitation reactions. Hypothetical diagrams of the proteins associating with WT and M8 (RBM) mutant RNP particles. Green circles, ORF1p-FLAG. Blue Oval, ORF2p. Red circle, purple squared oval, and green rectangle, host factors that might associate with ORF1p-FLAG and/or L1 RNPs. b The ORF1p (M8 [RBM]) mutant does not efficiently interact with Poly(A) Binding Protein Cytoplasmic 1 (PABPC1). HeLa-JVM cells were transfected with either pJM101/L1.3 (no FLAG), pJM101/L1.3FLAG (WT ORF1p-FLAG), or pALAF008 (ORF1p-FLAG [M8 [RBM]] mutant). An anti-FLAG antibody was used to immunoprecipitate ORF1p-FLAG. Western blots detected ORF1p (anti-FLAG), PABPC1 (anti-PABC1), and GAPDH (anti-GAPDH) in the input and IP fractions. GAPDH served as a sample processing control for the input fractions and a negative control in the IP experiments. c Separation of proteins associated with the WT and mutant ORF1p-FLAG proteins. The WT and M8 (RBM) mutant ORF1p-FLAG IP complexes were separated by SDS-PAGE using a 4-15% gradient gel and silver staining visualized the proteins. Protein size standards (kDa) are shown at the left of the gel. Black arrowhead, the expected molecular weight of ORF1p-FLAG. d Gene Ontology (GO) analysis identifies cellular proteins enriched in IP WT ORF1p-FLAG vs. the mutant ORF1p-FLAG complex. Cellular proteins present in the WT ORF1p and (M8 [RBM])-FLAG mutant IP complexes were identified using LC-MS/MS. Proteins having a >0.5 log₂ abundance ratio at any p-value in the WT ORF1p vs. M8 [RBM]) complexes were subjected to DAVID gene ontology analysis. Listed are the “functional annotation of UniProt Keyword GO biological process” terms. X-axis, protein count, the number of proteins identified by mass spectrometry that are included in each respective GO term. Y-axis, GO term. Circle size, −log₁₀FDR. Larger circles indicate higher confidence based on the FDR for each GO term. Red lettering, viral related GO terms. e GSEA preranked analysis identifies interferon-related gene sets enriched upon WT ORF1p-FLAG immunoprecipitation. Gene Set Enrichment Analysis (GSEA) of log₂ abundance ratio of cellular proteins immunoprecipitated in the WT ORF1p-FLAG vs. (M8 [RBM])-FLAG IP complexes was performed using hallmark gene sets in the Molecular Signatures Database (MSigDB: https://www.gsea-msigdb.org/gsea/msigdb/), followed by Leading Edge Analysis to determine gene set enrichment scores. The top six hallmark gene sets with the highest normalized enrichment score (NES) are sorted in descending values. X-axis, NES. Y-axis, hallmark gene sets. f The expression of engineered L1s modestly up-regulates IFN-α expression. HEK293T were transfected with either pCEP4 (an empty vector control), pJM101/L1.3FLAG (WT), pJM105/L1.3 (RT-), or pALAF008 (M8 [RBM]). RT-qPCR was used to quantify IFN-α (primer set: IFN-α, which amplifies IFN-α1 and IFN-α13) and L1 expression (primer set: mneoI [Alu or L1]) ~96 h post-transfection. IFN-α and L1 expression levels were normalized using β-actin (ACTB) as a control (primer set: Beta-actin). X-axis, name of constructs. Control, pCEP4. Y-axis, relative RNA expression levels normalized to the pCEP4 empty vector control. Red bars, normalized IFN-α expression levels. Gray bars, normalized L1 expression levels. Values from three independent biological replicates ± SEM are depicted in the graph. The p-values were calculated using a one-way ANOVA followed by Bonferroni-Holm post-hoc tests: pairwise comparisons of IFN-α relative to the pCEP4 control, p = 0.00028*** (WT); 0.00011*** (RT-); 3.14 × 10⁻⁶*** (M8 [RBM]). Pairwise comparisons of IFN-α: WT vs. RT-, p = 0.21^ns; WT vs. M8 (RBM), p = 0.00036***. Pairwise comparisons of L1 relative to WT, p = 0.87^ns (RT-), p = 0.10^ns (M8 [RBM]); ns: not significant; *** p < 0.001. For (d) and (e), the Source data are provided as a Source Data file.

Fig. 3. A network of ISGs that potentially affect WT L1 retrotransposition.

a STRING database analysis of WT ORF1p-FLAG associated proteins. Proteins with >0.5 log₂ abundance ratios in the WT ORF1p-FLAG vs. M8/RBM-FLAG complexes that exhibited a >5-fold change in expression upon induction by type I, II, and III IFNs were subjected to STRING analysis. Red and blue spheres, proteins annotated in UniProt as antiviral defense and innate immunity proteins, respectively. Thickness of the inter-connecting lines, the strength of association based on the number of independent channels supporting the putative interactions. The black dotted box indicates a group of proteins that closely associate (i.e., a putative ISG network); the majority are annotated as antiviral defense proteins in UniProt. The proteins in the box are listed at the top of the figure (follow dotted arrow) based upon whether they have been reported to regulate L1 retrotransposition (left, Reported ISG), or not (right, Unreported ISG). The top black dotted hexagon shows the proteins used for further analyses. b Volcano plot of WT ORF1p-FLAG vs. M8 (RBM) ORF1p-FLAG label-free quantitative mass spectrometry analysis. Data from the WT ORF1p-FLAG vs. M8 (RBM) ORF1p-FLAG IP-LC/MS experiments were analyzed to identify cellular proteins that preferentially associate with ORF1p-FLAG. X-axis, log₂ abundance ratios of WT ORF1p-FLAG vs. M8 (RBM) ORF1p-FLAG. Y-axis, −log₁₀ p-values of the abundance ratios. The ORF1p amounts obtained in the WT and M8 (RBM), which is indicated in the middle of the plot (0 abundance ratio) were used to normalize protein abundance ratios. Cutoffs of >0.5 log₂ abundance ratio and <0.05 p-values are shown as references for the enrichment of proteins in the WT ORF1p-FLAG fraction (red rectangle) or M8 (RBM) ORF1p-FLAG fraction (green rectangle). Blue dotted lines, proteins enriched in WT ORF1p-FLAG complexes (i.e., HELZ2, HERC5, DDX60L, IFIT1, and OASL). The p-values of the abundance ratios were calculated using Tukey Honestly Significant Difference test (post hoc) after an analysis of variance (ANOVA) test. c Independent confirmation that ISG proteins interact with WT ORF1p-FLAG. HEK293T cells were co-transfected with either pJM101/L1.3 (no tag) or pJM101/L1.3FLAG (ORF1p-FLAG) and the following individual carboxyl-terminal 3xMYC epitope-tagged ISG expression vectors: pALAF015 (HELZ2), pALAF016 (IFIT1), pALAF021 (DDX60L), pALAF022 (OASL), or pALAF023 (HERC5). The input and anti-FLAG IP reactions were analyzed by western blotting using an anti-FLAG (to detect ORF1p-FLAG) or an anti-MYC (to detect ISG proteins) antibody. For (a) and (c), the Source data are provided as a Source Data file.

Fig. 4. A subset of ISG proteins affect steady state L1 RNA levels, ORF1p cytoplasmic foci formation, and/or L1 retrotransposition.

a Overexpression of HERC5, HELZ2, or OASL inhibit L1 retrotransposition. HeLa-JVM cells were co-transfected with pJJ101/L1.3, which contains the mblastI retrotransposition indicator cassette, and either pCMV-3Tag-8-Barr or one of the following carboxyl-terminal 3xMYC epitope-tagged ISG protein expression plasmids: pALAF015 (HELZ2), pALAF016 (IFIT1), pALAF021 (DDX60L), pALAF022 (OASL), pALAF023 (HERC5), or pALAF024 (MOV10) according to the timeline shown at the top of the figure. A blasticidin expression vector (pcDNA6) was co-transfected into cells with either pCMV-3Tag-8-Barr or an individual ISG protein expression plasmid (see plates labeled control [pcDNA6]) to assess cell viability. The retrotransposition efficiencies then were normalized to the respective toxicity control. X-axis, name of the control (pCMV-3Tag-8-Barr) or ISG protein expression plasmid. Y-axis, relative retrotransposition efficiency normalized to the pJJ101/L1.3 + pCMV-3Tag-8-Barr co-transfected control. Representative results of the retrotransposition (see plates labeled mblastI [pJJ101]) and toxicity (see plates labeled control [pcDNA6]) assays are shown below the graph. Pairwise comparisons relative to the pJJ101/L1.3 + pCMV-3Tag-8-Barr control: p = 8.0 × 10⁻⁵** (HERC5); 4.4 × 10⁻⁶*** (HELZ2); 4.9 × 10⁻⁶*** (OASL); 0.011* (IFIT1); 0.12 ^ns (DDX60L); and 1.7 × 10⁻⁷*** (MOV10). MOV10 served as a positive control in the assay. b Expression of the ISG proteins in HeLa-JVM cells. HeLa-JVM cells were co-transfected with pTMF3, which expresses a version of ORF1p containing a T7 gene 10 carboxyl epitope tag (ORF1p-T7), and either a pCMV-3Tag-8-Barr (control) or the individual ISG-expressing plasmids used in panel (a). Whole cell extracts were subjected to western blot analysis 48 h post-transfection. ISG proteins were detected using an anti-MYC antibody. ORF1p was detected using an anti-T7 antibody. GAPDH served as a sample processing control. The relative band intensities of ORF1p-T7 are indicated under the ORF1p-T7 blot; they were calculated using ImageJ software and normalized to the respective GAPDH bands. c HELZ2 expression leads to a reduction in the steady state level of L1 RNA. HeLa-JVM cells were transfected as in panel (b). L1 RNA levels were determined by performing RT-qPCR using a primer set specific to RNAs derived from the transfected L1 (primer set: L1 [SV40]) and then were normalized to ACTB RNA levels (primer set: Beta-actin). X-axis, name of the constructs. Y-axis, relative level of L1 RNA normalized to the ORF1-T7 + pCMV-3Tag-8-Barr control. Pairwise comparisons relative to the control: p = 0.032^* (HERC5); 1.7 × 10⁻⁵*** (HELZ2): 0.14^ns (OASL); 0.29^ns (IFIT1); 0.20^ns (DDX60L); and 4.4 × 10^−4*** (MOV10). d Differential effects of ISG proteins on ORF1p-FLAG cytoplasmic foci formation. HeLa-JVM cells were co-transfected with pJM101/L1.3FLAG (WT ORF1p-FLAG) and either a pCEP4 empty vector (control) or one of the following carboxyl-terminal 3xMYC epitope-tagged ISG protein expression plasmids: pALAF015 (HELZ2); pALAF027 (HELZ2 WA1&2); pALAF022 (OASL); pALAF023 (HERC5); or pALAF024 (MOV10) to visualize WT ORF1p-FLAG cytoplasmic foci and co-localization between WT ORF1p-FLAG and the candidate ISG protein. Shown are representative fluorescent microscopy images. White scale bars, 20 µm. e Quantification of L1 cytoplasmic foci formation. X-axis, name of the constructs co-transfected with pJM101/L1.3FLAG (WT ORF1p-FLAG); control, pCEP4. Y-axis, percentage of transfected cells with visible ORF1p signal exhibiting ORF1p-FLAG cytoplasmic foci. The numbers (n) within the green rectangles indicate the number of analyzed cells in each experiment. Pairwise comparisons relative to the pJM101/L1.3FLAG (WT ORF1p-FLAG) + pCEP4 control: p = 8.6 × 10⁻⁴*** (HERC5); 1.2 × 10⁻⁷*** (HELZ2); 0.098^ns (HELZ2 WA1&2); 1.0 × 10⁻¹⁰*** (OASL); 2.7 × 10⁻⁹*** (MOV10). Values represent the mean ± SEM from three (in panels [a] and [e]) or six (in panel [c]) independent biological replicates. The p-values were calculated using one-way ANOVA followed by Bonferroni–Holm post-hoc tests; ns: not significant; * p < 0.05; ** p < 0.01; *** p < 0.001.

Fig. 5. The HELZ2 helicase activity is critical for L1 inhibition.

a Schematic of the HELZ2 protein domains. HELZ2 contains two putative helicase domains (helicase 1 and helicase 2), which surround a putative RNB exonuclease domain. Open triangles, positions of missense mutations in conserved amino acids within the Walker A (WA) boxes in the helicase 1 and helicase 2 domains (K550A [WA1] and K2180A [WA2], respectively). Red arrowheads, relative positions of the 3xMYC carboxyl-terminal epitope tag in the HELZ2 expression constructs. b The effect of mutations in the Walker A box on L1 retrotransposition in HEK293T cells. HEK293T cells were co-transfected with cepB-gfp-L1.3, which contains an mEGFPI retrotransposition indicator cassette, and either pCMV-3Tag-8-Barr (control), pALAF015 (WT HELZ2), or one of the following HELZ2 expression plasmids that contain a mutation(s) in the Walker A box: pALAF025 (WA1); pALAF026 (WA2); or pALAF027 (WA1&2). Cells co-transfected with cepB-gfp-L1.3RT(-) intronless and either pCMV-3Tag-8-Barr, pALAF015 (WT HELZ2), or a mutant HELZ2 plasmid served as transfection normalization and toxicity controls. Top, timeline of the assay for experiments shown in panels (b) and (c). X-axis, name of HELZ2 expression constructs co-transfected into cells with cepB-gfp-L1.3; control, pCMV-3Tag-8-Barr. Y-axis, relative retrotransposition efficiency normalized to the cepB-gfp-L1.3 + pCMV-3Tag-8-Barr control. Pairwise comparisons relative to the cepB-gfp-L1.3 (mEGFPI) + pCMV-3Tag-8-Barr control: p = 2.5 × 10⁻¹¹*** (WT HELZ2); 3.5 × 10⁻¹¹*** (WA1); 1.7 × 10⁻⁶*** (WA2); and 0.070^ns (WA1&2). c The effect of mutations in the Walker A box on L1 retrotransposition in HeLa-JVM cells. HeLa-JVM cells were co-transfected as in panel (b). Retrotransposition efficiencies were calculated as described in panel (b). Pairwise comparisons relative to the control: p = 0.00087*** (WT HELZ2); 0.26^ns (WA1); 0.32^ns (WA2); and 0.32^ns (WA1&2). d Mutations in the HELZ2 helicase domains reduce the ability to inhibit L1 ORF1p and RNA. HeLa-JVM cells were transfected with pTMF3 (L1 ORF1p-T7), denoted by + symbol, and either pCMV-3Tag-8-Barr (control), pALAF015 (WT HELZ2), or an individual HELZ2 expression plasmid containing a mutation(s) in the Walker A box: pALAF025 (WA1), pALAF026 (WA2), or pALAF027 (WA1&2). Top: L1 RNA levels were determined by RT-qPCR using primers directed against sequences in the transfected L1 RNA (primer set: L1 [SV40]) and then were normalized to ACTB RNA levels (primer set: Beta-actin). Pairwise comparisons relative to the pTMF3 (L1 ORF1p-T7) + pCMV-3Tag-8-Barr control: p = 9.5 × 10−⁹*** (WT); 1.9 × 10⁻⁸*** (WA1); 7.3 × 10⁻⁷*** (WA2); and 1.5 × 10⁻⁶*** (WA1&2). Pairwise comparisons relative to the pTMF3 (L1 ORF1p-T7) + WT HELZ2: p = 0.56^ns (WA1); 5.9 × 10⁻⁴*** (WA2); 1.9 × 10⁻⁴*** (WA1&2). Bottom: western blot image displaying ORF1p-T7 bands. HELZ2 expression was detected using an anti-MYC antibody. ORF1p was detected using an anti-T7 antibody. Pan-actin served as a sample processing control. e Small-interfering RNA (siRNA)-mediated knockdown of endogenous HELZ2 increases L1 retrotransposition. Top, timeline of the assay conducted in HeLa-JVM cells. Cells were transfected with a non-targeting siRNA control (siCtrl), siRNA targeting HELZ2 (siHELZ2), or siRNA targeting MOV10 (siMOV10). Middle left panel, HELZ2 RNA levels in siRNA treated cells. Middle right panel, MOV10 RNA levels in siRNA treated cells. X-axes, name of the siRNA. HELZ2 and MOV10 RNA levels were determined using RT-qPCR (primer sets: HELZ2 and MOV10, respectively) and then were normalized to ACTB RNA levels (primer set: Beta-actin). Y-axes, relative HELZ2 or MOV10 RNA levels normalized to the siCtrl. A two-tailed, unpaired Student’s t-test was used to calculate the p-values relative to the siRNA control: p = 3.1 × 10⁻⁵*** (siHELZ2); and 5.2 × 10⁻⁵*** (siMOV10). Bottom panel, HeLa-JVM cells were transfected with either siCtrl, siHELZ2, or siMOV10, followed by transfection with either cepB-gfp-L1.3 or cepB-gfp-L1.3RT(-) intronless, which was used to normalize transfection efficiencies. X-axis, name of the siRNA. Y-axis, relative retrotransposition efficiency. Pairwise comparisons relative to the non-targeting siRNA control: p = 2.9 × 10⁻⁴*** (siHELZ2); and 2.0 × 10⁻⁷*** (siMOV10). All the reported values represent the mean ± SEM from three independent biological replicates. The p-values, except for the RT-qPCR experiment shown in panel (e), were calculated using a one-way ANOVA followed by a Bonferroni–Holm post hoc tests. ns: not significant; * p < 0.05; *** p < 0.001.

Fig. 6. HELZ2 destabilizes L1 RNA through recognition of the L1 5′UTR sequence, leading to attenuation of L1-mediated IFN-α induction.

a The association between ORF1p and HELZ2 is RNA-dependent. HEK293T cells were co-transfected with pALAF015 (HELZ2-3xMYC) and either pJM101/L1.3FLAG (WT ORF1p-FLAG) or pJM101/L1.3 (no tag). The input and anti-FLAG IP fractions were analyzed by western blot using an anti-FLAG antibody to detect ORF1p-FLAG or an anti-MYC antibody to detect HELZ2-3xMYC. Shown are short (top blots) and longer (bottom blots) chemiluminescence western blot exposures. b HELZ2 expression reduces steady state levels of L1 RNA and ORF1p independent of ORF1p RNA binding. HeLa-JVM cells were co-transfected with pJM101/L1.3FLAG (WT ORF1p-FLAG) or the pALAF008 ORF1p-FLAG (M8 [RBM]) mutant expression plasmid and either pCEP4 (control) or pALAF015 (HELZ2). Top: L1 RNA amounts were determined by RT-qPCR (primer set: L1 [SV40]) and then were normalized to ACTB RNA levels (primer set: Beta-actin). The L1 RNA values were normalized to the WT L1 or ORF1p-FLAG (M8 [RBM]) + pCEP4 control transfections. Pairwise comparisons (in parentheses) relative to the (WT L1 + control) are shown: p = 7.1 × 10⁻⁷*** (WT L1 + HELZ2); 0.090^ns (M8 [RBM] + control); 6.7 × 10⁻⁷*** (M8 [RBM] + HELZ2). Pairwise comparisons of (WT L1 + HELZ2) vs. (M8 [RBM] + HELZ2), p = 0.92^ns. Bottom: ORF1p-FLAG and HELZ2 protein levels were detected by western blot using anti-MYC and anti-FLAG antibodies, respectively. GAPDH served as a sample processing control. c HELZ2 expression inhibits Alu retrotransposition. HeLa-HA cells were co-transfected with pTMO2F3_Alu (which expresses an Alu element marked with neo-based retrotransposition indicator cassette and monocistronic version of L1 ORF2p [see Methods]), pTMO2F3D145AD702A_Alu (which expresses an Alu element marked with neo-based retrotransposition indicator cassette and an EN-/RT- mutant version of L1 ORF2 [see Methods]), or phrGFP-C (a transfection normalization control) and either pCMV-3Tag-8-Barr (control), pALAF015 (WT HELZ2), or pALAF024 (WT MOV10). X-axis, name of constructs. Y-axis, the percentage of G418-resistant foci, indicative of Alu retrotransposition, relative to the pTMO2F3_Alu + pCMV-3Tag-8-Barr control (see “Methods” for more detail). Representative images of G418-resistant foci are shown below the graph. Pairwise comparisons relative to the pTMO2F3_Alu + pCMV-3Tag-8-Barr control: p = 7.8 × 10⁻⁵*** (HELZ2); 1.8 × 10⁻⁷*** (MOV10); and 1.6 × 10⁻⁷*** (EN-/RT-). d HELZ2 expression leads to a reduction in monocistronic ORF2 L1 RNA and ORF2p levels. HeLa-HA cells were co-transfected with pTMO2H3_Alu (ORF2p-3xHA and Alu) and either pCMV-3Tag-8-Barr (control) or pALAF015 (HELZ2). Top: ORF2 (gray bars) and Alu RNA (blue bars) levels were determined using RT-qPCR (primer sets: L1 [SV40] and mneoI [Alu or L1], respectively) and normalized to ACTB RNA levels (primer set: Beta-actin). X-axis, co-transfected constructs name. Y-axis, relative RNA level normalized to the pTMO2H3_Alu (ORF2p-3xHA and Alu) + pCMV-3Tag-8-Barr control. L1 ORF2 RNA pairwise comparison (ORF2/Alu + control vs. ORF2/Alu + HELZ2), p = 7.2 × 10⁻⁸***. Alu RNA pairwise comparison (ORF2/Alu + control vs. ORF2/Alu + HELZ2), p = 0.018*. Bottom: Western blotting using an anti-HA antibody was used to detect ORF2p. GAPDH served as a sample processing control. e The L1 5′UTR is required for HELZ2-mediated reduction of L1 RNA levels. HeLa-JVM cells were co-transfected with L1 (WT), L1 (∆5′UTR), or Fluc (a firefly luciferase gene flanked by the L1 5′ and 3′UTRs) and either pCMV-3Tag-8-barr (control) or pALAF015 (HELZ2). Schematics of the constructs are above the bar charts. RNA levels were determined by RT-qPCR using the following primer sets: L1 (SV40) (for L1 WT and L1[∆5′UTR]) or Luciferase (for Fluc) and then were normalized to GAPDH RNA levels (primer set: GAPDH). X-axis, name of respective constructs co-transfected with pCMV-3Tag-8-Barr (control) or pALAF015 (HELZ2); Y-axis, the relative amount of L1 or Fluc-based RNA relative to the relevant pairwise control (e.g., the L1 expression plasmid + pCMV-3Tag-8-Barr or the Fluc-based plasmid + pCMV-3Tag-8-Barr). Two-tailed, unpaired Student’s t-tests: p = 3.9 × 10⁻⁷*** (left plot); 0.35 ^ns (middle plot); 7.1 × 10−⁵*** (right plot). f HELZ2 expression represses L1-induced IFN-α expression. HEK293T cells were transfected with only the pCEP4 empty vector (control); or co-transfected with pCEP4 empty vector and pALAF015 (control + HELZ2); pJM101/L1.3FLAG and a pCEP4 empty vector (L1 + control); or pJM101/L1.3FLAG and pALAF015 (L1 + HELZ2). IFN-α (red bars) and L1 RNA (gray bars) levels were determined by RT-qPCR (using a primer set against IFNs [IFN-α, which amplifies IFN-α1 and IFN-α13] or the primer set mneoI [Alu or L1], respectively) and normalized to ACTB RNA levels (primer set: Beta-actin). The RNA levels then were normalized to the pCEP4 only (control) for IFN-α, and (L1 + control) for L1. L1 RNA pairwise comparison: (L1 + control vs. L1 + HELZ2), p = 1.4 × 10⁻¹⁰***. IFN-α RNA pairwise comparisons: (control vs. control + HELZ2), p = 1.4 × 10⁻⁴***; (control vs. L1 + control), p = 7.2 × 10⁻⁷***; (L1 + control vs. L1 + HELZ2), p = 5.7 × 10⁻⁸***. All values are reported as the mean ± SEM of three independent biological replicates. With the exception of panel (e), the p-values were calculated using a one-way ANOVA followed by Bonferroni–Holm post hoc tests. ns: not significant; * p < 0.05; *** p < 0.001.

Fig. 7. A working model hypothesizing a negative feedback loop between L1 RNA levels and ISG proteins.

L1 RNAs and/or RNPs can be detected by cytoplasmic RNA sensors, which elicit the secretion of type I interferons (IFNs); ORF1p RNA binding might shield L1 RNA from the sensors. IFN-binding to the extracellular IFN cell surface receptors then activates a signaling cascade, which induces the expression of ISGs, including HELZ2, HERC5, and OASL. These ISG proteins appear to inhibit L1 retrotransposition at different steps in the L1 retrotransposition cycle. HELZ2 appears to recognize RNA sequences and/or RNA structures within the L1 5′UTR, independently of ORF1p RNA binding, leading to the degradation of L1 RNA and subsequent blunting of the IFN response.