DNA hypomethylation results in activation of L1TD1 expression and loss of L1TD1 affects cell viability in HAP cells.

(A) Quantification of DNA methylation levels at the L1TD1 promoter in HAP1 wildtype (WT), DNMT1 KO and DNMT1/L1TD1 DKO cells using the MethyLight assay. DNA methylation is shown as percentage of methylation ratio (PMR). (B) qRT-PCR analysis of L1TD1 mRNA expression in HAP1 wildtype (WT), DNMT1 KO and DNMT1/L1TD1 DKO cells. GAPDH was used as a normalization control and relative L1TD1 mRNA levels in DNMT1 KO cells were set to 1. Data are shown as a mean of ± SD of 3 biological replicates. (C) Western blot analysis of L1TD1 levels in HAP1 WT, DNMT1 KO and DNMT1/L1TD1 DKO cells. β-actin was used as loading control. (D) Western blot analysis of L1TD1 protein expression in HAP1 WT and DNMT1 KO cells and OV-90 cells. β-actin was used as loading control. (E) Indirect immunofluorescence staining of L1TD1 (red) in HAP1 DNMT1 KO and DNMT1/L1TD1 DKO cells and OV-90 cells. In merged images nuclear DNA was stained with DAPI (blue). (F) Cell viability analysis of HAP1 WT, DNMT1 KO and DNMT1/L1TD1 DKO cells using the CellTiter-Glo assay measured over 96 h (n=6). (G) Bar graph representing the percentage of apoptotic cells of cultured cell lines quantified by flow cytometry analysis of cleaved caspase 3. (H) Western blot analysis of ɣH2AX levels in HAP1 WT, DNMT1 KO and DNMT1/L1TD1 DKO cells in nuclear extracts. Antibodies specific for histone H3 C-terminus and LAMIN B (LMNB) were used as loading controls. (A-B, F-G) Statistical significance was determined using one-way ANOVA with Tukey’s multiple comparison correction. * p≤0.05, ** p ≤0.01, *** p ≤0.001, **** p ≤0.0001, ns = not significant.

RIP-seq identifies a set of RNAs and transposon transcripts associated with L1TD1.

(A) Schematic representation of the RNA immunoprecipitation sequencing (RIP-seq) method (panel created with BioRender.com/x79t726). L1TD1-RNA complexes were isolated from HAP1 KO cell extracts with an L1TD1-specific antibody, RNA was isolated from complexes and input and cDNA libraries were prepared using the Smart-seq3 protocol. Sequencing data was analyzed by DEseq2 and TEtranscript software, separately. (B) Volcano plot showing L1TD1-associated transcripts as a result of DESeq2 analysis (cut-off log2FC > 2 and adj. p-value < 0.05). Selected hits are highlighted in blue. (C) RIP-qPCR analysis confirms L1TD1 interaction with the transcripts L1TD1, ARMC1, YY2. The bar graphs represent the fold enrichments of the transcripts in the IP samples of DNMT1 KO relative to DNMT1/L1TD1 DKO cells (set to 1) and normalized to input samples in the indicated cells. GAPDH was used as a negative control for the RIP-qPCR analysis. (D) Volcano plot showing L1TD1-associated transposon transcripts as result of the TEtranscript analysis with a log2FC > 1 and adj. p-value < 0.05. LINE1 elements are highlighted in green, ERV elements in yellow and other associated transposon transcripts in red. (E) RIP-qPCR analysis confirms the association of L1TD1 with L1 transcripts. Statistical significance was determined using paired two-tailed t-test. All data in the figure are shown as a mean of ± SD of 3 biological replicates. * p ≤0.05, ** p ≤0.01, *** p ≤0.001, **** p ≤0.0001, ns = not significant.

L1TD1 cross-talk with its ancestor L1 ORF1p.

(A) Volcano plot displaying the comparison of the proteomes of HAP1 DNMT1 KO and DNMT1/L1TD1 DKO cells determined by mass spectrometry. Differentially abundant proteins were plotted as DNMT1/L1TD1 DKO over DNMT1 KO (log2FC ≥ 1, adj. p-value < 0.05 (red) and log2FC ≤ −1, adj. p-value < 0.05 (blue)). (B) Volcano plot illustrating the DESeq2 analysis of RNA-seq performed with HAP1 DNMT1 KO and DNMT1/L1TD1 DKO cells. Differentially expressed genes are plotted as DNMT1/L1TD1 DKO over DNMT1 KO (log2FC ≥ 1, adj. p-value < 0.05 (red) and log2FC ≤ −1, adj. p-value < 0.05 (blue)). (C) Western blot analysis illustrating protein levels of L1TD1 and L1 ORF1p in HAP1 WT, DNMT1 KO and DNMT1/L1TD1 DKO cells. β-actin was used as loading control. (D) Physical interaction of L1 ORF1p and L1TD1. L1 ORF1p was immunoprecipitated with an L1 ORF1p-specific antibody from whole cell extracts prepared from DNMT1 KO and DNMT1/L1TD1 DKO cells. Precipitated L1 RNP complexes and inputs were analyzed on a Western blot. IgG was used as a negative IP control and β-actin was used as loading control. (E) Confocal microscopy images of indirect immunofluorescence co-stainings using mouse ORF1p (green) and rabbit L1TD1 (red) antibodies in DNMT1 KO, DNMT1/L1TD1 DKO and HAP1 WT cells. In merged images nuclear DNA was stained with DAPI.

L1TD1 promotes L1 retrotransposition.

(A) Schematic representation of plasmids used for retrotransposition (Figure modified from [35] and created with BioRender.com/k03r500). The pJJ101/L1.3 construct contains the full length human L1.3 element with a blasticidin deaminase gene (mblast) inserted in antisense within the 3’UTR. The mblast gene is disrupted by an intron and mblast expression occurs only when L1 transcript is expressed, reverse transcribed and inserted into the genome. The pJJ105/L1.3 plasmid contains a mutation in the reverse transcriptase (RT) resulting in defective retrotransposition. The backbone plasmid pCEP4 was used as additional negative control. The blasticidin deaminase gene containing plasmid pLenti6.2 was used as transfection/selection control. (B) Workflow of retrotransposition assay. DNMT1 KO and DNMT1/L1TD1 DKO cells were separately transfected with pJJ101 and control plasmids. Equal number of cells were seeded for each condition. Blasticidin selection (10 µg/ml) was started at day 4 and resistant colonies were counted on day 13. Panel created with BioRender.com/e85e605). (C) Bar graph showing the average number of retrotransposition events per 106 cells seeded in three independent experiments. Blasticidin resistant colonies in pLenti6.2 transfected cells were used for normalization. Statistical significance was determined using unpaired t-test. All data in the figure are shown as a mean of ± SD of three independent experiments, **** p ≤0.0001. (D) Representative pictures of bromophenol blue stainings of blasticidin resistant colonies for each genotype and each transfection.

Deregulation of the HAP1 transcriptome upon loss of DNMT1.

(A) Volcano plot illustrating the DESeq2 analysis of RNA-seq of HAP1 DNMT1 KO and wildtype (WT) cells. Differentially expressed genes are plotted as DNMT1 KO over WT (log2FC ≥ 1, adj. p-value < 0.05 (red) and log2FC ≤ −1, adj. p-value < 0.05 (blue). L1TD1 is highlighted in blue, KRAB- containing zinc finger proteins in green and de novo DNMTs in yellow. (B and C) DAVID Gene Ontology enrichment analysis of top upregulated transcripts DESeq2 comparisons (DNMT1 KO versus input and DNMT1 KO versus WT HAP1, log2FC ≥ 2, adj. p-value < 0.05) in terms of biological processes (B) and protein domains (C).

DNMT1 ablation results both in DNA hypomethylation at L1 elements and expression of L1 ORF1p transcripts.

(A) Quantification of DNA methylation levels at the L1TD1 promoter in HAP1 wildtype (WT) and DNMT1 KO cells using the MethyLight assay. DNA methylation is shown as percentage of methylation ratio (PMR). (B) qRT-PCR analysis of L1 mRNA expression in HAP1 wildtype (WT) and DNMT1 KO cells. GAPDH was used as a normalization control. All data in the figure are shown as a mean of ± SD of 3 biological replicates. * p ≤0.05, *** p ≤0.001.

L1TD1 interacts with a specific set of mRNAs and transposon transcripts.

(A) Volcano plot showing the L1TD1-associated transcripts enriched in DNMT1 KO IP over DNMT1/L1TD1 DKO IP (log2FC ≥ 2 and adj. p-val < 0.05). The transcripts of L1TD1, SYNJ2BP, ARMC1 are highlighted in green. (B) and (C) DAVID Gene Ontology enrichment analysis of 228 common transcripts identified as L1TD1 targets in both DESeq2 comparisons (DNMT1 KO versus input and DNMT1 KO versus DNMT1/L1TD1 DKO) in terms of biological processes (B) and protein domains (C). (D) Classification of TEs associated with L1TD1 (log2FC > 1, adj. p-val < 0.05) according to their frequency (taken from Suppl. Table S3).

GSEA analysis of proteins upon loss of L1TD1.

(A) Gene ontology enrichment analysis of proteins differentially expressed in HAP1 DNMT1/L1TD1 DKO versus DNMT1 KO cells. (B) Venn plot of the overlap of mRNAs identified in the RIP-seq analysis and differentially expressed proteins (up and down) identified in the mass spectrometry analysis of HAP1 DNMT1/L1TD1 DKO and DNMT1 KO cells. The single common mRNA is L1TD1.

Ablation of L1TD1 leads to changes in the transcriptome of HAP1 cells.

(A and B) qRT-PCR analyses of selected L1TD1-associated transcripts. Statistical significance was determined using unpaired two-tailed t-test. Data are shown as mean of ± SD of 3 biological replicates. ** p ≤0.01, ns not significant. (C) Volcano plot illustrating TEtranscript analysis of RNA-seq performed in HAP1 DNMT1 KO and DNMT1/L1TD1 DKO cells. Positively enriched TEs are shown in red and negatively enriched TEs are shown in blue. Alu elements are highlighted in green.

RNA-independent interaction of L1TD1 and L1 ORF1p in HAP1 and OV-90 cells.

(A) Western blot analysis of L1 ORF1p IPs with and without RNase A/T1 digestion in DNMT1 KO HAP1 cells using antibodies specific for L1 ORF1p and L1TD1 together with input samples. DNMT1/L1TD1 DKO HAP1 cells were used as a negative IP control and β-actin was used as Western blot loading control. (B) Western blot analysis of ORF1p IPs with and without RNase A/T1 digestion in OV-90 cells using antibodies specific for L1 ORF1p and L1TD1 in OV-90 cells. Mouse IgG was used as a negative control for the IP and β-actin was used as Western blot loading control. (C) Indirect immunofluorescence analysis of ORF1p (green) and L1TD1 (red) in OV-90 cells. In merged images nuclear DNA was stained with DAPI (blue).

L1TD1 enhances L1 retrotransposition.

(A) The bar graphs separately show the number of retrotransposition events per 106 cells seeded for three independent biological replicates. Blasticidin resistant colonies in pLenti6.2 transfected cells were used for normalization. Statistical significance was determined using unpaired t-test. All data in the figure are shown as a mean of ± SD of 3 technical replicates. ** p ≤0.01, **** p ≤0.0001. (B) Representative pictures of bromophenol blue stainings of blasticidin resistant colonies for transfection with the backbone plasmid pCEP4 used as negative control and the blasticidin deaminase gene containing plasmid pLenti6.2 used as transfection/selection control. (C) Potential mechanism of facilitated L1 retrotransposition by L1TD1. By association with L1-RNPs L1TD1 might enhance the chaperone function of L1ORF1p resulting in more efficient L1 retrotransposition. Panel C created with BioRender.com/r80m690.