Over-expression of ADAR1 in mice does not initiate or accelerate cancer formation in vivo

Abstract

Adenosine to inosine editing (A-to-I) in regions of double stranded RNA (dsRNA) is mediated by adenosine deaminase acting on RNA 1 (ADAR1) or ADAR2. ADAR1 and A-to-I editing levels are increased in many human cancers. Inhibition of ADAR1 has emerged as a high priority oncology target, however, whether ADAR1 overexpression enables cancer initiation or progression has not been directly tested. We established a series of in vivo models to allow overexpression of full-length ADAR1, or its individual isoforms, to test if increased ADAR1 expression was oncogenic. Widespread over-expression of ADAR1 or the p110 or p150 isoforms individually as sole lesions was well tolerated and did not result in cancer initiation. Therefore, ADAR1 overexpression alone is not sufficient to initiate cancer. We demonstrate that endogenous ADAR1 and A-to-I editing increased upon immortalization in murine cells, consistent with the observations from human cancers. We tested if ADAR1 over-expression could co-operate with cancer initiated by loss of tumour suppressors using a model of osteosarcoma. We did not see a disease potentiating or modifying effect of overexpressing ADAR1 or its isoforms in the models assessed. We conclude that increased ADAR1 expression and A-to-I editing in cancers is most likely a consequence of tumor formation.

Graphical Abstract

Figure 1.

Proposed model of ADAR1 action in cancer. (A) Proposed models of the action of ADAR1 in cancer. In the ‘ADAR1 as an oncogene’ model, elevated ADAR1 leads to increased A-to-I editing and this acts as a tumor initiating event or promotes tumor establishment and maintenance. In the alternate ‘ADAR1 as a passenger model’ ADAR1 is elevated as a result of changes in the tumor transcriptome and environment, leading to increased ADAR1 as a secondary consequence. (B) Schematic of the constructs used to overexpress murine Adar1 cDNA from the Rosa26 locus in mice.

Figure 2.

Overexpression of ADAR1 in vivo. (A) Schematic outline of in vivo validation experiment. (B) Representative flow cytometry histograms of peripheral blood leukocytes for GFP expression following 14 days in vivo tamoxifen treatment. (C) Western blot analysis of ADAR1 expression in the spleen (left panel) and liver (right panel) following 14 days in vivo tamoxifen treatment using anti-ADAR1 and anti-Flag antibodies. (D) Adar1 and Adarb1 expression as measured from RNA-seq of the liver following 14 days in vivo tamoxifen treatment (n = 3–4 per genotype; note Adar1-Za mutation samples were not sequenced).

Figure 3.

Overexpression of ADAR1 in vivo increased A-to-I editing levels of cellular RNA. Gene expression analysis (MA plot, upper) and A-to-I editing levels (lower) of known individual sites following 14 days in vivo tamoxifen treatment from (A) Adar1 full length (FL); (B) Adar1p110; (C) Adar1p150; (D) Adar1E861A liver RNA-seq datasets compared to Ubc-CreER+ tamoxifen treated controls (n = 3 per genotype). Red dots in upper panels represent interferon-stimulated genes. Blue dots in lower panels represent significantly different editing at individual sites between the genotypes (Jacusa statistic (likelihood ratio of two samples) >5). (E) The repeat editing index (AEI) from each genotype. (F) IGV screen shot of Azin1 editing at the recoding p.S367G site; quantitation and statistical analysis of the editing frequency at the recoding site and average number of reads per sample for the site (expressed as mean ± sem for each allele). **P< 0.01, ***P< 0.001; Statistical comparisons using a two-way ANOVA with multiple comparisons correction using Prism software.

Figure 4.

Long-term in vivo overexpression of ADAR1 is well tolerated. (A) Schematic outline of in vivo experiment. (B) Kaplan–Meier survival plot of each genotype. Numbers as indicated in the inset, no significant difference in survival. Peripheral blood (C) leukocyte numbers, (D) red blood cell numbers and (E) platelet counts for each genotype; number per genotype indicated in panel C noting that the range indicates the minimum and maximum at any given time point per genotype. Due to restrictions during the pandemic the time points have been assigned as 0, 14/28 days, 84 days, 175 days and >580 days and the time points grouped to the closest of these for graphing. (F) GFP levels in the total leukocyte population. (G) Representative flow cytometry plots showing GFP levels in each genotype at the indicated time points. If no statistical significance indicated then no significant difference.

Figure 5.

Modest changes in hematopoiesis with overexpression of ADAR1. (A) The percentage contribution of peripheral blood GFP+ cells to each indicated lineage across each genotype. (B) Total bone marrow cellularity per femur. (C) The percentage contribution of GFP+ cells in the bone marrow to each indicated population. (D) Outline of hematopoiesis and the relationship between populations assessed. (E) Representative flow cytometry plots used to define GFP positive and negative fractions and (F) assess the hematopoietic stem and progenitor compartment. (G) The percentage contribution of GFP+ cells in the bone marrow to the lineage-cKit + Sca1+ (LKS+) population and the long-term and short-term hematopoietic stem cell populations (contained within the LKS+ fraction). (H) The percentage contribution of GFP+ cells in the bone marrow to the lineage-cKit+ Sca1– (LKS–) population and the megakaryocyte progenitors (MkP), granulocyte macrophage progenitors (GMP), pre-GM, pre-Megakaryocyte erythroid progenitors (preMegE), pre colony forming unit erythroid (preCFU-E) and CFU-E populations (contained within the LKS- fraction). (I) Total cellularity of the spleen and contribution of the GFP+ cells to the indicated cell populations. (J) Total thymus cellularity and contribution of the GFP+ cells to the indicated cell populations. Each circle indicates an individual animal; *P< 0.05, **P< 0.01, ***P< 0.001; statistical comparisons using a two-way ANOVA with multiple comparisons correction using Prism software.

Figure 6.

Adar1 expression and editing activity increases during cellular immortalization. (A) Schematic outline of in vitro experiment. Long bone osteoblasts were isolated from R26-CreER^T2Trp53^fl/fl mice and cultured with tamoxifen to induce deletion of p53. Cells were collected at day 7, 14 and 21 for analysis. Osteoblasts were isolated from three animals and cultured separately (biological replicates). (B) Expression (counts per million) of Adar1 and Adarb1 at day 7, 14 and 21 as determined by RNA-seq at each time point. (C) MA plot of gene expression comparing day 7 (p53 still expressed) and day 21 (p53 deficient) with significantly different genes indicated in blue and interferon stimulated genes (ISGs) indicated in red. (D) IGV screen shot of Azin1 (upper) and Cdk13 (lower) editing at the indicated recoding sites; quantitation of the editing frequency at each site (expressed as mean ± sem for each allele). The Alu/repeat editing index (AEI) from each timepoint derived from the RNA-seq. (E) A-to-I editing levels of known individual sites comparing editing levels at day 21 (y axis) to day 7 (x axis) following tamoxifen treatment. Blue dots represent significantly different editing at individual sites between the genotypes (Jacusa statistic (likelihood ratio of two samples) >5). (F) The repeat editing index (AEI) at each time point calculated from the RNA-seq dataset. ***P< 0.001; Statistical comparisons using a two-way ANOVA with multiple comparisons correction.

Figure 7.

Overexpression of ADAR1 does not accelerate or modify osteosarcoma behavior in vivo. (A) Schematic outline of in vivo osteosarcoma model. (B) Kaplan–Meier survival plot of each genotype. Numbers as indicated in the inset, no significant difference in survival. Number of animals per genotype indicated in inset. (C) Median days of age of each genotype, same cohort as represented in the KM plot; assessed by two-way ANOVA with multiple comparison correction. (D) Analysis of primary tumor location and metastatic spread in each genotype. (E) Representative histology of primary and metastatic lesions from each indicated genotype. The osteosarcoma model generates a fibroblastic osteosarcoma. (F) Western blot of ADAR1 (anti-ADAR1 antibody) in whole tumor pieces derived from the indicated genotypes.