Loss of function of Atrx recapitulates phenotypes of alternative lengthening of telomeres in a primary mouse model of sarcoma

Abstract

The development of a telomere maintenance mechanism is essential for immortalization in human cancer. While most cancers elongate their telomeres by expression of telomerase, 10-15% of human cancers utilize a pathway known as alternative lengthening of telomeres (ALT). ALT is commonly associated with loss-of-function mutations in ATRX. Here, we developed a genetically engineered primary mouse model of sarcoma in CAST/EiJ mice to investigate the extent to which telomerase deficiency and Atrx-inactivation lead to ALT induction. We observed increases in multiple ALT-associated phenotypic indicators in tumors with loss of function mutations of Atrx. Furthermore, we found that loss of Atrx leads to an increase in telomeric instability and telomere sister chromatid exchange. However, Atrx-deficient tumors did not show productive telomere length maintenance in the absence of telomerase. This primary mouse model of sarcoma could facilitate future investigations into the molecular features of ALT in vivo.

Keywords: Cancer; Cell biology; Molecular biology.

Graphical abstract

Figure 1

Genetically engineered mouse model of soft tissue sarcoma with loss of Atrx (A) CAST/EiJ mice expressing Cas9 (Rosa26-1loxp-Cas9/+) and either wild type mTR (mTR^+/+), heterozygous mTR loss (mTR^+/−), or homozygous mTR loss (mTR^−/−) are injected in the gastrocnemius muscle with a two-plasmid system to induce expression of oncogenic Kras^G12D and biallelic mutation of Trp53 (CAST KP model), with additional mutation of Atrx (CAST KPA model), followed by electroporation of the gastrocnemius muscle. Tumor initiation occurs approximately 60 days after electroporation of the plasmids. (B) Graphical representation of the Atrx mutation profile generated from amplicon sequencing, using DNA extracted from primary tumors. Tumors are organized into wild type CAST KP which did not receive the sgRNA for Atrx, CAST KPA^∗ which retain wild type function of Atrx after CRISPR/Cas9 editing, and CAST KPA which have loss of function mutations of Atrx. Genomic DNA was extracted from fresh frozen tumors, and next generation amplicon sequencing was performed for the site of Atrx targeted by CRISPR/Cas9 sgRNA (blue text). Examples of wild-type, non-frameshift (NFS), frameshift (FS), and large deletion (LD) mutation sequences are shown. (C–E) Immunohistochemical staining of ATRX on FFPE tumor sections from (C) CAST KP, (D) CAST KPA^∗, and (E) CAST KPA.

Figure 2

Histological characterization of soft tissue sarcomas H&E staining and immunohistochemistry of tumors from CAST KP (left), CAST KPA^∗ (middle), and CAST KPA (right). The tumors generated in this model are characterized by focal and patchy staining of desmin, focal staining of smooth muscle actin (SMA), focal staining of pan-cytokeratin (Pan-CK), and negative staining of myogenin.

Figure 3

Characterizing telomere length and telomerase function (A) Telomere restriction fragment assay of RsaI and HinfI restricted DNA from tumors “T” and normal liver controls “L” visualized using chemiluminescent detection. (B) Telomerase Repeated Amplification Protocol (TRAP) assay performed on cell lines derived from primary tumors. 143B is an osteosarcoma cell line with known telomerase activity (positive control), and U2OS is an osteosarcoma cell line without functional telomerase (negative control).

Figure 4

C-circle measurement in primary tumors (A) Rolling circle amplification of c-circles was performed using DNA extracted from fresh frozen tumors. Chemiluminescent detection of dot blot products from CAST KP (blue), CAST KPA^∗ (purple), and CAST KPA (red) and human control cell lines (black), with and without the addition of phi29 DNA polymerase. (B) Graphical representation of the average c-circle content of tumors by tumor mutation profile using a one-way ANOVA with Tukey’s modification, and error bars are represented as the mean ±95% confidence interval. (C) Graphical representation of the average c-circle content of tumors by mTR genotype with CAST KP, with error bars represented as the mean ±95% confidence interval (blue), CAST KPA^∗ (purple), and CAST KPA (red).

Figure 5

Quantifying ALT-associated PML bodies in primary tumors (A) Representative images of immunoFISH including DAPI (blue), PML (magenta), and telomere (yellow). The colocalization of ultra-bright telomere foci with PML protein is classified as an ALT-associated PML body (APB). Scale bars have a length of 2.5 um. (B) Graphical representation of the proportion of APB positive nuclei in individual tumors by Atrx mutation status, compared using a Welch’s t test. Error bars are represented as the mean ±95% confidence interval. (C) Graphical representation of the proportion of APB positive nuclei by mTR genotype with CAST KP (blue), CAST KPA^∗ (purple), and CAST KPA (red). Error bars are represented as the mean ±95% confidence interval.

Figure 6

Cytogenetic analysis of primary cell lines (A) Metaphase Telomere FISH from a CAST KP primary cell line, DAPI (blue) and telomere (yellow). (B) Metaphase Telomere FISH from a CAST KPA primary cell line. Scale bars have a length of 10um. (C) Examples of fragile telomeres. (D) Examples of undetectable telomeres. (E) Example of chromosome fusion. (F–I) Graphical representation of (F) undetectable telomeres, (G) fragile telomeres, and (H) chromosome fusions. CAST KP (blue) CAST KPA (red). Statistics performed using one-way ANOVA with Sidak’s multiple comparisons. Error bars are represented as the mean ±95% confidence interval (I) Q-FISH violin plot of individual telomere fluorescence values from the TFL-Telo program, compared using a one-way ANOVA with Sidak’s multiple comparisons. Error bars are represented as the mean ±95% confidence interval.

Figure 7

Chromosome orientation FISH (CO-FISH) (A) Representative image of a CO-FISH metaphase spread (B) Representative image of a CO-FISH metaphase spread with multiple telomere sister chromatid exchanges (tSCE). Scale bars have a length of 10um. (C) Examples of normal CO-FISH telomere staining. (D) Examples of CO-FISH telomere staining with tSCE. (E) Graphical representation of tSCE quantification in each cell line. Statistics performed using a one-way ANOVA with Sidak’s multiple comparisons, and error bars are represented as the mean ±95% confidence interval.