The ALK(F1174L) mutation potentiates the oncogenic activity of MYCN in neuroblastoma

Abstract

The ALK(F1174L) mutation is associated with intrinsic and acquired resistance to crizotinib and cosegregates with MYCN in neuroblastoma. In this study, we generated a mouse model overexpressing ALK(F1174L) in the neural crest. Compared to ALK(F1174L) and MYCN alone, co-expression of these two oncogenes led to the development of neuroblastomas with earlier onset, higher penetrance, and enhanced lethality. ALK(F1174L)/MYCN tumors exhibited increased MYCN dosage due to ALK(F1174L)-induced activation of the PI3K/AKT/mTOR and MAPK pathways, coupled with suppression of MYCN pro-apoptotic effects. Combined treatment with the ATP-competitive mTOR inhibitor Torin2 overcame the resistance of ALK(F1174L)/MYCN tumors to crizotinib. Our findings demonstrate a pathogenic role for ALK(F1174L) in neuroblastomas overexpressing MYCN and suggest a strategy for improving targeted therapy for ALK-positive neuroblastoma.

Figure 1. Overexpression of ALK^F1174L Potentiates the Oncogenic Activity of MYCN In Vivo

(A) The ALK^F1174L cDNA was ligated 3′ to the rat Th promoter to generate the p4.5 Th-ALK^F1174L transgenic construct (Th-ALK^F1174L). (B) RT-PCR analysis of ALK expression in cervical ganglia and adrenal glands of 7 to 10-day-old WT mice (M1–M3) or mice transgenic for ALK^F1174L (M4–M6) (left panels). Spl, spleen. Electropherogram showing the change in the phenylalanine codon (TTC) in WT ALK to leucine (CTC) in the ALK^F1174L mutation in ganglia (right panel). (C–D) Kaplan-Meier survival curves of founder lines resulting from intercrosses of Th-ALK^F1174L mice and Th-MYCN mice. (C) Founder 2 - ALK^F1174L/MYCN, n=31; MYCN, n=9; ALK^F1174L, n=4; WT, n=5. Founder 4 - ALK^F1174L/MYCN, n=25; MYCN, n=20; ALK^F1174L, n=13; WT, n=6. (D) Founder 1 - ALK^F1174L/MYCN, n=12; MYCN, n=6; ALK^F1174L, n=12; WT, n=3. Founder 3 - ALK^F1174L/MYCN, n=11; MYCN, n=10; ALK^F1174L, n=11; WT, n=5. See also Figure S1.

Figure 2. Combined Overexpression of ALK^F1174L and MYCN Results in Multifocal Neuroblastomas

(A) Gross appearance of representative ALK^F1174L/MYCN and MYCN tumors. Tumors (*) arise as multifocal primary lesions in ALK ^F1174L/MYCN animals: thoracic paraspinous (left panel) and abdominal (right panel) tumors. K, kidney, L, liver, Lu, Lung. (B) Quantitative RT-PCR analysis of murine and human ALK expression in spleen, adrenal glands and tumors from ALK^F1174L/MYCN or MYCN mice. Error bars indicate mean values ± 95% confidence interval (CI). (C) Immunoblotting of ALK ^F1174L/MYCN tumors (T, M1–M6) for ALK, MYCN (left panel) and pALK (right panel). pALK was detected using a human-specific ALK^Y1604 antibody. Matched spleens (S) were used as negative controls and GAPDH was used as a loading control. (D) Hematoxylin and eosin (H&E) and immunohistochemical staining for ALK and MYCN of representative ALK ^F1174L/MYCN murine and human neuroblastoma tumor sections. Scale bars, 20μm, insets 5μm. See also Figure S2.

Figure 3. ALK^F1174L/MYCN Tumors Exhibit Distinct Expression Profiles and Downstream Signaling Compared to MYCN Tumors

(A) Heat map representation of differentially regulated genes in ALK ^F1174L/MYCN vs. MYCN tumors (fold change ≥2.0; corrected p-value < 0.05). (B) Gene set enrichment analysis of PI3K/AKT/mTOR and MAPK pathway genes in transcriptional profiles of ALK ^F1174L/MYCN and MYCN tumors. The normalized enrichment score (NES) and the nominal p-value are indicated. (C) Immunoblotting of PI3K/AKT/mTOR and MAPK signaling pathways in ALK ^F1174L/MYCN and MYCN tumors. (D) Immunohistochemical staining for PI3K/AKT/mTOR (pS6K) and MAPK (pMEK and pERK) pathways in ALK ^F1174L/MYCN and MYCN murine and human tumors (Scale bars, 10 μm). See also Figure S3.

Figure 4. ALK ^F1174L Enhances MYCN Protein Stabilization

(A) Immunohistochemical staining for MYCN in ALK ^F1174L/MYCN and MYCN tumors. Scale bars, 20μm. (B) Western blot analysis of MYCN protein in Kelly human neuroblastoma cells after ALK^F1174L knockdown using three different siRNAs (1, 2, and 3). Ctl, cells expressing a control siRNA. (C) Western blot analysis of MYCN protein expression in Kelly cells transfected with an ALK or control shRNA followed by treatment with 25 μmol/L of cycloheximide (CHX) and harvested at the indicated time points. (D) Immunoblotting of ALK ^F1174L/MYCN and MYCN tumors for pMYCN^S62, pMYCN^T58 and pGSK3β (upper panel). Graph (lower panel) depicts quantification of pMYCN^T58 expression in the two types of tumors. ^*p = 0.05 by unpaired t-test. (E) In situ detection of MYCN:FBXW7 complex formation (arrowheads) in untreated Kelly cells (mock control), ALK^F1174L-depleted cells (siRNAs 1 & 3), or cells probed with secondary antibody only as a negative control (antibody control). Quantification of the number of MYCN:FBXW7 interactions in untreated Kelly cells and ALK^F1174L-depleted Kelly cells (right panel). Data are presented as means ± standard deviation (SD). ^***p = 0.009 (Student’s t-test).

Figure 5. ALK ^F1174L Induces Expression of Endogenous Murine MYCN

(A) Heat map representing the top 50 upregulated genes (corrected p-value < 0.05, fold change ≥ 2.0) in ALK ^F1174L/MYCN tumors compared with MYCN tumors. (B) qRT-PCR analysis of murine and human MYCN expression in adrenal glands and tumors from ALK^F1174L/MYCN and MYCN mice. Expression is relative to either murine or human MYCN levels in tumors from MYCN heterozygotic mice. Mean values ± 95% CI are indicated. (C) Fluorescence in situ (FISH) analysis of murine ALK ^F1174L/MYCN tumors using an ALK dual-color break-apart probe (upper panel) and a MYCN probe (lower panel) (Scale bar, 10μm). (D) qRT-PCR analysis of ALK ^F1174L and MYCN expression in Kelly cells in which ALK^F1174L expression was depleted using shRNA, compared with cells in which a control shRNA was used. Western blot showing level of ALK^F1174L knockdown (inset). Means ± SEM values for 3 independent experiments are shown. Ctl, cells expressing a control shRNA.

Figure 6. ALK^F1174L Promotes Anti-apoptotic Activity in ALK^F1174L/MYCN Neuroblastomas

(A) Immunohistochemical staining for Ki67 and TUNEL of transgenic murine and human ALK ^F1174L/MYCN and MYCN neuroblastomas. Scale bars, 20μm. (B) Heat map representation of the expression levels of pro-survival genes, Bcl2 and Bclw, and the pro-apoptotic genes Bak, Bax, Bik, Bid, Noxa and Trp53, in ALK ^F1174L/MYCN and MYCN tumors. (C) qRT-PCR analysis of Bclw and Bid expression in ALK ^F1174L/MYCN and MYCN tumors (n=10 in each group). Gene expression is relative to that in a MYCN tumor. Data are presented as mean ± SEM. (^*p = 0.03, ^** p = 0.005). (D) Immunoblotting of ALK^F1174L/MYCN and MYCN tumors depicting expression of BCLW, BCL2 and BID proteins in the ALK^F1174L/MYCN and MYCN tumors. See also Figure S4.

Figure 7. Combined Targeting of ALK^F1174L and mTOR is Effective against ALK^F1174L/MYCN Neuroblastomas

(A) MRI images depicting a representative tumor response to vehicle or crizotinib (100 mg/kg) and Torin2 (20 mg/kg) as single agents or in combination (at single-agent doses) after 7 days of treatment. Comb, combination treatment. (B) Quantitation of changes in tumor volume in animals treated with crizotinib, Torin2, or combination therapy as measured by MRI on days 0, 3 and 7. Data are presented as means ± SEM (4 animals per treatment group). Crizotinib vs. combination, ^**p = 0.003; Torin2 vs. combination; ^****p = 0.0002; vehicle vs. combination, ^**p=0.002; all by student’s t-test. (C) Western blot analysis of the indicated proteins in ALK ^F1174L/MYCN tumors in panels (A) and (B). Veh, vehicle; Comb, combination. (D) H&E and immunohistochemical staining of tumors in (A) and (B) as indicated. Comb, combination treatment. Scale bars, 10μm. (E) Kaplan-Meier survival analysis for ALK ^F1174L/MYCN animals (10 per group) treated for 14 days with crizotinib and Torin2 as single or combined agents. Crizotinib vs. combination, p = 0.007; Torin2 vs. combination, p = 0.02; both comparisons by log-rank test. See also Figure S5 and Table S1.