Oncogenic RABL6A promotes NF1-associated MPNST progression in vivo

Abstract

Background: Malignant peripheral nerve sheath tumors (MPNSTs) are aggressive sarcomas with complex molecular and genetic alterations. Powerful tumor suppressors CDKN2A and TP53 are commonly disrupted along with NF1, a gene that encodes a negative regulator of Ras. Many additional factors have been implicated in MPNST pathogenesis. A greater understanding of critical drivers of MPNSTs is needed to guide more informed targeted therapies for patients. RABL6A is a newly identified driver of MPNST cell survival and proliferation whose in vivo role in the disease is unknown.

Methods: Using CRISPR-Cas9 targeting of Nf1 + Cdkn2a or Nf1 + Tp53 in the mouse sciatic nerve to form de novo MPNSTs, we investigated the biological significance of RABL6A in MPNST development. Terminal tumors were evaluated by western blot, qRT-PCR, and immunohistochemistry.

Results: Mice lacking Rabl6 displayed slower tumor progression and extended survival relative to wildtype animals in both genetic contexts. YAP oncogenic activity was selectively downregulated in Rabl6-null, Nf1 + Cdkn2a lesions whereas loss of RABL6A caused upregulation of the CDK inhibitor, p27, in all tumors. Paradoxically, both models displayed elevated Myc protein and Ki67 staining in terminal tumors lacking RABL6A. In Nf1 + p53 tumors, cellular atypia and polyploidy were evident and increased by RABL6A loss.

Conclusions: These findings demonstrate that RABL6A is required for optimal progression of NF1 mutant MPNSTs in vivo in both Cdkn2a and p53 inactivated settings. However, sustained RABL6A loss may provide selective pressure for unwanted alterations, including increased Myc, cellular atypia, and polyploidy, that ultimately promote a hyper-proliferative tumor phenotype akin to drug-resistant lesions.

Keywords: MPNST; Myc; NF1; RABL6A; YAP.

Figure 1.

RABL6A promotes MPNST progression in both Nf1 + Cdkn2a and Nf1 + p53 targeted primary MPNST mouse models. Adenoviral CRISPR-Cas9 targeting of Nf1 + Cdkn2a or Nf1 + p53 was performed in the sciatic nerve of wildtype (WT) or Rabl6 knockout (KO) C57BL/6N mice to generate primary MPNSTs. (A) Schematic of CRISPR-Cas9 targeting constructs with sgRNAs against Nf1 with p16Ink4a and p19Arf (together comprise the Cdkn2a locus), designated NC, or sgRNAs against Nf1 and p53, designated NP. U6 and CMV promoters are indicated. (B) Fold change in tumor volume shows mice lacking Rabl6 display slower tumor growth kinetics in both NC and NP genetic settings. (C) Time (in days) for tumors to triple in size in WT versus Rabl6 KO mice. (D) Survival (time to maximum 2000 mm³ tumor volume) of WT versus Rabl6 KO mice. Error bars, SEM. B: P value determined by a generalized linear model to assess the difference between the curves. C–D: P value, Student’s t-test for KO versus WT comparisons per genotype (*, P < .05; ***, P < .001).

Figure 2.

Tumor initiation is not altered by RABL6A loss in Nf1 + Cdkn2a and Nf1 + p53 inactivated MPNSTs. Individual tumor initiation and growth kinetics from (A) Nf1 + Cdkn2a (NC) and (B) Nf1 + 53 (NP) targeted tumors in wildtype (WT) and Rabl6 KO mice. (C) Initial tumor detection, measured in days following CRISPR-Cas9 adenovirus injection in the sciatic nerve of each mouse within the indicated groups, shows faster MPNST formation in NP tumors. Error bars, SEM. C: P value, Two-way ANOVA with Tukey’s multiple comparisons test (***, P < .001; n.s., not significant).

Figure 3.

RABL6A-null tumors display reduced YAP activity in a context-dependent manner. Representative YAP IHC images (200X magnification) from WT and Rabl6 KO mice in (A) Nf1 + Cdkn2a and (B) Nf1 + p53 primary MPNSTs with H-Score quantification graphed on the left. (C) Relative mRNA levels of YAP target genes, Ccn2 (Ctgf) (top) and Ccn1 (Cyr61) (bottom) from WT versus Rabl6 KO NC (Nf1 + Cdk2na) and NP (Nf1 + p53) edited tumors. YAP expression and activity (measured by downstream target expression) is decreased only in Nf1 + Cdkn2a Rabl6 KO tumors compared to WT, whereas Nf1 + p53 tumors remain the same. Error bars, SEM. P value, Student’s t-test for KO versus WT comparisons per genotype (*, P < .05; **, P < .01; n.s., not significant).

Figure 4.

Sustained RABL6A loss leads to paradoxical molecular and pathological alterations indicative of increased malignancy. (A) Representative western blots confirming loss of RABL6A, p16, and p53 in respective conditions. Rabl6 KO mice displayed increased p27 and Myc protein expression in both Nf1 + Cdkn2a (NC) and Nf1 + p53 (NP) tumors. ImageJ quantification of (B) p27 and (C) Myc protein expression. (D) Representative Ki67 IHC images (200X magnification) and H-score quantification reveal Rabl6 KO tumors have enhanced proliferation compared to WT tumors in both NC and NP settings. KO tumors notably also have enlarged cells. (E) Representative H&E images (scale bar, 75 µm) of the indicated tumors reveal increased cancer cell atypia with scattered enlarged cells in Nf1 + p53 altered tumors, which was further enhanced by RABL6A loss. Arrows highlight scattered enlarged cells with large, misshapen and multiple nuclei (morphologically consistent with polyploidy cells), which are magnified in the insets. Below, the number of enlarged cells with polyploidy was quantified and averaged from 5 different fields per tumor. B,C,D,E: Error bars, SEM. B,C,D: P value, Student’s t-test for comparisons between WT and KO tumors for the indicated tumor genotype (*, P < .05). E: P value, Tukey’s multiple comparisons test (*, P < .05; ****, P < .0001).

Figure 5.

Model of molecular changes in de novo MPNSTs following early versus sustained RABL6A loss. RABL6A-regulated pathways proposed to mediate reduced MPNST growth in the early stages of RABL6A loss (left) versus the highly proliferative tumors that have become refractory to sustained RABL6A loss (right). Validated changes are highlighted in color. Left, initial loss of RABL6A downregulates YAP expression and signaling (*, selectively in Nf1-Cdkn2a inactivated tumors) along with activation of p27-RB1-mediated cell cycle arrest. Right, long-term RABL6A loss somehow leads to increased Myc protein expression, which would promote the transcription (txn) of its tumor-promoting gene targets, some of which cause genomic instability. **In Nf1 + p53 inactivated tumors, sustained loss of RABL6A enhances tumor cell atypia and polyploidy, which are outcomes of genome instability and reflect increased malignancy.