Figures and data
Msi2 is expressed in the stem cells of the distal lung and in lung adenocarcinoma.
A) Msi2 is expressed in the Lin-EpCAM+ lung epithelial cells of Msi2GFP/+ (M2GFP) reporter mice. Representative FACS plots shown for non-reporter (Msi2+/+, left) and M2GFP reporter mouse lungs (Msi2eGFP/+middle). Frequency of M2GFP-expressing cells in lung epithelia (right; n = 2 non-reporter, n = 5 reporter). Data represented as mean ± SD, one outlier identified and removed using the Grubb’s test, ***p < 0.001 by Student’s t-test. B-C) Msi2 is expressed in the known stem cell populations of the distal lung. B) Frequency of Msi2 expression in the Lin-EpCAM+Sca1+ enriched Club/BASC and the Lin-EpCAM+Sca1- enriched AT2 cell populations (n=5). C) Representative images of cytospins from M2GFP reporter mouse Lin-EpCAM+ lung epithelial cells showing expression of Msi2 expression in Club cells (top row, marked by CC10 staining) and AT2 cells (bottom row, marked by SPC staining). D) Representative images of Msi2 in tumors from KPf/f mice; some tumors display ubiquitous expression of Msi2 (left), while others display more heterogeneous expression (right).
Msi2-expressing cells preferentially favor tumor growth in lung adenocarcinoma.
A) Schematic of the strategy used to measure the tumor sphere forming capacity of Msi2-expressing and non-expressing cells in vitro. Msi2GFP/+; KrasG12D/+; p53fl/fl; RosaCreER/+ (M2GFP-KPf/f -RosaCreER/+) mice were treated with tamoxifen for 5 days to induce recombination of floxed alleles. Three days following the final dose of tamoxifen, lungs were dissociated and Msi2-expressing (GFP+) or non-expressing (GFP-) cells were isolated by FACS and then plated in a tumor sphere assay. B) Msi2-expressing cells isolated from KPf/f lung epithelia following tamoxifen treatment preferentially form tumor spheres over multiple cell passages in vitro as compared to non-expressing cells. Representative experiment shown, (n = 3). Data represented as mean ± SD, **p<0.01, ***p<0.001 by Student’s t-test. C-E) Tumor spheres formed by Msi2+ cells in vitro form aggressive tumors in vivo. Tumor spheres isolated from Msi2+ cells after quaternary passage in vitro and transplanted in vivo form highly aggressive flank tumors (C) that retain Msi2 expression (D) and are able to metastasize to the lung (E). F) Heatmap showing significant differentially expressed genes between Msi2-expressing cells and non-expressing cells. G-L) Gene Set Enrichment Analysis of Msi2-expressing normal lung epithelial cell gene signatures. Heatmaps of gene signatures and selected genes that are involved in developmental and stem cell signaling (G-H), oncogenic signaling (I-J), and therapy resistance (K-L). Red represents gene signatures or genes that are upregulated in the presence of Msi2, while blue represents gene signatures or genes that are downregulated in the absence of Msi2.
Msi2 is required for lung adenocarcinoma growth and progression.
A) Breeding scheme used to generate the Msi2-/-; KrasG12D/+; p53fl/fl (Msi2-/--KPf/f) model of lung adenocarcinoma. B) Representative images of wild type (left) and Msi2-/- (right) KrasG12D/+; p53fl/fl lungs gender and age matched mice at 14 weeks after tumor initiation. C-D) The number of tumors formed (C) and overall tumor burden (D) is significantly reduced in Msi2-/- KPf/f mice. E) Msi2-/- KPf/f mice have a significant reduction in the frequency of mid- and high-grade tumors. Data represented as mean ± SD, two outliers identified and removed for high-grade tumors using the Grubb’s test *p < 0.05, **p<0.01 by Student’s t-test. F) The tumor take rate is reduced in Msi2-/- KPf/f mice. While only 17% of AdCre-infected wild ype mice are tumor-free at the time of death, more than twice as many (39%) Msi2-/- KPf/f mice are tumor-ree at the time of death. G) Msi2-/- KPf/f have significantly increased survival (p = 0.04, hazard ratio = 1.8; n = 27 WT-KPf/f, n = 23 Msi2-/--KPf/f), with a median survival of 428 days compared to 209 days for wild type mice. Log-rank test was used to determine the difference in survival curves between wild type and Msi2-/- KPf/f mice. H) Lung tumors in Msi2-/- KPf/f mice express Msi2. Representative mmunohistochemical staining (left) for Msi2 (top) or IgG control (bottom) in lung tumors from a 16-week old Msi2-/- KPf/f mouse. Quantification of the frequency of Msi2-expressing tumors from Msi2-/- KPf/f mice shows hat while some tumors lack Msi2 expression the majority of tumors express Msi2 (right, n = 3). Data represented as mean ± SD, **p<0.01 by Student’s t-test.
Loss of Msi2 impairs growth of established cancer.
A) Mouse KPf/f cells transduced with shMsi2 and used for in vitro tumorsphere assays have an 83% reduction in Msi2 mRNA expression. B) Loss of Msi2 significantly impairs tumor sphere formation in vitro (n =3). C) Mouse KPf/f cells transduced with shMsi2 and used for in vivo flank transplant assays have an 83% reduction in Msi2 mRNA expression. D) Loss of Msi2 consistently impairs tumor growth in flank transplants in vivo. Data represented as mean ± SD, one outlier identified and removed using the Grubb’s test **p<0.01 by Student’s t-test. E) Schematic for testing the impact of Msi2 loss on growth of patient- derived xenografts in vivo. Patient-derived xenografts were harvested, dissociated, transduced with shMsi2 or shControl virus and split into two pools that were incubated for either 24 or 48 hours. After 24 hours, one pool of cells (containing a mixture of transduced and untransduced cells) was transplanted into the flanks of NSG mice. After 48 hours the remaining pool of cells was analyzed via FACS to determine the frequency of cells infected with shRNA. F) The frequency of infection was comparable for shControl and shMsi2 in two independent patient samples. Reduced Msi2 expression led to pronounced inhibition of tumor growth in two independent patient-derived xenografts.
Msi2 regulates oncogenic signaling in lung adenocarcinoma.
A) Principal component analysis of KP cells transduced with shMsi2 (Msi2KD, purple) or shControl (gray). BG) Gene Set Enrichment Analysis of Msi2KD gene signatures. Heatmaps of gene signatures and selected genes that are involved in developmental and stem cell signaling (B-C), DNA repair (D-E), and metabolism (F-G) and are downregulated (shown in blue) following the loss of Msi2. H-I) Known (H) and novel (I) regulators of lung cancer that are downregulated following loss of Msi2. B-I) Red represents genes or gene signatures that are upregulated in the presence of Msi2, while blue represents genes that are downregulated following the loss of Msi2. J) Confirmation of downregulated genes in Msi2KD cells using qRT-PCR analysis. K) In vitro functional analysis of novel effectors of lung cancer (Ptgds, Arl2bp, Rnf157, Sty11) downregulated by Msi2. KPf/f cells were transduced with shRNA to inhibit the genes of interest and analyzed for the resulting impact on tumor sphere formation in vitro. Sphere formation, n=3 per condition. Data represented as mean ± SD,**p<0.01, ***p<0.001, ****p<0.0001 by Student’s t-test.
FACS plots for PDXs transduced with shRNA
PDX cells transduced with either shControl or shMsi2 lentivirus have a similar frequency of infection. CD45-CD31-PDX cells from Patient 1 show a 27.4% frequency of infection when transduced with shControl, and a 28.6% frequency of infection when transduced with shMsi2 (top). CD45-CD31-PDX cells from Patient 2 show an 8.85% frequency of infection when transduced with shControl, and an 11% frequency of infection when transduced with shMsi2 (bottom). PDX cells that were untransduced and unstained for FACS antibodies were used as negative gating controls (shown at right for each patient sample).
Antibodies used Tables of antibodies used in this study for FACS and immunofluorescence and immunohistochemistry
