Data on tumor progression of c-mos deficiency in murine models of KrasG12D lung and ApcMin colorectal cancer

The c-mos proto-oncogene was one of the first proto-oncogenes to be cloned. Apart from its role in meiosis, many efforts have been made to illuminate the mechanisms by which c-mos might acts as an oncogene. Increased Mos expression was found in most human tumor tissues. However, a detailed role of c-mos in tumor progression remains unknown. In this study, we analyzed online databases to find out the correlation between Mos expression and poor survival rates in human cancer patients. Then, we crossed c-mos knockout mice with ApcMin or KrasG12D mice to generate intestinal cancer model and lung cancer model, respectively. Tumor progression was monitored, and the influence of c-mos deficiency on cancer formation was investigated.


Value of the data
These data describe the expression of Mos differed between human and mouse. These data provide a significant correlation between high Mos expression and poor survival rates in lung cancer patients.

These data give insights into c-mos in tumor progression in both Apc Min intestinal cancer model and
Kras G12D lung cancer model.
These data are useful to researchers interested c-mos acts as an oncogene.

Data
The c-mos proto-oncogene was one of the first proto-oncogenes to be cloned [1]. Apart from its role in meiosis, many efforts have been made to illuminate the mechanisms by which c-mos might acts as an oncogene [2][3][4][5][6][7][8][9][10]. c-mos or its coding messenger RNA have been confirmed in most somatic tissues at low levels. Increased Mos expression was found in most human tumor tissues. However, a detailed role of c-mos in tumor progression remains unknown. This study was aimed to investigate whether c-mos is involved in tumor progression, via online database analysis and animal tumor models.   2. c-mos expression in human normal and cancer tissues. (A) Analysis of GeneAtlas U133A, gcrma in the BioGPS database (http://biogps.org) revealed that c-mos expression is present in human tissues. (B) Analysis of the TCGA Lung 2 cohort in the Oncomine database (www.oncomine.org) revealed that c-mos expression was significantly upregulated in human lung adenocarcinoma samples than the non-tumorous lung tissues. (C) Correlation between c-mos expression and patient survival. The c-mos expression and overall survival data were obtained from Kaplan-Meier survival plotter datasets as of April 20, 2017. The high and low c-mos (221367_at) expressers were grouped using an arbitrary cutoff percentile of 50% (966 for low c-mos expressers, and 960 for high c-mos expressers). The Mantel-Cox Log-Rank tests were done using the GraphPad Prism 7 software. (D) Analysis of the TCGA Colorectal 2 cohort in the Oncomine database (www.oncomine.org) revealed that c-mos expression was significantly upregulated in human colon and rectal adenocarcinoma samples than the non-tumorous colon tissues. (E) Correlation between c-mos expression and patient survival. The c-mos expression and overall survival data were obtained from TCGA datasets (Nature 2012). The high and low c-mos expressers were grouped using an arbitrary cutoff percentile of 50% (110 for low c-mos expressers, and 109 for high c-mos expressers). The Mantel-Cox Log-Rank tests were done using the GraphPad Prism 7 software. We first took advantage of online BIOGPS database (http://www.biogps.org). c-mos was found to be expressed almost evenly in human tissues ( Fig. 2A), while it expressed significantly higher in ovaries than other tissues in mouse (Fig. 3F). Analysis of the TCGA Lung 2 cohort in the Oncomine database (www.oncomine.org) showed that Mos expression was significantly upregulated in human lung adenocarcinoma samples than in the non-tumorous lung tissues (P o0.01) (Fig. 2B). In addition, analysis of the datasets obtained from Kaplan-Meier survival plotter revealed a significant correlation between high Mos expression and poor survival rates (Po 0.01) (Fig. 2C). The expression of Mos in human CRCs implicated that both of human colon and rectal adenocarcinoma tissues had higher Mos Fig. 3. Genetic deletion of c-mos gene has no effect on intestine and lung morphogenesis. (A) Real-time PCR quantification of c-mos mRNA levels in mouse lung and intestine tissues with wild-type (n ¼3) and with c-mos deficiency (n ¼ 3). Data were presented as means 7 SEM. Statistical analyses were performed using Student's t-test (B-C) H&E staining of the lung from wild-type and c-mos À / À mice with regular architecture. (D-E) Representative H&E staining of intestine from wild-type and c-mos À / À mice intestine. Scale bars: 50 μm. (F) Analysis of GeneAtlas MOE430, gcrma in the BioGPS database (http://biogps. org) revealed that c-mos expression is present in most mouse tissues but significantly higher in ovaries. expression (Po 0.01) (Fig. 2D), although the correlation of high Mos expression with poor survival rates was no trend toward significance (P ¼0.25) (Fig. 2E).
Then, c-mos deficiency was confirmed in both of lung and small intestine tissues using Real-time PCR quantification (Fig. 3A). Morphological changes of lung and small intestine in both Mos À / À and WT mice were observed at the age of 12 months. No significant differences were shown with regard to the tissue size, weight, and macroscopic appearance. The histological structures of lung and small intestine in Mos À / À and WT mice were nearly identical ( Fig. 3B-E).
To investigate the biological function of Mos in characterized cancers, we cross Mos À /À with Kras G12D to generate classic mice lung cancer model. Kras G12D and Kras G12D , Mos À /À mice were treated with Adeno-Cre injected intranasally at 8 weeks of age. After 12 weeks, mice were sacrificed for gross inspection and histopathological. Lung tumors were dissected for histopathological analysis. However, there was no difference on the slowdown of tumor progression in c-mos deficient mice (Fig. 4A-B). Fig. 4C-D showed representative histological sections from the Kras G12D mice and the Kras G12D ; Mos À /À mice, which showed no difference on tumor tissue morphogenesis. Then, the role of mos-deficiency in tumor formation was investigated using the Apc Min/ þ mouse intestinal tumor model. We found that there was no inhibitory effect on colon tumor progression when the c-mos gene was absent (Fig. 4E-F), and the absence of c-mos gene had no influence on small intestinal tumor burdens in Apc Min/ þ mice neither (Fig. 4G-H).

Mouse tumor model
For de novo lung cancer mice model, Kras G12D (n ¼11, male), and Kras G12D , Mos À / À (n ¼9, male) mice were treated with 2 Â 10 6 plague-forming unites of Adeno-Cre injected intranasally at 8 weeks of age (weight around 18 g) as previously described [11,12] (Fig. 1). After 12 weeks, mice were sacrificed in CO2 Rodent Euthanasia Chamber for gross inspection and histopathological. Lung tumors were dissected for histopathological analysis. Intestine cancer mice models with Apc Min/ þ (n ¼5, male), Apc Min/ þ , Mos À / À (n ¼5, male) mice were housed for 20 weeks, then mice were sacrificed in CO 2 Rodent Euthanasia Chamber and intestine tissues were collected for histopathological examination. Tumor number and tumor size were measured. All mice were monitored twice a week until endpoint time of the experiment. No animals were excluded from the analysis.

Quantitative RT-PCR
Lung and intestine samples were collected and total RNA was isolated with RNeasy Plus Mini Kit (QIAGEN) according to the manufacturer's instructions. Complementary DNAs were synthesized from the above-mentioned collected RNAs using the iScript cDNA Synthesis Kit (Bio-Rad). Quantitative PCR was done using IQ™ SYBR Green super-mixes and CFX96™ Touch Real-Time PCR detection system (Bio-rad). For all quantitative PCR reactions, Gapdh was measured for an internal control and used to normalize the data. The PCR primers used were as follows: c-mos: 5 0 -CTCCGGAGATCCTGAAAGGA-3 0 (sense) and 5 0 -CAGTGTCTTTCCAGTCAGGG-3 0 (antisense). Gapdh: 5 0 -TGCCCCCATGTTTGTGATG-3 0 (forward) and 5 0 -TGTGGTCATGAGCCCTTCC 0 (reverse).

Histopathological analysis
Histopathological analysis was performed according to our previous study [11,12]. In short, after mice were sacrificed, lungs were inflated with 1 ml Bouin's solution (Sigma-Aldrich) at room temperature for 20 min and fixed in 20 ml 4% PFA at 4°C for 24 h. Fixed lung tissues were embedded in paraffin sectioned at 5 μm thickness for hematoxylin and eosin (HE) staining.

Study design and statistical analysis
Minimal group size for tumor progression studies was calculated using an online power calculator available from DSS Researcher's Toolkit (https://www.dssresearch.com/KnowledgeCenter/toolk itcalculators.aspx) with an α of 0.05 and power of 0.8. Animal groups were not blinded but randomized, and investigators were blinded to the tumor counting experiments. No samples or animals were excluded from the analysis. Hypothesis concerning the data which included normal distribution (D'Agostino-Pearson normality test) and similar variation between the experimental groups were examined for appropriateness before the conduct of statistical tests. All statistical analyses were performed with Student's t-test (two independent groups) or two-way ANOVA (multiple groups) using IBM SPSS version 21.0, software (IBM Corp, Armonk, NY).

) Correlation of Mos expression and patient survival in colorectal cancers
The Mos expression and overall survival data were obtained from TCGA datasets (The Cancer Genome Atlas Network, 2012). The high and low Mos expressers were grouped using an arbitrary cutoff percentile of 50% (110 for low Mos expressers, and 109 for high Mos expressers). Pearson's correlation coefficient for Mos expression and patients' mortality is À 0.34, p-value is 0.25. The Mantel-Cox Log-Rank tests and Correlation tests were done by using GraphPad Prism 7 (GraphPad Software, La Jolla, CA, USA).