Blimp1 induces transient metastatic heterogeneity in pancreatic cancer [RNA-seq]

Purpose: Due to its high metastatic proclivity, pancreatic ductal adenocarcinoma (PDAC) is one of the most deadly types of cancer. Therefore, it is imperative to better understand how the disease spreads as it progresses. Using a novel genetically engineered mouse model that allows us to isolate a subpopulation of cancer cells with superior metastatic capacity, we show that this aggressive phenotype correlates exclusively with a strong hypoxia signature. We subsequently identified the novel hypoxia-inducible gene Blimp1, which appears to play a critical role in regulating the hypoxic response upon its induction. Furthermore, genetic ablation of Blimp1 greatly reduces the level of metastasis in a PDAC mouse model. The nature of this Blimp1-regulated hypoxia signature is very unstable, since the seeded metastatic lesions mostly re-adopt similar transcriptomic profiles as the primary tumors. In conclusion, our results offer a potential mechanistic insight into how hypoxia drives metastasis in PDAC. Methods: Pure, paired GFP-negative/Tomato-positive and GFP-positive/Tomato-positive cancer cells or pure Tomato-positive cancer cells were sorted from primary PDAC samples from 6 KPC-colors mice or KPCT mice, respectively, with the following criteria: single cell based on FSC-A/H; CD45-negative; CD31-negative; Ter119-negative; F4/80-negative; DAPI-negative; and Tomato-positive. RNA were extracted from 10^4 to 5x10^4 freshly sorted cancer cells using AllPrep DNA/RNA Micro Kit (Qiagen). RNA quality was assessed with the RNA6000 PicoAssay kit by using the Bioanalyzer 2100 (Agilent). All ex vivo RNA samples used for RNA-seq analyses had an RIN > 8.0. Total RNA (15 ng/sample) was used for cDNA synthesis and amplification with the Ovation RNA-Seq system (NuGEN Technologies, Inc.). Subsequently, the amplified DNA samples were fragmented through sonication (Covaris model S1) and subjected to library preparation using the Illumina TruSeqTM DNA sample preparation kit (Low-Throughput protocol) according to manufacturer''s protocol. The quality of purified cDNA library products was confirmed by bioanalyzer and prepared for cluster generation on HiSeq paired-end flow cells using the CBot automated cluster generation system followed by sequencing on HiSeq 2000 machines. We obtained 101bp, paired-end reads from fragments of an average length of 250bp. Subsequently, RNA-Seq reads were aligned to the mouse genome (mm10) using the STAR aligner with standard input parameters (Dobin et al., 2013). The number of reads uniquely aligned to exons of individual genes were counted with HTSeq against the UCSC KnownGene (mm10) transcriptome (Anders et al., 2015). Results: Compared to the GFP-negative counterparts, GFP-positive pure PDAC cancer cells express higher levels of genes that are highly enriched with hypoxia signature. Additionally, compared to the GFP-negative counterparts, GFP-positive pure PDAC cancer cells express lower levels of cell cycle-related genes. In contrast, pure cancer cells isolated based on locations reveal few consistent differentially expressed genes between primary tumor and liver metastases; no consistent differentially expressed gene between primary tumor and lymph node metastases. Conclusions: Transcriptome profiles of both GFP-negative/positive PDAC cancer cells suggest that Hmga2/GFP-expressing cancer cells are highly enriched for signatures that correspond to cells residing within hypoxic enrivonment. Overall design: Freshly sorted GFP/Hmga2-positive and GFP/Hmga2-negative PDAC cancer cells derived from tumors of 6 KPCT;Hmga2-CK/+ (KPC-colors) mice were subjected to transcriptome profiling by paired-end RNA-Seq (total of 6 pairs of samples with overall 12 samples). Additionally, pure Tomato-positive PDAC cancer cells isolated from different anatomical locations were also subjected to transcriptome profiling by paired-end RNA-Seq (n = 23, not including technical replicates).

37

Chiou SH, Risca VI, Wang GX, Yang D, Grüner BM, Kathiria AS, Ma RK, Vaka D, Chu P, Kozak M, Castellini L, Graves EE, Kim GE, Mourrain P, Koong AC, Giaccia AJ, Winslow MM