br Acknowledgments br Introduction Cancer is the leading cau

Acknowledgments
Introduction Cancer is the leading cause of mortality in the United States for those between the ages of 40–79 years, surpassing heart diseases by more than 100,000 deaths [1]. The cause of most cancer subtypes remains idiopathic and may arise out of ‘bad luck’ among otherwise seemingly healthy individuals [2]. Less than 10% of all cancers have known heritable germline alterations that contribute to increased predisposition for cancer development. For a number of other cancers, a comprehensive molecular explanation does not exist to elucidate why they spontaneously occur. Autotaxin is a secreted ectonucleotide enzyme found in circulation and tissues throughout the body, where it metabolizes lysophosphatidyl choline to yield lysophosphatidic acid. In addition to its phospholipase functions, autotaxin is a phosphodiesterase involved in inflammation and cancer progression [[3], [4], [5], [6]]. The most critical role of autotaxin occurs in the developing embryo, whereby autotaxin-null mice suffer embryonic lethality from defects in vasculature system formation [7,8]. Even a single point mutation within autotaxin's enzymatic catalytic site, a substitution of threonine 210 for alanine, recapitulates this lethality [9]. Alterations in autotaxin expression are correlated with numerous human diseases. For example, among patient specimens, autotaxin is overexpressed in renal cell carcinoma, AR 231453 mg carcinoma, thyroid carcinoma and glioblastoma multiforme [[10], [11], [12]]. The Cancer Genome Atlas reveals additional cancer types with altered levels of autotaxin [[13], [14], [15]] (Supplementary Fig. 1). Experimentally, the overexpression of autotaxin in mice promotes late-onset metastatic breast carcinoma, an elevation of circulating lysophosphatidic acid, bleeding diathesis, thrombosis attenuation, hepatocellular carcinoma and chronic liver disease [5,16,17]. These observations coincide with autotaxin's regulation of the vasculature and angiogenesis [[18], [19], [20]]. Tumor cells release DNA, RNA, protein and microRNA (miRNA) into the circulation, suggesting the application of ‘liquid’ biopsies to eventually become clinically possible as more evidence about specific molecules and their interpretation are revealed [21]. A recent study tested CancerSEEK, which measures a combination of circulating DNA and protein, and positively detected ∼70% of eight cancer types, with 43% of stage I cancers detected [22]. As the ability to accurately measure small molecules advances, the potential impact of liquid biopsies increases. MiRNAs are nucleotides ∼22 in length that regulate mRNA, non-coding RNAs, pseudogenes and other types of transcripts through complementary binding interactions with 3′ or 5′ untranslated regions and other elements. There are approximately 3000 defined miRNAs in the human genome, but at least 3707 more novel ones likely exist and it is unclear how many could ultimately be discovered [23]. Since mi-RNAs have the ability to silence transcriptional gene expression across a number of targets, this suggests a much broader role for these entities than what is currently appreciated [24]. Previously, miRNAs have shown superior biomarker ability to delineate cancer tissue classification, progression and prognostic outcomes [25,26]. In addition, miRNAs possess a high degree of stability in circulation [27], suggesting the possibility that these molecules are ideal biomarkers for liquid biopsies.
Materials and methods
Results To assess whether circulating miRNA could distinguish between inbred littermates that developed tumors, as a result of autotaxin expression, from those that did not develop tumors, heterozygous mice with autotaxin under the control of the alpha-1 antitrypsin promoter primarily expressed in liver were utilized. These “AT-ATX” transgenic mice were developed on an FVB/N background and females were aged to determine whether spontaneous tumors would form [17,30]. The heterozygous autotaxin genotype was used because homozygous autotaxin overexpression and/or homozygous knockout leads to embryonic lethality from defects in the vasculature system [7,8,31]. Importantly, tumors developed in 21% of transgenic mice, but not in wild-type FVB/N females (Fig. 1A).