Pancreatic ductal adenocarcinoma (PDAC) is a prevalent and almost uniformly fatal malignancy. Several factors contribute to the poor outcome of PDAC patients, but the ability of cancer cells to leave the primary tumors and establish inoperable metastases remains a major impediment to successful therapy. Metastasis is a complex multistep process that is poorly understood at the molecular level. In particular, the molecular events that are required for a cancer cell to leave the blood and enter a secondary organ, a process called metastatic seeding, have remained elusive. The identification of tool compounds that inhibit metastatic seeding would not only illuminate the fundamental mechanisms that enable this process but also be directly translated into therapeutics to inhibit metastatic spread. While large in vitro compound screens have been used to address a variety of biological questions, it has remained difficult to use this approach for complex in vivo processes like metastasis, as it remains unclear which cell-based assays and readouts accurately reflect the in vivo process. High content in vivo chemical screens have historically been challenging due to the high cost of performing these assays in parallel. Our multi-disciplinary proposal will establish a novel method for in vivo multiplexed screening of covalent inhibitors to allow a chemical genetic dissection of the metastatic seeding process. To identify compounds that inhibit the initial steps of metastatic seeding we will screen a library of ~1,500 small molecule covalent inhibitors of hydrolytic enzymes. Importantly, these compounds contain electrophilic traps that irreversibly bind their targets, enabling sustained inhibition after in viro pretreatment of the cells without the need for continued dosing. To permit a multiplexed in vivo screening, we generated 96 variants of a metastasis-derived PDAC cell line in which each variant cell line contains a unique 6-nucleotide barcode (PDACBC). We will pretreat each PDACBC cell line in vitro with one irreversible inhibitor, then pool the 96 pretreated PDACBC cell lines. Barcode representation in the pre-injection and post-metastatic seeding cells will be determined using IlluminaTM sequencing of the barcode region. Barcodes that are underrepresented the post-seeding population will identify candidate inhibitors of metastatic seeding. To validate top candidates PDAC cells will be pretreated with top candidate compounds and assessed for their ability to seed metastases and form macro-metastasis. To identify the targets of these inhibitors we will generate tagged analogs of hit compounds to enable the purification of drug-target complexes followed by mass spectrometry-based protein identification. shRNA knockdown of these targets followed by intravenous injection will also be used to confirm the effect of the target on metastatic seeding. Our study will identify not only novel drug targets but also lead compounds potent enough to elicit a reduction in metastatic ability in vivo. Our study will advance our understanding of PDAC metastasis and provide insight into common mechanisms used by other cancer types.