During the progression of epithelial tumors, the extracellular matrix (ECM) accumulates collagen stabilized by covalent cross-links that enhance ECM stiffness and trigger tumor cell invasion and dissemination. However, the mechanisms by which tumor cells regulate collagen cross-link formation have not been fully defined. Here, we found that lung adenocarcinoma metastasis is driven by high expression of lysyl hydroxylase 2 (LH2), an enzyme that hydroxylates telopeptidyl lysine residues on collagen, and that LH2 expression in tumor cells increases hydroxylysine aldehyde-derived collagen cross-links (HLCCs) and decreases lysine aldehyde-derived collagen cross-links (LCCs) without changing the total amount of cross-links in the tumor stroma, indicating that malignant progression is associated with a change in the type of collagen cross-link. Furthermore, we showed that LH2 is a secreted protein, which challenges the current belief that LH2 localizes strictly on the endoplasmic reticulum. On the basis of these preliminary results, we postulate that LH2 expression in tumor cells functions as a regulatory switch, controlling the relative abundance of distinct types of collagen cross-links that influence the metastatic fate of tumor cells, and that,in addition to hydroxylating collagen ? chains on the endoplasmic reticulum, LH2 is secreted and promotes metastasis by hydroxylating collagen molecules in the extracellular space. To test these hypotheses, we will pursue 2 specific aims. Aim 1: Determine whether ectopic LH2 expression promotes metastasis in mice that develop lung adenocarcinoma from the expression of mutant K-ras. We will intratracheally deliver lentiviruses that co-express Cre recombinase and LH2 to K-rasLSL-G12D mice to induce ectopic LH2 expression and the recombination of a conditional K-rasG12D allele in the same lung epithelial cells. As a comparison, we will also administer lentiviruses co-expressing Cre and lysyl oxidase (LOX) or LOX-like-2 (LOXL2), which are collagen-modifying enzymes that have known pro-metastatic roles. Aim 2: Determine whether LH2 promotes metastasis by regulating the type of collagen cross-link in the tumor stroma. We will ascertain whether LH2-induced metastasis and stromal mechanical properties are attenuated by mutation of the LH2 hydroxylase domain or by deletion of the N-terminal signal peptide and are different from those induced by LOX, which regulates the total amount, but not the type, of collagen cross-link. In summary, the novelty of our proposal rests in the implementation of unique mouse modeling approaches to test an original hypothesis that links a particular type of collagen cross-link to metastasis regulation. If proven correct, these findings will challenge the current paradigm of the way in which tumor cells regulate collagen cross-linking to increase stromal stiffness and promote metastasis, and they will provide a rationale forfuture efforts to develop LH2 antagonists to test whether LH2 inhibition prevents metastasis in lung cancer patients.