The myxovirus resistance protein, MX1, interacts with cytoskeletal proteins and is secreted as a component of epithelial cell-derived exosomes: Evidence for novel cellular functions

For the past 50 years, the dairy industry has experienced an increase in milk production accompanied by a decrease in reproductive performance. This is evidenced by an increase in the number of inseminations required to establish a pregnancy, resulting in extended calving intervals. Approximately 40% of embryo loss occurs during a time when the conceptus is signaling to the dam to maintain the pregnancy. In ruminants, this signal is interferon-tau (IFN-tau), which acts to block the endometrial luteolytic mechanism, allowing the corpus luteum to continue producing progesterone, which supports pregnancy. IFN-tau also upregulates a number of interferon-stimulated genes (ISGs) in the endometrium, which are thought to play a role in endometrial differentiation and embryo implantation. Two of these genes encode the myxovirus resistance proteins MX1 and MX2.;Ott and associates has been studying the function of MX proteins in the endometrium during early pregnancy. Recently, it was demonstrated that MX1 was a secreted protein. It was then demonstrated that MX1 regulated secretion of another protein, ISG15. Interestingly, MX1 and ISG15 lack a secretory signal sequence and secretion was independent of ER/golgi. Therefore, it was hypothesized that MX1 regulates the secretion of proteins via one of the relatively uncharacterized "unconventional" secretory pathways. Research presented here characterized the mechanism of MX1 secretion and further described how MX1 may function in the uterine epithelial cell as a regulator of secretion.;The objectives of the first two studies were to identify proteins that interact with MX1. In the first approach, chemical crosslinkers were utilized to capture biologically relevant interactions in living ovine glandular epithelial cells (oGE) cells. Potential interacting proteins discovered in these experiments included cytoskeletal proteins tubulin and actin, actin-modifying protein profilin 1, heat shock proteins 83 and 90, and annexin 2. The identifications were based on single or double peptides using mass spectrometry, and need to be confirmed using alternate techniques. In the second experiment a recombinant protein was developed and used as bait on an affinity column incubated with oGE cell lysates. One protein interaction discovered using this technique was between MX1 and tubulin beta (TUBB). These results were confirmed using co-immunoprecipitation and co-immunofluorescence. Furthermore, MX1 interacted with TUBB during interphase and mitosis in oGE cells with or without IFN-tau. Collectively, these studies identified and confirmed the interaction between MX1 and the cytoskeletal protein TUBB, and also identified other candidate proteins that should be investigated further.;A third study examined the mechanism of MX1 secretion. Experiments were designed to test the hypothesis that MX1 was secreted via the multivesicular body (MVB)/exosome pathway in oGE cells. A population of exosomes containing MX1 protein was identified and it was hypothesized that MX1 is trafficking proteins, potentially along the microtubule network, to the site of exosome formation and is being incorporated into exosomes as a result. The objective of these studies was to characterize the MX1-containing population of exosomes to better understand the role of MX1 in their formation and secretion. Results showed that MX1 was contained inside the exosomes and, like another exosomal protein, Alix, was secreted via a ceramide-dependent pathway. It was also determined that MX1 was co-localized, at least partially, with the tetraspanin CD63 but not with MHC class 1 positive compartments, suggesting that there are multiple unique populations of exosomes secreted by these cells.;The experiments presented in this dissertation further defined the role of MX1 in unconventional secretion and provided clues to MX1 function through identification of protein interactions. Based on previous studies and those described here, the working hypothesis is that MX1 regulates unconventional secretion of proteins at the level of MVB and/or ILV formation in uterine epithelial cells. MX1 could be trafficking proteins that are required for formation and secretion of these vesicles, and this process would be highly upregulated during early pregnancy because of the high levels of MX1 expression. The free-floating preattachment embryo is completely dependent on uterine histotrophe (uterine secretions) and therefore MX1 could have a direct role in embryo survival. Since the results of these experiments suggest MX1 could be regulating the secretion of exosomes, it will be important to determine (1) what proteins are secreted associated with exosomes; and (2) their role in early pregnancy. Exosomes have been shown to have numerous functions depending on their cellular origin. Therefore it is likely that these uterine epithelial cell-derived exosomes have important biological functions during pregnancy, and MX1 may have a role in regulating these functions by controlling the release of exosomes from the cell.