UC San Diego

The secretory pathway is made up of a collection of mostly ER- and Golgi-resident proteins that work together to synthesize, post-translationally modify, transport and quality control the secreted proteins (secPs). This complex system can be leveraged to produce a variety of recombinant proteins such as antibodies, growth factors, and many other biotherapeutics. While the production yield of many recombinant proteins has seen a several-fold increase; many proteins still fail to express recombinantly despite bioprocess optimization. We start this doctoral dissertation by investigating the properties limiting production of difficult-to-express rProteins via analysis of proteogenomic data from popular mammalian cell hosts expressing different human proteins.

Decades of research has now better charted the secretory pathway, and the functional roles of individual proteins are better understood. However, many secPs experience production bottlenecks within the secretory pathway to various degrees. Incidentally, the nature of secP synthesis has been shown to be highly product-specific. As protein-protein interactions (PPIs) are one of the major modalities through which machinery proteins in the secretory pathway assist protein secretion, we charted the transient interaction partners of key secPs using proximity-based biotinylation and mass spectrometry analysis to better understand the rate-limiting steps unique to each secP within the secretory pathway. Additionally, we determined and verified the interactions that contribute to high recombinant protein yields via structural interaction modeling.

Perturbations to the secretory pathway result in misfolded proteins. Amyloid disorders, such as Alzheimer’s disease (AD), involve aggregation of secPs. However, it is largely unclear how secretory pathway proteins contribute to amyloid formation. We integrated expression data with PPI networks to estimate a tissue’s fitness for producing specific secreted proteins, and analyzed the fitness of the secretory pathway of various brain regions and cell types for synthesizing the AD-associated amyloid-precursor protein (APP). We associated Aβ aggregation with systemic dysregulation of the secretory pathway components proximal to APP and amyloidogenic secretases in AD. Our analyses suggest that perturbations from AD risk loci cascade through the APP secretory support network and into the endocytosis pathway, connecting amyloidogenesis to dysregulation of secretory pathway components supporting APP and suggesting novel therapeutic targets for AD.