Neurodegenerative diseases result from a failure to maintain proteostasis, and a major hallmark of neurodegeneration is the accumulation of aggregated protein deposits in the degenerating neurons of patients. In amyotrophic lateral sclerosis and frontotemporal dementia (ALS/FTD), one pathway by which proteins aggregate is through the mislocalization of specific nuclear RNA- binding proteins (RBPs) with prion-like domains (PrLDs) to the cytoplasm. Upon prolonged cytoplasmic mislocalization, RBPs with PrLDs aggregate, leading to loss-of-function and gain-of- toxicity consequences, both of which contribute to pathology. Recent work has established that nuclear-import receptors (NIRs) properly transport their cargo from the cytoplasm to the nucleus and act as molecular chaperones, preventing and reversing the self-association of their clients. Together, these activities allow NIRs to maintain proper folding and localization of many proteins. One NIR, Karyopherin β2 (Kapβ2), transports several RBPs with PrLDs that are associated with the development of ALS/FTD, including FUS and hnRNPA1. Kapβ2 is an effective chaperone against wild-type cargo proteins; however, when the proline-tyrosine nuclear-localization signal (PY-NLS) of its cargo is mutated, the affinity of the interaction between Kapβ2 and its cargo can be severely impaired. People with PY-NLS mutations in FUS develop severe juvenile forms of ALS. Here, to address the problem presented by disease-linked mutant cargo, we perform a structure-function analysis of Kapβ2 and use two strategies to potentiate Kapβ2 chaperone activity. First, we employ biochemical and in-cell assays to determine the chaperoning capacity of truncated forms of Kapβ2 and find that nuclear import and cargo-remodeling activities are closely linked. Then, we focus on the cargo-binding interface of Kapβ2 and utilize structural, computational, and evolutionary information to design engineered variants of Kapβ2 that chaperone disease-linked PY-NLS variants of FUS and hnRNPA1 more effectively than wild-type Kapβ2 in vitro, in cells, and in animal models of FUS-ALS. Finally, we take a distinct approach by altering the surface properties of Kapβ2 to enhance chaperone activity. The results presented here provide mechanistic insight into how Kapβ2 operates as a chaperone and lay the foundation for further development of NIR-based therapeutic strategies for neurodegenerative diseases.