Randall-type heavy chain deposition disease (HCDD) is a rare disorder characterized by glomerular and peritubular amorphous deposits of a truncated monoclonal immunoglobulin heavy chain (HC) bearing a deletion of the first constant domain (CH1). We created a transgenic mouse model of HCDD using targeted insertion in the immunoglobulin κ locus of a human HC extracted from a HCDD patient. Our strategy allows the efficient expression of the human HC in mouse B and plasma cells, and conditional deletion of the CH1 domain reproduces the major event underlying HCDD. We show that the deletion of the CH1 domain dramatically reduced serum HC levels. Strikingly, even with very low serum level of truncated monoclonal HC, histologic studies revealed typical Randall-type renal lesions that were absent in mice expressing the complete human HC. Bortezomib-based treatment resulted in a strong decrease of renal deposits. We further demonstrated that this efficient response to proteasome inhibitors mostly relies on the presence of the isolated truncated HC that sensitizes plasma cells to bortezomib through an elevated unfolded protein response (UPR). This new transgenic model of HCDD efficiently recapitulates the pathophysiologic features of the disease and demonstrates that the renal damage in HCDD relies on the production of an isolated truncated HC, which, in the absence of a LC partner, displays a high propensity to aggregate even at very low concentration. It also brings new insights into the efficacy of proteasome inhibitor-based therapy in this pathology.