Analysis of Molecular-Cellular Responses to Proteasome Inhibitors in Multiple Myeloma Patients; A Translational Approach of Proteasome Inhibitors In Vivo Effects from the Drosophila Experimental Model to Humans

Organisms require efficient surveillance of proteome functionality to prevent disruption of proteostasis. Central to the proteostasis ensuring network is the proteasome, which degrades both normal short-lived ubiquitinated proteins and damaged or mutated proteins. Over-activation of the proteasome seems to represent a hallmark of advanced tumors and thus, its selective inhibition provides a strategy for the development of novel anti-tumor therapies. This approach is applied in multiple myeloma (MM) that represents the second most common hematological malignancy. Specifically, proteasome inhibitors have demonstrated clinical efficacy in the treatment of MM and mantle cell lymphoma and are evaluated for the treatment of other malignancies. Nevertheless, the impact of proteasome dysfunction in normal human tissues (which relates to side effects in the clinic) remains poorly understood.

By using the fruit fly Drosophila melanogaster as an in vivo experimental platform to study proteasome physiology we found that proteasome functionality is sex-, tissue- and age-dependent. Oral administration of proteasome inhibitors (e.g. Bortezomib or Carfilzomib) in young flies suppressed proteasome activities in the somatic tissues; reduced motor function (recapitulating peripheral neuropathy of Bortezomib treatment in the clinic) and caused premature aging. It also increased oxidative stress and activated an Nrf2-dependent feedback regulatory circuit that upregulated proteasome genes in order to restore normal proteasome functionality. Moreover, in line with observations in the clinic, Carfilzomib was found to cause milder (as compared to Bortezomib) neuromusculatory toxicity and reduction of flies' lifespan.

To address the question whether these findings can be translated in humans we started characterizing proteasome physiology in both healthy donors, as well as in MM patients treated with therapeutic proteasome inhibitors. For our studies we used isolated red blood cells (RBCs; represent an anucleate relatively "long-lived" proteome) and peripheral blood mononucleated cells (PBMCs; represent cell lineages with active genomic responses). Our analyses in healthy donors of different ages revealed significant variability of basal proteasome peptidase activities in both cell types. PBMCs expressed (as compared to RBCs) higher basal proteasome activities and RBCs from females had higher chymotrypsin-like activity as compared to RBCs from males of similar age. Furthermore, as in the flies' somatic tissues, proteasome activities were found (independently of sex and cell type) to decline during aging.

Studies in RBCs and PBMCs isolated from MM patients treated with Bortezomib revealed donor-, cell type- and drug-specific readouts. In most (but not all) cases proteasome activities were suppressed in both cell types at 24-hrs post-drug administration. RBCs were particularly sensitive to the inhibitor and their proteasome activities remained low during the entire course of treatment. On the contrary, PBMCs were characterized by phases of rebound proteasome activities during the periods of no drug administration; these phases correlated with upregulation of proteasome genes expression, indicating that the feedback regulatory circuit which functions to restore proteasome activities in flies is also operational in humans. Additional gene expression analyses in PBMCs showed that proteasome inhibition also triggers the induction of genes involved in chaperon, autophagy, unfolded protein- and antioxidant-responses pathways; while, as in the fly model, the intensity of genes induction seems to decline during aging. Interestingly, in those patients who (despite treatment) showed no reduction of proteasome activities we found marginal gene expression alterations, suggesting that the observed gene induction largely depends on proteasome loss of function. Importantly, at the clinical level we observed a positive correlation between the degree of proteasome inhibition (in PBMCs or RBCS) and the depth of disease responses.

The similarities between the Drosophila pharmacological model and the MM patients indicate that the molecular responses to proteasome malfunction are largely conserved in higher metazoans. We foresee that our ongoing studies will support a more personalized clinical therapeutic approach in hematological malignancies.