Multiple myeloma (MM) is characterized by dissemination and accumulation of plasma cells in the bone marrow (BM), which promotes tumor cell growth and therapy resistance. ROBO1 is a conserved transmembrane receptor of the Ig superfamily with no intrinsic catalytic activity, and its role in MM pathogenesis is unknown.
Material and Methods
We first analyzed ROBO1 expression via western blot and/or immunohistochemistry (IHC). Gene expression profiling in a cohort of 170 newly diagnosed MM patients (IFM170) was used to compare ROBO1 expression across primary MM and BM stroma cells (BMSC), and normal BM plasma cells (PC). We used short hairpin RNA (shRNA) for stable ROBO1 knock down (KD) and CRISPR-Cas9 for ROBO1 knock out (KO). For protein structure-function and rescue studies, ROBO1 KO MM cells were transduced with a lentiviral vector expressing either full-length (FL) or truncated ROBO1 mutants devoid of extracellular (Cyt) or intracellular domain (DeltaCyt), including patient-derived truncating mutations, with a C-terminus triple FLAG tag. FLAG immunoprecipitation (IP) followed by mass spectrometry or western blotting and immunofluorescence (IF) were used to identify ROBO1 interacting partners and ROBO1 cellular localization. We used a hydrogel encapsulation system to study proliferation in a 3D system. To study extramedullary and intramedullary MM growth in vivo, WT and ROBO1 KO OPM2 were injected either subcutaneously (plasmacytoma model) or intra-medullary in femoral bones of donor mice which were then implanted subcutaneously in recipient SCID mice (µ-SCID model). PET-CT was used to assess tumor volume. Mouse tumors were retrieved for IHC and RNA extraction followed by RNA sequencing. To study dissemination and homing, KO and FL addback OPM2 cells were injected intravenously in SCID mice. Femurs and plasmacytoma were retrieved at endpoint for IHC.
ROBO1 is highly expressed in human MM cell lines and primary MM cells with highest expression in cells carrying the high risk t(4;14) cytogenetic and low/absent expression in normal PC. Of human cancer cell lines, ROBO1 expression was limited to late B cell lineage; and ROBO1 KD was selectively cytotoxic against MM, but not other hematologic cancers. ROBO1 KO significantly decreases proliferation in a 3D culture system and tumor growth in extramedullary (mean tumor volume KO versus WT plasmacytoma: 457 versus 1323 mm3, p value= 0.02) and intramedullary (mean tumor volume KO versus WT: 823 versus 2684 mm3, p value= 0.001) murine models of human MM. ROBO1 KO MM cells show decreased adhesion to BM endothelial and BMSC, which is fully rescued by FL ROBO1 addback. To address whether ROBO1 loss alters dissemination/homing of MM cells in vivo, we injected mice intravenously with ROBO KO or FL addback OPM2 cells. While ROBO1 KO resulted in a modest, non-statistically significant prolongation in mouse OS (90 versus 75 days, respectively, p value 0.2), the pattern of disease was strikingly different. As expected, ROBO1 FL mice developed hindlimb paralysis with extensive BM infiltration with MM. Importantly, ROBO1 KO mice demonstrated reduced BM infiltration and developed solitary plasmacytoma. We next showed that ROBO1 C-terminus domain is necessary and sufficient to rescue ROBO1 KO proliferative defect while expression of ROBO1 truncations, including patient-derived frameshift mutations, acted as dominant negative. IP showed avid interaction of ROBO1 with ABL1. Interestingly, we showed that the cytosolic domain of ROBO1 undergoes cleavage and translocates to the nucleus, where its function is now being studied.
We show that ROBO1 is necessary for MM homing to the BM niche and for MM growth within and outside the BM space. ROBO1 cytosolic domain undergoes proteolytic cleavage and translocates to the nucleus and is necessary and sufficient to rescue ROBO1 KO defective proliferation. Based on our data, we propose a dual model for ROBO1 in MM: the full transmembrane receptor is involved in regulating adhesion, dissemination and homing of MM cells within the BM niche; the cleaved intracellular C-terminus domain participates in transcriptional regulation, promoting MM proliferation. These data suggest that ROBO1 C-terminus may be a novel molecular target in MM.
Roccaro:Celgene: Membership on an entity's Board of Directors or advisory committees; Celgene: Membership on an entity's Board of Directors or advisory committees; Janssen: Membership on an entity's Board of Directors or advisory committees; Janssen: Membership on an entity's Board of Directors or advisory committees; Associazione Italiana per al Ricerca sul Cancro (AIRC): Research Funding; Associazione Italiana per al Ricerca sul Cancro (AIRC): Research Funding; European Hematology Association: Research Funding; Transcan2-ERANET: Research Funding; European Hematology Association: Research Funding; AstraZeneca: Research Funding; Transcan2-ERANET: Research Funding; Amgen: Membership on an entity's Board of Directors or advisory committees; Amgen: Membership on an entity's Board of Directors or advisory committees; AstraZeneca: Research Funding. Ghobrial:Takeda: Consultancy; Sanofi: Consultancy; Amgen: Consultancy; BMS: Consultancy; Celgene: Consultancy; Janssen: Consultancy. Anderson:Sanofi-Aventis: Other: Advisory Board; Bristol-Myers Squibb: Other: Scientific Founder; Oncopep: Other: Scientific Founder; Amgen: Consultancy, Speakers Bureau; Janssen: Consultancy, Speakers Bureau; Takeda: Consultancy, Speakers Bureau; Celgene: Consultancy, Speakers Bureau.
Asterisk with author names denotes non-ASH members.