Study of the molecular mechanism(s) underlying the antitumoral and highly pleiotropic functions of the human RNASET2 protein, a phylogenetically conserved extracellular RNase.

The RNASET2 gene encodes for the only human secreted acid ribonuclease of the T2 family. This gene maps in 6q27, a region that is consistently found rearranged in many solid and hematological tumors. Experimental data gathered in our laboratory have demonstrated the role of RNASET2 as a tumor suppressor gene which is endowed with several functions including chemotactic and possibly activating activities toward the monocyte/macrophage population. In a mouse xenograft model, we have recently reported that inoculation of human ovarian cancer-derived cells forced to overexpress RNASET2 led to the development of small, growth-suppressed tumors, characterized by a strong infiltrate of host-derived innate immunity cells, which have been identified as mainly M1 polarized macrophages. By contrast, control cells not expressing RNASET2, developed large, fast-growing tumors with no sign of macrophage infiltration. Further investigations proved that recombinant RNASET2 (produced in either baculovirus or Pichia pastoris expression system) displayed a marked chemotactic activity in vitro, most likely dependent on a subfamily of G-protein Coupled Receptors associated with inhibitory G protein. Taken together, these results led us to suggest that RNASET2-mediated in vivo tumor suppression is a non-cell-autonomous process which result in the recruitment of macrophages into the tumor mass. Based on such evidences, we decided to further investigate the relation between RNASET2 and cells belonging to the monocyte-macrophages lineage. Most of the experiments we planned to carry out involved the use of purified recombinant human RNASET2, hence, the first step of the project was the development of a reliable supply system. Two different sources of production and purification of human RNASET2 protein were already available in our lab (P. pastoris and BEVS), however, a full exploitation of these reagents was precluded by several limitations inherent to both systems. To overcome these limitation, we decided to focus our efforts on improving the P. pastoris expression system, which was more cost-effective and easier to handle in term of available facilities. To solve the problem, we added another 6XHis tag to the construct coding for the human RNASET2 and managed to significantly improve the recovery of the input protein. Considering the downstream applications of the purified human RNASET2, contamination of the preparation by metals (especially Ni, Cd, Zn) and endotoxin are important factor to control in order to properly assess a relationship between a specific treatment (in our case the addition of recombinant human RNASET2) and immune cell response. Applying the proper care in the preparation of the buffers and the handling of the sample we manage to achieve a low endotoxin content and an undetectable level of contamination by metals (Ni, Cd, Zn). As previously mentioned, several studies have reported the consistent tumor suppressive role for several members of the T2 RNase protein family and the recruitment of immune cells seems to be involved in this suppression. In order to improve our knowledge of these functional features of T2 RNases and at the same time to evaluate their evolutionary conservation, we tested the role of human recombinant RNASET2 in the activation and recruitment of immune cells by injecting recombinant human RNASET2 in the body wall of Hirudo verbena a useful model system for our purposes. After rRNASET2 injection, a significant increase in the production of collagen fibrils and a consequent remodeling of the muscle layers was observed. The resulting massive production of connective tissues is then used as a scaffold for immune cells migration and for proper orientation of growth of new vessels. Injection of human recombinant RNASET2 was also shown to induce a massive migration of cells belonging to the macrophages lineage (characterized as HmAIF-1+ and CD68+ cells) within 24 h, coupled to the formation of new blood vessels. The observed inflammatory response was specifically dependent of recombinant RNASET2 injection, since infiltrating macrophages and neo-vessel formation were not observed following injection of either PBS or rRNASET2 protein that was pre-incubated with a neutralizing anti-RNASET2 antibody. Taken together, these data strongly suggest that rRNASET2 injection in leeches induces a marked inflammatory response characterized by macrophage recruitment. The data gathered in our lab in several in vitro and in vivo experimental systems strongly suggest the occurrence of an RNASET2-based intercellular cross-talk, whose molecular mechanisms are nevertheless largely unknown. As a first step in the logical sequence of events culminating in the above mentioned RNASET2 biological activities, we hypothesized the occurrence of a putative receptor for RNASET2. The common idea behind all our approaches was to covalently link RNASET2 (either the endogenous protein or the human recombinant protein) to any interactor and, after purification of the complex, to perform a MS analysis to characterize the interactants. As a first approach, we carried out chemical modification of purified recombinant RNASET2 with different compounds (SDA, LC-SDA, Sulfo SDA, Sulfo LC-SDA) in order to add a diazirine ring to all primary ammines of the target protein. The resulting modified RNASET2 protein could then, in principle, be crosslinked to all electron donors available following UV light activation. The modified protein was then incubated with RNASET2-silenced OVCAR3 cells and analyzed by western blot following UV exposure. Despite several attempts with this approach, we could not observe any shift in the molecular weight of RNASET2. Considering that most of the result obtained show evidence of a tropism of RNASET2 for cells belonging to the monocyte/macrophage lineage, with our last approach we tested a promyelocytic cell line. We exposed the U937 cells preincubated with rRNASET2 to a chemical crosslinkers (BS3) able to covalently link two or more proteins in close proximity. By this approach we managed to see a shift in the molecular weight of the RNASET2, but unfortunately in the next step we couldn't purify any detectable complex. Considering that TAMs are usually forced by the tumor cells to acquire an M2-polarized pro-tumorigenic phenotype, a key finding of our research group is that tumor xenografts that are suppressed in their growth following RNASET2 overexpression display a strong infiltration of M1-polarized macrophages which are known to show anti-tumorigenic properties. Therefore, the last part of this PhD program was dedicated to investigate the molecular/cellular bases of the interaction between RNASET2 and immune cells, in particular to describe a possible involvement of this protein in driving macrophages polarization. The main idea was to develop a system to reliably produce M0, M1 and M2 macrophages that could then be used to test the effect of RNASET2 looking at their transcriptional profile. To this end, we chose a well-known cellular model (THP-1 cell line) and, in order to remove any bias from future analysis, we selected a pool of cells silenced for the expression of RNASET2. THP-1 is a promyelocytic cell line therefore in order to effectively produce polarized cells we needed to differentiate it into macrophages. Using the guidelines found in the literature we managed, by treating THP-1 cells with PMA 5 ng/ml for 48 h, to produce M0 macrophages. Subsequently, we developed a protocol to effectively drive naive macrophages to either M1 or M2 phenotype. We followed the suggested guidelines and assessed the efficacy of our system by testing a small panel of known marker of polarization (TNF, CXCL10, CCL19 for M1 polarization and CCL22, MRC1, MSR1 for M2 polarization). Considering the final aim of the developed assay, we tested whether our macrophages, especially M1 and M2, still retained the plasticity to change their profile according to external stimuli. M0 cells were exposed to both M1 and M2 stimuli; the resulting profile was neither M1 nor M2, suggesting that the cells prepared with our system still retained the characteristic/desired plasticity to properly respond to external signals. The data gathered so far support the idea that the system developed might be useful to study the involvement of RNASET2 in macrophages polarization. To summarize, during these three years of my PhD program I was involved in many different parts of a project aimed at deepen our understanding of the mechanism underlying RNASET2 tumor suppression. Although I wasn't able to identify any putative receptor for RNASET2 experimental evidence suggests the existence of a molecular mechanism involved in the “sensing” of the RNASET2 protein. Further studies are clearly needed in order to clarify this issue and the recent finding of RNASET2-containing exosomes is a promising avenue of investigation. The data collected in Hirudo strengthen our hypothesis of a close relationship between Immune cells (in particular cells belonging to monocyte/macrophages lineage) and RNASET2 but, more importantly these results open new research opportunities to study the involvement of T2 ribonucleases in inflammation. From a strictly technical point of view, I was able to improve the system of production and purification of the recombinant human protein, making it possible the use of RNASET2 in conditions previously inaccessible (interaction with immune cells/injection in organism). To conclude I contributed to laid the bases for future studies aiming at dissecting the involvement of RNASET2 in macrophages polarization and I did it by developing a simple assay that might test the "polarizing potential" of RNASET2 (but potentially of any compound).