CEACAM1 (biliary glycoprotein-1, CD66a) was reported as a strong clinical predictor of poor prognosis in melanoma. We have previously identified CEACAM1 as a tumor escape mechanism from cytotoxic lymphocytes. Here, we present substantial evidence in vitro and in vivo that blocking of CEACAM1 function with a novel monoclonal antibody (MRG1) is a promising strategy for cancer immunotherapy. MRG1, a murine IgG1 monoclonal antibody, was raised against human CEACAM1. It recognizes the CEACAM1-specific N-domain with high affinity (K D $ 2 nmol/L). Furthermore, MRG1 is a potent inhibitor of CEACAM1 homophilic binding and does not induce any agonistic effect. We show using cytotoxicity assays that MRG1 renders multiple melanoma cell lines more vulnerable to T cells in a dose-dependent manner, only following antigen-restricted recognition. Accordingly, MRG1 significantly enhances the antitumor effect of adoptively transferred, melanoma-reactive human lymphocytes using human melanoma xenograft models in severe combined immunodeficient/nonobese diabetic (SCID/NOD) mice. A significant antibody-dependent cell cytotoxicity response was excluded. It is shown that MRG1 reaches the tumor and is cleared within a week. Importantly, approximately 90% of melanoma specimens are CEACAM1 þ , implying that the majority of patients with melanoma could be amenable to MRG1-based therapy. Normal human tissue microarray displays limited binding to luminal epithelial cells on some secretory ducts, which was weaker than the broad normal cell binding of other anticancer antibodies in clinical use. Importantly, MRG1 does not directly affect CEACAM1 þ cells. CEACAM1 blockade is different from other immunomodulatory approaches, as MRG1 targets inhibitory interactions between tumor cells and late effector lymphocytes, which is thus a more specific and compartmentalized immune stimulation with potentially superior safety profile.

Cells of hematopoietic origin can be subdivided into cells of the lymphoid lineage and those of the myeloid lineage, among which are myeloid-derived suppressor cells (MDSCs). The MDSCs can be further divided into CD11bLy6GLy6C monocytic (Mo) MDSCs and CD11bLy6GLy6C polymorphonuclear (PMN) MDSCs. Both subtypes support tumor growth and suppress anti-tumor immunity. Their accumulation at the tumor site includes mobilization from the bone marrow to the blood followed by colonization at the tumor site. The present study examines the mechanism by which PMN-MDSCs are mobilized from the BM to the blood to later accumulate at the tumor site. We show that the chemokine receptor CCR5 is a key driver of this event. We also show that, beyond chemoattraction, the interaction between CCR5 and its ligands promotes the proliferation of CCR5 PMN-MDSCs at the BM and, later, potentiates their immune-suppressive activities at the tumor site in part by inducing arginase-1.

We previously proposed a mechanism for control of Rubisco expression and assembly during oxidative stress in Chlamydomonas reinhardtii. The N terminus of the large subunit (LSU) comprises an RNA recognition motif (RRM) that is normally buried in the protein, but becomes exposed under oxidizing conditions when the glutathione pool shifts toward its oxidized form. Thus, de novo translation and assembly of Rubisco LSU stop with similar kinetics and the unpaired small subunit (SSU) is rapidly degraded. Here we show that the structure of the N-terminal domain is highly conserved throughout evolution, despite its relatively low sequence similarity. Furthermore, Rubisco from a broad evolutionary range of photosynthetic organisms binds RNA under oxidizing conditions, with dissociation constant values in the nanomolar range. In line with these observations, oxidative stress indeed causes a translational arrest in land plants as well as in Rhodospirillum rubrum, a purple bacterium that lacks the SSU. We highlight an evolutionary conserved element located within a-helix B, which is located in the center of the RRM and is also involved in the intramolecular interactions between two LSU chains. Thus, assembly masks the N terminus of the LSU hiding the RRM. When assembly is interrupted due to structural changes that occur under oxidizing conditions or in the absence of a dedicated chaperone, the N-terminal domain can become exposed, leading to the translational arrest of Rubisco LSU. Taken together, these results support a model by which LSU translation is governed by its dimerization. In the case that regulation of type I and type II Rubisco is conserved, the SSU does not appear to be directly involved in LSU translation.

Material