Cardio-, hepato- and pneumoprotective effects of autophagy checkpoint inhibition by targeting DBI/ACBP

ABSTRACT

DBI/ACBP (diazepam binding inhibitor, also known as acyl coenzyme A binding protein), acts as a paracrine inhibitor of macroautophagy/autophagy. We characterized a monoclonal antibody neutralizing mouse DBI/ACBP (a-DBI) for its cytoprotective effects on several organs (heart, liver and lung) that were damaged by surgical procedures (ligation of coronary and hepatic arteries or bile duct ligation), a variety of different toxins (acetaminophen, bleomycin, carbon tetrachloride or concanavalin A) or a methionine/choline-deficient diet (MCD). In all these models of organ damage, a-DBI prevents cell loss, inflammation and fibrosis through pathways that are blocked by pharmacological or genetic inhibition of autophagy. The hepatoprotective effects of a-DBI against MCD are mimicked by three alternative strategies to block DBI/ACBP signaling, in particular (i) induction of DBI/ACBP-specific autoantibodies, (ii) tamoxifen-inducible knockout of the Dbi gene, and (iii) a point mutation in Gabrg2 (gamma-aminobutyric acid A receptor, subunit gamma 2; Gabrg2^F77I) that abolishes binding of DBI/ACBP. We conclude that a-DBI-mediated neutralization of extracellular DBI/ACBP mediates potent autophagy-dependent organ protection by on-target effects, hence unraveling a novel and potentially useful strategy for autophagy enhancement. “Autophagy checkpoint inhibition” can be achieved by targeting DBI/ACBP.

DBI/ACBP is a phylogenetically ancient intracellular protein that binds so-called activated fatty acids (acyl coenzyme A esters) to facilitate their transport and metabolism. The name DBI stems from its interaction with benzodiazepine receptors including the GABRG2/g2 subunit of GABR/GABA_A (gamma-aminobutyric acid (GABA) A receptor). Importantly, DBI/ACBP is a leaderless protein that can be released through an autophagy-dependent pathway and then acts on GABR to inhibit autophagy, thus participating to an inhibitory feedback loop that limits autophagy. This feedback loop or “autophagy checkpoint” (ACI) can be interrupted by knockdown or knockout of the Dbi gene, by neutralizing DBI/ACBP by means of autoantibodies or externally supplied antibodies, or by mutating the GABRG2/g2 subunit of GABR to abolish DBI/ACBP binding (Gabrg2^F77I). Hence, genetic removal of DBI/ACBP expression, neutralization of DBI/ACBP or mutation of its receptor all similarly enhance autophagic flux. In other words, ACI can be achieved by targeting DBI/ACBP. We have chosen this terminology, ACI, in analogy with immune checkpoint inhibition (ICI), which has become the backbone of most oncological treatments aiming to stimulate anticancer immune responses. ICI targets immunosuppressive ligand–receptor interactions (between CD274/PD-L1 and PDCD1/PD1 or between CTLA4 and its receptors) by suitable monoclonal antibodies against these proteins, thus releasing the break from the immune system, while ACI would target DBI/ACBP to release the break from autophagy.

One peculiarity of ACI targeting DBI/ACBP is the multiplicity of autophagy-inducing effects. Thus, beyond rapid autophagy induction achieved by neutralizing the interaction between DBI/ACBP and GABR/GABA_A expressed on multiple cell types, DBI/ACBP neutralization also reduces food intake and stimulates lipolysis with a consequent increase in free fatty acids and ketone bodies that can be expected to induce long-term effects favoring autophagy. We observed in the past that DBI/ACBP neutralization by means of a specific monoclonal antibody (a-DBI) prevents high-fat diet-induced obesity as well as adiposity, hepatosteatosis and type 2 diabetes (Figure 1A). However, from these experiments, it was not clear whether the liver-protective effects were secondary to the induction of autophagy, or the reduction of caloric uptake induced by DBI/ACBP. Driven by this consideration, we decided to investigate the capacity of a-DBI to prevent partially nonalcoholic steatohepatitis (NASH) induced by a protocol that usually causes weight loss (rather than weight gain) and that involves a hepatotoxic methionine/choline-deficient diet (MCD0. In this model, we observed that weekly a-DBI injections prevent NASH, as they partially antagonize the MCD-induced weight loss [1]. Conversely, inhibition of autophagy by daily injections of 3-hydroxychloroquine or knockout of ATg4b abolishes the anti-NASH effects of a-DBI. Hence, the hepatosteatosis-suppressive effect of a-DBI can be uncoupled from its anti-obesity action but appears to rely on autophagy induction. Accordingly, a-DBI induces the stigmata of autophagy in hepatocytes and reduces the local expression of pro-inflammatory and pro-fibrotic gene products, as well as the infiltration of the organ by macrophages (Figure 1B).

Figure 1.

Schematic effects of DBI/ACBP neutralization. (A) Anti-obesity effects. (B) Anti-steatohepatitis effects in the context of methionine/choline-deficient diet. (C) Other hepatoprotective effects. (D) Cardio- and pneumoprotective effects. Abbreviations: ACTA2: actin alpha 2, smooth muscle, aorta; ADGRE1/F4/80: adhesion G protein-coupled receptor E1; CAT: catalase; CCL2/MCP1: chemokine (C-C motif) ligand 2; CCl₄, carbon tetrachloride; CD68: CD68 antigen, ConA: concanavalin A; CPT1A: carnitine palmitoyltransferase 1a, liver; DBI/ACBP: diazepam binding inhibitor; HMOX1 (heme oxygenase 1; GPX: glutathione peroxidase; GSR: glutathione reductase; IL1B: interleukin 1 beta; IL6: interleukin 6; NFE2L2/NRF2: nuclear factor, erythroid derived 2, like 2; NASH: nonalcoholic steatohepatitis; NLRP3: NLR family, pyrin domain containing 3; PDGF: platelet derived growth factor; PPARA: peroxisome proliferator activated receptor alpha; PPARGC1A/PGC1A: peroxisome proliferator activated receptor, gamma, coactivator 1 alpha; SLC2A/GLUT: solute carrier family 2 (facilitated glucose transporter); SOD1: superoxide dismutase 1, soluble; SOD2: superoxide dismutase 2, mitochondrial; UCP1: uncoupling protein 1 (mitochondrial, protein carrier); TNF/Tnfα: tumor necrosis factor; VIM: vimentin.

Intrigued by these observations, we determined whether the effects of a-DBI would be truly on-target. For this, we used three alternative methods for blocking the DBI/ACBP system, namely (i) vaccination with adjuvanted DBI/ACBP to break self-tolerance and to induce DBI/ACBP-specific autoantibodies, (ii) ubiquitous tamoxifen-inducible knockout of exon 2 of the Dbi gene, and (iii) a point mutation in Gabrg2 (Gabrg2^F77I) that abolishes binding of DBI/ACBP. All three alternative methods similarly protect against MCD-induced NASH, supporting the idea that a-DBI acts solely on DBI/ACBP (rather than on additional cross-reactive targets) to achieve hepatoprotection (Figure 1B).

We also wondered whether a-DBI would have a broader hepatoprotective effect against mechanical insults (such as ischemia-reperfusion of the hepatic artery, which causes necrosis, and bile duct ligation, which leads to fibrosis) and toxins (such as acetaminophen and concanavalin A, which both trigger necrosis, and carbon tetrachloride which induces cell death and fibrosis). In all these conditions, a-DBI confers significant hepatoprotection that is reversed by 3-hydroxychloroquine injections, suggesting that it requires autophagy to be efficient. Moreover, a-DBI not only mediates prophylactic effects but also helps to reverse manifest MCD-induced NASH and carbon tetrachloride-induced liver fibrosis (Figure 1C).

In a final set of experiments, we determined the capacity of a-DBI to act on other organs than the liver. Indeed, we observed that a-DBI prevents the fibrosis of lungs induced by intranasal administration of bleomycin, coupled to a reduction in inflammation- and fibrosis-relevant genes. Moreover, a-DBI, which induces autophagy and mitophagy in the heart muscle, is capable of reducing the infarcted area of mice subjected to ligation of the left coronary artery. This cardioprotective effect of a-DBI is lost after cardiomyocyte-specific knockout of the essential autophagic gene Atg7, meaning that it depends on autophagy (Figure 1D).

In sum, in mouse models, ACI with a-DBI has broad cardio-, hepato- and pneumoprotective effects. Nonetheless, there are several challenges for the preclinical development of ACI. First of all, it remains to be determined whether other organs (e.g., gut, joints, kidney, retina, etc.) or systemic diseases (e.g., cystic fibrosis, lupus erythematosus, septic shock, etc.) that might be targeted by autophagy enhancers will profit from the cytoprotective and anti-inflammatory effects of a-DBI. Second, given that the a-DBI antibody used in our studies is mouse-specific (and hence does not recognize human DBI/ACBP), it will be important to develop a new series of monoclonal antibodies that target both mouse and human DBI/ACBP to increase the translational relevance of the studies.

Funding Statement

The work was supported by the Cancéropôle Île-de-France through a PhD Scholarship in Humanities and Social Sciences Fondation pour la Recherche Médicale Equipex Onco-Pheno-Screen European Joint Programme on Rare Diseases Gustave Roussy Odyssea European Union Horizon 2020 [Oncobiom]; European Union Horizon 2020 [Crimson]; Institut National du Cancer Institut Universitaire de France LabEx Immuno-Oncolo Cancer Research ASPIRE Award from the Mark Foundation RHU Immunolife Seerave Foundation SIRIC Stratified Oncology Cell DNA Repair and Tumor Immune Elimination SIRIC Cancer Research and Personalized Medicine Ligue contre le Cancer Agence Nationale de la Recherche

Disclosure statement

GK has been holding research contracts with Daiichi Sankyo, Eleor, Kaleido, Lytix Pharma, PharmaMar, Osasuna Therapeutics, Samsara Therapeutics, Sanofi, Sotio, Tollys, Vascage and Vasculox/Tioma. GK has been consulting for Reithera. GK is on the Board of Directors of the Bristol Myers Squibb Foundation France. GK is a scientific co-founder of EverImmune, Osasuna Therapeutics, Samsara Therapeutics and Therafast Bio. GK is the inventor of patents covering therapeutic targeting of aging, cancer, cystic fibrosis and metabolic disorders. Among these, patents were licensed to Bayer (WO2014020041-A1, WO2014020043-A1), Bristol Myers Squibb (WO2008057863-A1), Osasuna Therapeutics (WO2019057742A1), PharmaMar (WO2022049270A1 and WO2022048775-A1), Raptor Pharmaceuticals (EP2664326-A1), Samsara Therapeutics (GB202017553D0), and Therafast Bio (EP3684471A1).