The role of hematopoietic stem/progenitor cells (HSPCs) in the development of inflammation in non-alcoholic steatohepatitis (NASH)

Abstract

Non-alcoholic fatty liver disease (NAFLD) is a common chronic hepatic disease that affects about a quarter of the global population. Between 5 and 10% of patients with NAFLD develop non-alcoholic steatohepatitis (NASH), the inflammatory and progressive form that is characterized by liver cell injury/death and inflammation. NASH patients are at higher risks for developing liver fibrosis, cirrhosis, and hepatocellular carcinoma. Indeed NAFLD/NASH is the third leading indication for liver transplantation. NAFLD is also a systemic disease which is able to disrupt metabolic homeostasis and is an independent risk factor for cardiovascular disease and diabetes. Unfortunately, to date, there is no approved drug available for the treatment of patients with NAFLD or NASH. Hence, there is an urgent unmet need to understand the underlying mechanisms of disease in order to identify novel therapeutic targets. Many experimental and clinical data indicated that inflammatory circulating monocytes and monocyte-derived macrophages play a central role in the progression of both NASH and cardiovascular disease (CVD). Indeed, it is well established that the presence of NASH is an independent risk factor of CVD, though the mechanism underlying this effect is unknown. Importantly, multiple clinical trials have been initiated with the liver as the target organ, however, some of them have been terminated due to concerns of an increase in the risk of CVD. Hence, both the US Food and Drug Administration (FDA) and the American Association for the Study of Liver Disease (AASLD) has stated that novel NAFLD/NASH therapies should be at least neutral from a cardiovascular risk perspective and ideally also reduce the risk of CVD in NASH. Hence uncovering the link between these diseases can potentially provide new orthogonal therapeutic targets for NASH and possibly CVD. Orthogonal therapies by definition aim to fine-tune critical nodes involved in multiple related conditions, thus functioning as a rheostat. Bone marrow hematopoietic stem/progenitor cells (BM-HSPCs) are the primary source of myeloid cell production. The proliferation of HSPCs and myeloid-biased hematopoietic stem cells (clonal haematopoiesis) leads to greater myelopoiesis that is causally linked to the development of CVD and atherosclerotic plaque formation. Cholesterol accumulation in HPSCs can also stimulate HSPC proliferation and myelopoiesis. However, no study to date has reported on the relationship between liver pathology and the rate of bone marrow stem cell haematopoiesis during liver injury in NASH. The aim of this thesis was to elucidate the link between NASH progression, innate immunity and the hematopoietic system. In novel data, we were able to demonstrate cross-talk between the liver and the bone marrow hematopoietic system. Indeed, in multiple murine models, the induction of inflammation within the liver in NASH accelerated the rate of myeloid cell production (myelopoiesis) within the bone marrow. Feeding mice a cholesterol-rich (ChR) diet led to the infiltration of immune cells to the liver, the formation of inflammatory foci and higher levels of inflammatory cytokines and chemokines. Mass cytometry (Cytof) and flow cytometric analysis of the liver demonstrated broad infiltration of immune cells especially bone marrow-derived myeloid cells to the liver of this ChR diet-fed mice compared to those fed normal chow (NC), confirming the induction of a systemic inflammatory response. Subsequently, we investigated the profile of immune cells in primary sites such as the bone marrow (BM). Cytof on bone marrow confirmed that the direction of inflammatory changes that happens in the liver is replicated in the immune profile of BM with over-production of myeloid-derived cells at this site. Importantly, we detected the emergence of a population of myeloid cells in ChR diet fed mice which express markers of myeloid-derived cells that infiltrate liver. This enhanced production of immune cells in BM (myelopoiesis) has not been reported in NASH. We next examined the effects of the ChR diet on HSPC homeostasis. Interestingly, this demonstrated that the level of the HSPC population associated positively with the level of liver inflammation. The fact that the ChR diet-induced haematopoiesis might not be surprising, as cholesterol-related pathways in stem cells have been shown to be involved in haematopoiesis. To investigate whether cholesterol is the causal link between myelopoiesis and liver inflammation, we undertook studies using a methionine choline-deficient (MCD) diet and a diet supplemented with 0.1% 3, 5-diethoxycarbonyl-1, 4-dihydrocollidine (DDC). In both models, liver injury and inflammation were independent of cholesterol and mirrored many of the morphological features of NASH. In these experiments, we demonstrated that the mere presence of liver injury and an initial hepatic inflammatory response was sufficient to stimulate bone marrow stem cell proliferation and myelopoiesis. These findings demonstrate the existence of cross-talk between liver and bone marrow (BM) that could be a link to explain the simultaneous occurrence of NASH and CVD risk and CVD disease. To find an appropriate target that can reduce liver inflammation and restore hematopoietic homeostasis in BM (to treat NASH and its CVD complications), we first performed RNA-Seq on liver samples and built a biological network using weight gene co-expression network analysis (WGCNA). Biological networks represent a mathematical model of biological systems and their functions. Biological networks are modular which means a group of genes (modules) are involved in similar biological functions. Not surprisingly, we noticed upregulation of a module of genes related to inflammation and the immune response in ChR diet-fed mice. Screening of genes in this network indicated co-expression of genes related to the non-canonical NF-kB pathway and the enzyme pyruvate kinase M2 (PKM2). The latter is a rate-limiting enzyme of the glycolytic pathway. In addition, mining the literature in silico, we noticed that PKM2 plays a role in BM hematopoietic stem cell homeostasis and is hence a promising target. We confirmed the association of RelB of the non-canonical NF-kB pathway with PKM2 in publicly available human and mice datasets. It has been suggested that co-expressed genes in a module are usually co-regulated and are functionally relevant. Thus we investigated the co-regulation of these pathways in myeloid cells in vitro and in vivo in three models of liver injury in mice. To test the effect of modulation of PKM2 activity, we treated mouse bone marrow-derived macrophages (mBMDMs) and human monocytes derived macrophages (hMDMs) with TEPP-46 (which keeps PKM2 in the tetramer form with high glycolytic enzyme activity) and then stimulated the cells with LPS and palmitate as an inflammatory stimulus. Using qPCR, western blotting, ELISA and Seahorse we demonstrated the inhibition of inflammation in the cells treated with TEPP-46. Finally, we performed an intervention in three murine NASH models by oral gavage with TEPP-46 daily for 2 weeks. In these studies, we observed that in vivo modulation of PKM2 alleviates liver injury and restores homeostasis in the bone marrow and reduces inflammation in the liver. In conclusion, our results explore a novel concept of how NASH independently increases cardiometabolic risk. This link we believe at least partly is explained by: 1. Crosstalk between the liver and the bone marrow hematopoietic system that is initiated by the development of hepatic inflammation. 2. This hepatic inflammation and its persistence/progression are associated with alterations of the homeostatic balance in stem cell production, and differentiation of HSPCs towards the myeloid lineage (myelopoiesis) in the bone marrow. This we believe leads to greater production and delivery of inflammatory immune cells to peripheral organs such as the liver and the cardiovascular system, thereby perpetuating inflammatory injury in these tissues. 3. A metabolic regulator, the enzyme PKM2 is a unique modulator of this balance, creating a new class of orthogonal drugs to treat both NASH and cardiovascular disease. 4. Modulation of PKM2 activity can lessen inflammation both locally and can restore homeostasis of BM- hematopoietic stem cells. The following graphical abstract summarizes our findings: See legend on next page Graphical abstract 1: Interplay between the liver, the bone marrow hematopoietic system and immune cells during NASH progression. In response to injurious stimuli that result in NASH (such as a highly processed western diet), hepatic myeloid cells (especially Kupffer cells) become activated and switch their metabolism to glycolysis to yield lactate in the presence of dimeric PKM2 (a glycolytic enzyme). Dimeric PKM2 can translocate to the nucleus, induce the expression of RelB, and activate the non-canonical NF-kB pathway. As a result, there is a greater secretion of inflammatory cytokines (TNF-a, IL-1β, CCL2). This initial liver inflammation signals to the bone marrow to stimulate myeloid-biased hematopoietic stem cells (HSCs), increases myelopoiesis, and ultimately results in exacerbated liver inflammation. Bone marrow myelopoiesis is a critical link to CVD (atherosclerosis) due to the higher flux of myeloid immune cells to plaque. The enzyme PKM2 that is involved in glycolysis and gene transcription was identified as a key regulator of myelopoiesis Targeting PKM2 (by using TEPP-46) can dampen inflammation both locally and can restore normal bone marrow homeostasis of HSCs. Hence, PKM2 could be a potent orthogonal therapeutic target to treat NASH. Abbreviations: HSCs: Hematopoietic stem cells; HSPC: Hematopoietic stem and progenitor cell; GMP: granulocyte monocyte progenitor, PKM2: Pyruvate kinase M2, NF-kB: Nuclear factor kappa-light-chain-enhancer of activated B cells, CVD: Cardiovascular disease, TNF-a: Tumor necrosis factor-alpha, IL-1β: interleukin 1 beta, CCL2: Chemokine (C-C motif) ligand 2, RelB: avian Reticuloendotheliosis viral oncogene homolog B, TEPP-46: Thienopyrrolopyridazinone.