1. Introduction.
Metabolic reprogramming has emerged as a hallmark in cancer, impacting gene expression, cell de-differentiation and tumour microenvironment (Pavlova & Thompson, 2016). It is well established that cancer cells must rewire the cellular metabolism to satisfy the demands of growth and proliferation, most notably by providing energy, reducing equivalents and building blocks while several metabolites exert a signalling functions promoting tumour growth and progression (Danhier et al., 2017). The exploration of cancer metabolism for clinical benefits is required to identify metabolic pathways that are limiting for tumour progression (Vander Heiden & DeBerardinis, 2017). Cholesterol is a major lipid that is crucial as a building block for membrane formation and protein structuration (Grouleff, Irudayam, Skeby & Schiott, 2015). It is also the precursor of steroid hormones, bile acids and oxysterols (Schroepfer, 2000). Cholesterol biosynthesis is a multi-step process (Nes, 2011) that is subject to homeostasis and finely regulated in cells (Luo, Yang & Song, 2020). In cancer cells several deregulations have recently emerged, opening up new therapeutic strategies (Huang, Song & Xu, 2020; Kuzu, Noory & Robertson, 2016).
Recent epidemiological studies have shown that breast cancer (BC) still represent the world leading female cancer in terms of incidence and mortality (Bray, Ferlay, Soerjomataram, Siegel, Torre & Jemal, 2018; Global Burden of Disease Cancer et al., 2019). Thus, there is an urgent need to find and validate new therapeutic targets in order to improve patient survival and tumour recurrence. BC is a heterogeneous pathology and several molecular BC subtypes have been described driving therapeutic treatments. We can distinguish the following three major subtypes: 1) Estrogen receptor positive breast cancers (ER(+)BC) are the most frequent BC and are treated with selective ER modulators (SERM) and aromatase inhibitors (AI). SERM block the mitogenic effects of 17β-oestradiol (E2) at the ER level, and AI inhibit E2 neosynthesis in BC (Jordan & Brodie, 2007). 2) ER(-)negative and HER2(+) BC are treated using anti-HER2 therapeutic antibodies that block the activation of HER2-dependent mitogenic pathways with or without conventional chemotherapy (Goldhirsch et al., 2011); 3) triple negative BC (TNBC) that do not express steroid hormone receptors and HER2 are treated by conventional chemotherapy with non-selective cytotoxic drugs (Goldhirsch et al., 2011).
Epidemiological population studies have identified links between cholesterol and cancer. Meta-analysis of clinical trials have shown an inverse relationship between circulating cholesterol levels and BC (Touvier et al., 2015), while hypercholesterolemia has been proposed as a risk factor for BC recurrence (Nelson, 2018), implying that cholesterol metabolism deregulations occured in BC and that targeting cholesterol metabolism deregulations may be of interest for BC treatment and chemoprevention (Garcia-Estevez & Moreno-Bueno, 2019).
At the molecular level, recent studies have shpwn that certain oxysterols display either tumour promoter but also tumour suppressor properties (Fig 1A). 27-hydroxycholesterol (27-HC) has been shown to stimulate ER(+)BC proliferation and invasiveness through the modulation of ER and LXRβ respectively (Nelson et al., 2013) . It has also been shown that the pro-metastatic action of 27-HC in mice required myeloid immune cell functions such as polymorphonuclear-neutrophils and γδ-T cells at distal metastatic sites (Baek et al., 2017). In mice, it has been shown that the CXCR2 receptor was involved in this effect (Baek et al., 2017; Raccosta et al., 2013; Raccosta, Fontana, Traversari & Russo, 2013) . It has been shown that other side-chain oxysterols displayed similar properties, possibly after sulfation by the sulfotransferase SULT2B1b (Moresco et al., 2018; Raccosta et al., 2013). These observations led to the proposal that combination therapies associating the inhibition of 27-HC biosynthesis at the 27-hydroxylase level (CYP27A1) and the use of ERα and LXR antagonist could increase the efficacy of treatments against ER(+)-BC (Nelson, 2018). In human, clinical studies from the EPIC-Heidelberg cohort showed that high level of circulating 27-HC were associated with a decreased BC risk in postmenauposal women suggesting that 27-HC could prevent BC in these cases (Le Cornet et al., 2020; Lu, Le Cornet, Sookthai, Johnson, Kaaks & Fortner, 2019). In contrast,  strategies aiming to target 27HC biosynthesis as well as its effectors, as proposed by Nelson et al, should be limited to ER(+)-BC patients after an endocrine therapy to protect them against BC recurrence .
Certain B-ring oxysterols such as 7-hydroperoxycholesterol and 5,6-epoxycholesterol (5,6-EC) (Fig.1A) have retained the attention of researchers during the last century as they are major autoxidation and photo-oxidation products of cholesterol (Smith, 1981; Smith & Johnson, 1989), and are suspected to be alkylating substances and thus possibly mutagenic and carcinogenic. These oxysterols were shown to induce mutagenicity in some yeast strains (Ansari, Walker, Smart & Smith, 1982; Smith, Smart & Ansari, 1979) and chinese hamster V79 cells (Chang, Jone, Trosko, Peterson & Sevanian, 1988; Peterson, Peterson, Spears, Trosko & Sevanian, 1988; Sevanian & Peterson, 1984; Sevanian & Peterson, 1986) in vitro. However in vivo tests failed to show any carcinogenic potencies for 5,6-EC (el-Bayoumy et al., 1996). Meanwhile, recent studies have revealed that 5,6-EC are involved in a metabolic branch clearly involved in carcinogenesis and identified new 5,6-EC metabolites with opposite properties regarding BC oncogenesis. 1) 5,6α-EC can give metabolites with antiproliferative and cancer cell redifferentiation properties: 5,6α-EC can be sulphated by the sulfotransferase SULT2B1b in BC cells to produce 5,6α-epoxy-cholesterol-3β-sulfate (5,6-ECS) (Fig 1A) and 5,6-ECS was shown to induce BC cell death and BC cell redifferentiation activitiesin vitro . In normal breast tissue 5,6α-EC was shown to be conjugated to histamine to give Dendrogenin A (Fig 1A,B), a steroidal alkaloid that displays tumour suppressive properties (de Medina et al., 2013; Poirot & Silvente-Poirot, 2018; Segala et al., 2017; Silvente-Poirot & Poirot, 2014) (Fig 2A). 2) 5,6α-EC can be transformed into a tumour promoter: 5,6-ECs were shown to give a secondary metabolite named oncosterone (6-oxo-cholestan-3,6- diol, cholestan-3,6-diol-6-one, OCDO) (Fig 1A,C) with tumour promoter properties in ER(+)BC and TN BC (Poirot, Soules, Mallinger, Dalenc & Silvente-Poirot, 2018; Silvente-Poirot, Dalenc & Poirot, 2018; Voisin et al., 2017) (Fig 2B).