In detail

Bruxelles Woluwe

In medicinal chemistry, one of the big challenges remains the discovery of an original hit that can be easily tuned into a lead and then in a drug candidate. In this regard, our aim is to develop an innovative fragment-to-lead strategy for high-quality lead identification. This computationally-assisted approach involves the initial discovery of low-molecular weight molecules called fragments. Owing to their small-size, fragments are more likely to reach key pockets within a protein active site, and, once their interaction within the active site is clearly understood, they represent a unique possibility of designing a promising hit compound in an efficient way.

The originality of this approach resides in the compilation of available experimental information about structural motifs recognized by the target, and their use to guide the selection of fragments from large chemical databases using a computationally-assisted screening. The interaction of chosen fragments with the target protein can then be experimentally assessed by means of biochemical and biophysical techniques, such as NMR, surface plasmon resonance (SPR), mass spectrometry (MS) and/or X-ray diffraction (XRD). Once the binding experimentally confirmed, rational drug design can start and the selected fragments can be finely tuned to provide an original hit. This original computationally-assisted fragment-to-lead strategy offers the prospect of a more efficient approach to drug discovery – resulting in the generation of high-quality leads with a better chance of success in future development.

Tryptophan catabolism is an important mechanism of peripheral immune tolerance contributing to tumoral immune resistance, and indoleamine 2,3-dioxygenases (IDO and TDO) inhibition is a promising strategy for anticancer drug development. IDO and TDO are unrelated heme-containing enzymes catalyzing the oxidative cleavage of the indole ring of L-tryptophan (L-Trp), the first and rate-limiting step along the kynurenine pathway. The implication of IDO in the phenomenon of tumoral immune resistance is the focus of intense researches and the enzyme is now recognized as a validated target for anti-cancer therapy.

Therefore, number of groups, including us, are actively searching for novel original IDO inhibitors. In contrast, the effect of TDO expression on the immune response has only been recently investigated in detail. Indeed, we showed in collaboration with the group of Prof Van den Eynde that TDO was effectively overexpressed by a number of human tumors and that this expression prevented rejection of tumor cells. We designed a novel TDO inhibitor and proved, in a preclinical model, the concept that TDO inhibition promotes tumoral immune rejection. Interestingly, blocking both TDO and IDO to improve the efficacy of cancer immunotherapy would be complementary: in a series of 104 human tumor lines of various histological types, we showed that 20 tumors expressed only TDO, 17 only IDO and 16 expressed both enzymes. Therefore, targeting both IDO and TDO would allow reaching 51% of tumors instead of either 32% or 35% with a compound inhibiting IDO or TDO alone, respectively. The design of IDO, TDO or dual IDO/TDO inhibitors is thus of major importance. Interestingly, our fragment-based drug design strategy recently provided promising results for the discovery of new IDO. These preliminary data are very encouraging to pursue the search for new anticancer agents through a fragment-based drug design strategy.


The last ten years have witnessed an increased regain of interest for tumor metabolism. Recent advances in this field have shed light on how tumors fuel rapid growth by preferentially engaging biosynthetic pathways. Although cellular metabolic pathways are rich pickings for drug targets, pinpointing enzymes that critically contribute to tumor metabolism is key to establish a therapeutic window since most of metabolic enzymes also play important roles in normal tissues.

1. PHGDH (3-phosphoglycerate dehydrogenase) and PSAT1 (phosphoserine aminotransferase-1) represent ideal targets for new anticancer strategies. These enzymes catalyze the first and second steps in the serine biosynthetic pathway, respectively. This pathway diverts a relatively small fraction of 3-phosphoglycerate from glycolysis to generate serine as well as equimolar amounts of NADH and α-ketoglutarate (αKG). Interestingly, two simultaneous recent reports have recently identified the serine pathway as a vital source of αKG to fuel the TCA cycle in a variety of tumor cells. These two studies further documented that serine supplement could not rescue tumor cells in which PHGDH and PSAT1 genes were knocked down, thereby identifying the serine pathway as a process providing malignant cells with critical amounts of its intermediary synthetic products, αKG and possibly NADH, instead of the end product, serine (that may also be taken up from the extracellular fluid). In good agreement with the above statement on the rationale to identify specific targets to tackle tumor metabolism, this latter observation indicates that serine deficiency in healthy tissues and possible disorders associated with the inhibition of either PHGDH or PSAT1 could be treated by exogenous serine supplement, whereas treatment with such inhibitory compounds could take advantage of the strict addiction of tumors to the by-products resulting from PHGDH and PSAT1 activation.

In this project, we aim to understand the role of the serine pathway in tumor progression and in particular to develop pharmacological tools to evaluate the extent of tumor addiction to this metabolic path and their therapeutic potential by exploring potential side effects on healthy tissues. To this end, novel innovative pharmacological inhibitors of PHGDH and PSAT1, the two main enzymes of the serine pathway (see Figure 1), were designed and chemically synthesized. These
compounds are currently optimized by chemical modifications and are expected to help deciphering the exact roles of these enzymes in cancer progression and insights on their physiological roles (that could represent limitations to the clinical use of such inhibitors).

2. Tumor cells are also characterized by a remarkable metabolic plasticity allowing them to survive and proliferate in hypoxic and extracellular acidic environments. In tumor cells, this plasticity allows the coexistence and coordination of several metabolic phenotypes, leading to an optimal use or resources. Hypoxic cells uses glucose that is metabolized by anaerobic glycolysis. Lactate is secreted and diffuses, and can be subsequently used by oxygenated tumor cells as a preferred energetic source to glucose. The lactate oxidative pathway requires the entrance of lactate in oxidative cells via a process that is mainly facilitated by the Monocarboxylate Transporter MCT1 and the oxidation of lactate to pyruvate by the lactate dehydrogenase B (LDHB). The pyruvate can then fueled the Krebs cycle and NADH uses the malate-aspartate shuttle to directly fuel the mitochondrial respiration chain. The oxidative use of lactate in the oxygenated tumor compartment therefore optimizes the availability of glucose for cells of the hypoxic compartment, thus constituting a unique metabolic cooperation. If the use of lactate by oxidative cells is a proven fact, the advantage it gives them remains largely unknown. A first series of studies showed that lactate can act as a proangiogenic agent. This signaling activity also depends on the oxidation of lactate to pyruvate by LDHB, allowing pyruvate to inhibit enzymes of the prolylhydroxylase family and activate the hypoxia-inducible transcription factor factor-1 (HIF-1) independently of hypoxia. In addition, a recent collaborative led with the team of P. Sonveaux (IREC) has shown that the oxidative use of lactate promotes autophagy, ie, a process of degradation and recycling of proteins and organelles requiring formation of specialized structures, autophagosomes, and their fusion with lysosomes. To promote autophagy, LDHB physically interacts with VATPase, a proton pump located on the surface of lysosomes, which it feeds with the protons produced during the lactate + NAD+ -->/<--pyruvate + NADH + H+ reaction. This observation seems important to us as autophagy participates in tumor progression by recycling damaged proteins and organelles when cancer cells are exposed to oxidative stress, and because it provides cells with energy substrates under metabolic stress conditions.

All these observations suggest that LDHB may be a new target in cancer therapy. However, there is currently no specific inhibitor of this enzyme, and the consequences of systemic inhibition of LDHB activity remain largely unknown.

In this project our aim is thus to develop and validate a peptide inhibitor and a non-peptide inhibitor to selectively inhibit tetramerization of LDHB. Our strategy will involve the use of Protein-Protein Interaction Inhibitor (PPI) identification methods that is, a highly multidisciplinary approach involving molecular modeling studies (identification of "Hot Spots"), biochemical studies (in vitro and in vivo inhibition of LDHB tetramerization, selectivity study) and biophysical studies (nuclear magnetic resonance analysis of ligand-LDHB interaction). To achieve the goal of a selective inhibition of LDHB, we will use an innovative strategy targeting the tetramerization site of LDHB rather than the active site of the enzyme.

Regulated cell death is necessary for tissue homeostasis, embryonic development, immunity and other biological processes. Nevertheless, dysregulated cell death leads to pathological developments especially in neurodegenerative diseases. On the other hand, cancer cells tend to avoid death in order to enhance proliferation.

In the past decade, multiple ways of regulated cell death were identified and characterized. Among them, Ferroptosis was discovered through the study of a genotype selective antitumor agent, erastin, a small molecule able to kill cancer cells without presenting any morphological or biochemical features of known cell death pathways. Instead, ferroptotic dying cells show a mitochondrial disruption associated with high levels of free intracellular iron and lipid hydroperoxides.

Although the exact mechanism leading to cell death is not yet fully understood, this lipid peroxidation phenomenon initiates the process and is considered as the main hallmark of ferroptosis (see Fig). Free intracellular iron plays a crucial role in redox mechanism of ferroptosis by catalyzing Fenton reaction (see Fig, yellow pathway). Thus, an increase in free iron levels mediated by a dysregulation of its transporting or stocking directly contributes to the accumulation of lipid peroxides.

Once those lipid hydroperoxides are formed, only the reducing action of the glutathione peroxidase 4 (GPX4) is able to prevent their damaging outcome (see Fig, green pathway). Because GPX4 uses Glutathione (GSH) as cofactor, a depletion of the pseudo-tripeptide and/or cysteine, the limiting amino acid in the GSH synthesis, causes an increase in lipid peroxides levels involved in the cell death initiation.

Interestingly, erastin induces ferroptosis by inhibiting the Xc- antiporter. Ras synthetic lethal compounds (RSL) are other pharmacological tools currently used to induce ferroptosis by directly inhibiting GPX4. On the contrary, ferroptosis can be prevented by several different ways. For example, allosteric activators of GPX4 were identified to enhance the oxidative stress controlling activity of the peroxidase. It is also possible to inhibit ferroptosis by decreasing the intracellular free iron levels of iron with metal chelators such as deferiprone.

More recently, a genome-wide recessive genetic screen has pointed to an essential role for Acsl4 in ferroptosis. This gene codes for the acyl-CoA synthetase long chain family member 4 (ACSL4) which catalyzes the ligation of a Coenzyme A (CoA) on a long chain polyunsaturated fatty acid (see Fig, blue pathway). By preferentially activating arachidonic and adrenic acids, ACSL4 shapes cellular lipid composition and dictates ferroptosis. Indeed, part of the acyl-CoA formed are esterified into membrane phospholipids by LPCAT3 where they can be oxidized (for example by the lipoxygenases) and then contribute to the lipid peroxides pool necessary for the initiation of the cell death. It was shown that the extinction of ACSL4 or its inhibition by glitazones was enough to prevent ferroptosis pharmacologically induced.

Ferroptosis was studied in few experimental Ferroptosis was studied in few experimental models. Nevertheless, this regulated cell death seems to have a real therapeutic interest. Indeed, several studies have highlighted the potential benefit of inducing ferroptosis in cancer therapy. For example, it was shown that sorafenib, a multi-kinase inhibitor mainly used in renal carcinoma, induces ferroptosis in cancer cells by disrupting system Xc-. Simultaneously, ferroptosis studies have widened their scope and it seems particularly relevant that preventing ferroptis is promising therapeutic strategy in a neurodegenerative context. This idea is supported by the FAIRPARK-II project, a clinical assay that study the role of DFP in the neurodegeneration slow-down in Parkinson’s disease.

In this new project, we are thus developing novel pharmacological tools to study ferroptosis and particularly ACSL4 inhibitors.