Case Study 2: Understanding metabolic vulnerabilities in vivo

Summary

• Development of ‘Plasmax’ culture media to recapitulate physiological metabolite profiles

• Understanding how maintenance of intracellular metabolite levels supports growth of KRAS-driven colorectal cancer in vivo

• The use of modified diets to exploit tumour metabolic vulnerabilities in vivo

Strategic approach to identification and exploitation of the true metabolic vulnerabilities of cancers in vivo. 1) We have developed a culture medium – ‘Plasmax’ (marketed through Ximbio) - which accurately recapitulates metabolic microenvironments encountered by tumours in vivo and are collaborating with CRUK’s Functional Genomics Centre to screen for metabolic vulnerabilities of cancer cell lines; 2) Several studies at the Institute have combined GEMMs with state-of-the-art metabolomic imaging to dissect the mechanisms of cancer metabolic vulnerabilities in vivo; 3) our mechanistic studies have led to the establishment of ‘Faeth’, which is driving clinical trials in precision cancer nutrition.

Research over the last 10 years has identified metabolic vulnerabilities conferred by oncogenes, and how these may be targeted to oppose growth of transformed cells in culture. However, standard culture conditions do not faithfully recapitulate the metabolic microenvironment encountered by tumours in vivo. We have, therefore, adopted complementary approaches - including development of media to recapitulate physiological metabolite profiles and the use of disease-positioned GEMMs in combination with mass spectrometry imaging/in situ metabolomics - to identify and better understand the metabolic vulnerabilities of tumours in vivo. We have also identified in vivo metabolic vulnerabilities that may be targeted by modified diets, and these studies are informing the design of new clinical trials.

Tumour cells re-wire their metabolism to maintain energy supply and redox balance in the face of the metabolic stresses posed by oncogenic drivers. Although certain metabolic differences between normal and transformed cells were originally highlighted by Otto Warburg in the 1950s, it was not until the 2010s – when liquid chromatography-mass spectrometry metabolomic technology become more accessible - that details of oncogene-associated metabolic re-wiring became more established. Moreover, it was in this period that we began to understand why cancer cells might be more glycolytic, and how rapidly proliferating transformed cells maintain redox balance. Indeed, metabolomic studies of cultured cancer cells established a widely disseminated view that flux of carbons from glycolysis to the TCA cycle is not coordinated in tumours; rather tumour cells generate lactate from glycolysis whilst fuelling their TCA cycle with glutamine. This balances anaplerotic with cataplerotic flux and minimises metabolic stress in rapidly growing transformed cells. However, it is now becoming clear that the vulnerabilities that one would predict to be associated with such re-wired metabolism are not necessarily apparent in vivo and agents designed to target these vulnerabilities have not translated effectively into the clinic. The reasons for this are likely threefold: 1) the constitution of culture media used to study cancer cell metabolism does not represent the metabolic/nutrient environment encountered by cancers in vivo 2) the way in which oncogenes influence metabolism may differ markedly between cultured cancer cells and tumours in situ and 3) differences in the metabolism of different regions of tumours are not reflected by bulk metabolomic analyses of extracted metabolites. The CRUK Beatson Institute has led a series of studies to address these issues, thus defining and understanding how particular oncogenes drive tumour metabolism in vivo and, thereby, identifying the metabolic vulnerabilities of cancers in patients.

Plasmax – a physiologically-relevant medium to study tumour cell metabolism

By carefully characterising the nutrient and metabolite profiles of human plasma samples, the Tardito group have created a culture medium, named ‘Plasmax’, which accurately recapitulates the metabolic environment of tumours (1). Tumour cells display strikingly different metabolic vulnerabilities when cultured in Plasmax than they do in historic media, such as DMEM. Moreover, untargeted metabolic analyses conducted by the Tardito and Blyth labs revealed that breast cancer spheroids grown in Plasmax accurately recapitulate the metabolic profile of mammary tumours in vivo. Following this, Cancer Research Technologies registered the brand Plasmax™ in the UK and Europe, and it is now being distributed internationally through the CRUK repository Ximbio (now Cancertools.org). Moreover, a whole genome essentiality screen, performed in collaboration between the Tardito and Sansom labs and the CRUK’s Functional Genomic Centre in Cambridge, recently identified genes that are essential for cancer cell proliferation in the tumour-relevant environment provided by Plasmax, but not in DMEM. This collaboration is ongoing and currently investigating whether Plasmax enhances the predictive power of in vitro screens to identify in vivo metabolic vulnerabilities.

Understanding how maintenance of intracellular metabolite levels supports tumour growth in vivo

The Rosetta Cancer Grand Challenge consortium establishes close collaboration between the National Physics Laboratory, the Sansom lab and several other groups to develop and exploit mass spectrometry imaging (MSI) of metabolites. A principal scientific question being addressed by Rosetta is how specific oncogenic drivers influence metabolism in tumours in situ. In situ metabolic MSI, in combination with in-depth metabolomics, indicated that, when CRC tumourigenesis is driven by a combination of loss of APC and expression of mutant KRAS, the resulting tumours appear to consume particularly large quantities of glutamine (2). Conditional, tumour-specific knockout mice and organoid approaches then indicated that this is due to expression of the SLC7A5 antiporter that exports intracellular glutamine from cells to power the import of other amino acids thus maintaining protein synthesis and tumour growth. Consistently, KRAS mutations are associated with the CMS3 ‘metabolic’ CRC subtype, which has poor prognosis. This novel insight into the metabolic re-wiring driven by mutated KRAS in vivo demonstrates exciting possibilities for targeting specific amino acid antiporters in a cancer subtype with a currently poor prognosis.

Exploiting oxidative stress is now thought to be a potentially plausible route to identifying tumour metabolic vulnerabilities in vivo. The Murphy, Zanivan and Sansom labs have deployed GEMMs to identify the AMPK-related kinase, NUAK1 as an important component of the antioxidant stress response in CRC (3). Deletion of NUAK1 in established colorectal tumours leads their rapid regression. These findings suggest that NUAK1 inhibition represents an important metabolic vulnerability in CRC, and the combination of NUAK1 inhibition with treatments, such as chemo and radiotherapy that increase oxidative stress, are currently being evaluated.

Dietary approaches to expose cancer’s metabolic vulnerabilities

Serine/glycine restriction - Due to deficiencies in serine and glycine synthesis pathway components, such as phosphoglycerate dehydrogenase (PHGDH) and phosphoserine amino transferase-1 (PSAT1), many cancer cells are dependent on exogenous supplementation of these amino acids for growth (4). The Maddocks, Blyth and Sansom labs exploited this observation to show that dietary restriction of serine and glycine significantly improves survival in multiple preclinical models of cancer (including CRC and lymphoma), and this is further reinforced by targeting endogenous serine synthesis by administration of PHGDH inhibitors (4,5). Furthermore, combining dietary serine and glycine restriction with a genetic alteration that increases endogenous oxidative stress (to mimic chemo/radiotherapy) further improved survival. This work led to the founding of Faeth Therapeutics with the aim of translating this discovery science into the clinic. The company secured $3.25m in venture capital from Khosla Ventures and has now recruited its first staff and CEO (completed mid-2019). Oliver Maddocks is Head of Research for the company. In August 2020, Faeth Therapeutics completed its first human healthy volunteer study using amino acid modulation, and its first clinical trial in cancer patients opened in 2022.

Mannose supplementation - In vivo PET imaging consistently indicates that a high capacity for glucose uptake is a fundamental characteristic of many tumours. The Ryan lab sought to exploit this phenomenon by investigating the growth-suppressing characteristics of sugars that are imported into cells via the same transporters as glucose. This indicated that mannose, in contrast to other sugars, reduced tumour cell growth and it achieved this by inhibiting enzymes in proximal glucose metabolism, such as hexokinase and phosphoglycerate kinase (6). As mannose is not toxic to normal cells and is well-tolerated by humans and mice, the Ryan, Blyth and Sansom labs tested the ability of mannose supplementation to influence tumour growth. Indeed, mannose administration reduced growth of various tumour types including CRC, particularly when used in combination with chemotherapy. Furthermore, this study found that low levels of phosphomannose isomerase (PMI) predicts sensitivity to mannose supplementation. Observations that PMI levels are generally low in CRC have now led the Ryan lab to collaborate with Richard Wilson (University of Glasgow) to strategise a clinical trial to test the potential for mannose supplementation to improve outcomes in CRC.

References

1. Vande Voorde J et al. Improving the metabolic fidelity of cancer models with a physiological cell culture medium. Sci Adv. 2019; 5: eaau7314
2. Najumudeen AK et al. The amino acid transporter SLC7A5 is required for efficient growth of KRAS-mutant colorectal cancer. Nat Genet. 2021; 53: 16-26
3. Port J et al. Colorectal tumors require NUAK1 for protection from oxidative stress. Cancer Discov. 2018; 8: 632-47
4. Tajan M et al. Serine synthesis pathway inhibition cooperates with dietary serine and glycine limitation for cancer therapy. Nat Commun. 2021; 12: 366
5. Maddocks ODK et al. Modulating the therapeutic response of tumours to dietary serine and glycine starvation. Nature. 2017; 544: 372-6
6. Gonzalez PS et al. Mannose impairs tumour growth and enhances chemotherapy. Nature. 2018; 563: 719-23