The properties of mammalian cells depend on their location within organs. Gene expression in the liver varies between periportal and pericentral hepatocytes^1,2,3, and in the intestine from crypts to villus tips^4,5. A key element of tissue spatial organization is probably metabolic, but direct assessments of spatial metabolism remain limited. Here we map spatial metabolic gradients in the mouse liver and intestine. We develop an integrated experimental–computational workflow using matrix-assisted laser desorption/ionization (MALDI) imaging mass spectrometry (IMS), isotope tracing and deep-learning artificial intelligence. Most measured metabolites (>90%) showed significant spatial concentration gradients in the liver lobules and intestinal villi. In the liver, tricarboxylic acid (TCA)-cycle metabolites and their isotope labelling from both glutamine and lactate localized periportally. Energy-stress metabolites, including adenosine monophosphate (AMP), also localized periportally, consistent with a high periportal energy demand. In the intestine, the TCA intermediates malate (tip) and citrate (crypt) showed opposite spatial patterns, aligning with higher glutamine catabolism in tips and lactate oxidation in crypts based on isotope tracing. Finally, we mapped the fate of the obesogenic dietary sugar fructose. In the intestine, oral fructose was catabolized faster in the villus bottom than in the tips. In the liver, fructose-derived carbon accumulated pericentrally as fructose-1-phosphate and triggered pericentral adenosine triphosphate (ATP) depletion. Thus, we both provide foundational knowledge regarding intestine and liver metabolic organization and identify fructose-induced focal derangements in liver metabolism.

Neuroblastoma is a highly lethal childhood tumour derived from differentiation-arrested neural crest cells^1,2. Like all cancers, its growth is fuelled by metabolites obtained from either circulation or local biosynthesis^3,4. Neuroblastomas depend on local polyamine biosynthesis, and the inhibitor difluoromethylornithine has shown clinical activity⁵. Here we show that such inhibition can be augmented by dietary restriction of upstream amino acid substrates, leading to disruption of oncogenic protein translation, tumour differentiation and profound survival gains in the Th-MYCN mouse model. Specifically, an arginine- and proline-free diet decreases the amount of the polyamine precursor ornithine and enhances tumour polyamine depletion by difluoromethylornithine. This polyamine depletion causes ribosome stalling, unexpectedly specifically at codons with adenosine in the third position. Such codons are selectively enriched in cell cycle genes and low in neuronal differentiation genes. Thus, impaired translation of these codons, induced by combined dietary and pharmacological intervention, favours a pro-differentiation proteome. These results suggest that the genes of specific cellular programmes have evolved hallmark codon usage preferences that enable coherent translational rewiring in response to metabolic stresses, and that this process can be targeted to activate differentiation of paediatric cancers.

Despite decades of study, large parts of the mammalian metabolome remain unexplored¹. Mass spectrometry-based metabolomics routinely detects thousands of small molecule-associated peaks in human tissues and biofluids, but typically only a small fraction of these can be identified, and structure elucidation of novel metabolites remains challenging^2,3,4. Biochemical language models have transformed the interpretation of DNA, RNA and protein sequences, but have not yet had a comparable impact on understanding small molecule metabolism. Here we present an approach that leverages chemical language models^5,6,7 to anticipate the existence of previously uncharacterized metabolites. We introduce DeepMet, a chemical language model that learns from the structures of known metabolites to anticipate the existence of previously unrecognized metabolites. Integration of DeepMet with mass spectrometry-based metabolomics data facilitates metabolite discovery. We harness DeepMet to reveal several dozen structurally diverse mammalian metabolites. Our work demonstrates the potential for language models to advance the mapping of the mammalian metabolome.