Research
We develop experimental and computational approaches to achieve a predictive understanding of metabolism — how it is controlled, why it is organized the way it is, and how it breaks down in aging and disease.
Predictive Models of Human Cell Metabolism
Cells run thousands of chemical reactions simultaneously, and these reactions must be precisely controlled — like a thermostat that prevents overheating. When this control fails, diseases including diabetes (uncontrolled glucose production), cardiovascular disease (uncontrolled cholesterol biosynthesis), and fatty liver disease (uncontrolled fatty acid biosynthesis) result.
We are building large-scale biophysical models of human metabolic pathways — computational representations that can predict how metabolism behaves under different conditions. Decades of biochemistry have produced rich data on how individual metabolic enzymes are regulated by small molecules and posttranslational modifications. Yet the function of most of these regulators at the pathway level remains unknown — it was simply not feasible to systematically remove them one by one until modern genome editing. We integrate this enzyme kinetics data into models that predict what each regulator does for the pathway as a whole, and test these predictions experimentally. Our ultimate goal is to develop a model that will accurately predict human metabolism in any cell type, under any condition, and in response to any perturbation — from a drug treatment to a change in diet.
Using this approach, we have shown that allosteric regulation of hexokinase and phosphofructokinase by ATP, ADP, phosphate, and glucose-6-phosphate is specifically required to maintain high ATP levels and prevent runaway accumulation of glycolytic intermediates, while glycolysis rate is controlled by ATP demand through mass action (Choe et al. 2025). This result overturns the textbook view that these regulators control the rate of glycolysis; the corresponding paper has been selected as the 2025 Biophysical Journal Paper of the Year. In a separate study, we found that fructose-2,6-bisphosphate (F26BP) performs a distinct function: it controls the ratio between glycolytic and respiratory ATP production independently of cellular energy state. This enables cells to adjust glycolytic flux independent of energy state in response to hormones and biosynthetic needs. (Kober et al. 2026).
Selected publications related to this research direction:
1) Choe M, Einav T, Phillips R, Titov DV. Glycolysis model shows that allostery maintains high ATP and limits accumulation of intermediates. Biophysical Journal
2) Kober MM, Yang X,Titov DV
Design Principles of Cell Metabolism
Why is cell metabolism organized the way it is? Cells have access to multiple routes for producing energy and building blocks, yet consistently favor specific pathways under specific conditions. For example, rapidly growing cells — including cancer cells — rely heavily on glycolysis for ATP production even when oxygen is available, a phenomenon known as the Warburg effect. Similarly, the same glycolytic enzymes are allosterically regulated by the same metabolites in organisms as distant as bacteria and human cells, yet why evolution selected these particular enzymes and regulators rather than others remains unclear. Understanding why cells adopt these metabolic strategies, rather than just how they regulate them, is essential for a complete picture of cell metabolism.
We use a combination of computational modeling and quantitative experiments to uncover the design logic behind metabolic organization. Our work has shown that the Warburg effect arises because glycolysis produces ATP faster per gram of pathway protein than respiration (Kukurugya et al. 2024), and that this speed advantage is what allows glycolytic cells to grow rapidly (Kukurugya et al. 2025). By asking “why this pathway or regulatory mechanism instead of that one?” we aim to identify the fundamental constraints — speed, efficiency, resource competition — that shape how cells build and control their metabolic networks.
Selected publications related to this research direction:
1) Kukurugya MA, Rosset S, Titov DV. The Warburg Effect is the result of faster ATP production by glycolysis than respiration. Proc Natl Acad Sci U S A 2024 Nov 8;():e2409509121.
2) Kukurugya MA, Zhang S, Ha BT, Ekvik AE, Titov DV. Glycolytic ATP production enables rapid mammalian cell growth.
Molecular Basis of Aging
What are the biochemical mechanisms that control how long an organism lives? Research over the past several decades has identified dietary (e.g., caloric restriction), genetic (e.g., insulin-like growth factor signaling), and pharmacological (e.g., rapamycin) interventions that robustly extend lifespan across species — yet the downstream processes responsible remain poorly understood.
We use the nematode C. elegans as a powerful model system to identify the evolutionarily conserved biochemical processes that control lifespan downstream of these classical interventions. We observed that aging worms exhibit a dramatic accumulation of lipoproteins called vitellogenins that are homologs of human apolipoprotein B – the causal driver of cardiovascular disease. Caloric restriction prevents this buildup, and reducing vitellogenin levels alone extends lifespan by over 60%, with no additional benefit from caloric restriction, suggesting that caloric restriction extends lifespan in part by inhibiting vitellogenesis. This observation suggests that diet- and age-dependent accumulation of lipoproteins is a conserved mechanism of aging in humans and C. elegans. By identifying such conserved mechanisms, we aim to lay the groundwork for rationally designed interventions to extend human healthspan.
Selected publications related to this research direction:
1) Yang B*, Manifold B*, Han W*, DeSousa C, Zhu W, Streets A*, Titov DV*. SRS microscopy identifies inhibition of vitellogenesis as a mediator of lifespan extension by caloric restriction in C. elegans.
Tools for Manipulation of Metabolism
Diet and exercise change cells from the inside — shifting the balance of key energy currencies like ATP, NADH, and NADPH. But until recently, there was no way to change these metabolic parameters directly and ask: what does each one actually do?
To fill this gap, we have developed a suite of genetically encoded tools — molecular machines that directly alter specific intracellular metabolic parameters. These include ATPGobble (ATP/ADP), LbNOX (NADH/NAD+), TPNOX (NADPH/NADP+), and UCP1 (mitochondrial membrane potential). We are expanding this toolkit to additional metabolic parameters, allowing us to systematically dissect how specific metabolic changes — like those induced by exercise or fasting — affect cell behavior, and to use these measurements to build and refine our computational models of metabolism.
Selected publications related to this research direction:
1) Ekvik AE, Kober MM, Titov DV. Genetically encoded tool for manipulation of ATP/ADP ratio in human cells.
2) Choe M, Ekvik AE, Stalnaker G, Shin HR, Titov DV. Genetically encoded tool for manipulation of ΔΨm identifies its role as the driver of ISR induced by ATP synthase dysfunction.
3) Choe M, Titov DV. Genetically encoded tools for measuring and manipulating metabolism. Nature Chemical Biology. 2022 May;18(5):451–460.
4) Cracan V*, Titov DV*, Shen H, Grabarek Z, Mootha VK. Genetically encoded tool for manipulation of NADP+/NADPH ratio. Nature Chemical Biology. 2017 Oct;13(10):1088-1095.
5) Titov DV*, Cracan V*, Goodman RP, Peng J, Grabarek Z, Mootha VK. Complementation of mitochondrial electron transport chain by manipulation of the NAD+/NADH ratio. Science. 2016 Apr 8;352(6282):231-5.