FABP1 emerges as a missing variable in CYP2C9 drug metabolism models
Drug clearance models can miss badly when a liver protein is left out of the equation. In this preprint, FABP1, the cytosolic liver fatty acid binding protein, moved from background biology to a measurable determinant of CYP2C9-mediated metabolism in human liver systems.
The work focused on two classic CYP2C9 probe substrates, diclofenac and tolbutamide. Both are widely used to interrogate pathway behavior, which makes them useful stress tests for any model that claims to predict human hepatic clearance. Investigators found that the presence of FABP1 altered the kinetics of metabolite formation in human liver microsomes, a result that points to more than passive co-residence inside the hepatocyte. FABP1 appears to shape the way CYP2C9 handles substrate availability, access, or protein-level interaction inside the crowded intracellular environment.
One sentence captures the practical implication: a liver model that ignores FABP1 can misread CYP2C9.
Kinetic shifts with diclofenac and tolbutamide were not explained by CYP2C9 alone
The central hypothesis was straightforward: systematic overprediction or underprediction of metabolic clearance from human liver microsomes may reflect an incomplete account of how cytosolic proteins interact with drugs and cytochrome P450 enzymes. Rather than treating CYP2C9 as an isolated catalytic unit, the analysis tested whether FABP1 changes the enzyme’s apparent behavior in a way that matters for real-world prediction.
Using diclofenac and tolbutamide, the investigators observed that the best-fitting kinetic model included FABP1-CYP2C9 protein-protein interactions. That is an important mechanistic shift. It suggests the metabolic phenotype is not simply a function of enzyme abundance, but also of molecular proximity and intracellular trafficking within the hepatocyte. FABP1 is positioned to influence lipid handling and ligand movement in the cytosol, which can alter access to membrane-associated CYP enzymes and, in turn, change metabolite formation rates.
The biology is plausible enough to matter at the bedside, even if this work remains preclinical. Drug metabolism is not determined only by what a liver expresses; it is also shaped by how intracellular proteins organize the path from substrate entry to enzymatic oxidation. For drugs with narrow therapeutic windows or steep exposure-response curves, a missed modifier can distort dose selection and pharmacokinetic expectations.
Proteomic data linked FABP1 and CYP2C9 expression to hydroxylation output
The study extended beyond a pair of probe experiments. Statistical modeling across a panel of 61 human liver S9 fractions examined proteomic expression alongside drug hydroxylation. In that broader dataset, CYP2C9 and FABP1 expression together best predicted probe substrate hydroxylation.
That pairing matters because it moves FABP1 from a mechanistic curiosity to a quantitative predictor. A single-enzyme model can explain only so much variation when hepatic metabolism depends on the concentration of both catalytic and noncatalytic proteins. Here, the cytosolic binding protein added predictive value alongside the enzyme itself, implying that variability in human drug handling may reflect combined expression patterns rather than CYP2C9 abundance in isolation.
This is the kind of result that changes how pharmacology models are built. A clearance estimate based only on enzyme content assumes the liver behaves like a linear catalytic chamber. Human hepatocytes are not that simple. Subcellular binding partners, intracellular partitioning, and protein-protein interactions can all shift the apparent metabolic rate of a substrate before it ever reaches the enzyme active site.
Physiologically based liver models improved after FABP1 was added
The most clinically relevant result came when FABP1 expression and protein-protein interactions were incorporated into new physiologically based liver models. Those revised models improved the quantitative accuracy of in vivo hepatic clearance prediction.
This matters because model error in hepatic clearance can cascade into dose miscalculation, especially during drug development or when translating preclinical metabolism data into first-in-human expectations. A model that better reflects intracellular reality can reduce the risk of overconfident predictions about exposure, half-life, and metabolic route dominance. For CYP2C9 substrates, FABP1 may be one of the missing pieces that explains why conventional human liver microsome systems sometimes systematically overshoot or undershoot true clearance.
The work also speaks to a larger point in drug metabolism science: cytosolic proteins are not mere scaffolding. They can influence enzymatic output in ways that alter the shape of the kinetic curve and the accuracy of downstream prediction. The current findings suggest that FABP1 belongs in the same conversation as CYP2C9 when the goal is to model human hepatic metabolism with useful precision.
Why this preprint matters for translational pharmacology
CYP2C9 handles many clinically important drugs, and diclofenac and tolbutamide are standard tools for studying that pathway. When a preclinical model improves after adding FABP1, the result reaches beyond one enzyme-substrate pair. It raises the possibility that other hepatic pathways may also depend on cytosolic binding proteins that have been underweighted in current frameworks.
There is no claim here that FABP1 replaces CYP2C9 as the driver of metabolism. The enzyme still performs the oxidation. The new insight is that substrate handling in the human liver is shaped by a broader protein environment than many models currently allow. In practical terms, this can influence how hepatic clearance is interpreted from human liver microsomes and how physiologically based pharmacokinetic models are parameterized.
The study is a preprint and has not undergone peer review, so the findings should be read as a strong mechanistic and modeling signal rather than a settled standard. Even so, the direction is clear: for CYP2C9-mediated metabolism, ignoring FABP1 leaves a measurable gap. The next question is how broadly that gap extends across other cytochrome P450 pathways and whether other cytosolic proteins produce the same kind of hidden distortion in human liver prediction models.