Research report: Ambrisentan ameliorates methotrexate nephrotoxicity through modulation of RhoA/ROCK and PI3K/MDM2-associated signaling: network pharmacology-guided discovery and in vivo confirmation

Methotrexate is one of the most widely used drugs in modern medicine, deployed against cancers ranging from leukemia to osteosarcoma as well as against rheumatoid arthritis and other inflammatory diseases. Yet its usefulness has always carried a dangerous caveat: the drug can injure the kidneys, sometimes severely enough that treatment must be interrupted. Now, a research team from Egypt and the United Arab Emirates reports that ambrisentan, a drug already approved for pulmonary arterial hypertension, substantially protected rats from methotrexate-induced acute kidney injury when given before and during exposure to the chemotherapy agent. The study, published in Frontiers in Pharmacology, combined computational network pharmacology with an in vivo rat experiment to trace the protection to specific molecular signaling pathways.

The rationale rested on a known biochemical villain. Methotrexate stimulates the expression of endothelin-1, a potent vasoactive peptide produced mainly by endothelial cells and also by renal epithelial and mesangial cells. Excess endothelin-1 drives pro-inflammatory, pro-oxidative and pro-fibrotic responses in the kidney, and blocking its effects has already been shown to protect against ischemia-reperfusion-induced acute kidney injury in experimental models. Ambrisentan is a highly selective antagonist of the endothelin type A receptor, the molecular dock where endothelin-1 latches onto cells to trigger downstream damage. The team hypothesized that blocking this receptor might counteract endothelin-1-mediated suppression of PPAR-γ, a nuclear receptor transcription factor that suppresses oxidative stress and inflammation and thereby helps control renal injury.

Before touching any animals, the researchers ran a computational screen. They pulled molecular targets for ambrisentan from five databases and targets for methotrexate from four, then intersected the methotrexate target list with disease targets for nephrotoxicity and renal tubular dysfunction retrieved from GeneCards, Open Targets, CTD, MalaCards and OMIM. After removing duplicates, 2,542 targets were associated with methotrexate-induced kidney injury and 208 with ambrisentan; the two sets overlapped in 102 unique targets. Uploading these to the STRING database produced a protein-protein interaction network of 101 nodes connected by 872 edges. Filtering by degree and ranking by betweenness and closeness yielded 20 hub targets, including IL6, TNF, IL1B, TP53, CASP3, BCL2, NFKB1, ICAM1 and PPARG. Notably, the authors are transparent that PPARG was not the highest-degree node in the network; it was selected for experimental assessment as a biologically plausible target based on its inclusion in the overlapping set and its connectivity, in line with the pre-specified hypothesis about endothelin and PPAR-γ.

Kyoto Encyclopedia of Genes and Genomes pathway enrichment of the hub genes pointed toward RhoA/ROCK and PI3K-associated signaling, alongside inflammatory and apoptotic markers. The top enriched KEGG pathways included lipid and atherosclerosis, TNF signaling, and AGE-RAGE signaling in diabetic complications, with strong linkage between RhoA, ROCK, PI3K, ICAM-1, PPARG, NF-κB, Bax and Bcl-2. This mechanistic map set the stage for the wet-lab validation: if ambrisentan truly protects the kidney through endothelin receptor blockade, the predicted downstream cascades should shift in measurable ways in treated animals.

For the in vivo phase, twenty-four male albino rats were randomly divided into four groups of six. The control group received the vehicle, 0.5% carboxymethyl cellulose, orally for ten days. An ambrisentan-only group received the drug at 5 mg/kg/day orally for the same period. A methotrexate group received vehicle and a single intraperitoneal injection of methotrexate at 20 mg/kg on day seven. The final group received oral ambrisentan daily for ten days plus the same single methotrexate injection on day seven. Blood was collected on day eleven for renal function assays, and kidney tissue was fixed for histology or frozen for biochemical analysis. Crucially, ambrisentan was given prophylactically, beginning before the methotrexate dose, a design the authors explicitly frame as proof of concept for prevention rather than treatment of established injury.

The protective effect was unmistakable. Under the microscope, kidneys from methotrexate-treated rats showed distorted renal corpuscles with widened capsular spaces, congested intraglomerular capillaries, cytoplasmic vacuolation in the epithelial lining of proximal and distal convoluted tubules, blood vessel congestion and marked lymphocytic infiltration. Periodic acid-Schiff staining revealed thickened basement membranes and a disrupted or absent apical brush border in proximal tubules, the hallmark of proximal tubular injury, with the optical density of the PAS reaction falling by 0.6-fold versus controls. In the ambrisentan co-treatment group, renal architecture largely returned to normal, PAS optical density rose 1.6-fold compared with methotrexate alone, and tubular injury scores improved markedly. All image analyses were performed by two investigators blinded to treatment groups.

Blood chemistry told the same story. Methotrexate raised serum urea roughly 2.2-fold and creatinine about 2.3-fold relative to controls; ambrisentan co-treatment cut urea by about half and creatinine by about 40% relative to the methotrexate group. Kidney injury biomarkers moved in parallel: methotrexate elevated KIM-1 and NGAL by roughly four-fold and five-fold, respectively, and co-treatment reduced KIM-1 by 70% and NGAL by 45%. The oxidative balance of the tissue also shifted. Methotrexate increased malondialdehyde, a lipid peroxidation marker, by 4.7-fold while cutting superoxide dismutase-1 activity by 0.7-fold; ambrisentan lowered MDA by 0.6-fold and boosted SOD activity 1.9-fold compared with methotrexate alone. Ambrisentan by itself left all these parameters essentially at control levels.

The molecular readouts aligned with the network predictions. Methotrexate elevated renal ICAM-1 and NF-κB protein levels approximately 3.2-fold and 3.1-fold while depressing PPAR-γ by 0.7-fold; ambrisentan co-treatment lowered ICAM-1 and NF-κB by about 0.6-fold and 0.7-fold and restored PPAR-γ toward normal. RhoA and its downstream kinase ROCK1, which endothelin-1 is known to activate through the ETA receptor, rose about 1.9-fold and 3.2-fold after methotrexate and fell by roughly 50% and 60% with ambrisentan. PI3Kα and MDM2, which had dropped by 36% and to 0.4-fold of control after methotrexate, were restored by 45% and to 1.61-fold of methotrexate levels, respectively. The authors caution that they did not directly measure Akt or MDM2 phosphorylation, so the activation status of the PI3K/Akt/MDM2 axis remains inferred rather than demonstrated.

Apoptosis assays sealed the case. Immunohistochemistry showed methotrexate raised the mean area of Bax immunopositivity nine-fold and slashed Bcl-2 by 0.8-fold, a shift in the Bax/Bcl-2 ratio consistent with mitochondrial apoptotic signaling, mitochondrial outer membrane permeabilization and caspase activation. p53 immunoreactivity in tubular cells and glomeruli climbed to nearly twenty times control levels, and cleaved caspase-3, the executioner of apoptosis, reached about 41-fold of control. Ambrisentan co-treatment reduced Bax immunopositivity by 0.9-fold, lifted Bcl-2 almost 3.6-fold, cut p53 immunoreaction to about 7% of the methotrexate level and reduced cleaved caspase-3-positive cells by approximately 96%. The proposed mechanism ties these threads together: blocking the ETA receptor dampens RhoA/ROCK signaling, which is known to relieve inhibition of the PI3K/Akt pathway; activated Akt phosphorylates MDM2, promoting p53 degradation and thereby sparing tubular cells from apoptosis.

The authors are careful about what this does and does not prove. This is a small rat study with six animals per group, using a single methotrexate dose and a prophylactic ambrisentan regimen; it did not test ambrisentan after injury was established, did not assess whether the drug alters methotrexate exposure, renal elimination or antitumor efficacy, and did not examine chronic or fibrotic outcomes. In clinical practice, high-dose methotrexate nephrotoxicity is strongly influenced by drug precipitation, clearance, hydration status and pharmacokinetics, so any chemotherapy-adjunct claim would require studies of drug-drug interactions and preserved antineoplastic activity, ideally in models where treatment begins after kidney injury appears. Still, as a first demonstration that a selective endothelin A receptor antagonist can blunt methotrexate nephrotoxicity through coordinated effects on PPAR-γ, RhoA/ROCK and PI3K/MDM2 signaling, the findings offer a coherent mechanistic framework and a plausible candidate for repurposing, pending the rigorous follow-up work the authors themselves call for.

Subject of Research: Repurposing the endothelin A receptor antagonist ambrisentan to prevent methotrexate-induced acute kidney injury

Article Title: Ambrisentan ameliorates methotrexate nephrotoxicity through modulation of RhoA/ROCK and PI3K/MDM2-associated signaling: network pharmacology-guided discovery and in vivo confirmation

Article References: Baraka, S. A., Eissa, N., Nagib, M. M., Elmaghraby, A. M., El-Sheikh, A. A. K., & Darwish, S. F. (2026). Ambrisentan ameliorates methotrexate nephrotoxicity through modulation of RhoA/ROCK and PI3K/MDM2-associated signaling: network pharmacology-guided discovery and in vivo confirmation. Frontiers in Pharmacology, 17, Article 1976309. https://doi.org/10.3389/fphar.2026.1976309

Image Credits: AI Generated

DOI: 10.3389/fphar.2026.1976309

Keywords: methotrexate, ambrisentan, acute kidney injury, nephrotoxicity, endothelin-1, PPAR-gamma, RhoA/ROCK, PI3K/MDM2, network pharmacology, rat study, drug repurposing, apoptosis