Hypothesis

Trans‑crocin 4 functions as a brain‑penetrant prodrug that is hydrolyzed extracellularly by neuronal β‑glucosidase, releasing crocetin which then inhibits prolyl‑hydroxylase domain enzymes (PHDs) and stabilizes HIF‑1α in hippocampal neurons, thereby shifting metabolism toward glycolysis and conferring neuroprotection in aging.

Mechanistic Rationale

The glycosylated crocin (trans‑crocin 4) crosses the BBB after intraperitoneal administration, whereas its aglycone crocetin does not appear in brain parenchyma despite high plasma levels[trans‑crocin 4 crosses the BBB post-i.p. in mice]. This disparity suggests that the intact glycoside is the species that reaches the CNS. Once inside the extracellular space, neuronal or glial β‑glucosidase (e.g., GBA2) can cleave the glucose moieties, generating free crocetin locally. Crocetin’s polyene structure can chelate Fe²⁺ required by PHDs, mimicking the action of known HIF‑1α stabilizers such as dimethyloxalylglycine[crocetin’s iron‑chelating capacity]. Stabilized HIF‑1α drives transcription of glycolytic enzymes (GLUT1, LDHA) and angiogenic factors (VEGF), supporting neuronal energy homeostasis under age‑related mitochondrial decline.

This mechanism explains why phenotypically similar neuroprotection is observed for both crocin and crocetin in vitro[trans‑crocin 4 reduces BACE1, APP‑C99, PSEN1/2 and tau via GSK3β/ERK] while in vivo brain exposure diverges: the glycoside supplies the active aglycone precisely where it is needed, avoiding peripheral sequestration by gut microbiota.

Testable Predictions

1. Inhibition of neuronal β‑glucosidase will abolish the HIF‑1α‑stabilizing effect of trans‑crocin 4 in hippocampal slices, without affecting its antioxidant activity.
2. Direct application of crocetin to the extracellular medium will increase HIF‑1α protein levels and HIF‑dependent reporter activity in neurons, an effect blocked by iron supplementation.
3. Aged mice receiving trans‑crocin 4 will show elevated hippocampal HIF‑1α target gene expression (e.g., Vegfa, Glut1) compared with vehicle, whereas equimolar crocetin administered intraperitoneally will not.
4. CRISPR‑mediated knockout of GBA2 in forebrain neurons will reduce crocetin formation after trans‑crocin 4 treatment and diminish the glycolytic shift measured by lactate production.

Experimental Approach

• Prepare acute hippocampal slices from young and aged mice; treat with trans‑crocin 4 (± β‑glucosidase inhibitor conduritol B epoxide) and measure HIF‑1α protein by western blot and HIF‑responsive luciferase activity.
• Use LC‑MS/MS to quantify intracellular crocetin levels after glycoside treatment, confirming intracellular generation.
• In vivo, administer trans‑crocin 4 or crocetin to aged mice (n=8/group) and harvest hippocampi after 6 h for qPCR of HIF‑1α targets and lactate assay.
• Validate with GBA2 conditional knockout mice (Camk2a‑Cre; Gba2^fl/fl) to test necessity of neuronal deglycosylation.

If these predictions hold, the hypothesis establishes a clear pharmacokinetic‑pharmacodynamic link: BBB‑permeable trans‑crocin 4 serves as a depot that yields crocetin locally to engage the HIF‑1α metabolic switch, reconciling the BBB paradox and offering a mechanism distinct from the established amyloid‑tau pathways.