In a recent study published in Nature Communications, researchers engineered and characterized a blood-brain barrier (BBB) transport vehicle targeting the CD98 heavy chain (CD98hc, 4F2, or SLC3A2) of heterodimeric amino acid transporters (TVCD98hc).
The BBB poses a particular barrier in medication development for neurodegenerative illnesses by restricting drug concentrations. Platforms such as the transferrin receptor (TfR)-binding transport vehicle (TVTfR), which targets proteins heavily expressed on the BBB, might allow safe, effective, and non-invasive brain delivery of biotherapeutics.
CD98hc might be targeted to offer personalized brain delivery platforms with unique pharmacological properties for central nervous system (CNS) diseases.
About the study
In the present study, researchers developed a TVCD98hc- based platform for delivering biotherapeutics into the brain.
Cynomolgus monkeys and humanized CD98hc knock-in murine animals were used to test the pharmacokinetic and biodistribution features of a CD98hc antibody transport vehicle (ATVCD98hc). After chronic treatment, the researchers assessed brain uptake, ATVCD98hc clearance, biodistribution in the CNS, and safety.
The researchers also investigated ways in which CD98hc binding valencies, interaction with the Fc receptor (FcR), and antigen-binding fragments (Fab) affected these features. To construct antibody transport vehicles (ATVCD98hc) and investigate the possibility of CD98hc-regulated biotherapeutics’ delivery to the brain, an 11-amino acid library was designed and fused to Fabs.
The TV6.6-CD98hc interaction was structurally characterized, and TV involvement at cellular surfaces was simulated by superimposing the transport vehicle 6.6-CD98hc structure onto an existing CD98hc/L-type amino acid transporter 1 (LAT1) complex. ATVCD98hc variant localization was determined by immunohistochemistry (IHC).
The retention and trafficking of ATVCD98hc and ATVTfR were evaluated using human embryonic kidney (HEK)293 cells that express CD98hc and transferrin receptor endogenously to investigate potential mechanisms behind these differing in vivo behaviors.
To determine whether additional binding may influence CD98hc-mediated biodistribution and retention, Fabs that targeted and inhibited the activities of beta-secretase 1 (BACE1) protein were incorporated in the mono- and bi-valent ATVCD98hc.6.39 structures. ATVCD98hc:BACE1 variations were introduced intravenously (i.v.) into CD98hcmu/hu knock-in murine animals, and serological and brain pharmacokinetics were assessed.
To evaluate the brain pharmacokinetics of ATVCD98hc molecules and to simulate the temporal dynamics of ATVCD98hc molecules entering and departing the brain, mathematical modeling was used.
The team tested affinity maturation (AM)-type libraries for variations with increased affinity to cynomolgus and human CD98hc. The yeast surface display of CD98hc was used to select the libraries for enhanced cross-reactive binding.
To analyze TVCD98hc safety and exposure, 6.8:DNP molecules were produced with the presence and absence of mutations that reduce binding to FcRs. The quantities of ATVCD98hc in cell-associated fractions (including parenchymal and vascular cells) were compared to non-cell-associated fractions.
The clearance, uptake kinetics, and biodistribution features of ATVCD98hc molecules varied from those of the previously described ATVTfR. In comparison to other known Blood-brain-barrier platforms that targeted TfR, peripherally given ATVCD98hc displayed distinct brain delivery with much slower kinetic characteristics.
Specific biodistribution patterns throughout the parenchyma of the brain might be altered by inserting Fc amino acid substitutions on the CD98hc antibody transport vehicles that affect FcR engagement, modifying CD98hc binding valency, and varying the degree of target interaction with Fabs.
Humanized CD98hc knock-in mice and cynomolgus monkeys showed improved ATV delivery to the brain. ATVCD98hc variants have clearance values equivalent to control IgG after a single intravenous dosage.
The designed TV binding data agreed with structural determinants at the complicated interface. The cynomolgus and human CD98hc homologs share all CD98hc residues that interact with the TV segment on TV6.6. The affinity of TVs for Cynomolgus CD98hc was reduced by 3-10 fold when they were cross-reactive.
Although both monovalent and bivalent TVs could bind CD98hc on cell surfaces, bivalent ATVCD98hc:DNP demonstrated superior cell binding versus its monovalent counterpart. Tmax for ATVCD98hc variations in the brain was one week post-dose, with levels 6-8 fold greater than for control human immunoglobulin 1 (huIgG).
At three weeks post-dose, all variations’ concentrations were considerably higher (7-12 fold). BiATVCD98hc variants cleaned the brain slower than monoATVCD98hc variants, with biATVCD98hc concentrations lingering in the brain 14 days after serological concentrations were below the lower limit of quantification.
The cell-specific biodistribution of ATVCD98hc in the CNS was non-neuronal, limited to subsets of glial cells, and was modified by valency and FcR binding. When compared to control human IgG1, CD98hc or TfR binding increased ATV brain absorption. The persistence of biATVCD98hc in the brain after serological concentrations were undetectable indicated that retention, and absorption, might influence the brain exposure of CD98hc-binding antibody transport vehicles.
Plasma clearance was similar for monovalent and bivalent ATVCD98hc:BACE1 and ATVCD98hc:DNP, showing that BACE1 binding did not affect peripheral clearance. In the absence of Fab binding, ATVCD98hc retention in the brain triggered the extended brain exposures observed. Increased cell connection in the brain was caused by higher affinity and bivalent binding.
Overall, the study findings supported TVCD98hc as a modular brain delivery platform with favorable kinetic, biodistribution, and safety properties, enhancing brain uptake in cynomolgus monkeys.