Results

Predicted potential effects of CTPs on the BBB and passive diffusion penetration properties

The potential BBB impairment of CTPs

The structural formulas of the six CTPs were submitted to the ProTox 3.0 toxicity-prediction platform with all available predictive models activated, and data were visualized to generate the corresponding radar plots (Figure 2). The predictive outcomes suggest that all six CTPs possess the potential to interact with multiple receptors concurrently, and display particularly notable effects on the BBB. Within the ProTox 3.0 framework, receptors such as matrix metalloproteinase (MMP), peroxisome proliferator-activated receptor γ (PPAR-γ), the aryl hydrocarbon receptor (AhR), estrogen receptor alpha (ERα), estrogen receptor ligand-binding domain (ER-LBD), androgen receptor (AR), pregnane X receptor (PXR), constitutive androstane receptor (CAR), and CYP3A4 are currently recognized as key predictors of the potential of xenobiotics to penetrate the BBB. Moreover, all six CTPs exhibit toxicity probabilities exceeding 74% for eight of these receptors, excluding ERα.

Figure 2: Radar plots of predicted toxicity profiles for six CTPs(A) HPCTP, (B) PFPCTP, (C) HCCTP, (D) HFCTP, (E) EPFCTP, and (F) HMCTP. Each radar plot illustrates the probability scores across multiple pharmacological activity classes, where the red sector represents the molecule-specific prediction and the green region denotes the activity range for known active compounds.

These findings indicate that CTPs are highly likely to achieve BBB penetration by modulating these receptors and by impairing BBB integrity.

CTPs potentially penetrate the BBB via passive diffusion

As described in predicted potential effects of CTPs on the BBB and passive diffusion penetration properties, the ability of compounds to cross the BBB via passive diffusion may vary depending on their physicochemical properties, and toxicity prediction tools can effectively predict these characteristics.22,23,24 Following computational prediction of the six CTPs using the Maestro software package, 12 evaluation parameters were compiled and scored, according to the criteria established in Table 1. As summarized in Table 2, all six CTPs achieved scores of 8 or higher according to the scoring system described in predicted potential effects of CTPs on the BBB and passive diffusion penetration properties, suggesting their potential for passive diffusion across the BBB. It is worth noting that QPlogBB is regarded as the parameter most strongly associated with BBB penetration capability among these molecular descriptors. In this study, five CTPs (excluding HCCTP) exhibited QPlogBB values within the favorable range for BBB permeability (−3 to 1.2). Additionally, the calculated molecular weights of five compounds (excluding HPCTP) remained below the recommended threshold of 450 g/mol for CNS-targeting drugs (Table 2). Furthermore, these compounds displayed fewer than 5 hydrogen bond (HB) acceptors (acceptor HB) and completely lacked hydrogen bond donors (donor HB). This molecular characteristic favors desolvation prior to interaction with the lipophilic regions of cell membranes, thereby facilitating passive diffusion.27,29

Molecular descriptors | Range for BBB | Comments | Reference
QPlogBB | −3–1.2 | predicted brain/blood partition coefficient. predicts favorable BBB passive permeation for molecules, which range between −3.0 and 1.2 (based upon 95% of known drugs) | Eigenmann et al.25; Kelder et al.26
QPPMDCK | >500 nm/s | predicted MDCK cell line permeability in nm/s; MDCK cells are considered to be a good mimic for the BBB; above the recommended is consider great based upon 95% of known drug | Eigenmann et al.25
MW | <450 g/mol | molecular weight; compounds with high molecular weight are unable to passively cross the blood brain barrier; this can be related with volume | van de Waterbeemd et al.27
QPlogPo/w | 4 | predicted octanol/water partition coefficient | Kelder et al.26
Dipole | 1–12.5 | computed dipole moment of the molecule | Figueira et al.28
Volume | 500–2,000 A3 | total solvent-accessible volume in cubic angstroms | Figueira et al.28
Donor HB | 0 | estimated number of hydrogen bonds that would be donated by the solute to water molecules in an aqueous solution; no hydrogen bond donors facilitates their desolvation before entering the lipophilic phase of the cell membranes | Eigenmann et al.25
Acceptor HB | <5 | estimated number of hydrogen bonds that would be accepted by the solute from water molecules in an aqueous solution; low effective number of H-bond acceptors (<5) facilitates their desolvation before entering the lipophilic phase of the cell membranes | Eigenmann et al.25
QPPcaco | >500 nm/s | predicted apparent Caco-cell permeability in nm/s; above the recommended is consider great based upon 95% of known drugs | Eigenmann et al.25
QPlogKhsa | −1.5 to 1.5 | prediction of binding to human serum albumin | Figueira et al.28
PSA | 70–90 | van der Waals surface area of polar nitrogen and oxygen atoms and carbonyl carbon atoms; recommended below the thresholds of 90 Å2 (ref. a) and 70 Å2 (ref-b) | van de Waterbeemd et al.27; Kelder et al.26
Rotatable bonds | <6 | number of non-trivial (not CX3), non-hindered (not alkene, amide, small ring) rotatable bonds; less than 6 bonds recommended for CNS drugs | Brito-Sánchez et al.29

Molecule | QPlogBB | QPPMDCK | MW | QPlogPo/w | Dipole | Volume | DonorHB | AccptHB | QPPCaco | QPlogKhsa | PSA | #rotor | Score
HPCTP | −0.5 | 1,602.261 | 693.571 | 5.123 | 8.279 | 1,880.919 | 0 | 14.5 | 2,825.122 | −0.14 | 72.975 | 8 | 8
PFPCTP | 0.584 | 10,000 | 323.038 | 2.915 | 1.945 | 773.453 | 0 | 4.5 | 3,487.27 | −0.425 | 72.509 | 2 | 11
HCCTP | 1.279 | 10,000 | 347.659 | 3.149 | 0.004 | 722.046 | 0 | 4.5 | 4,745.452 | −0.49 | 58.279 | 0 | 8
HFCTP | 0.812 | 10,000 | 248.932 | 1.32 | 0.005 | 534.87 | 0 | 4.5 | 2,789.85 | −1.033 | 68.285 | 0 | 9
EPFCTP | 0.605 | 10,000 | 274.994 | 2 | 2.436 | 664.134 | 0 | 4.5 | 3,440.229 | −0.743 | 73.19 | 2 | 11
HMCTP | 0.102 | 4,446.693 | 321.146 | 2.921 | 0.042 | 904.371 | 0 | 4.5 | 7,626.467 | −0.215 | 86.485 | 6 | 9

To enhance prediction accuracy regarding the BBB penetration potential of CTPs, we integrated computational results from both Schrödinger software and the ADMETlab 3.0 database. A total of 10 molecular descriptors associated with BBB permeability were calculated for the six CTPs (Table 3). Among these descriptors, particular attention was given to the parameter labeled “BBB.” According to the official documentation, this metric quantifies the predicted probability (range: 0–1) of a chemical substance crossing the BBB.30 In our analysis, HPCTP, PFPCTP, HCCTP, and HFCTP all demonstrated relatively high probabilities, ranging from 0.616 to 1. Thus, the combined results indicate that HPCTP, PFPCTP, HCCTP, and HFCTP may penetrate the BBB via passive diffusion.

Molecule | BBB | Molecular weight | Volume | nHA | nHD | logP | Caco-2 permeability | MDCK permeability | TPSA
HPCTP | 0.616 | 693.13 | 669.074 | 9 | 0 | 3.284 | −4.958 | −4.709 | 92.46
PFPCTP | 0.906 | 322.96 | 218.91 | 4 | 0 | 1.494 | −4.969 | −4.705 | 46.31
HCCTP | 1 | 344.74 | 183.738 | 3 | 0 | 1.208 | −5.03 | −4.733 | 37.08
HFCTP | 0.974 | 248.92 | 128.877 | 3 | 0 | 0.488 | −5.073 | −4.734 | 37.08
EPFCTP | 0.215 | 274.96 | 166.192 | 4 | 0 | 0.745 | −4.876 | −4.688 | 46.31
HMCTP | 0 | 321.04 | 248.989 | 9 | 0 | −0.564 | −4.842 | −4.726 | 92.46

Nevertheless, it remains unclear whether CTPs and their in vivo metabolites may elicit more pronounced BBB toxicity and further affect the nervous system, which warrants further experimental investigation.

BBB penetrating mechanisms of CTPs

Transporter function at the BBB

The BBB is formed by cerebral capillary endothelial cells, which exhibit specialized physiological functions that support an active transport system rich in carrier proteins. This system includes membrane proteins encoded by the ATP-binding cassette (ABC) transporter gene family, which primarily function to hydrolyze ATP and efflux xenobiotics or drugs from the CNS or cerebral endothelial cells, thereby providing neuroprotection and detoxification.23,24,31,32,33,34 Another critical group of BBB transporters comprises membrane-bound proteins encoded by the solute carrier (SLC) gene family that mediate bidirectional transport of various substances such as carbohydrates, amino acids, hormones, and fatty acids, between the bloodstream and the CNS.23,31

The sophisticated transporter network within the BBB offers CTPs potential pathways beyond passive diffusion to enhance CNS penetration. These include carrier-mediated transport and modulation of transporter activity. Further investigation is needed to explore alternative mechanisms by which CTPs may cross the BBB. Network toxicology provides an advanced framework for elucidating toxicity targets and mechanisms of environmental contaminants. Therefore, to examine whether CTPs may affect transport functions throughout varies transporter systems, we performed a network toxicology analysis method.

HPCTP exhibits significant potential to disrupt substance transport across the BBB

Based on prior screening outcomes, four compounds—HPCTP, PFPCTP, HCCTP, and HFCTP—were selected for network toxicology analysis. Potential targets were identified by querying each compound against Similarity Ensemble Approach (SEA), TargetNet, Index, SwissTargetPrediction, and the ChEMBL database. After removing duplicates, 784, 804, 725, and 832 target genes were obtained for HPCTP, PFPCTP, HCCTP, and HFCTP, respectively. These gene clusters were intersected with 54 high-confidence neurotoxicity-related genes associated with neurological system disorders (NSDs) from the Comparative Toxicogenomics Database (CTD), yielding 17, 6, 3, and 3 overlapping genes, respectively (Figure 3A). These overlapping genes are implicated in CNS and BBB functions.

Figure 3: Differential gene analysis and functional enrichment of four CTP analogues(A) Venn diagrams illustrate the unique and shared differentially expressed genes (DEGs) induced by HPCTP, PFPCTP, HCCTP, and HFCTP relative to the NSD group, indicating compound-specific transcriptional responses.(B) GO enrichment analyses reveal that compound-specific DEGs are predominantly enriched in stress responses, programmed cell death, transmembrane transport, and metabolic regulation.(C) KEGG enrichment analyses show distinct pathway patterns among the four compounds, involving lipid metabolism regulation, apoptosis-related pathways, viral infection processes, and neurodegenerative disease pathways, suggesting multi-target toxicological mechanisms.

The intersecting target genes of the four compounds were subsequently submitted to the DAVID database for Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analyses. For HPCTP, GO analysis revealed significant enrichment in the biological process (BP) category for “response to hypoxia,” followed by “xenobiotic transport across BBB.” In the cellular component (CC) category, genes were enriched in “plasma membrane” and “apical plasma membrane.” Molecular function (MF) terms included “xenobiotic transmembrane transporter activity,” “protease binding,” and “ABC-type xenobiotic transporter activity” (Figure 3B). These findings suggest that HPCTP targets genes associated with the CNS and BBB, particularly localizing to plasma membrane structures and potentially influencing xenobiotic permeation and transporter functions on the BBB. For PFPCTP, HCCTP, and HFCTP, the intersecting target genes were primarily enriched in the BP terms “response to lipopolysaccharide” and “positive regulation of neuron apoptotic process,” with no significant enrichment observed in other categories (Figure 3B). This result suggests that PFPCTP, HCCTP, and HFCTP are potentially related to neuronal apoptosis.

KEGG pathway enrichment analysis indicated that the target genes of HPCTP were predominantly enriched in “lipid and atherosclerosis” pathways and exhibited associations with ABC transporters. Meanwhile, our data show that HPCTP and PFPCTP are associated with cell apoptosis, while HCCTP and HFCTP are related to Alzheimer disease (Figure 3C)).

These results imply that HPCTP is more likely than the other three compounds to modulate its own BBB permeability by interfering with multiple BPs, which include the expression of transporter genes, as well as the function and activity of transporters at the BBB. Moreover, HPCTP may enhance the penetration of other drugs and xenobiotics across the BBB.

Potential effects of CTPs in ABC efflux transporters

As indicated in Section transporter function at the BBB, the ABC efflux transporter family plays a critical role in eliminating xenobiotics from the CNS. Downregulation of ABC transporter gene expression or inhibition of their protein function may indirectly increase the retention of exogenous chemicals and drugs within the CNS.35 P-glycoprotein (P-gp/ABCB1), multidrug resistance-associated protein 1 (MRP1/ABCC1), and breast cancer resistance protein (BCRP/ABCG2) represent the principal ABC efflux transporters on the BBB.23,36,37

These proteins are expressed on the luminal membrane of cerebral capillaries, and P-gp is the predominant efflux pump at the BBB; therefore, P-gp is taken as the representative example.33 Given their function in transporting both endogenous and exogenous toxic substances and drugs from the CNS, a chemical identified as a P-gp substrate generally exhibits lower actual BBB penetration than predicted by its lipophilicity alone. Conversely, substances that inhibit P-gp may display enhanced BBB penetration due to reduced efflux activity, whereas increased drug lipophilicity on the BBB elevates the likelihood of a compound serving as a substrate for ABC efflux transporters.

As ABC transporters act as efflux pumps on the BBB, we speculate that the six CTPs may act both as P-gp substrates subject to efflux and as inhibitors that suppress P-gp activity, thereby reducing efflux and enhancing their likelihood of crossing the BBB. In our study, the ADMETlab 3.0 database was employed to predict the probability of the six CTPs serving as P-gp inhibitors, P-gp substrates, MRP1 inhibitors, or BCRP inhibitors. Results indicated that five CTPs (excluding HCCTP) showed high probabilities of acting as P-gp inhibitors (0.994–1.000). In contrast, all six CTPs exhibited negligible or absent potential as P-gp substrates (0–0.002). Regarding MRP1 inhibition, HPCTP demonstrated relatively low probability (0.195), whereas the other five CTPs displayed high probabilities (0.885–0.923). The above results were compiled and integrated to produce Table 4. The potential for BCRP inhibition was minimal across all six compounds. Therefore, all six compounds likely influence efflux transporters and may lead to their accumulation in the CNS, which is related to two potential mechanisms: inhibiting the function of specific ABC efflux transporters, or not being recognized as substrates by these transporters.

Molecule | Pgp-inhibitor | Pgp-substrate | MRP1 inhibitor | BCRP inhibitor
HPCTP | 1 | 0 | 0.195 | 0
PFPCTP | 0.998 | 0.001 | 0.885 | 0.001
HCCTP | 0 | 0 | 0.913 | 0
HFCTP | 0.994 | 0.002 | 0.895 | 0.001
EPFCTP | 0.999 | 0 | 0.923 | 0
HMCTP | 1 | 0 | 1 | 0.002

The preceding GO analysis indicated that HPCTP shows the strongest association with xenobiotic transport across the BBB. At the molecular function level, it is linked to xenobiotic transmembrane transporter activity and ABC-type transporter activity. We therefore hypothesize that HPCTP exerts a more pronounced influence on the ABC transporter system compared to other CTPs. Further examination has revealed that among the target genes related to CNS and BBB functions, HPCTP is associated with three ABC transporter family genes: Abcb1 (P-gp), Abcg2 (Bcrp), and Abcc2 (Mrp2) (Figure 4A). These genes belong to the three major subfamilies responsible for efflux transport at the BBB.

Figure 4: Pathway enrichment, molecular docking, and molecular dynamics analysis of HPCTP(A) Sankey and bubble plots summarize KEGG pathway enrichment of HPCTP-associated targets, highlighting significant involvement in apoptosis, ABC transporters, viral infection pathways, and cholinergic synapses.(B) Docking models of HPCTP with two representative proteins, identify key hydrogen bonds, hydrophobic contacts, and π-interactions contributing to ligand stabilization.(C) Molecular dynamics simulations indicate stable Rg trajectories for both HPCTP-SLC2A1 and HPCTP-SLC6A3 complexes, whereas the RMSD of HPCTP-SLC6A3 fluctuated markedly higher s; the SLC6A3 complex exhibits higher atomic contact counts and more favorable binding free energy, suggesting stronger interaction stability.

Combined with the previous predictive results, these findings suggest that HPCTP may modulate ABC transporter function not only through direct inhibition of their activity, but also by altering their gene expression. Thus, different evaluations revealed that HPCTP exhibited a more extensive range of toxicity compared to the other five CTPs. Such effects could potentially lead to HPCTP-induced damage to the BBB barrier function and further alter efflux of other xenobiotics in the CNS.

HPCTP disrupts BBB SLCs

Combined literature analysis with network toxicology prediction revealed HPCTP’s potential association with SLC2A1 and SLC6A3 SLCs on the BBB, which suggests an alternative pathway for BBB penetration via transporter density and activity. Specifically, SLC2A1 (GLUT1) localizes to both luminal and abluminal membranes of brain capillary endothelial cells, primarily facilitating glucose and hexose transport from blood to brain, whereas SLC6A3 (DAT) functions as a neurotransmitter transporter with dopamine as its primary substrate.38,39

To evaluate the potential carrier-mediated transport of HPCTP across the BBB, we performed molecular docking simulations with SLC2A1 and SLC6A3. The computational binding affinities for HPCTP with SLC2A1 and SLC6A3 were—10.71 and—9.68 kcal/mol, respectively, both values significantly below the −5 kcal/mol threshold, indicating strong spontaneous binding. Detailed interaction analysis of HPCTP revealed an extensive network of molecular contacts in the SLC2A1 binding pocket, and intermolecular interaction (chemical bonds) as shown in the Figure 4B. Similarly, the SLC6A3 binding pocket accommodated HPCTP through chemical bonds in the Figure 4B.

Collectively, these computational analyses indicate that HPCTP exhibits favorable binding affinity and stable interactions with both SLC2A1 and SLC6A3 that are members of the SLC transporter family, and potentially influencing transporter-related processes at the BBB. However, these observations are based on computational predictions, further studies employing dedicated BBB permeability assays and in vivo models are required to validate these predictions.

Stable binding of HPCTP with two SLC transporters

Molecular dynamics simulations provide enhanced resolution of molecular interactions and transport dynamics facilitated by specific BBB transporters.38,40 We conducted all-atom molecular dynamics simulations spanning 100 ns to characterize the binding stability of HPCTP-SLC2A1 and HPCTP-SLC6A3. Results show that the SLC2A1-HPCTP complex maintained consistently low root-mean-square deviation (RMSD) values throughout the simulation, and rapidly reached equilibrium with minimal structural fluctuations, indicating constrained conformational flexibility and improved structural integrity (Figure 4C).

In comparison, the SLC6A3-HPCTP complex displayed significantly higher RMSD amplitudes with marked oscillations. These dynamics suggest more frequent conformational rearrangements and comparatively reduced stability. Both molecular complexes demonstrated stable radius of gyration trajectories during the entire simulation period (Figure 4C). This consistency implies that HPCTP binding induces compact structural organization in both receptor proteins.

Interfacial contact analysis provided additional insights, showing that SLC2A1 maintained substantially more atomic contacts with HPCTP than SLC6A3. The mean contact values were 2,504.2 ± 184.9 for SLC2A1 compared to 2,419.9 ± 256.2 for SLC6A3 (Figure 4C). These quantitative distinctions indicate a larger buried surface area and more robust non-covalent interactions in the SLC2A1-HPCTP complex. Conversely, the SLC6A3 interface showed fewer contacts with greater variability, reflecting weaker molecular engagement.

Binding free energy decomposition using the MM/PBSA method revealed that complex stabilization primarily resulted from favorable van der Waals interactions and nonpolar solvation terms, while polar solvation effects partially counterbalanced electrostatic contributions. The SLC2A1-HPCTP complex demonstrated more favorable values for both van der Waals and nonpolar solvation components. This energetic profile translated to stronger overall binding affinity, as evidenced by more negative ΔGbinding values compared to the SLC6A3-HPCTP system (Figure 4C). The observed energetic advantage for SLC2A1 aligns with its lower RMSD fluctuations and higher contact density. Together, these computational findings consistently support the superior binding stability of HPCTP with SLC2A1 compared to SLC6A3.

Moreover, RMSD analysis revealed that after the substitution of the key residues with alanine, HPCTP still bound to SLC2A1 and SLC6A3, but the mutant complexes exhibited lower number of atomic contacts and slightly increased radius of gyration, indicating a looser packing of the binding interface (Figures 5A–5C). Consistently, the binding free energies became less favorable in the mutant systems, further demonstrating that mutation of these key residues weakens the stability and affinity of HPCTP binding to both transporters (Figure 5D). These observations confirm that the interaction pattern predicted by our molecular docking, in which these residues play a central role in HPCTP recognition, is reliable.

Figure 5: Computational identification of key HPCTP-binding residues in SLC2A1 and SLC6A3

HPCTP exposure affects the glucose uptake of hCMEC/D3 cells

We observed a dose-dependent increase in fluorescence intensity in hCMEC/D3 cells following exposure to HPCTP at concentrations of 0, 1, 10, and 100 nM (Figure 6A). Statistical analysis demonstrated that compared with the control group, the fluorescence intensities in the 1, 10, and 100 nM groups increased significantly with increasing concentration (Figure 6B). These findings suggest that HPCTP exposure may alter the glucose uptake capacity of hCMEC/D3 cells.

Figure 6: Effects of HPCTP exposure on glucose uptake in hCMEC/D3 cells(A) Representative fluorescence images of intracellular glucose uptake in hCMEC/D3 cells following exposure to different concentrations of HPCTP. Green fluorescence indicates intracellular glucose accumulation. Scale bars, 50 μm.(B) Quantitative analysis of fluorescence intensity in hCMEC/D3 cells after HPCTP exposure. ∗p < 0.05, n = 3, data are presented as mean ± SEM.

HPCTP exposure affected the glucose transport, metabolism and Slc2a1 gene expression in hCMEC/D3 cells

To further investigate the effects of HPCTP exposure on glucose transport across the BBB, we established a compact hCMEC/D3 cell monolayer to simulate the BBB for permeability assays. According to the light microscope images, before and after the exposure, there were no changes in the cell monolayer membranes, and the connections between cells were not damaged after the exposure (Figure 7A). Compared with the control group, the glucose concentration in the lower chamber, cell lysates, and upper chamber increased significantly in a dose-dependent manner. Since consumed glucose is primarily metabolized by cells, the observed increases in glucose concentrations suggest that HPCTP exposure may inhibit glucose utilization in hCMEC/D3 cells. Moreover, the dose-dependent increase in glucose concentration in the upper chamber medium indicates that HPCTP exposure may impair the transmembrane transport capacity of hCMEC/D3 cells for glucose (Figure 7B). Interestingly, HPCTP exposure significantly downregulated the expression of the Slc2a1 gene in hCMEC/D3 cells. Given the critical role of SLC2A1 in BBB glucose transport, this effect may represent an important mechanism by which HPCTP inhibits glucose transport across the BBB.

Figure 7: Morphological observation, quantitative analysis of cells under different treatment conditions, and RT-qPCR statistical results of cells after exposure(A) Bright-field microscopy observation of cell morphology in the control group (×10 magnification, scale bars, 100 μm) and experimental group (×20 magnification, scale bars, 50 μm).(B) Corresponding quantitative statistical analysis of net glucose permeation, cellular glucose uptake and upper layer glucose residue amount.(C) Relative expression of Slc2a1 gene in hCMEC/D3 cells. ∗p < 0.05, n = 3, data are presented as mean ± SEM.

Overview of predicted HPCTP BBB penetration

Based on our previous predictions, we propose that HPCTP may cross BBB through both passive diffusion and SLC-mediated transport, and may also function as an inhibitor of ABC transporters. SLC2A1 is abundantly expressed on both the luminal and abluminal membranes of cerebral capillary endothelial cells, and molecular docking analyses indicate favorable binding between HPCTP and SLC2A1. Accordingly, we speculate that HPCTP facilitates its own BBB penetration by interacting with the SLC2A1 transporter.

Network-toxicology GO enrichment analysis further suggests that HPCTP regulates the expression of genes within the SLC and ABC transporter families. Specifically, HPCTP may upregulate Slc2a1, thereby increasing the density of SLC2A1 transporters on endothelial surfaces and enhancing its own translocation across the BBB. Conversely, HPCTP may downregulate the ABC efflux genes Abcb1 (P-gp), Abcg2 (Bcrp), and Abcc2 (Mrp2), reducing the expression of the major ABC efflux subfamilies and consequently increasing intracellular HPCTP levels. In addition, ADMETlab 3.0 predicts that HPCTP acts as a P-gp inhibitor, further elevating endothelial HPCTP accumulation and increasing the likelihood of successful BBB penetration (Figure 8).

Figure 8: Molecular mechanism of HPCTP penetrating the BBB