Targeting Mutant Phenylalanine Hydroxylase With Pyrimidine–Triazole Conjugates: A Primary Framework for Candidate Chaperone‐Based Strategies in Phenylketonuria

ABSTRACT

Phenylketonuria (PKU) is a rare metabolic disorder caused by pathogenic mutations in the phenylalanine hydroxylase (PAH) gene, which impair the conversion of phenylalanine to tyrosine, leading to subsequent neurotoxicity. Dietary management, sapropterin dihydrochloride, and pegvaliase are the current therapies; however, their limited efficacy and adverse effects highlight the pressing need for pharmacological treatments that restore PAH activity. Therefore, in this work, we designed and synthesized pyrimidine–triazole derivatives (Pyr‐TZ) as candidate pharmacological chaperones targeting the catalytic domain of PAH. Molecular docking disclosed that Pyr‐TZ‐2O and Pyr‐TZ‐2S exhibited higher binding energies (−8.5 to −10.3 kcal/mol) for the wild‐type (WT) and mutant (R252Q) PAH compared to sapropterin (−7.0 kcal/mol). Molecular Mechanics with Generalized Born and Surface Area solvation (MM‐GBSA) was used here as comparative estimates suggested enhanced binding stability of Pyr‐TZ compounds, primarily driven by van der Waals and lipophilic interactions. Drug‐likeness and ADMET profiling indicated favorable pharmacokinetic properties. Biological validation in R252Q mutant cells demonstrated significant upregulation of PAH and key tetrahydrobiopterin (BH₄) pathway genes, including quinonoid dihydropteridine reductase (QDPR), sepiapterin reductase (SPR), and 6‐pyruvoyl‐tetrahydropterin synthase (PTS), following Pyr‐TZ‐2O treatment. Consistent with these findings, sandwich ELISA revealed a dose‐dependent increase in PAH protein abundance post‐Pyr‐TZ‐2O treatment. Conclusively, Pyr‐TZ‐2O can be considered a potential pharmacological chaperone capable of stabilizing mutant PAH and enhancing cofactor regeneration, offering a rational therapeutic approach for PKU.

Phenylketonuria (PKU) results from pathogenic mutations in the phenylalanine hydroxylase (PAH) gene, causing phenylalanine accumulation and neurotoxicity. We designed pyrimidine–triazole (Pyr‐TZ) derivatives as potential pharmacological chaperones for mutant PAH. Molecular docking and MM‐GBSA analyses showed stronger binding of Pyr‐TZ‐2O than sapropterin. Drug‐likeness and ADMET profiles were favorable. In R252Q mutant cells, Pyr‐TZ‐2O increased PAH and tetrahydrobiopterin (BH₄) pathway gene expression and enhanced PAH protein levels, in a dose‐dependent manner, identifying Pyr‐TZ‐2O as a promising PKU therapeutic candidate.

1. Introduction

Phenylketonuria (PKU; OMIM #261600) is one of the most common inherited inborn errors of amino acid metabolism, resulting from deficient activity of phenylalanine hydroxylase (PAH; EC 1.14.16.1), the hepatic enzyme responsible for converting phenylalanine (Phe) to tyrosine (Tyr) in the presence of tetrahydrobiopterin (BH₄) as an essential cofactor [1, 2, 3]. PKU is inherited in an autosomal recessive manner and is caused by pathogenic variants in the PAH gene located on chromosome 12q22‐24.1 [4]. To date, more than 1500 disease‐associated PAH variants have been identified, contributing to significant genetic and phenotypic heterogeneity among affected individuals [4, 5, 6]. Although classified as a rare disorder, PKU remains a major public health concern because of its lifelong disease burden and the severe neurological complications associated with delayed diagnosis and treatment. The incidence of PKU varies considerably across geographic regions and ethnic populations. The worldwide prevalence is estimated at approximately 1 in 23 930 newborns [7]. In the United States, PKU occurs in approximately 1 in 25000 live births [8]. In Europe, the incidence ranges from about 1 in 5360 to 112000 live births in various countries. Asia generally exhibits lower prevalence rates, including Thailand (1 in 227 273), Japan (1 in 125 000), the Philippines (1 in 116 006), and Singapore (1 in 83 333), although China represents an exception with an estimated incidence of 1 in 15 924 live births [8]. In India, the prevalence of PKU is not well established because of limited newborn screening programs, with available studies reporting an incidence of approximately 1 in 18 300–20 513 live births [9, 10, 11]. The primary risk factor for PKU is the inheritance of pathogenic PAH alleles from both carrier parents, while it significantly increases disease frequency in certain populations.

The pathophysiology of PKU is driven by impaired PAH‐mediated metabolism, resulting in excessive accumulation of Phe in blood and tissues. Elevated Phe levels exert neurotoxic effects through multiple mechanisms, including competition with other large neutral amino acids for transport across the blood–brain barrier via the LAT1 transporter, thereby reducing cerebral uptake of tyrosine and tryptophan, which are precursors for dopamine, norepinephrine, and serotonin biosynthesis [12, 13]. Consequently, neurotransmitter imbalance contributes to cognitive dysfunction, behavioral abnormalities, psychiatric manifestations, and developmental delay. In addition, chronic hyperphenylalaninemia promotes oxidative stress, mitochondrial dysfunction, impaired protein synthesis, altered neuronal energy metabolism, and defective myelination, collectively leading to progressive neurological damage [14, 15]. If left untreated, PKU can cause severe intellectual disability, epilepsy, autism spectrum‐like features, motor dysfunction, and substantially reduced quality of life [16, 17, 18]. According to residual PAH activity and blood phenylalanine concentrations, PAH deficiency is generally classified as mild hyperphenylalaninemia (120–600 μmol/L), mild PKU (600–1200 μmol/L), and classical PKU (> 1200 μmol/L) [19].

Current therapeutic strategies primarily focus on reducing systemic phenylalanine concentrations. Lifelong dietary restriction of Phe intake combined with Phe‐free amino acid supplementation remains the cornerstone of treatment [20]. However, strict dietary adherence is difficult to maintain throughout life and often imposes significant psychosocial and economic burdens on patients and caregivers [21]. An alternative therapeutic strategy involves targeting misfolded PAH proteins using pharmacological chaperones (PhCs), which stabilize mutant enzymes, promote proper folding, and restore residual enzymatic activity [22]. In 2007, the United States Food and Drug Administration (FDA) approved sapropterin dihydrochloride (Kuvan), a synthetic formulation of BH₄, for the treatment of BH₄‐responsive PKU (Figure 1A) [23, 24]. Although sapropterin enhances PAH activity in a subset of patients with residual enzyme function, only approximately 20%–30% of individuals exhibit a clinically meaningful response, and its efficacy is generally limited in patients carrying severe or null mutations [25, 26, 27]. In 2018, the FDA approved pegvaliase, a PEGylated recombinant phenylalanine ammonia lyase enzyme, as an enzyme substitution therapy for adults with PKU (Figure 1B) [28, 29]. Pegvaliase effectively lowers circulating phenylalanine levels independently of PAH activity; however, treatment is frequently associated with adverse immune‐mediated reactions, including hypersensitivity and anaphylaxis, which may compromise patient compliance and long‐term therapeutic success [30]. Therefore, despite important advances in disease management, substantial unmet clinical needs remain, particularly for patients who exhibit poor responsiveness to currently available therapies.

FIGURE 1.

Chemical structure of the compounds.

Therefore, to address the limitations of existing PhCs and identify new candidates capable of stabilizing and correcting mutant PAH, we synthesized a series of pyrimidine–triazole (Pyr‐TZ) compounds targeting the catalytic groove of the enzyme. These pyrimidine derivatives were designed based on the structural similarity with the commercially available PhC, sapropterin, which contains a ring system combining a pyrimidine ring and a pyrazine ring. Furthermore, pyrimidine derivatives are well‐known for their bioisosteric versatility and ability to engage in hydrogen bonding within enzyme active sites, while triazole linkers confer conformational stability and enhanced pharmacokinetic properties [31, 32]. From a library of 21 designed compounds, Pyr‐TZ‐2O (Figure 1C) and Pyr‐TZ‐2S (Figure 1D) exhibited the highest binding energies obtained via molecular docking analyses with both wild‐type (WT)‐PAH and the mutant PAH‐Arg252Gln (R252Q). Molecular Mechanics with Generalized Born and Surface Area solvation (MM‐GBSA) analysis showed higher total free binding energy (G _bind − 75.37 to −102.28 kcal/mol), driven predominantly by van der Waals and lipophilic interactions. Furthermore, drug‐likeness analysis using ADMET profiling indicated favorable pharmacokinetic properties. mRNA expression analyses in mutant (R252Q) cell lines derived from PKU patients revealed increased expression of PAH, QDPR, SPR, and PTS post‐Pyr‐TZ‐2O treatment, with preliminary activation of the BH₄‐dependent enzymatic pathway. Consistent with these findings, sandwich ELISA revealed a dose‐dependent increase in PAH protein abundance post‐Pyr‐TZ‐2O treatment. Collectively, our results highlighted Pyr‐TZ‐2O as a promising candidate for PKU treatment; however, there is a need to conduct more studies to evaluate its therapeutic potential.

2. Methods

2.1. Theoretical Studies

2.1.1. Retrieval of Protein Structure and Construction of Mutant Models

The three‐dimensional (3D) structure of PAH protein (PDB ID: 1J8U) was obtained from the Protein Data Bank (https://www.rcsb.org) [33, 34]. Chain A was utilized for generating the mutant model. Point mutation (R252Q) was made in the PAH protein using UCSF Chimera (version 1.17.3; https://www.cgl.ucsf.edu/chimera). The crystal structure was imported into Chimera, and relevant residues were selected using the sequence viewer and command‐line interface. Mutations were introduced through the Rotamers tool, and energy minimization was performed using the Minimize Structure function to refine stereochemistry and eliminate steric clashes [35, 36]. The minimized mutant structures were saved in PDB format for analyses. The structures were visualized, and analyzed using PyMol (https://www.pymol.org), and BIOVIA Discovery Studio Visualizer 2021 Client (https://discover.3ds.com/discovery‐studio‐visualizer‐download) [37]. Additionally, the SDF format of the sapropterin (ID: 135398654) was sourced from the PubChem database (https://pubchem.ncbi.nlm.nih.gov/) [38, 39]. Chemical structures of the molecules were drawn through ChemDraw Ultra 12.0, available at https://chemistry.com.pk/software/free‐download‐chemdraw‐ultra‐12/.

2.1.2. Structural Validations of Protein Models

The quality of protein models was evaluated using the Structural Analysis and Verification Server (SAVES) (http://nihserver.mbi.ucla.edu/SAVES). Three tools were used in the SAVES server: (i) PROCHECK, which analyzed stereochemical features and generated Ramachandran plots; (ii) VERIFY 3D, which checked the consistency between atomic (3D) models and amino acid sequences (1D); and (iii) ERRAT, which assessed the overall quality of crystallographic structures [6, 40]. These validation methods ensured the accuracy of protein models for subsequent research.

2.1.3. Docking Parameters for Proteins and Ligands

In this study, chain A of PAH (1J8U) and PAH‐R252Q protein receptors were utilized. The crystal structures were processed via removing the complexed ligand molecules, adding hydrogen atoms, and applying Gasteiger charges using AutoDock Vina version 1.1.2 (http://vina.scripps.edu) [41]. Subsequently, we used PyRx 0.8 (https://sourceforge.net/projects/pyrx) software to minimize energy, change ligand molecules to PDBQT format, and dock protein–ligands [5, 41]. This docking procedure was conducted at an exhaustiveness (E) level of 8, indicating the depth of the search within AutoDock Vina. A grid box was created for specific active site residues [42], with coordinates set to x, y, z (0.2441, 25.1985, 9.6884) for PAH and x, y, z (0.2441, 25.1685, 9.6828) for PAH‐R252Q on the receptor's surface, with grid dimensions of 25 × 25 × 25 Å.

2.1.4. Analysis of Protein–Ligand Complexes

The interactions between protein–ligand complexes, including hydrophobic (π‐π stacking, amide‐π stacking, and sigma), electrostatic, hydrogen bond (H‐bond), and van der Waals, were depicted. Their 3D and 2D plots were generated via 3D Discovery Studio (https://discover.3ds.com/discovery‐studio‐visualizer‐download) and LigPlot+ version 2.2 (https://www.ebi.ac.uk/thornton‐srv/software/LigPlus) [43].

2.1.5. Molecular Dynamics Simulation

MD simulations were performed to assess the atomic movements of protein–ligands over time based on Newton's classical equations of motion. We ran 100 ns MD simulations using Desmond, a software package provided by Schrödinger LLC (Desmond | Schrödinger; schrodinger.com) to evaluate ligand binding under physiological parameters. Molecular docking results were considered as the basis for the MD simulations. The initial configurations for protein–ligand interactions were made via the Protein Preparation Wizard within the Maestro Schrödinger Suite 2017, optimizing and refining the complexes (https://www.schrodinger.com/protein‐preparation‐wizard). The System Builder tool was utilized for simulation setup. The Optimized Potentials for Liquid Simulations_2005 (OPLS_2005) force field was used to define the molecular characteristics (coordinates and topology). Complexes were placed in a solvated environment using the Transferable Intermolecular Potential with 3 Points (TIP3P) water model. To provide adequate space for the simulation, these solvated complexes were positioned inside an orthorhombic box, maintaining a 10 Å distance from the protein atoms to the edges of the box. A hybrid approach that combined the steepest descent method (with a minimum of 10 steps) and the Limited‐memory Broyden–Fletcher–Goldfarb–Shanno (LBFGS) was used for the energy minimization of the system. To maintain a neutral charge in the system, counterions were added as necessary. A 0.15 M sodium chloride (NaCl) salt solution was added to create the physiological conditions, and simulations were executed under the isothermal‐isobaric ensemble (NPT) at 1 atm pressure and 300 K temperature. A relaxation phase for the models was ensured before the actual simulation. Trajectory data were saved at 10‐ps intervals through the 100 ns run [37]. Simulation stability was calculated using the root‐mean‐square‐fluctuations (RMSF), root‐mean‐square‐deviation (RMSD), radius of gyration (Rg), and solvent accessible surface area (SASA) trajectories over time. The protein backbone of the equilibrated complex served as the reference structure for the RMSD plot.

2.1.5.1. Free Energy Calculation Through MM‐GBSA Approach

The combination of molecular mechanical energies with the MM‐GBSA or Poisson‐Boltzmann techniques (MM‐PBSA) is crucial for estimating the binding free energy (ΔG _bind) of small molecules that interact with biological macromolecules [44]. In this study, the binding free energies were computed using the Prime MM‐GBSA method, which utilizes the OPLS_2005 force field along with the variable dielectric surface generalized Born 2.0 (VSGB) solvation model. The docking positions for these compounds were obtained through the Glide algorithm, accessible via the Schrodinger platform at schrodinger.com.

The following equation is the formula for the ΔG _bind:

$$\mathrm{\Delta }{G}_{\text{bind}}=\mathrm{\Delta }{G}_{\left(\text{solv}\right)}+\mathrm{\Delta }{E}_{\left(\mathrm{MM}\right)}+\mathrm{\Delta }{G}_{\left(\mathrm{SA}\right)}$$

In this equation, ΔG _(solv) denotes the change in solvation energies of the unbound protein, ligand, and their complex, analyzed using the GBSA method. ΔE _(MM) reflects the energy difference between the unbound protein and ligand in comparison to the minimized energy of their bound state. Finally, ΔG _(SA) accounts for the change in surface area energies between the unbound protein, ligand, and their corresponding complex [3, 45]. For a thorough evaluation, data from different trajectories, including RMSD, RMSF, Rg, SASA, and B.E., were plotted using GraphPad Prism version 11.0.1 (https://www.graphpad.com/updates/prism‐11‐0‐1) [46].

2.1.6. Physicochemical and Absorption, Distribution, Metabolism, Excretion, and Toxicity (ADMET) Properties of Ligands

The physicochemical properties of the selected molecules, such as molar refractivity, the number of hydrogen bond donors and acceptors, topological surface area (TPSA), and lipophilicity (log Po/w), were evaluated using the SWISS‐ADME server (http://www.swissadme.ch) [47].

Also, we examined ADMET characteristics, including human intestinal absorption (HIA), blood–brain barrier (BBB) permeability, skin permeability, interactions with Cytochrome P450 (CYP) enzymes, toxicity, and excretion by using the pkCSM server at http://structure.bioc.cam.ac.uk/pkcsm [48].

2.1.7. Protein–Protein Network Analysis of PAH Protein

The identification of the protein–protein network associated with the PAH protein was done through the STRING database (https://string‐db.org) [49].

2.1.8. Density Functional Theory (DFT) Calculation of the Compounds

DFT calculations were performed using the B3LYP functional with the 6‐31G (d,p) basis set. Geometry optimization was followed by a vibrational frequency calculation at the same level to verify the optimized structure and to obtain IR data. Dipole moment, HOMO–LUMO energies, and the corresponding energy gap were extracted from the final optimized wavefunction. Molecular electrostatic potential (MEP) surfaces were generated using the optimized geometry to analyze charge distribution. All computational analyses and graphical visualizations were carried out using the standard quantum‐chemical software Gaussian 09W, and all reported orbital energies are given in electron volts (eV) [50].

2.2. Chemistry

2.2.1. Preparation of Ethyl‐6‐methyl‐2‐oxo‐4‐(4‐((1‐(p‐tolyl)‐1H‐1,2,3‐triazol‐4‐yl) methoxy) phenyl)‐1,2,3,4‐tetrahydropyrimidine‐5‐carboxylate

The target compound 1d was prepared in the following steps:

2.2.1.1. Step‐1: Preparation of 4‐(Prop‐2‐yn‐1‐Yloxy) benzaldehyde 1(b)

In a 100 mL round‐bottom flask, 5 mmol (1 eq) 4‐hydroxybenzaldehyde (1a) was taken, and 20 mL of acetone was added and stirred for a few minutes. Thereafter, 7.5 mmol (1.5 eq) of potassium carbonate (K₂CO₃) was added and continued stirring for 20 min then 12.5 mmol (2.5 eq) of propargyl bromide was added and the reaction mixture was stirred for 24 h. After 24 h tlc were taken in 20% ethyl acetate n‐hexane then the reaction mixture was filtered, and the filtrate was evaporated then the formed product was dissolved in ethyl acetate and washed with brine. A brownish white crystals 1b were formed with 87.5% yield (Figure 2) [51].

FIGURE 2.

Synthetic scheme of the Pyr‐TZ‐2O.

2.2.1.2. Step‐2: Preparation of Ethyl 6‐methyl‐2‐oxo‐4‐(4‐(prop‐2‐yn‐1‐yloxy) phenyl)‐1,2,3,4‐tetrahydropyrimidine‐5‐carboxylate (1c)

In a 50 mL round‐bottom flask, 1 mmol (1 eq) of 1b was dissolved in ethanol, and 2 mmol (2 eq) of ethyl acetoacetate, 1.9 mmol (1.9 eq) of urea, and 0.1 mmol (0.1 eq) of P‐toluene sulphonic acid were added, followed by refluxing for 36 h. A white solid product 1c formed with 95.7% yield (Figure 2) [52].

2.2.1.3. Step‐3: Preparation of Ethyl 6‐methyl‐2‐oxo‐4‐(4‐((1‐(p‐tolyl)‐1H‐1,2,3‐triazol‐4‐yl) methoxy) phenyl)‐1,2,3,4‐tetrahydropyrimidine‐5‐carboxylate (1d)

In a 100 mL round‐bottom flask, 1 mmol (1 eq) of 1c and 1.25 mmol (1.25 eq) of ArN₃ were taken, and 10 mL DMF were added. Further, 0.1 mmol (0.1 eq) CuSO₄ and 0.55 mmol (0.55 eq) Na‐ascorbate were dissolved in 10 mL DMF. The reaction mixture was stirred for 3 h at room temperature. TLC was taken using 70% ethyl acetate in n‐hexane. The reaction mixture was poured into ice; a white precipitate formed and was washed 2–3 times with cold water, then dried. A white solid product, 1d, was formed with 84.5% yield (Figure 2) [53].

2.3. Biological Evaluations

2.3.1. Cell Culture and Drug Treatment

The B‐lymphocyte cell line (ID: GM01565, PKU‐R252Q mutant) was sourced from the Coriell Institute for Medical Research in New Jersey, USA, and was cultured in Roswell Park Memorial Institute (RPMI) 1640 medium (Gibco, Thermo Fisher Scientific, USA) with 10% fetal bovine serum (FBS) (Gibco, Thermo Fisher Scientific, USA). The cultures were kept in a humidified atmosphere with 5% CO₂ at 37°C [54]. For drug treatment, cells were plated at a density of 1 × 10⁵ cells per well in 24‐well plates and treated with varying concentrations of Pyr‐TZ‐2O for 24 h prior to validation. Following treatment, cell lysates and culture supernatants were collected and stored at −80°C for subsequent molecular and biochemical analyses.

2.3.2. RNA Extraction, cDNA Synthesis, and Quantitative Real‐Time PCR (qRT‐PCR)

RNA extraction was done through the Trizol reagent (Invitrogen; Thermo Fisher Scientific, USA) according to the manufacturer's instructions. All independent experiments were performed in biological triplicate [55].

One microgram of total RNA was used for the cDNA synthesis using a cDNA synthesis kit (Thermo Fisher Scientific, USA). The mRNA levels from the synthesized cDNA were quantified using SYBR Green master mix (Applied Biosystems; Thermo Fisher Scientific, USA) in a 10 μL total reaction volume [56]. The cycling conditions were an initial denaturation at 50°C for 2 min, followed by 40 cycles of 95°C for 15 s and 60°C for 1 min. The relative fold change (FC) observed in the qRT‐PCR analysis was calculated using the comparative Ct (2^−ΔΔCt) method, with GAPDH as the endogenous control. A list of primers used is provided in Table S1.

2.3.3. Sandwich Enzyme‐Linked Immunosorbent Assay (ELISA) Analysis of PAH Protein

PKU‐R252Q‐mutant cells treated with Pyr‐TZ‐2O (10, 25, 50, and 100 μg/mL) and untreated control cells were lysed using radioimmunoprecipitation assay (RIPA) buffer supplemented with a 1% protease inhibitor cocktail. Protein concentrations were determined using the Bradford assay, and equal amounts of total protein were loaded onto each well of the ELISA plate for subsequent analysis [57]. To assess PAH protein concentration following Pyr‐TZ‐2O treatment, Sandwich ELISA (EH3498; FineTest, Wuhan, China) was performed according to the manufacturer's instructions. Briefly, 100 μL of standards or total cell‐protein were added to wells pre‐coated with anti‐PAH antibody and incubated for 90 min at 37°C. Wells were washed twice, followed by the addition of 100 μL biotin‐labeled antibody working solution and incubation for 60 min at 37°C. After three wash cycles, 100 μL of horseradish peroxidase‐streptavidin conjugate (SABC) working solution was added and incubated for 30 min at 37°C. The wells were subsequently washed five times, and 90 μL of TMB substrate solution was added for color development in the dark for 15 min at 37°C. The reaction was terminated by adding 50 μL stop solution, and absorbance was measured immediately at 450 nm using a microplate reader. PAH concentrations in the samples were calculated from a standard calibration curve generated using known PAH standards. qRT‐PCR and ELISA data were analyzed and graphically represented using GraphPad Prism version 11.0.1 (https://www.graphpad.com/updates/prism‐11‐0‐1).

3. Results

3.1. Theoretical Studies

3.1.1. Structural Validations of Wild‐Type and Mutant Protein Models

The UCSF Chimera software was utilized to construct the mutant PAH (R252Q) model, which was subsequently validated through various structural assessment tools (Figures 3A,B). Analyses conducted via the SAVES server, including PROCHECK, ERRAT, and VERIFY‐3D, confirmed that both the wild‐type and mutant models exhibited high stereochemical quality. Ramachandran plots revealed that 94.4% of residues were located in the most favored regions, 5.6% in additionally allowed regions, and none in disallowed regions, indicating well‐refined backbone geometry (Figure 3C,D). Both models had no bad contacts and 100% planar groups, proving optimal residue geometry and peptide planarity. VERIFY‐3D results indicated that 90.55% of residues had a 3D–1D score > 0.1, signifying correct residue environment in the mutant model. ERRAT analysis yielded a quality factor of 94.667, suggesting high mutant model accuracy.

FIGURE 3.

Three‐dimensional structures of (A) wild‐type and (B) PAH‐R252Q mutant proteins with highlighted residues (orange). Ramachandran plots of (C) WT‐PAH and (D) mutant PAH generated using PROCHECK.

3.1.2. Analysis of the Two Best Ligand Molecules That Bind to WT‐PAH and Mutant‐PAH Protein Receptors

Molecular docking was performed to assess the ligand–protein interactions and their binding energies within the active site. In light of previous findings indicating the limited therapeutic efficacy of sapropterin, we synthesized novel analogs and checked their binding potential using PyRx‐based molecular docking. Both Pyr‐TZ‐2O and Pyr‐TZ‐2S exhibited higher binding energies toward WT and mutant PAH (ranging from −8.5 to −10.3 kcal/mol), compared to sapropterin (−7.0 kcal/mol). In the WT‐Pyr‐TZ‐2O complex (Figure 4A and Table S2), the ligand embedded within the active site formed hydrogen bonds with Tyr166 and hydrophobic interactions with Leu146, Leu153, Trp224, and Ala220, thereby stabilizing the complex (−9.7 kcal/mol). In the mutant‐Pyr‐TZ‐2O complex (Figure 4B), the ligand maintained stability through hydrophobic contacts involving Trp326, Phe331, and Leu249, supporting a binding energy of −9.5 kcal/mol. The WT‐Pyr‐TZ‐2S complex (Figure 4C) demonstrated stabilization through hydrogen bonds with Glu228 and Tyr275, and hydrophobic interactions with Leu146, Phe152, and Trp224, contributing to its highest binding affinity (−10.3 kcal/mol). In the mutant‐Pyr‐TZ‐2S complex (Figure 4D), H‐bond interactions with Tyr377, along with hydrophobic residues (Leu249, Phe254, and Trp326), maintained stability (−8.5 kcal/mol). The WT‐sapropterin complex (Figure 4E) formed H‐bonds with Ser149, Glu184, Tyr223, and His188, indicating moderate stabilization (−7.0 kcal/mol). The mutant‐sapropterin complex (Figure 4F) exhibited fewer hydrogen bonds, with Ser251, His290, and Glu286, suggesting weaker interaction strength (−7.0 kcal/mol). The hydrogen bonding and hydrophobic network formed by Pyr‐TZ‐2O and Pyr‐TZ‐2S demonstrate their capacity to stabilize the PAH catalytic pocket, even under mutation‐induced conformational shifts. These findings indicate that both molecules possess superior binding stability compared to sapropterin, highlighting them as promising candidates for restoring PAH function in PKU therapy.

FIGURE 4.

2D models of the protein–ligand interaction of (A) WT‐Pyr‐TZ‐2O, (B) mutant‐Pyr‐TZ‐2O, (C) WT‐Pyr‐TZ‐2S, (D) mutant‐Pyr‐TZ‐2S, (E) WT‐sapropterin, and (F) mutant‐sapropterin.

3.1.3. MD Simulation and Gibbs Free Energy Calculation

The RMSD trajectories were monitored over a period of 100 ns to evaluate the overall structural stability of the protein–ligand complexes. Among the four complexes, the mutant‐Pyr‐TZ‐2S (green) showed the lowest overall RMSD, fluctuating between 1.0 and 1.8 Å, which suggests a highly stable conformation with minimal structural variation from the initial structure. The mutant‐Pyr‐TZ‐2O (orange) complex also exhibited relative stability, keeping RMSD values around 1.2–2.0 Å post‐equilibration. In contrast, WT‐Pyr‐TZ‐2S (blue) and WT‐Pyr‐TZ‐2O (yellow) complexes displayed slightly higher deviations, ranging between 1.3 and 2.3 Å, reflecting a less stable protein–ligand binding compared to the mutants (Figure 5A). Overall, the RMSD analysis suggests that mutation‐induced structural changes favor a more stable interaction with both Pyr‐TZ‐2O and Pyr‐TZ‐2S ligands.

FIGURE 5.

The trajectories for (A) RMSD, (B) RMSF, (C) Rg, and (D) SASA of both wild‐type and mutant protein receptors interacting with the ligands Pyr‐TZ‐2O and Pyr‐TZ‐2S.

RMSF reveals the dynamic behavior of specific residues in flexible loops, terminal regions, and the ligand‐binding pocket of the protein backbone during the 100 ns MD simulation. The WT‐Pyr‐TZ‐2S (blue) and WT‐Pyr‐TZ‐2O (yellow) complexes showed higher RMSF fluctuations (~1.5–3.5 Å) in certain loop regions, indicating increased dynamic motion in the WT protein. Both mutant complexes (orange and green) showed reduced oscillations, with RMSF values below 2.0 Å across most residues, suggesting enhanced local rigidity and stronger structural stabilization upon ligand association (Figure 5B).

The Rg trajectories were assessed over the 100 ns simulation period to evaluate the structural integrity of the protein–ligand complexes. During the simulation, all systems showed stable Rg values, fluctuating around 19.5–25.5 Å, indicating no major conformational expansion or unfolding events. The decrease in Rg for all the complexes suggests improved structural compactness, contributing to their enhanced conformational stability observed in the RMSD analysis (Figure 5C).

The SASA trajectory was evaluated to monitor changes in protein surface exposure and compactness during the 100 ns MD simulation. SASA reflects the protein surface accessible to solvent molecules, providing insights into conformational changes and folding stability upon ligand binding. All systems showed moderate fluctuations in SASA within 13 500–15 500 Ų, indicating dynamic equilibrium during the trajectory. The WT complexes exhibited higher average SASA values (14 810.59 and 14 555.16 Ų for the WT‐Pyr‐TZ‐2O and WT‐Pyr‐TZ‐2S), suggesting a more solvent‐exposed conformation. The mutant complexes maintained lower (14 471.80 and 14 425.84 Ų for the mutant‐Pyr‐TZ‐2O and mutant‐Pyr‐TZ‐2S) and more stable SASA values, indicating compact structural organization and enhanced conformational stability (Figure 5D).

The ΔG _bind of both WT and mutant complexes was calculated using the MM‐GBSA approach. The total binding energy was further decomposed into its contributing components: van der Waals (ΔG _bind vdW), lipophilic (ΔG _bind Lipo), solvation (ΔG _bind SolvGB), and covalent (ΔG _bind Covalent) terms (Figure 6). Among the systems, WT‐Pyr‐TZ‐2O (yellow) and WT‐Pyr‐TZ‐2S (blue) exhibited the most negative ΔG _bind values, of −93.93 kcal/mol and −102.28 kcal/mol, respectively, indicating strong binding interactions. However, the mutant complexes, mutant‐Pyr‐TZ‐2O (−75.37 kcal/mol) and mutant‐Pyr‐TZ‐2S (−87.48 kcal/mol), showed slightly less negative binding energies, suggesting a reduction in binding strength upon mutation. Collectively, the MM‐GBSA analysis revealed that WT complexes exhibit superior binding energies compared to mutant complexes. The predominance of van der Waals and lipophilic interactions underscores the critical role of hydrophobic residues in stabilizing the Pyr‐TZ conjugate ligands within the active pocket.

FIGURE 6.

Binding free energies and the corresponding energy components (lipophilic, covalent, and generalized Born electrostatic solvation, and van der Waals) obtained via MM‐GBSA analysis of WT and mutant protein receptors in complex with ligands Pyr‐TZ‐2O and Pyr‐TZ‐2S.

3.1.4. The Physicochemical and Pharmacokinetic Properties of the Synthesized Compounds

Prior to selecting any molecule as the ideal candidate drug, we must evaluate its compliance with Lipinski's rule of five. An ideal candidate should not exceed five hydrogen bond donors, 10 hydrogen bond acceptors, have a molecular weight of < 500, and a logp value below 5 [58]. Notably, our two selected compounds adhere to these criteria and are the most promising for further investigation (Table 1).

TABLE 1.

Physicochemical characteristics of Pyr‐TZ‐2O and Pyr‐TZ‐2S.

Descriptor	Pyr‐TZ‐2O	Pyr‐TZ‐2S
M.W (g/mol)	448.20	474.49
Consensus Log P o/w	2.69	3.09
Rotatable bonds	8	8
H‐bond acceptors	6	6
H‐bond donors	2	1
MR	128.64	151.20
TPSA	107.30	138.67
Lipinski #violations	0	0
Ghose #violations	0	1
Veber #violations	0	0
Egan #violations	0	0
Muegge #violations	0	0
Bioavailability score	0.55	0.55
PAINS #alerts	0	0
Brenk #alerts	0	1
Lead likeness #violations	No; 2 violations: MW > 350, Rotors > 7	No; 2 violations: MW > 350, Rotors > 7

The subsequent step involved analyzing the ADMET properties of the synthesized drug candidates. The ADMET database offers insights into skin permeability, human intestinal absorption (HIA), CYP450 interactions (as inhibitors or substrates), median lethal dose (LD₅₀) toxicity, and overall drug clearance rates. Both molecules exhibited high absorption efficiency in the human gut (Abs > 30). However, they also displayed increased skin permeability (value > −3) (Table 2). The ATP‐binding cassette (ABC) superfamily includes P‐glycoprotein (P‐gp), a crucial drug transporter that influences the distribution, absorption, excretion, metabolism, and toxicity of its substrates in the gastrointestinal system [59]. Our results indicated that both compounds function as P‐gp substrates (Table 2).

TABLE 2.

ADMET characteristics of Pyr‐TZ‐2O and Pyr‐TZ‐2S.

Descriptor	Properties	Pyr‐TZ‐2O	Pyr‐TZ‐2S
Absorption	Water solubility (log mol/L)	−4.761	−4.937
Absorption	CaCo2 permeability (log papp in 10−6 cm/s)	0.921	1.085
Absorption	Human intestinal absorption (% absorbed)	99.29	94.514
Absorption	Skin permeability (log Kp)	−2.736	−2.742
Absorption	p‐glycoprotein substrate	Yes	Yes
Absorption	p‐glycoprotein inhibitor I	Yes	Yes
Absorption	p‐glycoprotein inhibitor II	Yes	Yes
Distribution	VDss (log L/kg)	−0.272	−0.031
Distribution	Human fraction unbound (Fu)	0.084	0.084
Distribution	BBB permeability (log BB)	−0.712	−0.701
Distribution	CNS permeability (log PS)	−3.41	−2.648
Metabolism	CYP2D6 substrate	No	No
Metabolism	CYP34A substrate	Yes	Yes
Metabolism	CYP1A2 inhibitor	No	No
Metabolism	CYP2C19 inhibitor	Yes	Yes
Metabolism	CYP2C9 inhibitor	Yes	Yes
Metabolism	CYP2D6 inhibitor	No	No
Metabolism	CYP34A inhibitor	Yes	Yes
Excretion	Total clearance (log mL/min/kg)	0.543	−0.134
Excretion	Renal OCT2 substrate	No	No
Toxicity	AMES toxicity	No	No
Toxicity	Maximum tolerated dose (human)	0.621	0.611
Toxicity	hERG I inhibitor	No	No
Toxicity	hERG II inhibitor	Yes	Yes
Toxicity	Oral rat acute toxicity (LD50) (mol/kg)	2.742	2.797
Toxicity	Oral rat chronic toxicity (LOAEL) (log mg/kg_bw/day)	2.202	0.602
Toxicity	Hepatotoxicity	Yes	Yes
Toxicity	Skin sensitization	No	No
Toxicity	T. pyriformis toxicity (log μg/L)	0.29	0.29
Toxicity	Minnow toxicity (log mM)	−0.571	−0.938

Next, we evaluated how the drugs distribute within the body, beginning with the steady‐state volume of distribution (VDss). Drugs that interact with plasma proteins exhibit a lower apparent volume of distribution, while those that bind to tissues present a higher apparent volume of distribution [60]. Our results indicated that the VDss for Pyr‐TZ‐2O and Pyr‐TZ‐2S are −0.272 and −0.031 log L/kg, respectively (Table 2). Besides, we estimated the capacity of these molecules to traverse the BBB. Compounds with a log BB > 0.3 can easily cross the BBB, whereas those with a log BB less than −1 have significant difficulty in doing so [61]. The two compounds exhibited log BB values of −0.712 and −0.701. Other relevant parameters, including unbound percentages and CNS permeability of the tested molecules, can be found in Table 2.

We also investigated how the body metabolizes these drugs. CYP450 enzymes are hemoproteins that function as monooxygenases, responsible for the breakdown of drugs, bile acids, steroids, fatty acids, and even carcinogens [62]. Thus, we investigated how different compounds interacted with these CYP enzymes. The results indicated that both compounds are CYP34A‐substrates. Regarding total clearance, the two candidates demonstrated overall rates ranging from −0.134 to 0.543 log mL/min/kg (Table 2).

Lastly, we evaluated the toxicity profiles of the compounds. Interestingly, none were expected to cause skin sensitization. In acute toxicity tests on rats, the compounds showed LD₅₀ of 2.742 and 2.797 mol/kg (Table 2). In summary, further experimental testing is necessary for these candidates before we can reliably confirm their therapeutic effects.

3.1.5. DFT Analysis of the Compounds

The dipole moments of Pyr‐TZ‐2O (3.2747 D) and Pyr‐TZ‐2S (2.7894 D) indicate a relatively high polarity and increased charge separation, aligning with their electronic properties as determined by DFT. These molecules have low HOMO and LUMO energy levels (Pyr‐TZ‐2O: −0.17561/−0.05048 eV; Pyr‐TZ‐2S: −0.18558/−0.05186 eV) and small HOMO–LUMO gaps (0.12513 and 0.13372 eV, respectively), indicating they are electronically soft systems with a strong tendency for charge transfer, Relatively low excitation energy and potentially increased chemical reactivity; characteristics that are significant for applications in optoelectronics, adsorption‐driven processes, or electron transfer. The molecular electrostatic potential (MEP) surfaces of both compounds further corroborate these observations, showing distinct red and blue regions corresponding to negative and positive electrostatic potential, respectively. These regions indicate the potential sites for electrophilic and nucleophilic interactions. The notable charge separation and extensive π‐electron delocalization across their conjugated frameworks are consistent with the moderate dipole moments, narrow energy gaps, and anticipated strong charge‐transfer behavior of both molecules (Figure 7).

FIGURE 7.

DFT analysis of Pyr‐TZ‐2O and Pyr‐TZ‐2S displaying their calculated IR spectra (A, E), molecular electrostatic potential (MEP) surfaces (B, F), optimized 3D geometries (C, G), and HOMO–LUMO energy gaps (D, H), collectively illustrating the electronic structure, stability, and reactivity of both molecules.

3.2. Chemistry

Based on the previously discussed theoretical studies, Pyr‐TZ‐2O appears to be a promising candidate for further analysis. Consequently, we synthesized, characterized Pyr‐TZ‐2O, and then conducted additional biological evaluations.

3.2.1. Characterization of Compound (1d)

1. ¹ H NMR Spectral data of ethyl 6‐methyl‐2‐oxo‐4‐(4‐((1‐(p‐tolyl)‐1H‐1,2,3‐triazol‐4‐yl) methoxy) phenyl)‐1,2,3,4‐tetrahydropyrimidine‐5‐carboxylate (1d)

¹ H NMR (500 MHz, DMSO‐D ₆ ):

δ 9.10 (s, 1H), 8.84 (s, 1H), 7.74 (d, J = 8.6 Hz, 2H), 7.63 (s, 1H), 7.36 (d, J = 7.8 Hz, 2H), 7.13 (d, J = 8.8 Hz, 2H), 6.98 (d, J = 8.9 Hz, 2H), 5.16 (s, 2H), 5.06 (d, J = 3.4 Hz, 1H), 3.94 (q, J = 7.1 Hz, 2H), 2.34 (s, 3H), 2.20 (s, 3H), 1.05 (t, J = 7.1 Hz, 3H) (Figure S1).

• b ¹³ C NMR Spectral data of ethyl 6‐methyl‐2‐oxo‐4‐(4‐((1‐(p‐tolyl)‐1H‐1,2,3‐triazol‐4‐yl) methoxy) phenyl)‐1,2,3,4‐tetrahydropyrimidine‐5‐carboxylate (1d)

¹³ C NMR (126 MHz, DMSO‐D ₆ ):

δ 165.89, 157.71, 152.65, 148.61, 144.28, 138.91, 138.07, 134.87, 130.77, 127.99, 123.23, 120.56, 115.09, 100.03, 61.57, 59.68, 53.87, 21.11, 18.30, 14.63 (Figure S2).

• c HRMS data of ethyl 6‐methyl‐2‐oxo‐4‐(4‐((1‐(p‐tolyl)‐1H‐1,2,3‐triazol‐4‐yl) methoxy) phenyl)‐1,2,3,4‐tetrahydropyrimidine‐5‐carboxylate (1d)

[M + H]⁺ (Calculated): 448.1979; [M + H]⁺ (Observed): 448.1983 (Figure S3).

3.3. Biological Evaluations

3.3.1. Protein–Protein Interaction (PPI) Network Analysis

The STRING‐based PPI analysis identified the network among PAH, PTS, QDPR, and SPR, pivotal enzymes in the BH₄ recycling activity (Figure 8F). The strong interconnection between these proteins suggests their involvement in phenylalanine hydroxylation and cofactor regeneration, based on well‐established interactions reported in the literature and curated in the STRING database [63]. The cells posttreatment (Figure 8B–E) maintained their specific rounded morphology with clear boundaries and minimal debris, compared to the untreated control (Figure 8A). These findings suggest that cellular integrity was preserved after treatment, indicating the absence of cytotoxicity and maintenance of cellular homeostasis, rather than direct confirmation of restored PAH enzymatic activity.

FIGURE 8.

(A–E) Representative bright‐field images showing post‐Pyr‐TZ‐2O treatment compared to control. Black arrows highlighted that a few cells showed a complete and rounded shape, indicating that the treatment maintained the structural integrity of the cells. (F) STRING‐derived protein–protein interaction network of PAH, and (G–J) posttreatment expression profiles of PAH‐associated genes in PKU mutant cells. Statistical significance was determined relative to the untreated control (*p < 0.05, **p < 0.01, ***p < 0.001).

Afterwards, we assessed mRNA expression of the aforementioned genes postdrug treatment. The results showed the significant upregulation in mRNA expression of these genes, emphasizing consistent activation of the BH₄‐dependent enzymatic network and regeneration pathways that are essential for PAH function (Figure 8G–J). These findings suggest that Pyr‐TZ‐2O induces a coordinated transcriptional response consistent with PAH pathway engagement through gene‐level restoration. In a nutshell, Pyr‐TZ‐2O shows promising pathway‐level activation that may help address limitations of current pharmacological chaperones in PKU, warranting further activity‐based validation.

3.3.2. Pyr‐TZ‐2O Treatment Increases PAH Protein Abundance in PKU‐R252Q Mutant Cells

To validate the effect of Pyr‐TZ‐2O treatment on PAH protein abundance in PKU‐R252Q mutant cells, a sandwich ELISA was performed following treatment with increasing concentrations of the compound, and PAH levels were quantified using a standard curve generated from known PAH standards (Figure 9A).

FIGURE 9.

(A) PAH standard curve used for sandwich ELISA‐based protein quantification. (B) PAH concentrations in untreated (Ctrl) and Pyr‐TZ‐2O‐treated PKU‐R252Q cells. Data are presented as mean ± SD. Statistical significance was determined relative to the untreated control (*p < 0.05, **p < 0.01).

As shown in Figure 9B, treatment with Pyr‐TZ‐2O resulted in a significant increase in PAH protein abundance compared with untreated control cells. The mean PAH concentration in control cells (Ctrl) was 75.76 pg/mL, whereas treatment with 10 μg/mL Pyr‐TZ‐2O increased PAH levels to 596.69 pg/mL (p = 0.022). A further increase was observed at 25 μg/mL, resulting in a PAH concentration of 627.55 pg/mL (p = 0.017). Higher concentrations of the compound produced a marked elevation in PAH abundance, reaching 1070.13 and 1078.95 pg/mL following treatment with 50 and 100 μg/mL Pyr‐TZ‐2O, respectively (p = 0.0013 and p = 0.0012). Notably, a progressive dose‐dependent increase in PAH protein abundance was observed across the treatment range, with the most pronounced effect detected at 50–100 μg/mL.

The observed increase in PAH levels across all treatment groups suggests that Pyr‐TZ‐2O may promote restoration of PAH expression in the mutant cellular model. However, as this represents a preliminary investigation, further studies are required to determine whether the increased PAH protein abundance translates into functional enzymatic activity and restoration of phenylalanine‐to‐tyrosine conversion.

4. Discussion

PKU is recognized as a conformational disorder in which numerous pathogenic PAH variants retain residual catalytic activity but exhibit impaired folding dynamics, reduced thermodynamic stability, defective tetramer assembly, and enhanced intracellular degradation [4]. These structural abnormalities reduce the abundance of functional PAH enzyme and ultimately result in chronic hyperphenylalaninemia and its associated neurological manifestations. Consequently, therapeutic approaches aimed at restoring protein stability have emerged as attractive alternatives to conventional phenylalanine‐lowering strategies [16]. Pharmacological chaperones are particularly promising because they directly target the molecular defects underlying PAH dysfunction by promoting proper folding, stabilization, and intracellular persistence of partially functional variants [22]. Despite the availability of sapropterin, therapeutic responsiveness remains limited to a subset of BH₄‐responsive patients [25], underscoring the need for alternative small molecules capable of stabilizing a broader range of disease‐associated PAH variants.

The rationale of the present study was therefore to identify novel pyrimidine–triazole derivatives with the potential to function as pharmacological chaperones for mutant PAH. The clinically relevant R252Q mutation destabilizes the PAH catalytic region by disrupting local electrostatic interactions, making it an ideal candidate for studying the efficacy of small‐molecule stabilizers. Notably, cell lines harboring the same mutation are available for testing in our lab [54].

The molecular docking analyses demonstrated that Pyr‐TZ‐2O and Pyr‐TZ‐2S exhibited stronger predicted binding affinities and more favorable interaction profiles than the reference pharmacological chaperone sapropterin. The docking observations were further supported by molecular dynamics simulations, which provided insights into the stability of the ligand‐bound complexes. Both Pyr‐TZ derivatives were associated with reduced structural fluctuations, decreased solvent accessible surface area, and lower radius of gyration values. Collectively, these findings suggest enhanced structural compactness and conformational rigidity in the presence of ligands, characteristics commonly associated with improved protein stability. Consistent with these observations, MM‐GBSA analyses revealed favorable binding free energies for both lead compounds, with energetic contributions primarily arising from van der Waals and lipophilic interactions.

In addition, favorable pharmacokinetic characteristics are essential for the successful development of therapeutics. Both Pyr‐TZ‐2O and Pyr‐TZ‐2S complied with Lipinski's rule of five and demonstrated acceptable ADMET profiles, including predicted oral bioavailability, favorable intestinal absorption, and appropriate clearance characteristics. Furthermore, DFT revealed narrow HOMO–LUMO energy gaps and extensive electronic delocalization, supporting the favorable molecular reactivity and adaptability within the PAH binding pocket. These electronic properties may contribute to the stability and versatility of the ligand–protein interactions observed throughout the computational analyses.

Importantly, the computational predictions were complemented by preliminary biological validation in R252Q mutant cells. Treatment with Pyr‐TZ‐2O resulted in significant upregulation of PAH and the BH₄‐pathway genes QDPR, SPR, and PTS, suggesting modulation of molecular networks associated with phenylalanine hydroxylation and cofactor regeneration. In continuation, with these transcriptional findings, sandwich ELISA analysis demonstrated a significant dose‐dependent increase in PAH protein abundance following Pyr‐TZ‐2O treatment, further supporting restoration of the PAH protein at the protein level. The coordinated transcriptional and protein‐level responses observed across multiple components of the PAH–BH₄ axis provide initial evidence supporting the biological relevance of these compounds. Therefore, the observed molecular effects should be regarded as preliminary biological support for the proposed mechanism, rather than as definitive evidence of pharmacological chaperone activity, and further studies are required to establish functional restoration of PAH‐mediated phenylalanine‐to‐tyrosine conversion.

From a translational perspective, this study highlights the potential utility of targeting conformational instability as a therapeutic strategy for PKU. Unlike approaches that solely reduce circulating phenylalanine concentrations, pharmacological chaperones aim to restore endogenous PAH function by correcting the structural defects associated with pathogenic mutations. Such mutation‐responsive therapies may offer advantages in terms of physiological regulation and long‐term management of diseases. Although the current findings are still in the early stages of development, they support the continued exploration of pyrimidine–triazole scaffolds as potential stabilizers of mutant PAH and provide a foundation for future optimization efforts.

4.1. Strengths of the Study

This study combined rational drug design, molecular docking, molecular dynamics simulations, MM‐GBSA analysis, ADMET prediction, synthetic chemistry, and biological validation within an integrated workflow to identify potential PAH‐targeting pharmacological chaperones. The use of the clinically relevant R252Q mutation enhances the translational relevance of the findings, while the pyrimidine–triazole scaffold was rationally designed to promote favorable protein–ligand interactions and drug‐like properties of the compound. Importantly, the lead compounds demonstrated favorable binding characteristics in both wild‐type (WT) and mutant PAH models, and preliminary validation in mutant cells provided experimental support for the computational predictions. Collectively, these findings establish a promising foundation for the development of mutation‐responsive therapeutic strategies for PKU.

4.2. Limitations of the Study

Despite these encouraging findings, several limitations should be acknowledged. First, the study was restricted to the R252Q mutation, whereas PKU comprises a highly heterogeneous spectrum of more than 1500 reported PAH variants; therefore, the broader applicability of Pyr‐TZ compounds remains to be established in other variants. Second, biological validation was limited to gene expression analysis, and the functional consequences of the observed transcriptional changes were not directly assessed. In particular, phenylalanine‐to‐tyrosine conversion has not been evaluated.

Accordingly, the present study should be considered a preliminary proof‐of‐concept study. Future investigations should focus on validating these findings across multiple PAH mutations and determining whether Pyr‐TZ compounds can enhance PAH stability, restore enzymatic activity, promote phenylalanine metabolism, and demonstrate efficacy in relevant in vitro and in vivo models of PKU. Such studies are essential to establish the therapeutic potential of PKU.

5. Conclusion

This study combined computational modeling and preliminary experimental validation to identify pyrimidine–triazole (Pyr‐TZ) derivatives as potential pharmacological chaperones for phenylketonuria. Among the synthesized compounds, Pyr‐TZ‐2O and Pyr‐TZ‐2S demonstrated favorable binding characteristics, stable protein–ligand interactions, and enhanced conformational stabilization of both WT and PAH‐R252Q mutant proteins. Their encouraging ADMET profiles further support their potential as drug‐like agents. At the biological level, treatment of R252Q mutant cells resulted in the upregulation of PAH and key BH₄‐pathway genes (QDPR, SPR, and PTS), suggesting the modulation of molecular networks involved in phenylalanine hydroxylation, cofactor regeneration (Figure 10), and Sandwich ELISA supporting restoration of the PAH at the protein level. Although these findings are limited to transcript‐ and protein‐level analyses, they provide preliminary biological support for computational predictions and highlight the potential of Pyr‐TZ derivatives as PAH‐targeting stabilizers. Collectively, this study establishes an initial proof‐of‐concept framework for developing mutation‐responsive pharmacological chaperones for PKU. By addressing protein destabilization, a fundamental pathogenic mechanism underlying many PAH variants, Pyr‐TZ compounds represent promising lead molecules for future studies. Further biochemical, functional, and in vivo studies are required to validate their therapeutic efficacy; however, the present findings provide a strong foundation for the rational development of next‐generation targeted therapies aimed at restoring endogenous PAH function in patients with phenylketonuria.

FIGURE 10.

Proposed mechanism for restoring the PAH–BH₄ metabolic axis in PKU‐R252Q mutant cells after treatment.

Author Contributions

S.C. collected the data and wrote the manuscript. S.M., S.K., and A.K. designed and synthesized the compounds. N.P., R.N., S.S., Y.P., S.T., A.P., R.K.G., S.K.M., G.T., A.T., and A.S. edited the manuscript. P.K.D. supported, managed, and supervised the work and revised the manuscript.

Funding

This study was supported by the Indian Council of Medical Research, Government of India (Grant No. EMDR/SG/11/2024‐01‐05326).

Conflicts of Interest

The authors declare no conflicts of interest.

Data Availability Statement

The data will be made available upon reasonable request.