Donniphat Dejsuphong

Poly(adenosine diphosphate-ribose) polymerase 1 (PARP1) is a pivotal target for treating homologous recombination-deficient cancers through the mechanism of synthetic lethality. While machine learning has accelerated the identification of novel inhibitors, many models lack experimental validation and high-resolution mechanistic insights. In this study, we evaluated the predictive robustness of the PARP1pred model using a hierarchical pipeline. Initial bioactivity predictions for candidates in unseen chemical space were validated through biochemical and cellular sensitivity assays using a panel of isogenic TK6 cell lines. Subsequently, molecular docking, 100-ns molecular dynamics simulations, and molecular mechanics Poisson–Boltzmann surface area (MM-PBSA) energetic analysis were performed to provide a structural and thermodynamic rationale for the observed inhibitory potencies. The workflow successfully identified ZINC49069486 as a highly potent nanomolar lead that induced selective synthetic lethality in BRCA1-deficient cells. Crucially, the pipeline correctly classified ZINC67913374 as biologically inactive [median inhibitory concentration (IC ₅₀ ) > 100 μM], successfully identifying a false positive previously proposed in the literature as a potential inhibitor. MM-PBSA analysis revealed that the inactivity of ZINC67913374 was driven by an excessive polar solvation penalty that outweighed its favorable docking score. While ZINC98208034 and ZINC8793749 showed moderate enzymatic inhibition, they failed to induce a synthetic lethality response. These results confirm that the PARP1pred-driven hierarchical framework effectively prioritizes experimentally validated lead compound while filtering out deceptive computational hits, providing an accessible and robust strategy for PARP1 inhibitor discovery.

Cholangiocarcinoma (CCA) is a biliary tract carcinoma that is challenging to treat due to its heterogeneity and limited treatment options. Genetic alterations in DNA damage response (DDR) pathways and homologous recombination (HR) defects are common in CCA. This has prompted interest in the use of ataxia telangiectasia and Rad3‑related protein (ATR) and poly(ADP‑ribose) polymerase (PARP) inhibitors to treat CCA. The present study investigated the impact of an ATR inhibitor and various PARP inhibitors, individually and in combination, on CCA cell lines with different DDR mutation profiles. DDR gene alterations in these cell lines were analyzed, and the responses of the cells to treatment with the PARP inhibitors olaparib, veliparib and talazoparib and/or the ATR inhibitor AZD6738 were evaluated. Assessments focused on cellular viability, clonogenic survival and the combination index, alongside changes in DNA damage assessed via the formation of micronuclei and γ‑H2A histone family member X foci. The results revealed that the CCA cell lines with more DDR mutations exhibited greater sensitivity to single and combination treatments. Talazoparib was found to be the most potent PARP inhibitor in the CCA cell lines. The combination of AZD6738 and talazoparib demonstrated varying synergistic effects depending on the genetic background of the CCA cells, with greater efficacy in the cell lines less sensitive to single drug treatments. Mechanistically, this combination promoted the accumulation of DNA damage, including DNA double‑strand breaks. Overall, the study underscores the importance of HR in CCA. It reveals an association between the extent of DDR mutations and the response to AZD6738 and PARP inhibitors in CCA, both as single agents and in combination. These findings highlight that the number of mutated genes influences variability in the drug response.

Next‐generation sequencing (NGS) has transformed pharmacogenomics (PGx), enabling thorough profiling of pharmacogenes using computational methods and advancing personalized medicine. The Thai Pharmacogenomic Database‐2 (TPGxD‐2) analyzed 948 whole genome sequences, primarily from the Electricity Generating Authority of Thailand (EGAT) cohort. This study is an extension of the previous Thai Pharmacogenomic Database (TPGxD‐1) and specifically focused on 26 non‐very important pharmacogenes (VIPGx) genes. Variant calling was conducted using Sentieon (version 201808.08) following GATK's best workflow practices. We then annotated variant call format (VCF) files using Golden Helix VarSeq 2.5.0. Star allele analysis was performed with Stargazer v2.0.2, which called star alleles for 22 of 26 non‐VIPGx genes. The variant analysis revealed a total of 14,529 variants in 26 non‐VIPGx genes, with TBXAS1 had the highest number of variants (27%). Among the 14,529 variants, 2328 were novel (without rsID), with 87 identified as clinically relevant. We also found 56 known PGx variants among the known variants (n = 12,201), with UGT2B7 (19.64%), CYP1B1 (8.9%), SLCO2B1 (8.9%), and POR (8.9%) being the most common. We reported a high frequency of intermediate metabolizers (IMs) in CYP2F1 (34.6%) and CYP4A11 (8.6%), and a high frequency of decreased functional alleles in POR (53.9%) and SLCO1B3 (34.9%) genes. This study enhances our understanding of pharmacogenomic profiling of 26 non‐VIPGx genes of notable clinical importance in the Thai population. However, further validation with additional computational and reference genotyping methods is necessary, and novel alleles identified in this study should undergo further orthogonal validation.

Computational methods analyze genomic data to identify genetic variants linked to drug responses, thereby guiding personalized medicine. This study analyzed 942 whole‐genome sequences from the Electricity Generating Authority of Thailand (EGAT) cohort to establish a population‐specific pharmacogenomic database (TPGxD‐1) in the Thai population. Sentieon (version 201808.08) implemented the GATK best workflow practice for variant calling. We then annotated variant call format (VCF) files using Golden Helix VarSeq 2.5.0 and employed Stargazer v2.0.2 for star allele analysis. The analysis of 63 very important pharmacogenes (VIPGx) reveals 85,566 variants, including 13,532 novel discoveries. Notably, we identified 464 known PGx variants and 275 clinically relevant novel variants. The phenotypic prediction of 15 VIPGx demonstrated a varied metabolic profile for the Thai population. Genes like CYP2C9 (9%), CYP3A5 (45.2%), CYP2B6 (9.4%), NUDT15 (15%), CYP2D6 (47%) and CYP2C19 (43%) showed a high number of intermediate metabolizers; CYP3A5 (41%), and CYP2C19 (9.9%) showed more poor metabolizers. CYP1A2 (52.7%) and CYP2B6 (7.6%) were found to have a higher number of ultra‐metabolizers. The functional prediction of the remaining 10 VIPGx genes reveals a high frequency of decreased functional alleles in SULT1A1 (12%), NAT2 (84%), and G6PD (12%). SLCO1B1 reports 20% poor functional alleles, while PTGIS (42%), SLCO1B1 (4%), and TPMT (5.96%) showed increased functional alleles. This study discovered new variants and alleles in the 63 VIPGx genes among the Thai population, offering insights into advancing clinical pharmacogenomics (PGx). However, further validation is needed using other computational and genotyping methods.

Cancer is the leading cause of death worldwide, resulting in the mortality of more than 10 million people in 2020, according to Global Cancer Statistics 2020. A potential cancer therapy involves targeting the DNA repair process by inhibiting PARP-1. In this study, classification models were constructed using a non-redundant set of 2018 PARP-1 inhibitors. Briefly, compounds were described by 12 fingerprint types and built using the random forest algorithm concomitant with various sampling approaches. Results indicated that PubChem with an oversampling approach yielded the best performance, with a Matthews correlation coefficient > 0.7 while also affording interpretable molecular features. Moreover, feature importance, as determined from the Gini index, revealed that the aromatic/cyclic/heterocyclic moiety, nitrogen-containing fingerprints, and the ether/aldehyde/alcohol moiety were important for PARP-1 inhibition. Finally, our predictive model was deployed as a web application called PARP1pred and is publicly available at https://parp1pred.streamlitapp.com, allowing users to predict the biological activity of query compounds using their SMILES notation as the input. It is anticipated that the model described herein will aid in the discovery of effective PARP-1 inhibitors.

The inhibition of poly(ADP‐ribose) polymerases (PARPs) and ataxia telangiectasia and Rad3‐related (ATR) would be an alternative approach for cancer treatments. The aim of this study is to investigate the synergy of the different combinations of PARP inhibitors (olaparib, talazoparib, or veliparib) and ATR inhibitor AZD6738. A drug combinational synergy screen that combines olaparib, talazoparib, or veliparib with AZD6738 was performed to identify the synergistic interaction, and the combination index was calculated to verify synergy. TK6 isogenic cell lines with defects in different DNA repair genes were used as a model. Cell cycle analysis, micronucleus induction, and focus formation assays of serine‐139 phosphorylation of the histone variant H2AX demonstrated that AZD6738 diminished G2/M checkpoint activation induced by PARP inhibitors and allowed DNA damage‐containing cells to continue dividing, leading to greater increases in micronuclei as well as double‐strand DNA breaks in mitotic cells. We also found that AZD6738 was likely to potentiate cytotoxicity of PARP inhibitors in homologous recombination repair deficiency cell lines. AZD6738 sensitized more genotypes of DNA repair‐deficient cell lines to talazoparib than to olaparib and veliparib, respectively. The combinational approach of PARP and ATR inhibition to enhance response to PARP inhibitors could expand the utility of PARP inhibitors to cancer patients without BRCA1/2 mutations.

Recent developments in chemotherapy focus on target-specific mechanisms, which occur only in cancer cells and minimize the effects on normal cells. DNA damage and repair pathways are a promising target in the treatment of cancer. In order to identify novel compounds targeting DNA repair pathways, two key proteins, 53BP1 and RAD54L, were tagged with fluorescent proteins as indicators for two major double strand break (DSB) repair pathways: non-homologous end-joining (NHEJ) and homologous recombination (HR). The engineered biosensor cells exhibited the same DNA repair properties as the wild type. The biosensor cells were further used to investigate the DNA repair activities of natural biological compounds. An extract from Phyllosticta sp., the endophyte isolated from the medicinal plant Garcinia cowa Roxb. ex Choisy, was tested. The results showed that the crude extract induced DSB, as demonstrated by the increase in the DNA DSB marker γH2AX. The damaged DNA appeared to be repaired through NHEJ, as the 53BP1 focus formation in the treated fraction was higher than in the control group. In conclusion, DNA repair-based biosensors are useful for the preliminary screening of crude extracts and biological compounds for the identification of potential targeted therapeutic drugs.

Ring finger protein 43 (RNF43) is an E3 ubiquitin ligase which is well‐known for its role in negative regulation of the Wnt‐signaling pathway. However, the function in DNA double‐strand break repairs has not been investigated. In this study, we used a lymphoblast cell line, DT40, and mouse embryonic fibroblast as cellular models to study DNA double‐strand break (DSB) repairs. For this purpose, we created RNF43 knockout, RNF43^−/− DT40 cell line to investigate DSB repairs. We found that deletion of RNF43 does not interfere with cell proliferation. However, after exposure to various types of DNA‐damaging agents, RNF43^−/− cells become more sensitive to topoisomerase II inhibitors, etoposide, and ICRF193, than wild type cells. Our results also showed that depletion of RNF43 results in apoptosis upon etoposide‐mediated DNA damage. The delay in resolution of γH2AX and 53BP1 foci formation after etoposide treatment, as well as epistasis analysis with DNAPKcs, suggested that RNF43 might participate in DNA repair of etoposide‐induced DSB via non‐homologous end joining. Disturbed γH2AX foci formation in MEFs following pulse etoposide treatment supported the notion that RNF43 also functions DNA repair in mammalian cells. These findings propose two possible functions of RNF43, either participating in NHEJ or removing the blockage of 5′ topo II adducts from DSB ends.

Tripartite motif‐containing protein 29 (TRIM29) is involved in DNA double‐strand break (DSB) repair. However, the specific roles of TRIM29 in DNA repair are not clearly understood. To investigate the involvement of TRIM29 in DNA DSB repair, we disrupted TRIM29 in DT40 cells by gene targeting with homologous recombination (HR). The roles of TRIM29 were investigated by clonogenic survival assays and immunofluorescence analyses. TRIM29 triallelic knockout (TRIM29^{−/−/−/+}) cells were sensitive to etoposide, but resistant to camptothecin. Foci formation assays to assess DNA repair activities showed that the dissociation of etoposide‐induced phosphorylated H2A histone family member X (ɣ‐H2AX) foci was retained in TRIM29^{−/−/−/+} cells, and the formation of etoposide‐induced tumor suppressor p53‐binding protein 1 (53BP1) foci in TRIM29^{−/−/−/+} cells was slower compared with wild‐type (WT) cells. Interestingly, the kinetics of camptothecin‐induced RAD51 foci formation of TRIM29^{−/−/−/+} cells was higher than that of WT cells. These results indicate that TRIM29 is required for efficient recruitment of 53BP1 to facilitate the nonhomologous end‐joining (NHEJ) pathway and thereby suppress the HR pathway in response to DNA DSBs. TRIM29 regulates the choice of DNA DSB repair pathway by facilitating 53BP1 accumulation to promote NHEJ and may have potential for development into a therapeutic target to sensitize refractory cancers or as biomarker of personalized therapies.

Spinal muscular atrophy (SMA) is one of the leading causes of death in infants and young children from heritable diseases. Patients diagnosed with SMA develop symmetrical progressive muscle weakness and atrophy from degeneration of alpha motor neurons. Approximately 95% of patients have a homozygous deletion of survival motor neuron 1 (SMN1) gene in exon 7 and inherited in autosomal recessive pattern. Considering the high prevalence of SMA carrier in many population, it is possible that SMA is one of the most common autosomal recessive disorders in Thailand and Southeast Asia. In this study, we analyzed DNA from peripheral blood of 505 healthy Thai adults using quantitative PCR-based for SMN1 gene exon 7 copy number analysis. Individual samples with heterozygous deletion of SMN1 gene were confirmed with MLPA. The result identified 9 samples (1.78%) with heterozygous deletion and 39 samples as more than 2 copies of SMN1. No homozygous deletion was detected in the samples. In conclusion, we established carrier frequency of SMA in selected Thai population at 1.8% from 505 participants. The prevalence coincides with prevalence in East Asia and Caucasian population. The result could be implemented for SMA carrier screening in couples at risk in the region.