In silico screening of drug Bank data base to PDE10: A drug repurposing approach

Drug repurposing has emerged as a promising strategy for expediting drug development by identifying new therapeutic applications for existing drugs. In this study employed in silico screening approach to explore the DrugBank database for potential phosphodiesterase 10 (PDE10) inhibitors with applications in neurological, psychiatric disorders and cancer treatment. PDE10 plays a crucial role in regulating cyclic nucleotide levels in the brain and has been implicated in various diseases, including schizophrenia, Parkinson’s, Huntington’s diseases, and certain types of cancer. Through molecular docking, we evaluated the interactions and energetics of 28 candidate inhibitors with PDE10. Notably, 17 candidates met all selection criteria, presenting excellent potential for further investigation. The theoretical inhibitors demonstrated favorable ADMETx properties, and their adverse effects were comparable or lower than controls. These findings indicate the viability of repurposing existing drugs, such as Nebivolol, Fluvastatin, Pioglitazone and others, for PDE10 inhibition in diverse pathologies. Validation of these candidates in preclinical studies may open new avenues for drug development and clinical applications, addressing unmet medical needs in various disorders and cancer treatment.


Introduction
In recent years, drug repurposing has emerged as a promising approach to expedite drug development and reduce costs by identifying new therapeutic applications for existing drugs (1) (2). The use of in silico screening approaches has emerged as a powerful tool for identifying potential drug candidates by computationally analyzing large databases of known drugs and their targets. One promising target for drug repurposing is phosphodiesterase 10 (PDE10), an enzyme that plays a crucial role in regulating the levels of cyclic nucleotides in the brain. PDE10 dysregulation was implicated in various neurological and psychiatric disorders, including schizophrenia (3), Parkinson's (4), and Huntington's diseases (5); recently was implicated in lung and breast cancer (6). Studies have implicated PDE10 in certain types of cancer, like the one mentioned before, because dysregulation of the expression of the enzyme has been linked to altered intracellular signaling pathways involved in cell proliferation, survival, and angiogenesis. Some studies have suggested that PDE10 inhibitors are largely preclinical, conducted in cell lines or animal models, and focused on specific types of cancer, such as colorectal or prostate cancer (7),

Figure 1 PDE10 inhibitors and their clinical uses
The current state of research in the field of PDE10 inhibitors ( Figure 1) is characterized by a growing body of literature that highlights the therapeutic potential of targeting this enzyme. However, the limited number of approved PDE10 inhibitors for clinical use indicates the need for further exploration and discovery of new drug candidates. Controversial and diverging hypotheses exist regarding the specific mechanisms underlying PDE10 inhibition and its potential therapeutic applications. There are three commercial inhibitors of PDE10, those are: Papaverine, Tofisopam and Dipyridamole. In papaverine, regarding neurodegenerative diseases, such as Alzheimer's disease (8) or Parkinson's disease (9), it has not been specifically investigated or recognized as a potential therapeutic agent for these conditions. However, some studies suggest that papaverine as PDE10 inhibitor primarily breaks down cyclic adenosine monophosphate (cAMP) and cyclic guanosine monophosphate (cGMP) in cells (10). A study suggests that Tofisopam could modulate cellular processes involved in neurodegeneration and amyloid-beta accumulation (11). And finally, dipyridamole has been investigated for its potential neuroprotective effects in various neurodegenerative diseases. As it has shown promise in experimental studies by modulating inflammatory responses, reducing oxidative stress, and promoting neuronal survival (12). Also, it has been found to have anti-cancer properties through multiple mechanism. However, further research is needed to understand the underlying mechanisms and explore their clinical utility in cancer treatment (13).
An in silico screening approach to explore the DrugBank database, a comprehensive collection of approved and investigational drugs, in search of potential PDE10 inhibitors. By utilizing computational methods, we aimed to identify existing drugs that have the potential to target PDE10 and repurpose them for the treatment of neurological and psychiatric disorders, and different types of cancer.

DrugBank Data Base Prepared.
All drugs in the DrugBank database (14) were prepared in LigPrep (15) under physiological conditions of pH 7.4 and stable protonation and tautomer states at this condition (16).

PDE10 enzyme prepared.
For the preparation of the study protein, the crystal with code 6IJH (17) from the protein data bank was used, cleaned of co-crystallized molecules and modeled at physiological conditions of pH=7.4 in Protein Preparation Wizard (18) according to the previously established protocol (16).

Docking Molecular
The validation of the crystal used was carried out with its co-crystal, obtaining an RMSD of 0.75 A. The database was studied by molecular docking by Glide (19) with standard precision and flexibility in the catalytic site with the movement of Ser, Thr and Try residues, as well as the formation of disulfide bridge bonds, according to the protocol previously reported (16).

ADMETx
ADME values were obtained from the DrugBank database and those not available were predicted with Schrodinger's QikProp (20). The toxicological data were obtained from the DrugBank database (14) and from bibliographic reports in medical records described in the table below.

Results and discussion
The PDE10 enzyme is responsible for the hydrolysis of cAMP and cGMP, the inhibitors work by blocking the catalytic site of this process to prevent hydrolysis. Given the diverse role of this enzyme, drugs are used for pathologies, however, they are not used based on their efficacy as a PDE10 inhibitor. Through molecular coupling we can observe the energy of each inhibitor with the PDE10, in Table 1 we can observe these inhibitors as well as the interactions of each one with the enzyme, finding those key residues of interaction as well as the energetic limits for selection of new candidates. These inhibitors that are in phase/clinical use present a large number of adverse effects, mainly nausea, vomiting, dizziness, drowsiness, and low blood pressure, given this, new inhibition alternatives are constantly sought, from the database by means of molecular coupling by Schrodinger, a total of 9823 drug interactions with PDE10, however, only 1225 were better than Dipyridamole (reference lower energy inhibitor), and only 176 better than Papaverine (best inhibitor found in silico), giving excellent candidates for study in a model in vitro or in vivo, considering that there are already toxicity studies of these as well as safety and effectiveness window studies. However, depending on the application that is sought to be given, there are other selection criteria, that is, if the application is towards the Central Nervous System (CNS), it must be able to reach it, as well as a minimally invasive administration, since they are longterm treatments; while for associations in cancer, the aim is to contrast the adverse effects of the drug against those of the same pathology. Table 2 shows the selected candidates with higher energy than the selected inhibitor, as well as their clinical use and route of administration, which also met the Lipinsky, Ghose and Veber criteria, resulting in 17 candidates and 9 candidates that do not meet all the criteria but do have high energy.
Intravenous/oral 0/0/0 Table 2 shows the 28 candidates for repositioning. However, Protokylol has no direct use and Indium In-111 pentetreotide and Light green SF yellowish are used as contrast dye, so they are not suitable for pathology treatments associated with PDE10. The rest of the candidates are viable a priori due to their current uses, including some that are already being used against pathologies related to PDE10, such as Nebivolol, Fluvastatin, Pioglitazone and Rosiglitazone used in cardiac pathologies and diabetes. While Raltitrexed, Erlotinib, Rosiglitazone and Pralatrexate are used against cancer, which allows us to infer that these drugs may have more than one molecular target. The remaining 25 candidates were selected based on the pathology on which they could act, according to the ADMETx properties (Table 3). For a drug directed at the CNS it is necessary that it have a good bioavailability, therefore a good adsorption by blood brain barrier (BBB) and also that it has an effect on the CNS, the latter predicted by QikProp, as a value -2 idodeno and -1 good for new drugs on the CNS, resulting in 6 candidates for this type of treatment, Fluvastatin, Pitavastatin, Mirabegron, Pioglitazone, Vilazodone and Plalatrexate, however, these last two have a very low Jm value indicating a high probability of bioaccumulation which rules them out as drugs for CNS, given the type of adverse effects that could present, therefore; 4 candidates would remain. BBB "blood brain barrier", CNS "Central nervous system", LogP "octanol/water partition coefficient", TPSA "surface area polar topology", Jm "Epidermal transfer coefficient" For candidates for repositioning for use against cardiovascular diseases and diabetes, it is necessary that they be adsorbed in blood, but it is preferable that they do not pass or influence CNS, so with the data in Table 3, resulting in 5 new candidates, Nebivolol, Drotaverine, Erlotinib, Rosiglitazone and Tiaganine, as well as Tofisopam, whose inhibitory effect on PDE10 is already known. Analyzing your Jm, Drotavarine, Rosiglitazone or Tagabine would not be recommended given their high predisposition to biocumulation resulting in Nebivolol and Erlotinib as new as well as Tofisopam as known. For the selection of candidates against Cancer, it is difficult to limit the candidates, given the aggressive nature of the pathology, however a low bioaccumulation and null effect on CNS is convenient, resulting in 6, Nebivolol, Prucalopride, Drotaverine, Erlotinib, Rosiglitazone and Dipyrithione, of which Erlotinib and Dipyrithione are already used against some types of cancer, and even Nebivolol, Drotaverine, Erlotinib and Rosiglitazone, were previously candidates against cardiovascular pathologies.
For the selection of candidates against cancer, it is convenient not to limit their selection given the aggressiveness of the pathology, however, it is important to consider a low bioaccumulation and no effect on the CNS. Our analysis yielded 6 candidates: Nebivolol, Prucalopride, Drotaverine, Erlotinib, Rosiglitazone and Dipyrithione, of which Erlotinib and Dipyrithione are currently used for some types of cancer and Nebivolol, Drotaverine, Erlotinib and Rosiglitazone were previously candidates against cardiovascular pathologies.  table 4 we can see that these are at the same level or lower than the controls, several of these adverse effects can even be explained by PDE10 inhibition, such as a drop in blood pressure, giving an overview of largest application of these compounds for further in vitro and in vivo studies.

Conclusion
The potential of in silico screening for drug repurposing to identify PDE10 inhibitors whit therapeutic applications in neurological and psychiatric disorders, as well as cancer treatment. The analysis of DrugBank database led to the discovery of 28 promising candidates, 17 of which fully met the selection criteria. These candidates showed favorable interactions with PDE10 and exhibited acceptable ADMETx properties, suggesting their viability for further investigation. Moving forward, further in vitro and in vivo studies are essentials to validate the efficacy and safety of these PDE10 inhibitors. If successful, these candidates could open new avenues for the treatment of a wide range of disorders, thereby improving patient outcomes and addressing unmet medical needs. Overall, the study contributes to the growing field of drug repurposing and highlights the importance of computational approaches in expediting drug discovery and development.