Abstract
Severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) is an RNA virus belonging to the coronavirus family responsible for coronavirus disease 2019 (COVID-19). It primarily affects the pulmonary system, which is the target of chronic obstructive pulmonary disease (COPD), for which many new compounds have been developed. In this study, phosphodiesterase 4 (PDE4) inhibitors are being investigated. The inhibition of PDE4 enzyme produces anti-inflammatory and bronchodilator effects in the lung by inducing an increase in cAMP concentrations. Piclamilast and rolipram are known selective inhibitors of PDE4, which are unfortunately endowed with common side effects, such as nausea and emesis. The selective inhibition of the phosphodiesterase 4B (PDE4B) subtype may represent an intriguing technique for combating this highly contagious disease with fewer side effects. In this article, molecular docking studies for the selective inhibition of the PDE4B enzyme have been carried out on 21 in-house compounds. The compounds were docked into the pocket of the PDE4B catalytic site, and in most cases, they were almost completely superimposed onto piclamilast. Then, in order to enlarge our study, drug-likeness prediction studies were performed on the compounds under study.
1 Introduction
Severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) is a member of the Coronavirus family responsible for the worldwide pandemic of human respiratory illness coronavirus disease 2019 (COVID-19) [1]. Several studies are addressed to find new drug targets and vaccines to combat this disease [2–6]. Among the different involved targets, cyclic nucleotide phosphodiesterases (PDEs) are being investigated, a large superfamily of enzymes that play a major role in intracellular signaling by controlling tissue cAMP (cyclic 3′,5′-AMP) and cGMP (cyclic 3′,5′-GMP) levels in response to receptor activation. Among the 11 subtypes of the PDE family, PDE4 is the principal cAMP-metabolizing enzyme found in the immune and inflammatory cells. The inhibition of PDE4 may induce an increase in cAMP [7,8] and prolongation of the anti-inflammatory effect with a consequent bronchodilator effect. Thus, PDE4 inhibitors have been proven as anti-inflammatory agents against different pulmonary disorders by inhibiting the release of inflammatory signals and cytokines [9]. This inhibition has been addressed for diverse diseases [10], including inflammatory diseases [11], the topical treatment of psoriasis [12] and atopic dermatitis [13], cancer [14], inflammatory bowel diseases [15,16], neurological disorders [17], diabetic nephropathy [18], and pulmonary dysfunctions in COVID-19, such as chronic obstructive pulmonary disease (COPD) [19,20]. Several studies are being carried out on PDE4 inhibitors in order to understand their mechanism of action. Our study was focused on rolipram and piclamilast (RP 73401) (Figure 1) [21,22]. Rolipram (4-(3-cyclopentyloxy-4-methoxyphenyl)-2-pyrrolidone, A, Figure 1) was designed as a COPD therapeutic target acting as a PDE4 inhibitor [23,24]. PDE4 inhibition by piclamilast (3-cyclopentyloxy-N-(3,5-dichloropyridin-4-yl)-4-methoxybenzamide, B, Figure 1) was effective in preventing blast-induced long-term potentiation deficits [25]. A possible mechanism of action responsible for the anti-inflammatory and immunomodulatory effects of this class of compounds has been recently suggested by Nguyen et al. [26] for tanimilast, a new potent and selective inhaled inhibitor of PDE4 in advanced clinical development for the treatment of COPD. The study was carried out in vitro on a model of dendritic cell activation by SARS-CoV-2 genomic ssRNA (SCV2-RNA). Tanimilast lowered the release of pro-inflammatory cytokines (TNF-α and IL-6), chemokines (CCL3, CXCL9, and CXCL10), and Th1-polarizing cytokines (IL-12, type I IFNs).
PyMOL version of rolipram (a) and piclamilast (b).
PDE4, which consists of different isoforms [27], comprises four subtypes, PDE4A, B, C, and D. Non-selective PDE4 inhibitors, which bind all four PDE4 subtypes simultaneously, produce many promising therapeutic benefits, accompanied, however, by undesired side effects, notably, nausea, diarrhea, and emesis [28]; they may also induce gastroparesis, as demonstrated in mice [29]. Thus, the selective inhibition of a single subtype may be addressed. In particular, our study was aimed at the PDE4B subtype, which has been related to diverse activities. PDE4B inhibitors have also been suggested as promising therapeutic targets for Pseudomonas aeruginosa, which is a frequent cause of hospital-acquired lung infections [30], and other molecules have been designed and studied as potential PDE4B inhibitors to reduce or block inflammatory processes of the respiratory tract [31,32,33]. Indeed, PDE4B regulates the pro-inflammatory toll receptor – tumor necrosis factor α pathway in monocytes, macrophages, and microglial cells. Interestingly, it has been demonstrated that PDE4B, which is the predominant form in the immune and respiratory systems, is involved in human respiratory illness COVID-19, the worldwide pandemic caused by SARS-CoV-2 [34,35]. Therefore, it represents a molecular target for anti-inflammatory and antiviral drugs. Unfortunately, to date, developing selective PDE4B inhibitors is not easy as the amino acid sequence of the PDE4 active site is identical in all PDE4 subtypes (PDE4A-D) [36]. However, recently, a modeling study was carried out on PDE4B and PDE4D active cavities [37]. Among the PDE4 selective inhibitors, roflumilast, which shows a higher affinity to PDE4B than PDE4A, C, and D, represents a potential and effective therapy for COVID-19 [38]. Moreover, a PDE4B selective inhibitor, BI 1015550, has been proposed as a clinical drug candidate for the oral treatment of idiopathic pulmonary fibrosis [39]. In this view, our work aimed at designing new selective inhibitors of PDE4B as potential therapeutic agents for COVID-19 disease. Specifically, 21 in-house molecules were examined and correlated with the structures of piclamilast and rolipram in order to seek greater possible interactions with the catalytic site of the A chain of PDE4 [40]. Recently, a modeling study was carried out on PDE4B and PDE4D active cavities [37]. Piclamilast and rolipram were chosen since, in the literature, they are among the major PDE4 inhibitors (IC50 = 1.1 and 1,200 nM for piclamilast and rolipram, respectively) [41]. The magnesium and zinc ions present at the PDE4B catalytic site were taken into account for the docking study using AutoDock Vina software, as they could be involved in PDE4B inhibitory activity. Drug-likeness predictions, i.e., physicochemical-pharmacokinetic and absorption, distribution, metabolism, and excretion (ADME) properties, of the studied compounds have also been performed. These studies will allow us to select the best compounds to be proposed for future in-depth in vitro and/or in vivo studies. We are sure that identifying potentially active molecules could be important for the discovery and development of new drugs for COVID-19 and at the same time useful for the study of the molecular mechanisms involved in this devastating disease.
2 Methodology
2.1 Molecular modeling studies
A molecular docking study was performed on piclamilast and rolipram in the catalytic site of PDE4B, located mainly on the A chain of PDE4B (PDB code: 1XM4) [42]. AutoDock Vina was used for docking [43]. The magnesium and zinc ions present at the catalytic site of PDE4B were also considered. With this study, it is possible to predict the structure of the ligand–protein intermolecular complex and determine what is known as Pose or Binding Mode, which is the position, orientation, and conformation that each ligand assumes on the surface of the biological receptor macromolecule. All the side chains of the flexible amino acid residues selected have been included in the flex file, that is, those side chains present in the active site on the A chain of PDE4B in that could have a probable interaction with the ligand (Figure 2). In particular, the selected residues are ARG‘409, ASN‘235, ASN‘283, ASN‘395, ASP‘275, ASP‘346, ASP‘392, CME‘432, GLN‘284, GLN‘443, GLU‘304, HIS‘234, HIS‘238, HIS‘274, HIS‘278, HIS‘307, ILE‘410, ILE‘450, LEU‘303, LEU‘393, MET‘347, MET‘411, MET‘431, PHE‘414, PHE‘446, SER‘229, SER‘236, SER‘282, SER‘348, SER‘429, SER‘442, THR‘345, THR‘407, TYR‘233, TYR‘403, TYR‘449. Also, for the ligand, a file with the extension.pdbqt has been created, in which its atoms have been defined, also its potentials and its degrees of freedom. This last process was obviously repeated for all ligands tested.
Catalytic domain of human PDE4B in complex with piclamilast (in magenta).
2.2 Drug-likeness predictions – physicochemical-pharmacokinetic/ADME properties
Drug-likeness is one of the qualitative ideas employed for predicting drug-like properties. It is designated as an intricate balance of diverse molecular and structural features that assesses qualitatively the chance for a molecule to become an oral drug with respect to bioavailability. The targeted molecules were evaluated for predicting drug-likeness based on five separate filters, namely, Egan, Ghose, Muegge, Veber, and Lipinski rules accompanying bioavailability and drug-likeness scores using the Molsoft software and Swiss ADME program (http://swissadme.ch) using the ChemAxon’s Marvin JS structure drawing tool [44–47]. Drug-likeness was established from structural or physicochemical inspections of development compounds advanced enough to be considered oral drug candidates.
3 Results and discussion
3.1 Piclamilast
Piclamilast is a selective second-generation PDE4 inhibitor with important anti-inflammatory effects [48]. It has been studied for its applications in treating conditions such as COPD, bronchopulmonary dysplasia, and asthma. It acts through the selective inhibition of the four PDE4 isoforms (PDE4A-D) while showing no inhibition of the other PDEs [49]. PDE4 isoforms are particularly important for inflammatory and immunomodulatory cells. They are the most common PDE in inflammatory cells such as mast cells, neutrophils, basophils, eosinophils, T lymphocytes, macrophages, and structural cells such as sensory nerves and epithelial cells. Inhibition of PDE4 blocks the hydrolysis of cAMP, thereby increasing the levels of cAMP within the cells. cAMP suppresses the activity of immune and inflammatory cells; thus, the inhibition of PDE4 in a mouse model of induced chronic lung disease demonstrated anti-inflammatory properties, determined the reduction of pulmonary fibrin deposition and alveolar vascular loss, and prolonged survival in hyperoxia-induced neonatal lung injury. Moreover, a PDE4 inhibition study in a mouse model of allergic asthma showed that piclamilast significantly improved lung function, airway inflammation, and goblet cell hyperplasia. Recently, the neuroprotective effect of piclamilast has been studied. It has been suggested that post-ischemia pharmacological treatment with piclamilast determines an improvement of cerebral ischemia–reperfusion injury in mice [50]. The docking study with piclamilast gave the following resulting poses (Table 1).
Docking study with piclamilast
| Pose | Binding energy “affinity” (kcal/mol) | RMSDL.b. | RMSDU.b. |
|---|---|---|---|
| 1 | −10.1 | 0.000 | 0.000 |
| 2 | −9.5 | 0.851 | 1.392 |
| 3 | −9.4 | 0.807 | 2.110 |
| 4 | −8.4 | 1.023 | 2.553 |
| 5 | −8.4 | 1.480 | 2.512 |
| 6 | −8.0 | 1.276 | 2.087 |
| 7 | −7.8 | 1.624 | 4.096 |
| 8 | −7.8 | 1.734 | 2.126 |
| 9 | −7.7 | 1.520 | 2.351 |
The average RMSD of the poses are compared with pose 1.
The docking results show a similar bonding energy for pose 2 and pose 3, equal to −9.5 and −9.4 kcal/mol, respectively. The nine poses obtained from the docking were compared to the pose of piclamilast obtained experimentally with PyMol.
3.2 Rolipram
Rolipram is a selective PDE4 inhibitor discovered and developed as a potential antidepressant drug in the early 1990s. It has been used as a prototype molecule for drug discovery and pharmaceutical research development by several companies. The study on rolipram was stopped after clinical trials showed that its therapeutic window was too narrow; it could not be used at levels high enough to be effective without causing significant gastrointestinal (GI) side effects. Nevertheless, rolipram has several activities that make it a continuing target for research [51–54]. Rolipram has been used to study whether PDE4 inhibition could be useful in autoimmune diseases, Alzheimer’s disease, cognitive enhancement, spinal cord injury, and respiratory diseases such as asthma and COPD [55]. In this study, rolipram was used to better understand the exact level of exhaustiveness to be used to set the data when starting molecular docking. The data obtained with an exhaustiveness of 16 are satisfactory and comparable to those obtained with piclamilast. The results of the docking study with rolipram gave the following poses (Table 2).
Docking study with rolipram
| Pose | Binding energy “affinity” (kcal/mol) | RMSDL.b. | RMSDU.b. |
|---|---|---|---|
| 1 | −8.4 | 0.000 | 0.000 |
| 2 | −8.4 | 1.124 | 2.156 |
| 3 | −8.2 | 1.322 | 3.113 |
| 4 | −8.2 | 1.291 | 2.763 |
| 5 | −8.1 | 1.355 | 2.998 |
| 6 | −8.1 | 1.602 | 2.875 |
| 7 | −7.8 | 1.237 | 1.902 |
| 8 | −7.7 | 1.354 | 2.961 |
| 9 | −7.5 | 1.458 | 2.124 |
The average RMSD of the poses are compared with pose 1.
The docking results show the same binding energy for pose 1 and pose 2, equal to −8.5 kcal/mol. The nine poses obtained from docking were compared with the data obtained experimentally by PyMol. The conformation that overlaps almost perfectly is pose 1 (Figure 3).
Comparison between the crystalline structure of piclamilast (in magenta) with pose 1 of the docking of rolipram (in blue). Part of the catalytic site of PDE4B with part of the flexible residues is visible in green.
From the data obtained with different degrees of exhaustiveness (4, 8, 12, 14, 16, 20) for piclamilast and rolipram, we concluded that level 16 was the most favorable, which gave the results most similar to the experimentally obtained structure, and it was therefore chosen for the docking of all the synthetic compounds.
3.3 Other synthetic compounds
Docking studies were subsequently carried out on 21 compounds designed in our laboratories and with structural similarity with both rolipram and piclamilast, with an exhaustiveness level of 16 for all the molecules under examination [56,57]. Some of them are described in the literature for their synthesis and biological activities [58–65]. Docking studies will help us verify the possible inhibitory activity of our synthetic molecules and predict the possible interaction on the catalytic site of PDE4 in vitro. The different energy levels of molecules under study, along with their molecular structure, are represented in Table 3.
Energy levels of the first 5 poses out of a total of 9 of each molecule (whose name is indicated in the first column) with an exhaustiveness level of 16, obtained by molecular docking compared to piclamilast and rolipram with the program AutoDock Vina
| Compound | Structure | 1 | 2 | 3 | 4 | 5 |
|---|---|---|---|---|---|---|
| Piclamilast |
|
−10.1 | −9.5 | −9.4 | −8.4 | −8.4 |
| Rolipram |
|
−8.4 | −8.4 | −8.2 | −8.2 | −8.1 |
| 1 |
|
−6.8 | −6.8 | −6.8 | −6.7 | −6.6 |
| 2 |
|
−9.1 | −8.6 | −8.3 | −8.3 | −8.0 |
| 3 |
|
−9.5 | −9.4 | −9.3 | −9.2 | −9.0 |
| 4 |
|
−10.2 | −10.0 | −9.7 | −9.6 | −8.6 |
| 5 |
|
−10.1 | −9.9 | −9.8 | −9.8 | −9.6 |
| 6 |
|
−9.8 | −9.5 | −9.5 | −9.1 | −7.9 |
| 7 |
|
−7.5 | −7.5 | −7.3 | −7.1 | −6.9 |
| 8 |
|
−7.4 | −7.1 | −7.1 | −7.1 | −7.0 |
| 9 |
|
−9.1 | −9.1 | −8.8 | −8.7 | −8.6 |
| 10 |
|
−7.5 | −7.3 | −7.3 | −7.3 | −7.1 |
| 11 |
|
−8.3 | −8.3 | −8.1 | −8.0 | −8.0 |
| 12 |
|
−10.0 | −9.1 | −8.9 | −8.8 | −8.6 |
| 13 |
|
−8.0 | −7.6 | −7.6 | −7.2 | −7.1 |
| 14 |
|
−7.2 | −7.0 | −6.9 | −6.8 | −6.2 |
| 15 |
|
−7.8 | −7.7 | −7.7 | −7.6 | −7.3 |
| 16 |
|
−10.0 | −9.9 | −9.5 | −9.4 | −9.3 |
| 17 |
|
−7.9 | −7.9 | −7.9 | −7.4 | −6.8 |
| 18 |
|
−7.5 | −7.2 | −7.0 | −6.7 | −6.5 |
| 19 |
|
−7.0 | −6.9 | −6.7 | −6.7 | −6.6 |
| 20 |
|
−7.0 | −6.5 | −6.4 | −6.1 | −6.1 |
| 21 |
|
−7.6 | −7.3 | −7.3 | −7.0 | −7.0 |
Both compounds 3 (pose 3) and 16 (pose 1) interact with PDE4B mainly through hydrophobic interactions (Figure 4a and b). In the docking solutions, the two compounds are deeply buried in the active site of PDE4B, which covers about 90% of their accessible surface area. They interact mostly with the same PDE4B residues and the two divalent cations. It is especially noteworthy that aromatic rings of both 3 and 16 form π–π interactions with F446, which together with Q443 bind cAMP as part of the active site. Compound 16 makes an additional hydrogen bond with S282 and a very strong bond with Mg2+ (Figure 4b). Compared to piclamilast and rolipram, which both form two hydrogen bonds with Q443 and otherwise bind many of the same residues and ions as compounds 3 and 16, these new compounds interact with many more residues in the active site, probably due to the potential of their slightly larger size.
Docking of compounds 3 and 16 in the active site of PDE4B. Compound 3 (pose 3 in (a)) is shown with carbons in green, and compound 16 (pose 1 in (b)) with carbons in orange. The molecular surface of PDE4B is displayed in light blue, and the flexible side chains with carbons in magenta, of which the ones interacting with the ligands are labeled. Strong interactions and hydrogen bonds are indicated with dashed yellow lines and π–π interactions with transversal yellow lines. Rolipram and piclamilast are shown in lines with carbons in yellow and magenta, respectively.
3.4 Drug-likeness predictions – physicochemical-pharmacokinetic/ADME properties
3.4.1 Physicochemical-Pharmacokinetic properties
Drug-likeness is examined as an important part that provides the base for the molecules to be powerful oral drug candidates. Various rules, namely, Lipinski, Ghose, Veber, Egan, and Muegge, were considered to measure drug-likeness of the candidate compounds to find out whether they can be bioactive oral drug candidates according to some acute criterion like molecular weight, LogP, number of hydrogen bond acceptors, and donors. The number of violations of the above-disclosed rules, along with bioavailability and drug-likeness scores, are given in Table 4.
Drug-likeness predictions and physicochemical-pharmacokinetic/ADME properties of tested compounds
| No | MW | Number of HBAa | Number of HBDb | Log P o/w (iLOGP)c | Log Sd | TPSAe | BBB permeantf | GI absorption | Lipinski, ghose, veber, egan, and muegge violations | Bioavailability score | Drug-likeness model score |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | 251.24 | 5 | 0 | 2.28 | Soluble | 81.35 | No | High | 0 | 0.55 | −1.20 |
| 2 | 318.21 | 1 | 0 | 3.09 | Poorly soluble | 14.16 | Yes | High | 0 | 0.55 | −1.11 |
| 3 | 293.41 | 2 | 1 | 3.46 | Moderately soluble | 20.20 | Yes | High | 0 | 0.55 | 0.40 |
| 4 | 320.45 | 2 | 0 | 0.00 | Moderately soluble | 11.41 | Yes | High | 0 | 0.55 | 0.09 |
| 5 | 461.38 | 2 | 0 | 0.00 | Poorly soluble | 11.41 | No | High | 0 | 0.55 | 0.40 |
| 6 | 288.18 | 0 | 0 | 3.02 | Poorly soluble | 4.93 | Yes | High | 0 | 0.55 | −1.60 |
| 7 | 328.16 | 4 | 1 | 2.84 | Moderately soluble | 60.55 | Yes | High | 0 | 0.55 | −0.46 |
| 8 | 249.26 | 4 | 0 | 3.00 | Soluble | 49.69 | Yes | High | 0 | 0.55 | −0.73 |
| 9 | 310.35 | 3 | 1 | 3.04 | Moderately soluble | 52.49 | Yes | High | 0 | 0.55 | 0.09 |
| 10 | 259.34 | 2 | 0 | 3.07 | Moderately soluble | 31.23 | Yes | High | 0 | 0.55 | −0.72 |
| 11 | 309.36 | 3 | 0 | 3.40 | Moderately soluble | 40.46 | Yes | High | 0 | 0.55 | −0.77 |
| 12 | 251.32 | 1 | 0 | 2.68 | Moderately soluble | 20.31 | Yes | High | 0 | 0.55 | −0.04 |
| 13 | 323.58 | 3 | 0 | 2.51 | Moderately soluble | 43.60 | Yes | High | 0 | 0.55 | −0.55 |
| 14 | 323.58 | 3 | 0 | 2.61 | Moderately soluble | 43.60 | Yes | High | 0 | 0.55 | −0.38 |
| 15 | 318.17 | 3 | 1 | 2.52 | Moderately soluble | 55.63 | Yes | High | 0 | 0.55 | 0.07 |
| 16 | 343.42 | 3 | 1 | 3.31 | Moderately soluble | 55.63 | Yes | High | 0 | 0.55 | 0.37 |
| 17 | 332.20 | 3 | 0 | 2.82 | Moderately soluble | 48.64 | Yes | High | 0 | 0.55 | 0.12 |
| 18 | 318.17 | 3 | 1 | 2.54 | Moderately soluble | 55.63 | Yes | High | 0 | 0.55 | 0.10 |
| 19 | 281.26 | 6 | 0 | 2.57 | Soluble | 90.58 | No | High | 0 | 0.55 | −1.18 |
| 20 | 203.19 | 3 | 1 | 1.65 | Soluble | 59.16 | Yes | High | 0 | 0.55 | −1.30 |
| 21 | 221.21 | 4 | 2 | 1.54 | Soluble | 71.55 | Yes | High | 0 | 0.55 | −1.10 |
aNumber of hydrogen bond acceptors; bnumber of hydrogen bond donors; clipophilicity; dwater solubility (SILICOS-IT [S = Soluble]); etopological polar surface area (Å2); fblood–brain barrier permeant.
The results revealed that none of the compounds violated any rule, and their bioavailability score was around 0.55. All the tested molecules could pass the blood–brain barrier (BBB) except compounds 1, 5, and 19. All compounds exhibited moderate to good drug-likeness scores ranging from −1.60 to 0.40. The bioavailability radar of best-predicted compounds is displayed in Figure 5. Compounds 3 and 16 appeared to be the most promising in the in silico predictions, with a drug-likeness score of 0.40 and 0.37, respectively, without any rule violation.
![Figure 5
Bioavailability radar of the tested compounds. The pink area represents the optimal range for each property for oral bioavailability (lipophilicity [LIPO]: XLOGP3 between −0.7 and +5.0, molecular weight [SIZE]: MW between 150 and 500 g/mol, polarity [POLAR] TPSA between 20 and 130 Å2, solubility (INSOLU): log S not higher than 6, saturation (INSATU): fraction of carbons in the sp3 hybridization not less than 0.25, and flexibility (FLEX): no more than nine rotatable bonds).](/document/doi/10.1515/bmc-2022-0033/asset/graphic/j_bmc-2022-0033_fig_005.jpg)
Bioavailability radar of the tested compounds. The pink area represents the optimal range for each property for oral bioavailability (lipophilicity [LIPO]: XLOGP3 between −0.7 and +5.0, molecular weight [SIZE]: MW between 150 and 500 g/mol, polarity [POLAR] TPSA between 20 and 130 Å2, solubility (INSOLU): log S not higher than 6, saturation (INSATU): fraction of carbons in the sp3 hybridization not less than 0.25, and flexibility (FLEX): no more than nine rotatable bonds).
Moreover, all the tested compounds displayed high GI absorption, and most of them are P-gp (p-glycoprotein) non-inhibitors. The predictions for the passive BBB permeation, HIA (human GI absorption), and P-gp substrates are displayed together in the BOILED-Egg diagram, as shown in Figure 6.
BOILED-Egg diagram of the selected compounds 1–21.
3.4.2 ADME properties
The early evaluation of ADMET properties of drug candidates plays an important role in the research and development of new drugs. Taking into account that the existing methods for evaluating ADME-Tox properties are expensive and time-consuming and usually require extensive animal testing the computer modeling techniques for ADME-Tox prediction are more preferable method in early drug discovery. Thus, “drug-like” molecules were evaluated in silico for their ADME properties (Table 5) in order to rapidly screen multiple properties [40]. Compounds that have been predicted to exhibit a high BBB, low water solubility, and poor Caco2-permeability were excluded from potential hits. The server pkCSM [45] was used for this purpose. pkCSM relies on graph-based signatures. These encode distance patterns between atoms to represent the small molecule and to train predictive models.
Main pharmacokinetic descriptors studied on pkCSM predictive models
|
WSol, water solubility in 25°C (mg/L); Caco2, permeability of Caco2 cell line (Papp in ×10−6 cm/s) high permeability of Caco2 would translate in values >0.90; HIA, human intestinal absorption (% absorbed); BBB, represents the BBB permeability as logBB (the logarithmic ratio of brain to plasma concentrations) LogBB > 0.3 cross the brain, while logBB < −1 is poorly distributed to the brain; CNS, blood–brain permeability-surface area product as (logPS) compounds with logPS > −2 are considered to penetrate the CNS; CYP2D6s, substrate for CYP450 isoform 2D6; CYP3A4s, substrate for CYP450 isoform 3A4.
4 Conclusions
In order to identify new molecules to be investigated in the fight against COVID-19, a docking study, through the AutoDock Vina program, was carried out. The possible interactions of 21 in-house compounds with the catalytic site of the A chain of PDE4B were investigated, identifying the poses of the molecules with high similarity to piclamilast and rolipram in the most favorable spatial conformation of interaction with the catalytic site of PDE4B. These compounds were inserted into the pocket of the catalytic site of the PDE4B, and in many cases, they overlapped piclamilast almost completely. The obtained results indicated that the designed compounds might represent promising ligands for PDE4B receptors, and therefore, they deserve further in-depth in vitro studies. Moreover, drug-likeness was carried out on these compounds. The most interesting compounds in the in silico predictions were 3 and 16, showing drug-likeness scores of 0.40 and 0.37, respectively, without any rule violation.
5 Future perspectives
This interesting study requires further investigations of the selected compounds in terms of in vitro cytotoxicity and in vivo studies. The promising in silico and ADME results obtained urge us to proceed with inhibition assays, which will be carried out on isolated PDE4B enzymes in the not-too-distant future. We are confident that these results may lead to recognizing molecules active and selective as PDE4B inhibitors to be developed as new agents active at the pulmonary level for COVID-19 patients.
-
Funding information: This work was supported by PRIN (Progetti di Rilevante Interesse Nazionale) Grant 2017M8R7N9_004 and 2020KSY3KL_005 from MUR, Italy (S.A.).
-
Author contributions: Conceptualization, C.S. and G.S.; writing – original draft preparation, M.G.B. and V.I.; modeling studies, F.G. and M.M.; methodology, S.A. and M.S.S.; validation, A.G.; writing – review and editing, A.C.; drug-likeness and ADME, A.P.; supervision, M.S.S. All authors have read and agreed to the published version of the manuscript.
-
Conflict of interest: Authors state no conflict of interest.
-
Data availability statement: The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.
References
[1] Iacopetta D, Ceramella J, Catalano A, Saturnino C, Pellegrino M, Mariconda A, et al. COVID-19 at a glance: An up-to-date overview on variants, drug design and therapies. Viruses. 2022;14:573.10.3390/v14030573Search in Google Scholar PubMed PubMed Central
[2] Anjum F, Mohammad T, Asrani P, Shafie A, Singh S, Yadav DK, et al. Identification of intrinsically disorder regions in non-structural proteins of SARS-CoV-2: New insights into drug and vaccine resistance. Mol Cell Biochem. 2022;477(5):1607–19.10.1007/s11010-022-04393-5Search in Google Scholar PubMed PubMed Central
[3] Shamsi A, Mohammad T, Anwar S, AlAjmi MF, Hussain A, Rehman M, et al. Glecaprevir and Maraviroc are high-affinity inhibitors of SARS-CoV-2 main protease: possible implication in COVID-19 therapy. Biosci Rep. 2020;40(6):BSR20201256.10.1042/BSR20201256Search in Google Scholar PubMed PubMed Central
[4] Shamsi A, Mohammad T, Anwar S, Amani S, Khan MS, Husain FM, et al. Potential drug targets of SARS-CoV-2: From genomics to therapeutics. Int J Biol Macromol. 2021;177:1–9.10.1016/j.ijbiomac.2021.02.071Search in Google Scholar PubMed PubMed Central
[5] Catalano A, Iacopetta D, Ceramella J, Di Maio AC, Basile G, Giuzio F, et al. Are nutraceuticals effective in COVID-19 and post-COVID prevention and treatment? Foods. 2022;1(18):2884.10.3390/foods11182884Search in Google Scholar PubMed PubMed Central
[6] Ceramella J, Iacopetta D, Sinicropi MS, Andreu I, Mariconda A, Saturnino C, et al. Drugs for COVID-19: An update. Molecules. 2022;27(23):8562.10.3390/molecules27238562Search in Google Scholar PubMed PubMed Central
[7] Conti M, Richter W, Mehats C, Livera G, Park JY, Jin C. Cyclic AMP-specific PDE4 phosphodiesterases as critical components of cyclic AMP signaling. J Biol Chem. 2003;278(8):5493–96.10.1074/jbc.R200029200Search in Google Scholar PubMed
[8] Yan KUO, Gao LN, Cui YL, Zhang YI, Zhou XIN. The cyclic AMP signaling pathway: Exploring targets for successful drug discovery. Mol Med Rep. 2016;13(5):3715–23.10.3892/mmr.2016.5005Search in Google Scholar PubMed PubMed Central
[9] Yougbare I, Morin C, Senouvo FY, Sirois C, Albadin R, Lugnier C, et al. NCS 613, a potent and specific PDE4 inhibitor, displays anti-inflammatory effects on human lung tissues. Am J Physiol Lung Cell Mol Physiol. 2011;301:L441–50.10.1152/ajplung.00407.2010Search in Google Scholar PubMed
[10] Liu Z, Liu M, Cao Z, Qiu P, Song G. Phosphodiesterase-4 inhibitors: a review of current developments (2013–2021). Exp Opin Ther Patents. 2022;32(3):261–78.10.1080/13543776.2022.2026328Search in Google Scholar PubMed
[11] Li H, Zuo J, Tang W. Phosphodiesterase-4 inhibitors for the treatment of inflammatory diseases. Front Pharmacol. 2018;9:1048.10.3389/fphar.2018.01048Search in Google Scholar PubMed PubMed Central
[12] Zebda R, Paller AS. Phosphodiesterase 4 inhibitors. J Am Acad Dermatol. 2018;78(3):S43–S52.10.1016/j.jaad.2017.11.056Search in Google Scholar PubMed
[13] Yang H, Wang J, Zhang X, Zhang Y, Qin ZL, Wang H, et al. Application of topical phosphodiesterase 4 inhibitors in mild to moderate atopic dermatitis: a systematic review and meta-analysis. JAMA Dermatol. 2019;155(5):585–93.10.1001/jamadermatol.2019.0008Search in Google Scholar PubMed PubMed Central
[14] Ye W, Xu JP, Wang HT, Li XF, Wang WY, Zhou ZZ. Discovery of novel trimethoxyphenylbenzo[d]oxazoles as dual tubulin/PDE4 inhibitors capable of inducing apoptosis at G2/M phase arrest in glioma and lung cancer cells. Eur J Med Chem. 2021;224:113700.10.1016/j.ejmech.2021.113700Search in Google Scholar PubMed
[15] Banner KH, Trevethick MA. PDE4 inhibition: a novel approach for the treatment of inflammatory bowel disease. Trends Pharmacol Sci. 2004;25(8):430–6.10.1016/j.tips.2004.06.008Search in Google Scholar PubMed
[16] Di Maio AC, Basile G, Iacopetta D, Catalano A, Ceramella J, Cafaro D, et al. The significant role of nutraceutical compounds in ulcerative colitis treatment. Curr Med Chem. 2022;29:4216–34.10.2174/0929867329666211227121321Search in Google Scholar PubMed
[17] Bhat A, Ray B, Mahalakshmi AM, Tuladhar S, Nandakumar DN, Srinivasan M, et al. Phosphodiesterase-4 enzyme as a therapeutic target in neurological disorders. Pharmacol Res. 2020;160:105078.10.1016/j.phrs.2020.105078Search in Google Scholar PubMed
[18] Ookawara M, Nio Y. Phosphodiesterase 4 inhibitors in diabetic nephropathy. Cell Sign. 2022;90:110185.10.1016/j.cellsig.2021.110185Search in Google Scholar PubMed
[19] Lugnier C, Al-Kuraishy HM, Rousseau E. PDE4 inhibition as a therapeutic strategy for improvement of pulmonary dysfunctions in COVID-19 and cigarette smoking. Biochem Pharmacol. 2021;185:114431.10.1016/j.bcp.2021.114431Search in Google Scholar PubMed PubMed Central
[20] Mokra D, Mokry J. Phosphodiesterase inhibitors in acute lung injury: What are the perspectives? Int J Mol Sci. 2021;22:1929.10.3390/ijms22041929Search in Google Scholar PubMed PubMed Central
[21] Aragon IV, Boyd A, Abou Saleh L, Rich J, McDonough W, Koloteva A, et al. Inhibition of cAMP-phosphodiesterase 4 (PDE4) potentiates the anesthetic effects of Isoflurane in mice. Biochem Pharmacol. 2021;186:114477.10.1016/j.bcp.2021.114477Search in Google Scholar PubMed PubMed Central
[22] Çifci G, Aviyente V, Akten ED. Molecular docking study based on pharmacophore modeling for novel phosphodiesterase iv inhibitors. Mol Inform. 2012;31(6–7):459–71.10.1002/minf.201100141Search in Google Scholar PubMed
[23] Lugnier C, Schoeffter P, Le Bec A, Strouthou E, Stoclet JC. Selective inhibition of cyclic nucleotide phosphodiesterases of human, bovine and rat aorta. Biochem Pharmacol. 1986;35(10):1743–51.10.1016/0006-2952(86)90333-3Search in Google Scholar PubMed
[24] Crocetti L, Floresta G, Cilibrizzi A, Giovannoni MP. An Overview of PDE4 inhibitors in clinical trials: 2010 to Early 2022. Molecules. 2022;27(15):4964.10.3390/molecules27154964Search in Google Scholar PubMed PubMed Central
[25] Vogel III EW, Morales FN, Meaney DF, Bass CR, Morrison III B. Phosphodiesterase-4 inhibition restored hippocampal long term potentiation after primary blast. Exp Neurol. 2017;293:91–100.10.1016/j.expneurol.2017.03.025Search in Google Scholar PubMed PubMed Central
[26] Nguyen HO, Schioppa T, Tiberio L, Facchinetti F, Villetti G, Civelli M, et al. The PDE4 inhibitor tanimilast blunts proinflammatory dendritic cell activation by SARS-CoV-2 ssRNAs. Front Immunol. 2022;12:797390.10.3389/fimmu.2021.797390Search in Google Scholar PubMed PubMed Central
[27] Paes D, Hermans S, van den Hove D, Vanmierlo T, Prickaerts J, Carlier A. Computational investigation of the dynamic control of cAMP signaling by PDE4 isoform types. Biophys J. 2022;121(14):2693–711.10.1016/j.bpj.2022.06.019Search in Google Scholar PubMed PubMed Central
[28] Boyd A, Aragon IV, Rich J, McDonough W, Oditt M, Irelan D, et al. Assessment of PDE4 Inhibitor-induced hypothermia as a correlate of nausea in mice. Biology. 2021;10(12):1355.10.3390/biology10121355Search in Google Scholar PubMed PubMed Central
[29] McDonough W, Aragon IV, Rich J, Murphy JM, Saleh LA, Boyd A, et al. PAN-selective inhibition of cAMP-phosphodiesterase 4 (PDE4) induces gastroparesis in mice. FASEB J. 2020;34:12533–48.10.1096/fj.202001016RRSearch in Google Scholar PubMed PubMed Central
[30] Azam MA, Tripuraneni NS. Selective phosphodiesterase 4B inhibitors: A review. Sci Pharm. 2014;82:453–81.10.3797/scipharm.1404-08Search in Google Scholar PubMed PubMed Central
[31] Angeletti S, Benvenuto D, Bianchi M, Giovanetti M, Pascarella S, Ciccozzi M. COVID-2019: The role of the nsp2 and nsp3 in its pathogenesis. J Med Virol. 2020;92:584–8.10.1002/jmv.25719Search in Google Scholar PubMed PubMed Central
[32] Naganuma K, Omura A, Maekawara N, Saitoh M, Ohkawa N, Kubota T, et al. Discovery of selective PDE4B inhibitors. Bioorg Med Chem Lett. 2009;19:3174–6.10.1016/j.bmcl.2009.04.121Search in Google Scholar PubMed
[33] Gorja DR, Mukherjee S, Meda CLT, Deora GS, Kumar KL, Jain A, et al. Novel N-indolylmethyl substituted olanzapine derivatives: their design, synthesis and evaluation as PDE4B inhibitors. Org Biomol Chem. 2013;11(13):2075–9.10.1039/c3ob27424aSearch in Google Scholar PubMed
[34] Pachetti M, Marini B, Benedetti F, Giudici F, Mauro E, Storici P, et al. Emerging SARS-CoV-2 mutation hot spots include a novel RNA-dependent-RNA polymerase variant. J Transl Med. 2020;18:179.10.1186/s12967-020-02344-6Search in Google Scholar PubMed PubMed Central
[35] Saleh LA, Boyd A, Aragon IV, Koloteva A, Spadafora D, Mneimneh W, et al. Ablation of PDE4B protects from Pseudomonas aeruginosa-induced acute lung injury in mice by ameliorating the cytostorm and associated hypothermia. FASEB J. 2021;35:e21797.10.1096/fj.202100495RSearch in Google Scholar PubMed PubMed Central
[36] Fox D 3rd, Burgin AB, Gurney ME. Structural basis for the design of selective phosphodiesterase 4B inhibitors. Cell Signal. 2014;26:657–63.10.1016/j.cellsig.2013.12.003Search in Google Scholar PubMed PubMed Central
[37] Singh P, Mishra M, Agarwal S, Sau S, Iyer AK, Kashaw SK. Exploring the role of water molecules in the ligand binding domain of PDE4B and PDE4D: virtual screening based molecular docking of some active scaffolds. Curr Comp-Aid Drug Des. 2019;15(4):334–66.10.2174/1573409914666181105153543Search in Google Scholar PubMed
[38] Pham MT, Yang AJ, Kao MS, Gankhuyag U, Zayabaatar E, Jin SC, et al. Gut probiotic Lactobacillus rhamnosus attenuates PDE4B-mediated interleukin-6 induced by SARS-CoV-2 membrane glycoprotein. J Nutr Biochem. 2021;98:108821.10.1016/j.jnutbio.2021.108821Search in Google Scholar PubMed PubMed Central
[39] Herrmann FE, Hesslinger C, Wollin L, Nickolaus P. BI 1015550 is a PDE4B inhibitor and a clinical drug candidate for the oral treatment of idiopathic pulmonary fibrosis. Front Pharmacol. 2022;13:838449.10.3389/fphar.2022.838449Search in Google Scholar PubMed PubMed Central
[40] El Tabaa MM, El Tabaa MM. New putative insights into neprilysin (NEP)-dependent pharmacotherapeutic role of roflumilast in treating COVID-19. Eur J Pharmacol. 2020;889:173615.10.1016/j.ejphar.2020.173615Search in Google Scholar PubMed PubMed Central
[41] Hirose R, Manabe H, Nonaka H, Yanagawa K, Akuta K, Sato S, et al. Correlation between emetic effect of phosphodiesterase 4 inhibitors and their occupation of the high-affinity rolipram binding site in Suncus murinus brain. Eur J Pharmacol. 2007;573(1–3):93–9.10.1016/j.ejphar.2007.06.045Search in Google Scholar PubMed
[42] Card GL, England BP, Suzuki Y, Fong D, Powell B, Lee B, et al. Structural basis for the activity of drugs that inhibit phosphodiesterases. Structure. 2004;12:2233–47.10.1016/j.str.2004.10.004Search in Google Scholar PubMed
[43] Trott O, Olson AJ. AutoDock Vina: Improving the speed and accuracy of docking with a new scoring function, efficient optimization, and multithreading. J Comput Chem. 2010;31:455–61.10.1002/jcc.21334Search in Google Scholar PubMed PubMed Central
[44] Krüger A, Gonçalves Maltarollo V, Wrenger C, Kronenberger T. ADME profiling in drug discovery and a new path paved on silica. In: Gaitonde V, Karmakar R, Trivedi A, Editors. Drug Discovery and Development—New Advances. Vol. 1. London, UK: IntechOpen; 2019. p. 1–32.10.5772/intechopen.86174Search in Google Scholar
[45] Yang H, Lou C, Sun L, Li J, Cai Y, Wang Z, et al. admetSAR 2.0: Web-service for prediction andoptimization of chemical ADMET properties. Bioinformatics. 2018;35:1067–9.10.1093/bioinformatics/bty707Search in Google Scholar PubMed
[46] Cheng F, Li W, Zhou Y, Shen J, Wu Z, Liu G, et al. AdmetSAR: A comprehensive source and free tool for assessment of chemical ADMET properties. J Chem Inf Model. 2012;52:3099–105.10.1021/ci300367aSearch in Google Scholar PubMed
[47] Pires DEV, Blundell TL, Ascher DB. pkCSM: Predicting small-molecule pharmacokinetic and toxicity properties using graph-based signatures. J Med Chem. 2015;58:4066–72.10.1021/acs.jmedchem.5b00104Search in Google Scholar PubMed PubMed Central
[48] Zhao Y, O’Donnell JM, Zhang H-T. Inhibitor binding to type 4 phosphodiesterase (PDE4) assessed using [3H]Piclamilast and [3H]Rolipram. J Pharmacol Exp Ther. 2003;305:565–72.10.1124/jpet.102.047407Search in Google Scholar
[49] Tenor H, Hedbom E, Häuselmann HJ, Schudt C, Hatzelmann A. Phosphodiesterase isoenzyme families in human osteoarthritis chondrocytes–functional importance of phosphodiesterase 4. Br J Pharmacol. 2002;135(3):609–18.10.1038/sj.bjp.0704480Search in Google Scholar PubMed PubMed Central
[50] Kaur A, Singh TG, Khan H, Kumar M, Singh N, Abdel-Daim MM. Neuroprotective effect of piclamilast-induced post-ischemia pharmacological treatment in mice. Neurochem Res. 2022;47:2230–43.10.1007/s11064-022-03609-wSearch in Google Scholar PubMed
[51] Zhu J, Mix E, Winblad B. The antidepressant and antiinflammatory effects of rolipram in the central nervous system. CNS Drug Rev. 2001;7(4):387–98.10.1111/j.1527-3458.2001.tb00206.xSearch in Google Scholar PubMed PubMed Central
[52] Varona S, Puertas L, Galán M, Orriols M, Cañes L, Aguiló S, et al. Rolipram prevents the formation of abdominal aortic aneurysm (AAA) in mice: PDE4B as a target in AAA. Antioxidants. 2021;10:460.10.3390/antiox10030460Search in Google Scholar PubMed PubMed Central
[53] Dong XL, Wang YH, Xu J, Zhang N. The protective effect of the PDE-4 inhibitor rolipram on intracerebral haemorrhage is associated with the cAMP/AMPK/SIRT1 pathway. Sci Rep. 2021;11(1):1–10.10.1038/s41598-021-98743-wSearch in Google Scholar PubMed PubMed Central
[54] Qiu Y, Yao J, Jia L, Thompson DA, Zacks DN. Shifting the balance of autophagy and proteasome activation reduces proteotoxic cell death: a novel therapeutic approach for restoring photoreceptor homeostasis. Cell Death Dis. 2019;10(8):1–14.10.1038/s41419-019-1780-1Search in Google Scholar PubMed PubMed Central
[55] Wu Y, Li Z, Huang YY, Wu D, Luo HB. Novel phosphodiesterase inhibitors for cognitive improvement in Alzheimer’s Disease: miniperspective. J Med Chem. 2018;61(13):5467–83.10.1021/acs.jmedchem.7b01370Search in Google Scholar PubMed
[56] Lan J, Ge J, Yu J, Shan S, Zhou H, Fan S, et al. Structure of the SARS-CoV-2 spike receptor-binding domain bound to the ACE2 receptor. Nature. 2020;581:215–20.10.1038/s41586-020-2180-5Search in Google Scholar PubMed
[57] Li G, De Clercq E. Therapeutic options for the 2019 novel coronavirus (2019-nCoV). Nat Rev Drug Disc. 2020;19:149–50.10.1038/d41573-020-00016-0Search in Google Scholar PubMed
[58] Sanz R, Escribano J, Pedrosa MR, Aguado R, Arnaiz FJ. Dioxomolybdenum (VI)‐catalyzed reductive cyclization of nitroaromatics. Synthesis of carbazoles and indoles. Adv Synth Catal. 2007;349(4–5):713–8.10.1002/adsc.200600384Search in Google Scholar
[59] Saturnino C, Palladino C, Napoli M, Sinicropi MS, Botta A, Sala M, et al. Synthesis and biological evaluation of new N-alkylcarbazole derivatives as STAT3 inhibitors: preliminary study. Eur J Med Chem. 2013;60:112–9.10.1016/j.ejmech.2012.11.004Search in Google Scholar PubMed
[60] Caruso A, Voisin-Chiret AS, Lancelot JC, Sinicropi MS, Garofalo A, Rault S. Novel and efficient synthesis of 5, 8-dimethyl-9H-carbazol-3-ol via a hydroxydeboronation reaction. Heterocycles. 2007;71(10):2203–10.10.3987/COM-07-11113Search in Google Scholar
[61] Aebly AH, Levy JN, Steger BJ, Quirke JC, Belitsky JM. Expedient synthesis of eumelanin-inspired 5, 6-dihydroxyindole-2-carboxylate ethyl ester derivatives. RSC Adv. 2018;8(50):28323–8.10.1039/C8RA06148CSearch in Google Scholar
[62] Pigza JA, Han JS, Chandra A, Mutnick D, Pink M, Johnston JN. Total synthesis of the lycopodium alkaloid serratezomine A using free radical-mediated vinyl amination to prepare a β-stannyl enamine linchpin. J Org Chem. 2013;78(3):822–43.10.1021/jo302333sSearch in Google Scholar PubMed PubMed Central
[63] Luo Z, Jiang Z, Jiang W, Lin D. C–H amination of purine derivatives via radical oxidative coupling. J Org Chem. 2018;83(7):3710–8.10.1021/acs.joc.8b00066Search in Google Scholar PubMed
[64] Zhang Q, Chang G, Zhang L. Synthesis and properties of novel heat-resistant fluorescent conjugated polymers with bisindolylmaleimide. Chin Chem Lett. 2018;29(3):513–6.10.1016/j.cclet.2017.08.019Search in Google Scholar
[65] Esposito F, Sechi M, Pala N, Sanna A, Koneru PC, Kvaratskhelia M, et al. Discovery of dihydroxyindole-2-carboxylic acid derivatives as dual allosteric HIV-1 Integrase and reverse transcriptase associated ribonuclease H inhibitors. Antivir Res. 2020;174:104671.10.1016/j.antiviral.2019.104671Search in Google Scholar PubMed
© 2023 the author(s), published by De Gruyter
This work is licensed under the Creative Commons Attribution 4.0 International License.
