Screening Libraries | Activity-based Libraries
Activity Based Libraries were designed to support our customers with convenient classification of Life Chemicals Stock Compound Collection by potential application in different fields of drug development. The libraries were designed to maximally fit specified activity areas using combination of validated descriptors and compound properties as well as 2D or 3D similarity searches to the referent set of highly active molecules.
Search for chemotherapeutic cancer treatment is one of the most acute problems of modern pharmacology. To pursue this issue Life Chemicals has prepared the Anticancer Library using similarity search within its proprietary stock collection of HTS compounds against publicly available databases of compounds possessing antitumor activity.
42,000 Compounds screened by National Cancer Institute as part of the DTP Human Tumor Cell Line Screen project were downloaded and merged with the corresponding screening data expressed as GI50.
The NCI renamed the IC50 value, the concentration that causes 50% growth inhibition, into the GI50value to emphasize the correction for the cell count at time zero; thus, GI50 is the concentration of test drugwhere 100 × (T - T0)/(C - T0) = 50. The optical density of the test well after a 48-h period of exposure to test drug is T, the optical density at time zero is T0, and the control optical density is C. The GI50 measures the growth inhibitory power of the test agent.
NCI library was filtered down to 3,151 most potent compounds showing GI50 values below 50 nM.
This set of compounds was tested in 60 human cancer cell lines.
Life Chemicals database was searched for compounds similar to each of the selected 3,151 NCI compounds using the 80% similarity cut-off (Tanimoto) on MDL public keys fingerprints. This effort yielded 28,000 hits that were included into Life Chemicals Anticancer Library.
Results could be traced back to the specific cell lines.
The library has been designed with similarity search that was carried out against reference set of compounds that possess activity in treatment of pathogenic bacteria. The information about the activity has been gathered from the most reliable compound activity databases (ChEMBL, BindingDB) and PubChem (in case when compounds were mentioned as potent pro-drugs in specialized literature). The final library included compounds within Tanimoto index range 0.8 - 0.85 chosen based on activity data and number of compounds in the reference subsets.
The reference set included inhibitors of the following pathogenic bacteria:
● Pseudomonas aeruginosa
● Neisseria meningitides
● Klebiela pneumonia
● Esherichia coli
● Acinetobacter baumannii
The Library was designed with 2D fingerprint similarity search against the reference set of 6,937 biologically active compounds from 33 therapeutically relevant viral assays representing the following species of viruses:
● Avian sarcoma virus
● Bluetongue virus
● Dengue virus
● Human herpesvirus 4 type 2
● Human immunodeficiency virus 1
● Human immunodeficiency virus 2
● Influenza A virus (-Puerto Rico-8-34(H1N1))
● Influenza A virus (A-Puerto Rico-8-34-Mount Sinai(H1N1))
● Influenza A virus (A-Shangdong-9-1993(H3N2))
● Influenza A virus (A-Singapore-1-1957(H2N2))
● Influenza A virus (A-tern-Australia-G70C-1975(H11N9))
● Influenza A virus (A-Tokyo-3-1967(H2N2))
● Influenza B virus (B-Lee-40)
● Influenza B virus (B-Memphis-3-93)
● Influenza B virus (B-Nashville-6-89)
● Influenza B virus (STRAIN B-VICTORIA-3-85)
● Influenza B virus
● Measles virus
● Rous sarcoma virus (strain Schmidt-Ruppin A)
● West Nile virus
Life Chemicals collection was searched for compounds similar to the downloaded dataset using MDL public keys and the Tanimoto similarity cut-off of 90%.
Over 23,100 potential antiviral agents were extracted into Life Chemicals' Antiviral Library.
Designed with receptor-based approach, this library comprises potential drug candidates possessing activity towards InhA enzyme - specific protein of M.tuberculosis that is responsible for cell wall synthesis of the bacteria, but is not present in mammals. The structure of InhA protein and binding mode of known inhibitors were studied based on the analysis of crystal structure records in PDB. This has given us a detailed understanding of the protein-ligand interaction mechanism.
At the next step, Life Chemicals’ in-stock compound collection was processed according to ADME requirements; also, all undesirable groups were filtered out. Resulted drug-like set of compounds was screened by molecular docking using Glide program (Schrödinger software). 3FNH and 2H7I PDB entries were selected for the docking studies due to the most favorable ligand binding and high resolution of the crystal structures. The referent set of active ligands1,2 was used for evaluation of the docking procedure. The presence of the NAD+ coenzyme was taken into account in virtual screening as it is involved into a ligand binding. After the docking, the compounds have been selected by ligand efficacy and comparison with binding mode of the referent inhibitors. A set of 3,700 potential antituberculosis agents capable of binding with InhA protein was obtained (Fig.1).
Fig.1. Ligand F2269-0132 forms strong hydrogen bond with Tyr158 and protein coenzyme NAD+. High hydrophobic interaction observed between fused aromatic rings and large hydrophobic pocket of InhA.
The Library comprises compounds resulted from similarity search against reference sets of compounds displaying activity towards the most dangerous protozoa. The reference sets were taken from the most reliable inhibitor databases (ChEMBL, BindingDB) and PubChem (in case when compounds were mentioned as potent pro-drugs in specialized literature). According to the activity data and number of compounds in the reference subsets the Tanimoto similarity index was set up in a range from 0.8 to 0.85.
List of targeted protozoal families and species:
● Plasmodium genus
● Cryptosporidium family
● Dysentery shigella
● Giardiasis intestinalis
● Amowbiasis hystolitica
● Toxoplasmosis gondii
● Trichomonas vaginalis
● Trypanosoma brucei
Ca 6,100 resulting potentially active compounds formed Life Chemicals’ Antiprotozoa Library.
Life Chemicals CNS library comprises over 22,700 carefully designed and diverse small organic molecules. Compounds for this library have been selected according to the steps listed below. Calculations were done in SYBYL-X and VolSurf+ software:
1. Life chemicals in-stock compound collection has been filtered to meet the following criteria:
MW ≤ 450
ClogP ≤ 3.6
H-Bond Acceptors ≤ 8
H-Bond Donors ≤ 5
TPSA ≤ 120
Rotatable Bonds ≤ 5
logBB -3.0 to 1.0
2. PAIN and ADME filters were applied.
3. CNS specific filters applied:
a) If N + O ≤ 5 then the compound has a high chance to cross blood-brain barrier1.
b) If ClogP – (N+O)>0 then LogBB is likely to be positive1.
c) MW should be kept below 450 to facilitate CNS permeation2.
d) CNS drugs tend to be more rigid (caused by lack of rotable bonds) compared to non CNS drugs3.
According to4, compounds with logBB > 0.3 are readily distributed to the brain, while compounds with logBB < -1.00 are poorly distributed to the brain. The implemented filter removes compounds with logBB <= 0.3. In addition to that it also removes salts and zwitterions. This approach can only predict BBB penetration of compounds in the absence of active transport.
Finally, substructure filters for detection and removal of Pan Assay Interference (PAINS) compounds have been applied.
Hepatitis C Virus
The hepatitis C virus belongs to the genus Hepacivirus a member of the family Flaviviridae and it is known as one of the most dangerous diseases with very expensive treatment. Three proteins of capside were used as target for our search for potential agents against HCV.
A reference database of 3,486 biologically active compounds with confirmed potency in 22 HCV assays was compiled using the data available from patents and literature publications. These compounds show activities in the following 3 targets of the virus:
● Core protein
Life Chemicals collection was searched for compounds similar to compounds from the reference database using MDL public keys and the Tanimoto similarity cut-off of 90%. This effort resulted in a selection of 16,200 compounds that were included in Life Chemicals' HCV Targeted Library.
References for CNS
1. Norinder, U. and Haeberlein, M. (2002). Computational approaches to the prediction of blood brain distribution. Adv. Drug Deliv. Rev. 54, 291-313
2. Van de Waterbeemd, H. et al. (1998). Estimation of blood-brain barrier crossing of drugs using molecular size and shape and H-bonding descriptors. J, Drug Target. 6, 151-165
3. Doan, K.M. et al. (2002). Passive permeability and P-glycoprotein mediated efflux differentiate central nervous system (CNS) and non CNS marketed drugs. J. Pharmacol. Exp. Ther. 303, 1029-1037
4. Abraham MH, Takács-Novák K, Mitchell RC. On the partition of ampholytes: application to blood-brain distribution. J Pharm Sci. 1997 Mar;86(3):310-5.
References for Antituberculosis
1. He X., Alian A., Stroud R., Ortiz de Montellano P. R. J. Med. Chem., 2006, 49, pp. 6308–6323. http://dx.doi.org/10.1021/jm060715y
2. Freundlich J. S., Wang F., Vilcheze C., Gulten, G., Langley R., Schiehser G. A., Jacobus D. P., Jacobs W. R., Sacchettini J. C. Chem. Med. Chem., 2009, 4, pp. 241–248. http://dx.doi.org/10.1002/cmdc.200800261