• Frequently Asked Questions





    1.


    1.1

    SuperPain is a database for pain stimulating and pain relieving compounds, which bind or potentially bind to ion channels, such as TRPV1, TRPM8, TRPA1, TREK1, hERG or voltage-gated sodium channels.

    1.2

    The database consists of about 8,000 ligands, which are characterized by experimentally measured binding affinities. Additionally, 100,000 putative ligands are included. Moreover, the database provides a 3D structure of TRPV1 and predicted ligand-binding poses. These binding poses and a structural classification scheme provide hints for the design of new analgesic compounds. A user-friendly graphical interface allows similarity searching, cluster views, visualization of ligands docked into the receptor etc.
    top of page


    2.


    2.

    SuperPain is stored in a relational MySQLdatabase. To handle the chemical information within the database, the MyChem package is used. MyChem relies for most of its functions on the Open Babel toolbox. The website is built using PHP; web access is enabled via Apache HTTP Server. We recommend a recent verion of Mozilla Firefox; Google Chrome, Microsoft Internet Explorer and Apple Safari were tested with some configurations. JavaScript has to be enabled.
    top of page


    3.


    3.1

    Pain is an unpleasant sensory and emotional experience associated with actual or potential tissue damage, or described in terms of such damage.
    Pain is more than an unpleasant sensory experience associated with actual or potential tissue damage, it is the most common reason for physician consultation and often dramatically affects the quality of life. The management of pain is often difficult and new targets are required for a more effective and specific treatment.

    3.2

    This excerpt of the declaration of Montréal underlines the importance of research on pain management.
    The delegates to the International Pain Summit of the Internationsl Association for the Study of Pain have given in-depth attention to the unrelieved pain in the world, finding that pain management is inadequate in most of the world because
    • There is inadequate access to treatment for acute pain caused by trauma, disease, and terminal illness and failure to recognize that chronic pain is a serious chronic health problem requiring access to management akin to other chronic diseases such as diabetes or chronic heart disease
    • There are major deficits in knowledge of health care professionals regarding the mechanisms and management of pain.
    • Chronic pain with or without diagnosis is highly stigmatized.
    • Most countries have no national policy at all or very inadequate policies regarding the management of pain as a health problem, including an inadequate level of research and education.
    • Pain Medicine is not recognized as a distinct specialty with a unique body of knowledge and defined scope of practice founded on research and comprehensive training programs.
    • The World Health Organization (WHO) estimates that 5 billion people live in countries with low or no access to controlled medicines and have no or insufficient access to treatment for moderate to severe pain.
    • There are severe restrictions on the availability of opioids and other essential medications, critical to the management of pain.


    3.3.1

    Nociceptors are receptors detecting noxious stimuli. These stimuli are transformed into electrical signals, which are conducted to the central nervous system. Mediators of inflammation, such as prostaglandins, bradykinin or H+ can stimulate these nociceptors, but there are also free nerve endings that can be stimulated by thermal, chemical or mechanical influences.


    3.3.2

    Nociceptive pain arises from damage to non-neural tissues and is activated through nociceptors. There are two types of nociceptive pain: somatic pain (from joints, bones, muscles - such as headaches or arthritis) and visceral pain (from internal organs – such as endometriosis or prostate pain)


    3.3.3

    Neuropathic pain is caused by a lesion or disease of the somatosensory nervous system. Possible reasons are nerve irritation or damage. The most common example for neuropathic pain is the peripheral neuropathy (also called diabetic neuropathy).


    3.3.4

    Psychogenic pain is caused by psychological disorder, such as depression or anxiety. Usually, psychogenic pain has no physical origin, so it is difficult to treat patients suffering from this sort of pain.


    3.3.5

    Idiopathic pain has no physical or psychological cause. This sort of pain often appears in patients with pre-existing pain disorders, such as TMJ disorders or fibromyalgia. It is often associated with depression, but may involve cerebral and peripheral physiological mechanisms. Pain treatment has to be multifactorial.


    3.4

    Nerve cells transmit pain signals to the central nervous system through ion channels. Ion channels are proteins forming a pore that allows the flow of ions across membranes. Ion channels are voltage- or second messenger-gated. Therefore, these channels are promising targets for the development of pain therapeutics.

    top of page


    4.


    4.1

    The database consists of about 8,000 experimentally determined 8,000 ligands and 100,000 putative ligands for ion channels that are involved in the pain pathway. The database provides a 3D homology model of TRPV1 and predicted ligand-binding poses. These binding poses COULD BE USED AS hints for the design of new analgesic compounds. A user-friendly graphical interface allows similarity searching, cluster views, ATC browsing, visualization of ligands docked into the receptor etc.

    4.2

    The SuperPain Database is a resource on ligands binding to ion channels that are associated with the transmission of pain. The database shelters 100,000 putative ligands. There are different options to search and find promising compounds.

    4.2.1

    In the navigation bar, click Compounds.

    4.2.1.1


    If you are looking for a particular compound, the "Compound Properties" search box will help you to find the compound you are looking for. Type in the name (or some letters of the name), PubChem ID, IUPAC or the SMILES code of the compound you are looking for. You can also search ligands for a specific target. We provide ligands for TRPV1, TRPM8, TRPA1, hERG, TREK and sodium channels.

    4.2.1.2


    If you are interested in several compounds with similar features, the "Feature selection" search box will help you to find compounds of interest. Features are purchasability of experimentally determined ligands, purchasability (all) of putative ligands, IC50, EC50, molweight etc.

    Examples:
    • Purchasability - shows vendors of experimentally determined ligands: yes / no
    • Purchasability (all) - shows vendors of known ligands and putative ligands: yes / no

    • For the other features, you can type in a value with ' < / > 10' or a range '0 - 10' or a specific value (10).

    • Kd (nM) Receptor Affinity - dissociation constant: 0 - 14,400
    • IC50 (nM) - concentration that is needed to reduce enzyme activity by 50%: 0 - 56,000,000
    • EC50/IC50 - concentration of a compound where 50% of the population exhibit a response: 0 - 2,500,000
    • Molweight - molecular weight of the molecule: 56 - 1,884
    • Rotatable Bonds - number of rotatable bonds in the molecule: 0 - 26
    • Heavy Atoms - isotopic form of atoms that contain more than the common number of neutrons: 4 - 126
    • H-Bond Donor - number of hydrogen bond donors: 0 - 31
    • H-Bond Acceptor - number of hydrogen bond acceptors: 0 - 33


    4.2.1.3


    If you are interested in drugs targeting ion channels, the pain-related ATC tree could be helpful. The ATC tree was introduced by the WHO and lists drugs according to their Anatomical, Therapeutic and Chemical properties in different groups. Click on the category you are interested in (e.g. N - Nervous System) and choose between Anesthetics, Antiepileptics, Psycholeptics etc., that are used in the treatment of (neuropathic) pain.

    4.2.2

    Once you have found your compounds of interest, you might be interested in finding similar structures, that are putative ligands for your desired target. On every ‘Results’ page you are enabled to perform a similarity search. You can see experimentally determined (co-)affinities, synonymous names, chemical information, a vendor list and a link to a cluster. Click on ‘Show similar structures’. A page with similar compounds (Tanimoto coefficient) will open.


    4.2.3

    Another option to find similar structures is the Cluster view. We created structural fingerprints of all ligands and performed a similarity scan with the help of the Tanimoto coefficient. The clustering resulted in 684 clusters that are displayed on Cluster Page . You can browse through the different clusters by clicking on the chemical structures or you can even search for specific compounds. You also have the option to jump to the clusters on each "Results" page.
    The displayed heat-maps are diagonally divided and compare similarity (Tanimoto) and affinity (IC50) of the structures with each other. Mouseover the heat-map shows the compared structures. If you click on a square, a new page opens and the two structures are compared with each other in detail. Additionally, a similarity search for putative compounds can be performed.

    4.3

    Currently, there is no experimental data (X-ray, NMR) available on the atomic structure of the different ion channels. We created a homology model of TRPV1 that might help to better understand binding mechanisms. Click on Receptor. If Java is installed, you can easily navigate through a homology model of the TRPV1 receptor. Additionally we provide information on the other channels.

    4.4

    On the Compounds "results" page, click on "Show docking results". A Javascript window opens.

    You can choose between 5 docking poses by clicking on one of them. The ligand can be displayed in different manners, as well as the receptor itself. Press the left mouse button while rotating. A right click leads you to a drop-down menu, for zooming, measurements, color adjustments etc.
    For Safari and IPad users: use your common gestures to zoom and rotate the molecule.

    4.5

    The extrarenal fraction (Q0) value is able to predict whether a drug is primarily excreted unchanged via kidneys or metabolized and/or removed through another pathway. Thereby is (1- Q0) the fraction which is removed unchanged via kidneys. High Q0 values mean mainly metabolized drug and kidney independent excretion. The elimination half-life (EHL) of a drug is the time in which it looses one half of its activity. The longer the half-life, the longer it will remain in the human body. The Cocktail tool displays the Q0 and elimination half-life (EHL) values to compare the pharmacological properties of drugs and their alternatives. Thereby, extrarenal fraction (Q0) value is able to predict whether a drug is primarily excreted unchanged via kidneys or metabolized and/or removed through another pathway. Thereby is (1- Q0) the fraction, which is removed unchanged via kidneys. High Q0 values stand for mainly metabolized drugs and/or kidney independent excretion. In order to prevent adverse side effects and toxic drug levels in diseased kidney patients the Q0 value should be taken into account to change the drug or adjust the dosage. The Q0 could also help to estimate the extent of CYP-drug interactions. Drugs with low Q0 values (<0.3) are excreted unchanged to a large extent and occupying the CYP system lesser. Their impact on interactions is lower than for drugs with higher Q0 values. Hence, consideration of Q0 values in finding alternative drugs is useful to reduce the interaction potential, if the function of kidneys is sufficient. However, high values do not necessarily mean more CYP reactions. Nevertheless, it provides a useful support to select the alternative drugs.
    top of page


    5.


    5.1

    Multiple approaches in order to find compounds, which interact on ion channels, were made. One of them was textmining all abstracts of publications in the PubMed database for drugs that have been published to be ligands of TRPV1, TRPM8, TRPA1, hERG, KCNK2, TREK1, P2X, hERG, sodium channel or ASIC.

    Therefore we first downloaded Medline/PubMed data from the NCBI FTP site in xml-format. Using the search engine library Apache Lucene (http://lucene.apache.org) and a tool kit for processing text with computational linguistics (http://alias-i.com/lingpipe) the data was indexed. Lists of synonyms for each channel and drugs, as well as a list of common affinity terms and units were created to find any available ligand and affinity values. The search engine, written in Java, dynamically queries the indexed data and results in a structured query language (SQL) file containing the textmining hits. An example for a query for the literature search is:

    (lidocaine [TI] AND ic50 [TI]) OR (lidocaine [abstract] AND ic50 [abstract]).

    Found hits were put in a preliminary database and are displayed with keywords highlighted in different colors. In the manual evaluation process, confirmed experimentally determined affinities were moved to the final database.
    Further information regarding the molecules was retrieved from the PubChem database. PubChem is a freely available database, which is provided by the National Center for Biotechnology Information (NCBI). It contains detailed information on about 30 million compounds. All those compounds and additional information were put in a MySQL database.
    Additional affinities were found via a search of BindingDB. BindingDB is a database on experimentally determined protein-ligand interactions. It provides about 1,051,955 binding data for 7,117 protein targets and 440,396 small molecules. We found about 8,700 ligands.

    5.2

    The experimentally determined ligands had to be clustered regarding their similarity to each other by means of a K-means algorithm. As in the K-means algorithm, the number of clusters has to be defined beforehand the algorithm was slightly modified. To ensure that the most similar compounds are members of one cluster, a neighbor-joining algorithm was used. The R package with heatmap.2 (gplots) was used to display the compound similarities in a heat map.

    A structural clustering of the 8,700 compounds with an internal similarity (Tanimoto) above 0.7 was performed and resulted in 681 clusters with member sizes up to 52 compounds. For a better visualization, the member size was limited to 30. The neighbor-joining algorithm resulted in 684 clusters. The similarities are displayed in interactive heat maps. These heat-maps compare the compounds with each other regarding similarity (Tanimoto coefficient) and affinity (IC50).

    5.3

    To find potential ligands for the channels, a similarity search was performed. We used the clustered ligands, which have experimentally determined binding affinities and the PubChem database. PubChem is a freely available database, which is provided by the National Center for Biotechnology Information (NCBI). It contains detailed information on about 30 million compounds.
    To estimate the degree of structural similarity between two molecules and to efficiently search in a large chemical database like PubChem, it has been found to be effective to calculate structural fingerprints. Fingerprints encode the presence of 2D substructural fragments in a molecule, and the similarity between a pair of molecules is a function of the number of fragments that they have in common. With the help of the calculated fingerprints we searched for similar structures on PubChem.
    The similarity of an experimentally determined ligand A to a potential ligand B was calculated with the help of the Tanimoto coefficient (T):

    A, B number of 1-bits within the fingerprints
    C number of 1-bits in A and B at the same place

    Depending on the chemical topological properties it gives values between zero (no similarity) and one (identical).
    Compounds with a coefficient > 0.90 were classified as putative ligands and included in the database.

    top of page


    6.


    6.

    Currently, no x-ray crystallographic structures are available for most of the receptors, but there is some information on the active sites of some of the molecules.

    TRPV1
    The general approach was made as described by Fernández-Ballester and Ferrer-Montiel (Molecular Models of TRPV1, 2008). As a template, the high-resolution transmembrane domain of the Kv1.2 rat potassium channel (2.4 Å) was downloaded (PDB code 2R9R), as well as the amino acid sequence in FASTA format. The alignment of the human TRPV1 channel and the rat potassium channel was refined with tools for the prediction of transmembrane domains (Accelrys Discovery Studio). With the help of PyMOL the subunits were assembled to a tetramer to form the channel pore. The homology modeling was focused on the transmembrane domain with its central pore, as well as the binding sites for agonists and antagonists. Therefore, the sequence was shortened at the C- and N-terminals from 839 to 289 residues. The homology model was obtained by using the homology-modeling server SWISS-MODEL in alignment mode.
    Afterwards, with the help of PyMOL the subunits were assembled to a tetramer to form the channel pore. For the fitting of the TM helices, gaps were removed from these regions and shifted into loop regions.
    The energy minimization was calculated in 'Accelrys Discovery Studio' with the help of CHARMm-force field and 100 cycles of each steepest descent and conjugate gradient algorithms until RMS was <0.05 kcal/mol*Å.

    P2X, ASIC
    There are some crystal structures for P2X (4DW0, 4DW1) and ASIC (PDB: 4FZ0, 4FZ1) available on PDB.

    hERG
    The homology model of the hERG channel was retrieved from a research group in Germany (Stary et.al).
    top of page


    7.


    7.

    For the preparation of the in silico docking pipeline the crystal structures of ASIC and P2X and the homology model of TRPV1 was imported into Accelrys Discovery Studio. All together about 6,000 ligands (for TRPV1, ASIC, P2X and hERG) were docked into the binding sites by using the integrated Docking Tool LibDock. LibDock is a high-throughput docking algorithm. Based on polar and apolar interaction sites up to 1,000 ligand conformations were positioned into the binding site.


    The figure shows the docking parameters of the LibDock protocol.
    The conformation search was performed using the FAST method of Catalyst which allows the conformation generation within a user-defined energy threshold. 100 polar and apolar hotspots were defined using a grid of the receptors' binding sites, to which conformers were docked. A RMSD tolerance of 0.25Å avoided the calculation of similar poses. After a BFGS pose optimization, the top 10 poses with a LibDock Score of at least 100 were saved. These steps were performed without hydrogen atoms.

    The five best-ranked poses regarding energetic conditions are displayed on the website. 'JSmol' was used to implement a molecular viewer, which is JavaScript-based.
    top of page