G-proteins act as switches for signal transduction from extracellular space into the cell. This is accomplished through their interaction with G-Protein Coupled Receptors (GPCRs). G-proteins form hetero-trimers composed of Gα, Gβ and Gγ subunits, and they also possess a binding site for a nucleotide (GTP or GDP). G-proteins are named after their α-subunits, which on the basis of their amino acid similarity and function are grouped into four families (Gαs, Gαi/o, Gαq, Gα12).

GPCRs form the major group of receptors in eukaryotes and they possess seven transmembrane α-helical domains. GPCRs are usually classified into several classes, according to the sequence similarity shared by the members of each class.Class A of GPCRs (rhodopsin-like GPCRs) contains the majority of GPCRs, including receptors for structurally diverse ligands (biogenic amines, nucleotides, peptides, glycoprotein hormones etc). Class B (secretin-like GPCRs) contains purely peptide receptors, whereas class C (metabotropic glutamate family receptors) contains metabotropic glutamate and GABA-B receptors and some taste receptors. Class D contains the fungal pheromone receptors, class E contains the cAMP receptors of Dictyostelium and last is the Frizzled/Smoothened class. There is also a number of putative classes of newly discovered GPCRs, whose nomenclature has not been accepted yet from the scientific community.

The stimulation of GPCRs leads to the activation of G-proteins, which dissociate into Galpha and Gbeta-gamma subunits. The subunits then activate several effector molecules that lead to many kinds of cellular and physiological responses.

Effectors form a diverse group of proteins, that, throught their interaction with G-proteins, either act as second messengers, or lead directly to a cellular and physiological response. Effectors have never being classified before. We classified them into families, subfamilies and types, based on their function.

Data annotation

The annotation regarding the interaction between GPCRs, G-proteins and effectors and the effect of the particular interaction was a result of an exhaustive and detailed literature search. We collected the available information from review articles and original research papers, which we provide as links in each entry page. A point that it was impossible to be explained in the manuscript (and thus is discussed in the online manual pages) is the fact that no entries are included in the database solely using a prediction system. On the contrary, interactions are inferred from orthologues. In particular, when we have a particular reference stating that protein X interacts with protein Y in organism Z, we search all the other closely related organisms for such pairs (X-Y). This search is not being performed on an automated fashion (i.e. a simple BLAST search) but instead we rely on family classification (from PFAM), the gene name, the function of the proteins etc.

For instance from the reference that used fused chimeric mutants of bovine ACI:

“Wittpoth C, Scholich K, Yigzaw Y, Stringfield TM, Patel TB. Regions on adenylyl cyclase that are necessary for inhibition of activity by beta gamma and G(ialpha) subunits of heterotrimeric G proteins. Proc Natl Acad Sci U S A. 1999;96(17):9551-6.”

we conclude that Gbeta-gamma dimer inhibits Adenylyl Cyclase I, and thus this information could be transferred to all the available (mostly mammalian) organisms possessing Gbeta-gamma and ACI.

From the paper describing another heterologous expression system:

“Marty C, Browning DD, Ye RD. Identification of tetratricopeptide repeat 1 as an adaptor protein that interacts with heterotrimeric G proteins and the small GTPase Ras. Mol Cell Biol. 2003;23(11):3847-58”

we conclude that Galpha-16 interacts with TRP1 and this information could be expanded to all organisms possessing Galpha-16 and TRP1. And similarly we proceed with the other interactions.

From the paper describing the selectivity of AT2 receptor in the Rat fetus:

“Zhang J, Pratt RE. The AT2 receptor selectively associates with Gialpha2 and Gialpha3 in the rat fetus. J Biol Chem. 1996 Jun 21;271(25):15026-33“

we conclude that AT2 receptor interacts with Galpha-I and this information could be transferred to all organisms possessing AT2 receptor and Galpha-i.

From the paper describing another heterologous expression system:

“Borowsky B, Adham N, Jones KA, Raddatz R, Artymyshyn R, Ogozalek KL, Durkin MM, Lakhlani PP, Bonini JA, Pathirana S, Boyle N, Pu X, Kouranova E, Lichtblau H, Ochoa FY, Branchek TA, Gerald C. Trace amines: identification of a family of mammalian G protein-coupled receptors. Proc Natl Acad Sci U S A. 2001 Jul 31;98(16):8966-71”

we conclude that TA1 receptor couples with Galpha-s and this information could be expanded to all organisms possessing TA1 and Galpha-s.

Of course, there are other more “simple” and straightforward situations such as the interactions of Galpha-s subunits that are known for years to stimulate adenylate cyclases, and so on.

Scope of the gpDB

GpDB is a publicly accessible, relational database, containing information about G-proteins, GPCRs and Effectors. It contains detailed information for 410 G-proteins (257 G-alpha, 86 G-beta and 67 G-gamma), 2795 GPCRs belonging to families with known coupling to G-proteins, and 1510 Effectors, that interact with specific G-proteins. The sequences are classified according to a hierarchy of different classes, families and sub-families, based on literature search. Particularly, effectors are classified into families, subfamilies and types. The main innovation besides the classification of G-proteins, GPCRs and effectors is the relational model of the database, describing the known coupling specificity of the GPCRs to their respective alpha subunit of G-proteins and also the specific interaction between the different subfamilies of G-proteins and particular effector types, a unique feature not available in any other database. There is full sequence information with cross-references to publicly available databases, and the user may submit advanced queries for text search. Furthermore there is interconnectivity with PRED-GPCR, PRED-TMR, TMRPres2D, a pattern search tool, and an interface for running BLAST against the database. The database will be very useful for the study of G-protein/GPCR and G-protein/Effectors interactions, and for future development of algorithms predicting this interaction. It can be accessed via a web-based browser at the URL: http://bioinformatics.biol.uoa.gr/gpDB

Navigation

Through the navigation tool, the user has the ability to browse the database following the hierarchy. The navigation could be performed on the GPCRs, the G-PROTEINS or the EFFECTORS hierarchy. Following the link of GPCRs, the user may be navigated through:

GPCR CLASSES

Top of the GPCR classes page

GPCR FAMILIES

We have classified GPCRs into 64 different families

Top of the GPCR families page

GPCR SUB-FAMILIES

Each family is further subdivided into different subfamilies, based mainly on TIPS classification scheme that takes into account the native ligand(s) that binds to a particular GPCR .

The GPCR SUBFAMILIES MENU enables the user to either view the individual receptors of the specific subfamily or to view the coupling specificity of the GPCR subfamily with G-protein subfamilies.

Viewing the receptors of the specific subfamily

By clicking on a specific subfamily the user is presented with a list of all individual receptors belonging to this subfamily

EXAMPLE

The user is able to click on a specific subfamily like 5-HT

Subfamily of the 5-HYDROXYTRYPTAMINE RECEPTOR

Subfamilies of 5-Hydroxytryptamine receptor family

The result page presents all the individual receptors of 5-HT subfamily

Receptors of 5-HT subfamily

Coupling between GPCRs - G-protein subfamilies

The user has the potential to see the coupling specificity of a GPCR subfamily with G-proteins subfamilies

EXAMPLE

As it is shown in the picture of the GPCR subfamilies the user has the alternative to click on the arrow button instead of selecting to click on a specific subfamily.

The user then is presented with a list of G-protein subfamilies that couple to the specific GPCR subfamily. The G-protein types of these subfamilies have known coupling specificity to the receptors of this specific GPCR subfamily.

The result page

Following the link of G-PROTEINS, the user may browse through:

G-PROTEIN CLASSES

G-PROTEIN FAMILIES

Families of Galpha class

G-PROTEIN SUB-FAMILIES

The G-PROTEIN SUBFAMILIES MENU enables the user to either view the protein types of the specific subfamily or to view the coupling specificity of the G-protein subfamily with GPCR subfamilies.

Viewing the types of the specific subfamily

By clicking on a specific subfamily the user is presented with a list of all G-protein types belonging to this subfamily

EXAMPLE

The user is able to click on a specific subfamily like Galpha-12/13 subfamily of Gi/o family

Subfamilies of G12/13 family

The result page presents all the G-protein types of Galpha-12 subfamily

G-PROTEIN TYPES

Types of Galpha-12 subfamily

Then by clicking on any type the user is presented with all individual G-proteins

Ending up to individual G-proteins.

Proteins of Galpha-12 type

Coupling between GPCRs -G-proteins subfamilies

The user has the potential to see the coupling specificity of a G-protein subfamily with GPCR subfamilies

EXAMPLE

As it is shown in the picture of the G-protein subfamilies the user has the alternative to click on one of the two arrow buttons instead of selecting to click on a specific subfamily.

Subfamilies of Gi/o family

The red button presents the user a list of GPCR subfamilies that couple to the specific G-protein subfamily. The GPCRs of these subfamilies have known coupling specificity to the receptors of this specific G-protein subfamily

The result page

The yellow button presents the user a list of effector types with whom this G-protein subfamily interacts.

The result page

Following the link of Effectors, the user may browse through:

EFFECTOR FAMILIES

Effector Families

EFFECTOR SUB-FAMILIES

Subfamilies of ion channels family

EFFECTOR TYPES

The Effector TYPES MENU enables the user to either view the entries of this protein type or to view the G-protein subfamilies, that interact with this type.

Viewing the entries of the specific type

By clicking on a specific type the user is presented with a list of all Effector types belonging to this subfamily

EXAMPLE

The user is able to click on a specific type like ATP-sensitive inward rectifier potassium channel-1 type of the ATP-sensitive inward rectifier potassium channel subfamily

Types of ATP-sensitive inward rectifier potassium channels subfamily

By clicking on a type the user is presented with all individual Effectors

Ending up to individual Effectors.

Proteins of ATP-sensitive inward rectifier potassium channel-1 type

Interaction between G-proteins subfamilies and Effector types

The user has the potential to see the interaction between G-proteins subfamilies and Effector types

EXAMPLE

As it is shown in the picture of the Effector types the user has the alternative to click on the arrow button instead of selecting to click on a specific type.

Types of ATP-sensitive inward rectifier potassium channels subfamily

The user then is presented with a list of G-protein subfamilies that interact with this particular Effector type.

The result page

At each point the user may navigate up or down to the hierarchy tree.

Figure 1. The relational model of the database

Text Search

In the Text Search area, the user can search for any text in the fields of his/her preference. The user can enter any word in one or more of the available boxes under the name: 'Protein Name', 'Species', ’Common Name’, 'Description', 'Gene Name' and 'Cross-References'. The user has also the ability to select if he wants to exclude fragments from the results.

Each expression may contain:

i) Text terms to be searched for,

ii) Parenthesis '(' ')' which groups one or more sub-expressions,

iii) Operator '&' for AND, which combines two (or more) sub-expressions in a single field and gives the user the opportunity to search for entries that satisfy all sub-expressions.

iv) Operator '|' for OR, which combines two (or more) sub-expressions in a single field and gives to the user the opportunity to search for entries that satisfy at least one of the sub-expressions.

v) Operator '!' for NOT, which can only be used at the beginning of an expression and does not connect two sub-expressions. It provides to the user the opportunity to search for entries that necessarily do not satisfy the expression after the '!'.

vi) Operator '&!' for AND NOT, which combines two (or more) sub-expressions in a single search field and gives to the user the opportunity to search for entries that satisfy only the sub-expression in the left of the operator.

Expressions in separate search fields are combined with the AND operator, so every entry of the result set will satisfy the expressions of all the search fields the user has chosen. The user has the option to choose whether the query will be performed against the GPCRs or the G-Proteins included in the database.

PROTEIN NAME

Corresponds to the field PROTEIN NAME of an entry

EXAMPLE

If the user wants to retrieve all Galpha proteins he/she has to use the name “Galpha” as a query in the PROTEIN NAME box and select G-protein from Search Target field.

The top of the result page is:

The user has to use the name “Gbeta” and “Ggamma” in order to search for Gbeta and Ggamma proteins additionally

SPECIES

Corresponds to the field SPECIES of an entry (the scientific name of a species)

EXAMPLE

1) If the user wants to retrieve all G-proteins of Drosophila he/she has to use the name “Drosophila” as a query in the SPECIES box and select G-protein from Search Target field.

The result page is:

2) If the user wants to retrieve all GPCRs of Drosophila he/she has to use the name “Drosophila” as a query in the SPECIES box and select GPCR from Search Target field.

The top of the result page is:

COMMON NAME

Corresponds to the field COMMON NAME of an entry (the common name of a species)

EXAMPLE

3) If the user wants to retrieve all G-proteins of Drosophila melanogaster he/she has the alternative to use the name “fruit fly” (The common name of Drosophila melanogaster) as a query in the COMMON NAME box and select G-protein from Search Target field.

The result page is:

REFERENCES

Corresponds to the field DESCRIPTION of an entry

The user is able to use any accession numbers and/or IDs from other databases such as SWISS_PROT, PIR, MIM, PRODOM, GENEW, PRINTS, INTERPRO etc.

EXAMPLE

If the user wants to retrieve a G-protein that has “P29348” as an accession number in SWISS_PROT he has to use “P29348” as a query and select G-protein from Search Target field.

The top of the result page is:

DESCRIPTION

Corresponds to the field DESCRIPTION of an entry

EXAMPLE

The top of the result page is:

GENE NAME

Corresponds to the field GENE of an entry

EXAMPLE

The result page is:

BLAST Search

With the BLAST search tool, the user may submit a sequence and search the database for finding homologues. The user has the option to choose whether to perform the BLAST search against GPCRs sequences, G-proteins sequences and/or Effectors sequences. The input for the BLAST application is the sequence in standard FASTA format.

Submitting a sequence

The output of the BLAST query consists of a list of sequences in the database having significant E-values in a local pairwise alignment, ranked by statistical significance. In the output, are also listed the range of residues in which the alignment occurs, in both the target and the query sequence, the number of identical and similar residues in the alignment and the E-value of the alignment.

By clicking the NAME button from each hit, the user may visualize the local alignment

The top of the result page

And from there, the user may retrieve the detailed view of the entry corresponding to the particular target sequence.

The entry of the particular target sequence

Pattern Search

Using the Pattern Search tool, the user may perform searches for finding specific patterns in the proteins of the database. The user has the option to choose whether to perform the Pattern search against the GPCRs sequences or the G-Proteins. The input of the Pattern Search tool is a regular expression pattern following the PROSITE syntax.

The output of the Pattern search application consists of a list of the sequences matching the particular pattern. gpDB ID(s) and the NAME of the target sequence(s) are listed in the output. The user has the option to check the entry or the entries that he/she wants to retrieve, and after pressing the appropriate button, to have them in the detailed view.

Detailed View of a Protein

The detailed view of an entry corresponds to the last level of the hierarchy. In the detailed view, the available information regarding a GPCR, a G-protein or an Effector sequence is presented.

The fields of the detailed view are the following :

Classification of the particular protein according to the hierarchy

Gpdb Name

Gpdb ID

The gene name

The scientific name of the species that the protein was found

The common name of the species that the protein was found

The sequence

A short description

Available links with cross-references to the publicly available databases (UNIPROT, PIR, PDB, InterPro, PRINTS, PFAM, MIM, NCBI Gene etc)

INFORMATION ABOUT DIMERIZATION (only GPCRs)

Scaffold proteins that couple to the particular GPCR subfamily

Finally and more important, there is a list with the available information regarding the coupling specificity between GPCRs and G-proteins and/or the interaction between G-proteins and Effectors.

Additionally GPCR entries have links to PRED-GPCR, PRED-COUPLE2 and HMM-TM

PRED-GPCR, PRED-COUPLE2 and HMM-TM are tools that were developed in our laboratory.

PRED-GPCR is a system based on a probabilistic method that uses family specific profile HMMs in order to determine to which GPCR family a query sequence belongs or resembles.

PRED-COUPLE2 is a system based on a refined library of highly-discriminative Hidden Markov Models in order to predict the coupling specificty of GPCRs to all families of G-proteins (including G12/13). Hits from individual profiles are combined by a feed-forward Artificial Neural Network to produce the final output.

HMM-TM is an algorithm for the prediction of the topology of transmembrane proteins using HMMs.

By clicking on the representation button the user gains access to another tool of our laboratory TMRPres2D

The 'TransMembrane protein Re-Presentation in 2 Dimensions' tool, automates the creation of uniform, two-dimensional, high analysis graphical images/models of alpha-helical or beta-barrel transmembrane proteins

The gpDB accession numbers and the names of the proteins with which that protein couples, are listed with the appropriate links. By clicking on any of these links, the user will be presented with the detailed view of the corresponding protein.

The detailed view, of GPCRs, G-Proteins and Effectors are completely analogous, with the only difference being the fact that the coupling relationship is of the type “many-to-many”. This means, that a particular G-protein, may couple to more than one receptor of the same organism (which is usually the case) but that particular GPCR may also couple to other G-proteins of the same organism (promiscuous coupling). This also happens between G-Proteins and Effectors. Especially for GPCRs, the user has also the option to submit their sequences to the PRED-GPCR server and retrieve prediction regarding the classification of the receptor

EXAMPLE

Complete entry of A1 Adenosine receptor of Homo sapiens.