The KNOTTIN database

News

• December 2011 :
  Database update:
   - 34 families, 167 structures, 2193 sequences.
  
  
• November 2010 :
  Database update:
   - 33 families, 161 structures, 1994 sequences.
  
  
• January 2010 :
  3D models for all knottins are available:
   - All knottins for which the 3D structure has not yet been determined were automatically modeled using a specific protocol ( Gracy and Chiche BMC Bioinformatics 2010, 11:535).
   - 3D models can be downloaded:
  
  • - Use "Browse Database > Sequences & 3D models" in the left menu
  •     or use "Search Database > Information Search" then the "Sequence database" button (top section)
  • - Click the "3D model" links in the "PDB" column of the resulting table to download the models
   - 3D models can be viewed or superimposed (Jmol):
  
  • - Use "Browse Database > Sequences & 3D models" in the left menu
  •     or use "Search Database > Information Search" then the "Sequence database" button (top section)
  • - To view only one structure, use the "Jmol" links in the Nomenclature column
  • - To view superimposed structures, use (a) the "3D Structural superimposition" button in the light blue form and (b) the 1st column checkboxes to select which PDB structure(s) or 3D model(s) will be displayed.
   - 3D models can be built for any knottin submitted sequence:
  
  • - Use the Knoter1d3d tool (Category = Tertiary structure modeling) of our Protein Analysis Toolkit, PAT webserver.
  • - Note that a valid MODELLER license key is needed.
  • - The maximum number of structural templates (1-20), and the number of models generated for each template combination (1-5) can be modified.
  • - The CPU time is approximately few minutes to 1 hour depending on the parameters and the server load.
  
  
• January 2010 :
  Menu modification:
   - For sake of clarity the left main menu has been simplified and a new cascading sub-menu has been added
  
  • - The "Browse Database" entry provides pre-compiled tables or sequence alignments
  • - The "Search database" entry allows users to query the database
  • - The "Tools" entry allows users to run prediction and analyse tools
  • - The "Bibliography" entry provides short reviews on various knottin topics
  
  
• September 2009 :
  Database update:
   - 30 families, 155 structures, 1621 sequences.
  Knottin classification:
   - Sequences are classified into families according to Cysteine IV position, origin and/or function.
  • - 3 new potential families have been added (Algae, Bacteria, Trichoplax)
  • - 2 previous families, Scorpion3 and Terebra, were merged with Scorpion2 and Conotoxin1, respectively.
  • - 2 previous families, Insect antimicrobial and Phenoloxidase inhibitor, were renamed as Insect1 and Insect2 respectively.
  
  
• June 2008 :
  Database update:
   - 28 families, 148 structures, 1246 sequences.
   - Since they are no more supported by the PDB, theoretical models have been removed.
  Cyclic knottins:
   - In the standardized alignments, cyclic knottins are now truncated to the postulated mature peptides.
  
  • This modification is not applied to fragments or linear analogs.
    
  • NB1 Standardized alignments are accessible via: 'Search Database > Selection > Get standardized alignments'.
    
  • NB2 Both the full sequence and the 2D representation (Collier de Perles) of the putative cyclic mature sequence are available on the individual knottin cards via: 'Search Database (Sequence database) > Green buttons '.
    
  
  
• October 2007 :
  Structures can be viewed with the Jmol applet:
   - Each structure can be displayed using the Jmol link in the 'Structure' tabular outputs.
   - Structural superimpositions are available:
     1 - Select structures from tabular outputs using checkboxes or the 'Select All' button
     2 - choose the "Structural superimposition" radio button
     3 - click on the "GET" button
   - Knotted cysteines are labeled, and loops are colored as follows:
     (a)=green; b=coral; c=aquamarine; (d)=lightblue; e=hotpink
  
  
• September 2007 :
  The Tool menu has been updated:
   - Knoter3d determines if a 3D structure is a knottin.
   - Knoter1d predicts if a protein sequence is a knottin.
   - Standardized alignments and "Colliers de Perles" are available.
  Alignments can now be downloaded as:
   - msf, selex, pir or fasta formatted text.
   - sequence logo (weblogo).
  All knottins in the database (sequences and 3D structures) are renumbered according to the standard knottin numbering.
   - To get standardized alignments use the 'Search Database' menu, then the 'Get alignment' button.
   - To get classical pre-compiled alignments use the 'Sequences' menu.
   - Sequences or structures can be searched or sorted based on nomenclature (loop lengthes).
  
  

What are knottins

Knottins are small disulfide-rich proteins characterized by a very special "disulfide through disulfide knot"

• This knot is achieved when one disulfide bridge crosses the macrocycle formed by the two other disulfides and the interconnecting backbone.
• The knot implies that knottins contain at least 3 disulfide bridges
• The structural family of knottins have the disulfide between cysteines III and VI (orange) going through disulfides I-IV and II-V (yellow).
• The growth factor cystine knots also contain a knot but the connectivity is different and they cannot be superimposed onto knottins. These proteins belong to a distinct structural family not described in this site.
• Knottins are sometime refered to as "Inhibitor Cystine Knots"

Disulfide connectivity in
Knottins

Disulfide connectivity in
Growth Factor cystine knots

Schematic representation of
Knottins

Schematic representation of
Growth Factor cystine knots

The knottin structural family includes several unrelated families

• Protease inhibitors from plants
• Peptides from the Rubiaceae and the Violaceae plant families
• Toxins from cone snail, spider, bug, horseshoe crab, scorpion
• Gurmarin-like peptides, human Agouti-related protein ...
• Antimicrobial peptides

A number of small disulfide-rich proteins may look like Knottins, but are not Knottins simply because there is no knot in there (see a list here).

The knottin fold is an attractive scaffold for drug design

• it is small and easily accessible to chemical synthesis
• it is very stable, thanks to the high disulfide content and the knotted topology
• it is strongly sequence tolerent