PROSITE documentation PDOC00913

Calcium-binding EGF-like domain signature

A sequence of about forty amino-acid residues long found in the sequence of epidermal growth factor (EGF) has been shown [1,2,3,4,5,6] to be present in a large number of membrane-bound and extracellular, mostly animal proteins (see <PDOC00021>). Many of these proteins require calcium for their biological function and a calcium-binding site has been found to be located at the N-terminus of some EGF-like domains [8]. Calcium-binding may be crucial for numerous protein-protein interactions. Proteins that are known or that are predicted to contain calcium-binding EGF-like domains are listed below.

• Bone morphogenic protein 1 (BMP-1), a protein which induces cartilage and bone formation and which expresses metalloendopeptidase activity (1-2 copies). Homologous proteins are found in sea urchin - suBMP (1 copy) - and in Drosophila - the dorsal-ventral patterning protein tolloid (2 copies).
• Caenorhabditis elegans developmental proteins lin-12 (2 out of 13 copies) and glp-1 (10 copies).
• Calcium-dependent serine proteinase (CASP) which degrades the extracellular matrix proteins type I and IV collagen and fibronectin (1 copy).
• Cartilage oligomeric matrix protein COMP (2 out of 4 copies).
• Coagulation factors VII, IX, and X (1 out of 2 copies).
• Complement C1r components (1 copy).
• Complement C1s components (1 copy).
• Complement-activating component of Ra-reactive factor (RARF) (1 copy).
• Crumbs, an epithelial development protein from Drosophila (12 out of 29 copies).
• Epidermal growth factor precursor (3 out of 9 copies).
• Fibrillin 1 and fibrillin 2 (43 out of 47 copies).
• Fibropellins IA (8 out of 21 copies) and III (6 out of 8 copies) from the apical lamina - a component of the extracellular matrix - of sea urchin.
• Fibulin-1 (8 out of 9 copies) and fibulin-2 (9-10 out of 10-11), two extracellular matrix proteins.
• Leucocyte antigen CD97 (2 out of 3 copies), cell surface glycoprotein EMR1 (5 out of 6 copies) and cell surface glycoprotein F4/80 (6 out 7 copies).
• LDL receptors, which bind and transport low-density lipoproteins (1 out of 3 copies).
• Neurogenic proteins Notch (21-22 out of 36), Xotch (21 out of 36), Motch (16 out of 34) and the human homolog Tan-1 (18 out of 36), Delta (2 out of 9 copies), Drosophila Serrate (5 out of 14 copies) and Slit (2 out of 7 copies).
• Nidogen (also called entactin), a basement membrane protein from sea squirt (0 out of 2) and mammals (2 out of 6).
• Proteoglycans versican (1 out of 2 copies) and chondroitin sulfate proteoglycan (gene PG-M) (1 out of 2 copies).
• S1-5, a human extracellular protein whose ultimate activity is probably modulated by the environment (5 copies).
• Thrombomodulin (fetomodulin), which together with thrombin activates protein C (2 out of 6 copies).
• Thrombospondins 1 and 2 (1 out of 3 copies), 3 and 4 (2 out of 4 copies), adhesive glycoproteins that mediate cell-to-cell and cell-to-matrix interactions.
• Thyroid peroxidase 1 (EC 1.11.1.8) (1 copy).
• Transforming growth factor β-1 binding protein (TGF-B1-BP) (14 out of 16 or 18 copies).
• Uromodulin (Tamm-horsfall urinary glycoprotein) (THP) (2 out of 3 copies).
• Vitamin K-dependent anticoagulant protein S (3 out of 4 copies).
• 63 Kd sperm flagellar membrane protein from sea urchin (1 out of 3 copies).
• 93 Kd protein (gene nel) from chicken (2 out of 5 copies).
• Hypothetical 337.6 Kd protein T20G5.3 from Caenorhabditis elegans (1 out of 44 copies).

For human coagulation factor IX it has been shown [7] that the calcium-ligands form a pentagonal bipyramid. The first, third and fourth conserved negatively charged or polar residues are side chain ligands. Latter is possibly hydroxylated (see <PDOC00010>) [8]. A conserved aromatic residue as well as the second conserved negative residue are thought to be involved in stabilizing the calcium-binding site.

Like in non-calcium binding EGF-like domains there are six conserved cysteines and the structure of both types is very similar as calcium-binding induces only strictly local structural changes [8].

                             +------------------+        +---------+
                             |                  |        |         |
               nxnnC-x(3,14)-C-x(3,7)-CxxbxxxxaxC-x(1,6)-C-x(8,13)-Cx
               ****|******************|**********
                   +------------------+

'n': negatively charged or polar residue [DEQN]
'b': possibly beta-hydroxylated residue [DN]
'a': aromatic amino acid
'C': cysteine, involved in disulfide bond
'x': any amino acid
'*': position of patterns.

We have used the N-terminal part of the EGF domain as a consensus pattern. It includes the negative N-terminus and the possible hydroxylation site.