******************************
* 'Trefoil' domain signature *
******************************

A cysteine-rich  domain of  approximately forty five   amino acid residues has
been  found in some proteins from   different biological   sources [1,2].  The
'trefoil' domain contains six cysteines that are most probably linked by three
disulfide bonds. The proteins that are known to contain this domain are:

 - Pig Pancreatic Spasmolytic Polypeptide (PSP) [3], a protein of 106 residues
   that inhibits  gastrointestinal motility and  gastric  acid secretion.  PSP
   could also be a growth factor.  PSP contains two  copies  of  the 'trefoil'
   domain.
 - Human protein pS2 [4], a protein secreted by the stomach mucosa, whose gene
   is induced by  estrogen.  The exact function of pS2 is not known.  pS2 is a
   protein of 84  residues  (including  a  signal peptide of 21 residues). pS2
   contains one copy of the 'trefoil' domain.
 - Xenopus preprospasmolysin [5], a skin  gland precursor protein of 400 amino
   acid residues. This protein contains 4 copies of the 'trefoil' domain.  Two
   of these domains are in the N-terminal of the protein, while the two others
   are  in  the  C-terminal end;  the  separating  space  consists  of  tandem
   threonine/proline-rich repeats. The function of this protein is not known.
 - Xenopus 'APEG' protein [6], a skin  gland  protein of about 400 amino acids
   that mostly consist of tandem repeats of the sequence G-[E/G]-[A/P](2,4)-A-
   E.  This protein contains,  at  its  C-terminal extremity,  a  copy  of the
   'trefoil' domain.

Structurally the 'trefoil' domain can be represented as shown below.

              +-------------------------+
              |         +--------------+|
              |         |              ||
            xxCxxxxxxRxxCG#xxxxxxxCxxxxCC#xxxxxxxxWC#xxxxxxxx
                     *************|*****           |
                                  |                |
                                  +----------------+

'C': conserved cysteine involved in a disulfide bond.
'#': large hydrophobic residue.
'*': position of the pattern.

-Consensus pattern: R-x(2)-C-G-[FY]-x(3)-[ST]-x(3)-C-x(4)-C
-Sequences known to belong to this class detected by the pattern: ALL.
-Other sequence(s) detected in SWISS-PROT: NONE.

-Note: the residue in  position 7 of the pattern is Pro, except  in  the third
 'trefoil' domain of Xenopus preprospasmolysin, where it is replaced by Ile.

-Last update: November 1990 / Text revised.

[ 1] Thim L.
     FEBS Lett. 250:85-90(1989).
[ 2] Baker M.E.
     Biochem. J. 253:307-308(1988).
[ 3] Thim L., Thomsen J., Christensen M., Jorgensen K.H.
     Biochim. Biophys. Acta 827:410-418(1985).
[ 4] Rio M.C., Bellocq J.P., Daniel J.Y., Tomasetto C., Lathe R.,
     Chenard M.P., Batzenschlager A., Chambon P.
     Science 241:705-708(1988).
[ 5] Hoffmann W.
     J. Biol. Chem. 263:7686-7690(1988).
[ 6] Gmachl M., Berger H., Thalhammer J., Kreil G.
     FEBS Lett. 260:145-148(1990).