***************************
* DnaJ domains signatures *
***************************

The prokaryotic heat shock protein dnaJ  interacts  with  the chaperone hsp70-
like dnaK  protein  [1].  Structurally,  the  dnaJ  protein  consists of an N-
terminal conserved  domain  (called  'J'  domain)  of  about 70 amino acids, a
glycine-rich region  ('G'  domain')  of  about  30  residues, a central domain
containing four  repeats  of  a CXXCXGXG motif ('CRR' domain) and a C-terminal
region of  120  to  170  residues.  Such a structure is shown in the following
schematic representation:

  +------------+-+-------+-----+-----------+--------------------------------+
  | N-terminal | | Gly-R |     | CXXCXGXG  | C-terminal                     |
  +------------+-+-------+-----+-----------+--------------------------------+

It has been shown [2] that the 'J' domain as well as the 'CRR' domain are also
found in other prokaryotic and eukaryotic proteins which are listed below.

a) Proteins containing both a 'J' and a 'CRR' domain:

 - Yeast protein MAS5/YDJ1 which seems to be involved in mitochondrial protein
   import.
 - Yeast  protein  MDJ1,  involved  in  mitochondrial  biogenesis  and protein
   folding.
 - Yeast protein SCJ1, involved in protein sorting.
 - Plants dnaJ homologs (from leek and cucumber).
 - Human HDJ2, a dnaJ homolog of unknown function.

b) Proteins containing a 'J' domain without a 'CRR' domain:

 - Rhizobium  fredii  nolC, a protein involved in cultivar-specific nodulation
   of soybean.
 - Escherichia coli cbpA [3], a protein that binds curved DNA.
 - Yeast  protein  SEC63/NPL1,  important  for  protein    assembly  into  the
   endoplasmic reticulum and the nucleus.
 - Yeast protein SIS1, required for nuclear migration during mitosis.
 - Plasmodium falciparium  ring-infected  erythrocyte  surface antigen (RESA).
   RESA, whose function is not known, is associated with the membrane skeleton
   of newly invaded erythrocytes.
 - Human HDJ1.
 - Human HSJ1, a neuronal protein.

We developed  a  signature  pattern  for  the  'J'  domain, based on conserved
positions in  the  C-terminal half of this domain. We also developed a pattern
for the 'CRR' domain, based on the first two copies of that motif.

-Consensus pattern: [FY]-x(2)-[LIVMA]-x(3)-[FYWH]-[DENQSA]-x-L-x-D-x(3)-[KR]-
                    x(2)-[FY]
-Sequences known to belong to this class detected by the pattern: ALL.
-Other sequence(s) detected in SWISS-PROT: budding yeast zuotin.

-Consensus pattern: C-[DESTHK]-x-C-x-G-x-[GK]-[AGSM]-x(2)-[GNK]-x(4,5)-C-
                    x(2,3)-C-x-G-x-G
-Sequences known to belong to this class detected by the pattern: ALL.
-Other sequence(s) detected in SWISS-PROT: NONE.

-Expert(s) to contact by email: Kelley W.
                                kelley@cmu.unige.ch

-Last update: June 1994 / Patterns and text revised.

[ 1] Cyr D.M., Langer T., Douglas M.G.
     Trends Biochem. Sci. 19:176-181(1994).
[ 2] Bork P., Sander C., Valencia A., Bukau B.
     Trends Biochem. Sci. 17:129-129(1992).
[ 3] Ueguchi C., Kaneda M., Yamada H., Mizuno T.
     Proc. Natl. Acad. Sci. U.S.A. 91:1054-1058(1994).