Protein-splicing mechanism
The current knowledge of the protein splicing mechanism is mostly the result of
work by the research groups of FB Perler (New England Biolabs)
and H Paulus (Boston Biomedical Research Institute). A series of papers in 1993-96
elucidated the biochemistry of the reaction. They were summarized in two papers
in 1996
(Xu and Perler, EMBO J '96,
Chong et al, JBC '96) and in later reviews
(Perler et al, Curr. Opin. Chem Biol.,
Paulus, Chem. Soc. Rev. 27:375 '98
).
A full bibliography
of protein splicing and intein papers
can be found in the Inteins Database maintained by
FB Perler at NEB. Here I illustrate the reaction based on these works.
Shmuel Pietrokovski
 |
*
The first step in protein splicing is a reversible transition of the
peptide bond between the amino end of the intein and its amino terminal
flank (N-extein) into an ester or thioester bond. This transition depends on
a nucleophilic attack of the bond by the side-chain of the Ser or Cys residues
at the amino terminal end of the intein (-OH or -SH respectively). This
reaction is termed N-O when the attacking atom is an Oxygen and N-S when this
atom is Sulfur. This scheme shows the reaction with a Cys in the intein amino
end. All inteins begin with either Ser or Cys residues,except for
the two klbA inteins in M.jannaschii and Pyrococcus horikoshii OT3.
These start with an Ala and if they are active it cannot be through this
step since the Ala side-chain is a methyl group (CH3) not capable of
nucleophilic attack.
|
 |
 |
*
In the next step the side-chain of the residue C-terminal to the intein
(the first residue of the Carboxy (C) extein) attacks the ester (or thioester)
bond at the amino end of the intein. Here too the attack is by a polar side
chain of a Ser, Thr (both OH) or Cys (SH).This leads to a transesterification
and formation of a branched intermediate with two amino ends, one of the
N-extein
and one of the intein. The intein is joined by peptide bond to the C-extein
and the two exteins are joined by a thio/ester bond. This reaction is also
reversible. All known inteins indeed have Ser, Thr or Cys directly
following their C-end. This scheme shows the reaction with a Ser following
the intein carboxy end.
|
|
 |
*
The branched intermediate is resolved by the cyclization of the C-terminal
intein residue. The intein is now fully excised from the N and C exteins that
are yet linked to each other by the thio/ester bond. This step is
ireversible driving the reaction
forward. Almost all inteins have Asn as their carboxy end (as shown in
this scheme) and its cyclization results in a succinimide ring. Two known
inteins have Gln in their carboxy end and
a variation of this step has been proposed to
account for this. In brief, the reaction proceeds through Gln cyclization
into a glutarimide ring.
|
 |
 |
*
The final steps consist of spontaneous shift of the thio/ester bond linking
the exteins into a peptide bond (S/O-N acyl rearrangement) and probably some
hydrolysis of the succinimide (or glutarimide) ring at the intein carboxy end
to Asn and iso-Asn. These reactions are ireversible too and form
the mature host protein, chemically identical to the product of an intein-less
gene. Not much is biochemically known on the fate of the excised intein. In
experiments where it is over-produced it seems to be rapidly degraded. However,
genetic and phylogenetic analysis show that some inteins are also responsible
for the homing of the intein gene into unoccupied intein integration points in
homologous genes. This horizontal-transfer gene conversion is mediated by the
homing endonuclease protein domain found in the central region of most inteins.
The reaction is totally independent of the protein splicing reaction depicted
here.
|
|
 |
 |
|