in a number of proteins that also contain a CARD domain, including ASC/Pycard,
Nod1/CARD4, and NALP1. The grown recognition for the importance of functional
domains in apoptosis proteins is reflected by the fact that the pyrin domain was
discovered independently by five different groups and published almost
simultaneously (Bertin and DiStefano, 2000; Fairbrother et al., 2001; Martinon et
al., 2001; Pawlowski et al., 2001; Staub et al., 2001). Three reports have suggested
the name ‘pyrin domain’, which is also used here, while PAAD and DAPIN are
alternative names. Up to now, no three-dimensional structure for the pyrin domain
has been available. Since several methods of secondary structure prediction and
threading analysis all suggest a six-helix topology similar to DD, DED, and CARD,
the grouping of the PYD with those domains appears to be appropriate.
The ASC/Pycard protein has the architecture of a typical adapter protein, with an
N-terminal pyrin domain and a C-terminal CARD. The pyrin domain of this protein
was shown to mediate homodimerization and also interaction with the PYD of the
NALP1 protein (Martinon et al., 2001). The current understanding of the physiologic
importance of these proteins and their interaction is still insufficient. Another protein
suggesting a functional relationship to other death adapter proteins is the CASPY
caspase of zebrafish. This protein has the architecture of a typical caspase, but uses a
PYD as the prodomain, while mammalian caspases use DEDs or CARDs instead.
NALP1 and NALP2 appear to be the founding members of a large protein family
with an analogous architecture consisting of an N-terminal pyrin domain, followed
by a NACHT-type ATPase (see Section 5.2) and a leucine-rich repeat (LRR) region.
An interesting link between the pyrin domain and a family of human diseases is its
occurrence in the proteins pyrin/marenostrin and cryopyrin. The gene for the former
protein is mutated in Mediterranean periodic fever syndrome (Consortium, T.F.F.,
1997; Consortium, T.I.F., 1997), while the gene for cryopyrin is mutated in Muckle-
Wells syndrome, another periodic fever disorder (Hoffman et al., 2001). Cryopyrin
is a NALP-like protein, while pyrin has a different architecture with an N-terminal
pyrin domain, followed by a B-box zinc finger and a C-terminal SPRY domain.
Interestingly, a third condition associated with periodic fever is caused by a mutation
in the TNF-receptor, providing a further link between the corresponding pathways.
A distinct subfamily of pyrin domains, the ‘IFI-domain’, is found in a family of
interferon-induced proteins. The original reports of the pyrin domain disagree in the
judgment of whether these domains should be considered genuine pyrin domains.
There is little doubt that these domains are all related and are much closer to each
other than they are to the other six-helix death adapter domains. Thus, a grouping
with the pyrin domain seems to be warranted. It would be interesting to know whether
these IFI-type domains are able to interact with the conventional pyrin domains. The
current protein sequence databases hold nine human pyrin proteins, which are listed
in Table 6. This number is likely to grow considerably, since the poorly analyzed part
of the human genome appears to contain many more proteins belonging to this
domain class.
86 GENETICS OF APOPTOSIS