domains form structurally similar a-helical
bundles yet have distinct signaling specificities
dependent also on homotypic interactions. In
line with this conceptual analogy, it was
demonstrated that the PFD of HET-S and the
R0 NWD2 amyloid motif can replace the PYD
domains of NLRP3 and ASC, underlying that
prion-like polymerization is equally a feature of
the death domains superfamily, although not
based on controlled amyloid folding or amyloid
propagation (Cai et al. 2014).
Besides the HRAMs, at least two other
signal-transducing amyloids have been identified and named PP (pseudo-palindromic) and
s (sigma). The s putative signaling amyloid
was identified on the N-terminus of an NLR
protein encoded by the gene het-eN in the
genome of Nectria haematococca (Daskalov
et al. 2012; Graziani et al. 2004). Adjacent to
het-eN two other genes are found, sesA and
sesB, encoding proteins carrying at their Ctermini the s putative amyloid domain. The
length of the s-motif is about 25 amino acid
residues. The amino acid composition of the
motif is biased toward predominantly G, N/Q,
and aromatic residues. Unlike the HRAMs and
their associated NLRs, the s-motif found on the
HET-eN NLR and its downstream effectors
(SESA and SESB) are of equivalent length.
Nonetheless, a common point with the
HRAMs is the identification of sub-motifs composing the various s-motifs. These sub-motifs,
called A and B, are usually composed of eight or
nine amino acids and arranged in three different patterns (AAA, ABA, ABB). There is currently no direct evidence for the amyloidogenic
prion properties of the s-motif; however the
gene cluster (het-eN/sesA/sesB) has been previously associated in N. haematococca with an
infectious phenomenon termed “secteur”
caused by the s infectious element (Graziani
et al. 2004). The hypothesis that this s infectious cytoplasmic element corresponds to a
prion form of SESA and SESB is a plausible one.
More direct experimental evidence supports the amyloid prion properties of the PP
signaling domain (Daskalov et al. 2016). In one
reported case, the PP domains of an NLR and
its effector protein have been shown to form
prion amyloids. The NLR protein is PNT1 (PPNACHT-TPR), and the effector protein is
HELLP (HeLo-like PP), both encoded by a
gene cluster in the genome of Chaetomium globosum, a species in the same taxonomic order
as P. anserina. Like the nwd2/het-S cluster, the
pnt1/hellp and het-eN/sesA/sesB gene clusters
are present in the genomes of a variety of
other ascomycete species. The distinct prion
specificity of the PP amyloids, revealed in the
lack of detected heterospecific interactions
(cross-seeding) with the PFD of HET-S, offers
experimental evidence that supports the existence of different signaling specificity for each
amyloid domain. The pseudo-palindromic consensus sequence NxGfQfGxN, centered on the
Q residue, is the most highly conserved central
part of the 17/18 amino acids PP-motif.
Remarkably, the PP-motif shows strong
sequence similarity to RHIM (RIP homotypic
interaction motif)—an amyloid controlling a
pro-inflammatory programmed cell death reaction in metazoans termed necroptosis, which
plays a role in innate immunity (Li et al.
2012). A solid-state NMR structure of a RHIM
hetero-amyloid has been recently reported,
where the RHIM motifs from two different proteins (RIPK1 and RIPK3) alternate to form two
fronting b-sheets with a unique serpentine fold
and a common hydrophobic core (Mompean
et al. 2018). The amyloid core is structured
around the association of the two central
GfQfG motifs in which the similarity to the
PP-motif lies, making it likely that fungal PP
amyloids adopt a similar fold, which in turn
would strengthen a model of long-term evolutionary conservation of programmed cell
death-related amyloid signaling from fungi to
animals.
VI. Evolutionary Origin of NLRs:
Parallel Versus Convergent
Evolution
When it was shown that the NOD1 protein in
mammals detects bacterial lipopolysaccharides
and thus functions as an immune receptor, the
striking conclusion that plants and animals
employ analogous receptors in immune func134
A. Daskalov et al.
bundles yet have distinct signaling specificities
dependent also on homotypic interactions. In
line with this conceptual analogy, it was
demonstrated that the PFD of HET-S and the
R0 NWD2 amyloid motif can replace the PYD
domains of NLRP3 and ASC, underlying that
prion-like polymerization is equally a feature of
the death domains superfamily, although not
based on controlled amyloid folding or amyloid
propagation (Cai et al. 2014).
Besides the HRAMs, at least two other
signal-transducing amyloids have been identified and named PP (pseudo-palindromic) and
s (sigma). The s putative signaling amyloid
was identified on the N-terminus of an NLR
protein encoded by the gene het-eN in the
genome of Nectria haematococca (Daskalov
et al. 2012; Graziani et al. 2004). Adjacent to
het-eN two other genes are found, sesA and
sesB, encoding proteins carrying at their Ctermini the s putative amyloid domain. The
length of the s-motif is about 25 amino acid
residues. The amino acid composition of the
motif is biased toward predominantly G, N/Q,
and aromatic residues. Unlike the HRAMs and
their associated NLRs, the s-motif found on the
HET-eN NLR and its downstream effectors
(SESA and SESB) are of equivalent length.
Nonetheless, a common point with the
HRAMs is the identification of sub-motifs composing the various s-motifs. These sub-motifs,
called A and B, are usually composed of eight or
nine amino acids and arranged in three different patterns (AAA, ABA, ABB). There is currently no direct evidence for the amyloidogenic
prion properties of the s-motif; however the
gene cluster (het-eN/sesA/sesB) has been previously associated in N. haematococca with an
infectious phenomenon termed “secteur”
caused by the s infectious element (Graziani
et al. 2004). The hypothesis that this s infectious cytoplasmic element corresponds to a
prion form of SESA and SESB is a plausible one.
More direct experimental evidence supports the amyloid prion properties of the PP
signaling domain (Daskalov et al. 2016). In one
reported case, the PP domains of an NLR and
its effector protein have been shown to form
prion amyloids. The NLR protein is PNT1 (PPNACHT-TPR), and the effector protein is
HELLP (HeLo-like PP), both encoded by a
gene cluster in the genome of Chaetomium globosum, a species in the same taxonomic order
as P. anserina. Like the nwd2/het-S cluster, the
pnt1/hellp and het-eN/sesA/sesB gene clusters
are present in the genomes of a variety of
other ascomycete species. The distinct prion
specificity of the PP amyloids, revealed in the
lack of detected heterospecific interactions
(cross-seeding) with the PFD of HET-S, offers
experimental evidence that supports the existence of different signaling specificity for each
amyloid domain. The pseudo-palindromic consensus sequence NxGfQfGxN, centered on the
Q residue, is the most highly conserved central
part of the 17/18 amino acids PP-motif.
Remarkably, the PP-motif shows strong
sequence similarity to RHIM (RIP homotypic
interaction motif)—an amyloid controlling a
pro-inflammatory programmed cell death reaction in metazoans termed necroptosis, which
plays a role in innate immunity (Li et al.
2012). A solid-state NMR structure of a RHIM
hetero-amyloid has been recently reported,
where the RHIM motifs from two different proteins (RIPK1 and RIPK3) alternate to form two
fronting b-sheets with a unique serpentine fold
and a common hydrophobic core (Mompean
et al. 2018). The amyloid core is structured
around the association of the two central
GfQfG motifs in which the similarity to the
PP-motif lies, making it likely that fungal PP
amyloids adopt a similar fold, which in turn
would strengthen a model of long-term evolutionary conservation of programmed cell
death-related amyloid signaling from fungi to
animals.
VI. Evolutionary Origin of NLRs:
Parallel Versus Convergent
Evolution
When it was shown that the NOD1 protein in
mammals detects bacterial lipopolysaccharides
and thus functions as an immune receptor, the
striking conclusion that plants and animals
employ analogous receptors in immune func134
A. Daskalov et al.