In the work reported here, a RI rat panel was used to identify the LT susceptibility locus for both rats and their macrophages. Analyses of LT sensitivity in several rat strains as well as the RI panel identified a complete correlation between macrophage and rat sensitivity to the toxin. This locus, on chromosome 10, contains rNlrp1, which is the homolog for the mouse mNlrp1b gene, previously proven to be critical for determining murine macrophage susceptibility to LT. Sequence analyses of rNlrp1 in primary macrophages from twelve rat strains identified five polymorphic alleles. Surprisingly, the few polymorphisms that correlated with LT macrophage and animal sensitivity in rats were located within the first 100 aa of rNlrp1, in an area of undefined function, and not within the previously described Nlrp1 functional domains (NACHT, LRR and CARD).
The mapping data strongly suggests (with a P
0.000001) that this rNlrp1-
containing region of chromosome 10 is the LT sensitivity locus for both rats and their macrophages. Among the approximately 250 microsatellite markers previously characterized for this RI set 
and new previously unpublished markers sets, we found that marker D10Rat77 on chromosome 10 had an absolute genotypic correlation with the LT sensitivity phenotype. SNP analysis in this region also confirms the marker data and shows perfect correlation for LT susceptibility within the locus containing rNlrp1
(genome-wide empiric p-value
0.001). Our mapping data does not rule out the possibility that another gene very closely linked to rNlrp1
could be mediating LT's effects in the rat. However, two additional lines of evidence argue against this possibility. First, rNlrp1 aa sequence variations between several inbred rat strains unrelated to the RI panel progenitors correspond perfectly with the sensitivity phenotypes. Second, an absolute correlation was found between macrophage sensitivity and animal susceptibility for 34 rat strains. Considering the established role of mNlrp1b
in control of murine macrophage sensitivity, it is unlikely that a different gene controls macrophage sensitivity in rats. However we cannot exclude the possibility that whole animal susceptibility is controlled by multiple closely-linked genes within the single chromosome 10 locus identified in this study, and that these genes are inherited in a fashion such that polymorphisms associated with sensitivity in rNlrp1
are also always found in another candidate gene. Historically, a similar issue plagued the identification of mNlrp1b
as the mouse macrophage sensitivity locus for LT. The sensitivity locus in mice was first identified as the closely linked Kif1c
gene, which presented almost perfect polymorphism correlations with sensitivity 
. In the absence of a transgenic rat model providing definitive proof linking the rNlrp1
gene to rat death, we believe the mapping and sequence data presented here strongly support rNlrp1
as the most likely determinant of LT sensitivity.
Gene predictions and BLAST searches identified a potential paralog immediately adjacent to rNlrp1
, but this paralog is not expressed at the mRNA level, at least in macrophages. Similarly, of the three tandem mNlrp1
paralogs found in mice, only mNlrp1b
was shown to be expressed in the LT-sensitive 129S1/SvImJ macrophage, and expression of this paralog was sufficient to confer LT sensitivity to resistant mouse macrophages and fibroblasts 
. However, the other two mouse paralogs are expressed in a number of inbred strains, further complicating analyses of mouse susceptibility. Curiously, phylogenetic analyses indicate that the predicted rNlrp1 paralog sequence is distant from the other rat Nlrp1 sequences and is more similar to the mouse and human Nlrp1 sequences (Figure S4
The highly polymorphic nature of the mNlrp1b
alleles has made it difficult to associate specific polymorphisms with the macrophage sensitivity phenotype in mice. Fortunately, there are far fewer differences in rNlrp1
between sensitive and resistant rat strains. Sequence differences that correlated with phenotypic differences were found only within the extreme N-terminal region of rNlrp1, and not in the domains (NACHT, LRR, and CARD) which have recognized roles in Nlrp1 function. This surprising finding draws attention to the N-terminal domains of rodent Nlrp1 proteins, absent in hNlrp1, which instead contain a pyrin domain at the N-terminus 
. The pyrin domain in hNlrp1 is required for association with the inflammasome adaptor protein ASC, which is not part of the LT-induced mNlrp1b inflammasome complex 
. Interestingly, all human macrophages tested to date have been LT-resistant (unpublished observations), a behavior that might relate to the absence of the N-terminal pyrin domain in rodent Nlrp1. However, hNlrp1 polymorphisms are now being identified and associated with a number of human diseases 
, so it may be necessary to test a larger number of donors to identify any LT-sensitive hNlrp1
alleles. As the N-terminus in hNlrp1 plays an important role in protein-protein interactions, it is tempting to postulate that the N-terminal 100 aa of rodent Nlrp1 proteins may also interact with other cellular components to modulate function.
The perfect correlation of rat macrophage LT sensitivity to that of the animals might at first suggest that the lysis of macrophages in vivo
causes the rapid death of LT-injected rats. However, this is unlikely to be the case, for several reasons. Rat macrophages begin to die only 2 h after treatment in vitro
with saturating toxin doses, whereas the rats may die in as little as 37 min 
. Initial studies in LT-treated mice were interpreted as showing that death (which occurs only after 2–3 days) resulted from cytokines released following macrophage lysis 
. However, more extensive later studies showed that mice harboring resistant macrophages also succumb to LT through a vascular collapse that is similar to that in mice with sensitive macrophages 
, and the correlation within mouse strains between the LT sensitivities of isolated macrophages and the animals is poor 
. Studies with mNlrp1b transgenic mice confirm that macrophage and animal susceptibility to LT are not correlated 
. Preliminary studies in our laboratory suggest that cell types other than macrophages control the lethal response to LT (data not shown). Consistent with this view, Nlrp1 has recently been demonstrated to play a functional role in a number of cell types, including neuronal cells 
. Furthermore, it should be noted that LT-induced death in both rats and mice has recently been associated with early changes in cardiac function 
. Thus, it is possible that LT targeting of rNlrp1 function in the heart plays a role in the rapid lethality phenotype. A better understanding of the distribution and function of different Nlrp1 isoforms in various cell types is needed to fully understand the mechanisms by which LT may influence Nlrp1 activity, and whether this gene alone is sufficient for control of animal susceptibility to toxin.
In summary, we present data mapping the control of rapid LT-induced rat death to a single chromosome 10 locus. This locus contains the polymorphic rNlrp1 gene, which is the best candidate for conferring sensitivity to macrophages, and possibly to animals. As such, this is the first suggestion that an inflammasome NLR protein may directly control animal lethality. While both the mechanistic basis for the rapid LT-induced lethality in the rat and direct proof of rNlrp1-mediated rat death require further experimentation, identification of the limited polymorphisms within rNlrp1 that correlate perfectly with LT sensitivity suggest a starting point for analysis of the possible role this protein may play in controlling rapid rat death in response to LT.