点击显示 收起
Section of Functional Genomics, Division of Genomic Medicine, Sheffield University Medical School, Sheffield, United Kingdom
To the EditorThe Asp299Gly single-nucleotide polymorphism (SNP) of the human TLR4 gene is the most common hitherto described and is apparently functional in that airway epithelial cells from patients who carry it are hyporesponsive to lipopolysaccharide (LPS) and exhibit reduced expression of Toll-like receptor 4 (TLR4), a loss of function that is restored by ectopic transfection of the wild-type gene [1]. Furthermore, compared with individuals with wild-type TLR4, people with the Asp299Gly allele have been shown to have lower resting plasma concentrations of some (but not other) markers of inflammation, including interleukin-6, procalcitonin, fibrinogen, and soluble vascular-cell adhesion molecule 1 [2]. Therefore, it is reasonable to postulate that people with the Asp299Gly polymorphism should be more likely to manifest disease after exposure to a gram-negative infection, in which LPS is a biologically significant driver of pathology.
The data presented here by Morré et al. [3], together with ours [4], suggest that this is not the case. We found, in our large study of patients with a history of meningococcal disease and blood-donor control subjects, that the polymorphism is not overrepresented in people with this prototypic gram-negative infection; the allele frequency of Asp299Glu was 5.9% in 879 blood-donor control subjects, 6.5% in 1047 patients with microbiologically proven meningococcal disease, and 4.1% among 86 patients who died of proven meningococcal disease [4]. These frequencies exhibited no significant partitioning among patients, including fatal cases. It is noteworthy that the overall allele frequencies reported by ourselves are strikingly similar to those observed by Morré et al., lending technical credence to each study.
Two studies originating from the laboratory that discovered the Asp299Gly polymorphism have reported overrepresentation of the SNP in patients with clinical disease. Kiechl et al. [2], in a study involving 810 middle-aged and older individuals, 53 of whom were heterozygous for the Asp299Gly allele, found that the frequency of putative, undefined bacterial infections was higher in carriers of the SNP and that the severity of atherogenesis was also diminished in carriers. Lorenz et al. [5] studied 91 patients with septic shock and 73 healthy blood-donor control subjects and found the Asp299Gly exclusively in the septic-shock cohort. The latter study was somewhat small, but the absence of the allele in any of the blood-donor control subjects was a little surprising, in light of the allele frequency that weand Morré et al.have reported.
Asp299Gly is not the only polymorphism in human TLR4; there is another, Thr399lle, that also impairs lipopolysaccharide signaling and inflammation [1]. We have sequenced full-length TLR4 of people who have experienced meningococcal disease, and we have found other, functionally insignificant polymorphisms that have no likely direct causal relationship with the disease (R.C.R., unpublished data). In C3H/HeJ mice, a point mutation within exon 3 modifies TLR4 within the cytoplasmic portion of the receptor, which is critical for signal transduction [6]. In a search for gene variation that is likely to be biologically significant, we examined a 500-bp segment of the same region of the gene in 51 patients with meningococcal disease, but we found no evidence of sequence variation [4].
Despite both the reported functional significance of Asp299Gly and the reported associations described elsewhere, we remain unconvinced that it has any effect on the manifestation of gram-negative infections, and we consider that inefficient TLR4 signaling may not be rate limiting during human responses to natural LPS-driven diseases. There are a considerable number of collateral pathways available for innate responses to infections, and 10 human TLRs have now been described, which differ in their cofactor requirements and in their pattern-recognition specificities [7]. In the case of Neisseria meningitidis, there is at least 1 non-TLR4 pathway for innate signal transduction; TLR2 can also signal the presence of the organism [8, 9], possibly in response to the presence of lipoprotein H8. It is likely that such pathways contribute to effective innate immunity even if TLR4 function or expression is impaired.
It is clear that there is some genetic influence on the expression of disease due to Chlamydia trachomatis [10, 11], but Morré et al. show that C. trachomatisseropositive patients with tubal disease do not have an Asp299Gly allele frequency that differs from what would be expected. Although they used the best serological test available, we would offer the cautionary note that, compared with nucleic acid detection, serological tests have poor specificity in the diagnosis of genital infection due to this organism [12].
References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.