Response from Sandra E. Guggino

Abstract

Listen

Translate article

The paper by Pier et al.1xPier, G.B. et al. Nature. 1998; 393: 79–82Crossref | PubMed | Scopus (225)See all References1 describes the use of the first extracellular loop, and not other parts of CFTR, by Salmonella typhi to obtain entry into epithelial cells. Only these data were addressed in my article.There are four classes of CFTR defect. The first two classes reduce the abundance of CFTR at the plasma membrane by frameshift or nonsense mutations, which stop protein production. There are also processing mutations, which keep CFTR under the surface of the plasma membrane. Only these two defects would be expected to decrease entry of S. typhi, because these defects decrease the surface expression of CFTR. However, there are two additional classes of CFTR defect that would not be consistent with the hypothesis of Pier et al. The G55ID defect results in a malfunction of CFTR gating (a decreased open probability resulting in less chloride secretion) but the surface expression of this defective type of CFTR is normal. Therefore, the entry domain for S. typhi (the first extracellular loop) would be expressed normally. Likewise, there are several defects in the first three transmembrane segments that result in normal surface expression of the CFTR molecule, although conduction of chloride through these CFTR mutants is decreased. Therefore, although all of the defects found in CFTR would decrease chloride secretion and hence provide protection against diarrhea, only those defects associated with decreased surface expression of CFTR would account for heterozygote advantage in S. typhi entry, if the first extracellular domain is an entry point. To assume that a G551D defect in the first nucleotide-binding domain, which controls gating of the channel and is situated beyond the sixth transmembrane segment, causes a conformational change in the first extracellular loop is unlikely. Although it is simplistically elegant to hypothesize that there is decreased binding to mutant CFTR in the lungs and therefore decreased clearance of Pseudomonas aeruginosa infection in CF individuals, this hypothesis does not account for airway infections in those individuals with CFTR still expressed on the apical membrane. Pier et al. remark that expecting multiple mutations expressed in the first extracellular domain represents a Lamarckian view of evolution. Nevertheless, if S. typhi entry was the only selective advantage of the CFTR mutant heterozygote, class 3 or class 4 defects such as G551D would not be expected to have a frequency of 2.4%, which is similar to those of frameshift mutations.As evidence against the selective advantage of the heterozygote for diarrhea produced by cholera, Pier et al. cite conflicting reports on heterozygote CF animals. But it may be too soon for final conclusions to be drawn from only two reports. It is clear, however, from several reports that secretion in the homozygote is reduced when the gut is challenged with many different toxins, including heat-stable toxin2xGoldstein, J.L. et al. Gastroenterology. 1994; 107: 950–956PubMedSee all References, 3xCuthbert, A.W. et al. Br. J. Pharmacol. 1994; 112: 131–136Crossref | Scopus (80)See all References. Pier et al. also contend that cholera in Europe occurred too recently to influence the abundance of the DF508 mutation. Although it might be true that over half of the European population did not die at this time, the fact that all types of diarrhea are reduced by any class of CFTR defect, regardless of where, how or when the >500 defects arose, still makes the diarrheal theory of heterozygote protection attractive. Finally, in their response, Pier et al. suggest that in vivo experiments are critical to these investigations. However, in in vivo experiments showing S. typhi crossing the gut wall, Pier et al. did not report whether or not S. typhi only enters across crypt cells where CFTR resides. The intestinal crypt is well documented to be the site of chloride secretion in diarrhea and it is the only site where CFTR is found.I believe the work by Pier et al.1xPier, G.B. et al. Nature. 1998; 393: 79–82Crossref | PubMed | Scopus (225)See all References1 represents a new theory on the evolution of CFTR defects, but more work is needed to verify the experiments and to distinguish the two theories.