Molecular Epidemiology of Candida

In an interesting report analyzing isolates from Japan by our PCR method (2), Tamura et al. (9) state that “Mercure et al. (4) reported that Candida albicans produced a well-characterized EcoRI restriction fragment polymorphism pattern whose bands … the dimorphic (3.7- and 4.2-kbp) fragment … shown to have originated from the … rDNA … has been used to classify C. albicans into two types (genotypes A and B).” However, the study that reported that finding was from our laboratories, and it was 6 years earlier (6) than the study (4) they cite. Tamura et al. then state, “Their further studies confirmed that the presence or absence of group I introns … accounts for the difference…,” and they cite a 1998 paper for this (5), when the appropriate citation there should have been Mercure et al. (4) (and should have omitted “their”). Tamura et al. then state, “Those studies led to the preparation … of a PCR primer pair that can demonstrate the … group I introns … (2).” However, it was our prior studies as well as those of others (4, 6, 7) that led to our PCR studies (2), not the study (5) they cite, as our publication on PCR (2) indicated.

Tamura et al. cite our 2000 paper (1) as reporting the first case of Candida dubliniensis in Japan, but the list of references given by Tamura et al. doesn't include our 1999 paper (3), which not only included notation of a C. dubliniensis isolate from Japan, but, more importantly, indicated a distribution of C. albicans genotypes A, B, and C in Japan of 30, 9, and 12, respectively, very similar to the distribution (172, 66, and 56, respectively) Tamura et al. now report. Tamura et al. then state that their “… present experimental data … suggests that C. dubliniensis is distributed all over the world,” but this conclusion has been reached previously by many prior studies from several laboratories, including our own (3, 8).

Tamura et al. indicate that “ … the dark blue color of colonies on CHROMagar Candida, which has been considered one of the more useful features for the identification of C. dubliniensis, was not confirmed… .” However, it is a dark green colony that is commonly seen for C. dubliniensis on this agar, and several laboratories have previously indicated the unreliability of this criterion for identifying C. dubliniensis, thus underscoring the need for genotypic identification.

Finally, Tamura et al. cite a study (4) correlating genotypic group and flucytosine susceptibility, an association Tamura et al. also explored. However, this association was first reported by us (7) 3 years prior to the work Tamura et al. cite.

Editor’s Note:

The authors of the published article declined to respond.