It is more than obvious from our report on three M. heckeshornense strains isolated from two patients that this species represents a look-alike of M. xenopi and that this might cause problems in the differential identification of these two scotochromogens if only phenon-specific characteristics were considered. The comments of Richter et al. on two additional strains of M. heckeshornense found in their laboratory, which yielded positive results in tests for arylsulfatase, pyrazinamidase, and nicotinamidase activities, point in this direction. Unfortunately, the authors do not exactly specify what testing methods they used. It is well known that results may differ when pyrazinamidase activity is determined either by the agar method or by the method introduced by Bönicke (1-4). M. heckeshornense shows a very slow growth rate at 37°C (and even at 45°C), and we are well advised to state how tests were performed, e.g., at higher incubation temperatures or with higher inoculum sizes. More importantly (amidase test results may indeed vary for this species), we want to emphasize that interpretation of the arylsulfatase test results demands judicious consideration of test incubation time and quantification of positive results. Typically, about 60 and 99% of M. xenopi isolates present a strongly positive result in the 3-day and the 10-day arylsulfatase tests, respectively (1-1, 1-4). Thus, careful attention to this test with respect to quantification may be the sole biochemical clue that a subject strain with a negative or weakly positive test result possibly represents a M. heckeshornense isolate. However, with the introduction of tools of molecular biology, we can now approach differential identification of bacteria in the routine practice by highly precise methods applicable to all species, including those exhibiting identical patterns for phenotypic markers. Genetic methods have been improved with respect to simplification and reliability, which makes them appropriate for routine use in clinical laboratories today. M. heckeshornense can easily be identified using molecular methods, such as RFLP-PCR or automated sequencing (1-2).
Richter and colleagues claim that the description of new species should be based on observation of at least 5 strains. What may not be generally understood is that the process of resolving questions of taxonomy is a complex one and goes through different stages in which description, classification, and identification are considered (1-3). Certainly, this approach, as it serves the needs of the medical microbiology community, must achieve a reasonable compromise between timeliness and breadth of coverage. It is nearly needless to mention that clinical meaningfulness, paralleled by the quest for precision, should be the key to timeliness (1-3). The purpose of our article, then, was to describe a new species in a timely fashion in view of the given clinical priority. If there is a need to pursue the problem of separating this species from M. xenopi because both fit an identical biochemical test pattern, we believe that it would be more fruitful to collect a large number of strains (definitely more than five) in due course and to submit these to a more comprehensive review using comparative analyses of well-described tests. Interestingly, we have recently retrospectively scanned our strain collection for atypical M. xenopi strains and have found four isolates with a negative and one with a very weak 10-day arylsulfatase activity test result (the results for testing of pyrazinamidase agar activity were variable, and testing of nicotinamidase activity was not done). All of these strains were identified as M. heckeshornense by means of genetic methods. It is worth brief mention that two of these strains were isolated from one patient with progressive cavitary lung disease and histologically confirmed epitheloid cell granulomatosis. He had been treated in 1996 and 1997 for suspected but never proven tuberculosis. The first isolate was obtained between the two treatment series in the mid-1990s, while the second one was subcultured from one of the very recent multiple smear-positive specimens submitted to our laboratory in April this year.
The comments on the missing citation of a previous entry in a sequence database are appreciated because they draw our attention to a technical problem that may flaw database searches. Of course, not mentioning this sequence was not a matter of citation ethics (as would be the case for citation amnesia or inertia); rather, it is explained by our choice of search parameters. As we found out from a query at the National Center for Biotechnology Information, the alignments attained after performing the BLAST search may have received a high Expect value due to the discrepancy in length between query (M. heckeshornense, 1.5 kbp) and hit (accession no. AF101243 gi:4092535, 249 bp [released in January 1999]). Therefore, the alignment did not meet the Expect value threshold for reporting output and was thus overlooked. Ultimately, 16S ribosomal DNA sequences with a length of at least 500 bp are regarded as a standard today, but workers dealing with ribosequencing should be aware of the need to either perform additional searches with smaller fragments of a long input sequence (preferably with those containing variable regions of interest) or adapt search parameters.
REFERENCES
-
1-1.Kent P T, Kubica G P. Public health mycobacteriology—a guide for the level III laboratory. U.S. Department of Health and Human Services Publication (CDC) 86-8230. Atlanta, Ga: Centers for Disease Control; 1985. [Google Scholar]
-
1-2.Roth A, Reischl U, Streubel A, Naumann L, Kroppenstedt R M, Habicht M, Fischer M, Mauch H. Novel diagnostic algorithm for identification of mycobacteria using genus-specific amplification of the 16S–23S rRNA gene spacer and restriction endonucleases. J Clin Microbiol. 2000;38:1094–1104. doi: 10.1128/jcm.38.3.1094-1104.2000. [DOI] [PMC free article] [PubMed] [Google Scholar]
-
1-3.Telenti A. More on ‘What’s in a name … '—pragmatism in mycobacterial taxonomy. Int J Tuberc Lung Dis. 1998;2:182–183. [PubMed] [Google Scholar]
-
1-4.Wayne L G. Mycobacterial speciation. In: Kubica G P, Wayne L G, editors. The mycobacteria: a sourcebook. Microbiology series. Vol. 15. New York, N.Y: Marcel Dekker, Inc; 1984. pp. 25–65. [Google Scholar]
