Letters to the Editor

Huber, B. E., Escudero, R., Busse, H. J., systematics. Proc Natl Acad Sci U S A 105, tularensis, a facultative intracellular pathogen. Seibold, E., Scholz, H. C., Anda, P., Ka¨ mpfer, P. 2504–2509. PLoS Pathog 5, e1000472. & Splettstoesser, W. D. (2010). Description of Lapage, S. P., Sneath, P. H. A., Lessel, E. F., Olsufiev, N. G., Emelyanova, O. S. & Dunayeva, Francisella hispaniensis sp. nov., isolated from Skerman, V. B. D., Seeliger, H. P. R. & Clark, T. N. (1959). Comparative study of strains of B. human blood, reclassification of Francisella W. A. (editors) (1992). International Code of tularense in the old and new world and their novicida (Larson et al. 1955) Olsufiev et al. 1959 Nomenclature of Bacteria (1990 Revision). taxonomy. J Hyg Epidemiol Microbiol Immunol as Francisella tularensis subsp. novicida comb. Bacteriological Code. Washington, DC: American 3, 138–149. nov., and emended description of the genus Society for Microbiology. Francisella. Int J Syst Evol Microbiol 60, 1887– Skerman, V. B. D., McGowan, V. & Sneath, 1896. Larson, C. L., Wicht, W. & Jellison, W. L. (1955). P. H. A. (editors) (1980). Approved lists of A new organism resembling P. tularensis isolated bacterial names. Int J Syst Bacteriol 30, 225–420. Koeppel, A., Perry, E. B., Sikorski, J., Krizanc, D., Warner, A., Ward, D. M., Rooney, from water. Public Health Rep 70, 253–258. Tindall, B. J., Ka¨ mpfer, P., Euze´ by, J. P. & A. P., Brambilla, E., Connor, N. & other authors Larsson, P., Elfsmark, D., Svensson, K., Oren, A. (2006). Valid publication of names of (2008). Identifying the fundamental units of Wikstro¨ m, P., Forsman, M., Brettin, T., Keim, P. prokaryotes according to the rules of bacterial diversity: a paradigm shift to & Johansson, A. (2009). Molecular evolutionary nomenclature: past history and current practice. incorporate ecology into bacterial consequences of niche restriction in Francisella Int J Syst Evol Microbiol 56, 2715–2720.

Objections to the transfer of Francisella novicida to the subspecies rank of Francisella tularensis – response to Johansson et al. The description of novel species requires experiments, F. novicida is genetically close the results from the literature and the the careful selection and use of a wide to F. tularensis (Hollis et al., 1989) and the results from our investigations, but also for variety of methodologies. As pointed out phenotypic differences observed (Huber et sake of consistency, it is obvious that our by Tindall et al. (2010), experience gained al., 2010) are in agreement with the proposal to assign F. novicida to F. over the past six decades has continued to subspecies concept. Another important tularensis as a novel subspecies is well demonstrate the value of comparing point supporting this taxonomic supported. different datasets and also of basing the rearrangement is the acceptance of the new Below are some additional replies to description and delineation of taxa on as combination within the scientific certain arguments proposed by Johansson wide a dataset as possible. A combination community. The use of this not yet validly et al. (2010) to support their stance against of data acquired from DNA-based published new combination may be related the reclassification of F. novicida. methods (DNA–DNA hybridization, gene to the fact that in Bergey’s Manual of sequences, genomic fingerprints) and Systematic Bacteriology (often erroneously It is argued, that: phenotyping (chemotaxonomic, considered as the ‘bible’ of bacterial ‘From a practical standpoint, sep- physiological and morphological traits) systematics by those interested in bacterial arate species names are useful in a provides a sound basis for the taxonomy of taxonomy), the transfer of F. novicida to microbiological laboratory or a the prokaryotes (Tindall et al., 2010). The Francisella tularensis subsp. novicida was clinical setting and also as a basis decision as to whether two bacteria are recommended in the chapter dealing with for regulations governing the hand- members of a single species is still based on the genus Francisella (Sjo¨stedt, 2005). ling of medically important organ- the results from DNA–DNA hybridizations Although this proposal was never formally isms. [...] Importantly, it is fairly (Wayne et al., 1987; Stackebrandt et al., recognized, numerous microbiologists are easy to distinguish F. novicida and 2002). In general, two bacterial strains are already using the name. An online search F. tularensis on the basis of their assigned to the same species if their DNAs survey in ‘Pubmed’ (http://www.ncbi.nlm. different growth and metabolic reassociate at levels greater than 70 % and nih.gov/sites/entrez?db=PubMed) indi- requirements on artificial media’. 5 % or less DTm (Wayne et al., 1987), but cates that in recent years there is no the latter criterion is only rarely applied. In significant difference in the frequencies of In contrast to tularaemia caused by F. addition, Wayne et al. (1987) pointed out the use of the names F. novicida and F. tularensis subsp. tularensis or F. tular- ‘Subspecies designations can be used for tularensis subsp. novicida. ensis subsp. holarctica, human or genetically close organisms that diverge in animal infections with strains of F. phenotype’. From our point of view, it is not consistent tularensis subsp. novicida are extre- to have a species F. tularensis with three mely rare and there are very few Our proposal to transfer Francisella subspecies supported by DNA–DNA publications reporting the isolation novicida as a novel subspecies to F. relatedness data but distinguishable by of this facultative pathogen. Most of tularensis subsp. novicida is in agreement phenotypic traits and a separate species F. these reports have shown that it was with the above-mentioned novicida that also shares high DNA–DNA very difficult to distinguish those recommendations. As demonstrated by the relatedness values (.85 %) but which is isolates from strains of F. tularensis, results from DNA–DNA reassociation phenotypically distinguishable. Based on not only for routine clinical laborat-

Downloaded from www.microbiologyresearch.org by 1718 International Journal of Systematic and Evolutionary Microbiology 60 IP: 54.70.40.11 On: Tue, 11 Dec 2018 06:22:18 Letters to the Editor ories, but also for a reference labor- four traits from one or the other be validly published according to the atory (Birdsell et al., 2009). We have subspecies (Huber et al., 2009). requirements of the Bacteriological the impression that the majority of Code. In fact the term ‘valid’ can only It is argued that: routine laboratories will not be able to be applied to the valid publication of distinguish between the two subspe- ‘In 1989, Hollis et al. (1989) showed the name. This does not mean that the cies on the basis of phenotypic char- that F. novicida and F. tularensis scientific conclusions of the paper of acteristics or metabolic requirements. could be considered one species as Hollis et al. (1989) are incorrect. judged from DNA–DNA hybridiza- It is also difficult to understand why It is argued that: tion experiments (Hollis et al., only separate species are useful in a 1989). Their publication was not ‘Notably, earlier publications con- microbiological laboratory or a clinical valid according to the requirements sidered F. novicida and F. tularensis setting but not separate subspecies. outlined in the Bacteriological Code to be separate species based on The same is true for the basis of (Lapage et al., 1992; Tindall et al., differences in phenotype including regulations governing the handling of 2006). As a result, the proposed chemotaxonomic markers, distinct medically important organisms. In fact, elimination of the species F. novicida ecological roles, different clinical there is a clear example showing that and the demotion of it to a biogroup and epidemiological characteris- subspecies differentiation can be of F. tularensis was not included tics, and differing abilities and applied for different classification of among prokaryotic names with modes of invasion and mechan- medically important bacteria. Accord- standing in nomenclature’. isms of tissue damage in mammals ing to EU guideline 2000/54/EC, which In the paper of Hollis et al. (1989), the (Larson et al., 1955; Olsufiev et al., had to be translated into national law authors presented data on which they 1959; Skerman et al., 1980)’. in 27 European countries, F. tularensis proposed the transfer of Yersinia Microbiologists, including bacterial subsp. tularensis [Francisella tularensis philomiragia to the genus Francisella taxonomists, have always made use of (Type A)] is classified as a BSL 3 as Francisella philomiragia comb. nov. the tools that are available to them at pathogen, while F. tularensis subsp. Furthermore, based on results from the time. Unfortunately, the paper of holarctica [Francisella tularensis (Type DNA–DNA hybridizations (.85 % Olsufiev et al. (1959) was not accessible B)] is classified as a BSL 2 pathogen reassociation) they also proposed to to us, but neither Larson et al. (1955) (European Communities, 2000; EU transfer Francisella novicida to the nor Skerman et al. (1980) mention any directive 2000/54/EC). species Francisella tularensis as the chemotaxonomic markers that they had It is argued, that: third biogroup Francisella tularensis examined. To the best of our know- biogroup novicida in addition to the ledge, studies in bacterial taxonomy in ‘Indeed,inTable2ofHuberet al. biogroups tularensis and palaeartica. It the 1950s were restricted to ecological, (2010) data are provided that con- is not clear why Hollis et al. (1989) did physiological, morphological and sero- tradict their own proposal by pre- not apply the taxonomy proposed by logical examinations. Analyses of low- senting 11 metabolic reactions that Olsufjev & Meshcheryakova (1983), molecular mass cell components such are distinct between F. novicida and who had subdivided F. tularensis into as fatty acids, polar lipids, quinones F. tularensis (Huber et al.,2010)’. three subspecies, Francisella tularensis (chemotaxonomic markers) or DNA- It is common knowledge that a subsp. tularensis, Francisella tularensis based studies were not yet established, bacterial species is composed of indi- subsp. holarctica and Francisella tular- and hence it also appears likely that viduals that are not necessarily ident- ensis subsp. mediasiatica, or why they Olsufiev et al. (1959) did not examine ical in their phenotypic traits. For did not refer to the paper of Olsufjev chemotaxonomic markers. In the fol- instance, in the entry for Pseudomonas & Meshcheryakova (1983). In lowing decades, with the introduction fluorescens in Bergey’s Manual of Validation List No. 32 (Hollis et al., of novel tools, new knowledge has been Systematic Bacteriology (Palleroni, 1990), the name Francisella philomir- gained which has led to numerous 2005), among 139 traits listed, fewer agia comb. nov. was validly published reclassifications of established taxa. than 50 % (62 traits) are identical whereas the name F. tularensis For instance, species assigned to the among the different biovars of this biogroup novicida was not. This was genus Pseudomonas in Bergey’s Manual species and 18 traits are listed in which because the Bacteriological Code does of Systematic Bacteriology (Palleroni, biovar I strains are variable. Hence, we not cover infraspecific names (e.g. 1984) have since been shown to be biogroups, biovars, pathovars etc.) do not think that 11 traits distinguish- distributed over the three classes other than subspecies names. ing F. tularensis subsp. novicida among Alphaproteobacteria, Betaproteobacteria 98 traits tested contradict our pro- Hence, there is no scientific argument and Gammaproteobacteria. As a result, posal. In this context, it has also to be to state that the paper of Hollis et al. species phylogenetically only distantly mentioned here that the type strain of (1989) was not valid according to the related to the genus Pseudomonas have Francisella tularensis subsp. holarctica requirements of the Bacteriological been reclassified as species of other differs from the other two subspecies Code. Only the proposed name F. genera. A similar fate has befallen in seven traits and in an additional tularensis biogroup novicida could not numerous other taxa. Hence, there is

Downloaded from www.microbiologyresearch.org by http://ijs.sgmjournals.org 1719 IP: 54.70.40.11 On: Tue, 11 Dec 2018 06:22:18 Letters to the Editor no reason why taxonomic conclusions Holger C. Scholz,3 Erik Seibold,3 Wolf C., Nano, F. E. & other authors (2010). concerning the genus Francisella pro- D. Splettstoesser3 and Peter Ka¨ mpfer4 Objections to the transfer of Francisella novicida posed decades ago should remain to the subspecies rank of Francisella tularensis. 1 Int J Syst Evol Microbiol 60, 1717–1718. untouchable when new findings sug- Institut fu¨r Bakteriologie, Mykologie und gesting reclassification are reported. Hygiene, Veterina¨rmedizinische Larson, C. L., Wicht, W. & Jellison, W. L. (1955). Universita¨t, A-1210 Wien, Austria A new organism resembling P. tularensis isolated from water. Public Health Rep 70, 253–258. We agree with the authors of the letter 2Laboratorio de Espiroquetas y that genomic data can provide an Pato´genos Especiales, Centro Nacional Olsufiev, N. G., Emelyanova, O. S. & Dunayeva, T. N. (1959). enormous amount of information. de Microbiologı´a Instituto de Salud Comparative study of strains of B. tularense in the old and new world and their We think that bacterial taxonomy Carlos III, Madrid, Spain may undergo significant changes on taxonomy. J Hyg Epidemiol Microbiol Immunol 3 3, 138–149. the basis of the understanding of Bundeswehr Institute of Microbiology, genomic and proteomic information. German Reference Laboratory for Olsufjev, N. G. & Meshcheryakova, I. S. (1983). Tularemia, Neuherbergstr. 11, D-80937 However, for the time being, we are Subspecific taxonomy of Francisella tularensis Munich, Germany McCoy and Chapin 1912. Int J Syst Bacteriol 33, far from able to interpret this informa- 872–874. tion well enough to draw decisive 4Institut fu¨r Angewandte Mikrobiologie, Justus-Liebig-Universita¨t Giessen, Palleroni, N. J. (1984). Genus 1. Pseudomonas conclusions. There are numerous open AL D-35392 Giessen, Germany Migula 1894 237 . In Bergey’s Manual of questions, e.g. ‘Which genes belong to Systematic Bacteriology, vol. 1, pp. 141–199. the conserved genome core considered Correspondence: Hans-Ju¨rgen Busse Edited by N. R. Krieg & J. G. Holt. Baltimore: useful to define a taxon and which ([email protected]) Williams & Wilkins. belong to accessory dispensible genetic Palleroni, N. J. (2005). Genus I. Pseudomonas elements?’ The impact of processes Birdsell, D. N., Stewart, T., Vogler, A. J., Migula 1894 237AL.InBergey’s Manual of such as lateral gene transfer, gene Lawaczeck, E., Diggs, A., Sylvester, T. L., Systematic Bacteriology, 2nd edn, vol. 2, part B, duplication, recombination and rear- Buchhagen, J. L., Auerbach, R. K., Keim, P. & pp. 323–379. Edited by D. J. Brenner, N. R. Wagner, D. M. (2009). Francisella tularensis rangements of genes in the genome is Krieg, J. T. Staley & G. M. Garrity. New York: subsp. novicida isolated from a human in Springer. not clear and it may be very different Arizona. BMC Res Notes 2, 223. in different lineages! For instance, the Sjo¨ stedt, A. B. (2005). Genus I. Francisella European Communities (2000). Directive 2000/ AL sizes of Escherichia coli genomes range Dorofe9ev 1947, 176 .InBergey’s Manual of 54/EC of The European Parliament and of Systematic Bacteriology, 2nd edn, vol. 2, part B, from 4.6 to 5.7 Mb. This observation The Council. Off J Eur Communities L262, 21– pp. 200–210. Edited by D. J. Brenner, N. R. 45. indicates that strains of a single species Krieg, J. T. Staley & G. M. Garrity. New York: may lose or acquire 20–25 % of their Hollis, D. G., Weaver, R. E., Steigerwalt, A. G., Springer. Wenger, J. D., Moss, C. W. & Brenner, D. J. genome from other sources without Skerman, V. B. D., McGowan, V. & Sneath, (1989). Francisella philomiragia comb. nov. affecting their taxonomic status. So, P. H. A. (editors) (1980). Approved lists of (formerly Yersinia philomiragia) and Francisella for the time being, it is more appro- bacterial names. Int J Syst Bacteriol 30, 225– tularensis biogroup novicida (formerly 420. priate to rely on the generally applied Francisella novicida) associated with human approaches for delineating taxa includ- disease. J Clin Microbiol 27, 1601–1608. Stackebrandt, E., Frederiksen, W., Garrity, G. M., ing species and subspecies (for a review Hollis, D. G., Weaver, R. E., Steigerwalt, A. G., Grimont, P. A. D., Ka¨ mpfer, P., Maiden, M. C. J., see Tindall et al., 2010). Reclassification Wenger, J. D., Moss, C. W. & Brenner, D. J. Nesme, X., Rossello-Mora, R., Swings, J. & other of F. novicida as F. tularensis subsp. (1990). Francisella philomiragia comb. nov. In authors (2002). Report of the ad hoc committee novicida is in agreement with these Validation of the Publication of New Names and for the re-evaluation of the species definition in recommendations and also avoids con- New Combinations Previously Effectively bacteriology. Int J Syst Evol Microbiol 52, 1043– 1047. fusion among those interested in bac- Published Outside the IJSB, List no. 32. Int J Syst Bacteriol 40, 105–106. terial taxonomy. Adhering to the Tindall, B. J., Rossello´ -Mo´ ra, R., Busse, H.-J., Ludwig, W. & Ka¨ mpfer, P. (2010). species F. novicida might raise the Huber, B. E., Escudero, R., Busse, H.-J., Notes on the Seibold, E., Scholz, H. C., Anda, P., Ka¨ mpfer, P. characterization of prokaryote strains for question of why the other subspecies & Splettstoesser, W. D. (2010). Description of taxonomic purposes. Int J Syst Evol Microbiol 60, of F. tularensis have not been elevated Francisella hispaniensis sp. nov., isolated from 249–266. to species level, though at least some human blood, reclassification of Francisella Wayne, L. G., Brenner, D. J., Colwell, R. R., arguments for the species F. novicida novicida (Larson et al. 1955) Olsufiev et al. 1959 Grimont, P. A. D., Kandler, O., Krichevsky, M. I., could also apply to the three recognized as Francisella tularensis subsp. novicida comb. Moore, L. H., Moore, W. E. C., Murray, R. G. E. & subspecies of F. tularensis. nov., and emended description of the genus other authors (1987). International Committee Francisella. Int J Syst Evol Microbiol 60, 1887– on Systematic Bacteriology. Report of the ad hoc 1896. committee on reconciliation of approaches to 1 1 Hans-Ju¨ rgen Busse, Birgit Huber, Johansson, A., Celli, J., Conlan, W., Elkins, K. L., bacterial systematics. Int J Syst Bacteriol 37, 463– Pedro Anda,2 Raquel Escudero,2 Forsman, M., Keim, P. S., Larsson, P., Manoil, 464.

Downloaded from www.microbiologyresearch.org by 1720 International Journal of Systematic and Evolutionary Microbiology 60 IP: 54.70.40.11 On: Tue, 11 Dec 2018 06:22:18