The Genera Chromobacterium and Janthinobacterium

Purple-pigmented bacteria have been described since the end of the 19th century; they were reported as discoloring a variety of natural materials and occasionally as the causative agent of septicemia in humans and animals. Bacteria producing purple and violet colonies due to the production of a nondiffusible pigment, violacein, were classified in a redefined genus Chromobacterium by Buchanan (1918). The structure of violacein, an indole derivative produced via the oxidation of tryptophan, is shown in Fig. 1. It can be easily identified (Johnson and Beer, 1971) spectrophotometrically from the following properties: 1) in ethanolic solution it has an absorption maximum at 579 nm and a minimum at 430 nm; 2) by adding 10% (v/v) H₂SO₄ the solution turns green with an absorption maximum at 700 nm; and 3) when NaOH is added to an ethanolic solution, the solution turns green and afterwards reddish brown. Recently, violacein has also been characterized by nuclear magnetic resonance spectroscopy and mass spectrometry (Riveros et al., 1988). The pigment is only abundantly produced when tryptophan is available in the culture medium.

Although violacein is produced by only a few groups of bacteria, its presence does not necessarily indicate a close relationship between these organisms. An extensive phenotypic study (Sneath, 1956, 1960, 1974) within Chromobacterium, as defined by Buchanan (1918), provided evidence of two groups within this genus: a fermentative and mesophilic one (growth at 37°C but not at 4°C) and a nonfermentative and psychrophilic one (growth at 4°C but not at 37°C), for which he proposed two species, Chromobacterium violaceum and [Chromobacterium] lividum, respectively. However, these two species have been reported as being not closely related (Moffet and Colwell, 1968; Sneath, 1974) and since the latter was misnamed, in this chapter its epithet and all other misnamed epithets shall be enclosed in square brackets.

Violacein is not a good taxonomic marker as shown by the fact that not only have nonpigmented Chromobacterium violaceum and [Chromobacterium] lividum strains been isolated, but also since marine, violacein-producing strains were isolated, which were assigned on phenotypic grounds to the genus Alteromonas (Gauthier, 1976, 1982). Later, Van Landschoot and De Ley (1983) confirmed that these marine strains were genuine Alteromonas.

The relationships between the violacein-producing bacteria classified in Chromobacterium (Sneath, 1974) could finally be elucidated by De Ley et al. (1978). DNA-rRNA hybridizations within and between the violaceum and the lividum taxons and with different other Gram-negative genera demonstrated clearly that these purple-pigmented bacteria represent two tight but separate groups within rRNA superfamily III (see also De Ley’s introduction to the Proteobacteria in The Proteobacteria: Ribosomal RNA Cistron Similarities and Bacterial Taxonomy in the second edition) which are less related to each other than to other genera from this rRNA superfamily. In consequence, a new genus Janthinobacterium with one species J. lividum was created for the lividum taxon and the genus name Chromobacterium was restricted to the mesophilic, fermentative violaceum taxon (De Ley et al., 1978). Chromobacterium constitutes a separate rRNA branch on which all violaceum strains cluster with a T_m(e) of 76.5 to 80°C. Janthinobacterium constitutes a separate rRNA subbranch which is closely related to the rRNA subbranch containing [Pseudomonas] rubrisubalbicans and a yet unnamed group of bacteria isolated from clinical material (see Fig. 6 in The Proteobacteria: Ribosomal RNA Cistron Similarities and Bacterial Taxonomy in the second edition). The Janthinobacterium-[Pseudomonas] rubrisubalbicans rRNA branch is more closely related to the solanacearum rRNA branch than to Chromobacterium, emphasizing the deep phylogenetic gap between Chromobacterium and Janthinobacterium. Cataloging results (Woese et al., 1984) confirm the separate position of both genera in the beta group (Stackebrandt et al., 1988). In this chapter, Chromobacterium and Janthinobacterium are treated together merely on historical grounds.

Moss et al. 1978 described a new group of violacein-containing, fermentative, psychrophilic, fresh water bacteria and assigned them to a new species fluviatile in the genus Chromobacterium. It differs from C. violaceum by a thin flat, spreading colony type, some other phenotypic features (see below) and % GC values of 50 to 52. Within C. violaceum the GC content varies from 65 to 68 mol%, indicating at once that these new organisms cannot be members of Chromobacterium. DNA-rRNA hybridizations (Moss and Bryant, 1982) demonstrated that [Chromobacterium]*fluviatile is not closely related to the C. violaceum rRNA cluster. With a T_m(e) of 70 to 71°C vs. the C. violaceum rRNA probe, they occupy a separate position and are further removed from C. violaceum than the latter species is from the Neisseriaceae (Rossau et al., 1989). Comparison of their 16S rRNA sequences corroborates this conclusion (Dewhirst et al., 1989). Therefore, these organisms cannot belong in Chromobacterium; they have thus been misnamed and shall not be discussed here further. The differentiation of these three groups of violacein-producing organisms is given in Table 1.

Table 1. Characteristics differentiating the groups of violacein-producing organisms.