Update on the genus Robertmurraya: a bacterial genus honoring Dr. Robert G.E. Murray (with some personal reminiscences)

The genus Robertmurraya was created by my group in 2020 during the reclassification of species from highly polyphyletic genus Bacillus into multiple novel genera (Gupta et al. 2020; Patel and Gupta 2020). Members of this genus branch distinctly from the emended genus Bacillus and all other Bacillales genera in different phylogenetic trees. In addition, the demarcation of Robertmurraya as a distinct genus is supported by several newly identified molecular synapomorphies, consisting of conserved signature indels (CSIs) in protein sequences, which are uniquely found in all genome-sequenced species from this genus (Gupta et al. 2020). The genus Robertmurraya is composed of Gram-positive, rod-shaped, endospore-forming, and motile bacteria (Gupta et al. 2020). The description of this genus stated that it is “named after the Canadian microbiologist Dr. Robert George Everitt Murray, University of Western Ontario, for his contributions and leadership in the field of bacterial taxonomy” (Gupta et al. 2020).

The genus Robertmurraya presently contains seven species with validly published names, i.e., R. siralis (type species), R. andreesenii, R. beringensis, R. crescens, R. korlensis, R. kyonggiensis, and R. massiliosenegalensis (Parte 2018). Genome sequences are available for four of these species (R. siralis, R. korlensis, R. kyonggiensis, and R. massiliosenegalensis). As noted above, species from the genus Robertmurraya can be distinguished from other bacteria based on several rare genetic changes consisting of CSIs in protein sequences. One example of a CSI specific for this genus is shown in Fig. 1A, where a 1aa deletion in a conserved region of a protein of unknown function is commonly shared by all four genome-sequenced Robertmurraya species. Interestingly, in addition to these Robertmurraya species, this CSI is also present in two Bacillus species with effectively published names (“B. yapensis” and “B. dakarensis”) (Sarr et al. 2020; Xu et al. 2020) and one uncharacterized Bacillus strain (Bacillus sp. Y1). Earlier work on CSIs demonstrates that these molecular characteristics exhibit high degree of predictive ability to be found in other members of specific taxa (Bhandari et al. 2013; Barbour et al. 2017; Gupta et al. 2018a, 2020; Rudra and Gupta 2021). Based on this property of the CSIs, we have recently developed a web-based tool/server (AppIndels.com) that, using the information regarding the presence of known CSIs in a genome sequence, can predict its taxonomic affiliation (Gupta and Kanter-Eivin 2023). In our recent work, this server correctly predicted the taxonomic affiliation of >650 uncharacterized Bacillus spp. into 29 different genera/families (Gupta and Kanter-Eivin 2023). Of these analyzed strains, Bacillus sp. Y1 was predicted to belong to the genus Robertmurraya. We have also analyzed the genome sequences of “B. yapensis” and “B. dakarensis” using the AppIndels server. Results from the server show that the genome sequences of both these species also contain all three CSIs specific for the genus Robertmurraya (results not shown); thus, they are predicted to belong to this genus.

To confirm the inferences from the AppIndels server regarding the taxonomic affiliation of “B. yapensis”, “B. dakarensis”, and Bacillus sp. Y1, phylogenetic trees were constructed for these species/strains and other Robertmurraya species based on concatenated sequences of 87 proteins that are conserved among Bacillota (Wu et al. 2013) (Fig. 1B), and 16S rRNA gene sequences (Fig. 1C). In both these trees, all three of these species/strains reliably grouped within the clade consisting of other Robertmurraya species. Based on the results from phylogenetic analyses and the shared presence of Robertmurraya-specific CSIs in these species/strains, it can be reliably inferred that the species “B. yapensis” and “B. dakarensis” are part of the genus Robertmurraya and thus should accordingly be reclassified. Neither of these species have validly published names, and for “B. dakarensis”, a type strain is not available from two culture collections. However, based upon the available information, I am proposing “Bacillus yapensis” as a novel species Robertmurraya yapensis sp. nov. within the genus Robertmurraya. A new name is also proposed for “Bacillus dakarensis” so that it should be recognized as “Robertmurraya dakarensis.” This name could be validated once deposits of the proposed type strain in two publicly accessible culture collections have been made.

Robertmurraya yapensis (yap’ensis. N.L. masc. adj. yapensis, pertaining to Yap trench, which is the geographical position where the first isolate of this species was obtained).

The description of this species is essentially the same as reported by Xu et al. (2020) for B. yapensis. Cells are Gram-positive, motile, endospore-forming, and rod-shaped. Growth occurs in a medium with 0%–6% NaCl (optimum 0%–0.5%), 10–45 °C (optimum 37 °C), and pH 6–9 (optimum pH 7.0), at 0.1–50 MPa (optimum 0.1 MPa). The DNA G + C content of the type strain is 38.2 mol%. The type strain (XXST-01^T) was isolated from a deep-sea sediment sample at 6300 m of the Yap Trench, Pacific Ocean. Other biochemical and chemotaxonomic characteristics of this species are as described by Xu et al. (2020). The type strain from this species branches with other Robertmurraya species in 16S rRNA gene tree and in a tree based on concatenated sequences for multiple conserved proteins. Additionally, genome sequence of the type strain of this species also contains three CSIs found in the following proteins (sugar-binding transcriptional regulator, hypothetical protein (accession number WP_066048120), and asparagine synthase (glutamine-hydrolyzing)), which except for isolated exceptions are distinctive characteristics of the genus Robertmurraya (Gupta et al. 2020).

The GenBank accession number of the 16S rRNA gene sequence of strain XXST-01T is MK243676, and its draft genome sequence accession number is PRJNA533716.

The type strain is XXST-01^T (=MCCC 1A14143^T = JCM33181^T).

“Robertmurraya dakarensis” (da.ka.ren’sis. N.L. masc. adj. dakarensis of Dakar, the name of the capital of Senegal where the stool sample was collected).

The description of this species is as provided by Sarr et al. (2020) for “Bacillus dakarensis”.

The type strain is Marseille-P3515T (=CSURP3515).

As this paper is part of a tribute to Dr. Robert Murray, it would be appropriate to briefly describe my interactions with him. I first became acquainted with Dr. Robert Murray (Bob Murray) in an unusual manner. In 1996, my lab submitted a manuscript to Journal of Bacteriology describing our work on cloning of the gene for Hsp70 (DnaK) protein from Deinococcus proteolyticus and Thermomicrobium roseum and conducting evolutionary relationships based on this protein for prokaryotic organisms (Gupta et al. 1997). The manuscript also described several conserved indels shared by specific groups of organisms. The comments received on the manuscript from one of the reviewers were particularly interesting as they focused on the importance of molecular synapomorphies in understanding ancient evolutionary relationships. The use of the term “synapomorphy” to refer to the conserved indels was new to me and thus the reviewer’s comments were especially interesting. Hence, I wrote to the Reviewing Editor of this manuscript, the late Dr. Terry Beveridge, whether he could reveal the identity of this reviewer. As expected, Dr. Beveridge indicated that he could not do so without the consent of the reviewer but that he will check with the reviewer. In a few days, I received a postcard from Dr. Robert G.E. Murray indicating that he enjoyed reviewing this manuscript and he asked that I keep him informed of my future work along these lines. As my work prior to mid-90s was not in the areas of microbial evolution or systematics, I was unaware of the earlier important contributions that Dr. Murray had made to these fields (Murray 1984, 1986, 1988, 1992; Stackebrandt et al. 1988; Murray et al. 1990, 1999; Murray and Stackebrandt 1995). Nonetheless, my serendipitous introduction to Dr. Murray (Bob) led to my extensive interactions with him, mainly through letters and emails, which lasted for more than 10 years. Most of these correspondences focused on the evolutionary significances of multiple CSIs identified by our work, found in important conserved proteins, which demarcated key branch points in bacterial evolution. His response to one of my reviews/articles is noted in Fig. 2A (Gupta 1998). Based on the CSIs identified in our work, it was possible to delineate the branching order of the main bacterial phyla, which was not possible based on phylogenetic trees using 16S rRNA or other markers (Olsen and Woese 1993; Gupta 1998, 2001).

Bob recognized and appreciated the significance of the identified CSIs in understanding bacterial evolution and became a strong supporter of our work. In 1999, Bob brought to my attention an article published by Ludwig and Schleifer in the ASM News (Ludwig and Schleifer 1999), where these authors emphasized that the 16S rRNA-based phylogenies provide the best available method for understanding the evolutionary relationships among Bacteria and downplayed the inferences based on other conserved molecular characteristics, which were inconsistent with the 16S rRNA trees. In response, I published a letter in the ASM News noting several limitations of the 16S rRNA-based phylogenies, particularly regarding understanding the branching order of bacteria phyla, and also the need for more reliable means for demarcation of different taxa (Gupta 2000). I also indicated how rare genetic changes such as the CSIs provide novel means to bridge these shortcomings. Dr. Murray, on his own, also wrote a letter to the ASM News indicating that the article by Ludwig and Schleifer paid minimal attention to approaches other than those based on 16S rRNA sequences (referring specifically to our work on CSIs), which were providing convincing and thought-stimulating conclusions based on characteristics in highly conserved proteins (Murray 2000). He ended this letter by stating “Research in evolutionary biology is certainly opening our eyes to exciting facts of biology and a new aspect of taxonomy, but we have to beware of excluding well-founded novelties that are counter to current and strongly held interpretations. It is worth remembering that science is an ever-moving set of approximations to the truth we strive to discover” (Murray 2000).

Dr. Murray also discussed our work on the evolutionary significances of the identified CSIs with Dr. Peter H. Sneath, who was his close family friend and similar to him had made major contributions towards the advancement of the field of prokaryotic taxonomy (Sneath and Sokal 1973; Skerman et al. 1980; Murray 1986; Lapage et al. 1992). Dr. Sneath was a cofounder of the field of numerical taxonomy, which uses shared presence of specific characteristics to infer evolutionary relationships (Sneath and Sokal 1973; Sneath 2001). Bob through his letter to me (Fig. 2B) suggested that “(Dr. Peter Sneath) is your best bet as a solid proponent of the indel theory in BMT (Bergey’s Manual Trust).” Following Bob’s advice, I sent some reprints of our work to Dr. Sneath. Dr. Sneath’s (Peter) response was exceedingly positive (Fig. 2C), and it formed the basis of our working together on determining the applicability of the character compatibility approach to molecular sequence data. Peter informed me that the inferences that I had drawn based upon the identified CSI were supported by his analysis using the compatibility approach. To extend this work further, I stopped over in London in 2004 to visit Dr. Sneath. Peter picked me up from Leicester train station and we spent several hours at his residence discussing the work (grateful for the warm hospitality extended by his wife Joan). Peter shared with me the results of his analyses and explained the working of the character compatibility approach. We agreed that for a more effective demonstration of this approach to molecular sequence data (which until that time was not done), a larger dataset was needed than provided by the sequences of then available CSIs. We decided to examine the branching order of Proteobacterial subdivisions (classes) using sequence data for several conserved proteins. The results of these studies demonstrating the usefulness of the character compatibility approach to generalized molecular data were published in 2007 in Journal of Molecular Evolution (Gupta and Sneath 2007). This paper constituted the last scientific contribution of Dr. Peter Sneath, and this collaboration was made possible by Dr. Robert E. Murray.

Lastly, it should be mentioned that I met Bob only on two occasions, once at the University of Guelph in 1997 during a visit by the late Dr. Lynn Margulis (whom I knew well) and another time in 1998, when he invited me to give a seminar at the University of Western Ontario. However, despite my limited in-person interactions with Bob, I felt a close bond with him. I believe this feeling to some extent was mutual. In one of his last emails to me (5 January 2008), in response to my New Year’s greetings, he wrote

I am thankful to Bob for his strong support and appreciation for our work. My naming the genus Robertmurraya, honoring his contributions to the field of Microbial Systematics, is a small token of thanking him and paying him a personal tribute. Although I was unable to contact Bob to obtain his approval for the naming of this genus, I know he would have appreciated it. He would have also been pleased with the progress that we have made in using the “molecular synapomorphies” for clarifying the evolutionary relationships and classification of several important taxa of prokaryotes (Gupta 2016; Gupta et al. 2018a, 2018b, 2020; Patel and Gupta 2020; Saini and Gupta 2021; Gupta and Kanter-Eivin 2023).

I thank Bashudev Rudra for assistance in the construction of 16S rRNA gene tree and Dr. Herb Schellhorn for comments on the manuscript.