Objective

Comparative genomic analysis is a valuable tool to explore new genomes and metagenomes to search for functional processes, phylogenetic relationships, and evolutionary traits among organisms. An example of this comparative analysis is finding that proteins involved in the Spo0 pathway constitute of at least a sensor kinase, the phosphotransferases Spo0F and Spo0B, and the response regulator Spo0A [1], were suggested to be ancestral since it was found not only in Bacilli but also in some Clostridia [2]. While an original article from our group that describes the distribution and prevalence of the Spo0 pathway and also confirms that this pathway is ancestral was under review, the results mentioned above were published by Davidson and collaborators using a set 84 Firmicutes [2]. Therefore, in this note, we provide a more extensive list of manually curated Firmicutes possessing different forms of the Spo0 pathway initiating sporulation in Firmicutes.

Data description

The data are a collection of predicted protein architectures defining the proteins shaping the Spo0 pathway in Firmicutes. As a guide, we used the proteins shaping the sporulation cascades in the model organisms shown in Data File 1 [3]. Using the well-curated profiles of the PfamA database [4], we constructed the architectures shown in Table 1, Data File 1 [3]. The architectures were then used to inspect the genomes of 325 Firmicutes extracted from the Kyoto Encyclopedia of Genes and Genomes (KEGG) genome database [5,6,7], using the hmmscan program from the HMMER suite [8]. We preserved those hits with an expectation value > 0.001. We discarded sequences that do not preserve the same kind of domains and with a length not longer than twice the length of the model. We used the genome neighborhood as a parameter of selection to discriminate homologous proteins not belonging to the Spo0 pathway. To this end, we inspected three genes upstream and downstream from the query gene. Frequently, neighbor genes encode proteins that were organized in a cluster of orthologous genes (COG) [9]. COGs were assigned by finding homologous proteins for each query and neighbors using a hidden Markov model (HMM) search using the hmmsearch program [8]. This HMM search process employs a previously constructed model set that represents each of the 4873 COGs [9, 10]. The genomes tested were annotated using Operon Mapper, which is able to classify genes into COGs [10, 11]. The list of the COGs assigned to the Spo0 proteins and their neighbors is available in Table 1, Data File 2 [12].

Table 1 Overview of data files/data sets

To detect orphan histidine kinases (HKo), defined as kinases not having as a neighboring response regulator [13], we performed a hmmscan using the profiles of the PfamA shown in Table 1, Data File 1 [3] and kept those hits with an expectation value > 0.001 that fulfill the architectures described in Table 1, Data File 1 [3]. As an additional condition, we discarded those sequences that did not have the same number of domains and with a length no more than twice the length of the model. As the architectures were constructed using proteins experimentally proven to participate in sporulation (Table 1, Data File 1) [3], we considered for this analysis the architecture found in Clostridium thermocellum, which is composed of a HisKA, an HATPase, and a response regulator domain (Table 1, Data File 1) [3].

Spo0 pathways found for each endospore former are shown in Table 1, Data File 3 [14]. These results show the architectures describing twelve HKos, three architectures representing transferases, and two architectures describing the response regulator Spo0A. The final table includes 185 Bacilli, 134 Clostridia, 4 Negativicutes, and 2 Erysipelotrichia species.

Limitations

The data presented in this note were not published as a research article since the main findings were published while our manuscript was under review. Nonetheless, the data are still important and useful since they are a collection of well-curated Firmicutes species for which a Spo0 pathway was present or absent.

  • Experimental evidence of a sporulation phenotype is not available for a vast number of strains since the conditions have not been identified.

  • New protein profiles should be constructed to identify novel orphan kinases.

  • A study should be performed in newly sequenced genomes to increase the collection; nonetheless, the groups presented in this note are probably enough to confirm the main conclusions found by both groups.