Skip to main content

Genomic characterization of bacteria from the ultra-oligotrophic Madison aquifer: insight into the archetypical LuxI/LuxR and identification of novel LuxR solos



To characterize the bacterial community of Wind Cave’s Madison aquifer through whole-genome sequencing, and to better understand the bacterial ecology by identifying genes involved in acyl-homoserine lactone (AHL) based quorum-sensing (QS) systems.


Genome-based taxonomic classification revealed the microbial richness present in the pristine Madison aquifer. The strains were found to span eleven genera and fourteen species, of which eight had uncertain taxonomic classifications. The genomes of strains SD129 and SD340 were found to contain the archetypical AHL QS system composed of two genes, luxI and luxR. Surprisingly, the genomes of strains SD115, SD129, SD274 and SD316 were found to contain one to three luxR orphans (solos). Strain SD129, besides possessing an archetypical AHL QS luxI-luxR pair, also contained two luxR solos, while strain SD316 contained three LuxR solos and no luxI-luxR pairs. The ligand-binding domain of two LuxR solos, one each from strains SD129 and SD316, were found to contain novel substitutions not previously reported, thus may represent two LuxR orphans that detection and response to unknown self-produced signal(s), or to signal(s) produced by other organisms.


Due to difficulties in access, the microbial life in subsurface aquifers are an under-explored area of microbiology [1]. A recent study has demonstrated that the Madison aquifer, accessed directly by travel through Wind Cave, Wind Cave National Park (WCNP), had a greater bacterial diversity compared to nearby wells that intersected the same aquifer [2]. This discrepancy was shown to be due to contamination of the well water by bacterial species from overlaying rock units, meaning that the microbiology of the aquifer itself could only be accurately assessed via the cave. Without the influence of the well-water microbiology, it was found that the microbiology of the ultra-oligotrophic Madison aquifer was more complex than previously anticipated [2].

Quorum sensing (QS) is a bacterial cell–cell signaling system that employs small compound signals and regulates group behaviors for bacterial-bacterial and bacterial-host interactions [3, 4]. In one QS system, bacteria produce and secrete signals, called acyl-homoserine lactones (AHLs), into the surrounding environment. A typical AHL-QS system contains a LuxI (the AHL signal synthase) and a LuxR (transcriptional regulator). These proteins are usually encoded adjacent to each other on the chromosome [5]. In addition to the canonical luxI/luxR pair, many bacteria also contain extra copies of luxR transcriptional regulators that are not proximal to any luxI synthase gene [5].

An unpaired luxR gene is termed a luxR solos/orphan and similarly encodes for QS LuxR-type transcriptional regulators consisting of a signal (ligand)-binding domain at the N terminus and a DNA-binding helix-turn-helix (HTH) domain at the C terminus [6,7,8]. Some solos respond to endogenously produced AHLs to expand their regulatory range. Others “eavesdrop” on other bacterial species, changing their gene expression in response to the foreign AHL signals. There are even examples of LuxR solos responding to other chemical signals entirely, including those produced by species in other kingdoms of life. Such a subfamily of LuxR solos has been identified in plant-associated bacteria (PAB), which respond to plant-produced signals, thus forming an interkingdom signaling circuits [9].

We recently described the whole-genome sequences (wgs) of eight Ensifer sp. isolated from two different caves including strain SD006, from the Madison aquifer of WCNP [1]. The genome of SD006 is 427,000 bp larger than the largest of the other seven Ensifer sp. isolated from a dry limestone surface of the Lechuguilla Cave in New Mexico [10]. We are not aware of other studies that report bacterial wgs obtained from a subterranean aquifer accessed by a cave with insights on AHL quorum sensing.

In this work, first we provide wgs, de novo genome assembly and annotation of fourteen diverse bacterial strains isolated from the Madison aquifer accessed via Wind Cave [2]. Second, we provide insight utilizing these wgs with various genome-mining and proteomic tools to resolve the questions of strain classification and identity of quorum-sensing genes of the AHL class, luxI and luxR homologs, using a systematic bioinformatic approach [5, 9, 10]; and finally, we identified seven new LuxR solos from four WCNP strains, SD115, SD129, SD274 and SD316.

Main text

Materials and methods

SD strains were isolated from calcite lake in Wind Cave, which represents the piezometric surface of the Madison aquifer where it is intersected by the cave at a depth of 200 m below the surface [2]. The strains were maintained on half-strength tryptic soy agar medium (Merck, Germany).

Genomic DNA was isolated from 2.0 ml of two-day-old broth cultures using Sigma-Aldrich DNA extraction kit according to the manufacturer’s recommendations. Then, 1 ng of DNA from each isolate as quantified using PicoGreen (ThermoFisher Scientific) was processed using the Nextera XT library prep kit (Illumina) followed by sequencing on the Illumina MiSeq (2 × 250 paired-end run configuration).

Adapter-trimmed paired-end reads were assembled de novo using Unicycler tool [11]. The whole genome assemblies were then uploaded to the antibiotics and secondary metabolite analysis shell (antiSMASH) [12], in order to predict and identify secondary metabolite biosynthetic pathways. The assemblies were also uploaded to JSpeciesWS for identification via Tetra correlation search in conjunction with ANIb [13]. Other genes of interest were searched for using tblastn multiple alignment, using reference proteins as query sequences [14]. Phylogenomic analysis was carried-out using PhyloPhlAn [15].

To test for quorum sensing cell–cell communication mechanism of the acyl-homoserine lactone (AHL) class, antiSMASH analysis [11] was performed on each of the SD series genomes to identify secondary metabolites which include luxI homologs that encode for the production of AHLs. Tblastn multiple sequence alignment was used to test each genome for the presence LuxR homologies, using query sequence AFP89744.1. Alignments with a MaxScore of 50 or greater were considered putative LuxR homologs.

Putative LuxI and LuxR homologs were first identified based on the presence of proteins containing the hidden markov model PF00765 and PF03472 respectively. Interproscan [15] was used to validate each of the HMM matches. Proteins matching with PF00765 (putative LuxIs) were checked for domains IPR001690 and IPR018311, while proteins matching with PF03472 (putative LuxRs) were checked for domains IPR016032, IPR005143, IPR000792, and IPR036388. These domains are present in nearly all functional LuxI and LuxR proteins respectively. All validated homologs were further scrutinized by aligning them with canonical LuxR and LuxI proteins respectively via ClustalOmega [16]. Certain residues in the alignment were compared against conserved sites identified [17] for further characterization of homology and functionality [18]. To determine the status of LuxR solos in the SD series strains, 10 kbp regions centered around each of the validated luxR homologs were analysed for the presence of luxI homologs and visualized by Easyfig [19].

Results and discussion

The genome sizes of the strains sequenced in this study range between 2.3 to 6.9 megabases with GC content and N50 values ranging from 36.10 to 73.22% and 22,000 to 1,041,000 bp, respectively (Table 1). To classify each strain, the 5S, 16S, 23 s rRNA gene sequences were extracted from each genome using BARRNAP ( and searched against the NCBI database using BLASTN. Species-level identification was also performed using JSpecies [13]. If the output of the BLASTN search corresponded to a species within the JSpecies or NCBI database then the genome in the database was used to calculate the ANI value. The taxonomy information is present in Table 1.

Table 1 Genome annotation information for the SD strains

ANI analysis and JSpecies package [13] were used to investigate the species circumscriptions of the fourteen SD strains (Table 1). An ANI value in the range of 95% to 96% is the accepted cut-off threshold for species-species delineation [13]. Only five of the fourteen genomes produced an ANI value at > 96%, those being strains SD018, SD090, SD226, SD274, SD316. Strains SD075 and SD083 had ANI values approximately 94 to 95%, setting these two strains in the transitionary zone [13]. The remaining seven SD strains (072, 088, 115, 129, 287, 291, and 340) produced ANI values spanning 71% to 89% within the different species zone (uncertain taxonomic status) putatively indicating that these strains could represent new species (Table 1). Ten monophyletic groups encompass the fourteen WCNP strains, of which three are located in the Firmicutes, four in the Actinobacteria and seven in the Proteobacteria phyla (Fig. 1). This genomic information warrants further re-classification investigations.

Fig. 1

Evolutionary placement of fourteen WCNP strains. The fourteen strains sequenced in this study are denoted by a red star next to the strain name. Phylogeny was based on PhyloPhlAN analysis of four hundred of conserved domains scattered throughout the genome sequences. These strains showed phylogenetic diversity, spanning Proteobacteria, Actinobacteria, and Firmicutes. An archaebacterial clade, Methanosarcina, was selected as an outgroup

Strain SD340, an Acidovorax species, was found to have an abnormality in one of its canonical LuxI/R QS systems. This abnormality is with regards to luxI homolog localized on contig 4, which was initially discounted due to the missing autoinducer synthesis conserved site, IPR018311. Further analysis indicated, however, that this LuxI could still be a functional autoinducer synthase. PFAM analysis identified the protein as being in the "autoinducer synthase family", achieving a bit score of 81.3 with e-value 6.0e-23. Furthermore, this protein, when aligned with the canonical LuxI proteins, demonstrated complete consensus with the conserved residues as described in Fuqua and Greenberg [20]. Further evidence provided by Lim et al. [21] confirmed the existence of functional LuxI proteins lacking the IPR018311 domain. Due to complete consensus of the conserved residues and validation in clinical isolate Pandoraea pnomenusa RB38 of the ppnI [21], we propose that the luxI on contig 4 of Acidovorax sp. strain SD340 is an authentic AHL synthase gene (Additional file 1). Further investigations into the AHL synthase activity encoded by this luxI are currently underway.

A total of seven luxR solos have been identified in SD115, SD129, SD274 and SD316 and their gene neighborhoods are shown (Fig. 2a, Additional file 2). The three conserved residues of the DNA-binding domains E178, L182 and G188 are conserved in all seven SD strain LuxR solo homologs (Fig. 2b). Alignment of the identified LuxR solo homologs from SD strains shows substitution in the LuxR homolog (vjbR) from SD316 (contig 2_994) in the highly conserved amino acids in the regulatory domains W57M and Y61W that is similarly reported in PAB LuxR solos (Fig. 2b). PAB LuxR solos e.g., NesR, XagR, OryR, PsoR and others (Fig. 2c) form a robust monophyletic group with LuxR solo of SD316 (contig2_994). The W and M substitutions may be involved in binding to plant-based compounds, as the substitutions are present in OryR and partially present in PsoR, from two PAB known to have an inter-kingdom exchange with plants [22, 23].

Fig. 2

Detection and analysis of LuxR solos. a 10 kbp genetic region surrounding identified luxR homologs (luxR solos) (red) having no corresponding luxI homologs. b Alignment of putative LuxR homologs with canonical LuxR homologs using clustalOmega. Highlighted residues correspond to invariant sites in authentic LuxR-type AHL-mediated transcriptional regulators [21]. Residues highlighted in yellow match with the consensus, whereas those highlighted in green vary from the consensus. Regions marked with a blue diamond are involved in ligand-binding. Regions marked with a pink diamond are involved in DNA-binding. Residues are numbered based on the sequence of TraR. c Evolutionary placement of LuxR family transcriptional regulators. LuxR homologs from the WCNP strains are denoted by a red star. LuxR homologs surrounded by a green line box are LuxR solos from plant-associated bacteria (PAB). LuxR homologs surrounded by a blue line box are neither AHL-like or PAB

All of the seven putative SD strain LuxR solos contain the conserved amino acids D70, P71, and E178, L182 and G188. A LuxR solo identified from SD316, on contig 2_994, has substitutions W85M and Y61W (Fig. 2b), identical to the LuxR solo, PsrR, from the plant endophyte Kosakonia sp. PsrR belongs to the PAB subfamily of LuxR solos and was shown to be involved in root endosphere colonization [24]. Furthermore, substitutions were observed in two of the seven LuxR solo homologs from SD129 (contig 10_52) and SD316 (contig 6_72) in which the conserved amino acids in regulatory domains contained substitutions W85R and for G113 residue, V and T, respectively (Fig. 2b). These amino acid substitutions represent novel changes not reported in other LuxR solo proteins and may reflect specificities required for the unknown binding molecule(s) for these two LuxR solo regulatory proteins. Building on this trend, Coutinho and coworkers showed that an ethanolamine derivative from cottonwood tree leaf macerates activates the Pseudomonas sp. GM79 pipA expression at extremely low concentrations (10 pM) and that the LuxR solo, PipR is required for pipA activation [25, 26].

Comparison of the Ochrobactrum pseudogrignonense strains SD129 and SD340 with those species in the NCBI database show a staple pattern with three luxR genes and one luxI gene. The luxI, whenever present, appears to always have a proximal luxR (Additional file 3).


We hypothesize that the LuxR solos reported here could potentially be responsive to AHLs or different signals produced by neighboring species or signals in the aquifer water and coordinate regulation of gene expression, thus potentially playing important roles in the ecology and persistence of these species in this pristine aquifer.


This work is from draft genome assembly of bacterial strains.

The possible presence of plasmids in strain cannot be clearly identified.

Availability of data and materials

The genome sequences of the strains described in this study have been deposited in the GenBank database. The accession numbers and annotation features are presented in Table 1.

Direct links are below:

















Acyl homoserine lactones


Antibiotics and secondary metabolite analysis shell




Plant-associated bacteria


Wind Cave National Park


  1. 1.

    Beam JP, Becraft ED, Brown JM, Schulz F, Jarett JK, Bezuidt O, Poulton NJ, Clark K, Dunfield PF, Ravin NV, Spear JR, Hedlund BP, Kormas KA, Sievert SM, Elshahed MS, Barton HA, Stott MB, Eisen JA, Moser DP, Onstott TC, Woyke T, Stepanauskas R. Ancestral absence of electron transport chains in patescibacteria and DPANN. Front Microbiol. 2020;11:1848.

    Article  PubMed  PubMed Central  Google Scholar 

  2. 2.

    Hershey OS, Kallmeyer J, Wallace A, Barton MD, Barton HA. High microbial diversity despite extremely low biomass in a deep karst aquifer. Front Microbiol. 2018;2018(9):2823.

    Article  Google Scholar 

  3. 3.

    Fuqua WC, Winans SC, Greenberg EP. Quorum sensing in bacteria: the LuxR-LuxI family of cell density-responsive transcritptional regulators. J Bacteriol. 1994;176:269–75.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  4. 4.

    Waters CM, Bassler BL. Quorum-sensing: cell-cell communication in bacteria. Ann Rev Cell Dev Biol. 2005;21:319–46.

    CAS  Article  Google Scholar 

  5. 5.

    Gan HM, Gan HY, Ahmad NH, Aziz NA, Hudson AO, Savka MA. Whole genome sequencing and analysis reveal insights into the genetic structure, diversity and evolutionary relatedness of luxI and luxR homologs in bacteria belonging to the Sphingomonadaceae family. Front Cell Infect Microbiol. 2015.

    Article  PubMed  PubMed Central  Google Scholar 

  6. 6.

    Fuqua C. The QscR quorum-sensing regulon of Pseudomonas aeruginosa: an orphan claims it identity. J Bacteriol. 2006;188:3169–71.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  7. 7.

    Gonzalez JF, Venturi V. A novel widespread interkingdom signaling circuit. Trends Plant Sci. 2013;18:167–74.

    CAS  Article  PubMed  Google Scholar 

  8. 8.

    Brotherton CA, Medema MH, Geenberg EP. 2018. luxR homolog-linked biosynthetic gene clusters in Proteobacteria. mSystems. 2018;3(3):e00208–17.

  9. 9.

    Coutinho BG, Mevers E, Schaefer AL, Pelletier DA, Harwood CS, Clardy J, Greenberg EP. A plant-responsive bacterial-signaling systems senses an ethanolamine derivative. PNAS. 2018;115(39):9785–90.

    CAS  Article  PubMed  Google Scholar 

  10. 10.

    Kumar HKS, Gan HM, Tan MH, Eng WWH, Barton HA, Hudson AO, and Savka MA. Genomic characterization of eight Ensifer strains isolated from pristine caves and whole genome phylogeny of Ensifer (Sinorhizobium). J Genomics. 2017; 5:12–15. doi:

  11. 11.

    Wick RR, Judd LM, Gorrie CL, Holt KE. Unicycler: resolving bacterial genome assemblies from short and long sequencing reads. PLoS Comput Biol. 2017;13:e1005595.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  12. 12.

    Blin K, Shaw S, Steinke K, Villebro R, Ziemert N, Lee SY, Medema MH, Weber T. antiSMASH 5.0: updates to the secondary metabolite genome mining pipeline. Nucleic Acids Res. 2019.

    Article  PubMed  PubMed Central  Google Scholar 

  13. 13.

    Richter M, Rosselló-Móra R, Glöckner FO, and Peplies J. 2015. JSpeciesWS: a web server for prokaryotic species circumscription based on pairwise genome comparison.

  14. 14.

    Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ. Basic local alignment search tool. J Mol Biol. 1990;215:403–10.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  15. 15.

    Jones P, Binns D, Chang HY, Fraser M, Li W, Mcanulls C, et al. InterProScan 5: genome-scale protein function classification. Bioinformatics. 2014;30:1236–40.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  16. 16.

    Seemann T. Prokka:rapid prokaryotic genome annotation. Bioinformatics. 2014;30:2068–9.

    CAS  Article  PubMed  Google Scholar 

  17. 17.

    Jones P, Binns D, Chang HY, et al. InterProScan 5: genome-scale protein function classification. Bioinformatics. 2014;30(9):1236–40.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  18. 18.

    Bolger AM, Lohse M, Usadel B. Trimmomatic: a flexible trimmer for Illumina sequence data. Bioinformatics. 2014;30:2114–20.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  19. 19.

    Sievers F, Wilm A, Dineen DG, Gibson TJ, Karplus K, Li W, Lopez R, McWilliam H, Remmert M, Sding J, Thompson JD, Higgins DG. Fast, scalable generation of high-quality protein multiple sequence alignments using Clustal Omega. Mol Syst Biol. 2011;7:539.

    Article  PubMed  PubMed Central  Google Scholar 

  20. 20.

    Sullivan MJ, Petty NK, Beatson SA. Easyfig: a genome comparison visualizer. Bioinformatics. 2011;27(7):1009–10.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  21. 21.

    Fuqua C, Greenberg EP. Listening in on bacteria: acyl-homoserine lactone signaling. Nat Rev Mol Cell Biol. 2002;3(9):685–95.

    CAS  Article  PubMed  Google Scholar 

  22. 22.

    Lim YL, Ee R, How KY, Lee SK, Yong D, Tee KK, Yin WF, Chan KG. Complete genome sequencing of Pandoraea pnomenusa RB38 and molecular characterization of its N-acyl homoserine lactone synthase gene ppnI. PeerJ. 2015;3:e1225.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  23. 23.

    Gangming X. Evolution of LuxR solos in bacterial communication: receptors and signals. Biotechnol Lett. 2019.

    Article  Google Scholar 

  24. 24.

    Subramoni S, Gonzalez JF, Johnson A, Pechy-Tarr M, Rochat L, Paulsen I, Loper JE, Keel C, Venturi V. Bacterial subfamily of LuxR regulators that respond to plant compounds. AEM. 2011;77(13):4579–88.

    CAS  Article  Google Scholar 

  25. 25.

    Mosquito S, Meng K, Devescovi G, Bertani I, Geller AM, Levy A, Myers MP, Bez C, Govaceuszach S, Venturi V. LuxR solos in the Plant Endophyte Kosakonia sp. Strain KO348. 2020.

    Article  Google Scholar 

  26. 26.

    Coutinho CG, Mevers E, Schaefer AL, Pelletier DA, Harwood CS, Clardy J, Greenberg EP. A plant-responsive bacterial-signaling system senses an ethanolamine derivative. Proc Natl Acad Sci USA. 2018;115(39):9785–90.

    CAS  Article  PubMed  Google Scholar 

Download references


Not applicable.


The authors acknowledge the Thomas H. Gosnell School of Life Sciences (GSoLS) and the College of Science (COS) at the Rochester Institute of Technology (RIT) for ongoing support. PCW was supported by a 2019 RIT COS Summer Undergraduate Research Fellowship.

Author information




Conceived and designed experiments: PCW, NHW, MAS. Performed the experiments PCW, NHW. Analyzed the data: AOH, MAS, NHW, PCW. Isolated strains HAB. AOH, HAB, HMG, MAS, PCW wrote the paper. All authors read and approved the final manuscript.

Corresponding author

Correspondence to Michael A. Savka.

Ethics declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors have declared that no competing interests exist.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Additional file 1.

Detection and analysis of LuxI synthases. (A) 10kbp genetic region surrounding identified luxR homologs (red) having corresponding LuxI homologs (blue) in SD129 and SD340. (B) Interproscan output of a successfully validated luxI homolog. Each accession number corresponds to a detected protein domain. (C) Alignment of putative LuxI homologs with canonical LuxI homologs using clustalOmega. Residues highlighted in yellow are invariant sites in validated LuxI-type autoinducer synthases (Fuqua and Greenberg, 2002). Residues are numbered based on the sequence of TraI.

Additional file 2.

Interproscan output of a successfully validated luxR homolog. Each accession number corresponds to a detected protein domain.

Additional file 3.

Genomic analyses of eight Ochrobactrum pseudogrignonense strains. Analysis of strains available on NCBI and comparison to SD129 and SD316 reveal a commonality in the presence of luxR and luxI genes1.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit The Creative Commons Public Domain Dedication waiver ( applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Wengert, P.C., Wong, N.H., Barton, H.A. et al. Genomic characterization of bacteria from the ultra-oligotrophic Madison aquifer: insight into the archetypical LuxI/LuxR and identification of novel LuxR solos. BMC Res Notes 14, 175 (2021).

Download citation


  • Madison aquifer bacteria
  • Karst aquifer
  • Ultra-oligotrophic bacteria
  • Pristine cave
  • Wind Cave National Park
  • LuxI
  • LuxR
  • LuxR solo