Molecular Characterization of Mercury Resistant Bacteria Isolated from Tannery Wastewater

Aatif Amin1,*, Zakia Latif2, Arslan Sarwar1, Basit Zeshan1 and Mushtaq A. Saleem1

1Department of Microbiology, Faculty of Life Sciences, University of Central Punjab, Lahore

2Department of Microbiology and Molecular Genetics, University of the Punjab, Lahore-54590

ABSTRACT

Mercury resistant (HgR) bacteria were isolated from heavy metal polluted wastewater and soil, collected from the proximity of some tanneries from Kasur, Pakistan. Three out of 30 bacterial strains were screened out on the basis of resistance level against various concentrations of HgCl2. Bacterial isolates AZ-1, AZ-2 and AZ-3 showed resistance up to 40 μg/mL of HgCl2 and mercury sensitive (HgS) isolate ZA-15 was taken as a negative control. 16S rDNA ribotyping and phylogenetic analysis were performed for the characterization of isolates as Bacillus sp. AZ-1 (KT270477), B. cereus AZ-2 (KT270478), B. cereus AZ-3 (KT270479) and Enterobacter cloacae ZA-15 (KJ728671). Phylogenetic relationship on the basis of merA nucleotide sequence confirmed 51-100% homology with the corresponding region of the merA gene of already reported mercury resistant Gram positive bacteria. Restriction fragment length polymorphism (RFLP) analysis was applied to the amplification products of 16S rRNA and merA genes and a specific restriction patterns was successfully obtained after treatment with different endonucleases. A small scale reservoir of Luria Bertani (LB) medium supplemented with 30 μg/mL of HgCl2 was designed to check the detoxification ability of the selected strains. The results demonstrated 83% detoxification of mercury by both B. cereus AZ-2 and B. cereus AZ-3, and 76% detoxification by Bacillus sp. AZ-1 (p<0.05).

Article Information

Received 26 April 2019

Revised 22 June 2019

Accepted 27 August 2019

Available online 27 November 2020

Authors’ Contribution

AA designed and executed the experiments. ZL designed the experiments and supervised the work. AA executed the experimental work and analyzed the data. MS proofread the manuscript.

Key words

Bacillus spp., Hg-detoxification, RFLP, 16S rRNA, merA.

DOI: https://dx.doi.org/10.17582/journal.pjz/20190426100430

* Corresponding author: aatif.amin@ucp.edu.pk

0030-9923/2021/0001-0033 $ 9.00/0

Copyright 2021 Zoological Society of Pakistan

Introduction

Mercury (Hg) pollution is a global problem due to its toxicity to human, plants and animals. In the environment, the level of mercury pollution is increasing day by day due to the natural and anthropogenic sources. Natural sources include volcanoes, forest fires, cinnabar (ore) and fossil fuels whereas anthropogenic activities include the release of industrial waste from chlor-alkali industries, metal-mining, emissions from coal-using power plants and incineration of municipal and medical waste (Steenhuisen and Wilson, 2015). Mercury occurs in two forms as ionic and organic in the environment which cause cytotoxicity and neurotoxicity to humans and animals (Santos-Gandelman et al., 2014b).

Bacillus sp. RC607, first time identified to have chromosomal resistance against mercury, was isolated from contaminated site, Boston Harbor, USA (Sedlmeier and Altenbuchner, 1992). In later studies, 74 Bacillus spp. isolated from Minimata Bay sediment, Japan and other bacterial isolates from Russian environment showed nucleotide sequences like the strain RC607. It was suggested the possible worldwide distribution of RC607 mer operon. Plasmids containing mer operon were found in several Bacillus strains analyzed in the Russian study (Nakamura and Silver, 1994; Bogdanova et al., 1998).

Microorganisms have evolved mercury detoxification systems against mercury containing compounds due to the existence of mercury in the environment and its toxic effects. Bacteria showed the potential to detoxify the toxic form of mercury (Hg+2) into non-toxic elemental form (Hg0) through a cytoplasmic enzyme, mercuric reductase which was encoded by merA gene of mer operon (Lund et al., 1986). A lot of research analysis was done on this cytoplasmic enzyme, a member of flavin containing NADPH dependent dithiol oxidoreductase and was isolated from Gram positive as well as Gram negative bacteria (Moore et al., 1990). Gram positive and some of Gram negative bacteria also possess organomercurial lyase encoded by merB gene of mer operon which cleaves the C–Hg bond of many mercury containing compounds. The enzyme, organomercurial lyase confers broad spectrum resistance against mercuric compounds which results the elemental mercury (Hg0) diffused out of the cell (Huang et al., 1999).

Bacterial 16S rRNA possesses mosaic composition of phylogenetically conserved and variable regions which make this gene a common target for taxonomic purposes. The molecular technique in which difference in DNA sequences is detected by the presence of fragments of different lengths after digestion of the DNA samples of interest with specific restriction enzymes is known as restriction fragment length polymorphism (RFLP). A 16S rRNA-RFLP method for the identification of Arcobacter species has previously been described by Figueras et al. (2008). Many researchers have targeted the 16S, 23S rRNA and merA genes for RFLP analysis in order to identify species belonging to a variety of genera and species (Jeng et al., 2001). The effects of stress and genetic parameters like genetic diversity, gene transfer and evolution could be evaluated by mercury resistance determinants. Osborn et al. (1997) have employed merA gene to check the diversity in Gram positive mercury resistant determinants because of its key role in mercury detoxification system.

Keeping in mind the available information about Hg-resistant bacteria and their potential to detoxify mercury, the current study is designed to determine the detoxification efficiency of selected Hg-resistant bacteria isolated from mercury-contaminated sites at lab scale. Furthermore, characterization and diversity of selected Hg- resistant bacteria is determined on the basis of RFLP analyses of 16S rRNA and merA genes.

 

Materials and methods

Isolation and screening of HgR bacteria

Mercury resistant bacteria were isolated from wastewater and contaminated soil samples obtained from some tanneries around Kasur city of Pakistan. Serial dilutions of 10-1, 10-2, 10-3 and 10-4 were prepared from 1% of the original water and soil samples. For the isolation of individual colonies of Hg-resistant bacteria, LB agar plates supplemented with different concentrations of HgCl2 ranging from 1-40 μg/mL were spread by 100 μL of 10-3 and 10-4 dilutions. All the plates were placed in an incubator set at 37°C for 24 h. After incubation, the isolated Hg-resistant bacterial colonies were selected and then re-streaked on new LB agar plates without HgCl2 to get purified colonies (Amin and Latif, 2017a). All the purified Hg-resistant bacterial cultures were stored as glycerol stocks at -80°C.

PCR-amplification of 16S rDNA and merA gene

The bacterial isolates AZ-1, AZ-2 and AZ-3, showing high resistance against HgCl2 were characterized by 16S rDNA ribotyping. The bacterial genomic DNA was extracted by Pure Link® Genomic DNA Mini Kit (www.lifetechnologies.com). The universal primers were used for the amplification of 16S rRNA gene; 16S-F (5´AGAGTTTGATCCTGGCTCAG3´) and 16S-R (5´AAGGAGGTGATCCAGCCGCA3´) (Normand, 1995) and for merA gene; merAF: 5’TGGGTGGAACTTGCGTTAA3’ and merAR: 5’TTATCCAGCACAGCAAGATA3’ using BIOER XP-Thermal Cycler. The amplification conditions for both genes were set as 35 cycles of a denaturation step (94°C, 1 min), an annealing step (55°C, 1 min) and an extension step (72°C, 2 min) with a final 10 min chain elongation at 72°C (Amin and Latif, 2017b; Amin et al., 2019). 16S rRNA gene sequencing was carried out by Macrogen sequencing core facility, Seoul, Korea. The obtained sequences were checked for homology through nucleotide blast and submitted to GenBank.

Phylogenetic relationship

The multiple sequence alignment through clustalW was used to check phylogeny among bacterial species of Bacillus genera on the basis of 16S rRNA and merA genes through neighbor joining method using MEGA 5 software. The percentage of homology among different clade was checked with bootstrap test at value of 1000 replica as shown next to the branches. The tree is drawn to scale which represent the nucleotide change. The units of branch lengths and the evolutionary distances were same as used to infer the phylogenetic tree.

RFLP analysis of 16S rRNA and merA genes

PCR products of partially amplified 16S rRNA and merA genes were subjected to RFLP. Two restriction endonucleases EcoR1 and Taq1 were used for 16S rRNA gene while HinF1 and HaeIII (BsuR1) were used for merA gene. The reaction mixture contained 3.0 μL of 1X buffer (R-buffer for HaeIII (BsuRI) and HinF1, and unique-buffer for EcoR1 and Taq1), 15.0 μL PCR products (approximately 1.0 μg), 1 μL of specific endonuclease and 11 μL of deionized water with total volume of 30 μL. The reaction mixtures were incubated at their specific temperatures as recommended by Fermentas restriction enzymes buffer activity chart (www.fermentas.com). The restriction fragments were separated along with 100bp plus DNA ladder on 1.5% w/v agarose gel.

Detoxification of Hg2+ by selected bacteria at a lab scale

For the detoxification/reduction of toxic ionic mercury (Hg+2) into less toxic elemental Hg0, 5 mL of LB medium was inoculated with each 5 μL of Hg-resistant bacterial glycerol stock of AZ-1, AZ-2 and AZ-3 and ZA-15 (Hg-sensitive, used as -ve control). After overnight incubation at 37°C, 1.5 mL of each starter culture (O.D. 2.0 at 600 nm) was taken to inoculate four flasks (4th flask was taken as –ve control), each containing 30 mL of LB medium supplemented with 30 μg/mL of HgCl2 in triplicate. All flasks were incubated at 37°C for 24 h at 120 rpm of agitation. After incubation, the pH of culture medium for each strain was adjusted to 0.4 by adding concentrated H2SO4 and then cultures were spun at 12000 × g for 15 min. After centrifugation, the supernatant was separated and then transferred to separating funnel. To the cooled supernatant, 4 mL of 6N acetic acid and 2.5 mL chloroform were added and vigorously shaked for 1 min. The phases were allowed to separate completely and took out the chloroform layer and discarded. Freshly prepared 5 mL of 0.001% dithizone solution was added in the remaining solution of each strain and again shaked vigorously for 1 min. The layers were allowed to separate by putting cotton in the tips of separating funnels and eluted the dithizone-Hg+2 complex. The elemental mercury (Hg0) was vaporized due the reduction (Hg+2 → Hg0) by selected Hg-resistant bacteria through mercuric reductase (encoded by merA gene). On the reduction of all given concentration of Hg+2 to Hg0, no significant color was produced with dithizone. In the presence of Hg+2, an orange color was observed in the organic layer and the color became more pronounced with higher concentration of Hg+2. The optical density (O.D.) of water–free chloroform extracts of each strain was determined at 500 nm against reagent blank (dithizone) to estimate the detoxification of LB medium (Elly, 1973; Humaira et al., 2005).

Statistical analysis

All data of small scale Hg-detoxification experiment were subjected to mean ± standard deviation. The analysis of variance (ANOVA) was performed by using statistical package for social sciences (SPSS) windows V.20 software.

 

Table I.- Growth of bacterial strains at different concentrations of HgCl2.

Isolates	Growth against HgCl2
	10 µg/mL	20 µg/mL	30 µg/mL	40 µg/mL
AZ-1	+	+	+	+
AZ-2	+	+	+	+
AZ-3	+	+	+	+
ZA-15	-	-	-	-

 

Results and discussion

The interaction between heavy metals and heavy metal resistant microorganisms has been determined by bacterial transformation and the conversion of metallic ion form to elemental form by reduction (Chang and Huang, 1998). Some of these microorganisms have shown promising activity to decrease mercury from contaminated environments. In the present study, three bacterial strains out of 30 were selected on the basis of resistance to higher HgCl2 concentrations (MIC: upto 1-40 μg/mL) and one strain ZA-15 was taken as a negative control (Table I). In literature, 100 μM, 50 ppm, 100 μg/mL and 20 μg/mL were found to be the MIC for mercury in Bacillus spp. (Sathyavathi et al., 2013; Santos-Gandelman et al., 2014a; Dash and Das, 2015; Amin and Latif, 2017).

Mercury resistant bacterial isolates (AZ-1, AZ-2 and AZ-3) were characterized by 16S rDNA ribotyping (≈ 1.5 kb) and identified as Bacillus sp. (KT270477), B. cereus (KT270478), and B. cereus (KT270479), respectively. Other close matches to Bacillus sp. (KT270477) included B. anthracis HQ200405 and Bacillus sp. JN593078 and GU566355. The bacterial isolates also showed similarity (97%) among themselves. In the same clade, B. cereus AZ-1 and AZ-2 showed 97% similarity with already reported B. thuringiensis KP306751, Bacillus sp. JQ691603 and B. cereus KP407139, KR303714, KP202304 and LK392517 (Fig. 1). Phylogenetic analysis of selected bacterial strains on the basis of merA gene showed 100% homology with already reported sequences of merA genes in B. cereus NG034957, B. megaterium NG035056, Bacillus sp. LC015492, Paenibacillus sp. LC015492. Likewise, B. macroides Y09906 showed 99% homology, Bacillus sp. RC607 AF138877 and Clostridium butyricum AB024961 as 88%, respectively by B. cereus AB066362 as shown in Figure 2.

 

16S rRNA gene is an excellent molecular marker that is widely being used for the study of bacterial community composition due to the possession of variable regions in its nucleotide sequence. Many researchers such as Osborn et al. (1997), Narita et al. (2003) and Scheidegger et al. (2009) described PCR-RFLP based analysis of 16S rRNA and merA genes to find the genetic variability and diversity in bacterial community. In the present study, selected bacteria already characterized by 16S rDNA ribotyping were further characterized and confirmed at species level by RFLP analyses of both 16S rRNA and merA genes. PCR-amplified products of 16S rRNA (≈ 1.5 kb) and merA (≈ 1.3 kb) genes of four bacterial strains were digested by EcoR1 and TaqI, and by HinF1 and HaeIII, respectively. Different restricted fragments of 16S rRNA gene showing different sizes of four bacterial strains, AZ-1, AZ-2, AZ-3 and ZN-15 were resolved on 1% agarose by comparing with 100bp plus DNA ladder. Similarly, restriction pattern of merA gene of three bacterial strains, AZ-1, AZ-2 and AZ-3 (except ZN-15 used as a negative control for merA gene) was observed on 1% agarose gel. The number and size of bands of both genes after treatment with restriction endonucleases varied on agarose gel as shown in Figure 3A and 3B. It was resulted from PCR-RFLP experiment that selected Hg-resistant bacteria belonged to same genera as Gram +ve Bacillus on the basis of restriction pattern of 16S rRNA gene except E. cloacae ZN-15 (Gram -ve). Likewise, RFLP analysis of merA gene amplified from characterized bacterial strains confirmed the presence of Gram +ve mer operon due to the similar digestion pattern.

Review of literature showed that certain species such as cyanobacteria, Klebsiella aerogenes, K. aerogenes NCTC418 and Pseudomonas putida were involved in bio-reduction of toxic mercury compounds under laboratory conditions (Glendinning et al., 2005; Oehmen et al., 2009). The species of Enterobacter, Cronobacter and Pseudomonas showed the detoxification potential of mercury up to 95% in yeast extract medium (YEM). Nakamura et al. (1999) reported the decrease of mercury from various mercury compounds like mercuric chloride (HgCl2) as 88.9%, methyl mercuric chloride (MeHg) 95.4%, ethylmercuric chloride (EtHg) 83.8%, thimerosal 91.9%, fluorescein mercuric acetate (FMA*) 74.6%, phenylmercuric acetate (PMA) 5.7%, and p-chloromercuric benzoate (p-CMB) 92.3% by Pseudoalteromonas haloplanktis strain M-1.

 

 

 

The bio-sorption capability of the immobilized B. cereus cells was described by Sinha et al. (2012) as 104.1 mg/g Hg2+ at 30°C and pH 7.0, biomass concentration of 0.02 g/L and contact time of 72 h. In this study, a lab scale experiment was designed to analyze the detoxification potential of mercury by selected bacterial strains. The results indicated 23 (76%), 25 (83%), 25 (83%) and 1 (2.5%) μg/mL detoxification of Hg2+ out of 30 μg/mL (100%) by the selected bacterial strains as Bacillus sp. AZ-1, B. cereus AZ-2, B. cereus AZ-3 and E. cloacae ZA-15, respectively (Fig. 4).

 

Conclusion

In the present study, the detoxification potential of Hg-resistant bacteria, Bacillus sp. AZ-1, B. cereus AZ-2 and B. cereus AZ-3 was checked (E. cloacae ZA-15 used as –ve control) and then their molecular identification and phylogenetic analysis were described on the basis of 16S rDNA ribotyping and merA gene sequence. Furthermore, the nucleotide sequence variations in 16S rRNA and merA genes of selected bacteria were determined by RFLP analyses. The results showed that mercury resistant bacterial strains, B. cereus AZ-2 and B. cereus AZ-3 isolated from Kasur city of Pakistan can be used for the Hg-detoxification of industrial effluents due to the presence of merA gene and its manipulation into mercuric reductase expression system of bacteria.

 

Acknowledgments

The authors are profoundly grateful to the Higher Education Commission (HEC), Pakistan and University of the Punjab for providing financial support to conduct the current research work.

 

Statement of conflict of interest

The authors have declared no conflict of interest.

 

References

Amin, A. and Latif, Z., 2017a. Screening of mercury-resistant and indole-3-acetic acid producing bacterial-consortium for growth promotion of Cicer arietinum L. J. Basic Microbiol., 57: 204-217. https://doi.org/10.1002/jobm.201600352

Amin, A. and Latif, Z., 2017b. Cloning, expression, isotope labeling, and purification of transmembrane protein MerF from mercury resistant Enterobacter sp. AZ-15 for NMR studies. Front. Microbiol., 8: 1250-1261. https://doi.org/10.3389/fmicb.2017.01250

Amin, A., Sawar, A., Saleem, M.A., Latif, Z. and Opella, S.J., 2019. Expression and purification of transmembrane protein MerE from mercury-resistant Bacillus cereus. J. Microbiol. Biotechnol., 29: 274-282. https://doi.org/10.4014/jmb.1704.04062

Bogdanova, E., Bass, I., Minakhin, L., Petrova, M., Mindlin, S., Volodin, A., Kalyaeva, E., Tiedje, J., Hobman, J. and Brown, N., 1998. Horizontal spread of mer operons among Gram-positive bacteria in natural environments. Microbiology, 144: 609-620. https://doi.org/10.1099/00221287-144-3-609

Chang, J.S. and Huang, J.C., 1998. Selective adsorption/recovery of Pb, Cu, and Cd with multiple fixed beds containing immobilized bacterial biomass. Biotechnol. Prog., 14: 735-741. https://doi.org/10.1021/bp980070y

Dash, H.R. and Das, S., 2015. Bioremediation of inorganic mercury through volatilization and biosorption by transgenic Bacillus cereus BW-03 (PW-05). Int. Biodeterior. Biodegrad., 103: 179-185. https://doi.org/10.1016/j.ibiod.2015.04.022

Elly, C.T., 1973. Dithizone procedure for mercury analysis. J. Water Pollut. Contr. Fed., 45: 940-945.

Figueras, M.J., Collado, L. and Guarro, J., 2008. A new 16S rDNA-RFLP method for the discrimination of the accepted species of Arcobacter. Diagn. Microbiol. Infect. Dis., 62: 11-15. https://doi.org/10.1016/j.diagmicrobio.2007.09.019

Glendinning, K., Macaskie, L. and Brown, N., 2005. Mercury tolerance of thermophilic Bacillus sp. and Ureibacillus sp. Biotechnol. Lett., 27: 1657-1662. https://doi.org/10.1007/s10529-005-2723-8

Huang, C.C., Narita, M., Yamagata, T. and Endo, G., 1999. Identification of three merB genes and characterization of a broad-spectrum mercury resistance module encoded by a class II transposon of Bacillus megaterium strain MB1. Gene, 239: 361-366. https://doi.org/10.1016/S0378-1119(99)00388-1

Humaira, K., Mohammad, J. and Mohammad, I., 2005. A simple spectrophotometric determination of trace level mercury using 1, 5-diphenylthiocarbazone solubilized in micelle. Anal. Sci., 21: 507-512. https://doi.org/10.2116/analsci.21.507

Jeng, R.S., Svircev, A.M., Myers, A.L., Beliaeva, L., Hunter, D.M. and Hubbes, M., 2001. The use of 16S and 16S-23S rDNA to easily detect and differentiate common Gram-negative orchard epiphytes. J. Microbiol. Meth., 44: 69-77. https://doi.org/10.1016/S0167-7012(00)00230-X

Lund, P.A., Ford, S.J. and Brown, N.L., 1986. Transcriptional regulation of the mercury-resistance genes of transposon Tn501. J. Gen. Microbiol., 132: 465-480. https://doi.org/10.1099/00221287-132-2-465

Moore, M.J., Distefano, M.D., Zydowsky, L.D., Cummings, R.T. and Walsh, C.T., 1990. Organomercurial lyase and mercuric ion reductase: Nature’s mercury detoxification catalysts. Accounts chem. Res., 23: 301-308. https://doi.org/10.1021/ar00177a006

Nakamura, K., Hagimine, M., Sakai, M. and Furukawa, K., 1999. Removal of mercury from mercury-contaminated sediments using a combined method of chemical leaching and volatilization of mercury by bacteria. Biodegradation, 10: 443-447. https://doi.org/10.1023/A:1008329511391

Nakamura, K. and Silver, S., 1994. Molecular analysis of mercury-resistant Bacillus isolates from sediment of Minamata Bay, Japan. Appl. environ. Microbiol., 60: 4596-4599.

Narita, M., Chiba, K., Nishizawa, H., Ishii, H., Huang, C.C., Kawabata, Z., Silver, S. and Endo G., 2003. Diversity of mercury resistance determinants among Bacillus strains isolated from sediment of Minamata Bay. FEMS Microbiol. Lett., 223: 73-82. https://doi.org/10.1016/S0378-1097(03)00325-2

Normand, P., 1995. Utilisation des séquences 16S pour le positionnement phylétique d’un organisme inconnu. Oceanis, 21: 31-56.

Oehmen, A., Fradinho, J., Serra, S., Carvalho, G., Capelo, J., Velizarov, S., Crespo, J. and Reis, M., 2009. The effect of carbon source on the biological reduction of ionic mercury. J. Hazard. Mater., 165: 1040-1048. https://doi.org/10.1016/j.jhazmat.2008.10.094

Osborn, A.M., Bruce, K.D., Strike, P. and Ritchie, D.A., 1997. Distribution, diversity and evolution of the bacterial mercury resistance (mer) operon. FEMS Microbiol. Rev., 19: 239-262. https://doi.org/10.1111/j.1574-6976.1997.tb00300.x

Santos-Gandelman, J.F., Cruz, K., Crane, S., Muricy, G., Giambiagi-Demarval, M., Barkay, T. and Laport, M.S., 2014a. Potential application in mercury bioremediation of a marine sponge-isolated Bacillus cereus strain Pj1. Curr. Microbiol., 69: 374-380. https://doi.org/10.1007/s00284-014-0597-5

Santos-Gandelman, J.F., Giambiagi-Demarval, M., Muricy, G., Barkay, T. and Laport, M.S., 2014b. Mercury and methylmercury detoxification potential by sponge-associated bacteria. Antonie Leeuwenhoek, 106: 585-590. https://doi.org/10.1007/s10482-014-0224-2

Sathyavathi, S., Manjula, A., Rajendhran, J. and Gunasekaran, P., 2013. Biosynthesis and characterization of mercury sulphide nanoparticles produced by Bacillus cereus MRS-1. Indian J. exp. Biol., 51: 973-978.

Scheidegger, E.M., Fracalanzza, S.A., Texeira, L.M. and Caradarelli-Leite, P., 2009. RFLP analysis of a PCR-amplified fragment of the 16S rRNA gene as a tool to identify Enterococcus strains. Mem. Inst. Oswaldo Cruz., 104: 1003-1008. https://doi.org/10.1590/S0074-02762009000700011

Sedlmeier, R. and Altenbuchner, J., 1992. Cloning and DNA sequence analysis of the mercury resistance genes of Streptomyces lividans. Mol. Gen. Genet., 236: 76-85.

Sinha, A., Pant, K.K. and Khare, S.K., 2012. Studies on mercury bioremediation by alginate immobilized mercury tolerant Bacillus cereus cells. Int. Biodeterior. Biodegrad., 71: 1-8. https://doi.org/10.1016/j.ibiod.2011.12.014

Steenhuisen, F. and Wilson, S.J., 2015. Identifying and characterizing major emission point sources as a basis for geospatial distribution of mercury emissions inventories. Atmospheric Environ., 112: 167-177. https://doi.org/10.1016/j.atmosenv.2015.04.045

Tamura, K., Nei, M. and Kumar, S., 2004. Prospects for inferring very large phylogenies by using the neighbor-joining method. Proc. natl. Acad. Sci. U.S.A., 101: 11030-11035. https://doi.org/10.1073/pnas.0404206101