Two cGAS-like receptors induce antiviral immunity in Drosophila.

In mammals, cyclic GMP-AMP (cGAMP) synthase (cGAS) produces the cyclic dinucleotide 2'3'-cGAMP in response to cytosolic DNA and this triggers an antiviral immune response. cGAS belongs to a large family of cGAS/DncV-like nucleotidyltransferases that is present in both prokaryotes1 and eukaryotes2-5. In bacteria, these enzymes synthesize a range of cyclic oligonucleotides and have recently emerged as important regulators of phage infections6-8. Here we identify two cGAS-like receptors (cGLRs) in the insect Drosophila melanogaster. We show that cGLR1 and cGLR2 activate Sting- and NF-κB-dependent antiviral immunity in response to infection with RNA or DNA viruses. cGLR1 is activated by double-stranded RNA to produce the cyclic dinucleotide 3'2'-cGAMP, whereas cGLR2 produces a combination of 2'3'-cGAMP and 3'2'-cGAMP in response to an as-yet-unidentified stimulus. Our data establish cGAS as the founding member of a family of receptors that sense different types of nucleic acids and trigger immunity through the production of cyclic dinucleotides beyond 2'3'-cGAMP.

Classification, characterization and structural analysis of sugar nucleotidylyltransferase family of enzymes.

Sugar Nucleotidyl Transferases (SNTs) constitute a large family of enzymes that play important metabolic roles. Earlier, for one such SNT, termed N-acetylglucosamine-1-phosphate uridyltransferase- GlmU, we had established that two magnesium ions - Mg2+A and Mg2+B - catalyze the sugar-nucleotidyl transfer reaction. Despite a common structural framework that SNTs share, we recognized key differences around the active-site based on the analysis of available structures. Based on these differences, we had classified SNTs into two major groups, Group - I & II; and further, variation in 'Mg2+A-stabilizing motifs' led us to sub-classify them into five distinct sub-groups. Since group specific conservation of 'Mg2+A-stabilizing motifs' was based only for 45 available structures, here we validate this via an exhaustive analysis of 1,42,025 protein sequences. Previously, we had hypothesized that a metal-ion-catalyzed mechanism would be operative in all SNTs. Here, we validate it biochemically and establish that Mg2+ is a strict requirement for nucleotidyl transfer reactions in every group or sub-group and that a common metal ion dependent mechanism operates in SNTs. Further, mutating Mg2+A coordinating residue in each sub-group led to abolished catalysis, indicating an important role for both of these residues and suggest that SNTs employ variations over 'a conserved catalytic mechanism mediated by Mg2+ ion(s)', to bring about functional diversity. This would constitute a comprehensive study to establish the catalytic mechanism across the family of SNTs.

Urm1: an essential regulator of JNK signaling and oxidative stress in Drosophila melanogaster.

Ubiquitin-related modifier 1 (Urm1) is a ubiquitin-like molecule (UBL) with the dual capacity to act both as a sulphur carrier and posttranslational protein modifier. Here we characterize the Drosophila melanogaster homologues of Urm1 (CG33276) and its E1 activating enzyme Uba4 (CG13090), and show that they function together to induce protein urmylation in vivo. Urm1 conjugation to target proteins in general, and to the evolutionary conserved substrate Peroxiredoxin 5 (Prx5) specifically, is dependent on Uba4. A complete loss of Urm1 is lethal in flies, although a small number of adult zygotic Urm1 (n123) mutant escapers can be recovered. These escapers display a decreased general fitness and shortened lifespan, but in contrast to their S. cerevisiae counterparts, they are resistant to oxidative stress. Providing a molecular explanation, we demonstrate that cytoprotective JNK signaling is increased in Urm1 deficient animals. In agreement, molecular and genetic evidence suggest that elevated activity of the JNK downstream target genes Jafrac1 and gstD1 strongly contributes to the tolerance against oxidative stress displayed by Urm1 (n123) null mutants. In conclusion, Urm1 is a UBL that is involved in the regulation of JNK signaling and the response against oxidative stress in the fruit fly.

Molecular evolutionary and structural analysis of the cytosolic DNA sensor cGAS and STING.

Cyclic GMP-AMP (cGAMP) synthase (cGAS) is recently identified as a cytosolic DNA sensor and generates a non-canonical cGAMP that contains G(2',5')pA and A(3',5')pG phosphodiester linkages. cGAMP activates STING which triggers innate immune responses in mammals. However, the evolutionary functions and origins of cGAS and STING remain largely elusive. Here, we carried out comprehensive evolutionary analyses of the cGAS-STING pathway. Phylogenetic analysis of cGAS and STING families showed that their origins could be traced back to a choanoflagellate Monosiga brevicollis. Modern cGAS and STING may have acquired structural features, including zinc-ribbon domain and critical amino acid residues for DNA binding in cGAS as well as carboxy terminal tail domain for transducing signals in STING, only recently in vertebrates. In invertebrates, cGAS homologs may not act as DNA sensors. Both proteins cooperate extensively, have similar evolutionary characteristics, and thus may have co-evolved during metazoan evolution. cGAS homologs and a prokaryotic dinucleotide cyclase for canonical cGAMP share conserved secondary structures and catalytic residues. Therefore, non-mammalian cGAS may function as a nucleotidyltransferase and could produce cGAMP and other cyclic dinucleotides. Taken together, assembling signaling components of the cGAS-STING pathway onto the eukaryotic evolutionary map illuminates the functions and origins of this innate immune pathway.

A widespread bacteriophage abortive infection system functions through a Type IV toxin-antitoxin mechanism.

Bacterial abortive infection (Abi) systems are 'altruistic' cell death systems that are activated by phage infection and limit viral replication, thereby providing protection to the bacterial population. Here, we have used a novel approach of screening Abi systems as a tool to identify and characterize toxin-antitoxin (TA)-acting Abi systems. We show that AbiE systems are encoded by bicistronic operons and function via a non-interacting (Type IV) bacteriostatic TA mechanism. The abiE operon was negatively autoregulated by the antitoxin, AbiEi, a member of a widespread family of putative transcriptional regulators. AbiEi has an N-terminal winged-helix-turn-helix domain that is required for repression of abiE transcription, and an uncharacterized bi-functional C-terminal domain, which is necessary for transcriptional repression and sufficient for toxin neutralization. The cognate toxin, AbiEii, is a predicted nucleotidyltransferase (NTase) and member of the DNA polymerase ß family. AbiEii specifically bound GTP, and mutations in conserved NTase motifs (I-III) and a newly identified motif (IV), abolished GTP binding and subsequent toxicity. The AbiE systems can provide phage resistance and enable stabilization of mobile genetic elements, such as plasmids. Our study reveals molecular insights into the regulation and function of the widespread bi-functional AbiE Abi-TA systems and the biochemical properties of both toxin and antitoxin proteins.

GDP-D-mannose pyrophosphorylase from Pogonatherum paniceum enhances salinity and drought tolerance of transgenic tobacco.

Pogonatherum paniceum is a highly drought- and salt-tolerant plant species that is typically used for ecological restoration and the conservation of soil and water in many countries. Understanding the molecular mechanisms underlying plant abiotic stress responses, especially to salinity and drought stresses, in species such as P. paniceum could be important to broader crop improvement efforts. GDP-D-mannose pyrophosphorylase (GMPase) is the limiting enzyme in the synthesis of L-ascorbic acid (AsA), which plays a crucial role in the detoxification of reactive oxygen species (ROS). We have cloned and characterized the cDNA of the PpGMP gene of P. paniceum encoding a GMPase. The full-length cDNA sequence contains 1411 nucleotides encoding a putative protein with 361 amino acid residues and an approximate molecular mass of 39.68 kDa. The GMPase transcript was up-regulated in P. paniceum plants subjected to salinity and drought stress, respectively. Transgenic tobacco expressing PpGMPase exhibited enhanced salinity and drought resistance, a higher seed germination rate, better growth performance, a higher AsA content, a more stable redox state, higher superoxide dismutase (SOD) activity, and lower levels of malonaldehyde (MDA) and H2O2 under drought and salinity stress. Taken together, expression of PpGMPase in tobacco conferred salinity and drought stress tolerance by increasing the content of AsA, thereby enhancing ROS-detoxifying functions. Thus, PpGMP is a potential candidate gene for crop improvement.

RNA decay via 3' uridylation.

The post-transcriptional addition of non-templated nucleotides to the 3' ends of RNA molecules can have a profound impact on their stability and biological function. Evidence accumulated over the past few decades has identified roles for polyadenylation in RNA stabilisation, degradation and, in the case of eukaryotic mRNAs, translational competence. By contrast, the biological significance of RNA 3' modification by uridylation has only recently started to become apparent. The evolutionary origin of eukaryotic RNA terminal uridyltransferases can be traced to an ancestral poly(A) polymerase. Here we review what is currently known about the biological roles of these enzymes, the ways in which their activity is regulated and the consequences of this covalent modification for the target RNA molecule, with a focus on those instances where uridylation has been found to contribute to RNA degradation. Roles for uridylation have been identified in the turnover of mRNAs, pre-microRNAs, piwi-interacting RNAs and the products of microRNA-directed mRNA cleavage; many mature microRNAs are also modified by uridylation, though the consequences in this case are currently less well understood. In the case of piwi-interacting RNAs, modification of the 3'-terminal nucleotide by the HEN1 methyltransferase blocks uridylation and so stabilises the small RNA. The extent to which other uridylation-dependent mechanisms of RNA decay are similarly regulated awaits further investigation. This article is part of a Special Issue entitled: RNA Decay mechanisms.

tRNAHis-guanylyltransferase establishes tRNAHis identity.

Histidine transfer RNA (tRNA) is unique among tRNA species as it carries an additional nucleotide at its 5' terminus. This unusual G(-1) residue is the major tRNA(His) identity element, and essential for recognition by the cognate histidyl-tRNA synthetase to allow efficient His-tRNA(His) formation. In many organisms G(-1) is added post-transcriptionally as part of the tRNA maturation process. tRNA(His) guanylyltransferase (Thg1) specifically adds the guanylyate residue by recognizing the tRNA(His) anticodon. Thg1 homologs from all three domains of life have been the subject of exciting research that gave rise to a detailed biochemical, structural and phylogenetic enzyme characterization. Thg1 homologs are phylogenetically classified into eukaryal- and archaeal-type enzymes differing characteristically in their cofactor requirements and specificity. Yeast Thg1 displays a unique but limited ability to add 2-3 G or C residues to mutant tRNA substrates, thus catalyzing a 3' → 5' RNA polymerization. Archaeal-type Thg1, which has been horizontally transferred to certain bacteria and few eukarya, displays a more relaxed substrate range and may play additional roles in tRNA editing and repair. The crystal structure of human Thg1 revealed a fascinating structural similarity to 5' → 3' polymerases, indicating that Thg1 derives from classical polymerases and evolved to assume its specific function in tRNA(His) processing.

Comparative analysis of a large dataset indicates that internal transcribed spacer (ITS) should be incorporated into the core barcode for seed plants.

A two-marker combination of plastid rbcL and matK has previously been recommended as the core plant barcode, to be supplemented with additional markers such as plastid trnH-psbA and nuclear ribosomal internal transcribed spacer (ITS). To assess the effectiveness and universality of these barcode markers in seed plants, we sampled 6,286 individuals representing 1,757 species in 141 genera of 75 families (42 orders) by using four different methods of data analysis. These analyses indicate that (i) the three plastid markers showed high levels of universality (87.1-92.7%), whereas ITS performed relatively well (79%) in angiosperms but not so well in gymnosperms; (ii) in taxonomic groups for which direct sequencing of the marker is possible, ITS showed the highest discriminatory power of the four markers, and a combination of ITS and any plastid DNA marker was able to discriminate 69.9-79.1% of species, compared with only 49.7% with rbcL + matK; and (iii) where multiple individuals of a single species were tested, ascriptions based on ITS and plastid DNA barcodes were incongruent in some samples for 45.2% of the sampled genera (for genera with more than one species sampled). This finding highlights the importance of both sampling multiple individuals and using markers with different modes of inheritance. In cases where it is difficult to amplify and directly sequence ITS in its entirety, just using ITS2 is a useful backup because it is easier to amplify and sequence this subset of the marker. We therefore propose that ITS/ITS2 should be incorporated into the core barcode for seed plants.

The ß-d-manno-heptoses are immune agonists across kingdoms.

Bacterial small molecule metabolites such as adenosine-diphosphate-d-glycero-ß-d-manno-heptose (ADP-heptose) and their derivatives act as effective innate immune agonists in mammals. We show that functional nucleotide-diphosphate-heptose biosynthetic enzymes (HBEs) are distributed widely in bacteria, archaea, eukaryotes, and viruses. We identified a conserved STTR5 motif as a hallmark of heptose nucleotidyltransferases that can synthesize not only ADP-heptose but also cytidine-diphosphate (CDP)- and uridine-diphosphate (UDP)-heptose. Both CDP- and UDP-heptoses are agonists that trigger stronger alpha-protein kinase 1 (ALPK1)-dependent immune responses than ADP-heptose in human and mouse cells and mice. We also produced ADP-heptose in archaea and verified its innate immune agonist functions. Hence, the ß-d-manno-heptoses are cross-kingdom, small-molecule, pathogen-associated molecular patterns that activate the ALPK1-dependent innate immune signaling cascade.

Evolutionary history of the GH3 family of acyl adenylases in rosids.

GH3 amino acid conjugases have been identified in many plant and bacterial species. The evolution of GH3 genes in plant species is explored using the sequenced rosids Arabidopsis, papaya, poplar, and grape. Analysis of the sequenced non-rosid eudicots monkey flower and columbine, the monocots maize and rice, as well as spikemoss and moss is included to provide further insight into the origin of GH3 clades. Comparison of co-linear genes in regions surrounding GH3 genes between species helps reconstruct the evolutionary history of the family. Combining analysis of synteny with phylogenetics, gene expression and functional data redefines the Group III GH3 genes, of which AtGH3.12/PBS3, a regulator of stress-induced salicylic acid metabolism and plant defense, is a member. Contrary to previous reports that restrict PBS3 to Arabidopsis and its close relatives, PBS3 syntelogs are identified in poplar, grape, columbine, maize and rice suggesting descent from a common ancestral chromosome dating to before the eudicot/monocot split. In addition, the clade containing PBS3 has undergone a unique expansion in Arabidopsis, with expression patterns for these genes consistent with specialized and evolving stress-responsive functions.

Comprehensive classification of nucleotidyltransferase fold proteins: identification of novel families and their representatives in human.

This article presents a comprehensive review of large and highly diverse superfamily of nucleotidyltransferase fold proteins by providing a global picture about their evolutionary history, sequence-structure diversity and fulfilled functional roles. Using top-of-the-line homology detection method combined with transitive searches and fold recognition, we revised the realm of these superfamily in numerous databases of catalogued protein families and structures, and identified 10 new families of nucleotidyltransferase fold. These families include hundreds of previously uncharacterized and various poorly annotated proteins such as Fukutin/LICD, NFAT, FAM46, Mab-21 and NRAP. Some of these proteins seem to play novel important roles, not observed before for this superfamily, such as regulation of gene expression or choline incorporation into cell membrane. Importantly, within newly detected families we identified 25 novel superfamily members in human genome. Among these newly assigned members are proteins known to be involved in congenital muscular dystrophy, neurological diseases and retinal pigmentosa what sheds some new light on the molecular background of these genetic disorders. Twelve of new human nucleotidyltransferase fold proteins belong to Mab-21 family known to be involved in organogenesis and development. The determination of specific biological functions of these newly detected proteins remains a challenging task.

A prototypical cytidylyltransferase: CTP:glycerol-3-phosphate cytidylyltransferase from bacillus subtilis.

BACKGROUND: The formation of critical intermediates in the biosynthesis of lipids and complex carbohydrates is carried out by cytidylyltransferases, which utilize CTP to form activated CDP-alcohols or CMP-acid sugars plus inorganic pyrophosphate. Several cytidylyltransferases are related and constitute a conserved family of enzymes. The eukaryotic members of the family are complex enzymes with multiple regulatory regions or repeated catalytic domains, whereas the bacterial enzyme, CTP:glycerol-3-phosphate cytidylyltransferase (GCT), contains only the catalytic domain. Thus, GCT provides an excellent model for the study of catalysis by the eukaryotic cytidylyltransferases. RESULTS: The crystal structure of GCT from Bacillus subtilis has been determined by multiwavelength anomalous diffraction using a mercury derivative and refined to 2.0 A resolution (R(factor) 0.196; R(free) 0.255). GCT is a homodimer; each monomer comprises an alpha/beta fold with a central 3-2-1-4-5 parallel beta sheet. Additional helices and loops extending from the alpha/beta core form a bowl that binds substrates. CTP, bound at each active site of the homodimer, interacts with the conserved (14)HXGH and (113)RTXGISTT motifs. The dimer interface incorporates part of a third motif, (63)RYVDEVI, and includes hydrophobic residues adjoining the HXGH sequence. CONCLUSIONS: Structure superpositions relate GCT to the catalytic domains from class I aminoacyl-tRNA synthetases, and thus expand the tRNA synthetase family of folds to include the catalytic domains of the family of cytidylyltransferases. GCT and aminoacyl-tRNA synthetases catalyze analogous reactions, bind nucleotides in similar U-shaped conformations, and depend on histidines from analogous HXGH motifs for activity. The structural and other similarities support proposals that GCT, like the synthetases, catalyzes nucleotidyl transfer by stabilizing a pentavalent transition state at the alpha-phosphate of CTP.

Monophyly of class I aminoacyl tRNA synthetase, USPA, ETFP, photolyase, and PP-ATPase nucleotide-binding domains: implications for protein evolution in the RNA.

Protein sequence and structure comparisons show that the catalytic domains of Class I aminoacyl-tRNA synthetases, a related family of nucleotidyltransferases involved primarily in coenzyme biosynthesis, nucleotide-binding domains related to the UspA protein (USPA domains), photolyases, electron transport flavoproteins, and PP-loop-containing ATPases together comprise a distinct class of alpha/beta domains designated the HUP domain after HIGH-signature proteins, UspA, and PP-ATPase. Several lines of evidence are presented to support the monophyly of the HUP domains, to the exclusion of other three-layered alpha/beta folds with the generic "Rossmann-like" topology. Cladistic analysis, with patterns of structural and sequence similarity used as discrete characters, identified three major evolutionary lineages within the HUP domain class: the PP-ATPases; the HIGH superfamily, which includes class I aaRS and related nucleotidyltransferases containing the HIGH signature in their nucleotide-binding loop; and a previously unrecognized USPA-like group, which includes USPA domains, electron transport flavoproteins, and photolyases. Examination of the patterns of phyletic distribution of distinct families within these three major lineages suggests that the Last Universal Common Ancestor of all modern life forms encoded 15-18 distinct alpha/beta ATPases and nucleotide-binding proteins of the HUP class. This points to an extensive radiation of HUP domains before the last universal common ancestor (LUCA), during which the multiple class I aminoacyl-tRNA synthetases emerged only at a late stage. Thus, substantial evolutionary diversification of protein domains occurred well before the modern version of the protein-dependent translation machinery was established, i.e., still in the RNA world.

Evolutionary relationship between the bacterial HPr kinase and the ubiquitous PEP-carboxykinase: expanding the P-loop nucleotidyl transferase superfamily.

Similarities between protein three-dimensional structures can reveal evolutionary and functional relationships not apparent from sequence comparison alone. Here we report such a similarity between the metabolic enzymes histidine phosphocarrier protein kinase (HPrK) and phosphoenolpyruvate carboxykinase (PCK), suggesting that they are evolutionarily related. Current structure classifications place PCK and other P-loop containing nucleotidyl-transferases into different folds. Our comparison of both HPrK and PCK to other P-loop containing proteins reveals that all share a common structural motif consisting of an alphabeta segment containing the P-loop flanked by an additional beta-strand that is adjacent in space, but far apart along the sequence. Analysis also shows that HPrK/PCK differ from other P-loop containing structures no more than they differ from each other. We thus suggest that HPrK and PCK should be classified with other P-loop containing proteins, and that all probably share a common ancestor that probably contained a simple P-loop motif with different protein segments being added or lost over the course of evolution. We used the structure-based sequence alignment containing residues specific to HPrK/PCK to identify additional members of this P-loop containing family.

Enzymatically active 2',5'-oligoadenylate synthetases are widely distributed among Metazoa, including protostome lineage.

2',5'-Oligoadenylate synthetases (OASs) belong to the nucleotidyl transferase family together with poly(A) polymerases, CCA-adding enzymes and the recently discovered cyclic-GMP-AMP synthase (cGAS). Mammalian OASs have been thoroughly characterized as components of the interferon-induced antiviral system. The OAS activity and the respective genes were also discovered in marine sponges where the interferon system is absent. In this study the recombinant OASs from several multicellular animals and their closest unicellular relative, a choanoflagellate, were expressed in a bacterial expression system and their enzymatic activities were examined. We demonstrated 2-5A synthesizing activities of OASs from the marine sponge Tedania ignis, a representative of the phylogenetically oldest metazoan phylum (Porifera), from an invertebrate of the protostome lineage, the mollusk Mytilus californianus (Mollusca), and from a vertebrate species, a cartilaginous fish Leucoraja erinacea (Chordata). However, the expressed proteins from an amphibian, the salamander Ambystoma mexicanum (Chordata), and from a protozoan, the marine choanoflagellate Monosiga brevicollis (Choanozoa), did not show 2-5A synthesizing activity. Differently from other studied OASs, OAS from the marine sponge T. ignis was able to catalyze the formation of oligomers having both 2',5'- and 3',5'-phosphodiester linkages. Our data suggest that OASs from sponges and evolutionarily higher animals have similar activation mechanisms which still include different affinities and possibly different structural requirements for the activating RNAs. Considering their 2'- and 3'-specificities, sponge OASs could represent a link between evolutionarily earlier nucleotidyl transferases and 2'-specific OASs from higher animals.

A family of poly(U) polymerases.

The GLD-2 family of poly(A) polymerases add successive AMP monomers to the 3' end of specific RNAs, forming a poly(A) tail. Here, we identify a new group of GLD-2-related nucleotidyl transferases from Arabidopsis, Schizosaccharomyces pombe, Caenorhabditis elegans, and humans. Like GLD-2, these enzymes are template independent and add nucleotides to the 3' end of an RNA substrate. However, these new enzymes, which we refer to as poly(U) polymerases, add poly(U) rather than poly(A) to their RNA substrates.

DNA polymerase C of the thermophilic bacterium Thermus aquaticus: classification and phylogenetic analysis of the family C DNA polymerases.

Bacterial family C DNA polymerases (DNA pol IIIs), the major chromosomal replicative enzymes, have been provisionally classified based on primary sequences and domain structures into three classes: class I (Escherichia coli DNA pol C-type), class II (Bacillus subtilis DNA pol C-type), and class III (cyanobacterial DNA pol C-type), respectively. We have sequenced the structural gene encoding the DNA pol C catalytic subunit of the thermophilic bacterium Thermus aquaticus. This gene, designated the Taq DNA pol C gene, contains a 3660-bp open reading frame which specifies a polypeptide of molecular weight of 137,388 daltons. Comparative sequence analyses revealed that Taq DNA pol C is a class I family C DNA polymerase. The Taq DNA pol C is most closely related to the Deinococcus radiodurans DNA pol C. Although a phylogenetic tree based on the class I family C DNA pols is still in the provisional stage, some important conclusion can be drawn. First, the high-G+C and the low-G+C Gram-positive bacteria are not monophyletic. Second, the low-G+C Gram-positive bacteria contain multigenes of family C DNA pols (classes I and II). Third, the cyanobacterial family C DNA pol, classified as class III because it is encoded by a split gene, forms a group with the high-G+C Gram-positive bacteria.

The hyperthermophilic bacterium Thermotoga maritima has two different classes of family C DNA polymerases: evolutionary implications.

Bacterial DNA polymerase III (family C DNA polymerase), the principal chromosomal replicative enzyme, is known to occur in at least three distinct forms which have provisionally been classified as class I ( Escherichia coli DNA pol C-type), class II ( Bacillus subtilis DNA pol C-type) and class III (cyanobacteria DNA pol C-type). We have identified two family C DNA polymerase sequences in the hyperthermophilic bacterium Thermotoga maritima. One DNA polymerase consisting of 842 amino acid residues and having a molecular weight of 97 213 belongs to class I. The other one, consisting of 1367 amino acid residues and having a molecular weight of 155 361, is a member of class II. Comparative sequence analyses suggest that the class II DNA polymerase is the principal DNA replicative enzyme of the microbe and that the class I DNA polymerase may be functionally inactive. A phylogenetic analysis using the class II enzyme indicates that T.maritima is closely related to the low G+C Gram-positive bacteria, in particular to Clostridium acetobutylicum, and mycoplasmas. These results are in conflict with 16S rRNA-based phylogenies, which placed T.maritima as one of the deepest branches of the bacterial tree.