Queuosine Salvage in Bartonella henselae Houston 1: A Unique Evolutionary Path

Queuosine (Q) stands out as the sole tRNA modification that can be synthesized via salvage pathways. Comparative genomic analyses identified specific bacteria that showed a discrepancy between the projected Q salvage route and the predicted substrate specificities of the two identified salvage proteins: 1) the distinctive enzyme tRNA guanine-34 transglycosylase (bacterial TGT, or bTGT), responsible for inserting precursor bases into target tRNAs; and 2) Queuosine Precursor Transporter (QPTR), a transporter protein that imports Q precursors. Organisms like the facultative intracellular pathogen Bartonella henselae, which possess only bTGT and QPTR but lack predicted enzymes for converting preQ1 to Q, would be expected to salvage the queuine (q) base, mirroring the scenario for the obligate intracellular pathogen Chlamydia trachomatis. However, sequence analyses indicate that the substrate-specificity residues of their bTGTs resemble those of enzymes inserting preQ1 rather than q. Intriguingly, mass spectrometry analyses of tRNA modification profiles in B. henselae reveal trace amounts of preQ1, previously not observed in a natural context. Complementation analysis demonstrates that B. henselae bTGT and QPTR not only utilize preQ1, akin to their Escherichia coli counterparts, but can also process q when provided at elevated concentrations. The experimental and phylogenomic analyses suggest that the Q pathway in B. henselae could represent an evolutionary transition among intracellular pathogens—from ancestors that synthesized Q de novo to a state prioritizing the salvage of q. Another possibility that will require further investigations is that the insertion of preQ1 has fitness advantages when B. henselae is growing outside a mammalian host.


Introduction
tRNA modifications fine-tune translation by various mechanisms such as modulating the efficiency or accuracy of translation or affecting tRNA stability (1).Recent studies have revealed that modifications can play key roles in bacterial pathogenesis (2,3).Queuosine (Q) is a modification of the wobble base in tRNAs with GUN anticodons of many bacteria and eukaryotes that can affect both translational efficiency and accuracy depending on the organism (4,5).The in vivo significance of this modification has remained enigmatic for decades as it has been lost repeatedly during evolution (6), but recent studies have suggested that it may act as a regulatory component in the translation of proteins derived from genes enriched in TAT codons compared with TAC codons (7,8).In bacteria, Q modification was shown to have roles in oxidative stress, metal homeostasis, and virulence (9)(10)(11)(12).
Q is synthesized from guanosine triphosphate (GTP) by bacteria in a complex eight-step pathway fully elucidated in E. coli (Fig. 1A).Four enzymes (GCHI, QueD, QueE, and QueC) convert GTP into 7cyano-7-deazaguanine (preQ0).QueF then reduces preQ0 to 7-aminomethyl-7-deazaguanine (preQ1) that is inserted into tRNAs by tRNA guanine-34 transglycosylase (bacterial TGT, or bTGT) (4).The inserted base preQ1 is converted to Q by two additional steps involving QueA and QueG or QueH, depending on the organism (4,13).It should be noted that Q is the only tRNA modification that can be salvaged or recycled (5).Eukaryotes only use the salvage route, and their TGT enzyme (eTGT), a heterodimeric QTRT1/QTRT2 complex, incorporates the queuine (q) base directly into the target tRNAs (5).The salvage routes in bacteria vary greatly.Some organisms lack the preQ0 or preQ1 pathway genes but encode all the downstream genes and import these precursor bases to finalize Q synthesis in vivo (14).This salvage route is also observed in organisms like E. coli that can synthesize Q de novo.Other bacteria lack all canonical synthesis genes except tgt and can salvage q like eukaryotes.Two q salvage routes have been identified in bacteria, to date (15), including the direct salvage route found in the intracellular pathogen Chlamydia trachomatis, or the indirect salvage route found in the gut microbe Clostridioides difficile (Fig. 1B and C, respectively).In the direct route, the substrate specificity of the C. trachomatis bTGT enzyme has shifted to insert q instead of preQ1 like most bacterial homologs (15).In the indirect route, a recently discovered radical enzyme (QueL) can regenerate the preQ1 intermediate from a q precursor that is imported directly or derived from the hydrolysis of the Q nucleoside by QueK (15).Only a few transporters involved in Q salvage pathways have been identified and experimentally characterized.The first was the YhhQ /COG1738 family, now renamed QPTR (Queuosine Precursor Transporter), involved in the transport of preQ0 and preQ1 in E. coli and q in C. trachomatis (14,15).Members of the Energy-Coupling Factors (ECFs) family had been predicted to be involved in preQ0 and preQ1 transports (15)) and two of the C. difficile specificity components (or ECF-QueT) were shown to transport a variety of Q precursors in a reconstituted system (15).
The proportion of Q-modified tRNAs can change with the developmental stage in several eukaryotic parasites that undergo complex cycles that switch between hosts such as Trypanosoma cruzei and Entamoeba histolytica (16,17).Very little is known about the role of Q in bacterial pathogens that also switch between mammalian and insect hosts such as the facultative intracellular pathogen Bartonella henselae.This bacterium uses fleas and possibly ticks as vectors during blood feeding.Feces of these insects can also infect cats when they are scratched into a break in the skin.Once inside the cat, the bacteria enter the bloodstream, primarily residing within endothelial cells, where they multiply (18).B. henselae exhibits a high level of heterogeneity (19) and 16S rRNA sequence analyses led to the identification of two serotypes, Houston-1, and Marseille (20,21).Metabolic reconstruction of the Q synthesis pathway of B.
henselae Houston 1 suggested it utilized a direct q salvage pathway like C. trachomatis.However, analysis of B. henselae bTGT and QTPR sequences did not match this prediction as these were more similar to the bTGT and QPTR enzymes that recognize preQ1.To resolve this discrepancy, we set out to characterize experimentally the Q salvage enzymes in this facultative intracellular pathogen.

Metabolic reconstruction and sequence analyses of queuosine salvage genes give contradictory results.
We previously showed that QPTR proteins have different substrate specificities, shifting from preQ0 and preQ1 in E. coli to q in C. trachomatis (14,15).To better understand the molecular determinants that drive this change in specificity, we constructed a Sequence Similarity Network (SSN) of the QPTR family (PF02592).We then colored the SSN based on the presence/absence of the Q synthesis genes in the corresponding genome as an indirect way to predict the substrate specificity of a given QPTR protein.As shown in Fig. 2, we were able to generate an SSN that separated the QPTR's predicted to salvage preQ0 and preQ1 (in yellow or red) from those predicted to salvage q (in blue).However, a few exceptions stood out even at a stringent alignment score of 70.The QPTRs found in organisms that encode only tgt, and hence predicted to salvage q, cluster with QPTR proteins predicted to salvage preQ0/ preQ1 because their genomes encode QueA and QueG or QueH (circled in Fig. 2).These include QPTR proteins from Bartonella species such as B. henselae Houston-1 (UniProt id A0A0H3M726_BARHE).
Following up on this discrepancy, we retrieved all Q synthesis proteins from the InterPro database with their IPR protein family ID (see Materials and Methods) and counted the numbers of each protein encoded in each genome at the level of different taxonomic ranks, including order, class, and family (Table S1).Then, we extracted the TGT sequences from all orders, classes, families, and genera that encode only tgt as described in the methods section, aligned them, and compared the predicted substrate-binding residues (Table S2).As we previously reported (15), TGTs that salvage q typically contains GG[LS][AS]G in the substrate-binding pocket (Fig. 3).Interestingly, bTGTs in Bartonella and Pelagibacter contain a GGLAVG site like the E. coli enzyme (Fig. 3) and are phylogenetically distant from other bTGTs found in genomes that only encode tgt (Fig. 3).We then examined the substrate-binding pocket in a modeled structure of B. henselae bTGT (Bh bTGT, UniProt ID A0A0R4J8M4_BARHE) aligned with the structure of the human TGT catalytic subunit QTRT1 in complex with q (PDB ID: 6H45) (22).The aspartate residues and G216GLAVGE222 that are conserved in bTGT proteins of the Hyphomicrobiales order are in proximity to queuine (Fig. S1B and C).The predicted distance between V220 and the cis-diol groups of q is less than 1 Angstrom (Fig. S1C), suggesting that it may prevent the binding of q to B. henselae bTGT as in the Zymomonas mobilis TGT (23).We previously showed that C. trachomatis TGT salvages q and that its substrate-binding pocket can accommodate the larger substrate (15).Here, we propose that the bTGT protein and the QPTR transporter in Bartonella and Pelagibacter species salvage preQ1, even though they lack, like C. trachomatis, the enzymes that make Q-tRNAs from preQ1-tRNAs.
B. henselae bTGT and QPTR proteins preferentially salvage preQ1 but also q and Q with low affinity.
The gene encoding the B. henselae Bh bTGT protein complemented the Q-phenotype of a queDF tgt deletion mutant of E. coli when expressed in trans and in the presence of exogenous preQ1 even in low concentrations (down to 10 nM) (Fig. 4A).Similarly, Bh QPTR transported both preQ1 and preQ0 in an E. coli strain auxotrophic for preQ1 and preQ0 (Fig. 4B).These results confirmed our predictions based on the SSN that the Bartonella bTGT and QPTR proteins use preQ1 as a substrate but do not match the metabolic reconstruction that predicted q salvage in B. henselae.Hence, we tested if this organism's bTGT and QPTR could have evolved a broader substrate specificity and use q as a substrate.
To test whether the QPTR and bTGT proteins of B. henselae can use q as a substrate, we used an E. coli strain that expresses the QPTR and/or the bTGT proteins of C. trachomatis that can only use q as a substrate (Fig. 1) (15).Salvage of q was observed when expressing both Bh tgt and yhhQ genes at concentrations of q over 500 nM (Fig. 4C).When overexpressing the E. coli yhhQ gene, no such salvage of q was observed even at concentrations of 5 M (Fig. 4C).However, if the C. trachomatis bTGT and QPTR proteins can salvage q when present at 100 nM, the Bh QPTR cannot (Fig. S2).These results showed that the Bh QPTR protein has acquired the capacity to transport q while retaining the preQ1/preQ0 specificity, but it is still not as efficient as the C. trachomatis QPTR transporter that is specific for q.
We then tested whether the Bh QPTR could transport the Q ribonucleoside by modeling with an E. coli strain expressing the C. difficile Q hydrolase (Cd queK) and q lyase (Cd queL) genes that allow Q to be salvaged by E. coli (Fig. 1) (15).Expressing only Bh QPTR allowed the salvage of Q only at extremely high concentrations (5 M) (Fig. 4D).When the C. difficile Q transporter (Cd ECF_TAA', QueT) is expressed in this strain, Q can be salvaged when present at concentrations of 100 nM [( 15) and (Fig. 4D].

Traces of preQ1 can be detected in endogenous B. henselae tRNAs
The natural habitat of this intracellular pathogen (mammalian vasculature) should be richer in q than in preQ1.We, therefore, analyzed by LC-MS/MS the tRNA modification profile of bulk tRNA extracted from B. henselae cells grown in sheep blood agar (HIBB) medium.The experiments were performed three times independently with conflicting results.The first two experiments were done to test the tRNA extraction protocols with an intracellular bacterium, using only one sample each time.Small amounts of preQ1 and minute amounts of Q were detected the first time (but neither was present the second time (data not shown).
Because cells were grown in the presence of sheep blood and serum, we had little control over the sources of Q or preQ1; we, therefore, repeated the experiment a third time with 5 independent samples, adding 100 nM preQ1 in three of the samples.As shown in Fig. 5, Fig. S3 and Table S3, small amounts of preQ1 and Q were detected in all samples (around 1000 times less than in a typical E. coli sample).The exogenous addition of 100 nM preQ1 did not make any difference.It is not possible to determine if the observed Q was derived from B. henselae tRNAs or from contaminating mammalian tRNAs, as eukaryotic-specific tRNA modifications such as m 2 2G, are detected in similar quantities (Table S3) and great variations in Q levels were observed between samples.Nevertheless, the presence of preQ1 cannot be explained by any contamination as mammalian host tRNAs never harbor this modification and other bacteria would not accumulate it, as preQ1 had only been detected previously in queA mutants of E. coli (24).The low amounts detected suggest that the tRNAs are not fully modified, as the preQ1 levels are 11%-40% of cmnm 5 s 2 U levels and 1.1%-2.8% of k 2 C levels; two well-conserved bacterial modifications (25)(Fig.S4 and Table S3).
Early loss of Q pathway genes in the Bartonellaceae family within the Hyphomicrobiales order B. henselae belongs to the Alphaproteobacteria class (25).This is a diverse Gram-negative taxon comprised of several phototrophic genera, several genera metabolizing C1-compounds (e.g., Methylobacterium spp.), symbionts of plants (e.g., Rhizobium spp.), endosymbionts of arthropods (Wolbachia) and intracellular pathogens (e.g., Rickettsia) (26).To better understand the evolution of the Q synthesis and salvage pathway in Alphaproteobacteria, we performed a phylogenetic distribution analysis of the corresponding genes in 2,127 different species with complete genome sequences from the class in the BV-BRC database as described in the methods section (Fig. S5).The tree suggests that there were three events involving the loss of preQ1 synthesis genes: one occurred after the split between Brucella and Bartonella species (Fig. 6 and ).This analysis revealed that most Alphaproteobacteria are prototrophic for Q, as 1,414 (64%) of the species analyzed encode the complete pathway.In addition, if the loss of the preQ1 synthesis genes occurred sporadically in different branches, all the organisms analyzed, with the exception of Bartonella, harbored tgt, queA, and queG genes and hence were not predicted to salvage q.Focusing more specifically on the Bartonella genus using a similar analysis revealed a very different pattern (Fig. 6).
Among the 65 organisms in the Bartonella genus with available complete genomes, as of October 2023, only Bartonella sp.HY038, branching at the root of the genus, encoded the canonical de novo Q synthesis pathway.Most, like B. henselae, have lost all the genes but tgt.In addition, a more in-depth analysis revealed fragmented tgt genes of several B. quintana strains (boxed in Fig. 6), suggesting these organisms have totally lost the capacity to make Q-modified tRNAs.This evolutionary scenario seems to be a recurring theme in intracellular bacteria like the Rickettsiales.While nearly all rickettsiae branching closer to the root retain the full pathway except for queD (collapsed in Fig. S6), other rickettsiae such as Anaplasma, Ehrlichia, and Wolbachia, have lost nearly the full pathway.However, cases of fragmented tgt genes are rare and only observed in a Wolbachia endosymbiont of Cimex lectularius (box in Fig. S5).
In summary, the phylogenic distribution analysis suggests that the direct ancestor of Bartonella species must have harbored the full Q pathway but that it was lost very early in the evolution of this clade.
Most bacteria in this clade are predicted to transport and insert preQ1 but without further conversion to Q.
In addition, some species like B. quintana have lost the pathway completely (Fig. 6).

Discussion
Q is an ancient modification predicted to be present in the ancestors of bacteria (27).It is still present in most extant bacteria even if minimalist genetic codes can exist without this complex modification (28).
Independent analyses of the genomes of bacteria in the human microbiome have shown that 90 to 95% of these organisms maintain the capacity to synthesize or salvage Q with around half encoding the full synthesis pathway (15,29).Many of the bacteria that have lost Q are organisms that have undergone a genome reduction process, where their genetic material has been streamlined over evolutionary time such as the parasitic Mycoplasma spp. or insect endosymbionts such as Riesia pediculicola (30,31).
Obligate intracellular human pathogens tend to have reduced genomes compared to their free-living ancestors as their metabolisms have adapted to a nutritionally rich niche (32).Regarding Q, the scenarios that one can envision in the transition to a strict intracellular lifestyle with access to the queuine precursor from the mammalian host are: 1) keeping the ancestral pathway; 2) losing the modification; or 3) switching to a queuine salvage route.We performed the metabolic reconstruction of Q metabolism in genera of strict intracellular human pathogens such as Chlamydia spp., members of the order Rickettsiales (Anaplasma spp., and Rickettsia spp.) and Coxiella burnetii (33,34).Indeed, examples of these three possible paths were observed (Fig. 7): Rickettsia (Fig. 7A) and Coxiella (Table S5 line 267) have kept the full Q synthesis pathway; Anaplasma spp.(Fig. 7A), Borrelia (Fig. 7B), Ehrlichia spp.(Table S5  except tgt (Fig. 7C).
The situation seen in B. henselae is not commonly observed in other intracellular bacteria and no other organisms in the Hyphomicrobiales order seem to follow the same pattern (Fig. 7D).Indeed, the presence of bTGT and QPTR homologs and the absence of QueA and QueG or QueH (Fig. 6) would suggest that these organisms salvage q like C. trachomatis, but the corresponding enzymes have retained their substrate specificity towards preQ1 (Figs. 3 and 4).PreQ1 is not a molecule found in mammalian cells (Brian Green, personal communication), and the fact that we were able to detect a small proportion of tRNAs carrying that modification when extracted from B. henselae cells grown in HIBB suggests this pathway is functional even if the source of preQ1 remains a mystery and could be due to contamination with a preQ1 synthesizing organism, an unknown source of preQ1 in the culture medium, or from the activity of a yet-tobe-discovered q hydrolase (intracellular or possibly extracellular as q transport is not efficient as discussed below) (Fig. 1D).The low amount of preQ1 modification (Table S3) suggests it does not play an important role in decoding accuracy under these conditions.The fact that tgt gene decay is observed in several organisms in this clade such as B. quintana reinforces this idea.A primary difference between B. quintana and B. henselae is their reservoir ecology.B. quintana uses only humans as a reservoir, whereas B. henselae is more promiscuous and frequently isolated from both cats and humans (35).
Another intriguing finding of this study is that the QPTR and bTGT enzymes of B. henselae can use q as a substrate when present at high concentrations (>500 nM), whereas the E. coli orthologs cannot.
It is difficult to establish if such concentrations could be physiological, and we could not show with certainty that B. henselae tRNA extracted from HIBB-grown cells harbored Q.One can propose several evolutionary explanations for this broadening of substrate specificity of the B. henselae salvage proteins.In one, the enzyme and transporter specificities would become more relaxed in B. henselae as they are no longer under selective pressure to maintain efficient preQ1 selection, and we are observing an intermediate stage along the evolutionary loss of the whole pathway.An argument against that hypothesis is that if preQ1 synthesis genes and queA, queG and queH genes seem to have been lost very early in the clade, tgt is often the last maintained and we did not find any examples where tgt was lost with the other genes maintained, suggesting a fitness advantage of maintaining the tgt gene.In addition, both B. henselae QPTR and bTGT encoding genes are expressed and the proteins detected in vitro [see Table S5 of (36)].The other possibility is that we are observing a transition of a preQ1 salvage to a q salvage that is working poorly in human cells but could be efficient in an environment with more q/Q.Could the insect vector provide such an environment?
Answering these questions will require further studies, including additional quantitative data on q/preQ1 levels in different environments and tRNA modification profiles along the pathogen's life cycle.In summary, the study sheds light on the diverse and adaptable nature of queuosine metabolism in various bacteria, particularly in intracellular pathogens.The unique characteristics of Q salvage observed in B.
henselae raise intriguing questions about its role in different host/vector environments.Further investigations are warranted to unravel the complexities of Q salvage and its implications to Bartonella's virulence.

Comparative genomics and bioinformatics
The BLAST tools (37) S6.Complete genomes of Alphaproteobacteria with good quality were retrieved from BV-BRC (38).
For metabolic reconstruction analyses in each taxonomic rank, all protein members were retrieved from the InterPro database (41) using the following IPR family ID: FolE1, IPR001474; FolE2, IPR003801; QueD, IPR007115; QueE, IPR024924; QueC, IPR018317; QueF, IPR00029500; TGT, IPR004803; QueA, IPR003699; QueG, IPR004453; QueH, IPR003828.A universal single-copy small ribosomal protein uS2, (IPR001865) was used to estimate the number of organisms in each rank.The number of proteins per taxonomic rank was computed and the criteria used for filtering the groups ranks encoding just TGT were the following: 1) the number of TGT proteins was no less than 10 so we were not polluted with small taxonomic sample size; 2) the number of each of the QueDECFAGH proteins was no more than a fifth of the number of TGT proteins.To analyze the conserved residues in the substrate-binding pockets, the sequences of TGT proteins from select taxonomic groups were retrieved from UniProt and aligned using MUSCLE v5.1 (39).The conserved residues were visualized using weblogo3 (42).

Sequence Similarity Networks (SSNs)
The Enzyme Function Initiative (EFI) suite of web tools was used to generate the SSN (45).Visualization of SSNs was carried out using Cytoscape 3.10.1ape(46).7,625 PF02592 family sequences were retrieved from UniProt using the family option with fraction of 3 and submitted to EFI.The initial SSN was generated with an alignment score cutoff set such that each connection (edge) represents a sequence identity of above approximately 40%.The obtained SSN was first colored according to the configurations for salvaging preQ1, preQ0, queuine, and Queuosine de novo synthesis.Then more stringent SSNs were created by increasing the alignment score cutoff in small increments (usually by 5).This process was repeated until most clusters were homogeneous in their colors.The UniProt IDs were associated with the genome ID including GenBank/EMBL, RefSeq nucleotide, BV-BRC genome ID, Ensembl genome ID, using homemade scripts (scripts available upon request).The UniProt IDs of PF02592 family sequences in the SSN are listed in Table S7 as well as corresponding presence of Q pathway genes.The connection between UniProt IDs and genome information was performed by querying UniProt ID mapping file using homemade scripts (scripts available upon request).

Strains, media, and growth conditions
Strains and plasmids used in this study are listed in Table S9.LB medium (tryptone 10 grams/liter, yeast extract 5 grams/liter, sodium chloride 5 grams/liter) was routinely used for E. coli growth at 37°C.The medium was solidified using 15 g/L of agar.As needed, kanamycin (50 g/mL), ampicillin (100 g/mL), and chloramphenicol (25 g/mL) were added.In the presence of exogenous Q precursors as previously described (51), cells were cultured in M9-defined medium containing 1% glycerol (Thermo Fisher Scientific, Waltham, MA, USA) for the purpose of eliminating background Q-tRNA.After cells reached an optical density at 600 nm (OD600nm) of 0.1-0.2,0.2% arabinose was added to induce the expression of genes under the pBAD promoter.After cells reached an OD600nm of 0.2, DMSO, preQ0, preQ1, q, or Q were added.The transport reaction was stopped at time points of 30 or 60 min after supplementing with DMSO or different Q precursors by placing samples on melting ice and then centrifuging, followed by tRNA extraction.Q was purchased from Epitoire (Singapore), q from Santa Cruz Biotechnology, preQ1 and preQ0 from Sigma-Aldrich.
B. henselae Houston I was obtained from the American Type Culture Collection (ATCC 49882) and cultivated as previously described (52)  was added to a final concentration of 100 nM.Following harvest into ice-cold heart infusion broth, tRNA was collected from the bacterial cells.

Construction of E. coli strains and plasmids
B. henselae yhhQ gene (Bh yhhQ) was chemically synthesized (without optimization) in pTWIST _ Kan vector.Xbal and HindIII restrictions sites were added at the 5′ and 3′ ends, respectively (Twist Bioscience HQ) (Table S10).Bh yhhQ DNA sequence was amplified using two primers pairs (F_Bh  A0A0R4J8M4_BARHE, respectively.E. coli transformations were performed using the CaCl2 chemical transformation procedure (53).Transformants were selected on LB agar supplemented with ampicillin or chloramphenicol (100 µg/mL).The clones were validated through sequencing and PCR analyses using primers designed specifically for Bh yhhQ and Bh tgt genes.All primers used in this study are listed in Table S10.

tRNA extraction and migration
Cells were harvested by centrifugation at 16,000 x g for 2 minutes at 4°C.Immediately after pelleting, the cells were resuspended in 1 mL of Trizol (Thermo Fisher Scientific, Waltham, MA, USA).According to the manufacturer's instructions, small RNA was extracted with the PureLink Tm miRNA Isolation kit (Thermo Fisher Scientific, Waltham, MA, USA).50 μL of RNase-free water were used to elute the purified RNAs.Quantification of prepared tRNA was performed using a Nanodrop 1000 spectrophotometer.We loaded 150 ng of tRNAs per well on a denaturing 8 M urea, 8% polyacrylamide gel containing 0.5% 3-(Acrylamido) phenylboronic acid (Sigma-Aldrich) after resuspending in a 2X RNA Loading Dye (NEB).
Migration was performed in a mixture of 1X TAE at 4°C.With a wet transfer apparatus in 1X TAE at 150 mA at 4°C for 90 minutes, tRNAs were transferred onto a Biodyne B precut nylon membrane (Thermo Scientific).The membrane was UV irradiated in a UV crosslinker (Fisher FB-UVXL-1000) at a preset UV energy dosage of 120 mJ/cm2.The North2South Chemiluminescent Hybridization and Detection Kit (Thermo) was used to detect tRNA Asp .As the DIG Easy Hyb (Roche) drastically reduces the background noise, it was used as the initial membrane-blocking buffer instead of the North2South kit's membraneblocking buffer.Hybridization was done at 60°C, using the specific biotinylated primer for tRNA Asp GUC (14) (5' biotin-CCCTGCGTGACAGGCAGG 3' for E. coli added to a final concentration of 50 ng/mL.S9.Cells were grown in minimal media in the presence of exogenous preQ0, preQ1, q or Q as noted.DMSO was used as control when no deazapurine was supplemented.The comparative view of the corresponding truncated and full-length tgt gene variants in different Bartonella species in the tree (right).Red, tgt; yellow, queA; blue, genes encoding a transporter-like protein; black, hypothetical genes; gray, other genes.The fragmented tgt genes in B. quintana strains are boxed.
Gene IDs are provided in supplementary Table S4.Only taxonomic ranks that contain more than 15 TGTs are shown.

Fig. S5 node 1 )
Fig. S5 node 1), one occurred after the split of Pelagibacteraceae (Fig. S5 and Fig S6 node 2), and one yhhQ_XbaI_PBAD33 and R_Bh yhhQ_HindIII_PBAD33) by PCR with the addition of restrictions sites XbaI and HindIII at their 5′ and 3′ ends, respectively.Bh yhhQ was cloned into the XbaI and HindIII sites of pBAD33.B. henselae tgt (Bh tgt) was amplified by PCR from B. henselae genomic DNA using the KpnI-RBS-TGTBh-F_PBAD24 and TGTBh-SbfI-R _PBAD24 primers and cloned into the KpnI and SbfI sites of pBAD24.The UniProt IDs for Bh yhhQ and Bh tgt are A0A0H3M726_BARHE and

Figure 3 .
Figure 3.Comparison of the phylogenetic tree of taxonomic ranks of bacteria that harbor tgt only

Figure 5 .
Figure 5. Quantification of preQ1 and Q in B. henselae.

Figure 6 .
Figure 6.Phylogenetic analysis of Queuosine biosynthesis genes in Bartonellaceae.The clade of

Figure 7 .
Figure 7.The prevalence of Q pathway proteins in selected taxonomic ranks in Uniprot database.
line 52), and Chlamydia (Table S5 line 212), kept only tgt.Most species in Mycobacteriales have totally lost the pathway genes on HIBB agar plates [Bacto heart infusion agar (Becton, Dickinson, Sparks, MD) supplemented with 4% defibrinated sheep blood and 2% sheep serum (Quad Five, Ryegate, MT) by volume] for 4 days at 37 o C, 5% CO2 and 100% relative humidity.When required, preQ1 The blot was visualized by the iBright™ Imaging Systems.MgCl2 and 5 mM Tris-HCL buffer pH 8.0.The digestion mixture was incubated at 37 °C for 6 h.After digestion, all samples were analyzed by chromatography-coupled triple-quadrupole mass spectrometry (LC-MS/MS).For each sample, 600 ng of hydrolysate was injected for two technical replicates.Using synthetic standards, HPLC retention times of RNA modifications were confirmed on a Waters Acuity BEH C18 column (50 × 2.1 mm inner diameter, 1.7 µm particle size) coupled to an Agilent 1290 HPLC system and an Agilent 6495 triple-quad mass spectrometer.The Agilent sample vial insert was used.The HPLC system was operated at 25 °C and a flow rate of 0.3 mL/min in a gradient TableS11with Buffer A (0.02% formic acid in double distilled water) and Buffer B (0.02% formic acid in 70% acetonitrile).The HPLC column was coupled to the mass spectrometer with an electrospray ionization source in positive mode with the following parameters: Dry gas temperature, 200 °C; gas flow, 11 L/min; nebulizer, 20 psi; sheath gas temperature, 300 °C; sheath gas flow, 12 L/min; capillary voltage, 3000 V; nozzle voltage, 0 V. Multiple reaction monitoring (MRM) mode was used for detection of product ions derived from the precursor ions for all the RNA modifications with instrument parameters including the collision energy (CE) optimized for maximal sensitivity for the modification.Based on synthetic standards Signal intensities for each ribonucleoside were normalized by dividing by the sum of the UV signal intensities of the four canonical ribonucleosides recorded with an in-line UV spectrophotometer at 260 nm.