Hypermutator strains of Pseudomonas aeruginosa reveal novel pathways of resistance to combos of cephalosporin antibiotics and beta-lactamase inhibitors

To characterize mutations that emerged below CZA choice in Khil and colleagues [19], we carried out Illumina whole-genome sequencing (WGS) of advanced isolates. Three lineages (MPAO1-WT, MPAO1-mutS^Tn, and a comparator medical P. aeruginosa isolate, PT) have been grown in CZA concentrations starting from 0.5 μg/mL to 256 μg/mL (see Strategies and Khil and colleagues [19] for particulars). Liquid cultures that displayed detectable progress have been plated, and as much as 3 colonies per passage have been chosen for sequencing with a give attention to isolates straddling the clinically related CZA concentrations. Following high quality management, WGS knowledge from 303 isolates have been included within the evaluation, with a median of three.7 million reads and median protection of 51× per pattern (S1 and S2 Tables). To prioritize variants that have been doubtless targets of choice and to account for potential mutations launched throughout lab manipulations previous to sequencing in hypermutator strains, we targeted on SNVs that have been retained in terminal isolates as CZA resistance advanced. These “fastened” variants have been outlined as those who have been present in at the least 2 isolates of a given lineage together with at the least 1 terminal isolate, however that weren’t current within the parental (beginning) pressure of the lineage. For this goal, the set of variants current within the parental MPAO1-mutS^Tn was assessed by sequencing of the majority beginning bacterial inventory resolution and 6 isolates spanning 2 of the lineages on the finish of the primary passage day. This method was chosen to have in mind the excessive spontaneous mutation price in these lineages and revealed 54 “early” variants that have been dominant earlier than the top of the primary passage day (S1 Information, Strategies). Notably 50/50 (100%) of the early SNVs that occurred have been transitions, in line with a MutS deficiency-driven hypermutation spectrum [36–39].

We recognized a complete of 14, 21, and 139 fastened variants that emerged below CZA choice within the PT, MPAO1-WT, and MPAO1-mutS^Tn lineages, respectively, representing a imply of three.5, 5.3, and 46.3 fastened variants per lineage (Figs 1A and S1). In settlement with earlier work [36–39], fastened mutations within the MutS-deficient lineages demonstrated a powerful transition bias (119/120 SNV), in comparison with the WT lineages (5/11 SNV; Fisher’s precise check p-value < 0.0001). To guage the worldwide properties of the noticed fastened mutations, we utilized a comparative genomics method utilizing the Pseudomonas ortholog database and in contrast common amino acid identification versus gene conservation for the coding sequences with non-synonymous substitutions [40,41] (see Strategies). This evaluation demonstrated a broad distribution of mutated genes within the hypermutator lineages and confirmed that mutations weren’t confined to poorly conserved accent genes however have been distributed throughout genes in any respect conservation ranges (Fig 1B). Additional Gene Ontology evaluation of the variants within the hypermutator lineages demonstrated enrichment in plenty of purposeful lessons together with purine nucleotide metabolism and cell wall biosynthesis (S2 Fig).

Fig 1. Common traits of fastened genomic variants that emerged throughout CZA choice.

(A) Proportions and kinds of fastened variants acquired through the course of CZA adaptation for every genotype. The whole variety of fastened variants acquired throughout lineages per genotype is indicated in parenthesis. (B) Evolutionary conservation of genes with fastened variants acquired in MPAO1-mutS^Tn- (yellow) and MPAO1-WT (pink) CZA adaptation experiments within the Pseudomonas genus. Every P. aeruginosa PAO1 CDS is plotted as a dot representing common amino acid identification and the proportion of genomes containing the given CDS (breadth of protection) amongst a set of 100 full Pseudomonas genomes (see S6 Information and Strategies). Genes with out fastened mutation are represented in blue. The marginal density plots and histograms present the distribution of a corresponding conservation measure in all CDSs (grey) and genes containing fastened variants (blue), respectively. The underlying knowledge to generate this determine will be present in S2 Information. CZA, ceftazidime/avibactam; WT, wild kind.

https://doi.org/10.1371/journal.pbio.3001878.g001

Nearly all of fastened variants in lineages from each WT strains (16/21 in MPAO1-WT and 9/14 in PT) have been positioned inside genes beforehand described in affiliation with beta-lactam resistance, together with efflux pumps and their transcriptional regulators (mexR, MexA, MexB, nalD, mexE, mexF, oprN) or within the Ω-loop of the PDC chromosomal cephalosporinase (Fig 2 and S3 Desk). Of word, a 7 kilobase deletion involving 8 genes, together with a part of mexF, oprN, PA2496, PA2497, PA2498 (yahD), PA2499 (ykoA), PA2500 (cynX), PA2501, and a part of PA2502 (PAO1 Reference coordinates 2,812,525 to 2,819,849), emerged in passage 6 within the MPAO1-WT 1D lineage. Lineage 3B within the PT isolate acquired a 21-nucleotide deletion within the gene encoding the PDC beta-lactamase gene, which has beforehand been demonstrated to confer CZA and ceftolozane-tazobactam (C/T) resistance [27]. Mutations within the ATP-binding subunit of the Clp protease clpA, in addition to within the PhoP/PhoQ phosphorelay system have been additionally noticed in impartial lineages (clpA in all WT, PT, and mutS lineages besides WT lineage 1B; phoQ in lineages 1A, 1D, 2A, 2B, and 2D, see Fig 2 and S3 Desk).

Fig 2. Wild-type lineages demonstrated mutations in canonical resistance genes, whereas MMR-deficient lineages developed mutations in genes not beforehand related to cephalosporin resistance.

Fastened variants are plotted versus passage (horizontal axis tick marks) and CZA MIC (μg/mL), with a stuffed tile indicating presence of the variant on the corresponding passage. The shade of blue represents the proportion of isolates (allele frequency) in a given passage carrying the corresponding variant. As much as 3 isolates per lineage per passage underwent WGS. The sidebars to the best of every plot symbolize purposeful traits of the variant and its protein targets as follows: coding/noncoding (NC) variant; variant affect (IM) as predicted by SnpEff; conservation rating (CN) as relative conservation of a mutated AA place inside every protein outlined as a within-CDS percentile of the Jensen–Shannon divergence scores calculated over a set of 100 Pseudomonas genomes; breadth (BR) as the share of 100 Pseudomonas genomes wherein the CDS is current; common AAI of the CDS with respect to its orthologs. Genes marked with a Ψ image have been additionally mutated within the no antibiotic management experiments. Genes marked with a diamond ◆ are these for which experimental proof helps a task in third technology cephalosporin resistance. An asterisk (*) represents a passage with no sequenced isolates. FS = Frameshift variant. ΔOprN = 7-kb deletion together with the oprN gene (partial deletion of mexF, oprN, PA2496, PA2497, PA2498 (yahD), PA2499 (ykoA), PA2500 (cynX), PA2501 and a part of PA2502; PAO1 Reference coordinates 2,812,525–2,819,849). CZA MIC was decided by E-test. The underlying knowledge to generate this determine will be present in S2 Information. AAI, amino acid identification; CZA, ceftazidime/avibactam; MMR, mismatch restore; WGS, whole-genome sequencing.

https://doi.org/10.1371/journal.pbio.3001878.g002

We subsequent regarded on the fastened variants within the MPAO1-mutS^Tn lineages as CZA resistance advanced. Strikingly—and in distinction to the conduct of the WT isolates—virtually all fastened variants acquired as CZA resistance advanced by the medical breakpoint of 16/4 μg/mL have been positioned in genes that haven’t been beforehand described to be concerned in CZA or third technology cephalosporin resistance (Fig 2). Although mutations in these lineages didn’t happen within the “classical” resistance genes, we recognized 9 mutations throughout 6 genes for which some degree of prior experimental proof exists supporting a task in resistance to 3rd technology cephalosporins: PA1436 (mexN of MexMN efflux pump), acsB, ftsL, clpA, argJ, PA0478-PA0479 [30,42–47]. All of those mutations first appeared in passages with CZA MIC equal to or larger than the medical resistance breakpoint of 16 μg/mL, and thus contributed to resistance solely at these increased MICs. Notably, one of many 54 “early” variants current within the parental pressure of the MPAO1-mutS^Tn lineages was positioned inside MexB (W753R). This variant, mentioned extra totally under, was current by the top of the primary day of passage in all sequenced isolates.

Given the excessive mutation charges within the hypermutators, numerous background mutations that don’t enhance health mutations are anticipated to be co-selected together with mutations that enhance health. We thus count on that some proportion of the noticed fastened variants (presumably the bulk) didn’t contribute on to antibiotic resistance. As a primary step in the direction of understanding which subset of fastened variants within the MPAO1-mutS^Tn isolates have been these contributing to antibiotic resistance, we prolonged the comparative genomic evaluation by including measures of relative conservation and predicted affect at every place [40,48] (Fig 2, Strategies). This evaluation predicted a average to excessive affect for plenty of fastened variants within the MPAO1-mutS^Tn lineages, indicating plenty of potential purposeful targets. A further indicator of the significance of a particular variant or gene goal for CZA resistance is impartial choice and fixation in numerous lineages. General, 8 genes have been mutated in additional than 1 MPAO1-mutS^Tn lineage: PA1545 (hypothetical protein belonging to Pseudomonas ortholog group 3815) [40]; glucose-6-phosphate isomerase, pgi; pdtA (phosphate depletion regulated TPS companion A); the β-subunit of RNA polymerase, rpoB; mexW (subunit of the MexVW RND-type efflux pump); the phoQ subunit of the 2-component sensor PhoP/Q; ftsI (PBP3); and the ATP-binding subunit of the Clp protease, clpA (S3 Fig). A further noncoding variant was shared throughout the three lineages, positioned 82 nucleotides upstream from the MexVW operon (place 4903384 within the PAO1 reference). 5 of the variants in these 8 frequent targets (RpoB D521G, Pgi G164D, a frameshift in PA1545, MexW E36K, and a SNV 82nt upstream from MexV) have been equivalent between at the least 2 lineages, suggesting that these variants could have been current at low proportions within the heterogeneous beginning inhabitants from which every lineage was derived within the microevolution experiment, versus impartial de novo prevalence throughout adaptation. One other risk is early-stage cross-contamination between lineages within the experiment. We have been unable to seek out supportive genomic proof figuring out a cross-contamination occasion to clarify the sharing of those mutations, although this case could also be arduous to differentiate from a variant that’s shared by the starter populations at low frequency. However, the truth that these variants had initially low inhabitants frequency however subsequently turned dominant in a number of lineages below growing CZA choice means that they could have supplied health benefits.

To look at whether or not among the noticed variants could have been chosen by situations within the serial passaging experiment apart from antibiotic stress, we in contrast variants that occurred below CZA choice to the variants that occurred below the identical situations within the absence of CZA for 11 sequential passages. We discovered 5 shared genes with at the least 1 variant amongst isolates passaged with and with out CZA (S4 Desk). Out of 12 variants in these 5 shared genes (pdtA, amaB, PA2462, pilY, and rfaE), only one variant was equivalent between isolates passaged with and with out CZA (G insertion at place 1114264, within the amaB gene). The pdtA and PA2462 genes have unusually lengthy open studying frames (>10 kb), which will increase the probability of observing repeated mutations in these genes by likelihood and may clarify why mutations have been noticed in these genes below each situations. The 5 variants seen in pdtA, amaB, PA2462, and pilY consisted of quick indels in GC-rich contexts, which endure increased charges of mutation within the mismatch-repair poor P. aeruginosa isolates [49,50]. None of those variants have been shared between a number of advanced CZA-resistant lineages.

To discover whether or not the CZA-adapted MMR-deficient isolates demonstrated gross progress defects as a result of mutational accumulation, we measured progress curves in LB broth for CZA-evolved isolates at completely different factors alongside antibiotic adaptation from all 3 MPAO1-mutS^Tn lineages and 1 MPAO1-WT lineage for comparability (Strategies, S4 Fig). Whereas there’s a discount within the progress price in lineage 2D at passage 12 and a progressive pattern to slower progress alongside passages in lineage 2A, the expansion curves of most isolates weren’t considerably completely different from these of MPAO1-WT 1A.

Given the shocking discovering that the CZA-resistant MPAO1-mutS^Tn lineages largely lacked fastened mutations in genes recognized to be concerned in CZA and third technology cephalosporin resistance, we subsequent sought to look at whether or not the noticed mutations had resulted in overexpression of recognized resistance genes such because the MexAB-OprM efflux pump and PDC beta lactamase by sudden transcriptional management mechanisms. We thus carried out RNA-seq on 3 advanced CZA-resistant MPAO1-WT isolates representing 3 completely different WT lineages, in addition to on a collection of MPAO1-mutS^Tn isolates with completely different MICs over the clinically related vary from the three hypermutator lineages and their respective ancestors. At complete of 61 mid-log part RNA-seq datasets have been obtained from 22 isolates with a median of 6.1 million reads per transcriptome (S3 Information).

Comparability of transcriptomes demonstrated substantial variations between the advanced CZA-resistant MPAO1-WT and MPAO1-mutS^Tn isolates. Whereas the general numbers of differentially expressed genes (DEGs) weren’t considerably completely different between the terminal WT and MMR-deficient teams (S5 Fig), principal part evaluation (S6 Fig) demonstrated that their content material differed, and transcriptomes from the two teams of isolates have been nicely separated within the first 2 principal parts. These international variations in DEG content material between WT and MMR-deficient lineages can also be evident from a heatmap illustration (Fig 3A). Evaluation of expression of recognized resistance genes revealed up-regulation of the MexAB-OprM efflux pump parts within the MPAO-WT transcriptomes (mexA >1.5 and mexB >1.3 log₂ fold change (LFC) in all 3 WT lineages versus WT ancestor, p < 0.001) (Fig 3B). These expression modifications present an evidence for CZA resistance within the WT isolates that’s in line with each preliminary genomic evaluation and beforehand described mechanisms. In distinction, the MPAO1-mutS^Tn isolates didn’t present modifications in expression of mexA, mexB, oprM or in different beforehand described main resistance genes, together with oprD, ftsI, or the PDC beta-lactamase gene (Fig 3B). Moreover, there have been no baseline variations in expression of the 6 main resistance genes listed above between the MPAO1-WT and MPAO1-mutS^Tn, utilizing thresholds of p_alter < 0.01 and/or LFC >1 (S7 Fig). To slender down potential mechanisms conferring CZA resistance, we recognized 91 genes that have been persistently up-regulated and 142 persistently down-regulated in MPAO1-mutS^Tn lineages (S8 Fig). This evaluation revealed an vital element: The MexVW operon demonstrated elevated expression with early evolution of CZA resistance in all 3 MPAO1-mutS^Tn lineages, and this coincided with acquisition of fastened mutations on this operon, together with the E36K mutation in MexW with average to excessive conservation and affect scores (Fig 2). For comparability, we examined the expression of different RND transporter techniques in these isolates and located that MexVW was the one transporter that was overexpressed persistently throughout lineages (Fig 3C). This elevated expression of mexV and mexW was statistically vital when evaluating the MPAO1-mutS^Tn terminal isolates with the MPAO1-WT terminal isolates (LFC > 3; p_alter < 4 × 10⁻¹⁵). We additionally famous that mexG, mexH, and mexI have been down-regulated in all MPAO1-mutS^Tn lineages.

Fig 3. Dissimilar transcriptome profiles spotlight lack of involvement of main cephalosporin resistance genes in CZA resistance in MMR-deficient lineages and reveal MexVW overexpression.

Three lineages of MPAO1-WT and MPAO1-mutS^Tn have been chosen for transcriptomic evaluation and three organic replicates from 1 isolate per passage underwent RNA sequencing. (A) A clustered heatmap of the highest 750 most variable genes throughout CZA-resistant isolates is proven. Rlog normalized ranges of expression of every gene (rows) have been scaled inside every row. Isolates (columns) are grouped by lineage and organized by genotype with WT on the left and hypermutators on the best. Inside every lineage, isolates are organized from left to proper so as of accelerating variety of passages below CZA choice, with the corresponding MIC indicated on the backside. 1X and 2X symbolize the MPAO1-WT and MPAO1-mutS^Tn parental strains of the respective lineages. (B) LFC with respect to their corresponding ancestor of the terminal MPAO1-WT isolates (first 3 columns) and CZA-adapted MPAO1-mutS^Tn isolates (final 9 columns). The rows symbolize 6 main beta-lactam resistance genes in P. aeruginosa from 4 main resistance pathways (encoding the MexAB-OprM efflux pump, OprD outer membrane porin, PDC beta-lactamase, and penicillin-binding-protein 3 FtsI). Information for two differentially expressed RND efflux pumps within the MPAO1-mutS^Tn lineages (MexVW and MexGHI-OpmD) are moreover proven. (C) Expression ranges of RND efflux pumps and outer membrane porins in MPAO1-mutS^Tn lineages through the course of the CZA adaptation experiment. The vertical axis represents the LFC of expression ranges in comparison with the early isolates, and the horizontal axis exhibits the passage quantity. CZA MIC is indicated with coloration tiles below the axis. Strains are coloured and labeled if the respective gene reached a LFC > 1.0 within the terminal isolates of the lineage. The underlying knowledge to generate this determine will be present in S2 Information. CZA, ceftazidime/avibactam; MMR, mismatch restore; WT, wild kind.

https://doi.org/10.1371/journal.pbio.3001878.g003

Primarily based on the foregoing mixture of genomic and transcriptomic evaluation, we targeted on the MexVW efflux pump as probably contributing to CZA resistance within the hypermutators. To check the consequences of the two recognized mutations, we launched them individually and together into an MPAO1-WT pressure to generate 3 by-product strains: MPAO1 mexV -82:T>C, MPAO1 MexW E36K, and MPAO1 mexV -82:T>C + MexW E36K. We carried out susceptibility testing to quite a lot of lessons of antimicrobials and classes of beta lactams on all 3 strains (Figs 4A and S9 and S5 Desk). This demonstrated that every mutation individually conferred a 1.5- to 2-fold improve in CZA MIC, and the two mutations together conferred a statistically vital 4- to 6-fold improve in MICs to ceftazidime, CZA, C/T, and cefepime. Extra testing demonstrated no modifications in susceptibility to aminoglycosides, meropenem or fluoroquinolones with respect to WT MPAO1 (S5 Desk and S9 Fig). Normal exams of AMR have been supplemented with an evaluation of progress kinetics within the presence of CZA. These outcomes corroborated the power of engineered mutants to develop below increased CZA concentrations, with the double mutant with the ability to develop at CZA concentrations 4 occasions that of the wild kind (Fig 4B). The relative contributions of the mexV -82:T>C and MexW E36K to resistance to every of the examined antibiotics are displayed in Fig 4C. Prolonged antimicrobial susceptibility testing was additionally carried out on the advanced hypermutator isolates for comparability, with the caveat that interpretation is sophisticated by the expansion of satellite tv for pc colonies throughout the zones of inhibition (see S10 Fig and S6 Desk).

Fig 4. MexVW efflux pump mutations launched right into a WT MPAO1 pressure confer resistance to cephalosporins and cephalosporin/beta-lactamase mixture antibiotics.

(A) Gradient diffusion (E-test) MICs for MPAO1-WT, mexV -82:T>C, MexW E36K, and double mutant (n = 10 per antibiotic/genotype mixture). The containers show the median with the decrease and higher hinges equivalent to the primary and third quartiles. Brackets with asterisks above the plot point out statistical significance with respect to MPAO1-WT (Wilcoxon 2-sided p-value < 0.05). All pairwise comparisons between mutants have been statistically vital as nicely (S7 Desk). (B) Development curves of MexVW engineered strains at 37°C in LB broth with a spread of ceftazidime concentrations with fastened 4 μg/mL avibactam focus. Strains symbolize the median OD₆₀₀ of seven organic replicates per time level, with shaded envelope representing the usual error of the imply. (C) Scatterplot displaying the relative contribution of the mexV -82:T>C and MexW E36K mutations to resistance to every antibiotic. Every coloured level represents the imply ratio of gradient diffusion MIC within the corresponding mutant pressure to that within the parental MPAO1 pressure, with error bars symbolize the corresponding customary error of the imply. A dashed line of equality is included. The underlying knowledge to generate this determine will be present in S2 Information. AZN, aztreonam; CAZ, ceftazidime; CTX, cefotaxime; CZA, ceftazidime/avibactam; FEP, cefepime; MEM, meropenem; PIP, piperacillin; PIP/TAZ, piperacillin/tazobactam; C/T, ceftolozane/tazobactam; WT, wild kind.

https://doi.org/10.1371/journal.pbio.3001878.g004

Provided that mutations within the genes encoding the MexVW efflux pump haven’t generally been related to resistance to cephalosporin/beta-lactamase inhibitor combos, we sought to know the frequency with which these variants occurred in publicly out there P. aeruginosa genomes. BLAST was used to seek for sequence homologues of the mexW CDS and intergenic sequence upstream from mexV in a set of seven,492 P. aeruginosa genomes from the NCBI Pathogen Detection database [51] (see S8 Desk and Strategies). Neither of the two particular variants recognized in our examine was discovered on this search, and the neighboring amino acid and nucleotide sequences have been comparatively conserved (S8 Desk). The obvious rarity of the particular mutations uncovered on this adaptive evolution experiment means that sequence house could harbor a big universe of potential resistance mechanisms out there to hypermutators.

As famous above, the MPAO1-mutS^Tn pressure was discovered to vary from the MPAO1-WT pressure by 54 variants (50 SNVs and 4 indels), with 33 of those variants current at a majority frequency within the beginning inventory and a further 21 that occurred through the first passage, however earlier than CZA resistance appeared (S1 Information). All 50 of the SNVs have been transitions, in line with the anticipated consequence of MutS-deficient hypermutation. Although these mutations in combination don’t account for the CZA resistance that subsequently developed, it was potential that they could have performed unappreciated roles that influenced the probability of evolving resistance by main efflux pathways. Importantly, these mutations included a transition variant within the mexB gene within the MPAO1-mutS^Tn isolate (ensuing within the W753R substitution), and we hypothesized that this substitution could have inactivated the MexAB-OprM efflux pump, blocking additional evolution that might have used up-regulation of this transporter to confer resistance. To check whether or not this mutation would have been appropriate with the MexAB-OprM-based CZA resistance that in the end advanced within the WT strains, we launched it into 3 completely different strains: (1) an MPAO1-WT pressure; (2) an advanced CZA-resistant WT isolate with elevated MexAB-OprM expression (1C12-1); and (3) an engineered MPAO1 mexV -82:T>C + MexW E36K background. The aforementioned 1C12-1 isolate was obtained throughout serial CZA passage and incorporates an intergenic SNV between mexR and mexA (471932 C>T), upstream from the MexAB-OprM operon. This intergenic variant seems to be liable for the rise in MexAB-OprM expression seen on this lineage by RNA-seq (Fig 3B).

The introduction of the MexB W753R substitution considerably lowered the aztreonam MIC in all 3 strains, in line with disruption of MexAB-OprM operate. Within the MexAB-OprM-overexpressing 1C12-1 isolate, the introduction of MexAB W753R additionally decreased the CZA and ceftazidime MIC to pre-evolved WT baseline ranges (S11 Fig), suggesting that this isolate’s CZA and CAZ resistance have been fully as a result of MexAB-OprM overexpression, and that the MexB W753R mutation was totally inactivating. These findings help the speculation that MexB W753R was not appropriate with the evolutionary mechanisms the WT isolates used to generate resistance to CZA below choice, and means that this early inactivating mutation within the MexAB-OprM efflux pump, a main mediator of CZA resistance, probably drove an evolutionary seek for different resistance mechanisms. Curiously, the CZA and CAZ resistance weren’t affected by this mexB mutation when launched in a mexV -82:T>C + MexW E36K background (S11 Fig), offering additional help that MexVW can confer cephalosporine resistance by a MexAB-OprM-independent mechanism.

To guage how generally hypermutation phenotypes and MexAB-OprM efflux pump deficiency happen in P. aeruginosa, we used the NCBI Pathogen Genome Database to seek for disruptive mutations (e.g., frameshifts and cease codons, mutations that lead to full loss or extreme truncation of the affected protein) within the corresponding genes (see Strategies). Inactivation of plenty of completely different genes can lead to hypermutator phenotypes in P. aeruginosa, together with MMR genes mutS, mutL, and uvrD, base excision restore (BER) genes mutT, mutM and mutY, and others. We recognized severely disruptive variants in at the least one in all these genes in 260 out of 6,805 assemblies (3.8%) (S9 Desk). Provided that some isolates will be clonal, and this may inflate this estimate, a conservative decrease certain estimate of 155/6,805 assemblies (2.3%) was calculated assuming that each one repeated variants are clonal. We equally estimated the frequency of disruptive variants in mexB, mexA, or oprM that might be anticipated to lead to truncated or mistranslated proteins with purposeful inactivation of the MexAB-OprM efflux pump. We discovered 86 distinctive disruptive variants within the mexA gene, 148 within the mexB gene, and 33 within the oprM gene (S10 Desk). In complete, 629 (9.2%) isolates had a variant predicted to be severely disruptive in at the least 1 of the three genes leading to a nonfunctional MexAB-OprM complicated. Once more, a conservative decrease certain estimate of 267/6,820 (3.9%) assemblies was calculated assuming that each one isolates carrying a given variant represented a single clonal lineage.

A caveat to those calculations is that they’re primarily based solely on NCBI genome sequences with out purposeful verification, and we can not rule out sequencing errors and meeting artifacts, although the prevalence must be terribly excessive to clarify these findings. One other caveat is that isolates from cystic fibrosis respiratory specimens, the place hypermutators are frequent, could also be overrepresented within the genomic databases [4,12–18]. Nevertheless, even when hypermutators and MexAB-OprM efflux poor isolates are considerably much less frequent than our estimates of ≥2.3% and ≥3.9%, this may nonetheless counsel that the co-occurrence of hypermutation and preexisting MexAB-OprM purposeful deficiencies could also be considerably extra frequent than appreciated. We thus requested what number of assemblies within the database could be predicted to be each putative hypermutators and have a nonfunctional MexAB-OprM. Surprisingly, 91 out of 260 putative hypermutators (35%) have been additionally discovered to have at the least 1 sequence variant suggestive of an inactive MexAB-OprM, elevating the likelihood that hypermutation and MexAB-OprM inactivation could the truth is happen steadily collectively. As soon as once more, with a view to alter for clonality on this subset of assemblies, we conservatively assumed every particular person disruptive variant (n = 47) to symbolize a single clonal lineage, acquiring a decrease certain estimate of 47/260 (18%). This evaluation means that concurrent hypermutator phenotypes and MexAB-OprM efflux pump disruption coexist in a small, however simply detectable, proportion of the clinically related P. aeruginosa inhabitants. These isolates could also be anticipated to evolve different resistance mechanisms below choice as did the isolates within the experiments on this work. Nevertheless, detailed examination of resistance mechanisms in these isolates is restricted as a result of lack of susceptibility testing knowledge. Out of seven,492 isolates within the NCBI Pathogen Detection database, solely a small minority (149 genomes, 2%) had any antimicrobial susceptibility knowledge, with 27 (0.3%) having CZA susceptibility knowledge out there. Solely 7 out of the 260 putative hypermutator isolates as outlined above had AST phenotypic knowledge out there, and only one/7 had CZA knowledge. These knowledge don’t permit particular conclusions about resistance phenotypes in these isolates to be drawn.

P. aeruginosa pressure MPAO1 (MPAO1-WT) and its mutS transposon insertion mutant (MPAO1-mutS^Tn, isolate PW7149) have been obtained from the U. WA Colin Manoil Lab Two Allele Library [65]. An elevated spontaneous mutation price within the MPAO1-mutS^Tn isolate was confirmed experimentally by way of a rifampin reversion assay, as beforehand printed [19]. P. aeruginosa isolate “PT” was a multidrug resistant, CAZ-susceptible, non-hypermutator medical isolate from our establishment, as beforehand reported [19]. All strains have been prone to CZA at baseline (Fig 2).

In a beforehand printed examine [19], lineages of MPAO1-WT, MPAO1-mutS^Tn, and PT have been serially passaged as follows. For every pressure, the frozen shares have been inoculated on sheep blood agar (SBA, TSA with 5% sheep blood) (Remel, Lenexa, Kansas, United States of America) and incubated at 35°C for 18 to 24 h. A single colony was inoculated in 20 mL Luria–Bertani (LB) (BD Difco, Franklin Lakes, New Jersey, USA) broth and incubated at 35°C for 12 h in a shaking incubator at 250 RPM, after which diluted to an preliminary OD₆₀₀ of 0.05, equivalent to a CFU rely of roughly 1.5 × 10⁸ CFU/mL. This CFU rely ends in a comparatively massive preliminary important inhabitants dimension (roughly 1.5 × 10⁷ CFU) however was chosen to make sure strong survival of the inhabitants throughout passaging below antibiotic choice. Quadruplicate lineages from this inventory have been serially passaged on Luria–Bertani (LB) broth with a gradient of ceftazidime (Sigma-Aldrich, St. Louis, Missouri, USA) focus in doubling dilutions starting from 0.5 to 512 μg/mL with a hard and fast avibactam (Sigma-Aldrich, St. Louis, Missouri, USA) focus of 4μg/mL. The passage interval was roughly 24 h, for a complete of 20 passages. At every interval, cultures have been aliquoted from the very best antibiotic focus nicely with an OD₆₀₀ >0.2 after background subtraction, diluted by 1:100, and re-inoculated in contemporary media with antibiotic gradient. To verify mutant stability and to acquire a measure of MIC in pure tradition for a subset of consultant isolates, susceptibility testing was carried out by E-test (BioMerieux, Durham, North Carolina, USA) from 1 colony per lineage per passage after progress on SBA for twenty-four h following the E-test producer’s directions. As a no-selection management, an equivalent scheme with quadruplicate lineages per pressure was carried out on LB broth with no antibiotic and passaging after 1:100 dilution.

Three isolates from every lineage/passage proven in S4 Fig have been streaked for isolation from glycerol shares frozen at −80°C onto Remel TSA with 5% sheep blood agar and incubated at 35°C for 20 h. 1 / 4 of a ten mL inoculation loopful of colonies have been collected and suspended in contemporary LB broth. These cell mixtures have been used to inoculate 200 mL LB to a goal OD₆₀₀ of 0.01 as measured on a Biotek Cytation 5 instrument utilizing the Gen5 Microplate Reader and Imager Software program (Agilent Applied sciences, Santa Clara, California, USA). Hourly OD₆₀₀ measurements for every pressure have been collected at 37°C for 21 h and plotted utilizing GraphPad Prism v.8 software program (San Diego, California, USA).

All engineered mutants described in addition to the MPAO1-WT underwent testing with an prolonged panel of antibacterials. Ampicillin, aztreonam, cephalothin, cefoxitin, ceftolozane/tazobactam (C/T), cefepime, CZA, ceftazidime, piperacillin, piperacillin/tazobactam, and meropenem MICs have been obtained for n = 10 organic replicates with E-test for the MPAO1-WT, mexV -82:T>C, MexW E36K, and mexV -82:T>C + MexW E36K mutants. Testing with levofloxacin, ciprofloxacin, amikacin, tobramycin, and gentamicin was carried out in n = 5 organic replicates in these mutants by the Kirby–Bauer technique (Hardy Diagnostics Santa Maria, California, USA and BD Diagnostics, Franklin Lakes, New Jersey, USA). Moreover, the MPAO1-WT, MPAO1-WT + MexB W753R, MPAO1 1C12-1, MPAO1 1C12-1 + MexB W753R, mexV -82:T>C + MexW E36K, and mexV -82:T>C + MexW E36K + MexB W753R have been examined with CZA, ceftazidime, and aztreonam by E-test in n = 6 organic replicates. All susceptibility testing was carried out following CLSI tips [66]. Briefly, isolates have been plated from frozen into LB agar plates and incubated for twenty-four h at 35°C, after which ≥5 colonies per isolate have been resuspended in 0.85% NaCl to a turbidity of 0.5 McFarland and inoculated as a garden utilizing a sterile cotton-tip swab onto Mueller–Hinton agar plates (Hardy Diagnostics Santa Maria, California, USA and BD Diagnostics, Franklin Lakes, New Jersey, USA). After drying for 10 to fifteen min, antibiotic check strips and discs have been utilized to the agar floor with tweezers. Plates have been learn after incubation for 18 to twenty h at 35°C.

A desk displaying the discrepancies between measured CZA E-test MICs in a consultant isolate per passage and the antibiotic focus inhibiting progress in that passage through the antibiotic adaptation scheme is proven in S4 Information.

For every lineage, WGS was carried out on 3 particular person colony subcultures from passages straddling at the least the clinically related vary of CZA concentrations. Single colonies from the beginning MPAO1-mutS^Tn and MPAO1-WT shares have been subcultured for WGS as nicely. DNA extraction was carried out with easyMAG (Biomerieux, Durham, North Carolina, USA) or DNeasy Blood and Tissue equipment (QIAGEN, Hilden, Germany). DNA libraries have been ready with both Nextera XT, Nextera Flex (Illumina, San Diego, California, USA), New England Biolabs NEBNext or iGenomX Riptide (iGenomX, Carlsbad, California, USA) [67] strategies. Sequencing was carried out utilizing 3 Illumina platforms: MiSeq with 300-bp paired-end reads, iSeq 100, or HiSeq 2500 or NextSeq 550 with 150-bp paired-end reads. Reads have been aligned to PAO1 GenBank reference (RefSeq accession quantity NC_002516.2) for the MPAO1-WT and MPAO1-mutS^Tn; or our de novo draft meeting for the PT pressure [19], utilizing BWA v0.7.17 (https://github.com/lh3/bwa) with default parameters, and filtered for reads with “correct pairs” utilizing Samtools v. 1.6 (https://github.com/samtools/samtools). With the intention to have impartial affirmation of SNVs discovered within the terminal isolates, they have been moreover sequenced utilizing Nextera XT or NEBNext Extremely II library preparation. For this goal, terminal isolates have been outlined as isolates from the newest passage of CZA adaptation that have been recoverable for sequencing (S11 Desk).

One consultant isolate from every of 4 lineages per pressure (MPAO1-WT and MPAO1-mutS^Tn), passaged with out antibiotic choice was sequenced after 11 passages. DNA extraction was carried out with DNeasy Blood and Tissue equipment (QIAGEN, Hilden, Germany) and DNA libraries have been ready with Nextera Flex. WGS was carried out with an Illumina iSeq instrument (150-bp PE reads). Downstream variant evaluation was carried out in an equivalent means as with the CZA-adapted isolates. A listing of variants referred to as in these isolates will be discovered within the S5 Information.

A complete of 376 genomic DNA libraries from lineages passaged below CAZ choice have been initially sequenced. Libraries with fewer than 1,000,000 reads (iGenomX) or fewer than 200,000 reads (Nextera) have been excluded from evaluation. These completely different cutoffs have been used to account for the inherent distinction in homogeneity of protection between the library preparation strategies, in addition to the truth that Nextera libraries have been sequenced with 300-bp paired-end reads, versus 150-bp paired-end reads for the iGenomX method. High quality management was carried out with Picard Instruments v2.25.0 (https://broadinstitute.github.io/picard/) utilizing the CollectWgsMetrics and CollectAlignmentSummaryMetrics instruments, and samples with fewer than 90% of bases reaching at the least 5X protection or with <98% reads aligned to reference have been excluded from additional evaluation. A complete of 303 genomes that met the above standards have been used for all additional evaluation.

Variants have been analyzed utilizing the vcftools v.0.1.16 (https://github.com/vcftools/vcftools) and R v. 4.0 (https://www.r-project.org/) softwares. The beginning MPAO1-WT and MPAO1-mutS^Tn strains share 55 frequent variants relative to the PAO1 reference, which have been subtracted from evaluation. For the MPAO1-mutS^Tn lineages, a further set of 54 variants have been current by the top of the primary day of passage, denoted as “early variants” all through the textual content. These included 33 variants current within the MPAO1-mutS^Tn isolate beginning inventory tradition plus a further 21 variants that have been current by the top of the primary passage. The latter have been primarily based on sequencing 3 isolates every from passage 1 of lineages 2A and 2D (S1 Information). We assessed for giant deletions by analyzing learn counts aligning to every annotated gene within the PAO1 reference genome, normalized by gene size, pattern, and run protection. Areas with normalized learn rely values under 0.1 supported in terminal isolates have been thought of as signatures of true massive deletions and have been included within the evaluation.

Some genome areas had persistently low protection within the iGenomX libraries, leading to problem differentiating some true variants from technical artifact. To verify fastened variants recognized within the iGenomX libraries, genomic DNA libraries from the terminal isolates of every lineage have been moreover sequenced with a second library preparation technique (Nextera XT or NEBNext Extremely II). We targeted our evaluation on fastened variants, which have been outlined as variants current in (1) at the least 2 completely different passages or the ultimate passage inside a single lineage; and (2) current within the terminal isolates independently sequenced with Nextera XT or NEBNext Extremely II for the lineage. The traits of every fastened variant have been analyzed with respect to passage of look, pressure, and CZA E-test MIC utilizing the tidyverse (https://www.tidyverse.org/) R bundle. Genomic evaluation detected potential cross-contamination of MPAO1-mutS^Tn lineage 2B into 2C at passage quantity 4, affecting 12 isolates sequenced on this lineage. Although the ensuing lineage 2C then was passaged independently and advanced a 6 doubling dilution improve in CZA MIC over these passages, we selected to take away it from combination genomic evaluation at this stage as a conservative method.

Variant annotation was carried out with snpEff v. 4.3 (https://pcingola.github.io/SnpEff/) with default parameters, utilizing native PAO1 and PT reference databases generated from Genbank (.gbk) information, utilizing the “java -jar snpEff.jar construct” command. The NCBI PAO1 reference (RefSeq accession quantity NC_002516.2) was used for MPAO1-WT and MPAO1-mutS^Tn, whereas the.gbk file generated by the Prokka software program (https://github.com/tseemann/prokka) was used for the PT lineages. Variants have been then extracted and aggregated for evaluation with snpSift after transformation to 1 entry per line (command: $SNPEFF_HOME/scripts/vcfEffOnePerLine.pl | java -Xmx10g -jar $SNPSIFT_JAR extractFields—CHROM POS REF ALT “ANN[*].ALLELE” “ANN[*].EFFECT” “ANN[*].IMPACT:” “ANN[*].FEATURE” “ANN[*].FEATUREID” “ANN[*].CDS_POS” “ANN[*].CDS_LEN” “ANN[*].AA_POS” “ANN[*].AA_LEN” “ANN[*].DISTANCE” “ANN[*].ERRORS”).

The allele frequency for a hard and fast variant in a given passage was calculated as the share of isolates possessing the variant over the variety of isolates sequenced in that passage. Calculated conservation, Jensen–Shannon divergence rating, breadth, and variant affect scores have been calculated as follows. All protein sequences have been downloaded from the 100-genome Pseudomonas genome ortholog database (https://pseudomonas.com/rbbh/pairs) (S6 Information). MAFFT v7.475 was used to generate a number of sequence alignments for every of the PAO1 reference genome proteins with their respective orthologs. A beforehand described algorithm [48] was then used to assign every amino acid residue a rating primarily based on the Jensen–Shannon divergence and to calculate a percentile throughout all different residues in the identical protein. The “breadth” for every protein was outlined as the share of Pseudomonas isolates on this set of 100 genomes containing an ortholog of the given protein. The “Variant affect” parameter was collected from the SnpEff output, outlined roughly as: Excessive affect = frameshift or nonsense variants; Average affect = missense variants or in-frame deletions; Low affect = synonymous variants; Modifier = noncoding variants.

Amongst 139 fastened variants in MPAO1-MutSTn, we recognized 78 distinct genes with at the least 1 non-synonymous variant. To carry out purposeful annotation, the GO phrases for the whole set of PAO1 loci have been obtained from the Pseudomonas genome database [40]. A complete of 4,042 PAO1 loci have been annotated with GO phrases and have been appropriate for evaluation. We then utilized the topGO v2.46.0 R bundle [68] to carry out an enrichment evaluation utilizing the weight01 algorithm with the set of 78 mutated genes. Enriched GO phrases with a 1-tailed Fisher’s precise check p < 0.05 are proven on S2 Fig.

Twenty-two isolates from the in-vitro evolution experiment have been chosen for RNA-seq evaluation. The set included 2 × 0, ancestor pressure of the hypermutator lineages from the adaptive evolution experiment, and all terminal and intermediate isolates listed in S3 Information for every lineage (2A, 2B, and 2D). The ancestor of the WT lineages within the adaptive evolution experiment, 1 × 0, was additionally chosen, together with all terminal isolates in 3 WT lineages (1A, 1C, 1D).

Ultimate cDNA libraries have been quantified utilizing a Qubit 4.0 Fluorometer (Invitrogen, Waltham, Massachusetts, USA), and common fragment lengths have been decided utilizing the Agilent Fragment Analyzer (Agilent Applied sciences, Santa Clara, California, USA). Samples for which libraries had a measured dsDNA focus <5 ng/dL or common peak dimension <300 bp have been deemed insufficient for evaluation and re-harvested as described above. Ultimate libraries have been diluted to 4 nM for single-end sequencing on an Illumina NextSeq utilizing the NextSeq 500/550 Excessive Output cartridge v2.5 75-cycle equipment (Illumina, San Diego, California, USA), acquiring a complete of two to 13 million reads per pattern (S3 Information).

To quantify gene expression, the DESeq2 v1.30.1 bundle (https://bioconductor.org/packages/launch/bioc/html/DESeq2.html) was used to acquire rlog-normalized learn counts for all samples (DESeq2 rlog operate) that have been then used for PCA plot (DESeq2 plotPCA operate). To research differential gene expression, the DEseq2 (DESeq operate with components ~ lineage) was used to acquire log2-fold estimates of expression and adjusted false discovery charges for every advanced isolate in comparison with its respective ancestor. These values have been shrunk utilizing the lfcShrink operate with kind = apeglm. Every differential gene expression desk was cross-referenced with the per-gene metadata out there from the Pseudomonas genome database (https://www.pseudomonas.com/) in addition to the genomic variant, passage, and CZA MIC knowledge from the in vitro evolution experiment. Volcano plots and linear plots have been constructed utilizing the ggplot2 bundle (https://ggplot2.tidyverse.org/). Venn diagrams displaying the frequent up/down-regulated genes per advanced terminal isolate per lineage have been generated with the ggVenn R bundle (https://github.com/yanlinlin82/ggvenn). For heatmap building, the subset of the highest 750 genes with best variance was used as per defaults of the genefilter bundle (https://bioconductor.org/packages/launch/bioc/html/genefilter.html). For show, a Euclidean distance matrix of the rlog-normalized expression for every gene and isolate was clustered row-wise utilizing hierarchical clustering and plotted/annotated utilizing the ComplexHeatmap bundle (https://github.com/jokergoo/ComplexHeatmap).

With the intention to engineer mutations associated to the MexVW efflux pump into the WT MPAO1, we used a beforehand printed 2-step allelic alternate technique [69]. First, we constructed 2 single mutants. The primary contained the one level mutation in mexW (4904752 G>A) to generate the E36K substitution within the translated MexW protein, hereafter referred to as “mexW E36K” for brevity. The second mutation (T>C) was launched at place 4,903,384 upstream of mexV into the WT MPAO1 background. These 2 mutations have been then mixed to acquire a double mutant. To assemble the allelic alternate vector pEX18Gm (hereafter: vector), we first synthesized the mutant allele gene fragment (hereafter: fragment) (gBlocks HiFi Gene Fragments, Built-in DNA Applied sciences) containing the specified mutation with 500-bp upstream and 500-bp downstream areas. With the intention to clone the fragment into the alternate vector, we added XbaI and HindIII restriction websites on the 5′ and three′ ends of the fragment, respectively. The alternate vector and the synthesized gene fragments have been digested with XbaI and HindIII (New England Biolabs) and subsequently ligated and reworked into the NEB 5-alpha Competent E. coli (New England Biolabs). The insertion of the mutant allele into pEX18Gm and sequence have been verified by Sanger sequencing utilizing M13F and M13R common primers. We subsequent reworked the engineered alternate vector into E. coli S17.1 (λ pir+) and carried out biparental mating with WT P. aeruginosa MPAO1. Single recombinant mutants have been chosen on Vogel-Bonner minimal medium containing 60 μg/mL gentamicin. Double recombinant mutants have been chosen on no salt LB agar plates containing 15% (w/v) sucrose. The colonies have been screened for the purpose mutation by PCR amplification (Phusion Excessive-Constancy DNA polymerase, Thermo Fisher Scientific) and Sanger sequenced utilizing applicable primers (S12 Desk). For the development of the double mutant (place 4903384 T>C and mexW E36K), the pEX18Gm:: mexW E36K assemble was conjugationally transferred to the P. aeruginosa MPAO1 place 4903384 T>C single level mutant and chosen single and double crossover mutants colonies as described above.

The two-step allelic alternate technique utilizing the pEX18Gm vector with few modifications was used to assemble the mexB (475447 T>C) mutation yielding the W753R substitution within the translated protein, within the WT MPAO1, MPAO1 mexV -82:T>C + mexW E36K, and MPAO1-1C12-1 pressure backgrounds. First, we synthesized the mutant allele gene fragment (gBlocks HiFi Gene Fragments) containing the specified mutation with 400-bp upstream and 400-bp downstream areas and cloned into allelic alternate vector pEX18Gm as described above. The allelic alternate vector pEX18Gm:: mexB W753R was reworked into E. coli S17.1 (λ pir+) and additional used to switch the vector pEX18Gm:: mexB W753R into completely different P. aeruginosa strains by biparental matting and chosen single and double recombinants as described above. Single recombinant merodiploid for MPAO1 1C12-1 have been chosen on Pseudomonas isolation agar (BD Biosciences) containing 60 μg/mL gentamicin. Double recombinant mutants have been screened for the specified level mutation by PCR amplification (Phusion Excessive-Constancy DNA polymerase, Thermo Fisher Scientific) and Sanger sequencing utilizing applicable primers (S12 Desk).

All the constructed mutants above underwent whole-genome sequencing to confirm the specified mutations and absence of any further mutations. Briefly, particular person colonies from every mutant have been chosen after in a single day progress on LB agar, DNA extracted and libraries ready utilizing the Nextera Flex/Illumina DNA (Illumina, San Diego, California, USA) library preparation equipment, with sequencing carried out on an iSeq 100 platform with 150-bp PE reads. Reads have been aligned to the PAO1 GenBank reference (RefSeq accession quantity NC_002516.2) with a further contig for the pEX18Gm cloning vector, utilizing BWA software program v0.7.17 (https://github.com/lh3/bwa) with default parameters. The alignment was filtered for reads with “correct pairs” utilizing Samtools v. 1.6 (https://github.com/samtools/samtools). Variant calling was carried out with Freebayes v. 1.1.0 (https://github.com/ekg/freebayes) in opposition to the corresponding reference, utilizing haploid mode, a minimal allele fraction >0.8, minimal alternate rely >5, mapping high quality >100, and customary high quality filters. Referred to as variants have been inspected manually within the Broad Institute Integrative Genomics Viewer (IGV) [70,71].

FASTA nucleotide sequences for 7,493 P. aeruginosa genomes have been downloaded from the NCBI Pathogens Database (accessed on Might 14, 2021). All FASTA information have been annotated with Prokka v1.14.6 (https://github.com/tseemann/prokka) and native BLASTn and BLASTp databases have been created with the NCBI BLAST [72] makeblastdb command. A BLASTn search was carried out utilizing as queries the intergenic area between PA4373-mexV in addition to the whole nucleotide sequence of mexW (PA4375). Equally, the protein sequence of MexW was used as a question for a BLASTp search within the native protein database. A concatenated FASTA for all hits with p.c identification > 90% and E worth < 10⁻⁶ was used to generate MAFFT a number of sequence alignments. To seek out and quantify the variants of curiosity, alphabet frequencies have been calculated across the residues of curiosity, utilizing the Biostrings v0.32.0 (https://bioconductor.org/packages/launch/bioc/html/Biostrings.html) R bundle.

The proportion of transitions / transversions in MPAO1-WT versus MPAO1-mutS^Tn was analyzed utilizing a 2-tailed Fisher’s precise check utilizing the GraphPad software program (https://www.graphpad.com/quickcalcs/contingency1.cfm). Different statistical evaluation of differential gene expression was carried out with DESeq2 v1.26.0. The Wilcoxon/Mann–Whitney check was carried out with the R built-in operate Wilcox-test.