Proteomics reveal temperature-coupled cobalamin homeostasis and pathogenicity in Pseudomonas aeruginosa

(250 W) 14 15 The opportunistic pathogen Pseudomonas aeruginosa is highly adaptable to different 16 environmental conditions due to its versatile sensing and metabolic capabilities . Both 17 external temperature and metal availability have a strong influence on the virulence and 18 pathogenicity of P. aeruginosa , but the coupling between these two factors is not well 19 understood. While iron is recognized as major player in nutritional immunity, the role of 20 cobalt and the cobalt -containing vitamin B₁₂ (cobalamin) during host infection remains 21 unclear. Here, we investigate the environmental isolate P. aeruginosa PA254 using high-22 resolution global proteomics and cellular cobalamin measurements over a temperature 23 gradient spanning environmental, host -associated, and heat-stress conditions (22–42 °C). 24 PA254 occupies a continuum between an ambient-temperature virulent state characterized 25 by versatile secreted factors, exopolysaccharide-rich biofilms, and planktonic swimmers 26 and surface swarmers; and a host-associated virulent state characterized by potent secretion 27 effectors, alginate -dominated biofilms, and a strong proportion of surface twitching 28 motility. Pathway analyses indicate a shift toward carbon sparing, energy conservation, 29 redox control, and metabolic maintenance during a host -adapted lifestyle, along with the 30 strong overexpression of alternative iron acquisition strategies relying on heme and 31 siderophores. Proteins of the cobalamin biosynthetic pathway declined significantly above 32 ambient temperatures, despite constant intracellular B12 concentrations across all 33 conditions. This decoupling of biosynthesis from cellular pools implies prioritization and 34 recycling within B12-dependent processes, while the lack of B12 production at human body 35 temperatures creates avenues for therapeutics interfering with B12 supply. Altogether, this 36 work highlights a gradual rather than stepwise reprogramming of the P. aeruginosa 37 proteome in response to environmental cues, and highlights proteomics as a tool to 38 investigate system level mechanisms of challenging pathogens. 39 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 27, 2026. ; https://doi.org/10.64898/2026.01.27.701802doi: bioRxiv preprint

(550 W) 40 41 Pseudomonas aeruginosa is a ubiquitous high-priority pathogen that poses a major global 42 health threat.(1) Its ability to persist in a wide range of settings, from medical facilities to 43 urban soils and the open ocean, creates a pervasive exposure risk for humans and a 44 widespread source of initial infections.(2) P. aeruginosa can cause acute or chronic sepsis, 45 pneumonia, enterocolitis, and dermatitis, as well as opportunistic cross -infections of 46 existing conditions.(3) Due to its resistance against multiple antibiotics and host defense 47 strategies, it is especially dangerous in nosocomial infections of immunocompromised 48 patients.(1, 4, 5) 49 50 The pathogenicity of P. aeruginosa is closely linked to its ability to form facultatively 51 anaerobic biofilms, as well as to external temperature and metal availability. P. aeruginosa 52 biofilms – sessile multicellular communities that attach to surfaces through a cohesive 53 matrix built from extracellular polymeric substances (EPS) – are responsible for the 54 majority of human infections due to enhanced multidrug resistance and defense 55 protection.(6) The transition from planktonic twitching, swimming, or swarming motility 56 to biofilms is governed by complex gene regulation mechanisms that dynamically respond 57 to environmental stimuli, e.g. through two -component sensing systems, (7, 8) and induce 58 quorum sensing circuits which further advance biofilm maturation.(9) 59 60 External temperature is one important cue for bacterial host colonialization. Indeed, the 61 switch from ambient (22 °C) to host-associated (37 °C) temperatures does not only shape 62 the structural integrity of biofilm architecture,(6) but also dictates other adaptations of P. 63 aeruginosa to the human body through, for instance, thermoregulated RNA folding. (10) 64 These include the upregulation of bacterial secretion systems , the direct expression of 65 virulence factors and quorum sensing molecules, the acquisition of essential nutrients, and 66 the speed of metabolism and multiplication.(10-14) 67 68 Metal ion supply and homeostasis are other crucial determinant s for pathogenicity. Iron 69 (Fe), for example, is essential for biofilm formation, increases antibiotic resistance, and 70 contributes to host tissue damage within the secreted compound pyocyanin .(15) In fact, 71 nutritional immunity is a defense strategy of the human host to withhold Fe from bacterial 72 foci,(16) making P. aeruginosa compete for Fe through siderophore expression and heme 73 degradation.(17-19) In addition, Zinc (Zn) is necessary in metalloenzymes that function as 74 virulence agents, such as tissue degrading proteases ,(20, 21) and new insights about Zn 75 and manganese (Mn) in host nutritional immunity against P. aeruginosa are starting to 76 emerge.(22, 23) 77 78 In contrast, the roles of Cobalt (Co) and the Co-containing vitamin B12 at the host-microbe 79 interface during P. aeruginosa infections remains poorly understood. Co is an essential 80 cofactor in methionine synthetases, ribonucleotide reductases (RNRs), and other metabolic 81 enzymes.(24) Although P. aeruginosa only encodes genes for aerobic B12 biosynthesis, 82 B12-dependent RNRs were shown to be required for anaerobic biofilm growth,(25) and the 83 mechanisms of B12 homeostasis in oxygen -devoid biofilms is still unclear .(26, 27 ) 84 Interestingly, Co can partially compensate for Zn in starved P. aeruginosa and restore some 85 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 27, 2026. ; https://doi.org/10.64898/2026.01.27.701802doi: bioRxiv preprint virulence traits,(28) further demonstrating the need to understand the function of Co in 86 human infections. 87 88 Global high-resolution proteomics is a powerful tool to investigate the functional status of 89 pathogens under physiologically relevant conditions. To date, most system analyses of P. 90 aeruginosa have relied on transcriptomics, often comparing two discrete growth 91 temperatures (low vs. high), or focused on metal biology without accounting for 92 temperature. Here, we apply in-depth proteomics to illuminate how metal homeostasis and 93 pathogenicity are reshaped in P. aeruginosa across a broad temperature gradient, and to 94 investigate various metal homeostasis mechanisms including cobalamin. 95 96

(2,050 W) 97 98 Strain Cultivation 99 Pseudomonas aeruginosa 2-54 (PA254) was isolated in the Central Pacific Ocean on the 100 research expedition METZYME (KM1128) aboard the R/V Kilo Moana, as described 101 previously.(27) PA254 was cultured in autoclaved marine broth containing 1 L 0.2 µm -102 filtered coastal seawater (Vineyard Sound), 5 g peptone (Fisher Scientific), and 1 g yeast 103 extract (BD Difco), with or without 1.5% agar (Fisher Scientific). 104 105 Growth Curves 106 To record growth curves at different temperatures (22, 25, 27, 30, 35, 37, 40, 42 °C), PA254 107 was grown in 12-well plates in a SpectraMax M3 plate reader (Molecular Devices). First, 108 an acclimated pre -culture of PA254 in 5 mL marine broth was prepared from a single 109 colony or glycerol stock, and shaken at 600 rpm at the desired temperature overnight. 110 Before inoculation of the 12 -well plate, 20 mL marine broth (supplemented with 1 µM 111 CoCl2) as well as the plate reader chamber were pre-adjusted to the desired temperature for 112 at least 15 minutes. After that, 1.5 mL marine broth and 15 µL PA254 pre -culture were 113 dispensed into each well. Optical density was then recorded at 600 nm every 10 minutes 114 for 16-42 hours until reaching stationary phase, with orbital shaking for 5 s before every 115 measurement. To prevent condensation, plate lids were pre -treated with 5% Triton -X in 116 ethanol, and dried in a laminar flow hood overnight. 117 118 Genome Annotation 119 Genomes of bacterial isolates from the METZYME expedition were sequenced at the Johns 120 Hopkins Deep Sequencing and Microarray Core Facility, as reported previously. (27) 121 Isolates 1-54 and 2-54 were identified to be the same strain of P. aeruginosa, and PA154 122 yielded a circular chromosome (6,455,702 base pairs). JSpeciesWS tetra correlation search 123 (TCS)(29) showed P154 closely related to P. aeruginosa P-14 and BWH047 (Z 0.99984); 124 however, whole-genome alignments with these hits suggested a distinct and novel strain. 125 Translation initiation sites were predicted with the gene finding software Prodigal 126 (PROkaryotic DynamIc programming Genefinding Algorithm).(30) Open reading frames 127 (6,523) were then translated into amino -acid code, and a diamond -search was performed 128 against the NCBI nr (non -redundant) protein database using blastp, resulting in 5,752 129 functionally annotated and 793 hypothetical proteins. Of the 793 hypothetical proteins, 269 130 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 27, 2026. ; https://doi.org/10.64898/2026.01.27.701802doi: bioRxiv preprint could be annotated using eggNOG (evolutionary genealogy of genes: Non -supervised 131 Orthologous Groups) mapper(31) and were integrated into the final annotated proteome. 132 133 Proteomics 134 Extraction. Protein extraction and digestion were performed using a modified version of 135 the Protifi S -trap mini spin column protocol recommended by the manufacturer 136 (https://protifi.com/protocols/). All solvents were LC -MS grade (Fisher Optima), unless 137 otherwise noted. Briefly, 9 mL of a PA254 culture were harvested in a 2 mL EtOH-washed 138 microtube through repeated centrifugation at 12,000 rpm for 5 minutes, and frozen at -80 139 °C prior to extraction. Upon thawing, the pellet was resuspended in 300 µL lysis buffer 140 (2% SDS, 50 mM tetraethylammonium bromide, pH 8.5) and incubated at 95 °C for 10 141 minutes. The samples were cooled on ice, treated with 2 µL benzonase nuclease (25.5 u/µL, 142 Novagen), shaken at 350 rpm for 30 minutes at 37 °C, and centrifuged at 12,000 rpm for 143 10 minutes. The supernatant was transferred to a fresh 2 mL EtOH-washed microtube and 144 total protein quantified via micro-BCA protein assay (Thermo Scientific). 145 146 Reduction and Alkylation. An aliquot of 100 µg extracted proteins in lysis buffer (2% SDS, 147 50 mM tetraethylammonium bromide, pH 8.5) was prepared for reduction and alkylation. 148 First, 4 µL 500 mM dithiothreitol (in 50 mM ammonium bicarbonate) were added, and 149 samples incubated at 45 °C for 30 minutes. Then, 12 µL 500 mM iodoacetamide (in 50 150 mM ammonium bicarbonate) were added, and samples incubated at room temperature for 151 30 minutes in the dark. Excess iodoacetamide was quenched by addition of 4 µL 500 mM 152 dithiothreitol (in 50 mM ammonium bicarbonate). Afterwards, samples were treated with 153 23 µL 12% phosphoric acid, incubated at room temperature for 5 minutes, and diluted with 154 1.7 mL binding buffer (100 mM tetraethylammonium bromide, pH 7.1, 90% MeOH). 155 Proteins were loaded onto pre-rinsed S-trap mini-spin columns in increments of 600 µL at 156 4,000 rpm for 30 s, with each flow-through being loaded a second time. The samples were 157 washed with 8 -10 times with 600 µL binding buffer (100 mM tetraethylammonium 158 bromide, pH 7.1, 90% MeOH) at 4,000 rpm for 30 s, and one time with 600 µL 90% MeOH 159 at 12,000 rpm for 2 minutes. 160 161 Digestion. Trypsin digestion was performed on-column with a protein:trypsin ratio of 25:1. 162 A solution of 4 µL trypsin (Trypsin Gold, Mass Spectrometry Grade, Promega) in 125 µL 163 50 mM ammonium bicarbonate was allowed to permeate the S-trap mini spin columns by 164 centrifuging at 1,000 rpm for 30 seconds, and reloading any flow -through. The digestion 165 was carried out at 37 °C for 14 hours. 166 167 Peptide preparation . Digested proteins were eluted into a fresh 2 mL EtOH -washed 168 microtube stepwise with 80 µL 50 mM Ammonium bicarbonate, 80 µL 0.2% formic acid, 169 and 80 µL 50% acetonitrile and 0.2% formic acid at 12,000 rpm for 1 -5 min each. 170 Combined eluents were then quantified via micro-BCA protein assay (Thermo Scientific) 171 before drying in a SpeedVac (ThermoSavant) at room temperature. The residue was 172 redissolved in LC-MS buffer (2% acetonitrile, 0.1% formic acid) to a final concentration 173 of 1 µg/µL peptides. To avoid particulates, peptides were centrifuged at 12,000 rpm for 20 174 minutes, and the supernatant diluted to 0.1 µg/µL for analysis. 175 176 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 27, 2026. ; https://doi.org/10.64898/2026.01.27.701802doi: bioRxiv preprint Peptide analysis . Tryptic peptides were analyzed using liquid chromatography coupled 177 with tandem mass spectrometry (LC/MS/MS) by injecting 1 µg on a Thermo Dionex NC-178 3500RS HPLC coupled to a Thermo Scientific Astral Orbitrap mass spectrometer with a 179 Thermo Flex source. Each sample was concentrated onto a trap column (0.2 x 10 mm ID, 180 5 μm particle size, 120 Å pore size, C18 Reprosil -Gold, Dr. Maisch GmbH) and rinsed 181 with 100 µL LC -MS buffer (2% acetonitrile, 0.1% formic acid) before elution through a 182 reverse phase C18 column (0.1 x 150 mm ID, 3 μm particle size, 120 Å pore size, C18 183 Reprosil-Gold, Dr. Maisch GmbH). Chromatography was performed at 0.5 mL/min flow 184 rate over a 70 minute nonlinear gradient from 5% to 95% acetonitrile and 0.1% formic acid 185 in water. Mass spectrometry was performed in DDA mode with MS1 scans spanning 380 186 to 1280 m/z in 240 K resolution. The top n ions underwent MS2 scans with a 1.6 m/z 187 isolation window and a 7 s dynamic exclusion time. 188 189 Proteomics Informatics 190 Raw mass spectra were searched against the PA254 annotated proteome fasta, generated 191 as described under ‘genome annotation’, using the program FragPipe (Version 23). The 192 parameters used were a precursor mass tolerance of 10 ppm, a fragment mass tolerance of 193 20 ppm, and up to 2 missed cleavages. Differential expression analysis was conducted 194 using FragPipe-Analyst(32) based on limma (Linear Models for Microarray and Omics 195 Data). MaxLFQ intensity values were processed using a Benjamini Hochberg adjusted p-196 value < 0.05, a log2 fold -change cutoff of 1, and no imputation, to yield overexpressed 197 (positive), underexpressed (negative), and differentially expressed (total) proteins 198 (Supplemental Dataset S1). Volcano plots of differently expressed proteins were prepared 199 by plotting -log10 unadjusted p-value against log2 fold -change using python. For 200 visualization purposes, a reduced log2 fold-change cutoff of 0.5 was used. Heatmap plots 201 of proteins versus temperature were prepared in the following manner: ANOVA was used 202 to test for the dependence of protein abundance on temperature using three biological 203 replicates each temperature point. Proteins where ANOVA p < 0.05 were visualized as 204 heatmaps using R studio using unique spectral counts max-normalized to a scale of 0-1. 205 206 Pathway Analysis 207 For metabolic pathway analysis, overrepresentation analysis (ORA), and gene set 208 enrichment analysis (GSEA), the proteome of PA254 was first mapped onto Kyoto 209 Encyclopedia of Genes and Genomes (KEGG) pathway maps. For this purpose, the KEGG 210 pathway maps of the P. aeruginosa PAO1 proteome were used, which are publicly 211 available ( https://www.kegg.jp/kegg-bin/get_htext?pae00001). A diamond search using 212 blastp was applied to assign known PAO1 pathways to the corresponding PA254 proteins, 213 leading to 6,036 proteins with pathways. For the remaining 487 proteins without pathways, 214 additional pathway information for 197 proteins was obtained using eggNOG mapper(31) 215 and integrated into the final pathway database (Supplemental Dataset S 2). Differentially 216 expressed proteins per pathway per temperature were then plotted as percent of total 217 detected proteins per pathway per temperature. 218 Overrepresentation analysis (ORA) was performed on the over - and underexpressed 219 proteins of different temperatures vs. 22 °C (foreground), obtained as described under 220 ‘proteomics informatics’, against total detected proteins per pathway per temperature 221 comparison (background). Using python, the statistical significance of pathway 222 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 27, 2026. ; https://doi.org/10.64898/2026.01.27.701802doi: bioRxiv preprint overrepresentation was assessed using Fisher’s exact test with a one -sided alternative 223 hypothesis, and p-values were corrected for multiple testing using the Benjamini –224 Hochberg false discovery rate (FDR) method. Pathways with an FDR < 0.05 were 225 considered significantly overrepresented, and plotted against -log10 FDR. 226 Gene set enrichment analysis (GSEA) was performed on the total expressed proteins of 227 different temperatures, obtained as described under ‘proteomics informatics’, by ranking 228 all log2 fold-changes for each temperature comparison. Using python, pathways for PA254 229 proteins were then converted into a Gene Matrix Transposed (GMT) format. GSEA was 230 conducted using the package gseapy with 1,000 permutations per analysis. Pathways 231 containing 500 proteins were excluded. Pathways with an FDR < 0.05 were 232 considered significantly enriched, and plotted against the normalized enrichment score 233 (NES). 234 235 Cellular Cobalamin 236 To quantify the cellular cobalamin content, 9 mL of a PA254 culture were harvested in a 2 237 mL microtube through repeated centrifugation at 12,000 rpm for 5 minutes. Cell pellets 238 were washed twice with 1 mL 0.2 µm -filtered and autoclaved coastal seawater (Vineyard 239 Sound) to remove dissolved cobalamin in the medium, and frozen at -80 °C prior to 240 extraction. All solvents were LC-MS grade (Fisher Optima), unless otherwise noted. Upon 241 thawing, samples were treated with 1.5 mL MeOH and 10 µL 10% HCl (for roughly 10 242 mM or pH 2 final concentration), and shaken at 800 rpm for 30 minutes at room 243 temperature in the dark. The supernatant was collected through centrifuging at 12,000 rpm 244 for 2 minutes, and the pellet extracted a second time as above. The combined extraction 245 fractions were then dried in a SpeedVac (ThermoSavant) at room temperature in the dark 246 before resuspension in 50 µL LC-MS buffer (2% acetonitrile, 0.1% formic acid). To avoid 247 particulates, samples were centrifuged at 14,000 rpm for 20 minutes in the dark, transferred 248 to a 0.2 mL microtube, and centrifuged again. The supernatant was transferred to an amber 249 microvial and measured via LC/MS/MS on a Thermo Dionex NC-3500RS HPLC coupled 250 to a Thermo Scientific Fusion Orbitrap mass spectrometer with a Thermo Flex source. Each 251 sample was concentrated onto a trap column (0.2 x 10 mm ID, 5 μm particle size, 120 Å 252 pore size, C18 Reprosil-Gold, Dr. Maisch GmbH) and rinsed with 100 µL LC -MS buffer 253 (2% acetonitrile, 0.1% formic acid) before elution through a reverse phase C18 column 254 (0.1 x 150 mm ID, 3 μm particle size, 120 Å pore size, C18 Reprosil -Gold, Dr. Maisch 255 GmbH). Chromatography was performed at 0.5 mL/min flow rate over a 40 minute 256 nonlinear gradient from 5% to 95% acetonitrile and 0.1% formic acid in water. Mass 257 spectrometry targeted the observable masses of cyanocobalamin (m/z = 1354.57, 678.29), 258 methylcobalamin (m/z = 1343.59, 672.80), and hydroxocobalamin (m/z = 1345.57, 259 673.79). Total MS2 fragment areas for each form of cobalamin (CN, Me, OH) were 260 obtained using Skyline (Version 23.1) and used to calculate absolute concentrations. A 6-261 point calibration curve for each cobalamin (CN, Me, OH) was prepared in the range of 1 -262 100 µg/L in LC -MS buffer (2% acetonitrile, 0.1% formic acid). External standards 263 containing 5 µg/L of each cobalamin (CN, Me, OH) were included with each set of 264 unknown samples. A series of blanks was run after each standard or sample, and cobalamin 265 peak areas eluting in blanks (if any) were added to the standard or sample peak areas until 266 blanks washed out clean. 267 268 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 27, 2026. ; https://doi.org/10.64898/2026.01.27.701802doi: bioRxiv preprint Dissolved Cobalamin 269 Dissolved cobalamin in the spent media of PA254 cultures was measured according to a 270 previously published method (33). Due to representing only a small fraction of total 271 cobalamin, dissolved cobalamin was omitted from the discussion. 272 273 Data Availability 274 The sequenced P. aeruginosa genome for the isolate 1 -54 is available at Zenodo (doi: 275 10.5281/zenodo.15336754). The annotated proteome has been submitted to PRIDE and 276 Uniprot. The raw global proteomic spectra are available ProteomeXchange and PRIDE 277 (review access available upon request). The processed global proteomic data are available 278 as Supplemental Dataset 1. 279 280

(1,650 W) 281 282 Growth at temperatures 283 In order to investigate the dependence of metal homeostasis and pathogenicity in P. 284 aeruginosa PA254 on external temperature, the organism was cultivated across a range 285 from 22-42 °C (Fig. 1). A pre-culture acclimated to the desired temperature over night was 286 used for inoculation. PA254 was capable of growing to a dense culture (OD 600 > 0.35) in 287 all conditions tested, albeit with a longer lag phase at lower and a lower maximum OD600 288 at higher temperatures (Fig. 1a). The growth rates continuously increased from 0.10 hr-1 at 289 22 °C to 0.45 hr-1 at 42 °C (Fig. 1b). 290 291 292 Figure 1. Temperature-dependent growth curves (a) and growth rates (b) of Pseudomonas 293 aeruginosa PA254 cultivated at environmental (22-30 °C), host-associated (35-37 °C), and 294 heat stress conditions (40-42 °C). 295 296 Global proteomics metrics 297 Cultures in stationary phase were evaluated through global proteomics analyses. We 298 identified 4,730 unique proteins in this study, representing a proteome coverage of 72.5 % 299 across all temperatures (Supplemental Dataset S1). Individual samples were annotated with 300 3,400-4,700 unique features ( Fig. S1 ). The Coefficient of Variation, Pairwise Pearson 301 correlation, and log2 centered intensity showed strong reproducibility among replicates 302 (Figs. S2-S4). Principal component analysis with 34.8% (PC1) and 21.1% (PC2) indicated 303 that experimental treatments accounted for the majority of observed variance (Fig. S5). Up 304 to 13.5 % of all detected proteins were differentially expressed (Fig. S6), with the largest 305 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 27, 2026. ; https://doi.org/10.64898/2026.01.27.701802doi: bioRxiv preprint amount between 22 °C vs. 42 °C (638 proteins; Fig. S7), and the smallest amounts between 306 25 °C vs. 30 °C, 30 °C vs. 35 °C, and 37 °C vs. 40 °C (< 3 proteins). 307 308 Differential protein expression 309 Major metabolic and physiological traits known to be temperature -responsive in P. 310 aeruginosa were mirrored by significant changes in the proteome ( Fig. 2). Classic heat 311 stress proteins were strongly upregulated at 42 °C, including the molecular chaperones 312 ClpB, GroEL, DnaJ, and DnaK ( Fig. 2a ), similar to the heat shock response of other 313 Pseudomonads.(14, 34) 314 PA254 expression of several established virulence factors was highly temperature 315 dependent and partitioned into two groups (Fig. 2b). One set of proteins was upregulated 316 at 25 °C and contained the sugar antigen producing rhamnosyltransferase WbpX, (35) 317 several members of the pyoverdine gene cluster Pvd,(36) as well as effectors released by 318 type 1-3 bacterial secretion systems (T1-3SS),(37) namely the surface structures AprE and 319 AprF (T1SS), the exotoxin LipA (T2SS), and the secreted product PemB (T3SS). The 320 second set of proteins was upregulated at 37 -42 °C and included the secreted tissue -321 degrading protease LasA and corresponding regulator LasR, and select effectors of T3SS, 322 such as exotoxin ExoU and the pore-forming proteins PopB/D, and the needle tip protein 323 PcrV.(37) 324 Proteins implicated in biofilm formation were present throughout the temperature curve. 325 The EPS biofilm matrix of P. aeruginosa largely consists of exopolysaccharides 326 synthesized via the Psl operon, which was upregulated between 25-35 °C (Fig. 2c), and is 327 associated with early cell aggregation and biofilm initiation. (38) Alginate, on the other 328 hand, is a major constituent of chronic mucoid biofilms; both AlgB/R upregulate the 329 AlgACD gene locus which directly controls the biosynthesis of alginate precursors. (38) 330 Both AlgB/R, and the negative regulators MucA/B protecting P. aeruginosa from alginate 331 overproduction,(38) were upregulated at 37-42 °C. 332 Bacterial swimming motility operates via rotating flagella , where the individual 333 components are constructed using the large gene clusters Fli (rotor, filament and cap) and 334 Flg (hooks, rings and rod). Several of these core genes were underexpressed above 30 °C, 335 including the flagellin building block ( Fig. 2d). Bacterial twitching motility, on the other 336 hand, is a flagella-independent movement along moist, solid surfaces with the help of hair-337 like pili. More than ten proteins of the associated Pil operon were strongly overexpressed 338 above 37 °C, in addition to the minor pili assembly subunit protein FimUT (Fig. 2e).(39) 339 In addition, the response regulator FleR implicated in swarming was significantly 340 overexpressed at 22 -25 °C ( Supporting Data S1 ),(40) a behavior connected to biofilm 341 development and antibiotic resistance.(41) 342 Under anaerobic conditions, P. aeruginosa operates via denitrification using nitrate as 343 terminal electron acceptor for energy production. Denitrification proteins occurred across 344 the temperature gradient and exhibited diverse trends ( Fig. 2d ). Several nitrate/nitrite 345 transporters, nitrate respiration regulators, and nitrate reductases were overexpressed at 22-346 25 °C. Nitrite reductases and oxide reductases peaked at 30 °C, and the nitrite/formate 347 transporter YfdC at 37 -42 °C. However, select proteins (NarG/H, NorB/C) were highly 348 expressed at two or more non-contiguous temperature regimes. 349 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 27, 2026. ; https://doi.org/10.64898/2026.01.27.701802doi: bioRxiv preprint 350 Figure 2. Temperature-dependent expression of proteins related to virulence ( a), heat 351 shock ( b), biofilm formation ( c), twitching motility ( d), swimming motility ( e), and 352 denitrification ( f) in Pseudomonas aeruginosa PA254. Displayed are only entries that 353 showed a significant correlation of protein abundance to temperature gradient (ANOVA p 354 < 0.05), and that were differentially expressed between at least two temperature points 355 (Benjamini-Hochberg adjusted p < 0.05). For gene abbreviations, see Supplemental 356

Table S1. 357 358 Pathway analyses 359 To identify functional programs of P. aeruginosa responding to temperature stimuli, 360 proteins were mapped to the corresponding cellular pathways of the Kyoto Encyclopedia 361 for Genes and Genomes (KEGG) Pathway Database ( Supplemental Dataset S2). A 362 comparison between the differentially expressed proteins at temperature minimum vs. 363 maximum (22 °C vs. 42 °C) showed that changes concentrated mostly within the pathways 364 of xenobiotics degradation (16.5 %), cell motility (15.7 %), lipid metabolism (14.8 %), and 365 bacterial infectivity (14.3 %; Fig. S8). 366 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 27, 2026. ; https://doi.org/10.64898/2026.01.27.701802doi: bioRxiv preprint Both overrepresentation analysis (ORA) and gene set enrichment analysis (GSEA) were 367 applied to identify statistically significant pathway trends (Fig. 3). ORA detects pathways 368 based on differentially expressed proteins, while GSEA captures broad trends across the 369 entire ranked proteome. ORA showed an enrichment in a subset of 23 and GSEA of 24 370 metabolic pathways, with partial overlap. 371 Several coordinated metabolic shifts were captured with both ORA and GSEA approaches. 372 The metabolism of certain amino acids was negatively enriched at temperatures above 37 373 °C, e.g. glycine, serine, threonine, histidine, and tyrosine . Biosynthesis of the virulent 374 pigment phenazine showed strong, coherent enrichment between 25 -40 °C compared to 375 both temperature extremes (22 or 42 °C). Quorum sensing proteins were similarly enriched 376 throughout intermediate temperatures but not at extremes. Starch and sucrose metabolism 377 were negatively enriched below 37 °C and positively enriched above 37 °C. Under 378 prolonged heat stress conditions (40 –42 °C), genetic information processing pathways 379 started to decline, including DNA replication, base excision repair, mismatch repair, and 380 homologous recombination. 381 382 383 384 Figure 3. Temperature-dependent pathway analysis of Pseudomonas aeruginosa PA254 385 via a) ORA (overrepresentation analysis) and b) GSEA (gene set enrichment analysis) . 386 Downregulated proteins at variable temperatures vs. 22 °C displayed in blue, and 387 upregulated proteins in orange. For ORA, proteins with a Benjamini-Hochberg adjusted p 388 < 0.05 are displayed as -log10 FDR (false discovery rate). For GSEA, proteins with an 389 FDR q-value < 0.05 are displayed as absolute NES (normalized enrichment score). 390 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 27, 2026. ; https://doi.org/10.64898/2026.01.27.701802doi: bioRxiv preprint 391 ORA further highlighted an overrepresentation of carbon assimilation and energy 392 production pathways at host -associated temperatures (35 –37 °C), namely glyoxylate, 393 propanoate, methane, and lipoic acid metabolism. Heat stress (40–42 °C) resulted in a 394 decline in bacterial chemotaxis and motility pathways, while chaperone and folding 395 catalysts were overrepresented. 396 GSEA revealed a negative enrichment in the catabolism of complex carbohydrates (pentose 397 and glucuronate interconversion, ascorbate and aldarate metabolism) at high temperatures 398 (40–42 °C). In addition, catabolic pathways for xenobiotics (c hlorocyclohexane, 399 chlorobenzene, benzoate , and fluorobenzoate degradation ) also showed a concerted 400 negative enrichment and temperatures above 35 °C. Ribosomal proteins were continuously 401 enriched between 25–35 °C. 402 A small subset of pathways exhibited contrasting trends between ORA and GSEA. The 403 biosynthesis of polyketides and non -ribosomal peptides (NRPs) was negatively enriched 404 in ORA, but positively enriched at lower temperatures using GSEA (25 –30 °C) ; these 405 secondary metabolites include antibiotics and siderophores.(36) 406 407 Metal homeostasis 408 Iron uptake systems in P. aeruginosa are directly correlated with host infection, (15-19) 409 and 13 proteins associated with these mechanisms changed significantly with temperature 410 in this study (Fig. 4a). Among these, alternative iron uptake systems(42) were significantly 411 overexpressed above 37 °C when compared to 22 °C: the outer membrane heme receptor 412 PhuR, the hemopexin-binding importer HxuA, and the ferri-enterobactin transporter PirA. 413 FemA, responsible for f erri-mycobactin uptake, and FoxA and FiuA, both involved in 414 ferrichrome/ferrioxamine uptake, were significantly underexpressed at 22 -30 °C when 415 compared to 42 °C. In addition, FoxA was also significantly overexpressed at 37 -42 °C 416 when compared to 25 °C. The FpvAB system transports Fe-bound pyoverdine and was 417 significantly elevated at all temperatures above 22 °C. 418 Pseudomonads possess multiple redundant uptake strategies for other biologically relevant 419 metal ions and their complexes (Co, Zn, Mn, copper (Cu)). (21, 42 ) Only two Zn 420 transporters(21) and one Mn uptake protein(43) varied significantly with temperature, but 421 were not differentially expressed at host -adapted conditions ( Fig. 4b-c). No significant 422 responses to host temperatures were found for Cu uptake systems (Fig. 4d). While no Co 423 metal ion-specific transporter in P. aeruginosa is known, Co is often co-imported through 424 other channels (e.g., with Zn), and cobalamin imported through TonB -dependent BtuB 425 receptors.(44) Three BtuB transporters were identified in the PA254 proteome ; however, 426 no Co or B 12 import system was temperature -sensitive (Fig. 4e). Proteins related to the 427 Zn/Co siderophore pseudopaline(45) were not detected. 428 Besides extracellular cobalamin uptake, P. aeruginosa is capable of aerobic B 12 429 biosynthesis through the Cob cluster. Several members of the biosynthetic Cob pathway as 430 well as CysG varied strongly with temperature, and the abundance these proteins 431 diminished significantly after 22 °C (Fig. 4f). CobA/F/L/ST/V were not observed in this 432 dataset; CobC/G/K/M -O/Q/T/U did not change significantly. The corrinoid 433 adenosyltransferase PduO catalyzing the regeneration of active cobalamin cofactors was 434 highest at 37 °C, but not differentially expressed at that temperature (unadjusted p = 0.018). 435 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 27, 2026. ; https://doi.org/10.64898/2026.01.27.701802doi: bioRxiv preprint Cobalamin is employed as cofactor for the methionine synthase MetH and the 436 ribonucleotide reductase NrdJ. The cobalamin -free counterparts of these enzymes are 437 MetE, NrdA/B under oxic conditions, and NrdD under anoxic conditions, respectively.(26) 438 B12-dependent MetH was detected with no significant changes throughout the temperature 439 gradient, along with the methionine synthesis regulator MetR ( Fig. 4g ). Both B 12-440 dependent and independent NrdA/D/J changed significantly with temperature and were 441 overexpressed at low temperatures versus 35 °C. NrdJa was also overexpressed at 25 vs. 442 40-42 °C. The anoxic protein NrdD was not detected at 25 and 37-40 °C. 443 444 445 446 Figure 4. Temperature-dependent metal homeostasis in Pseudomonas aeruginosa PA254. 447 Heatmaps show expression of proteins related to Fe uptake (a), Zn uptake (b), Mn uptake 448 (c), Cu uptake (d), cobalamin uptake (e), cobalamin biosynthesis and regeneration (f), and 449 cobalamin usage ( g). Displayed are only entries that showed a significant correlation of 450 protein abundance to temperature gradient (ANOVA p < 0.05), and that were differentially 451 expressed between at least two temperature points (Benjamini -Hochberg adjusted p < 452 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 27, 2026. ; https://doi.org/10.64898/2026.01.27.701802doi: bioRxiv preprint 0.05). For gene abbreviations, see Supplemental Material Table S1 . Bar graph shows the 453 intracellular cobalamin concentrations with dashed line denoting the average (h). 454 455 Cobalamin concentrations 456 The absolute concentrations of cobalamin within P. aeruginosa PA254 cells were 457 measured using LC/MS/MS. For that purpose, the amounts of the analogs cyanocobalamin 458 (CN-B12), methylcobalamin (Me -B12), and hydroxocobalamin (OH -B12; Fig. S 9) were 459 quantified individually and added as total B12. To ensure detectable B12 concentrations, the 460 growth medium was supplemented with 1 M CoCl 2 (Fig. S10). Intracellular cobalamin 461 through the temperature curve averaged at 0.384 ± 0.026 pM OD -1 when normalized to 462 optical density of the PA254 culture, and the values at individual temperature s were not 463 statistically different (Fig. 4h). The majority of observed cobalamin was OH -B12 at all 464 temperatures (Fig. S11). Extracellular cobalamin in the growth medium represented only 465 a small fraction (<10%) of the total B12 and was thus omitted from this study. It was found 466 that PA254 produced B 12 until late stationary phase, and that intracellular B 12 did not 467 decline within aged cultures (Fig. S12). 468 469

(1,050 W) 470 471 Infectious strains of P. aeruginosa have been reported to operate under two distinct 472 temperature regimes: an environmental lifestyle (~20-25 °C), favoring biofilm growth plus 473 the degradation of complex natural sugars and xenobiotics, and at human body 474 temperatures (37 °C), favoring rapid proliferation plus adaptation to nutrients scavenged 475 in the host.(6, 10, 12 -14, 46, 47 ) Here, this study shows a continuous transition of these 476 two modes over a temperature gradient, and associated adaptations in pathogenic traits, 477 metal homeostasis, and metabolism throughout. 478 479 Continuous modes of temperature-dependent pathogenic physiology 480 481 Experiments with the virulent reference strain PAO1 establish that many quorum sensing 482 products, secretion systems , and virulence factors are strongly induced at 37 °C ,(12) 483 especially factors reliant on the transcriptional regulator rhlR such as pyocyanine .(10) 484 PAO1 also upregulates select virulence agents at ambient temperatures: protease IV, (46) 485 pyoverdine,(12) and T1SS and T2SS products (e.g. apr).(47) The multidrug resistant strain 486 PA14 also showed enhanced transcription of the highly virulent T3SS effectors (exo, pop, 487 pcr) and phenazine biosynthesis at 37 °C.(11) In contrast, while PA254 also overexpresses 488 these core T3SS products at 37 °C, other T3SS effectors (i.e. PemB) were significantly 489 higher at 25 °C, together with the rhlR-dependent WbpX. Overall, differentially expressed 490 virulence factors in PA254 were equally distributed between low and high temperatures 491 (see Fig. 2a), and the expression of RhlR, pyocyanine, and protease IV not regulated. These 492

suggest that PA254 may exhibit stronger virulence at ambient conditions than 493 typical clinical strains, and that rhlR quorum sensing control and the type 3 secretion 494 system are gradually integrated over the temperature spread instead of localized to host 495 conditions. Supporting this, the pathways for quorum sensing and phenazine synthesis 496 were strongly enriched throughout 25-40 °C (see Fig. 3). 497 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 27, 2026. ; https://doi.org/10.64898/2026.01.27.701802doi: bioRxiv preprint PAO1, PA14, and several clinical isolates display the highest biofilm mass below 22 498 °C.(13) While biofilms are also formed above that , low temperature results in distinct 499 biofilm architecture and protein enrichment. (6) Biofilm formation was also strain 500 dependent: sessile biomass and EPS producing genes (pel, psl, alg) of most clinical strains 501 declined with rising temperature, but PAO1’s biofilm formation and alginate production 502 partially recovered at 37 °C. (13) While the biofilm mass of PA254 was not directly 503 quantified here, associated proteins were partitioned equally between initial biofilms at 504 moderate (25-35 °C) and chronic biofilms at high (37-42 °C) temperatures. Both swarming 505 and twitching, as well as pathway enrichment ( see Fig. 3b), confirmed biofilm processes 506 at various temperatures. 507 Together with trends in motility, PA254 occupies a continuum between a) an ambient -508 temperature virulent state characterized by a subset of type 1-3 secreted factors, pyoverdine 509 production, exopolysaccharide -rich biofilms, and a strong proportion of planktonic 510 swimmers and surface swarmers; and b) a host-associated virulent state characterized by a 511 subset of highly potent type 2-3 secreted factors, alginate-dominated biofilms, and a strong 512 proportion of surface twitching motility. 513 514 Thermal reprogramming of macro- and micronutrient metabolism 515 516 External temperature resulted in the restructuring of major metabolic pathways across 517 different temperature transitions, with pronounced changes surrounding a host -adapted 518 lifestyle. Above 37 °C, PA254 shifts from the degradation of complex environmental 519 xenobiotics and sugar acids to the metabolism of host -derived carbon sources such as 520 glycans, fatty acids, and reduced small molecules .(48) Especially glyoxylate shunt 521 upregulation is a signature of pathogen adaptation to host nutrients, oxidative stress, and 522 biofilm conditions. (49) This is accompanied by a shift from environmental nitrate 523 consumption to reduced nitrite and nitrous oxide typical for oxidative stress and biofilm 524 conditions (see Fig. 2f).(50) In conjunction with reduced amino acid metabolism , these 525 trends point toward a host-adapted state of carbon sparing, energy conservation, redox 526 balance control under partial denitrification, and metabolic maintenance under chronic 527 biofilm conditions. 528 In a transcriptomics study , the P. aeruginosa strain PAO1 exhibited slightly contrasting 529 trends. A comparison of 37 °C vs . 22 °C showed d ownregulated transcript s for the 530 metabolism of starch, sucrose, alcohol, and pyoverdine , and u pregulated transcripts for 531 pyochelin metabolism, TCA cycle and glucose assimilation.(12) These differences might 532 stem from the two individual strains used, from the 5-fold less yeast extract in our culturing 533 medium, or from the distinct pathways captured by DNA- versus protein-based ‘omics. 534 While biofilm and virulence traits of PA254 were present across the temperature gradient, 535 uptake of alternative Fe sources was strongly coupled to 37 °C. Treatments with human 536 calprotectin, which sequesters metals during host infection, have demonstrated the aerobic 537 metal starvation response of PAO1 and PA14 to differ from the anaerobic response. (51) 538 For example, Fe-acquiring phuR was only upregulated under anaerobic metal scarcity . 539 Here, significant changes in PhuR expression together with partial denitrification and 540 biofilm formation indicators suggest existing redox transitions in PA254 cultures, and 541 highlights the elaborate metabolic control within hypoxic gradients commonly encountered 542 in host-associated niches(52) rather than strict oxic or anoxic conditions. 543 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 27, 2026. ; https://doi.org/10.64898/2026.01.27.701802doi: bioRxiv preprint 544 Decoupling of cobalamin biosynthesis from cellular reservoirs 545 546 During biofilm formation, P. aeruginosa experiences transitions between oxygenated and 547 anoxic conditions.(52) Earlier studies have shown that ribonucleotide reductases (RNRs) 548 are vital for anaerobic growth, and that the cobalamin -dependent class II RNR dominates 549 in biofilms. (53) Exogenous B 12 addition was shown to be required for full anaerobic 550 growth of RNR-II dependent P. aeruginosa due to absence of the genes for anoxic B 12 551 production.(25) This led to the hypothesis that aerobic or microaerophilic B12 synthesis via 552 the CobN enzyme provides enough cobalamin for subsequent anoxic biofilm layers.(26) 553 Contrary to that assumption, previous metalloproteomics analysis of PA254 revealed 554 CobN to be more abundant under oxygen -depleted conditions. (27) Here, CobN was 555 constant throughout the temperature gradient independent of changes in biofilm or redox 556 marker proteins, with a significant decline only under heat stress (Supplementary Dataset 557 S1). Yet the overall cobalamin biosynthetic pathway was impaired above 22 °C (see Fig. 558 4f), suggesting that CobN abundance alone might not be an accurate proxy for B 12 559 production. CobN also did not correlate with NrdJ, which was significantly underexpressed 560 at 35-42 °C. 561 Overall, this dataset shows partitioning into a subset of temperature -independent 562 cobalamin parameters (B12 transporters, B12-dependent methionine synthase, intracellular 563 B12) and a subset of temperature -regulated proteins (B 12 biosynthesis, B12 recycling, B12-564 dependent RNRs), the latter not always responding to the same temperature. We assume 565 that due to a decline of new cobalamin production, the organism prioritizes certain B 12 566 functions (MetH) at the expense of others (NrdJ), while utilizing recycling mechanisms to 567 keep the cellular cobalamin reservoir constant. Since the B 12 biosynthesis pathway is 568 strongly downregulated at human body temperatures, P. aeruginosa is possibly susceptible 569 to Co nutritional immunity strategies by the host, thus creating new avenues for potential 570 drug discovery. 571 572

(170 W) 573 574 By combining high-resolution global proteomics with direct measurements of intracellular 575 cobalamin, this study provides an integrated view of how temperature shapes the functional 576 landscape of Pseudomonas aeruginosa PA254. It was observed that virulence, motility, 577 metabolism, and metal homeostasis were responsive to a gradient between environmental 578 and host-relevant temperatures. In comparison to other strains, PA254 expresses various 579 proteins involved in biofilm formation and virulent secretion systems throughout the 580 temperature curve, rather than switching between two discrete states. Importantly, the 581 thermal sensitivity of iron acquisition and cobalamin synthesis decoupled from cellular 582 cobalamin pools serve as starting point for nutritional immunity research. In a previous 583 transcriptomic study, 6.4% of the Pseudomonas aeruginosa genome was found to be 584 differentially regulated between 22 and 37 °C. (12) Between the same treatments, the 585 proteomics approach applied here uncovers 8.9% of differentially expressed proteins 586 among the entire proteome, and thus represents an important complementary tool for 587 systems biology analysis. Further studies should integrate temperature gradients with the 588 direct manipulation of metal content and control of oxygen availability for this pathogen. 589 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 27, 2026. ; https://doi.org/10.64898/2026.01.27.701802doi: bioRxiv preprint 590

591 592 The authors report no conflicts of interest. The research was supported by the NIH grant 593 R01GM135709 and the Simons Foundation Microbial Oceanography Project Award to 594 M.A.S. We thank Fadime Stemmer for her help with the code to generate volcano plots. 595 596

597 598 1. W. H. Organization, WHO bacterial priority pathogens list, 2024: bacterial 599 pathogens of public health importance, to guide research, development, and 600 strategies to prevent and control antimicrobial resistance . (World Health 601 Organization, 2024), pp. 72. 602 2. S. Crone, M. Vives -Flórez, L. Kvich, A. M. Saunders, M. Malone, M. H. 603 Nicolaisen, E. Martínez -García, C. Rojas -Acosta, M. Catalina Gomez -Puerto, H. 604 Calum, M. Whiteley, R. Kolter, T. Bjarnsholt, The environmental occurrence of 605 Pseudomonas aeruginosa. APMIS 128, 220-231 (2020). doi: 10.1111/apm.13010. 606 3. G. P. Bodey, R. Bolivar, V. Fainstein, L. Jadeja, Infections Caused by 607 Pseudomonas aeruginosa . Rev. Infect. Dis. 5, 279 -313 (1983). doi: 608 10.1093/clinids/5.2.279. 609 4. G. M. Rossolini, E. Mantengoli, Treatment and control of severe infections caused 610 by multiresistant Pseudomonas aeruginosa . Clin. Microbiol. Infect. 11, 17 -32 611 (2005). doi: 10.1111/j.1469-0691.2005.01161.x. 612 5. J. A. Driscoll, S. L. Brody, M. H. Kollef, The Epidemiology, Pathogenesis and 613 Treatment of Pseudomonas aeruginosa Infections. Drugs 67, 351-368 (2007). doi: 614 10.2165/00003495-200767030-00003. 615 6. K. Bisht, J. L. Moore, R. M. Caprioli, E. P. Skaar, C. A. Wakeman, Impact of 616 temperature-dependent phage expression on Pseudomonas aeruginosa biofilm 617 formation. npj Biofilms Microbiomes 7, 22 (2021). doi: 10.1038/s41522 -021-618 00194-8. 619 7. H. Mikkelsen, M. Sivaneson, A. Filloux, Key two -component regulatory systems 620 that control biofilm formation in Pseudomonas aeruginosa. Environ. Microbiol. 13, 621 1666-1681 (2011). doi: 10.1111/j.1462-2920.2011.02495.x. 622 8. C.-Y. Chang, Surface Sensing for Biofilm Formation in Pseudomonas aeruginosa. 623 Front. Microbiol. 8, (2018). doi: 10.3389/fmicb.2017.02671. 624 9. T. Bjarnsholt, M. Givskov, The role of quorum sensing in the pathogenicity of the 625 cunning aggressor Pseudomonas aeruginosa. Anal. Bioanal. Chem. 387, 409-414 626 (2007). doi: 10.1007/s00216-006-0774-x. 627 10. M. V. Grosso -Becerra, G. Croda -García, E. Merino, L. Servín -González, R. 628 Mojica-Espinosa, G. Soberón -Chávez, Regulation of Pseudomonas aeruginosa 629 virulence factors by two novel RNA thermometers. Proc. Natl. Acad. Sci. 111, 630 15562-15567 (2014). doi: 10.1073/pnas.1402536111. 631 11. O. Wurtzel, D. R. Yoder -Himes, K. Han, A. A. Dandekar, S. Edelheit, E. P. 632 Greenberg, R. Sorek, S. Lory, The Single-Nucleotide Resolution Transcriptome of 633 Pseudomonas aeruginosa Grown in Body Temperature. PLoS Pathogens 8, 634 e1002945 (2012). doi: 10.1371/journal.ppat.1002945. 635 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 27, 2026. ; https://doi.org/10.64898/2026.01.27.701802doi: bioRxiv preprint 12. M. Barbier, F. H. Damron, P. Bielecki, M. Suárez-Diez, J. Puchałka, S. Albertí, V. 636 M. dos Santos, J. B. Goldberg, From the Environment to the Host: Re -Wiring of 637 the Transcriptome of Pseudomonas aeruginosa from 22°C to 37°C. PLoS One 9, 638 e89941 (2014). doi: 10.1371/journal.pone.0089941. 639 13. S. Kim, X. -H. Li, H. -J. Hwang, J. -H. Lee, Thermoregulation of Pseudomonas 640 aeruginosa Biofilm Formation. Appl. Environ. Microbiol. 86, e01584 -01520 641 (2020). doi: 10.1128/AEM.01584-20. 642 14. K.-G. Chan, K. Priya, C. -Y. Chang, A. Y. Abdul Rahman, K. K. Tee, W. -F. Yin, 643 Transcriptome analysis of Pseudomonas aeruginosa PAO1 grown at both body and 644 elevated temperatures. PeerJ 4, e2223 (2016). doi: 10.7717/peerj.2223. 645 15. A. G. Oglesby-Sherrouse, L. Djapgne, A. T. Nguyen, A. I. Vasil, M. L. Vasil, The 646 complex interplay of iron, biofilm formation, and mucoidy affecting antimicrobial 647 resistance of Pseudomonas aeruginosa . Pathogens and Disease 70, 307 -320 648 (2014). doi: 10.1111/2049-632x.12132. 649 16. E. M. Zygiel, C. E. Nelson, L. K. Brewer, A. G. Oglesby -Sherrouse, E. M. Nolan, 650 The human innate immune protein calprotectin induces iron starvation responses in 651 Pseudomonas aeruginosa . J. Biol. Chem. 294, 3549 -3562 (2019). doi: 652 10.1074/jbc.RA118.006819. 653 17. A. P. Kaur, I. B. Lansky, A. Wilks, The Role of the Cytoplasmic Heme -binding 654 Protein (PhuS) of Pseudomonas aeruginosa in Intracellular Heme Trafficking and 655 Iron Homeostasis. J. Biol. Chem. 284, 56 -66 (2009). doi: 656 10.1074/jbc.M806068200. 657 18. E. M. Zygiel, A. O. Obisesan, C. E. Nelson, A. G. Oglesby, E. M. Nolan, Heme 658 protects Pseudomonas aeruginos and Staphylococcus aureus from calprotectin -659 induced iron starvation. J. Biol. Chem. 296, (2021). doi: 660 10.1074/jbc.RA120.015975. 661 19. K. D. Barker, K. Barkovits, A. Wilks, Metabolic Flux of Extracellular Heme 662 Uptake in Pseudomonas aeruginosa Is Driven by the Iron -regulated Heme 663 Oxygenase (HemO). J. Biol. Chem. 287, 18342 -18350 (2012). doi: 664 10.1074/jbc.M112.359265. 665 20. M. R. Gonzalez, V. Ducret, S. Leoni, K. Perron, Pseudomonas aeruginosa zinc 666 homeostasis: Key issues for an opportunistic pathogen. Biochim. Biophys. Acta, 667 Gene Regul. Mech. 1862, 722-733 (2019). doi: 10.1016/j.bbagrm.2018.01.018. 668 21. A. C. S. Meyer, M. R. McIlvin, P. Lopez, B. C. Searle, M. A. Saito, Proteomic 669 profiling of zinc homeostasis mechanisms in Pseudomonas aeruginosa through 670 data-dependent and data-independent acquisition mass spectrometry. Metallomics 671 17, (2025). doi: 10.1093/mtomcs/mfaf020. 672 22. S. Wang, J. Cheng, Y. Niu, P. Li, X. Zhang, J. Lin, Strategies for Zinc Uptake in 673 Pseudomonas aeruginosa at the Host –Pathogen Interface. Front. Microbiol. 12, 674 (2021). doi: 10.3389/fmicb.2021.741873. 675 23. T. E. Kehl-Fie, E. P. Skaar, Nutritional immunity beyond iron: a role for manganese 676 and zinc. Curr. Opin. Chem. Biol. 14, 218 -224 (2010). doi: 677 10.1016/j.cbpa.2009.11.008. 678 24. S. Okamoto, L. D. Eltis, The biological occurrence and trafficking of cobalt. 679 Metallomics 3, 963-970 (2011). doi: 10.1039/c1mt00056j. 680 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 27, 2026. ; https://doi.org/10.64898/2026.01.27.701802doi: bioRxiv preprint 25. A. Crespo, J. Gavaldà, E. Julián, E. Torrents, A single point mutation in class III 681 ribonucleotide reductase promoter renders Pseudomonas aeruginosa PAO1 682 inefficient for anaerobic growth and infection. Sci. Rep. 7, 13350 (2017). doi: 683 10.1038/s41598-017-14051-2. 684 26. A. Crespo, N. Blanco -Cabra, E. Torrents, Aerobic Vitamin B12 Biosynthesis Is 685 Essential for Pseudomonas aeruginosa Class II Ribonucleotide Reductase Activity 686 During Planktonic and Biofilm Growth. Front. Microbiol. 9, (2018). doi: 687 10.3389/fmicb.2018.00986. 688 27. M. A. Saito, M. R. McIlvin, The iron metalloproteome of Pseudomonas aeruginosa 689 under oxic and anoxic conditions. Metallomics 17, (2025). doi: 690 10.1093/mtomcs/mfaf023. 691 28. E. Michetti, V. Secli, M. L. Astolfi, C. Demingo, F. Pacello, S. Ammendola, A. 692 Battistoni, Cobalt mitigates zinc -starvation effects in Pseudomonas aeruginosa . 693 BioMetals, (2025). doi: 10.1007/s10534-025-00769-4. 694 29. M. Richter, R. Rosselló-Móra, F. Oliver Glöckner, J. Peplies, JSpeciesWS: a web 695 server for prokaryotic species circumscription based on pairwise genome 696 comparison. Bioinformatics 32, 929 -931 (2015). doi: 697 10.1093/bioinformatics/btv681. 698 30. D. Hyatt, G. -L. Chen, P. F. LoCascio, M. L. Land, F. W. Larimer, L. J. Hauser, 699 Prodigal: prokaryotic gene recognition and translation initiation site identification. 700 BMC Bioinformatics 11, 119 (2010). doi: 10.1186/1471-2105-11-119. 701 31. C. P. Cantalapiedra, A. Hernández -Plaza, I. Letunic, P. Bork, J. Huerta -Cepas, 702 eggNOG-mapper v2: Functional Annotation, Orthology Assignments, and Domain 703 Prediction at the Metagenomic Scale. Mol. Biol. Evol. 38, 5825-5829 (2021). doi: 704 10.1093/molbev/msab293. 705 32. Y. Hsiao, H. Zhang, G. X. Li, Y. Deng, F. Yu, H. V. Kahrood, J. R. Steele, R. B. 706 Schittenhelm, A. I. Nesvizhskii, Analysis and Visualization of Quantitative 707 Proteomics Data Using FragPipe-Analyst. J. Proteome Res. 23, 4303–4315 (2024). 708 doi: 10.1021/acs.jproteome.4c00294. 709 33. M. Mars Brisbin, A. Schofield, M. R. McIlvin, A. I. Krinos, H. Alexander, M. A. 710 Saito, Vitamin B12 conveys a protective advantage to phycosphere -associated 711 bacteria at high temperatures. ISME Communications 3, (2023). doi: 712 10.1038/s43705-023-00298-6. 713 34. F. Ito, T. Tamiya, I. Ohtsu, M. Fujimura, F. Fukumori, Genetic and phenotypic 714 characterization of the heat shock response in Pseudomonas putida . 715 MicrobiologyOpen 3, 922-936 (2014). doi: 10.1002/mbo3.217. 716 35. J. Melamed, A. Kocev, V. Torgov, V. Veselovsky, I. Brockhausen, Biosynthesis of 717 the Pseudomonas aeruginosa common polysaccharide antigen by D -718 Rhamnosyltransferases WbpX and WbpY. Glycoconjugate J. 39, 393-411 (2022). 719 doi: 10.1007/s10719-022-10040-4. 720 36. J. Ravel, P. Cornelis, Genomics of pyoverdine -mediated iron uptake in 721 pseudomonads. Trends Microbiol. 11, 195 -200 (2003). doi: 10.1016/S0966 -722 842X(03)00076-3. 723 37. C. Liao, X. Huang, Q. Wang, D. Yao, W. Lu, Virulence Factors of Pseudomonas 724 aeruginosa and Antivirulence Strategies to Combat Its Drug Resistance. Front. 725 Cell. Infect. Microbiol. 12, (2022). doi: 10.3389/fcimb.2022.926758. 726 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 27, 2026. ; https://doi.org/10.64898/2026.01.27.701802doi: bioRxiv preprint 38. G. Laverty, S. P. Gorman, B. F. Gilmore, Biomolecular Mechanisms of 727 Pseudomonas aeruginosa and Escherichia coli Biofilm Formation. Pathogens 3, 728 596-632 (2014). doi: 10.3390/pathogens3030596. 729 39. Y. Nguyen, S. Sugiman -Marangos, H. Harvey, S. D. Bell, C. L. Charlton, M. S. 730 Junop, L. L. Burrows, Pseudomonas aeruginosa Minor Pilins Prime Type IVa Pilus 731 Assembly and Promote Surface Display of the PilY1 Adhesin. J. Biol. Chem. 290, 732 601-611 (2015). doi: 10.1074/jbc.M114.616904. 733 40. A. M. Kollaran, S. Joge, H. S. Kotian, D. Badal, D. Prakash, A. Mishra, M. Varma, 734 V. Singh, Context -Specific Requirement of Forty -Four Two-Component Loci in 735 Pseudomonas aeruginosa Swarming. iScience 13, 305 -317 (2019). doi: 736 10.1016/j.isci.2019.02.028. 737 41. J. Tremblay, E. Déziel, Gene expression in Pseudomonas aeruginosa swarming 738 motility. BMC Genomics 11, 587 (2010). doi: 10.1186/1471-2164-11-587. 739 42. I. J. Schalk, O. Cunrath, An overview of the biological metal uptake pathways in 740 Pseudomonas aeruginosa. Environ. Microbiol. 18, 3227 -3246 (2016). doi: 741 10.1111/1462-2920.13525. 742 43. A. Haritha, A. Rodrigue, P. M. Mohan, A comparative analysis of metal 743 transportomes from metabolically versatile Pseudomonas. BMC Research Notes 1, 744 88 (2008). doi: 10.1186/1756-0500-1-88. 745 44. D. A. Rodionov, A. G. Vitreschak, A. A. Mironov, M. S. Gelfand, Comparative 746 Genomics of the Vitamin B12 Metabolism and Regulation in Prokaryotes. J. Biol. 747 Chem. 278, 41148-41159 (2003). doi: 10.1074/jbc.M305837200. 748 45. J. S. McFarlane, A. L. Lamb, Biosynthesis of an Opine Metallophore by 749 Pseudomonas aeruginosa . Biochemistry 56, 5967 –5971 (2017). doi: 750 10.1021/acs.biochem.7b00804. 751 46. R. E. Robinson, J. K. Robertson, S. M. Prezioso, J. B. Goldberg, Temperature 752 controls LasR regulation of piv expression in Pseudomonas aeruginosa. mBio 16, 753 e00541-00525 (2025). doi: 10.1128/mbio.00541-25. 754 47. E. Termine, G. P. Michel, Transcriptome and secretome analyses of the adaptive 755 response of Pseudomonas aeruginosa to suboptimal growth temperature. 756 International Microbiology 12, 7-12 (2009). doi: 10.2436/20.1501.01.76. 757 48. G. T. Walker, M. Raffatellu, Salmonella respiration turns the tables on propionate. 758 Trends Microbiol. 30, 206-208 (2022). doi: 10.1016/j.tim.2022.01.011. 759 49. S. Ahn, J. Jung, I. -A. Jang, E. L. Madsen, W. Park, Role of Glyoxylate Shunt in 760 Oxidative Stress Response. J. Biol. Chem. 291, 11928 -11938 (2016). doi: 761 10.1074/jbc.M115.708149. 762 50. S. S. Yoon, R. F. Hennigan, G. M. Hilliard, U. A. Ochsner, K. Parvatiyar, M. C. 763 Kamani, H. L. Allen, T. R. DeKievit, P. R. Gardner, U. Schwab, J. J. Rowe, B. H. 764 Iglewski, T. R. McDermott, R. P. Mason, D. J. Wozniak, R. E. W. Hancock, M. R. 765 Parsek, T. L. Noah, R. C. Boucher, D. J. Hassett, Pseudomonas aeruginosa 766 Anaerobic Respiration in Biofilms: Relationships to Cystic Fibrosis Pathogenesis. 767 Developmental Cell 3, 593-603 (2002). doi: 10.1016/S1534-5807(02)00295-2. 768 51. J. M. Weiner, W. H. Lee, E. M. Nolan, A. G. Oglesby, Calprotectin elicits aberrant 769 iron starvation responses in Pseudomonas aeruginosa under anaerobic conditions. 770 J. Bacteriol. 207, e00029-00025 (2025). doi: 10.1128/jb.00029-25. 771 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 27, 2026. ; https://doi.org/10.64898/2026.01.27.701802doi: bioRxiv preprint 52. D. Worlitzsch, R. Tarran, M. Ulrich, U. Schwab, A. Cekici, K. C. Meyer, P. Birrer, 772 G. Bellon, J. Berger, T. Weiss, K. Botzenhart, J. R. Yankaskas, S. Randell, R. C. 773 Boucher, G. Döring, Effects of reduced mucus oxygen concentration in airway 774 Pseudomonas infections of cystic fibrosis patients. The Journal of Clinical 775 Investigation 109, 317-325 (2002). doi: 10.1172/JCI13870. 776 53. A. Crespo, L. Pedraz, J. Astola, E. Torrents, Pseudomonas aeruginosa Exhibits 777 Deficient Biofilm Formation in the Absence of Class II and III Ribonucleotide 778 Reductases Due to Hindered Anaerobic Growth. Front. Microbiol. Volume 7 - 779 2016, (2016). doi: 10.3389/fmicb.2016.00688. 780 781 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted January 27, 2026. ; https://doi.org/10.64898/2026.01.27.701802doi: bioRxiv preprint