https://doi.org/10.5061/dryad.fj6q5742z

Description of the data and file structure

All data were collected using the same microscope and imaging settings, as detailed in the associated manuscript. The dataset is provided as a compressed folder containing all data used to generate the main and Supplementary Figures.

The folder is organized by figure number. Within each subfolder, compiled raw data—processed following segmentation of microcolonies, individual cells, and aggregates—are available in .csv or .mat formats.

MATLAB scripts used for data processing and visualization are included in .m format. When a separate supplementary data file is not provided, the relevant data are embedded in the initial lines of the corresponding MATLAB script. For epifluorescence microscopy images, kindly contact the authors.

Files and variables

File: Dryad_DataRepository.zip

Description: 

Single cell segmentation. Scripts for single cells analysis and sample images are provided to extract the single cell from .tiff images. The parameters used for this study are the number of cells segmented and the average fluorescence per cell. This information was then compiled into .mat or .xlsx files and used to quantify the amount of fluorescence signal, number of cells fluorescing, and proportion of cells fluorescing.

Aggregate segmentation. Scripts developed for aggregate identification are also provided for the data in the supporting information. Phase contrast images were used for aggregates segmentations, mCherry fluorescence channel was used for fluorescence extraction, and Brightfield for RGB color parameters.

ICP-MS. Aqueous manganese was measured using ICP-MS and the results are reported in .xlsx files, including the sample name, original dilution, dilution factors, amounts of acid added, ICP-MS reads, and calculation to concentrations in µM. When file is not provided, the ICP-MS data is contained in the .m scripts.

Flase color images. The script and sample images are provided to generate false color of the segemented cells using the mask obtained during single cell segmentation analysis. 

Name convention:

• Bioreporter for mnxG promoter: PmnxG, P2447, P24
• Bioreporter for mcoA promoter: PmcoA, P2665, P26
• mnxG Gene deletion: Delta2447, Delta_mnxG
• mcoA Gene deletion: Delta2665, Delat_mcoA
• mnxG and mcoA double gene deletion: Delta2447_2665, DeltamnxGmcoA
• Mcherry or m-cherry: mCherry fluorescence channel
• PC: Phase contrast channel
• WT: Wild-type
• t0xy001: time point 0, position 1
• Mn: Manganese

All fluorescence data has arbitrary units (a.u.) that are specific to the epifluorescence microscopy settings (described in the materials and methods of the corresponding publication). Manganese concentration data was collected in ppb and converted to µM. 

Missing values are indicated by NaN. 

Specific Folder content

Figure 1_Bimodal Promoter activation. This folder contains the TIFF images and scripts to extract the proportion of cell fluorescing. This file is structured by Mn concentration with folder "1 uM", following by subfolders for the strains studied: "P2447", "P2665", and "WT". Each of these folder is then organized by time point, where original TIFF images in the mCherry, and Phase contrast (PC) channels, "FalseColorMap.m" MATLAB script, "qqplot.svg", "Segmentation_txy_seg.mat", and a final falsed color TIFF figure "FalseColor.tiff", can be found for each time point. The time points are 16h, 18h, 20h, and 20h for the strain P2447; 18h, 20h, and 24h for strain P2665; and 18h, 20h for WT strain. 

The script "FalseColorMap.m" processes phase-contrast and mCherry fluorescence images to quantify the proportion of a cell population exhibiting reporter fluorescence. The precomputed cell segmentation mask (e.g., "Segmentation_t0xy002_seg.mat") is used to isolate individual cells, estimate background intensity, and compute per-cell mean fluorescence values. Cells are classified as fluorescent or non-fluorescent using a fixed intensity threshold, and a false-color image is generated to visualize classification results. The script reports the percentage of fluorescent cells and exports a high-resolution false-color overlay. Quantile–quantile (QQ) plots are additionally generated to assess fluorescence intensity distributions and support threshold selection across single images and pooled single-cell datasets.

Figure 2a_Average Fluorescence. In this folder, input files (.mat) and the script "AverageFluoProcessing48h.m" can be found. This script consolidates single-cell fluorescence measurements from three biological replicates (R1–R3) for two reporter strains (P2447 / PmnxG and P2665 / PmcoA) across a Mn(II) dose series, then visualizes and models the combined response. It loads replicate .mat files (e.g., FLUO_R1.mat) plus supplemental datasets for specific concentrations (notably 400 µM and updated/thresholded 500 µM, and separate 0 and 1 µM “Report” exports), inserts those data into the appropriate positions of the FLUO cell arrays, and pools all replicate values by condition to compute mean and standard deviation per Mn level. The script generates plots: (i) combined swarm/summary plots comparing both promoters across Mn(II) concentrations, (ii) strain-specific swarm and boxchart variants, and (iii) log-scale dose–response scatter plots with linear regression on log([Mn]) to estimate response slope and goodness-of-fit (R²) for each promoter, exporting figures (PNG/SVG/EPS) for downstream use.

• "FLUO_R1.mat", "FLUO_R2.mat", "FLUO_R3.mat": Replicate fluorescence datasets. Each file contains a FLUO cell array with per-cell reporter fluorescence values for different Mn(II) concentrations and strains. These are the primary single-cell fluorescence inputs.
• "Mn400_P24A_ExportmChFlo2.mat", "Mn400_P24B_ExportmChFlo2.mat", "Mn400_P24C_ExportmChFlo2.mat": Replicate fluorescence datasets for strain P2447 at 400 µM Mn(II). Contain per-cell mCherry fluorescence values (Flo2) extracted from image analysis.
• "Mn400_P26A_ExportmChFlo2.mat", "Mn400_P26B_ExportmChFlo2.mat", "Mn400_P26C_ExportmChFlo2.mat" : Replicate fluorescence datasets for strain P2665 at 400 µM Mn(II). Same structure and purpose as the P2447 files.
• "P24_400.mat", "P26_400.mat": Additional or pooled fluorescence data for P2447 and P2665 at 400 µM Mn(II), used to supplement or merge with replicate datasets.
• "P24_500.mat", "P26_500.mat": Fluorescence data for P2447 and P2665 at 500 µM Mn(II). These datasets are already thresholded and replace one replicate in the combined analysis.
• "Report_P24_0uM.mat", "Report_P26_0uM.mat": Single-cell reporter fluorescence measurements for P2447 and P2665 at 0 µM Mn(II) (baseline / control condition).
• "Report_P24_1uM.mat", "Report_P26_1uM.mat": Single-cell reporter fluorescence measurements for P2447 and P2665 at 1 µM Mn(II) (low-Mn condition).

Figure 2b&3a_Average Population active promoter. This folder contains "PopulationKinetics.m" processing scripts for replicate 1 and 2, and "PopulationKineticsRep3Reprocessed.m" for replicate 3 processing. The first script aggregates single-cell fluorescence data from multiple biological replicates and manganese (Mn²⁺) concentrations to quantify reporter activity across conditions. It loads pre-processed fluorescence measurements for different promoters (P2447 / PmnxG and P2665 / PmcoA), combines replicates, computes mean and variance per Mn²⁺ level, and visualizes the results using swarm plots, boxplots, and log-scaled dose–response regressions. The script also fits linear models to log-transformed Mn²⁺ concentrations to estimate response rates and generates publication-ready figures comparing promoter activity across strains and conditions. The second script compiles time-series data for multiple biological replicates (promoter-reporter strains and wild type) across a range of initial Mn(II) concentrations, then computes replicate-averaged trajectories (mean ± variability with NaN-safe handling) for (i) the proportion of the population fluorescing and (ii) dissolved Mn(aq) kinetics. The script generates multi-panel figures comparing promoters/strains and WT over time for each Mn condition, exports publication-ready graphics (SVG/PNG), and calculates steady-state activation metrics from the final time points to produce summary plots (linear and log-scaled Mn axes) and bar charts. The workflow assumes replicate matrices are already “onset-corrected” where indicated and requires external helper functions (e.g., mseb) available on the MATLAB path.

The file "R3.mat" is the third biological replicate of single-cell fluorescence measurements, imported separately. 

Figure 3b_Mn oxidation kinetics. This folder contains the processing script "MnKinetics.m", with raw data and processing within the script. This MATLAB script processes Mn(II) time-series kinetics data across multiple initial Mn doses (1, 10, 50/60, 100/120, 250/300, 400, 500, and 600 µM), stored as hard-coded matrices (K1, K10, K50, K100, K250, K400, K500, K600) where column 1 is time (h) and subsequent columns are replicate Mn(aq) measurements (µM). For each dose, it computes the mean (aK*) and an uncertainty band (kSTD*, based on replicate standard error scaled by a t-multiplier), removes rows with missing averages, and generates stacked/tiled plots of Mn(aq) vs time with scatter points and shaded error envelopes (requires mseb, added via addpath(genpath(...))). The script also produces diagnostic linearized plots for first-order (log([Mn])) and second-order (1/[Mn]) transforms, then estimates apparent removal rates by fitting linear models (fitlm) over selected time windows for each replicate and dose, extracting slopes and R² values. Figures are exported as SVG and PNG (e.g., KineticsProp.svg/.png, KineticsProp_1stOrder.*, KineticsProp_2stOrder.*, plus shortened summary variants). No external .mat input (e.g., R3.mat)**** is used in this script—all kinetics data are defined directly in the code. 

Figure 4_Kinetics of Promoter activation. This folder contains the data and script to compute the kinetics of promoter activation. This script quantifies and visualizes the initial rate of promoter activation (population “turn-on” dynamics) as a function of Mn(II) concentration, and compares these biological rates to Mn oxide precipitation rates derived independently (e.g., from Michaelis–Menten / kinetics fits). The script defines promoter-activation time series for two promoter reporters (P_mnxG and P_mcoA) across a range of Mn(II) treatments (≈0, 1, 10, 50–60, 100–120, 250–300, 400, 500 µM), with replicate measurements stored as matrices (e.g., Mn1, Mn10, Mn50, P1, P10, P50). For each treatment, it estimates initial activation rates (% h⁻¹) using linear regression (fitlm) over an early time window selected per replicate (to focus on the onset/linear rise phase), then compiles mean rates and standard deviations (rP2447, stdP2447, rP2665, stdP2665). In parallel, the script loads/defines Mn oxide precipitation rates (aMn2, stdMn2) and produces figures that (i) overlay promoter activation rates for both reporters against Mn(II) on a log-scaled x-axis with Mn oxide rates on a secondary axis, (ii) show an alternative two-panel subplot separating promoter activation from Mn oxide rates, (iii) optionally summarize selected conditions (e.g., 50, 250, 400 µM) as bar/jitter plots, and (iv) generate a log–log version of the Mn oxide rate plot (excluding zero values). Output figures are saved as SVG files (e.g., Rate_ProportionPromoter_MnOx.svg, Rate_ProportionPromoter_MnOx_V2.svg, Rate_ProportionPromoter_MnOx_V3.svg). The script requires MATLAB’s Statistics toolbox (fitlm) and uses mseb for shaded error bars (ensure mseb is on the MATLAB path via the addpath(genpath(...)) call).

Figure 5_MnKineticsDel. In this folder, the raw ICP-MS results of the Mn concentration and kinetic processing can be found. The follwing spreadsheet summarize ICP-MS manganese (Mn) measurements and the dilution/normalization steps used to convert instrument output back to the original Mn concentration for each sample. Each row is a single sample ID in the Name column (e.g., t14_WT_50, t16_del24_250B), where the naming convention encodes the time point (e.g., t14, t16, t24), strain/genotype (e.g., WT, del24, del26), and the nominal Mn treatment level (e.g., 50, 250). Real expected concentration (µM) provides the intended starting Mn concentration for comparison, while [Target] and ppm track the working target concentration used for the ICP-MS run and/or standards. The middle columns document all dilution bookkeeping: DF (dilution factor for transferring sample into a final 5 mL digest), Volume sample (mL) in 5 mL and Volume total (final digest volume), and the acidification/digestion volumes (Target volume HNO3, Weighed HNO3 1%) alongside weight sample. Any additional upstream dilution is recorded under Predilution, and Final dilution factor captures the total effective dilution applied prior to analysis. ICP results reports the instrument readout for Mn in the measured solution, which is then converted to Mn_uM (Mn concentration in µM in the measured/diluted solution) and finally to True Mn, which represents the Mn concentration back-calculated to the original sample using the listed dilution factors (and is the main value to use for comparing across treatments and replicates).

• "ICP_del24_del26_Kinetics_Run2.csv": ICP-MS results for sorption kinetic experiment using the double knockout strain, second replicate
• "ICP_Sorption_Mn50_deldel.csv": ICP-MS results for sorption kinetic experiment using the double knockout strain.
• "ICP-MS_MnKineticsDel_may26.csv": ICP-MS results for wild-type (WT), and single knockout strains at 400 µM MnCl₂. Letters a,b, and c represent triplicate results.
• "ICP230203_WT_del24_del26.csv":  ICP-MS results for wild-type (WT), and single knockouts strains at 50 and 250 µM Mn(II). Time points where measured at 14h, 16h, 18h, 20h, 24h, and 39h.
• "ICPRerun_Oxidized samples.csv": Rerun of ICP-MS on previous samples that did not pass QA/QC.
• "MnKineticsdel.csv":  ICP-MS results for wild-type (WT), and single knockouts strains at 50 and 250 µM Mn(II). Time points where measured at 14h, 16h, 18h, 20h, 24h, 41h, and 48h. This dataset correspond to an other biological replicate.
• "ODkinetics.csv": Optical Density (OD) point measurements at 14h. The units are given in absorbance units (abs.).
• "Sorption400_ICP.csv": ICP-MS results for wild-type (WT), and single knockout strains at 400 µM MnCl₂. Letters a,b, and c represent triplicate results. "t" represents the time point in hours (e.g., t18 = 18h time point).

Next, the folder contains two subfolder. The "500uM" subfolder contains the file "ICP_Kinetics_Mnc_del500_WT300_uM.csv", which are the ICP-MS results of the WT and single knockout strains for the 500 µM and 300 µM Mn(II) conditions. The folder "Figure" contains the scripts and compiled spreadsheets-input file to generate the kinetic figures:

• "Kinetic50.csv": contains the processed data with time point (in hours) and amount of Mn(II) in solution (in µM) for the WT and both single knockout strains. The second table calculates the amount of Mn oxide present by mass balance: Mn oxides = Mn total - Mn(aq), with Mn total as 55 µM.
• "Kinetic250.csv": contains the processed data with time point (in hours) and amount of Mn(II) in solution (in µM) for the WT and both single knockout strains at 250 µM initial Mn(II).
• "Kinetic400_R1.csv": contains the processed data with time point (in hours) and amount of Mn(II) in solution (in µM) for the WT and both single knockout strains at 400 µM initial Mn(II) for Replicate 1.
• "Kinetic400_R2.csv": contains the processed data with time point (in hours) and amount of Mn(II) in solution (in µM) for the WT and both single knockout strains at 400 µM initial Mn(II) for Replicate 2.
• "Kinetic400_R3.csv": contains the processed data with time point (in hours) and amount of Mn(II) in solution (in µM) for the WT and both single knockout strains at 400 µM initial Mn(II) for Replicate 3.

The script "FigureKin50uM_250_400.m" compiles and visualizes Mn(II) oxidation kinetics, with a focus on generating the 50 µM time-course figure (and parallel analyses for 250 µM and 400 µM). It defines time vectors (hours) and replicate measurements for WT, ΔmcoA, ΔmnxG, and the double knockout (ΔmcoA ΔmnxG), then converts the raw measurements to Mn oxide produced by subtracting values from the initial Mn(II) pool (e.g., 55 - data for the 50 µM experiment, 250 - data for 250 µM, and 400 - data for 400 µM). For each condition, the script calculates the mean and standard deviation across replicates using nanmean / nanstd (NaNs preserved for missing datapoints). It generates kinetics plots using scatter points with error bars (and optional shaded-error plotting via mseb), performs linear fits over the early time points (using polyfit and fitlm) to estimate initial oxidation rates and report R², and saves figures as SVG files (e.g., 1_Kinetics50uM_V2.svg, 2_Kinetics250uM_V2.svg, 3_Kinetics400uM_V2.svg). A final section repeats the 50 µM workflow on normalized kinetics (MnOx / MnOx_tot), enabling direct comparison of oxidation progression independent of absolute oxide yield.

The script "FigureKineticsDel.m" reproduces the Mn(II) oxidation / MnOx precipitation kinetics analysis for a knockout experiment across multiple starting Mn(II) concentrations (50, 250, 400, and 500 µM). It begins by loading time‐series kinetics tables from Excel files (Kinetic50.csv, Kinetic250.csv, and three independent 400 µM replicates: Kinetics400_R1.csv, Kinetics400_R2.csv, Kinetics400_R3.csv). The tables are converted to arrays (K50, K250, K400_1–3) and supplemented with manually defined replicate matrices for select conditions (e.g., WT50, del2450, del2650, Mn250, WT250, and Mn400WT). The script first plots raw dissolved Mn(aq) kinetics as scatter + dashed line traces for WT and knockout strains (Δ2447 / Δ2665) at 50 and 250 µM, and displays three separate 400 µM replicate panels (WT shown; mutant plotting for 400 µM is present but commented out). It then computes maximum stepwise oxidation rates using first differences (diff) and extracts average MnOx precipitation rates by fitting linear regressions (fitlm) over user-selected time windows corresponding to the “linear” portion of each trajectory. Slopes from multiple replicates are assembled into matrices (S50, S250, and Mx for 400 µM) and summarized as mean ± SD precipitation rates for each strain and concentration; 500 µM rates are included from precomputed slope vectors (wt_500, del26_500, del24_500). Finally, the script generates a bar plot of mean precipitation rates (50/250/400/500 µM) with overlaid jittered points showing replicate slopes, exports figures (KinBar, KinBar.svg), and performs basic statistics (one-way ANOVA with anova1 and post-hoc multcompare) plus a variance test (vartestn, Levene absolute) on selected replicate slope sets.

The script "FigureMnOx.m" generates the Mn oxide precipitation figure for the knockout experiment by compiling replicate MnOx measurements at four starting Mn(II) concentrations (50, 250, 400, and 500 µM) and comparing strains (WT, ΔmcoA / “del24”, ΔmnxG / “del26”, and the double knockout / “deldel”). The MnOx data are imported directly as four matrices (A50, A250, A400, A500), where columns correspond to strain (WT, del24, del26, deldel) and rows are replicate measurements; missing replicates are stored as NaN and ignored during averaging. The script computes the mean MnOx produced for each strain at each Mn level (a50, a250, a400, a500), stacks these into an average matrix, and then creates a grouped bar plot (one group per Mn level, one bar per strain). Individual replicates are overlaid as jittered scatter points on top of each bar to show replicate variability. Finally, the figure is formatted (colors, labels, axis limits) and exported as MnOxBar and MnOx.svg. A one-way ANOVA is run (as written, on A400) followed by a post-hoc multiple comparison (multcompare) to assess differences among strains at the 400 µM condition.

The script "Script_RateCalculationV2.m" calculates Mn oxide precipitation (or Mn loss) rates for knockout strains by fitting linear regressions to selected time windows of Mn concentration time series and extracting the slope as the rate (µM h⁻¹). It first evaluates higher-concentration datasets (e.g., ~400 µM) for ΔmcoA (K1) and ΔmnxG (K2), visualizing replicate trajectories with quick diagnostic plots and then using fitlm() on user-defined index ranges (chosen to represent the most linear portion of the kinetics). For each replicate column, the script records the regression R² and the slope (mdl.Coefficients.Estimate(2)), then summarizes replicate-to-replicate variability using nanmean and nanstd (handling missing values). The workflow is repeated for lower concentration (50 µM) datasets using separate K1/K2 matrices and short early-time windows, and again for higher concentration (500 µM) datasets with time windows shifted later in the experiment to capture the linear regime. Overall, this file is a “rate extraction” utility: it does not produce final publication plots by itself, but it standardizes how slopes (rates) are computed across genotypes, concentrations, and replicate sets by explicitly documenting the regression windows used.

Figure 6_Michaelis-Menten Equation. This folder contains the scripts and .mat input to run Michaelis-Menten models on enzymatic Mn(II) oxidation. The script "MMequationKnockouts.m" compares experimentally derived Mn oxide precipitation rates to Michaelis–Menten (MM) kinetic models for WT and knockout strains. It assembles rate observations across multiple Mn(II) concentrations by pooling replicate rate measurements into matrices (r3, MM) and computing mean ± standard deviation of absolute rates. Using published/previously fit MM parameters (Vmax and Km; plus separate parameter sets for ΔmnxG, ΔmcoA, and pooled data), it generates modeled rate curves using the MM form v = (Vmax·S)/(Km + S) over a concentration grid (S). The script produces multiple figures: (i) MM curve + raw rate points for WT, (ii) MM fits for ΔmnxG and ΔmcoA with replicate scatter, (iii) a combined comparison plot, and (iv) a version using mean ± SD with optional shaded confidence intervals from nonlinear fits (fit + confint). It also generates a log–log plot of rate vs Mn concentration to estimate apparent reaction order via linear regression in log space. Figures are saved as SVG/PNG (e.g., MM_KineticsWT.svg, MM_KineticsCombinedaverage.png, Log_MnRates.svg), providing publication-ready summaries of saturation behavior and genotype-specific kinetic differences.

The file "vmax_Mnx.mat" is a MATLAB data file that stores pre-computed kinetic parameters used for Michaelis–Menten modeling of Mn(II) oxidation rates. It contains fitted values (or intermediate results) related to Vmax, Km, or rate matrices derived from prior analyses, which are reused in this script rather than recalculated.

Figure SI 1_WildType Autofluorescence. Autofluorescence signal measured at 1 µM and 400 µM Mn(II). In this folder, the raw fluorescence data per cell is contained in the CSV files. Each column corresponds to a time point in hours (14h, 16h, 18h, 20h, 24h, 48h, 72h). All fluorescence values per cells have been background subtracted (as described in the methods of the corresponding manuscript). The files "P24_400A.csv", "P24_400B.csv", and "P24_400C.csv" are biological triplicates of the strain P2447 at 400 µM Mn(II). The files "P24_1A.csv", "P24_1B.csv", and "P24_1C.csv" are biological triplicates of the strain P2447 at 1 µM Mn(II). The files "WT_400A.csv", "WT_400B.csv", and "WT_400C.csv" are biological triplicates of the wild-type strain at 400 µM Mn(II). The files "WT_1A.csv", "WT_1B.csv", and "WT_1C.csv" are biological triplicates of the wild-type strain at 1 µM Mn(II). 

The files "Strain3.mat", "Strain1.mat", and "WTdata.mat" contain the data structures for fluorescence signal at all time point tested and for all the strains used in this study. 

The processing scrip "WTprocessing.m" processes time-resolved single-cell fluorescence data for wild-type strains across multiple Mn concentrations. It loads background-corrected raw fluorescence data from three biological replicates, merges replicates into a unified structure, and imports additional measurements at 1 µM and 400 µM Mn from Excel files. The script filters outliers and negative values, summarizes fluorescence distributions over time using box and swarm plots, and computes high-quantile (99.5%) and mean fluorescence values. It then quantifies the proportion of the population exceeding a defined fluorescence threshold at each condition, enabling comparison of autofluorescence and active subpopulations across Mn concentrations. The output consists of publication-ready figures showing temporal fluorescence distributions, threshold-based population fractions, and calibration plots used for downstream kinetic and population analyses.

The script "BackgroundAnalysis.m" aggregates and quality-controls single-cell fluorescence data for the wild-type strain across three biological replicates, then summarizes autofluorescence/background behavior across Mn(II) conditions. It loads preprocessed (background-corrected) strain structures from WTdata.mat, Strains1.mat, and Strain3.mat, reorganizes them into a unified WT replicate structure, and imports additional validation datasets from CSV files for WT and P2447 at 1 µM and 400 µM Mn(II). For each replicate/timepoint, it removes NaNs and negative values, filters extreme outliers using a sigma-based cutoff, and visualizes distributions over time using boxcharts (with the 99.5th percentile overlaid). It then pools populations at 14 h (and related early timepoints), appends the 1 µM and 400 µM datasets, computes the fraction of cells above a fluorescence threshold (default 550 a.u., with a different threshold used for the 400 µM case), and generates swarm plots to support threshold calibration across Mn(II) concentrations. Outputs include figures such as CalibrationThreshold.svg and WT_autofluorescence.svg.

Figure SI 2_Average Aggregates Activation. The script "PostprocessingAggregateSinglecells.m" compiles postprocessed fluorescence/aggregate-intensity measurements across Mn(II) concentrations for three strains (WT, P2447, P2665), then summarizes and visualizes how signal changes with Mn dose. It loads hard-coded replicate matrices where the first row contains Mn(II) concentrations (0.04–500 µM) and subsequent rows contain replicate measurements (with NaNs for missing values). The script computes mean ± SD across replicates for each strain and generates two figures: (1) a log-scaled concentration–response plot (mean ± SD) comparing absolute fluorescence intensity among strains, and (2) the same plot after subtracting the WT mean at each concentration to remove baseline signal (WT-corrected response). Figures are saved as SVG files (AggPromoter.svg and AggPromoter_Log.svg).

Figure SI 4_Growth Rate Mnc. The script "GrowthrateMn.m" is used to analyze how microbial growth rate varies with dissolved Mn(II) concentration. It loads preprocessed growth-rate datasets from "RargMn.mat", "Rarg2.mat", and "RateSorted.mat", computes mean growth rates and variability across replicates, and evaluates trends using linear regression and polynomial fitting. Growth rates are plotted against Mn concentration on both linear–log and log–log scales to assess concentration-dependent responses, including nonlinearity at higher Mn levels. The script outputs publication-ready figures showing growth rate versus Mn concentration and logarithmic fits used to interpret kinetic behavior.

Figure SI 7_Promoter rate vs Mn rate. The script "RatesvsPromters.m" compares Mn oxide precipitation rates with promoter activation rates for the PmnxG (P2447) and PmcoA (P2665) strains across multiple Mn concentrations. It loads precomputed Mn precipitation rates and promoter activation datasets ("RatesMn.mat"), reorganizes and sorts the data to align replicates, and performs linear regressions to quantify correlations between biological activity (promoter activation, % h⁻¹) and Mn oxidation rates (µM h⁻¹). The script generates multiple scatter plots with fitted regression lines, including forward and inverted axes, and separately analyzes low-Mn and high-Mn concentration regimes to assess changes in coupling strength. Output figures are saved as SVG and PNG files.

Figure SI 9_XAS single knockouts. This folder contains two subfolders: "Postprocessed Data" and "Pure-valence references_XANES(mu)". The first subfolder contains the XANES scans of the wild-type (WT), single mnxG knockout (del24), and single mcoA knockout (del26) for Mn oxides produced in Lept growth medium with 500 µM Mn(II) (e.g., "Lept500_....mu") and MST growth medium with 50 µM Mn(II) (e.g., "MST50_...mu"). The labels "avXscans" represent the number of scans averaged (e.g., "_av3scans" are an average of three). The label "fl" signifies that the XANES were acquired with fluorescence detector. 

The next folder "Pure-valence references_XANES(mu)" contains all the reference minerals as .mu files for manganese oxides. The Mn speciation in the file name corresponds to a pure mineral phase with this exact speciation. This is used to compare our samples against these pure references.

The scipt "PostProcessing_XASknockout.m" imports Mn K-edge XANES fluorescence spectra for wild type and single-knockout strains (ΔmcoA, ΔmnxG) measured in two culture conditions (Lept at 500 µM Mn and MST/DMOx at 50 µM Mn), along with pure-valence reference spectra (Mn(IV) δ-MnO₂, Mn(III) acetate, and Mn(II) rhodocrosite). It generates stacked/vertically shifted overlay plots to visually compare sample spectra to oxidation-state standards, annotates each trace, and exports figures (SVG/PNG).

The file "TableXANES_fitting.csv" summarizes the linear combination fitting (LCF) results of Mn K-edge XANES spectra for Lept and MSTA strains, including ΔmnxG, ΔmcoA, and wild-type. Reported values are the fitted fractional contributions (weights ± fitting error) of δ-MnO₂ (Mn(IV)) and biogenic birnessite formed in the presence of glucose, along with the associated R-factor indicating goodness of fit. The results highlight distinct differences in Mn oxidation state distributions across strains and genetic backgrounds, with ΔmnxG samples showing higher δ-MnO₂ contributions, while ΔmcoA and wild-type samples are dominated by birnessite-like Mn phases.

Figure SI 10_Alphafold-multimer. This file contains the full-length amino acid sequence of the manganese-oxidizing multicopper oxidase MnxG ("Pputida_GB1_MnxG.txt") from Pseudomonas spp., compiled from identical annotations across multiple strains (100% identity over 1944 aa). The sequence was used as an input for AlphaFold-Multimer analyses in combination with McoA ("Pputida_GB1_McoA.txt") to test whether MnxG and McoA are capable of forming a stable multimeric complex. The BLAST alignment confirms complete sequence conservation, supporting the use of a single representative MnxG sequence for structural modeling and protein–protein interaction predictions.

Number of cells segmented. This folder contains the summary CSV file "NumberCellsSeg.csv". This table summarizes the total number of individual cells successfully segmented and analyzed for the wild-type (WT) strain across all experimental conditions in this study. Cell counts are reported for three biological replicates (Rep 1–3) at multiple time points (12–72 h) and Mn(II) concentrations (0–500 µM). Because replicates were not always acquired at identical time points, counts are listed per replicate with the corresponding acquisition time, and a combined total (“Sum”) is provided for each condition and time point.

Code/software

MATLAB version R2021b