Bottom-up reconstruction of synthetic pyrenoids provides insights into the mechanisms and evolution of carbon concentration by EPYC1 proteins

Reagents and supplies

Except in any other case acknowledged, chemical compounds had been obtained from Sigma-Aldrich and Carl Roth within the highest purity and high quality. NaH¹⁴CO₃ and Okay¹⁴CN had been obtained from Hartmann Analytics and PerkinElmer. Reagents for cloning had been obtained from New England Biolabs, Macherey-Nagel, Qiagen and Thermo Fisher Scientific. For first exams, industrial D-RuBP sodium salt hydrate (>90% purity, R0878, Sigma-Aldrich) was used. For detailed characterizations, enzymatically synthetized RuBP following printed protocols⁵⁵ was employed. For the willpower of specificity constants, [1-³H] RuBP was synthetized enzymatically from D-[2-³H] glucose (PerkinElmer) following printed protocols⁵⁶. The Rubisco inhibitor CABP was beforehand synthetized in two batches utilizing both chilly or Okay¹⁴CN utilizing printed protocols⁵⁵. The focus of the chilly CABP batch was beforehand decided by inhibiting a Rubisco inventory containing a identified energetic web site focus (as quantified by [¹⁴C]-CABP) with various concentrations of chilly CABP and used with the acknowledged focus. Except in any other case acknowledged, all information had been analysed and processed utilizing Python 3.10 with the packages matplotlib 3.8.2 (ref. ⁵⁷), pandas 2.1.4 (ref. ⁵⁸), scipy 1.11.4 (ref. ⁵⁹), numpy 1.26.2 (ref. ⁶⁰), openCV 4.9.0 (ref. ⁶¹), scikits-odes 2.7.0 (refs. ^62,63,64), logomaker0.8 (ref. ⁶⁵) and statsmodels 0.14.1 (ref. ⁶⁶).

Phylogenetics and ancestral sequence reconstruction

To deduce the evolutionary relationships of EPYC1_CR and homologues, amino acid sequences had been gathered by on-line BLASTP on 6 September 2022 utilizing the amino acid sequence of EPYC1_CR from C. reinhardtii (XP_001690584.2) because the preliminary question. Sequences had been aligned utilizing MUSCLE 3.8.31 (ref. ⁶⁷) adopted by tree inference by IQ-TREE 2.2.0.3 (ref. ⁶⁸) and a best-fit mannequin of evolution as decided by Modelfinder 2.2.0.3 (ref. ⁶⁹), with gamma-distributed among-site price variation and stuck base frequencies. Two fashions of evolution had been recognized by IQ-TREE and examined: first, Mitochondrial Inverterbrate⁷⁰ when contemplating all amino acid evolution fashions, and second, Jones–Taylor–Thornton⁷¹ when proscribing to probably the most generally chosen amino acid fashions of LG⁷², Jones–Taylor–Thornton and WAG⁷³. Each approaches led to the identical inferred sequences with respect to the pure variation.

Ancestral sequences and posterior possibilities of ancestral states had been reconstructed at inside nodes as applied in IQ-TREE, utilizing the identical evolutionary mannequin as per tree inference. Hole project of ancestral sequences was decided utilizing Fitch parsimony with PastML (v.1.9.34)⁷⁴, yielding our closing tree. Ancestral sequences include states with the best posterior possibilities in any respect websites chosen.

The ensuing evolutionary tree was rooted utilizing timetree.org⁴² to the oldest node. Estimates of atmospheric fuel concentrations had been straight tailored from the web site.

Protein manufacturing and purification

Protein-coding DNA was codon-optimized and produced by Twist Biosciences or Azenta Life Sciences. Constructs had been cloned to the C terminus (Gibson meeting) or the N and C termini (Golden Gate cloning utilizing PaqCI) of the R. rubrum CbbM coding area in a His-bdSUMO plasmid acquired from Addgene⁷⁵ utilizing OneShot Top10 cells (Invitrogen). All plasmids and primers are provided in Supplementary Information 1 and 2. The constructs had been expressed in OneShot BL21-AI (Invitrogen) cells in TB media. An ON seed tradition was used to inoculate 6 l of manufacturing tradition. Cells had been harvested by centrifugation at 5,000g in a Beckman Coulter Avanti JXN-26 centrifuge outfitted with a JLA-8.100 rotor. The cells had been resuspended in 50 mM Tris, 10 mM MgCl₂, 500 mM NaCl, 2 mM 2-β-mercaptoethanol (2-BME) and 10 mM imidazole at pH 8.5. Cells had been damaged utilizing a microfluidizer at 18,000 psi with 4 cycles. Subsequently, the cell particles was sedimented utilizing ultracentrifugation at 100,000g (Sorvall WX+, Thermo Scientific, with T-647.5 fastened angle rotor 6 × 50 ml). The supernatant was then filtered utilizing 0.2-nm syringe cap filters. The filtrate was subsequently utilized to a 50- or 150-ml superloop (Cytiva) and from there on two stacked 5-ml HisTrap columns (Cytiva). The protein on the column was washed with 50 mM imidazole, 50 mM Tris, 500 mM NaCl and a pair of mM 2-BME, pH 8.5, till the UV 280-nm sign was steady. Subsequently, 5 ml of two mg ml⁻¹ bdSNEP1 in 50 mM imidazole, 50 mM Tris, 500 mM NaCl and a pair of mM 2-BME, pH 7.5, supplemented with 2 mM dithiothreitol (DTT) was utilized to the columns and incubated for 5 hours at 20 °C. The cleaved protein was then eluted utilizing 10 ml of 10 mM imidazole, 50 mM Tris, 500 mM NaCl and a pair of mM 2-BME, pH 8.5, and captured within the superloop. Subsequently, the protein was utilized to a 16/600 200-pg Superdex SEC-column in 50 mM Tris, 500 mM NaCl and 10 mM MgCl₂, pH 8.5, with 10 % v/v glycerol, isocratically eluted, and 2-ml fractions had been collected. The fractions had been analysed for protein content material and impurities utilizing SDS–PAGE after which concentrated to ~100 μM utilizing 10-kDa or 30-kDa MWCO concentrators (Amicon Extremely Filters, Merck). The protein was aliquoted into 20-μl volumes, frozen in liquid nitrogen and saved at −80 °C till additional use. For cryoEM construction elucidation inside condensates, the purest SEC fraction (with out glycerol) of CbbM–EPYC1_CRΔα was used. For crystallization, the CbbM fraction with the best purity from SEC utilizing 20 mM Tris and 100 mM NaCl, pH 8.5, was concentrated to five mg ml⁻¹. For each cryoEM and crystallography, the enzyme fractions had been activated utilizing 10 mM NaHCO₃ within the presence of a twofold molar extra of CABP for about 30 min.

RbcLS_CR was natively purified from C. reinhardtii tailored from Wunder et al.²¹. For this, the cells had been grown with 17.4 mM acetate-supplemented TAP media for a number of days till most optical density (3–3.5) was reached. Cells had been harvested by centrifugation at 5,000g in a Beckman Coulter Avanti JXN-26 centrifuge outfitted with a JLA-8.100 rotor. The cells had been resuspended in 100 mM Tris, pH 8.0, 100 mM NaCl, 0.5 mM EDTA, pH 8.0, 5 mM DTT and EDTA-free protease inhibitor tablets (Roche). Cells had been damaged utilizing a microfluidizer at 18,000 psi with 4 cycles. Subsequently, the cell particles was sedimented utilizing ultracentrifugation at 47,000g (Sorvall WX+, Thermo Scientific, with T-647.5 fastened angle rotor 6 × 50 ml). The soluble supernatant was utilized to a Q-HP anion trade column (Cytiva) mounted on an Äkta Pure methods (Cytiva) equilibrated in 50 mM NaCl, 20 mM Tris-HCl, pH 8.0, and a pair of.5 mM DTT and fractionated utilizing a linear gradient between 50 mM and 1 M NaCl. The right fractions had been decided by way of SDS–PAGE and subsequently desalted utilizing PD10 desalting drip columns (Cytiva) in 50 mM NaCl, 20 mM Tris, pH 8.0, and a pair of.5 mM DTT. The pattern was then utilized to a 10-ml self-packed ResourceQ15 (Cytiva) and fractionated with the identical gradient as earlier than. Rubisco-containing samples had been decided by way of SDS–PAGE, pooled and utilized to a Superdex 200 Improve 10/300 (Cytiva) equilibrated in 20 mM Tris (pH 8.0), 50 mM NaCl, 1 mM DTT and 5% v/v glycerol. Rubisco-containing fractions had been decided by way of SDS–PAGE, pooled, concentrated utilizing 10-kDa MWCO Amicon spin filters (Merck), flash-frozen in liquid nitrogen and saved at −70 °C.

Radiometric carboxylation assays

Time-curve assays

The assays had been tailored from Schulz et al.⁷⁶. Prior to every assay, protein inventory focus was measured by way of BCA assay (Thermo Scientific), and the condensate formation was assessed by way of mild microscopy (Supplementary Fig. 2) below response circumstances. Reactions had been performed with NaH¹⁴CO₃ or NH₄H¹⁴CO₃ shares with particular radioactive actions between 4 and eight mCi mmol⁻¹. The solubility of CO₂ in water (mol l⁻¹ atm⁻¹) was calculated utilizing the Henderson–Hasselbalch equation utilizing beforehand printed values⁷⁷. Time-curve assays had been carried out aerobically in 50–100-μl volumes in 1.5-ml crimp-neck amber vials (VWR) at 25 °C and 300 rpm utilizing a Thermomixer Professional (CellMedia). For every dataset, 4 time factors with three replicates every for each CbbM and CbbM–IDP in the identical shaker had been measured. To this finish, 1 μM CbbM/CbbM–IDP, 0.1 mg ml⁻¹ CA and 1 mM MgCl₂ was activated for five min utilizing 5 mM ¹⁴CO₃²⁻ shares in 50 mM Tris, pH 7.5–8.0 (relying on condensate stability). Subsequently, 0.3 mM RuBP was added to start out the response. After 2, 10, 30 and 90 min, the reactions had been quenched utilizing 10% closing v/v of fifty% v/v formic acid. The reactions had been left to evaporate on a heating block at 95 °C ON. Residual materials was resuspended in 500 μl ddH₂O, transferred to 5-ml scintillation vials, and combined with 4.5 ml scintillation cocktail (ROTISZINT HighCapacity, Carl Roth). The quantity of acid-stable radioactivity was quantified utilizing a Beckman LS 6000 scintillation counter. Preliminary charges had been computed utilizing the two min time level and normalized with the respective energetic web site content material. Lively web site content material was quantified by way of [¹⁴C]-CABP binding. To this finish, the identical response mixtures with chilly CO₃²⁻ and CABP as a substitute of RuBP had been incubated within the response buffer supplemented with 500 mM NaCl to stop condensate formation for 1 h. The mixtures had been centrifuged for 10 min at 10,000g. Subsequently, 35 μl supernatant was utilized to a SERAgel Professional 250 column (ISERA) on an Agilent 1260 HPLC system and isocratically eluted in the identical buffer. Fractions containing the protein had been collected (2 ml complete quantity), combined with 12 ml scintillation cocktail and measured on the scintillation counter. Lively web site content material was computed from the identified counts/focus of the [¹⁴C]-CABP shares. For complete carboxylation, the ratios of the maximally reached scintillation depend within the time curves of Rubisco and Rubisco fusion had been used. If vital, a number of datasets had been mixed by first computing the ratios of particular person datasets after which averaging the ratios to account for fluctuations in assays. Errors had been propagated to the ratios and subsequently to the ultimate ratio by way of empirical error propagation. For anaerobic preliminary price comparisons, assay buffers, vials and shares had been degassed with CO₂ free nitrogen for not less than 3 h.

³H specificity assays

The assays had been carried out precisely as reported in Schulz et al.⁷⁶. Briefly, CbbM and CbbM–IDPs had been incubated in 20-ml septum capped glass scintillation vials containing 1 ml 50 mM Tris, pH 7.5–8.0, 1 mM MgCl₂ and 0.1 mg ml⁻¹ carbonic anhydrase. Response mixtures had been equilibrated in outlined fuel mixtures containing 995,009 ppm O₂ and 4,991 ppm CO₂ (Air Liquide) earlier than the reactions had been initiated by the addition of 1 mM [1-3H]-RuBP. After 1 h, the merchandise had been dephosphorylated utilizing alkaline phosphatase (10 U per response) for one more 30 min and subsequently separated on an HPX-87H column (Bio-Rad). ³H-glycerate and ³H-glycolate peaks had been quantified utilizing liquid scintillation counting. The specificity issue S_C/O was calculated as described beforehand⁵⁶. Response buffers had been tailored to condensate-forming circumstances utilizing the identical buffers as for the time-curve assays.

LC–MS/MS specificity assays

The assays had been tailored from Schulz et al.⁷⁶. Buffers and vials had been degassed with CO₂ free nitrogen and equilibrated in 995,009 ppm O₂ and 4,991 ppm CO₂ (Air Liquide) previous to response begin and through the response. Reactions had been run in 0.5-ml volumes in 1.5-ml crimp-neck amber vials (VWR) outfitted with gas-tight conical rubber stoppers in a Thermomixer Professional (CellMedia) at 25 °C. For every response, 1 μM CbbM and CbbM–IDPs had been used. The reactions had been supplemented with 0.1 mg ml⁻¹ CA after which activated for five min earlier than the addition of 1 mM RuBP. The reactions had been run in quadruplicates for 30 min and quenched utilizing 50 μl formic acid. Subsequently, 35 μl centrifuged for two min at 3,000g and submitted for LC–MS/MS measurement.

LC–MS/MS technique

Quantitative willpower of RuBP, 2-PG and 3-PG was carried out utilizing an LC–MS/MS system. The chromatographic separation was carried out on an Agilent Infinity II 1290 HPLC system utilizing a SeQuant ZIC-pHILIC column (150 mm × 2.1 mm, 5-μm particle dimension, peek coated, Merck) linked to a guard column of comparable specificity (20 mm × 2.1 mm, 5-μm particle dimension, Phenomoenex) and a continuing circulation price of 0.1 ml min⁻¹ with cell phases consisting of 10 mM ammonium acetate in water, pH 9, supplemented with medronic acid to a closing focus of 5 μM (A) and 10 mM ammonium acetate in 90:10 acetonitrile to water, pH 9, supplemented with medronic acid to a closing focus of 5 μM (B) at 40 °C. The injection quantity was 0.5 μl. The cell part profile consisted of the next steps and linear gradients: 0–1 min fixed at 75% B, 1–6 min from 75% to 40% B, 6 to 9 min fixed at 40% B, 9–9.1 min from 40% to 75% B, 9.1 to twenty min fixed at 75% B. An Agilent 6470 mass spectrometer was utilized in damaging mode with an electrospray ionization supply and the next circumstances: ESI spray voltage, 2,000 V; nozzle voltage, 1,000 V; sheath fuel, 300 °C at 12 l min⁻¹; nebulizer stress, 20 psi; and drying fuel, 150 °C at 11 l min⁻¹. Compounds had been recognized on the premise of their mass transition and retention time in comparison with requirements. Chromatograms had been built-in utilizing MassHunter software program (Agilent). Absolute concentrations had been calculated by way of normal addition, spiking every pattern with a combined inventory answer of the goal analytes at completely different concentrations (0 μM, 10 μM, 25 μM, 50 μM, 100 μM and 250 μM) and analysing every pattern six instances. Mass transitions, collision energies, cell accelerator voltages and dwell instances had been optimized utilizing chemically pure requirements. The parameter settings of all targets are given in Supplementary Desk 5.

HPLC–SEC measurement for willpower of protein quantity in condensates

CbbM–IDP quantities within the steady part and in non-phase-separated circumstances had been measured utilizing a HPLC–SEC–UV/Vis system. First, the condensates had been fashioned in 50 mM Tris, pH 7.5, 0 mM NaCl and 1 mM MgCl₂ in triplicates at 1 µM in 200 µl complete quantity. To measure the overall quantity of protein within the system, the identical experiment was performed in non-condensate-forming circumstances in 50 mM Tris, pH 8.5, 500 mM NaCl and 10 mM MgCl₂. The samples had been incubated at room temperature for 30 min and subsequently sedimented at 2,000g for 30 min. The supernatant was then sampled utilizing 35 µl. The chromatographic separation was carried out on an Agilent Infinity II 1290 HPLC system utilizing an AdvanceBio analytical isocratic SEC column (300 mm × 4.6 mm, 2.7-μm particle dimension, 500-Å pore dimension, Agilent) at a continuing circulation price of 0.2 ml min⁻¹ in 50 mM Tris, pH 8.5, 500 mM NaCl and 10 mM MgCl₂ buffer for 35 min. The injection quantity was 2.5 μl. Chromatograms had been exported and built-in utilizing Python code based mostly on rampy 0.4.9 (ref. ⁷⁸), scipy 1.11.4 (ref. ⁷⁹) and numpy 1.26.2 (ref. ⁶⁰). The built-in peaks had been in contrast between the supernatants of phase-separated and homogeneous samples, and the relative quantity of complete protein contained in the condensates was calculated by way of (1-frac{{{rm{Space}}}_{{rm{condensates}}}}{{{rm{Space}}}_{{rm{homogeneous}}}}). Errors had been propagated utilizing empirical error propagation.

Crystallization and X-ray construction willpower

Crystallization was carried out by sitting-drop vapour diffusion at 16 °C. Purified inactive CbbM in 20 mM Tris buffer (pH 8.5) containing 100 mM NaCl was combined with a situation comprising 1.2 M ammonium chloride, 100 mM MES (pH 6) and 20% (w/v) PEG6000 in a 1:1 ratio with a closing drop quantity of 1 µl. The mom liquor was supplemented with 30% (v/v) PEG200 for about 1 min, earlier than the crystals had been looped and plunge-frozen in liquid nitrogen. Purified activated CbbM was crystallized in a buffer containing 20 mM Tris (pH 8.5), 100 mM NaCl, 0.2 mM CABP and 10 mM sodium bicarbonate combined with a situation containing 200 mM Bis-Tris propane (pH 7.3) and 20% (w/v) PEG4000 in a 1:1 ratio with a closing drop quantity of 1 µl. The mom liquor was supplemented with 33% (v/v) PEG200 for about 1 min, earlier than the crystals had been looped and plunge-frozen in liquid nitrogen.

X-ray diffraction information (Supplementary Desk 5) had been collected on the PETRA III beamline P13 of DESY (Deutsches Elektronen-Synchrotron, Hamburg). The info had been processed with XDS⁸⁰. Constructions had been solved by molecular substitute utilizing Phaser of the Phenix package deal 1.21 (ref. ⁸¹) utilizing an AlphaFold2 2.2.0 (ref. ⁸²) mannequin as a search template and refined with Phenix.Refine. Guide modelling, refinement and ligand becoming in addition to water selecting had been carried out in Coot 0.9.8.3 (ref. ⁸³). Ultimate positional and B-factor refinements had been carried out utilizing Phenix.Refine. Structural fashions for the inactive kind and activated CABP certain kind had been deposited to the Protein Information Financial institution in Europe below accession numbers 9I27 and 9I26, respectively (Supplementary Desk 6). The figures had been made utilizing PyMOL 2.5 (PyMOL Molecular Graphics System, Schrödinger (www.pymol.org)).

CryoEM pattern preparation, information acquisition and single-particle reconstruction

The CbbM–EPYC1_CRΔα inventory was activated utilizing 10 mM NH₄HCO₃ and a twofold extra of CABP for 30 min previous to pattern preparation. Subsequently, 2.7 μl 50 mM Tris and 0 mM NaCl, pH 7.5, was pipetted onto C-Flat R2/1 300 copper mesh grids that had been glow-discharged for 45 s instantly earlier than use, and 0.3 μl of the protein inventory was added to the drop, leading to a closing protein focus of ~1 μM.

The pattern was blotted for 3 s with blot power 20 at 100% relative humidity and 4 °C utilizing a Vitrobot Mark IV (Thermo Scientific). Grids had been plunge-frozen in liquid ethane cooled by liquid nitrogen, clipped into Autogrid cartridges and used for information assortment instantly.

CryoEM information had been acquired on a Titan Krios G3i electron microscope (Thermo Scientific), operated at an acceleration voltage of 300 kV and outfitted with a BioQuantum-K3 imaging filter (Gatan). Information had been measured in electron counting mode at a nominal magnification of ×130,000 (0.655 Å per pixel) with a complete dose of 60 e⁻ A⁻² (60 fractions), utilizing the aberration-free image-shift correction in EPU (Thermo Scientific). 5 photos had been acquired per foil gap, and the nominal defocus vary for information assortment was −1.2 to 2.4 μm. Foil holes had been chosen on the premise of capturing a sufficiently skinny portion of the condensates across the too-dense components.

Micrographs had been processed in cryoSPARC 4.4.1 (Structura Biotechnology Inc.). The primary 11,525 micrographs had been motion- and CTF-corrected, after which low-quality micrographs had been excluded, yielding 7,116 remaining micrographs. A complete of 45,233 particles had been manually picked because the small dimension of the structured area precluded dependable automated selecting in cryoSPARC. Utilizing denoised micrographs, additional template-based particle searches and several other classification steps, the ultimate construction reconstruction was completed with 209,691 high-quality particles. Detailed single-particle evaluation is proven in Supplementary Figs. 20 and 21.

CryoEM map becoming was initially carried out in UCSF-ChimeraX 1.7 (ref. ⁸⁴) utilizing an AlphaFold2 2.2.0 (ref. ⁸²) mannequin as a template. The ensuing mannequin was manually refined in Coot 0.9.8.3 (ref. ⁸³). Computerized refinement of the construction was carried out utilizing phenix.real_space_refine in Phenix (v.1.21)⁸¹. The ensuing mannequin and electron density map had been deposited within the Protein Information Financial institution in Europe and the Electron Microscopy Information Financial institution with the accession codes 9I28 and EMD-52579, respectively. Mannequin statistics are listed in Supplementary Desk 6.

Confocal microscopy for fluorescence detection of carboxylation, polarity estimation, pH estimation and carbonic anhydrase partitioning

The exercise within the condensates was qualitatively noticed utilizing a fluorescence coupled assay (Supplementary Fig. 8). The exact concentrations are displayed in Supplementary Desk 8. Pgmi was produced in-house, horseradish peroxidase was acquired from Sigma (P8375), enolase from Sigma (E6126), pyruvate kinase from Roche (10109045001), carbonic anhydrase from Sigma (C2624) and pyruvate oxidase from Sigma (P4591). First, all parts besides Rubisco and RuBP had been pipetted into 384 glass-bottom plates (Cellvis, 384-well glass-bottom plate P384-1.5H-N). The protein was then added, and the combination was checked for condensate formation. After Rubisco activation for a number of minutes, RuBP was added to look at the rise in fluorescence in a confocal epi-fluorescence microscope (Leica TCS SP8 X, Diode 405 nm laser/WLL mixture, transmission PMT and HyD detector). Pictures had been recorded at 1,024 × 1,024 pixels with a pixel dimension of 80.17 nm and a price of 0.644 photos per second for 25 minutes. Samples had been excited at 2% LP at 570 nm and recorded with a 585–600 nm detection window.

The polarity of condensates was estimated utilizing the assay reported in Küffner et al.⁵⁰. Condensates had been fashioned with 1 μM of CbbM–IDP fusion in 50 mM Tris, 0 mM NaCl and pH 7.5. Pictures had been recorded at 1,024 × 1,024 pixels with a pixel dimension of 41.01 nm and a price of 0.375 photos per second. Samples had been excited at 2% LP with 405 nm and detected in a window from 425 nm to 784 nm with a 5-nm bandwidth and 5-nm steps. Subsequently, the intensities had been extracted, and a gamma distribution was fitted to the information to estimate the maximum-intensity wavelength. Polarity was inferred utilizing an ordinary curve utilizing reported literature information (Supplementary Fig. 13)^85,86.

In analogy to the polarity estimation contained in the condensates, we tailored an intra-condensate pH estimation assay from Stoffel et al.⁸⁷. An ordinary curve from pH 6 to 10 utilizing 50 mM Bis-Tris (pH 6 and 6.5) and Tris buffers with 50 µM 5-(and-6)-Carboxy SNARF-1 (Thermo Scientific) in a Tecan Infinity platereader (Tecan) at excitation of 520 nm and emission from 530 nm to 750 nm was recorded (Supplementary Fig. 14). Condensates had been fashioned in pH 7.5, 50 mM Tris, 1 mM MgCl₂ and 50 µM 5-(and-6)-Carboxy SNARF-1. Spectra of the condensates had been recorded by way of lambda stacks within the confocal microscope. Pictures had been recorded at 2% LP at 405 nm and an emission window from 518 to 706 nm utilizing a 5-nm step dimension and a 5-nm detection window. Areas of curiosity of the condensates had been extracted, averaged and in contrast in opposition to the usual curve to estimate the inner pH.

Carbonic anhydrase partitioning was assessed utilizing Atto425-tagged carbonic anhydrase. To this finish, we incubated 2 mg ml⁻¹ carbonic anhydrase from bovine erythrocytes (Sigma-Aldrich) with a ~2-fold molar extra of Atto425-Malemide (ATTO-TEC) for 12 h on ice. Subsequently, we separated the tagged protein from unreacted Atto425 with a ten/30 Superdex 200 pg enhance (Cytiva) isocratically in 50 mM Tris, 500 mM NaCl and a pair of mM 2-BME, pH 8.5. The tagged protein was concentrated utilizing 500 μl Amicon 10 kDa spin filters. Condensates had been fashioned utilizing 1 μM fusion protein in 50 mM Tris and 0 mM NaCl, pH 8.0. After 5–10 min, a combination of 10% CA-ATTO425 and 90% CA with a complete closing focus of 0.1 mg ml⁻¹ was added to the condensates. After a few minutes, photos had been recorded at 1,024 × 1,024 pixels with a pixel dimension of 80.17 nm, a price of 0.241 photos per second and a body averaging of 8. Samples had been excited at 2% LP with 405 nm and recorded with a 465–505 nm detection window.

Mass photometry dimension evaluation

Microscope coverslips (1.5 H, 24 mm × 60 mm, Carl Roth) and CultureWell Reusable Gaskets (CW-50R-1.0, 50-3 mm diameter × 1 mm depth) had been cleaned thrice with MQ-H₂O and isopropanol adopted by drying below a stream of pressurized air. Gaskets had been dried additional at room temperature in a single day. Immersion oil (Immersion Oil 518F, Zeiss) was utilized to the mass photometer (MP) goal, and dry gaskets had been subsequently assembled on coverslips and positioned on the stage of a One^MP MP (Refeyn). Then, 18 μl of SEC buffer was utilized to the gasket wells. After focusing, 2 μl of pattern was added and quickly combined, and measurements had been began (<15 s after pattern addition). Information had been acquired for 60 s at 100 frames per second utilizing Purchase^MP 1.2.1 (Refeyn 1.2.1). Samples had been ready by diluting purified protein to 0.25 μM focus in 50 mM Tris, 500 mM NaCl and 10 mM MgCl₂, pH 8.5. MP distinction values had been calibrated to molecular lots utilizing industrial NativeMark Unstained Protein Customary (Thermo Fisher). The instrument was calibrated utilizing a 30-fold dilution of NativeMark, which was diluted an extra 10× on the instrument. Acquired photos had been processed utilizing DiscoverMP 1.2.3 (Refeyn).

Mass motion and mass switch mannequin simulations and information match

Differential equation methods had been applied in Python 3.13 and solved utilizing scikits.odes based mostly on the SUNDIALS suite^62,63,64. Information had been fitted utilizing most chance estimators and globally optimized utilizing a differential evolution algorithm from scipy.optimize^88,89. The differential equation system for Rubisco was tailored from earlier simulation work and is proven under⁴⁹:

the place ({c}_{i}^{{mathrm{in}}}) and ({c}_{i}^{{mathrm{out}}}) denote the concentrations of compound i inside and out of doors of the condensates; okay and Okay_i denote the mass switch coefficient and the partition coefficient of the respective compound; r(c_i) denotes the speed of the respective compound; E denotes the energetic web site focus; okay_C and okay_O denote the catalytic constants of carboxylation and oxygenation, respectively; and ({Okay}_{{mathrm{m}},{{mathrm{CO}}}_{2}}), ({Okay}_{{mathrm{m}},{{mathrm{O}}}_{2}}) and ({Okay}_{{mathrm{m}},{{mathrm{RuBP}}}}) denote the Michaelis fixed of CO₂, O₂ and RuBP, respectively. The diffusion coefficient (D) was assumed to be 10 μm² s⁻¹ on the premise of small-molecule diffusion throughout condensates⁴⁹, and the common condensate radius (R) was assumed to be 10 μm at a part quantity fraction Φ of 1% on the premise of estimations from earlier measurements⁴⁹. The typical radius was assumed to linearly scale with the part quantity fraction in a primary approximation.

Reporting abstract

Additional data on analysis design is out there within the Nature Portfolio Reporting Summary linked to this text.