Comparative differential cuproproteomes of Rhodobacter capsulatus reveal novel copper homeostasis related proteins

Abstract

Copper (Cu) is an essential, but toxic, micronutrient for living organisms and cells have developed sophisticated response mechanisms towards both the lack and the excess of Cu in their environments. In this study, we achieved a global view of Cu-responsive changes in the prokaryotic model organism Rhodobacter capsulatus using label-free quantitative differential proteomics. Semi-aerobically grown cells under heterotrophic conditions in minimal medium (∼0.3 μM Cu) were compared with cells supplemented with either 5 μM Cu or with 5 mM of the Cu-chelator bathocuproine sulfonate. Mass spectrometry based bottom-up proteomics of unfractionated cell lysates identified 2430 of the 3632 putative proteins encoded by the genome, producing a robust proteome dataset for R. capsulatus. Use of biological and technical replicates for each growth condition yielded high reproducibility and reliable quantification for 1926 of the identified proteins. Comparison of cells grown under Cu-excess or Cu-depleted conditions to those grown under minimal Cu-sufficient conditions revealed that 75 proteins exhibited statistically significant (p < 0.05) abundance changes, ranging from 2- to 300-fold. A subset of the highly Cu-responsive proteins was orthogonally probed using molecular genetics, validating that several of them were indeed involved in cellular Cu homeostasis.

Graphical Abstract

Cuproproteome of model bacterium Rhodobacter capsulatus reveals 75 Cu-responsive proteins that are strongly influenced (2–300 fold) by Cu availability.

Fig. 1

Rhodobacter capsulatus proteome. (A) Venn diagram showing experimentally identified proteins among those encoded by the R. capsulatus genome, and among the identified proteins, those that are quantified as described in the Experimental procedures. The total number of uncharacterized proteins and their distributions among the identified and the quantified proteins subcategories are indicated. (B) Distribution of R. capsulatus genome proteins among the various cellular locations, with the experimentally identified proteins indicated by the shaded areas. Percentages refer to those identified by MS in this study. Localization information was obtained from PsortB (version 3.0.2; ; www.psort.org). (C) TIGR main role categories for the identified proteins with the assigned TIGRFAM annotations (; www.tigrfams.jcvi.org release 15.0, 2014), extracted from UniprotKB. The top main roles are shown, with the number of proteins indicated for each category. Only 881 proteins, covering 36% of the identified proteins could be categorized using the TIGRFAM annotations.

Fig. 2

LFQ rank profile of quantified proteins for Cu-sufficient (control) samples. The log₂(LFQ) intensity (mean of biological and technical repeats) was plotted against the protein rank based on the LFQ values, showing the distribution of the 1926 quantified proteins in the Cu-sufficient (control) samples. Note that log₂(LFQ) spans from ∼−5.3 to +7.7, corresponding to about 8000-fold variation. The insets on the upper right and lower left corners show the identities of the 20 highest- and 20 lowest-LFQ proteins. The top 20 proteins: Rcc00147, elongation factor Tu tuf1; Rcc02533, light-harvesting protein B-800/850, gamma chain pucDE; Rcc02478, 60 kDa chaperonin groL; Rcc02971, ATP synthase subunit beta atpD; Rcc02973, ATP synthase subunit alpha atpA; Rcc00659, photosynthetic reaction center, H subunit puhA; Rcc00911, propionyl-CoA carboxylase, alpha subunit pccA; Rcc02160, glyceraldehyde-3-phosphate dehydrogenase gap3; Rcc00718, malate dehydrogenase mdh; Rcc01244, polyamine ABC transporter potD1; Rcc03024, TRAP dicarboxylate transporter dctP3; Rcc01369, ABC transporter, P substrate-binding protein; Rcc02008, trigger factor tig; Rcc00706: oligopeptide ABC transporter oppA1; Rcc01125:30S ribosomal protein S1 rpsA; Rcc00296, elongation factor G fusA1; Rcc02959, polyamine ABC transporter potD5; Rcc00290. 50S ribosomal protein L7/L12 rplL; Rcc00906, propionyl-CoA carboxylase pccB; Rcc01887, isocitrate dehydrogenase [NADP] icd. The bottom 20 proteins: Rcc00039, nucleotidyltransferase family protein; Rcc02217, glycosyl transferase, group 1; Rcc02433, uncharacterized protein; Rcc01906, uncharacterized protein; Rcc01278, conserved domain protein; Rcp00105, transcriptional regulator, XRE family; Rcc02968, uncharacterized protein; Rcc03380, SsrA-binding protein (small protein B; smpB); Rcc01028, iron siderophore/cobalamin ABC transporter; Rcc00886, ABC transporter, ATP-binding protein; Rcp00074, nitrous oxide maturation protein nosD; Rcc02033, Precorrin 3B synthase cobZ; Rcc02488, uncharacterized protein; Rcc00889, NosL family protein; Rcc03532, formamidopyrimidine-DNA glycosylase mutM; Rcc02140, extracellular ligand-binding receptor; Rcc00890, uncharacterized protein; Rcc02781, FAD dependent oxidoreductase; Rcp00068, peptidase, M4 family; Rcc02110, multicopper oxidase family protein.

Fig. 3

Proteomic data quality for R. capsulatus cells grown under various Cu concentrations. (A) Color-coded Pearson correlations and multi-scatter LFQ plots depicting the three biological replicates (1–3) and the three Cu (+BCS, +Cu, Ctrl (Cu-sufficient)) concentrations used. Only the reliably quantified 1926 proteins are included, with each biological replicate representing mean LFQ values of the technical repeats. The correlations observed are uniformly high (>0.959) throughout all conditions, and are highest (0.978 to 0.994) among the biological replicates for a fixed Cu level, and lowest (0.959 to 0.974) between the two extreme Cu levels, +BCS and +Cu. (B) Principle component analysis (PCA) of the LFQ based protein expression values. Note that the biological replicates (1–3) group together within each Cu growth condition, and the largest separation corresponds to the two extreme Cu concentrations, +Cu and +BCS. Technical repeats were averaged, and 1797 of the 1926 proteins with non-zero LFQ values were used for the PCA analysis. Both the Pearson correlation and PCA analyses were performed prior to missing data imputation (Perseus, ; www.coxdocs.org).

Fig. 4

Volcano plots showing quantitative enrichment or depletion of R. capsulatus proteins by Cu availability. The t-test −log p value versus log₂ fold change for Cu supplementation (+Cu/Ctrl) (A) and Cu depletion (+BCS/Ctrl) (B) as compared to Cu-sufficient control (Ctrl) conditions. The vertical and horizontal dashed lines indicate the 2 fold-changes and the p value of 0.05, respectively. The differentially affected proteins that satisfy both of these significance cutoff criteria (p value <0.05 and fold-change >2) are indicated in red and blue circles for +Cu and +BCS, respectively. See Table 2 for descriptions of the proteins indicated by the Rcc numbers.

Fig. 5

Hierarchical clustering of significantly changed proteins. A heat map of z-scored log₂ LFQ intensities of the 75 Cu-responsive proteins with differential abundance changes in response to different Cu amounts was generated using unsupervised Euclidean hierarchical clustering (Perseus, ; www.coxdocs.org). The samples are identified by Cu condition (Ctrl, +Cu and +BCS), biological repeat (B1, B2, B3) and technical repeat (R1, R2, R3, R4). The proteins segregated into three main clusters, indicated as 1, 2 and 3 on the dendrogram (left), separated by dotted lines and shown by the differentiated blue-yellow color groups in the heat map.

Fig. 6

LFQ abundance of overproduced and underproduced proteins relative to the control (Cu-sufficient) in cells exposed to different concentrations of Cu. The mean LFQ values of the 75 differentially regulated proteins, under +Cu or +BCS conditions, were compared to their counterparts in Cu-sufficient control cells in the protein LFQ rank plot of Fig. 2. Black dots represent log₂ LFQ in Ctrl (Cu-sufficient) conditions; red and blue circles represent the LFQ for selected proteins in Cu-excess (+Cu), and Cu-depleted (+BCS) conditions, respectively. Proteins that are highly overproduced or underproduced (p-value <0.01, fold-change >4) are labeled. Note that the LFQ value of a given protein does not necessarily indicate its cellular amount, and only used here for differential comparison purposes.

Fig. 7

Fold-changes of strongly Cu-responsive R. capsulatus proteins. The fold-changes of the 18 most highly Cu-responsive (p-value <0.01 and fold-change >4) proteins that are overproduced or underproduced in cells grown under (A) Cu-excess (+Cu/Ctrl), and (B) Cu-depletion (+BCS/Ctrl) conditions. Cellular localizations of these proteins are color-coded as indicated.

Fig. 8

Cellular distribution of Cu-related proteins in the presence or absence of Cu in the growth medium. Proteins of unchanged abundance are shown in gray, and those that are predicted but not identified in this study are in dotted gray. Proteins that exhibited differential abundance changes in +Cu and +BCS growth conditions are colored using different shades of red-yellow and blue, respectively, based on the exhibited fold changes with respect to Cu-sufficient (control) cells. For each protein, up or down arrows refer to its overproduction or underproduction, respectively, in response to +Cu or +BCS. The R. capsulatus proteins that are known, or predicted (based on homology) to be involved in Cu homeostasis are indicated as circles, those that have attributed functions as ellipsoids, and those that are uncharacterized (including family members) as rectangles. Proteins that are functionally related to each other, such as enzyme subunits or assembly factors, are grouped near to each other. Of the Cu-responsive proteins listed in Table 2, those that are related to chemotaxis or motility are not included in the figure.