Skip to main page content
U.S. flag
See More

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2011 Apr;23(4):1273-92.
doi: 10.1105/tpc.111.084400. Epub 2011 Apr 15.

Systems biology approach in Chlamydomonas reveals connections between copper nutrition and multiple metabolic steps

Madeli Castruita  1 , David Casero, Steven J Karpowicz, Janette Kropat, Astrid Vieler, Scott I Hsieh, Weihong Yan, Shawn Cokus, Joseph A Loo, Christoph Benning, Matteo Pellegrini, Sabeeha S Merchant
Affiliations

Systems biology approach in Chlamydomonas reveals connections between copper nutrition and multiple metabolic steps

Madeli Castruita et al. Plant Cell. 2011 Apr.

Abstract

In this work, we query the Chlamydomonas reinhardtii copper regulon at a whole-genome level. Our RNA-Seq data simulation and analysis pipeline validated a 2-fold cutoff and 10 RPKM (reads per kilobase of mappable length per million mapped reads) (~1 mRNA per cell) to reveal 63 CRR1 targets plus another 86 copper-responsive genes. Proteomic and immunoblot analyses captured 25% of the corresponding proteins, whose abundance was also dependent on copper nutrition, validating transcriptional regulation as a major control mechanism for copper signaling in Chlamydomonas. The impact of copper deficiency on the expression of several O₂-dependent enzymes included steps in lipid modification pathways. Quantitative lipid profiles indicated increased polyunsaturation of fatty acids on thylakoid membrane digalactosyldiglycerides, indicating a global impact of copper deficiency on the photosynthetic apparatus. Discovery of a putative plastid copper chaperone and a membrane protease in the thylakoid suggest a mechanism for blocking copper utilization in the chloroplast. We also found an example of copper sparing in the N assimilation pathway: the replacement of copper amine oxidase by a flavin-dependent backup enzyme. Forty percent of the targets are previously uncharacterized proteins, indicating considerable potential for new discovery in the biology of copper.

PubMed Disclaimer

Figures

Figure 1.
Figure 1.
Analysis of DGE-TAG and RNA-Seq Data at the Chlamydomonas CTH1 Locus. This figure shows the UCSC browser view (http://genomes.mcdb.ucla.edu/CreCopper/) for the Chlamydomonas CTH1 locus. The browser view is centered at scaffold 11 at the base positions 114,500 to 1,150,500. (A) UCSC browser view of DGE-TAG coverage for the CTH1 locus. JGI gene models are shown in green. The position of NlaIII sites in the genome are indicated with black arrows, and the positions to which 17-bp DGE-TAGs map are indicated with a vertical line where the height of the line indicates the intensity of the signal (on a log scale) and the color indicates the condition analyzed (blue for copper deficient and red for copper replete). The DGE-TAG method occasionally has unique hits that map to the antisense strand instead of the sense strand. Darker versus lighter coloration is used to show the DNA strand to which the tags align; darker coloration is used for the strand that goes 5′ to 3′ (from left to right), whereas the lighter color refers to the complementary strand. For example, the CTH1 locus is oriented 3′ to 5′ from left to right, and the majority of the tags are therefore lighter colored since they map on the bottom strand. DGE-TAG tracks shown in (A) represent the average tag intensity of replicate cultures for wild-type copper-replete (+Cu) and copper-deficient (–Cu) cultures. (B) UCSC browser view of RNA-Seq coverage for the CTH1 locus. The track shown in black represents the normalized intensity, plotted on a log scale, of all RNA-Seq reads from 28 lanes mapped on to the genome. The tracks shown in gray and blue represent a single experiment: +Cu represents the RNA-Seq coverage for the locus from a wild-type copper-replete culture, while –Cu represents the coverage from a copper-deficient culture.
Figure 2.
Figure 2.
Copper Deficiency Targets under Photoheterotrophic and Photoautotrophic Conditions. (A) RNA-Seq identifies copper deficiency and CRR1 targets. RNA was extracted from batch cultures of Chlamydomonas cells. Genotypes, growth conditions, and comparisons A, B, C, and D are illustrated (top part). The crr1 and crr1:CRR1 cells were grown in the presence of acetate for comparison C. Note that the uncomplemented crr1 mutant grows very poorly in copper-deficient medium (pale green color). Cells were collected at a density of 3 × 106 cell mL−1. The Venn diagrams (bottom part) identify the number of differentially accumulating transcripts at ≥2-fold change (FDR < 0.05) in each pairwise comparison and the overlap of different sets. For comparison D, the number of differentially accumulating transcripts was computed at ≥5-fold change (FDR < 0.05). Transcripts with RPKM values <10 were not used in any of the analyses. The 63 putative CRR1 targets were used to retrieve sequences upstream of the transcription start site of those genes to identify CuREs (see text). (B) Copper deficiency targets group into three categories. Principal component analysis of RNA-Seq data group the experiments into three different categories according to growth condition or genetic background (first component) and two different groups according to copper metabolism (second component). The type of library (single-end v1, v2, and paired end [PE]) is shown to be unimportant.
Figure 3.
Figure 3.
Validation of mRNA Abundance Changes. (A) Comparison of fold difference in RNA abundance between copper-replete versus copper-deficient cells estimated from RNA-Seq versus real-time PCR experiments. Each data point represents the change in abundance of one mRNA (average of two independent experiments). For real-time PCR, each sample was analyzed separately in technical triplicates. For RNA-Seq, the same two samples were pooled prior to preparation of libraries. The correlation coefficient was calculated from the regression line derived with a 95% confidence interval. (B) The abundance of mRNAs of select CRR1 targets upon exposure of copper-deficient cells to copper. Top part: Copper-deficient cells, sampled at time 0, were exposed to 100 nM CuEDTA at a density of 3 × 106 cells mL−1 and sampled 30, 60, 120, 180, and 240 min later for preparation of total RNA as illustrated. Bottom part: The fold change in mRNA abundance at each time point was estimated by real-time PCR relative to the t = 0 RNA sample (100% expression) and is normalized to the endogenous reference gene CBLP, whose abundance does not change in this situation. The data presented are averages of three independent experiments.
Figure 4.
Figure 4.
Transcript Length Distribution after Manual Gene Model Correction. (A) UCSC browser view for a differentially expressed locus. Paired-end reads used to establish exon connectivity are shown as black rectangles joined by a line. The RNA-Seq coverage, on a log scale, from –Cu versus +Cu cells of one experiment is shown in blue or gray, respectively. The original JGI gene models, from FM3.1 of the Chlamydomonas genome, used to compute fold changes shown in Supplemental Data Set 3 online are shown in green. RNA-Seq data were used to inform the manual construction of a single gene model, shown in red. (B) Distribution of transcript lengths. Models for 149 genes whose transcripts responded to copper deficiency were manually investigated and improved using both single- and paired-end RNA-seq reads. The availability of millions of new ESTs from 454 sequencing (http://genomes.mcdb.ucla.edu/Cre454/) allowed further improvements. The distribution of transcript lengths is shown for the uncorrected gene models in the FM3.1 frozen catalog (black squares) versus the gene models after data-dependent correction (red circles). A polynomial best-fit line is shown for both data sets.
Figure 5.
Figure 5.
Predicted Protein Domain Function for Copper Deficiency and CRR1 Targets. The deduced protein sequences of the transcript set of 149 and 109 copper deficiency targets from Supplemental Data Set 3 online (acetate, photoheterotrophic; top left) and Supplemental Data Set 4 online (CO2, photoautotrophic; top right), respectively, were analyzed at the Pfam site (E-value ≤1 × 10−4). The identified protein domains were grouped according to their function into five categories: redox (red), transport (orange), protease (green), unknown (gold), and other (blue). Unknowns refer to protein sequences for which a domain could not be identified or if the domain was identified as having an unknown function (DUF). Domains that could not be classified as redox, transport, or protease were grouped into the “other” category. The distribution of protein domain functions in the subset of CRR1 targets (63 genes, bottom left) identified under photoheterotrophic conditions (see Figure 2A) is also shown. Chlamydomonas gene model set FM3.1 was used to estimate the genome-wide distribution of protein functions in the same categories. For the analysis of the complete genome (bottom right), the protein sequences with a predicted domain function with a bit score ≤10 were labeled unknown.
Figure 6.
Figure 6.
Abundance of mRNAs at Each Step in the Tetrapyrrole Biosynthesis Pathway. (A) The transcripts(s) encoding enzyme(s) at each step of the pathway are shown (Lohr et al., 2005; Tanaka and Tanaka, 2007). The heme branch is highlighted in beige, the chlorophyll branch in green, and the linear tetrapyrrole branch in blue. Transcripts whose abundance is impacted by copper nutrition are indicated in red (increased in –Cu) or blue (increased in +Cu). (B) Relative mRNA abundances, in RPKM, are shown for copper-replete (+Cu) and copper-deficient (–Cu) photoheterotrophic (acetate) and photoautotrophic (CO2) cultures sampled at 3 × 106 cells mL−1. Three light-independent isoform proteins encoded by the chloroplast genes chlB, chlL, and chlN are not listed because RNA-Seq reads were not mapped to the chloroplast genome.
Figure 7.
Figure 7.
Changes in Fatty Acid Composition and Ferrodoxin5 Abundance Parallel Changes in mRNA Abundance. (A) Total fatty acids profile of whole CRR1 cells. (B) The fatty acid composition of DGDG. For (A) and (B), lipids were isolated from +Cu and −Cu CRR1 cells. Lipids were separated by thin layer chromatography and quantified by GC. The bars show lipid composition in mol % and indicate the means (±sd) of six independent experiments. Fatty acid types are shown as number of carbons:number and positions (Δ) of double bonds. Values marked with an asterisk are significantly different from each other (P < 0.05, nonpaired two-sample t test). (C) For the immunodetection of ferredoxin5, proteins were separated on a nondenaturing polyacrylamide gel (15%) and transferred to PVDF for immunoblot analysis with anti-Fdx5. Protein loading was normalized by Coomassie blue staining.
Figure 8.
Figure 8.
Promoter Analysis of CRR1 Targets. Motif-enrichment analysis of proximal promoters (~250 bp upstream of the transcription start site) of CRR1 targets shows overrepresentation of GTAC-core motifs (see Supplemental Methods online for details). The logo displays the sequence conservation 10 bp upstream and downstream of GATC sites. [See online article for color version of this figure.]
Figure 9.
Figure 9.
Copper Status Determines the Fate of Plastocyanin. (A) Sequence alignment of the PCC1 proteins from Chlamydomonas and V. carteri to the N-terminal domain of Arabidopsis PAA2 (GenBank AY297817). The putative copper binding motif (MxCxxC) is highlighted in red. Residues identical in all sequences are indicated by a black background, and conserved residues are indicated by a gray background. (B) Our proposed pathway for changing the steady state abundance of plastocyanin by changing the expression of the hypothetical copper delivery protein PCC1 versus the putative protease RSEP1 as a function of copper nutrition (+Cu versus –Cu). Under copper-replete conditions (top), the expression of RSEP1 is repressed resulting in the accumulation of plastocyanin. Under copper-deficient conditions (bottom), the expression of PCC1 is repressed and RSEP1 actively degrades plastocyanin, thus liberating copper (blue spheres).
See this image and copyright information in PMC

References

    1. Abdel-Ghany S.E., Müller-Moulé P., Niyogi K.K., Pilon M., Shikanai T. (2005). Two P-type ATPases are required for copper delivery in Arabidopsis thaliana chloroplasts. Plant Cell 17: 1233–1251 - PMC - PubMed
    1. Abdel-Ghany S.E., Pilon M. (2008). MicroRNA-mediated systemic down-regulation of copper protein expression in response to low copper availability in Arabidopsis. J. Biol. Chem. 283: 15932–15945 - PMC - PubMed
    1. Allen M.D., del Campo J.A., Kropat J., Merchant S.S. (2007a). FEA1, FEA2, and FRE1, encoding two homologous secreted proteins and a candidate ferrireductase, are expressed coordinately with FOX1 and FTR1 in iron-deficient Chlamydomonas reinhardtii. Eukaryot. Cell 6: 1841–1852 - PMC - PubMed
    1. Allen M.D., Kropat J., Merchant S.S. (2008). Regulation and localization of isoforms of the aerobic oxidative cyclase in Chlamydomonas reinhardtii. Photochem. Photobiol. 84: 1336–1342 - PubMed
    1. Allen M.D., Kropat J., Tottey S., Del Campo J.A., Merchant S.S. (2007b). Manganese deficiency in Chlamydomonas results in loss of photosystem II and MnSOD function, sensitivity to peroxides, and secondary phosphorus and iron deficiency. Plant Physiol. 143: 263–277 - PMC - PubMed

Publication types

MeSH terms