Extraction of RNA and synthesis of first-strand cDNA

The total RNA was extracted from the frozen cells using the Trizol reagent (Invitrogen, USA) according to the protocol provided by the manufacturer. Deoxyribonuclease I (Invitrogen, USA) was used to remove trace of genomic DNA. The first-strand cDNA was synthesized using M-MLV Reverse Transcriptase following the recommended protocols (Promega, USA). The RNA was subsequently digested using RNase H (MBI Fermentas, Canada).

Microarray construction and hybridization

The yeast genome 70-mer oligonucleotides microarrays were obtained from CapitalBio, and the hybridization was done by the CapitalBio Company (China) (Zhang et al., 2002). The microarrays were monitored on a GenePix scanner (Axon Instrument, USA). For each test and control sample, two hybridizations were performed by using a reverse fluorescence strategy, and 1.5-fold averaged over the two biological replicates and q-value ≤ 1% was set for differentially expressed genes.

Trascriptional analysis of MSN4/MSN2 under different concentration of iron-surplus condition

S. cerevisiae S288C was inoculated in 150 ml YPD medium at 30°C. The yeast at an OD_{600 nm} of 0.1 was transferred into six flasks, with FeCl₃ added to a final concentration of 0, 0.5, 1, 2, 5 mM respectively. The yeast cells were collected after being cultured for 4 h. The total RNA was extracted as described above. The amount of RNA was quantified by measuring the absorbance at 260 nm. Semi-quantitative reverse transcription-polymerase chain reaction (RT-PCR) was performed as described previously (Frohloff et al., 2001). MSN4 primers used are (forward: 5′-CTCATAACAACAACAATGGTAAGGTTC-3′; reverse: 5′-GATGTTGTGATAAATTGTCACTTCTAC-3′); MSN2 primers used are (forward: 5′-GAAGGAAAGAAGGCCAAGTTACAG-3′; reverse: 5′-GTCTCCATGTTTTTTATGAGTCTTG-3′) and house-keeping gene β-ACTIN primers used are (forward: 5′-AAACCGCTGCTCAATCTTC-3′; reverse: 5′-CATTCTTTCGGCAATACCTG-3′). And in the quantitative RT-PCR (qRT-PCR), immunoQuantArray 96-well plates were prepared by the addition of 0.4 μl of 10 μM forward & reverse primers per well. To each well was then added 10 μl of SYBR Premix Ex Taq, 2 μl of cDNA and 6.8 μl of ddH2O. QRT-PCR data were collected on the ABI Prism 7,000 Sequence Detection System software using the default reaction conditions (1 cycle at 95°C for 5 min, 40 cycles at 95°C for 15 s and at 60ºC for 1 min). The values of threshold cycle (Ct) were exported for Global Pattern Recognition Algorithm (GPR) analysis (Akilesh et al., 2003; Livak et al., 2001), that is, ΔCt_gene = Ct_gene − Ct_normalizer, ΔΔCt = ΔCt_gene − ΔCt_control, amount of target gene = 2^−ΔΔCt.

Transcriptional profiles of yeast in response to surplus iron

To compare the profiles of the yeast transcriptomic response to surplus iron, early logarithmic yeast S288C grown in liquid YPD medium were exposed to 5 mM Fe³⁺ for 1 or 4 h. The number of genes that were expressed differentially compared to the control at indicated time points (1 h and 4 h) is in Fig. 2A. Among the 5935 genes covered by the DNA microarray, the expression levels of 48 genes changed more than 2-fold and 135 genes showed a changed between 1.5- and 2-fold at 1 h, whereas 346 genes changed more than 2-fold and 487 genes showed a change between 1.5- and 2-fold at 4 h. The expression profiles indicated that the yeast response to surplus iron is multiphasic. Only 183 stress-related genes responded promptly by 1 h. At 4 h, 833 genes involved in metabolic pathways were expressed. Venn diagrams revealed that two genes were upregulated more than 2-fold, compared to 25 genes that were commonly downregulated by more than 2-fold both at 1 and 4 h (Fig. 2B). These downregulated genes were mainly related to iron transport and homeostasis, and consisted of 17 open reading frames (ORFs) at 1 h and 26 ORFs at 4 h (Supplementary Tables 1 and 2). For instance, the genes FIT1, FIT2, and FIT3 encode cell wall-anchored mannoproteins that can retain siderophore-iron in the cell wall. The high-affinity iron transporter (FET3 and FTR1) and low-affinity iron transporter (FET4) are involved in iron uptake through the plasma membrane. Several downregulated genes were also associated with iron metabolism, such as the iron-sulfur cluster synthesizing gene ISU2 and the ferric reductases encoded by genes FRE6/7. All these genes indicated that surplus iron downregulated the transcription of iron-related genes, thus maintaining the proper intracellular iron concentration.

Open in a separate window

Fig. 2.

(A) Statistical comparisons of the transcriptomic responses upon 5 mM Fe³⁺. Black bars indicate genes that strongly responded (≥ 2.0-fold), whereas gray bars indicate genes that weakly responded (1.5~2.0-fold). (B) Venn diagrams of up-regulated genes and down-regulated genes of S. cerevisiae upon surplus iron in 1 h versus 4 h. The genes are shown here whose transcriptional level changed more than 2.0-fold.

When yeast cells were exposed to surplus iron for a considerable time such as 4 h, genes involved in the tricarboxylic acid cycle, electron transport, and oxidative phosphorylation were all upregulated, compared to the downregulated genes of anaerobic respiration. These data showed that surplus iron drove yeast cells to perform aerobic respiration and thus facilitated faster growth, concomitant with the upregulation of genes inamino acid/protein biosynthesis, assembly and modification, sugar/lipid biosynthesis, and cell wall formation (Supplementary Table 2). Furthermore, several stress-responsive genes were also upregulated in surplus iron at 4 h, such as HSP12, HOR7, LSP1, SOD2, GRX4, CCC1, GSY1, and HSP31. In particular, HSP12, HOR7, SOD2, CCC1 and HSP31 contain a stress response element (STRE) consensus sequence in their promoter regions, and all are target genes of the stress-resistant transcription factors Msn2/4 (Gauci et al., 2009; Martinez-Pastor et al., 1996). For instance, Hsp12 is a plasma membrane protein induced by stress conditions that probably initially perceives extracellular surplus iron. The vacuolar iron transporter Ccc1 that is induced by Msn2/4 facilitates the storage of surplus intracellular iron in the vacuole to alleviate high-iron toxicity. In addition to Msn2/4-dependent stress-responsive genes, many upregulated genes in other metabolic pathways are also targets of Msn2/4 (Table 1) (Gauci et al., 2009). Therefore, we further checked changes in the transcription of MSN2/4, with the result that MSN4 increased 1.5-fold and 1.75-fold under 5 mM of Fe³⁺ at 1 and 4 h, respectively, compared to the stable transcription of MSN2 in iron-surplus conditions.

To further confirm this finding, RT-PCR assays were performed (Figs. 3A and ​and3B).3B). The 2^−ΔΔCt was calculated to indicate relative changes in gene expression. The results showed that transcription of MSN4 was significantly upregulated upon increased iron concentration in the medium, up by approximately 2-fold at 5 mM of Fe³⁺, using YPD medium as the background control.

Open in a separate window

Fig. 3.

The relative amounts of mRNA of MSN2/MSN4 determined by RT-PCR (A) and qRT-PCR (B) upon different concentration of surplus iron. The transcription of MSN4 was increased dependent on iron concentration, while MSN2 was not. β-ACTIN (house-keeping gene) was as internal control. The results represent three independent assays with error bars.

This work was supported by the 973 project from the Ministry of Science and Technology of China (Project 2012CB911002) and the Chinese National Natural Science Foundation (Grant 30870490). We also thank Prof. Shingo Izawa (Kyoto University) for providing the plasmid pAdh1-Msn4-GFP.