Regulation of ribonucleotide reductase in response to iron deficiency

Abstract

Ribonucleotide reductase (RNR) is an essential enzyme required for DNA synthesis and repair. Although iron is necessary for class Ia RNR activity, little is known about the mechanisms that control RNR in response to iron deficiency. In this work, we demonstrate that yeast cells control RNR function during iron deficiency by redistributing the Rnr2-Rnr4 small subunit from the nucleus to the cytoplasm. Our data support a Mec1/Rad53-independent mechanism in which the iron-regulated Cth1/Cth2 mRNA-binding proteins specifically interact with the WTM1 mRNA in response to iron scarcity and promote its degradation. The resulting decrease in the nuclear-anchoring Wtm1 protein levels leads to the redistribution of the Rnr2-Rnr4 heterodimer to the cytoplasm, where it assembles as an active RNR complex and increases deoxyribonucleoside triphosphate levels. When iron is scarce, yeast selectively optimizes RNR function at the expense of other non-essential iron-dependent processes that are repressed, to allow DNA synthesis and repair.

Figure 1. Yeast dATP and dCTP levels increase upon nutritional or genetic iron deficiency

Wild-type BY4741 and fet3Δfet4Δ cells were grown for 7 hours in SC (Fe) or SC with 100 µM BPS (−Fe), and dATP and dCTP levels determined by the DNA polymerase-based enzymatic assay. The values of dATP and dCTP are represented as relative to those of the wild-type strain grown under Fe-sufficient conditions. The average and standard deviation of at least 3 independent experiments is represented.

Figure 2. Yeast Rnr2 and Rnr4 redistribute to the cytoplasm in response to nutritional or genetic Fe deficiencies

(A,B) Wild-type BY4741 cells were grown for 6 hours in SC (Fe) or SC with 100 µM BPS (−Fe). Then DNA was stained with 1 µg/mL DAPI for 3 min, and Rnr2 and Rnr4 localization determined by IMF with anti-Rnr2 (A) and anti-Rnr4 (B) antibodies, respectively. An overlap of DAPI and IMF signals is also shown (Merge). (C,D) Quantitative analysis of Rnr2 (C) and Rnr4 (D) subcellular localization patterns in wild-type cells grown as indicated in panels A and B. Percentages of cells with distinct localization patterns were represented as follows: black bars, cells with a predominantly nuclear IMF signal; grey bars, cells with both nuclear and cytoplasmic IMF signal; and white bars, cells with a predominantly cytoplasmic IMF signal. (E,F) Quantitation of Rnr2 (E) and Rnr4 (F) subcellular localization patterns in fet3Δfet4Δ cells grown for 6 hours in SC (Fe) or SC with 100 µM FAS (+Fe). Cells were processed and the results represented as indicated in panels C and D. In all cases, at least 200 cells were counted for each independent experiment performed by triplicate. The average and the standard deviation are represented. See also Supplemental Figure S1.

Figure 3. The Mec1 and Rad53 kinases do not participate in the redistribution of Rnr2 and Rnr4 to the cytoplasm in response to Fe deficiency

The phosphorylation stage of Rad53 (A) and Dun1 (B) proteins was determined. Wild-type BY4741 (A) and Dun1-MYC13 (B) cells were grown on SC without (Fe) or with 100 µM BPS (−Fe) for 6 hours, SC with 0.04 % MMS for 2 hours, or SC with 0.2 M HU for 2 hours. Proteins were extracted and analyzed by Western blotting with anti-Rad53 (A) and anti-c-Myc (B) antibodies. Pgk1 was used as a loading control. Wild-type Y300 (C,D), mec1Δsml1Δ (E,F), and rad53Δsml1Δ cells (G,H) were grown as described in panel A, and cells were processed and the data analyzed as described in Figure 2. See also Supplemental Figure S2.

Figure 4. Cells defective in CTH1 and CTH2 exhibit a defect in the redistribution of Rnr2 and Rnr4 to the cytoplasm in response to Fe deficiency

cth1Δcth2Δ cells co-transformed with pRS416-CTH1 and pRS415-CTH2 (CTH1 + CTH2), pRS416 and pRS415 (cth1Δ + cth2Δ), or pRS416-CTH1-C225R and pRS415-CTH2-C190R (CTH1-C225R + CTH2-C190R), were grown for 6 hours in SC (Fe) or SC with 100 µM BPS (−Fe), and then processed and the data analyzed as described in Figure 2. See also Supplemental Figure S3.

Figure 5. Cth1 and Cth2 promote the down-regulation of Wtm1 levels in response to Fe deficiency

cth1Δcth2Δ cells co-transformed with pRS416 and pRS415 (vector + vector), pRS416-CTH1 and pRS415-CTH2 (CTH1 + CTH2), or pRS416-CTH1-C225R and pRS415-CTH2-C190R (C225R + C190R), were grown for 8 hours in SC-ura-leu (+) or SC-ura-leu with 100 µM BPS (−). (A) Total RNA was extracted and analyzed by Northern blotting with a WTM1 specific probe. ACT1 was used as a loading control. (B) Proteins were extracted and analyzed by Western blotting with anti-c-Myc. Ponceau staining was used as a loading control.

Figure 6. Cth1 and Cth2 proteins specifically interact with WTM1 mRNA to promote Rnr2 and Rnr4 redistribution to the cytoplasm and sustain RNR function in response to Fe deficiency

(A) The Y3H assay was used to monitor in vivo interactions between Cth1 or Cth2 proteins and the ARE-containing fragment of the WTM1 3’-UTR mRNA. L40-coat cells were co-transformed with (1) pIIIA/MS2-1 containing the 3’-UTR of WTM1, WTM1-mt1, WTM1-mt2, WTM1-mt3, SDH4 (as a positive control), or vector alone (as a negative control); and (2) pACT2 vector alone or fused to CTH1, CTH1-C225R, CTH2 or CTH2-C190R. Cells were grown on SC-ura-leu (+His), and SC-ura-leu-his (−His) containing 250 or 750 µM 3-aminotriazol (3-AT) plates for 2–6 days at 30°C. (B) Yeast cells mutagenized in both WTM1 ARE patches exhibit a defect in the redistribution of Rnr2 and Rnr4 to the cytoplasm in response to Fe deficiency similar to CTH1- and CTH2-defective cells. WTM1-mt3 cells were grown for six hours in SC (Fe) or SC with 100 µM BPS (−Fe), and then processed and the data analyzed as described in Figure 2. (C) Determination of dATP and dCTP levels in cth1Δcth2Δ and WTM1-mt3 cells. Yeast cells were grown for seven hours in SC (Fe) or SC with 100 µM BPS (−Fe), and analyzed as described in Figure 1. The values of dATP and dCTP from the wild-type strain, previously shown in Figure 1, have been included for clarity. See also Supplemental Figure S4.

Figure 7. Cth1 and Cth2 control the levels of Rnr2 and Rnr4 subunits in response to iron deficiency

The levels of Rnr2 and Rnr4 mRNAs (A) and proteins (B) were determined in cth1Δcth2Δ cells co-transformed, grown, and processed as detailed in Figure 5. The levels of RNA and protein were quantified and normalized to SCR1 (RNA loading control) and Pgk1 (protein loading control), respectively. The Y3H assay was used to monitor in vivo interactions between Cth1 and Cth2 (C) proteins and the ARE-containing fragment of the RNR1 and RNR2 3’-UTR mRNAs. L40-coat cells were co-transformed with (1) pIIIA/MS2-1 containing the 3’-UTR of RNR2, RNR4, SDH4, and vector alone; and (2) pACT2 vector alone or fused to CTH1, CTH1-C225R, CTH2 or CTH2-C190R. Cells were grown as described in Figure 6. See also Supplemental Figure S5.