The nonsense-mediated mRNA decay (NMD) pathway differentially regulates COX17, COX19 and COX23 mRNAs

Abstract

The differential regulation of COX17, COX19 and COX23 mRNAs by the nonsense-mediated mRNA decay (NMD) pathway was investigated. The NMD pathway regulates mRNAs that aberrantly terminate translation. This includes mRNAs harboring premature translation termination codons and natural mRNAs. Most natural mRNAs regulated by NMD encode fully functional proteins involved in various cellular processes. However, the cause and targeting of most of these mRNAs by the pathway is not understood. Analysis of a set of mRNAs involved in copper homeostasis showed that a subset of these mRNAs function in mitochondrial copper homeostasis. Here, we examined the regulation of COX17, COX19 and COX23 mRNAs by NMD. These mRNAs encode homologous mitochondrial proteins involved in metallation of cytochrome c oxidase. We found that COX17, COX19 and COX23 mRNAs are differentially regulated by NMD depending on environmental copper levels. A long 3'-UTR contributes to the direct regulation of COX19 mRNA by the pathway. Alternatively, COX23 mRNA contains a long 3'-UTR, but is indirectly regulated by the pathway under two conditions tested here. Analysis of the functionality of the NMD targeting features in COX23 mRNA showed that the COX23 3'-UTR is sufficient to trigger NMD. The regulation of mRNAs involved in mitochondrial copper metabolism by NMD is physiologically significant because excess copper enhances growth of NMD mutants on a non-fermentable carbon source. These findings suggest that regulation of mRNAs encoding homologous proteins by NMD can be differential depending on environmental copper levels. Furthermore, these findings suggest copper ion homeostatic mechanisms in the mitochondria occur at the mRNA level via the NMD pathway.

Keywords: Mitochondrial copper homeostasis; Nonsense-mediated mRNA decay; Saccharomyces cerevisiae; mRNA; mRNA decay.

Fig. 1

COX17 and COX19 mRNAs are direct NMD targets in rich media while COX23 mRNA is an indirect target. Schematic representations of COX17, COX19 and COX23 mRNAs (a–c, respectively). Representative half-life northern blots were measured with total RNA extracted from wild-type strain AAY334 (UPF1 rpb1-1; Kebaara et al. 2003) and NMD mutant strain AAY335 (upf1Δ rpb1-1; Kebaara et al. 2003). Yeast cells were harvested over a 35-min period at different time points indicated above the northern blots. The northern blots were probed with radiolabeled DNA from the COX23, COX19 and COX17 ORFs, respectively. For controls, the membranes were probed with CYH2 and SCR1 DNA. The half-lives were determined using SigmaPlot graphs and are shown to the right of each northern blot. Typical SigmaPlot graphs showing the decay rate of COX17 (a), COX19 (b) and COX23 (c) mRNAs in UPF1 and upf1Δ yeast strains are shown to the right of the half-life northern blots. The half-life measurements are an average of at least three independent experiments and are calculated based on the time required for half of the original RNA to degrade

Fig. 2

The NMD pathway is functional under low copper conditions and the CTR1 mRNA is not regulated by NMD under these conditions. Representative steady-state mRNA accumulation levels were measured with RNA extracted from isogenic wild-type strain W303 (UPF1) and NMD mutant AAY320 (upf1Δ) yeast strains grown in complete minimal media containing bathocuproinedisulfonic acid (BCS). Half-life northern blots with total RNA extracted from wild-type strain AAY334 (UPF1 rpb1-1; Kebaara et al. 2003) and NMD mutant strain AAY335 (upf1Δ rpb1-1; Kebaara et al. 2003). Yeast cells were harvested over a 35-min time period at different time points indicated above the northern blots. The northern blots were sequentially probed with radiolabeled DNA from the CTR1 (a), CYH2 and SCR1 (b). CTR1 encodes a high-affinity copper transporter of the plasma membrane. Reduced copper levels result in increased CTR1 mRNA expression. Typical SigmaPlot graphs showing the decay rate of CTR1 (a) and CYH2 pre-mRNA in UPF1 and upf1Δ yeast strains are shown below the half-life northern blots. The half-life measurements are an average of at least three independent experiments

Fig. 3

COX23 mRNA is immune to NMD under low copper conditions; COX17 mRNA is indirectly regulated by the pathway, while COX19 mRNA is a direct NMD target under this conditions. Representative steady-state mRNA accumulation levels (a–c, top panels) were measured as described in Fig. 2. Half-life northern blots (a–c, bottom panels) with total RNA extracted from wild-type strain AAY334 (UPF1 rpb1-1; Kebaara et al. 2003) and NMD mutant strain AAY335 (upf1Δ rpb1-1; Kebaara et al. 2003) were measured as described in Fig. 2. The northern blots were probed with radiolabeled DNA from the COX23, COX19, and COX17, respectively. The half-lives were determined using SigmaPlot and are shown to the right of each northern blot. The half-life measurements are an average of at least three independent experiments

Fig. 4

The NMD pathway is functional under 100 µM copper and the CUP1 mRNA is not an NMD target under this conditions. Steady-state mRNA accumulation levels were measured with RNA extracted from isogenic wild-type strain W303 (UPF1) and NMD mutant AAY320 (upf1Δ) yeast strains grown in media containing 100 µM copper (top panels). Half-life northern blots with total RNA extracted from wild-type strain AAY334 (UPF1 rpb1-1; Kebaara et al. 2003) and NMD mutant strain AAY335 (upf1Δ rpb1-1; Kebaara et al. 2003) grown in media containing 100 µM copper (bottom panels). Yeast cells were harvested as described in Fig. 2 and probed with radiolabeled DNA from the CUP1 (a), CYH2 and SCR1 (b). CUP1 encodes a metallothionein that binds copper. The bottom band on the CYH2 mRNA/pre-mRNA steady-state and half-live northern is the CUP1 mRNA, which is highly overexpressed under high copper conditions. The CUP1 gene is induced by the Ace1 transcription factor when yeast cells are exposed to elevated copper levels. Increases in copper levels result in increased CUP1 expression. Half-lives were determined using SigmaPlot, by measuring the time it takes for half of the original mRNA levels to degrade. The half-life graphs are shown below each half-life northern blot. The half-life measurements are an average of at least three independent experiments

Fig. 5

COX19 mRNA is a direct NMD target under high copper conditions, while COX23 mRNA is an indirect target and COX17 mRNA is immune to degradation by the pathway. Representative steady-state mRNA accumulation levels (a–c, top panels) were measured with RNA as described in Fig. 4. Half-life northern blots (a–c, bottom panels) with total RNA extracted from wild-type strain AAY334 (UPF1 rpb1-1; Kebaara et al. 2003) and NMD mutant strain AAY335 (upf1Δ rpb1-1; Kebaara et al. 2003). Yeast cells were harvested as described in Fig. 4 and probed with radiolabeled DNA from the COX23, COX19, and COX17 ORFs, respectively. Half-lives were determined using SigmaPlot, by measuring the time it takes for half of the original mRNA levels to degrade. The half-life graphs are shown to the right of each half-life northern blot. The half-life measurements are an average of at least three independent experiments

Fig. 6

The COX23 3′-UTR is sufficient to target CYC1 mRNA to NMD. Schematic representation of CYC1COX23 3′-UTR mRNA (a) and representative northern blots of the CYC1COX23 3′-UTR steady-state mRNA accumulation levels (b) and half-lives (c). The northern blots were probed with DNA specific to the 5′-UTR and ORF of CYC1. All yeast cells for b and c were grown in synthetic complete media lacking leucine and were harvested as described in Fig. 1. Steady-state and half-life mRNA measurements are an average of at least three independent experiments. The half-lives were determined using SigmaPlot. CYH2 and SCR1 are controls

Fig. 7

NMD mutants’ respiratory impairments on non-fermentable carbon sources is recovered by elevated copper levels. UPF1 (W303a), upf1Δ (AAY320), upf2Δ (HFY1300) and upf3Δ (HFY861) yeast cells were grown to mid-log phase on complete minimal media. Tenfold serial dilutions of the cells were spotted onto complete minimal medium with either glucose (a) or lactate (b) as the carbon source. Additionally, the four yeast strains were spotted on media containing either 100 µM copper (middle panels) or 1 mM copper (right panels) and incubated 30 °C for 3 days

Fig. 8

Overexpression of COX19 enhances growth of wild-type yeast strains on media containing elevated copper levels. UPF1 (W303a) and upf1Δ (AAY327) yeast strain were transformed with pGT74/T3 (a), pG188/T1 (b) and pG84/T1 (c) plasmids and grown to mid-log phase on complete minimal media lacking uracil. Tenfold serial dilutions of the cells were spotted onto complete minimal medium lacking uracil (left panels), containing either 100 µM copper (middle panels) or 1 mM copper (right panels) and incubated 30 °C for 4 days