RTG-dependent mitochondria-to-nucleus signaling is regulated by MKS1 and is linked to formation of yeast prion [URE3]

Abstract

An important function of the RTG signaling pathway is maintenance of intracellular glutamate supplies in yeast cells with dysfunctional mitochondria. Herein, we report that MKS1 is a negative regulator of the RTG pathway, acting between Rtg2p, a proximal sensor of mitochondrial function, and the bHLH transcription factors Rtg1p and Rtg3p. In mks1 Delta cells, RTG target gene expression is constitutive, bypassing the requirement for Rtg2p, and is no longer repressible by glutamate. We show further that Mks1p is a phosphoprotein whose phosphorylation pattern parallels that of Rtg3p in response to activation of the RTG pathway, and that Mks1p is in a complex with Rtg2p. MKS1 was previously implicated in the formation of [URE3], an inactive prion form of a negative regulator of the nitrogen catabolite repression pathway, Ure2p. rtg Delta mutations induce [URE3] and can do so independently of MKS1. We find that glutamate suppresses [URE3] formation, suggesting that the Mks1p effect on the formation of [URE3] can occur indirectly via regulation of the RTG pathway.

Figure 1

Inactivation of MKS1 reverses the block in CIT2 expression in rtg2Δ cells. β-Galactosidase activities were determined in triplicate in whole-cell extracts of the indicated strains grown to midlog phase in YNBR + cas medium. Each strain contained a CIT2-lacZ reporter gene integrated at the CIT2 locus.

Figure 2

Effects of inactivation of MKS1 in wild-type and rtg2Δ cells on Rtg1p and Rtg3p. (A) Subcellular localization of C-terminal GFP derivatives of Rtg1p and Rtg3p in ρ⁺ wild-type (WT) and mutant derivatives of strain PSY142. Constructs encoding the GFP derivatives were transplaced into the respective RTG1 and RTG3 loci. (B) Rtg3p is dephosphorylated in mks1Δ cells grown in YNBD + cas medium. Rtg3p was detected by Western blotting with a polyclonal anti-Rtg3p antibody raised against recombinant Rtg3p (Sekito et al., 2000). Lane 2 shows that the mobility of Rtg3p collapses to a faster migrating species when an extract of wild-type cells was pretreated with 5 U of calf intestinal phosphatase (CIP).

Figure 3

Inactivation of MKS1 alters glutamate regulation of the RTG pathway. (A) Glutamate auxotrophy of rtg2Δ but not of rtg3Δ cells is reversed by the mks1Δ mutation. Derivatives of PSY142 cells were plated on YNBD medium without or with 0.01% glutamate. (B) CIT2-lacZ reporter gene expression is not significantly repressed by glutamate in mks1Δ cells. β-Galactosidase assays were carried out on whole-cell extracts of the indicated strains derived from PSY142 grown in YNBD medium with or without supplementation with 0.2% glutamate. (C) Glutamate-dependent phosphorylation of Rtg3p is reversed by the mks1Δ mutation. The increase in phosphorylation of Rtg3p is observed when wild-type PSY142 ρ⁺ cells are grown in YNBD medium containing 0.2% glutamate but not in mks1Δ or mks1Δ rtg2Δ derivatives.

Figure 4

Mks1p is a phosphoprotein. (A) Mks1(HA)₃ was expressed in ρ⁺ mks1Δ cells grown in YNBD medium containing 0.2% glutamate. Mks1p(HA)₃ was detected by Western blotting with a monoclonal anti-HA antibody either with or without pretreatment with λ phosphatase (λ PPase). (B) Western blot analysis of Mks1p(HA)₃ in extracts of ρ⁺ or ρ^ο cells grown in YNBD medium with or without the addition of 1% casamino acids or 0.2% glutamate as indicated. All strains were derived from PSY142. The asterisk indicates the position of a more phosphorylated species.

Figure 5

Effects of rapamycin on Rtg3p and Mks1p. (A) Effects of rapamycin and glutamate on the subcellular localization of Rtg3p-GFP and Mks1p-GFP. PSY142 ρ⁺ cells expressing Rtg3p-GFP or Mks1p-GFP from constructs transplaced into the respective chromosomal RTG3 or MKS1 loci treated for 30 min with 200 ng/ml rapamycin (+) or vehicle alone (−) were grown in YNBD with or without the addition of 0.2% glutamate. (B) Rapamycin induces dephosphorylation of Rtg3p and Mks1p. Extracts from PSY142 ρ⁺ cells grown in YNBD supplemented with 0.01% glutamate and treated with rapamycin or vehicle as described in A were analyzed by Western blotting with Rtg3p polyclonal anti-Rtg3p antiserum or monoclonal anti-myc antibody. Lanes 3 and 4 are extracts prepared from rtg3Δ cells (top) or from cells not containing the construct encoding Mks1(myc)₃ (bottom).

Figure 6

Coimmunoprecipitation of Mks1p and Rtg2p. (A) Northern blot analysis of MKS1 transcripts were determined on total RNA prepared from PSY142 cells (B) Coimmunoprecipitation of (myc)₃- and (HA)₃-tagged derivatives of Mks1p and Rtg2p, respectively.

Figure 7

Effects of rtg mutations and glutamate on [URE3] formation. Two independent colonies of an MLY42 derivative (MATα ura2) were analyzed in parallel as indicated by the unshaded and shaded bars. (A) Effects of inactivation of RTG and MKS1 on [URE3] formation. Wild-type (derived from MLY42, a Σ1278b background strain) and isogenic rtg2Δ, rtg3Δ, mks1Δ, rtg2Δ mks1Δ, and mks1Δ rtg3Δ mutant derivatives (all contain an ura2Δ::kanMX4 mutation) were grown in YPD medium for 2 d to saturation. Cells were pelleted and washed twice with sterile water. For each strain, 10⁶ or 10⁷ cells were plated on each of seven YNBD plates supplemented with 200 μg/ml ureidosuccinic acid and 0.01% glutamate. The number of USA⁺ colonies per plate was determined after 5 d of incubation at 30°C. Two independent colonies of each strain were analyzed as indicated by the shaded and unshaded bars. Error bars indicate SD of colony numbers from seven plates. CIT2 expression refers to the qualitative level of CIT2 expression in the various strains grown in YPD medium. (B) Glutamate inhibits [URE3] formation. Cells were cultured and plated in the same way as in A except that YNBD plates supplemented with 200 μg/ml ureidosuccinic acid and the indicated amounts of glutamate were used. Error bars indicate SD of colony numbers from seven plates.

Figure 8

Model linking the negative feedback control of glutamate levels by the RTG pathway and [URE3] formation. Glutamate (or glutamine) could inhibit [URE3] formation directly or through an intermediary sensing mechanism.