Yeast Mpk1 mitogen-activated protein kinase activates transcription through Swi4/Swi6 by a noncatalytic mechanism that requires upstream signal

Abstract

The cell wall integrity mitogen-activated protein kinase (MAPK) cascade of Saccharomyces cerevisiae drives changes in gene expression in response to cell wall stress. We show that the MAPK of this pathway (Mpk1) and its pseudokinase paralog (Mlp1) use a noncatalytic mechanism to activate transcription of the FKS2 gene. Transcriptional activation of FKS2 was dependent on the Swi4/Swi6 (SBF) transcription factor and on an activating signal to Mpk1 but not on protein kinase activity. Activated (phosphorylated) Mpk1 and Mlp1 were detected in a complex with Swi4 and Swi6 at the FKS2 promoter. Mpk1 association with Swi4 in vivo required phosphorylation of Mpk1. Promoter association of Mpk1 and the Swi4 DNA-binding subunit of SBF were codependent but did not require Swi6, indicating that the MAPK confers DNA-binding ability to Swi4. Based on these data, we propose a model in which phosphorylated Mpk1 or Mlp1 forms a dimeric complex with Swi4 that is competent to associate with the FKS2 promoter. This complex then recruits Swi6 to activate transcription. Finally, we show that human ERK5, a functional ortholog of Mpk1, is similarly capable of driving FKS2 expression in the absence of protein kinase activity, suggesting that this mammalian MAPK may also have a noncatalytic function in vivo.

FIG. 1.

Cell wall stress induction of FKS2 expression is dependent on an SBF-binding site and SBF. (A) The SBF-binding site in the FKS2 promoter is required for cell wall stress-induced activation. FKS2-lacZ reporter plasmids bearing point mutations in a predicted SBF-binding site (wild-type, p2052; −386T, p2053; −388C, p2054; and −389G, p2055) were transformed into wild-type yeast strain 1788. Transformants were grown to saturation at 23°C in SD-uracil medium. Cultures were diluted into 3 ml of YEPD medium so that subsequent incubation at 23°C, 39°C, or 23°C with 50 μg/ml Congo red for 15 h resulted in mid-log-phase cultures (A₆₀₀ of 1.0 to 1.5). β-Galactosidase (β-Gal) activity was measured in crude extracts. (B) Mutants in SWI4 and SWI6 are defective for thermal activation of FKS2 expression. An FKS2-lacZ reporter plasmid (p2052) was transformed into wild-type (1788), swi4Δ (DL3145), or swi6Δ (DL3148) strains. Transformants were treated as described in panel A, except that cultures were diluted into YEPD medium containing 10% sorbitol for osmotic support. (C) Mutants in SWI4 and SWI6 are competent for thermal activation of PRM5 expression. A PRM5-lacZ reporter plasmid (p1366) was transformed into wild-type (1788), swi4Δ (DL3145), or swi6Δ (DL3148) strains. Transformants were treated as described in panel A, except that cultures were diluted into YEPD medium containing 10% sorbitol for osmotic support. (D) A cell cycle-regulated SCB reporter is not induced in response to cell wall stress. A CLN2-lacZ reporter plasmid (p2066) and its parent vector (p904) were transformed into the wild-type strain (1788). Transformants were treated as described in panel A. Each value represents the mean and standard deviation from three independent transformants. Sp. Act., specific activity; U, unit.

FIG. 2.

CWI requirements for thermal induction of FKS2 and PRM5. (A) Induction of FKS2 expression is dependent on either Mpk1 or its pseudokinase paralog, Mlp1. An FKS2-lacZ reporter plasmid (p2052) was transformed into the wild-type (DL3193), bck1Δ (DL3529), mkk1Δ mkk2Δ (DL3541), mpk1Δ (DL3195), mlp1Δ (DL3194), or mpk1Δ mlp1Δ (DL3196) strain. Transformants were cultured as described in the legend of Fig. 1 in the presence of YEPD medium containing 10% sorbitol for osmotic support. β-Galactosidase (β-Gal) activity was measured in crude extracts. (B) Mpk1 does not require catalytic activity for thermal induction of FKS2, but Mpk1 and Mlp1 must be phosphorylated. An FKS2-lacZ reporter plasmid (p2052) was cotransformed with centromeric plasmids bearing MPK1 (p2188), mpk1-TA/YF (p2190), mpk1-K54R (p2193), MLP1 (p2346), mlp1-Y192F (p2347), or vector (pRS315) into an mpk1Δ mlp1Δ strain (DL3196). Transformants were treated as above. (C) Mpk1 catalytic activity is required for thermal induction of PRM5. A PRM5-lacZ reporter plasmid (p1366) was cotransformed with the indicated plasmids from panel B into an mpk1Δ mlp1Δ strain (DL3196). Transformants were treated as above, except that culture time was 5 h rather than 15 h. Each value represents the mean and standard deviation from three independent transformants. Sp. Act., specific activity; U, unit.

FIG. 3.

Phosphorylated Mpk1 and Mlp1 bind to the FKS2 promoter. (A) ChIP analysis of the FKS2 promoter with Mpk1-HA. A wild-type yeast strain (1788) was cotransformed with FKS2-lacZ reporter plasmids (p2052, wild-type; or p2053, SBF-site mutant −386T) and multicopy plasmids expressing the indicated HA-tagged MPK1 allele (wild-type, p777; mpk1-TA/YF, p778; mpk1-K54R, p2119; or vector, YEp351). Transformants were cultivated in YEPD medium at 23°C or subjected to thermal stress at 39°C for 15 h prior to ChIP analysis using primers designed to detect only the plasmid-borne FKS2 promoter. PCRs from whole-cell extracts (WCE) are also shown. (B) ChIP analysis of the FKS2 promoter with Mlp1-HA. Wild-type yeast strain 1788 was cotransformed with FKS2-lacZ reporter plasmids from panel A and multicopy plasmids expressing the indicated HA-tagged MLP1 allele (wild-type, p2022; mlp1-Y192F, p2024; or vector, YEp351). Transformants were cultivated and subjected to thermal stress prior to ChIP analysis, as above. WT, wild type; TAYF, T190A Y192F.

FIG. 4.

Interdependent recruitment of Mpk1 and Swi4 to the FKS2 promoter. (A) Genomic ChIP demonstrates that association of Mpk1-HA with the FKS2 promoter is dependent on Swi4 but not Swi6. Yeast strains (wild-type, DL3187; swi4Δ, DL3405; or swi6Δ, DL3233) were transformed with a multicopy plasmid bearing the indicated MPK1-HA allele and subjected to thermal stress prior to ChIP analysis as described in the legend of Fig. 3, except that strains were cultivated in the presence of 10% sorbitol. An additional primer pair for PCR amplification DYN1 sequence was included as a negative control. (B) Association of Swi4-HA with the FKS2 promoter is dependent on Mpk1/Mlp1 but not Swi6. Yeast strains (wild-type, DL3187; swi6Δ, DL3233; or mpk1Δ mlp1Δ, DL3183) were transformed with a multicopy plasmid bearing SWI4-HA (p2339) and treated as above. (C) Association of Swi6 with the FKS2 promoter is dependent on both Swi4 and Mpk1/Mlp1. Yeast strains (wild-type, DL3187; swi4Δ, DL3405; or mpk1Δ mlp1Δ, DL3183) were transformed with a multicopy plasmid bearing SWI6-HA (p2341) and treated as above. WCE, whole-cell extract; WT, wild type; α, anti; TAYF, mpk1-T190A Y192F.

FIG. 5.

Physical association of activated Mpk1 with Swi4. (A) Two-hybrid association of Mpk1 with Swi4 is dependent on activation of Mpk1. A Gal4DBD fusion to the catalytic domain of Mpk1, Mpk1(1-328) (plasmid p2248), was tested for interaction with a Gal4AD fusion to Swi4 (p2352) in yeast two-hybrid strain SFY526. Transformants were cultivated in selective medium for 15 h either at 23°C or 39°C. Vector controls (Gal4DBD vector, p1172; Gal4AD vector, p1173) are included for each fusion. Each value represents the mean and standard deviation from three independent transformants. (B) Coimmunoprecipitation (IP) of Swi4 with Mpk1 requires phosphorylation of Mpk1 but not its protein kinase activity. Yeast strain DL454 (mpk1Δ) was cotransformed with multicopy plasmids bearing the indicated FLAG-tagged MPK1 allele (wild-type, p2313; mpk1-TA/YF, p2316; mpk1-K54R, p2317) and His-tagged SWI4 allele (p2418). Transformants were cultivated to mid-log phase in selective medium lacking methionine (to induce expression of Swi4-His₆) and either subjected to heat stress for 1 h at 39°C or maintained at 25°C. Cell extracts (Input) and immunoprecipitates with anti-FLAG M2 affinity gel (Sigma) or protein A affinity gel (Sigma) or no-antibody controls were subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis and analyzed by immunoblotting with anti-FLAG or anti-His (Qiagen) antibodies. α, anti; WT, wild type; β-Gal, β-galactosidase; Sp. Act., specific activity; U, unit; Ab, antibody.

FIG. 6.

Model for CWI-regulated induction of FKS2 expression. (A) Cell wall stress results in phosphorylation and activation of the Mpk1 MAPK and the Mlp1 pseudokinase. Activated Mpk1 or Mlp1 binds Swi4. (B) The dimeric complex binds to the SBF-binding site within the FKS2 promoter. (C) Swi6 engages the ternary complex to initiate transcription. (D) Model for activation of transcription by CWI signaling. Cell wall stress activates the Mpk1 MAPK and the Rlm1 transcription factor. MLP1 expression is under the transcriptional control of Rlm1. Stress-induced Mlp1 is phosphorylated by MEKs (Mkk1/2) and stimulates FKS2 transcription redundantly with Mpk1 by noncatalytic activation of SBF. P, phosphate.

FIG. 7.

The noncatalytic transcriptional function of Mpk1 and Mlp1 is conserved in human ERK5. FKS2-lacZ reporter plasmids (wild-type, p2052; SBF-binding site mutant, p2053) were cotransformed with ERK5 expression plasmids into an mpk1Δ mlp1Δ strain (DL3196). ERK5 plasmids were wild-type (ERK5; p2349), ERK5-T219A Y221F (TAYF; p2350), ERK5-K84R (p2351), or vector (pUT34; p2348). Transformants were cultured as described in the legend of Fig. 1 in the presence of YEPD medium containing 10% sorbitol for osmotic support. β-Galactosidase (β-Gal) activity was measured in crude extracts. Each value represents the mean and standard deviation from three independent transformants. Sp. Act., specific activity; U, unit.