Palmitoylation of CYSTM (CYSPD) proteins in yeast

Abstract

A superfamily of proteins called cysteine transmembrane is widely distributed across eukaryotes. These small proteins are characterized by the presence of a conserved motif at the C-terminal region, rich in cysteines, that has been annotated as a transmembrane domain. Orthologs of these proteins have been involved in resistance to pathogens and metal detoxification. The yeast members of the family are YBR016W, YDL012C, YDR034W-B, and YDR210W. Here, we begin the characterization of these proteins at the molecular level and show that Ybr016w, Ydr034w-b, and Ydr210w are palmitoylated proteins. Protein S-acylation or palmitoylation, is a posttranslational modification that consists of the addition of long-chain fatty acids to cysteine residues. We provide evidence that Ybr016w, Ydr210w, and Ydr034w-b are localized to the plasma membrane and exhibit varying degrees of polarity toward the daughter cell, which is dependent on endocytosis and recycling. We suggest the names CPP1, CPP2, and CPP3 (C terminally palmitoylated protein) for YBR016W, YDR210W, and YDR034W-B, respectively. We show that palmitoylation is responsible for the binding of these proteins to the membrane indicating that the cysteine transmembrane on these proteins is not a transmembrane domain. We propose renaming the C-terminal cysteine-rich domain as cysteine-rich palmitoylated domain. Loss of the palmitoyltransferase Erf2 leads to partial degradation of Ybr016w (Cpp1), whereas in the absence of the palmitoyltransferase Akr1, members of this family are completely degraded. For Cpp1, we show that this degradation occurs via the proteasome in an Rsp5-dependent manner, but is not exclusively due to a lack of Cpp1 palmitoylation.

Keywords: Akr1; CYSPD; CYSTM; Erf2; palmitoylation; yeast.

Figure 1

GFP-Cpp1, GFP-Cpp2, and GFP-Cpp3 are localized to the plasma membrane and they are palmitoylated.A, protein sequence alignment and secondary structure prediction for the yeast CYSPD family members. Conserved cysteines are highlighted in green. B, fluorescence images of yeast cells expressing GFP-tagged Cpp1, Cpp2, and Ccp3. The fusions localize to the plasma membrane (PM). GFP-Cpp1 is polarized to the new buds, and GFP-Cpp2 and GFP-Ccp3 are also polarized, to a lesser extent (upper panels). The polarized distribution of GFP-CCp1 is dependent on endocytic cycling since it is lost in a sla1Δ strain (middle panels). GFP-Cpp1 is lost from the PM in the recycling mutant ric1Δ and it is found in intracellular dots. GFP-Cpp2 and GFP-Cpp3 are also found inside the cell in a ric1Δ, but a substantial amount of fluorescence remains at the PM (lower panels). The scale bar represents 2 μm. C, acyl-biotin exchange assay on GFP-tagged Cpp1, Cpp2, and Cpp3 shows that these proteins are palmitoylated. In the hydroxylamine (HA)-treated samples, palmitates are exchanged for biotin, the proteins are pulled down using streptavidin beads and detected on a Western blot using anti-GFP antibodies. Negative controls are treated with Tris buffer instead of hydroxylamine. D, detection of palmitoylation by metabolic labeling and click chemistry. Yeast cells expressing GFP-Cpp1 or GFP-Cpp2 were grown in the presence (+) or absence of azido palmitate (azido-palm), lysed, and the azido palmitate was modified by biotin alkyne. Palmitoylated proteins were pulled down with streptavidin beads and analyzed by Western blot using antibodies against GFP (upper panel), the palmitoylated protein Vac8 (middle panel), or the nonpalmitoylated protein Tom40 (lower panel). CPP, cysteine-rich palmitoylated protein; CYSPD, cysteine-rich palmitoylated domain.

Figure 2

GFP-Cpp1, GFP-Cpp2, and GFP-Cpp3 are bound to membranes by palmitoylation.A, GFP-Cpp1-3 are released from membranes upon hydroxylamine treatment. Membrane fractions of yeast cells expressing GFP-Cpp1-3 were treated with 1 M hydroxylamine, which cleaves the thioester bonds between fatty acids and cysteine residues, or Tris as a control, and the presence of GFP-Cpp1 in the supernatant (S) or the membrane pellet (P) was assessed by Western blot. The endogenous transmembrane protein Tlg1 was detected in the same blots as a negative control. As positive control cells expressing GFP-Yck2 were treated as indicated above and analyzed by Western blot (bottom left panel). A control experiment using the non-palmitoylable version of GFP-Cpp1 (5xΔcys) was also included in the analysis (bottom right panel). B, the signals from bands as in A were quantified and the graph shows a scatter plot displaying the average pellet/supernatant (P/S) ratio ± SD for each protein from three independent experiments. C, fluorescence microscopy images of GFP-Cpp1 induced for 2 h by the addition of galactose, in the presence or absence of 20 μM of the palmitoylation inhibitor 2-bromopalmitate (2-BP) (right panels). As a negative control, Gal-driven GFP-Sso1, a transmembrane protein was included (left panels). Cells expressing the palmitoylated protein GFP-Yck2 were assessed as a positive control for the experiment. The scale bar represents 2 μm. CPP, cysteine-rich palmitoylated protein.

Figure 3

GFP-Cpp1 insertion in the membrane is independent of the GET and EMC complexes. Fluorescence microscopy images showing the membrane localization of GFP-Cpp1 in WT and mutant strains lacking either GET1, GET2, EMC3, or EMC6 (lower panels). As controls, the tail-anchored transmembrane proteins GFP-Sso1 (upper panels) and GFP-Snc1 (middle panels) were included. The scale bar represents 2 μm. CPP, cysteine-rich palmitoylated protein.

Figure 4

GFP-Cpp1 is degraded in the absence of Ark1.A, fluorescence microscopy images showing GFP-Cpp1 expressed in strains lacking each of the yeast PATs. Fluorescence is partially lost in erf2Δ and completely lost in akr1Δ. The scale bar represents 2 μm. B, the scatter plot graph shows the average values ± SD of the fluorescence intensity of the strains as in A. Total Fluorescence was measured in each cell using the same threshold for all the strains. Each average value was relativized to the average fluorescence intensity of the WT group (n = 40). C, fluorescence microscopy images of GFP-Cpp1 expressed in erg6Δ and akr1Δ erg6Δ strains in the presence or absence of the proteasome inhibitor MG132. Degradation of GFP-Cpp1 is prevented by the proteasome inhibitor. Experiments are carried out in an erg6Δ background to improve permeability to the drug. The scale bar represents 2 μm. D, scatter plot graph showing the fluorescence intensity average ± SD of GFP-Cpp1 in the double mutant akr1Δ erg6Δ strain, measured in the presence or absence of MG132. Each average value was relativized to the average fluorescence intensity of the control cells erg6Δ (n = 20). a. u: arbitrary units; CPP, cysteine-rich palmitoylated protein; PAT, palmitoyltransferase; RFI, relative fluorescence intensity.

Figure 5

GFP-Cpp1 degradation is dependent on Rsp5.A, fluorescence microscopy images of GFP-Cpp1 expressed in WT (upper panel), akr1Δ (middle panel), and akr1Δ rsp5hm (lower panel) strains. In the absence of Akr1, the expression and localization of GFP-Cpp1 are rescued by lowering the levels of Rsp5. The scale bar represents 2 μm. B, acyl-biotin exchange assay of GFP-Cpp1 expressed in WT and akr1Δ rsp5hm strains, showing that the protein remains palmitoylated when degradation is precluded. C, Western blot analysis showing the expression levels of GFP-Cpp1 in WT and indicated mutant strains. GFP-Cpp1 is partially degraded in erf2Δ, completely degraded in akr1Δ and this degradation is partially precluded in akr1Δ rsp5hm strain. Tlg1 was used as a loading control. D, bar graph showing the average values ± SD from the quantification of three independent experiments as in C. CPP, cysteine-rich palmitoylated protein; CYSPD, cysteine-rich palmitoylated domain; HA, hydroxylamine.

Figure 6

Palmitoylation and localization of GFP-Cpp1 cysteine mutants.A, amino acid sequences of WT and cysteine mutants in Cpp1 CYSPD. Cysteines are colored in green and introduced alanines are colored in red. B, fluorescence microscopy images of GFP-Cpp1 and the GFP Cpp1 cysteine mutants. The scale bar represents 2 μm. C, acyl-biotin exchange (ABE) assay of WT GFP-Cpp1 and indicated cysteine mutants, showing that cysteines in both halves of the domain can be palmitoylated. CPP, cysteine-rich palmitoylated protein; CYSPD, cysteine-rich palmitoylated domain; HA, hydroxylamine.

Figure 7

Degradation of GFP-Cpp1 in akr1Δ is independent of the CYSPD domain.A, Western blot analysis of GFP-Cpp1 and mutant versions with either all the cysteines mutated to alanine (5xΔCys), lacking the CYSPD domain (ΔCYSPD) or with the CYSPD replaced by the transmembrane domain of Sso1 Cpp1-(Sso1), were expressed in WT and akr1Δ strains (left panel). The band intensities were quantified and graphed. Plots show the mean values ± SD from three independent experiments (right panel). B, Western blot analysis of GFP-Cpp1 and mutant versions with either all the cysteines mutated to alanine (5xΔCys), lacking the CYSPD domain (ΔCYSPD) or with the CYSPD replaced by the transmembrane domain of Sso1 Cpp-(Sso1), were expressed in WT and erf2Δ strains (left panel). The band intensities were quantified and graphed. Plots show the mean ± SD values from three independent experiments (right panel). Tlg1 was used as a loading control. CPP, cysteine-rich palmitoylated protein; CYSPD, cysteine-rich palmitoylated domain.