Degradation of citrate synthase lacking the mitochondrial targeting sequence is inhibited in cells defective in Hsp70/Hsp40 chaperones under heat stress conditions

Abstract

Most nucleus‐encoded mitochondrial precursor proteins are synthesized in the cytosol and imported into mitochondria in a post‐translational manner. In recent years, the quality control mechanisms of nonimported mitochondrial proteins have been intensively studied. In a previous study, we established that in budding yeast a mutant form of citrate synthase 1 (N∆Cit1) that lacks the N-terminal mitochondrial targeting sequence, and therefore mislocalizes to the cytosol is targeted for proteasomal degradation by the SCF^Ucc1 ubiquitin ligase complex. Here, we show that Hsp70 and Hsp40 chaperones (Ssa1 and Ydj1 in yeast, respectively) are required for N∆Cit1 degradation under heat stress conditions. In the absence of Hsp70 function, a portion of N∆Cit1-GFP formed insoluble aggregates and cytosolic foci. However, the extent of ubiquitination of N∆Cit1 was unaffected, implying that Hsp70/Hsp40 chaperones are involved in the postubiquitination step of N∆Cit1 degradation. Intriguingly, degradation of cytosolic/peroxisomal gluconeogenic citrate synthase (Cit2), an endogenous substrate for SCF^Ucc1-mediated proteasomal degradation, was not highly dependent on Hsp70 even under heat stress conditions. These results suggest that mitochondrial citrate synthase is thermally vulnerable in the cytosol, where Hsp70/Hsp40 chaperones are required to facilitate its degradation.

citrate synthase, TCA cycle, glyoxylate cycle, ubiquitin–proteasome system, degradation, Hsp70, Saccharomyces cerevisiae

Introduction

Mitochondria consist of several hundred different proteins, most of which are synthesized in the cytosol as precursors with an N-terminal mitochondrial targeting sequence, also known as a presequence. If synthesis of the precursor is completed in the cytosol, cytosolic factors including molecular chaperones may bind to the precursor, keep it in an unfolded, import-competent state, and facilitate its transfer to mitochondrial receptor proteins before it is imported into mitochondria through the translocator of the outer membrane and translocator of the inner membrane channels (Pfanner et al. 2019, Bykov et al. 2020, Drwesh and Rapaport 2020, Araiso et al. 2022).

However, the import process is affected by mitochondrial defects, premature folding or misfolding of import substrates, problems in the cytosol, and cytosolic aggregation of pathological protein species including mutant huntingtin/polyQ protein, α-synuclein, and β-amyloid (Mossmann et al. 2014, Yano et al. 2014, Cenini et al. 2016, Di Maio et al. 2016, Bauerlein et al. 2017, Sorrentino et al. 2017, Boos et al. 2019, Martensson et al. 2019, Szego et al. 2019, Yablonska et al. 2019, Nowicka et al. 2021). Impairment of mitochondrial import results in accumulation of nonimported proteins in the cytosol, where there is a potential risk of misfolding due to the absence of suitable maturation factors. To cope with such deleterious situations, cells have evolved protein quality control mechanisms via which cytosolic molecular chaperones maintain the solubility and prevent the aggregation of nonimported proteins in the cytosol, and terminally misfolded proteins are eliminated by the ubiquitin–proteasome system (UPS) (Wang and Chen 2015, Wrobel et al. 2015, Itakura et al. 2016, Hansen et al. 2018, Weidberg and Amon 2018, Zurita Rendon et al. 2018, Boos et al. 2019, 2020, Martensson et al. 2019, Su et al. 2019, Metzger et al. 2020, Shakya et al. 2021, Song et al. 2021). Failure of chaperone systems and/or the UPS may result in excessive accumulation of nonimported proteins, which form toxic aggregates and/or associate with other proteins or organelles in the cell, leading to proteostasis disturbance, cellular growth defects, neurodegenerative diseases, or aging (Maruszczak et al. 2023). However, it remains unclear whether and how cytosolic chaperones and other factors help to discriminate between import, degradation, and folding of newly synthesized mitochondrial precursor proteins in vivo.

In a previous study, we established that the mutant form of mitochondrial citrate synthase lacking the N-terminal targeting sequence (N∆Cit1) mislocalizes to the cytosol and is ubiquitinated by the SCF^Ucc1 (Skp1–Cdc53–F-box protein Ucc1) ubiquitin ligase complex for proteasomal degradation (Nishio et al. 2023). Similarly, the precursor form of Cit1 (preCit1) that accumulates upon import inhibition is also degraded in a manner dependent on the SCF^Ucc1 complex and the proteasome (Nishio et al. 2023).

Herein, to further characterize the degradation pathway of nonimported mitochondrial proteins, we focused on the essential SSA class of Hsp70 chaperones along with the cochaperone Hsp40, and analyzed their role in N∆Cit1 degradation. We demonstrate that Hsp70 and Hsp40 chaperones (Ssa1 and Ydj1 in yeast, respectively) are required for N∆Cit1 degradation under heat stress conditions. In the absence of Hsp70 function, a portion of N∆Cit1-GFP formed insoluble aggregates and cytosolic foci; however, the extent of ubiquitination of N∆Cit1 was unaltered, leading to the hypothesis that Hsp70/Hsp40 chaperones are involved in N∆Cit1 degradation at the postubiquitination step. Intriguingly, degradation of cytosolic/peroxisomal gluconeogenic citrate synthase (Cit2), an endogenous substrate for SCF^Ucc1-mediated proteasomal degradation (Nakatsukasa et al. 2015), was only slightly affected in cells defective in Hsp70 even under heat stress conditions. We suggest that mitochondrial citrate synthase is thermally vulnerable in the cytosol, where Hsp70/Hsp40 chaperones are required to facilitate its degradation.

Materials and methods

Yeast strains and plasmids

The yeast strains, plasmids, and oligonucleotide primers used in this study are listed in Tables 1, 2, and 3, respectively. All gene deletion and tagged strains were constructed using standard homologous recombination methods (Burke et al. 2000). The UCC1 gene was disrupted by transforming yeast strains with a ucc1Δ::CgHIS cassette that was amplified by PCR from BYP1804 (National BioResource Project) using primers OKN1718 and OKN1719. The SSA2 gene was disrupted by transforming the yeast strain (ssa1Δ::Kan^R, Thermo Fisher Scientific) with a ssa2Δ::CgHIS cassette that was amplified by PCR from BYP1804 using primers OKN2301 and OKN2302. Correct disruption was confirmed by colony PCR using primers OKN817/OKN818 (ucc1Δ) and OKN2303/OKN2304 (ssa2Δ).

Table 1.

Yeast strains used in this study.

Name	Genotype	Reference/source
JN516	MATα, his3-11,15, leu2-3 112, ura3-52, trp1-Δ1, lys2, ssa2-1(LEU2), ssa3-1(TRP1), ssa4-2(LYS2)	Becker et al. (1996)
JB67	MATα, his3-11,15, leu2-3 112, ura3-52. trp1-Δ1, lys2, ssa1-45, ssa2-1(LEU2), ssa3-1(TRP1), ssa4-2(LYS2)	Becker et al. (1996)
W303-1b	MATα, ade2-1, can1-100, ura3-1, trp1-1, his3-11,15, leu2-3 112	Caplan et al. (1992)
ydj1-151	MATα, ade2-1, can1-100, ura3-1, trp1-1, his3-11,15, leu2-3 112, ydj1-2::HIS3, ydj1-151::LEU2	Caplan et al. (1992)
BY4741	MATa, his3Δ1, leu2Δ0, met15Δ0, ura3Δ0	S. Michaelis
MHY3-6C	MATa, his3Δ1, leu2Δ0, met15Δ0, ura3Δ0, ssa1Δ::KanR ssa2Δ::CgHIS3	This study
W303-1a	MATa, ade2-1, can1-100, ura3-1, trp1-1, his3-11,15, leu2-3 112	T. Kamura
MHY4-5G	MATα, his3-11,15, leu2-3 112, ura3-52, trp1-Δ1, lys2, ssa2-1(LEU2), ssa3-1(TRP1), ssa4-2(LYS2), ucc1Δ::CgHIS	This study
MHY4-5E	MATα, his3-11,15, leu2-3 112, ura3-52. trp1-Δ1, lys2, ssa1-45, ssa2-1(LEU2), ssa3-1(TRP1), ssa4-2(LYS2), ucc1Δ::CgHIS	This study
TKY2168	MATa, ade2-1, can1-100, ura3-1, trp1-1, his3-11,15, leu2-3 112, pdr5∆::HPH, ucc1∆::KanR	Nakatsukasa et al. (2015)

Table 2.

Plasmids used in this study.

Name	Description	Reference/source
pRS316 pKN434	CEN/ARS, URA3, AmpR p416: PCIT1-CIT1(∆2–42)-HA-TCYC1	Sikorski and Hieter (1989) Nishio et al. (2023)
pKT11 pKT12	pRS316: PCIT1-CIT1-EGFP-TCYC1 pRS316: PCIT1-CIT1(∆2–42)-EGFP-TCYC1	Nishio et al. (2023) Nishio et al. (2023)
pKN194 pKN545 pKN372 pRS423mycUb pSM989	p415: PCIT2-CIT2-HA-SKL-TCIT2 p415: PGAL1-CIT2-HA-SKL-TCYC1 2 μm, LEU2, PCUP1-8xHis–tagged Ub-TCYC1 2 μm, HIS3, PCUP1-myc–tagged Ub 2 μm, TRP1, ubi::myc	Nakatsukasa et al. (2015) Nishio et al. (2023) This study Nakatsukasa et al. (2008) Hochstrasser et al. (1991)
pKN485	pRS316: PCIT1-CIT1(∆2–42, D173A, S429A)-HA-TCYC1	Nishio et al. (2023)

Table 3.

Oligonucleotide primers used in this study.

Name	Sequence
OKN1718	GTT TGT ACT ATG CCT ATT GGC GCA AAG AAG ACA GAG TGT GCA AAC AAG AGG TTG TAA AAC GAC GGC CAG T
OKN1719	GCA TAG ACG TAT ATA CAC ATA TAT TAT AGA CTA ATT GAT AAA TTT TTT TTC AGG AAA CAG CTA TGA CCA T
OKN817	TTC AAA GTC TCT GCC CCT CTA TTG
OKN818	GTC AAT AGG TAC GAT ATA ATG AGT GAG CTG
OKN2301	TTG ATT AAT TCC AAC AGA TCA AGC AGA TTT TAT ACA GAA ATA TTT ATA CAG TTG TAA AAC GAC GGC CAG T
OKN2302	TTA ATC AAC TTC TTC GAC AGT TGG ACC TTC AGC TTC TGG AGC TGG AGG AGC AGG AAA CAG CTA TGA CCA T
OKN2303	TGA ATT GAC TGC ATG GCA CTG C
OKN2304	GCA GAC GGT ATC TAC ACT GCT GTA C

Growth medium

Yeast strains were grown in YP-rich medium [1% yeast extract, 1% peptone, and adenine hydrochloride (100 mg/l)] or synthetic complete medium [0.67% yeast nitrogen base without amino acids, adenine hydrochloride (100 mg/l), and all standard amino acids] containing 2% glucose, 2% raffinose, or 2% galactose.

Cycloheximide chase experiment

The cycloheximide chase protein degradation assay was performed as described previously. Basically, cells were grown to mid-log phase at 26°C or 30°C, and time = 0 samples were collected immediately before 100 µg/ml cycloheximide was added to the medium. Where indicated, the temperature was shifted to 37°C in a water bath shaker. Where indicated, cells were grown at 30°C after cycloheximide was added. At the indicated time points, cells were collected and suspended in 10%−20% trichloroacetic acid on ice. Cells were then collected by centrifugation, and the pellet was stored at −30°C. The cell pellet was suspended in 300 µl of 20% trichloroacetic acid and lysed by vigorous vortexing with glass beads for ∼30 min (TAITEC Max Mixer EVR-032), with occasional inversion of the tube to prevent cells accumulating at the bottom. The cell lysate was added to 900 µl of 5% trichloroacetic acid, the tube was inverted several times, and then 900 µl of the suspension was transferred to a new tube. Proteins were precipitated by centrifugation at 20 000 g for 15 min at 4°C. The precipitates were dissolved in KNTCASB [80 mM Tris-HCl (pH 7.5), 8 mM EDTA (pH 8.0), 12.5% glycerol, 8 M urea, 4% SDS, 200 mM DTT, Tris (0.8 mg/ml), and 0.1% bromophenol blue]. Samples were heated at 55°C for 15 min and cleared by centrifugation at 20 000 g for 1 min at room temperature before SDS-polyacrylamide gel electrophoresis (SDS-PAGE) and western blotting analyses were performed. In Fig. 6(E), protein samples were run on an SDS-PAGE gel containing 8 M urea.

Western blotting

Proteins were transferred from polyacrylamide gels to Immobilon-P membranes (Millipore, IPVH00010) in blotting buffer (25 mM Tris, 192 mM glycine, and 10% methanol) using the GENIE electrophoretic transfer device (Idea Scientific Company) at a constant current of 650 mA. The membranes were washed with TBS-T buffer [20 mM Tris-HCl (pH 7.5), 150 mM NaCl, and 0.1% Tween 20] and, when necessary, blocked with 3% skim milk in TBS-T buffer for 30 min. Subsequently, the membranes were incubated with primary antibodies in TBS-T buffer, TBS-T buffer supplemented with 3% skim milk, or Can Get Signal (TOYOBO, NKB-201) at 4°C overnight. Cit2-HA was detected using an anti-HA antibody diluted in Can Get Signal in Fig. 6(A) and (E), but with the same antibody diluted in TBS-T buffer supplemented with 3% skim milk in Fig. 6(C). The membranes were then washed three times with TBS-T buffer (10–60 min per wash), incubated with secondary antibodies [Sigma-Aldrich, antirabbit IgG (whole molecule)-peroxidase antibody, A6154-1ML; antimouse IgG (whole molecule)-peroxidase antibody, A4416-1ML] for ∼60 min, and washed three times with TBS-T buffer. Finally, the membranes were incubated with Chemi-Lumi One (Nacalai Tesque, 07880) or Luminata Forte Western HRP substrate (Merck Millipore, WBLUF0500) and exposed to X-ray film. Band intensities were quantified with ImageJ (National Institutes of Health) directly from films scanned at high resolution (EPSON, GT-X980) in TIF file format (600 dpi). Anti-HA (180–3 TANA2) and anti-Pgk1 (ab113687, 22C5D8) antibodies were purchased from MBL (MEDICAL & BIOLOGICAL LABORATORIES) and Abcam, respectively. An anti-GFP antibody (AB0020-200) was purchased from Sicgen. The anti-Dpm1 (5C5A7) antibody was purchased from Invitrogen. An anti-Cdc48 antibody was a gift from T. Endo (Kyoto Sangyo University).

Subcellular fractionation analysis

Cells were grown at 26°C until the mid-log phase. Where indicated, cells were treated with 100 µg/ml cycloheximide immediately prior to shifting the temperature to 37°C for 30 min. Cells were collected and suspended in buffer 88 at pH 7.4 (20 mM HEPES (pH 7.4), 150 mM KOAc, 250 mM sorbitol, 5 mM MgOAc) supplemented with 1 × PIC and disrupted using glass beads. Unbroken cells and aggregated materials were removed by centrifugation at 300 g at 4°C for 3 min before the cell homogenates were centrifuged at 20 000 g at 4°C for 30 min. The supernatant proteins (cytosol fraction) were precipitated in 10% trichloroacetic acid and dissolved in KNTCASB. The resulting pellets (membrane fraction) were washed in buffer 88 (pH 7.4) and proteins were again precipitated in 10% trichloroacetic acid and dissolved in KNTCASB. Thereafter, protein samples were subjected to SDS-PAGE and western blotting.

In vitro pull-down analysis

Some of the following methods are similar to those previously published (Nishio et al. 2023). The His6–3xFLAG–Ucc1–Skp1 complex was purified from Sf21 cells that had been infected with a baculovirus encoding His6–3xFLAG–Ucc1 and Skp1 using Ni-NTA agarose (FUJIFILM Wako, catalog number 141–09764) chromatography (Nakatsukasa et al. 2015). This complex was dialyzed against dialysis buffer [40 mM Tris-HCl (pH 7.5), 60 mM NaCl, and 10% glycerol] and stored at −80°C. The complex was then immobilized to anti-FLAG M2 affinity gel (Sigma-Aldrich, A2220) in buffer [40 mM Tris-HCl (pH 7.5), 150 mM NaCl, 10% glycerol, and 0.1% Triton X-100] by rotating at 4°C for ∼30 min. Subsequently, the beads were washed three times with the same buffer. In parallel, cells (ucc1Δ background) expressing NΔCit1-HA under the control of the CIT1 promoter from a low-copy plasmid were cultured in SD medium overnight at the permissive temperature, diluted into YP-rich medium containing glucose, and grown to log phase. Where indicated, cells were shifted to 37°C for 40 min. Cells were then lysed in buffer [40 mM Tris-HCl (pH 7.5), 150 mM NaCl, and 0.5% Triton X-100] supplemented with 1 × protease inhibitor cocktail (PIC) using glass beads in a round-bottom plastic tube by eight rounds of vigorous vortexing for 30 s with 30 s intervals on ice. The suspension was added to 400 µl of the same buffer and transferred to a new tube. The glass beads were washed with 400 µl of buffer and pooled in the same tube. The lysates were cleared by centrifugation at 20 000 g for 10 min at 4°C. A portion of the supernatant was saved, and proteins were precipitated using trichloroacetic acid, dissolved in KNTCASB, and used as an input control. The remaining lysate was incubated with anti-FLAG beads that had been loaded with the His6–3xFLAG–Ucc1–Skp1 complex (see above) and rotated at 4°C for 20–40 min. The beads were then washed five times with buffer [40 mM Tris-HCl (pH 7.5), 150 mM NaCl, 10% glycerol, and 0.1% Triton X-100]. The proteins were eluted with KNTCASB at 55°C for 30 min, separated by SDS-PAGE, and subjected to western blotting.

In vivo ubiquitination assay

Some of the following methods are similar to those previously published (Nishio et al. 2023). Cells expressing NΔCit1-HA from a low-copy plasmid were transformed with the pRS423mycUb, pKN372, or pSM989 plasmid encoding ubiquitin under the control of the copper-inducible promoter. The cells were then cultured in SD medium overnight at the permissive temperature and diluted into YP-rich medium containing glucose. His-tagged ubiquitin expression was induced by adding 100 µM CuSO₄ to the medium. After 5–10 h, when OD₆₀₀ reached ∼1.0, cells were shifted to 37°C for 30 min. Typically, ∼30 OD₆₀₀ equivalents of cells (∼70 OD₆₀₀ in Fig. 5F) were treated with 10 mM azide and the cell pellets were stored at −80°C. Subsequently, the cells were lysed in 250 µl of buffer [20 mM Tris-HCl (pH 7.5), 150 mM NaCl, 0.5% Triton X-100, 0.1% SDS, and 10 mM N-ethylmaleimide] supplemented with 1 × complete EDTA-free PIC (Roche, catalog number 11873580001) using glass beads in a round-bottom plastic tube by eight rounds of vigorous vortexing for 30 s with 30 s intervals on ice. The same buffer (500 µl) was then added, and the suspension was transferred to a new tube. The glass beads were washed again with 500 µl of buffer and pooled in the same tube. The collected lysates were cleared by centrifugation at 300 g for 10 min at 4°C, and the cleared lysates were incubated with 1 µl of an anti-HA antibody and Dynabeads Protein G (Invitrogen) at 4°C for 2 h. The beads were then washed four times with buffer [20 mM Tris-HCl (pH 7.5), 150 mM NaCl, 0.5% Triton X-100, 0.05% SDS, and 10 mM N-ethylmaleimide]. Immunoprecipitated proteins were eluted with KNTCASB at 55°C for 30 min, separated by SDS-PAGE, and subjected to western blotting. Polyubiquitin chains were detected with a mono- and polyubiquitinated conjugates monoclonal antibody (FK2) (HRP conjugate) (BioMol, PW0150).

Fluorescence microscopy

Cells expressing NΔCit1-EGFP under the control of the CIT1 promoter from a low-copy plasmid (CEN/ARS) (pKT12) were inoculated into SD medium and grown at 26°C. Where indicated, cells were treated with cycloheximide (100 µg/ml) for 15 or 30 min. Cells were fixed with 10% formaldehyde for 30 min and washed with TBS buffer [20 mM Tris-HCl (pH 7.5) and 150 mM NaCl]. Where indicated, cells were incubated with 5 µg/ml Hoechst 33342 (DOJINDO, 346–07951) for 10 min at room temperature and washed twice with TBS buffer. Cells were observed under an Axio Observer inverted microscope equipped with an Axiocam 503 mono camera (Carl Zeiss) manipulated by ZEN software. GFP was excited at 475 nm (Colibri 7 Type RYB-UV), and the emission was passed through a 524/50 nm band-pass filter (92 HE light-emitting diode (LED) [4′,6-diamidino-2-phenylindole (DAPI)/GFP/mCherry]). Hoechst 33342 was excited at 385 nm (Colibri 7 Type RYB-UV), and the emission was passed through a 425/30 nm band-pass filter [92 HE LED (DAPI/GFP/mCherry)].

Aggregation assay

Cells (∼15 OD₆₀₀ equivalent) were suspended in 300 µl of lysis buffer [30 mM Tris-HCl (pH 7.5), 20 mM KCl, 150 mM NaCl, 1 mM EDTA, and 0.5% Triton X-100] supplemented with 1 × PIC before they were disrupted in the presence of glass beads (Yasui Kikai) in a round-bottom plastic tube by eight rounds of vigorous vortexing for 30 s with 30 s intervals on ice. The same buffer (450 µl) was then added, and the suspension was transferred to a new tube. The glass beads were washed again with 450 µl of lysis buffer and pooled in the same tube. The collected lysates were cleared by centrifugation at 300 g for 3 min at 4°C. The supernatant (∼600 µl) was ultracentrifuged at 100 000 g for 30 min at 4°C (Beckman coulter, Optima MAX-XP). The resultant supernatant (∼300 µl) was transferred to a new tube and proteins were precipitated with 10% trichloroacetic acid before they were dissolved in 100 μl of KNTCASB-T (50 µl of KNTCASB was supplemented with 50 μl of 1 M Tris). The pellet obtained by ultracentrifugation was rinsed with 600 µl of lysis buffer before it was dissolved in 100 μl of KNTCASB.

Coimmunoprecipitation assay

Cells (∼90 OD₆₀₀ equivalent) were grown in synthetic complete medium containing glucose overnight at 26°C and diluted into YP-rich medium containing glucose. When OD₆₀₀ reached ∼0.7 (it usually took ∼6 h), cells were shifted to 37°C for 15 min before they were treated with 10 mM azide. The collected cells were immediately dissolved in buffer [20 mM Tris-HCl (pH 7.5), 150 mM NaCl, 10% glycerol, and 0.5% Triton X-100] supplemented with 1 × PIC and disrupted with glass beads by four rounds of vigorous vortexing for 30 s with 30 s intervals at 4°C using a multibeads shocker (Yasui Kikai). The cell homogenates were centrifuged at 20 000 g for 15 min at 4°C to remove unbroken cells and aggregated materials. The supernatants were further cleared by centrifugation at 20 000 g for 15 min at 4°C. The cleared lysates were incubated with 1 µl of an anti-HA antibody and Dynabeads Protein G (Invitrogen) at 4°C for 50 min. The beads were then washed three times with buffer [20 mM Tris-HCl (pH 7.5), 150 mM NaCl, 10% glycerol, and 0.5% Triton X-100]. Immunoprecipitated proteins were eluted with KNTCASB at 55°C for 30 min, separated by SDS-PAGE, and subjected to western blotting.

Results

The SSA class of Hsp70 chaperones and the Hsp40 cochaperone Ydj1 are required for NΔCit1 degradation under heat stress conditions

In a previous study, we analyzed the stability of a mitochondrial Cit1 mutant lacking amino acids 2–42, the region that contains the N-terminal mitochondrial targeting sequence (hereafter, we call this mutant NΔCit1). We established that NΔCit1 mislocalizes to the cytosol and is ubiquitinated by the SCF^Ucc1 ubiquitin ligase for proteasomal degradation (Nishio et al. 2023). To further elucidate the requirements for NΔCit1 degradation, we focused on the essential SSA class of Hsp70 chaperones, which comprises four SSA paralogs in yeast: SSA1, SSA2, SSA3, and SSA4 (Werner-Washburne et al. 1987). The SSA Hsp70 chaperones are partially but not completely redundant in terms of substrate activity and expression (Werner-Washburne et al. 1987, 1989, Boorstein and Craig 1990a, b, Nelson et al. 1992, Sharma et al. 2009). Ssa1 and Ssa2 are normally expressed at high levels during log phase, although SSA1 expression is further induced by heat shock. Ssa3 and Ssa4 are stress-induced variants (Werner-Washburne et al. 1989). The ssa1Δssa2Δ strain has been established to study the role of Ssa1/Ssa2 in degradation of misfolded proteins in the cytosol and nucleus (Heck et al. 2010, Prasad et al. 2010, 2012, 2018, Jones et al. 2020); therefore, we first performed a cycloheximide chase assay to monitor turnover of NΔCit1 in ssa1Δssa2Δ cells. NΔCit1 was degraded in ssa1Δssa2Δ cells to a similar extent as in SSA1SSA2 (isogenic wild-type) cells at the normal growth temperature for budding yeast (30°C) (Fig. 1A and B), suggesting that Ssa1 and Ssa2 are not essential for NΔCit1 degradation. Intriguingly, however, when cells were shifted to 37°C immediately after cycloheximide was added, the rate of NΔCit1 degradation was considerably slowed in ssa1Δssa2Δ cells (Fig. 1C and D), suggesting that Ssa1/Ssa2 are required for NΔCit1 degradation under heat stress conditions.

Figure 1.

Cycloheximide chase analysis of N∆Cit1 in cells lacking Ssa1 or Ydj1. (A) The indicated cells expressing N∆Cit1-HA under the control of its own promoter from a low-copy plasmid (CEN/ARS) were grown overnight at 30°C. The next morning, the cell density was adjusted to OD₆₀₀ = 0.2 and cells were further grown for 5–6 h. When OD₆₀₀ reached 0.7–0.8, cycloheximide was added to the medium immediately after the time = 0 samples were collected. Samples were further collected at the indicated time points and subjected to western blotting. Cdc48 served as a loading control. (B) Quantification of the results in (A) is shown. (C) The cycloheximide chase experiment was performed for the indicated strains as in (A) except that cells were shifted to 37°C immediately after cycloheximide was added. (D) Quantification of the results in (C) is shown. (E) The cycloheximide chase experiment was performed for the indicated strains. Cells were grown to mid-log phase overnight at 26°C. The time = 0 samples were collected immediately before cycloheximide was added to the medium and the temeprature was shifted to 37°C. (F) Quantification of the results in (E) is shown. (G) The cycloheximide chase experiment was performed for the indicated strains as in (E). (H) Quantification of the results in (G) is shown. The quantified results are means ± SE of four (A)–(F) and three (G) and (H) independent experiments. Uncropped or substantial portion of blots are presented in Figure S2 (Supporting Information).

To further verify whether Ssa1 is involved in NΔCit1 degradation, we also measured NΔCit1 turnover in ssa1-45 temperature-sensitive mutant cells, in which the SSA2, SSA3, and SSA4 genes are disrupted by gene insertional mutagenesis and the SSA1 gene is replaced with the temperature-sensitive allele ssa1-45 (Becker et al. 1996). Cells expressing NΔCit1 were grown to log phase at the permissive temperature of 26°C and shifted to the restrictive temperature of 37°C immediately after cycloheximide was added to the medium. NΔCit1 was considerably stabilized in ssa1-45 cells but was degraded over time in SSA1 cells (Fig. 1E and F). A similar trend was observed for ydj1-151 temperature-sensitive mutant cells (Caplan et al. 1992) (Fig. 1G and H). These results suggest that the SSA class of Hsp70 chaperones along with the Hsp40 cochaperone Ydj1 are required for NΔCit1 degradation under heat stress conditions.

Inactivation of Ssa1 leads to formation of cytosolic foci of N∆Cit1-GFP at 37°C

We next analyzed the subcellular localization of NΔCit1 in ssa1Δssa2Δ cells. Cells expressing C-terminally GFP-tagged NΔCit1 were grown to log phase at the permissive temperature of 26°C and observed under a fluorescence microscope (time point (a) in Fig. 2A). In addition, to observe the localization of NΔCit1-GFP when its degradation was inhibited, cycloheximide was added and cells were shifted to 37°C for 30 min before the localization of NΔCit1-GFP was observed (time point (b) in Fig. 2A). NΔCit1-GFP was distributed in the cytosol in both wild-type and ssa1Δssa2Δ cells at 26°C (Fig. 2B–D, SSA1SSA2 (a) and ssa1Δssa2Δ (a), respectively). However, while NΔCit1-GFP was distributed in the cytosol in wild-type cells that had been shifted to 37°C (SSA1SSA2 (b)), it formed clear foci in ssa1Δssa2Δ cells that had been shifted to 37°C (ssa1Δssa2Δ (b)). These foci were observed in ∼28.9% (84 of 291 cells) of ssa1Δssa2Δ cells at 37°C, but only in ∼5.2% (15 of 290 cells) of wild-type cells. When the GFP-tagged wild-type Cit1, which contains the N-terminal mitochondrial targeting sequence, was expressed in ssa1Δssa2Δ cells, it did not form foci similar to those formed by N∆Cit1-GFP (Fig. 2E).

Figure 2.

A portion of N∆Cit1-GFP forms foci in ssa1∆ssa2∆ cells at 37°C. (A) Schematic of the preparation of cells for microscopic analysis. Cells were first grown to log phase overnight at 26°C (time point “a”). Thereafter, 100 µg/ml cycloheximide was added and cells were shifted to 37°C for 30 min (time point “b”). Cells at time points (a) and (b) were fixed with formaldehyde and subjected to microscopic analysis. (B) Localization of N∆Cit1-GFP in SSA1SSA2 (wild-type) and ssa1∆ssa2∆ cells. White arrows indicate foci of N∆Cit1. (C) Differential interference contrast (DIC) images of the fields of view observed in (B). (D) Enlarged views of the red boxed areas in (B). (E) Localization of Cit1-GFP in SSA1SSA2 (wild-type) and ssa1∆ssa2∆ cells. Cells were grown at 26°C until the mid-log phase. Where indicated, cells were moved to 37°C for 30 min before they were subjected to the microscopic analysis. Scale bars, 5 µm.

Since the foci signals and the Hoechst 33342 signal did not overlap, this would indicate that they are cytosolic and not nuclear (Fig. 3A). To confirm this finding, we performed biochemical subcellular fractionation analysis. Here, the cells were disrupted with glass beads in the absence of detergent, then, the homogenate was centrifuged at 20 000 g to separate it into the membrane (pellet) and the cytosolic (supernatant) fractions. The membrane fraction supposedly contains most of the mitochondrial and endoplasmic reticulum membranes (Rieder and Emr 2001). A considerable amount of NΔCit1-GFP was fractionated into the supernatant prepared from wild-type cells, which were grown at 26°C or 37°C, in addition to that prepared from ssa1Δssa2Δ cells grown at 26°C (Fig. 3B, lanes 1–6). It should be noted that the amount of NΔCit1-GFP decreased in wild-type cells that had been shifted to 37°C because of its degradation (Fig. 3B, lanes 3–4, SSA1SSA2 (b)). Importantly, the distribution of the NΔCit1-GFP remained unchanged in the pellet/supernatant fractions, even after the ssa1Δssa2Δ cells were moved to 37°C (Fig. 3B, lanes 7–8, ssa1Δssa2Δ (b)). Although we cannot fully exclude the possibility that the foci are loosely associated with membranous organelles, these data support the idea that the observed foci are cytosolic.

Figure 3.

A portion of N∆Cit1-GFP forms insoluble aggregates in ssa1∆ssa2∆ cells under heat stress conditions. (A) Localization of N∆Cit1-GFP in SSA1SSA2 and ssa1∆ssa2∆ cells. Cells were grown as in Fig. 2(A) before they were subjected to the microscopic analysis. Nuclei were stained using Hoechst 33342. DIC images are shown on the right. (B) Subcellular fractionation analysis of SSA1SSA2 (wild-type) and ssa1∆ssa2∆ cells expressing N∆Cit1-GFP. Cells were grown as in Fig. 2(A) before they were subjected to the microscopic analysis. Pgk1 served as a cytosolic maker, while Dpm1 (dolichyl-phosphate mannose synthase 1), which resides in the endoplasmic reticulum, served as a membrane marker. An asterisk indicates nonspecific bands. (C) An aggregation assay was performed of N∆Cit1-GFP in wild-type and ssa1∆ssa2∆ cells at time points (a) and (b) (see Fig. 2A). The whole cell lysate was fractionated by centrifugation at 100 000 g before the pellet (P100) and supernatant (S100) were subjected to western blotting with an anti-GFP antibody. Coomassie Brilliant Blue (CBB) staining of the membrane is shown below. Uncropped or substantial portion of blots are presented in Fig. S3 (Supporting Information).

We next analyzed the detergent solubility of NΔCit1-GFP. A total lysate was prepared from cells grown as in Fig. 2(A) in the presence of 0.5% Triton X-100 and was centrifuged at 100 000 g. A portion of NΔCit1-GFP was fractionated in the pellet fraction (P100) prepared from ssa1Δssa2Δ cells that had been shifted to 37°C (ssa1Δssa2Δ (b)), while most molecular species were fractionated in the supernatant when ssa1Δssa2Δ cells were grown at 26°C (ssa1Δssa2Δ (a)) and when wild-type cells were grown at 26°C (SSA1SSA2 (a)) or shifted to 37°C (SSA1SSA2 (b)) (Fig. 3C). It should be noted that the amount of NΔCit1-GFP decreased in wild-type cells that had been shifted to 37°C because of its degradation (SSA1SSA2 (b)).

We also analyzed the subcellular localization of NΔCit1-GFP in ssa1-45 cells. Cells were prepared for microscopic analysis as in Fig. 4(A). Consistent with the results obtained with ssa1Δssa2Δ cells, ∼39% of ssa1-45 cells contained NΔCit1-GFP foci when they were shifted to 37°C (Fig. 4B–D, ssa1-45 (b)), whereas less than 1% of ssa1-45 cells contained foci at 26°C (ssa1-45 (a)). No foci were observed in SSA1 cells at either temperature (SSA1 (a) and SSA1 (b)). Furthermore, a portion of NΔCit1-GFP formed insoluble aggregates only when ssa1-45 cells were shifted to 37°C (Fig. 4E, ssa1-45 (b)). These results demonstrate that under heat stress conditions, the SSA class of Hsp70 chaperones is required to maintain the solubility of N∆Cit1, which will otherwise form insoluble cytosolic foci.

Figure 4.

A portion of N∆Cit1-GFP forms insoluble foci in ssa1-45 cells at 37°C. (A) A schematic of the preparation of cells for microscopic analysis. Cells were first grown to log phase overnight at 26°C (time point “a”). Thereafter, 100 µg/ml cycloheximide was added and cells were shifted to 37°C for 15 min (time point “b”). Cells at time points (a) and (b) were fixed with formaldehyde and subjected to microscopic analysis. (B) Localization of N∆Cit1-GFP in SSA1 and ssa1-45 cells. White arrows indicate foci of N∆Cit1-GFP. (C) DIC images of the fields of view observed in (B). (D) Enlarged views of the red-boxed areas in (B). (E) An aggregation assay was performed of N∆Cit1-GFP in SSA1 and ssa1-45 cells at time points (a) and (b). The whole cell lysate was fractionated by centrifugation at 100 000 g before the pellet (P100) and supernatant (S100) were subjected to western blotting with an anti-GFP antibody. CBB staining of the membrane is shown below. Scale bars, 5 m. Uncropped or substantial portion of blots are presented in Fig. S4 (Supporting Information).

The degree of N∆Cit1 ubiquitination is unaffected in cells defective in Ssa1 and Ydj1

To investigate the step(s) at which Hsp70/Hsp40 chaperones act during NΔCit1 degradation, we first analyzed the interaction between NΔCit1 and Ucc1 using a pull-down assay (Nakatsukasa et al. 2015, Nishio et al. 2023). An anti-FLAG affinity gel loaded with the recombinant 3xFLAG–Ucc1–Skp1 complex was incubated with yeast lysate prepared from cells expressing NΔCit1-HA. NΔCit1 bound to the 3xFLAG–Ucc1–Skp1 complex (Fig. 5A, lane 4). To investigate if Ssa1 is involved in recognition of NΔCit1 by Ucc1, ssa1-45 cells expressing NΔCit1 were shifted to 37°C before the lysate for the pull-down assay was prepared. However, NΔCit1 from ssa1-45 cells robustly bound to Ucc1 (Fig. 5B). A similar result was obtained even when the lysate was further preincubated at 37°C before it was subjected to the pull-down assay (Fig. 5C). These results suggest that NΔCit1 can bind to Ucc1 even when Ssa1 is inactivated.

Figure 5.

N∆Cit1 ubiquitination is unaltered in cells defective in Ssa1 or Ydj1. (A) The interaction between 3xFLAG–Ucc1–Skp1 and NΔCit1-HA was analyzed by a pull-down assay. Cells (ucc1∆ background) expressing N∆Cit1-HA under the control of its own promoter from a low-copy plasmid (CEN/ARS) (pKN434) were grown to log phase at 30°C. The whole cell lysate was prepared and incubated with anti-FLAG beads that had been loaded with the 3xFLAG–Ucc1–Skp1 complex. Note that 3xFLAG–Ucc1–Skp1 appears as two bands (αFLAG), possibly due to processing in host insect cells (Nakatsukasa et al. 2015). (B) Cells (ssa1-45ucc1∆ and SSA1ucc1∆) expressing N∆Cit1-HA were grown to log phase at 26°C before they were shifted to 37°C for 40 min. The lysate was prepared and subjected to the pull-down assay as in (A). (C) The pull-down assay was performed as in (B) except that the lysate was preincubated at 37°C for 40 min before it was subjected to the assay. (D) An in vivo ubiquitination assay was performed of ssa1-45 and SSA1 cells expressing N∆Cit1-HA or N∆Cit1 (DS/AA)-HA. Cells were first grown at 26°C and shifted to 37°C for 30 min. N∆Cit1-HA or N∆Cit1 (DS/AA)-HA was immunoprecipitated with an anti-HA antibody under denaturing conditions and subjected to western blotting with an anti-HA or anti-Ub antibody. (E) An in vivo ubiquitination assay was performed of ssa1∆ssa2∆ and wild-type cells expressing N∆Cit1-HA. Cells were grown as in (D). (F) An in vivo ubiquitination assay was performed of ydj1-151 and wild-type cells expressing N∆Cit1-HA. Cells were grown as in (D). Asterisks indicate immunoglobulin G (IgG) and/or nonspecific bands. IP, immunoprecipitation; WB, western blotting. Uncropped or substantial portion of blots are presented in Figs S5 and S6 (Supporting Information).

We next analyzed the ubiquitination state of NΔCit1. NΔCit1 was immunoprecipitated from cells under denaturing conditions and the degree of its ubiquitination was assessed by western blotting with an antiubiquitin antibody. To better visualize the ubiquitinated species, ubiquitin was overexpressed under the control of the copper-inducible promoter. Consistent with a previous study, while NΔCit1 ubiquitination was observed in SSA1 wild-type cells (Fig. 5D, lane 6), its stable mutant variant, NΔCit1-DS/AA, which has D173A and S429A mutations and is thus not recognized by Ucc1 (Nishio et al. 2023), was less efficiently ubiquitinated (Fig. 5D, lane 7), suggesting that NΔCit1 ubiquitination in SSA1 cells is Ucc1-dependent. We then analyzed NΔCit1 ubiquitination in ssa1-45 cells. The degree of NΔCit1 ubiquitination was lower in ssa1-45 cells than in SSA1 wild-type cells (Fig. 5D, compare lanes 5 and 6). However, the expression level of NΔCit1 was lower in ssa1-45 cells than in SSA1 wild-type cells under these conditions (Fig. 5D, WB: α-HA, compare lanes 1 and 2). Accordingly, less NΔCit1 was immunoprecipitated from ssa1-45 cells than from SSA1 wild-type cells (Fig. 5D, IP: α-HA, WB: α-HA, compare lanes 1 and 2). While we do not know the exact reason for this phenomenon, this result suggests that the ssa1-45 strain is not appropriate to assess the degree of NΔCit1 ubiquitination. By contrast, NΔCit1 was immunoprecipitated from ssa1Δssa2Δ cells as efficiently as from SSA1SSA2 wild-type cells (Fig. 5E, compare lanes 1 and 2). The degree of NΔCit1 ubiquitination was unaffected in the absence of Ssa1 and Ssa2 (Fig. 5E, compare lanes 4 and 5). A similar finding was made in ydj1-151 cells (Fig. 5F). These results suggest that the SSA class of Hsp70 chaperones and the Hsp40 cochaperone Ydj1 are not essential for SCF^Ucc1-mediated recognition and ubiquitination of NΔCit1 under heat stress conditions.

A cytosolic/peroxisomal citrate synthase, Cit2, is robustly degraded in ssa1∆ssa2∆ cells under heat stress conditions

Cit1 shares ∼75% primary sequence identity and ∼86% similarity with Cit2, a cytosolic/peroxisomal gluconeogenic citrate synthase (Nishio et al. 2023). Moreover, the three-dimensional structures of Cit1 and Cit2 are almost identical, and we previously established that Cit2 is targeted for SCF^Ucc1-mediated proteasomal degradation (Nakatsukasa et al. 2015). We, therefore, investigated whether degradation of Cit2, which localizes to both the cytosol and peroxisomes in glucose-grown cells (Nakatsukasa et al. 2015), is also dependent on Ssa1. To this end, we performed cycloheximide chase analysis of Cit2 in ssa1Δssa2Δ cells. Cells expressing Cit2 under the control of its own promoter were grown at the permissive temperature of 26°C and shifted to 37°C immediately after cycloheximide was added to the medium. Cit2 was robustly degraded in ssa1Δssa2Δ cells (Fig. 6A and B). A similar finding was made when Cit2 was expressed under the control of the GAL1 promoter, although Cit2 was slightly stabilized in ssa1Δssa2Δ cells (Fig. 6C and D). As a control, Cit2 degradation was assessed in ucc1Δ cells and Cit2 was almost completely stabilized (Fig. 6E).

Figure 6.

Cit2 is robustly degraded in ssa1∆ssa2∆ cells at 37°C. (A) Wild-type (BY4741) and ssa1∆ssa2∆ cells expressing Cit2-HA under the control of its own promoter from a low-copy plasmid (CEN/ARS) (pKN194) were grown to log phase overnight at 26°C. The “time = 0” samples were collected and cycloheximide was added to the medium immediately before cells were shifted to 37°C. Samples were further collected at the indicated time points and subjected to western blotting as described in the section “Materials and Methods.” Pgk1 served as a loading control. (B) Quantification of the results in (A) is shown. (C) Cells transformed with pKN545, a CEN/ARS plasmid encoding Cit2-HA under the control of the GAL1 promoter, were first grown in raffinose-containing medium overnight at 26°C to mid-log phase. Expression of Cit2-HA was induced by adding 2% galactose for 1 h. A cycloheximide chase experiment was performed as in (A). Pgk1 served as a loading control. (D) Quantification of the results in (C) is shown. (E) A cycloheximide chase experiment was performed of Cit2-HA in ucc1∆ cells grown at 30°C. Cdc48 served as a loading control. (F) and (G) An aggregation assay was performed of Cit2-HA in wild-type and ssa1∆ssa2∆ cells as in Fig. 3. Cit2-HA expression was induced as in (C). The pellet (P100) and supernatant (S100) were subjected to western blotting with an anti-HA antibody. CBB staining of the membrane is shown below. The quantified results are means ± SE of three independent experiments. Uncropped or substantial portion of blots are presented in Fig. S7 (Supporting Information).

We next analyzed the detergent solubility of Cit2. A total lysate was prepared from cells grown as in Fig. 6(F) in the presence of 0.5% Triton X-100 and centrifuged at 100 000 g. Most Cit2 species fractionated in the supernatant even from total lysate prepared from ssa1Δssa2Δ cells that had been shifted to 37°C (ssa1Δssa2Δ (b)) (Fig. 6G). These results suggest that Cit2 is thermally more robust than N∆Cit1 in the cytosol and that Cit2 degradation is not highly dependent on Ssa1/Ssa2 even under heat stress conditions.

Discussion

In this study, we demonstrated that NΔCit1, a mutant form of mitochondrial citrate synthase that lacks the N-terminal mitochondrial targeting signal, was robustly degraded in ssa1∆ssa2∆ cells under the normal growth condition at 30°C. However, its degradation was inhibited in ssa1∆ssa2∆ cells as well as in ssa1-45 and ydj1-151 cells when cells were shifted to the restrictive temperature of 37°C. These results suggest that Hsp70/Hsp40 are not essential for NΔCit1 degradation under normal conditions, but are essential for its degradation under heat stress conditions. How do Hsp70/Hsp40 facilitate NΔCit1 degradation under heat stress conditions?

Hsp70/Hsp40 chaperones are generally assumed to prevent aggregation of misfolded substrates and facilitate recognition by the ubiquitination machinery. They may also act as substrate-recognition factors and serve as intermediaries between the substrates and ubiquitination machinery (Balchin et al. 2016). Based on these assumptions, we initially hypothesized that Hsp70/Hsp40 chaperones act before NΔCit1 ubiquitination and prevent its aggregation under heat stress conditions to facilitate its recognition and ubiquitination by Ucc1. However, although a portion of NΔCit1 formed insoluble aggregates in ssa1∆ssa2∆ and ssa1-45 cells that were shifted to 37°C, Ucc1-mediated recognition and ubiquitination of NΔCit1 were robust in cells lacking Ssa1/Ssa2 or Ydj1 activities. One explanation is that these chaperones increase the accessibility of NΔCit1 to the proteasome after ubiquitination. In this case, the steady-state level of NΔCit1 ubiquitination was expected to increase in cells defective in Hsp70/Hsp40. However, it was almost unaffected, although we cannot exclude the possibility that there was a concomitant increase in the activity of a deubiquitination enzyme or in the amount of aggregated protein. Nonetheless, the current simplest model is that Hsp70/Hsp40 maintain the solubility of NΔCit1 after its ubiquitination.

Based on this scenario, we suspect that cytosolic foci formed by NΔCit1 in ssa1∆ssa2∆ and ssa1-45 cells under heat stress conditions contain ubiquitinated species of NΔCit1, and that ubiquitinated NΔCit1 accumulates in the pellet fraction in the aggregation assay. However, there are several technical difficulties that hamper the point of action of Hsp70/Hsp40 during NΔCit1 degradation being fully defined. First, the pellet obtained following ultracentrifugation in the aggregation assay was heavily aggregated and thus high concentrations of SDS and a reducing reagent were needed to solubilize it. Therefore, it was difficult to use this fraction in the immunoprecipitation experiment to detect ubiquitinated species. Second, we failed to detect binding of Ssa1 to NΔCit1 by the regular coimmunoprecipitation experiment, possibly due to their weak interaction (Fig. S1, Supporting Information). Third, the ssa1-45 and ssa1∆ssa2∆ strains each have their own characteristics. For example, in the ssa1-45 strain, the SSA2, SSA3, and SSA4 genes are inactivated by insertional mutagenesis and N-terminal fragments of each protein may be expressed (Craig and Jacobsen 1984, Werner-Washburne et al. 1987). Expression of these fragments may be upregulated at the elevated temperature (Werner-Washburne et al. 1989, Gasch et al. 2000). On the contrary, the ssa1∆ssa2∆ strain survives by inducing expression of SSA3 and SSA4 but is nonetheless slow-growing and temperature-sensitive (Craig and Jacobsen 1984, Werner-Washburne et al. 1987). Furthermore, ubiquitin-system genes including UBI4 and UBC4 are transcriptionally upregulated in the ssa1∆ssa2∆ strain upon the shift to high temperature (Baxter and Craig 1998). In our cycloheximide chase experiment, cells were shifted from the permissive to the restrictive temperature immediately after cycloheximide was added. Therefore, the effects of upregulation of SSA fragments in the ssa1-45 strain and of ubiquitin-system genes in the ssa1∆ssa2∆ strain are likely limited because their translation was inhibited by cycloheximide, although we cannot exclude possible basal effects at the permissive temperature. To fully define the direct role of Hsp70/Hsp40 in NΔCit1 degradation, in addition to conventional biochemical and genetic approaches, we suggest that real-time in vivo assays to monitor binding of NΔCit1 to chaperones and its partitioning into cytosolic foci are important.

An intriguing finding was that Cit2 degradation was less dependent on Hsp70/Hsp40 chaperones. A previous report suggested that purified Cit1 and Cit2 have different physical properties despite their high sequence similarity (Kispal and Srere 1991). For example, Cit2 activity is more strongly inhibited by NADH, ADP, and AMP than Cit1 activity. In addition, Cit2 loses its activity when stored without glycerol, incubated at pH 8.1, or heated at 40°C. By contrast, Cit1 is much more stable at pH 8.1 and loses its activity much more slowly at 40°C than Cit2, suggesting that Cit2 is a more labile enzyme than Cit1 in vitro (Kispal and Srere 1991). Therefore, it could be argued that Cit2 may be more thermally vulnerable than Cit1 and thus chaperones might be needed more for Cit2 degradation to prevent its aggregation than for Cit1 degradation. However, the opposite was true; Cit2 degradation was not highly dependent on Ssa1/Ssa2, whereas NΔCit1 degradation required these chaperones. We do not know the exact reason for this inconsistency, but it should be noted that these previous analyses were performed with purified enzymes in vitro. In cells, the environment in the cytosol differs from that in the mitochondrial matrix and lacks suitable maturation factors that are likely required for matrix proteins. The cytosolic enzyme fructose-1,6-bisphosphatase (Fbp1) binds to Ssa1 and its degradation is reportedly inhibited in ssa1-45 cells when they are shifted from a respiration to a fermentative carbon source (Juretschke et al. 2010). Therefore, we assume that the involvement of chaperones in degradation of metabolic enzymes is dependent on various factors including but not limited to their physical properties, localizations, and expression levels, and the cellular metabolic status. For NΔCit1, chaperones are required for its degradation, most likely because it is prone to aggregate at high temperatures in vivo.

Although we focused on NΔCit1 in this study, it is important to elucidate the role of Hsp70/Hsp40 and other factors in preCit1 degradation upon mitochondrial import failure. Since the accumulation of excess preCit1 is toxic to the cell, it undergoes degradation by the proteasome (Nishio et al. 2023). However, it is reasonable to hypothesize that a mild accumulation of preCit1, which is supposed to exhibit a catalytic activity (Nishio et al. 2023), could function as a backup system to maintain the flux in the TCA cycle, even if the flux bypasses the cytosol, under the condition where the mitochondrial import is weakly inhibited. This line of study would also be valuable for the field of fermentation and metabolic engineering of microorganisms. During the industrial fermentation process, microorganisms are often exposed to severe stresses such as high and low temperature, freezing, desiccation, chemicals, osmotic pressure, high and low pH, limited nutrition, and accumulation of misfolded proteins. Therefore, microorganisms including yeast must appropriately respond and adapt to changes in the environment for their survival. A previous study suggested a possible physiological role of the cell cycle in regulation of mitochondrial protein import (Harbauer et al. 2014). It is important to analyze how industrial fermentation stress might affect mitochondrial import and lead to accumulation of precursor proteins. Understanding the quality control of mitochondrial proteins during the fermentation process will help to develop an industrially beneficial microorganism.

Acknowledgments

We thank Kazuya Nishio for valuable discussion, the National BioResource Project for plasmids, T. Kamura for yeast strains, T. Endo for anti-Cdc48 antibodies, S. Michaelis for plasmids, and J.L. Brodsky for anti-Ssa1 antibodies. We acknowledge the assistance of the Research Equipment Sharing Center at Nagoya City University.

Conflict of interest

The authors declare no conflict of interest.

Funding

This work was supported by the Japan Society for the Promotion of Science KAKENHI grants (19H02923 and 23H02176to K.N.), the Japan Science and Technology Agency SPRING grant (JPMJSP2130to T.K.), the DENTSU Scholarship Foundation to M.H., the Toray Science Foundation to K.N., the Institute for Fermentation, Osaka to K.N., and the Toyoaki Scholarship Foundation to K.N.