Yeast and plasmids methods

Yeast cultures were grown at 30 °C in standard growth medium as previously described [37]. Plasmids were introduced into yeast using the lithium acetate transformation method [38]. A complete list of yeast strains and plasmids used in this study is provided in S1 and S2 Tables, respectively. Gene deletions and epitope tagging were performed using PCR-based homologous recombination as described [39]. Plasmids pRH469, pRH1377, pRH2058, and pRH2997 were generously provided by the Randy Hampton lab. Plasmids pRH1377, pRH2058, or pRH2997 were directly transformed into the indicated yeast strains. Plasmid pRH469 was linearized with StuI and integrated into the URA3 locus of the specified strains. ER localization was monitored using Pho88-GFP, a GFP-tagged ER membrane protein [40].

For inositol depletion experiments, yeast strains were initially grown on inositol-containing plates. Cells were then cultured overnight in inositol-free medium (prepared using yeast nitrogen base without amino acids and inositol) to reach mid-exponential phase. Cultures were freshly diluted into inositol-free medium and incubated for an additional 4–5 hours prior to fluorescence imaging or cycloheximide chase assays. In control experiments, 100 μM inositol was added to the inositol-free medium.

Fatty acid–supplemented growth media were prepared as previously described [41]. Briefly, standard SC medium was modified to contain 0.1% glucose and 1.5% Brij L23 (Sigma-Aldrich), and supplemented with 16 mM fatty acids (oleic acid or linoleic acid; Sigma-Aldrich). Control media were prepared identically but without the addition of fatty acids.

Growth test

Yeast growth was assessed using a serial dilution spot assay. Yeast strains were cultured overnight in standard growth medium at 30°C with shaking until they reached mid-log phase. Cultures were then adjusted to an OD₆₀₀ of 1.0, and a series of 5-fold serial dilutions were prepared in sterile water. Five microliters of each dilution were spotted onto appropriate solid agar plates using a multichannel pipette. Plates were incubated at the indicated temperatures and time points, and growth was documented by imaging. Representative images were taken after the appropriate incubation period to compare growth differences between strains.

Fluorescence microscopy imaging

Fluorescence microscopy was performed using a Zeiss Axiovert 200M fluorescence microscope (Carl Zeiss MicroImaging) equipped with standard Zeiss filter sets for DAPI, FITC/GFP and TRITC/RFP. Yeast cells were grown to mid-log phase in the indicated media, harvested by centrifugation, and resuspended in phosphate-buffered saline (PBS) prior to imaging. Live-cell imaging was performed using a 100 × 1.3 NA oil immersion objective, and images were captured with a monochrome digital camera (Axiocam; Carl Zeiss MicroImaging). For each field, z-stacks spanning 5 µm were collected at 0.2 µm step intervals. Following image acquisition, the stacks were deconvolved and projected to generate 2D images for subsequent quantitation. Exposure times and imaging settings were kept constant across all samples to ensure comparability. Image acquisition, deconvolution, and analysis were carried out using Zeiss Zen software.

To quantify cortical ER (cER) inheritance, more than 100 budded cells were analyzed per condition. Cells were classified into three categories based on bud index (the ratio of bud size to mother cell diameter), and the presence or absence of cER in the bud was scored, as previously described [42].

Nuclear DNA was stained with DAPI to visualize nuclear position and facilitate identification of perinuclear ER localization in live cells. Cells were incubated with 1 µg/mL DAPI in PBS for 5–10 minutes at room temperature, followed by a PBS wash prior to imaging.

IET tubule measurement protocols

To quantify initial ER tubules (IET)—a recently introduced term describing nascent ER tubules that emerge from the perinuclear ER and extend into the bud [43]—fluorescence microscopy images were analyzed using ImageJ software. Yeast cells expressing fluorescently tagged ER markers were grown to mid-log phase in appropriate media and imaged using a Zeiss fluorescence microscope equipped with a 100 × oil immersion objective. Images were acquired under identical exposure settings to ensure consistency across samples.

For image analysis, raw fluorescence images were opened in ImageJ, and background subtraction was performed using the “Subtract Background” function. Tubules were manually traced using the “Freehand Line” tool or the “Segmentation” tool for more precise edge detection. The “Measure” function was then used to determine tubule length in pixels, which was converted to micrometers based on the microscope’s calibration settings. At least 50 tubules per condition were measured from multiple fields of view. To quantify tubule fluorescence intensity, tubules were selected as regions of interest (ROIs) using the “Freehand Selection” or “Polygon Selection” tool, ensuring that the entire tubule was encompassed. The “Measure” function was then used to determine the mean fluorescence intensity and area of the ROI. Background fluorescence was subtracted by measuring an adjacent non-fluorescent region. At least 50 tubules per condition were analyzed across multiple fields of view. Statistical analysis was conducted using Microsoft Excel, with data analyzed using a student’s t-test. Results are presented as mean ± standard error of the mean (SEM).

UPRE-GFP assay and quantitation

The unfolded protein response element (UPRE)-GFP reporter assay was used to assess UPR activation. Yeast cells carrying the UPRE-GFP reporter plasmid were grown to mid-log phase in selective media at 30°C. Cells were then either left untreated or subjected to ER stress by treatment with 1ug/ml tunicamycin for 90mins. Following treatment, cells were harvested by centrifugation, washed with phosphate-buffered saline (PBS), and resuspended in PBS for fluorescence microscopy imaging.

Fluorescence images were captured using a Zeiss fluorescence microscope with a 100 × oil immersion objective. To quantify GFP intensity, images were analyzed using ImageJ software. Regions of interest (ROIs) were manually selected to measure fluorescence intensity within individual cells. Background fluorescence was subtracted, and mean fluorescence intensity (MFI) was calculated for each condition. Data were normalized to cell count and analyzed for statistical significance using Excel t-test.

Cycloheximide chase degradation assay

The degradation of epitope-tagged proteins was assessed using a cycloheximide chase assay as described [44]. Briefly, yeast cultures grown to the logarithmic phase were treated with 100 μg/ml cycloheximide to inhibit protein synthesis. Samples were then collected at various time points post-treatment, lysed, and subjected to immunoblotting to evaluate protein degradation, following the protocol outlined by [44]. GFP-tagged and HA-tagged proteins were detected using anti-GFP (Roche, catalogue number 11814460001) and anti-HA monoclonal antibodies (16B12, Covance, catalog number MMS-101P-500). Pgk1 is used as loading control and anti-Pgk1 is provided from Invitrogen (catalog number 459250).

ER lipid saturation via OLE1 regulates cortical ER inheritance at late stages

To investigate whether alterations in ER lipid saturation influence the unfolded protein response (UPR), the ER surveillance (ERSU) pathway, and ER-associated degradation (ERAD), we employed a previously established system in Saccharomyces cerevisiae that modulates lipid saturation through promoter variants controlling OLE1 expression [36] (Figs 1A and S1A). OLE1 encodes the sole fatty acid desaturase in yeast, introducing double bonds into saturated fatty acids (FA) which results in changes of the membrane fluidity. Gradual reduction of OLE1 expression was achieved by promoter mutagenesis, generating four yeast strains (SFA1-SFA4) with progressively increased membrane phospholipid (PL) saturation and a decreasing phosphatidylethanolamine (PE)/phosphatidylcholine (PC) ratio. Mass spectrometry analysis from the previous study reported that SFA1 featured a PL composition similar to WT cells, which was followed by SFA2 and SFA3 with moderate increases, and finally, SFA4 displayed the most pronounced increase in saturated FA, which was characterized by a marked rise in fully saturated phospholipid species and shortened acyl chain length (decreasing from ~C16–C18 species toward shorter chains) [36]. In addition, the PE/PC ratio decreased across the SFA strains, reflecting broader remodeling of ER membrane lipid composition [36].

Download:

• PNG
  larger image
• TIFF
  original image

Fig 1. Lipid Saturation level has effects on ER inheritance.

(A–B) Analysis of ER lipid saturation levels in SFA1–SFA4 strains. (A) Representative fluorescence images showing localization of the ER LipSat sensor in wild type (MNY3414, W303 background) and SFA1-SFA4 strains. Cells were grown in standard SC medium containing 2% glucose. A grayscale bar above the images illustrates gradual changes in levels of saturated fatty acids. Blue arrows mark membrane-associated ER LipSat sensor, while red arrowheads indicate nucleoplasmic localization. (B) Quantification of membrane-bound versus nucleoplasmic ER LipSat sensor localization. Fluorescence distribution was measured from >100 cells per strain. Scale bar, 5 μm. (C) The ER LipSat sensors enable assessment of fatty acid (FA) saturation in the ER membrane. Mga2, an ER membrane localized transcription factor, undergoes proteolytic cleavage when membrane saturation increases, allowing its N-terminal portion to translocate into the nucleus to activate transcription. Therefore, the relative distribution of nuclear versus ER-localized Mga2 serves as a readout for ER membrane saturation. Without elevated FA saturation, ER LipSat sensor stays on the ER Membrane. Diagram was adapted from Romanauska et al., 2021 [41]. (D) Cortical ER (cER) inheritance defects in SFA4. ER was visualized with a well-established ER marker Pho88-GFP. Representative Live-cell images of wild-type and SFA1–SFA4 strains expressing Pho88-GFP are shown. For comparison, wild-type cells treated with 1 μg/mL tunicamycin (Tm) for 90 min were included to induce ER stress. Yellow arrowheads indicate normal cER in daughter cells, while white arrows point to defective cER. Cells were classified into categories I–III as previously described [29]. Scale bar, 2 μm. (E)-(F) Quantification of the cER inheritance in class I, II and III for WT (E) and SFA1-SFA4 cells (F). Over 100 cells per class were analyzed, with data representing the average of three independent experiments. Error bars denote standard deviation (SD).

https://doi.org/10.1371/journal.pone.0345910.g001

To visualize lipid saturation levels within different ER sub-regions, particularly the inner nuclear membrane (INM) and perinuclear ER (pnER), we applied Mga2-GFP–based lipid saturation (LipSat) reporters based on Mga2-GFP [41]. Mga2 is an ER-resident transcription factor whose GFP-tagged form localizes to both cortical ER (cER) and pnER [13] (S1B and S1C Fig). Under elevated lipid saturation, Mga2 undergoes proteolytic cleavage within its transmembrane domain, releasing a nuclear-targeted fragment that accumulates in the nucleoplasm [45,46]. Conversely, under reduced saturation, Mga2 remains membrane-bound without cleavage. Quantification of GFP signals in the nucleoplasm versus ER membranes (cER/pnER) thus provides a rapid readout of the relative distribution of saturated versus unsaturated lipids in ER membrane or INM. In wild-type control cells, ~ 25% of ER LipSat sensor GFP fluorescence localized to the membrane fraction, while ~75% appeared in the nucleus (S1D Fig). Supplementing cultures with oleic acid (C18:1), which moderately reduces lipid saturation, led to a slight decrease in Mga2 cleavage and nuclear GFP accumulation. Linoleic acid (C18:2), which more strongly reduces saturation, further decreased Mga2 cleavage and favored membrane localization of the sensor (S1D Fig). Comparable patterns were observed for the INM LipSat sensor GFP fluorescence (S1E Fig). These outcomes mirrored prior studies and validated the reporter’s performance in our system [41].

Applying this ER LipSat reporter to the SFA strains, we observed that SFA1 behaved similarly to wild type (~60% membrane bound, ~ 40% nucleoplasmic), while SFA2, SFA3, and particularly SFA4 showed progressively higher Mga2 nuclear signals, demonstrating elevated lipid saturation (Fig 1A and 1B). We next asked how these lipid property changes influenced cortical ER (cER) inheritance, a process we assayed using Pho88-GFP, an integrated ER reporter [29]. As previously described, ER inheritance proceeds in a sequential manner in budding yeast [17,47]: during early cell cycle stages (class I), an initial ER tubule (IET) extends from the pnER toward the bud neck, enters the daughter cell, and anchors at the polarisome at the bud tip. This is followed by lateral spreading along the plasma membrane to establish the cortical ER beneath the daughter cortex (class II–III cells). Consistent with earlier reports, tunicamycin-induced ER stress abolished cER inheritance in many class I cells (Fig 1C and 1D). When cells encounter stress much later (class III), cells completed division, but cER inheritance was blocked in the next cycle [29].

In the SFA background, cER inheritance defects correlated with the degree of lipid saturation (Fig 1C, and 1E). SFA1 and SFA2 resembled wild-type. In contrast, SFA3 class I cells, and SFA4 class I–II cells, displayed reduced cER inheritance, indicating that early stages of inheritance are particularly sensitive to increased lipid saturation. In SFA4, the reduction was more noticeable in class II cells, although still less severe than the inheritance blocks triggered by tunicamycin (Figs 1D, 1E, and S2A). SFA3 and SFA4 thus exhibited unique inheritance profiles, with SFA4 showing the clearest impairment, especially in class II daughters, distinguishing it from the relatively intact inheritance seen in WT, SFA1, and SFA2.

To investigate whether early steps of ER inheritance were disrupted in SFA4, we analyzed IET initiation and orientation in synchronized class I or II cells. Cells were arrested at G2/M with hydroxyurea and released into G1, capturing populations enriched for incipient bud formation (Fig 2A and 2B). Using calcofluor white (CFW) staining to visualize bud scars, we monitored IET orientation relative to the bud neck. Operationally, IETs were defined as tubular extensions emerging from the pnER toward the bud. Quantitative imaging revealed no significant differences between WT and SFA4 in IET frequency, timing, length, or intensity, nor in the proportion of tubules reaching into the bud (Fig 2B–2F). Thus, early ER targeting events remained largely intact in SFA4.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 2. Effects of Lipid Saturation level on different stages of cortical ER (cER) inheritance.

(A) Experimental diagram for hydroxyurea (HU) synchronization release into cell cycle. Yeast cells were synchronized in S-phase by treatment with 100 mM hydroxyurea (HU) for 2 hours. After HU treatment, cells were washed to remove HU and allowed to recover in fresh medium. Samples were collected at 20, 50, and 80 min post-wash to monitor progression through the cell cycle. Imaging for initial ER tubules (IET) was performed at 80 minutes after HU release. (B) The formation of initial ER tubules (IET) and their entry into the new bud in early stages of cER inheritance are not affected in SFA4 cells. Calcofluor White (CFW) staining marks cytokinetic remnants (CRMs). Three representative cells are shown for each indicated strain, with images from the GFP channel, merged GFP and DAPI channels, and DIC. Arrowheads indicate an IET extending from the perinuclear ER (pnER) in the mother cell toward the newly forming bud. Scale bar, 5 µm. (C–D) Quantification of IET tubule length and fluorescence intensity across different strains. Measurements were performed using ImageJ as described in Materials and Methods. Over 50 cells were analyzed per strain. (E) Representative images showing initial ER tubules (IET) extending into the bud at different stages of cER inheritance in wild-type and SFA4 cells. Yellow arrowheads indicate cells with an IET extending a half way into the daughter cell but not reach to the bud tip, while white arrows highlight IET extension reaching fully to bud tip. Scale bar, 5 µm. (F) Quantification of cells exhibiting different ER tubule positions. Depending on the inheritance stage, ER tubules either dynamically extend to the bud tip or fail to do so. Over 50 cells were analyzed per strain. (G) cER in large buds during the late stages of cell cycle does not extend all the way through the cortex of the daughter cell in SFA4 cells. Yellow arrowheads indicate normal cER distribution in wild-type cells, while white arrows highlight ER accumulation at the bud tip and a lack of cER around the bud cortex in SFA4 cells. Scale bar, 2 µm. (H) Quantification of cER accumulation at the bud tip in large-budded wild-type and SFA4 cells. Over 50 cells were analyzed per strain.

https://doi.org/10.1371/journal.pone.0345910.g002

Striking defects emerged instead in class II cells. Compared to ~30% of WT daughters, nearly 80% of SFA4 class II buds accumulated ER markers abnormally at the bud tip, with little or no cortical ER along the bud cortex (Fig 2G and 2H). Although these inheritance defects were milder than those observed under acute ER stress, SFA4 uniquely exhibited stronger impairment in class II than in class I or III cells. Taken together, these findings indicate that excess ER lipid saturation does not disturb the earliest targeting events of ER inheritance but selectively disrupts later stages, particularly the maturation and distribution of cortical ER in class II daughter cells.

Differential eeffects of ER lipid saturation on ERAD substrate degradation

Recognizing the essential role of phospholipid homeostasis in ER function and cellular stress adaptation, we next explored how the saturation state of ER membrane lipids modulates the efficiency of ER-associated degradation (ERAD). Specifically, we quantified the half-lives of distinct ERAD substrates by monitoring their steady-state protein levels over time following cycloheximide (CHX)–mediated inhibition of protein synthesis—a well-established method to determine protein degradation dynamics (Fig 3). By arresting new protein synthesis with CHX, we could precisely track the decrease in substrate abundance as a direct readout of degradation, allowing robust determination of protein half-life and stability under varying lipid saturation conditions. Using this cycloheximide chase assay, we probed the fate of three well-established ERAD substrates: Hmg2-GFP—an integral membrane protein processed via the ERAD-M pathway (mutation in the membrane domain), CPY*-HA—a mutated soluble luminal protein (G255R in carboxypeptidase Y) cleared by the ERAD-L pathway (luminal domain misfolding), and Ste6–166-3xHA-GFP (Ste6*-HA-GFP)—a mutant membrane protein with a misfolded cytosolic domain, handled by the ERAD-C pathway (cytosolic domain misfolding) (Fig 3A–3F). Under these conditions, Hmg2-GFP degradation was modestly but reproducibly accelerated in both SFA1 and SFA4 mutants compared with wild-type cells, with the effect more pronounced in SFA1 and less evident in SFA4 (Fig 3A and 3B). The degradation of Ste6*-HA-GFP was also enhanced in both mutants but was most pronounced in SFA4 (Fig 3C and 3D). In contrast, the clearance of CPY*-HA remained essentially unchanged across wild-type, SFA1, and SFA4 backgrounds (Fig 3E and 3F).

Download:

• PNG
  larger image
• TIFF
  original image

Fig 3. Effects of Lipid Saturation level on ERAD.

(A)-(E) A cycloheximide (CHX) chase of Hmg2-GFP (A-B), CPY*-HA (C and D), and Ste6*-HA-GFP (E and F) expressed in Wild-type, lipid saturation mutants, SFA1 and SFA4. Upon CHX treatment, equal numbers of cells were collected to prepare total cell extracts for SDS-PAGE, followed by western blot analysis to detect ERAD membrane (ERAD-M) substrate Hmg2-GFP, ERAD luminal substrate (ERAD-L) CPY*-HA, and ERAD cytosolic substrate (ERAD-C), Ste6*-HA-GFP. Pgk1 was used as a loading control. The mean percentage of each ERAD substrate remaining in the cells for at least three biological replicates is plotted. Error bars represent the standard deviation (SD). A one-tailed t-test was used to determine the significance of the difference between WT and SFA4 cells or between WT and SFA1 cells. p < 0.05, *; p < 0.005, **; p < 0.001, ***. (G) Schematic summary of ERAD substrate degradation in SFA1 or SFA4 yeast strains. In SFA1 yeast cells, membrane lipids display wild-type like saturation with slightly shortened acyl chain length. In contrast, SFA4 cells exhibit both markedly reduced acyl chain length and increased saturation of lipid membrane. These changes in membrane composition selectively enhance the degradation of ERAD-C and ERAD-M substrates–Ste6*-HA-GFP and Hmg2-GFP, respectively– while having no effect on the degradation of the ERAD-L substrate CPY*-3xHA. ERAD-C and ERAD-M substrates are recognized by the Doa10 and Hrd1 ubiquitin ligase complexes, respectively, and extracted by the Cdc48-Ubx2 complex. ERAD-L substrate degradation depends on the Hrd1-Hrd3 complex and the ER luminal chaperones Kar2 and Yos9.

https://doi.org/10.1371/journal.pone.0345910.g003

These results reveal a specificity: membrane lipid saturation may selectively accelerate the turnover of ERAD substrates with transmembrane or cytosolic misfolded domains, while leaving the degradation rate of soluble luminal proteins largely unaltered. Notably, SFA1 cells—enriched in unsaturated lipids— had a particularly strong impact on the stability of Hmg2-GFP, whereas SFA4 cells—with increased membrane saturation—most significantly reduced the half-life of Ste6*-HA-GFP (Fig 3G). Together, our findings point to potentially distinct and substrate-specific influences of ER membrane lipid saturation on ERAD efficiency, suggesting the lipid saturation change might act as a modulator of protein quality control within the ER.

Lipid saturation levels of the ER membrane drive distinct UPR activation dynamics

To determine whether increased ER membrane lipid saturation influences UPR activity, we monitored UPR induction using a UPRE-GFP reporter in wild-type (WT), SFA1, and SFA4 yeast cells. Cells were transformed with a well-established reporter plasmid expressing GFP under the control of a UPR transcriptional element (UPRE), and GFP fluorescence was quantified using ImageJ. As expected, treatment with tunicamycin (Tm; 1 µg/mL for 1.5 hours) led to an approximately two-fold increase in GFP fluorescence in WT cells, consistent with previous findings [22]. SFA1 cells treated with Tm exhibited GFP levels similar to those of Tm-treated WT cells (Fig 4A and 4B).

Download:

• PNG
  larger image
• TIFF
  original image

Fig 4. Effects of lipid saturation level on UPR pathways.

(A) Expression of GFP in yeast cells carrying the UPRE-GFP reporter, where the unfolded protein response element (UPRE) was fused to GFP and expression of GFP reports levels of UPR activation by ER stress. Wild-type (WT), SFA1 or SFA4 yeast cells harboring the UPRE-GFP plasmid were incubated with control DMSO or 1 µg/ml Tm for 90 minutes for examining the ER stress activation levels. Representative images for each condition are shown. Scale bar, 5 µm. (B) Quantitating over 300 cells carrying UPRE-GFP, in either WT, SFA1 or SFA4 cells, with or without Tm treatment. GFP expression in SFA4 cells even prior to Tm treatment shows that SFA4 cells induce ER stress and activate UPR. Mean fluorescence intensity with standard error of the mean is shown. A one-tailed t-test was used to determine the significance of the difference between WT and SFA4 cells or between SFA4 cells with and without Tm treatment. p < 0.05, *; p < 0.01, **; p < 0.002, ***. (C) Growth phenotypes of SFA1 - SFA4 yeast strains with or without Tm. Serial dilutions of overnight cultures from the indicated strains were plated on YPD plates or YPD plates containing 0.6 µg/ml Tm and grown at 30°C for 2–3 days. The red box highlights the improved growth of the SFA4 strain on YPD plates containing 0.6 µg/ml Tm. (D) cER inheritance in SFA1 and SFA4 cells upon treatment with Tm, an ER stress inducer. SFA1 or SFA4 strains expressing Pho88-GFP were treated with control DMSO or Tm for 90 minutes and then examined by fluorescence microscopy. Representative images of GFP or overlay of DIC with GFP of class I, II, and III cells are shown. Yellow arrowheads indicate normal cER in the bud, and white arrows point to the discontinuous cER under the cortex of the daughter cell. Scale bars, 5 µm. (E) Quantification of the cER inheritance of the cell shown in (D). For each condition, over 100 cells were analyzed. Results represent analyses from three independent experiments. Error bars denote standard deviation (SD). A one-tailed t-test was used to determine the significance of the difference between SFA1 and SFA4 cells with and without Tm treatment, p < 0.01, **; p < 0.05, *; ns, not significant.

https://doi.org/10.1371/journal.pone.0345910.g004

Remarkably, SFA4 cells displayed elevated GFP fluorescence even without Tm treatment, reaching levels comparable to those observed in Tm-treated WT or SFA1 cells. This strong basal activation indicates that increased ER membrane lipid saturation in SFA4 is sufficient to induce UPR activation independently of external ER stressors. Upon Tm treatment, GFP levels in SFA4 cells increased further, demonstrating that the UPR machinery remains responsive to additional ER stress (Fig 4A and 4B). These results support the idea that ER lipid saturation perturbs ER homeostasis and activates the UPR independently of unfolded protein accumulation, likely through direct sensing of membrane lipid composition by UPR transducers Ire1 in yeast. To corroborate the findings from the UPRE-GFP reporter assay, we evaluated the growth phenotypes of WT and SFA strains under ER stress conditions (Fig 4C). Under unstressed conditions, SFA1, SFA2, and SFA3 strains exhibited growth comparable to WT. In contrast, SFA4 cells showed slower growth, consistent with a basal stress state. Upon ER stress induction with tunicamycin (Tm), all strains except SFA4 displayed a growth defect. Interestingly, SFA4 cells exhibited improved growth in the presence of Tm relative to their untreated condition, and even outperformed other SFA strains under ER stress. These results suggest that Tm-induced proteotoxic stress activates the UPR in a way that may partially compensate for the lipotoxic defects in SFA4 cells.

To further investigate the unexpected behavior of SFA4, we investigated if/how ER stress alters the cortical ER (cER) inheritance when compared to that observed in unstressed SFA4 cells. In all classes of cells for SFA4, induction of ER stress with Tm restored cER inheritance, enabling all cells to properly inherit cER despite significant alterations in overall ER morphology (Fig 4D, 4E, and S2B). Notably, unstressed SFA4 class II cells exhibited a pronounced decrease in cER inheritance (Fig 4E, class II, gray bar). However, upon induction of ER stress with Tm, the reduced cER inheritance in SFA4 class II cells was restored to nearly normal levels (Figs 4D, 4E, and S2B). This was in contrast to SFA1 cells; when we assessed cER inheritance across all three classes with ER stress, cER inheritance was blocked in class I, II, and III buds in a manner similar to that observed in wild-type cells (Fig 4D and 4E). These findings suggest that elevated lipid saturation in SFA4 impacts ER inheritance efficiency, while ER stress-induced changes in lipid composition could facilitate improved cER inheritance. Alternatively, ER stress may alter the biophysical characteristics of yeast ER lipids in a way that enhances the effectiveness of cER inheritance.

Impact of inositol depletion on cortical ER inheritance and ERAD substrate degradation

Outcomes of above studies have motivated us to test other types of lipid changes on ER inheritance. Inositol is a vital precursor for the synthesis of phosphatidylinositol (PI) in yeast. And lipid saturation levels modulate how inositol or inositol-containing lipids are incorporated and/or function in membranes [48]. Furthermore, previous research has demonstrated that inositol depletion activates the unfolded protein response (UPR) and leads to significant alterations in cellular lipid composition [49,50]. To investigate whether membrane alterations caused by inositol depletion affect ER quality control, we employed LipSat sensors (S1B and S1C Fig) to assess ER or INM membrane lipid saturation under inositol-depleted conditions (Fig 5A and 5B and S3A-S3B). Our analysis revealed that inositol deprivation does not cause major changes to lipid saturation levels at the ER or INM, suggesting that the observed effects on the ER homeostasis are not due to altered membrane saturation.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 5. Inositol depletion affects cER inheritance and ERAD substrate degradation.

(A) ER membrane lipid saturation levels in response to inositol depletion. Live cell imaging of wild-type cells expressing the plasmid-based ER LipSat sensor under conditions with or without 100 µM inositol supplementation. Red arrowheads indicate nucleoplasmic localization, and white arrows point to nuclear envelope localization. Scale bar, 5 µm. (B) Localization of the ER LipSat sensor was quantitated to assess lipid saturation levels. Sensor distribution was classified into two phenotypes: membrane-bound (indicative of lipid unsaturation level) or nucleoplasmic (indicative of lipid saturation level). Data are presented as mean values and standard deviations. A one-tailed t-test shows no significant difference (ns, p > 0.1) for ER lipid saturation between the two growth conditions. (C) Inositol depletion leads to defects in cER inheritance. Wild-type cells expressing the ER membrane reporter, Pho88-GFP, were analyzed by fluorescence microscopy under normal or inositol-depleted conditions. Representative images show abnormal cER inheritance, with yellow arrowheads indicating normal cER formation in the bud and white arrows indicating defective cER. Scale bar, 5 µm. (D) Quantification of cER inheritance defects. More than 100 cells were analyzed under conditions of inositol supplementation (+Inositol) or depletion (-Inositol). Data represent the average of three independent experiments, with error bars indicating standard deviation (SD). Statistical significance of cER inheritance differences in wild-type cells grown with or without inositol was assessed using a one-tailed t-test. p < 0.002, ***; p < 0.1, *; ns, not significant. (E) Cycloheximide (CHX) chase degradation assay was performed for ERAD-M substrate, Hmg2-GFP, and analyzed by quantitating remaining Hmg2-GFP at each time point following the release from cycloheximide chase. (F) Quantification of ERAD-M (Hmg2-GFP) levels shows the percent of Hmg2-GFP remaining at each time point from the cycloheximide chase assay in panel (E). (G)-(H) Cycloheximide chase degradation assay was performed for ERAD-L substrate, CPY*-3xHA and analyzed by quantitating remaining CPY*-3xHA at each time point following the release from cycloheximide chase. (I)-(J) Cycloheximide chase degradation assay for ERAD-C substrate, Ste6*-HA-GFP, In all cases (E-H), Pkg1 was used for loading controls for all the quantitation. In all cases, means of percent remaining for each ERAD substrate from at least three biological replicates are plotted, with error bars representing standard deviation (SD). A one-tailed t-test was used to determine the significance of the difference between WT cells grown with and without inositol. p < 0.05, *; p < 0.005, **; p < 0.001, ***.

https://doi.org/10.1371/journal.pone.0345910.g005

We next examined the impact of inositol depletion on cER inheritance and ERAD pathways. Under inositol-depleted conditions, early-stage (Class I) cells successfully formed ER tubules emerging from the pnER, extending into the bud and anchoring at the bud tip. However, in later-stage (Class II) cells, while ER tubules persisted, cER formation was partially inhibited or delayed. By the final stage (Class III), despite proper pnER being established in the bud, cER failed to uniformly distribute along the bud cortex (Figs 5C and S3C). Quantitative analysis confirmed significant defects in cER inheritance due to inositol depletion, although these defects were less severe than those induced by tunicamycin (Tm)–mediated ER stress (Fig 5D, in comparison to Fig 1D).

To assess the influence of inositol depletion on ERAD, we measured the turnover of three distinct ERAD substrates. Under inositol-depleted conditions, degradation of the ERAD-M substrate Hmg2-GFP accelerated (Fig 5E and 5F). In contrast, both the ERAD-L substrate CPY*-HA (Fig 5G and 5H) and the ERAD-C substrate Ste6*-HA-GFP (Fig 5I and 5J) exhibited mild stabilization, with only a slight decrease in their degradation rates, suggesting a modest impairment in the breakdown of these substrates. Previous studies have reported that inositol depletion alters ER membrane phospholipid composition—specifically decreasing PI levels while elevating cardiolipin (CL) and phosphatidylglycerol (PG) levels [51–53] (S4A Fig). These changes modulate overall membrane lipid homeostasis, which collectively enhances the degradation of ERAD-M substrates. In contrast, the degradation of ERAD-L and ERAD-C substrates remains largely unaffected, indicating that lipid perturbation under inositol depletion selectively impacts the ERAD-M pathway. To determine whether inositol depletion–induced membrane aberrations act synergistically with increased ER membrane lipid saturation, we assessed the growth of wild-type and SFA1–SFA4 strains on medium with or without inositol. Wild-type, SFA1, and SFA2 cells displayed no growth defects under inositol deprivation. However, the SFA3 strain grew slightly slower at 3^0°C under inositol-depleted growth condition and somewhat increased sensitivity at elevated temperature of 37^°C (S4B Fig, blue box). This trend became more pronounced for SFA4, where growth at 30°C on -inositol plates was reduced markedly (S4B Fig, red box). These results indicate a synthetic interaction between inositol depletion and elevated ER membrane lipid saturation.