1. Cell Biology
Download icon

Glucose restriction drives spatial reorganization of mevalonate metabolism

  1. Sean Rogers
  2. Hanaa Hariri
  3. N Ezgi Wood
  4. Natalie Ortiz Speer
  5. W Mike Henne  Is a corresponding author
  1. Department of Cell Biology, University of Texas Southwestern Medical Center, United States
  2. Department of Biological Sciences, Wayne State University, United States
Research Article
  • Cited 0
  • Views 1,730
  • Annotations
Cite this article as: eLife 2021;10:e62591 doi: 10.7554/eLife.62591

Abstract

Eukaryotes compartmentalize metabolic pathways into sub-cellular domains, but the role of inter-organelle contacts in organizing metabolic reactions remains poorly understood. Here, we show that in response to acute glucose restriction (AGR) yeast undergo metabolic remodeling of their mevalonate pathway that is spatially coordinated at nucleus-vacuole junctions (NVJs). The NVJ serves as a metabolic platform by selectively retaining HMG-CoA Reductases (HMGCRs), driving mevalonate pathway flux in an Upc2-dependent manner. Both spatial retention of HMGCRs and increased mevalonate pathway flux during AGR is dependent on NVJ tether Nvj1. Furthermore, we demonstrate that HMGCRs associate into high-molecular-weight assemblies during AGR in an Nvj1-dependent manner. Loss of Nvj1-mediated HMGCR partitioning can be bypassed by artificially multimerizing HMGCRs, indicating NVJ compartmentalization enhances mevalonate pathway flux by promoting the association of HMGCRs in high molecular weight assemblies. Loss of HMGCR compartmentalization perturbs yeast growth following glucose starvation, indicating it promotes adaptive metabolic remodeling. Collectively, we propose a non-canonical mechanism regulating mevalonate metabolism via the spatial compartmentalization of rate-limiting HMGCR enzymes at an inter-organelle contact site.

Introduction

The complexity of eukaryotic metabolism requires spatial organization, so membrane-bound and membraneless organelles compartmentalize enzymes into distinct sub-cellular regions. Enzyme partitioning into spatially defined domains is a cellular organizational principle observed in cytoplasmic assemblies (Chan et al., 2015), enzyme homo-polymers (Meredith and Lane, 1978) and biomolecular condensates (Banani et al., 2017; Jin et al., 2017). Such enzymatic assemblies, or metabolons, enhance or fine-tune metabolic flux, locally sequester toxic intermediates, or limit product shunting to competing pathways. Enzyme partitioning has also been exploited in synthetic bioengineering to artificially force enzymes into close proximity to drive local reactions or metabolic channeling through enzymatic cascades (Dueber et al., 2009).

Inter-organelle contact sites also serve as platforms for enzymatic organization and lipid synthesis reactions. In particular, the endoplasmic reticulum (ER) partitions lipogenic processes like non-vesicular lipid transport and organelle biogenesis at inter-organelle junctions (Lev, 2012; Hariri et al., 2018; Friedman et al., 2011). In yeast and humans, mitochondrial-associated membranes with the ER (MAMs) harbor enhanced metabolic activity to support reactions such as the biosynthesis and transport of phosphatidylserine (Vance, 1990; Gaigg et al., 1995). More recent work has highlighted the role of ER-mitochondrial contacts defining distinct sub-domains within mitochondria that support the formation of multi-enzyme complexes that drive Coenzyme Q synthesis (Subramanian et al., 2019; Eisenberg-Bord et al., 2019). In yeast, the nuclear envelope (continuous with the ER network) also makes extensive contact with the vacuole (equivalent to the lysosome), forming a nucleus-vacuole junction (NVJ). Formed through hetero-dimerization of nuclear envelope protein Nvj1 and vacuole-associated protein Vac8, the NVJ is a disc-shaped inter-organelle contact that acts as a multi-functional platform to organize lipid transport, fatty acid synthesis, and lipid droplet (LD) biogenesis, particularly during metabolic stress (Hariri et al., 2018; Murley et al., 2015; Kvam et al., 2005). Despite these insights, few studies mechanistically dissect how metabolic cues regulate enzyme recruitment and compartmentalization at inter-organelle contacts. Furthermore, how inter-organelle junctions enable enzyme spatial organization to fine-tune or enhance metabolic flux remains poorly described. Such inter-organelle crosstalk is particularly pertinent to understanding adaptive responses to stresses such as glucose starvation, which is characterized by drastic decreases in cellular ATP as well as changes in cytoplasmic pH and fluidity that alter macromolecular trafficking (Joyner et al., 2016; Munder et al., 2016).

Here, we use budding yeast as a genetically enabling model system to dissect the role of ER-lysosome contacts (e.g. the NVJ) as organizational platforms in mevalonate metabolism. We find that in response to acute glucose restriction (AGR), yeast actively partition and selectively retain HMG-CoA Reductase (HMGCR) enzymes at the NVJ in a Nvj1 and Upc2-dependent manner. This enzyme sub-compartmentalization enhances mevalonate pathway flux and promotes mevalonate and ultimately sterol-ester biosynthesis during AGR-induced metabolic remodeling. Furthermore, we find loss of HMGCR NVJ partitioning impacts metabolic remodeling following glucose starvation; failure to compartmentalize HMGCRs affects resumption of yeast growth following re-introduction to glucose containing media, but can be rescued by addition of exogenous mevalonate. We also find that NVJ compartmentalization of HMGCRs is accompanied by an Nvj1-dependent association of HMGCRs into high-molecular-weight complexes. Interestingly, both the growth resumption delay and mevalonate production phenotypes observed in Nvj1 knock-out cells can be rescued by the artificial multimerization of HMGCRs via a tetramerizing tag.

Results

Yeast Hmg-CoA reductases (HMGCRs) inducibly partition at the nucleus-vacuole junction (NVJ) in response to acute glucose restriction (AGR)

Given that ER-mediated contact sites in Saccharomyces cerevisiae act as lipogenic domains, we visually screened GFP-tagged neutral lipid metabolism enzymes for signs of compartmentalization at ER inter-organelle contact sites in yeast exposed to AGR. Note that AGR is operationally defined here as culturing yeast in synthetic complete (SC) media containing 2% glucose, then briefly centrifuging them and placing them into SC media containing 0.001% glucose. Screening revealed that endogenously GFP-tagged HMG-CoA Reductases (HMGCR) Hmg1 and Hmg2, which localize throughout ER network and particularly on the nuclear envelope (NE), enriched at the regions where the nuclear surface was in contact with the vacuole following AGR (Figure 1A). This nucleus-vacuole interface was confirmed to be the NVJ as Hmg1-GFP co-localized with NVJ tether Nvj1-mRuby3 following AGR treatment for 4 hr (Figure 1B). As Hmg1 and Hmg2 are functionally redundant in mevalonate synthesis, a central metabolite that supplies diverse cellular pathways including sterol biogenesis, we chose to dissect the mechanisms underlying Hmg1 NVJ partitioning. To investigate whether the NVJ was required for Hmg1 partitioning, we imaged Hmg1-GFP in the absence of NVJ tethers Nvj1 and Vac8 (Pan et al., 2000), and found they were required for Hmg1-GFP partitioning during AGR (Figure 1C). Furthermore, time-lapse imaging revealed that Hmg1-GFP NVJ partitioning peaked after ~4 hr of introducing yeast to AGR (Figure 1D–F). During this period, Hmg1-GFP was ~15 times enriched at the NVJ relative to other NE regions. Collectively, this suggests that AGR induces the time-dependent compartmentalization of Hmg1/2 at the NVJ in a Nvj1 and Vac8 dependent manner.

The HMG-CoA Reductases, Hmg1 and Hmg2, enrich at the NVJ in an Nvj1-dependent manner during AGR.

(A) Confocal microscopy images of yeast expressing endogenously tagged Hmg1-GFP or Hmg2-GFP (green) grown in SC media with 2% glucose to an OD of 0.5 (Log), or in 0.001% glucose for 4 hr (AGR). Vacuoles (magenta) were visualized by staining with FM4-64. Red arrows represent the relative location of the Nucleus-Vacuole Junction (NVJ), or where the nuclear envelope is closely apposed to the vacuole in nvj1Δ cells. Scale bars = 5 μm. (B) Epifluorescence microscopy images showing the overlap of Hmg1-GFP with Nvj1-mRuby3 after cells were exposed to four hours of AGR. Red arrows represent the location of the NVJ. Scale bars = 5 μm (C) Confocal microscopy images of yeast expressing endogenously tagged Hmg1-GFP (green) in wild type (WT), nvj1Δ, or vac8Δ yeast. Red arrows represent the location of NVJ. (D) Epifluorescence microscopy images of WT or nvj1Δ yeast expressing endogenously tagged Hmg1-GFP growing in 2% glucose to an OD of 0.5 (Log) or in 0.001% glucose for 2 or 4 hr (AGR). (E) Quantification of Hmg1 partitioning measured by line scan across the nuclear envelope (NE) and plotted as the ratio of Hmg1-GFP intensity at the NVJ relative to Hmg1-GFP intensity opposite the NVJ. Scale bars represent 5 μm. (Brown-Forsyth and Welch ANOVA. N > 51 cells. ***p-value<0.001). (F) Quantification of percentage of cells displaying Hmg1 NVJ partitioning per field of view. Any cell with partitioning >2.0 was considered as displaying Hmg1 NVJ partitioning. (Brown-Forsyth and Welch ANOVA. N > 51 cells. ***p-value<0.001).

NVJ compartmentalization of HMGCRs is independent of other mevalonate pathway enzymes and lipid trafficking proteins Osh1 and Ltc1

Inter-organelle contacts are implicated as domains coordinating the recruitment of supra-molecular enzyme complexes that constitute metabolic pathways. The HMGCRs are the rate-limiting enzymes of the mevalonate pathway that generates ergosterol, the cholesterol analog of yeast. To dissect whether other proteins along the mevalonate pathway also partition at the NVJ during AGR, we imaged 20 endogenously mNeonGreen (mNG)-tagged proteins involved in ergosterol biosynthesis under ambient and AGR stress, but surprisingly none detectably enriched at the NVJ (Figure 2A). As the NVJ has also been proposed to function in sterol transport between the vacuole and ER network, we also examined whether loss of the NVJ-resident proteins Osh1 and Lam6/Ltc1, implicated in the inter-organelle trafficking of lipids and/or sterols, would impact Hmg1-GFP NVJ recruitment (Murley et al., 2015; Kvam and Goldfarb, 2004). Hmg1-GFP efficiently partitioned at the NVJ following AGR in both osh1Δ and lam6/ltc1Δ yeast (Figure 2B). Collectively, this suggests that detectable Hmg1/2 NVJ enrichment is unique among enzymes involved in ergosterol metabolism, and is therefore unlikely to constitute a ‘classical’ multi-enzyme metabolon.

Hmg1 NVJ partitioning is specific amongst ergosterol biosynthesis proteins and independent of NVJ-associated sterol transport proteins.

(A) Epifluorescence microscopy images of twenty ergosterol biosynthesis proteins endogenously tagged with mNeonGreen (mNg) imaged in log phase and after 4 hr of AGR. Cells were co-stained with FM4-64 (magenta) to visualize vacuoles. Cellular compartments occupied by each protein are designated in white text in the top right corner of each image. Scale bars = 5 μm. (B) Epifluorescence microscopy images of cells endogenously tagged with Hmg1-GFP in wild type (WT), osh1Δ, or ltc1Δ cells exposed to 4 hr of AGR. Scale bars = 5 μm.

Hmg1-GFP is selectively retained at the NVJ in a Nvj1-dependent manner

Next, we investigated the mechanistic basis for NVJ compartmentalization of Hmg1-GFP. We began by interrogating whether the NVJ-partitioned and non-partitioned Hmg1-GFP pools were physically segregated along the NE. Using fluorescence recovery after photobleaching (FRAP), we determined that the non-partitioned Hmg1-GFP pool can enter the NVJ (Figure 3A,B); therefore, there is likely no diffusion barrier at the NVJ allowing selective entry of only specific pools of Hmg1-GFP protein. Next we interrogated whether Hmg1-GFP was selectively retained after entry into the NVJ region. Fluorescence loss in photobleaching (FLIP) revealed the average lifetime of NVJ-partitioned Hmg1-GFP was >100 s, compared to only ~25 s for Hmg1-GFP along the rest of the NE (Figure 3C,D). Given that most of the Hmg1-GFP signal within the NVJ-partitioned population was not photobleached to completion, calculated halftimes may underestimate the actual halftime. Collectively, this indicates that upon AGR, Hmg1-GFP is recruited and selectively retained at the NVJ. Surprisingly, the halftime of Hmg1-GFP was also significantly increased during AGR in nvj1Δ cells. This may be connected to a change in cytoplasmic viscosity which has been examined previously (Joyner et al., 2016); however, further analysis is required to determine if this reduced mobility is specific to Hmg1-GFP or a general consequence of AGR.

Figure 3 with 1 supplement see all
Hmg1 is selectively retained at the NVJ and requires a luminal domain of Nvj1.

(A) Spinning disk confocal microscopy images from FRAP movies of yeast expressing endogenously tagged Hmg1-GFP and stained with FM4-64 (magenta). Portions of the NE corresponding to the NVJ were photobleached (white arrows) and recovery was monitored in both WT and nvj1Δ cells grown to log phase or after 4 hr of AGR. Red arrows represent the location of the NVJ. (B) Quantification of Hmg1-GFP FRAP halftimes after photobleaching. (Brown-Forsyth and Welch ANOVA. N > 26 cells) (C) Spinning disk confocal microscopy images taken from FLIP movies. Conditions and strains were the same as in (A). One region of the NE that lies opposite the NVJ was photobleached every 5 s (white arrows) and Hmg1-GFP signal at the NVJ was monitored for loss of fluorescence. Time above the images indicates seconds after the first bleach pulse. Red arrows represent the location of the NVJ (D) Quantification of FLIP halftimes. (Brown-Forsyth and Welch ANOVA. N > 29 cells. *p-value>0.05 ***p-value<0.001). (E) Cartoon of Hmg1-GFP constructs endogenously expressed in yeast with both the cytosolic catalytic domain (blue) and TM/luminal domains (black) (Hmg1FL) or the TM/luminal domains alone (Hmg11-525). (F) Epifluorescence microscopy images of yeast expressing endogenously tagged Hmg1FL-GFP or Hmg11-525-GFP. (G) Cartoon of Nvj1-mNeonGreen constructs expressed in nvj1Δ cells. Nvj1FL-mNg contains all domains of Nvj1 including the N-terminal luminal anchor (red), the TM domain (black), and the Vac8-binding domain (VBD) (blue). Chimeric constructs either replaced the VBD with a vacuole-binding PX domain (Nvj1PX) or replaced the endogenous TM domain with the TM domain of Nvj2 (Nvj1NVJ2TM). Images of these constructs can be found in Figure 3—figure supplement 1A. Truncations of Nvj1 removed either residues 15–24 (Nvj115-24Δ), 15–30 (Nvj115-30Δ), or mutated residues 28 and 29 to alanine (Nvj1RK→AA). (H) Epifluorescence microscopy images of yeast expressing truncation constructs depicted in (G). Cells were co-expressing endogenously tagged Hmg1-mRuby3 (magenta). Scale bars represent 5 μm. (I) Quantification of Hmg1 partitioning at the NVJ for yeast expressing truncations of Nvj1-mNg and Hmg1-mRuby3. (Brown-Forsyth and Welch ANOVA. N > 61 cells. *p-value<0.05 ***p-value<0.001).

Next, we dissected which regions of Hmg1 and Nvj1 were required for Hmg1-GFP NVJ retention. We generated truncated versions of GFP-tagged Hmg1 containing only its integral transmembrane (TM) region and lacking its cytoplasm-exposed catalytic domain. This Hmg11-525-GFP truncation was sufficient to localize to the NVJ during AGR, indicating the integral membrane region of Hmg1 was sufficient for NVJ recruitment (Figure 3E,F). To dissect the Nvj1 regions required for Hmg1-GFP recruitment, we expressed mNg-tagged Nvj1 fragments in yeast co-expressing Hmg1-mRuby3 (Figure 3G). A Nvj1-mNG chimera lacking its cytoplasm-exposed region and containing a vacuole-binding PX domain in place of its Vac8-binding domain (Nvj1-PX) was sufficient to tether between the nucleus and vacuole, and was sufficient to partition Hmg1-mRuby3 at the NVJ during AGR, indicating the Nvj1 cytoplasmic domain is not required for Hmg1 recruitment (Figure 3—figure supplement 1A). Similarly, a Nvj1-mNG chimera with its TM region replaced with the TM of Nvj2 (Nvj1Nvj2TM) also partitioned Hmg1-mRuby3 at the NVJ, indicating the TM and cytoplasmic regions are not required for Hmg1-mRuby3 recruitment (Figure 3—figure supplement 1A). However, expression of a full length Nvj1-mNG lacking residues 15–30 (Nvj1Δ15-30) of its luminal region formed an NVJ contact but failed to recruit Hmg1-mRuby3. Re-addition of six residues to this construct (Nvj1Δ15-24) rescued Hmg1-mRuby3 recruitment. Furthermore, mutation of arginine and lysine residues at positions 28/29 to alanines (Nvj1RK→AA) also ablated AGR-induced recruitment of Hmg1 to the NVJ, suggesting residues 25–30 of the Nvj1 luminal domain play a key role in Hmg1-mRuby3 NVJ partitioning (Figure 3G–I). Collectively, these data indicate that the ER embedded region of Hmg1 is recruited to the NVJ in a manner requiring the ER-luminal Nvj1 region, and that Hmg1 partitioning is not dependent on the Hmg1 cytoplasmic catalytic domain nor the Nvj1 cytoplasmic/vacuole binding region.

NVJ partitioning of Hmg1 is Upc2 dependent and correlates with sterol-ester biosynthesis

The NVJ was previously identified as a site for starvation-induced microautophagy of the nucleus (Roberts et al., 2003), where NVJ-associated proteins accumulate and are subsequently engulfed by the vacuole. HMGCRs are also known to undergo proteasomal degradation during specific metabolic cues. To determine if NVJ partitioning of Hmg1-GFP regulated Hmg1 protein abundance or turnover, we performed immunoblot analysis of Hmg1-GFP in log-phase and 4 hr AGR-treated cells. AGR increased steady-state Hmg1-GFP protein levels, indicating glucose restriction promoted Hmg1 protein accumulation (Figure 4A). Interestingly, nvj1Δ yeast displayed a more marked increase in Hmg1-GFP protein compared to wildtype. To determine the influence of proteasomal degradation on Hmg1 protein levels during AGR, we treated cells with proteasome inhibitor MG132 and observed only a moderate increase in steady state Hmg1-GFP protein levels (Figure 4B). In contrast, treatment with the translation inhibitor cycloheximide during AGR returned Hmg1-GFP protein abundance to levels comparable with log-phase cells. Collectively, this analysis suggests that proteasomal degradation does not contribute significantly to controlling Hmg1-GFP protein abundance during AGR, and the increase is likely due to de novo protein synthesis during glucose restriction.

Figure 4 with 1 supplement see all
Hmg1 protein accumulates in nvj1Δ yeast via enhanced synthesis, which coincides with Upc2-dependent Hmg1 clustering at the NVJ.

(A) Immunoblot of cells expressing endogenously tagged Hmg1-GFP grown to log-phase or treated with AGR for 4 hr. Tubulin and Sec61 antibodies were used as loading controls. (B) Immunoblot of Hmg1-GFP expressing cells grown to log-phase or treated with AGR for 4 hr. AGR-treated cells were also co-treated with either 100 μg/mL cycloheximide or 25 μM MG132. Tubulin was used as a loading control. (C) Epifluorescence microscopy images of cells expressing endogenously tagged Hmg1-GFP in either a wild type (WT) or Snf1 knock-out (snf1Δ) background. (D) Epifluorescence microscopy images of cells expressing endogenously tagged Hmg1-GFP in either a wild type (WT) or ecm22Δ background. Scale bars = 5 μm. (E) Epifluorescence microscopy images of endogenously tagged Hmg1-GFP (green) with FM4-64-stained vacuoles (magenta) grown to log phase or after 4 hr of AGR in wild type (WT) and upc2Δ yeast. Scale bars = 5 μm. Red arrows indicate relative position of the NVJ. (F) Quantification of Hmg1-GFP partitioning at NVJ from images represented in (E). (Brown-Forsyth and Welch ANOVA. N > 49 cells. ***p-value<0.001). (G) Schematic representing abbreviated ergosterol biogenesis pathway. (H) Epifluorescence microscopy images of wild type and nvj1Δ cells expressing endogenously tagged Hmg1-GFP treated with 10 μg/mL of terbinafine for 4 hr. Scale bars represent 5 μm. (I) Quantification of Hmg1-GFP partitioning at the NVJ from images shown in (H). (Brown-Forsyth and Welch ANOVA. N > 51 cells. ***p-value<0.001).

To further explore the metabolic cues governing AGR-induced Hmg1 synthesis and spatial partitioning, we monitored Hmg1-GFP localization in yeast lacking the major glucose-sensing kinase Snf1, the yeast AMPK homolog that regulates metabolic remodeling during changes in glucose availability (Hedbacker and Carlson, 2008). Surprisingly, snf1Δ yeast maintained AGR-induced Hmg1-GFP partitioning at the NVJ, indicating Hmg1-GFP recruitment to the NVJ was not dependent on Snf1/AMPK signaling (Figure 4C). We next examined whether Hmg1-GFP NVJ partitioning required the ergosterol-sensing transcription factor Upc2 that controls yeast sterol synthesis in a manner similar to mammalian SREBP signaling (Yang et al., 2015). Indeed upc2Δ yeast failed to partition Hmg1-GFP at the NVJ during AGR, and this was specific to upc2Δ, as Hmg1-GFP maintained NVJ partitioning in ecm22Δ yeast, a Upc2 paralog (Figure 4D–F). To determine whether loss of Upc2 was influencing Hmg1 partitioning through Nvj1, we imaged Nvj1-mNg in wild type or upc2Δ cells. We found that neither Nvj1 localization, Nvj1 expression, nor NVJ size was affected in the absence of Upc2 (Figure 4—figure supplement 1A–C); therefore, loss of Hmg1 NVJ partitioning in upc2Δ cells occurs independently of Nvj1. Additionally, we monitored Hmg1-GFP expression in upc2Δ cells and found no significant changes when compared to Hmg1-GFP wild type cells (Figure 4—figure supplement 1D). These data suggest that there are yet unidentified factors that are needed for Hmg1 NVJ partitioning that require the presence of Upc2, as well as the luminal region of Nvj1, and that partitioning might be uncoupled from Hmg1 protein expression.

Control of Hmg1 partitioning via Upc2 implied that Hmg1-GFP NVJ partitioning correlated with alterations in cellular ergosterol levels. In line with this, Hmg1-GFP NVJ partitioning was also induced by treatment with the squalene epoxidase inhibitor terbinafine, which blocks de novo ergosterol biogenesis (Figure 4G–I). To directly examine cellular sterols, we conducted thin layer chromatography (TLC). Indeed, following 4 hr of AGR when Hmg1-GFP is NVJ partitioned, yeast exhibited ~30% more steady-state levels of sterol-esters (SE), while free ergosterol levels were unchanged (Figure 5A,B). Co-treatment with the HMGCR inhibitor lovastatin during AGR suppressed this SE elevation, indicating the elevated SE pool originated from de novo SE synthesis rather than esterification of pre-existing ergosterol. Altogether, this suggests that AGR induces an Upc2-dependent compartmentalization of Hmg1 at the NVJ, which correlates with elevated de novo SE synthesis.

Figure 5 with 1 supplement see all
Hmg1 NVJ partitioning coincides with de novo sterol-ester production and increased Hmg1 catalytic efficiency.

(A) Quantification of SEs in wild type and nvj1Δ cells collected at log phase, after 4 hr of AGR, or after a 4 hr co-treatment of AGR in the presence of 20 μg/mL lovastatin. Neutral lipids were extracted, separated by TLC, and visualized using Cu(II) Sulfate spray and charring. Quantification performed by densitometry. (Brown-Forsyth and Welch ANOVA. N = 6. **p-value<0.01) (B) Quantification of free sterols using the same strains and methodology as in (A). (Brown-Forsyth and Welch ANOVA. N = 6. ***p-value<0.001). (C) Scintillation counting quantification of radiolabeled HMG-CoA* produced after a 15 min pulse with 14C-acetate in intact WT or nvj1Δ cells grown for 4 hr in AGR. After quenching the radio-labeling reaction, soluble metabolites were isolated and endogenous HMG-CoA was converted into mevalonate using an in vitro HMGCR reaction. Mevalonate was subsequently separated and isolated by TLC, visualized by autoradiography, and quantified by scintillation counting. (Brown-Forsyth and Welch ANOVA. N = 6. **p-value<0.01). (D–G) Quantifications of radiolabeled DAG and mevalonate-derived lipids. Radio-labeling was performed as described in (C). Lipids were extracted and separated by TLC and visualized by autoradiography. Squalene and SE bands were quantified by scintillation counting. Ergosterol and DAG bands were quantified by densitometry. (Brown-Forsyth and Welch ANOVA. N = 6. ***p-value<0.001). (H) Growth curves of WT and nvj1Δ cells. Cells were treated with AGR for 10 hr in the absence or presence of 10 μg/mL mevalonate, and subsequently diluted to OD600 = 0.1 in SC media containing 2% glucose lacking mevalonate. OD600 measurements were taken every hour following dilution in fresh glucose-containing media. (I) Resumption probability graph depicting probability of cells producing a daughter bud as a function of time following an SC to SC+glucose transition (N > 85 cells).

NVJ partitioning of HMGCR enhances mevalonate pathway flux

Given that nvj1Δ yeast accumulate more Hmg1-GFP protein than wild-type cells (Figure 4A), we were surprised to find that nvj1Δ yeast produce similar steady-state levels of SEs following AGR. This implied that HMGCR enzymes in nvj1Δ yeast could be catalytically less efficient. If true, we would expect: (1) accumulation of the HMGCR substrate HMG-CoA and (2) a decrease in downstream mevalonate pathway products such as squalene, ergosterol, and SEs. To interrogate whether Hmg1 spatial compartmentalization influenced Hmg1 enzymatic activity and/or mevalonate pathway flux, we used 14C-acetate pulse-radiolabeling to monitor these mevalonate pathway components. Indeed, nvj1Δ yeast exhibited significantly elevated 14C-labeled HMG-CoA after a 15-min radio-pulse, and contained significantly less 14C-labeled squalene, ergosterol, and SE (Figure 5C–F, Figure 5—figure supplement 1A). However, 14C-diacylglycerol (DAG) was not significantly impacted, suggesting these alterations were specific to mevalonate pathway flux and not due to a non-specific dilution of the 14C-acetate radiolabel in nvj1Δ cells (Figure 5G).

To investigate whether these changes in mevalonate metabolism affected yeast fitness or growth, we monitored yeast growth after a 10 hr period of AGR stress, followed by re-introduction to SC-media containing 2% glucose (SC+glucose). Indeed, nvj1Δ yeast displayed slower growth in culture following an AGR to SC+glucose transition compared to WT (Figure 5H). Notably, this was rescued by addition of exogenous mevalonate, suggesting the growth defect was attributed to defects in the mevalonate pathway (Figure 5H). To dissect the nature of this delayed cell growth, we conducted single-cell time-lapse imaging of yeast exposed to 10 hr of SC media lacking glucose (which induces cell cycle arrest and a halt to cell budding), followed by a glucose replenishment phase when yeast were again re-supplied with SC+glucose media, which promoted budding resumption. Quantifying the time required for growth resumption following this SC to SC+glucose transition, as measured by the appearance of the first daughter bud, showed that nvj1Δ cells had a significant delay in growth resumption (Figure 5I). In line with this, doubling times of cells were not affected by either loss of Nvj1 or addition of mevalonate following growth resumption (Figure 5—figure supplement 1B). Together, these findings indicate that nvj1Δ yeastunable to spatially compartmentalize Hmg1 manifest alterations in mevalonate pathway flux, and that nvj1Δ yeast manifest growth defects following AGR that can be rescued by exogenous mevalonate addition.

Altered mevalonate pathway flux is not due to the general loss of the NVJ

To rule out the possibility that complete ablation of the NVJ contact site itself was responsible for reduced mevalonate pathway flux in nvj1Δ cells, we subjected nvj1Δhmg2Δ cells expressing either Nvj1FL or Nvj1RK→AA to radio-pulse analysis of the mevalonate pathway. Because the Nvj1RK→AA mutant still formed NVJs but did not recruit Hmg1, we were able to more precisely determine the role of Hmg1 partitioning specifically on mevalonate pathway flux. As seen for nvj1Δ cells, Nvj1RK→AA cells accumulated labeled HMG-CoA after a short pulse with 14C-acetate (Figure 6A). Furthermore, 14C-labeled squalene, ergosterol, and sterol-esters were significantly decreased in the Nvj1RK→AA mutant, while DAG production was unaffected (Figure 6B–D; Figure 6—figure supplement 1A,B). We also observed by single-cell time lapse imaging that the Nvj1RK→AA mutant cells experience growth resumption delay following an SC to SC+glucose media transition, similar to nvj1Δ cells (Figure 6E,F). It is unlikely that these effects can be contributed to different Hmg1 expression in the given backgrounds, as Hmg1-mRuby3 fluorescence intensity in Nvj1FL and Nvj1RK→AA cells were comparable during AGR (Figure 6—figure supplement 1C). Overall, this further supports a model where Hmg1 partitioning at the NVJ, and not simply the presence of an NVJ contact, contributes to increased mevalonate pathway flux and resumption of growth following glucose starvation.

Figure 6 with 1 supplement see all
AGR-induced compartmentalization of Hmg1 increases enzymatic efficiency.

(A) Quantification of radiolabeled HMG-CoA* from nvj1Δhmg2Δ yeast expressing either Nvj1FL or Nvj1RK→AA. Growth and radiolabeling was performed as described above. (Brown-Forsyth and Welch ANOVA. N = 6. ***p-value<0.001). (B–D) Quantification of radiolabeled squalene, ergosterol, and SE produced from a 15 min radio-labeling pulse in nvj1Δhmg2Δ yeast expressing either Nvj1FL or Nvj1RK→AA. (Brown-Forsyth and Welch ANOVA. N = 6. **p-value<0.01, ***p-value<0.001). (E) Quantification of growth resumption times during SC to SC+glucose transition in nvj1Δhmg2Δ yeast expressing either Nvj1FL or Nvj1RK→AA (N > 82 cells, ***p<0.0001, Kolmogorov-Smirnov test). (F) Resumption probability graph depicting probability of cells producing a daughter bud as a function of time following an SC to SC+glucose transition (N > 82 cells). (G) Cartoon representation of ADHpr:Hmg1-GFP strains. Strains were generated by removal of endogenous Hmg1 and Hmg2, and replacing with ADH promoter-driven Hmg1-GFP (ADHpr:Hmg1-GFP). (H) Immunoblot of Hmg1-GFP isolated from ADHpr:Hmg1-GFP strains before or after treatment with AGR. Tubulin was used as a loading control. (I) Epifluorescence imaging of ADHpr:Hmg1-GFP strains showing that Hmg1 NVJ partitioning can be uncoupled from endogenous protein expression. Red arrows indicate relative position of the NVJ. Scale bars = 0.5 μm. (J) Quantification of radiolabeled HMG-CoA* from yeast expressing ADHpr:Hmg1-GFP in wildtype (NVJ1+) or nvj1Δ yeast. Growth and radiolabeling was performed as described above. (Brown-Forsyth and Welch ANOVA. N = 6. **p-value<0.01). (K–M) Quantification of squalene, ergosterol, and SE produced from a 15 min radio-labeling pulse in ADHpr:Hmg1-GFP strains. (Brown-Forsyth and Welch ANOVA. N = 6. ***p-value<0.001).

Uncoupling Hmg1 abundance from compartmentalization indicates a role for NVJ partitioning in mevalonate pathway flux

The AGR-induced increases in Hmg1 protein abundance made it more challenging to specifically dissect the role of Hmg1 spatial compartmentalization in mevalonate metabolism. To mechanistically dissect this compartmentalization and uncouple it from Hmg1 protein level, we generated yeast strains lacking endogenous Hmg1 and Hmg2 and expressing Hmg1-GFP from a non-native ADH promoter (Figure 6G). This new strain (ADHpr:Hmg1-GFP) maintained similar Hmg1-GFP protein levels with and without AGR treatment, and in the presence or absence of Nvj1 (Figure 6H). Critically, ADHpr:Hmg1-GFP yeast still partitioned Hmg1-GFP at the NVJ during AGR in an Nvj1-dependent manner, indicating we had uncoupled Hmg1 protein levels from Hmg1-GFP spatial compartmentalization (Figure 6I).

Next, we dissected whether loss of NVJ spatial compartmentalization in this ADHpr:Hmg1-GFP yeast impacted mevalonate pathway flux. Radio-pulse analysis revealed that nvj1Δ yeast with ADHpr:Hmg1-GFP displayed exacerbated alterations in mevalonate pathway flux, exhibiting elevated 14C-labeled HMG-CoA levels and decreased 14C-squalene, ergosterol, and SEs (Figure 6J–M, Figure 6—figure supplement 1D). Once again, 14C-DAG levels were unaffected, indicating that nvj1Δ yeast manifested specific defects in mevalonate metabolism and not other ER-associated lipid metabolic pathways (Figure 6—figure supplement 1E). This suggests that Hmg1-GFP NVJ spatial partitioning, independent of Hmg1 protein levels, promotes mevalonate pathway flux.

Loss of Hmg1 NVJ partitioning can be bypassed by compartmentalizing Hmg1 via artificial multimerization

Bioengineering studies indicate that HMGCR enzymatic domains exhibit enhanced catalytic activity when forced into close proximities via multi-valent flexible scaffolds (Dueber et al., 2009). Similarly, human HMGCR requires tetramerization for catalytic activity (Istvan et al., 2000). We hypothesized that Hmg1 NVJ partitioning may enhance enzymatic activity by promoting the probability of close physical associations between HMGCR enzymes as they are retained in the NVJ. Indeed, when we analyzed Hmg1-3HA using blue native PAGE (BN-PAGE) we found that Hmg1 associated in high-molecular-weight (HMW) species corresponding to an approximate molecular mass of 720 kDa (Figure 7A,B). Appearance of Hmg1 HMW species was dramatically increased during AGR, and HMW species were significantly reduced in nvj1Δ cells. Differences in HMW species abundance was not simply due to Hmg1-3HA expression, as the trend was observed even when Hmg1-3HA loading was adjusted evenly as observed by SDS-PAGE from the same samples (Figure 7A,B).

Figure 7 with 1 supplement see all
Artificial multimerization of Hmg1 rescues enzymatic deficiencies of nvj1Δ cells.

(A) Blue native PAGE (BN-PAGE) and SDS-PAGE analysis of Hmg1-3HA expressed in either WT or nvj1Δ cells under log-phase or AGR growth conditions. Samples were loaded to mitigate differences in total Hmg1 loading such that differences in the BN-PAGE could be easily interpreted. Arrows on the side indicate relative molecular mass in corresponding region as determined by molecular weight standards. (B) Quantification of BN-PAGE and SDS-PAGE analysis as shown in (A). For quantification of relative Hmg1-3HA abundance in HMW species, fold change relative to WT log samples were first calculated for BN-PAGE species and subsequently normalized by total expression of Hmg1 in each sample/condition as determined by SDS-PAGE. (Brown-Forsyth and Welch ANOVA. N = 3. **p-value<0.01). (C) Cartoon representation of Hmg1 artificial tetramerization by DsRed2 tagging in nvj1Δ cells. (D) Scintillation counting quantification of radiolabeled HMG-CoA* from yeast expressing endogenously tagged monomeric Hmg1-mRuby3 or tetrameric Hmg1-DsRed2 in wildtype (NVJ1+) or nvj1Δ yeast. Quantification performed as in (Figure 6A). (Brown-Forsyth and Welch ANOVA. N = 6. **p-value<0.01 ***p-value<0.001). (E–G) Quantification of radiolabeled squalene, ergosterol, and SE produced in Hmg1-mRuby3 or Hmg1-DsRed2 yeast. Radio labeling and quantifications performed as in (Figure 6B–D). (Brown-Forsyth and Welch ANOVA. N = 6. ***p-value<0.001). (H) Average growth resumption times measured by single-cell time lapse microscopy. (WT: N = 110, nvj1Δ: N = 175, WT Hmg1-DsRed2: N = 94, nvj1Δ Hmg1-DsRed2: N = 96). (N > 94 cells, **p<0.001, Kolmogorov-Smirnov test). (I) Resumption probability graph depicting probability of cells producing a daughter bud as a function of time following an SC to SC+glucose transition. (N > 94 cells). (J) Cartoon representation of Nvj1 and Upc2-mediated Hmg1 partitioning at the NVJ. Spatial partitioning of Hmg1 at the NVJ coincides with increased production of mevalonate-derived metabolites.

To test whether loss of Hmg1 HMW species in nvj1Δ cells was contributing to lower mevalonate pathway flux, we fused Hmg1 to a constitutively tetrameric fluorescent protein, DsRed2, and determined whether this artificial multimerization could bypass the loss of NVJ-mediated compartmentalization by comparing it to Hmg1 tagged with a monomeric fluorescent protein, mRuby3 (Figure 7C). Both Hmg1-mRuby3 and Hmg1-DsRed2-tagged strains manifested similar increases in Hmg1 abundance in AGR, suggesting this tagging did not affect AGR-induced Hmg1 protein increase (Figure 7—figure supplement 1D). Furthermore, Hmg1-DsRed2 protein migrated at higher than expected molecular weight even in SDS-PAGE gels compared to Hmg1-mRuby3, suggesting the DsRed2 tag promoted self-association (Figure 7—figure supplement 1E). Hmg1-DsRed2 foci were also noted on the NE during AGR, consistent with stabilized Hmg1 multimers (Figure 7—figure supplement 1A).

As expected, nvj1Δ cells tagged with mRuby3 still manifested elevated 14C-labeled HMG-CoA and reduced squalene, ergosterol, and SE labeling associated with NVJ loss (Figure 7D–G). Strikingly, these perturbations were rescued in the nvj1Δ Hmg1-DsRed2 strain, closely mirroring wildtype 14C-labeled levels of Hmg-CoA, squalene, ergosterol, and SE (Figure 7D–G, Figure 7—figure supplement 1B,C). Once again, appending either mRuby3 or DsRed2 tags to Hmg1 did not significantly alter expression of the protein from what was observed with other tags, and our data cannot be explained by differences in Hmg1 expression alone (Figure 7—figure supplement 1D). Collectively, this indicates that artificial multimerization of Hmg1 can bypass the loss of NVJ-mediated Hmg1 compartmentalization.

Next, we interrogated whether Hmg1-DsRed2 tagging would rescue the growth resumption delay observed for nvj1Δ yeast following 10 hr exposure to SC media lacking glucose. As expected, growth resumption delay was observed in nvj1Δ yeast expressing Hmg1-mRuby3. Remarkably, expression of Hmg1-DsRed2 in nvj1Δ cells rescued this defect (Figure 7H,I). Collectively, this is consistent with a model where NVJ-dependent Hmg1 spatial compartmentalization via NVJ partitioning or artificial multimerization enhances mevalonate pathway flux as well as defects in growth resumption following glucose starvation.

Discussion

Acute changes in nutrient availability trigger metabolic remodeling that is characterized by alterations in metabolic pathway flux and metabolite supply and demand. How cells spatially and temporally coordinate this remodeling remains poorly characterized, yet critical to our understanding to cell adaptation and survival. Changes in nutrient availability can drive the subcellular re-distribution of enzymes within cells, as well as the formation of enzyme assemblies or complexes that promote or fine-tune metabolic pathways. Here, we present evidence that AGR in yeast enhances mevalonate pathway flux, and propose the mechanism underlying this requires the spatial compartmentalization of the rate-limiting HMGCR enzymes at the yeast NVJ (Figure 7J).

Through time-lapse and FRAP/FLIP-based imaging, we find that Hmg1 is recruited and selectively retained at the NVJ in a manner requiring basic residues within the Nvj1 luminal domain. We also demonstrate that Hmg1 partitioning at the NVJ can be uncoupled from Nvj1 binding of Vac8, which further supports a model where Hmg1 partitioning is being controlled predominantly from the ER. One model is that Hmg1 is selectively recruited as an enzyme ‘client’ to the Nvj1 ‘scaffold’ during AGR, but we cannot rule out the possibility that other factors may be involved together with Nvj1 in Hmg1 selective retention at the NVJ. Indeed, Upc2, a sterol-sensing transcription factor required for Hmg1 NVJ partitioning, appears to influence Hmg1 independent of either Nvj1 expression or localization, which may suggest additional factors required for Hmg1 partitioning at the NVJ. Both the requirement of Upc2 and the sufficiency of the Erg1 inhibitor terbinafine to induce Hmg1 NVJ partitioning is consistent with AGR stress producing a cellular demand for mevalonate. In line with this, we observe an increase in de novo SE synthesis during AGR; however, we cannot rule out that other mevalonate-derived metabolites may play important roles during AGR. In fact, we would expect AGR to promote a metabolic switch to mitochondrial respiration, which in part utilizes at least one mevalonate-derived metabolite, coenzyme Q, to function as an electron carrier. In support of this, we do not observe significant decline of steady-state SE levels in nvj1Δ cells, suggesting that mevalonate may be shunted into several pathways during AGR. Remarkably, nvj1Δ yeast manifest alterations in mevalonate pathway flux and growth defects when faced with glucose starvation, which can be rescued by the addition of exogenous mevalonate (the HMGCR enzymatic product).

We have additionally uncovered that mevalonate pathway flux can be rescued in nvj1Δ cells by artificially multimerizing Hmg1 using a tetrameric fluorophore. In line with this, we have observed that Hmg1 has a propensity to assemble into HMW species > 720 kDa during AGR in a Nvj1-dependent manner. Thus, mevalonate pathway flux during AGR appears tightly correlated with the ability to form HMW Hmg1 species at the NVJ, rather than the total Hmg1 protein level in the cell. To more fully dissect the role of HMGCR spatial compartmentalization on mevalonate metabolism, we generated an artificial yeast strain that maintains more constant Hmg1 protein levels and inducibly accumulates at the NVJ, thus functionally uncoupling Hmg1 protein abundance from NVJ spatial compartmentalization. Remarkably, this strain exhibits defects in mevalonate pathway flux when Hmg1 cannot be NVJ partitioned, again underscoring the role of Hmg1 spatial compartmentalization in fine-tuning mevalonate metabolism. Collectively, we present a model where AGR-induced metabolic remodeling of mevalonate metabolism is spatially coordinated at the NVJ via selective retention of HMGCRs and their subsequent incorporation into HMW species. An intriguing possibility is that, in general, inter-organelle contact sites act as platforms that slow the diffusion of proteins that enter them via their interactions with resident tether ‘scaffolds’, thus increasing the local concentrations of proteins and enzymes sufficient to fine-tune metabolic flux.

Glucose and other nutrient limitations can induce autophagy and have previously been shown to induce micro-lipophagy, where lipid droplets are engulfed by the yeast vacuole (Seo et al., 2017). Indeed, we have observed that LDs accumulate in close proximity to the NVJ during periods of glucose limitation (Hariri et al., 2018; Hariri et al., 2019). The relationship between LD biogenesis and HMGCR partitioning at the NVJ is not fully resolved here and requires further study. An intriguing possibility is that the spatial re-organization of HMGCRs may contribute to LD biogenesis or compositional remodeling during the general metabolic remodeling required during glucose restriction.

In addition to inducing the compartmentalization of HMGCRs at the NVJ, yeast glucose starvation has also been shown to induce the formation of several cytoplasmic enzyme assemblies involved in nucleotide (Ura7), amino acid (Gly1), and lipid (Acs1) metabolism (Munder et al., 2016). These cytoplasmic assemblies are thought to reduce enzymatic activity and promote the transition into cellular dormancy. In contrast, other protein assemblies promote enzymatic activity. In the liver, the protein Mig12 binds to cytosolic acetyl-CoA carboxylases (Acc) and promotes their polymerization and catalytic activity, which elevates fatty acid synthesis and eventual triglyceride accumulation in hepatocytes (Kim et al., 2010). Our work provides evidence that HMGCRs form similar spatial assemblies at yeast ER-lysosome contact sites, implicating a role for inter-organelle junctions in the spatial coordination of metabolically-induced enzymatic assembly formation. With this, we have added to a long, and growing, list for roles fulfilled by inter-organelle contact sites in metabolic regulation.

Materials and methods

Key resources table
Reagent type
(species) or
resource
DesignationSource or
reference
IdentifiersAdditional
information
Strain, strain background (Saccharomyces cervisiae)W303This paperW303See Supplementary file 1 for all yeast strains
Transfected construct (Saccharomyces cervisiae)pRS305 vector backbone (Leu selection)This paperpRS305See Supplementary file 2 for all yeast expression plasmids
AntibodyAnti-GFP
(rabbit polyclonal)
AbcamCat# abcam290WB (1:10,000)
Antibodyanti-mRuby3
(rabbit polyclonal)
InvitrogenCat# R10367WB (1:1,000)
Antibodyanti-DsRed2
(mouse monoclonal)
OrigeneCat#: TA180005WB: (1:1,000)
AntibodyAnti-tubulin
(rat monoclonal)
AbcamCat#: ab6160WB: (1:15,000)
AntibodySec61
(rabbit poly)
J. Friedman lab (UTSW)N/AWB: (1:15,000)
Software, algorithmSegmentation and tracking code for single-cell time-lapse imaging of yeast buddingfrom N. E. Wood, A. Doncic, A fully-automated, robust, and versatile algorithm for long-term budding yeast segmentation and tracking. PLoS ONE. 14, e0206395 (2019). Folch et al., 1957---

Strains, plasmids, and yeast growth conditions

Request a detailed protocol

W303 (leu2-3,112 trp1-1 can1-100 ura3-1 ade2-1 his3-11,15) was used as the wild-type parental strain for all experiments and cloning in this study. All strains and plasmids can be found in Supplementary files 1 and 2, respectively. Deletion of endogenous genes, C-terminal tagging at endogenous loci, and transformation of plasmids was accomplished by the traditional lithium acetate method. Endogenous knockins and knockouts were validated by genomic PCR, and plasmids were validated by sequencing. Plasmids generated for this study were created using RepliQa HiFi assembly (Quantabio cat. 95190) following the manufacturer’s protocol. For cloning into pRS305 plasmids, destination vectors were cut with XhoI and SacI enzymes, while linearization of pFa6a plasmids was accomplished by digestion with AscI and PacI enzymes. Synthetic complete media supplemented with amino acids was used for all experiments, except where leucine was omitted from the media to accommodate growth of strains carrying pRS305 plasmids. All yeast were grown in 30°C incubators shaking at 210 RPM. Log phase yeast were grown to an OD600 of 0.5 in the presence of 2% dextrose. AGR-treated yeast were grown to OD600 of 0.5 in the presence of 2% dextrose, collected, washed with dextrose-free media, and resuspended in media containing 0.001% dextrose for the indicated time period. For experiments where TG lipolysis was induced, cells were grown to stationary phase over 24 hr and then diluted to an OD600 of 0.5 in the presence of 2% dextrose and 10 μg/mL Cerulenin (Sigma C2389). Cells were incubated with cerulenin for 4 hr prior to collection. Where indicated, Terbinafine (Sigma T8826), Lovastatin (Sigma 1370600), Cycloheximide (Sigma C7698), or MG132 (Sigma M7449) were added to the media at the beginning of AGR treatment to final concentrations of 10 μg/mL, 20 μg/mL, 100 μg/mL, and 25 μM, respectively.

Fluorescence microscopy

Request a detailed protocol

For confocal microscopy, cells were grown as described above, collected by centrifugation at 3000xg for 2 min, and resuspended in glucose-free media at approximately one one-hundredth of original volume. All images were taken as single slices at approximately mid-plane using a Zeiss LSM880 inverted laser scanning confocal microscope equipped with Zen software. Images were taken with a 63x oil objective NA = 1.4 at room temperature. Prior to imaging, cells were incubated for three hours with 0.5 μg/mL FM4-64 dye (Invitrogen T13320) to visualize vacuoles.

For epifluorescence microscopy, cells were grown, stained, and collected as described above. Vacuoles were also stained with 5 μg/mL CMAC (ThermoFisher C2110) dye for two hours prior to imaging, where indicated. Imaging was performed on an EVOS FL Cell Imaging System at room temperature. Hmg1 NVJ partitioning was quantified using Fiji software. For quantification, RGB images were converted to 16-bit and a background subtraction was performed by subtracting original images by a duplicate ‘Gaussian blur’ filtered image (sigma (radius) = 5.0). Five-pixel line scans were taken across the nuclear envelope toward the NVJ, and the ‘plot profile’ function was used in Fiji to produce a fluorescence histogram of nuclear envelope signal. The sum area under each curve was calculated and plotted as the ratio of fluorescence intensity of NVJ-associated signal by fluorescence intensity of Non-NVJ associated NE signal.

FRAP and FLIP analysis

Request a detailed protocol

Yeast used for FRAP and FLIP were grown and collected as described above, and imaging was conducted for 1 hr after collection. Photobleaching movies were taken on an Andor spinning disk confocal microscope through a 63x oil objective (NA = 1.4). The microscope is equipped with an Andor Ultra EMCCD and Metamorph software. For FRAP measurements, a single circular ROI of 0.77 μm2 that corresponds to the NVJ was selected and bleached with a 408 nm laser at 100% power and 100 ms dwell time. One image was taken before the bleach, and subsequent images were taken every 500 ms for a total movie length of 25 s. For FLIP measurements, single circular ROIs of 0.77 μm2 were selected, taking care to select an area of the NE that was furthest from the NVJ. Each bleaching cycle consisted of a pre-bleach image, a single bleach with a 408 nm laser with a 100 ms dwell time, and four post-bleach images taken 500 ms apart. Each movie captured a total of 50 bleach cycles, which corresponds to 300 s movies. Fiji software was used to quantify bleaching curves and halftimes. Pre-processing of images included background subtraction as described above and 3D Gaussian smoothing (sigma = 0.5). FRAP quantification was performed using the double normalization method as previously described (Phair and Misteli, 2000). Briefly, intensity was measured for all time points in an ROI corresponding to the bleached region (Ifrap) and an ROI corresponding to the whole cell (Iwhole-cell), and normalized intensities were generated for each timepoint (Inormalized(t)) using Equation 1:

(1) Inormalizedt= Iwhole-cell-preIwhole-cell(t)*Ifrap(t)Ifrap-pre

In the equation above, the ‘pre’ subtext indicates the timepoint preceding ROI bleaching. Intensity recovery curves were created for each movie by further normalizing values such that pre-bleach intensities were set to one and post-bleach intensity of an ROI was set to 0. Full normalization was accomplished using Equation 2 followed by subtracting the normalized Ifrap-bleach value from all timepoints.

(2) Inormalized-fullt= Inormalized(t)Ifrap-pre-Ifrap-bleach

In Equation 2, Ifrap-bleach indicates the intensity of the photobleached ROI at the time of photobleaching. Halftimes were calculated from individual fluorescence recovery curves using Graphpad Prism eight software and fitting the data to a one-phase exponential association. Pre-processing for FLIP images was the same as for FRAP images. FLIP movies were quantified using Fiji to monitor the intensity of NVJ-associated signal over time. Intensity measurements were normalized using Equation 3.

(3) IFLIPt= INVJ(t)INVJ-pre-bleach

In Equation 3, IFLIP(t) is the relative fluorescence at the NVJ at time t, INVJ(t) is the raw intensity value at the NVJ at time t, and INVJ-pre-bleach is the intensity at the NVJ before bleaching occurred. Halftimes were calculated from FLIP decay curves, which were generated in Graphpad Prism eight software by fitting the data to a one-phase exponential decay.

Lipid extraction and thin layer chromatography

Request a detailed protocol

For lipid extraction, approximately 50OD units of cells were collected for each sample, and pellet wet weight was normalized and noted prior to extraction. Lipid extraction was performed using a modified Folch method (Folch et al., 1957). Briefly, cell pellets were resuspended in MilliQ water with glass beads and lysed by three one-minute cycles on a bead beater. Chloroform and methanol were added to the lysate to achieve a 2:1:1 chloroform:methanol:water ratio. Samples were vortexed, centrifuged to separate the organic and aqueous phases, and the organic phase was collected. Extraction was repeated a total of three times. Prior to thin layer chromatography, lipid samples were dried under a stream of argon gas and resuspended in 1:1 chloroform:methanol to a final concentration corresponding to 4 μL of solvent per 1 mg cell pellet wet weight. Isolated lipids were spotted onto heated glass-backed silica gel 60 plates (Millipore Sigma 1057210001), and neutral lipids were separated in a mobile phase of 80:20:1 hexane:diethyl ether:glacial acetic acid. TLC bands were visualized by spraying dried plates with cupric acetate in 8% phosphoric acid and baking at 140°C for an hour. To quantify TLC bands, all plates were run with an internal neutral lipid standard. Densitometry of bands was performed in Fiji.

Batch culture growth curves

Request a detailed protocol

Cells were treated with AGR for 10 hr, as described above. Some samples were also co-treated with 10 μg/mL mevalonate (Sigma 50838) at the beginning of AGR treatment. After a 10-hr treatment with AGR, cultures were diluted to an OD600 of 0.1 in SC media containing 2% glucose. The OD600 of cultures was measured each hour and plotted in Prism eight software.

Single-cell time-lapse microscopy

Request a detailed protocol

Cells were imaged using a Zeiss Observer Z1 microscope equipped with automated hardware focus, motorized stage, temperature control, a Zeiss EC Plan-Apochromat 63 × 1.4 or 40 × 1.3 oil immersion objectives, and an AxioCam HRm Rev three camera. Exposure times for experiments shown in Figure 5—figure supplement 1B: Phase contrast 20 ms, Whi5-mKoκ 150 ms, Erg6-mTFP1 20 ms, Vma1-mNeptune2.5 75 ms, Nvj1-mRuby3 75 ms, Msn2-mNeonGreen 40 ms. Exposure times for experiments shown in Figure 4 and Figure 7—figure supplement 1: Phase contrast: 40 ms, Hmg1-mRuby3 or Hmg1-DsRed2: 100 ms.

All experiments were performed with a Y04C Cellasic microfluidic device (http://www.cellasic.com/) using 1 psi flow rate. Images are taken every 6 min. Prior to loading into the microfluidics chamber, cells were sonicated and mixed with 50 μL SCD media to achieve an OD600 of approximately 0.1. In the chamber, cells are first grown for 2 hr in SCD. Next, they are exposed to acute glucose restriction (AGR) by switching to SC for 10 hr which is followed by glucose replenishment by 4 hr SCD. To determine time of growth resumption, cells are segmented and tracked as described previously (Doncic et al., 2013). Next, the time of resumption is annotated semi-automatically using a custom MATLAB software by determining the time of bud growth or new bud emergence during the 4-hr glucose replenishment following AGR.

Immunoblotting

Request a detailed protocol

Approximately 50OD units of cells were collected for protein extraction. Prior to protein extraction, cell pellet wet weights were normalized. Protein extraction was accomplished by precipitating proteins with 20% TCA for 30 min on ice, followed by three washes of the pellet with cold 100% acetone. The protein pellet was dried for fifteen minutes in a speed-vac to remove residual acetone, and all pellets were resuspended in 2x SDS sample buffer (65.8 mM Tris-HCl, pH 6.8; 2% SDS; 25% glycerol; 10% 2-mercaptoethanol; 0.01% bromophenol blue). Resuspended protein samples were heated at 70°C for 10 min prior to being loaded onto a homemade 4–15% polyacrylamide gel and separated by electrophoresis. Proteins were transferred to a 0.45 μm nitrocellulose membrane in Towbin SDS transfer buffer using a Criterion tank blotter with plate electrodes (BioRad 1704070). Immediately following transfer, membranes were stained with PonceauS and cut using a clean razor blade. Membranes were blocked with 5% milk dissolved in TBS-T buffer, and primary antibodies were allowed to bind overnight at 4°C. Primary antibodies used for determining Hmg1 protein expression are as follows: GFP (Abcam ab290; 1:10,000 dilution), mRuby3 (Invitrogen R10367; 1:1000 dilution), and DsRed2 (OriGene TA180005; 1:1000 dilution). Rat monoclonal antibody to tubulin (Abcam ab6160; 1:15,000 dilution) and rabbit polyclonal antibody to Sec61 (Jonathan Friedman lab; 1:5000 dilution) were used as loading controls. Immunoblots were developed by binding HRP-conjugated anti-rabbit IgG (Sigma A0545; 1:10,000), anti-rat IgG (Abcam ab97057; 1:10,000), or anti-mouse IgG (Abcam ab6728; 1:2,000) secondary antibodies to the membrane for 1 hr in the presence of 5% milk followed by three washes in TBS-T and developing with ECL substrate (BioRad cat. 1705061). Signal was captured by X-ray film.

Radiolabeling, metabolite extraction, and metabolite separation

Request a detailed protocol

Approximately 100OD units of cells were used for radiolabeling experiments. All cells were grown and treated with AGR as previously described. Prior to radiolabeling, cells were collected by centrifugation and washed with dextrose-free media. All liquid was removed from cell pellets prior to labeling. To start radiolabeling, 1.0 mL of dextrose-free media containing 5 μCi/mL 14C-Acetate was quickly added to each tube, followed by mixing with pipetting. Tubes were tumbled in a 30°C rotating incubator for fifteen minutes. To quench the radiolabeling reaction, and wash the cells, samples were pipetted into 20 mL of −40°C quenching buffer (60% methanol; 1 mM tricine pH 7.4). Cells were incubated in quenching buffer for three minutes, centrifuged at 3000xg at −10°C, and washed with 20 mL of −40°C quenching buffer. Cells were again pelleted by centrifugation, and pellets for lipid extraction were stored at −80°C. For pellets undergoing soluble metabolite extraction, all quenching buffer was thoroughly removed and 1.5 mL of 80°C 75% ethanol was added to each sample followed by a three-minute incubation at 80°C and a subsequent 5-min incubation on ice. Debris was removed from metabolite extract by centrifuging at 20,000xg for 1 min. HMG-CoA labeled during the pulse was separated and quantified as mevalonate. To convert endogenous HMG-CoA to mevalonate, ethanol was thoroughly evaporated from isolated metabolites under argon gas, and each sample was resuspended in HMGCR buffer (50 mM Tris-HCl, pH 6.8; 1 mM NaCl; 1 mM MgCl2; 1 mM DTT; 100 mM glucose-6-phosphate; 1 mM NADP+; 1 mM NADPH). The samples were split into two tubes, one tube would be treated with enzymes, and the untreated tubes acted as blanks for endogenous mevalonate labeled during the pulse. For treated tubes, 2U of HMGCR enzyme (Sigma H7039) and 2U of glucose-6 phosphate dehydrogenase (Sigma G6378) were added. Reactions were carried out over night at 37°C. Prior to loading samples onto TLC plates, total radioactivity in each sample was quantified by scintillation counting and loading was adjusted accordingly. HMGCR Reactions were spotted onto Silica gel G plates (Miles scientific P01911) and separated with a mobile phase of 70:25:5 Diethyl ether:glacial acetic acid:water. Each lane was doped with 5 μg of unlabeled mevalonate to act as a tracer for downstream scraping/quantification. TLC plates were developed overnight by autoradiography and visualized in an Amersham Typhoon FLA 9500 developer. To visualize the unlabeled tracer mevalonate, plates were sprayed with p-anisaldehyde reagent and baked at 140°C for 10 min. Individual bands were scraped, mixed with 6 mL of EcoLume scintillation cocktail (VWR cat. IC88247001), and quantified by scintillation counting in a Beckman LS 6500 instrument. Mevalonate from the untreated samples was averaged and subtracted from the final values of the treated samples. For lipid extracts, an aliquot of cell lysate was taken immediately following bead beating and radioactivity was quantified by scintillation counting to serve as a normalization standard prior to loading samples onto the TLC plate. Lipid extraction was performed as described above. To separate squalene, ergosterol, DAG, and sterol-esters, total lipid extracts were spotted onto silica gel 60G plates (Fisher cat. NC9825743), and developed in a mobile phase of 55:35:10:1 hexane:petroleum ether:diethyl ether:glacial acetic acid. Autoradiography and scintillation counting was performed as indicated above. Prior to TLC separation, unlabeled squalene, ergosterol, and sterol-ester was added to each lane of the plate as tracer. DAG and ergosterol were quantified by densitometry in Fiji.

Native protein isolation

Request a detailed protocol

Approximately 100OD units of cells were collected for BN-PAGE analysis. Cell pellets were collected into 2 mL screw-cap tubes and frozen at −80°C until further processing. For cell lysis, one large scoop of glass beads and 300 μL of lysis buffer was added to each tube (50 mM HEPES-NaOH, pH 7.0; 50 mM NaCl; 250 mM sorbitol; 10% glycerol; 20 mM arginine; 20 mM glutamic acid; 1.5 mM MgCl2; 1.0 mM CaCl2; 1 mM EDTA; 1 mM DTT; 50 μM Mg132; 1x protease/phosphatase inhibitor cocktail (ThermoFisher cat. 78429)). Tubes containing frozen cell pellet, glass beads and lysis buffer were agitated for one minute using a bead beater followed by 2 min on ice and another minute of agitation with a bead beater. After glass bead lysis, another 700 μL of lysis buffer was added to each tube followed by a brief mixing via vortex. Cell debris was cleared by a 5 min 1000xg spin at 4°C. Supernatants were transferred to a new 1.5 mL tube and centrifuged again at 21,000xg for 30 min at 4°C to isolate a crude microsomal fraction. The supernatants from the first spin were discarded and crude microsomes were washed with 1000 μL of cold lysis buffer followed by another spin at 21,000xg for 30 min at 4°C. Supernatants were again discarded and crude microsomes were resuspended in 60 μL of lysis buffer containing digitonin at a final concentration of 1.5%. Tubes were kept on ice for 30 min and gentle pipetting was applied every 10 min to aid in solubilization of the microsomes. To clear insoluble materials, tubes were centrifuged at 100,000xg for 30 min at 4°C in a Beckman TLA-55 fixed angle rotor. Supernatants were transferred to new tubes on ice after the clearance spin. Approximately 30 μL was reserved for BN-PAGE and 20 µL for SDS-PAGE. The remaining 10 μL was used to quantify protein concentrations in duplicate with the Pierce detergent compatible Bradford kit (ThermoFisher cat. 23246). For SDS-PAGE samples, 4x NuPAGE LDS sample buffer (cat. NP0007) was added to each sample for a final concentration of 1x. Each sample was supplemented with 2-mercaptoethanol and urea at final concentrations of 5% and 8M, respectively. SDS-PAGE samples were heated at 37°C for 40 min and loaded onto homemade 4–15% polyacrylamide gels. SDS-PAGE and immunoblotting was performed as described above.

Blue native PAGE and immunoblotting

Request a detailed protocol

Prior to BN-PAGE, 3.3 μL of BN-PAGE loading buffer was added to each sample (100 mM Bis-Tris, pH 7.0; 500 mM 6-aminocaprioc acid; 5% Coomassie G-250) and gentle mixing was applied via vortex. Samples were loaded onto NativePAGE 3–12% Bis-Tris gels (ThermoFisher cat. BN1001BOX) alongside a NativeMark unstained protein standard (ThermoFisher cat. LC0725). Separation was performed in a mini gel tank (ThermoFisher cat. A25977) with 1x NativePAGE running buffer (ThermoFisher cat. BN2001) in the anode and 1x NativePAGE running buffer containing 1x cathode buffer additive (ThermoFisher cat. BN2002) in the cathode. Electrophoresis was performed at 150V for 30 min, after which cathode buffer was replaced with 1x NativePAGE running buffer containing 0.1x cathode buffer additive and ran at 150V for another 30 min. After 1 hr at 150V, voltage was increased to 300V until the dye front was at the bottom of the gel. Once gel case was removed, the wells and dye front were excised and discarded from the gel with a clean razor blade. The ladder was excised and stained separately in GelCode blue stain (ThermoFisher cat. 24594) according to the manufacturer’s instructions. The remaining gel was incubated in denaturing buffer (300 mM Tris; 100 mM acetic acid; 1% SDS) for 20 min on a rocker, after which, the gel was placed between two glass plates wetted with denaturing buffer and incubated for one hour. Proteins were transferred to a 0.2 μm PVDF membrane overnight by wet transfer at 25V in a cold room with BN-PAGE transfer buffer (150 mM Tris; 50 mM acetic acid). After overnight transfer, the membrane was incubated in 8% acetic acid for five minutes followed by a 5-min incubation in MiilliQ water. The membrane was dried for 1 hr at room temperature and then washed quickly four times with 100% methanol followed by a quick wash with MilliQ water. The membrane was then placed in 5% milk for 1 hr, and subsequent steps for immunoblotting were performed as described above. For detection of Hmg1-3HA, an anti-HA primary antibody was used (Abcam ab9110; 1:2000 dilution) followed by rabbit secondary described above.

Statistical analysis

Request a detailed protocol

T-tests and one-way ANOVA tests were performed using Graphpad Prism8 software. Kolmogorov-Smirnov test was conducted using kstest2 MATLAB function. All unpaired t-tests were performed with Welch’s correction. For one-way ANOVA, Brown-Forsyth and Welch ANOVA was performed to account for non-uniform standard deviations, followed by Turkey’s post-hoc test to extract p-values. For all t-tests and ANOVAs. * p-value<0.05, **p-value<0.01, ***p-value<0.001. Dotted lines in violin plots represent the median of the data, while upper and lower dotted lines represent the upper and lower quartiles. Bar graphs show mean of the data with error bars indicating the standard deviation.

Data availability

All data generated or analysed during this study are included in the manuscript and supporting files.

References

Decision letter

  1. Benoît Kornmann
    Reviewing Editor; University of Oxford, United Kingdom
  2. Suzanne R Pfeffer
    Senior Editor; Stanford University School of Medicine, United States
  3. Benoît Kornmann
    Reviewer; University of Oxford, United Kingdom

In the interests of transparency, eLife publishes the most substantive revision requests and the accompanying author responses.

Acceptance summary:

This study shows that an important biosynthetic pathway is influenced by the sequestration of some of its enzymes at a membrane contact site. This finding uncovers a novel way to control metabolism, besides transcriptional, translational and posttranslational control; the controlled doing and undoing of contact sites.

Decision letter after peer review:

Thank you for submitting your article "Glucose restriction drives spatial reorganization of mevalonate metabolism and liquid-crystalline lipid droplet biogenesis" for consideration by eLife. Your article has been reviewed by three peer reviewers, including Benoît Kornmann as the Reviewing Editor and Reviewer #1, and the evaluation has been overseen by Suzanne Pfeffer as the Senior Editor.

The reviewers have discussed the reviews with one another and the Reviewing Editor has drafted this decision to help you prepare a revised submission.

As the editors have judged that your manuscript is of interest, but as described below that additional experiments are required before it is published, we would like to draw your attention to changes in our revision policy that we have made in response to COVID-19 (https://elifesciences.org/articles/57162). First, because many researchers have temporarily lost access to the labs, we will give authors as much time as they need to submit revised manuscripts. We are also offering, if you choose, to post the manuscript to bioRxiv (if it is not already there) along with this decision letter and a formal designation that the manuscript is "in revision at eLife". Please let us know if you would like to pursue this option. (If your work is more suitable for medRxiv, you will need to post the preprint yourself, as the mechanisms for us to do so are still in development.)

This study shows that the yeast HMG-CoA reductase, which normally localizes throughout the ER, is, upon acute glucose removal (AGR), sequestered at a contact site between the nuclear ER and the lysosome, called nuclear-vacuole junctions (NVJ). AGR is accompanied by an increase in sterol ester synthesis, and failure to generate NVJs hampers the flux in the steryl ester biosynthesis pathway, suggesting that the sequestration of the HMG-CoA reductase impacts its activity positively. One of the strengths of the manuscript is that it takes a great care in removing any transcriptional influence by observing changes in HMG-CoA reductase activity when the protein is expressed from constitutive promoters, and also that it attempts to recreate the activity change by artificially clustering of the protein at defined positions. This novel finding stresses the importance of subcellular localization for proper wiring of metabolic fluxes. Additionally, AGR causes the appearance of liquid-crystalline structures in lipid droplets, an interesting side observation.

All three reviewers found the work exciting and important. There are nevertheless ambiguities in the present manuscript that need to be addressed, and a decision taken about what to make of the side observation with liquid crystalline structures in LDs (see below).

Points that require new experimental work:

1. The first point is about the artificial clustering. This is a very important point of the study, but the result is however yet not entirely clear.

– The human HMG-CoA reductase works as a tetramer. The oligomerization state of yeast Hmg1 is unclear. There is a strong focus on the manuscript on multimerization without data supporting it. Monomeric mRuby and tetrameric DsRed are used to manipulate the oligomeric state Hmg1 but there is no information on the oligomeric state of these fusion proteins. Moreover it is not tested whether the fusion to different tags affect the protein's activity and/or abundance. In order to make the point that it is indeed the oligomerization status change that causes the change in activity, it is necessary to rule out changes in abundance or activity due to the tags being appended to the protein.

– It is unclear why AGR should cause stabilized Hmg1-dsRed2 multimers in nvj1-delete cells (lines 243-244, figure 6I). Here the multimerization is supposed to be constitutive and driven by the dsRed. The factor (Nvj1) that is supposed to cause AGR-dependent clustering of Hmg1 is absent. Does it mean that Hmg1 has a tendency to cluster during AGR in a way that is mechanistically unclear, and in order for it to happen efficiently, it needs to be boosted, either by being sequestered at the NVJ, or by being fused to a tetramerizing tag? This definitely requires further tests and explanations.

2. The second point is Nvj1's influence on Hmg1.

– One likely explanation that is strongly implied is that Nvj1 and Hmg1 interact via the 10 a.a. stretch identified therein. Can this interaction be shown by pull-down? Is the interaction induced by AGR, or is it constitutive (in which case Hmg1 simply follows Nvj1 relocalisation to NVJs in AGR)? Also several conditions prevent Hmg1 recruitment and sequestration to NVJs (e.g. lovastatin, upc2 deletion). Do these conditions affect the making of NVJs and the localisation of Nvj1 therein or are they specific to Hmg1 recruitment?

– There is a possibility that the effects of the Nvj1 deletions are indirect through the absence of NVJ, and the absence of recruitment therein of other lipid homeostasis factors (e.g. Osh1, Tsc13, Vps13). It is however possible to decouple NVJ generation and Hmg1 recruitment by using the 10 a.a. deletion mutant. Using such a mutant would be much cleaner and ideally, all experiments using Nvj1 delete cells should be replaced or complemented with the 10 a.a. mutant. This would imply a large amount of work, and the reviewers discussed that it should be sufficient to just do the growth assays, as well as the data in figure 5A-G with this setup (see also point below).

– It is emphasized throughout that Nvj1's influence on Hmg1 activity is limited to AGR conditions. This is consistent with Hmg1 sequestrations at NVJs only in these conditions. It is however not tested.

All the data from figure 5C-G and 6D-G,J-M are in AGR only. Yet, Nvj1 might also influence the mevalonate pathway in log phase. Since it will be anyway necessary to redo the experiments of figure 5A-G with the 10 a.a. nvj1 mutant (see point above), it will be necessary to add log phase quantification. If nvj1 mutation or deletion had an effect on log phase, a condition where Hmg1 is not clustered at NVJs, then this would imply a much more indirect mechanism. To make the claim water-tight, these experiments should ideally be performed in an Hmg2-delete background to discount any variations in Hmg2 expression during AGR.

Points that might be addressed by amendments in the text, or by new experimental work:

3. The part on Liquid crystalline layered LDs is disconnected from the rest. It is not tested whether Nvj1-dependent clustering of Hmg1 to NVJs is necessary, sufficient or unrelated to the formation of these structure. While all three reviewers found it to be an interesting observation, they found that, as presently written, the manuscript is ambiguous at stating the lack of connection between the two phenomena. The title itself "Glucose restriction drives spatial re-organization of mevalonate metabolism and liquid-crystalline lipid droplet biogenesis" could suggest that the latter is the result of the former, which is not tested. It would be crucial to either strongly emphasize the absence of connection between the two stories (including likely remove reference to the LCL-LDs in the title), find a link between the stories (e.g. show that LCL-LDs are not made with the 10 a.a. mutant), or remove this part entirely to keep it for another, more focused manuscript.

4. The effect of Upc2 on Hmg1 is unclear. The paragraph starts with the observation of increased Hmg1 accumulation in AGR, then looks for factor affecting this induction and homes in on Upc2. There, however, instead of protein accumulation, the subcellular localization is assessed. Does Upc2 deletion blunt Hmg1 induction in AGR or does it only affect its localization? If the latter, is it because of a general imbalance in the sterol biosynthesis pathway, akin to lovastatin-treated cells? Does it also affect Njv1 localization to the vacuole, or does it affect Hmg1-Nvj1 interaction (if not constitutive, see above)? While deciphering the exact mechanistic role of Upc2 in this context is probably beyond the scope of this manuscript, it should however be discussed clearly in the text.

5. Seo and colleagues (eLife 2016) showed that AGR triggers LD autophagy. While the two phenotypes observed are not incompatible, the apparent differences of AGR in LD formation and/or consumption should at least be discussed.

6. The reduced ability of nvj1 deleted cells to cope with AGR is attributed to reduced sterylester synthesis. Since Nvj1-delete cells might be suffering from several other ailments, a more direct way to test the importance of sterylester biosynthesis in AGR survival could be to look at are1,2 deleted cells.

Please also have a careful look at and address the points of the individual reviews appended below.

Reviewer #1:

This study reports that acute glucose restriction (AGR) causes the sequestration of HMG-Coa reductases at Nucleus-Vacuole Junctions (NVJs), contact sites between the nuclear ER and lysosomal membranes in yeast. This sequestration is shown to enhance the metabolic activity of the enzyme, something that is beneficial to survive AGR. In a second part, AGR is shown to cause phase separation of steryl-esters in lipid droplets, but the connection to NVJ and HM-Coa reductase sequestration is unclear. The authors hammer their points home with elegant genetics where they artificially cluster HMG-Coa reductase and remove any transcriptional influence.

This is a very interesting story that shows that mere subcellular repositioning of enzymes can affect their activity and change metabolic fluxes. It is one of the first study that pinpoints a role for membrane contact sites beyond the classical lipid and calcium exchanges.

Below are some ideas to strengthen the main claims of the paper:

1. Throughout, the importance of HMG-CoA reductase sequestration is assessed by comparing wild-type strains to a NVJ1-delete strain, which is incapable of sequestering HMG-CoA reductase to NVJs. However the deletion of Nvj1 affects much more than HMG-CoA sequestration. It affects the whole making of NVJs and the sequestration therein of several factors which could have indirect roles in the phenotypes observed, for instance Osh1, Vps13 and Tsc13. Incidentally, the authors present a much cleaner way to assess the role of HMG-CoA reductase sequestration; a mutant lacking just 5 amino acids from Nvj1, which is still able to make NVJs and presumably still able to recruit Osh1, Vps13 and Tsc13 therein. It would be much more informative if all experiments using NVJ1-delete cells were performed while rescuing the phenotype with either WT Nvj1 or the 5-aminoacid deletion thereof. This would tease apart what from the phenotype is due to a lack in HMG CoA reductase sequestration and what is due to lack of NVJs in general.

2. There is a confusion regarding Upc2's activity in regulating Hmg1. It is proposed that this transcription factor is responsible for the increased accumulation of Hmg1 in AGR. Yet, the test for this idea, does not look at Hmg1 accumulation, but at its distribution. Accumulation and distribution are two separate processes that can be teased apart as shown in figure 6A-C. It appears, from figure 4E that Upc2 rather affects the distribution of Hmg1. We do not know if it affects its accumulation, since a western blot (or any other quantification) has not been performed. Even if Upc2 was necessary for transcriptional induction of Hmg1, a change in distribution of the preexisting pool of Hmg1 should be observed, unless only the newly synthesized pool is susceptible to sequestration. This should be tested, and is not quite consistent with the experiments where Hmg1 is under the control of the ADH1 promoter. An alternate hypothesis is that Upc2 is required for the expression of another factor that is necessary for Hmg1 sequestration. This hypothesis appears quite likely and can be tested by deleting UPC2 in the strain where Hmg1 is under the Adh1 promoter.

3. The experiments with artificial clustering are only valid if the different fusion proteins are expressed at the same levels. This is not tested.

Reviewer #2:

In this manuscript, Rogers et al. show how in response to acute glucose restriction, HMG-CoA Reductases re-localize from the nuclear envelope to the nucleus vacuole junction (NVJ), a contact site at which the nucleus and vacuole are physically attached to each other. They describe in detail the molecular determinants underlying this behavior, and uncover a functional role of spatial enzyme compartmentalization in regulation of flux through the mevalonate pathway, and a link to formation of liquid-crystalline layered sterol ester-rich lipid droplets.

These are conceptually novel findings that will likely be of high interest to the broad readership of eLife. The data is of very high quality, and supports the claims raised by the authors.

I support publication of this exciting manuscript.

Reviewer #3:

This manuscript describes that during acute glucose restriction (AGR) the rate limiting enzyme of the mevalonate pathway Hmg1 concentrates at the NVJ. Hmg1 relocalization requires the NVJ components Nvj1, Vac8, the ergosterol-sensing transcription factor Upc2 and the activity of the sterol enzyme Erg1. Hmg1 partitioning is mediated through its transmembrane domain and requires a 10 amino acid region in Nvj1 luminal domain. It is claimed that Hmg1 NVJ partitioning promotes increased flux through the mevalonate pathway and accumulation of steryl ester to promote growth during AGR recovery. Some of the observation described are very interesting. Moreover, understanding how metabolic pathways and membrane contacts sites are rewired in response to environmental changes is an issue of fundamental importance. However, the study presents a number of critical flaws and some of the claims are not convincingly demonstrated.

1. A major assumption of this study is that Hmg1 does not normally function as a tetramer as mammalian HMGR. In mammals HMGR tetramerization is mediated by its cytosolic catalytic domain which is more than 60% similar to Hmg1. This complicates the interpretation of several experiments. For example, the authors the authors use monomeric mRuby and tetrameric DsRed to "manipulate" the oligomeric state Hmg1 but there is really no information on the oligomeric state of these fusion proteins. The results presented in Figure 6 can easily be explained by differences in the levels or functionality of the two fusion proteins (not tested in both cases).

2. It is nicely shown that Hmg1 NVJ localization requires its membrane domain and a few residues in the luminal domain of Nvj1. But how exactly is this regulated? Do these proteins interact? If yes, is this interaction constitutive and driven by the increase in Nvj1 upon AGR? or is it regulated in response to AGR?

3. Still related with the previous point, the trigger for Hmg1 NVJ localization is unclear. The upc2D and terbinafine results suggest the flux through the pathway is important but some additional information would be good. There are drugs/mutants affecting almost all the steps along the pathway so this could be easily resolved.

4. In figure 5A-B, it is shown that inhibition of Hmg1 with Lovastatin reverses the changes in steryl esters as expected. Does this treatment also affect Hmg1 NVJ accumulation and/or Nvj1 distribution?

5. All the measurements of mevalonate pathway intermediates and products were performed in cells expressing Hmg2 and the levels of the protein under the conditions tested is unknown. Is it possible that Nvj1 affects the levels of Hmg2? To avoid this complications, the same measurements should have been done in hmg2D cells. Also the data shown in Figure 5 and 6 does not show that the effect is specific to AGR

6. The authors propose that steryl esters are necessary for optimal AGR response. However, this possibility was not formally tested by preventing steryl ester formation, which can be easily done by deleting Are1 and Are2 acyl transferases.

7. In Figure 6I, why does Hmg1-DsRed2 responds to AGR even in nvj1D cells. Presumably the tetramerization of DsRed is independent of glucose levels.

8. The data on LCL-LDs is very interesting but it does not seem to belong to this story.

9. Seo and colleagues (eLife 2016) showed that AGR triggers LD autophagy. While the two phenotypes observed are not incompatible, the apparent differences of AGR in LD formation and/or consumption should at least be discussed.

https://doi.org/10.7554/eLife.62591.sa1

Author response

All three reviewers found the work exciting and important. There are nevertheless ambiguities in the present manuscript that need to be addressed, and a decision taken about what to make of the side observation with liquid crystalline structures in LDs (see below).

Points that require new experimental work:

1. The first point is about the artificial clustering. This is a very important point of the study, but the result is however yet not entirely clear.

– The human HMG-CoA reductase works as a tetramer. The oligomerization state of yeast Hmg1 is unclear. There is a strong focus on the manuscript on multimerization without data supporting it. Monomeric mRuby and tetrameric DsRed are used to manipulate the oligomeric state Hmg1 but there is no information on the oligomeric state of these fusion proteins. Moreover it is not tested whether the fusion to different tags affect the protein's activity and/or abundance. In order to make the point that it is indeed the oligomerization status change that causes the change in activity, it is necessary to rule out changes in abundance or activity due to the tags being appended to the protein.

We thank the reviewers for this excellent point. We have now analyzed both Hmg1-mRuby3 and Hmg1-DsRed2 protein expression in WT and nvj1-KO backgrounds by Western blot, and found very similar trends for Hmg1-GFP levels in WT and nvj1-KO backgrounds. That is to say, Hmg1 expression increases in AGR very similarly in both strains, suggesting that our radio-pulse data cannot be simply explained by differential protein expression. Unfortunately, due to lack of antibodies recognizing untagged Hmg1, we cannot compare Hmg1 protein expression directly between Hmg1-mRuby3 and Hmg1-DsRed2 strains. We attempted several ways to do this, including use of multiple antibodies raised against human HMGCR, but these would not recognize yeast Hmg1. We also contacted other members of the yeast community that have studied Hmg1 in the past, but no antibodies were available.

Secondly, to address the point about Hmg1 oligomerization: (1) We now directly observe Hmg1 high molecular weight (HMW) species by blue native PAGE (Figure 7A,B). Importantly, these HMW species are induced by AGR, and are Nvj1-dependent. This supports a model where AGR induces Hmg1 to partition at the NVJ, which promotes Hmg1 incorporation and/or assembly into a HMW species. (2) To be as precise on our interpretation as possible, we have changed the phrasing used in this paper to describe Hmg1 entering into ‘High Molecular Weight Species’ rather than explicitly stating that these are Hmg1 tetramers.

Third, regarding the use of DsRed2 to modulate Hmg1 multimerization: we attempted several different methods to directly monitor Hmg1-DsRed2 oligomeric status. Indeed, by BN gels, we observed that Hmg1-DsRed2 ran at high molecular weights, but these gels were often of lower quality as the Hmg1-DsRed2 bands were quite smeary. We also noted that Hmg1-DsRed2 ran at a higher than expected molecular weight even in SDS-PAGE gels, indicating that this tag promoted SDS-resistant multimers. We have now added a comparative SDS-PAGE gel showing this and comparing it to Hmg1-mRuby3 in Supp Figure 5E. You can also note the smear-like pattern of the Hmg1-DsRed2 in Supp Fig5D.

Fourth, regarding any changes to Hmg1 abundance in different genetic backgrounds: We noted above that Hmg1-DsRed2 and Hmg1-mRuby3 both display comparative increases in protein abundance with AGR, suggesting these tags to not alter the AGR-induced protein increases for Hmg1. Additionally, because we now have an Nvj1 mutant (Nvj1 RK-28,29-AA) that exhibits mevalonate pathway defects, we quantified Hmg1-mRuby3 fluorescence intensities when expressed in Nvj1 WT and Nvj1 RK-28,29-AA backgrounds. The Hmg1-mRuby3 fluorescence intensity distributions are near identical in both backgrounds, suggesting the alterations to mevalonate pathway flux in the Nvj1 RK-28,29-AA background are not due to differences in Hmg1 protein levels (Figure 6—figure supplement 1C).

– It is unclear why AGR should cause stabilized Hmg1-dsRed2 multimers in nvj1-delete cells (lines 243-244, figure 6I). Here the multimerization is supposed to be constitutive and driven by the dsRed. The factor (Nvj1) that is supposed to cause AGR-dependent clustering of Hmg1 is absent. Does it mean that Hmg1 has a tendency to cluster during AGR in a way that is mechanistically unclear, and in order for it to happen efficiently, it needs to be boosted, either by being sequestered at the NVJ, or by being fused to a tetramerizing tag? This definitely requires further tests and explanations.

Thank you for pointing this out. We agree: our point here was that the DsRed2 tag is designed to generate constitutive multimers for Hmg1. Indeed, we observe that this DsRed2 tag promotes Hmg1-DsRed2 to run at high molecular weight even in SDS-PAGE gels in Log or AGR conditions (Figure 7—figure supplement 1E). We do see more Hmg1-DsRed2 foci by imaging during AGR, but this may be due to the increase in total Hmg1-DsRed2 protein induced by AGR, which is similar to other tags (Figure 7—figure supplement Supp 1D). We have amended the text to try to articulate this as precisely as possible. Our current model for how Hmg1 multimerizes, which we now state in the Discussion, is that NVJ compartmentalization of Hmg1/2 may promote the probability of these enzymes, which normally diffuse throughout the ER network, to engage each other and multimerize. In essence, the NVJ may serve as an ER sub-domain where enzymes may have increased probability to engage one another.

2. The second point is Nvj1's influence on Hmg1.

– One likely explanation that is strongly implied is that Nvj1 and Hmg1 interact via the 10 a.a. stretch identified therein. Can this interaction be shown by pull-down? Is the interaction induced by AGR, or is it constitutive (in which case Hmg1 simply follows Nvj1 relocalisation to NVJs in AGR)? Also several conditions prevent Hmg1 recruitment and sequestration to NVJs (e.g. lovastatin, upc2 deletion). Do these conditions affect the making of NVJs and the localisation of Nvj1 therein or are they specific to Hmg1 recruitment?

Thank you for this suggestion. Working with Hmg1 in biochemical experiments has proven very challenging as Hmg1 is prone to aggregation. Thus, we were concerned with co-IP data. However, we now present new data showing that mutation of Nvj1 residues RK-28,29 to alanine (RK— 28,29—AA) inhibits Hmg1 recruitment to the NVJ while maintaining NVJ patch formation (Figure 3G-I). This Nvj1 mutant enabled us to show that loss of Hmg1 NVJ recruitment perturbs mevalonate pathway flux (Figure 6A-D), and growth resumption after glucose starvation (Figure 6E,F) even while the NVJ is intact. We also now show that loss of Upc2 does not affect NVJ formation (Figure 4—figure supplement 1A-C). We did not use lovastatin for these experiments. We also now comment in the Discussion that we cannot rule out the fact that the Nvj1-Hmg1 interaction may be indirect via another protein, but fully elucidating this interaction is beyond the scope of the study.

– There is a possibility that the effects of the Nvj1 deletions are indirect through the absence of NVJ, and the absence of recruitment therein of other lipid homeostasis factors (e.g. Osh1, Tsc13, Vps13). It is however possible to decouple NVJ generation and Hmg1 recruitment by using the 10 a.a. deletion mutant. Using such a mutant would be much cleaner and ideally, all experiments using Nvj1 delete cells should be replaced or complemented with the 10 a.a. mutant. This would imply a large amount of work, and the reviewers discussed that it should be sufficient to just do the growth assays, as well as the data in figure 5A-G with this setup (see also point below).

As mentioned above, we now show that a mutant Nvj1 (Nvj1 RK-28,29-AA) shows very similar defects in mevalonate pathway flux AND growth resumption in a background where the NVJ is still intact (Figure 6A-F). This strongly suggests that the effects we observe are due to Hmg1 not being recruited to the NVJ, and not from the general loss of the NVJ itself.

– It is emphasized throughout that Nvj1's influence on Hmg1 activity is limited to AGR conditions. This is consistent with Hmg1 sequestrations at NVJs only in these conditions. It is however not tested.

All the data from figure 5C-G and 6D-G,J-M are in AGR only. Yet, Nvj1 might also influence the mevalonate pathway in log phase. Since it will be anyway necessary to redo the experiments of figure 5A-G with the 10 a.a. nvj1 mutant (see point above), it will be necessary to add log phase quantification. If nvj1 mutation or deletion had an effect on log phase, a condition where Hmg1 is not clustered at NVJs, then this would imply a much more indirect mechanism. To make the claim water-tight, these experiments should ideally be performed in an Hmg2-delete background to discount any variations in Hmg2 expression during AGR.

We attempted to monitor mevalonate pathway flux during Log phase via radiolabel pulse-chase experiments. Surprisingly, the uptake of 14C-acetate is very low in Log phase. Yeast in Log phase growth thus exhibit very low 14C-acetate incorporation into mevalonate pathway products. For comparison, we’ve attached a cpm histogram for yeast in Log and AGR phases in Author response image 1. This low 14C-acetate uptake made monitoring 14-C labeled mevalonate pathway products essentially impossible.

Author response image 1

Regarding whether nvj1 mutation or deletion has an effect in Log phase, it should be noted that the Nvj1 RK-28,29-AA mutant experiments in AGR demonstrate that defects in mevalonate pathway flux still occur when the NVJ is intact. This also addresses the indirect effect issue.Regarding variation from Hmg2 expression, we have conducted experiments in our artificial strain that only expresses Hmg1-GFP on an artificial promoter (Figure 6H-M), as well as a strain background expressing only Hmg1 (Hmg2 also deleted here) with a mutant Nvj1 that cannot recruit Hmg1 (Figure 6A-D). Both experiment sets indicate that loss of Hmg1 recruitment to the NVJ perturbs mevalonate pathway flux.

Points that might be addressed by amendments in the text, or by new experimental work:

3. The part on Liquid crystalline layered LDs is disconnected from the rest. It is not tested whether Nvj1-dependent clustering of Hmg1 to NVJs is necessary, sufficient or unrelated to the formation of these structure. While all three reviewers found it to be an interesting observation, they found that, as presently written, the manuscript is ambiguous at stating the lack of connection between the two phenomena. The title itself "Glucose restriction drives spatial re-organization of mevalonate metabolism and liquid-crystalline lipid droplet biogenesis" could suggest that the latter is the result of the former, which is not tested. It would be crucial to either strongly emphasize the absence of connection between the two stories (including likely remove reference to the LCL-LDs in the title), find a link between the stories (e.g. show that LCL-LDs are not made with the 10 a.a. mutant), or remove this part entirely to keep it for another, more focused manuscript.

We thank the reviewers for these constructive comments. We have now removed the data related to the liquid-crystalline LDs in this revision. The revision is thus focused on dissecting the role of Hmg1 compartmentalization on mevalonate pathway flux during AGR.

4. The effect of Upc2 on Hmg1 is unclear. The paragraph starts with the observation of increased Hmg1 accumulation in AGR, then looks for factor affecting this induction and homes in on Upc2. There, however, instead of protein accumulation, the subcellular localization is assessed. Does Upc2 deletion blunt Hmg1 induction in AGR or does it only affect its localization? If the latter, is it because of a general imbalance in the sterol biosynthesis pathway, akin to lovastatin-treated cells? Does it also affect Njv1 localization to the vacuole, or does it affect Hmg1-Nvj1 interaction (if not constitutive, see above)? While deciphering the exact mechanistic role of Upc2 in this context is probably beyond the scope of this manuscript, it should however be discussed clearly in the text.

We now show that Upc2 deletion does not alter the behavior of Hmg1 protein levels (Figure 4figure supplement 1D). We also now show that Upc2-ko does not impact NVJ formation or Nvj1 levels (Figure 4—figure supplement 1A-C). We agree that completely dissecting the interplay between Upc2, Nvj1, and Hmg1 is beyond the scope of this manuscript, but have revised the text to say that the full mechanism will be explored in future studies.

5. Seo and colleagues (eLife 2016) showed that AGR triggers LD autophagy. While the two phenotypes observed are not incompatible, the apparent differences of AGR in LD formation and/or consumption should at least be discussed.

Thank you for this important discussion point. We now discuss the relationship between glucosestarvation induced autophagy and glucose restriction in the Discussion. We speculate that HMGCR partitioning at the NVJ may impact aspects of LD biogenesis or composition, which may ultimately impact the micro-autophagy of LDs. These are a part of ongoing studies.

6. The reduced ability of nvj1 deleted cells to cope with AGR is attributed to reduced sterylester synthesis. Since Nvj1-delete cells might be suffering from several other ailments, a more direct way to test the importance of sterylester biosynthesis in AGR survival could be to look at are1,2 deleted cells.

We find that the growth resumption defects associated with nvj1-ko can be rescued by culturing the yeast with mevalonate, the enzymatic product of HMGCRs. Therefore, we attribute the defects with perturbed mevalonate synthesis itself, rather than sterol-ester synthesis, per se. Mevalonate is a key metabolite needed for several metabolic pathways including sterol metabolism but also coenzyme Q and heme production, which are needed for mitochondrial metabolism. We speculate that growth resumption defects may be related to mitochondrial metabolism, but fully resolving this is beyond the scope of this current study.

Please also have a careful look at and address the points of the individual reviews appended below.

Reviewer #1:

This study reports that acute glucose restriction (AGR) causes the sequestration of HMG-Coa reductases at Nucleus-Vacuole Junctions (NVJs), contact sites between the nuclear ER and lysosomal membranes in yeast. This sequestration is shown to enhance the metabolic activity of the enzyme, something that is beneficial to survive AGR. In a second part, AGR is shown to cause phase separation of steryl-esters in lipid droplets, but the connection to NVJ and HM-Coa reductase sequestration is unclear. The authors hammer their points home with elegant genetics where they artificially cluster HMG-Coa reductase and remove any transcriptional influence.

This is a very interesting story that shows that mere subcellular repositioning of enzymes can affect their activity and change metabolic fluxes. It is one of the first study that pinpoints a role for membrane contact sites beyond the classical lipid and calcium exchanges.

Below are some ideas to strengthen the main claims of the paper:

1. Throughout, the importance of HMG-CoA reductase sequestration is assessed by comparing wild-type strains to a NVJ1-delete strain, which is incapable of sequestering HMG-CoA reductase to NVJs. However the deletion of Nvj1 affects much more than HMG-CoA sequestration. It affects the whole making of NVJs and the sequestration therein of several factors which could have indirect roles in the phenotypes observed, for instance Osh1, Vps13 and Tsc13. Incidentally, the authors present a much cleaner way to assess the role of HMG-CoA reductase sequestration; a mutant lacking just 5 amino acids from Nvj1, which is still able to make NVJs and presumably still able to recruit Osh1, Vps13 and Tsc13 therein. It would be much more informative if all experiments using NVJ1-delete cells were performed while rescuing the phenotype with either WT Nvj1 or the 5-aminoacid deletion thereof. This would tease apart what from the phenotype is due to a lack in HMG CoA reductase sequestration and what is due to lack of NVJs in general.

Thank you. As discussed above, we now show that Nvj1 mutant RK-28,29-AA, which can form an NVJ but does not recruit Hmg1, manifests defects in mevalonate pathway flux (Figure 6A-D) and growth resumption following glucose starvation (Figure 6E,F). This indicates that these effects are specific for Hmg1 recruitment, and not due to general loss of the NVJ.

2. There is a confusion regarding Upc2's activity in regulating Hmg1. It is proposed that this transcription factor is responsible for the increased accumulation of Hmg1 in AGR. Yet, the test for this idea, does not look at Hmg1 accumulation, but at its distribution. Accumulation and distribution are two separate processes that can be teased apart as shown in figure 6A-C. It appears, from figure 4E that Upc2 rather affects the distribution of Hmg1. We do not know if it affects its accumulation, since a western blot (or any other quantification) has not been performed. Even if Upc2 was necessary for transcriptional induction of Hmg1, a change in distribution of the preexisting pool of Hmg1 should be observed, unless only the newly synthesized pool is susceptible to sequestration. This should be tested, and is not quite consistent with the experiments where Hmg1 is under the control of the ADH1 promoter. An alternate hypothesis is that Upc2 is required for the expression of another factor that is necessary for Hmg1 sequestration. This hypothesis appears quite likely and can be tested by deleting UPC2 in the strain where Hmg1 is under the Adh1 promoter.

We now show that upc2-ko does not effect Hmg1 protein levels (Figure 4—figure supplement 1D). We also now show that upc2-ko does not alter NVJ formation nor Nvj1 levels (Figure 4—figure supplement 1A-C). Regarding other factors Upc2 may regulate the expression of, we agree, and have modified the text to state that the full mechanism of Upc2’s role in the Nvj1-Hmg1 interaction requires further study beyond the scope of this study.

3. The experiments with artificial clustering are only valid if the different fusion proteins are expressed at the same levels. This is not tested.

We now provide evidence that the Hmg1 protein levels undergo the same changes in abundance during Log and AGR, suggesting these differences are not due to changes in abundance. Furthermore, our new experiments with Nvj1 RK-28,29-AA show that loss of Hmg1 recruitment to the NVJ causes defects in mevalonate pathway flux. These yeast contain the same Hmg1-mRuby3 protein levels (Figure 6—figure supplement 1C), suggesting that alterations in Hmg1 partitioning, and not protein level, cause defects im mevaloante pathway flux.

Reviewer #3:

This manuscript describes that during acute glucose restriction (AGR) the rate limiting enzyme of the mevalonate pathway Hmg1 concentrates at the NVJ. Hmg1 relocalization requires the NVJ components Nvj1, Vac8, the ergosterol-sensing transcription factor Upc2 and the activity of the sterol enzyme Erg1. Hmg1 partitioning is mediated through its transmembrane domain and requires a 10 amino acid region in Nvj1 luminal domain. It is claimed that Hmg1 NVJ partitioning promotes increased flux through the mevalonate pathway and accumulation of steryl ester to promote growth during AGR recovery. Some of the observation described are very interesting. Moreover, understanding how metabolic pathways and membrane contacts sites are rewired in response to environmental changes is an issue of fundamental importance. However, the study presents a number of critical flaws and some of the claims are not convincingly demonstrated.

1. A major assumption of this study is that Hmg1 does not normally function as a tetramer as mammalian HMGR. In mammals HMGR tetramerization is mediated by its cytosolic catalytic domain which is more than 60% similar to Hmg1. This complicates the interpretation of several experiments. For example, the authors the authors use monomeric mRuby and tetrameric DsRed to "manipulate" the oligomeric state Hmg1 but there is really no information on the oligomeric state of these fusion proteins. The results presented in Figure 6 can easily be explained by differences in the levels or functionality of the two fusion proteins (not tested in both cases).

Thank you for this important point. First, as discussed above, we now present data that Hmg1 enters into high molecular weight (HMW) assemblies that we observe by Blue Native Gel (Figure 7A,B). These HMW assemblies are induced by AGR, and are Nvj1 dependent. Second, we have now amended our text to say that Hmg1 “enters into high molecular weight assemblies” rather than make any claims that Hmg1 forms tetramers. Third, we have conducted new Western blots showing that Hmg1-mRuby3 and Hmg1-DsRed2 tagged proteins undergo very similar protein increased from Log to AGR conditions, suggesting tagging does not affect overall protein levels. Fourth, we now provide evidence that Hmg1-DsRed2 runs at higher than predicted MWs even in SDS-PAGE gels, indicating the DsRed2 tetramer promotes oligomerization (Figure 7—figure supplement 1E). Fifth, we also now show that in a yeast background where the Hmg1 protein levels are the same (Figure 6—figure supplement 1C), when Hmg1 cannot be recruited to the NVJ (as in Nvj1 RK-28,29-AA), there are defects in mevalonate pathway flux (Figure 6A-D).

2. It is nicely shown that Hmg1 NVJ localization requires its membrane domain and a few residues in the luminal domain of Nvj1. But how exactly is this regulated? Do these proteins interact? If yes, is this interaction constitutive and driven by the increase in Nvj1 upon AGR? or is it regulated in response to AGR?

As discussed above, we now present data that this Nvj1-Hmg1 interaction requires residues RK28,29, and use a double alanine mutant to show that this mutant cannot recruit Hmg1, has defects in mevalonte metabolism flux, but still forms an NVJ (Figure 6A-D, Figure 3H,I). We have also modified the text to state that we do not fully understand this mechanism of Upc2-Nvj1-Hmg1 interaction, but this is part of an ongoing investigation beyond the scope of this manuscript.

3. Still related with the previous point, the trigger for Hmg1 NVJ localization is unclear. The upc2D and terbinafine results suggest the flux through the pathway is important but some additional information would be good. There are drugs/mutants affecting almost all the steps along the pathway so this could be easily resolved.

We now show that upc2-ko does not affect Hmg1 protein levels (Figure 6—figure supplement 1D). We also more further defined the interaction as requiring Nvj1 residues RK-28,29. As mentioned above, we do not fully understand the mechanism of this interaction, but it is beyond the scope of this current paper.

4. In figure 5A-B, it is shown that inhibition of Hmg1 with Lovastatin reverses the changes in steryl esters as expected. Does this treatment also affect Hmg1 NVJ accumulation and/or Nvj1 distribution?

We did not comment on this in the manuscript, but this treatment does not affect Hmg1 recruitment to the NVJ.

5. All the measurements of mevalonate pathway intermediates and products were performed in cells expressing Hmg2 and the levels of the protein under the conditions tested is unknown. Is it possible that Nvj1 affects the levels of Hmg2? To avoid this complications, the same measurements should have been done in hmg2D cells. Also the data shown in Figure 5 and 6 does not show that the effect is specific to AGR

Thank you. We have now done controlled mevalonate pathway flux experiments in a background with only Hmg1-GFP under a controlled ADH promoter (Figure 6J-M). We have also done these experiments in a mutant Nvj1 RK-28,29-AA that does not have Hmg1 (only Hmg1 here), and an intact NVJ is present. Here, there are still defects in mevalonate pathway flux, and Hmg1-mRuby3 protein levels are the same in Nvj1 WT and Nvj1 RK-28,29-AA backgrounds (Figure 6A-D; Figure 6figure supplement 1C). This strongly suggests that defects in mevalonate pathway flux and not due to Hmg1 protein levels, nor from Hmg2 alterations.

6. The authors propose that steryl esters are necessary for optimal AGR response. However, this possibility was not formally tested by preventing steryl ester formation, which can be easily done by deleting Are1 and Are2 acyl transferases.

As discussed above, we have modified the text to suggest that defects in AGR response my be due to defects in mevalonate production, which is necessary for many aspects of lipid and mitochondrial metabolism. In line with this, addition of mevalonate rescues NVJ1-ko associated growth defects following AGR (Figure 5H). Additionally, we now show that the Nvj1 RK-28,29-AA, which has an NVJ but cannot recruit Hmg1 to it, and has defects in mevalonate pathway flux, also has growth resumption defects (Figure 6E,F).

7. In Figure 6I, why does Hmg1-DsRed2 responds to AGR even in nvj1D cells. Presumably the tetramerization of DsRed is independent of glucose levels.

We do not completely understand this, but it is possible that the organization of the nuclear envelope and/or cytoplasm changes during AGR. It is also known that the fluidity of the yeast cytoplasm changes during glucose starvation, which may affect general macromolecular mobility, and could induce clustering of high molecular weight complexes (Joyner, eLife, 2016; https://elifesciences.org/articles/09376). This is part of an ongoing investigation.

8. The data on LCL-LDs is very interesting but it does not seem to belong to this story.

Thank you. We have removed this section and put in another ongoing project.

9. Seo and colleagues (eLife 2016) showed that AGR triggers LD autophagy. While the two phenotypes observed are not incompatible, the apparent differences of AGR in LD formation and/or consumption should at least be discussed.

Thank you. We now discuss the Seo et al. study and its relationship to our work in the Discussion.

https://doi.org/10.7554/eLife.62591.sa2

Article and author information

Author details

  1. Sean Rogers

    Department of Cell Biology, University of Texas Southwestern Medical Center, Dallas, United States
    Contribution
    Conceptualization, Data curation, Formal analysis, Visualization, Methodology, Writing - original draft
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0003-1173-6526
  2. Hanaa Hariri

    1. Department of Cell Biology, University of Texas Southwestern Medical Center, Dallas, United States
    2. Department of Biological Sciences, Wayne State University, Detroit, United States
    Contribution
    Conceptualization, Data curation, Supervision
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0002-2953-7536
  3. N Ezgi Wood

    Department of Cell Biology, University of Texas Southwestern Medical Center, Dallas, United States
    Contribution
    Conceptualization, Data curation, Formal analysis, Visualization, Methodology
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0002-1874-9226
  4. Natalie Ortiz Speer

    Department of Cell Biology, University of Texas Southwestern Medical Center, Dallas, United States
    Contribution
    Conceptualization, Visualization, Methodology
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0002-0982-3025
  5. W Mike Henne

    Department of Cell Biology, University of Texas Southwestern Medical Center, Dallas, United States
    Contribution
    Conceptualization, Formal analysis, Supervision, Funding acquisition, Investigation, Writing - original draft, Project administration
    For correspondence
    mike.henne@utsouthwestern.edu
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0002-2135-2799

Funding

Welch Foundation (I-1873)

  • W Mike Henne

National Institute of General Medical Sciences (GM119768)

  • W Mike Henne

National Institute of General Medical Sciences (5T32GM008297)

  • Sean Rogers

National Institute of Diabetes and Digestive and Kidney Diseases (DK126887)

  • W Mike Henne

The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.

Acknowledgements

The authors want to thank members of the Henne and Nicastro labs for help in the completion of this project. The authors thank Joel Goodman for the tgl3,4,5,Δ yeast strain. W.M.H. is supported by funds from the Welch Foundation (I-1873), the NIH NIGMS (GM119768), the Ara Parseghian Medical Research Fund, and the UT Southwestern Endowed Scholars Program. S.R. is supported in part by a NIH T32 training grant (5T32GM008297).

Senior Editor

  1. Suzanne R Pfeffer, Stanford University School of Medicine, United States

Reviewing Editor

  1. Benoît Kornmann, University of Oxford, United Kingdom

Reviewer

  1. Benoît Kornmann, University of Oxford, United Kingdom

Publication history

  1. Received: August 29, 2020
  2. Accepted: April 6, 2021
  3. Accepted Manuscript published: April 7, 2021 (version 1)
  4. Version of Record published: April 20, 2021 (version 2)

Copyright

© 2021, Rogers et al.

This article is distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use and redistribution provided that the original author and source are credited.

Metrics

  • 1,730
    Page views
  • 242
    Downloads
  • 0
    Citations

Article citation count generated by polling the highest count across the following sources: Crossref, PubMed Central, Scopus.

Download links

A two-part list of links to download the article, or parts of the article, in various formats.

Downloads (link to download the article as PDF)

Download citations (links to download the citations from this article in formats compatible with various reference manager tools)

Open citations (links to open the citations from this article in various online reference manager services)

  1. Further reading

Further reading

    1. Cancer Biology
    2. Cell Biology
    Kerrie-Ann McMahon et al.
    Research Article

    Caveolae-associated protein 3 (cavin3) is inactivated in most cancers. We characterized how cavin3 affects the cellular proteome using genome-edited cells together with label-free quantitative proteomics. These studies revealed a prominent role for cavin3 in DNA repair, with BRCA1 and BRCA1 A-complex components being downregulated on cavin3 deletion. Cellular and cell-free expression assays revealed a direct interaction between BRCA1 and cavin3 that occurs when cavin3 is released from caveolae that are disassembled in response to UV and mechanical stress. Overexpression and RNAi-depletion revealed that cavin3 sensitized various cancer cells to UV-induced apoptosis. Supporting a role in DNA repair, cavin3-deficient cells were sensitive to PARP inhibition, where concomitant depletion of 53BP1 restored BRCA1-dependent sensitivity to PARP inhibition. We conclude that cavin3 functions together with BRCA1 in multiple cancer-related pathways. The loss of cavin3 function may provide tumor cell survival by attenuating apoptotic sensitivity and hindering DNA repair under chronic stress conditions.

    1. Cell Biology
    2. Neuroscience
    Haining Zhong et al.
    Tools and Resources Updated

    Precise and efficient insertion of large DNA fragments into somatic cells using gene editing technologies to label or modify endogenous proteins remains challenging. Non-specific insertions/deletions (INDELs) resulting from the non-homologous end joining pathway make the process error-prone. Further, the insert is not readily removable. Here, we describe a method called CRISPR-mediated insertion of exon (CRISPIE) that can precisely and reversibly label endogenous proteins using CRISPR/Cas9-based editing. CRISPIE inserts a designer donor module, which consists of an exon encoding the protein sequence flanked by intron sequences, into an intronic location in the target gene. INDELs at the insertion junction will be spliced out, leaving mRNAs nearly error-free. We used CRISPIE to fluorescently label endogenous proteins in mammalian neurons in vivo with previously unachieved efficiency. We demonstrate that this method is broadly applicable, and that the insert can be readily removed later. CRISPIE permits protein sequence insertion with high fidelity, efficiency, and flexibility.