Fatty acid trafficking to mitochondria and peroxisomes in Tetrahymena thermophila, a new frontier for lipid droplet biology

Abstract

Lipid droplets are increasingly recognized as necessary organelles. However, the cellular pathways that regulate lipid droplets have only been defined in select fungi, algae, plants, and animals. Our experiments expand the study of lipid droplets to an evolutionarily distinct model organism, the ciliate Tetrahymena thermophila. We identify conserved pathways that promote lipid droplet homeostasis while also uncovering features that suggest adaptation. We show that Tetrahymena accumulate lipid droplets in response to nutrient deprivation, including starvation and the stationary phase. Pulse-chase experiments with a fluorescent fatty acid analogue demonstrate lipid trafficking to lipid droplets in starved cultures. Unlike other cell types, starved Tetrahymena appear to use both peroxisomes and mitochondria (not vacuoles) for further fatty acid catabolism. We observe cooccurence of the fluorescent fatty acid analogue with markers of peroxisomes and a subpopulation of mitochondria, suggesting specialized catabolic roles for both organelles. We demonstrate a decrease in survival following starvation in the presence of inhibitors of mitochondrial fatty acid import or peroxisomal fatty acid metabolism. Together, our experiments add Tetrahymena to the expanding list of eukaryotes that increase lipid droplets in response to nutrient depletion while also uncovering important and distinct roles for mitochondrial and peroxisomal catabolism in survival pathways.

• Lipid droplets are crucial for cellular energy storage, but detailed knowledge of their synthesis and degradation is limited to a few evolutionarily related organisms. This leaves a significant gap in understanding how diverse life forms manage lipid metabolism.

• Using the evolutionarily distinct ciliate Tetrahymena thermophila, the authors show that lipid droplets accumulate during nutrient deprivation. Fatty acids traffic to mitochondria and peroxisomes in starved Tetrahymena, and both organelles are essential for survival.

• This work establishes a new model organism for studying lipid biology and opens new avenues for research into the specialized roles of organelle subpopulations in cellular energy regulation.

INTRODUCTION

All eukaryotic cells, and some bacteria, have the capacity to store lipid in intracellular lipid droplets. Lipid droplet homeostasis impacts biofuel and food production, metabolic health and disease, and cellular responses to infection and nutritional challenge (reviewed in Henne et al., 2018). A growing recognition of the importance of lipid droplets has expanded investigation of the mechanisms that form and degrade lipid droplets and the cellular processes that require stored lipids, including examples of bacteria, fungi, animals, diatoms, plants, and algae (Lundquist et al., 2020).

This research shares the first analysis of lipid storage mechanisms in Tetrahymena thermophila. Despite the recognition that lipid droplets form in diverse cell types, all previous knowledge of the pathways that control lipid droplet homeostasis derives from research on organisms that together represent only a few branches on the evolutionary tree of life. Tetrahymena have been classified as belonging to the SAR (Stramenopile–Alveolata–Rhizaria) evolutionary supergroup with other mostly unicellular, free-living or parasitic organisms (Adl et al., 2012). The SAR supergroup is notably distinct from the Opisthokonta (including fungi and animals) or Archaeplastida (including plants and algae) lineages. For context, the Alveolata and Opisthokonta shared a last common ancestor ∼1.5 billion years ago (Adl et al., 2012; Eme et al., 2014). Identification of lipid storage mechanisms in Tetrahymena thus offers an opportunity to both affirm the importance of such pathways in all living systems and to uncover alternative metabolic strategies brought about by evolution.

Recent experiments have discovered that lipid droplets form under conditions of cell stress, including (paradoxically) nutrient deprivation. Yeasts build lipid droplets in response to acute loss of glucose (Seo et al., 2017). Diatoms and microalgae, like Chlamydomonas, accumulate lipid droplets when nitrogen or phosphorus is depleted (Minhas et al., 2016; Leyland et al., 2020). Lipid droplets appear in Arabidopsis leaves following periods of darkness or nitrogen deprivation (Gidda et al., 2016; Coulon et al., 2020). Cultured mammalian cells accumulate lipid droplets when starved of amino acids (Nguyen et al., 2017). Although the initial trigger for lipid droplet synthesis is cell type dependent (e.g., loss of glucose, nitrogen, or amino acids), these reports together contribute to a growing recognition of the importance of lipid droplets in the cellular response to nutrient deficiency.

Lipid droplets may fulfill two important functions in a starving cell. First, lipid droplets may sequester lipotoxic species released upon autophagic digestion of cellular membranes. Second, lipid droplets may supply lipid substrates to catabolic reactions within cells, thus fueling continued cell activity. Rambold and colleagues have shown that this second function is critically important in nutrient-deprived mouse embryonic fibroblasts, where hydrolysis of neutral lipid stored in lipid droplets supports mitochondrial fatty acid oxidation (Rambold et al., 2015).

The reactions in β-oxidation that are responsible for fatty acid catabolism are located in distinct organelles in different organisms. Plants and yeast cells perform β-oxidation exclusively in peroxisomes, whereas mammalian cells have enzymes for β-oxidation in both mitochondria and peroxisomes. In mammals, evidence suggests that mitochondria and peroxisomes have specialized to catabolize lipids of different lengths (reviewed in Kunau et al., 1995; Kumar et al., 2024). Here, we examine the role of lipid catabolism in both peroxisomes and mitochondria in the survival of starved Tetrahymena. We provide evidence that, like mammalian cells (Rambold et al., 2015), starved Tetrahymena traffic fatty acids to mitochondria with a time course consistent with prior incorporation into lipid droplets. We additionally uncover fatty acid trafficking to peroxisomes in starved Tetrahymena. Furthermore, we show that peroxisomes are important for cell survival in starved cells, but not in cells growing under nutrient-rich conditions. Together, our data demonstrate that Tetrahymena utilize both peroxisomal and mitochondrial fatty acid catabolism to support cell survival when nutritionally challenged.

RESULTS AND DISCUSSION

Tetrahymena accumulate lipid droplets following nutrient deprivation

Lipid droplets accumulate in cells exposed to nutritional stress (reviewed in Henne et al., 2018). To establish T. thermophila as a new model for the study of lipid droplet homeostasis, we starved cells in 10 mM Tris, pH 7.5, and followed lipid droplet accumulation with fluorescent microscopy and the lipophilic dye BODIPY 493/503. We also monitored changes in the intracellular mass of triacylglycerol, the major form of stored lipid in Tetrahymena (Thompson and Nozawa, 1972), by lipid extraction and thin layer chromatography. We observed marked increases in both triacylglycerol mass and the abundance of lipid droplets in cells starved in 10mM Tris, pH 7.5, compared with cells grown in nutrient-rich Neff media (0.25% proteose peptone, 0.25% yeast extract, 0.5% dextrose, 33.3 µM FeCl₃) (Figure 1). We additionally observed significant accumulation of lipid droplets in the stationary phase of growth, where nutrients are depleted beyond what supports continued cell division (Figure 1). Thus, despite their distinct evolutionary history, Tetrahymena respond to nutritional challenge with an increase in fat storage.

FIGURE 1:

Tetrahymena accumulate lipid droplets following nutrient deprivation. (A) Cells in log phase, stationary phase, or starved for 22 h in 10 mM Tris, pH 7.5 were incubated for 30 min with the lipid droplet marker BODIPY 493/503. Lipid droplets were imaged using a 40x objective and FITC filter; scale bar, 20 µm. (B and C) Lipids extracted from cells in log phase, stationary phase, or starved for 14 h in 10 mM Tris, pH 7.5 were separated by thin-layer chromatography. The intensity of the spots corresponding to the triacylglycerol (TG) standard was quantified by ImageJ and expressed relative to pixel intensity from log-phase samples. Data include averages ± SD from five experiments (B) and a representative thin layer chromatogram (C).

Because these cells are cultured with a total lack of external nutrients, we hypothesize that the increase in triacylglycerol mass results from a pool of fatty acids generated by hydrolysis of membrane lipids. Consistent with this model, Nilsson (1984) previously reported a decrease in Tetrahymena cell size with starvation. We also report notably smaller cells following 22 h in 10 mM Tris, pH 7.5 (Figure 1). Lipid droplet formation may therefore be an evolutionarily conserved strategy that repurposes membrane lipids to create an energy source that supports survival during nutritional stress.

Fatty acids traffic to lipid droplets in starved Tetrahymena

New lipid droplets bud from the ER membrane, the site of triacylglycerol synthesis. Pillai and colleagues and our own unpublished results locate Tetrahymena triacylglycerol synthesis enzymes to the ER (Pillai et al., 2017). Our data show that the synthesis of triacylglycerol and the budding of lipid droplets lagged the onset of starvation in Tetrahymena. At 3 h, changes in triacylglycerol mass were minimal (1.17 ± 0.04 fold increase in starved vs. log-phase cells) and lipid droplets stained with the lipophilic dye BODIPY 493/503 were largely undetectable (Figure 2, column 1). The earliest accumulation of BODIPY 493/503-stained lipid droplets was observed after cells were starved for 6 h in 10 mM Tris, pH 7.5. By 9 h, the BODIPY 493/503-stained lipid droplets were bright and distinct, with further accumulation of lipid droplets continuing through 20 h. This timing closely matches that observed in mammalian cells (Nguyen et al., 2017), suggesting that it may represent the period required for cells to detect nutrient deprivation and subsequently increase lipid droplet formation.

FIGURE 2:

Fatty acids traffic to lipid droplets in starved Tetrahymena. Tetrahymena were pulsed with the fluorescent fatty acid analogue Red C12 in nutrient-rich media overnight before chasing with 10 mM Tris buffer pH 7.5, for 0 to 24 h. Lipid droplets were labeled with BODIPY 493/503 before imaging; scale bar, 20 µm.

To observe the trafficking of fatty acids to newly formed lipid droplets, we used a pulse-chase technique in which cells in nutrient-rich media were pulsed overnight with a fluorescent fatty acid analogue (Red C12) and then washed and incubated under starvation conditions in 10 mM Tris, pH 7.5 lacking the fluorescent dye. At the beginning of the starvation period (chase), the fluorescent signal from Red C12 co-migrated with the triacylglycerol spot on the thin layer chromatogram, suggesting that overnight incubation of nutrient-rich media were sufficient for incorporation of the Red C12 label into the triacylglycerol pool (not shown). Despite the detection of Red C12 in triacylglycerol, we did not detect accumulation of BODIPY-stained lipid droplets in cells that were starved for 0 or 3 h (Figure 2). The earliest stages of lipid droplet synthesis were detected at 6 h, when much of the Red C12 staining colocalized with the small, faint BODIPY 493/503 stained puncta. By 9 h, accumulation of the Red C12 fatty acid analogue was evident in the bright, distinct BODIPY 493/503-stained lipid droplets. Some overlap between the pattern of BODIPY 493/503 and Red C12 staining continued in cells starved of nutrients for 24 h, although the overall number of lipid droplets per cell was noticeably diminished after 20 h. The observed decrease in lipid droplet number following an initial accumulation phase suggests a biphasic response to starvation, characterized by an early stage of triacylglycerol accumulation accompanied by increased lipid droplet formation, followed by a later stage involving triacylglycerol degradation and a corresponding reduction in lipid droplet abundance.

Lipase activity is required for the survival of starved Tetrahymena

Lipid droplets in starved cells may provide a source of energy to support prolonged periods of nutrient deprivation. Indeed, we detected a decrease in Red C12 fluorescence comigrating with triacylglycerol on TLC after 3, 8.5, and 24 h of starvation and a decrease in BODIPY 493/503-stained lipid droplets following 24 h starvation (data not shown and Figure 2, respectively). Both results suggest triacylglycerol turnover in nutrient-deprived Tetrahymena. Two primary mechanisms for mobilizing fatty acids from lipid droplets have been proposed: lipophagy and cytosolic lipases. During lipophagy, lipid droplets are engulfed by autophagosomes that fuse with lysosomes. Hydrolytic enzymes digest the engulfed lipid droplets, releasing free fatty acids. Alternatively, in lipolysis, cytoplasmic triacylglycerol lipases bind to the surface of lipid droplets and directly hydrolyze the stored triacylglycerol.

In our experiments, starving Tetrahymena for 12 h and treating with 10 mM of the lipophagy inhibitor 3-methyladenine (3-MA) did not alter viability. Less than 5% of starved cells treated with 3-MA or vehicle for 1 or 4 h were stained with trypan blue. In contrast, treating starved cells with 25 µM of the pan lipase inhibitor bromoenol lactone (BEL) led to near complete cell death (96±2% of cells stained with trypan blue after 1 h of BEL, compared with <2% of vehicle-treated cells). The severity of this effect is likely exacerbated by the myriad roles of lipases in cells, including a hypothesized role for phospholipases in remodeling membrane lipids for triacylglyerol synthesis. Until triacylglycerol lipases are identified in Tetrahymena, we cannot say unequivocally that lipolysis controls fatty acid release from lipid droplets. Of note, there are no apparent homologues for adipose triglyceride lipase (ATGL) in the Tetrahymena genome, and treating starved Tetrahymena with the ATGL inhibitor ATGListatin does not decrease cell viability (data not shown). Thus, although our data uncover a critical role for lipase activity in starved Tetrahymena, future experiments are needed to identify relevant enzymes and understand their function.

Fatty acids traffic to mitochondria in starved Tetrahymena

Experiments in starved mouse embryonic fibroblasts suggest that fatty acids mobilized from lipid droplets travel to the mitochondria for β-oxidation (Rambold et al., 2015). To determine whether a similar pathway occurs in Tetrahymena, we utilized Red C12 to follow fatty acid trafficking to mitochondria in starved cells. Mito-Tracker identified two populations of mitochondria in starved Tetrahymena (Figure 3A). Cortical mitochondria appeared in rows near the basal bodies at the base of each cilium. An additional pool of centrally localized mitochondria was also detected when we examined the intracellular focal plane. Only the centrally localized mitochondria colocalized with the fatty acid analogue Red C12 (Figure 3A), raising the possibility that subsets of mitochondria may be selectively participating in mitochondrial β-oxidation in starved Tetrahymena.

FIGURE 3:

Fatty acids traffic to mitochondria in starved Tetrahymena. (A) Tetrahymena were starved in 10 mM Tris buffer pH 7.5, for 16 h in the presence of Red C12. Mitochondria were labeled with Mito-Tracker Green before imaging; scale bar, 20 µm. (B) Tetrahymena were pulsed with the fluorescent fatty acid analogue Red C12 in nutrient-rich Neff media overnight before chasing with 10 mM Tris buffer pH 7.5, for 0 to 24 h. Mitochondria were labeled with Mito-Tracker Green before imaging. Colocalization between Red C12 and the mitochondrial marker was observed 8 to 24 h after starvation; scale bar, 20 µm. (C) Quantification of the correlation between the Red C12 and Mito-Tracker signals in the experiment described in B. As a control, Mito-Tracker images were rotated at 180°. Data include mean ± SEM, n = 10 (16 h) or 11 (4 h). Welch's t test; ****; P < 0.0001.

A pulse-chase experiment demonstrated the time-course of Red C12 trafficking to the intracellular subpopulation of mitochondria (Figure 3, B and C). Early timepoints (0–4 h) showed little to no overlap between Mito-Tracker and Red C12 stains. However, at later timepoints (8–24 h), Red C12 was consistently observed in small, numerous Mito-Tracker-stained puncta. The timing of Red C12 appearance in a subpopulation of mitochondria is consistent with a model of fatty acid trafficking to mitochondria after lipid droplet synthesis and hydrolysis of stored lipid. In starved mammalian cells, fatty acids appear to traffic from lipid droplets to the mitochondria (Rambold et al., 2015). Our data support a similar model where lipid droplets serve as a source of lipid for mitochondrial β-oxidation in Tetrahymena.

In Saccharomyces cerevisiae (Seo et al., 2021) and mouse embryonic fibroblasts (Gomes et al., 2011; Rambold et al., 2015), nutrient deprivation initiates massive mitochondrial fusion and remodeling. Tetrahymena show more modest changes to mitochondrial size and organization. Levy and Elliot report an increase in autophagic degradation of mitochondria with starvation (Levy and Elliott, 1968). We observed an increase in the number of small puncta in the intracellular focal plane without evidence of mitochondrial fusion (Figure 3B). The population of mitochondria located at the basal bodies became slightly more disorganized (data not shown), but again failed to show significant evidence of mitochondrial fusion. We consistently observed larger, round structures that stained brightly with Mito-tracker, but not with Red C-12 (Figure 3B). Because we have observed similar large, round structures with unrelated fluorescent markers, and their appearance did not change over the course of our experiments, we did not pursue them further.

Fatty acids traffic to peroxisomes in starved Tetrahymena

Our experiments identify a second critical path for fatty acid catabolism in starved Tetrahymena. Peroxisomes house their own β-oxidation enzymes and serve as the sole source of these reactions in many organisms, including plants and S. cerevisiae. Peroxisomal β-oxidation typically plays a minor role in the oxidation of lipids in animals; however, nutritional stressors, including starvation, may increase flux through these reactions (Orellana et al., 1992). Although peroxisomes have been previously shown to be abundant organelles in Tetrahymena (Baudhuin et al., 1965), the functional importance of these structures has not been elucidated.

In our experiments, we explored whether Tetrahymena might oxidize fatty acids in peroxisomes. Previous identification of peroxisomes in Tetrahymena utilized electron microscopy in fixed cells (Nilsson, 1984). We have generated the molecular tools to identify peroxisomes in living Tetrahymena cells by fusing the peroxisomal targeting signal (PTS1) to the C-terminus of an mCherry or GFP cassette under control of an inducible promoter. This strategy has been used in a variety of other cell types where the machinery for targeting proteins to peroxisomes is conserved (Metz et al., 2017). We confirmed that the pattern of fluorescence observed with mCherry-PTS1 and GFP-PTS1 was consistent with that of a tagged Tetrahymena peroxisomal protein (ABCD1-YFP) (data not shown).

Labeling peroxisomes in Tetrahymena with mCherry-PTS1 or GFP-PTS1 provided opportunities to examine peroxisome dynamics in living, starved cells. Our experiments demonstrate an increase in peroxisome number following nutrient deprivation (Figure 4A). Trafficking of the Red C12 fatty acid analogue to peroxisomes also suggests an important role for Tetrahymena peroxisomes in starved cells. Localization of Red C12 at the peroxisome was visible after 4 h of starvation (Figure 4, B and C), before we observed significant lipid droplet accumulation. At 16 h, we continue to observe some localization of Red C12 at peroxisomes (Figure 3C). Thus, unlike trafficking to mitochondria, lipids may move to peroxisomes during the early stages (4 h) of starvation before Red C12 has localized to lipid droplets. Consistent with this model, there is no overwhelming evidence that Tetrahymena peroxisomes engulf lipid droplets during starvation (Supplemental Figure S1, A and B). Analysis of the BODIPY 493/503 stain and peroxisome marker with Pearson's correlation coefficients suggests there may be increasing overlap, an effect that is difficult to discern visually.

FIGURE 4:

Peroxisomes increase in number and accumulate fatty acids in starved Tetrahymena. (A) Tetrahymena strains expressing the peroxisome marker mCherry-PTS1 were grown to log phase or starved for 18 h in 10 mM Tris, pH 7.5. The number of peroxisomes/cell was counted from images of 75 starved or log-phase cells from four experiments. Representative images are shown; scale bar, 20 µm. (B) Tetrahymena expressing a fluorescent peroxisome label (GFP-PTS1) were pulsed with the fluorescent fatty acid analogue Red C12 in Neff media overnight before chasing in 10 mM Tris buffer pH 7.5, for 4 to 16 h. We saw the most colocalization between Red C12 and our peroxisome marker at early (4–12 h) stages of starvation; scale bar, 20 µm. (C) Quantification of the correlation between the Red C12 and GFP-PTS1 signals in the experiment described in B. As a control, GFP-PTS1 images were rotated at 180^o. Data include mean ± SEM, n = 10. Welch's t test; ****, P < 0.0001; ns, not significant.

Both peroxisomes and mitochondria support cell survival during Tetrahymena starvation

Our research supports a model in which fatty acids traffic to both peroxisomes and mitochondria in nutrient-deprived Tetrahymena. To determine whether fatty acid trafficking to both mitochondria and peroxisomes is functionally important, we utilized chlorpromazine, an inhibitor of peroxisomal fatty acid metabolism, and etomoxir, an inhibitor of mitochondrial fatty acid import (Gotoh et al., 1990; Skorin et al., 1992). Under nutrient-rich conditions, only inhibition of mitochondrial fatty acid import, not peroxisomal fatty acid metabolism, slowed cell growth (Figure 5, A–C). In contrast, both mitochondrial and peroxisomal pathways supported cell survival following starvation (Figure 5D).

FIGURE 5:

Fatty acid catabolism in both peroxisomes and mitochondria contributes to Tetrahymena survival during nutrient stress. Tetrahymena cultures were incubated in Neff media (A–C) or 10 mM Tris, pH 7.5 (D) with 10 µM peroxisomal inhibitor (chlorpromazine, Cpz), 50 µM mitochondrial inhibitor (etomoxir, Eto), no treatment, or both inhibitors. Because the etomoxir was solubilized in DMSO, we used equivalent concentrations of this vehicle for the DMSO control. We followed the growth of cells in nutrient-rich Neff media for 72 h using absorbance to estimate culture density. A representative growth curve is shown in A. The carrying capacities (B) and maximum growth rates (C) were determined from the growth curves. Growth rates are the slopes of regression lines on the curves (change in absorbance every 30 min), and carrying capacity is the maximum absorbance reached by the culture. Carrying capacities and maximum growth rates are shown as the mean ± SD from three separate experiments (P values: ** < 0.05, *** < 0.01).

Peroxisomal and mitochondrial β-oxidation may address different functions in starved Tetrahymena. Mitochondria have traditionally been viewed as catabolic organelles, with the acetyl CoA produced by mitochondrial β-oxidation providing a substrate for the citric acid cycle and further energy production. On the other hand, the output of peroxisomal β-oxidation feeds into anabolic pathways, with products used for glycogenesis. Tetrahymena (like plants and yeast) convert fatty acid to carbohydrate through reactions that occur in the peroxisomes. The glyoxylate pathway converts the acetyl group of acetyl CoA, the 2-carbon product of fatty acid oxidation, to substrates for glycogenesis. Our experiments showing the trafficking of fatty acids to peroxisomes suggest a mechanism for the synthesis of glycogen in starving Tetrahymena. Intriguingly, Levy and Elliot reported that starved Tetrahymena degrade lipid droplets before glycogen stores (Levy and Elliott, 1968), perhaps suggesting that glycogen, not stored triacylglycerol, could serve as the long-term energy depot in Tetrahymena. The significance of this assertion is not yet known.

With these experiments, we expand the study of lipid droplets to include the single-celled ciliate T. thermophila. In our working model, lipid droplets form following autophagy of cellular membranes. Thus, in the starved cell, lipid droplets may serve to protect cells from lipotoxicity (caused by bulk release of fatty acids from autophagy) and allow for regulated control of the rate of subsequent triacylglycerol hydrolysis. Lipids may traffic directly to peroxisomes or flow from lipid droplets to mitochondria. We do not find evidence of lipid droplets or the fluorescent fatty acid marker Red C12 in Tetrahymena vacuoles (Supplemental Figures S2 and S3). This is in contrast with yeast, which moves lipid droplets into vacuoles for degradation during periods of both nutrient depletion (acute glucose restriction, stationary phase growth) and excess (oleate feeding) (van Zutphen et al., 2014; Wang, 2014; Seo et al., 2017). Our results do not exclude lipophagy as a path of lipid droplet degradation in Tetrahymena; however, these reactions are not occurring in the compartment that is labeled with yeast vacuolar stains. On the other hand, as previously discussed, lipases are clearly important in starved Tetrahymena. Bromoenol lactone, a pan lipase inhibitor, is toxic to starved Tetrahymena. The specific neutral lipid lipases that regulate lipid homeostasis in Tetrahymena have not been identified.

Collectively, these experiments introduce Tetrahymena as a model organism that is evolutionarily distinct from those with defined pathways for lipid mobilization and utilization. Tetrahymena provide a eukaryotic cell system for investigating contemporary questions, such as the functional significance and biological machinery supporting distinct roles for mitochondrial subpopulations and the role of organelle contact sites in lipid movement within cells. Additionally, these cells provide an opportunity to explore the functional significance of fatty acid metabolism in peroxisomes and mitochondria. Further elucidation of these and other questions relating to lipid trafficking and storage in Tetrahymena will likely identify both conserved and novel pathways.

MATERIALS AND METHODS

Request a protocol through Bio-protocol

Strains and culture conditions

The wild-type CU428.2 strain of T. thermophila (Tetrahymena Stock Center at Cornell University) was grown at 30°C in modified Neff medium (0.25% proteose peptone, 0.25% yeast extract, 0.5% dextrose, and 33.3 µM FeCl₃). Log phase was defined as cell cultures at a concentration between 2 × 10⁵ and 5 × 10⁵ cells/ml; stationary phase cells were grown for an additional 24 h after reaching log phase and were at a density of ∼1 × 10⁶ cells/ml. For starvation, cells in log phase were washed twice and resuspended in 10 mM Tris, pH 7.5, at a density of ∼3 × 10⁵ cells/ml.

Quantification of triacylglycerol mass

Lipids were extracted from either 1.5 million (for 3-h timepoint) or 500,000 (for 14-h timepoint) starved, stationary, or log-phase cells. Cell pellets were resuspended in 2 ml hexane:isopropanol:water (80:20:2). After 30 min, 0.5 ml hexane:diethyl ether (1:1) was added, and samples vortexed. After 10 min, 1 ml DIH₂O was added, samples vortexed, and incubated for 15 min. The top layers of each sample were collected, dried under nitrogen, and resuspended in chloroform:methanol (1:1). Lipid species were separated by TLC (PE SIL G plates, Whatman) using a solvent of petroleum ether:ethyl ether:acetic acid (80:20:1) and visualized with iodine vapor to show all lipids, or imaged with a LiCor Odyssey imager to locate Red C12 fluorescence.

A triolein standard was used to determine the migration of triacylglycerol. Pixel intensities of bands of interest were determined by ImageJ and expressed relative to log-phase cells. Data are averages ± SD from five experiments

Construction and expression of GFP-PTS1 and mCherry-PTS1

Two DNA oligonucleotides, 5′-CACCCCTTTACATTCTAAATTATGAGGATCC and 5′-GGATCCTCATAATTTAGAATGTAAAGGGGTG (IDT), encoding amino acids PLHSKL (one letter code) were annealed and directionally TOPO cloned into the pENTR-D-TOPO entry vector (Thermo Fisher Scientific) using CACC/GTGG that was added to the 5′ end of each oligonucleotide. The pENTR clone was recombined using the LR Clonase (Thermo Fisher Scientific) reaction into the Gateway-based T. thermophila destination expression vectors, pBS-mCherry-gtw and pBS-GFP-gtw, a gift from Doug Chalker (Malone et al., 2008), which contained an N-terminal MTT1-inducible fluorescent protein (GFP or mCherry) expression cassette cloned upstream of a cycloheximide-resistant Tetrahymena rpl29 locus. Before biolistic transformation, plasmid constructs were digested with HindIII or SacI/PvuI to produce linear plasmid and transformed into Tetrahymena cells starved for 18 h. Transformed cells were selected in Neff medium containing 12.5 µg/ml cycloheximide. To induce expression, 0.5 µg/ml CdCl₂ was added to cell cultures.

Fluorescent microscopy and image processing

Lipid droplets and mitochondria were visualized following 30 min incubation with 1 µg/ml BODIPY 493/503 (Thermo Fisher Scientific) and 100 nM Mito-Tracker Green (Thermo Fisher Scientific). Peroxisomes were visualized following expression of GFP-PTS1 or mCherry-PTS1 by induction for 4 h with 0.5 µg/ml CdCl₂. For fatty acid localization, cells were incubated for 16 h with 2.5 µg/ml BODIPY 558/56 C12 (Red C12, Thermo Fisher Scientific). For vacuole staining, CellTracker CMAC Blue (Thermo Fisher Scientific) was added to 100 µM for 5 min before visualization. For pulse-chase experiments to follow fatty acid trafficking, cells were grown to log phase in modified Neff medium (0.25% proteose peptone, 0.25% yeast extract, 0.5% dextrose, 33.3 µM FeCl₃) in the presence of 1 µg/ml BODIPY 558/56 C12 (Red C12, Thermo Fisher Scientific). The cells were washed twice and resuspended in 10 mM Tris, pH 7.5, for indicated chase times. For co-staining lipid droplets, 1 µg/ml BODIPY 493/503 (Thermo Fisher Scientific) was added for the final 30 min of chase. For all fluorescent microscopy experiments, live cells were examined using the 40X objective on an Olympus BX53 microscope equipped with a FITC or TRITC filter and an ORCA-Flash 4.0LT+ camera. Images were captured using CellSens software.

For Red C12 localization at either mitochondria or peroxisomes and lipid droplet localization at peroxisomes, images were cropped and background was subtracted using a rolling ball radius of 12 pixels using ImageJ. ImageJ “JACoP” software was used to determine the Pearson's Correlation Coefficient, turning one of the two channels 180^o as a control.

Growth curves

Tetrahymena cultures in log-phase were diluted to 12,500 cells/ml in Neff media. A total of 300 µl was placed in eight replicate wells in a 96-well plate. Absorbance at 540 nm was read with a BioTek Epoch 2 plate reader every 30 min for 72 h. Where indicated, 10 µm chlorpromazine (Cayman Chemical), 50 µm etomoxir (Cayman Chemical), or an equivalent volume of DMSO was added to cell cultures at the start of the experiment (time 0).

Cell viability

Cells in log phase were washed and resuspended in 10 mM Tris, pH 7.5. Samples were left untreated, or incubated with 10 µm chlorpromazine (Cayman Chemical), 50 µm etomoxir (Cayman Chemical), both drugs, or an equivalent volume of DMSO (as a control for etomoxir treatment). After 24 h at room temperature, the density of all cells remaining in each culture was determined with a hemocytometer. Dead cells were identified following staining with an equivalent volume of 0.4% trypan blue (Thermo Fisher Scientific).

In other experiments, cells in log phase were washed and incubated in 10 mM Tris, pH 7.5, for 12 h at 30°C. The cells remained in 10 mM Tris, pH 7.5, while treated with 25 µM bromoenol lactone (BEL, Cayman) or 10 mM 3-methyladenine (3-MA, Sigma) for an additional 1 (BEL) or 4 h (3-MA) before trypan blue staining as above.

Abbreviations used:

3-MA

3-methyladenine

BEL

bromoenol lactone

PTS

peroxisomal targeting signal.